SPL12 Regulates AGL6 and AGL21 to Modulate Nodulation and Root Regeneration under Osmotic Stress and Nitrate Sufficiency Conditions in Medicago sativa

The highly conserved plant microRNA, miR156, affects root architecture, nodulation, symbiotic nitrogen fixation, and stress response. In Medicago sativa, transcripts of eleven SQUAMOSA-PROMOTER BINDING PROTEIN-LIKE, SPLs, including SPL12, are targeted for cleavage by miR156. Our previous research revealed the role of SPL12 and its target gene, AGL6, in nodulation in alfalfa. Here, we investigated the involvement of SPL12, AGL6 and AGL21 in nodulation under osmotic stress and different nitrate availability conditions. Characterization of phenotypic and molecular parameters revealed that the SPL12/AGL6 module plays a negative role in maintaining nodulation under osmotic stress. While there was a decrease in the nodule numbers in WT plants under osmotic stress, the SPL12-RNAi and AGL6-RNAi genotypes maintained nodulation under osmotic stress. Moreover, the results showed that SPL12 regulates nodulation under a high concentration of nitrate by silencing AGL21. AGL21 transcript levels were increased under nitrate treatment in WT plants, but SPL12 was not affected throughout the treatment period. Given that AGL21 was significantly upregulated in SPL12-RNAi plants, we conclude that SPL12 may be involved in regulating nitrate inhibition of nodulation in alfalfa by targeting AGL21. Taken together, our results suggest that SPL12, AGL6, and AGL21 form a genetic module that regulates nodulation in alfalfa under osmotic stress and in response to nitrate.


Introduction
Whereas root system architecture in plants is critical, because of its role in controlling nutrient cycling, water use efficiency, and resistance to biotic and abiotic stresses, balanced nutritional elements can also be helpful in combatting environmental stress [1,2]. The availability of nitrogen is a major limiting factor determining crop growth and productivity [3]. Although synthetic nitrogen fertilizers have increased crop yields, they are disadvantaged by their significant human health and environmental costs [4]. Among nitrogen-fixing crops, alfalfa (Medicago sativa L.) is the most widely cultivated forage crop around the world [5], and is one of the most important crops commercially grown for various purposes, including crop rotations, livestock feed, and soil improvement. This is due in part to its ability to form a symbiotic relationship with rhizobial bacteria [6][7][8]. The symbiotic nitrogen fixation of legumes takes place in root nodules [9], where rhizobia reduce atmospheric di-nitrogen to ammonia by expressing the nitrogenase enzyme [10].
As nitrogenase is exceptionally rich in sulfur [11], this element becomes limiting during symbiosis. There is a high demand for sulfur in nodulated legumes, and nitrogen fixation is more sensitive to sulfur deficiency than to nitrate uptake [12]. An abundant supply of sulfur in plants markedly increases nodulation and nitrogen fixation [12,13]. On the other hand, sulfur-deficiency in plants leads to a decrease in nodulation, nodule metabolism, and nitrogenase biosynthesis and activity, presumably because of the low-availability of   3 (n = 14-25 plants). * indicates significant differences relative to WT using Student's t-test (p < 0.05). Error bar indicates standard deviation.

Effect of Nitrate on Expression of SPL12 and AGL21
To shed light on the possible role of SPL12 in nitrate inhibition of nodulation, we determined whether the transcript levels of SPL12 and AGL21 were regulated by nitrate. AGL21 in alfalfa is closely related to the ANR1 clade in Arabidopsis (Figure 2A), and AtANR1, a member of the ANR1 clade, is involved in the nitrate regulation of root development in Arabidopsis [39,40]. Additionally, AtAGL21, another member of this clade, is upregulated by nitrogen deprivation [41]. Based on these findings, we hypothesized that AGL21 is involved in the nitrate regulation of nodulation in alfalfa. To test this hypothesis, we investigated changes in the transcript levels of AGL21 and SPL12 under nitrate treatment ( Figure 2B). The results showed that the transcript levels of AGL21 were increased after 5 h of nitrate treatment in WT plants, but the transcript levels of SPL12 was not affected throughout the treatment period. Based on these results, we propose that SPL12 is involved in regulating nitrate inhibition of nodulation in alfalfa by targeting AGL21. under nitrate treatment (Figure 2b). The results showed that the transc were increased after 5 hrs of nitrate treatment in WT plants, but the SPL12 was not affected throughout the treatment period. Based o propose that SPL12 is involved in regulating nitrate inhibition of nod targeting AGL21.  (B) Relative gene transcript levels of SPL12 and AGL21 were analyzed in WT at 0, 5, and 24 h after 20 mM nitrate treatment. Statistical groupings across different time points were determined separately for SPL12 and AGL21 transcript abundance using one-way ANOVA with the Tukey HSD post hoc test. Significant difference in post hoc Tukey multiple comparisons test is indicated with different letters ("a" for SPL12 expression level and "p, q, and r" for AGL21 expression level) Error bar indicates standard deviation. To investigate the expression profile of SPL12 in alfalfa, we measured its transcript levels in various organs (leaf, stem, and root) of 21-day-old WT alfalfa plants. The transcript levels of SPL12 were detected at similar levels in all three tissues ( Figure 3A). The transcript levels of AGL21 were also determined in the same tissues, and found to be nearly undetectable in leaf and stem tissues and highly expressed in roots ( Figure 3B). The low level of AGL21 in the leaf is consistent with previous reports that the Arabidopsis ANR1-like genes were expressed primarily in roots [45]. AGL79 expression was also nearly undetectable in leaf tissues in Arabidopsis [46]. letters ("a" for SPL12 expression level and "p, q, and r" for AGL21 expression level) Error bar indicates standard deviation.
To investigate the expression profile of SPL12 in alfalfa, we measured its transcript levels in various organs (leaf, stem, and root) of 21-day-old WT alfalfa plants. The transcript levels of SPL12 were detected at similar levels in all three tissues (Figure 3a). The transcript levels of AGL21 were also determined in the same tissues, and found to be nearly undetectable in leaf and stem tissues and highly expressed in roots (Figure 3b). The low level of AGL21 in the leaf is consistent with previous reports that the Arabidopsis ANR1-like genes were expressed primarily in roots [45]. AGL79 expression was also nearly undetectable in leaf tissues in Arabidopsis [46].

SPL12 is a Direct Regulator of AGL21
Previous transcriptomic analysis of miR156 overexpression (miR156-OE) plant A17, revealed 8,373 differentially expressed genes between roots of WT and miR156-OE [27]. Of the many genes differentially expressed in miR156-OE plant A17 relative to WT, AGL21 (MS.gene069166, MS.gene068633, MS.gene70086, and MS.gene027842), a gene that encodes a yet-to-be characterized alfalfa MADS box protein was significantly upregulated in A17 [27]. This gene had also significantly high transcript levels in SPL12-RNAi based on the transcriptomic analysis of SPL12-RNAi genotypes [29], which was confirmed by RT-qPCR ( Figure 4a). These results suggested that AGL21 might be regulated by SPL12, thus, further characterization was carried out by ChIP-qPCR using 35S::SPL12m-GFP plants to determine if AGL21 is a direct target of SPL12. DNA sequence analysis revealed that the promoter (2000 bp) of alfalfa AGL21 has four putative SPL binding sequences with a core GTAC sequence that are distributed in three regions (I, II, III) (Figure 4b), and three of them (in regions I and III) possess the typical NNGTACR SBP-binding consensus ( Figure S2). The three regions were tested for SPL12 occupancy using ChIP-qPCR analysis of 35S::SPL12m-GFP plants. Compared to WT, 35S::SPL12m-GFP plants showed significantly higher SPL12 binding to the AGL21 promoter region (Figure 4c), and occupancy in the three regions was substantially higher than that at the negative control

SPL12 Is a Direct Regulator of AGL21
Previous transcriptomic analysis of miR156 overexpression (miR156-OE) plant A17, revealed 8,373 differentially expressed genes between roots of WT and miR156-OE [27]. Of the many genes differentially expressed in miR156-OE plant A17 relative to WT, AGL21 (MS.gene069166, MS.gene068633, MS.gene70086, and MS.gene027842), a gene that encodes a yet-to-be characterized alfalfa MADS box protein was significantly upregulated in A17 [27]. This gene had also significantly high transcript levels in SPL12-RNAi based on the transcriptomic analysis of SPL12-RNAi genotypes [29], which was confirmed by RT-qPCR ( Figure 4A). These results suggested that AGL21 might be regulated by SPL12, thus, further characterization was carried out by ChIP-qPCR using 35S::SPL12m-GFP plants to determine if AGL21 is a direct target of SPL12. DNA sequence analysis revealed that the promoter (2000 bp) of alfalfa AGL21 has four putative SPL binding sequences with a core GTAC sequence that are distributed in three regions (I, II, III) ( Figure 4B), and three of them (in regions I and III) possess the typical NNGTACR SBP-binding consensus ( Figure S2). The three regions were tested for SPL12 occupancy using ChIP-qPCR analysis of 35S::SPL12m-GFP plants. Compared to WT, 35S::SPL12m-GFP plants showed significantly higher SPL12 binding to the AGL21 promoter region ( Figure 4C), and occupancy in the three regions was substantially higher than that at the negative control LOB1 ( Figure 4C), indicating that the SPL12 protein can bind to multiple regions in the AGL21 promoter to regulate its expression.  (Figure 4c), indicating that the SPL12 protein can bind to multiple regions in the AGL21 promoter to regulate its expression.  , is used as a negative control. * and ** indicate significant differences relative to WT using Student's t-test (n = 3) p < 0.05, p < 0.01, respectively. Error bar indicates standard deviation.

Effect of SPL12 Silencing in Response to Osmotic Stress
The involvement of miR156 in regulating drought and salinity responses was previously demonstrated in alfalfa [20,21]. To determine whether SPL12 is regulated in response to osmotic stress, the SPL12 transcript levels were assessed in five-week-old WT alfalfa plants treated with mannitol (400 mM) for two weeks. The transcript abundance of SPL12 was significantly increased (1.4 fold) under osmotic stress compared to the well-watered control treatment ( Figure 5A). To understand the role of SPL12 in osmotic tolerance response, additional experiments were performed on SPL12-RNAi and WT alfalfa plants, where phenotypic parameters of plants were recorded. After three weeks of osmotic stress treatment in WT and SPL12-RNAi plants, root length, lateral root number, and main root number were affected to various degrees depending on the genotype ( Figure 5C-E). SPL12-RNAi plants appeared to tolerate stress better than WT because, after three weeks of stress, viable green leaves were observed in SPL12-RNAi plants but not in WT ( Figure 5B). To investigate possible differences in their ability to grow under osmotic stress, the difference in root length before and after stress was measured. Only WT showed a decrease in root length due to osmotic stress, whereas the SPL12-RNAi plants maintained root growth ( Figure 5C). Maintenance of root growth by SPL12-RNAi also included the number of adventitious (main) roots regenerated from the stems under osmotic stress. In fact, the number of main roots in SPL12-RNAi plants did not show any significant changes in comparison between the two conditions (control and osmotic stress), while WT plants showed a reduction over the three weeks of stress compared to the well-watered WT plants ( Figure 5D). Furthermore, relative to WT, an increase in lateral root numbers was observed in one of the SPL12-RNAi genotypes (RNAi12-7) under the control condition, and in all of the SPL12-RNAi transgenic plants under stress ( Figure 5E).
35S::SPL12m-GFP and WT plants from means of n = three individual plants where LATERAL OR-GAN BOUNDARES-1, LOB1, is used as a negative control. * and ** indicate significant differences relative to WT using Student's t-test (n = 3) p < 0.05, p < 0.01, respectively. Error bar indicates standard deviation.

Effect of SPL12 Silencing in Response to Osmotic Stress
The involvement of miR156 in regulating drought and salinity responses was previously demonstrated in alfalfa [20,21]. To determine whether SPL12 is regulated in response to osmotic stress, the SPL12 transcript levels were assessed in five-week-old WT alfalfa plants treated with mannitol (400 mM) for two weeks. The transcript abundance of SPL12 was significantly increased (1.4 fold) under osmotic stress compared to the wellwatered control treatment (Figure 5a). To understand the role of SPL12 in osmotic tolerance response, additional experiments were performed on SPL12-RNAi and WT alfalfa plants, where phenotypic parameters of plants were recorded. After three weeks of osmotic stress treatment in WT and SPL12-RNAi plants, root length, lateral root number, and main root number were affected to various degrees depending on the genotype (Figure 5c-e). SPL12-RNAi plants appeared to tolerate stress better than WT because, after three weeks of stress, viable green leaves were observed in SPL12-RNAi plants but not in WT ( Figure  5b). To investigate possible differences in their ability to grow under osmotic stress, the difference in root length before and after stress was measured. Only WT showed a decrease in root length due to osmotic stress, whereas the SPL12-RNAi plants maintained root growth ( Figure 5c). Maintenance of root growth by SPL12-RNAi also included the number of adventitious (main) roots regenerated from the stems under osmotic stress. In fact, the number of main roots in SPL12-RNAi plants did not show any significant changes in comparison between the two conditions (control and osmotic stress), while WT plants showed a reduction over the three weeks of stress compared to the well-watered WT plants ( Figure 5d). Furthermore, relative to WT, an increase in lateral root numbers was observed in one of the SPL12-RNAi genotypes (RNAi12-7) under the control condition, and in all of the SPL12-RNAi transgenic plants under stress ( Figure 5e).  [11][12][13][14]. * and ** indicate significant differences relative to WT using Student's t-test p < 0.05, p < 0.01, respectively. Error bar indicates standard deviation.

SPL12 Silencing Mitigates Nodulation Inhibition under Osmotic Stress
To gain an insight into the function of SPL12 in nodulation under osmotic stress, two weeks after cutting, the rooted SPL12-RNAi transgenic plants were inoculated with S. meliloti and also treated with mannitol (400 mM) for three weeks (21 dai). When comparing the number of nodules in well-watered (treated with distilled water) and mannitoltreated plants ( Figure 6A), WT plants showed a decrease in nodule numbers while SPL12-RNAi genotypes maintained nodulation after three weeks of osmotic stress ( Figure 6B).
of WT alfalfa treated with mannitol (400 mM) n = 3; (B) representative WT and SPL12-RNAi plants that were treated with mannitol (400 mM) for three weeks; (C) root length; (D) number of main roots; and (E) number of lateral roots of WT and SPL12-RNAi alfalfa under control and osmotic stress conditions (n = [11][12][13][14]. * and ** indicate significant differences relative to WT using Student's t-test p < 0.05, p < 0.01, respectively. Error bar indicates standard deviation.

SPL12 Silencing Mitigates Nodulation Inhibition under Osmotic Stress
To gain an insight into the function of SPL12 in nodulation under osmotic stress, two weeks after cutting, the rooted SPL12-RNAi transgenic plants were inoculated with S. meliloti and also treated with mannitol (400 mM) for three weeks (21 dai). When comparing the number of nodules in well-watered (treated with distilled water) and mannitol-treated plants (Figure 6a), WT plants showed a decrease in nodule numbers while SPL12-RNAi genotypes maintained nodulation after three weeks of osmotic stress (Figure 6b). . * and ** indicate significant differences relative to WT using Student's t-test p < 0.05, p < 0.01, respectively. Error bar indicates standard deviation. . * and ** indicate significant differences relative to WT using Student's t-test p < 0.05, p < 0.01, respectively. Error bar indicates standard deviation.
Considering the improved nodule numbers in SPL12-RNAi plants at 14 dai [29], we also tested the nodulation capacity of SPL12-RNAi plants at 14 dai under osmotic stress. In line with this, under the well-watered condition, SPL12-RNAi transgenic plants produced significantly more nodules compared to WT. Following 400 mM mannitol treatment, the nodule number was reduced in WT compared to well-watered condition ( Figure 6C), but transgenic SPL12-RNAi plants maintained nodulation, in fact, nodule numbers was not noticeably affected by 400 mM mannitol treatment in these transgenic plants ( Figure 6C). To shed light on molecular events associated with SPL12 function under stress conditions, we investigated the effect of mannitol treatment on the transcript levels of AGL6, AGL21 (regulated by SPL12) and CLE13 (that inhibits nodulation [47]) in WT and SPL12-RNAi alfalfa. The results showed that there were significant differences between plants under stress and control conditions ( Figure 7). As expected, the transcript level of AGL21 was significantly higher in all of the SPL12-RNAi plants compared to WT under the control condition ( Figure 7A). AGL21 was also significantly upregulated under stress in two SPL12-RNAi genotypes compared to WT. However, AGL21 was downregulated in WT plants under stress, whereas two of the SPL12-RNAi genotypes showed no significant differences (RNAi12-24 and RNAi12-29) compared to plants grown under control conditions ( Figure 7A). For AGL6, significantly lower transcript levels were detected in WT and SPL12-RNAi transgenic plants under the stress condition compared to counterpart plants grown under the control condition ( Figure 7B). However, no significant changes in AGL6 transcript levels were detected in SPL12-RNAi genotypes (except for RNAi-24) compared to WT under either control or osmotic stress conditions ( Figure 7B). Given that, under osmotic stress, SPL12-RNAi plants at 21 dai produced more nodules compared to WT (Figure 6b), we analyzed the transcript levels of CLE13 and found a Given that, under osmotic stress, SPL12-RNAi plants at 21 dai produced more nodules compared to WT ( Figure 6B), we analyzed the transcript levels of CLE13 and found a decrease in transcript levels under osmotic stress in all genotypes ( Figure 7C). Under the control condition, CLE13 transcript levels were higher in SPL12-RNAi plants relative to WT, while under osmotic stress, there was no significant change in transcript levels of CLE13 ( Figure 7C). This result is consistent with results on nodulation in SPL12-RNAi and WT at 21 dai, where there was no significant difference between WT and SPL12-RNAi plants, CLE13 was significantly upregulated in the SPL12-RNAi plants at 21 dai [29].

Sulfate Transporters Are Enhanced in SPL12-Silenced Plants
There is a high demand for sulfur in nodulating legumes, and in fact, nitrogen fixation is more sensitive to sulfur deficiency than to nitrate uptake [12,48]. A good supply of sulfur enhances nodulation and nitrogen fixation [12,13]. RNA-seq analysis from a previous study [29] revealed that two Group3 SULTR genes, SULTR3.4 and SULTR3.5, were significantly upregulated in SPL12-RNAi plants, a finding that was validated by RT-qPCR ( Figure 8A,B).  5 in WT and SPL12-RNAi alfalfa exposed to three weeks of osmotic stress (400 mM mannitol). * and ** indicate significant differences relative to WT using Student's t-test (n = 3) p < 0.05, p < 0.01, respectively. Error bar indicates standard deviation.
Since SULTR3.4 and SULTR3.5 are members of Group3 SULTRs, which are strongly regulated by abiotic stress in plant roots [17], we aimed to investigate their transcript levels under osmotic stress. WT alfalfa plants had higher SULTR3.4 levels under osmotic stress compared to the unstressed control, and there was a change in transcript levels be-

in WT and
SPL12-RNAi alfalfa exposed to three weeks of osmotic stress (400 mM mannitol). * and ** indicate significant differences relative to WT using Student's t-test (n = 3) p < 0.05, p < 0.01, respectively. Error bar indicates standard deviation.
Since SULTR3.4 and SULTR3.5 are members of Group3 SULTRs, which are strongly regulated by abiotic stress in plant roots [17], we aimed to investigate their transcript levels under osmotic stress. WT alfalfa plants had higher SULTR3.4 levels under osmotic stress compared to the unstressed control, and there was a change in transcript levels between treatments in SPL12-RNAi plants ( Figure 8C). It was noted that SULTR3.4 expression in RNAi12-7 and RNAi12-29 was higher than in WT under control conditions. WT and RNAi12-29 plants showed a decrease in SULTR3.5 abundance in response to osmotic stress, whereas RNAi12-7 and RNAi12-24 plants were able to maintain their levels of SULTR3.5 ( Figure 8D). When considering the plants under the stress condition only, RNAi12-7 and RNAi12-24 had an enhanced SULTR3.5 transcript level compared to WT. SULTR3.5 expression in well-watered SPL12-RNAi plants was higher than in WT, and also relative to counterparts under osmotic stress ( Figure 8D).

Effect of Mannitol Treatment on Expression of Stress-Related Genes
The effect of drought on the expression of antioxidant-related glutathione synthase (GSH) [49] and the stress-responsive transcription factor WD40-1 [50] was previously reported in alfalfa. Enhanced levels of GSH and WD40-1 in miR156-OE alfalfa under drought stress in leaves and roots, respectively, were also previously reported [20,21]. In the current study, the transcript abundance of GSH and WD40-1 was examined to determine whether SPL12 serves to maintain the transcript levels of these genes in alfalfa exposed to osmotic stress. While the transcript levels of GSH were higher in well-watered RNAi12-7 and RNAi12-24 compared to WT plants ( Figure 9A), they did not show a change in response to osmotic stress in SPL12-RNAi and WT plants. In fact, GSH was at lower levels in RNAi12-7 under stress relative to control ( Figure 9A). The effect of drought on the expression of antioxidant-related glutathione synthase (GSH) [49] and the stress-responsive transcription factor WD40-1 [50] was previously reported in alfalfa. Enhanced levels of GSH and WD40-1 in miR156-OE alfalfa under drought stress in leaves and roots, respectively, were also previously reported [20,21]. In the current study, the transcript abundance of GSH and WD40-1 was examined to determine whether SPL12 serves to maintain the transcript levels of these genes in alfalfa exposed to osmotic stress. While the transcript levels of GSH were higher in well-watered RNAi12-7 and RNAi12-24 compared to WT plants (Figure 9a), they did not show a change in response to osmotic stress in SPL12-RNAi and WT plants. In fact, GSH was at lower levels in RNAi12-7 under stress relative to control (Figure 9a).  Similarly, for WD40-1 transcript levels, SPL12i-7 and SPL12i-29 showed higher levels under control treatment compared to WT plants ( Figure 9B), but there was no change between SPL12-RNAi and WT plants under osmotic stress, with WD40-1 even showing a decrease in RNAi12-7 and SPL12i-29 under stress relative to the control ( Figure 9B).

AGL6 Silencing Maintains Nodulation under Osmotic Stress
Given that AGL6 is a direct target of SPL12, and with the observed reduction in transcript levels of AGL6 in SPL12-RNAi genotypes during osmotic stress, we set out to investigate the potential role of AGL6 in nodulation under this stress. Two-week-old rooted WT and AGL6-RNAi transgenic plants (L9, L13A, and L13B) were inoculated with S. meliloti and treated with mannitol (400 mM) for two weeks (14 dai) or three weeks (21 dai). The number of nodules was compared in both the control and mannitol-treated plants ( Figure 10A). At 14 dai, AGL6-RNAi transgenic plants produced significantly more nodules compared to WT under the well-watered condition (Figure 10b). Upon treatment with 400 mM mannitol, the nodule number was reduced in WT, but AGL6-RNAi plants maintained nodulation after two weeks of osmotic stress (Figure 10b). At 21 dai, stressed WT plants had a reduced nodule number when compared to the well-watered WT and the stressed . * and ** indicate significant differences relative to WT using Student's t-test p < 0.05, p < 0.01, respectively. Error bar indicates standard deviation. At 14 dai, AGL6-RNAi transgenic plants produced significantly more nodules compared to WT under the well-watered condition ( Figure 10B). Upon treatment with 400 mM mannitol, the nodule number was reduced in WT, but AGL6-RNAi plants maintained nodulation after two weeks of osmotic stress ( Figure 10B). At 21 dai, stressed WT plants had a reduced nodule number when compared to the well-watered WT and the stressed AGL6-RNAi plants, while AGL6-RNAi genotypes maintained nodulation after three weeks of stress ( Figure 10C), thus confirming the likely involvement of AGL6 in regulating nodulation under osmotic stress in alfalfa.

How Nitrate Availability Affects Nodulation through the SPL12-AGL21-Regulatory Pathway
To conserve energy, plants inhibit nodulation under conditions of nitrate abundance in the rhizosphere, resulting in a decrease in nodule numbers, nodule mass, and nitrogen fixation, as well as an acceleration of nodule senescence [44]. This regulation of nodulation by nitrate is a part of the AON signaling pathway [43,51]. As the SPL12-RNAi and AGL6-RNAi plants showed an increase in nodulation, we tested the relationship between nitrate and the miR156/SPL12 regulatory system. Under nitrate-sufficient conditions, the rhizobiainoculated roots of the SPL12-RNAi plants developed more active nodules relative to WT, demonstrating the role of the miR156/SPL12-mediated system in controlling the rhizobiaalfalfa symbiosis. In Phaseolus vulgaris (the common bean), miR172c acts as a signaling component of the nitrate-dependent AON, and decreases the sensitivity of nodulation to inhibition by nitrate. Common bean plants overexpressing miR172 showed an increase in active nodules in the presence of nitrate [52]. AtSPL9 was shown to be a potential nitrate-regulatory hub in Arabidopsis, where it may target primary nitrate-responsive genes [53]. AtSPL9 expression is affected by nitrate, and the transcript levels of AtNRT1.1, AtNIA2, and AtNiR significantly increased in response to nitrate in AtSPL9 overexpression Arabidopsis plants [53]. In tomato plants (Solanum lycopersicum), it was reported that an SPL transcription factor, LeSPL-CNR, directly binds to the promoter of SlNIA, resulting in its repressing its expression and activity [54]. LeSPL-CNR was further shown to negatively regulate SlNIA transcription levels in response to cadmium (cd) stress in tomato plants [54].
Based on our findings, we propose that SPL12 regulates nodulation under a high concentration of nitrate in alfalfa by downregulating AGL21. Here, we showed that AGL21 is upregulated in SPL12-RNAi alfalfa plants. AGL21 is an ANR1 MADS box protein-coding gene, and AtANR1 MADS box proteins were previously shown to mediate the effect of externally applied nitrate on lateral root development in Arabidopsis [39,40]. In rice, two MADS box genes, OsMADS25 and OsMADS27, are involved in the regulation of root development in response to nitrate [55]. In Arabidopsis, AtAGL21 is expressed in different tissues, but most strongly in roots, where AtAGL21 plays an important role in lateral root development under nitrogen deficiency [41]. In the common bean, PvAGL21 is expressed in nodules, and its expression is higher in roots compared to pods, seeds, and stems [56]. These results are consistent with our finding that alfalfa AGL21 is highly expressed in the roots and that its expression is induced by nitrate. Future research should focus on generating and analyzing AGL21-silencing and overexpressing alfalfa plants to determine the effect on root architecture, nodulation and nitrogen fixation.

Role of SPL12 and AGL6 in Regulating Nodulation under Osmotic Stress in Alfalfa
Legume crops can adjust their root architecture in response to environmental conditions, not only by branching out, but also by forming a symbiosis with rhizobial bacteria to form nitrogen-fixing nodules [57]. Soil salinity is a major abiotic stress that causes nutrients to become unavailable to plants, and it leads to a nutrient-deprived situation, or nutrient stress, affecting plant yield and root growth (reviewed by Sindhu et al. [58]). Not only does miR156 regulate nodulation in alfalfa, it is also involved in the response to abiotic stress. The miR156-mediated regulation of response to drought, heat, and salinity was previously demonstrated in alfalfa [18][19][20][21], where miR156 targets a number of SPL genes for silencing by transcript cleavage in [22,23]. Specifically, SPL13, SPL9, and SPL8 have been investigated for their role in drought tolerance in this plant [20,21,25,26]. Downregulating SPL13, SPL9 and SPL8 in transgenic plants resulted in alfalfa plants that were less susceptible to drought [20,21,25,26]. SPL12 was shown to be upregulated in response to mild and severe salinity stress in alfalfa, but was suppressed in all miR156-OE genotypes [19]. In this study, we observed a significant increase in the transcript levels of SPL12 in WT under osmotic stress as opposed to control conditions. The upregulation of SPL12 under osmotic stress is consistent with a previous report that showed an increase in SPL13 transcript levels in WT alfalfa plants under drought [20].
The roots are the first plant organ to encounter changes in response to a soil water deficit. Studies in Arabidopsis showed initiation and elongation of lateral roots in droughttolerant genotypes that led to improved water uptake and drought adaptation [59,60]. In this study, a significant increase in root length accompanied by higher lateral root numbers was observed in alfalfa SPL12-RNAi plants under osmotic stress ( Figure 5A,E). A previous study by Arshad et al. [20] showed increased root length in miR156-OE and SPL13-RNAi alfalfa genotypes under drought stress. Moreover, the miR156-SPL10 module was reported to be involved in root development by silencing AtAGL79 to control root length and lateral root numbers in Arabidopsis [46]. Therefore, it appears that improved root architecture may help SPL12-RNAi alfalfa plants to better access water from deeper in the soil under water scarcity conditions. The symbiotic interaction between legume plants and rhizobacteria can be negatively impacted by drought, resulting in reduced nodule numbers and diminished nitrogenase activity [61][62][63]. Nitrogenase activity in root nodules of M. truncatula was decreased by 18% and 66% after two and four days of water withdrawal, respectively [64]. It has been shown that in M. truncatula, both symbiotic plant components and S. meliloti bacteria residing in the root nodules adjust their gene expression profiles in response to drought stress [64]. Our results showed a decrease in the nodule numbers in WT plants under osmotic stress conditions, while SPL12-RNAi genotypes maintained nodulation under this stress. The transcript levels of CLE13 decreased under osmotic stress in all genotypes, while they increased in SPL12-RNAi plants under control conditions. This is consistent with increasing nodulation under osmotic stress in SPL12-RNAi genotypes. The AGL6 transcript level did not show any significant changes under the control condition, while its expression was increased only in RNAi12-24 under osmotic stress compared with WT. A previous study showed that SPL12 positively regulates AGL6. In fact, the transcript level of AGL6 was increased in 35S::SPL12 genotypes, but no change was detected in SPL12-RNAi plants [29]. Given the functional redundancy of some SPLs, silencing only one SPL gene may not be sufficient to affect the expression of AGL6 in SPL12-RNAi genotypes. Here, it was shown that AGL6 transcript levels were also lower under osmotic stress compared to no treatment, resulting in AGL6-RNAi genotypes maintaining their nodulation activity. These observations that SPL12-RNAi and AGL6-RNAi plants maintained nodulation under osmotic stress suggest a role for SPL12/AGL6 in regulating nodulation in alfalfa under osmotic stress.
In nodulating legumes, sulfur supply plays an important role in symbiotic nitrogen fixation, as sulfur deficiency causes a decrease in nodulation, inhibition of nitrogen fixation, and a slowing down of nodule metabolism [14]. Accordingly, sulfate transport and metabolism also positively affect nitrogen fixation and nodulation [14]. A sulfate transporter in the symbiosomal membrane of L. japonicus, LjSST1, was the first indication of sulfate exchange between the two symbiotic partners [15]. LjSST1 is specifically and highly expressed in the nodules, suggesting a crucial role for this protein in the transport of sulfate from the plant to the bacteroids [15]. The sst1 mutants developed smaller nodules and displayed symptoms of nitrogen deficiency only under symbiotic conditions. The nodules of the sst1 mutant plants showed a reduction of approximately 90% in the rate of nitrogen fixation [15]. In the current study, two of the Group3 SULTR genes, SULTR3.4 and SULTR3.5, were significantly upregulated in SPL12-RNAi plant roots compared to WT. MtSULTR3.5 in M. truncatula, a homolog of LjSST1, is strongly expressed in nodules [16]. Another study showed that MtSULTR3.5 expression is strongly upregulated in M. truncatula roots subjected to salt stress [17,65]. Of the sulfate transporters, Group3 SULTRs specifically operate under abiotic stress conditions, and they are among salt-and drought-responsive genes in both Arabidopsis and M. truncatula [17,[65][66][67]. Interestingly, SULTR3.1 and SULTR3.4 genes are upregulated in the roots of both Arabidopsis and M. truncatula plants subjected to drought stress [17]. Given the above findings, we measured the transcript levels of SULTR3.4 and SULTR3.5 in alfalfa root tissues under osmotic stress. The maintenance of SULTR3.4 and SULTR3.5 transcript levels under osmotic and control conditions in SPL12-RNAi roots indicates that SPL12 must be involved in SULTR3.4 and SULTR3.5 regulation. Although the five AtSULTR3 transporters have been functionally characterized in Arabidopsis [68], further studies are still needed to understand the contribution of nodule sulfate transporters to salt stress response in legumes.

Plant Material and Growth Conditions
Medicago sativa L. (alfalfa) clone N4.4.2 [69] was obtained from Daniel Brown (Agriculture and Agri-Food Canada, London, ON, Canada) and was used as the wild-type (WT) genotype. Alfalfa genotypes with reduced expression levels of SPL12 and AGL6, SPL12-RNAi (RNAi12-7, RNAi12-24, and RNAi12-29) and AGL6-RNAi (L9, L13A, and L13B), respectively, and 35S::SPL12-GFP [29] were used in this study. The transgenic alfalfa plants were generated previously in Dr. Hannoufa's laboratory using the WT clone N4.4.2 [29]. WT and transgenic alfalfa plants were grown under greenhouse conditions at 21-23 • C, 16 h light/8 h dark per day, light intensity of 380-450 W/m 2 (approximately 500 W/m 2 at high noon time), and a relative humidity of 56% for the duration of all experiments. Because of the obligate outcrossing nature of alfalfa, WT and transgenic alfalfa were propagated by rooted stem cuttings to maintain the genotype throughout the study. Stem-cutting propagation and morphological characterization of alfalfa plants were carried out as described previously [28].

Phenotypic Analysis of Nodule Development
To determine the number of nodules, plants were examined at 14 and 21 days after inoculation (dai) with Sinorhizobium meliloti Sm1021. To eliminate potential microbial contamination, equipment was surface-sterilized using 1% sodium hypochlorite, while vermiculite and water were sterilized by autoclaving for 1 h. S. meliloti Sm1021 strain was cultured on a yeast-extract broth agar [70] for two days at 28 • C. A single colony was then inoculated in liquid TY medium and incubated at 28 • C to an optical density OD 600 nm of 1.5. The alfalfa rooted stems were inoculated by applying 5 mL of bacterial suspension or sterilized water (non-inoculated control) as described previously [28] and allowed to grow for two or three additional weeks.

Nitrate Treatment
To explore if SPL12-related regulation of nodulation is affected by nitrate, the nodulation test was performed upon treatment with this nutrient. WT and SPL12-RNAi alfalfa stem cuttings were grown on vermiculite for 14 days and were then inoculated with S. meliloti Sm1021 and treated with KCl or KNO 3 . For this, the 14-day-old inoculated transgenic and WT plants were watered with 3, 8, or 20 mM KNO 3 or KCl twice a week for two or three weeks. The entire experiment was repeated twice under the same growth and nitrate treatment conditions to test the reproducibility of the results. Effects on nodulation were studied by counting the number of the active (pink) nodules.
To investigate whether treatment with KNO 3 affects expression of SPL12 and AGL21 genes, WT and SPL12-RNAi alfalfa plants were grown on vermiculite for 14 days, then the plants were transferred to Murashige & Skoog Modified Basal Salt Mixture without nitrogen (M531, PhytoTechnology Laboratories®, KS, USA) liquid media and left overnight at room temperature. For the nitrate signaling test, the samples were treated with 20 mM KNO 3 for 0, 5, and 24 h, then roots were collected and flash frozen in liquid nitrogen and stored at −80 • C for later transcript analysis of SPL12 and AGL21.

Mannitol Treatment
To investigate whether SPL12 affects nodulation when plants are grown under osmotic stress, WT, SPL12-RNAi, and AGL6-RNAi alfalfa plants were grown on vermiculite for 14 days and were then inoculated with S. meliloti Sm1021 for two days, followed by treatment with mannitol (to mimic osmotic stress). For mannitol treatment, 16-day-old inoculated WT and transgenic plants were watered with 400 mM mannitol or distilled water once a week for two or three weeks. The below-ground phenotypic parameters were measured according to Aung et al. [27]. The phenotypes included in the characterization were number of main roots, lateral roots, and root length. The roots directly emerging from the stem were considered as main roots, while those that emerged from the main roots were counted as lateral roots. Root length was considered as the length of the longest root. The entire experiment was repeated twice under the same growth and osmotic stress conditions to test the reproducibility of the results. Root samples were harvested from SPL12-RNAi and WT plants under osmotic and control conditions and were flash frozen in liquid nitrogen and kept at −80 • C for later transcript analysis of SPL12, AGL21, AGL6, CLE13, SULTR3.4, SULTR3.5, GSH, and WD40-1.

RNA Extraction, Reverse Transcription and RT-qPCR
Different alfalfa tissues (stems, leaves and roots) were collected and flash frozen in liquid nitrogen and stored at −80 • C until further use. Approximately 100 mg fresh weight was used for total RNA extraction using RNeasy Plant Mini-prep Kit (Qiagen, Hilden, Germany, Cat # 1708891) for leaf and stem samples, and Total RNA Purification Kit (Norgen Biotek Corp., Thorold, ON, Canada, Cat # 25800) for roots. Tissue was homogenized using a PowerLyzer ® 24-bench top bead-based homogenizer (Cat # 13155) according to the manufacturer's manual. Approximately 500 ng of Turbo DNase (Invitrogen, CA, US, Cat # AM1907)-treated RNA was used to generate cDNA using the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA, USA, Cat # 1708891). Transcript levels were analyzed by RT-qPCR using a CFX96 TouchTM Real-Time PCR Detection System (Bio-Rad) and SsoFast™ EvaGreen ® Supermixes (Bio-Rad Cat # 1725204) using gene specific primers. Each reaction consisted of 2 µL of cDNA template, 0.5 µL forward and reverse gene-specific primers (10 µM each) (Table S1), 5 µL SsoFast Eva green Supermix and topped up to 10 µL with ddH 2 O. For each sample three or four biological replicates were analyzed, and each biological replicate was tested using three technical replicates. Transcript levels were analyzed relative to three reference genes: CYCLOPHILIN (Cyclo) [71], β-actin (ACTB) [72], and ACTIN DEPOLYMERIZING FACTOR (ADF) [71,72] (primers are listed in Table S1).

ChIP-qPCR Analysis
Shoot tips of alfalfa plants overexpressing SPL12 tagged with GFP driven by the 35S promoter (p35S:SPL12m-GFP) were used as materials for ChIP-qPCR analyses, which were performed based on a previously described protocol [29] with the Chromatin Immunoprecipitation Assay kit (Lot:2382621, Millipore, Billerica, MS, USA). Briefly, nuclei were purified from shoot tips that proteins bound to DNA were cross-linked using 1% formaldehyde under a vacuum for 20 min and the mixtures were ground in liquid nitrogen. The chromatin solution was then sonicated twice at power 3 for 15 s on ice into 500-1000 bp fragments using a Sonic Dismembrator (Fisher Scientific, PA, USA). Ab290 GFP antibody was added to the chromatin solution and protein A-agarose beads were added to recover immune complexes. The precipitated DNA was extracted using phenol: chloroform (1:1, v:v) and resuspended in distilled water to be used for ChIP-qPCR analysis using qnMsAGL21 as listed in Table S1. SPL12 occupancy on AGL21 was estimated by comparing the fold enrichment in p35S:SPL12m-GFP and WT plants. A DNA fragment containing a SBP binding consensus was amplified from a LATERAL ORGAN BOUNDARES-1, LOB1, gene [73] for use as a negative control.

Phylogenetic Tree Construction
The phylogenetic tree was constructed based on an alignment of the MADS-box domain and using publicly available sequences of M. sativa, M. truncatula, and Arabidopsis. Amino acids were aligned by visualization and nucleotides were subjected to ClustalW alignment analysis. The phylogenetic tree was constructed using the neighbor-joining method of phylogenetic tree construction using MEGA7 [74].

Statistical Analysis
Statistical analyses were performed using Microsoft Excel spreadsheet software. Paiwise comparisons were made using a Student's t-test with either equal or unequal variance. The significant differences between sample means for three or more data sets were calculated using the one-way analysis of variance (ANOVA) where appropriate.

Conclusions
Our investigation of the SPL12 function revealed that SPL12 and its direct target, AGL6, regulate nodulation under osmotic stress, as plants with reduced SPL12 and AGL6 showed an enhanced number of nodules under this stress, resulting in the maintenance of nodulation in SPL12-RNAi and AGL6-RNAi plants despite the adverse stress conditions. This study, together with the previous observations that miR156-OE plants had increased tolerance to drought [20,21], and improved nodulation and nitrogen fixation [27], provided evidence that miR156-targeted SPL12 is a regulator of nodulation under osmotic stress in alfalfa. Moreover, maintenance of nodulation by AGL6-RNAi suggests a role for AGL6 in the control of nodulation in alfalfa under osmotic stress ( Figure 11). Our results showed that the role of SPL12 in alfalfa is not only restricted to regulating nodulation under normal conditions, but also controls this process under nitrate-sufficient conditions. Rhizobia-inoculated alfalfa roots with reduced levels of SPL12 were found to develop more active nodules, relative to WT under nitrate-sufficient conditions, demonstrating the role of the miR156/SPL12-mediated system in controlling rhizobia-alfalfa symbiosis. SPL12 regulates nodulation under nitrate treatment in alfalfa by targeting AGL21. AGL21 is an ANR1 MADS box protein-coding gene. AtANR1 MADS box proteins were previously shown to mediate the effect of externally applied nitrate on lateral root Our results showed that the role of SPL12 in alfalfa is not only restricted to regulating nodulation under normal conditions, but also controls this process under nitrate-sufficient conditions. Rhizobia-inoculated alfalfa roots with reduced levels of SPL12 were found to develop more active nodules, relative to WT under nitrate-sufficient conditions, demonstrating the role of the miR156/SPL12-mediated system in controlling rhizobia-alfalfa symbiosis. SPL12 regulates nodulation under nitrate treatment in alfalfa by targeting AGL21. AGL21 is an ANR1 MADS box protein-coding gene. AtANR1 MADS box proteins were previously shown to mediate the effect of externally applied nitrate on lateral root development in Arabidopsis [39,40]. Previously, RNAseq and gene ontology analysis showed AGL21 to be upregulated in SPL12-RNAi alfalfa roots [29], where its transcript levels were induced by nitrate. As a negative regulator of AGL21, SPL12 silencing upregulates AGL21 and enhances the production of active nodules under high nitrate condition ( Figure 11). We also determined the direct binding of SPL12 to the AGL21 promoter. Taken together, our results suggest that SPL12 along with AGL6 and AGL21 modulate alfalfa nodulation under osmotic stress and sufficient nitrate conditions. Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/plants11223071/s1, Figure S1: Effect of nitrate on nodulation in SPL12-RNAi plants; Figure S2: Effect of nitrate on nodule phenotype in SPL12-RNAi plants; Figure S3: Promoter sequence of the alfalfa AGL21 gene with putative SBD binding elements; Table S1: List of primers used in this study.