# Source paper archive - Source title: Anionic lipids stabilize etioplast prolamellar bodies and facilitate thylakoid biogenesis during etioplast-to-chloroplast differentiation - Author(s): Yoshihara, A.; Uwatoko, R.; Kobayashi, K.; Nagata, N.; Kobayashi, K. - Source landing URL: https://www.biorxiv.org/content/10.64898/2026.10.31.763292v1 - PDF URL: https://www.biorxiv.org/content/10.64898/2026.10.31.763292v1.full.pdf - PDF SHA-256: 7ef7026cfcfac1db7dd002f0f0642274b890736a3399680fc18ef1b3c4162a1f - Converter: pymupdf4llm 0.0.27 - Limits: Automated PDF text extraction; reading order, equations, tables and formatting may be incomplete or incorrect. Images are omitted, and no OCR was performed. Markup and links are neutralized for safe downloading. Consult the original PDF for the authoritative version. --- [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **Date of submission:** October 31, 2026 **Title:** Anionic lipids stabilize etioplast prolamellar bodies and facilitate thylakoid biogenesis during etioplast-to-chloroplast differentiation **Authors:** Akiko Yoshihara [1)† ] (akiko.yoshihara@riken.jp), Risa Uwatoko [2)††] (ruwatoko@jeol.co.jp), Keiko Kobayashi [2)] (kkobayashi@fc.jwu.ac.jp), Noriko Nagata [2)] (n-nagata@fc.jwu.ac.jp), Koichi Kobayashi [1,3)] (kkobayashi@omu.ac.jp) **Institutions:** 1) Department of Biology, Graduate School of Science, Osaka Metropolitan University, 132-1 Morinomiya 2-chome, Joto-ku, Osaka-shi, Osaka 536-0025, Japan 2) Department of Chemical and Biological Sciences, Faculty of Science, Japan Women’s University, Bunkyo-ku, Tokyo 112-8681, Japan 3) Faculty of Liberal Arts, Science and Global Education, Osaka Metropolitan University, 132-1 Morinomiya 2-chome, Joto-ku, Osaka-shi, Osaka 536-0025, Japan **Present addresses:** - RIKEN Center for Sustainable Resource Science (CSRS), 1-7-22 Suehiro-cho, Tsurumi-ku, Yokohama, Kanagawa 230-0045, Japan - † JEOL Ltd., Akishima, Tokyo 196-8558, Japan **Corresponding author: Koichi Kobayashi** Faculty of Liberal Arts, Science and Global Education, Osaka Metropolitan University, 2-1-132 Morinomiya, Joto-ku, Osaka-shi, Osaka 536-8525, Japan Phone: +81 6 6167 1221 Email: kkobayashi@omu.ac.jp ORCID ID: https://orcid.org/0000-0003-3726-4790 **Number of tables:** 0 **Number of figures:** 8 **Word count:** 5175 words, including Introduction, Results, Discussion, and Acknowledgements **Supplementary data:** Supplementary Video S1, Supplementary Tables S1-S6. **Running title:** Anionic lipids stabilize PLBs and facilitate thylakoid biogenesis 1 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 1 **Anionic lipids stabilize etioplast prolamellar bodies and facilitate thylakoid biogenesis** 2 **during etioplast-to-chloroplast differentiation** 3 Akiko Yoshihara [1)], Risa Uwatoko [2)], Keiko Kobayashi [2)], Noriko Nagata [2)], Koichi Kobayashi [1,3)*] 4 1) Department of Biology, Graduate School of Science, Osaka Metropolitan University, 132-1 5 Morinomiya 2-chome, Joto-ku, Osaka-shi, Osaka 536-0025, Japan 6 2) Department of Chemical and Biological Sciences, Faculty of Science, Japan Women’s 7 University, Bunkyo-ku, Tokyo 112-8681, Japan 8 3) Faculty of Liberal Arts, Science and Global Education, Osaka Metropolitan University, 132-1 9 Morinomiya 2-chome, Joto-ku, Osaka-shi, Osaka 536-0025, Japan 10 *Correspondence: kkobayashi@omu.ac.jp 11 12 **Abstract** 13 Etioplasts of dark-grown angiosperm seedlings contain prolamellar bodies (PLBs), lattice-like 14 membranes that are quickly transformed into thylakoids upon illumination. 15 Phosphatidylglycerol (PG) and sulfoquinovosyldiacylglycerol (SQDG) are major anionic lipids 16 in these membranes, but their roles in etioplast-to-chloroplast differentiation remain unclear. We 17 analyzed _Arabidopsis thaliana_ mutants with reduced PG biosynthesis ( _pgp1-1_ ), complete 18 SQDG deficiency _(sqd1_ ), and the _pgp1-1 sqd1_ double mutant. Although _pgp1-1_ etioplasts 19 showed a loosened PLB lattice when fixed under white light, they retained regular, 20 wild-type-like lattices when fixed under infrared light, suggesting that their PLBs were more 21 susceptible to light-induced disassembly. The rapid PLB disassembly in _pgp1-1_ upon 22 illumination was accompanied by rapid conversion of dark-accumulated protochlorophyllide to 23 chlorophyll and lamellar membrane formation, suggesting a link between membrane remodeling 24 and chlorophyll formation. However, subsequent chlorophyll accumulation through de novo 25 biosynthesis was retarded in illuminated _pgp1-1_ seedlings, whereas photosynthetic electron 26 transport was quickly activated. Although _sqd1_ showed no obvious defects, _pgp1-1 sqd1_ 27 exhibited more severe defects than _pgp1-1_, particularly in internal membrane formation, 28 chlorophyll accumulation, photosynthetic activation, and photosynthetic gene expression. These 29 findings suggest that PG is crucial for membrane-associated processes during 30 etioplast-to-chloroplast differentiation, while SQDG may help maintain chloroplast 31 development when PG availability is limited. 32 **Key words:** chlorophyll, chlorophyllide, chloroplast, etioplast, phosphatidylglycerol, 33 photosynthesis, prolamellar body, protochlorophyllide, sulfoquinovosyldiacylglycerol, thylakoid 34 membrane 35 2 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 36 **Introduction** 37 Cotyledon cells of dark-grown angiosperm seedlings develop chloroplast precursors called 38 etioplasts. Etioplasts contain a three-dimensional lattice of branched membrane tubules termed 39 prolamellar bodies (PLBs). Protochlorophyllide (Pchlide), an intermediate of chlorophyll (Chl) 40 biosynthesis, accumulates in PLBs instead of Chl. Most Pchlide is bound at the catalytic site of 41 light-dependent NADPH-Pchlide oxidoreductase (LPOR), which also functions as a 42 membrane-remodeling protein essential for PLB formation (Schoefs and Franck, 2003; 43 Solymosi and Schoefs, 2010; Nguyen _et al._, 2021). LPOR forms oligomeric ternary complexes 44 with Pchlide and NADPH, and these complexes helically coat the PLB membranes (Floris and 45 Kühlbrandt, 2021). 46 Upon illumination, Pchlide is instantaneously converted into chlorophyllide (Chlide) by 47 LPOR, followed by the gradual degradation of LPOR, leading to the disassembly of PLBs into 48 lamellar membranes (Solymosi and Schoefs, 2010; Kowalewska _et al._, 2016). Although Chl 49 synthase, which catalyzes the esterification of Chlide with a phytol chain, is considered inactive 50 when the PLB structure is highly organized, it becomes activated upon PLB disassembly 51 (Lindsten _et al._, 1990). In parallel, illumination upregulates the expression of 52 photosynthesis-associated genes, thereby promoting Chl and photosynthetic protein biosynthesis 53 (Ma _et al._, 2001; Pipitone _et al._, 2021). Along with LPOR degradation, the accumulation of 54 photosynthetic membrane protein complexes such as light-harvesting complex II (LHCII) drives 55 the transformation of the tubular PLB lattice into stacked thylakoid membranes and the 56 differentiation of etioplasts to chloroplasts (Fujii _et al._, 2019 _b_ ). 57 The composition of pigments and proteins substantially changes during the 58 PLB-to-thylakoid transformation (Blomqvist _et al._, 2008; Plöscher _et al._, 2011; Garmash _et al._, 59 2013; Pipitone _et al._, 2021). By contrast, the lipid composition of PLBs is similar to that of 60 thylakoid membranes (Selstam and Sandelius, 1984; Dorne _et al._, 1990), and thus PLBs are 61 regarded as a lipid reservoir for the formation of initial thylakoid membranes (Solymosi and 62 Schoefs, 2010; Fujii _et al._, 2019 _b_ ). The PLB lipid bilayer is mainly composed of two 63 galactolipids, monogalactosyldiacylglycerol (MGDG) and digalactosyldiacylglycerol (DGDG), 64 which account for ~50% and ~30% of total PLB lipids, respectively (Selstam and Sandelius, 65 1984). These galactolipids are required for etioplast development (Fujii _et al._, 2017, 2018) and 66 for etioplast-to-chloroplast differentiation (Fujii _et al._, 2019 _a_ ). In addition, PLBs contain 67 sulfoquinovosyldiacylglycerol (SQDG), a sulfur-containing glycolipid, and the phospholipid 68 phosphatidylglycerol (PG), each accounting for ~10 mol% of total PLB lipids (Selstam and 69 Sandelius, 1984). 70 PG and SQDG are anionic lipids whose polar head groups are negatively charged at 71 physiological pH. PG is indispensable for photosynthesis in oxygenic photosynthetic organisms 3 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 72 including cyanobacteria, algae, and plants (Wada and Murata, 2007; Kobayashi _et al._, 2024). 73 Many PG molecules are specifically associated with photosystem I (PSI) and photosystem II 74 (PSII), where they likely play essential roles in the structure and function of these complexes 75 (Yoshihara and Kobayashi, 2022). Accordingly, in _Arabidopsis thaliana_, a T-DNA insertional 76 knockout mutation ( _pgp1-2_ ) of phosphatidylglycerophosphate synthase 1 (PGP1), which results 77 in the complete loss of plastidic PG biosynthesis, causes severe impairments of photosynthetic 78 reactions and thylakoid membrane development (Hagio _et al._, 2002; Kobayashi _et al._, 2016). 79 Meanwhile, an approximately 80% reduction in PGP1 activity caused by a single amino acid 80 substitution ( _pgp1-1_ ) results in a ~30% decrease in leaf PG content and a pale-green phenotype, 81 without a severe impairment of photosynthetic capacity (Xu _et al._, 2002). However, additional 82 loss of SQDG biosynthesis in the _pgp1-1_ background leads to defects in photosynthetic electron 83 transport, Chl accumulation, thylakoid development, and growth under light conditions (Yu and 84 Benning, 2003; Yoshihara _et al._, 2021). Because the loss of SQDG biosynthesis alone does not 85 impair these processes in _A. thaliana_ (Yu _et al._, 2002; Yoshihara _et al._, 2021), SQDG likely 86 compensates for some functions of PG when PG levels are insufficient. Indeed, SQDG content 87 increases as PG content decreases under phosphate-deficient conditions, thereby maintaining the 88 total level of anionic lipids (Essigmann _et al._, 1998). Because mutants lacking SQDG 89 biosynthesis exhibit decreased photosynthetic activity under phosphate-deficient conditions 90 (Yoshihara _et al._, 2021), the compensatory role of SQDG may be particularly important during 91 phosphate deficiency. 92 We have recently reported that PG also plays an important role in etioplast development, 93 whereas SQDG supports the function of PG (Yoshihara _et al._, 2024). Etioplasts in _pgp1-1_ 94 cotyledons grown in the dark show a loosened PLB lattice structure, impaired Pchlide 95 biosynthesis particularly at the Mg-insertion step, and altered behavior of 96 Pchlide-LPOR-NADPH complexes. In contrast, the _sqd1_ and _sqd2_ mutants, which completely 97 lack SQDG biosynthesis due to knockout mutations in _SQD1_ and _SQD2_, respectively, show no 98 obvious defects in etioplast development. However, the _pgp1-1 sqd1_ and _pgp1-1 sqd2_ double 99 mutants exhibit more severe defects than _pgp1-1_ during etioplast development. These findings 100 suggest that anionic lipids, particularly PG, are required for membrane-protein organization 101 even before the accumulation of photosynthetic membrane proteins in the dark (Yoshihara _et al._, 102 2024). 103 A recent review proposed a model in which lipid biosynthesis triggers Chl and 104 photosynthetic protein biosynthesis and promotes thylakoid membrane biogenesis and 105 photosynthetic activation (Fujii _et al._, 2025). During etioplast-to-chloroplast differentiation, 106 galactolipid biosynthesis is required for the efficient organization of the Chl biosynthesis 107 pathway and for photosynthetic gene expression in plastids and the nucleus (Fujii _et al._, 2019 _a_, 4 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 108 2024). In contrast, it remains unclear whether anionic lipids play specific roles in these 109 processes. In this study, we investigated the de-etiolation processes of dark-grown seedlings of 110 _pgp1-1_, _sqd1_, and the _pgp1-1 sqd1_ double mutant, including rapid conversion of Pchlide to Chl 111 via Chlide, PLB disassembly, de novo Chl biosynthesis, photosynthetic activation, and 112 photosynthetic gene expression. The results highlight the specific roles of plastid anionic lipids, 113 particularly PG, in etioplast-to-chloroplast differentiation. 114 115 **Materials and methods** 116 **Plant materials and growth conditions** 117 The wild type Col-0 (WT), _sqd1_ (Okazaki _et al._, 2009), _pgp1-1_ (Xu _et al._, 2002), and _pgp1-1_ 118 _sqd1_ plants (Yoshihara _et al._, 2021), all in the Columbia background of _A. thaliana_, were used 119 in this study. 120 Surface-sterilized seeds were sown on half-strength Murashige and Skoog medium 121 (pH 5.7 adjusted with KOH) containing 1% (w/v) sucrose and solidified with 0.8% (w/v) agar, 122 cold-treated at 4 °C for 3–4 d in the dark, and then exposed to white light ( ∼ 80 μ mol photons 123 m � ² s � ¹) at 23 °C for 4 h to synchronize germination. Seedlings grown in the dark for 4 d were 124 either used for experiments or subsequently illuminated with continuous white light ( ∼ 80 μ mol 125 photons m � ² s � ¹) at 23 °C for deetiolation. 126 127 **Transmission electron microscopy and scanning electron microscopy analysis** 128 The samples were processed for electron microscopy based on the procedure reported by Fujii _et_ 129 _al._ (2017), with minor modifications. Except for samples fixed under infrared light, cotyledons 130 were dissected immediately after sampling and immersed in a fixative containing 4% 131 glutaraldehyde and 4% paraformaldehyde at 4°C overnight. After fixation, the samples were 132 postfixed in 2% OsO 4 supplemented with 1.5% K 3 [Fe(CN) 6 ] at 4°C for 2 h, dehydrated through 133 an ethanol series, transferred to propylene oxide, infiltrated with Spurr’s resin, and embedded. 134 For fixation under infrared light, dark-grown seedlings were handled under 850-nm 135 infrared illumination (GR10-IR, SHODENSHA, Osaka, Japan) using an infrared camera 136 (GR200HD2-IR, SHODENSHA). Because fine dissection was difficult under these conditions, 137 whole seedlings were first immersed in the same fixative, wrapped in aluminum foil, and stored 138 at 4°C for 2 days to allow fixation without exposure to visible light. The fixed seedlings were 139 then dissected and processed for postfixation, dehydration, resin infiltration, and embedding as 140 described above. 141 Ultrathin sections (80 nm thick) were stained with uranyl acetate and lead citrate and 142 observed by transmission electron microscopy (TEM; JEM-1400, JEOL, Tokyo, Japan) at high 143 magnification to examine fine structural details of plastid internal membranes. For scanning 5 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 144 electron microscopy (SEM), which was used to examine larger tissue areas containing 145 numerous plastid profiles, the stained sections were further coated with OsO 4 and observed by 146 field-emission SEM (SU8220, Hitachi, Tokyo, Japan). Plastid profiles in the SEM images were 147 manually classified and quantified based on their internal membrane organization. Plastid 148 profiles in which internal membranes were absent or could not be clearly distinguished were 149 excluded from the analysis. In addition, to assess the three-dimensional organization of _pgp1-1_ 150 _sqd1_ plastids, serial ultrathin sections, each 80 nm thick, were prepared and observed by SEM. 151 152 **In situ low-temperature fluorescence spectroscopy** 153 For analysis of the time-dependent changes in low-temperature (77 K) fluorescence spectra 154 during dark incubation after a single flash, approximately 30 etiolated seedlings grown for 4 d 155 in the dark were placed in a clear microtube under dim green light, exposed to a 0.7-ms single 156 flash (PZ42X, Sunpak), and kept in darkness at 13 °C for the indicated time. The microtube was 157 then placed in a liquid nitrogen-filled Dewar flask positioned in the cuvette holder of a 158 fluorometer (RF-6000, Shimadzu, Kyoto, Japan). Fluorescence emission spectra were recorded 159 in liquid nitrogen under 440-nm excitation. 160 For analysis of light-induced transitions of low-temperature fluorescence spectra, 161 approximately 30 etiolated seedlings were placed in a clear microtube, exposed to continuous 162 white light ( ∼ 80 μ mol photons m � ² s � ¹) at 23 °C for the indicated time after a flash, and used 163 for recording fluorescence emission spectra under 435-nm excitation as described above. 164 165 **Immunoblot analysis** 166 Seedlings grown for 4 d in the dark were exposed to white light ( ∼ 80 μ mol photons m [−2] s [−1] ) for 167 5 minutes and then frozen in liquid nitrogen. Immunoblot analysis, from total protein extraction 168 to detection and quantification of total LPOR proteins, was performed as described previously 169 (Yoshihara _et al._, 2024). 170 171 **Determination of porphyrin content** 172 To determine the ratio of Chl _a_ to Chlide _a_ after a flash, 4-d-old etiolated seedlings were 173 incubated in the dark for the indicated time after a single flash, immediately frozen in liquid 174 nitrogen, homogenized in N,N-dimethylformamide, and analyzed as described previously (Fujii 175 _et al._, 2017) using the high � performance liquid chromatography (HPLC) system (Yoshihara _et_ 176 _al._, 2024). The ratio of Chl _a_ to Chlide _a_ was determined from respective fluorescence peak 177 areas measured at 670 nm under 430-nm excitation. 178 To determine the porphyrin composition after continuous illumination, 4-d-old 179 etiolated seedlings were exposed to white light ( ∼ 80 μ mol photons m [−2] s [−1] ) for the indicated 6 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 180 time, extracted in N,N-dimethylformamide at 4 °C in the dark for 1 d, and analyzed by HPLC. 181 Pchlide content was determined by measuring the fluorescence at 634 nm under 440 nm 182 excitation and using a known concentration of Pchlide as a standard. To determine Chlide _a_ and 183 Chl _a_ contents, peak areas in the fluorescence chromatograms recorded at 670 nm under 430-nm 184 excitation were converted to amounts using a calibration based on the photoactive Pchlide 185 content determined before illumination, which corresponds to the Chl(ide) _a_ content formed 186 after the flash. 187 To analyze Chl accumulation, 4-d-old etiolated seedlings were exposed to white light 188 ( ∼ 80 μ mol photons m [−2] s [−1] ) for the indicated time and then incubated in 80% (v/v) acetone at 189 4 °C in the dark for 1 d. Chl content was determined by measuring fluorescence emission at 670 190 nm under 435-nm excitation using a spectrofluorometer (RF-6000), with a known concentration 191 of Chl as a standard. At 0 h, the amount of Pchlide described in Yoshihara _et al._ (2024) was 192 used instead of Chl content. The Chl accumulation rate was calculated by dividing the Chl 193 content accumulated during the indicated time period by the duration of the period. 194 195 **Photosynthetic efficiency analysis** 196 Chl fluorescence parameters were analyzed using an imaging pulse amplitude modulation 197 (PAM) fluorometer (IMAGING-PAM MINI, Walz, Effeltrich, Germany) and ImagingWinGigE 198 software. Etiolated seedlings exposed to continuous white light ( ∼ 80 μ mol photons m [−2] s [−1] ) for 199 the indicated time were dark-adapted for 5 min. Minimal fluorescence ( _F_ _o_ ) was determined 200 using the lowest measuring light (460 nm, ~0.2 μ mol photons m [−2] s [−1] at a frequency of 2 Hz), 201 and maximal fluorescence ( _F_ _m_ ) was obtained by applying a saturating pulse (460 nm, ~3400 202 μ mol photons m [−2] s [−1] for 720 ms). The seedlings were then illuminated with actinic light (460 203 nm, ∼ 75 μ mol photons m [−2] s [−1] ) for 10 min, and a second saturating pulse was applied to 204 determine the steady-state fluorescence ( _F_ _s_ ) and maximal fluorescence under actinic 205 illumination _(F_ ′ _m_ ), respectively. The maximum quantum yield of PSII ( _F_ _v_ / _F_ _m_ ) was calculated as 206 ( _F_ _m_ − _F_ _o_ )/ _F_ _m_ . The effective quantum yield of PSII ( Φ _PSII_ ) and the quantum yields of regulated 207 ( Φ _NPQ_ ) and non-regulated energy dissipation ( Φ _c_ ) were calculated according to Guadagno _et al._ 208 (2010) as follows: Φ _PSII_ = ( _F_ ′ _m_ − _F_ _s_ )/ _F_ ′ _m_, Φ _NPQ_ = _F_ _s_ / _F_ ′ _m_ − _F_ _s_ / _F_ _m_, and Φ _c_ = _F_ _s_ / _F_ _m_ . 209 210 **Reverse transcription quantitative PCR analysis** 211 mRNA level was quantified as described (Yoshihara et al., 2024) with some modifications. 212 Complementary DNA was amplified by 2-step thermal cycling consisting of an initial 213 denaturation step at 95 °C for 30 s followed by 40 cycles of 5 s at 95 °C and 30 s at 60 °C with 214 the use of the THUNDERBIRD Next SYBR qPCR Mix (TOYOBO, Osaka, Japan) and 200 nM 215 gene-specific primers (Supplementary Table S1). The thermal cycling and signal detection were 7 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 216 performed in duplicate using the CFX Connect (Bio-Rad, Hercules, CA, USA). The relative 217 abundance of all transcripts amplified was normalized to the mean expression levels of the 218 reference genes _ACTIN8_ and _UBIQUITIN11_ according to Pfaffl (2001). 219 220 **Statistical analysis** 221 For multiple comparison tests, one-way ANOVA with Tukey’s honestly significant difference 222 post hoc test was performed using the multcomp R package (Hothorn _et al._, 2008). A 223 significance level of P < 0.05 was used. 224 225 **Accession numbers** 226 Sequence data of the genes investigated in this article can be found in The Arabidopsis 227 Information Resource under the following accession numbers: _PGP1_ (AT2G39290), _SQD1_ 228 (AT4G33030), _ACT8_ (AT1G49240), _UBQ11_ (AT4G05050), _CHLH_ (AT5G13630), _HEMA1_ 229 (AT1G58290), _LHCA4_ (AT3G47470), _LHCB6_ (AT1G15820), _psaA_ (ATCG00350), and _psbA_ 230 (ATCG00020). 231 232 **Results** 233 **PLBs deficient in anionic lipids are prone to light-induced disassembly** 234 We previously reported that cotyledon cells of dark-grown _pgp1-1_ seedlings contained loosened 235 PLBs in etioplasts (Yoshihara _et al._, 2024), resembling those observed in WT etioplasts after ~1 236 h of illumination (Kowalewska _et al._, 2016). In the previous study, etiolated cotyledons were 237 fixed under white light, which took around 5 min (Yoshihara _et al._, 2024). We hypothesized that 238 the short-term exposure to white light during fixation caused rapid PLB disassembly in _pgp1-1_ 239 but not in WT. To test this hypothesis, we fixed etiolated seedlings under infrared light (850 nm), 240 at a wavelength well beyond the absorption bands of Pchlide–LPOR complexes (Shibata, 1957). 241 In dark-grown WT and _sqd1_ seedlings fixed under infrared light, etioplasts exhibited 242 highly regular PLB lattices similar to those observed under white-light fixation (Fig. 1A). Even 243 in _pgp1-1_ seedlings, most etioplasts showed regular lattice structures of PLBs under infrared 244 fixation, in contrast to their loosened appearance under white light (Yoshihara _et al._, 2024). 245 These data suggest that _pgp1-1_ forms WT-like PLBs in darkness, but that their lattice structures 246 are rapidly disassembled upon brief light exposure during fixation. Nevertheless, quantitative 247 classification of internal membrane structures revealed that 5.4% (13 of 242) etioplast profiles 248 in ultrathin sections of _pgp1-1_ exhibited disordered internal membrane networks instead of 249 lattice structures (Fig. 1B), suggesting that decreased PG biosynthesis results in heterogeneous 250 defects in PLB lattice formation among etioplasts. 251 Our previous study also showed that etioplasts in the _pgp1-1_ _sqd1_ double mutant fixed 8 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 252 under white light exhibited disordered membrane networks or dense membrane aggregates 253 instead of well-organized PLB lattices (Yoshihara _et al._, 2024). When fixed under infrared light, 254 94.8% (183 of 193) etioplast profiles in ultrathin sections of _pgp1-1_ _sqd1_ exhibited only 255 irregular membrane structures lacking PLB lattices, whereas 3.1% (6 of 193) showed irregular 256 membranes containing partial lattice structures and 2.1% (4 of 193) exhibited regular PLB 257 lattices throughout the structure (Fig. 1). However, three-dimensional analysis based on serial 258 sections revealed a higher proportion of lattice-containing structures: 22.1% (43 of 194) of 259 etioplasts contained partial lattice structures, whereas 5.6% (11 of 194) exhibited only regular 260 lattices. Serial section analysis further showed that the partial lattice structures were not formed 261 as a central core enclosed by irregular membranes, but instead occupied part of the internal 262 membrane system, extending from the periphery to the interior (Supplementary Video S1). 263 Considering that no lattice structures were observed in _pgp1-1_ _sqd1_ etioplasts when fixed under 264 white light (Yoshihara _et al._, 2024), these lattices are likely disassembled rapidly during the 265 fixation process, as in _pgp1-1_ . 266 267 **Anionic lipid deficiency accelerates dissociation of oligomeric Chlide-LPOR complexes** 268 **and Chlide esterification** 269 We previously reported that the peak shift of Chlide fluorescence following flash-induced 270 Pchlide-to-Chlide conversion occurred more rapidly in _pgp1-1_ and _pgp1-1_ _sqd1_ than in WT 271 (Yoshihara _et al._, 2024), suggesting a relationship between structural changes in PLBs and the 272 dissociation of oligomeric Chlide-LPOR-NADP [+] complexes (Franck _et al._, 1999; Schoefs and 273 Franck, 2008). To examine how anionic lipids affect the initial phase of this dissociation process, 274 etiolated seedlings of WT and anionic lipid mutants were kept in the dark at a reduced 275 temperature (13 °C) for 1, 30, 60, or 120 s after a flash, and their fluorescence spectra were 276 measured at 77 K. 277 In all samples, a fluorescence peak at ~690 nm, originating from the 278 Chlide-LPOR-NADP [+] oligomer (Schoefs, 2001), appeared 1 s after a flash (Fig. 2A), indicating 279 rapid photoconversion of Pchlide in the oligomeric ternary complex to Chlide, as observed 280 previously (Yoshihara _et al._, 2024). In WT and _sqd1_, the 690-nm peak was red-shifted to ~695 281 nm after 30 s (Fig. 2B), which presumably corresponds to fluorescence from the oligomeric 282 Chlide-LPOR-NADPH complexes (El Hamouri _et al._, 1981; Oliver and Griffiths, 1982). In 283 _pgp1-1_ and _pgp1-1 sqd1_, the peak wavelengths after 30 s appeared to be shorter than those in 284 WT and _sqd1_, but the differences were not significant. In WT, the peak had blue-shifted to 693.5 285 nm by 120 s, consistent with gradual dissociation of the oligomeric complexes. Similar peak 286 shifts were observed in the anionic lipid mutants. However, the peak wavelengths in _pgp1-1_ and 287 particularly _pgp1-1 sqd1_ were significantly shorter than that in WT, whereas _sqd1_ showed a 9 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 288 trend toward a longer peak wavelength than WT. In addition, a shoulder band at around 675 nm, 289 which may originate from free Chlide released from the LPOR active site (Franck _et al._, 1999; 290 Schoefs and Franck, 2008), appeared in all samples 30 s after illumination (Fig. 2A). The band 291 was more intense in _pgp1-1_ and even more pronounced in _pgp1-1 sqd1_ than in WT and _sqd1_ 292 (Fig. 2C). In parallel with these band shifts, a small shoulder or peak was observed at ~653 nm 293 after 60 s (Fig. 2A), indicating the regeneration of photoactive Pchlide-LPOR complexes 294 (Franck _et al._, 1999; Schoefs, 2001). The band was more distinct in _pgp1-1_ and _pgp1-1 sqd1_ 295 than in WT and _sqd1_ . These data suggest that the dissociation or structural rearrangement of 296 Chlide-LPOR-NADPH complexes, release of free Chlide, and regeneration of photoactive 297 Pchlide-LPOR complexes after a flash are accelerated in _pgp1-1_ and particularly _pgp1-1 sqd1_ . 298 Structural states of PLBs have also been reported to influence the activity of Chl 299 synthase in esterifying Chlide with a phytyl chain (Lindsten _et al._, 1990, 1993). To evaluate 300 Chlide esterification, we determined the Chl _a_ /Chlide _a_ ratio in etiolated seedlings after 0.5 and 301 1 min of dark incubation following the flash (Fig. 2D). At 1 min after the flash, _pgp1-1_ and 302 _pgp1-1 sqd1_ showed more than a twofold higher Chl _a_ /Chlide _a_ ratio than WT and _sqd1_, 303 suggesting accelerated phytylation of Chlide _a_ to Chl _a_ in these mutants. 304 305 **Anionic lipid deficiency causes rapid PLB-to-lamellar membrane transformation upon** 306 **illumination** 307 To investigate whether plastid anionic lipid deficiency affects PLB-to-thylakoid membrane 308 transformation during etioplast-to-chloroplast differentiation, we performed time-course 309 observations of plastid morphology upon illumination (Fig. 3A). In WT and _sqd1_, lamellar 310 membranes gradually extended from the periphery of loosened PLBs, accompanied by 311 progressive reduction in the size of PLB remnants over 6 h of illumination. While PLB 312 remnants were still present in some plastids in WT and _sqd1_ after 6 h, almost all PLB lattices 313 had already transformed into lamellar membranes by 3 h in _pgp1-1_ . After 9 h of illumination, no 314 marked differences in the morphology of developing thylakoids were observed among WT, _sqd1_, 315 and _pgp1-1_ . These observations suggest that reduced PG biosynthesis accelerates 316 PLB-to-lamellar membrane transformation but does not markedly affect subsequent thylakoid 317 development, whereas loss of SQDG biosynthesis has little effect on these processes. In _pgp1-1_, 318 disordered internal membrane networks and dense membrane aggregates, similar to those 319 observed in dark-grown etioplasts, were also detected during illumination. Even after 24 h of 320 illumination, 2.0% (6 of 296) of plastid profiles in ultrathin sections of _pgp1-1_ cotyledons 321 exhibited these abnormal membrane structures without forming thylakoids (Fig. 3B). 322 Rapid transformation from PLBs to lamellar membranes was also observed in _pgp1-1_ 323 _sqd1_, with extended lamellar membranes and no detectable PLB remnants as early as 3 h after 10 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 324 illumination (Fig. 3A). However, after 24 h of illumination, 26.8% (364 of 1358) of plastid 325 profiles in ultrathin sections of _pgp1-1_ _sqd1_ exhibited abnormal membrane structures instead of 326 thylakoid membranes (Fig. 3B), suggesting that a subset of plastids retained abnormal 327 membrane structures rather than developing normal thylakoid membranes, as observed in a few 328 plastids in _pgp1-1_ . 329 330 **Anionic lipid deficiency accelerates Pchlide-to-Chl conversion upon illumination** 331 LPOR is gradually degraded upon illumination (Kay and Griffiths, 1983; Armstrong _et al._, 332 1995), and this degradation is thought to be associated with PLB disassembly (Fujii _et al._, 333 2019 _b_ ). To assess whether the abnormally rapid PLB disassembly in _pgp1-1_ and _pgp1-1_ _sqd1_ 334 results from accelerated LPOR degradation, we determined total LPOR levels after 5 min of 335 illumination (Fig. 4A). LPOR content in _pgp1-1_ _sqd1_ was 40% lower than in WT, whereas that 336 in _sqd1_ and _pgp1-1_ was comparable to WT. A similar 40% reduction was previously observed in 337 dark-grown _pgp1-1_ _sqd1_ seedlings without illumination (Yoshihara _et al._, 2024), indicating that 338 light-induced LPOR degradation is not notably accelerated in these mutants. 339 In etioplasts, Pchlide exists in photoactive and nonphotoactive forms (Schoefs, 2001). 340 Immediately after a flash, both Chlide (derived from photoactive Pchlide) and Pchlide 341 (nonphotoactive forms) were detected, with a small fraction of Chlide further converted to Chl 342 (Fig. 4B). Total porphyrin levels after the flash were lower in _pgp1-1_ and _pgp1-1_ _sqd1_ than in 343 WT and _sqd1_, consistent with reduced accumulation of Pchlide in these mutants during dark 344 growth (Yoshihara _et al._, 2024). After 5 min of illumination following the flash, most porphyrin 345 intermediates were converted to Chl in _pgp1-1_ and _pgp1-1_ _sqd1_, with no detectable Pchlide 346 remaining, whereas substantial and minor amounts of Chlide and Pchlide, respectively, 347 remained in WT and _sqd1_ . After 10 min, neither Pchlide nor Chlide was detected in _pgp1-1_ and 348 _pgp1-1_ _sqd1_, although small amounts of these intermediates were still detected in WT and _sqd1_ 349 even after 20 min. These results indicate that Pchlide-to-Chl conversion via Chlide occurs more 350 rapidly in _pgp1-1_ and _pgp1-1_ _sqd1_ . 351 352 **Rapid PLB transformation in anionic lipid mutants does not enhance chloroplast** 353 **development** 354 To assess whether accelerated PLB-to-lamellar membrane transformation in _pgp1-1_ and _pgp1-1_ 355 _sqd1_ promotes the establishment of chloroplast functions, we performed time-course analyses of 356 Chl accumulation and photosynthetic capacity during deetiolation (Fig. 5). In _pgp1-1_, the rate of 357 Chl accumulation was lower than those in WT and _sqd1_, particularly during the early stage of 358 deetiolation, resulting in consistently lower Chl content throughout the illumination period. 359 Furthermore, the _pgp1-1_ _sqd1_ double mutant showed a markedly reduced rate of Chl 11 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 360 accumulation and consequently a substantially lower Chl content after 24 h of illumination. 361 Next, we evaluated photosynthetic capacity during deetiolation using PAM Chl 362 fluorescence measurements (Fig. 6). During the first 3 h of illumination, Chl fluorescence was 363 too weak in all plants to yield reliable photosynthetic parameters. Between 6 and 24 h after 364 illumination, _F_ _v_ _/F_ _m_ increased in _sqd1_ and _pgp1-1_ to levels similar to those in WT. Meanwhile, 365 _pgp1-1_ _sqd1_ seedlings consistently showed lower _F_ _v_ _/F_ _m_ than WT. Φ _PSII_, which represents PSII 366 photochemical efficiency under actinic light and is affected by the overall performance of the 367 photosynthetic electron transport system (Baker, 2008), was also lower in _pgp1-1_ _sqd1_ . In 368 contrast, _pgp1-1_ showed higher Φ _PSII_ than WT after 6 and 9 h of illumination, but no significant 369 difference from WT at 24 h. In _pgp1-1_ and _pgp1-1_ _sqd1_, Φ _NPQ_, which reflects regulated thermal 370 dissipation of absorbed light energy, remained low throughout the illumination period, whereas 371 in WT and _sqd1_, it was initially high and subsequently decreased to levels comparable to those 372 in _pgp1-1_ and _pgp1-1_ _sqd1_ by 24 h. In _pgp1-1_ _sqd1_, Φ _C_, which reflects non-regulated energy 373 dissipation, was substantially higher than in WT throughout the measurement, whereas Φ _C_ in 374 _pgp1-1_ was only slightly higher during the first 9 h of illumination. These results suggest that, 375 during the early stage of deetiolation, partial PG deficiency may even increase photochemical 376 utilization of absorbed light energy along with decreased regulated energy dissipation, whereas 377 additional loss of SQDG in the PG-deficient background severely perturbs the establishment of 378 photosynthetic functions. 379 To investigate how the loss of plastid anionic lipid biosynthesis affects the behavior of 380 Chl(ide)-protein complexes during deetiolation, we analyzed low-temperature fluorescence 381 spectra of intact cotyledons after continuous illumination of etiolated seedlings (Fig. 7). WT and 382 _sqd1_ seedlings illuminated for 5 min showed a sharp fluorescence band peaking at 693 nm and a 383 small shoulder around 680 nm, which are presumed to originate mainly from LPOR-bound and 384 released Chl(ide), respectively (Lindsten _et al._, 1993; Gabruk _et al._, 2026). These plants 385 exhibited a shift of the Chl(ide) peak to ~685 nm within 20 min, followed by a gradual red shift 386 to 689 nm during 6 h of illumination. In contrast, _pgp1-1_ and _pgp1-1 sqd1_ seedlings illuminated 387 for 5 min showed a single fluorescence band peaking at 682 nm and 679 nm, respectively, 388 consistent with the faster progression of the blue shift of Chl(ide) fluorescence after a single 389 flash in these mutants (Fig. 2) (Yoshihara et al., 2024). After 10 min of continuous illumination, 390 these mutants also exhibited a gradual red shift of the Chl(ide) peak, although _pgp1-1_ _sqd1_ 391 consistently showed shorter peak wavelengths than _pgp1-1_ . Because PSII-LHCII complexes 392 emit fluorescence at around 685-695 nm at 77 K (Lamb _et al._, 2018), the red shift of Chl 393 fluorescence toward ~690 nm after 1 h of illumination suggests the incorporation of Chl 394 molecules into PSII-LHCII complexes. After 2 h of illumination, another fluorescence band 395 appeared around 730 nm in all plants, which is known to originate from PSI-LHCI. At this time 12 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 396 point, the peak wavelength of the PSI-LHCI band in _pgp1-1_ and _pgp1-1_ _sqd1_ was around 727 397 nm, shorter than those in WT and _sqd1_ (~731 nm). After 6 h of illumination, the PSI-LHCI band 398 in _pgp1-1_ and _pgp1-1_ _sqd1_ shifted to ~733 nm, similar to those in WT and _sqd1_, although the 399 fluorescence intensity in _pgp1-1_ _sqd1_ remained lower than in WT and _sqd1_ . These results 400 suggest that the organization and functional properties of PSI-LHCI complexes, including their 401 interactions with PSII-LHCII, in _pgp1-1_ and _pgp1-1 sqd1_ differ from those in WT and _sqd1_ 402 during deetiolation. 403 404 **SQDG is required for photosynthesis-associated gene expression when PG biosynthesis is** 405 **reduced** 406 Plastid lipid biosynthesis and subsequent membrane formation have been hypothesized to 407 trigger photosynthesis-associated gene expression in both plastids and the nucleus (Fujii _et al._, 408 2025). To elucidate how decreased anionic lipid biosynthesis affects photosynthesis-associated 409 genes encoded in plastids and the nucleus during deetiolation, we examined mRNA levels of 410 key Chl biosynthesis genes _HEMA1_ and _CHLH_, encoding the major isoform of glutamyl-tRNA 411 reductase and the H subunit of Mg-chelatase, respectively; _LHCA4_ and _LHCB6_, encoding 412 subunits of the light-harvesting complexes of PSI and PSII, respectively; and _psaA_ and _psbA_, 413 encoding core subunits of the reaction centers of PSI and PSII, respectively (Fig. 8). _HEMA1_, 414 _CHLH_, _LHCA4_, and _LHCB6_ are encoded in the nucleus, whereas _psaA_ and _psbA_ are encoded in 415 plastids. In WT, mRNA levels of all genes increased in response to illumination, and _sqd1_ 416 showed a similar upregulation pattern. In _pgp1-1_, the mRNA levels of some genes tended to be 417 lower than those in WT at certain time points. Furthermore, _pgp1-1 sqd1_ generally showed 418 lower mRNA levels than WT across all examined genes during illumination. 419 420 **Discussion** 421 In this study, we demonstrated that a partial reduction in plastid PG biosynthesis affects multiple 422 processes during etioplast-to-chloroplast differentiation. Although etioplasts in _pgp1-1_ seedlings 423 were capable of forming apparently normal PLB lattice structures in darkness, these structures 424 were highly susceptible to disassembly upon illumination, indicating reduced stability of the 425 PLB architecture (Fig. 1A) (Yoshihara _et al._, 2024). Accordingly, the transformation from PLBs 426 to lamellar membranes upon illumination (Fig. 3), along with the dissociation of Chlide-LPOR 427 complexes and subsequent phytylation resulting in Chl (Fig. 2 and 4), proceeded more rapidly 428 in _pgp1-1_ than in WT. Furthermore, the more rapid rise in Φ _PSII_ after the onset of illumination in 429 _pgp1-1_ than in WT may reflect a faster establishment of the photosynthetic electron transport 430 machinery (Fig. 6B). However, impaired development of regulated energy dissipation during the 431 early stage of deetiolation cannot be ruled out as a cause of the increased Φ _PSII_ in _pgp1-1_ (Fig. 13 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 432 6C). In contrast, Chl accumulation during illumination was delayed (Fig. 5), and the 433 fluorescence properties of PSII-LHCII and PSI-LHCI, potentially reflecting changes in their 434 development and interactions, were altered in _pgp1-1_ (Fig. 7). Thus, although _pgp1-1_ rapidly 435 developed lamellar membranes from PLBs upon illumination, the development of the Chl 436 biosynthetic pathway and photosynthetic apparatus differed markedly from that in WT. In 437 addition, in _pgp1-1_ cotyledon cells, a subset of plastids failed to develop thylakoid membranes 438 and instead retained abnormal membrane structures during illumination (Fig. 1B and 3B). These 439 findings suggest that PG is important not only for the structural integrity of PLBs but also for 440 the coordinated progression of membrane remodeling and photosynthetic development during 441 deetiolation. 442 The formation of regular PLB lattices despite their high susceptibility to light-induced 443 disassembly indicates that PG is important for maintaining PLB structural stability rather than 444 for their assembly. Upon illumination, Pchlide-LPOR-NADPH is instantaneously converted to 445 Chlide-LPOR-NADP [+], followed by rapid replacement of bound NADP [+] by NADPH (Schoefs, 446 2001). The formation of Chlide-LPOR-NADPH complexes after photoreaction has been 447 suggested to transiently stabilize oligomeric assemblies and delay PLB disassembly (Gabruk _et_ 448 _al._, 2026). Subsequently, these oligomeric complexes gradually dissociate, as reflected by the 449 fluorescence peak shift from ~695 nm to ~680 nm at 77 K (the Shibata shift) (Shibata, 1957; 450 Schoefs, 2001). In _pgp1-1_ etioplasts, this fluorescence shift was accelerated (Fig. 2A) 451 (Yoshihara _et al._, 2024), consistent with more rapid dissociation of the oligomeric 452 Chlide-LPOR-NADPH complexes. Moreover, _pgp1-1_ showed the rapid emergence of a 453 fluorescence band around 675 nm, which presumably originates from free Chlide released from 454 the complexes (Schoefs, 2001; Schoefs and Franck, 2008). Considering that PG increases the 455 affinity between LPOR and NADPH in vitro (Gabruk _et al._, 2017), reduced PG biosynthesis in 456 _pgp1-1_ may destabilize the Chlide-LPOR-NADPH oligomers formed after illumination, thereby 457 promoting rapid disassembly of the PLB lattice structure and release of free Chlide. Conversely, 458 PG may contribute to delaying these processes by stabilizing oligomeric Chlide-LPOR-NADPH 459 complexes. 460 The accelerated Shibata shift in _pgp1-1_ is accompanied by faster Chlide-to-Chl 461 conversion than in WT (Fig. 2D and Fig. 4B). In etioplasts, Chl synthase is present in both 462 PLBs and prothylakoids, but the PLB-associated form is maintained in an inactive state 463 (Lindsten _et al._, 1990). Because the addition of NADPH to isolated PLBs stabilizes oligomeric 464 Chlide-LPOR-NADPH complexes and thereby inhibits the esterification of Chlide by Chl 465 synthase, oligomerization of these complexes and the resulting PLB compaction are likely to 466 interfere with Chl synthase activity (Lindsten _et al._, 1990, 1993). In _pgp1-1_ etioplasts, the rapid 467 dissociation of these oligomeric complexes and consequent disassembly of PLBs may allow 14 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 468 PLB-localized Chl synthase to become active more rapidly, resulting in faster Chlide-to-Chl 469 conversion. In addition, the rapid release of free Chlide after photoconversion may contribute to 470 its rapid phytylation by Chl synthase, while the alternative possibility that Chl synthase 471 activation promotes Chl(ide) release cannot be excluded (Schoefs and Franck, 2008). 472 Despite the more rapid conversion of dark-accumulated Pchlide to Chl via Chlide in 473 _pgp1-1_, de novo Chl biosynthesis after illumination was slower in this mutant than in WT (Fig. 474 5). Transcript levels of _HEMA1_ and _CHLH_, which are representative genes tightly co-expressed 475 with other key Chl biosynthesis genes (Kobayashi and Masuda, 2016), did not markedly 476 decrease in _pgp1-1_ (Fig. 8), implying that reduced expression of these genes does not account 477 for the slower Chl accumulation. Meanwhile, we previously reported that Pchlide biosynthesis 478 is retarded in _pgp1-1_ etioplasts, particularly at the step of Mg insertion into protoporphyrin IX 479 (Yoshihara _et al._, 2024). This impairment in the Chl biosynthetic pathway may persist during 480 etioplast-to-chloroplast differentiation, resulting in reduced Chl accumulation rates despite 481 accelerated activation of Chl synthase associated with PLB disassembly in _pgp1-1_ . 482 In contrast to the slower Chl accumulation, Φ _PSII_ in _pgp1-1_ rose more rapidly than in 483 WT during the first 9 h of illumination (Fig. 6B), whereas Φ _NPQ_ was substantially lower (Fig. 484 6C). Because _F_ _v_ / _F_ _m_ increased similarly in WT and _pgp1-1_ (Fig. 6A), the more rapid rise in Φ _PSII_ 485 in this mutant may reflect differences in the operation of the photosynthetic electron transport 486 system rather than differences in the functional integrity of PSII. One possibility is that the 487 photosynthetic electron transport components and/or associated metabolic processes, including 488 ATP and NADPH synthesis and consumption, become functional more rapidly in _pgp1-1_, 489 allowing more efficient electron transport and alleviating the conditions that trigger the 490 development of non-photochemical quenching (NPQ) mechanisms. In this case, the rapid 491 transformation of internal membranes in _pgp1-1_ plastids during deetiolation might contribute to 492 the more rapid establishment of these processes. Alternatively, mechanisms to induce NPQ 493 development at the early stage of deetiolation may be impaired in _pgp1-1_, which could result in 494 an increase in excitation energy input into PSII. If the photosynthetic electron transport system 495 and downstream metabolic processes have sufficient capacity to accommodate this increased 496 input, it could contribute to the higher Φ _PSII_, while the remaining excess energy would be 497 dissipated through non-regulated pathways, consistent with the slightly higher Φ _C_ in _pgp1-1_ 498 (Fig. 6D). In _Triticum aestivum_, energy-dependent NPQ and violaxanthin cycle operation 499 increase during deetiolation, potentially helping to protect the developing photosynthetic 500 apparatus from photooxidative damage (Garmash _et al._, 2013). A similar photoprotective role of 501 NPQ may exist in _A. thaliana_ during deetiolation, and its development could be impaired in 502 _pgp1-1_ . Further analyses are required to elucidate the molecular mechanisms underlying the 503 altered partitioning of absorbed light energy in _pgp1-1_ during the early stages of deetiolation. 15 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 504 Unlike _pgp1-1_, the _sqd1_ mutant did not exhibit notable differences from the WT 505 during etioplast-to-chloroplast differentiation, suggesting that SQDG alone is not essential for 506 this process. However, the _pgp1-1_ _sqd1_ double mutant showed more severe defects than _pgp1-1_ . 507 Many plastids in the double mutant failed to form regular PLB lattice structures even under 508 infrared light fixation (Fig. 1B), and a substantial proportion retained abnormal membrane 509 structures without developing thylakoid membranes after illumination (Fig. 3B). These 510 observations suggest that a certain level of anionic lipids is required to maintain internal 511 membrane integrity during plastid development and that SQDG plays a compensatory role 512 under conditions of reduced PG biosynthesis. 513 Another prominent feature of _pgp1-1_ _sqd1_ is its notably low _F_ _v_ / _F_ _m_, indicating reduced 514 PSII competence, which may also account for the low Φ _PSII_ and high Φ _C_ under light (Fig. 6). 515 The shorter-wavelength peak of PSII-LHCII-derived fluorescence at 77 K after 6 h of 516 illumination in _pgp1-1 sqd1_ suggests altered energetic interactions between PSII and LHCII 517 and/or within each component (Fig. 7, G and H). Structural studies of PSII-LHCII complexes 518 have demonstrated that PG and SQDG molecules occupy multiple sites in PSII, including 519 cavities around the plastoquinone binding sites, the monomer-monomer interface of the PSII 520 dimer, and the interfaces between PSII and LHCII proteins (Wei _et al._, 2016; Su _et al._, 2017). In 521 the thermophilic cyanobacterium _Thermosynechococcus elongatus_, the complete absence of 522 SQDG causes partial impairment of PSII activity, with PG probably occupying all 523 SQDG-binding sites (Nakajima _et al._, 2018), suggesting that PG can partially substitute for 524 SQDG in maintaining PSII structure. Although it remains unclear to what extent SQDG can 525 compensate for PG deficiency in the PSII-LHCII complexes, it is possible that PG and SQDG 526 play partially complementary roles in preserving their structural integrity. If so, the complete 527 absence of SQDG combined with partial PG deficiency in _pgp1-1 sqd1_ may alter the lipid 528 environment at anionic-lipid-binding sites in PSII-LHCII, thereby compromising the structural 529 integrity of these complexes, as suggested by its low _F_ _v_ / _F_ _m_ (Fig. 6A). In addition, the attenuated 530 light induction of _psaA_ and _psbA_, and presumably of co-expressed genes encoding other 531 photosystem proteins, might contribute to the impairment of photosynthetic establishment in 532 _pgp1-1 sqd1_ during deetiolation (Fig. 8). 533 The _pgp1-1 sqd1_ mutant also exhibited substantially reduced Chl accumulation even 534 compared with _pgp1-1_ (Fig. 5). This reduction can be at least partly explained by a more severe 535 defect in the Mg insertion step of the Chl biosynthetic pathway in _pgp1-1_ _sqd1_ than in _pgp1-1_ 536 (Yoshihara _et al._, 2024). Although direct evidence is still lacking, the possibility that Chl 537 biosynthesis is regulated through the thylakoid membrane as a scaffold has been discussed 538 (Fujii _et al._, 2025). How changes in the lipid environment within plastids affect the 539 functionalities of enzymes associated with the membrane-bound Chl biosynthetic pathway 16 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 540 remains to be investigated. In addition, _pgp1-1 sqd1_ showed strong attenuation of light-induced 541 expression of _HEMA1_ and _CHLH_ (Fig. 8), which also likely contributes to the decreased Chl 542 accumulation. 543 Previous studies using the knockout _pgp1-2_ mutant indicate that PG biosynthesis is 544 required for the light-induced expression of photosynthesis-associated plastid-encoded genes 545 (PhAPGs) and nuclear-encoded genes (PhANGs) (Fujii _et al._, 2022). Moreover, decreased 546 biosynthesis of either MGDG or DGDG attenuates the upregulation of PhAPGs and PhANGs 547 during deetiolation, suggesting the importance of membrane lipids for these processes (Fujii _et_ 548 _al._, 2019 _a_, 2024). The partial reduction in MGDG biosynthesis did not noticeably disrupt PLB 549 formation in etioplasts or the initial PLB-to-lamellar membrane transformation during 550 deetiolation (Fujii _et al._, 2017, 2019 _a_ ), despite its negative effects on photosynthesis-associated 551 gene expression. In contrast, overexpression of _CURVATURE THYLAKOID1A_ disrupts 552 PLB-to-thylakoid transformation without affecting PhAPG and PhANG expression 553 (Sandoval-Ibáñez _et al._, 2021). These results suggest that the induction of PhAPG and PhANG 554 expression during deetiolation is associated with membrane lipid biosynthesis but is not 555 primarily dependent on the morphological alterations of pre-existing internal membranes. 556 Considering that lipids are synthesized in the plastid envelope, de novo lipid biosynthesis at the 557 envelope may contribute to triggering photosynthesis-associated gene expression at the onset of 558 chloroplast development. Of note, loss of plastid PG biosynthesis in _pgp1-2_ specifically disrupts 559 the expression of PhAPGs that are transcribed by plastid-encoded RNA polymerase, including 560 _psaA_ and _psbA_, suggesting that PG biosynthesis is required for gene expression mediated by this 561 polymerase (Fujii _et al._, 2022). Because the attenuation of _psaA_ and _psbA_ induction was more 562 apparent in _pgp1-1_ _sqd1_ than in _pgp1-1_ (Fig. 8), SQDG may partially compensate for PG 563 function in this process. In addition, plastid-to-nuclear retrograde signaling downstream of 564 PhAPG expression has been implicated in the decreased PhANG expression observed in plastid 565 lipid mutants (Fujii _et al._, 2022, 2024, 2025). Although different classes of plastid lipids play 566 distinct roles in chloroplast development, all lipid classes are involved in upregulating 567 photosynthesis-associated gene expression, suggesting that lipid biosynthesis itself, or the 568 formation of an initial lipid bilayer resulting from this process, may trigger 569 photosynthesis-associated gene expression during chloroplast development. Future studies will 570 provide insights into the molecular mechanisms underlying this triggering process. 571 572 **Acknowledgments** 573 We are grateful to our colleagues in the plastid dynamics research community (plastid.jp) for 574 helpful and fruitful discussions. 575 17 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 576 **Author contributions** 577 A.Y. performed most experiments, analyzed and interpreted data, and prepared the first draft of 578 the manuscript. R. U., Ke.K., and N.N. performed transmission electron microscopic 579 experiments. Ko.K. designed and conducted experiments, analyzed and interpreted data, and 580 wrote and edited the manuscript. 581 582 **Conflict of interest** 583 None declared. 584 585 **Funding** 586 This work was supported by the Japan Society for the Promotion of Science (KAKENHI nos. 587 18H03941, 20K06691, 22H05076 to Ko.K. and Grant-in-Aid for JSPS Fellows no. JP23KJ1853 588 to A. Y.). A part of this research was carried out under the Cooperative Research Project of 589 Research Center for Biomedical Engineering. 590 591 **Data availability** 592 The data underlying this article are available in the article and in its online supplementary 593 material. 594 595 **Supplemental data** 596 **Supplementary Video S1.** Sequential transmission electron microscopy images of etioplasts in 597 a cotyledon of a 4-d-old dark-grown _pgp1-1 sqd1_ seedling fixed under infrared light. The video 598 comprises 63 images of consecutive ultrathin sections, each 80 nm thick. 599 **Supplementary Table S1.** Sequences of oligonucleotide primers used for quantitative reverse 600 transcription PCR analysis. 601 **Supplementary Table S2.** Statistical analysis of the data shown in Fig. 2B and 2C. 602 **Supplementary Table S3.** Statistical analysis of the data shown in Fig. 5C. 603 **Supplementary Table S4.** Statistical analysis of the data shown in Fig. 6. 604 **Supplementary Table S5.** Statistical analysis of the data shown in Fig. 7H and 7I. 605 **Supplementary Table S6.** Statistical analysis of the data shown in Fig. 8. 606 18 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **References** **Armstrong GA, Runge S, Frick G, Sperling U, Apel K** . 1995. Identification of NADPH:protochlorophyllide oxidoreductases A and B: a branched pathway for light-dependent chlorophyll biosynthesis in _Arabidopsis thaliana_ . Plant Physiology **108**, 1505–1517. **Baker NR** . 2008. Chlorophyll fluorescence: a probe of photosynthesis in vivo. Annual Review of Plant Biology **59**, 89–113. **Blomqvist LA, Ryberg M, Sundqvist C** . 2008. Proteomic analysis of highly purified prolamellar bodies reveals their significance in chloroplast development. Photosynthesis Research **96**, 37–50. **Dorne AJ, Joyard J, Douce R** . 1990. Do thylakoids really contain phosphatidylcholine? Proceedings of the National Academy of Sciences of the United States of America **87**, 71–74. **El Hamouri B, Brouers M, Sironval C** . 1981. Pathway from photoinactive _P_ 633−628 protochlorophyllide to the _P_ 696−682 chlorophyllide in cucumber etioplast suspensions. Plant Science Letters **21**, 375–379. **Essigmann B, Güler S, Narang RA, Linke D, Benning C** . 1998. Phosphate availability affects the thylakoid lipid composition and the expression of _SQD1_, a gene required for sulfolipid biosynthesis in _Arabidopsis thaliana_ . Proceedings of the National Academy of Sciences of the United States of America **95**, 1950–1955. **Floris D, Kühlbrandt W** . 2021. Molecular landscape of etioplast inner membranes in higher plants. Nature Plants **7**, 514–523. **Franck F, Bereza B, Böddi B** . 1999. Protochlorophyllide-NADP [+] and protochlorophyllide NADPH complexes and their regeneration after flash illumination in leaves and etioplast membranes of dark-grown wheat. Photosynthesis Research **59**, 53–61. **Fujii S, Kobayashi K, Lin Y-C, Liu Y, Nakamura Y, Wada H** . 2022. Impacts of phosphatidylglycerol on plastid gene expression and light induction of nuclear photosynthetic genes. Journal of Experimental Botany **73**, 2952–2970. **Fujii S, Kobayashi K, Nagata N, Masuda T, Wada H** . 2017. Monogalactosyldiacylglycerol facilitates synthesis of photoactive protochlorophyllide in etioplasts. Plant Physiology **174**, 2183–2198. **Fujii S, Kobayashi K, Nagata N, Masuda T, Wada H** . 2018. Digalactosyldiacylglycerol is essential for organization of the membrane structure in etioplasts. Plant Physiology **177**, 19 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) 1487–1497. **Fujii S, Nagata N, Kobayashi K** . 2025. Membrane lipid-driven thylakoid biogenesis coordinating chlorophyll synthesis and expression of photosynthetic proteins. Plant and Cell Physiology **66**, 1645–1665. **Fujii S, Nagata N, Masuda T, Wada H, Kobayashi K** . 2019 _a_ . Galactolipids are essential for internal membrane transformation during etioplast-to-chloroplast differentiation. Plant and Cell Physiology **60**, 1224–1238. **Fujii S, Wada H, Kobayashi K** . 2019 _b_ . Role of galactolipids in plastid differentiation before and after light exposure. Plants **8**, 357. **Fujii S, Wada H, Kobayashi K** . 2024. Orchestration of photosynthesis-associated gene expression and galactolipid biosynthesis during chloroplast differentiation in plants. Plant and Cell Physiology **65**, 1014–1028. **Gabruk M, Desfosses A, Estrozi LF, Pintscher S, Rawski M, Wa** ż **ny G, Garbacz A,** **Zbyradowski M, Kruk J, Fiedor L** . 2026. Structures of LPOR-Chlide complexes reveal the structural basis of membrane remodeling and photocatalysis. Nature Communications **17**, 9427. **Gabruk M, Mysliwa-Kurdziel B, Kruk J** . 2017. MGDG, PG and SQDG regulate the activity of light-dependent protochlorophyllide oxidoreductase. The Biochemical Journal **474**, 1307–1320. **Garmash EV, Dymova OV, Malyshev RV, Plyusnina SN, Golovko TK** . 2013. Developmental changes in energy dissipation in etiolated wheat seedlings during the greening process. Photosynthetica **51**, 497–508. **Guadagno CR, Virzo De Santo A, D’Ambrosio N** . 2010. A revised energy partitioning approach to assess the yields of non-photochemical quenching components. Biochimica et Biophysica Acta **1797**, 525–530. **Hagio M, Sakurai I, Sato S, Kato T, Tabata S, Wada H** . 2002. Phosphatidylglycerol is essential for the development of thylakoid membranes in _Arabidopsis thaliana_ . Plant and Cell Physiology **43**, 1456–1464. **Hothorn T, Bretz F, Westfall P** . 2008. Simultaneous inference in general parametric models. Biometrical Journal **50**, 346–363. **Kay SA, Griffiths WT** . 1983. Light-induced breakdown of NADPH-protochlorophyllide 20 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) oxidoreductase in vitro. Plant Physiology **72**, 229–236. **Kobayashi K, Endo K, Wada H** . 2016. Multiple impacts of loss of plastidic phosphatidylglycerol biosynthesis on photosynthesis during seedling growth of Arabidopsis. Frontiers in Plant Science **7**, 336. **Kobayashi K, Jimbo H, Nakamura Y, Wada H** . 2024. Biosynthesis of phosphatidylglycerol in photosynthetic organisms. Progress in Lipid Research **93**, 101266. **Kobayashi K, Masuda T** . 2016. Transcriptional regulation of tetrapyrrole biosynthesis in _Arabidopsis thaliana_ . Frontiers in Plant Science **7**, 1811. **Kowalewska Ł, Mazur R, Suski S, Garstka M, Mostowska A** . 2016. Three-dimensional visualization of the tubular-lamellar transformation of the internal plastid membrane network during runner bean chloroplast biogenesis. The Plant Cell **28**, 875–891. **Lamb JJ, Røkke G, Hohmann-Marriott MF** . 2018. Chlorophyll fluorescence emission spectroscopy of oxygenic organisms at 77 K. Photosynthetica **56**, 105–124. **Lindsten A, Welch CJ, Schoch S, Ryberg M, Rüdiger W, Sundqvist C** . 1990. Chlorophyll synthetase is latent in well preserved prolamellar bodies of etiolated wheat. Physiologia Plantarum **80**, 277–285. **Lindsten A, Wiktorsson B, Ryberg M, Sundqvist C** . 1993. Chlorophyll synthetase activity is relocated from transforming prolamellar bodies to developing thylakoids during irradiation of dark-grown wheat. Physiologia Plantarum **88**, 29–36. **Ma L, Li J, Qu L, Hager J, Chen Z, Zhao H, Deng XW** . 2001. Light control of Arabidopsis development entails coordinated regulation of genome expression and cellular pathways. The Plant Cell **13**, 2589–2608. **Nakajima Y, Umena Y, Nagao R, Endo K, Kobayashi K, Akita F, Suga M, Wada H, Noguchi T,** **Shen J-R** . 2018. Thylakoid membrane lipid sulfoquinovosyl-diacylglycerol (SQDG) is required for full functioning of photosystem II in _Thermosynechococcus elongatus_ . Journal of Biological Chemistry **293**, 14786–14797. **Nguyen HC, Melo AA, Kruk J, Frost A, Gabruk M** . 2021. Photocatalytic LPOR forms helical lattices that shape membranes for chlorophyll synthesis. Nature Plants **7**, 437–444. **Okazaki Y, Shimojima M, Sawada Y,** _**et al.**_ 2009. A chloroplastic UDP-glucose 21 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) pyrophosphorylase from Arabidopsis is the committed enzyme for the first step of sulfolipid biosynthesis. The Plant Cell **21**, 892–909. **Oliver RP, Griffiths WT** . 1982. Pigment-protein complexes of illuminated etiolated leaves. Plant Physiology **70**, 1019–1025. **Pfaffl MW** . 2001. A new mathematical model for relative quantification in real-time RT–PCR. Nucleic Acids Research **29**, e45. **Pipitone R, Eicke S, Pfister B, Glauser G, Falconet D, Uwizeye C, Pralon T, Zeeman SC,** **Kessler F, Demarsy E** . 2021. A multifaceted analysis reveals two distinct phases of chloroplast biogenesis during de-etiolation in Arabidopsis. eLife **10**, e62709. **Plöscher M, Reisinger V, Eichacker LA** . 2011. Proteomic comparison of etioplast and chloroplast protein complexes. Journal of Proteomics **74**, 1256–1265. **Sandoval-Ibáñez O, Sharma A, Bykowski M,** _**et al.**_ 2021. Curvature thylakoid 1 proteins modulate prolamellar body morphology and promote organized thylakoid biogenesis in _Arabidopsis thaliana_ . Proceedings of the National Academy of Sciences of the United States of America **118**, e2113934118. **Schoefs B** . 2001. The protochlorophyllide–chlorophyllide cycle. Photosynthesis Research **70**, 257–271. **Schoefs B, Franck F** . 2003. Protochlorophyllide reduction: Mechanisms and evolution. Photochemistry and Photobiology **78**, 543–557. **Schoefs B, Franck F** . 2008. The photoenzymatic cycle of NADPH: protochlorophyllide oxidoreductase in primary bean leaves ( _Phaseolus vulgaris_ ) during the first days of photoperiodic growth. Photosynthesis Research **96**, 15–26. **Selstam E, Sandelius AS** . 1984. A comparison between prolamellar bodies and prothylakoid membranes of etioplasts of dark-grown wheat concerning lipid and polypeptide composition. Plant Physiology **76**, 1036–1040. **Shibata K** . 1957. Spectroscopic studies on chlorophyll formation in intact leaves. The Journal of Biochemistry **44**, 147–173. **Solymosi K, Schoefs B** . 2010. Etioplast and etio-chloroplast formation under natural conditions: the dark side of chlorophyll biosynthesis in angiosperms. Photosynthesis Research **105**, 143–166. 22 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **Su X, Ma J, Wei X, Cao P, Zhu D, Chang W, Liu Z, Zhang X, Li M** . 2017. Structure and assembly mechanism of plant C2S2M2-type PSII-LHCII supercomplex. Science **357**, 815–820. **Wada H, Murata N** . 2007. The essential role of phosphatidylglycerol in photosynthesis. Photosynthesis Research **92**, 205–215. **Wei X, Su X, Cao P, Liu X, Chang W, Li M, Zhang X, Liu Z** . 2016. Structure of spinach photosystem II-LHCII supercomplex at 3.2 � Å resolution. Nature **534**, 69–74. **Xu C, Härtel H, Wada H, Hagio M, Yu B, Eakin C, Benning C** . 2002. The _pgp1_ mutant locus of Arabidopsis encodes a phosphatidylglycerolphosphate synthase with impaired activity. Plant Physiology **129**, 594–604. **Yoshihara A, Kobayashi K** . 2022. Lipids in photosynthetic protein complexes in the thylakoid membrane of plants, algae, and cyanobacteria. Journal of Experimental Botany **73**, 2735–2750. **Yoshihara A, Kobayashi K, Nagata N, Fujii S, Wada H, Kobayashi K** . 2024. Anionic lipids facilitate membrane development and protochlorophyllide biosynthesis in etioplasts. Plant Physiology **194**, 1692–1704. **Yoshihara A, Nagata N, Wada H, Kobayashi K** . 2021. Plastid anionic lipids are essential for the development of both photosynthetic and non-photosynthetic organs in _Arabidopsis thaliana_ . International Journal of Molecular Sciences **22**, 4860. **Yu B, Benning C** . 2003. Anionic lipids are required for chloroplast structure and function in Arabidopsis. The Plant Journal **36**, 762–770. **Yu B, Xu C, Benning C** . 2002. Arabidopsis disrupted in _SQD2_ encoding sulfolipid synthase is impaired in phosphate-limited growth. Proceedings of the National Academy of Sciences of the United States of America **99**, 5732–5737. 23 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **Figure legends** **Figure 1. Membrane structures of etioplasts in 4-d-old etiolated cotyledons of wild-type** **(WT) and anionic lipid mutants. (A)** Ultrastructures of etioplasts (upper panel) and the enlarged views of prolamellar body (PLB) lattices (lower panel) in cotyledons of dark-grown WT, _sqd1_, _pgp1-1_, and _pgp1-1 sqd1_ seedlings fixed under infrared light. Bars = 1.0 μ m (upper panel) and 0.5 μ m (lower panel). In _pgp1-1 sqd1_, only 2.1% of etioplast profiles in ultrathin sections exhibited regular PLB lattices throughout the structure. **(B)** Ultrastructures of etioplasts containing disordered membrane networks in cotyledons of _pgp1-1_ and _pgp1-1 sqd1_ seedlings fixed under infrared light. Disordered internal membrane networks, rather than regular lattice structures, were observed in 5.4% of _pgp1-1_ and 94.6% of _pgp1-1 sqd1_ etioplasts in ultrathin sections. Bars = 1.0 μ m. All images are high-resolution montages constructed from multiple images. **Figure 2. Rearrangement of oligomeric LPOR-pigment complexes and pigment** **metabolism in darkness following a flash in 4-d-old etiolated seedlings of wild type (WT)** **and anionic lipid mutants. (A)** _In situ_ low-temperature (77 K) fluorescence spectra of cotyledons after 1, 30, 60, and 120 s of dark incubation at 13 � following a single flash. Fluorescence intensities were normalized to the mean fluorescence intensity over the 600–800 nm range. The insets show enlarged views of the spectra from 645 to 665 nm. Data are means from 3 independent experiments. **(B)** Changes in the fluorescence peak wavelength in the 690–696 nm region and **(C)** relative fluorescence intensity at 675 nm, normalized to the peak fluorescence intensity around 690 nm, during dark incubation. Data are means ± SE from 3 independent experiments. Statistical significance was assessed using one-way ANOVA followed by Tukey’s honestly significant difference (HSD) test. The results of the post hoc tests are provided in Supplementary Table S2. **(D)** Chl _a_ /Chlide _a_ ratio in etiolated seedlings after 0.5 and 1 min of dark incubation following a single flash. Data are means ± SE from 3 independent experiments. Different lowercase letters indicate significant differences among genotypes at each time point (p < 0.05, one-way ANOVA followed by Tukey’s HSD test). **Figure 3. Time-course observation of plastid morphology in 4-d-old etiolated cotyledons** **of wild-type (WT) and anionic lipid mutants after illumination. (A)** Ultrastructures of cotyledon plastids after 1.5, 3, 6, 9, and 24 h of continuous illumination in WT, _sqd1_, _pgp1-1_, and _pgp1-1 sqd1_ . **(B)** Ultrastructures of plastids containing disordered membrane structures in _pgp1-1_ and _pgp1-1 sqd1_ after 24 h of illumination. Approximately 2% and 27% of plastid profiles in ultrathin sections of _pgp1-1_ and _pgp1-1_ _sqd1_, respectively, exhibited abnormal 24 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) membrane structures without forming thylakoid membranes. All images are high-resolution montages constructed from multiple images. Bars = 1.0 μ m. **Figure 4. LPOR levels and porphyrin content in 4-d-old etiolated cotyledons of wild-type** **(WT) and anionic lipid mutants after illumination. (A)** Immunodetection of total LPOR proteins (~37 kDa), with Coomassie brilliant blue (CBB) staining shown as a loading control, and quantification of LPOR protein levels relative to the WT level in etiolated seedlings after 5 min of continuous illumination. Data are means ± SE from 3 independent experiments. Different lowercase letters indicate significant differences among genotypes (p < 0.05, one-way ANOVA followed by Tukey’s honestly significant difference test). **(B)** Chlorophyll _a_, chlorophyllide _a_, and protochlorophyllide content in 4-d-old etiolated seedlings after 0, 5, 10, and 20 min of continuous illumination following a single flash. Data are means ± SE from 3 independent experiments. **Figure 5. Chlorophyll (Chl) accumulation in 4-d-old etiolated cotyledons of wild-type** **(WT) and anionic lipid mutants during illumination. (A)** Greening phenotype, **(B)** rate of Chl accumulation, and **(C)** Chl content in etiolated seedlings of WT, _sqd1_, _pgp1-1_, and _pgp1-1_ _sqd1_ after 1.5, 3, 6, 9, and 24 h of continuous illumination. Chl content represents the sum of Chl _a_ and Chl _b_ . In **(C)**, the inset shows an enlarged view of the data at 0, 1.5, and 3 h. Protochlorophyllide content from Yoshihara _et al._ (2024) is shown for 0 h instead of Chl content. Data are means ± SE from 16 independent experiments at 0 h (Yoshihara _et al._, 2024) and 6 independent experiments at each time point from 1.5 to 24 h. Statistical significance was assessed using one-way ANOVA followed by Tukey’s honestly significant difference (HSD) test. In **(B)**, Different lowercase letters indicate significant differences among genotypes for each time period. For **(C)**, the results of the HSD tests are provided in Supplementary Table S3. **Figure 6. Establishment of the photosynthetic machinery in 4-d-old etiolated cotyledons** **of wild-type (WT) and anionic lipid mutants after illumination. (A)** The maximum quantum yield of photosystem II ( _F_ _v_ / _F_ _m_ ), **(B)** the effective quantum yield of photosystem II ( Φ _PSII_ ), **(C)** the quantum yield of regulated energy dissipation ( Φ _NPQ_ ), and **(D)** the quantum yield of non-regulated energy dissipation ( Φ _c_ ) in etiolated seedlings of WT, _sqd1_, _pgp1-1_, and _pgp1-1 sqd1_ after 6, 9, and 24 h of continuous illumination. Data are means ± SE from approximately 50 individuals per genotype. Statistical significance was assessed using one-way ANOVA followed by Tukey’s honestly significant difference (HSD) test. The results of the post hoc tests are provided in Supplementary Table S4. 25 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **Figure 7. In situ low-temperature fluorescence spectra in 4-d-old etiolated seedlings of** **wild type (WT) and anionic lipid mutants after continuous illumination.** 77 K fluorescence spectra of cotyledons after **(A)** 5, **(B)** 10, **(C)** 20, **(D)** 60, **(E)** 120, **(F)** 180, and **(G)** 360 min of continuous illumination. Fluorescence intensities were normalized to the mean fluorescence intensity over the 600–800 nm range. **(H, I)** Changes in fluorescence peak wavelength in the **(H)** 678–694 nm and **(I)** 725–735 nm regions during the illumination. Data are means ± SE from 3 independent experiments. Statistical significance was assessed using one-way ANOVA followed by Tukey’s honestly significant difference (HSD) test. The results of the post hoc tests are provided in Supplementary Table S5. **Figure 8. Induction of photosynthesis-associated gene expression in 4-d-old etiolated** **cotyledons of wild-type (WT) and anionic lipid mutants during illumination.** The mRNA levels of photosynthesis-associated genes in etiolated seedlings before (0 h) and after (3, 6, 12 h) illumination are expressed as fold changes relative to the WT level at 0 h, after normalization to the reference genes _ACTIN8_ and _UBIQUITIN11_ . Data are means ±SE from 3 independent experiments. Statistical significance was assessed using one-way ANOVA followed by Tukey’s honestly significant difference (HSD) test. The results of the post hoc tests are provided in Supplementary Table S6. 26 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **Figure 1. Membrane structures of etioplasts in 4-d-old etiolated cotyledons of wild-type (WT) and anionic lipid mutants. (A)** Ultrastructures of etioplasts (upper panel) and the enlarged views of prolamellar body (PLB) lattices (lower panel) in cotyledons of dark-grown WT, _sqd1_, _pgp1-1_, and _pgp1-1 sqd1_ seedlings fixed under infrared light. Bars = 1.0 μm (upper panel) and 0.5 μm (lower panel). In _pgp1-1 sqd1_, only 2.1% of etioplast profiles in ultrathin sections exhibited regular PLB lattices throughout the structure. **(B)** Ultrastructures of etioplasts containing disordered membrane networks in cotyledons of _pgp1-1_ and _pgp1-1 sqd1_ seedlings fixed under infrared light. Disordered internal membrane networks, rather than regular lattice structures, were observed in 5.4% of _pgp1-1_ and 94.6% of _pgp1-1 sqd1_ etioplasts in ultrathin sections. Bars = 1.0 μm. All images are high-resolution montages constructed from multiple images. [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **Figure 2. Rearrangement of oligomeric LPOR-pigment complexes and pigment metabolism in darkness following a flash in 4-** **d-old etiolated seedlings of wild type (WT) and anionic lipid mutants. (A)** _In situ_ low-temperature (77 K) fluorescence spectra of cotyledons after 1, 30, 60, and 120 s of dark incubation at 13℃ following a single flash. Fluorescence intensities were normalized to the mean fluorescence intensity over the 600–800 nm range. The insets show enlarged views of the spectra from 645 to 665 nm. Data are means from 3 independent experiments. **(B)** Changes in the fluorescence peak wavelength in the 690–696 nm region and **(C)** relative fluorescence intensity at 675 nm, normalized to the peak fluorescence intensity around 690 nm, during dark incubation. Data are means ± SE from 3 independent experiments. Statistical significance was assessed using one-way ANOVA followed by Tukey’s honestly significant difference (HSD) test. The results of the post hoc tests are provided in Supplementary Table S2. **(D)** Chl _a_ /Chlide _a_ ratio in etiolated seedlings after 0.5 and 1 min of dark incubation following a single flash. Data are means ± SE from 3 independent experiments. Different lowercase letters indicate significant differences among genotypes at each time point (p < 0.05, one-way ANOVA followed by Tukey’s HSD test). [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **Figure 3. Time-course observation of plastid morphology in 4-d-old etiolated cotyledons of wild-type (WT) and anionic lipid** **mutants after illumination. (A)** Ultrastructures of cotyledon plastids after 1.5, 3, 6, 9, and 24 h of continuous illumination in WT, _sqd1_, _pgp1-1_, and _pgp1-1 sqd1_ . **(B)** Ultrastructures of plastids containing disordered membrane structures in _pgp1-1_ and _pgp1-1 sqd1_ after 24 h of illumination. Approximately 2% and 27% of plastid profiles in ultrathin sections of _pgp1-1_ and _pgp1-1_ _sqd1_, respectively, exhibited abnormal membrane structures without forming thylakoid membranes. All images are high-resolution montages constructed from multiple images. Bars = 1.0 μm. [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **Figure 4. LPOR levels and porphyrin content in 4-d-old etiolated cotyledons of wild-type (WT) and anionic lipid mutants after** **illumination. (A)** Immunodetection of total LPOR proteins (~37 kDa), with Coomassie brilliant blue (CBB) staining shown as a loading control, and quantification of LPOR protein levels relative to the WT level in etiolated seedlings after 5 min of continuous illumination. Data are means ± SE from 3 independent experiments. Different lowercase letters indicate significant differences among genotypes (p < 0.05, one-way ANOVA followed by Tukey’s honestly significant difference test). **(B)** Chlorophyll _a_, chlorophyllide _a_, and protochlorophyllide content in 4-d-old etiolated seedlings after 0, 5, 10, and 20 min of continuous illumination following a single flash. Data are means ± SE from 3 independent experiments. [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **Figure 5. Chlorophyll (Chl) accumulation in 4-d-old etiolated cotyledons of wild-type (WT) and anionic lipid mutants during** **illumination. (A)** Greening phenotype, **(B)** rate of Chl accumulation, and **(C)** Chl content in etiolated seedlings of WT, _sqd1_, _pgp1-1_, and _pgp1-1 sqd1_ after 1.5, 3, 6, 9, and 24 h of continuous illumination. Chl content represents the sum of Chl _a_ and Chl _b_ . In **(C)**, the inset shows an enlarged view of the data at 0, 1.5, and 3 h. Protochlorophyllide content from Yoshihara _et al._ (2024) is shown for 0 h instead of Chl content. Data are means ± SE from 16 independent experiments at 0 h (Yoshihara _et al._, 2024) and 6 independent experiments at each time point from 1.5 to 24 h. Statistical significance was assessed using one-way ANOVA followed by Tukey’s honestly significant difference (HSD) test. In **(B)**, Different lowercase letters indicate significant differences among genotypes for each time period. For **(C)**, the results of the HSD tests are provided in Supplementary Table S3. [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **Figure 6. Establishment of the photosynthetic machinery in 4-d-old etiolated cotyledons of wild-type (WT) and anionic lipid** **mutants after illumination. (A)** The maximum quantum yield of photosystem II ( _F_ _v_ / _F_ _m_ ), **(B)** the effective quantum yield of photosystem II ( _Φ_ _PSII_ ), **(C)** the quantum yield of regulated energy dissipation ( _Φ_ _NPQ_ ), and **(D)** the quantum yield of non-regulated energy dissipation ( _Φ_ _c_ ) in etiolated seedlings of WT, _sqd1_, _pgp1-1_, and _pgp1-1 sqd1_ after 6, 9, and 24 h of continuous illumination. Data are means ± SE from approximately 50 individuals per genotype. Statistical significance was assessed using one-way ANOVA followed by Tukey’s honestly significant difference (HSD) test. The results of the post hoc tests are provided in Supplementary Table S4. [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **Figure 7. In situ low-temperature fluorescence spectra in 4-d-old etiolated seedlings of wild type (WT) and anionic lipid** **mutants after continuous illumination.** 77 K fluorescence spectra of cotyledons after **(A)** 5, **(B)** 10, **(C)** 20, **(D)** 60, **(E)** 120, **(F)** 180, and **(G)** 360 min of continuous illumination. Fluorescence intensities were normalized to the mean fluorescence intensity over the 600–800 nm range. **(H, I)** Changes in fluorescence peak wavelength in the **(H)** 678–694 nm and **(I)** 725–735 nm regions during the illumination. Data are means ± SE from 3 independent experiments. Statistical significance was assessed using one-way ANOVA followed by Tukey’s honestly significant difference (HSD) test. The results of the post hoc tests are provided in Supplementary Table S5. [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763292; this version posted November 4, 2026. The copyright holder for this preprint](https://doi.org/10.64898/2026.10.31.763292) (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 [available under a CC-BY-NC 4.0 International license.](http://creativecommons.org/licenses/by-nc/4.0/) **Figure 8. Induction of photosynthesis-associated gene expression in 4-d-old etiolated cotyledons of wild-type (WT) and anionic** **lipid mutants during illumination.** The mRNA levels of photosynthesis-associated genes in etiolated seedlings before (0 h) and after (3, 6, 12 h) illumination are expressed as fold changes relative to the WT level at 0 h, after normalization to the reference genes _ACTIN8_ and _UBIQUITIN11_ . Data are means ±SE from 3 independent experiments. Statistical significance was assessed using one way ANOVA followed by Tukey’s honestly significant difference (HSD) test. The results of the post hoc tests are provided in Supplementary Table S6.