Seeds of A. thaliana, ecotype Columbia-0 (Col-0) and of the transgenic lines COPT1^OE and COPT3^OE were surface-sterilized and stratified for 2 days at 4°C and were germinated in ½ MS medium (Sigma) plates including 1% sucrose (Murashige and Skoog, 1962) (½ MS) or supplemented with 10 μM CuSO₄ (½ MS + 10 Cu) for the microarray analysis. Seedlings were grown as previously described (Andrés-Colás et al., 2010) for 7 days with a 12 h neutral photoperiod (65 μmol m⁻² of cool-white fluorescent light) at 23°C/16°C temperature cycle.

In order to obtain ½ MS medium with the indicated concentrations of either Cu or Fe, the solution was prepared by adding macronutrients (Sigma) and micronutrients consisted in a mix of 50 μM H₃BO₃, 36.6 μM MnSO₄ H₂O, 15 μM ZnSO₄ 7H₂O, 0.57 μM NaMoO₄ 2H₂O, 0.25 mM KI, and 0.05 μM CoCl₂ 6H₂O. Finally, 0.05% MES, 1% sucrose, and 0.8% phytoagar was added, and the pH was adjusted to 5.7–5.8 with diluted KOH. The Cu concentration in the Cu deficiency (0 μM CuSO₄) media including commercial phytoagar (Duchefa Biochemie) measured by ICP-MS is <0.008 μM Cu. To study the effects that the different Cu and Fe content have in plants, 50 μM Fe-citrate and 1 μM CuSO₄ 5H₂O were added to the medium for Cu and Fe sufficiency conditions. Moreover, seedlings were grown in Cu deficiency (0 μM CuSO₄) and Cu excess (10 μM CuSO₄). On the other hand, Fe-sufficient and slight or severe Fe deficiency medium was supplemented with 50, 10 and 0 μM Fe-citrate, respectively. Other metal concentrations and treatments were used for specific experiments as indicated in the Supplementary Figure Legends.

The chlorophyll content in seedlings and leaves was determined by the trichlorometric method (Parsons and Strickland, 1965). Root length was measured using the Image J 1.42q software (http://rsb.info.nih.gov./ij). Values represent the arithmetic mean ± standard deviation (SD) of three biological replicates (n = 3).

For the determination of ferroreductase activity, three seedlings of 7-day-old were collected and weighed. Next, a 1:1 mixture made with 300 μM bathophenantroline disulfonate (BPDS, Sigma) and 100 μM Fe III-EDTA was added to the seedlings and incubated at 30°C with stirring 225 rpm in the dark. After 30 min, the solution was collected and absorbance A₅₃₅ was measured in a spectrophotometer (Grillet et al., 2014). Values represent the arithmetic mean ± standard deviation (SD) of three biological replicates (n = 3).

For root Fe³⁺ detection by Perl’s staining, four to five seedlings of 11-day-old were vacuum infiltrated with equal volumes of 4% (v/v) HCl and 4% (w/v) K- ferrocyanide (Perl´s stain solution for 15 min) and incubated at room temperature for 30 min (Stacey et al., 2008). One representative photograph is shown in the figure. This method is based on the Fe³⁺ dependent conversion of ferrocyanide into insoluble crystals of Prussian blue under acidic conditions. Localization of Fe³⁺ was observed and analyzed with a (Olympus CX41) microscope equipped with (Leyca MC170HD) camera and (LAS V4.10) software.

Seven-day-old seedlings of the WT and the COPT1^OE line were grown in the 12 h neutral photoperiod and three biological replicates were obtained for the (½ MS) treatment and four biological replicates were used as 10 μM CuSO₄, (½ MS + 10 Cu) samples. Total RNA was isolated using the RNeasy Plant Mini Kit (Qiagen) and aRNA was amplified using the MessageAmp™ II aRNA Amplification kit (Ambion). Long oligonucleotide microarrays were provided by Dr. David Galbraith (University of Arizona, http://www.ag.arizona.edu/microarray/). The hybridization and analysis were performed as described elsewhere (Bueso et al., 2007). The expression values (log₂) were obtained using the GenePix Pro 6.0 microarray-analysis software (Molecular Devices, Sunnyvale CA) and normalized with the GenePix Pro 6.0 and Acuity 4.0 software (Molecular Devices, Sunnyvale CA). Differential genes were identified with significance analysis of microarray (SAM) (Tusher et al., 2001) with false discovery rate (FDR) of <6% and 2-fold change (log₂ ≤|1|). Biological processes were identified with the Gene Ontology (GO) annotation (Ashburner et al., 2000), performed by the GeneCodis2.0 (http://genecodis.dacya.ucm.es/) (Carmona-Saez et al., 2007; Nogales-Cadenas et al., 2009) program ( Table 1 ). The total differentially regulated genes are shown as Supplementary material ( Supplementary Tables SI and SII ). The microarray raw data were deposited in the NCBI’s Gene Expression Omnibus (Edgar, 2002) and are accessible through GEO Series accession number GSE143857.

Total Arabidopsis RNA was isolated using the RNeasy Plant Mini Kit (Qiagen), was quantified by UV spectrophotometry and its integrity was visually assessed on ethidium bromide-stained agarose gels. After treatment with Dnase I Amp Grade (Invitrogen), cDNA was generated by retro-transcriptase SSII (Invitrogen) as previously described (Andrés-Colás et al., 2006). Real-time quantitative PCR (RT-qPCR) was carried out with SYBR-Green qPCR Super-Mix-UDG with ROX (Invitrogen) with the specific primers detailed in Table SVI in a CFX96 Touch™ Real Time PCR Detection System (BioRad), with one cycle of 95°C for 2 min and 40 cycles consisting in 95°C for 30 s and 60°C for 30 s. Expression values were normalized to UBQ10 and to the WT in deficiency conditions using the 2^{−ΔΔ Ct} method. Values represent the arithmetic mean ± standard deviation (SD) of three biological replicates (n = 3).

To study O₂ consumption, 14–16 roots of 10-day-old seedlings, grown in different conditions, were used. The roots were cut with a scalpel and resuspended in 1.5 ml of ½ MS liquid medium, as it is described above but without sucrose. The roots were transferred to an airtight chamber and the measurement of O₂ consumption for a minimum of 5 min was performed using a Clark type electrode (Oxyview system). The rate of decrease of O₂, referenced to fresh weight (F.W.) of the roots (nmol O₂/OD₆₀₀ × F.W.) was taken as an index of respiratory capacity. Values represent the arithmetic mean ± standard deviation (SD) of three biological replicates (n = 3).

Cu and Fe contents were determined by ICP-MS as described previously (Andrés-Colás et al., 2006; Carrió-Seguí et al., 2015) at the Servei Central de Suport a la Investigació Experimental (SCSIE) of the Universitat de València.

For the determination of the content of ABA, indol-3-acetic acid (IAA), and JA, 8-day-old seedlings were lyophilized, processed and analyzed by UHPLC (ultra-high-pressure liquid chromatography) Q-Exactive (ThermoFisher Scientific) as described previously (da Silva et al., 2017) in the plant hormone quantification service of the Institute of Molecular and Cellular Plant Biology (IBMCP, Valencia). Values represent the arithmetic mean ± standard deviation (SD) of three biological replicates (n = 3).

The statistical analysis of the relative gene expression was performed by the pair wise fixed reallocation randomization test (p-value <0.05) (Pfaffl, 2002). For the remaining parameters, the analysis was carried out using one or two-way ANOVA with the means compared by the Duncan test or a Kruskal–Wallis (p-value <0.05) test for a non-parametric measurements using the InfoStat software, version 2010 (http://www.infostat.com.ar) (Di Rienzo et al., 2011).

To identify at a molecular level the global effects caused by the deregulation of Cu homeostasis in Arabidopsis, we performed a comparative transcriptomic analysis of 7-day-old wild-type (WT) and a previously generated COPT1 overexpressing (COPT1^OE ) line (Andrés-Colás et al., 2010), grown under Cu deficiency (½ MS) and mild Cu excess (½ MS + 10 Cu) conditions. The induction of COPT1 expression under Cu deficiency in the WT was corroborated by RT-qPCR, as well as its overexpression in COPT1^OE seedlings, both under deficiency and excess Cu conditions, in the samples that were used in the hybridizations of the DNA microarrays (data not shown). From the global analysis of gene expression, a total of 583 differentially expressed genes were identified with a log₂ ratio of ≥|1| in the COPT1^OE versus WT ( Supplementary Figure S1 and Tables SI and SII ). These were distributed in a total of 482 induced (ratio ≥1) and of 101 repressed genes (ratio ≥−1) in COPT1^OE seedlings for the two growth conditions tested ( Supplementary Figure S1 and Supplementary Tables SI and SII ).

The number of genes induced was greater than that of genes repressed in the two conditions. On the other hand, the mild Cu excess condition showed the greatest number of genes with differential expression compared to Cu deficiency, with 393 genes (312 induced and 81 repressed) compared to 160 genes (142 induced and 18 repressed), respectively. The analysis of the gene ontology (GO) of differentially regulated genes between WT and COPT1^OE seedlings indicated that several the Biological Processes categories were overrepresented, including those linked to photosynthesis, photomorphogenesis, transport of metals such as Fe, Zn, manganese, cadmium, and phosphate, oxidative stress and responses to abscisic acid (ABA) and cold stress ( Supplementary Table SIII ). On the other hand, the Molecular Function of genes differentially regulated in COPT1^OE included the chlorophyll and tetrapyrrole binding ( Supplementary Table SIV ). GO analysis also revealed that the chloroplast was the subcellular compartment most affected by the changes observed in COPT1^OE , as 39% of the significantly enriched GO categories of Cellular Compartment are related to chloroplast (chloroplast, chloroplast envelope, thylakoid membrane, chloroplast membrane, stroma, photosystems I and II, and light-harvesting complex) ( Supplementary Table SV ). The Fe-related genes studied in this work, which expression was affected in COPT1^OE plants, are summarized in Table 1 .

One of the most relevant result of the transcriptomic analysis in COPT1^OE is the altered expression of Fe-related genes ( Table 1 ), such as IRT1 and FRO2, involved in the strategy I of Fe uptake ( Table 1 ). Expression of the Fe transporter IRT1 was reduced in COPT1^OE seedling in both Cu deficiency (½ MS) or excess (½ MS + 10 Cu). In contrast, the genes encoding FRO reductases showed a different behavior; expression of FRO2 was induced but that for FRO3 was repressed in COPT1^OE seedlings only in Cu excess. In order to further assess the effects of Cu on their expression, seedlings were grown on hand-made ½ MS medium with three different CuSO₄ concentrations, including Cu deficiency (0 μM CuSO₄), sufficiency (1 μM CuSO₄), and mild Cu excess (10 μM CuSO₄). Furthermore, in addition to COPT1^OE , plants overexpressing the COPT3 transporter (COPT3^OE ) (Andrés-Colás et al., 2010) were also included to further assess the effect of deregulated Cu⁺ entrance ( Figure 1 ). IRT1 expression slightly decreased in the WT seedlings as Cu increased in the medium, being significantly lower under Cu sufficiency and excess compared to deficiency conditions ( Figure 1A ). Similar to the global transcriptomic analysis, IRT1 expression in COPT^OE (COPT1^OE and COPT3^OE lines) was lower than in WT under both Cu deficiency and excess. Since genes encoding reductases FRO2 and FRO3 displayed distinct regulation in COPT1^OE seedlings ( Table 1 ), total ferroreductase activity was measured in the roots of 7-day-old COPT^OE seedlings ( Figure 1B ). No changes in ferroreductase activity were observed under Cu deficiency and sufficiency. However, a slightly lower ferroreductase activity was observed under Cu excess in the COPT1^OE plants compared to the WT ( Figure 1B ).

COPT1^OE and COPT3^OE seedlings were previously shown to incorporate more Cu, both under metal deficiency and excess in the growth medium (Andrés-Colás et al., 2010). Now, the Fe content was determined by ICP-MS in both shoots and roots from 7-day-old WT and COPT^OE seedlings grown under the different Cu conditions ( Figures 1C, D ). We observed that COPT^OE seedlings had slightly lower endogenous Fe concentration than the WT both in shoots and roots but only significant under Cu sufficiency and excess. This decrease was more evident in the roots of the COPT3^OE seedlings ( Figure 1D ), which is consistent with their exacerbated reduction in IRT1 expression ( Figure 1A ).

To corroborate the expression changes of other Fe-related genes involved in Fe storage and metabolism in COPT1^OE seedlings, we further investigated the expression of genes included in these processes, such as FER1, FER3, and At-NEET ( Table 1 and Figure 2 ). Previous data have shown that the expression of FER1 is higher among the four genes (FER1, FER2, FER3 and FER4) that encode ferritins (Petit et al., 2001). FER1 and FER3 expression was down-regulated under Cu excess compared to Cu deficiency in WT and this decrease was more exacerbated in COPT^OE seedlings ( Table 1 and Figures 2A, B ). This pattern is also in agreement with the lower Fe content in the COPT^OE under Cu excess ( Figure 1C ). At-NEET encodes a Fe/S protein involved, among other processes, in the Fe metabolism and ROS homeostasis (Nechushtai et al., 2012; Mittler et al., 2019). At-NEET expression was induced under Cu deficiency in the WT and reduced in the COPT^OE seedlings under all the Cu conditions studied ( Table 1 and Figure 2C ). The expression of FER1, FER3, and At-NEET was also analyzed under different Cu and Fe contents in WT seedlings ( Supplementary Figure S2 ). Whereas all of them were induced under Cu deficiency compared to control conditions, only FER1 expression was repressed under Fe deficiency. FER3 and At-NEET were not regulated by Fe levels ( Supplementary Figure S2 ). Altogether, these results indicate that the deregulated Cu⁺ entrance in COPT^OE prevents the induction of genes involved in Fe uptake, leading ultimately to a reduced endogenous Fe content, and suggesting that overexpression of COPT may alter the local response to Fe deficiency in roots.

Certain Fe deficiency responses are under the control of the Cu-responsive SPL7 transcription factor (Kastoori Ramamurthy et al., 2018). To determine whether the inhibition of IRT1 expression observed in COPT^OE could be a SPL7-mediated response, we checked the expression in COPT^OE seedlings of SPL7 and SPL7-regulated markers of Cu deficiency responses, such as COPT2 and FSD1 ( Supplementary Figure S3 ), the latter encoding FeSOD (Yamasaki et al., 2009). SPL7 expression remained mostly unaffected in COPT^OE . Whereas COPT2 expression was down-regulated in COPT^OE seedlings, FSD1 expression was not affected under Cu deficiency ( Supplementary Figures S3B, C ). Furthermore, other SPL7 targets were not differentially expressed in the COPT1^OE seedlings under Cu deficiency ( Tables SI and SII ), indicating that the SPL7 factor was properly functioning in COPT^OE seedlings and it was not differentially affecting Cu deficiency responses in these plants. On the other hand, the expression of CSD2, encoding the chloroplastic Cu/Zn SOD CSD2, was reduced in COPT^OE seedlings under Cu sufficiency and excess ( Supplementary Figure S3D ).

To further address the role of Cu in the lack of Fe-deficiency response, we analyzed the expression of Fe-related bHLH transcription factors and other regulators ( Table 1 and Figure 3 ). A slight increase and a lower expression were observed for FIT under Cu deficiency and excess, respectively, in COPT1^OE seedlings compared to the WT ( Figure 3A ). Consistent with FIT inhibition under Cu excess, the expression of two of its targets, IRT1 and COPT2, was slightly repressed in COPT^OE seedlings ( Figure 1A and Supplementary Figure S3B ), which suggests that high Cu in these plants represses the activity of this master Fe uptake regulator. The expression of the subgroup Ib of Fe-related bHLH (bHLH38, bHLH39, bHLH100 and bHLH101) transcription factors was also analyzed ( Figures 3B, C , Supplementary Figures S4A, B ). We observed that, in the WT plants, expression of bHLH-Ib factors was in general greater as Cu increased in the growth medium. Remarkably, its expression was highly increased in COPT1^OE seedlings under Cu excess conditions ( Figure 3 and Supplementary Figure S4 ). Therefore, the opposite regulation in COPT1^OE seedlings under Cu excess was observed for FIT and the rest of regulatory bHLH-Ib factors. Whereas FIT expression was reduced under Cu excess in COPT^OE respect to the WT, bHLH-Ib expression was increased under Cu excess ( Figure 3 and Supplementary Figure S4 ).

Among the most induced Fe-related genes in the COPT1^OE line under Cu excess were the Fe peptide regulators FEP2 and FEP3 ( Table 1 ). The expression of these responds to Fe deficiency in a FIT-independent manner (Hirayama et al., 2018). In agreement with low Fe levels in the COPT1^OE plants, FEP2 and FEP3 were clearly induced under Cu excess ( Figures 3D, E ). This result indicated that the FIT-independent signaling pathway for FEP2 and FEP3 induction was not affected by deregulated Cu uptake in COPT1^OE under Cu excess.

Taken together these results indicate that strategy I local responses to Fe deficiency cannot take place in COPT^OE seedlings under Cu excess despite the increased expression in bHLH-Ib transcription factors and FEP genes. This is probably due to the inhibition of FIT expression ( Figure 3A ), whereas the opposite occurs under Cu deficiency where, despite slightly enhanced FIT expression in the COPT1^OE seedlings, the low bHLH-Ib expression ( Figure 3 and Supplementary Figure S4 ) was precluding assimilatory Fe deficiency responses.

Regarding to upstream regulators of Fe homeostasis that were repressed in COPT1^OE ( Table 1 ), we analyzed the expression of the subgroup IVc members of bHLH (bHLH105/IRL3 and bHLH115) transcription factors. IRL3 and bHLH115 were repressed in COPT3^OE ( Supplementary Figures S4C, D ). On the other hand, BTS expression encoding an E3 ubiquitin ligase that participates in Fe sensing (Long et al., 2010; Kobayashi et al., 2013) was enhanced under Cu excess in WT plants and further increased in COPT1^OE ( Figure 3F ). This increased expression in BTS and the reduced Fe uptake response under Cu excess ( Figure 1 ) was according to the BTS role as a negative regulator of Fe assimilatory responses (Hindt et al., 2017).

Plant hormones are involved in regulating the expression of Fe deficiency-responsive genes (Kobayashi, 2019). In order to check if hormone contents were significantly affected in COPT1^OE seedlings, abscisic acid (ABA), jasmonic acid (JA), and indol acetic acid (IAA) contents were determined ( Supplementary Figure S5 ). Whereas the ABA levels were not significantly modified with respect to controls, JA and IAA levels decreased in COPT1^OE seedlings as Cu increased in the medium ( Supplementary Figure S5 ).

Given the previous data, we studied the phenotype of COPT^OE seedlings under Fe deficiency ( Figure 4 ). Thus, 7-day-old WT, COPT1^OE , and COPT3^OE seedlings were grown in media with Fe sufficiency (50 μM FeSO₄) and slight Fe deficiency (10 μM FeSO₄) ( Figure 4 ). Whereas WT seedlings increased their root length under slight Fe deficiency conditions, COPT^OE presented a similar length of their roots in both media ( Figures 4A, B ). Root shortening in the COPT^OE with respect to WT under slight Fe deficiency was around 50%. No changes were observed in the seedlings fresh weight under sufficiency, while under slight Fe deficiency, the weight was reduced in COPT1^OE ( Figure 4C ). Moreover, the total chlorophyll content did not show significant changes in the WT, while COPT^OE seedlings displayed around 50% lower chlorophyll content in Fe deficiency ( Figure 4D ). Taken together, these data indicate that COPT^OE seedlings are more sensitive to slight Fe deficiency conditions than WT.

Then, we analyzed the expression of Fe homeostasis-related genes under Fe sufficiency and deficiency. IRT1 expression increased under Fe deficiency in both WT and COPT^OE ( Supplementary Figure S6A ). On the other hand, an increase in FIT expression was observed under Fe deficiency in the WT plants, while in COPT^OE this increase was exacerbated ( Supplementary Figure S6B ). Furthermore, FEP2 and FEP3 expression was induced in WT plants under Fe deficiency but it was also highly increased in COPT1^OE ( Supplementary Figures S6C, D ). Taken together, these results indicated that COPT^OE seedlings properly responded to Fe deficiency under Cu sufficiency and that their defects in Fe homeostasis were restricted to Cu deficiency and excess conditions.

The expression of the Fe homeostasis regulators FIT, bHLH38, bHLH100, and UPB1 was analyzed under different metal deficiency and excess conditions in the WT plants ( Supplementary Figure S7 ). FIT, bHLH38, and bHLH100 where induced under Fe deficiency whereas UPB1 was repressed. Curiously, all of them but FIT were regulated in the same sense under Cu excess and Fe deficiency, either induced (bHLH38 and bHLH100) or repressed (UPB1), suggesting that maybe there are certain similarities or common steps in signal transduction between the responses to Cu excess and Fe deficiency, leading to the induction of these transcriptional factors.

GO analysis of the global transcriptomic analysis revealed that the chloroplast was the subcellular compartment most affected in COPT1^OE ( Supplementary Table SV ). GO categories differentially enriched among genes differentially regulated in COPT1^OE seedlings were photosynthesis-associated nuclear genes (PhANG) and genes related to photomorphogenesis, and oxidative stress ( Table 1 and Supplementary Tables SIII and SV ). Therefore, we decided to assess the alteration of the expression of genes in the tetrapyrrole biosynthesis pathway and the retrograde signaling components ( Table 1 ). The LHCB2.3 expression, a representant among the multiple PhANG, which are down-regulated in the COPT1^OE seedlings ( Table SII ), was confirmed to be repressed ( Figure 5A ). Moreover, the expression of GUN4, a regulator of the plastidic tetrapyrrole signaling (Davison et al., 2005), was also repressed in most conditions ( Figure 5B ). In addition, CHL27/CDR1 (Bang et al., 2008) was also down-regulated in COPT^OE seedlings ( Figure 5C ), further confirming the involvement of tetrapyrrole signaling in Cu-dependent Fe deficiency responses. The regulation of a chloroplastic sigma factor SIG1 involved in retrograde signaling (Macadlo et al., 2019) was also down-regulated under Cu deficiency in COPT^OE seedlings ( Figure 5D ). Taken together, these results indicated that the tetrapyrrole signaling pathway was affected under both Cu deficiency and excess, probably as a consequence of the common Fe deficiency conditions faced by COPT^OE seedlings.

Since the respiratory electron transport chain is one of the most metal requiring processes, O₂ consumption was measured in 7-day-old roots of WT and COPT1^OE seedlings grown in deficiency (0 μM CuSO₄), sufficiency (1 μM CuSO₄) and excess CuSO₄ (10 μM CuSO₄) ( Figure 6A ). The results showed that O₂ consumption in WT seedlings did not exhibit significant changes between the three media, while COPT1^OE displayed lower O₂ consumption when the Cu conditions were not optimal, in both deficiency and in excess, with a decrease of 42 and 46% respectively, compared to WT ( Figure 6A ). In agreement with a defective functioning of the respiratory electron transport chain, expression of the ALTERNATIVE OXIDASE 1D (AOX-1D), aimed to restore electron flow in a non-phosphorylating bypass (Clifton et al., 2006; Selinski et al., 2018), was increased in COPT1^OE ( Figure 6B ). Moreover, other AOX genes, such as AOX1A (+2.46) and AOX2 (+2.08), were also induced under Cu excess in COPT^OE ( Supplementary Table SI ). On the other hand, expression of the LOW SULPHUR UPREGULATED1 (LSU1) has been shown to prevent chloroplastic ROS production by interacting with FSD2 (Garcia-Molina et al., 2017). LSU1 was also greatly induced under Cu excess ( Figure 6C ), as well as LSU2 and LSU3 ( Supplementary Table SI ). Inhibition of FSD1 and CSD2 expression in the COPT^OE plants ( Supplementary Figure S3 ) suggested a defective antioxidant capacity by an increase in the radical superoxide (O₂ ^.− ) and a decrease in hydrogen peroxide (H₂O₂).

UPB1 spatially regulates ROS distribution in the root transition zone where the balance between O₂ ^.− and H₂O₂ controls the transition between root cell proliferation and differentiation (Tsukagoshi et al., 2010). UPB1 expression was induced under Cu deficiency and sufficiency in COPT1^OE seedlings compared to WT ( Figure 6D ). Accordingly, two of its direct targets, PER39 and PER40, were also up-regulated under Cu deficiency ( Supplementary Figures S8A, B ). Furthermore, UBP1 expression under different metal stress conditions in the WT indicated that was repressed under Cu excess as well as when Fe is low ( Supplementary Figure S7D ). These responses indicated clearly different responses of Fe regulators to low and high Cu levels.

Since these results would lead to a further decrease in H₂O₂, seedlings were grown in Cu-deficient media in the presence of H₂O₂ and the reductants ascorbic acid (AsH) and 1,4-ditiothreitol (DTT) ( Supplementary Figure S9 ). COPT1^OE seedlings showed shorter roots than controls under Cu deficiency. Whereas no clear effects were observed on root length for the H₂O₂ treatment, reductants such as AsH and DTT, improved the root growth of COPT1^OE compared to WT ( Supplementary Figure S9 ) further pointing to increased oxidative stress in the overexpressing line under Cu deficiency.

Additionally, since the Fe content was lower in COPT^OE , a Fe excess treatment was also provided with no differential effect compared to controls ( Supplementary Figure S9 ). In order to ascertain the Fe redox state in the roots, 7-day-old seedlings grown in the same media with different Cu content (0, 1, and 10 μM CuSO₄) were stained with the Perl´s blue to detect the presence of Fe³⁺. The intensity of stained roots indicated that Fe³⁺ increased as Cu rise in the media from 0 to 10 μM Cu in all seedlings´ genotypes. Moreover, COPT^OE seedlings were more intensely stained than WT in all the conditions ( Figure 7 ) and this result is reproducible in 11-day-old seedlings ( Supplementary Figure S10 ), indicating that Cu⁺ uptake through COPT is affecting the redox state of incorporated Fe leading to an increase in the Fe³⁺/Fe²⁺ ratio in the COPT^OE roots since the lower total Fe concentration ( Figure 1D ).

These results suggested that the reduction in Fe content observed in COPT^OE could be due to the increased Fe³⁺/Fe²⁺ ratio when Cu⁺ entrance was enhanced and deregulated in COPT^OE seedlings. Thus, a model for the Cu excess effects on Fe homeostasis in COPT^OE seedlings has been proposed ( Figure 8 ). The deregulated Cu⁺ entrance in COPT^OE plants under high Cu activated the expression of the bHLH-Ib factors but not FIT. This effect redounded in Fe deficiency symptoms in COPT^OE probably as a consequence of the lack of Fe assimilatory response. Cu excess increased the Fe³⁺/Fe²⁺ ratio in COPT^OE roots, probably leading to signaling pathways that could further inhibit Fe uptake.