# Source paper archive - Source title: NRT2.2 is essential for nitrate uptake and carbon-nitrogen balance in Nannochloropsis oceanica - Author(s): Santin, A., Verdesca, L., Bellan, A., Minio, M., Gerotto, c., Morosinotto, T., Perin, G., Alboresi, A. - Source landing URL: https://www.biorxiv.org/content/10.64898/2026.11.02.763162v1 - PDF URL: https://www.biorxiv.org/content/10.64898/2026.11.02.763162v1.full.pdf - PDF SHA-256: 21b5d443fe47575bf86e56a7d857e43d81d1e174edd782e46c040afb47ad8bfe - 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/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in [perpetuity.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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 1 **TITLE** 2 NRT2.2 is essential for nitrate uptake and carbon-nitrogen balance in _Nannochloropsis oceanica_ 3 4 **AUTHOR NAMES AND AFFILIATIONS** 5 Anna Santin [1,a], Laura Verdesca [1,b], Alessandra Bellan [1], Miles Minio [2], Caterina Gerotto [2], Tomas 6 Morosinotto [1], Giorgio Perin [1], Alessandro Alboresi [1 ] 7 8 1 Department of Biology, University of Padova, 35131, Padova, Italy 9 2 Department of Life and Environmental Sciences, Marche Polytechnic University, 60131 Ancona, 10 Italy 11 12 a present address: UMR7141- Photobiology and Physiology of Plastids and Microalgae (P3M), 13 CNRS, Sorbonne University, 75005, Paris, France 14 b present address: Stazione Zoologica Anton Dohrn, Villa Comunale, 80121, Naples, Italy 15 16 - Corresponding author: Alessandro Alboresi. Department of Biology, University of Padova, Italy. 17 [Email: alessandro.alboresi@unipd.it](mailto:alessandro.alboresi@unipd.it) 18 19 20 **Running head:** NRT2.2 and nitrate uptake in _Nannochloropsis_ 21 22 23 **ORCID** 24 [Anna Santin: https://orcid.org/0000-0002-2950-7046; Laura Verdesca: https://orcid.org/0009-0001-](https://orcid.org/0000-0002-2950-7046) 25 [6220-9972;](https://orcid.org/0009-0001-6220-9972) Alessandra Bellan: [https://orcid.org/0009-0008-2346-5560;](https://orcid.org/0009-0008-2346-5560) Miles Minio: 26 [https://orcid.org/0009-0002-8522-2368; Caterina Gerotto:](https://orcid.org/0009-0002-8522-2368) [https://orcid.org/0000-0001-7142-3535;](https://orcid.org/0000-0001-7142-3535) 27 [Tomas Morosinotto: https://orcid.org/0000-0002-0803-7591; Giorgio Perin: https://orcid.org/0000-](https://orcid.org/0000-0002-0803-7591) 28 [0002-9204-4080; Alessandro Alboresi: https://orcid.org/0000-0003-4818-7778](https://orcid.org/0000-0002-9204-4080) 29 1 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 30 **ABSTRACT** 31 Nitrate is the dominant nitrogen source for marine microalgae, yet how it is taken up remains poorly 32 understood in many ecologically and biotechnologically relevant species. Here we show that 33 NITRATE TRANSPORTER 2.2 (NRT2.2) is the principal nitrate transporter of the oleaginous 34 marine microalga _Nannochloropsis oceanica_ . _N. oceanica_ encodes only two NRT2 proteins and, 35 unlike land plants and green algae, lacks a recognizable homolog of NITRATE ASSIMILATION 36 RELATED 2 (NAR2), the partner protein that green-lineage NRT2s require for activity and plasma 37 membrane targeting. CRISPR knock-outs and confocal imaging showed that NRT2.2 is located at the 38 plasma membrane and is required for nitrate uptake, whereas its paralog NRT2.1 cannot compensate 39 for its loss. When nitrate was resupplied after nitrogen starvation, _nrt2.2_ mutants did not take up 40 nitrate and did not resume growth, while they recovered normally with ammonium. Their 41 photosystem II efficiency fell below that of fully nitrogen-starved wild-type cells, and chlorophyll 42 content was reduced, while total protein remained stable. Fixed carbon was redirected into storage, 43 with lipids and carbohydrates accumulating above the levels of all strains after 3 days of nitrogen 44 starvation. Loss of NRT2.2 therefore mimics a prolonged nitrogen starvation, even though nitrate is 45 available. Because nitrate was present but could not be used, recovery from starvation depends on the 46 capacity to take up nitrate, not only on its availability. In the ocean, where nitrate is depleted for long 47 periods and resupplied episodically, NRT2.2 is thus likely to set how quickly _Nannochloropsis_ can 48 use new nitrate inputs and resume growth. 49 50 **KEYWORDS** 51 Nitrate transporters; nitrogen starvation; carbon-nitrogen balance; lipid accumulation; marine 52 microalgae; _Nannochloropsis oceanica_ ; NRT2 53 2 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 54 **INTRODUCTION** 55 Nitrogen (N) is an essential macronutrient for all photosynthetic organisms, required for the synthesis 56 of proteins, nucleic acids, chlorophylls and numerous metabolic cofactors. Its availability therefore 57 directly determines growth and productivity, as its assimilation is tightly coordinated with carbon 58 fixation and energy metabolism (Simionato et al., 2013; Santin et al., 2026). While the molecular 59 networks controlling N uptake and signaling are well established in land plants (Wang et al., 2018; 60 Maghiaoui et al., 2020; Cassan et al., 2023), the mechanisms of N transport in marine microalgae 61 remain poorly understood (McCarthy et al., 2017; Busseni et al., 2019; Smith et al., 2019) . 62 Marine phytoplankton can use a variety of inorganic and organic N sources, including nitrate (NO 3 [-] 63 ), ammonium (NH 4 [+] ), nitrite and, in several taxa, urea and amino acids. The relative contribution of 64 these sources depends on the species, their environmental availability and the energetic cost of their 65 assimilation (Glibert, 2016). Although NH 4 [+] is energetically cheaper to assimilate, its concentration 66 in seawater is very low, from ~0.3 µM in the euphotic zone to ~0.01 µM below it, whereas NO 3 [-], 67 although often below 1 µM in surface water, is the largest pool of bioavailable N in the ocean (Gruber, 68 2008; Sanz-Luque et al., 2015). In both cases, concentrations are lower and more variable than in 69 soil, where they can reach the millimolar range (Kabala et al., 2017). Marine microalgae must 70 therefore acquire NO 3 [-] efficiently even at very low concentrations, which makes membrane transport 71 a critical step in N acquisition, storage and assimilation. 72 NO 3 [-] transport across biological membranes relies on proton-coupled active transporters of the major 73 facilitator superfamily (MFS), which are predicted to contain twelve transmembrane helices (von 74 Wittgenstein et al., 2014; Rogato et al., 2015). Within the MFS, NO 3 [-] transport is primarily mediated 75 by members of two evolutionarily distinct families: the low-affinity NITRATE TRANSPORTER 76 1/PEPTIDE TRANSPORTER FAMILY (NRT1/NPF) and the high-affinity NITRATE 77 TRANSPORTER 2 (NRT2) family. The two families differ in their transport properties and 78 physiological roles. NRT1/NPF proteins generally mediate NO 3 [-] uptake at higher external 79 concentrations and have diversified to transport a broad range of substrates, including peptides, 80 hormones and secondary metabolites in addition to NO 3 [-] (Fan et al., 2017; Santin et al., 2021; Santin 81 et al., 2024). NRT2 proteins, in contrast, are primarily associated with high-affinity NO 3 [-] uptake in 82 plants and green algae (Fernandez and Galvan, 2008; Xu et al., 2024), making them particularly 83 important when NO 3 [-] is scarce. In land plants and in the freshwater alga _Chlamydomonas reinhardtii_, 84 high-affinity NO 3 [-] uptake by NRT2 requires the accessory protein NITRATE ASSIMILATION 85 RELATED 2 (NAR2, also known as NRT3) (Zhou et al., 2000; Rexach et al., 2002). In contrast, 86 NAR2 homologs are absent from algae with secondary plastids (Rogato et al., 2015), suggesting that 87 NRT2-mediated NO 3 [-] transport operates differently in these lineages. However, NRT2 mutants have 3 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 88 not yet been characterized in any secondary endosymbiont, so how NRT2 proteins work in these 89 organisms is still unclear. 90 _Nannochloropsis_ is a genus of marine microalgae with a secondary red plastid, widely studied for its 91 high lipid productivity and as a model for marine photosynthesis and N metabolism (Simionato et al., 92 2013; Fattore et al., 2021). Hereafter, _Nannochloropsis_ refers to the genus in the broad sense, 93 including _Microchloropsis_ (formerly _N. gaditana_ and _N. salina_ ). In _Nannochloropsis_, N deprivation 94 induces extensive metabolic and structural changes, including reorganization of the photosynthetic 95 apparatus and the accumulation of storage lipids, highlighting the central role of N availability in 96 controlling cellular physiology (Simionato et al., 2013; Fattore et al., 2021). Genomic and 97 transcriptomic analyses have identified candidate genes for NO 3 [-] assimilation in _Nannochloropsis_, 98 including NRT2 transporters (Radakovits et al., 2012), but their function and subcellular localization 99 are unknown, so it is still unclear how NO 3 [-] enters the cell. 100 Here, we characterized the two NRT2 transporters of _N. oceanica_ and showed that NRT2.2 is located 101 at the plasma membrane and mediates NO 3 [-] uptake from the medium. Loss of NRT2.2 prevents cells 102 from resuming growth on NO 3 [-] after N starvation, establishing a direct link between NRT2-dependent 103 NO 3 [-] uptake and recovery from N limitation. Cells lacking NRT2.2 behave as if N-starved even 104 though NO 3 [-] is available: they lose chlorophyll and photosynthetic efficiency and store the fixed C as 105 lipids and carbohydrates. 106 4 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 107 **MATERIALS AND METHODS** 108 **Identification of** _**Nannochloropsis oceanica**_ **NRT2 sequences** 109 _N. oceanica_ NRT2 sequences were identified by tBLASTn searches in NanDeSyn 110 (https://nandesyn.single-cell.cn/), using the six NRT2 proteins of _Phaeodactylum tricornutum_ as 111 queries against the _N. oceanica_ IMET1 genome (Wang et al., 2014); redundant hits were then 112 removed. The retrieved sequences were analyzed with TMHMM 2.0 113 (https://services.healthtech.dtu.dk/services/TMHMM-2.0/) to verify the presence of the 10-12 114 transmembrane helices (TMHs) typical of NRT2 transporters. Genomic and coding sequences were 115 confirmed by PCR amplification and sequencing (Table S1 and Data Set S1). The number of NRT2 116 genes was further verified by BLASTp against all gene models and tBLASTn against the genome 117 assembly of _N. oceanica_ CCMP1779 (JGI PhycoCosm, https://phycocosm.jgi.doe.gov/), which 118 retrieved only _NRT2.1_ and _NRT2.2_ . 119 To search for NAR2/NRT3 partner proteins, _Arabidopsis thaliana_ (AtNAR2.1 ID AT5G50200.1 and 120 AtNAR2.2 ID AT4G24730.1) and _Chlamydomonas reinhardtii_ (Cre09.g410900) proteins were used 121 as queries for BLASTp against all available _Nannochloropsis_ proteomes. NRT2.1 and NRT2.2, 122 retrieved at 100% identity, served as positive controls of the proteome search. 123 **Phylogenetic analysis of NRT2 proteins** 124 NRT2 homologs were retrieved by tBLASTn searches with default parameters and an E-value 125 threshold of 1 × 10 [-5], using the NRT2 amino acid sequences of _N. oceanica_ and diatoms as queries, 126 against NCBI (https://blast.ncbi.nlm.nih.gov/), the JGI Genome Portal 127 (https://genome.jgi.doe.gov/portal/), Ensembl Protists (https://protists.ensembl.org/index.html) and 128 PLAZA Diatoms 1.0 (https://bioinformatics.psb.ugent.be/plaza/versions/plaza_diatoms_01/). To 129 cover all major eukaryotic lineages, this dataset was complemented with NRT2 sequences from 130 UniProt (prasinophytes, trebouxiophytes, Chlorophyceae, red algae, haptophytes and cryptophytes) 131 and from the JGI PhycoCosm genomes of _Microchloropsis gaditana_ CCMP1894 and _Vischeria_ sp. 132 C74, the closest relatives of _Nannochloropsis_ with sequenced genomes. The _Tribonema minus_ 133 sequence A0A835YPG2 was excluded because of a probable annotation error (an internal tandem 134 duplication of about 30 residues). The final dataset included 94 NRT2 sequences (Supplementary 135 Data Set S2). 136 Sequences were aligned with MAFFT v.7 using the L-INS-i algorithm (Katoh and Standley, 2013), 137 and gap-rich, poorly aligned positions were removed with ClipKIT in gappy mode (gap threshold 0.4; 138 (Steenwyk et al., 2020), yielding 469 amino acid positions, 435 of which were parsimony 139 informative. A maximum likelihood (ML) tree was inferred with IQ-TREE v.2.0.7 (Minh et al., 2020) 5 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 140 under the LG+F+R7 model, selected as the best-fitting model by ModelFinder according to the 141 Bayesian information criterion (Kalyaanamoorthy et al., 2017). Branch support was assessed with 142 1000 ultrafast bootstrap replicates (Hoang et al., 2018) and 1000 replicates of the SH-like 143 approximate likelihood ratio test (SH-aLRT; Guindon _et al._, 2010). The tree was rooted at the 144 midpoint. Alternative hypotheses on the origin of the two _N. oceanica_ NRT2s (a duplication specific 145 to _Nannochloropsis_ _versus_ a duplication before the divergence of _Nannochloropsis_ and 146 _Microchloropsis_ ) were compared with the approximately unbiased (AU) test (Shimodaira, 2002) 147 implemented in IQ-TREE. 148 **Structural prediction of** _**N. oceanica**_ **NRT2s** 149 As no experimental structure of an NRT2 transporter is available, structural models of _N. oceanica_ 150 NRT2s were built with Phyre v.2.2 (Powell et al., 2025), using as template the crystal structure of the 151 major facilitator superfamily (MFS) glycerol-3-phosphate transporter GlpT of _Escherichia coli_ 152 (SCOP domain: d1pw4a; PDB: 1PW4). Models were visualized with UCSF ChimeraX (Goddard et 153 al., 2018). To identify residues putatively involved in NO 3 [−] binding, the _N. oceanica_ NRT2s were 154 aligned with _Aspergillus nidulans_ NrtA (formerly CrnA), whose key residues have been characterized 155 by mutagenesis (Unkles et al., 2004), using the MUSCLE algorithm (Edgar, 2004) implemented in 156 MEGA v.11.0.11 (Tamura et al., 2021). 157 _**N. oceanica**_ **strain and culture** 158 An axenic culture of _Nannochloropsis oceanica_ strain IMET1 was obtained from the Provasoli 159 Guillard National Center for Marine Algae and Microbiota (www.bigelow.org). Cultures were 160 maintained in F/2 medium (Guillard, 1975), prepared with 0.22 μm-filtered and autoclaved seawater, 161 at 21 °C and 100 rpm, under continuous white fluorescent light (90 μmol photons m [−2] s [−1] ). Light 162 intensity was measured with a LI-250A light meter (LI-COR, Lincoln, NE, USA). 163 **Generation and screening of** _**N. oceanica**_ _**nrt2**_ **knock-out and NRT2.2-mVenus lines** 164 Guide RNAs (gRNAs) targeting _N. oceanica_ _NRT2.1_ and _NRT2.2_ for Cpf1 (Cas12a) genome editing 165 were designed with CRISPOR (http://crispor.tefor.net/crispor.py) on the _N. oceanica_ IMET1 166 reference genome. Three crRNAs were selected for each gene, targeting regions predicted to encode 167 TMHs to increase the likelihood of disrupting NRT2 function (the complete list of gRNAs is given 168 in Table S1). The gRNAs were purchased from IDT (https://eu.idtdna.com/pages) and co-transformed 169 with a linearized antibiotic resistance cassette, following the protocol of Perin _et al._ (2015). For 170 selection, the _Aph7_ gene from _Streptomyces hygroscopicus_ (kindly provided by Prof. Krishna K. 171 Niyogi (Vieler et al., 2012), conferring resistance to hygromycin B, was used under the control of the 6 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 172 endogenous LDSP_E (Lipid Droplet Surface Protein E) promoter and the 35S terminator (Berthold 173 et al., 2002; Vieler et al., 2012). 174 To generate NRT2.2-mVenus lines, _N. oceanica_ IMET1 was co-transformed by electroporation with 175 two cassettes: one for hygromycin resistance (Berthold et al., 2002; Vieler et al., 2012) required for 176 selection, and one for NRT2.2-mVenus expression. The NRT2.2-mVenus plasmid was generated by 177 Gibson Assembly (Gibson et al., 2009). In this plasmid, the NRT2.2 coding sequence was fused at 178 its 3’ end to mVenus, followed by a His-tag, under the control of the endogenous _LDSP_E_ promoter 179 (Berthold et al., 2002; Vieler et al., 2012). Cloning primers are listed in Table S1. Both cassettes were 180 excised from their plasmids and purified on a 0.8% (w/v) agarose gel. Transformation followed the 181 protocol of Perin _et al._ (2015). Resistant colonies were picked after 4-5 weeks and screened by PCR 182 with the primers listed in Table S1. 183 **Subcellular localization of NRT2.2** 184 Cells were imaged with a Zeiss LSM 700 confocal laser scanning microscope (Carl Zeiss, Jena, 185 Germany) equipped with a Plan-Apochromat 63x/1.40 Oil DIC M27 objective. mVenus was excited 186 at 488 nm and detected at 524-536 nm, and chlorophyll _a_ autofluorescence was excited at 640 nm 187 and detected at 655-667 nm. Laser power, detector gain and offset were identical for all acquisitions. 188 Single optical sections were acquired with ZEN Blue v.3.14 software (Carl Zeiss) at 1024 × 1024 189 pixels (0.099 µm pixel [-1] ; WT and line #11) or 512 × 512 pixels (0.198 µm pixel [-1] ; line #9). 190 Fluorescence distribution was quantified on the original 16-bit images with custom R scripts (image 191 quantification) and Python scripts (profile analysis and figure assembly). In each image, the median 192 background was subtracted from both channels and intensities were divided by the pixel dwell time 193 (counts µs [-1] ), so that images acquired with different settings could be compared. Bleed-through of 194 chlorophyll _a_ autofluorescence into the mVenus channel was estimated on WT cells, which do not 195 express mVenus, as a constant fraction of the chlorophyll signal (4.1%), and subtracted from the 196 mVenus channel of all images. In total, 32 cells from 15 independent fields were analyzed (WT, n = 197 10; NRT2.2-mVenus #9, n = 9; NRT2.2-mVenus #11, n = 13). In cells showing a peripheral mVenus 198 ring, the cell center and ring diameter were determined by fitting a circle to the ring. Intensity profiles 199 of mVenus and chlorophyll _a_ were extracted along 18 diameters through the cell center and averaged 200 for each cell. Peripheral enrichment was calculated as the ratio between the mean mVenus intensity 201 within ±0.3 µm of the ring radius and that within the inner half of the radius. Co-localization of 202 mVenus with chlorophyll _a_ was assessed with Pearson’s correlation coefficient calculated on the 203 bleed-through-corrected images. 7 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 204 **Growth under different N sources and concentrations** 205 _N. oceanica_ wild-type (WT) and knock-out (KO) lines were maintained in autoclaved filtered F/2 206 medium (Guillard, 1975) enriched in NH 4 [+] (and 882 µM NH 4 Cl) in addition to the standard addition 207 of 882 µM NaNO 3 . After three days (mid-exponential phase), cells were transferred to N-free F/2 208 medium for 72 h to deplete residual N. After 72 h of N starvation, cells were diluted to 1 × 10 [7] cells 209 mL [-1] in F/2 medium supplemented with 882 μM NaNO 3, 250 μM NaNO 3 or 882 μM NH 4 Cl. The 210 experiment was performed first in 20 mL Erlenmeyer flasks and then in 80 mL tubes in a Multi 211 Cultivator MC 1000-OD (Photon Systems Instruments, Drásov, Czech Republic), under atmospheric 212 CO 2 and continuous white light (35 and 80 μmol photons m [-2] s [-1], respectively). Each condition was 213 tested in at least three biological replicates. 214 Growth was monitored as optical density at 750 nm (OD 750 ) with a Spark multimode microplate 215 reader (Tecan, Männedorf, Switzerland) and confirmed by cell counting with an automated cell 216 counter (Cellometer Auto X4, Nexcelom Bioscience, Lawrence, MA, USA). In the Multi-Cultivator, 217 OD 680 was recorded automatically every hour. 218 To measure extracellular NO 3 [−] consumption, 1 mL of culture was filtered by vacuum filtration onto 219 glass microfiber Whatman GF/F filters with nominal pore size 0.7 µm (Sigma-Aldrich, St. Louis, 220 MO, USA), and the absorbance of the filtrate was measured at 220 nm with a Cary 60 UV-Vis 221 spectrophotometer (Agilent Technologies, Santa Clara, CA, USA). NO 3 [−] concentrations were 222 calculated from a standard curve (10 μM to 1 mM NO 3 [−] in N-free F/2 medium), according to 223 McCarthy _et al._ (2017). 224 Data are presented as mean ± SD of independent biological replicates. Differences between each KO 225 line and the WT were tested with an unpaired two-tailed Student’s t-test (for growth and NO 3 [−] uptake, 226 at day 4). Statistical analyses were performed with Python (SciPy). 227 **Pigments and protein extraction and quantification** 228 Chlorophylls _a_ (Chl _a_ ) and total carotenoids (Car) were extracted in the dark from culture samples 229 with 100% N,N-dimethylformamide (DMF) (Sigma-Aldrich, St. Louis, MO, USA), following Liistro 230 _et al._ (2026). Pigments in the extracts were quantified with an Agilent Cary 60 spectrophotometer 231 (Agilent Technologies, Santa Clara, CA, US) using specific extinction coefficients (Moran, 1982). 232 Absorbance at 750 nm was subtracted from all readings to correct for light scattering, and pigment 233 content was expressed per cell using the cell concentration of each culture. 234 A total of 2 × 10 [8] cells were harvested from each culture, washed twice, and resuspended in 180 µL 235 B1 buffer (20 mM Tris-KOH, pH 7.8, 400 mM NaCl, 2 mM MgCl 2, 3 mM benzamidine, 1 mM 236 PMSF). Cells were disrupted with acid-washed 150-212 µm glass beads (Sigma-Aldrich, St. Louis, 8 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 237 MO, USA) using a Bullet Blender Storm Pro (Next Advance, Inc., Troy, NY, USA) for four cycles 238 of 30 sec at 3500 rpm, with 3 min on ice between cycles. Subsequently, 80 µL of 3X sample buffer 239 (45 mM Tris-HCl, pH 6.8, 30 mM DTT, 3% SDS, 10% glycerol) were added, followed by an 240 additional 30 sec disruption step. 241 After extraction, proteins were precipitated with 4 volumes of cold 100% acetone for 1 h at -20 °C. 242 Samples were centrifuged for 10 min at 15000 g at 4°C. The pellet was dried and then resuspended 243 in 80 µl of B1 buffer (without protease inhibitors). Protein quantification was performed using a 244 Bradford assay. For each sample, 33 µl were added to 1 ml of Bradford Reagent (Sigma-Aldrich, St. 245 Louis, MO, USA) and incubated for 10 min at room temperature in the dark. Absorbance was 246 measured at 595 nm. Protein concentrations were calculated from a bovine serum albumin (BSA) 247 standard curve and expressed per cell, based on the 2 × 10 [8] cells extracted per sample. 248 **Determination of photosynthetic efficiency** 249 Chlorophyll fluorescence was measured _in vivo_ with a Dual-PAM-100 fluorometer (Heinz Walz, 250 Effeltrich, Germany). After 25 min of dark adaptation, the maximum quantum yield of Photosystem 251 II (PSII) was calculated as F v /F m = (F m −F 0 )/F m (Maxwell and Johnson, 2000), where F 0 is the 252 minimum fluorescence of dark-adapted cells and F m the maximum fluorescence measured during a 253 saturating light pulse (~7000 μmol photons m [−2] s [−1], 600 ms). 254 **Quantification of intracellular neutral lipids** 255 Intracellular neutral lipids were quantified by Nile Red staining (Sigma-Aldrich, St. Louis, MO, 256 USA). 1 × 10 [7] cells were collected, resuspended in 200 µL of F/2 medium and then stained with 0.625 257 µg mL [−1] Nile Red (DMSO stock solution) for 15 min at 37 °C in the dark with shaking (100 rpm). 258 Fluorescence was measured with the Tecan microplate reader described above (excitation 485 nm, 259 emission 580 nm) and corrected for background fluorescence using unstained samples and water 260 blanks. 261 **Elemental C and N composition** 262 For C and N quantification, cell pellets were washed three times with an isosmotic ammonium 263 formate solution (0.5 M) and freeze-dried. Capsules containing 0.8-1.2 mg of dried sample were 264 analyzed with a CHN elemental analyzer (ECS 4010, Costech, Italy) coupled to an ID Micro EA 265 isotope ratio mass spectrometer (Compact Science Systems, Newcastle-under-Lyme, UK) to 266 determine C and N content as a percentage of sample dry weight. Urea was used as the analytical 267 standard. 9 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 268 In the _nrt2.2_ KO lines grown with NO 3 [−], N was below the detection limit of 4% N/DW in all but one 269 replicate; as the single measurable value was itself above the detection limit, these samples were 270 reported as having %N below this threshold (Fig. 3C). Because N could not be detected, the 271 instrument software did not calculate %C for these samples. %C was therefore calculated from the C 272 peak areas as follows, after verifying that this calculation reproduced the values automatically 273 reported for the samples in which %C had been provided: 274 C sample = C urea × (A sample / A urea ) / (A ref,sample / A ref,urea ) 275 %C sample = C sample × 100 / DW 276 **Fourier-transform infrared (FTIR) spectroscopy** 277 About 1 mg of dried cells, previously washed three times with the isosmotic ammonium formate 278 solution, was weighed and resuspended in 110 μL of grade 1 water; 50 μL of the suspension were 279 placed on a silicon disk and dried at 80 °C. FTIR spectra of whole cells were acquired in a range from 280 400 to 4000 cm [−1], sampling every 1.93 cm [−1] with a Tensor 27 FTIR spectrometer (Bruker Optics 281 GmbH, Ettlingen, Germany). FTIR bands were assigned to cellular pools of proteins, carbohydrates, 282 and lipids as described in Giordano _et al._ (2001). The band at about 1656 cm [-1] was assigned to protein, 283 the one at about 1740 cm [-1] to lipids and the ones in the range 1160-1030 cm [-1] to carbohydrates. Band 284 integrals of the deconvolved spectra were obtained using the OPUS v.6.5 software (Bruker Optics 285 GmbH, Ettlingen, Germany) and used to calculate the relative abundances of proteins, carbohydrates 286 and lipids. For each strain and condition, three biological replicates were analyzed in technical 287 duplicates, and technical duplicates were averaged before statistical analysis. 288 10 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 289 **RESULTS** 290 **1.** **Identification, evolution and structure of** _**N. oceanica**_ **NRT2 transporters** 291 tBLASTn searches against the _N._ _oceanica_ IMET1 genome (Wang et al., 2014) identified two NRT2 292 sequences, NO30G01230 and NO14G02450, hereafter referred to as NRT2.1 and NRT2.2 293 respectively (Data Set S1). Both gene models were supported by available transcriptomic data and 294 confirmed by PCR. 295 To place the _N. oceanica_ NRT2 proteins in an evolutionary context, we reconstructed a maximum 296 likelihood phylogeny of 94 NRT2 proteins from stramenopiles, the green lineage, red algae, 297 haptophytes, cryptophytes and fungi (Fig. 1A; Supplementary Data Set S2). The tree resolved two 298 strongly supported groups: one comprised sequences from land plants and green algae together with 299 those of haptophytes, red algae and cryptophytes, whereas the other included stramenopiles and fungi 300 (Fig. 1A). In the tree, NRT2.2 groups with the single NRT2 of _Microchloropsis_, whereas NRT2.1 301 branches outside this pair (Fig. 1A). Together with an AU test rejecting a duplication within 302 _Nannochloropsis_ (p = 0.002), this suggests that the two _N. oceanica_ NRT2s arose from a duplication 303 that occurred before the divergence of _Nannochloropsis_ and _Microchloropsis_, followed by the loss 304 of NRT2.1 in _Microchloropsis_ . 305 We generated structural models of _N. oceanica_ NRT2.1 and NRT2.2 using Phyre2.2 (template: 306 bacterial MFS transporter d1pw4a, PDB: 1PW4). Both models predicted 12 transmembrane helices 307 (TMHs), with the N- and C-termini in the cytoplasm (Fig. 1B). Since the two models were highly 308 similar, NRT2.2 is shown as representative (Fig. 1B), and the corresponding NRT2.1 residues are 309 reported in Fig. S1. Mutagenesis studies of _Aspergillus nidulans_ NrtA (previously CrnA) (Unkles et 310 al., 2004) identified key residues required for NO 3 [-] binding. Two positively charged arginine residues, 311 R86 in TMH2 and R302 in TMH8, are conserved in NRT2.2 and correspond to residues required for 312 NO 3 [-] binding in NrtA (Fig. 1B; Fig. S1 for NRT2.1). Three further conserved residues were identified 313 in the same putative binding site: N167 in TMH4, which may contribute to substrate stabilization; 314 E265, whose negative charge may balance the positive charges of R86 and R302 through a salt bridge; 315 and N390, likely involved in positioning NO 3 [-] or maintaining the conformation of the translocation 316 pathway (Fig. 1B) (Unkles et al., 2004). All five residues are also conserved in NRT2.1 (Fig. S1). 317 The conservation of the complete binding-site architecture in both paralogs indicates that both 318 proteins could, in principle, transport NO 3 [-] . 319 11 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 320 321 **Figure 1. Phylogeny and predicted structure of** _**Nannochloropsis**_ _**oceanica**_ **NRT2s.** A) Maximum322 likelihood phylogeny of 94 NRT2 proteins from 50 photosynthetic and non-photosynthetic 323 eukaryotes. Filled and open circles indicate nodes with ultrafast bootstrap (UFBoot) support ≥ 95% 324 and 70-94%, respectively; SH-aLRT/UFBoot values are shown for the eustigmatophyte clade and for 325 the NoNRT2.2- _Microchloropsis_ clade. Colors indicate taxonomic groups; _N. oceanica_ NRT2.1 and 326 NRT2.2 are in orange. Sequences and accession numbers are listed in Supplementary Data Set S1. 327 B) Predicted structural model of _N. oceanica_ NRT2.2 with 12 transmembrane helices and, in the 328 zoomed view, the putative NO 3 [-] binding site: R86 and R302 (green), E265 (magenta), and N167 and 329 N390 (blue). 330 12 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 331 **2.** **Loss of NRT2.2 abolishes NO** **3** **[-]** **uptake in** _**N. oceanica**_ 332 To dissect the functional role of the two NRT2 transporters, we generated _N. oceanica_ _nrt2.1_ and 333 _nrt2.2_ knock-out (KO) mutants using a CRISPR-Cpf1 genome editing approach, targeting each gene 334 with three different gRNAs (Fig. S2). Amplification and sequencing of the target loci confirmed the 335 mutations (Fig. S2). In the _nrt2.1_ KO lines #1 and #2, the hygromycin resistance cassette was 336 integrated at the target locus. _nrt2.2_ KO line #1 carries the cassette in the antisense orientation relative 337 to NRT2.2, whereas line #2 carries the cassette in the sense orientation (Fig. S2). RT-PCR analysis 338 confirmed the absence of detectable _NRT2.1_ or _NRT2.2_ transcripts in two independent KO lines per 339 gene (Fig. 2A), which were selected for phenotypic characterization. 340 The KO lines were isolated and maintained in medium containing both NO 3 [-] and NH 4 [+], in which 341 _nrt2.1_ and _nrt2.2_ KO strains grew similarly to the wild type (WT) (data not shown), as NH 4 [+] can 342 sustain growth independently of NO 3 [-] transport. To test NO 3 [-] utilization specifically, cells were 343 starved of N for 3 days to deplete intracellular N reserves and then resupplied with either NO 3 [-] or 344 NH 4 [+] as the sole N source (Fig. 2B). Upon resupply of 882 µM NO 3 [-], extracellular NO 3 [-] progressively 345 decreased in WT and _nrt2.1_ KO cultures, to <100 µM after 4 days, whereas it remained unchanged 346 in both _nrt2.2_ KO lines, demonstrating a complete loss of NO 3 [-] uptake (Fig. 2C). In WT cultures 347 ~20% of the NO 3 [-] was already taken up after 1 day, before any increase in cell number (Fig. 2C,D), 348 indicating that the lack of NO 3 [-] uptake in _nrt2.2_ KO lines is not a consequence of impaired growth. 349 Accordingly, WT and _nrt2.1_ KO cultures resumed growth, tripling cell density within 4 days, whereas 350 both _nrt2.2_ KO lines showed no increase in cell number (Fig. 2D). The same phenotype was observed 351 at a lower NO 3 [-] concentration (250 µM): _nrt2.2_ KO lines neither took up NO 3 [-] nor grew, whereas WT 352 and _nrt2.1_ KO cells reduced extracellular NO 3 [-] by ~70% within 4 days (Fig. S3A,B). 353 In contrast, when NH 4 [+] was resupplied as the sole N source, _nrt2.2_ KO strains recovered similarly to 354 WT and _nrt2.1_ KO (Fig. 2E,F), indicating that their growth defect is specific to NO 3 [-] and is not a 355 general effect of the mutations. These findings identify NRT2.2 as an essential component of the 356 NO 3 [-] uptake machinery in _N. oceanica_, required for NO 3 [-] acquisition and NO 3 [-] -dependent growth. 13 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 357 358 **Figure 2. Impaired growth and nitrate uptake in** _**Nannochloropsis oceanica**_ _**nrt2.2**_ **KO mutants.** 359 A) RT-PCR of NRT2.1 and NRT2.2 transcripts in the wild type (WT) and in two _nrt2.1_ and two 360 _nrt2.2_ knock-out (KO) lines (#1, #2); ACT (actin), reference transcript. Primers are listed in Table 361 S1. B) Experimental setup. Cells were grown under atmospheric CO 2 (400 ppm) and continuous white 362 light (35 µmol photons m [-2] s [-1] ) in 882 µM nitrate (NO 3 [-] ) and 882 µM ammonium (NH 4 [+] ) until 363 exponential phase, starved of N for 3 days and then resupplied with NO 3 [-] or NH 4 [+] . C) Residual NO 3 [-] 364 in the medium and D) cell concentration during 4 days in 882 µM NO 3 [-] . E) Cell concentration in 882 365 µM NH 4 [+] (control). F) Cultures after 4 days of NO 3 [-] or NH 4 [+] repletion. Data are mean ± SD of at 366 least three biological replicates. Asterisks: unpaired two-tailed Student's t-test vs WT at day 4; * p < 367 0.05, ** p < 0.01, *** p < 0.001. 368 369 **3.** **Intracellular C and N partitioning following loss of NRT2.2** 370 To understand the consequences of the loss of NO 3 [-] uptake on cell physiology, we characterized the 371 C and N composition, storage metabolism and photosynthetic performance of WT, _nrt2.1_ KO and 372 _nrt2.2_ KO strains. To obtain enough biomass, cells were grown in a Multi-Cultivator photobioreactor 14 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 373 under the same starvation-repletion regime and sampled at the end of the 3-day N depletion (hereafter 374 -N) and after 4 days of repletion with NH 4 [+] or NO 3 [-] (hereafter +NH 4 [+] and +NO 3 [-] ), the same time point 375 used in flasks, when WT cultures had consumed most of the NO 3 [-] (Fig. 2C). As in flasks, _nrt2.2_ KO 376 lines neither grew (Figs. 3A and S3C,D) nor consumed NO 3 [-] (Fig. S4) in +NO 3 [-], whereas all strains 377 behaved like WT in -N and +NH 4 [+] (Fig. S4). Since _nrt2.2_ KO cells cannot use NO 3 [-], at the end of the 378 +NO 3 [-] treatment they had been without an available N source for 7 days (3 days of depletion followed 379 by 4 days of repletion), compared with 3 days for -N samples. 380 The lack of NO 3 [-] uptake caused a strong intracellular N deficiency. In +NO 3 [-], _nrt2.2_ KO cells showed 381 a slight increase in total carbon (C) content relative to WT and _nrt2.1_ KO (Fig. 3B) and a strong 382 reduction in N content, with five of six samples falling below the detection limit of the assay (Fig. 383 3C), yielding a sharply elevated C/N ratio (Fig. S5). Since most cellular N is allocated to proteins, we 384 then quantified total protein content, which was unaffected across genotypes and conditions (Figs. 385 3D and S6). _nrt2.2_ KO cells therefore maintained their protein content per cell while accumulating 386 C. Since N is expressed per dry weight, its decrease largely reflects dilution by C-rich compounds. In 387 parallel, maximum PSII efficiency (F v /F m ) was strongly reduced in _nrt2.2_ KO cells in +NO 3 [-] (0.43 388 0.49 _vs_ 0.65 in WT; Fig. 3E), well below the values of all strains in -N (~0.61) and +NH 4 [+] (Fig. S6), 389 without an available N source (7 vs 3 days). Chlorophyll, a major N sink of the photosynthetic 390 apparatus, was reduced by up to ~50% in _nrt2.2_ KO cells, to levels similar to −N (Figs. 3F and S7), 391 and the overall pigment composition shifted (Fig. 3G). The decrease in F v /F m in _nrt2.2_ KO cells was 392 thus accompanied by a loss of chlorophyll, while total protein per cell was maintained. 15 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 393 394 **Figure 3. Altered intracellular N content and photosynthetic performance in** _**Nannochloropsis**_ 395 _**oceanica**_ _**nrt2.2**_ **knock-out mutants.** Wild-type (WT), two _nrt2.1_ and two _nrt2.2_ knock-out (KO) 396 lines were grown in the Multi-Cultivator system under atmospheric CO 2 (400 ppm) and continuous 397 white light (80 µmol photons m [-2] s [-1] ) and analyzed after 4 days of repletion with 882 µM nitrate (NO 3 [-] 398 ); growth curves are shown in Figure S3 and the control conditions, nitrogen (N) deprivation and 399 NH 4 [+] repletion, in Figures S4-S7. A) Biomass production. B) Carbon (C) and C) N content, as 400 percentage of dry weight. The detection limit for N is 4% of dry weight (grey area): N was not 401 detected (n.d.) in five of six _nrt2.2_ KO samples, and the only measurable sample ( _nrt2.2_ KO #2) is 402 shown. D) Total protein content. E) Maximum photosynthetic efficiency, expressed as F v /F m, and F) 403 chlorophyll content. G) Relative pigment composition (% chlorophyll and total carotenoids). Data 404 are mean ± SD of at least three biological replicates. Asterisks: unpaired two-tailed Student’s t-test 405 vs WT; * p < 0.05, ** p < 0.01, *** p < 0.001. 406 407 We next asked where the excess C was stored. To answer this, whole cells were analyzed by Fourier 408 transform infrared (FTIR) spectroscopy, which measures the absorption of the main functional groups 409 of cellular macromolecules: the amide I band of proteins (~1650 cm [-1] ), the ester C=O band of lipids 410 (~1740 cm [-1] ) and the C-O-C bands of carbohydrates (950-1200 cm [-1] ) (Fig. 4A-C). Because total 411 protein content was stable across strains and conditions (Figs. 3D and S6), spectra were normalized 412 to the amide I band, and lipid/protein and carbohydrate/protein ratios were quantified by fitting the 413 spectra with Lorentzian bands (Fig. 4D,E). After 3 days of N depletion, all strains accumulated lipids, 414 with a lipid/protein ratio about twice that in +NH 4 [+], while the carbohydrate/protein ratio did not 415 increase (Fig. 4A,B,D,E). In both conditions, the ratios of _nrt2.2_ KO lines were similar to WT; the 416 only differences, in single _nrt2.1_ KO lines, were not confirmed in the second line (Fig. 4D,E). In 16 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 417 +NO 3 [-], by contrast, _nrt2.2_ KO spectra diverged from all other strains, with a clear increase in both 418 bands (Fig. 4C). The lipid/protein ratio of _nrt2.2_ KO lines was ~3.7-fold higher than in WT (Fig. 4D), 419 and the carbohydrate/protein ratio ~1.4-fold higher (Fig. 4E). 420 Neutral lipids were also measured independently by Nile Red staining, a fluorescent dye that binds 421 neutral lipids such as triacylglycerols in intact cells. In all strains, Nile Red fluorescence was ~2.5 422 fold higher after N depletion than in +NH 4 [+] (Fig. 4F). In +NO 3 [-], _nrt2.2_ KO lines showed ~3-fold 423 higher fluorescence than WT, reaching or slightly exceeding N-depleted levels (Fig. 4F). C fixation 424 therefore continues in _nrt2.2_ KO cells, but in the absence of NO 3 [-] uptake fixed C cannot be 425 incorporated into N-containing compounds and is redirected into lipid and carbohydrate storage: an 426 N-starvation signature, although NO 3 [-] is available in the medium. 427 428 **Figure 4. Increased lipid and carbohydrate content in** _**Nannochloropsis oceanica**_ _**nrt2.2**_ **knock-** 429 **out mutants.** Fourier Transform Infrared (FTIR) spectra of _N. oceanica_ wild-type (WT), two _nrt2.1_ 430 and two _nrt2.2_ knock-out (KO) lines grown in the Multi-Cultivator system, under atmospheric CO 2 431 (400 ppm) and continuous white light (80 µmol photons m [-2] s [-1] ) A) after 3 days of nitrogen (N) 432 depletion, B) after 4 days of repletion with 882 µM ammonium (NH 4 [+] ), and C) after 4 days of 433 repletion with 882 µM nitrate (NO 3 [-] ). Spectra are normalized to the protein amide I peak at ~1650 434 cm-1; total protein did not differ between strains (Fig. S6). Shaded areas mark the spectral regions of 435 the fitted bands used for quantification: lipid ester C=O (1770-1710 cm [-1] ) and carbohydrate C-O-C 436 (1200-950 cm [-1] ); colored asterisks refer to the band ratios in D and E. D) Lipid (ester C=O) and E) 437 carbohydrate (C-O-C) band areas relative to the protein amide I band (~1656 cm [-1] ), obtained by fitting 17 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 438 the spectra with Lorentzian bands (lipids: ester C=O band at ~1740 cm [-1] ; carbohydrates: sum of the 439 bands between ~1030 and ~1160 cm [-1] ). F) Neutral lipids, measured as Nile Red fluorescence, in the 440 same conditions. Data are mean ± SD of at least three biological replicates (FTIR technical duplicates 441 were averaged). Asterisks: unpaired two-tailed Student’s t-test vs WT; * p < 0.05, ** p < 0.01, *** p 442 < 0.001. 443 444 **4.** **Subcellular localization of NRT2.2 on the plasma membrane** 445 Since loss of NRT2.2 caused the strongest phenotype under NO 3 [-] supply, we focused on this 446 transporter and determined its subcellular localization. We generated stable lines expressing NRT2.2 447 fused at its C-terminus to mVenus (NRT2.2-mVenus) (Fig. 5A). After transformation, independent 448 lines were screened by PCR for integration of the construct (Fig. 5B), and two positive lines (#9 and 449 #11) were selected for confocal microscopy. 450 In both lines, mVenus formed a ring at the cell periphery, flanking the chlorophyll signal (Fig. 5C,D). 451 mVenus fluorescence at the cell edge was 3.9 ± 1.2-fold (line #9, n = 4 cells) and 4.3 ± 0.9-fold (line 452 #11, n = 5 cells) higher than in the cell interior, and the diameter of the fluorescent ring (3.1 ± 0.4 and 453 3.4 ± 0.5 µm, respectively) was consistent with the size of _N. oceanica_ cells. In addition, the 454 correlation between mVenus and chlorophyll signals within each cell was close to zero or slightly 455 negative (mean Pearson’s r = −0.13 in both lines; Fig. 5E), indicating that the fusion protein is 456 excluded from the chloroplast. Together, these data localize NRT2.2 to the plasma membrane, at the 457 interface between the cell and the external medium, consistent with its role in NO 3 [-] uptake from 458 seawater in _N. oceanica_ . 18 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 459 460 **Figure 5. Subcellular localization of** _**Nannochloropsis oceanica**_ **NRT2.2 to the plasma** 461 **membrane.** A) Expression construct: NRT2.2 coding sequence fused at its 3’ end to the yellow 462 fluorescent protein mVenus, co-transformed with a hygromycin resistance cassette. B) Genomic PCR 463 confirming the integration of the construct (1834 bp amplicon, primers for/rev in A) in lines #9 and 464 #11; WT, wild-type parental strain; C+, transformation plasmid. C) Confocal images of WT and 465 NRT2.2-mVenus cells: bright field (BF), chlorophyll _a_ autofluorescence (magenta), mVenus (green) 466 and merge. mVenus is corrected for chlorophyll bleed-through and shown with the same intensity 467 range in all lines. Scale bars = 5 μm. D) Mean intensity profiles (± SD) of mVenus (green) and 468 chlorophyll _a_ (magenta, normalized to the maximum of each profile) across the cell (18 diameters 469 per cell; position normalized to the ring radius, ring at ±1; WT scaled by the mean ring radius, 1.63 470 µm), in WT cells (n = 10) and in NRT2.2-mVenus cells with a peripheral ring (#9: n = 4; #11: n = 5); 471 mVenus is on a separate scale for each line. E) Pearson’s correlation between mVenus and 472 chlorophyll _a_ signals after bleed-through correction in the NRT2.2-mVenus lines #9 and #11; each 473 dot is one cell, bars are means. 474 19 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 475 **DISCUSSION** 476 N availability in marine environments fluctuates in time and space, and N is often the primary limiting 477 nutrient for marine primary production (Gruber, 2008; Sanz-Luque et al., 2015). Here we identified 478 NRT2.2 as the plasma membrane transporter that mediates NO 3 [-] uptake in _N. oceanica_ . NRT2.2 is 479 required for NO 3 [-] -dependent growth, and its loss disrupts the balance between C and N metabolism, 480 driving cells into a physiological state that resembles prolonged N starvation, even though NO 3 [-] is 481 available. 482 **NRT2.2 is the main gateway for nitrate acquisition in** _**N. oceanica**_ 483 _N. oceanica_ possesses only two _NRT2_ genes, _NRT2.1_ and _NRT2.2_, whereas the closely related 484 _Microchloropsis_ species carry a single one (Radakovits et al., 2012), although annotation 485 inconsistencies in these genomes call for caution when estimating gene number (Fig. 1A). The single 486 NRT2 retained in _Microchloropsis_ is the ortholog of NRT2.2 (Fig. 1A), suggesting that the 487 duplication occurred before the split between the two genera and was followed by the loss of NRT2.1 488 in _Microchloropsis_ . This is consistent with NRT2.2 representing the core NO 3 [-] transporter of the 489 lineage, and NRT2.1 an accessory paralog. However, NRT2 sequences cluster mainly by taxonomic 490 lineage (Rogato et al., 2015; Busseni et al., 2019), so phylogenetic position alone cannot predict the 491 function of each paralog. This compact organization contrasts with that of land plants and green algae, 492 in which NO 3 [-] uptake relies on multiple, partially redundant NRT2 members with distinct regulatory 493 profiles, such as AtNRT2.1 and AtNRT2.2, the major components of the high-affinity NO 3 [-] transport 494 system in _Arabidopsis thaliana_ (Filleur et al., 2001; Li et al., 2007; Kotur et al., 2012), and the 495 multiple, differentially regulated NRT2 homologs of _Chlamydomonas reinhardtii_ (Fernandez and 496 Galvan, 2008). 497 In land plants and _C. reinhardtii_, high-affinity NO 3 [-] uptake by NRT2 requires the partner protein 498 NAR2 (Zhou et al., 2000; Rexach et al., 2002), which is absent from _N. oceanica_ and from other 499 algae with secondary plastids (Rogato et al., 2015). The complete loss of NO 3 [-] uptake in the _nrt2.2_ 500 KO lines (Fig. 2C) shows that NRT2.2 is functional _in vivo_ without NAR2, either alone, like the 501 fungal transporter NrtA, or with a partner protein not yet identified. However, whether NRT2.2 502 operates as a high-affinity transporter remains to be tested: the uptake assays used here were 503 performed at 250-882 µM NO 3 [-], well above typical seawater concentrations, and kinetic 504 measurements at low micromolar NO 3 [-] will be needed to determine its affinity. Beyond copy number, 505 NRT2 proteins are conserved across eukaryotic lineages, and this conservation extends to the 506 structural level. Although no eukaryotic NRT2 structure has been experimentally resolved, the 507 predicted models of _N. oceanica_ NRT2 showed the typical 12-TMH architecture and the conservation 20 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 508 of charged and polar residues, suggesting that the molecular determinants required for NO 3 [-] 509 recognition and membrane transport remain highly conserved (Fig. 1B). 510 The knock-out mutants showed that one of the two copies has the main role: loss of NRT2.2 511 completely abolished NO 3 [-] uptake (Fig. 2C) and growth on NO 3 [-] (Fig. 2D), whereas growth on NH 4 [+] 512 was comparable to the WT (Fig. 2E). The defect is therefore specific to NO 3 [-] acquisition and identifies 513 NRT2.2 as the major gateway for extracellular NO 3 [-] in this species. The absence of compensation by 514 NRT2.1 indicates that, unlike in plants and green algae, NO 3 [-] acquisition in _N. oceanica_ relies on a 515 single dominant transporter rather than on partially redundant family members. This specialization 516 may reflect the ecological conditions of marine microalgae: unlike land plants, which face NO 3 [-] 517 gradients largely determined by soil properties and root architecture (Kabala et al., 2017), marine 518 phytoplankton experience rapidly changing nutrient conditions, where NO 3 [-] can shift from complete 519 depletion in stratified, nutrient-poor waters to sudden enrichment after upwelling or mixing (Gruber, 520 2008; Sanz-Luque et al., 2015). In this setting, because NRT2.2 controls NO 3 [-] entry, its capacity may 521 determine how quickly _N. oceanica_ resumes growth when NO 3 [-] becomes available again. 522 No clear phenotype was found in the _nrt2.1_ KO lines, and three explanations are possible. First, 523 NRT2.2 may fully compensate for the loss of NRT2.1 (but not vice versa), as documented for NRT2 524 paralogs in vascular plants (Xu et al., 2024; Berman et al., 2026). Second, the two transporters may 525 differ in affinity or kinetic properties (Liu et al., 1999). The NO 3 [-] concentrations tested here (250-882 526 µM, much higher than in seawater), set by the resolution of the assay, may not have captured the 527 activity range of NRT2.1. Third, the expression of _N. oceanica_ NO 3 [-] transporters responds to 528 environmental conditions, including timing of N starvation (Li et al., 2014), light and C supply (Wei 529 et al., 2019), so NRT2.1 may become relevant only under conditions not examined here. 530 Finally, the plasma membrane localization of NRT2.2 (Fig. 5) places it at the interface between the 531 cell and its environment, consistent with its role in the uptake of extracellular NO 3 [-] . 532 **Loss of NRT2.2 disrupts the C/N balance and triggers a starvation-like metabolic state** 533 The consequences of impaired NO 3 [-] uptake extended beyond growth inhibition and highlighted a 534 central role of NRT2.2 in maintaining cellular metabolic homeostasis. Disrupting the main NO 3 [-] entry 535 point produces an internal N deficit in cells that remain surrounded by NO 3 [-], and the resulting 536 phenotype can be read as the metabolic signature of N starvation, imposed here genetically rather 537 than environmentally. 538 The most direct consequence of NRT2.2 disruption was the decrease of chlorophyll content and 539 photosynthetic performance (Figs. 3F,G and 6), indicating that N limitation primarily affected the 540 abundance and functional capacity of the photosynthetic apparatus. Chlorophyll is an N-rich 21 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 541 molecule, and the light-harvesting complexes bind it in large amounts. Reducing pigment content 542 could therefore represent a direct way to lower N demand without dismantling the protein scaffold of 543 the photosystems (Wang et al., 2024). This is consistent with N deprivation triggering chloroplast 544 remodeling, degradation of photosynthetic pigments and reduction of photosystem activity in 545 microalgae (Bender et al., 2014; Fattore et al., 2021). In N-starved WT _Nannochloropsis_, this 546 remodeling is an acclimation response: photosynthesis remains operational and supplies energy and 547 reducing power for cell maintenance and lipid synthesis (Simionato et al., 2013; Wang et al., 2024). 548 The _nrt2.2_ KO lines instead face a permanent block of NO 3 [-] uptake, and their phenotype reflects a 549 chronic N deficiency that parallels, but does not necessarily reproduce, the response of WT cells to 550 N starvation. 551 In _N. oceanica_, long-term N deprivation drives a global down-regulation of protein expression, with 552 the induction of N-scavenging enzymes, proteases and proteasome components to recycle N from 553 existing cellular components, and autophagy of plastids (Dong et al., 2013; Simionato et al., 2013). 554 In the _nrt2.2_ KO, by contrast, total protein content remained stable (Figs. 3D and 6): these lines 555 reproduce chlorosis and reduced photosynthetic performance, but not the proteome-wide N recycling 556 of N-starved cells, which in WT cells appears only after longer starvation. The sharp decrease in N 557 content despite stable protein levels reflects the different normalization of the two measurements: N 558 is expressed per dry weight and protein per cell, so the accumulation of C-rich lipids and 559 carbohydrates (Fig. 4) increases cell dry weight and dilutes N, while non-protein N pools such as 560 chlorophyll also decrease (Fig. 3F). Two non-exclusive interpretations can be proposed. First, N 561 allocation may follow a hierarchy in which pigments are sacrificed before the protein pool, preserving 562 the core cellular machinery. Second, the absence of bulk protein remobilization may indicate that 563 _nrt2.2_ KO cells do not fully engage the canonical N-recycling program of N-starved cells (Dong et 564 al., 2013; Wang et al., 2024), so that the chronic deficit is not buffered by protein degradation. This 565 makes the _nrt2.2_ KO a useful tool to dissect which components of the starvation response depend on 566 the perception of external N and which on the internal N deficit. 567 Another metabolic consequence of NRT2.2 loss was the accumulation of C-rich storage compounds, 568 including neutral lipids and carbohydrates (Figs. 4 and 6). This matches the classical response to N 569 deficiency and reflects the metabolic coupling between C and N. The reduction of NO 3 [-] to NH 4 [+] and 570 its subsequent incorporation into amino acids through the GS-GOGAT cycle consume C skeletons 571 (2-oxoglutarate), ATP and reducing power from photosynthesis. In the _nrt2.2_ KO, NO 3 [-] cannot enter 572 the cell and none of these demands arises, while light-driven C fixation continues, albeit probably at 573 a reduced capacity (Figs. 3C and 6) (Santin et al., 2026). The fixed C and reducing power that would 574 normally feed amino acid synthesis are therefore redirected (McCarthy et al., 2017; Smith et al., 2019) 22 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 575 toward storage compounds (Simionato et al., 2013; Fattore et al., 2021). This response is shared by 576 the _nrt2.2_ KO and by N-starved WT cells even though the two states differ in their pigment and 577 protein responses, suggesting that C storage is a robust consequence of N deficiency irrespective of 578 how the deficiency is imposed. Triacylglycerols (TAG) allow cells to store excess C in an osmotically 579 neutral form (Goncalves et al., 2016). 580 Carbohydrates also increased in the _nrt2.2_ KO lines (Fig. 4E). Carbohydrate accumulation is a 581 common response to N limitation in microalgae (Pancha and Imamura, 2026). _Nannochloropsis_, 582 however, stores C as β-1,3-glucan rather than starch (Vogler et al., 2018), and typically favors lipid 583 over carbohydrate accumulation during N stress (Recht et al., 2012). The simultaneous increase of 584 both pools in _nrt2.2_ KO strains therefore suggests that the C surplus exceeded what the lipid pathway 585 alone could absorb, or that the chronic _nrt2.2_ block, with no demand for C and reducing power for 586 NO 3 [-] assimilation from the start, used both sinks. The two pools may also play distinct roles: 587 carbohydrate synthesis is a rapid, low-cost sink for fixed C, whereas TAG synthesis requires more 588 ATP and reducing power and represents a more reduced, long-term store (Valledor et al., 2014; Wang 589 et al., 2024). Time-resolved measurements of both pools would clarify whether carbohydrates act as 590 an early sink that is progressively complemented by lipids. Overall, the strong accumulation of both 591 compounds indicates that the disruption of NO 3 [-] uptake places cells in a prolonged C-rich, N-deficient 592 metabolic state that cannot be relieved just by NO 3 [-] supply (Fig. 2). 593 594 23 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 595 **Figure 6. Model of NRT2.2-mediated NO** **3** **[−]** **uptake in** _**Nannochloropsis**_ _**oceanica**_ **and** 596 **consequences of its loss.** A) Wild type (WT). NO 3 [−] enters through the plasma-membrane transporter 597 NRT2.2, is reduced to NO 2 [−] by nitrate reductase (NR) in the cytosol and to NH 4 [+] by nitrite reductase 598 (NiR) in the chloroplast, where it is assimilated into amino acids (GS/GOGAT) using carbon 599 skeletons from the Calvin-Benson-Bassham (CBB) cycle. Amino acids sustain protein synthesis. 600 NRT2.1 (grey) is not required for NO 3 [−] uptake under these conditions. B) _nrt2.2_ knock-out (KO). 601 Without NRT2.2, NO 3 [−] is not taken up and remains in the medium, and cells become N-limited 602 although NO 3 [−] is available: chlorophyll _a_ decreases (paler thylakoids) together with F v /F m, cell 603 division stops and fixed carbon is diverted to storage lipids and carbohydrates (Fig. 4), increasing the 604 C/N ratio; protein content per cell is unaffected. NH 4 [+], taken up independently of NRT2.2 (AMT, 605 ammonium transporter), restores growth. Black arrows, unchanged; thick red arrows, increased; 606 dashed blue arrows, decreased; grey, absent or no phenotype. The areas of the lipid droplets (yellow) 607 and carbohydrate symbols (hexagons) are proportional to the FTIR ratios in +NO 3 [-] (Fig. 4D,E). ATP 608 and NADPH supply to the CBB cycle was not measured (question mark). The outermost chloroplast 609 membrane (dashed) is continuous with the endoplasmic reticulum. 610 **NRT2.2 at the interface between the ocean and the cell** 611 NRT2.2 is the plasma membrane gateway for NO 3 [-] in _N. oceanica_ . From a physiological perspective, 612 these findings place NO 3 [-] uptake upstream of the coordination between C and N metabolism. By 613 controlling NO 3 [-] entry, it determines whether fixed C is channelled into growth or diverted into 614 storage, placing it at the core of the metabolic response to N availability (Fig. 6). Because _N. oceanica_ 615 relies on a single dominant transporter, its metabolism has little buffering against the loss of this 616 uptake system. 617 This tight coupling also has ecological implications. In the ocean, NO 3 [-] is the largest pool of 618 bioavailable N, yet it is depleted for long periods in surface waters and resupplied episodically 619 (Gruber, 2008). Under these conditions, the capacity to acquire NO 3 [-] can strongly limit primary 620 productivity, and the dependence of _N. oceanica_ on a single dominant transporter makes NRT2.2 a 621 strong candidate to determine how this alga uses transient NO 3 [-] inputs and how it then allocates C 622 between growth and storage, consequently affecting the whole community. Moreover, 623 macromolecular composition determines cell density and, in non-motile phytoplankton, sinking: 624 carbohydrates and proteins increase density, whereas lipids lower it (Liefer et al., 2019; Wu et al., 625 2025). In nutrient-depleted surface waters, protein and carbohydrate ballast has been proposed to help 626 cells sink toward deeper, nutrient-rich layers (Wu et al., 2025). The concurrent accumulation of 627 carbohydrates and lipids in the _nrt2.2_ KO lines may therefore alter buoyancy, thus impairing sinking 628 behaviour. By linking nutrient supply to the allocation of fixed C, NRT2.2 connects the chemistry of 629 seawater to the biochemical composition of the cell, with potential consequences for the C:N 630 stoichiometry of the organic matter produced in the ocean (Liefer et al., 2019). 631 Together, these results show that NRT2.2 is the gateway through which NO 3 [-] enters the cell and sets 632 the balance between growth and C storage, providing a mechanistic link between the N supply of the 24 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 633 ocean and the metabolism of phytoplankton in a fluctuating environment. Defining how this gateway 634 is regulated and how widely it is shared among lineages will be key to predicting how phytoplankton 635 cope with a variable N supply and, given the biotechnological relevance of _Nannochloropsis_, to 636 redirect C partitioning toward storage compounds such as lipids. 637 25 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 638 **COMPETING INTERESTS** 639 The authors declare no competing interests. 640 641 **DATA AVAILABILITY STATEMENT** 642 All data have been deposited in the Research Data UniPD repository under the following accession 643 [ID: https://researchdata.cab.unipd.it/id/eprint/1961.](https://researchdata.cab.unipd.it/id/eprint/1961) 644 645 **FUNDING** 646 AA and CG acknowledge funding from MUR PRIN22 Project “INPHOMARE” (2022KT9X3S). GP 647 acknowledges the Department of Biology of the University of Padova for the BIRD2023 Project 648 ‘ORGANISM’, the MUR for the PRIN2022 Project 2022CWZNZC_002 and the PRIN-PN22 Project 649 P2022WL8TS_003. 650 651 **ACKNOWLEDGEMENTS** 652 The authors wish to thank the Imaging Light Microscopy Facility and the Plant Genome Editing and 653 Phenotyping Facility of the Biology Department of the University of Padova (Italy). AI was used to 654 assist with figure assembly and the scripts used to plot the data without altering the image data, and 655 to revise the English language of the manuscript without changing its scientific content. 656 657 **AUTHOR CONTRIBUTIONS** 658 Anna Santin: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, 659 Visualization, Writing - original draft, Writing - review & editing 660 Laura Verdesca: Formal analysis, Investigation, Methodology, Writing - review & editing 661 Alessandra Bellan: Formal analysis, Investigation, Methodology, Writing - review & editing 662 Miles Minio: Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing 663 review & editing 664 Caterina Gerotto: Data curation, Funding acquisition, Investigation, Resources, Supervision, 665 Validation, Writing - review & editing 666 Tomas Morosinotto: Funding acquisition, Supervision, Writing - review & editing 667 Giorgio Perin: Funding acquisition, Supervision, Writing - review & editing 668 Alessandro Alboresi: Conceptualization, Data curation, Funding acquisition, Investigation, Project 669 administration, Resources, Supervision, Validation, Writing - review & editing 670 26 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 671 **REFERENCES** 672 **Bender SJ, Durkin CA, Berthiaume CT, Morales RL, Armbrust EV** (2014) Transcriptional 673 responses of three model diatoms to nitrate limitation of growth. Front Mar Sci. doi: 674 10.3389/fmars.2014.00003 675 **Berman A, Zylberberg I, Mayrose I, Shani E** (2026) Navigating genetic redundancy in plant 676 genomes: insights for research and breeding. Trends Plant Sci **31** : 961–973 677 **Berthold P, Schmitt R, Mages W** (2002) An Engineered _Streptomyces hygroscopicus aph 7_ ″ Gene 678 Mediates Dominant Resistance against Hygromycin B in _Chlamydomonas reinhardtii_ . Protist 679 **153** : 401–412 680 **Busseni G, Rocha Jimenez Vieira F, Amato A, Pelletier E, Pierella Karlusich JJ, Ferrante MI,** 681 **Wincker P, Rogato A, Bowler C, Sanges R, et al** (2019) Meta-Omics Reveals Genetic 682 Flexibility of Diatom Nitrogen Transporters in Response to Environmental Changes. Mol Biol 683 Evol **36** : 2522–2535 684 **Cassan O, Pimparé L-L, Dubos C, Gojon A, Bach L, Lèbre S, Martin A** (2023) A gene regulatory 685 network in Arabidopsis roots reveals features and regulators of the plant response to elevated 686 CO2. New Phytol **239** : 992–1004 687 **Dong H-P, Williams E, Wang D, Xie Z-X, Hsia R, Jenck A, Halden R, Li J, Chen F, Place AR** 688 (2013) Responses of _Nannochloropsis oceanica_ IMET1 to Long-Term Nitrogen Starvation 689 and Recovery. Plant Physiol **162** : 1110–1126 690 **Edgar RC** (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. 691 Nucleic Acids Res **32** : 1792–1797 692 **Fan X, Naz M, Fan X, Xuan W, Miller AJ, Xu G** (2017) Plant nitrate transporters: from gene 693 function to application. J Exp Bot **68** : 2463–2475 694 **Fattore N, Bellan A, Pedroletti L, Vitulo N, Morosinotto T** (2021) Acclimation of photosynthesis 695 and lipids biosynthesis to prolonged nitrogen and phosphorus limitation in _Nannochloropsis_ 696 _gaditana_ . Algal Res **58** : 102368 697 **Fernandez E, Galvan A** (2008) Nitrate Assimilation in _Chlamydomonas_ . Eukaryot Cell **7** : 555–559 698 **Filleur S, Dorbe MF, Cerezo M, Orsel M, Granier F, Gojon A, Daniel-Vedele F** (2001) An 699 arabidopsis T-DNA mutant affected in Nrt2 genes is impaired in nitrate uptake. FEBS Lett 700 **Gibson DG, Young L, Chuang R-Y, Venter JC, Hutchison CA, Smith HO** (2009) Enzymatic 701 assembly of DNA molecules up to several hundred kilobases. Nat Methods **6** : 343–345 702 **Giordano M, Kansiz M, Heraud P, Beardall J, Wood B, McNaughton D** (2001) Fourier 703 Transform Infrared Spectroscopy as a Novel Tool to Investigate Changes in Intracellular 704 Macromolecular Pools in the Marine Microalga _Chaetoceros Muellerii_ (bacillariophyceae). J 705 Phycol **37** : 271–279 706 **Glibert PM** (2016) Margalef revisited: A new phytoplankton mandala incorporating twelve 707 dimensions, including nutritional physiology. Harmful Algae **55** : 25–30 27 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 708 **Goddard TD, Huang CC, Meng EC, Pettersen EF, Couch GS, Morris JH, Ferrin TE** (2018) 709 UCSF ChimeraX: Meeting modern challenges in visualization and analysis. Protein Sci **27** : 710 14–25 711 **Goncalves EC, Koh J, Zhu N, Yoo M-J, Chen S, Matsuo T, Johnson JV, Rathinasabapathi B** 712 (2016) Nitrogen starvation-induced accumulation of triacylglycerol in the green algae: 713 evidence for a role for ROC40, a transcription factor involved in circadian rhythm. Plant J **85** : 714 743–757 715 **Gruber N** (2008) Chapter 1 - The Marine Nitrogen Cycle: Overview and Challenges. Nitrogen Mar. 716 Environ. Second Ed. Academic Press, San Diego, pp 1–50 717 **Guillard RRL** (1975) Culture of Phytoplankton for Feeding Marine Invertebrates. _In_ WL Smith, MH 718 Chanley, eds, Cult. Mar. Invertebr. Anim. Springer US, Boston, MA, pp 29–60 719 **Guindon S, Dufayard J-F, Lefort V, Anisimova M, Hordijk W, Gascuel O** (2010) New 720 Algorithms and Methods to Estimate Maximum-Likelihood Phylogenies: Assessing the 721 Performance of PhyML 3.0. Syst Biol **59** : 307–321 722 **Hoang DT, Vinh LS, Flouri T, Stamatakis A, von Haeseler A, Minh BQ** (2018) MPBoot: fast 723 phylogenetic maximum parsimony tree inference and bootstrap approximation. BMC Evol 724 Biol **18** : 11 725 **Kabala C, Karczewska A, Gałka B, Cuske M, Sowiński J** (2017) Seasonal dynamics of nitrate and 726 ammonium ion concentrations in soil solutions collected using MacroRhizon suction cups. 727 Environ Monit Assess **189** : 304 728 **Kalyaanamoorthy S, Minh BQ, Wong TKF, von Haeseler A, Jermiin LS** (2017) ModelFinder: 729 fast model selection for accurate phylogenetic estimates. Nat Methods **14** : 587–589 730 **Katoh K, Standley DM** (2013) MAFFT Multiple Sequence Alignment Software Version 7: 731 Improvements in Performance and Usability. Mol Biol Evol **30** : 772–780 732 **Kotur Z, Mackenzie N, Ramesh S, Tyerman SD, Kaiser BN, Glass ADM** (2012) Nitrate transport 733 capacity of the _Arabidopsis thaliana_ NRT2 family members and their interactions with 734 AtNAR2.1. New Phytol **194** : 724–731 735 **Li J, Han D, Wang D, Ning K, Jia J, Wei L, Jing X, Huang S, Chen J, Li Y, et al** (2014) 736 Choreography of Transcriptomes and Lipidomes of _Nannochloropsis_ Reveals the 737 Mechanisms of Oil Synthesis in Microalgae. Plant Cell **26** : 1645–1665 738 **Li W, Wang Y, Okamoto M, Crawford NM, Siddiqi MY, Glass ADM** (2007) Dissection of the 739 AtNRT2.1:AtNRT2.2 inducible high-affinity nitrate transporter gene cluster. Plant Physiol 740 **143** : 425–433 741 **Liefer JD, Garg A, Fyfe MH, Irwin AJ, Benner I, Brown CM, Follows MJ, Omta AW, Finkel** 742 **ZV** (2019) The Macromolecular Basis of Phytoplankton C:N:P Under Nitrogen Starvation. 743 Front Microbiol. doi: 10.3389/fmicb.2019.00763 744 **Liistro E, Calcinoni A, Agostini A, Pressi S, Marotta E, La Rocca N, Carbonera D** (2026) 745 Enhanced carotenoid photoprotection in Far-Red light acclimated _Chroococcidiopsis_ 746 _thermalis_ . Plant Physiol Biochem **233** : 111265 28 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 747 **Liu K-H, Huang C-Y, Tsay Y-F** (1999) CHL1 Is a Dual-Affinity Nitrate Transporter of _Arabidopsis_ 748 Involved in Multiple Phases of Nitrate Uptake. Plant Cell **11** : 865–874 749 **Maghiaoui A, Gojon A, Bach L** (2020) NRT1.1-centered nitrate signaling in plants. J Exp Bot **71** : 750 6226–6237 751 **Maxwell K, Johnson GN** (2000) Chlorophyll fluorescence—a practical guide. J Exp Bot **51** : 659– 752 668 753 **McCarthy JK, Smith SR, McCrow JP, Tan M, Zheng H, Beeri K, Roth R, Lichtle C,** 754 **Goodenough U, Bowler CP, et al** (2017) Nitrate Reductase Knockout Uncouples Nitrate 755 Transport from Nitrate Assimilation and Drives Repartitioning of Carbon Flux in a Model 756 Pennate Diatom. Plant Cell **29** : 2047–2070 757 **Minh BQ, Schmidt HA, Chernomor O, Schrempf D, Woodhams MD, von Haeseler A, Lanfear** 758 **R** (2020) IQ-TREE 2: New Models and Efficient Methods for Phylogenetic Inference in the 759 Genomic Era. Mol Biol Evol **37** : 1530–1534 760 **Moran R** (1982) Formulae for Determination of Chlorophyllous Pigments Extracted with _N,N_ 761 Dimethylformamide. Plant Physiol **69** : 1376–1381 762 **Pancha I, Imamura S** (2026) Microalgal starch: regulation of its quantity and roles. Front Plant Sci. 763 doi: 10.3389/fpls.2026.1736651 764 **Perin G, Bellan A, Segalla A, Meneghesso A, Alboresi A, Morosinotto T** (2015) Generation of 765 random mutants to improve light-use efficiency of _Nannochloropsis gaditana_ cultures for 766 biofuel production. Biotechnol Biofuels **8** : 161 767 **Powell HR, Islam SA, David A, Sternberg MJE** (2025) Phyre2.2: A Community Resource for 768 Template-based Protein Structure Prediction. J Mol Biol 168960 769 **Radakovits R, Jinkerson RE, Fuerstenberg SI, Tae H, Settlage RE, Boore JL, Posewitz MC** 770 (2012) Draft genome sequence and genetic transformation of the oleaginous alga 771 _Nannochloropis gaditana_ . Nat Commun **3** : 686 772 **Recht L, Zarka A, Boussiba S** (2012) Patterns of carbohydrate and fatty acid changes under nitrogen 773 starvation in the microalgae _Haematococcus pluvialis_ and _Nannochloropsis_ sp. Appl 774 Microbiol Biotechnol **94** : 1495–1503 775 **Rexach J, Llamas A, Fernández E, Galván A** (2002) The activity of the high-affinity nitrate 776 transport system I (NRT2;1, NAR2) is responsible for the efficient signalling of nitrate 777 assimilation genes in _Chlamydomonas reinhardtii_ . Planta **215** : 606–611 778 **Rogato A, Amato A, Iudicone D, Chiurazzi M, Ferrante MI, d’Alcalà MR** (2015) The diatom 779 molecular toolkit to handle nitrogen uptake. Mar Genomics **24** : 95–108 780 **Santin A, Caputi L, Longo A, Chiurazzi M, Ribera d’Alcalà M, Russo MT, Ferrante MI, Rogato** 781 **A** (2021) Integrative omics identification, evolutionary and structural analysis of low affinity 782 nitrate transporters in diatoms, diNPFs. Open Biol **11** : rsob.200395, 200395 783 **Santin A, Russo MT, de los Ríos LM, Chiurazzi M, d’Alcalà MR, Lacombe B, Ferrante MI,** 784 **Rogato A** (2024) The tonoplast localized protein PtNPF1 participates in the regulation of 785 nitrogen response in diatoms. New Phytol **241** : 1592–1604 29 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 786 **Santin A, Storti M, Mezzavilla M, Fortunato A, Fiorin F, Gaboriau DCA, Arcudi F, Flores E,** 787 **Morosinotto T, Perin G** (2026) Photosynthesis regulation impacts carbon and nitrogen 788 assimilation in the diazotrophic cyanobacterium _Anabaena_ sp. PCC 7120. New Phytol **251** : 789 94–110 790 **Sanz-Luque E, Chamizo-Ampudia A, Llamas A, Galvan A, Fernandez E** (2015) Understanding 791 nitrate assimilation and its regulation in microalgae. Front. Plant Sci. 6: 792 **Shimodaira H** (2002) An Approximately Unbiased Test of Phylogenetic Tree Selection. Syst Biol 793 **51** : 492–508 794 **Simionato D, Block MA, La Rocca N, Jouhet J, Maréchal E, Finazzi G, Morosinotto T** (2013) 795 The Response of _Nannochloropsis gaditana_ to Nitrogen Starvation Includes _De Novo_ 796 Biosynthesis of Triacylglycerols, a Decrease of Chloroplast Galactolipids, and 797 Reorganization of the Photosynthetic Apparatus. Eukaryot Cell **12** : 665–676 798 **Smith SR, Dupont CL, McCarthy JK, Broddrick JT, Oborník M, Horák A, Füssy Z, Cihlář J,** 799 **Kleessen S, Zheng H, et al** (2019) Evolution and regulation of nitrogen flux through 800 compartmentalized metabolic networks in a marine diatom. Nat Commun **10** : 4552 801 **Steenwyk JL, Iii TJB, Li Y, Shen X-X, Rokas A** (2020) ClipKIT: A multiple sequence alignment 802 trimming software for accurate phylogenomic inference. PLOS Biol **18** : e3001007 803 **Tamura K, Stecher G, Kumar S** (2021) MEGA11: Molecular Evolutionary Genetics Analysis 804 Version 11. Mol Biol Evol **38** : 3022–3027 805 **Unkles SE, Rouch DA, Wang Y, Siddiqi MY, Glass ADM, Kinghorn JR** (2004) Two perfectly 806 conserved arginine residues are required for substrate binding in a high-affinity nitrate 807 transporter. Proc Natl Acad Sci **101** : 17549–17554 808 **Valledor L, Furuhashi T, Recuenco-Muñoz L, Wienkoop S, Weckwerth W** (2014) System-level 809 network analysis of nitrogen starvation and recovery in _Chlamydomonas reinhardtii_ reveals 810 potential new targets for increased lipid accumulation. Biotechnol Biofuels **7** : 171 811 **Vieler A, Wu G, Tsai C-H, Bullard B, Cornish AJ, Harvey C, Reca I-B, Thornburg C,** 812 **Achawanantakun R, Buehl CJ, et al** (2012) Genome, Functional Gene Annotation, and 813 Nuclear Transformation of the Heterokont Oleaginous Alga _Nannochloropsis oceanica_ 814 CCMP1779. PLOS Genet **8** : e1003064 815 **Vogler BW, Brannum J, Chung JW, Seger M, Posewitz MC** (2018) Characterization of the 816 _Nannochloropsis gaditana_ storage carbohydrate: A 1,3-beta glucan with limited 1,6817 branching. Algal Res **36** : 152–158 818 **Wang B, Xu Y, Lin L, Xiao J, Xie Y, Dai C, Jia J** (2024) Remodeling of central carbon metabolism 819 of _Nannochloropsis oceanica_ in response to gradual nitrogen depletion. Algal Res **82** : 103658 820 **Wang D, Ning K, Li J, Hu J, Han D, Wang H, Zeng X, Jing X, Zhou Q, Su X, et al** (2014) 821 Nannochloropsis Genomes Reveal Evolution of Microalgal Oleaginous Traits. PLOS Genet 822 **10** : e1004094 823 **Wang Y-Y, Cheng Y-H, Chen K-E, Tsay Y-F** (2018) Nitrate Transport, Signaling, and Use 824 Efficiency. Annu Rev Plant Biol **69** : 85–122 30 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 825 **Wei L, El Hajjami M, Shen C, You W, Lu Y, Li J, Jing X, Hu Q, Zhou W, Poetsch A, et al** 826 (2019) Transcriptomic and proteomic responses to very low CO 2 suggest multiple carbon 827 concentrating mechanisms in _Nannochloropsis oceanica_ . Biotechnol Biofuels **12** : 168 828 **von Wittgenstein NJ, Le CH, Hawkins BJ, Ehlting J** (2014) Evolutionary classification of 829 ammonium, nitrate, and peptide transporters in land plants. BMC Evol Biol **14** : 11 830 **Wu Y, Kouadio VK, Usherwood TR, Li J, Bisher M, Aurora R, Lam AZ, Lam AR, Lytton-Jean** 831 **AKR, Manalis SR, et al** (2025) Diverse biophysical and molecular mechanisms drive 832 phytoplankton sinking in response to starvation. PLOS Biol **23** : e3003508 833 **Xu N, Cheng L, Kong Y, Chen G, Zhao L, Liu F** (2024) Functional analyses of the NRT2 family 834 of nitrate transporters in Arabidopsis. Front Plant Sci **15** : 1351998 835 **Zhou J-J, Fernández E, Galván A, Miller AJ** (2000) A high affinity nitrate transport system from 836 _Chlamydomonas_ requires two gene products. FEBS Lett **466** : 225–227 837 838 31 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 839 **FIGURE LEGENDS** 840 841 **Figure 1. Phylogeny and predicted structure of** _**Nannochloropsis**_ _**oceanica**_ **NRT2s.** A) Maximum842 likelihood phylogeny of 94 NRT2 proteins from 50 photosynthetic and non-photosynthetic 843 eukaryotes. Filled and open circles indicate nodes with ultrafast bootstrap (UFBoot) support ≥ 95% 844 and 70-94%, respectively; SH-aLRT/UFBoot values are shown for the eustigmatophyte clade and for 845 the NoNRT2.2- _Microchloropsis_ clade. Colors indicate taxonomic groups; _N. oceanica_ NRT2.1 and 846 NRT2.2 are in orange. Sequences and accession numbers are listed in Supplementary Data Set S1. 847 B) Predicted structural model of _N. oceanica_ NRT2.2 with 12 transmembrane helices and, in the 848 zoomed view, the putative NO 3 [-] binding site: R86 and R302 (green), E265 (magenta), and N167 and 849 N390 (blue). 850 ALT TEXT: Phylogenetic tree of NRT2 transporters and predicted 3D structure of _Nannochloropsis_ 851 NRT2.2. 852 853 **Figure 2. Impaired growth and nitrate uptake in** _**Nannochloropsis oceanica**_ _**nrt2.2**_ **KO mutants.** 854 A) RT-PCR of NRT2.1 and NRT2.2 transcripts in the wild type (WT) and in two _nrt2.1_ and two 855 _nrt2.2_ knock-out (KO) lines (#1, #2); ACT (actin), reference transcript. Primers are listed in Table 856 S1. B) Experimental setup. Cells were grown under atmospheric CO 2 (400 ppm) and continuous white 857 light (35 µmol photons m [-2] s [-1] ) in 882 µM nitrate (NO 3 [-] ) and 882 µM ammonium (NH 4 [+] ) until 858 exponential phase, starved of N for 3 days and then resupplied with NO 3 [-] or NH 4 [+] . C) Residual NO 3 [-] 859 in the medium and D) cell concentration during 4 days in 882 µM NO 3 [-] . E) Cell concentration in 882 860 µM NH 4 [+] (control). F) Cultures after 4 days of NO 3 [-] or NH 4 [+] repletion. Data are mean ± SD of at 861 least three biological replicates. Asterisks: unpaired two-tailed Student's t-test vs WT at day 4; * p < 862 0.05, ** p < 0.01, *** p < 0.001. 863 ALT TEXT: Gel, scheme and growth curves showing that _nrt2_ .2 knock-out lines do not take up nitrate 864 and do not grow on it. 865 866 **Figure 3. Altered intracellular N content and photosynthetic performance in** _**Nannochloropsis**_ 867 _**oceanica**_ _**nrt2.2**_ **knock-out mutants.** Wild-type (WT), two _nrt2.1_ and two _nrt2.2_ knock-out (KO) 868 lines were grown in the Multi-Cultivator system under atmospheric CO 2 (400 ppm) and continuous 869 white light (80 µmol photons m [-2] s [-1] ) and analyzed after 4 days of repletion with 882 µM nitrate (NO 3 [-] 870 ); growth curves are shown in Figure S3 and the control conditions, nitrogen (N) deprivation and 871 NH 4 [+] repletion, in Figures S4-S7. A) Biomass production. B) Carbon (C) and C) N content, as 872 percentage of dry weight. The detection limit for N is 4% of dry weight (grey area): N was not 873 detected (n.d.) in five of six _nrt2.2_ KO samples, and the only measurable sample ( _nrt2.2_ KO #2) is 874 shown. D) Total protein content. E) Maximum photosynthetic efficiency, expressed as F v /F m, and F) 875 chlorophyll content. G) Relative pigment composition (% chlorophyll and total carotenoids). Data 876 are mean ± SD of at least three biological replicates. Asterisks: unpaired two-tailed Student’s t-test 877 vs WT; * p < 0.05, ** p < 0.01, *** p < 0.001. 878 ALT TEXT: Graphs showing higher C/N and lower photosynthetic efficiency in _nrt2.2_ knock-out 879 lines. 880 881 **Figure 4. Increased lipid and carbohydrate content in** _**Nannochloropsis oceanica**_ _**nrt2.2**_ **knock-** 882 **out mutants.** Fourier Transform Infrared (FTIR) spectra of _N. oceanica_ wild-type (WT), two _nrt2.1_ 883 and two _nrt2.2_ knock-out (KO) lines grown in the Multi-Cultivator system, under atmospheric CO 2 884 (400 ppm) and continuous white light (80 µmol photons m [-2] s [-1] ) A) after 3 days of nitrogen (N) 885 depletion, B) after 4 days of repletion with 882 µM ammonium (NH 4 [+] ), and C) after 4 days of 886 repletion with 882 µM nitrate (NO 3 [-] ). Spectra are normalized to the protein amide I peak at ~1650 887 cm-1; total protein did not differ between strains (Fig. S6). Shaded areas mark the spectral regions of 888 the fitted bands used for quantification: lipid ester C=O (1770-1710 cm [-1] ) and carbohydrate C-O-C 32 [bioRxiv preprint doi: https://doi.org/10.64898/2026.11.02.763162; this version posted November 3, 2026. The copyright holder for this](https://doi.org/10.64898/2026.11.02.763162) preprint (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-ND 4.0 International license.](http://creativecommons.org/licenses/by-nc-nd/4.0/) 889 (1200-950 cm [-1] ); colored asterisks refer to the band ratios in D and E. D) Lipid (ester C=O) and E) 890 carbohydrate (C-O-C) band areas relative to the protein amide I band (~1656 cm [-1] ), obtained by fitting 891 the spectra with Lorentzian bands (lipids: ester C=O band at ~1740 cm [-1] ; carbohydrates: sum of the 892 bands between ~1030 and ~1160 cm [-1] ). F) Neutral lipids, measured as Nile Red fluorescence, in the 893 same conditions. Data are mean ± SD of at least three biological replicates (FTIR technical duplicates 894 were averaged). Asterisks: unpaired two-tailed Student’s t-test vs WT; * p < 0.05, ** p < 0.01, *** p 895 < 0.001. 896 ALT TEXT: FTIR spectra and plots showing more lipids and carbohydrates in _nrt2.2_ knock-out lines. 897 898 **Figure 5. Subcellular localization of** _**Nannochloropsis oceanica**_ **NRT2.2 to the plasma** 899 **membrane.** A) Expression construct: NRT2.2 coding sequence fused at its 3’ end to the yellow 900 fluorescent protein mVenus, co-transformed with a hygromycin resistance cassette. B) Genomic PCR 901 confirming the integration of the construct (1834 bp amplicon, primers for/rev in A) in lines #9 and 902 #11; WT, wild-type parental strain; C+, transformation plasmid. C) Confocal images of WT and 903 NRT2.2-mVenus cells: bright field (BF), chlorophyll _a_ autofluorescence (magenta), mVenus (green) 904 and merge. mVenus is corrected for chlorophyll bleed-through and shown with the same intensity 905 range in all lines. Scale bars = 5 μm. D) Mean intensity profiles (± SD) of mVenus (green) and 906 chlorophyll _a_ (magenta, normalized to the maximum of each profile) across the cell (18 diameters 907 per cell; position normalized to the ring radius, ring at ±1; WT scaled by the mean ring radius, 1.63 908 µm), in WT cells (n = 10) and in NRT2.2-mVenus cells with a peripheral ring (#9: n = 4; #11: n = 5); 909 mVenus is on a separate scale for each line. E) Pearson’s correlation between mVenus and 910 chlorophyll _a_ signals after bleed-through correction in the NRT2.2-mVenus lines #9 and #11; each 911 dot is one cell, bars are means. 912 ALT TEXT: Confocal images showing NRT2.2-mVenus at the cell periphery. 913 914 **Figure 6. Model of NRT2.2-mediated NO** **3** **[−]** **uptake in** _**Nannochloropsis**_ _**oceanica**_ **and** 915 **consequences of its loss.** A) Wild type (WT). NO 3 [−] enters through the plasma-membrane transporter 916 NRT2.2, is reduced to NO 2 [−] by nitrate reductase (NR) in the cytosol and to NH 4 [+] by nitrite reductase 917 (NiR) in the chloroplast, where it is assimilated into amino acids (GS/GOGAT) using carbon 918 skeletons from the Calvin-Benson-Bassham (CBB) cycle. Amino acids sustain protein synthesis. 919 NRT2.1 (grey) is not required for NO 3 [−] uptake under these conditions. B) _nrt2.2_ knock-out (KO). 920 Without NRT2.2, NO 3 [−] is not taken up and remains in the medium, and cells become N-limited 921 although NO 3 [−] is available: chlorophyll _a_ decreases (paler thylakoids) together with F v /F m, cell 922 division stops and fixed carbon is diverted to storage lipids and carbohydrates (Fig. 4), increasing the 923 C/N ratio; protein content per cell is unaffected. NH 4 [+], taken up independently of NRT2.2 (AMT, 924 ammonium transporter), restores growth. Black arrows, unchanged; thick red arrows, increased; 925 dashed blue arrows, decreased; grey, absent or no phenotype. The areas of the lipid droplets (yellow) 926 and carbohydrate symbols (hexagons) are proportional to the FTIR ratios in +NO 3 [-] (Fig. 4D,E). ATP 927 and NADPH supply to the CBB cycle was not measured (question mark). The outermost chloroplast 928 membrane (dashed) is continuous with the endoplasmic reticulum. 929 ALT TEXT: Model comparing carbon and nitrogen metabolism in wild-type and _nrt2.2_ knock-out 930 cells. 931 33