Transcriptional Regulation of zma-MIR528a by Action of Nitrate and Auxin in Maize

In recent years, miR528, a monocot-specific miRNA, has been assigned multifaceted roles during development and stress response in several plant species. However, the transcription regulation and the molecular mechanisms controlling MIR528 expression in maize are still poorly explored. Here we analyzed the zma-MIR528a promoter region and found conserved transcription factor binding sites related to diverse signaling pathways, including the nitrate (TGA1/4) and auxin (AuxRE) response networks. Accumulation of both pre-miR528a and mature miR528 was up-regulated by exogenous nitrate and auxin treatments during imbibition, germination, and maize seedling establishment. Functional promoter analyses demonstrated that TGA1/4 and AuxRE sites are required for transcriptional induction by both stimuli. Overall, our findings of the nitrogen- and auxin-induced zma-MIR528a expression through cis-regulatory elements in its promoter contribute to the knowledge of miR528 regulome.


Introduction
Global analyses performed on several plant species have revealed that up to 90% of their genome is transcribed, although only a small fraction of the transcripts correspond to protein-coding RNAs (mRNAs) [1]. The remaining transcripts (non-coding RNAs, ncRNAs) have emerged as central regulators of different molecular programs implicated in plant development, growth, and stress response [2]. Several types of ncRNAs with different functions have been described, among which microRNAs (miRNAs) of 20 to 24 nt in length are known to control gene expression at the post-transcriptional level by inhibiting the accumulation or translation of specific mRNA targets [3]. Like protein-coding genes, the transcription of MIR genes is controlled by the action of the DNA-dependent RNA polymerase II (Pol II) along with general and specific transcription factors (TFs; extensively reviewed by [3]). A capped and polyadenylated primary miRNA transcript (pri-miRNA) is processed inside dicing bodies (D-bodies) formed by proteins such as DICER LIKE-1 (DCL1), double-stranded RNA-binding protein HYPONASTIC LEAVES (HYL1), and the structural protein SERRATE (SE), among others. Sequential processing originates the stemloop precursor (pre-miRNA) that eventually produces a mature miRNA associated with the ARGONAUTE 1 (AGO1) protein to assemble the RNA-induced silencing complex (RISC; broadly reviewed in [3,4]).
Most plant miRNAs have been studied in their mature form with a primary focus on their target prediction and validation [5][6][7]. Their contribution to plant development and growth has been denoted through regulating particular targets that act as key players in 2 of 20 controlling phase transition, biotic and abiotic stress response, nutrient homeostasis, and many other processes [8][9][10][11][12]. In contrast, just a few studies have focused on the regulation of MIR gene expression [13][14][15].
So far in maize (Zea mays), multiple miRNAs have been described as part of diverse cellular and developmental processes. One of them is zma-miR528, a monocot-specific miRNA, initially reported in rice. This miRNA displays a wide spectrum of functions by targeting distinct mRNA transcripts in different species [16][17][18][19][20][21]. Its function and transcriptional regulation have been more extensively studied in rice [15,17]. Os-miR528 is codified by a single gene and becomes highly accumulated upon vegetative to reproductive transition, opposite to the miR156 accumulation pattern [17]. In this work, the authors nicely demonstrated that Os-miR528 is regulated during plant development at the transcriptional level by the miR156 target Os-SPL9 and at the post-transcriptional level by aging-dependent alternative splicing of the pri-miRNA. They also found that cis-elements in miR528 gene promoters vary among different rice accessions, correlating with variable expression levels of the mature miRNA. Such complex and dynamic control of miR528 transcription awaits further exploration.
In maize, mature miR528 levels significantly increase during dry seed imbibition and drop upon germination completion [22]. It is also affected by external nitrogen supply, as seedlings grown under excess nitrogen conditions presented higher levels of this miRNA than those grown in a nitrogen-deficient media [19]. The regulation of specific miR528 targets, copper-containing laccase transcripts (ZmLAC3 and ZmLAC5), affects lignin biosynthesis and correlates with plants being prone to lodging when miR528 accumulates due to high nitrate growing conditions. Furthermore, higher miR528 levels were observed during maize embryo development and maturation [23], as well as over the course of embryo dedifferentiation [24] during maize somatic embryogenesis (SE). Likewise, in wellestablished embryogenic calli, miR528 regulates the abundance of target mRNAs involved in oxidative stress, auxin accumulation, and differentiation. However, its level is reduced upon hormone depletion during in vitro plantlet regeneration [25].
Even though miR528 works as a central regulatory component in several pathways [26], little is known about the regulatory elements driving its expression in maize. Here we found that during seed germination of the maize cultivar Tuxpeño VS-535, miR528 is mostly derived from the expression of the zma-MIR528a locus. Exploration of the zma-MIR528a promoter region revealed the presence of conserved transcription factor binding sites (TFBS) related to diverse signaling pathways, including the nitrate (TGA1/4) and auxin (AuxRE) response networks. Accumulation of both pre-miR528a and mature miR528 was up-regulated by exogenous nitrate and auxin treatments during imbibition, germination, and seedling establishment. Partial 5 promoter and specific TFBS deletion analyses demonstrated that TGA1/4 and AuxRE boxes are needed for transcriptional induction under both stimuli. Our findings contribute to understanding the regulatory network that controls zma-MIR528 expression in response to previously reported stimuli and settles specific miRNA accumulation profiles in maize. This work also broadens the knowledge of factors regulating MIR gene expression in plants.

Conservation of pre-miR528 in Monocots
To unveil the conservation among miR528 precursors in different species, we retrieved from PmiREN the sequences annotated as pre-miR528 in 21 monocot species, maize included (Supplementary Material, Table S1). Multiple sequence alignment of precursors revealed a high degree of conservation, especially for the miRNA-5p, as all species presented the same sequence. On the contrary, the miRNA-3p region displayed more differences across the analyzed sequences. In addition, the alignment revealed a few conserved bases in the stem region (Supplementary Material, Figure S1). Therefore, we performed a phylogenetic analysis illustrating the evolutive relationships between precursors among species. Pre-miR528a and pre-miR528b from maize clustered in a well-defined clade together with species from the Panicoideae subfamily, like Sorghum bicolor (Sbi), Panicum virgatum (Pvi), Setaria italic (Sit), and Saccharum hybrid cultivar ( Figure 1). Conversely, precursors, such as the one present in the banana (Mac-premiR528), showed a more distant evolutive relationship with sequences from maize, in agreement with previous reports [20]. The high degree of conservation is also noticeable by exploring the precursor folding, where sequences closely related to zma-premiR528a deployed mispairing at the 12th and 16th positions of the duplex region, leading to similar secondary structures (Supplementary Material, Figure S1). Altogether, these results indicate that precursors and mature miR528 are highly conserved among monocots at the sequence and structural levels, contrasting with their target diversity between species. differences across the analyzed sequences. In addition, the alignment revealed a few conserved bases in the stem region (Supplementary Material, Figure S1). Therefore, we performed a phylogenetic analysis illustrating the evolutive relationships between precursors among species. Pre-miR528a and pre-miR528b from maize clustered in a well-defined clade together with species from the Panicoideae subfamily, like Sorghum bicolor (Sbi), Panicum virgatum (Pvi), Setaria italic (Sit), and Saccharum hybrid cultivar ( Figure 1). Conversely, precursors, such as the one present in the banana (Mac-premiR528), showed a more distant evolutive relationship with sequences from maize, in agreement with previous reports [20]. The high degree of conservation is also noticeable by exploring the precursor folding, where sequences closely related to zma-premiR528a deployed mispairing at the 12th and 16th positions of the duplex region, leading to similar secondary structures (Supplementary Material, Figure S1). Altogether, these results indicate that precursors and mature miR528 are highly conserved among monocots at the sequence and structural levels, contrasting with their target diversity between species. Figure 1. Phylogenetic analysis of pre-miR528 sequences found in monocots. Phylogenetic reconstruction was made with MEGA X using the Maximum Likelihood (ML) method and the best-adjusted model (3-parameter Tamura) with 1000 bootstrap iterations. Twenty-eight monocot miR528 Figure 1. Phylogenetic analysis of pre-miR528 sequences found in monocots. Phylogenetic reconstruction was made with MEGA X using the Maximum Likelihood (ML) method and the best-adjusted model (3-parameter Tamura) with 1000 bootstrap iterations. Twenty-eight monocot miR528 precursor sequences were clustered in different subclades using the ggtree package in R studio. Arabidopsis thaliana miR408 precursor was used as an outlier with an equivalent function, as this miRNA shared several targets with miR528 [20].

Identification of Promoter cis-Acting Elements in zma-MIR528a
Although multiple miRNAs have been described in maize, regulatory regions for most MIR genes remain elusive. According to miRBase (release 22.1) there are two different precursors for miR528 in maize, corresponding to zma-MIR528a and zma-MIR528b located on chromosomes 1 and 9, respectively [21]. Our first approach for their promoter description was to experimentally identify the transcription start site (TSS) by taking advantage of 5 -CAP modification in the primary transcript (pri-miR528a/b). Decapping and adapter ligation of imbibed seed RNA followed by RT and nested PCRs showed a fragment of about 300 nt, absent in the un-decapped sample (minus TAP control, Figure 2A). Cloning and sequencing of the amplification products allowed us to define the TSS position on the annotated sequence for MIR528a. The amplification product corresponding to pri-miR528b was not recovered, even though the 5 RLM-RACE protocol was performed using primers matching both genes. Sequenced clones showed three alternative start sites with a preference for the adenine located 89 bp upstream of the pre-miR528a sequence previously annotated in databases ( Figure 2B). To corroborate our findings, we analyzed public CAGE-Seq (Cap Analysis Gene Expression and deep Sequencing) data [27]. We found a sharp cluster of CAGE-TSS reads upstream of the pre-miR528a sequence, with the dominant TSS position concurrent with our 5 RLM-RACE results (Supplementary Material, Figure S2A). miRNA shared several targets with miR528 [20].

Identification of Promoter cis-acting Elements in zma-MIR528a
Although multiple miRNAs have been described in maize, regulatory regions for most MIR genes remain elusive. According to miRBase (release 22.1) there are two different precursors for miR528 in maize, corresponding to zma-MIR528a and zma-MIR528b located on chromosomes 1 and 9, respectively [21]. Our first approach for their promoter description was to experimentally identify the transcription start site (TSS) by taking advantage of 5′-CAP modification in the primary transcript (pri-miR528a/b). Decapping and adapter ligation of imbibed seed RNA followed by RT and nested PCRs showed a fragment of about 300 nt, absent in the un-decapped sample (minus TAP control, Figure 2A). Cloning and sequencing of the amplification products allowed us to define the TSS position on the annotated sequence for MIR528a. The amplification product corresponding to pri-miR528b was not recovered, even though the 5′ RLM-RACE protocol was performed using primers matching both genes. Sequenced clones showed three alternative start sites with a preference for the adenine located 89 bp upstream of the pre-miR528a sequence previously annotated in databases ( Figure 2B). To corroborate our findings, we analyzed public CAGE-Seq (Cap Analysis Gene Expression and deep Sequencing) data [27]. We found a sharp cluster of CAGE-TSS reads upstream of the pre-miR528a sequence, with the dominant TSS position concurrent with our 5′ RLM-RACE results (Supplementary Material, Figure S2A). Once the TSS was defined, we analyzed the promoter sequences of zma-MIR528a and homologous genes from different species to find putative TFBS (Supplementary Material, Figure S3A; Table S2). Only five of the analyzed sequences (zma-MIR528a included) presented a TATA box core element. In contrast, a higher number of genes displayed a CATT regulatory box within their promoter sequence (Supplementary Material, Figure  S3B). Furthermore, several MIR528 promoter regions (~1500 bp upstream of TSS) displayed conserved binding motifs for TFs involved in cytokinin response (ARABIDOPSIS RESPONSE REGULATORS, ARR1), ABA signaling (ABA Response Elements, ABRE) [28], copper homeostasis (CuRE) [29], and vascular tissue formation and function (DNA-binding with one finger, Dof TFs) [30] ( Figure 2C; Supplementary Material, Figure S3B).
Notably, the zma-MIR528a promoter harbors two TGACG motifs known as binding sites for the TGACG-BINDING FACTORS 1/4 (TGA1/4), which are important regulatory components for nitrate responses in A. thaliana [31]. Also, the zma-MIR528a promoter presented one TGTCTC motif ( Figure 2C), defined as an auxin response element (AuxRE), recognized by AUXIN RESPONSE FACTOR (ARF) proteins [32]. Both nitrate and auxin signaling routes are directly involved in reported experimental models where the accumulation of zma-miR528 is significantly enhanced [19,33]. Considering this, the TGA1/4 and AuxRE elements were more closely inspected in this study.

Accumulation of pre-miR528a Decreases as Seedling Establishment Takes Place
To address how zma-MIR528a is transcriptionally regulated, the presence of precursor transcripts should be evaluated. Previous reports indicate that mature miR528 highly accumulates throughout maize embryonic development, with further increase upon seed imbibition and decline at post-germinative stages [21,22]. Thus, we explored whether miR528 precursors could be detected during germination and seedling establishment until 72 h after seed imbibition ( Figure 3A). Although precursors from both zma-MIR528a and zma-MIR528b were evaluated (positive amplification from genomic DNA; gDNA), only pre-miR528a (123 nt) was detected by end-point RT-PCR ( Figure 3B). Relative quantification by qRT-PCR showed that pre-miR528a abundance continuously drops from dry seed to initial seedling stages ( Figure 3C). By contrast, mature miR528 accumulated towards the first 24 h of imbibition and then went down ( Figure 3D), consistent with the aforementioned reports. Overall, these observations allowed us to select germination and seedling establishment as our working model to evaluate the effect of diverse stimuli upon zma-MIR528a expression by quantifying its precursor.

Exogenous Nitrate and Auxin Treatments during Maize Seed Imbibition Trigger Increases in pre-miR528a and Mature miR528 Levels
After establishing the experimental model, we tested two different exogenous treatments during germination and early seedling growth to evaluate their effect on the precursor and mature miR528 accumulation ( Figure 4A). Incubation with 30 mM KNO 3 positively impacted the seed germination rate ( Figure 4B top and Supplementary Material, Figure S4A). In agreement with previous reports [34], extended incubation with nitrate accelerated seedling growth and shoot expansion (Supplementary Material, Figure S4B). On the other hand, NPA-treated seedlings subjected to a short pulse (6 h) of the auxin NAA did not show evident growth differences with respect to its control (NPA alone) and contrasted with H 2 O imbibition ( Figure 4B bottom and Supplementary Material, Figure S4C). However, extended incubation with NAA promoted the emergence of multiple lateral roots (Supplementary Material, Figure S4D), confirming the reported physiological effect of exogenous auxin [35].
As expected according to previous studies [19], significant increments of pre-miR528a and miR528 were observed in the presence of nitrate solution relative to H 2 O at 24, 48, and 72 h upon seed imbibition ( Figure 4C). The treatment with NAA also increased their levels, with the highest accumulation observed at 4 and 6 h after the auxin pulse on seedlings pre-treated with NPA. On the other hand, inhibition of polar auxin transport by NPA alone did not cause significant changes on either precursor or mature miRNA levels, since at 72 h of imbibition they showed considerably reduced levels ( Figures 3C,D and 4D).
Owing to this, we tested if the observed increments were due to transcriptional events. As shown in Supplementary Material Figure S5, nitrate and auxin treatments involved transcriptional activity to promote the highest accumulation of pre-miR528a and mature miR528. The inclusion of alpha-Amanitin (inhibitor of RNA pol II, a-aman) exhibited a significant reduction in the levels of both molecules with respect to the treatment alone. Subsequently, we tested whether up-regulation of zma-MIR528a caused by nitrate and auxin altered the accumulation of previously confirmed mRNA targets [21]. The abundance of BASIC HELIX-LOOP-HELIX 152 (bHLH152; Zm00001d016873), MULTIDRUG AND TOXIC COMPOUND EXTRUSION/BIG EMBRYO 1 (MATE/BIGE1; Zm00001d012883), and SUPEROXIDE DISMUTASE 1a (SOD1a; Zm00001d031908) transcripts significantly decreased in samples treated either with KNO 3 or NAA (Supplementary Material, Figure  S6). Overall, these results show that nitrate and auxin trigger transcription of zma-MIR528a, resulting in significant precursor and mature miRNA accumulation as well as target downregulation during maize seed imbibition.  Oligonucleotides were designed to amplify both pre-miR528a (123 nt) and pre-miR528b (78 nt), as shown for genomic DNA (gDNA). (C,D) Quantification by real-time PCR of pre-miR528a and mature miR528 in maize embryonic axes at the indicated stages of seed imbibition. Fold change represents the abundance relative to dry seed (0 h) normalized either by 18S rRNA (precursor) or U6 snRNA internal control (mature miRNA). Error bars represent the standard error of the mean from three biological replicates with three technical replicates for each one (n = 9). Data were analyzed by one-way ANOVA with multiple comparisons by the Tukey post hoc test. Boxes that do not share at least one identical letter differ significantly (p < 0.05) from each other.

Exogenous Nitrate and Auxin Treatments during Maize Seed Imbibition Trigger Increases in pre-miR528a and Mature miR528 Levels
After establishing the experimental model, we tested two different exogenous treatments during germination and early seedling growth to evaluate their effect on the precursor and mature miR528 accumulation ( Figure 4A). Incubation with 30 mM KNO3 positively impacted the seed germination rate ( Figure 4B top and Supplementary Material, Figure S4A). In agreement with previous reports [34], extended incubation with nitrate accelerated seedling growth and shoot expansion (Supplementary Material, Figure S4B). Oligonucleotides were designed to amplify both pre-miR528a (123 nt) and pre-miR528b (78 nt), as shown for genomic DNA (gDNA). (C,D) Quantification by real-time PCR of pre-miR528a and mature miR528 in maize embryonic axes at the indicated stages of seed imbibition. Fold change represents the abundance relative to dry seed (0 h) normalized either by 18S rRNA (precursor) or U6 snRNA internal control (mature miRNA). Error bars represent the standard error of the mean from three biological replicates with three technical replicates for each one (n = 9). Data were analyzed by one-way ANOVA with multiple comparisons by the Tukey post hoc test. Boxes that do not share at least one identical letter differ significantly (p < 0.05) from each other.  The expression levels were normalized using 18S rRNA (for precursor) and U6 snRNA (for mature miRNA) accumulation. Error bars represent standard errors of means from at least three independent replicate experiments. * p < 0.05, ** p > 0.01, and *** p < 0.001 (Data were analyzed by one-way ANOVA with Tukey post hoc test comparing treated samples to control for each time point).

zma-MIR528a Promoter Harbors Regulatory Elements Responsible for Transcription Enhancement in Response to Nitrate and Auxins
Given that nitrate and auxin treatments triggered zma-MIR528a expression, we sought to analyze the regions within the promoter responsible for such induction. From our previous TFBS analysis, the presence of TGA1/4, AuxRE, and TATA boxes was considered to design nested 5 deletion segments of the promoter fused to the eGFP/GUS reporters ( Figure 5A). The resulting constructs were used to perform transient activity assays in maize protoplasts and showed activity for both reporters under the full-length promoter (pMIR_1, −1180 bp from TSS; Supplementary Material, Figure S7A,B). This indicated that the zma-MIR528a promoter works as an integral and functional sequence. However, as the sequence was narrowed, transcriptional activity decayed. Deletion of 814 nt (pMIR_3) and 1157 nt (pMIR_4) upstream of the TSS significantly reduced the promoter activity under control conditions (Supplementary Material, Figure S7C). Next, we assayed the activity of each construct when transfected protoplasts were incubated with nitrate (10 mM KNO 3 ) and auxin (1 µM NAA) treatments. An increase in GUS reporter expression under nitrate stimulus was observed for protoplasts transfected with the full-length (pMIR_1) and narrowed p_MIR2 (-843 bp from TSS) promoter versions ( Figure 5B). In contrast, only the full-length promoter (pMIR_1) presented significant induction with NAA ( Figure 5C). When nitrate and auxin were used in a combined treatment, the pMIR_1 and pMIR_2 constructs also significantly increased their transcriptional activity ( Figure 5D). Since the pMIR_2 construct lacks one TGA1/4 box and the AuxRE but still harbors the proximal TGA1/4 box, this result suggests that this motif is sufficient to induce expression at least in response to high nitrate concentration.

TFBS within the zma-MIR528a Promoter Contribute Differentially to Nitrate and Auxin Induction
Although transient expression assays evidenced that the zma-MIR528a promoter harbors cis-acting sequences involved in transcriptional activation in response to nitrate and auxin treatments, the contribution of each TGA1/4 box and the AuxRE element was not completely clear due to the likely presence of additional response elements within the

TFBS within the zma-MIR528a Promoter Contribute Differentially to Nitrate and Auxin Induction
Although transient expression assays evidenced that the zma-MIR528a promoter harbors cis-acting sequences involved in transcriptional activation in response to nitrate and auxin treatments, the contribution of each TGA1/4 box and the AuxRE element was not completely clear due to the likely presence of additional response elements within the promoter. Therefore, we tested the effect of individual TFBS deletions in the context of a full-length promoter. In contrast to sequential deletions, single TFBS deletions did not affect basal transcription of the reporter (mock in Figure 6), supporting a role of other elements within the region deleted in pMIR_2 that guide expression under control conditions. Interestingly, transcriptional induction under WT promoter was significantly greater for combined nitrate and auxin, as compared to single treatments ( Figure 6). Deleting the distal TGA1/4 site (−1065 position; ∆TGA1) still showed significant induction by nitrate, auxin, or combined treatments. Conversely, deletion of the TGA1/4 site at position −738 (∆TGA2) abolished the induction by either of the individual treatments but still allowed significant, although minor, increase by the combined action of nitrate and auxin ( Figure  6). Altogether, these results indicate that the TGA1/4 site proximal to the TSS plays a pivotal role in zma-MIR528a transcriptional activation by either nitrate or auxin. Finally, the mutation of the TGTCTC sequence (∆AuxRE, position −875) negatively affected only the induction by NAA alone, proving it is required for proper activation of the zma-MIR528a promoter by the auxin signaling pathway ( Figure 6).

ARF34 Contributes to Transcriptional Activation of the zma-MIR528a Promoter
To further explore the role of auxin response machinery on zma-MIR528a promoter activation, we co-transfected maize protoplasts with pMIR_1 or mutant reporter constructs along with an effector plasmid containing the ARF34 coding sequence (Zm00001eb031700, Figure 7A), which is a well-studied transcriptional activator ARF [36]. Significantly higher induction was evident for protoplasts co-transfected with this effector and pMIR_1 reporter ( Figure 7B and Supplementary Material, Figure S8). An increase in reporter activity was not detected for protoplasts co-transfected with ARF4 Figure 6. Differential contribution of TFBS within zma-MIR528a promoter to nitrate and auxin induction. eGFP-GUS reporter constructs under full-length zma-MIR528a promoter or promoters harboring deleterious mutations at either of two TGA1/4 (AS-1) or the AuxRE sites were transfected in maize protoplasts and assayed under control (mock), nitrate (10mM KNO 3 ), auxin (1 µM 1-NAA), or combined treatment conditions. Deleted TFBS are shown as red-crossed boxes in each construct. Values represent GUS activity determined by fluorometry. 35S:GUS and vector lacking a promoter region (Empty) were used as positive and negative controls, respectively. Data represent mean ± S.E.M. of at least three independent experiments. One-way analysis of variance (ANOVA) and post hoc Tukey's test were used to determine the significant difference between treatments and mock for each construct. (* p < 0.05; ** p < 0.01).
Interestingly, transcriptional induction under WT promoter was significantly greater for combined nitrate and auxin, as compared to single treatments ( Figure 6). Deleting the distal TGA1/4 site (−1065 position; ∆TGA1) still showed significant induction by nitrate, auxin, or combined treatments. Conversely, deletion of the TGA1/4 site at position −738 (∆TGA2) abolished the induction by either of the individual treatments but still allowed significant, although minor, increase by the combined action of nitrate and auxin ( Figure 6). Altogether, these results indicate that the TGA1/4 site proximal to the TSS plays a pivotal role in zma-MIR528a transcriptional activation by either nitrate or auxin. Finally, the mutation of the TGTCTC sequence (∆AuxRE, position −875) negatively affected only the induction by NAA alone, proving it is required for proper activation of the zma-MIR528a promoter by the auxin signaling pathway ( Figure 6).

ARF34 Contributes to Transcriptional Activation of the zma-MIR528a Promoter
To further explore the role of auxin response machinery on zma-MIR528a promoter activation, we co-transfected maize protoplasts with pMIR_1 or mutant reporter constructs along with an effector plasmid containing the ARF34 coding sequence (Zm00001eb031700, Figure 7A), which is a well-studied transcriptional activator ARF [36]. Significantly higher induction was evident for protoplasts co-transfected with this effector and pMIR_1 reporter ( Figure 7B and Supplementary Material, Figure S8). An increase in reporter activity was not detected for protoplasts co-transfected with ARF4 (Zm00001eb067270) activator [36], suggesting that the zma-MIR528a promoter could be particularly responsive to ARF34. Furthermore, induction was observed for co-transfections with ∆TGA1 or ∆TGA2 constructs, but not with the construct without the AuxRE site (∆AuxRE + ARF34; Figure 7B). Interestingly, the induction was significantly lower for ∆TGA2 than the one observed for pMIR_1 or ∆TGA1 reporters. This suggests that the proximal TGA1/4 and AuxRE motifs might cooperate for full auxin inducibility of the zma-MIR528a promoter. Finally, the presence of a 1 µM NAA further increased the transcriptional activity of the wild-type promoter (pMIR_1) in the presence of ARF34 ( Figure 7C), supporting the involvement of additional auxin signaling pathway components in zma-MIR528a regulation. Albeit our results exposed the relevance of AuxRE and proximal TGA1/4 sites for NAA-mediated activation of the zma-MIR528a promoter, direct interaction between these elements and ARF34 or other transcription factors will need further experimental demonstration. (Zm00001eb067270) activator [36], suggesting that the zma-MIR528a promoter could be particularly responsive to ARF34. Furthermore, induction was observed for co-transfections with ∆TGA1 or ∆TGA2 constructs, but not with the construct without the AuxRE site (∆AuxRE + ARF34; Figure 7B). Interestingly, the induction was significantly lower for ∆TGA2 than the one observed for pMIR_1 or ∆TGA1 reporters. This suggests that the proximal TGA1/4 and AuxRE motifs might cooperate for full auxin inducibility of the zma-MIR528a promoter. Finally, the presence of a 1 µM NAA further increased the transcriptional activity of the wild-type promoter (pMIR_1) in the presence of ARF34 ( Figure  7C), supporting the involvement of additional auxin signaling pathway components in zma-MIR528a regulation. Albeit our results exposed the relevance of AuxRE and proximal TGA1/4 sites for NAA-mediated activation of the zma-MIR528a promoter, direct interaction between these elements and ARF34 or other transcription factors will need further experimental demonstration.

Discussion
Although several miRNAs have been studied in plants, there is a lack of information about promoter regions and regulatory elements that coordinate their expression. Their analysis will help to better understand their regulation and how different stimuli control their expression. We experimentally validated the transcription start site (TSS) for the zma-MIR528a gene at 89 bp upstream from the mapping site of the precursor pre-miR528a. Comparable lengths have been reported for A. thaliana and rice MIR genes, as most miRNAs display less than 200 nt between the TSS and the stem-loop-structured precursor [37]. We found additional start sites adjacent to the predominant TSS, consistent with next-generation data previously reported [27]. Narrow clustering of multiple TSS is a common feature among genes with tissue-specific expression [27], which could explain the specific accumulation patterns reported for miR528 in the vascular tissue of maize leaves, stems, and immature embryos [19,21,38]. The TSS determination enabled us to limit and more accurately scrutinize the promoter region for putative cis-regulatory elements (CREs). Interestingly, both zma-MIR528a and zma-MIR528b displayed core promoter regulatory elements analogous to those present in other MIR genes found in rice [39] and A. thaliana [40]. However, only zma-MIR528a presented a TATA box element nearby the TSS region. This element has been reported to affect the promoter strength and increase the expression of endogenous and synthetic genes in plants [41] and is over-represented in MIR plant promoters [42]. Therefore, we hypothesized that TATA box presence in MIR528a could be involved with the differential expression rates between both MIR528 genes in maize, explaining our lack of detection for pre-miR528b. Also, expression of MIR528b may be restricted to particular tissues or developmental stages, even though diverse expression data sources have only collected reads mapping against the MIR528a genomic region [43][44][45]. Moreover, numerous CAAT sites were found close to the TSS and throughout the promoter for both genes in maize. CAAT boxes are conserved sites believed to determine the efficiency of transcription [46]. Indeed, the insertion of multiple CAAT boxes is a proposed model for acquisition of promoter activity for some MIR genes such as MIR1444, MIR058, and MIR12112 in Vitis vinifera [47]. In addition, these core promoter elements were conserved in homologous genes, which suggests they may have a general role in guiding the expression of MIR528 in monocots.
A higher degree of conservation was found among pre-miRNAs at the sequence and structural levels, as each precursor leads to identical mature miRNA in the analyzed sequences. This shared secondary structure might influence the processing and the accumulation of miR528 among species, as precise cleavage of pre-miRNA requires structural and sequence determinants, but their relative contributions are still being studied [48,49].
The presence of specific TFBS within the promoter provided hints about the metabolic and physiological pathways commanding zma-MIR528a expression. We found two TGA1/4 (also known as activation sequence-1, AS-1) binding motifs in the zma-MIR528a promoter that were highly conserved in other MIR528 genes. The TGA1/4 transcription factors are members of the basic leucine zipper (bZIP) family and participate in complex transcriptional networks associated with nitrate response and salicylic acid biosynthesis [31,50,51]. zma-MIR528a also exhibited an auxin response element (AuxRE), even though its presence was less conserved among homologs. The auxin responsiveness is achieved by the interaction of ARFs with AuxREs at promoter regions of diverse genes involved in several developmental processes [32,52,53]. TGA1/4 and AuxRE sites are found in the promoters of several Zea mays sequenced accessions (Supplementary Material, Figure S9A) supporting their relevance for zma-MIR528a expression regulation in response to external stimuli. In addition, multiple TFs that potentially recognize TGA1/4 and AuxRE sites exhibited expression profiles similar to MIR528a expression in distinct tissues (Supplementary Material, Figure S9B). Accordingly, we found that high nitrate conditions and exogenous auxin application increased the zma-MIR528a precursor and mature miRNA levels at early seedling stages of Tuxpeño VS-535 maize. Previous reports indicate that mature miR528 levels are affected in maize roots under nitrate starvation [19,38], while nitrogen luxury enhanced its expression [19] in different maize lines. High nitrate and auxin concentrations are commonly used for callus induction and proliferation during maize somatic embryogenesis (SE), which might explain the high miR528 abundance in dedifferentiated tissues [21,25,33].
The responsiveness of miRNAs to nitrate and auxin is well-documented in diverse plants and experimental models [38,54]. For instance, treatments with indoleacetic acid (IAA) promote transcription of ath-MIR393b in aerial seedling organs [55]. Furthermore, miR393 also accumulates two hours after incubation in 5 mM potassium nitrate solution and induces target down-regulation [56,57]. Interestingly, during SE induction in A. thaliana, expression of MIR393a and MIR393b increase to enable explant sensitivity to auxins [58], which mirrors the miR528 response in maize SE.
Transient expression assays with the zma-MIR528a promoter confirmed that elements responsive to nitrate and auxins are localized within the analyzed sequence. The distribution of TGA1/4 sites in zma-MIR528a is similar to footprints detected in promoters of genes transcriptionally activated by nitrate in Arabidopsis [59]. However, only the removal of the TGA1/4 site nearest to the TSS (∆TGA2) significantly affected the transcription enhancement by external stimuli, evidencing a differential contribution of each site. This is congruent with studies reporting that higher gene expression correlates with TGA1/4 binding motifs located closer to TSS [59,60]. In addition, our results suggest that the proximal TGA1/4 site in the zma-MIR528a promoter also cooperates with auxin induction. Enhancement of activity by NAA was not observed for the ∆TGA2 construct, in spite of the presence of AuxRE. This is not surprising, since several TFs that bind the TGACG motif guide gene expression in response to synthetic auxins, methyl jasmonate, and H 2 O 2 , acting as molecular linkers between different signaling pathways [61][62][63][64].
Early auxin response genes tend to have two or more adjacent TGTCTC/TGTCGG motifs separated by a variable spacer sequence to allow ARFs' dimerization and proper gene activation [32,65,66]. Nonetheless, nearly 86% of the ARFs' targeted sites in maize are composed of two or more TGTC core motifs with less than 50 nucleotides of spacing [36,67]. Interestingly, the zma-MIR528a promoter displays a TGTC core motif located 36 nt downstream of the analyzed AuxRE site, presenting architectural features analogous to other auxin-induced genes in maize [36]. Transcriptional activator ARF34 modulates the activation of the zma-MIR528a promoter through the AuxRE motif, as shown by our transactivation assays in protoplasts. Hence, alteration of the architecture comprising AuxRE and TGTC sites might abolish auxin signaling over zma-MIR528a. Similar results were reported for Ath-MIR390a, as deletion of a 36 bp portion containing an inverted AuxRE motif suppressed the response of this gene to auxin treatments [14].
Overall, the analysis of the zma-MIR528a promoter suggests that nitrate and auxin pathways cooperate to induce MIR528a expression (Figure 8). This synergism may occur due to an overlap of the molecular regulatory routes previously reported for NO 3 − and auxins. The primary nitrate response proceeds through master regulators (TCP-NLP) that directly control TFs such as TGA1/4 to localize their binding sites in promoters of target genes. They are also implicated in auxin homeostasis and signaling by regulating TRYPTOPHAN AMINOTRANSFERASE OF ARABIDOPSIS 1 (TAA1), involved in IAA biosynthesis [68], and promoting the accumulation of AUXIN SIGNALING F-BOX 3/TRANSPORT INHIBITOR RESPONSE 1 (AFB3/TIR1), required for degradation of ARFs' repressor Aux/IAA [68,69]. Considering this, the co-existence of high auxin and nitrate conditions enhances zma-MIR528a expression through localizing TGA1/4 and ARF TFs at cognate motifs in the promoter region, which results in the reduction of mRNA targets levels ( Figure 8). Such outcomes have been reported for the stimuli studied in this work, resulting in maize plant lodging under high nitrate growth conditions [19] or callus induction and proliferation from immature embryos [21,25]. Moreover, auxins accumulate during late embryo development and regulate seed dormancy acquisition in several plants [70,71]. This auxin increment might be guiding the transcriptional activation of zma-MIR528a, as the precursor is highly accumulated in dry maize seeds. However, additional regulatory routes, such as the ROS or the cold stress response pathways could also control zma-MIR528a expression by activating other cis-regulatory elements. Furthermore, the regulatory mechanism might depend on the species, developmental stage, or condition being studied [15,17]. additional regulatory routes, such as the ROS or the cold stress response pathways could also control zma-MIR528a expression by activating other cis-regulatory elements. Furthermore, the regulatory mechanism might depend on the species, developmental stage, or condition being studied [15,17]. Both stimuli activate MIR528a expression through TFs' interaction with their binding sites. As a consequence of such coordinated regulation, miR528 level increases to down-regulate its mRNA targets.
Transcriptional activation of zma-MIR528a during late embryogenesis correlates with high pre-miR528 levels in dry seeds and mature miR528 transient accumulation upon seed imbibition, followed by rapid decrease after germination. This is accompanied Figure 8. A proposed model for nitrate-and auxin-mediated transcriptional regulation over zma-MIR528a. Perception of high nitrate triggers primary signaling guided by master regulators to activate several TFs. TGA1, TGA4, and other TFs would recognize cis-regulatory elements in the promoter of zma-MIR528a to enhance its expression in response to the stimulus. Furthermore, elevated auxin concentration promotes the activation of AFB3 and TIR1 to polyubiquitinate the repressor Aux/IAA, setting free the clade A ARFs (transcriptional activators, such as ARF34) to interact with AuxRE at the promoter and stimulate transcription. Both pathways could overlap through TCP/NLP-mediated regulation of both auxin biosynthesis and response pathways. The nitrate signal induces auxin biosynthesis (dashed lines) as master regulators of this route promote the expression of genes involved in tryptophan metabolism to obtain IAA. These regulators and the nitrate-sensing machinery also control the nitrate-AFB3/TIR1 auxin perception to activate ARFs (solid line). Both stimuli activate MIR528a expression through TFs' interaction with their binding sites. As a consequence of such coordinated regulation, miR528 level increases to down-regulate its mRNA targets.
Transcriptional activation of zma-MIR528a during late embryogenesis correlates with high pre-miR528 levels in dry seeds and mature miR528 transient accumulation upon seed imbibition, followed by rapid decrease after germination. This is accompanied by fine target regulation during the process [21], which possibly contributes to appropriate developmental and stress responses in early germination [72]. However, zma-MIR528a knock-out maize mutants are required to confirm its physiological role in these processes. Nevertheless, this work provides critical information on the transcription of a developmentally relevant microRNA and identifies the key TF binding sites that contribute to the auxin and nitrogen response of MIR528a. Future work should also explore additional layers of regulation for MIR528 expression to increase knowledge about the relationship between factors determining miR528 accumulation and target regulation, as well as the implication of complex regulatory networks on the growth and development of different monocot plants.

Plant Material and Growth Conditions
The Mexican maize (Zea mays L.) cultivar Tuxpeño VS-535 was used due to its agronomic relevance and the available previous characterization of miR528 in SE [21]. Seeds were disinfected with 6% commercial NaClO solution under constant agitation (250 rpm) for about 5 min and rinsed five times with 100 mL of fresh sterilized water. Imbibition, vertical germination, and seedling establishment were performed as previously reported [35]. Briefly, seeds were placed in paper towel rolls and incubated vertically in cylindrical containers with a 12 h light/12 h dark photoperiod at 28 • C. Nitrate stimulus application was conducted following previous reports [19] with some modifications. Rolls were soaked in 150 mL of either sterile water (control condition; mock) or 30 mM KNO 3 solution for 72 h, with sampling after 24, 48, and 72 h of continuous treatment. For exogenous auxin addition, experiments were performed as recommended by [35]. Seeds were imbibed for 72 h in water (mock) or 50 µM N-1-naphthylphthalamic acid (NPA, auxin transport inhibitor). After this time, a portion of the NPA-treated seedlings was transferred to 50 µM 1-Naphthalene Acetic Acid (NAA), and samples were collected at 0, 2, 4, and 6 h after the NAA pulse. Embryo axes and coleoptiles were extracted from seeds and seedlings and immediately frozen in liquid nitrogen until further use. Extended treatment with 50 µM NAA for 48 h was performed for phenotype registration. Controls remained under H 2 O or 50 µM NPA alone for each time point. All reagents were purchased from Merck, Sigma-Aldrich, St. Louis, MO, unless otherwise stated.

RNA Isolation, Size Fractionation, and Purification
Total RNA was extracted from collected tissues using TRIzol reagent (Invitrogen, Thermo Fisher Scientific Inc. Waltham, MA, USA) as previously reported [73]. DNase I treatment, size fractionation, and concentration of large (>200 nt) and small (<200 nt) RNAs were performed using the RNA Clean and Concentrator kit (Zymo Research, Irvine, CA, USA) following the manufacturer's instructions.

Precursor, Mature microRNA, and Target Quantification by RT-qPCR
For pre-miRNA detection, 2 µg of the fraction of small RNAs (<200 nt) were polyadenylated using the Poly(A) tailing kit (Invitrogen, Waltham, MA, USA). Reverse transcription (RT) reaction was carried out with oligo (dT) and the Improm-II reverse transcription system (Promega, Madison, WI, USA). Primers for each miR528 precursor were designed using Primer3Plus [74], with qPCR settings activated. Both precursors (pre-miR528a/b) were accessible for amplification using a combination of premiR_Fw and preMIR528_Rv3 oligonucleotides, as they were designed to align to conserved sites presented in both sequences (Supplementary Material, Table S3). For the mature miRNA, previously reported stem-loop and specific forward primers were used along with pulsed stem-loop RT reactions for miR528 and U6 snRNA using the sRNA fractions [21]. For target quantification, total RNA was reverse transcribed (RT) using oligo (dT) and the Improm-II reverse transcription system (Promega, Madison, WI, USA). For each target, previously reported primers were used [21]. qPCR was performed using the Maxima SYBR Green/ROX qPCR Master Mix in a 7500 Real-time PCR System (Applied Biosystems, Bedford, MA, USA). Relative expression was calculated by the 2 −∆∆Ct method with normalization to 18S rRNA (precursor and targets) or U6 snRNA (for miRNA) as internal housekeeping gene controls. Data from each experiment (three independent experiments with three technical replicates each) were subjected to one-way analysis of variance (one-way ANOVA) with Tukey's Multiple Comparison post hoc test for statistical significance.

Experimental Validation of the Transcription Start Site (TSS) by 5 RLM-RACE
Mapping of TSS was performed following a reported protocol [75] with slight modifications. About 10 µg of total RNA was treated with the QuickCIP enzyme (NEB, Ipswich, MA, USA) at 37 • C for 1 h. Then, 15 µL of 5 M ammonium acetate was added to stop each reaction; treated RNA was purified by phenol-chloroform extraction and isopropanol precipitation. Decapping enzyme (NEB, Ipswich, MA) was used for 5 -CAP removal of the treated RNAs. Minus TAP reactions were also assembled as ligation controls. Ligation of 5 RACE adapter (Supplementary Material, Table S3) was performed employing 2 µL of CIP-treated and decapped RNA and 2 µL of T4 RNA ligase (2.5 U/µL) for 1 h at 37 • C. The ligation products were reverse transcribed using oligo (dT) and the Improm-II reverse transcription system (Promega, Madison, WI, USA). Consecutive amplification reactions (nested PCR) were conducted using adapter-specific and gene-specific primers (Supplementary Material, Table S3). The final PCR products were separated by agarose gel electrophoresis and purified using the Wizard ® SV Gel and PCR Clean-Up System (Promega, Madison, WI, USA). Finally, the purified product was cloned into the pGEM-T easy vector (Promega Madison, WI, USA), and the resulting clones were sequenced.

Phylogenetic Analysis of miR528 Precursors and Promoters in Monocots
Precursor sequences of miR528 were obtained for 21 monocot species (Supplementary Material, Table S1) using miRBase release 22.1 [76] and the Plant microRNA Encyclopedia PmiREN version 2.0 [77]. Upstream regions (≈2000 nt from the precursor mapping site) were also retrieved from the Ensembl Plant database release 52 [78] for promoter homology comparisons. Sequences were aligned using the nucleotide-optimized MUSCLE algorithm [79] with 50 iteration cycles. The phylogenetic reconstructions were conducted using the MEGA X [80] and Maximum Likelihood (ML) method with the best-adjusted model (3-parameter Tamura) and 1000 bootstrap replications. In addition, folding of some precursor sequences was predicted using the webserver RNAFold from the University of Vienna (http://rna.tbi.univie.ac.at/cgi-bin/RNAWebSuite/RNAfold.cgi, accessed on 9 May 2022).

Generation of Plasmid Constructions
Genomic DNA was extracted from maize seedlings using a reported CTAB method [83]. Specific primers for the full-length MIR528a promoter (pMIR_1; Supplementary Material, Table S3) were used to amplify the region between the position −1510 upstream of TSS and the precursor sequence (209 nt beyond TSS). Amplification was accomplished using the Phusion High-Fidelity DNA Polymerase (NEB, Ipswich, MA, USA) following the manufacturer's protocol. The amplicon was gel-purified using the Wizard ® SV Gel and PCR Clean-Up System (Promega, Madison, WI, USA) kit and cloned into the pGEM-T easy vector (Promega, Madison, WI, USA) for sequencing. Sequential deletion fragments of the promoter were amplified with specific primers (pMIR_1-4; Supplementary Material, Table S3) using the pMIR_1 plasmid as template and were cloned into the pENTR-D-TOPO vector (ThermoFisher, Waltham, MA, USA). Proper orientation was verified by PCR. Correct entry clones were used for sub-cloning into the pBGWFS7.0 vector employing the Gateway LR Clonase II Enzyme Mix (Invitrogen, Waltham, MA, USA). The integrity of constructs and reporter genes was assayed by sequencing. Deletion of binding sites (TGA1/4 and AuxRE) was achieved by site-directed mutagenesis using the overlapping PCR method, as reported by [84]. WT and mutant expression cassettes were cloned into the pGEM-T easy vector for Sanger sequencing and subsequent promoter activity assays.
Effector plasmids for co-transfection assays were obtained by Gateway recombination of the entry clones pUT6075 and pUT3104 (acquired from the ABRC) corresponding to ZmARF34 (Zm00001eb031700) and ZmARF4 (Zm00001eb067270), respectively, with pEarleyGate 102 as the destination vector using the LR-clonase enzyme mix (Invitrogen, Waltham, MA, USA). Correct insertion and integrity of the corresponding ARF coding sequences downstream of the CaMV 35S promoter were checked by Sanger sequencing.

Maize Leaf Protoplast Isolation and Transfection
Protoplast isolation was carried out following the protocol reported by [85] with modifications. Briefly, the fully expanded second leaves from etiolated maize plants were cut into ≈ 0.5 cm strips. Leaf strips were soaked, vacuum infiltrated with the enzyme solution (1.5% cellulase R10, 0.75% macerozyme R10 (Yakult, Tokyo, Japan), 0.6M mannitol, 10 mM MES (pH 5.7), 10 mM CaCl 2 , and 0.1% BSA), and incubated for 3 h with constant agitation (50 rpm) in darkness. Protoplasts were collected by filtration and centrifugation (200× g for 4 min), washed out twice with W5 solution (154 mM NaCl, 125 mM CaCl 2 , KCl 5 mM, and 2 mM MES pH5.7), and resuspended into cold MMg solution (15 mM MgCl 2 , 0.4 M mannitol, 4 mM MES, pH 5.7) to a final concentration of 10 6 mL −1 protoplasts. For transfection, about 2 × 10 5 protoplasts were combined with 50 µg of each plasmid DNA and PEG transfection solution (40% w/v PEG 4000, 0.6 M mannitol, and 0.1 M CaCl 2 ) and incubated for 30 min at 25 • C covered from light. Next, the transfected cells were washed with W5 buffer and incubated at 25 • C in the dark for 16 h for basal expression detection. To avoid protoplast swelling and bursting, lower concentrations of each treatment (10 mM KNO 3 , 1 µM NAA, or both) were applied for the last 6 h of incubation as reported by [86] and [87]. After this, the protoplasts were harvested for further analysis. Transfection efficiency and normalization were assessed by absolute qPCR based on the number of copies of recombinant DNA successfully delivered into protoplasts after transfection, as previously reported [88].

Promoter Activity Analysis
Promoter activity was determined for each construct by measuring GUS enzymatic activity with 4-Methylumbelliferyl-β-D-glucuronide (4-MUG) as substrate. After incubation and treatments, protoplasts were harvested, and protein extracts were collected following the protocol reported by [89]. Fluorescence was measured based on a 4-methylumbelliferone (4-MU) standard curve using the Varioskan LUX Multimode microplate reader (Ther-moFisher, Waltham, MA, USA). In each sample, the total protein concentration was estimated by Bradford assay. GUS activity was calculated based on the 4-MU standard curve and expressed in nmol 4-MU min −1 mg −1 .
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/ijms232415718/s1 Figure S1. Sequence and structure conservation between miR528 precursors. Figure S2. CAGE TSS data analysis and the promoter sequence of zma-MIR528a. Figure S3. Conserved cis regulatory elements present in MIR528 promoters in monocots. Figure S4. Physiological effects of the application of nitrate or auxin treatment during maize seed germination and seedling establishment. Figure S5. The up-regulation of pre-miR528a and mature miR528 levels by nitrate and auxin is significantly reduced by inhibition of RNA pol II activity. Figure  S6. Negative impact of nitrate and auxin treatments on miR528 target accumulation. Figure S7. Zma-MIR528a promoter drives basal reporter expression in maize protoplasts. Figure S8. Protoplasts co-transfection assays. Figure S9. The TATA box, TGA1/4, and AuxRE sites of the Zma-MIR528a promoter are conserved among lines. Table S1. miR528 precursors genomic localization in monocots using PmiREN database. Table S2. Putative binding sites identified in the zma-MIR528a promoter using NewPLACE database. Table S3. Oligonucleotides used in the work. Funding: This work was funded by Universidad Nacional Autónoma de México: PAPIIT 218921, and Facultad de Química PAIP 5000-9118. Neither institution had any role in the study design, collection, analysis, or interpretation of the data, writing of the manuscript, or the decision to submit the paper for publication.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.