# Source paper archive - Source title: Putrescine-responsive post-transcriptional feedback regulation of Arabidopsis ADC2 mediated by 5'-UTR elements - Author(s): Hiragori, Y., Yasumuro, M., Karino, T., Kaido, A., Sakihama, Y., Goto, Y., Yamashita, Y., Naito, S., Onouchi, H. - Source landing URL: https://www.biorxiv.org/content/10.64898/2026.10.31.763274v1 - PDF URL: https://www.biorxiv.org/content/10.64898/2026.10.31.763274v1.full.pdf - PDF SHA-256: f06596f520ba582c122d2b70e34750cced396935ca80afdba6639f004c7cef93 - Converter: pymupdf4llm 0.0.27 - Limits: Automated PDF text extraction; reading order, equations, tables and formatting may be incomplete or incorrect. Images are omitted, and no OCR was performed. Markup and links are neutralized for safe downloading. Consult the original PDF for the authoritative version. --- [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in [perpetuity.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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 1 2 # 3 TITLE 4 Putrescine-responsive post-transcriptional feedback regulation of Arabidopsis _ADC2_ mediated 5 ×� (�-UTR elements 6 ## 7 RUNNING TITLE 8 Putrescine-responsive regulation of _AtADC2_ 9 10 Yuta Hiragori [1,2] *, Miharu Yasumuro [3], Taihei Karino [1], Atsushi Kaido [1], Yasuko Sakihama [1,2], 11 Yuya Goto [1], Yui Yamashita [1,2], Satoshi Naito [1,2], and Hitoshi Onouchi [1,2] 12 13 1 Graduate School of Agriculture, Hokkaido University, Sapporo 060-8589, Japan 14 2 Research Faculty of Agriculture, Hokkaido University, Sapporo 060-8589, Japan 15 3 School of Agriculture, Hokkaido University, Sapporo 060-8589, Japan 16 *Corresponding authors: 17 Yuta Hiragori (hiragori@agr.hokudai.ac.jp) 18 Hitoshi Onouchi (onouchi@agr.hokudai.ac.jp) 19 1 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in [perpetuity.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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) # 20 ABSTRACT 21 Putrescine (Put) is a bioactive polyamine that contributes to plant growth, development, and 22 stress responses, whereas excessive Put accumulation impairs growth. However, how plants 23 maintain Put levels within an appropriate range remains poorly understood. In _Arabidopsis_ 24 _thaliana_, expression of the arginine decarboxylase 2 ( _ADC2_ ) gene is a major determinant of Put 25 7×W=O7=⁵6s F6≥6, ‡6 ⁷T−‡ ₙT7ₙ ₙT6 (� W=ₙ≥7=⁷I7ₙ6O ≥6’₁−= e(�-UTR) of _ADC2_ mediates Put 26 responsive feedback regulation and contributes to Put homeostasis in Arabidopsis. In transient 27 expression assays using an emerald luciferase (Eluc) gene placed downstream of the _ADC2_ (� 28 UTR ( _ADC2_ (�-UTR: _Eluc_ ), Eluc activity was reduced by exogenously applied Put but increased 29 when endogenous Put biosynthesis was impaired by _adc2_ mutations or treatment with an 30 arginine decarboxylase inhibitor. In transgenic plants carrying _ADC2_ (�-UTR: _Eluc_, Put 31 treatment modestly decreased _Eluc_ mRNA abundance but substantially reduced Eluc activity 32 and translation efficiency, suggesting that this Put-responsive repression occurs mainly at the 33 translational level. Among the _cis_ -regulatory elements in the _ADC2_ (�-UTR, mutations 34 affecting either the conserved AUU/CUG-initiated upstream ORF or the predicted RNA 35 secondary structure in the ADC-box, a conserved GC-rich motif, abolished Put-responsive 36 repression, indicating that these elements are required for this regulation. Two genome-edited 37 lines carrying deletions spanning these elements exhibited delayed germination and impaired 38 growth, and the more severely affected line showed increased Put accumulation. Together, 39 these findings reveal a feedback mechanism that regulates _ADC2_ 6�9≥6⁷⁷₁−= ₙT≥−W’T ₁ₙ⁷ (�-UTR 40 and support a role for this mechanism in Put homeostasis in Arabidopsis. 41 2 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) # 42 INTRODUCTION 43 Polyamines are essential and ubiquitous molecules in living cells. Owing to their cationic 44 nature, polyamines interact with DNA, RNA, ATP, and other negatively charged cellular 45 molecules, thereby influencing multiple stages of gene expression, including translation 46 (Kusano et al. 2008; Igarashi and Kashiwagi 2010). The major polyamines in many species are 47 the diamine putrescine (Put), the triamine spermidine (Spd), and the tetraamine spermine (Spm) 48 (Fig. 1). Plants also produce thermospermine (Tspm), a structural isomer of Spm (Blázquez 49 2024). At appropriate levels, polyamines are required for normal plant growth and 50 development, and contribute to tolerance to biotic and abiotic stresses (Chen et al. 2019; 51 Blázquez 2024). By contrast, excessive accumulation of polyamines has adverse effects. For 52 example, transgenic tobacco and Arabidopsis lines overexpressing Put biosynthetic genes 53 exhibit Put overaccumulation and severe growth defects, such as leaf chlorosis and necrosis in 54 tobacco and dwarfism and late flowering in Arabidopsis (Masgrau et al. 1997; Alcázar et al. 55 2005). The dual nature of Put suggests the presence of mechanisms that tightly maintain its 56 homeostasis, yet these mechanisms remain poorly understood in plants. 57 Put occupies a central position in plant polyamine biosynthesis as the precursor to other 58 polyamines, Spd, Spm, and Tspm (Fig. 1). In many plant species, Put is synthesized either from 59 L-arginine (Arg) through a pathway initiated by arginine decarboxylase (ADC) or directly from 60 L-ornithine (Orn) by ornithine decarboxylase (ODC). In the ADC pathway, ADC first converts 61 Arg to agmatine. Agmatine is then converted to Put either via _N_ -carbamoylputrescine (Janowitz 62 et al. 2003; Piotrowski et al. 2003) or directly by arginases with agmatinase activity as reported 63 in Arabidopsis and soybean (Patel et al. 2017). In the ODC pathway, ODC directly converts 64 Orn to Put (Michael et al. 1996). In _Arabidopsis thaliana_, however, the genome lacks a 65 functional _ODC_ gene, and Put biosynthesis therefore depends solely on the ADC pathway 66 (Hanfrey et al. 2001; Fig. 1). The Arabidopsis genome contains two genes encoding ADC, 3 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 67 _ADC1_ (At2g16500, also called _SPE1_ ) and _ADC2_ (At4g34710, also called _SPE2_ ). Histochemical 68 analyses using �-glucuronidase staining showed that _ADC1_ promoter activity was generally low 69 during vegetative development, whereas the _ADC2_ promoter was active during seed 70 germination and root and leaf development (Hummel et al. 2004). Moreover, loss of _ADC2_ 71 significantly reduces Put levels, whereas loss of _ADC1_ does not (Urano et al. 2004; Rossi et al. 72 2015), supporting a predominant role for _ADC2_ in determining Put levels in Arabidopsis. 73 Feedback regulation of polyamine biosynthetic genes contributes to polyamine 74 homeostasis in eukaryotes. A well-characterized example is _S_ -adenosylmethionine 75 decarboxylase (SAMDC), which produces decarboxylated _S_ -adenosylmethionine, an 76 aminopropyl donor required for Spd, Spm, and Tspm synthesis (Fig. 1). _SAMDC_ mRNAs 77 contain an upstream open reading frame (uORF) with an evolutionarily conserved amino acid 78 sequence (Franceschetti et al. 2001). uORFs can repress translation of the downstream main 79 ORF by causing ribosome dissociation after uORF translation or by inducing ribosome stalling 80 during uORF translation (von Arnim et al. 2014; Hinnebusch et al. 2016; Dever et al. 2020). In 81 some cases, they can also trigger mRNA decay through quality-control pathways such as 82 nonsense-mediated decay (NMD) (Gaba et al. 2005; Nyikó et al. 2009; Uchiyama-Kadokura et 83 al. 2014). In mammals and plants, elevated Spd/Spm levels induce ribosome stalling during 84 translation of the _SAMDC_ uORFs, thereby repressing _SAMDC_ translation. This ribosome 85 stalling depends on the uORF-encoded peptide, although the conserved peptide sequences differ 86 between mammals and plants (Ruan et al. 1996; Law et al. 2001; Raney et al. 2002; Hanfrey et 87 al. 2002; Hanfrey et al. 2005; Uchiyama-Kadokura et al. 2014). In addition to this translational 88 feedback regulation, _ACAULIS5_ ( _ACL5_ ) mRNA, which encodes Tspm synthase, is subject to 89 transcriptional feedback regulation by Tspm (Kakehi et al. 2008; Yariuchi et al. 2021). 90 However, feedback regulation of genes encoding Put biosynthetic enzymes has not been 91 established in plants. 4 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 92 Previous studies have identified several _cis_ -regulatory elements in the (� W=ₙ≥7=⁷I7ₙ6O 93 regions ((�-UTRs) of plant _ADC_ genes that repress expression of the downstream main ORF. 94 Earlier work showed that a short uORF in the carnation _ADC_ gene represses translation of the 95 main ORF (Chang et al. 2000). More recently, a conserved GC-≥₁⁵T ;−ₙ₁3 ₙ6≥;6O ₙT6 �–αM 96 ×−�� ‡7⁷ ₁O6=ₙ₁3₁6O ₁= ₙT6 (�-UTRs of plant _ADC_ genes and shown to repress translation of the 97 main ORF in tomato. The ADC-box was predicted to form a stable RNA secondary structure 98 that may contribute to repression by impeding ribosome scanning (Wu et al. 2019). 99 Additionally, we recently identified a conserved non-AUG-initiated uORF in plant _ADC_ genes 100 and demonstrated its repressive effect in Arabidopsis _ADC2_ . In the _ADC2_ uORF, two 101 noncanonical start codons, AUU and CUG, contribute to repression of the main ORF (Hiragori 102 et al. 2023)s P−ₙ7×I�, ₙT6 (�-UTR of Arabidopsis _ADC2_ contains other putative _cis_ -regulatory 103 elements, including a short AUG-initiated uORF and the ADC-box. However, the physiological 104 significance of the cis-regulatory elements in the _ADC2_ (�-UTR remains unclear. 105 Here, we investigated whether the Arabidopsis _ADC2_ (�-UTR mediates Put-responsive 106 regulation and contributes to Put homeostasis. Reporter assays showed that the _ADC2_ (�-UTR 107 represses main ORF expression in response to exogenous Put, whereas inhibition of 108 endogenous Put biosynthesis has the opposite effect. Analyses of transgenic reporter lines 109 further indicated that translational repression makes a major contribution to this response. 110 Mutational analyses revealed that Put-responsive repression requires two elements within the 111 _ADC2_ (�-UTR: the AUU/CUG-initiated uORF and the predicted RNA secondary structure of 112 the ADC-box. Genome-edited _ADC2_ (�-UTR mutants exhibited delayed germination and 113 impaired growth, and the more severely affected mutant overaccumulated Put. Together, these 114 findings reveal a Put-responsive feedback mechanism mediated by the _ADC2_ (�-UTR and 115 support a role for the regulatory functions of this region in maintaining Put homeostasis and 116 normal growth in Arabidopsis. 5 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) # 117 RESULTS ## 118 The Arabidopsis ADC2 ��-UTR regulates downstream main ORF 119 expression in response to intracellular Put levels 120 To examine whether the Arabidopsis _ADC2_ (�-UTR mediates repression of downstream main 121 ORF expression in response to Put, we performed transient expression assays in Arabidopsis 122 mesophyll protoplasts. The _ADC2_ (�-UTR and the first 5 nt of its main ORF were fused to the 123 coding sequence of emerald luciferase (Eluc) carrying a PEST degradation sequence ( _Eluc_ 124 PEST), under the control of the cauliflower mosaic virus 35S RNA promoter (35Spro). This 125 construct was designated as _ADC2_ 5�-UTR: _Eluc_ . A control reporter carried only a short 126 unrelated sequence in place of the _ADC2_ (�-UTR (Fig. 2A). A plasmid expressing _Renilla_ 127 luciferase (Rluc) from the same promoter was co-introduced, and Eluc activity was normalized 128 to Rluc activity. After 24 h of incubation with 0�4 mM Put, normalized Eluc activity from 129 _ADC2_ (�-UTR: _Eluc_ showed a strong negative correlation with exogenous Put concentration 130 (Fig. 2B). The control construct produced substantially higher Eluc activity than _ADC2_ (� 131 UTR: _Eluc_ and showed no significant response to 2 mM Put, whereas activity from _ADC2_ (� 132 UTR: _Eluc_ was reduced by this treatment (Fig. 2C). The absence of a significant response in the 133 control construct indicates that the observed Put-responsive repression was not a general effect 134 of Put on the 35Spro or the _Eluc_ -PEST reporter system. Thus, the _ADC2_ (�-UTR conferred both 135 strong basal repression and repression in response to exogenous Put. 136 We next tested whether perturbation of endogenous Put biosynthesis affected Eluc 137 activity from the _ADC2_ (�-UTR reporter. We first used protoplasts prepared from two _adc2_ 138 mutants carrying T-DNA insertions in _ADC2_, _adc2-1_ (SALK_073977) and _adc2-2_ 139 (WiscDsLoxHs065_09H; Supplemental Fig. S1A). The _adc2-1_ mutant has been reported to 140 reduce endogenous _ADC2_ expression and Put levels (Kim et al. 2013). Reverse transcription 141 quantitative PCR (RT-qPCR) analysis showed that _ADC2_ transcript abundance was markedly 6 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 142 reduced in both mutants compared with Columbia-0 (Col-0) wild-type (WT), with only one of 143 three biological replicates of _adc2-1_ and none of those of _adc2-2_ yielding quantifiable values 144 (Supplemental Fig. S1B), suggesting that endogenous Put levels may also be reduced in _adc2-_ 145 _2_ . In protoplasts prepared from both mutants, relative Eluc activity from _ADC2_ (�-UTR: _Eluc_ 146 was higher than that in Col-0 in the absence of exogenous Put. After treatment with 2 mM Put, 147 relative Eluc activity was reduced in all genotypes, and the differences observed in untreated 148 conditions were no longer detected (Fig. 2D). 149 We further examined the effect of inhibiting endogenous Put biosynthesis in protoplasts 150 prepared from Arabidopsis MM2d suspension cells (Menges and Murray 2002). Protoplasts 151 transfected with _ADC2_ (�-UTR: _Eluc_ - 6≥6 ₁=⁵W×7ₙ6O ‡₁ₙT −≥ ‡₁ₙT−Wₙ t ;G � 152 difluoromethylarginine (DFMA), an ADC inhibitor reported to reduce endogenous Put levels 153 (Birecka et al. 1985; Fallon and Phillips 1988; Fig. 2E). Unlike in mesophyll protoplasts, 1 mM 154 Put alone did not significantly reduce relative Eluc activity in MM2d protoplasts (Fig. 2, B and 155 F). Because increases in Put levels have been reported during callus induction and cellular 156 reprogramming (Desai and Mehta 1985; Iwase et al. 2018), this difference may reflect 157 differences in endogenous Put levels associated with the distinct differentiation states of the 158 source cells. DFMA alone significantly increased Eluc activity, whereas addition of 1 mM Put 159 in the presence of DFMA significantly reduced it (Fig. 2F). These results support the 160 interpretation that the DFMA-induced increase in Eluc activity resulted from a decrease in 161 endogenous Put levels. Together, the responses to exogenous Put, _adc2_ mutations, and ADC 162 inhibition demonstrate feedback regulation mediated by the _ADC2_ (�-UTR in response to 163 intracellular Put levels. 164 7 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) ## 165 Put-responsive repression mediated by the ADC2 ��-UTR involves 166 regulation at both the mRNA and translational levels 167 We next examined the regulatory level at which the _ADC2_ (�-UTR mediates Put-responsive 168 repression. Changes in luciferase activity alone cannot distinguish altered translation from 169 altered reporter mRNA abundance. We therefore generated Arabidopsis transgenic lines 170 carrying the _ADC2_ (�-UTR: _Eluc_ construct and conducted reporter assays together with RT 171 qPCR. Two independent homozygous _ADC2_ (�-UTR: _Eluc_ lines were analyzed. As a negative 172 ⁵−=ₙ≥−I, ‡6 W⁷6O 7 T−;−��’−W⁷ I₁=6 ⁵7≥≥�₁=’ ₙT6 (�-UTR of _MASP1_ (At4g03260) in the same 173 reporter vector (Fig. 3A). Ten-day-old seedlings were incubated with or without 2 mM Put for 174 24 h, after which Eluc activity and _Eluc_ mRNA abundance were measured. Put treatment 175 significantly reduced Eluc activity in both independent _ADC2_ (�-UTR: _Eluc_ lines to 176 approximately 40% of their respective untreated levels but had no significant effect on Eluc 177 activity in the _MASP1_ control line (Fig. 3B). 178 Put treatment also caused a modest but statistically significant decrease in _Eluc_ mRNA 179 abundance in both _ADC2_ (�-UTR: _Eluc_ lines, to approximately 70% to 80% of their respective 180 untreated levels, whereas no significant change was detected in the _MASP1_ control line (Fig. 181 3C). 182 We calculated relative Eluc translational efficiency (TE) as the ratio of relative Eluc 183 activity to relative _Eluc_ mRNA abundance. Put treatment significantly reduced relative Eluc 184 TE in both _ADC2_ (�-UTR: _Eluc_ lines to approximately 50% to 60% of their respective untreated 185 levels, whereas no significant change was detected in the _MASP1_ control line (Fig. 3D). The 186 reduction in relative Eluc TE (approximately 40%�50%) was greater than the reduction in 187 reporter mRNA abundance (approximately 20%�30%). These results support translational 188 repression as a major component of the Put-responsive regulation mediated by the _ADC2_ (� 189 UTR, with a smaller contribution from reduced reporter mRNA abundance. 8 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 190 ## 191 The non-AUG-initiated uORF is required for Put-responsive regulation 192 mediated by the ADC2 ��-UTR 193 We next determined which _cis_ -elements within the _ADC2_ (�-UTR are required for Put 194 responsive regulation. A short AUG-initiated uORF and a conserved non-AUG-initiated uORF 195 have been identified in plant _ADC_ (�-UTRs (Chang et al. 2000; Hiragori et al. 2023), and both 196 uORFs are present in the Arabidopsis _ADC2_ (�-UTR (Fig. 4A; Supplemental Fig. S2). The 197 conserved ADC-box motif is also present in the _ADC2_ (�-UTR and overlaps with the 198 AUU/CUG-initiated uORF (Fig. 4A; Supplemental Fig. S2). 199 To test whether translation of these uORFs is required for Put-responsive regulation, the 200 start codons of these uORFs in the _ADC2_ (�-UTR: _Eluc_ reporter were replaced by codons that 201 were not expected to function as start codons (AUG to AAA, AUU to GUU, and CUG to CUC) 202 (Fig. 4B). Our previous study using Arabidopsis suspension cells suggested that both the AUU 203 and CUG codons of the non-AUG-initiated uORF contribute to repression of the main ORF 204 (Hiragori et al. 2023). We therefore mutated either or both of these codons. WT and mutant 205 reporter constructs were introduced into Arabidopsis mesophyll protoplasts, and Eluc activity 206 was compared in the presence and absence of exogenous Put. Mutation of the AUG codon 207 increased Eluc activity in the absence of exogenous Put, indicating that the AUG-initiated 208 uORF represses expression of the main ORF (Fig. 4C). However, this mutation did not affect 209 the Put-responsive reduction in Eluc activity (Fig. 4D). Thus, the AUG-initiated uORF 210 contributes to basal repression of the main ORF but is not required for Put-responsive 211 regulation. 212 Under untreated conditions, mutation of the CUG codon or simultaneous mutation of 213 the AUU and CUG codons increased Eluc activity relative to the WT reporter, whereas 214 mutation of the AUU codon alone had no significant effect on Eluc activity (Fig. 4E). By 9 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 215 contrast, none of these non-AUG-codon mutants showed a significant reduction in Eluc activity 216 following Put treatment (Fig. 4F). Thus, disruption of either the AUU or CUG initiation codon 217 abolished the Put-responsive reduction in Eluc activity. Because the AUU and CUG codons are 218 located close to the ADC-box, we assessed whether these mutations altered the predicted 219 secondary structure of the ADC-box region using an _in silico_ RNA secondary structure 220 prediction tool (Sato et al. 2021). The single-codon mutations of either AUU or CUG had little 221 effect on the predicted RNA secondary structure, whereas simultaneous mutation of both 222 codons altered the predicted structure (Supplemental Fig. S3). Although the effects of the 223 substitutions on RNA structure cannot be completely excluded, the loss of Put-responsive 224 regulation in each of the single mutants, whose predicted structures remained largely 225 unchanged, supports a requirement for both initiation codons in the response under these assay 226 conditions, consistent with a role for uORF translation. 227 ## 228 Predicted RNA secondary structure in the ADC-box region contributes to 229 Put-responsive repression 230 We next tested whether the ADC-box is also required for Put-responsive regulation mediated 231 by the _ADC2_ (�-UTR. To distinguish the contribution of its conserved nucleotide sequence from 232 ₙT7ₙ −3 ₁ₙ⁷ 9≥6O₁⁵ₙ6O ²P– ⁷6⁵−=O7≥� ⁷ₙ≥W⁵ₙW≥6, ‡6 ’6=6≥7ₙ6O 7= �W=97₁≥6O� ;Wₙ7=ₙ O6⁷₁’=6O ₙ− 233 disrupt base pairing within the ADC-×−� 7=O 7 �97₁≥6O� ;Wₙ7=ₙ ⁵7≥≥�₁=’ 7OO₁ₙ₁−=7I nucleotide 234 substitutions designed to restore the predicted secondary structure (Fig. 5, A and B). Although 235 the unpaired and paired mutations altered part of the amino acid sequence of the overlapping 236 AUU/CUG-initiated uORF, neither mutation changed the position or length of the uORF. Under 237 untreated conditions, the unpaired and paired mutants showed approximately eightfold and 238 fourfold the Eluc activity of the WT reporter, respectively (Fig. 5C). However, the unpaired 239 mutant showed no significant Put-dependent decrease in Eluc activity, whereas the paired 10 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 240 mutant retained Put-responsive repression comparable to that of the WT reporter, with activity 241 decreasing to approximately 25% of the untreated level (Fig. 5D). The paired mutant therefore 242 restored Put responsiveness despite extensive changes in its primary sequence. Together, these 243 results indicate that the RNA secondary structure formed in the ADC-box region is critical for 244 Put-responsive regulation. 245 ## 246 Regulation mediated by the ADC2 ��-UTR contributes to normal growth and 247 Put homeostasis in Arabidopsis 248 The expression level of _ADC2_ is a major determinant of Put levels in Arabidopsis (Urano et al. 249 2004; Rossi et al. 2015). Put-responsive repression of _ADC2_ expression may therefore function 250 as a negative feedback mechanism that contributes to Put homeostasis. To examine the 251 importance of _ADC2_ (�-UTR-mediated regulation _in planta_, we introduced mutations into the 252 genomic region corresponding to the _ADC2_ (�-UTR using the CRISPR-Cas9 system. We 253 obtained two mutant lines, #3-7 and #11-8, harboring 25- and 31-nt deletions, respectively (Fig. 254 6A; Supplemental Fig. S4A). In both mutants, the deleted regions span the AUU/CUG-initiated 255 uORF and the ADC-box. Both deletions shifted the reading frame of the AUU/CUG-initiated 256 uORF, changing the position of the stop codon 17 nt downstream of the original stop codon 257 (Fig. 6A; Supplemental Fig. S4A). Because the deleted nucleotides participate in base pairing 258 within the ADC-box RNA secondary structure, we predicted the RNA secondary structures 259 surrounding the ADC-box in each mutant sequence _in silico_ . The predicted structures of both 260 mutant sequences differed substantially from that of the WT sequence (Supplemental Fig. S4B). 261 When grown on half-strength Murashige and Skoog (MS) medium, #3-7 and #11-8 262 showed significantly reduced shoot fresh weight and primary root length compared with Col 263 0, with #11-8 exhibiting the more severe growth phenotype (Fig. 6, B�D). Radicle emergence 11 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 264 and cotyledon expansion were delayed in both lines, and the delay in cotyledon expansion was 265 particularly pronounced in #11-8 (Supplemental Fig. S5, A and B). 266 We quantified polyamines in Col-0 and #11-8 by high-performance liquid 267 chromatography (HPLC). In #11-8, Put abundance was approximately 3.5 times that in Col-0, 268 whereas Spd and the combined Spm+Tspm fraction showed no significant differences (Fig. 6E; 269 Supplemental Fig. S6). Put accumulation in #11-8, the line with the more severe growth and 270 germination phenotypes, suggests an association between Put accumulation and these 271 developmental defects. Together, these results support a contribution of _ADC2_ (�-UTR 272 mediated regulation to Put homeostasis and normal germination and growth in Arabidopsis. 273 12 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) # 274 DISCUSSION ## 275 Put-responsive feedback regulation mediated by the ADC2 ��-UTR 276 Our findings identify the Arabidopsis _ADC2_ (�-UTR as a _cis_ -regulatory module that mediates 277 negative feedback regulation in response to Put availability. Exogenous Put reduced Eluc 278 activity in mesophyll protoplasts, whereas _adc2_ mutations or pharmacological inhibition of 279 ADC increased it; Put supplementation counteracted these increases (Fig. 2). In transgenic 280 plants, Put reduced Eluc activity more strongly than reporter mRNA abundance and decreased 281 calculated relative Eluc TE, supporting a major contribution from translational repression (Fig. 282 3). 283 Exogenous Put reduced the abundance of reporter mRNA modestly but significantly in 284 an _ADC2_ (�-UTR-dependent manner (Fig. 3C). Because Put did not affect the abundance of 285 _MASP1_ reporter mRNA driven by the same 35Spro, this reduction may reflect _ADC2_ (�-UTR 286 dependent mRNA destabilization rather than a change in transcription. Translation termination 287 at a uORF or ribosome stalling can engage mRNA quality control pathways, including NMD 288 (Gaba et al. 2005; Nyikó et al. 2009; Uchiyama-Kadokura et al. 2014). Changes in ribosome 289 behavior on the _ADC2_ (�-UTR could therefore couple repression of main ORF translation to 290 reduced mRNA stability. 291 ## 292 Roles of the cis -regulatory elements in the ADC2 ��-UTR and potential 293 mechanisms of Put-responsive regulation 294 Mutational analyses showed that the AUU/CUG-initiated uORF and the predicted RNA 295 secondary structure of the ADC-box region are central to Put-responsive regulation, whereas 296 the short AUG-initiated uORF is dispensable for this response (Figs. 4 and 5). Elimination of 297 either the AUU or CUG initiation codon abolished the Put-responsive repression, as did 13 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 298 disruption of predicted base pairing in the ADC-box region (Fig. 4F; Fig. 5D). Conversely, a 299 compensatory mutation predicted to reconstitute the ADC-box structure reinstated the Put 300 responsive repression despite extensive changes in primary sequence (Fig. 5), supporting the 301 importance of the RNA structure rather than the conserved sequence itself. Notably, the effects 302 of these mutations on Put responsiveness did not parallel their effects on basal repression. 303 Mutation of the AUG-initiated uORF increased basal Eluc activity without abolishing Put 304 responsive repression; mutations of the AUU/CUG-initiated uORF abolished Put 305 responsiveness despite differing effects on basal activity; and both the unpaired and paired 306 ADC-box mutants increased basal activity, although only the unpaired mutant lost Put 307 responsiveness (Figs. 4 and 5). Together, these findings indicate that Put-responsive regulation 308 does not simply reflect the strengthening or alleviation of basal repression. 309 Because the AUU/CUG-initiated uORF overlaps the ADC-box (Fig. 4A; Supplemental 310 Fig. S2), their contributions to Put responsiveness cannot be completely separated. 311 Nevertheless, the paired mutant restored Put responsiveness despite containing more extensive 312 sequence changes within the overlapping uORF than the unpaired mutant, arguing that the loss 313 of responsiveness in the unpaired mutant cannot be explained simply by changes in the uORF 314 ⁷6⁸W6=⁵6 eR₁’s (.s M−=†6≥⁷6I�, ₙT6 �–// 7=O �M/q ;Wₙ7ₙ₁−=⁷ 7×−I₁⁷T6O µWₙ ≥6⁷9−=⁷₁†6=6⁷⁷, 315 although both mutations were predicted to have little effect on the ADC-box RNA structure, 316 making it unlikely that their effects resulted from disruption of this structure (Fig. 4F; 317 Supplemental Fig. S3). These observations support a model in which both the AUU/CUG 318 initiated uORF and the ADC-box RNA structure are required for Put-responsive repression. 319 One possible mechanism underlying this Put-responsive regulation is that Put stabilizes 320 an RNA structure in the ADC-box region. Owing to their cationic nature, polyamines can 321 stabilize RNA structures by interacting with the negatively charged phosphate backbone 322 (Kusano et al. 2008; Igarashi and Kashiwagi 2010). A stabilized ADC-box structure could 14 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 323 repress main ORF translation in at least two ways. First, because it lies near and downstream 324 of the AUU/CUG initiation sites, it could slow scanning by a 43S preinitiation complex and 325 thereby enhance recognition of either of the non-AUG start codons, consistent with previous 326 reports that downstream RNA structures can promote initiation at upstream near-cognate 327 codons (Kozak 1990; Wang et al. 2022). Increased uORF initiation would reduce the number 328 of ribosomes reaching the main ORF. Another possibility is that the ADC-box structure could 329 slow elongation or termination of ribosomes translating the uORF, obstructing subsequent 330 scanning or translating ribosomes and reducing the number of ribosomes that reach the main 331 ORF (Dever et al. 2020; Bao et al. 2022). In either model, cooperation between the ADC-box 332 RNA structure and the AUU/CUG-initiated uORF could explain the effects of the mutations on 333 Put responsiveness (Figs. 4 and 5), although the current data do not distinguish between these 334 models. 335 Put could also act directly on ribosomes to selectively modulate _ADC2_ translation 336 ₙT≥−W’T ₁ₙ⁷ (�-UTR. Put has been reported to associate with actively translating ribosomes and 337 to alter their ribosomal protein composition (Murillo et al. 2025). Tspm can also bind to the 338 peptidyl transferase center in ribosomes and modulate translation of specific genes (Ko et al. 339 2026). These findings leave open the possibility that direct interaction between Put and 340 ribosomes modulates translation on the _ADC2_ (�-UTR, although direct evidence for such a 341 mechanism is currently lacking. 342 ## 343 Physiological significance of regulation mediated by the ADC2 ��-UTR 344 The genome-edited lines #3-7 and #11-8 provide a physiological link between regulation 345 mediated by the _ADC2_ (�-UTR and Put homeostasis. Both lines exhibited impaired growth and 346 delayed germination, with more pronounced defects in #11-8 (Fig. 6; Supplemental Fig. S5). 15 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 347 The increased Put level in this more severely affected line suggests that Put overaccumulation 348 may be associated with the developmental defects (Fig. 6E). 349 The phenotypes of the _ADC2_ (�-UTR mutants are consistent with a previous report 350 showing that plants constitutively overexpressing _ADC2_ overaccumulate Put and show growth 351 defects associated with reduced levels of bioactive gibberellins and their precursors (Alcázar et 352 al. 2005). Because gibberellins are phytohormones that regulate the timing of seed germination, 353 tT6 O6I7�6O ’6≥;₁=7ₙ₁−= −3 ₙT6 (�-UTR mutants could involve altered gibberellin metabolism. 354 Together, these findings support a contribution of the regulatory functions of the _ADC2_ (�-UTR 355 to Put homeostasis and normal germination and growth _in planta_ . 356 The absence of corresponding increases in the other major polyamines (Fig. 6E) 357 suggests that regulation mediated by the _ADC2_ (�-UTR is particularly important for restricting 358 Put accumulation rather than uniformly controlling downstream polyamine levels. Together 359 with feedback mechanisms regulating _SAMDC_ and _ACL5_ expression, this finding supports a 360 model in which polyamine homeostasis is controlled separately at multiple steps of the 361 biosynthetic pathway. 362 ## 363 Potential conservation and lineage-dependent importance of the ADC ��- 364 UTR-mediated regulation 365 Conservation of the ADC-box and the non-AUG-initiated uORF in multiple plant _ADC_ genes 366 raises the possibility that feedback regulation mediated by _ADC_ 5�-UTRs contributes to Put 367 homeostasis beyond Arabidopsis (Wu et al. 2019; Hiragori et al. 2023). However, the 368 physiological contribution of this feedback regulation may depend on the organization of Put 369 biosynthesis in each lineage. Arabidopsis lacks a functional ODC, and _ADC2_ is a major 370 determinant of its Put levels (Fig. 1; Hanfrey et al. 2001; Urano et al. 2004; Rossi et al. 2015), 371 potentially making feedback through the _ADC2_ (�-UTR particularly consequential. In plants 16 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 372 possessing both ADC and ODC pathways, regulation of _ODC_ genes could complement or partly 373 substitute for regulation mediated by _ADC_ (�-UTRs. A more complete understanding of Put 374 homeostasis in plants will require comparative analyses of the regulation of genes involved in 375 Put biosynthesis and metabolism across species with different biosynthetic architectures. 376 17 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) # 377 MATERIALS AND METHODS ## 378 Plant materials and growth conditions 379 _Arabidopsis thaliana_ Col-0 was used as WT. The T-DNA insertion lines _adc2-1_ 380 (SALK_073977) and _adc2-2_ (WiscDsLoxHs065_09H), both in the Col-0 background, were 381 obtained from the Arabidopsis Biological Resource Center (Columbus, OH, USA). 382 Homozygous T-DNA insertions were confirmed by the absence of a gene-specific PCR product 383 and the presence of a gene-specific/T-DNA-border PCR product. Primers used for genotyping 384 are listed in Supplemental Table S1. MM2d suspension cells were derived from Landsberg 385 _erecta_ . Stable lines carrying an _ADC2_ (�-UTR or _MASP1_ (�-UTR reporter and two genome 386 edited lines #3-7 and #11-8, carrying 25- and 31-nt deletions in the _ADC2_ (�-UTR, respectively, 387 were generated in this study (see below). 388 Seeds were surface-sterilized with sodium hypochlorite and stratified at 4°C in the dark 389 for 4�6 d for transgenic reporter assays or 6 d for germination and growth assays. Seedlings 390 were grown at 22°C under continuous LED light (75�too �;−I ;�� ⁷��. −= T7I3-strength MS 391 ;6O₁W; ⁵−=ₙ7₁=₁=’ tn e‡h†. ⁷W⁵≥−⁷6 7=O os( ’ U�� l-(N-morpholino)ethanesulfonic acid 392 (MES; pH 5.7), solidified with 0.8% (w/v) agar or, for growth measurements, 0.5% (w/v) gellan 393 gum. Mesophyll protoplast donors were grown for 4 weeks in Jiffy-7 pellets (Sakata Seed 394 Corporation, Yokohama, Japan) at 22°C under continuous fluorescent light (50�βo �;−I ;�� 395 ⁷��.s 396 ## 397 Plasmid construction 398 Plasmids used or generated in this study and primers used for plasmid construction are listed in 399 Supplemental Tables S2�S4. Reporter plasmids used for transient expression assays were 400 derived from pKM56, which contains a 35Spro, an _Eluc_ -PEST, and the Arabidopsis _HSP18.2_ 401 terminator ( _HSP_ ter) in pUC19 (Hayashi et al. 2017; Hiragori et al. 2023). pKM56 served as the 18 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 402 control plasmid. For the _ADC2_ 5�-UTR: _Eluc_ reporter plasmid pYH21, the _ADC2_ (�-UTR and 403 first 5 nt of the main ORF were amplified from Col-0 cDNA and inserted between the XbaI and 404 SalI sites of pKM56 by seamless ligation cloning extract (SLiCE) assembly (Motohashi 2017). 405 Mutations were introduced by overlap PCR (Ho et al. 1989). All PCR-amplified regions were 406 verified by Sanger sequencing. The reference plasmid pKN17, which carries _Rluc_ under the 407 control of 35Spro, was generated by inserting annealed complementary oligonucleotides 408 �RLUC 5UTR Fw/Rv� between the XbaI and SalI sites of pKM5 (Hayashi et al. 2017) and was 409 used in all transient assays. 410 The binary vector pYH74 was derived from pKI1.1R (Tsutsui and Higashiyama 2017). 411 An annealed �GG to CC IF for/rev� fragment was inserted into SbfI/PmeI-digested pKI1.1R by 412 In-Fusion assembly (Takara, Shiga, Japan) to remove SbfI and SalI sites. The resulting plasmid 413 was digested with KpnI and StuI. A fragment amplified from pKI1.1R by sequential PCRs with 414 �M1¹–zK3−≥lK=6‡�h�M1¹–zK≥6†′ 7=O �M1¹–zK3−≥�h�CBSAE_rev� was inserted into the 415 KpnI/StuI sites by In-Fusion assembly, replacing the Cas9/guide RNA (gRNA) cassette with 416 an SbfI�BamHI�EcoRI�SmaI�XmaI multiple cloning site (MCS). To generate the _MASP1_ (� 417 UTR reporter plasmid pYH76, the SbfI/EcoRI fragment containing the 35Spro:: _MASP1_ (� 418 UTR: _Eluc_ -PEST:: _HSP_ ter cassette was excised from pYH17 (Hiragori et al. 2023) and ligated 419 into SbfI/EcoRI-digested pYH74. To create the _ADC2_ (�-UTR reporter plasmid pYH75, the 420 SbfI/BglII fragment containing the _ADC2_ (�-UTR: _Eluc_ construct was excised from pYH21 and 421 ligated into SbfI/BglII-digested pYH76. 422 ## 423 Generation of transgenic and genome-edited lines 424 Constructs were introduced into _A. thaliana_ plants by the floral dip method (Clough and Bent 425 1998) using _Agrobacterium_ _tumefaciens_ strain GV2260 (Deblaere et al. 1985). For transgenic 426 reporter lines, hygromycin B-resistant T1 plants were selected. Then, T2 lines carrying the 19 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 427 transgene at a single locus were selected based on a 3:1 segregation ratio of hygromycin B 428 resistance, and homozygous lines were established in the T3 generation by confirming that no 429 segregation occurred. Two _ADC2_ (�-UTR reporter lines #1 and #2 originated from different T0 430 plants. 431 For genome-edited lines, annealed �SPE2 gR1 F/R� or �SPE2 gR2 F/R� 432 oligonucleotides (Supplemental Table S5) were ligated into AarI-digested pKI1.1R to generate 433 pYH70 (gRNA1) or pYH71 (gRNA2). T1 transformants were selected by hygromycin B 434 resistance. CRISPR/Cas9-induced mutations were detected by heteroduplex mobility assays 435 (Foster et al. 2019) in T1 and T2 plants and verified by Sanger sequencing. Homozygous T3 436 plants or their progeny were used for analyses. Homozygous mutant lines #3-7 and #11-8 were 437 generated using gRNA2 and gRNA1, respectively. 438 ## 439 Mesophyll protoplast preparation and transfection 440 Mesophyll protoplasts were isolated following Wu et al. (2009), with some modifications. 441 Lower epidermal layers of rosette leaves were removed using masking tape (Kamoi Kakoshi, 442 Okayama, Japan). Twenty leaves were incubated in 20 mL of enzyme solution (1% Cellulase 443 �Onozuka� R10 and 0.25% Macerozyme R10 (Yakult, Tokyo, Japan), 0.4 M mannitol, 10 mM 444 M7MI�, lo ;G ΔMI, ostn bovine serum albumin, and 20 mM MES-KOH, pH 5.6) for 2�3 h at 445 room temperature and 40 rpm. Centrifugation during washing and after polyethylene glycol 446 (PEG) treatment was performed at 1,000 rpm for 3 min at 4°C using a Sorvall ST 8FR centrifuge 447 (Thermo Fisher Scientific, Waltham, MA, USA) with a swinging-bucket rotor. W5 buffer 448 ⁵−=ₙ7₁=6O t(i ;G P7MI, tl( ;G M7MI�, ( ;G ΔMI, ( ;G ’IW⁵−⁷6, 7=O l ;G Gz¹ e9F (sp.s 449 Each transfection contained 1.5 pmol each of the reporter and reference plasmids in 10 450 �U ‡7ₙ6≥, ts( � to� 9≥−ₙ−9I7⁷ₙ⁷ ₁= too �U G7G’ ⁷−IWₙ₁−= eosi G ;7==₁ₙ−I, t( ;G G’MI�, 7=O 451 4 mM MES-Δ0F, 9F (sx., 7=O tto �U µzq ⁷−IWₙ₁−= eion µzqiooo, ost G M7MI�, 7=O osl G 20 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 452 mannitol). After 15 min at room temperature, 800 µL W5 was added. Protoplasts were pelleted 453 and resuspended in 200 µL WI medium (0.5 M mannitol, 20 mM KCl, and 4 mM MES-KOH, 454 pH 5.6). Put was added to the protoplast cultures at 0�4 mM for dose-response assays or at 0 or 455 2 mM for other comparisons. The protoplasts were then incubated at 22°C for 24 h in the dark. 456 ## 457 MM2d suspension cell protoplast preparation and transfection 458 MM2d protoplast isolation and transfection followed Hiragori et al. (2023). Briefly, cells 459 harvested 3 d after subculture were digested in modified Linsmaier and Skoog (LS) medium 460 containing 1% Cellulase �Onozuka� RS (Yakult), 0.1% Pectolyase Y23 (Kyowakasei, Osaka, 461 Japan), and 0.4 M mannitol at 26°C. Each transfection contained 3 µg of pYH21 and pKN17 in 462 x �U ‡7ₙ6≥, g � to� 9≥−ₙ−9I7⁷ₙ⁷ ₁= too �U G7G’ ⁷−IWₙ₁−= eosi G ;7==₁ₙ−I, t( ;G G’MI�, 7=O 463 5 mM MES, pH 5.8), and 106 µL PEG solution. After PEG treatment, cells were resuspended 464 in 200 µL modified LS medium containing 0.4 M mannitol. Put and/or DFMA were added to 465 final concentrations of 1 mM each immediately before incubation at 22°C for 24 h in the dark. 466 ## 467 Luciferase assay 468 For transient expression assays, protoplasts were frozen in liquid nitrogen and extracted by 469 vortexing for 2 min in 100 or 200 µL buffer containing 100 mM sodium phosphate (pH 7) and 470 5 mM dithiothreitol (DTT). Eluc and Rluc activities were measured using the Dual-Luciferase 471 Reporter Assay System (Promega, Madison, WI, USA), and Eluc activity was normalized to 472 Rluc activity. 473 For transgenic reporter assays, 15 ten-day-old seedlings per flask were submerged in 10 474 mL liquid half-strength MS medium supplemented with 0 or 2 mM Put and shaken at 80 rpm 475 for 24 h. Whole seedlings from each flask were pooled, ground in liquid nitrogen with a mortar 476 and pestle, and divided for RNA and protein extraction. For protein extraction, the samples 21 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 477 were vortexed for 2 min in the same extraction buffer. Eluc activity was measured using the 478 Luciferase Assay System (Promega) and normalized to total protein determined with Pierce 479 660 nm Protein Assay Reagent (Thermo Fisher Scientific). 480 ## 481 RNA analysis by RT-qPCR 482 RNA was extracted using TRIzol (Thermo Fisher Scientific) and treated with RQ1 RNase-Free 483 DNase (Promega). RNA (500 ng) was reverse-transcribed using SuperScript IV (Thermo Fisher 484 Scientific) according to the manufacturer�s protocol, except that RNaseOUT was omitted, 100 485 ng of oI₁’−eO%.����� 9≥₁;6≥ ‡7⁷ W⁷6O, 7=O reverse transcription was performed at 50°C for 60 486 min. RT-qPCR used LightCycler FastStart DNA MasterPLUS SYBR Green I and a LightCycler 487 2.0 instrument (Roche, Mannheim, Germany). _Eluc_ transcript abundance was normalized to 488 _����_ (At5g60390) in the same cDNA sample. For analysis of endogenous _ADC2_ expression 489 in Col-0 and the _adc2_ mutants, three fresh rosette leaves from an untreated 4-week-old plant 490 were pooled per biological replicate (n = 3 per genotype). _ADC2_ transcript abundance was 491 normalized to _����_ . Primers used for RT-qPCR are listed in Supplemental Table S6. 492 ## 493 RNA secondary structure prediction 494 WT and mutant RNA structures were predicted and drawn using the MXFold2 Server 495 (https://ws.sato-lab.org/mxfold2/; Sato et al. 2021) with default settings. Inputs comprised the 496 ADC-box region and 100 nt on each side. 497 ## 498 Growth and germination assays 499 For growth assays, seedlings that had germinated by day 2 after stratification were selected and 500 transferred to fresh gellan-gum plates on day 5. On day 10, shoots were cut at the root� 22 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 501 hypocotyl boundary and weighed individually in 1.5-mL tubes. Primary root length was 502 measured from images using ImageJ. 503 For germination assays, seeds stored dry with silica gel in the dark for more than one month 504 were sown on three agar plates. The same seeds were observed over 7 d after stratification. The 505 percentage in each plate was treated as one replicate at each time point (n = 3). 506 ## 507 Polyamine quantification 508 Whole seedlings grown for 14 d on half-strength MS agar under continuous light 509 e799≥−�₁;7ₙ6I� p( �;−I ;�� ⁷��. ‡6≥6 9−−I6O ×� 9I7ₙ6 e799≥−�₁;7ₙ6I� lo ⁷66OI₁=’⁷, too ;’ 96≥ 510 sample; n = 3 per genotype). Frozen, ground tissue was extracted with four volumes (v/w) of 511 5% perchloric acid. A 200-µL supernatant aliquot was supplemented with 1,7-diaminoheptane 512 (Tokyo Chemical Industry, Tokyo, Japan) to 50 µM in the extract, mixed with 400 µL of 10 513 ;’ ;U�� O7=⁷�I chloride (Tokyo Chemical Industry) in acetone and 200 µL saturated sodium 514 carbonate, and incubated overnight at room temperature in the dark. After adding 100 µL of 515 too ;’ ;U�� 7⁸W6−W⁷ 9≥−I₁=6 7=O ₁=⁵W×7ₙ₁=’ 3−≥ go ;₁=, ⁷7;9I6⁷ ‡6≥6 ;₁�6O ‡₁ₙT l(o �U 516 benzene. The organic phase was recovered, dried under reduced pressure, dissolved in 100 µL 517 acetonitrile:water (1:1), and filtered through a Millex syringe filter (0.45 µm; Merck Millipore, 518 Burlington, MA, USA). Mixed Put, Spd, and Spm standards (each at 1, 10, 20, 40, or 80 µM in 519 perchloric acid solution) were processed identically. 520 HPLC used a Puresil 5-µm C18 column (120 Å, 4.6 × 150 mm; Waters, Milford, MA, 521 USA) with a Resolve Silica Guard-Pak precolumn (Waters), an ELITE LaChrom L-2130 pump 522 and L-2400 UV detector, and D-2000 Elite software (Hitachi, Tokyo, Japan). Injection volume 523 - 7⁷ io �U, O6ₙ6⁵ₙ₁−= ‡7⁷ 7ₙ gio =;, 7=O 3I−‡ ≥7ₙ6 ‡7⁷ t ;U ;₁=��s –⁵6ₙ−=₁ₙ≥₁I6f‡7ₙ6≥ was 524 60:40 for 0�15 min, increased to 100% acetonitrile over 15�30 min, maintained for 30�45 min, 525 and returned to 60:40 for 45�60 min. 23 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 526 ## 527 Sequence identifiers 528 The Arabidopsis locus identifiers discussed in this study are At4g34710 ( _ADC2_ ), At2g16500 529 ( _ADC1_ ), At4g03260 ( _MASP1_ ), and At5g60390 ( _����_ ). 530 531 24 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) # 532 SUPPLEMENTARY MATERIAL 533 Supplemental Figure S1. T-DNA insertion alleles of _ADC2_ . 534 Supplemental Figure S2. Sequence and regulatory elements of the _ADC2_ (�-UTR. 535 Supplemental Figure S3. Predicted RNA secondary structures of non-AUG start-codon 536 mutants. 537 Supplemental Figure S4. Deletion sequences and predicted RNA structures of _ADC2_ (�-UTR 538 genome-edited alleles. 539 Supplemental Figure S5. Germination and cotyledon expansion of _ADC2_ (�-UTR genome 540 edited lines. 541 Supplemental Figure S6. HPLC chromatograms and calibration curves for polyamine 542 quantification. 543 Supplemental Table S1. Primers used for mutant genotyping. 544 Supplemental Table S2. Plasmids used in transient assays and primers used for plasmid 545 construction. 546 Supplemental Table S3. Plasmids used for generation of transgenic or genome-edited lines and 547 oligo DNA used for plasmid construction. 548 Supplemental Table S4. Primers used for plasmid construction. 549 Supplemental Table S5. Oligo DNA used for guide RNA cloning. 550 Supplemental Table S6. Primers used for RT-qPCR. 551 # 552 ACKNOWLEDGMENTS 553 The authors thank Ms. Saeko Yasokawa for technical assistance, and Ms. Sayo Aono and Ms. 554 Aki Kosugi for general assistance. 555 25 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) # 556 AUTHOR CONTRIBUTIONS 557 Y.H. and H.O. designed research. Y.H., M.Y., T.K., A.K., Y.S., and Y.G. performed the 558 experiments. Y.H., Y.S., Y.Y., S.N., and H.O. analyzed data. Y.H., S.N., and H.O. wrote the 559 manuscript. 560 # 561 FUNDING 562 This work was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI 563 [Grant Nos. JP22KJ0031 and JP24K23113 to Y.H., JP23K0579903 and JP26K0873306 to 564 H.O.]. 565 _Conflicts of interest statement._ None declared. 566 # 567 REFERENCES 568 Alcázar R, García-Martínez JL, Cuevas JC, Tiburcio AF, Altabella T. 2005. Overexpression 569 of ADC2 in Arabidopsis induces dwarfism and late-flowering through GA deficiency. Plant 570 Journal. 43(3):425�436. https://doi.org/10.1111/j.1365-313X.2005.02465.x 571 von Arnim AG, Jia Q, Vaughn JN. 2014. Regulation of plant translation by upstream open 572 reading frames. Plant Science. 214:1�12. https://doi.org/10.1016/j.plantsci.2013.09.006 573 Bao C, Zhu M, Nykonchuk I, Wakabayashi H, Mathews DH, Ermolenko DN. 2022. Specific 574 length and structure rather than high thermodynamic stability enable regulatory mRNA stem575 loops to pause translation. Nat Commun. 13(1):988. https://doi.org/10.1038/s41467-022576 28600-5 577 Birecka H, Bitonti AJ, McCann PP. 1985. Activities of Arginine and Ornithine 578 Decarboxylases in Various Plant Species. Plant Physiol. 79(2):515�519. 579 https://doi.org/10.1104/pp.79.2.515 580 Blázquez MA. 2024. Polyamines: Their Role in Plant Development and Stress. Annual 581 Review of Plant Biology. 75:95�117. https://doi.org/10.1146/annurev-arplant-070623-110056 582 Chang KS, Lee SH, Hwang SB, Park KY. 2000. Characterization and translational regulation 583 of the arginine decarboxylase gene in carnation (Dianthus caryophyllus L.). Plant Journal. 584 24(1):45�56. https://doi.org/10.1046/j.0960-7412.2000.00854.x 26 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 585 Chen D, Shao Q, Yin L, Younis A, Zheng B. 2019. Polyamine function in plants: 586 Metabolism, regulation on development, and roles in abiotic stress responses. Frontiers in 587 Plant Science. 9(January):1�13. https://doi.org/10.3389/fpls.2018.01945 588 Clough SJ, Bent AF. 1998. Floral dip: a simplified method for Agrobacterium-mediated 589 transformation of Arabidopsis thaliana. The Plant Journal. 16(6):735�743. 590 https://doi.org/10.1046/j.1365-313x.1998.00343.x 591 Deblaere R, Bytebier B, De Greve H, Deboeck F, Schell J, Van Montagu M, Leemans J. 592 1985. Efficient octopine Ti plasmid-derived vectors for Agrobacterium-mediated gene 593 transfer to plants. Nucleic Acids Res. 13(13):4777�4788. 594 https://doi.org/10.1093/nar/13.13.4777 595 Desai HV, Mehta AR. 1985. Changes in Polyamine Levels during Shoot Formation, Root 596 Formation, and Callus Induction in Cultured _Passiflora_ Leaf Discs. Journal of Plant 597 Physiology. 119(1):45�53. https://doi.org/10.1016/S0176-1617(85)80214-5 598 Dever TE, Ivanov IP, Sachs MS. 2020. Conserved Upstream Open Reading Frame Nascent 599 Peptides That Control Translation. Annual Review of Genetics. 54:237�264. 600 https://doi.org/10.1146/annurev-genet-112618-043822 601 Fallon KM, Phillips R. 1988. Polyamines in Relation to Growth in Carrot Cell Cultures. Plant 602 Physiol. 88(1):224�227. https://doi.org/10.1104/pp.88.1.224 603 Foster SD, Glover SR, Turner AN, Chatti K, Challa AK. 2019. A mixing heteroduplex 604 mobility assay (mHMA) to genotype homozygous mutants with small indels generated by 605 CRISPR-Cas9 nucleases. MethodsX. 6:1�5. https://doi.org/10.1016/j.mex.2018.11.017 606 Franceschetti M, Hanfrey C, Scaramagli S, Torrigiani P, Bagni N, Burtin D, Michael AJ. 607 2001. Characterization of monocot and dicot plant S-adenosyl-l-methionine decarboxylase 608 gene families including identification in the mRNA of a highly conserved pair of upstream 609 overlapping open reading frames. Biochem J. 353(Pt 2):403�409. 610 https://doi.org/10.1042/0264-6021:3530403 611 Gaba A, Jacobson A, Sachs MS. 2005. Ribosome occupancy of the yeast CPA1 upstream 612 open reading frame termination codon modulates nonsense-mediated mRNA decay. 613 Molecular Cell. 20(3):449�460. https://doi.org/10.1016/j.molcel.2005.09.019 614 Hanfrey C, Elliott KA, Franceschetti M, Mayer MJ, Illingworth C, Michael AJ. 2005. A dual 615 upstream open reading frame-based autoregulatory circuit controlling polyamine-responsive 616 translation. Journal of Biological Chemistry. 280(47):39229�39237. 617 https://doi.org/10.1074/jbc.M509340200 618 Hanfrey C, Franceschetti M, Mayer MJ, Illingworth C, Michael AJ. 2002. Abrogation of 619 Upstream Open Reading Frame-mediated Translational Control of a Plant S620 Adenosylmethionine Decarboxylase Results in Polyamine Disruption and Growth 621 Perturbations. Journal of Biological Chemistry. 277(46):44131�44139. 622 https://doi.org/10.1074/jbc.M206161200 623 Hanfrey C, Sommer S, Mayer MJ, Burtin D, Michael AJ. 2001. Arabidopsis polyamine 624 biosynthesis: Absence of ornithine decarboxylase and the mechanism of arginine 27 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 625 decarboxylase activity. Plant Journal. 27(6):551�560. https://doi.org/10.1046/j.1365626 313X.2001.01100.x 627 Hayashi N, Sasaki S, Takahashi H, Yamashita Y, Naito S, Onouchi H. 2017. Identification of 628 Arabidopsis thaliana upstream open reading frames encoding peptide sequences that cause 629 ribosomal arrest. Nucleic Acids Res. 45(15):8844�8858. https://doi.org/10.1093/nar/gkx528 630 Hieno A, Naznin HA, Hyakumachi M, Sakurai T, Tokizawa M, Koyama H, Sato N, 631 Nishiyama T, Hasebe M, Zimmer AD, et al. 2014. ppdb: plant promoter database version 3.0. 632 Nucleic Acids Research. 42(D1):D1188�D1192. https://doi.org/10.1093/nar/gkt1027 633 F₁==6×W⁷⁵T –q, )†7=−† )µ, ¹−=6=×6≥’ Ps lotxs %≥7=⁷I7ₙ₁−=7I ⁵−=ₙ≥−I ×� (�-untranslated 634 regions of eukaryotic mRNAs. Science. 352(6292):1413�1416. 635 https://doi.org/10.1126/science.aad9868 636 Hiragori Y, Takahashi H, Karino T, Kaido A, Hayashi N, Sasaki S, Nakao K, Motomura T, 637 Yamashita Y, Naito S, Onouchi H. 2023. Genome-wide identification of Arabidopsis non638 AUG-initiated upstream ORFs with evolutionarily conserved regulatory sequences that 639 control protein expression levels. Plant Mol Biol. 111:37�55. https://doi.org/10.1007/s11103640 022-01309-1 641 Ho SN, Hunt HD, Horton RM, Pullen JK, Pease LR. 1989. Site-directed mutagenesis by 642 overlap extension using the polymerase chain reaction. Gene. 77(1):51�59. 643 https://doi.org/10.1016/0378-1119(89)90358-2 644 Hummel I, Bourdais G, Gouesbet G, Couée I, Malmberg RL, El Amrani A. 2004. Differential 645 gene expression of ARGININE DECARBOXYLASE ADC1 and ADC2 in Arabidopsis 646 thaliana: characterization of transcriptional regulation during seed germination and seedling 647 development. New Phytologist. 163(3):519�531. https://doi.org/10.1111/j.1469648 8137.2004.01128.x 649 Igarashi K, Kashiwagi K. 2010. Modulation of cellular function by polyamines. The 650 International Journal of Biochemistry & Cell Biology. 42(1):39�51. 651 https://doi.org/10.1016/j.biocel.2009.07.009 652 Iwase A, Mita K, Favero DS, Mitsuda N, Sasaki R, Kobayshi M, Takebayashi Y, Kojima M, 653 Kusano M, Oikawa A, et al. 2018. WIND1 induces dynamic metabolomic reprogramming 654 during regeneration in _Brassica napus_ . Developmental Biology. 442(1):40�52. 655 https://doi.org/10.1016/j.ydbio.2018.07.006 656 Janowitz T, Kneifel H, Piotrowski M. 2003. Identification and characterization of plant 657 agmatine iminohydrolase, the last missing link in polyamine biosynthesis of plants. FEBS 658 Letters. 544(1�3):258�261. https://doi.org/10.1016/S0014-5793(03)00515-5 659 Kakehi J, Kuwashiro Y, Niitsu M, Takahashi T. 2008. Thermospermine is Required for Stem 660 Elongation in Arabidopsis thaliana. Plant and Cell Physiology. 49(9):1342�1349. 661 https://doi.org/10.1093/pcp/pcn109 662 Kozak M. 1990. Downstream secondary structure facilitates recognition of initiator codons by 663 eukaryotic ribosomes. Proceedings of the National Academy of Sciences of the United States 664 of America. 87(21):8301�8305. https://doi.org/10.1073/pnas.87.21.8301 28 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 665 Kusano T, Berberich T, Tateda C, Takahashi Y. 2008. Polyamines: essential factors for 666 growth and survival. Planta. 228(3):367�381. https://doi.org/10.1007/s00425-008-0772-7 667 Law GL, Raney A, Heusner C, Morris DR. 2001. Polyamine Regulation of Ribosome Pausing 668 at the Upstream Open Reading Frame of S-Adenosylmethionine Decarboxylase. Journal of 669 Biological Chemistry. 276(41):38036�38043. https://doi.org/10.1074/jbc.M105944200 670 Masgrau C, Altabella T, Farrás R, Flores D, Thompson AJ, Besford RT, Tiburcio AF. 1997. 671 Inducible overexpression of oat arginine decarboxylase in transgenic tobacco plants. The 672 Plant Journal. 11(3):465�473. https://doi.org/10.1046/j.1365-313X.1997.11030465.x 673 Menges M, Murray JAH. 2002. Synchronous Arabidopsis suspension cultures for analysis of 674 cell-cycle gene activity. Plant Journal. 30(2):203�212. https://doi.org/10.1046/j.1365675 313X.2002.01274.x 676 Michael AJ, Furze JM, Rhodes MJ, Burtin D. 1996. Molecular cloning and functional 677 identification of a plant ornithine decarboxylase cDNA. Biochem J. 314(Pt 1):241�248. 678 https://doi.org/10.1042/bj3140241 679 Motohashi K. 2017. Seamless Ligation Cloning Extract (SLiCE) Method Using Cell Lysates 680 from Laboratory Escherichia coli Strains and its Application to SLiP Site-Directed 681 Mutagenesis BT - In Vitro Mutagenesis: Methods and Protocols.Reeves A, editor. Methods 682 Mol Biol. 1498:349�357. https://doi.org/10.1007/978-1-4939-6472-7_23 683 Murillo E, Martínez-Seidel F, Atanasov KE, Gentry-Torfer D, Pereira Firmino AA, Erban A, 684 Nie S, Leeming MG, Suwanchaikasem P, Boughton BA, et al. 2025. Polyamines and flg22 685 reshape the ribosomal protein composition of actively translating ribosomes in plants. Plant 686 Physiology and Biochemistry. 220:109585. https://doi.org/10.1016/j.plaphy.2025.109585 687 Nyikó T, Sonkoly B, Mérai Z, Benkovics AH, Silhavy D. 2009. Plant upstream ORFs can 688 trigger nonsense-mediated mRNA decay in a size-dependent manner. Plant Mol Biol. 71(4� 689 5):367�378. https://doi.org/10.1007/s11103-009-9528-4 690 Patel J, Ariyaratne M, Ahmed S, Ge L, Phuntumart V, Kalinoski A, Morris PF. 2017. Dual 691 functioning of plant arginases provides a third route for putrescine synthesis. Plant Science. 692 262(January):62�73. https://doi.org/10.1016/j.plantsci.2017.05.011 693 Piotrowski M, Janowitz T, Kneifel H. 2003. Plant C-N Hydrolases and the Identification of a 694 Plant _N_ -Carbamoylputrescine Amidohydrolase Involved in Polyamine Biosynthesis. Journal 695 of Biological Chemistry. 278(3):1708�1712. https://doi.org/10.1074/jbc.M205699200 696 Raney A, Law GL, Mize GJ, Morris DR. 2002. Regulated Translation Termination at the 697 Upstream Open Reading Frame in _S_ -Adenosylmethionine Decarboxylase mRNA. Journal of 698 Biological Chemistry. 277(8):5988�5994. https://doi.org/10.1074/jbc.M108375200 699 Rossi FR, Marina M, Pieckenstain FL. 2015. Role of Arginine decarboxylase (ADC) in 700 Arabidopsis thaliana defence against the pathogenic bacterium Pseudomonas viridiflava. Plant 701 Biology. 17(4):831�839. https://doi.org/10.1111/plb.12289 702 Ruan H, Shantz LM, Pegg AE, Morris DR. 1996. The Upstream Open Reading Frame of the 703 mRNA Encoding S-Adenosylmethionine Decarboxylase Is a Polyamine-responsive 29 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 704 Translational Control Element *. Journal of Biological Chemistry. 271(47):29576�29582. 705 https://doi.org/10.1074/jbc.271.47.29576 706 Sato K, Akiyama M, Sakakibara Y. 2021. RNA secondary structure prediction using deep 707 learning with thermodynamic integration. Nat Commun. 12(1):941. 708 https://doi.org/10.1038/s41467-021-21194-4 709 Tsutsui H, Higashiyama T. 2017. pKAMA-ITACHI Vectors for Highly Efficient 710 CRISPR/Cas9-Mediated Gene Knockout in Arabidopsis thaliana. Plant and Cell Physiology. 711 58(1):46�56. https://doi.org/10.1093/pcp/pcw191 712 Uchiyama-Kadokura N, Murakami K, Takemoto M, Koyanagi N, Murota K, Naito S, 713 Onouchi H. 2014. Polyamine-responsive ribosomal arrest at the stop codon of an upstream 714 open reading frame of the AdoMetDC1 gene triggers nonsense-mediated mRNA decay in 715 Arabidopsis thaliana. Plant and Cell Physiology. 55(9):1556�1567. 716 https://doi.org/10.1093/pcp/pcu086 717 Urano K, Yoshiba Y, Nanjo T, Ito T, Yamaguchi-Shinozaki K, Shinozaki K. 2004. 718 Arabidopsis stress-inducible gene for arginine decarboxylase AtADC2 is required for 719 accumulation of putrescine in salt tolerance. Biochemical and Biophysical Research 720 Communications. 313(2):369�375. https://doi.org/10.1016/j.bbrc.2003.11.119 721 Wang J, Shin B-S, Alvarado C, Kim J-R, Bohlen J, Dever TE, Puglisi JD. 2022. Rapid 40S 722 scanning and its regulation by mRNA structure during eukaryotic translation initiation. Cell. 723 185(24):4474-4487.e17. https://doi.org/10.1016/j.cell.2022.10.005 724 Wu D, von Roepenack-Lahaye E, Buntru M, de Lange O, Schandry N, Pérez-Quintero AL, 725 Weinberg Z, Lowe-Power TM, Szurek B, Michael AJ, et al. 2019. A Plant Pathogen Type III 726 Effector Protein Subverts Translational Regulation to Boost Host Polyamine Levels. Cell 727 Host and Microbe. 26(5):638-649.e5. https://doi.org/10.1016/j.chom.2019.09.014 728 Wu F-H, Shen S-C, Lee L-Y, Lee S-H, Chan M-T, Lin C-S. 2009. Tape-Arabidopsis 729 Sandwich - a simpler Arabidopsis protoplast isolation method. Plant Methods. 5(1):16. 730 https://doi.org/10.1186/1746-4811-5-16 731 Yariuchi Y, Okamoto T, Noutoshi Y, Takahashi T. 2021. Responses of Polyamine-Metabolic 732 Genes to Polyamines and Plant Stress Hormones in Arabidopsis Seedlings. Cells. 733 10(12):3283. https://doi.org/10.3390/cells10123283 734 ## 735 Abbreviations 736 (�-/%², (� W=ₙ≥7=⁷I7ₙ6O ≥6’₁−=“ g(¹9≥−, cauliflower mosaic virus 35S RNA promoter; ACL5, 737 ACAULIS5; ADC, arginine decarboxylase; ADC1, arginine decarboxylase 1; ADC2, arginine 738 decarboxylase 2; Arg, L-arginine; Col-0, Columbia-0; αRG–, �-difluoromethylarginine; Eluc, 739 emerald luciferase; gRNA, guide RNA; HPLC, high-performance liquid chromatography; 30 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 740 NMD, nonsense-mediated decay; ODC, ornithine decarboxylase; Orn, L-ornithine; Put, 741 putrescine; Rluc, Renilla luciferase; RT-qPCR, reverse transcription quantitative PCR; 742 SAMDC, _S_ -adenosylmethionine decarboxylase; Spd, spermidine; Spm, spermine; TE, 743 translation efficiency; Tspm, thermospermine; uORF, upstream open reading frame; WT, wild 744 type. 745 31 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) # 746 LEGENDS ## 747 Figure 1. Polyamine biosynthetic pathways relevant to Put homeostasis in 748 Arabidopsis. 749 Put is produced from Arg through arginine decarboxylase (ADC), followed by agmatine 750 iminohydrolase (AIH) and _N_ -carbamoylputrescine amidohydrolase (CPA), or by the 751 agmatinase activity of arginase (ARGAH). ARGAH also converts Arg to Orn. The dashed 752 ornithine decarboxylase (ODC) route is shown for comparison; Arabidopsis lacks a functional 753 ODC pathway. Put is converted to Spd, which is used to synthesize Spm or Tspm. SPDS, 754 spermidine synthase; SPMS, spermine synthase; TSPMS, thermospermine synthase. _S_ 755 adenosylmethionine decarboxylase (SAMDC) generates decarboxylated _S_ -adenosylmethionine 756 (dcSAM), the aminopropyl donor, from _S_ -adenosylmethionine (SAM). 757 ## 758 Figure 2. The ADC2 ��-UTR mediates reporter responses to Put supply and 759 inhibition of Put biosynthesis. 760 (A) Schematic representations of reporter constructs. _ADC2_ (�-UTR: _Eluc_ (upper) contains the 761 _ADC2_ (�-UTR (thick blue line) and the first 5 nt of its main ORF (blue box) followed by _Eluc_ 762 PEST (gray box). The control construct (lower) contains a short unrelated sequence in place of 763 the _ADC2_ -derived sequence. Both reporters are driven by 35Spro (white arrows). (B) Relative 764 Eluc activity in mesophyll protoplasts carrying the _ADC2_ (�-UTR: _Eluc_ plasmid after 24 h of 765 treatment with 0�4 mM Put (n = 4 per concentration). The line was fitted by least-squares 766 regression; Pearson�s correlation coefficient ( _r_ ) and _P_ value are _r_ D �osmt 7=O _P_ D ts( � to��, 767 respectively. (C) Relative Eluc activity from the _ADC2_ (�-UTR: _Eluc_ and control plasmids in 768 mesophyll protoplasts treated with 0 or 2 mM Put (n = 4 per group). Note the discontinuous y 769 axis. (D) Relative activity of the _ADC2_ (�-UTR reporter in mesophyll protoplasts from Col-0, 32 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 770 _adc2-1_, and _adc2-2_, with 0 or 2 mM Put (n = 3 per group). (E) Chemical structure of DFMA 771 and schematic inhibition of ADC-dependent Put biosynthesis. (F) Relative activity of the _ADC2_ 772 (�-UTR reporter in MM2d suspension cell protoplasts treated with 0 or 1 mM Put and 0 or 1 773 mM DFMA (n = 3 per group). In all transient assays, protoplasts were incubated at 22°C for 24 774 h in the dark. Eluc activity was normalized to Rluc activity from the reference plasmid and 775 scaled to the mean of untreated _ADC2_ reporter control in B and C, untreated Col-0 in D, or the 776 no-addition control in F, within each experiment. Bars show means ± SD. In C, data were 777 compared between the 0 and 2 mM Put treatments within each construct using one-sided, 778 W=97₁≥6O ¹ₙWO6=ₙ�⁷ _t_ tests assuming equal variances, with Holm correction for the two 779 comparisons (***, adjusted _P_ < 0.001; n.s., not significant, adjusted _P_ � oso(.s In D and F, 780 different letters indicate significant differences among groups within each panel (two-way 781 ANOVA followed by Tukey�s test, _P_ < 0.05). 782 ## 783 Figure 3. Put reduces reporter activity, mRNA abundance, and TE in stable 784 ADC2 ��-UTR reporter lines. 785 (A) Schematic representations of the reporter constructs used for transgenic reporter assays. 786 These constructs contain the (�-UTR and the first 5 nt of the main ORF from either _ADC2_ (blue 787 thick line and box) or _MASP1_ (black thick line and box) upstream of _Eluc_ -PEST (gray box), 788 driven by the 35Spro (white arrow). (B) Relative Eluc activity in two independent _ADC2_ 789 reporter lines ( _ADC2_ #1 and _ADC2_ #2) and the _MASP1_ control line ( _MASP1_ #1). Ten-day-old 790 seedlings were transferred to liquid half-strength MS medium and treated with 0 or 2 mM Put 791 for 24 h. Eluc activity was normalized to total protein. (C) Relative _Eluc_ mRNA abundance, 792 normalized to _����_ mRNA in the same cDNA sample. (D) Relative Eluc TE, calculated for 793 each biological replicate as relative Eluc activity divided by relative _Eluc_ mRNA abundance 794 measured in the same sample. Values are expressed relative to the untreated condition for each 33 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 795 line. Bars show means ± SD of three biological replicates, each comprising 15 seedlings pooled 796 from one flask. Treated and untreated samples were compared within each line using one-sided, 797 unpaired ¹ₙWO6=ₙ�⁷ _t_ tests assuming equal variances. *, _P_ < 0.05; **, _P_ < 0.01; ***, _P_ < 0.001; 798 n.s., _P_ � oso(s 799 ## 800 Figure 4. Mutational analysis of the AUG- and non-AUG-initiated uORFs 801 in the ADC2 ��-UTR. 802 (A) Organization of the _ADC2_ (�-UTR. The short AUG-initiated uORF is shown in red, the 803 AUU/CUG-initiated uORF in orange, and the ADC-box in blue. Positions are shown relative 804 ₙ− ₙT6 ;7₁= 0²R ⁷ₙ7≥ₙs ;�q, (� ⁵79“ e–.�, 9−I�e–. ₙ7₁Is e1. 2%, �–/q, �–//, �M/q, 7=O 805 �–//�M/q ≥69−≥ₙ6≥ ⁵−=⁷ₙ≥W⁵ₙ⁷s %T6 αP– ⁷ₙ7≥ₙ ⁵−O−= –%q ‡7⁷ ⁵T7=’6O ₙ− ––– ₁= �–/q“ 806 ATT was changed to GTT and CTG to CTC in the non-AUG mutants, individually or together. 807 Red letters indicate substitutions. Schematics use DNA notation, whereas construct names refer 808 to RNA codons. (C, E) Relative Eluc activity in untreated mesophyll protoplasts, with WT set 809 ₙ− t, 3−≥ �–/q eM. 7=O ₙT6 =−=-AUG mutants (E). (D, F) Relative Eluc activity after treatment 810 with 0 or 2 mM Put, with the untreated value set to 1 for each construct. Assays were incubated 811 at 22°C for 24 h in the dark. Eluc activity was normalized to Rluc activity from the reference 812 plasmid before fold normalization. Bars show means ± SD (WT, n = 4; each mutant, n = 3 per 813 treatment). One-sided, unpaired ¹ₙWO6=ₙ�⁷ _t_ tests assuming equal variances were used for C; the 814 same tests with Holm correction within each graph were used for D�F. Comparisons were with 815 WT in C and E and between 0 and 2 mM Put within each construct in D and F. *, _P_ < 0.05; **, 816 _P_ < 0.01; ***, _P_ < 0.001; n.s., _P_ � oso(s 817 34 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) ## 818 Figure 5. Effects of disrupting and restoring the predicted RNA structure in 819 the ADC-box region on Put-responsive repression. 820 (A) Reporter design and WT, unpaired, and paired ADC-box sequences. Substituted nucleotides 821 are shown in red, and connecting lines indicate predicted base-paired segments. The mutations 822 overlap the non-AUG-initiated uORF. (B) RNA secondary structures predicted using 823 MXFold2. Blue arrows mark the boundaries of the ADC-box; base colors distinguish nucleotide 824 identities, and red outlines mark substituted nucleotides. (C) Relative Eluc activity under 825 untreated conditions, with WT set to 1. (D) Relative Eluc activity after treatment with 0 or 2 826 mM Put, with the untreated condition set to 1 for each construct. Reporters were tested in 827 mesophyll protoplasts incubated at 22°C for 24 h in the dark. Eluc activity was normalized to 828 Rluc activity from the reference plasmid before scaling to the control-group mean within each 829 experiment. Bars show means ± SD (n = 3 per group). One-sided, unpaired ¹ₙWO6=ₙ�⁷ _t_ tests 830 assuming equal variances were used with Holm correction within each graph: each mutant was 831 compared with WT in C, and treated and untreated samples were compared within each 832 construct in D. *, _P_ < 0.05; **, _P_ < 0.01; ***, _P_ < 0.001; n.s., _P_ � oso(s 833 ## 834 Figure 6. ADC2 ��-UTR genome-edited lines exhibit growth defects, and 835 #11-8 accumulates Put. 836 (A) Organization of the WT and edited _ADC2_ (�-UTRs. Lines #3-7 and #11-8 contain 25- and 837 31-nt deletions, respectively, overlapping with the non-AUG-initiated uORF and ADC-box. 838 Codons are shown in DNA notation. (B) Seedlings of Col-0, #3-7, and #11-8 grown vertically 839 on half-strength MS medium solidified with gellan gum, photographed on day 10 after 840 stratification. Scale bar, 1 cm. (C) Shoot fresh weight and (D) primary root length measured on 841 day 10. Seedlings that had germinated by day 2 were transferred on day 5 to three fresh plates, 842 with five seedlings per genotype per plate (n = 15 per genotype). Points represent individual 35 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 843 seedlings; thick horizontal lines indicate means and thin lines indicate SD ranges. Different 844 letters indicate significant differences among genotypes (one-way ANOVA followed by 845 Tukey�s test, _P_ < 0.05). (E) Put, Spd, and combined Spm+Tspm abundance in Col-0 and #11 846 8, determined by HPLC in 14-day-old whole seedlings grown on half-strength MS agar medium 847 and expressed per gram fresh weight (F.W.). The combined Spm+Tspm peak was quantified 848 using the Spm calibration curve. Bars show means ± SD of three samples per genotype, each 849 comprising approximately 20 seedlings pooled from one plate (approximately 100 mg). 850 Comparisons with Col-0 used one-sided, unpaired ¹ₙWO6=ₙ�⁷ _t_ tests assuming equal variances, 851 with Holm correction across the comparisons in E. *, _P_ < 0.05; **, _P_ < 0.01; ***, _P_ < 0.001; 852 n.s., _P_ � oso(s 853 ## 854 Supplemental Figure S1. T-DNA insertion alleles of ADC2 . 855 (A) Locations of the T-DNA insertions in _adc2-1_ (SALK_073977) and _adc2-2_ 856 (WiscDsLoxHs065_09H). Black arrowheads indicate the positions and orientations of the 857 primers used for RT-qPCR. (B) Relative _ADC2_ mRNA levels measured by RT-qPCR in the 858 indicated genotypes. Each biological replicate comprised fresh rosette leaves from a 4-week 859 old plant. _ADC2_ transcript abundance was normalized to _����_ and expressed relative to Col 860 0. The bar and error bars for Col-0 represent the mean ± SD (n = 3). In _adc2-1_ e�., −=I� −=6 −3 861 the three biological replicates yielded a quantifiable value, shown as a single dot, whereas the 862 remaining two showed target-specific amplification below the lowest point of the standard 863 curve. In _adc2-2_ e�., ₙ‡− ≥69I₁⁵7ₙ6⁷ ⁷T−‡6O ₙ7≥’6ₙ-specific amplification below the lowest point 864 of the standard curve, while only non-specific amplification was detected in the remaining 865 replicate. Thus, no quantifiable values were obtained for _adc2-2_ . 866 36 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) ## 867 Supplemental Figure S2. Sequence and regulatory elements of the ADC2 ��- 868 UTR. 869 The nucleotide sequence is shown in DNA notation, with positions numbered from the first 870 nucleotide. The first nucleotide shown in red marks the transcription start site (TSS) peak 871 annotated in Plant Promoter Database 3.0 (http://ppdb.agr.gifu-u.ac.jp; Hieno et al. 2014). The 872 short AUG-initiated uORF is highlighted in red and the AUU/CUG-initiated uORF in orange. 873 The ADC-box is underlined in blue. Boxes mark the uORF initiation codons; the beginning of 874 the main ORF is highlighted in dark gray. 875 ## 876 Supplemental Figure S3. Predicted RNA secondary structures of non-AUG 877 start-codon mutants. 878 P≥6O₁⁵ₙ6O ⁷ₙ≥W⁵ₙW≥6⁷ 7≥6 ⁷T−‡= 3−≥ 2%, �–//, �M/q, 7=O �–//�M/q ⁷6⁸W6=⁵6⁷ 879 encompassing the ADC-box region and 100 nt on each side, calculated and drawn using 880 MXFold2 Server with default settings. Blue arrows mark the ADC-box boundaries, base colors 881 distinguish nucleotide identities, and red outlines mark substituted nucleotides. 882 ## 883 Supplemental Figure S4. Deletion sequences and predicted RNA structures 884 of ADC2 ��-UTR genome-edited alleles. 885 (A) Alignment of the WT region with the 25-nt deletion in #3-7 and the 31-nt deletion in #11 886 8. Orange shading marks the uORFs; the frameshifted mutant uORFs terminate 17 nt 887 downstream of the WT stop position. Black and gray boxes indicate guide RNA target regions 888 and protospacer adjacent motifs (PAMs), respectively. Blue letters mark the ADC-box, red 889 dashed segments indicate deleted nucleotides, and black underlines and connecting lines 890 indicate predicted base pairing. (B) Structures predicted using MXFold2 for the WT and edited 37 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 891 ADC-box regions with 100 nt on each side. Blue arrows mark the ADC-box boundaries, and 892 base colors distinguish nucleotide identities. Red triangles mark deletion sites, with the numbers 893 indicating deleted lengths. 894 ## 895 Supplemental Figure S5. Germination and cotyledon expansion of ADC2 ��- 896 UTR genome-edited lines. 897 Germination scored by radicle emergence (A) and cotyledon expansion (B) in Col-0, #3-7, and 898 #11-8 over 7 d after 6 d of stratification. Seeds were grown on half-strength MS agar medium 899 at 22°C under continuous LED illumination (75�too �;−I ;�� ⁷��.s q6≥;₁=7ₙ₁−= ‡7⁷ ⁷⁵−≥6O 900 when the seed coat had split and the radicle protruded; cotyledon expansion was scored when 901 the cotyledons were fully exposed and separated. Percentages were calculated using all seeds 902 sown in one plate as the denominator. Each plate contained 76 seeds per genotype, and each 903 plate-level proportion was one replicate (n = 3 plates). Values are means ± SD. 904 ## 905 Supplemental Figure S6. HPLC chromatograms and calibration curves for 906 polyamine quantification. 907 Chromatograms are shown for mixed Put, Spd, and Spm standards and for three samples each 908 of Col-0 and #11-8, labeled rep1�rep3, corresponding to the analysis in Fig. 6E. Each sample 909 comprised approximately 20 whole seedlings pooled from one plate at 14 d of growth. 910 Dansylated polyamines were detected by UV absorbance at 340 nm. Peaks are designated 0, 911 1,7-diaminoheptane (internal standard); 1, Put; 2, Spd; and 3, Spm in the standards and the 912 unresolved Spm+Tspm fraction in plant samples. Relative peak area is the polyamine peak area 913 divided by the internal-standard peak area. Calibration plots show relative peak area against 914 concentration for the five nonzero standard concentrations, with fitted equations and 38 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) 915 coefficients of determination displayed in the figure. All sample relative peak areas were within 916 the calibration ranges. Std, standard. 917 ## 918 Supplemental Table S1. Primers used for mutant genotyping. 919 Primer names, targets, and sequences are listed for the _ADC2_ genome-edited region and the two 920 T-DNA insertion lines. The gene-specific pairs SALK_073977 F/R and 921 WiscDsLoxHs065_09H F/R were used together with gene-specific/T-DNA border primer 922 combinations; LBb1.3 and LB4 modified were used for SALK_073977 and 923 WiscDsLoxHs065_09H, respectively. 924 ## 925 Supplemental Table S2. Plasmids used in transient assays and primers used 926 for plasmid construction. 927 PI7⁷;₁O ₁O6=ₙ₁3₁6≥⁷, ≥69−≥ₙ6≥ O6⁷₁’=⁷, 7=O ⁵I−=₁=’ 9≥₁;6≥ =7;6⁷ 7≥6 I₁⁷ₙ6Os �0≥₁’₁=7I 9I7⁷;₁O� 928 identifies the vector backbone used for insertion. All listed plasmids are based on pUC19. 929 Primer sequences are provided in Supplemental Table S4. 930 ## 931 Supplemental Table S3. Plasmids used for generation of transgenic or 932 genome-edited lines and oligo DNA used for plasmid construction. 933 _ADC2_ and _MASP1_ reporter constructs for transformation and CRISPR-Cas9 guide constructs 934 for genome editing are listed with their vector and oligonucleotide identifiers. Guide RNA 935 cloning oligonucleotide sequences are provided in Supplemental Table S5. 936 39 [bioRxiv preprint doi: https://doi.org/10.64898/2026.10.31.763274; this version posted November 2, 2026. The copyright holder for this](https://doi.org/10.64898/2026.10.31.763274) 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 4.0 International license.](http://creativecommons.org/licenses/by/4.0/) ## 937 Supplemental Table S4. Primers used for plasmid construction. 938 µ≥₁;6≥ ⁷6⁸W6=⁵6⁷ 7≥6 ’₁†6= ₁= ₙT6 (�-to-g� O₁≥6⁵ₙ₁−=s µ≥₁;6≥ =7;6⁷ ⁵−=ₙ7₁=₁=’ �¹µzl� 7≥6 939 retained to match the original laboratory records. 940 ## 941 Supplemental Table S5. Oligo DNA used for guide RNA cloning. 942 S6⁸W6=⁵6⁷ −3 ¹µzl ’²t Rh² 7=O ¹µzl ’²l Rh² 7≥6 I₁⁷ₙ6O ₁= ₙT6 (�-to-g� O₁≥6⁵ₙ₁−=s z7⁵T 943 oligonucleotide pair was annealed and ligated into _Aar_ I-digested pKI1.1R to generate pYH70 944 (gRNA1) or pYH71 (gRNA2), respectively (Supplemental Table S3). 945 ## 946 Supplemental Table S6. Primers used for RT-qPCR. 947 Primer names, target genes, and sequences are listed for _Eluc_, _ADC2_, and _����_ (At5g60390). 948 _����_ was used as the reference transcript for _Eluc_ mRNA measurements in the stable reporter 949 lines and for _ADC2_ mRNA measurements in Supplemental Fig. S1B. 40 SAM SAMDC dcSAM L-ornithine ~~ODC~~ N-carbamoylputrescine ARGAH L-arginine **ADC** agmatine AIH ARGAH CPA **Put** dcSAM SPDS **Spd** dcSAM dcSAM SPMS TSPMS **Spm** **Tspm** **Figure 1** ## **A** 250 200 150 100 1.00 _**ADC2**_ **5′-UTR:** _**Eluc**_ **Control** ## **B** 1.00 0.75 0.50 0.25 0.00 _ADC2_ 5′-UTR 35Spro _Eluc_ -PEST ## **C** Put 0 mM 2 mM n.s. a 0 1 2 3 4 Put [mM] ## **D** ## **E** 0.75 0.50 0.25 0.00 DFMA L-Arg **[ADC]** **Put** ## **F** 2.0 1.5 1.0 0.5 0.0 Col−0 _adc2−1_ _adc2−2_ 3 2 1 0 Put [mM] DFMA [mM] 0 1 0 1 0 0 1 1 **Figure 2** ## **A** **B** _**ADC2**_ **5′-UTR** _**MASP1**_ **5′-UTR** _**ADC2**_ **#1** _**ADC2**_ **#2** _**MASP1**_ **#1** 1.0 0.5 1.0 0.5 0.0 0 mM 2 mM Put 0.0 0 mM 2 mM Put 1.0 0.5 0.0 0 mM 2 mM Put ## **C ADC2 #1 ADC2 #2 MASP1 #1** n.s. 1.0 0.5 1.0 0.5 0.0 0 mM 2 mM Put 0.0 0 mM 2 mM Put 1.0 0.5 0.0 0 mM 2 mM Put ## **D** _**ADC2**_ **#1** _**ADC2**_ **#2** _**MASP1**_ **#1** 1.0 0.5 1.0 0.5 1.0 0.5 0.0 0 mM 2 mM Put 0.0 0 mM 2 mM Put 0.0 0 mM 2 mM Put **Figure 3** ## **A** AUG AUU CUG m [7] G _ADC2_ (A) n ADC-box -500 -400 -300 -200 -100 0 [nt] ## B ATT |Col1|Col2|CT|Col4|Col5|Col6| |---|---|---|---|---|---| |35Spro|||||_Eluc_-PEST| |35Spro|||||| AAA |35Spro|Col2|Col3|Col4|Eluc-PEST| |---|---|---|---|---| |35Spro||||| GTT |Col1|Col2|CT|Col4|Col5| |---|---|---|---|---| |35Spro||||_Eluc_-PEST| |35Spro||||| ATT |Col1|Col2|CT|Col4|Col5| |---|---|---|---|---| |35Spro||||_Eluc_-PEST| |35Spro||||| GTT **∆AUU∆CUG** 35Spro |GTT|Col2|CTC T|Col4|Col5| |---|---|---|---|---| |35Spro|35Spro|CTC<br>_Eluc_-PEST|CTC<br>_Eluc_-PEST|CTC<br>_Eluc_-PEST| |35Spro|35Spro|CTC<br>_Eluc_-PEST|CTC<br>_Eluc_-PEST|_Eluc_-PEST| |35Spro||||| ## **C D** 2.5 1.5 1.0 0.5 0.0 1.5 1.0 Put 0 mM 2 mM 2.0 1.5 1.0 0.5 0.0 WT ∆AUG ## **E F** 5 4 3 2 Put 0 mM 2 mM WT ∆AUG 1 0 0.5 0.0 **Figure 4** ## **A** _ADC2_ 5′-UTR CTG WT CTCCTGAGGGGGTAGCCGAGGCTCCGGCCTCGGCGGTTTTTAAACCCCTACCT 0.0 WT unpaired paired **Figure 5** ## **B** ## **C** **unpaired** **paired** 10.0 7.5 5.0 2.5 0.0 CTCCTGAGGGGGTAGCCGAGGCTCCGCGGAGCCCGGTTTTTAAAGGGGTACCT CTCCTGACCCCGTAGGGCTCCGTCCGCGGAGCCCGGTTTTTAAAGGGGTACCT **WT** **unpaired** **paired** ## **D** 1.0 0.5 Put 0 mM 2 mM CTG ATT ADC-box ## **A** ATG _ADC2_ Col-0 |Col1|Col2| |---|---| ||| 25 nt 31 nt ## B Col-0 #3-7 #11-8 _ADC2_ _adc2_ 5′-UTR #3-7 _ADC2_ _adc2_ 5′-UTR #11-8 Put Spd Spm+Tspm Col−0 #11−8 ## **C** ## **D** c **Figure 6** a 6 4 2 0 a b b c 8 6 4 2 0 ## **E** 200 150 100 50 0