Comparative Study of Several Fe Deficiency Responses in the Arabidopsis thaliana Ethylene Insensitive Mutants ein2-1 and ein2-5

Iron (Fe) is an essential micronutrient for plants since it participates in essential processes such as photosynthesis, respiration and nitrogen assimilation. Fe is an abundant element in most soils, but its availability for plants is low, especially in calcareous soils. Fe deficiency causes Fe chlorosis, which can affect the productivity of the affected crops. Plants favor Fe acquisition by developing morphological and physiological responses in their roots. Ethylene (ET) and nitric oxide (NO) have been involved in the induction of Fe deficiency responses in dicot (Strategy I) plants, such as Arabidopsis. In this work, we have conducted a comparative study on the development of subapical root hairs, of the expression of the main Fe acquisition genes FRO2 and IRT1, and of the master transcription factor FIT, in two Arabidopsis thaliana ET insensitive mutants, ein2-1 and ein2-5, affected in EIN2, a critical component of the ET transduction pathway. The results obtained show that both mutants do not induce subapical root hairs either under Fe deficiency or upon treatments with the ET precursor 1-aminocyclopropane-1-carboxylate (ACC) and the NO donor S-nitrosoglutathione (GSNO). By contrast, both of them upregulate the Fe acquisition genes FRO2 and IRT1 (and FIT) under Fe deficiency. However, the upregulation was different when the mutants were exposed to ET [ACC and cobalt (Co), an ET synthesis inhibitor] and GSNO treatments. All these results clearly support the participation of ET and NO, through EIN2, in the regulation of subapical root hairs and Fe acquisition genes. The results will be discussed, taking into account the role of both ET and NO in the regulation of Fe deficiency responses.


Introduction
Iron (Fe) is very abundant in most soils, mainly as Fe 3+ , although its availability to plants is low, especially in calcareous soils [1][2][3]. On the other hand, excessive Fe accumulation by the plant may lead to toxic effects [4,5]. Therefore, Fe acquisition is highly regulated. Dicot (Strategy I) plants, such as Arabidopsis, need to reduce Fe 3+ to Fe 2+ by means of a plasma membrane ferric reductase, encoded by the FRO2 gene, prior to its root absorption through a Fe 2+ transporter, encoded by the IRT1 gene [4,6,7]. When grown under Fe deficiency, dicot plants induce several physiological and morphological responses (mainly in their roots) aimed at facilitating the mobilization and acquisition of this nutrient [4,6,7]. Among the physiological responses, dicot plants enhance both ferric reductase activity (FRA; due to increased expression of AtFRO2-like genes) and Fe 2+ uptake capacity (due to increased expression of AtIRT1-like genes) [4,7]. In addition to physiological responses, dicot plants can develop some morphological responses in their activity (FRA; due to increased expression of AtFRO2-like genes) and Fe 2+ uptake capacity (due to increased expression of AtIRT1-like genes) [4,7]. In addition to physiological responses, dicot plants can develop some morphological responses in their roots, such as subapical root hairs, root epidermal transfer cells and cluster roots, also named proteoid roots [7,8]. All these root modifications enhance Fe uptake by increasing the contact surface of roots with soil and by chemically modifying the soil environment [9].
The regulation of the Fe-related genes associated with physiological responses, such as FRO2 and IRT1, is not totally understood, but in the last few years several transcription factors (TFs) that participate in their activation have been found. The master regulator of most Fe acquisition genes in Arabidopsis is the FIT (bHLH29) TF [4,6,7,10,11]. The Arabidopsis fit mutant is very chlorotic and lacks the ability to activate most Fe deficiency responses in roots [7]. In Arabidopsis, the FIT regulatory network comprises other bHLH TFs, such as bHLH38, bHLH39 and others [10,11]. Some of these TFs, such as bHLH38 and bHLH39, can interact with FIT to form heterodimers that activate the expression of the Fe acquisition genes FRO2 and IRT1 [4,10,11]. FIT is induced in roots in response to Fe deficiency, while other bHLHs are induced in both roots and leaves in response to Fe deficiency [4].
Once adequate Fe has been absorbed, Fe deficiency responses need to be switched off to minimize Fe toxicity and energy costs. Their regulation is not fully understood, but several hormones and signaling substances, such as auxin, ethylene (ET) and nitric oxide (NO), which increase their production in Fe-deficient roots, have been proposed to participate in the activation of most responses in dicot plants [4,7,[12][13][14][15][16][17][18][19][20][21]. Auxin, ET and NO are closely interrelated in a complex manner since each one can affect the production and/or distribution of the other ones [19,[22][23][24][25]. For example, it has been found that the expression of genes encoding enzymes implicated in ET synthesis, and ET itself, are enhanced by NO treatments, while NO accumulates in the subapical region of roots treated with the ET precursor 1-aminocyclopropane-1-carboxylic acid (ACC) [24,26]. Besides ET synthesis, Fe deficiency also affects ET responsiveness by altering the expression of genes implicated in ET signaling, such as EIN2, EIN3, EILs and ERFs [7,18,19]. EIN2 is upregulated in Arabidopsis roots under Fe deficiency [27] and also in plants overexpressing the PDF1.1 gene, which causes upregulation of Fe acquisition genes [28]. In contrast to auxin, ET and NO, other substances, such as cytokinins, jasmonic acid (JA) and brassinosteroids, have been involved in the suppression of Fe deficiency responses [4,7,16,19].
ET is synthesized from methionine via a pathway that requires several enzymes, such as ACC synthases and ACC oxidases, and where ACC is the immediate ET precursor [29]. ET synthesis can be blocked by using several ethylene inhibitors, such as cobalt (Co), or promoted by applying the ET precursor ACC [7,12,13]. Although ethylene's mode of action is not fully understood, a linear canonical signaling pathway has been proposed in Arabidopsis [30][31][32]: In this pathway, CTR1 is a kinase, EIN2 is a transmembrane protein located in the endoplasmic reticulum (ER) membrane, and EIN3, EILs and ERFs are TFs [30][31][32].
EIN2 possesses a Nramp-like transmembrane domain and a cytosolic COOH end (CEND; Figure 1) domain [32][33][34][35]. The role of the Nramp transmembrane domain is not yet clear, while two roles have been proposed for the CEND portion [32][33][34][35]. The Nramp portion shows similarity to the Nramp family of metal-ion transporters [32,36]. However, no metal transport activity has been detected in heterologously expressed EIN2 [32], although its possible role as a sensor of divalent metals has been suggested [36]. In the absence of ET, CTR1 phosphorylates EIN2, preventing the cleavage and translocation of CEND into the nucleus. In the presence of ET, CTR1 is inactivated, resulting in dephosphorylation of EIN2 and its cleavage. CEND is then translocated into the nucleus, where it interacts with the EIN2 Nuclear Associated Protein 1 (ENAP1) and potentially ET receptors → CTR1 Plants 2021, 10, x FOR PEER REVIEW 2 o activity (FRA; due to increased expression of AtFRO2-like genes) and Fe 2+ uptake capac (due to increased expression of AtIRT1-like genes) [4,7]. In addition to physiological sponses, dicot plants can develop some morphological responses in their roots, such subapical root hairs, root epidermal transfer cells and cluster roots, also named proteo roots [7,8]. All these root modifications enhance Fe uptake by increasing the contact su face of roots with soil and by chemically modifying the soil environment [9].
The regulation of the Fe-related genes associated with physiological responses, su as FRO2 and IRT1, is not totally understood, but in the last few years several transcripti factors (TFs) that participate in their activation have been found. The master regulator most Fe acquisition genes in Arabidopsis is the FIT (bHLH29) TF [4,6,7,10,11]. The A bidopsis fit mutant is very chlorotic and lacks the ability to activate most Fe deficien responses in roots [7]. In Arabidopsis, the FIT regulatory network comprises other bHL TFs, such as bHLH38, bHLH39 and others [10,11]. Some of these TFs, such as bHLH and bHLH39, can interact with FIT to form heterodimers that activate the expression the Fe acquisition genes FRO2 and IRT1 [4,10,11]. FIT is induced in roots in response to deficiency, while other bHLHs are induced in both roots and leaves in response to Fe d ficiency [4].
Once adequate Fe has been absorbed, Fe deficiency responses need to be switch off to minimize Fe toxicity and energy costs. Their regulation is not fully understood, b several hormones and signaling substances, such as auxin, ethylene (ET) and nitric oxi (NO), which increase their production in Fe-deficient roots, have been proposed to part ipate in the activation of most responses in dicot plants [4,7,[12][13][14][15][16][17][18][19][20][21]. Auxin, ET and NO a closely interrelated in a complex manner since each one can affect the production and distribution of the other ones [19,[22][23][24][25]. For example, it has been found that the expr sion of genes encoding enzymes implicated in ET synthesis, and ET itself, are enhanc by NO treatments, while NO accumulates in the subapical region of roots treated with t ET precursor 1-aminocyclopropane-1-carboxylic acid (ACC) [24,26]. Besides ET synthes Fe deficiency also affects ET responsiveness by altering the expression of genes implicat in ET signaling, such as EIN2, EIN3, EILs and ERFs [7,18,19]. EIN2 is upregulated in A bidopsis roots under Fe deficiency [27] and also in plants overexpressing the PDF1.1 ge which causes upregulation of Fe acquisition genes [28]. In contrast to auxin, ET and N other substances, such as cytokinins, jasmonic acid (JA) and brassinosteroids, have be involved in the suppression of Fe deficiency responses [4,7,16,19].
ET is synthesized from methionine via a pathway that requires several enzymes, su as ACC synthases and ACC oxidases, and where ACC is the immediate ET precursor [2 ET synthesis can be blocked by using several ethylene inhibitors, such as cobalt (Co), promoted by applying the ET precursor ACC [7,12,13]. Although ethylene's mode of tion is not fully understood, a linear canonical signaling pathway has been proposed in A bidopsis [30][31][32]: In this pathway, CTR1 is a kinase, EIN2 is a transmembrane protein located in t endoplasmic reticulum (ER) membrane, and EIN3, EILs and ERFs are TFs [30][31][32].
EIN2 possesses a Nramp-like transmembrane domain and a cytosolic COOH e (CEND; Figure 1) domain [32][33][34][35]. The role of the Nramp transmembrane domain is n yet clear, while two roles have been proposed for the CEND portion [32][33][34][35]. The Nram portion shows similarity to the Nramp family of metal-ion transporters [32,36]. Howev no metal transport activity has been detected in heterologously expressed EIN2 [32], a hough its possible role as a sensor of divalent metals has been suggested [36]. In the a sence of ET, CTR1 phosphorylates EIN2, preventing the cleavage and translocation CEND into the nucleus. In the presence of ET, CTR1 is inactivated, resulting dephosphorylation of EIN2 and its cleavage. CEND is then translocated into the nucle where it interacts with the EIN2 Nuclear Associated Protein 1 (ENAP1) and potentia EIN2 → EIN3/EILs → ERFs → ET responses (1) In this pathway, CTR1 is a kinase, EIN2 is a transmembrane protein located in the endoplasmic reticulum (ER) membrane, and EIN3, EILs and ERFs are TFs [30][31][32].
EIN2 possesses a Nramp-like transmembrane domain and a cytosolic COOH end (CEND; Figure 1) domain [32][33][34][35]. The role of the Nramp transmembrane domain is not yet clear, while two roles have been proposed for the CEND portion [32][33][34][35]. The Nramp portion shows similarity to the Nramp family of metal-ion transporters [32,36]. However, no metal transport activity has been detected in heterologously expressed EIN2 [32], although its possible role as a sensor of divalent metals has been suggested [36]. In the absence of ET, CTR1 phosphorylates EIN2, preventing the cleavage and translocation of CEND into the nucleus. In the presence of ET, CTR1 is inactivated, resulting in dephosphorylation of EIN2 and its cleavage. CEND is then translocated into the nucleus, where it interacts with the EIN2 Nuclear Associated Protein 1 (ENAP1) and potentially Histone Acetyl Transferases (HATs), resulting in histone acetylation. This causes an uncompacting of chromatin, resulting in more EIN3/EIL1 binding to target genes and ultimately transcription activation ( Figure 1) [32,35,37,38]. In addition to this role in the nucleus, CEND in the cytosol can block the mRNA translation of the F-box proteins EBF1 and EBF2, which participate in the proteasomal degradation of EIN3 and EIL1, thus promoting their accumulation [32][33][34]. EIN3 and EIL1 further regulate the expression of the ET response factor (ERF) family TFs [30][31][32]39]. The ERF TFs act downstream of EIN3, and EIL1, to activate or repress ET-responsive genes, although some ERFs can be activated by ET-independent TFs, not related to EIN3 [40]. In Arabidopsis, as in other plant species, many mutants altered in the different ET signaling genes have been identified. Among them, the ein2 mutants are insensitive to most responses to ET [30,35,36]. More recently, the existence of an alternate "non canonical" route for ET signaling, besides the one including EIN2, is further supported by different experimental results ( Figure 1; [7,30,32,41]). For example, the expression of several Fe acquisition genes in response to either ET precursors or inhibitors is altered in the Arabidopsis ein2-1 mutant [27,42]. Moreover, Kim et al. [43] have shown that this mutant is capable of responding to ET if JA levels or signaling are low. Figure 1. Model for the role of ethylene, through EIN2, on the upregulation of the Fe acquisition genes FRO2 and IRT1 in Arabidopsis. Fe deficiency causes enhanced ET production in roots. In the presence of ET, the CEND portion of the EIN2 protein is cleaved and shuttled into the nucleus where it increases the transcriptional activity of the EIN3/EIL1 TFs (see Introduction for more details). The accumulation of these latter TFs trigger ET responses, such as enhanced FIT transcription and, consequently, FRO2 and IRT1 upregulation. The transcription of FIT, FRO2 and IRT1 could also be triggered through alternate pathways, such as the "non canonical" ET-signaling route involving AHPs and ARRs or the ones involving other hormones, such as auxin (dashed lines). In addition to FIT, the different routes could also upregulate FRO2 and IRT1 through other TFs, such as ERF1. Abbreviations: AHPs, Arabidopsis Histidine-containing Phosphotransmitters; ARRs, Arabidopsis Response Regulators; ER, Endoplasmic Reticulum; ERF1, Ethylene Response Factor1; ET, Ethylene; NM, Nuclear Membrane; TFs, Transcription Factors. Based on [7,17,[32][33][34][35][36]44,45].
The implication of ET in the regulation of Fe deficiency responses in dicot plants is supported by many experimental results. For example, the addition of ET inhibitors, such as Co, to Fe-deficient plants inhibits the induction of most Fe deficiency responses, such as enhanced FRA and subapical root hairs, while the addition of ET or ACC to Fe-sufficient plants induces them [7,12,13,19,27,46,47]. Furthermore, ET also affects the expression of the genes controlling these responses [7,19,27,44,[46][47][48]. For example, ET upregulates FIT expression and, consequently, the Fe acquisition genes FRO2 and IRT1, activated by this key TF ( Figure 1) [27,46,48]. The link between ET and FIT has been recently reinforced. It has been shown that EIN3 and EIL1, two TFs in the ET signaling pathway, interact with MED16 (Mediator) and MED25 to form a complex implicated in FIT transcription [4,44]. Moreover, EIN3 and EIL1 can also influence the posttranscriptional stability of FIT [48].
ET, along with auxin, NO, and other hormones and signaling substances, has also been implicated in the regulation of Fe deficiency morphological responses, such as subapical root hairs [7,8,13,14,19]. In addition to the results obtained with ET inhibitors and precursors, several ET mutants have been used to study the role of ET in the development of these morphological changes. In relation to the EIN2 protein, several mutants have been identified in Arabidopsis, such as ein2-1 and ein2-5 [33,36,43,49]. The ein2-5 mutation is located in the Nramp transmembrane domain, while the ein2-1 mutation is located in the CEND cytosolic portion ( Figure 1) [36]. It has been shown that the Arabidopsis ET insensitive ein2-1 mutant did not develop subapical root hairs either under Fe-deficiency or upon ET treatments, while the wild-type did [13,27]. Similarly, the ein2-5 mutant did not develop subapical root hairs upon ACC treatment [49]. Results obtained with the ein2-1 mutant also suggest that Fe deficiency physiological and morphological responses can be regulated through different ET signaling pathways. For instance, the development of subapical root hairs is impaired in the ein2-1 mutant (see above), while enhanced FRA and the expression of Fe acquisition genes is not [13,27,42].
In previous works, it was found that the addition of Co, an ET synthesis inhibitor, at 50 µM final concentration, to Fe-deficient wild-type Columbia plants drastically inhibited several Fe deficiency responses, such as FRA enhancement and FIT, FRO2 and IRT1 upregulation [27,46]. However, the application of Co, at 50 µM final concentration, to Fe-deficient ein2-5 plants inhibited FRA enhancement but did not inhibit FIT, FRO2 or IRT1 upregulation (unpublished results). The objective of this work was to look further into the above lack of inhibition of the ein2-5 mutant by comparing it with the ein2-1 mutant under different Co concentrations, and also under other treatments, such as ACC and S-nitrosoglutathione (GSNO) applications. GSNO is derived from glutathione (GSH) and NO and is regarded as an intracellular NO reservoir as well as a vehicle of NO throughout the cell [50]. Although endogenous GSNO, present at very low concentrations, does not behave exactly as NO [51,52], exogenous applied GSNO, at the concentrations used, is considered to be a NO donor [14,22,26].

Results
In wild-type Columbia plants, Fe deficiency induces a suite of genes in roots, including those involved in Fe acquisition, such as FRO2, IRT1 and FIT [7,27,46]. In previous works, it was shown that the upregulated expression of the Fe deficiency induced genes FRO2, IRT1 and FIT in wild-type Columbia plants was suppressed by ET inhibitors, such as cobalt (Co), applied at 50 µM final concentration [27,46]. However, preliminary experiments in our lab showed that the application of Co, at 50 µM final concentration, to Fe-deficient ein2-5 plants did not inhibit FIT, FRO2 or IRT1 upregulation (unpublished results).
In this work, we wanted to verify the effect of Co on the expression of the Fe-related genes described above, and also on ferric reductase activity (FRA), in the Arabidopsis ET insensitive ein2.5 mutant but also in the ein2.1 mutant. In addition, we wanted to compare the ability of both mutants to respond to ET and NO treatments. As described below, the behavior of the mutants in response to Co (ET synthesis inhibitor), ACC (ET precursor) and GSNO (NO donor) was different in the induction of physiological responses.   (Figure 2). The effect of Co (ET inhibitor) on the expression of the above genes in Fe-deficient plants was different depending on the mutant. While in the ein2.1 mutant, Co drastically inhibited the expression of all the genes studied, in the ein2.5 mutant, Co had no effect in any case except on FIT expression, where its relative expression decreased with the higher Co doses (75 and 100 µM; Figure 2).
FRA was greatly enhanced under Fe deficiency in both mutants and was drastically inhibited upon Co treatments in both mutants (Figure 3). Collectively, the results show that the Fe deficiency physiological responses studied (FRA and FRO2, IRT1 and FIT expression) are induced by Fe deficiency in both mutants. However, both mutants behave differently in response to the ET synthesis inhibitor Co: it drastically inhibits FRO2 and IRT1 expression in Fe-deficient ein2-1 plants but not in Fe-deficient ein2-5 plants.

Effect of ACC on Ferric Reductase Activity (FRA) and FRO2, IRT1 and FIT Expression
Previous results showed that the ein2.1 mutant was able to upregulate FRO2, IRT1 and FIT expression in response to ET treatments as the wild-type did [27]. The results obtained in this work agree with these previous ones, showing that these genes are upregulated by ACC (ET precursor) in both ein2 mutants (Figure 4). However, while FRO2 and IRT1 were greatly induced by ACC in Fe-sufficient ein2.1 mutant plants, mainly at 5 µM, the induction was much lower in Fe-sufficient ein2.5 mutant plants (Figure 4). Even at ACC 5 µM, the ein2.5 mutant did not reach the expression levels attained by the ein2.1 mutant at ACC 1 µM (Figure 4). However, these differences were not remarkable in relation to FIT expression ( Figure 4). In relation to FRA, at ACC 1 µM, its enhancement was higher in Fe-sufficient ein2.1 mutant plants than in Fe-sufficient ein2-5 mutant plants ( Figure 5). At ACC 5 µM, the enhancement was similar in both mutants ( Figure 5). Collectively, the results show that both mutants induce the Fe deficiency physiological responses studied (FRA and FRO2, IRT1 and FIT expression) in response to ACC. However, the intensity of the induction was generally lower in the ein2-5 mutant.

NO Accumulation in Wild-Type Columbia, ein2.1 and ein2.5 Roots in Response to Fe Deficiency and ACC Treatment
As shown in Figure 6, NO accumulated in wild-type Columbia roots under Fe deficiency and also in response to ACC treatments ( Figure 6). In the ein2.1 mutant, the results were similar to the wild-type, although they presented slightly lower NO accumulation, mainly upon ACC treatment ( Figure 6). Finally, the ein2.5 mutant was able to accumulate NO in response to Fe deficiency but not in response to ACC treatment ( Figure 6).
Collectively, the results show that both mutants accumulate NO in the subapical region of the roots in response to Fe deficiency, as occurred in the wild-type Columbia. However, while the ein2.1 mutant also accumulates NO in response to ACC treatment, as occurred in the wild-type Columbia, this accumulation does not occur in the ein2-5 mutant.

Effect of GSNO on Ferric Reductase Activity (FRA) and FRO2, IRT1 and FIT Expression
Under Fe sufficient conditions, FRO2, IRT1 and FIT expression was induced by GSNO treatment in the ein2.1 mutant (Figure 7), as occurred in the wild-type Columbia [26]. However, in the ein2.5 mutant, GSNO treatment did not induce the expression of any of these genes but inhibited it (Figure 7).
Under Fe-deficient conditions, the results were similar: FRO2, IRT1 and FIT expression was induced by GSNO treatment in the ein2.1 mutant but not in the ein2.5 mutant (Figure 8). In the ein2-1 mutant, FRO2 and IRT1 expression attained much higher values in the GSNO-treated plants under Fe deficiency (Figure 8) than under Fe sufficiency (Figure 7).   In relation to FRA, it was induced by the GSNO treatment in both mutants under Fe deficient conditions, with the enhancement being much higher in the ein2.1 mutant (Figure 9). Under Fe-sufficient conditions, there was no effect of GSNO on FRA in any of the mutants (data not shown).
Collectively, the results show that the ein2-1 mutant induces all the Fe deficiency physiological responses studied (FRA and FRO2, IRT1 and FIT expression) in response to GSNO, while the ein2-5 mutant only slightly induced FRA but not FRO2, IRT1 and FIT expression.

Effect of ACC and GSNO on the Development of Subapical Root Hairs
The addition of ACC (at 1 or 10 µM final concentration) or GSNO (at 100 or 500 µM final concentration) to Fe-sufficient ein2-1 and ein2-5 plants did not induce the development of subapical root hairs as occurred in the wild-type Columbia, even with lower ACC and GSNO doses ( Figure 10). None of the mutants induced the development of subapical root hairs under Fe deficiency, while the wild-type did (results not shown). After that, roots were excised and stained with toluidine blue (0.05% w/v). Pictures were taken by using a stereomicroscope.
In relation to the Fe nutrition of dicot (Strategy I) plants, it has been shown that some Fe deficiency responses are impaired in the Arabidopsis ET insensitive mutant ein2-1. For example, this mutant does not develop subapical root hairs either under Fe deficiency or upon ACC treatment, while the wild-type Columbia does Figure 10 [13]. Moreover, ferric reductase activity (FRA) and FRO2 and IRT1 expression are induced in the ein2-1 mutant both under Fe deficiency and upon ET treatments [27,42,53]. However, FRO2 and IRT1 expression attained much lower levels, and was more delayed, in this mutant than in the wild-type Columbia [53].
The results obtained in this work show that both Arabidopsis ET insensitive mutants, ein2-1 and ein2-5, are able to respond to Fe deficiency by inducing some physiological responses, such as FRA enhancement and FRO2, IRT1 and FIT upregulation (Figures 2 and 3), which agrees with previous results obtained with the ein2-1 mutant [27,42]. However, none of the mutants induced subapical root hairs either under Fe deficiency or upon ACC/GSNO treatments (Figure 10), which also matches with previous results obtained with both mutants upon ACC treatment, under Fe deficiency and under other nutrient deficiencies [13,24,27,49,54]. At first, these results suggest that physiological and morphological responses may be regulated through different ET signaling pathways, as previously suggested [13,27,42]. In relation to the physiological responses, the similar behavior of both mutants in their response to Co for FRA enhancement (Figure 3) but their different behavior in response to Co for FRO2 and IRT1 expression (Co, at the concentrations used, inhibits their upregulation in ein2-1 but not in ein2-5; Figure 2) suggests that the ein2-1 mutation has distinct consequences than the ein2-5 one. Furthermore, in the ein2.5 mutant, Co did not inhibit FRO2 expression (responsible for FRA; Figure 2d) but inhibited FRA enhancement (Figure 3b), which would suggest a posttranscriptional regulation of FRO2, as previously proposed by Connolly et al. [55]. These results also suggest a possible implication for ET in such a process. It should be noted that the addition of Co, at 50 µM final concentration, to Fe-deficient wild-type Columbia plants drastically inhibited both FRA enhancement and FIT, FRO2 and IRT1 upregulation [27,46]. However, Co, even at 100 µM final concentration, did not inhibit either FRO2 or IRT1 expression in the ein2-5 mutant (Figure 2d,e).
To further analyze the differences between the ein2-1 and ein2-5 mutations, we then compared the behavior of both mutants in response to ACC (ET precursor) and GSNO (NO donor) treatments. As shown in Figures 4 and 5, both mutants induced all the Fe deficiency physiological responses studied (FRA and FRO2, IRT1 and FIT expression) in response to ACC treatments, as occurred in wild-type Columbia plants [27,46]. These results agree with previous ones obtained with ein2-1 mutant plants treated with either ACC or ET itself [27,42]. However, the intensity of FRO2 and IRT1 upregulation, and FRA enhancement, in response to ACC treatment was lower in the ein2-5 mutant than in the ein2-1 mutant, again suggesting that both mutations have different consequences.
Since NO has also been involved in the activation of Fe deficiency physiological responses [14,15,26,27], we studied the effect of GSNO application on the induction of these responses in both mutants. Similarly to ACC, GSNO also induced all the Fe deficiency physiological responses studied (FRA and FRO2, IRT1 and FIT expression) in the ein2-1 mutant (Figures 7 and 8), as occurred in wild-type Columbia plants [26,27]. FRA was only induced in Fe-deficient plants (Figure 9), which further supports the existence of Fe-related repressive signals [46,51]. In contrast to the ein2-1 mutant, GSNO had no positive effect on the ein2-5 mutant (Figures 7 and 8); only a slight FRA enhancement in Fe-deficient plants (Figure 9). This again shows that the ein2-5 mutant is less responsive to the GSNO treatment, similarly to what occurs with ACC (see above paragraph).
In previous works, NO accumulation in the subapical region of roots either under Fe deficiency or upon ACC treatment has been found [14,24,26]. To test this possibility, both mutants, and also wild-type Columbia plants, were subjected to either Fe deficiency or Fe sufficiency with ACC addition. The ein2-1 mutant accumulated NO under both kinds of treatments, as occurred in the wild-type Columbia, but the ein2-5 mutant only accumulated NO under Fe deficiency but not upon ACC treatment ( Figure 6). This again shows that this latter mutant is less responsive to ACC than the ein2-1 mutant. In addition, the results indicate that EIN2 plays an important role in the accumulation of NO upon ACC treatment, but that perhaps this role could be overridden under Fe deficiency.
The results obtained in this work clearly support the participation of ET, through EIN2, in the regulation of the Fe acquisition genes FRO2 and IRT1. Furthermore, the results do support the existence of additional routes for ET signaling besides the linear canonical one including EIN2 ( Figure 1) [7,30,32,41]. Even more, since the less responsive mutant to ET (ACC) treatment, ein2-5, it still able to upregulate Fe acquisition genes under Fe deficiency (Figure 2), the existence of additional signaling routes, ET-independent, for the activation of the genes would be also possible ( Figure 1). In this sense, it should be noted that ACC itself has been proposed as a signaling molecule [56] and other hormones, such as auxin, have also be involved in the upregulation of Fe acquisition genes ( Figure 1) [17,22]. There are also hormones affected by Fe deficiency, such as jasmonic acid (JA), that could negatively interact with the ET signaling pathway. Kim et al. [43] showed that both ein2-1 and ein2-5 mutants become responsive to ET if JA levels or signaling are low. In accordance with these results, it should be noted that a role for JA in the suppression of Fe deficiency responses has been proposed [57,58]. Another possibility could be the existence of additional FIT-independent pathways to control FRO2 and IRT1 expression. In this way, Balparda et al. [45] have recently proposed that FRO2 and IRT1 expression could be directly controlled by the ERF1 TF (also associated with ET; see Introduction). The possibility exists that these alternate pathways ( Figure 1) could be potentiated when the EIN2-dependent pathway is impaired.

Plant Materials, Growth Conditions and Treatments
Seeds of the Arabidopsis thaliana (L.) Heynh ecotype "Columbia" and the ET-insensitive mutants ein2-1 and ein2.5 were grown under controlled conditions as previously described [27]. Briefly, seeds were germinated in black peat and, when appropriate, seedlings were transferred to individual containers (of 70 mL volume) with complete nutrient solution continuously aerated. The nutrient solution without Fe had the following composition: macronutrients; 2 mM Ca(NO 3  Consequently, in the -Fe + GSNO treatment, the idea was to accelerate the induction of the responses with the GSNO treatment, while in the -Fe + Co treatments, the idea was to inhibit the induction of the responses by the Co treatment. Stock solution of GSNO was prepared as previously described [27]. After treatments, root ferric reductase activity (FRA) was determined as described in the next section. Finally, the roots were collected and kept at −80 • C to later analyze gene expression. Each treatment consists of six biological replications. In some cases, "Columbia", ein2-1 and ein2.5 plants growing in nutrient solution with 10 µM Fe-EDDHA were treated for 24 h with ACC (1 or 10 µM final concentration) or GSNO (100 or 500 µM final concentration). After that, roots were excised and stained with toluidine blue (0.05% w/v) and pictures were taken by using a stereomicroscope.

Ferric Reductase Activity Determination
FRA was determined as previously described [46]. Briefly, intact plants were pretreated for 30 min in plastic vessels with 50 mL of a nutrient solution without micronutrients, pH 5.5, and then placed into 20 mL of a Fe (III) reduction assay solution for 1 h. This assay solution consisted of nutrient solution without micronutrients, 100 µM Fe (III)-EDTA and 300 µM Ferrozine, pH 5.0 (adjusted with 0.1N KOH). The environmental conditions during the measurement of Fe (III) reduction were the same as the growth conditions described above. FRA was determined spectrophotometrically by measuring the absorbance (562 nm) of the Fe (II)-Ferrozine complex and by using an extinction coefficient of 29,800 M −1 cm −1 . After the reduction assay, roots were excised and weighed, and the results were expressed on a root fresh weight basis. The values represent the mean ± SE of six replicates.

Real-Time PCR Analysis
Real-time PCR analysis was carried out as previously described [53]. Briefly, roots were ground to a fine powder with a mortar and pestle in liquid nitrogen. Total RNA was extracted using the Tri Reagent solution (Molecular Research Center, Inc., Cincinnati, OH, USA) according to the manufacturer's instructions. cDNA synthesis was performed by using M-MLV reverse transcriptase (Promega, Madison, WI, USA) from 3 µg of DNase-treated root RNA as the template and random hexamers as the primers. The gene expression study by qRT-PCR was performed in a qRT-PCR Bio-Rad CFX connect thermal cycler and the following amplification profile: initial denaturation and polymerase activation (95 • C for 3 min), amplification and quantification repeated 40 times (90 • C for 10 s, 57 • C for 15 s and 72 • C for 30 s), and a final melting curve stage of 65 • C to 95 • C with increment of 0.5 • C for 5 s to ensure the absence of primer dimer or non-specific amplification products. PCR reactions were set up in 20 µL of SYBR Green Bio-RAD PCR Master Mix, following the manufacturer's instructions. Controls containing water instead of cDNA were included to check for contamination in the reaction components. Primer pairs designed by García et al. [50] were used to amplify FRO2, IRT1, and FIT cDNA. Standard dilution curves were performed for each primer pair to confirm appropriate efficiency of amplification (E = 100 ± 10%). Constitutively expressed SAND1 and YLS8 genes, which do not respond to changes in the Fe conditions [59], were used as reference genes to normalize qRT-PCR results. The relative expression levels were calculated from the threshold cycles (Ct) values and the primer efficiencies by the Pfaffl method [60]. Each PCR analysis was conducted on three biological replicates and each PCR reaction was repeated twice.

Statistical Analyses
All experiments were repeated at least twice and representative results are presented. The values of qRT-PCR represent the mean ± SE of three independent biological replicates. The values of FRA represent the mean ± SE of six replicates. Depending on the experiment, different tests were used. When comparing different treatments with a control (Figures 4 and 5), *, ** or *** indicate significant differences (p < 0.05, p < 0.01 or p < 0.001) among treatments using one-way analysis of variance (ANOVA) followed by a Dunnett's test. Different letters (Figures 2 and 3) indicate significant differences (p < 0.05) among treatments using one-way analysis of variance (ANOVA) followed by a Tukey multiple range test. In the GSNO experiments (Figures 7-9), *, ** or *** indicate significant differences (p < 0.05, p < 0.01 or p < 0.001) between the two treatments according to the Student's t test.

Conclusions
The results presented in this work show that Arabidopsis ethylene insensitive ein2 mutants are still able to respond to ET (ACC) treatments for the upregulation of Fe deficiency physiological responses, which agrees with previous results [27,42]. In the same way, they are able to respond to NO treatments. However, the ein2-5 mutant (altered in the Nramp-domain of EIN2) and the ein2-1 mutant (altered in the CEND portion of EIN2) differ in their response to ACC (ET precursor) and GSNO (NO donor) treatments, with ein2-5 being less responsive to both of them. This suggests that ET and NO are closely interrelated through EIN2. However, the reasons for the differences between ein2-1 and ein2-5 are not yet known.
The results obtained clearly support the participation of ET and NO, through EIN2, in the regulation of the Fe acquisition genes FRO2 and IRT1. In the same way, the results also support the implication of ET and NO, through EIN2, in the development of subapical root hairs. Furthermore, the results do support the existence of additional routes for ET signaling besides the linear canonical one including EIN2.
Since the less responsive mutant to ET (ACC) and NO (GSNO) treatments, ein2-5, is still able to upregulate Fe acquisition genes under Fe deficiency, this could suggest the existence of additional signaling routes, ET-independent, for the activation of these genes.
Finally, the results obtained show that cobalt (ET inhibitor) inhibits the enhanced ferric reductase activity provoked by Fe deficiency in the ein2-5 mutant, while the FRO2 gene, associated with this activity, is not inhibited. These results further support the posttranscriptional regulation of FRO2, suggesting a possible implication of ET in such a process. Funding: This research was supported by the European Regional Development Fund from the European Union, the "Ministerio de Economía y Competitividad" (Projects AGL2013-40822-R and RTI2018-097935-B-I00) and the "Junta de Andalucía" (Research Groups AGR115 and BIO159).

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: All the data included in this article are publicly available.