Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2021 Dec 27;233(4):1881–1899. doi: 10.1111/nph.17899

The miR156b‐GmSPL9d module modulates nodulation by targeting multiple core nodulation genes in soybean

Jinxia Yun 1,*, Zhengxi Sun 1,2,3,*, Qiong Jiang 3, Youning Wang 1, Can Wang 1, Yuanqing Luo 1, Fengrong Zhang 1, Xia Li 1,✉
PMCID: PMC9303946  PMID: 34862970

Summary

  • Symbiotic nodulation is initiated in the roots of legumes in response to low nitrogen and rhizobial signal molecules and is dynamically regulated by a complex regulatory network that coordinates rhizobial infection and nodule organogenesis.

  • It has been shown that the miR156‐SPL module mediates nodulation in legumes; however, conclusive evidence of how this module exerts its function during nodulation remains elusive.

  • Here, we report that the miR156b‐GmSPL9d module regulates symbiotic nodulation by targeting multiple key regulatory genes in the nodulation signalling pathway of soybean. miR156 family members are differentially expressed during nodulation, and miR156b negatively regulates nodulation by mainly targeting soybean SQUAMOSA promoter‐binding protein‐like 9d (GmSPL9d), a positive regulator of soybean nodulation. GmSPL9d directly binds to the miR172c promoter and activates its expression, suggesting a conserved role of GmSPL9d. Furthermore, GmSPL9d was coexpressed with the soybean nodulation marker genes nodule inception a (GmNINa) and GmENOD40‐1 during nodule formation and development. Intriguingly, GmSPL9d can bind to the promoters of GmNINa and GmENOD40‐1 and regulate their expression.

  • Our data demonstrate that the miR156b‐GmSPL9d module acts as an upstream master regulator of soybean nodulation, which coordinates multiple marker genes involved in soybean nodulation.

Keywords: GmENOD40‐1, GmNINa, GmSPL9d, miR156b, miR172c, nodulation, Soybean (Glycine max)

Introduction

Soybean is one of the best sources of protein and is a staple in both human and animal diets. Because soybean is high in protein, it has a much higher need for nitrogen (N) than other crops during development and reproduction (Nicolas et al., 2017). Soybean has established a symbiotic relationship with nitrogen‐fixing bacteria in soil that are able to convert atmospheric dinitrogen (N2) into ammonia, which can contribute to satisfying its high N demand. In exchange, soybean plants provide these bacteria with carbohydrates and other nutrients for rhizobial growth and proliferation (Desbrosses & Stougaard, 2011; Downie, 2014). Symbiotic N fixation is not only beneficial for increasing soybean productivity but also a powerful means through which to increase the N supply in arable land to improve the yield of other cereal crops (Fan et al., 2006; Li et al., 2009). Thus, symbiotic N fixation is the main factor associated with the yield and quality of soybean crops in particular and sustainable agriculture in general.

Symbiotic N fixation occurs in a specified root organ (e.g. root nodule) of soybean that hosts rhizobia. When N in soil is low, soybean releases isoflavones (e.g. genistein) to induce the production of nod factors (NFs) by rhizobia (Pueppke et al., 1998). NFs are perceived by soybean NF receptors (NFR1 and NFR5) and trigger the signalling cascade that regulates rhizobial infection and nodule organogenesis, which consist of a series of developmental processes, including dedifferentiation of root outer cortical cells, nodule primordia formation, nodule emergence and nodule maturation (Ferguson et al., 2010; Broghammer et al., 2012; Moling et al., 2014). It has been shown that the onset of rhizobial infection and nodule organogenesis is mediated by the nodulation signalling pathway, in which many key components have been identified in the past three decades. Among them, nodule inception (NIN), an RWP‐RK transcription factor and early nodulin gene (e.g. ENOD40), is the earliest identified regulator essential for nodulation in legumes, including soybean (Kouchi & Hata, 1993; Yang et al., 1993; Klein et al., 1996; Schauser et al., 1999; Vernie et al., 2015).

When soybean roots are inoculated with rhizobia or treated with NFs, the expression of ENOD40 is rapidly upregulated (Kouchi & Hata, 1993; Yang et al., 1993; Minami et al., 1996), and alteration of ENOD40 expression significantly influences nodulation (Charon et al., 1999; Kumagai et al., 2006; Wan et al., 2007). ENOD40 can act as a regulatory long noncoding RNA (Campalans et al., 2004) that also encodes two small peptides of 12 or 13 amino acids (Sousa et al., 2001), and these ENOD40 peptides can bind to and enhance the stability of sucrose synthase (Hardin et al., 2003). These results suggest that ENOD40s may modulate nodulation by either functioning as cell–cell signalling molecules or being directly involved in the regulation of carbon sinks during nodulation. Because of the pivotal role of ENOD40s in nodulation, they have been widely used as marker genes for nodule formation. Previously, we showed that GmENOD40 expression is repressed by Nodule Number Control 1 (NNC1), an AP2 transcription factor that is targeted by the noncoding micro‐RNA (miRNA) miRNA172c (miR172c) (Wang et al., 2014). However, it remains unknown how the transcription of ENOD40s is activated spatially and temporally. In comparison, NIN has attracted extensive attention because of its specific and indispensable role in legume nodulation (Schauser et al., 1999; Vernie et al., 2015). Orthologues of NIN in different legumes were specifically induced by rhizobia and are essential for legume–rhizobia symbiosis (Griesmann et al., 2018; van Velzen et al., 2018). The orthologous NIN proteins in Medicago trunctula and Lotus japonicus exert their function by directly targeting many genes, such as Nodulation Pectate Lyase (NPL) (Xie et al., 2012), Nuclear Factor‐Y (NF‐Y) NF‐YA1 and NF‐YA2 (Soyano et al., 2013), and Early Nodulin 11 (ENOD11) (Svistoonoff et al., 2010; Vernie et al., 2015; Kawaharada et al., 2017), which are involved in various processes of nodulation, including cell modification, biosynthesis of hormones (e.g. gibberellin), nutrient (e.g. N, P and S) uptake and assimilation (C. W. Liu et al., 2019). In soybean, the NIN orthologue GmNINa can directly bind to the miR172c promoter and activate miR172c transcription (Wang et al., 2014, 2019). In addition, NIN orthologues are also key components of autoregulation of nodulation (AON). NIN transcription factors can activate CLE genes in legumes (e.g. CLAVATA3/ESR (CLE)‐RELATED‐ROOT SIGNAL1 (CLE‐RS1), CLE‐RS2 in L. japonicus, GmRIC1/2 in soybean) that encode long‐distance signal peptides from roots to shoots to turn on the AON. As feedback regulators, the expression levels of NIN orthologue genes were inhibited by AON (Soyano et al., 2014; Wang et al., 2019). NIN proteins can interact with other transcription factors (e.g. NNC1 in soybean and NIN‐like protein 1 (NLP1) in M. truncatula) to modulate optimal nodule numbers under low and high levels of N (Soyano et al., 2014; Wang et al., 2019).

Because of the pivotal roles of NIN and ENOD40 in nodulation, the expression levels of NIN and ENOD40 are temporally and spatially controlled. In L. japonicas, NIN is activated by both CYCLOPS and ERF REQUIRED FOR NODULATION1 (ERN1) during rhizobial infection in infection thread formation (M. Liu et al., 2019), while in M. truncatula, NIN activation is regulated by the complex of CYCLOPS and two GRAS family transcription factors, NODULATION SIGNALING PATHWAY1 (NSP1) and NSP2 (Hirsch et al., 2009). A recent study has shown that cellular‐specific expression of Medicago NIN in epidermis, cortex and pericycle cells is finely controlled by a distant regulatory mechanism (J. Liu et al., 2019). These findings highlight the complexity of transcriptional regulation of NIN orthologue genes, which is also the case for ENOD40 in legumes. There are many questions related to the spatial and temporal regulation of NIN and ENOD expression that need to be answered.

miRNAs are master regulators of plant development and dynamically regulate the developmental transition of plants (Llave et al., 2002; Ha & Kim, 2014; Yu et al., 2017; Treiber et al., 2019). Accumulating evidence has shed light on the key regulatory roles of miRNAs in the nodulation of various leguminous plants (Subramanian et al., 2008; Y. Wang et al., 2009; Dong et al., 2013; Hofferek et al., 2014; Íñiguez et al., 2015; Aung et al., 2017). Strikingly, two well‐known pairs of miRNAs, miR156 and miR172, that sequentially regulate the transition from vegetative to reproductive growth in plants (J. W. Wang et al., 2009; Wu et al., 2009) are also involved in symbiotic nodulation in legumes (Yan et al., 2013; Wang et al., 2014, 2015; Aung et al., 2015, 2017; Íñiguez et al., 2015). Recent studies have proven the positive role of miR172s in the nodulation of various legumes (Reynoso et al., 2013; Yan et al., 2013; Hofferek et al., 2014; Wang et al., 2014, 2019; Holt et al., 2015; Nova‐Franco et al., 2015). In comparison, miR156s appear to play opposite roles in different leguminous plants. In alfafa, miR156 promotes nodulation (Aung et al., 2015, 2017), while in L. japonicus and soybean, miR156 represses nodulation (Yan et al., 2013; Wang et al., 2015). It has been shown that overexpression of miR156 affects the expression of miR172 during soybean nodulation (Yan et al., 2013), and LjmiR156a overexpression downregulates several nodulation genes in early nodulation of L. japonicus, such as the NIN and ENOD40 genes (Wang et al., 2015). These results implicate miR156 as one of the upstream regulators of early nodulation in legumes; however, it remains unknown how miR156s regulate these nodulation genes and which SQUAMOSA PROMOTER BINDING PROTEINLIKE (SPL) genes function downstream of miR156 to modulate nodulation in legumes.

In this study, we performed a systematic functional characterization of soybean miR156 and SPL families. We found that these miR156(s) and GmSPL(s) genes were differentially expressed during nodulation. Strikingly, our results identified the miR156b‐GmSPL9d module as a key upstream modulator that controls early nodulation by directly regulating several key regulatory genes in the nodulation signalling pathway, including GmNINa, miR172c and GmENOD40‐1 in soybean. These results reveal a novel regulatory network mediated by the miR156‐GmSPL9d module in coordinated regulation of nodule formation and nodule development.

Materials and Methods

Plant and rhizobia growth conditions

Soybean (Glycine max (L.) Merrill cv Williams 82 (W82)) and miR156bOE transgenic lines were used in our experiments. The plant growth conditions and procedures for rhizobial inoculation using Bradyrhizobium diazoefficiens strain USDA110 were minor modifications from our previous study (Y. Wang et al., 2009). Briefly, we suspended the rhizobia with deionized water and inoculated 30 ml of diluted rhizobia for each seedling.

Soybean hairy root transformation and B. diazoefficiens inoculation

Soybean hairy root transformation to generate composite plants was performed using Agrobacterium rhizogenes K599 according to methods described previously (Kereszt et al., 2007; Jian et al., 2009). For rhizobial inoculation, transgenic composite plants were transplanted to pots (10 × 10 cm) containing vermiculite, which was irrigated with the N‐deficient solution as described by Y. Wang et al. (2009). The plants were grown for a 1‐wk recovery period (16 h of light, 25°C and 50% relative humidity). The plants were then inoculated with a freshly prepared suspension of B. diazoefficiens strain USDA110 (OD600 = 0.08). At 28 d after inoculation (DAI), nodule numbers per hairy root or per root system were scored, and the roots or nodules were rinsed briefly in PBS (pH 7.5), immediately frozen in liquid nitrogen and stored at −80°C for gene expression analyses. For the phenotypic analysis of miR156b/f‐CRISPR/Cas9 composite plants, we first counted the number of nodules per hairy root of the composite plant, and then extracted DNA to detect the selectable marker bar gene. The putative transgenic roots were validated by Sanger sequencing. For the phenotypic analysis of GmSPL9d‐CRISPR/Cas9 roots, we identified the transgenic hairy roots using a portable fluorescence lamp (Luyor 3415; Luyor, Shanghai, China). The number of nodules per transgenic root was examined and the mutations of the transgenic roots were confirmed by Sanger sequencing.

Root hair deformation and infection assays

To detect infection events, the entire root of mir156b/f or EV‐2 hairy roots and 4 cm root segments of miR156b overexpression first‐order lateral roots (miR156bOE) or wild‐type (WT) below the root–hypocotyl junction were cut and harvested at 6 DAI and then briefly washed with sterile PBS to remove vermiculite. The roots were then stained with 0.01% methylene blue for 5 min and washed three times with deionized water. The stained transgenic roots were observed with an Olympus CX31(CX31; Olympus, Tokyo, Japan) biological microscope for curled root hairs.

RNA extraction and quantitative PCR analysis

Total RNA and small RNAs were extracted from the samples of plant roots using TRIzol reagent (Aidlab Biotechnologies Co. Ltd, Beijing, China). Total RNA was treated with gDNA Wiper Mix (Yeasen Biotech, Shanghai, China) to remove genomic DNA. cDNA strands were synthesized from the RNAs using Hifair® II 1st Strand cDNA Synthesis SuperMix for qPCR kit (Yeasen Biotech). Quantitative PCR was performed using a Hieff® qPCR SYBR® Green Master Mix kit (Yeasen Biotech) with gene‐specific primers (Supporting Information Tables S1, S2). GmELF1B was used as an internal control.

Stem‐loop‐specific reverse transcription for miR156 family members (e.g. miR172c and miR1520d) was performed as described previously (Chen et al., 2005). Quantitative reverse transcriptase PCR (qRT‐PCR) was conducted using the Hieff® qPCR SYBR® Green Master Mix kit (Yeasen Biotech) with the gene‐specific primers listed in Table S1.

To determine the specificity of the miR156 family of reverse primers and quantitative primers, we first purified the quantitative products shown in Fig. 1(c–l). These PCR fragments were ligated into the pEASY‐Blunt 3 vector (Transgene, Paris, France; CB301), and then two positive monoclonals were selected for sequencing. The sequencing results were compared with corresponding miR156 mature sequences to determine the specificity of the reverse primers and the quantitative primers (Fig. S1).

Fig. 1.

Fig. 1

Open in a new tab

Sequence alignments and expression analysis of soybean miR156 family members during nodulation. (a) All 25 sequences of the soybean miR156 family members were aligned using Mega5 software. Asterisks represent conserved nucleotides in all mature miRNAs. (b) Phylogenetic analysis of 53 precursor sequences of miR156 family members from four plant species. The pre‐miRNA sequences of nine miR156 family members in Arabidopsis thaliana (purple triangle), eight in Lotus japonicas (blue square), 11 in Medicago truncatula (green square) and 25 in soybean (red round) were used for the alignment, and the phylogenetic neighbour‐joining tree was reconstructed using Mega7 phylogenetic analysis software. Bootstrap values (percentages of 1000 replicates) are indicated on the nodes. (c–l) Expression of miR156 family members in uninoculated (−R) and inoculated (+R) soybean cv Williams 82 (W82) roots at the indicated time points after treatments. (c) miR156a/h/u/v/w/x/y, (d) miR156b, (e) miR156c/d/e/i/j/l/m, (f) miR156e, (g) miR156f, (h) miR156 g, (i) miR156k/n/o, (j) miR156p/t, (k) miR156q/s and (l) miR156r. Values are the average ± SD from three independent experiments.

Plasmid construction

For the miR156b overexpression construct, the sequence containing the pre‐miRNA fragment of miR156b (181 bp) was amplified and inserted into the plant expression vector pTF101.1 (EV‐1) under control of the CaMV35S promoter using the restriction enzymes XbaI and SacI. To reduce the activity of miR156b, the STTM156b‐88 construct was made using a method modified from a previous protocol reported by Yan et al. (2012). For expression of the GmSPL9d‐GFP fusion protein, the GmSPL9d coding sequence was amplified and inserted into the plant expression vector pEZRK (EV‐3) using the restriction enzymes EcoRI and BamHI under control of the CaMV35S promoter. To knock down GmSPL9d, an artificial miRNA targeting GmSPL9d was constructed. An artificial miRNA (amiRNA) targeting GmSPL9d was generated following the algorithm and the protocol described at http://wmd.weigelworld.org. We selected the best amiRNA sequences directed against GmSPL9d. Primers to create the complete amiRNA‐GmSPL9d precursor sequences were retrieved following the algorithm at http://wmd.weigelworld.org. Overlapping PCR was performed on an amiRNA template (plasmid pRS300, containing the Arabidopsis miR319a precursor) using the protocol provided by Weigelworld. PCR products were cloned in pDONOR207 (Invitrogen). amiRNA silencing and recovery sequences were inserted into the Gateway destination binary vector pMDC32 (EV‐4). All primers used for plasmid construction are listed in Table S3.

CRISPR/Cas9 technology was used to knock out miR156. First, the precursor of miR156b/f was analysed using the online software Crispr‐P (http://cbi.hzau.edu.cn/crispr/), and two most reliable single guide RNAs (sgRNAs) (GAAGAGAGAGAGCACAACCC and GATGATGACAGAGGAAGAGA) were obtained. For each target locus, a pair of DNA oligos were synthesized from Shanghai Sangon Biotechnology Co. Ltd (Shanghai, China) and annealed to generate dimers. The target sequences were cloned into the sgRNA expression cassettes of the pYLCRISPR/Cas9P35S‐B vector as previously described (Ma et al., 2015). The miR156b/f‐CRISPR/Cas9 construct was validated by sequencing, and the plasmids were then transformed into A. rhizogenes strain K599 for hairy root transformation. The genomic sequence of miR156b/f was amplified with validation primers miR156b‐criJD and miR156f‐criJD (Table S3) from individual transgenic hairy roots and confirmed for gene editing by sequencing. All of the primers used for plasmid construction are listed in Table S3.

CRISPR/Cas9 technology was used to knock out GmSPL9d. First, the nucleotide sequence of GmSPL9d was analysed using the online software Crispr‐P (http://cbi.hzau.edu.cn/crispr/), and the two most reliable sgRNAs, TCCTCTTCTGAGTCCCTCAACGG and CCTTCAACTTGACACCTGGGAGG, were selected. Then, two AtU6 promoter‐sgRNA‐AtU6 terminator cassettes were amplified using the vector pCBC‐DT1T2 as a template. Next, the two fragments were cloned by Golden Gate reaction into two BsaI sites of the backbone vector pKSE401‐GFP (EV‐5) and transformed into chemically competent Escherichia coli DH5α (Tang et al., 2018). The GmSPL9d‐CRISPR/Cas9 construction was validated by sequencing, and the plasmids were then transformed into A. rhizogenes strain K599 for hairy root transformation. The genomic sequence of GmSPL9d was amplified with the validation primer GmSPL9‐criJD (Table S3) from individual transgenic hairy roots and confirmed for gene editing by sequencing. The primers used for plasmid construction are listed in Table S3.

In situ hybridization

A 380 nucleotide (nt) fragment of the GmSPL9d 5′‐untranslated region, a 240 nt fragment of the GmENOD40‐1 coding sequence and a 163 nt fragment of the GmNINa coding sequence were amplified from Williams 82 cDNA using KOD Polymerase (Stratagene, San Diego, CA, USA) and cloned into pSPT 18 (Roche) using specific primers (Table S4). Digoxigenin‐labelled sense or antisense probes were synthesized with T7 or SP6 RNA Polymerase (Roche). Roots inoculated with rhizobia and nodules at different stages were collected and fixed in 4% paraformaldehyde. Paraffin‐embedded materials were sectioned at a thickness of 8 μm. After the sections had been deparaffinized and dehydrated, hybridization and detection were performed as described previously (Long et al., 1996).

Protein expression constructs and protein purification

For purification of the recombinant GST‐GmSPL9d protein, the open‐reading frame of GmSPL9d was cloned into pGEX‐4T‐1. The GST‐GmSPL9d expression plasmid was then transformed into E. coli strain BL21. Protein purification was performed using glutathione agarose (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions. The primers used to construct the plasmid are listed in Table S3.

Electrophoretic mobility shift assay

Electrophoretic mobility shift assays (EMSAs) were performed using the LightShift Chemiluminescent EMSA Kit (Pierce, Rockford, IL, USA) according to the manufacturer’s protocol and as described by Yoo et al. (2010). The DNA binding activity of the GST‐GmSPL9d protein was analysed using two oligonucleotides containing the GTAC motifs present in the GmNINa, miR172c and GmENOD40‐1 promoters labelled with biotin at the 5′ end (Sangon) (the primers used are shown in Table S5). After incubation for 45 min at 25°C, the protein probe mixture was separated on a 6% polyacrylamide gel and transplanted to a Biodyne B nylon membrane (Pall, New York, NY, USA). Migration of the biotin‐labelled probes was detected using streptavidin‐horseradish peroxidase conjugates that bind to biotin and the chemiluminescent substrate according to the manufacturer’s protocol.

ChIP‐qPCR assay

Two grams of 10 DAI stable transgenic plant roots containing 35S::GmSPL9d‐GFP and control W82 plants were selected for the chromatin immunoprecipitation (ChIP) assay. Roots were ground to a fine power in liquid nitrogen, and the nuclei were isolated. The nuclear extract was cross‐linked with 1% formaldehyde for 30 min under vacuum; cross‐linking was stopped with 0.125 M glycine. Immunoprecipitation was performed with an anti‐green fluorescent protein (GFP) antibody and protein G beads. Immunoprecipitation in the absence of anti‐GFP served as the control. qRT‐PCR analysis was performed using specific primers corresponding to different promoter regions of GmNINa, miR172c and GmENOD40‐1, and GmELF1B was used as an internal control (primers used are shown in Table S6).

Bioinformatics analysis

ClustalX 1.83 was used to compare the difference between miR156 family members based on mature sequences. Using Mega5 software, a phylogenetic tree was reconstructed based on precursor sequences (http://www.mirbase.org/) of miR156 family members. The amino acid sequences of GmSPL9 and AtSPL9 were aligned using ClustalX 1.83. The genome structure of GmSPL9 family members was determined via http://gsds.cbi.pku.edu.cn/.

Statistical analysis

All data were analysed using Sigmaplot 10.0 (Systat Software, Chicago, IL, USA) and GraphPad Prism 5 (GraphPad Software, La Jolla, CA, USA). The averages and SD of all data were calculated, and Student’s t‐test was performed to generate P values. In the figures, statistically significant differences are marked as follows: *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Results

miR156 family members are differentially expressed, and miR156b/f are downregulated during nodulation

To understand the role of miR156 family members in soybean nodulation, we first conducted a genome‐wide search for miR156 family members in the soybean genome. A total of 25 miR156 family members were found in soybean (Fig. 1a). The mature miR156 family members are 20–22 nt in length and share high sequence homology with changes of 1–5 nt (Fig. 1a). Some mature miR156 members, such as miR156a/h/u/v/w/x/y, miR156c/d/i/j/l/m, miR156k/n/o, miR156p/t and miR156q/s, have identical sequences. However, the sequences, lengths and structures of the precursor miR156(s) (pre‐miRNA156) are highly varied, with the maximum conservation among the sequences in the stem section of the hairpin structures where mature miR156(s) are produced (Fig. S2). Phylogenetic analysis revealed that the miR156 family is conserved among Glycine max, L. japonicus, M. trunctula and Arabidopsis thaliana but has markedly different family numbers. Soybean has the largest number of miR156 genes (25 miR156s), while others have c. 10 miR156 family members (Fig. 1b).

To identify the members of miR156 that are involved in soybean nodulation, we analysed the expression profiles of miR156 family members in soybean roots inoculated without or with compatible B. diazoefficiens strain USDA110 at the specified time points. Due to the identical sequences of some mature miR156(s) (e.g. miR156a/h/u/v/w/x/y, miR156c/d/i/j/l/m, miR156k/n/o, miR156p/t and miR156q/s), the expression of these miR156s was the sum of the identical mature miR156(s). As shown in Fig. 1(c–l), all the miR156(s) exhibited divergent expression patterns during nodulation. For example, expression levels of miR156a/h/u/v/w/x/y, miR156c/d/i/j/l/m, miR156e, miR156k/n/o and miR156p/t in noninfected roots gradually decreased, and expression in infected roots first increased and then decreased. Ixpression in infected roots on days 1 and 3 was higher than that in noninfected roots, indicating that rhizobia were induced on days 1 and 3 (Fig. 1c,e,f,i,j). Expression of miR156g in noninfected roots gradually increased, and that in infected roots first increased and was then downregulated. Expression was induced by rhizobia at 3 d (Fig. 1h). Expression of miR156s/q in infected and noninfected roots gradually increased, and expression was induced by rhizobia at 3 and 5 d (Fig. 1k). Expression of miR156r in infected and noninfected roots was first upregulated, then decreased and then increased (Fig. 1l). Notably, expression of miR156b and miR156f in infected roots was rapidly downregulated at 1 DAI, although their expression was also gradually downregulated over the duration of the experiment (Fig. 1d,g). Dynamic and differential expression of these miR156 members suggests different roles of miR156 family genes during early nodulation and response to rhizobia.

miR156b and miR156f are negative regulators of soybean nodulation

Previously, we showed that miR172c is continuously upregulated in roots within 10 DAI (Wang et al., 2014). It is well known that miR156 displays an opposite expression pattern to miR172 during the growth phase transition (Guo et al., 2017). The fact that the expression trends of miR156b and miR156f were opposite to that of miR172c during nodulation (Fig. 1d,g) suggests that these two miRNAs may play opposite roles to miR172c. Phylogenetic analysis revealed that miR156b and miR156f are highly linked miRNAs belonging to one clade with nearly identical mature sequences (Fig. 1a,b) and highly conserved precursor sequences (Fig. S3). To determine whether endogenous miR156b and miR156f are required for nodulation, we specifically knocked down miR156b and miR156f using short tandem target simulation (STTM) technology (Yan et al., 2012) (Fig. S4). Indeed, STTM156‐88‐expressing roots had a significantly reduced level of miR156b and miR156f expression (Figs 2a,b, S5), and these STTM156‐88 transgenic hairy roots produced a significantly increased number of root nodules compared with the EV control at 28 DAI (Fig. 2c,d). However, expression of other miR156 family members in miR156b STTM hairy roots was also reduced (Fig. S5). To prove the roles of miR156b and miR156f in soybean nodulation, we used the CRISPR/Cas9 system to specifically knock out both miR156b and miR156f in soybean roots (Fig. S6). The mir156b/f composite plants exhibited significantly increased root hair deformation (Fig. 2e–g) and nodule number compared with the control plants (Fig. 2h,i), thus confirming the inhibitory role of miR156b and miR156f in nodulation. Since rhizobial inoculation caused continuously decreased miR156b transcript levels (Fig. 1d), miR156b was selected for further analysis.

Fig. 2.

Fig. 2

Open in a new tab

miR156b and miR156f negatively regulate soybean nodulation. (a, b) Relative expression levels of miR156b (a) and miR156f (b) in the empty vector EV‐1 and STTM156‐88 transgenic roots. miR1520d was the reference gene. At least three transgenic roots were analysed for qRT‐PCR. Student’s t‐test was used to identify significant differences; ***, P < 0.001. (c) Nodule performance of individual hairy roots transformed with EV‐1 and STTM156‐88 at 28 d after inoculation (DAI). Bar, 2 cm. (d) Quantitative data of nodule number per root transformed with EV1 and STTM156‐88 at 28 DAI. Data are given as mean ± SD. More than 20 hairy roots were analysed in each individual experiment. Student’s t‐test was used to identify significant differences; **, P < 0.01. (e–g) Curled root hair number of the hairy roots transformed with the empty vector (EV‐2) and mir156b/f at 6 DAI. The asterisk marks the curled root hair. (e, f) Quantification of the curled root hairs in the transgenic lines (n ≥ 5). Values are the average ± SD from three independent experiments (g). Student’s t‐test was used to identify significant differences; **, P < 0.01. Bar, 20 μm. (h) Nodule phenotypes of mir156b/f roots or hairy roots transformed with the empty vector EV‐2 at 28 DAI. Bar, 2 cm. (i) Quantitative analysis of nodule number in mir156b/f and EV‐2 control roots. Data are given as mean ± SD. More than 10 hairy roots were analysed in each biological repeat. Student’s t‐test was used to identify significant differences; **, P < 0.01.

To test whether miR156b is involved in nodulation, we generated composite plants harbouring hairy roots overexpressing miR156b under control of the CaMV 35S (35S) promoter that were inoculated with rhizobia. The level of miR156b was markedly increased in the miR156b‐overexpressing hairy roots compared with that of the EV control roots (Fig. 3a). The transformed roots overexpressing miR156b formed significantly fewer nodules than the EV controls at 28 DAI (Fig. 3b,c). To further prove the role of miR156b, we evaluated the number of curled root hairs and nodule number of transgenic plants overexpressing miR156b under control of the CaMV 35S (35S) promoter (Sun et al., 2019). Expression levels of miR156b were markedly increased in both miR156bOE lines, and that of miR156b was much higher in miR156bOE5 (Fig. 3d), which is consistent with our previous results (Sun et al., 2019). The numbers of curled root hairs and nodules were significantly decreased in the miR156bOE lines compared to the WT control (Fig. 3e–i). Together, these data confirm that miR156b indeed plays a negative regulatory role in soybean nodulation.

Fig. 3.

Fig. 3

Open in a new tab

miR156b and miR156f negatively regulate soybean nodulation. (a) Relative expression levels of miR156b in EV‐1 and 35S::miR156b hairy roots. miR1520d was the reference gene. At least three transgenic roots were analysed for qRT‐PCR. Values are the mean ± SD from three independent experiments. Student’s t‐test was used to identify significant differences; ***, P < 0.001. (b) Nodule performance of individual hairy roots transformed with EV‐1 and 35S::miR156b at 28 DAI. Bar, 2 cm. (c) Quantitative data of nodule number per hairy root transformed with EV and 35S::miR156b at 28 DAI. Values are the mean ± SD. More than 20 hairy roots were analysed in each individual experiment. Student’s t‐test was used to identify significant differences; ***, P < 0.001. (d) Expression levels of miR156b in two miR156bOE5 and miR156bOE11 transgenic lines and wild‐type roots. miR1520d was used as the reference gene. Values are the mean ± SD from three independent experiments. Student’s t‐test was used to identify significant differences; ***, P < 0.001. (e–g) Curled root hair number in miR156bOE5 and WT plants at 6 DAI. The asterisk marks the curled root hair. (e, f) Quantification of the curled root hairs in miR156bOE5 and WT roots (n ≥ 5). Values are the average ± SD from three independent experiments (g). Student’s t‐test was used to identify significant differences; ***, P < 0.001. Bar, 20 μm. (h) Nodule performance of miR156bOE5 and miR156bOE11 lines and wild‐type plants (WT). Bar, 2 cm. (i) Quantitative analysis of nodule numbers in miR156bOE and WT plants. Values are the mean ± SD from three independent experiments. Student’s t‐test was used to identify significant differences between miR156bOE lines and WT: *, P < 0.05; ***, P < 0.001.

GmSPL9d positively regulates soybean nodulation and mediates miR156b inhibition of nodulation

Previously, we showed that GmSPL6c and SPL9d can be cleaved by miR156b and that miR156b overexpression caused a marked reduction in GmSPL6c and SPL9d expression in soybean mature nodules (Sun et al., 2019), which suggests that GmSPL6c and SPL9d are potential target genes of miR156b in nodulation. To identify the main target of miR156b in early nodulation, we analysed expression levels of all the GmSPL6 (GmSPL6a‐e) and GmSPL9 (GmSPL9a‐d) genes in the roots of the composite plants overexpressing miR156b or STTM156‐88 at 5 DAI. All the GmSPL genes were downregulated except GmSPL9a in miR156b‐overexpressing roots. Among them, gene expression levels of GmSPL6a, GmSPL9b and GmSPL9d were significantly lower in miR156‐overexpressing roots than those in the vector control samples (Fig. 4a). By contrast, all the GmSPL genes were significantly upregulated except GmSPL6b and GmSPL6d in STTM156‐88 roots. Notably, the levels of GmSPL6c, GmSPL6e, GmSPL9b and GmSPL9d mRNA in STTM156‐88 roots were remarkably higher than that in the vector controls (Fig. 4a). Next, we analysed the temporal expression patterns of these genes during early nodulation and found that only GmSPL9b and GmSPL9d expression was induced by rhizobia (Fig. 4b–g), and the expression pattern was opposite to that of miR156b (Fig. 1d). Both GmSPL9b and GmSPL9d contain one miR156b complementary site and can be effectively cleaved (Sun et al., 2019) (Fig. S7a). Taking this together, we speculated that GmSPL9b and GmSPL9d are the main target genes of miR156b in early nodulation. Since GmSPL9d showed the strongest induction by rhizobia (Fig. 4e,g) and displayed a stark opposite expression trend to miR156b, we selected GmSPL9d for further functional research.

Fig. 4.

Fig. 4

Open in a new tab

GmSPL9d is the main target gene of miR156b during soybean nodulation. (a) Expression analysis of GmSPL6 and GmSPL9 family members in 35S::miR156b, STTM156‐88 and empty vector (EV‐1) hairy roots at 5 d after inoculation (DAI). GmELF1B was the reference gene for qRT‐PCR analysis. Student’s t‐test was used to identify significant differences compared with EV‐1: *, P < 0.05; **, P < 0.01; ***, P < 0.001. (b–g) Expression levels of GmSPL6a (b), GmSPL6c (c), GmSPL6e (d), GmSPL9b (e), GmSPL9c (f) and GmSPL9d (g) in uninoculated (−R) and inoculated (+R) roots at the indicated time points. GmELF1B served as the reference gene for qRT‐PCR analysis. All experiments consisted of three independent biological replicates. Data are given as mean ± SD from three individual experiments.

The conserved SBP domain of GmSPL9d (Fig. S7b) and nuclear localization (Fig. S7c) indicate that it may function as an SBP transcription factor. To verify the function of GmSPL9d in nodulation, hairy roots with increased and reduced levels of GmSPL9d were generated and inoculated with B. diazoefficiens USDA110. The hairy roots overexpressing GmSPL9d formed significantly more nodules than the EV control roots (Fig. 5a–c). By contrast, the number of nodules per GmSPL9d knockdown transgenic root expressing an amiRNA (amiSPL9d) was significantly reduced compared with that of the roots expressing the empty vector (Fig. 5d–f). In addition, we generated GmSPL9d knockout hairy roots using CRISPR/Cas9 technology and found that the nodule number per gmspl9d root was dramatically reduced (Figs 5g,h, S8). The results demonstrated that GmSPL9d is a positive regulator of soybean nodulation.

Fig. 5.

Fig. 5

Open in a new tab

GmSPL9d positively regulates soybean nodulation. (a) Nodule performance of individual hairy roots expressing EV3 and 35S::GmSPL9d at 28 DAI. Bar, 2 cm. (b) Relative expression levels of GmSPL9d in EV3 and 35S::GmSPL9d hairy roots. GmELF1B was the reference gene. (c) Quantitative analysis of nodule number per transgenic root expressing EV3 and 35S::GmSPL9d. (d) Nodule performance of individual transgenic roots expressing EV4 and amiGmSPL9d at 28 d after inoculation (DAI). Bar, 2 cm. (e) Relative expression levels of GmSPL9d in EV4 and amiGmSPL9d transgenic roots. GmELF1B was the reference gene. (f) Quantitative data of nodule number per transgenic root expressing EV4 and amiGmSPL9d. (g) Nodule performance of hairy roots transformed with empty vector (EV) and GmSPL9d‐CRISPR (gmspl9d) at 28 DAI. Bar, 2 cm. (h) Quantitative analysis of nodule number in EV5 and gmspl9d roots. Data are the average of three biological repeats. Error bars indicate SD. At least three transgenic roots were used for qRT‐PCR analysis, and seven transgenic roots were used for nodule number data. Student’s t‐test was used to identify significant differences: *, P < 0.05; **, P < 0.01; ***, P < 0.001.

To prove the genetic relationship between miR156b and GmSPL9d in nodulation, we performed a genetic complementation test between miR156bOE and overexpression of GmSPL9d or 7mGmSPL9d (a mutant GmSPL9d DNA with seven mismatches at the miR156b complementary site without affecting amino acids) (Sun et al., 2019). Overexpressing GmSPL9d or miR156b‐resistant 7mGmSPL9d completely restored nodulation inhibition by miR156b overexpression regardless of whether they were driven by the 35S promoter or the native promoter of GmSPL9d (Fig. 6a–h). Together, these data suggest that GmSPL9d is the main target of miR156b in soybean nodulation.

Fig. 6.

Fig. 6

Open in a new tab

GmSPL9d recovers the nodulation capacity inhibited by miR156b. (a–f) Nodule phenotypes of Williams 82 (WT) hairy roots transformed with EV‐3 (a), 35S::miR156b and EV‐3 (b), 35S::miR156b and 35S::GmSPL9d (c), 35S::miR156b and 35S::7mGmSPL9d (d), 35S::miR156b and GmSPL9d::GmSPL9d (e), and 35S::miR156b and GmSPL9d::7mGmSPL9d (f). Bar, 2 cm. (g) Expression of GmSPL9d in the above transgenic hairy roots. (h) Quantitative analysis of nodule number per transgenic root expressing different gene combinations. All experiments consisted of three independent biological replicates. Data are the average of three biological repeats. Error bars indicate SD. Different letters indicate significant differences by the Student‐Newman‐Kuels test (P < 0.05).

The miR156b‐GmSPL9d module regulates multiple core nodulation genes

To uncover the molecular basis of soybean nodulation mediated by the miR156b‐GmSPL9d module, we analysed the expression profiles of symbiosis marker genes in soybean, including GmNINa, miR172c and GmENOD40‐1, in infected soybean roots with altered expression of miR156b and GmSPL9d. First, we detected the expression of these three marker genes in uninoculated (−R) and inoculated (+R) soybean W82 roots and found that these three genes were induced by rhizobia (Fig. 7a–c). Overexpression of miR156b resulted in reduced expression of GmNINa, miR172c and GmENOD40‐1 (Fig. 7d–f). By contrast, reduced miR156 in the miR156b‐STTM transgenic roots caused upregulation of these marker genes (Fig. 7d–f).

Fig. 7.

Fig. 7

Open in a new tab

GmSPL9d positively regulates GmNINa, miR172c and GmENOD40‐1. (a–c) Expression of GmNINa (a), miR172c (b) and GmENOD40‐1 (c) in uninoculated (−R) and inoculated (+R) soybean W82 roots. Roots were collected at 0, 1, 3 and 5 d after inoculation (DAI), and the corresponding uninoculated roots were collected. (d–f) qRT‐PCR analysis of GmNINa (d), miR172c (e) and GmENOD40‐1 (f) in transgenic roots transformed with empty vector (EV1), 35S::miR156b and STTM156‐88 at 5 DAI. Transcripts of GmNINa and GmENOD40‐1 in each sample were normalized to those of GmELF1B, while miR172c expression was normalized to that of miR1520d. (g–i) qRT‐PCR analysis of GmNINa (g), miR172c (h) and GmENOD40‐1 (i) in roots transformed with empty vector (EV3) and 35S::SPL9d at 5 DAI. (j–l) qRT‐PCR analysis of GmNINa (j), miR172c (k) and GmENOD40‐1 (l) in roots transformed with empty vector (EV4) and amiSPL9d at 5 DAI. GmNINa and GmENOD40‐1 in each sample were normalized to those of GmELF1B, while miR172c expression was normalized to that of miR1520d. Data are the mean of three biological repeats. Error bars indicate SD. At least three single transformed hairy root samples were collected for qRT‐PCR. Student’s t‐test was used to identify significant differences: *, P < 0.05; **, P < 0.01; ***, P < 0.001.

To test whether miR156b regulates GmNINa, miR172c and GmENOD40‐1 transcription through its target gene GmSPL9d, we also examined expression levels of these genes in 35S::GmSPL9d transgenic roots inoculated with rhizobia and uninoculated roots. Expression levels of miR172c, GmNINa and GmENOD40‐1 were markedly increased in the GmSPL9d‐overexpressing transgenic roots compared with those in EV control roots. Even after inoculation with water, these three genes were also induced (Fig. 7g–i). We also detected the expression of these three nodulation‐necessary genes in amiSPL9d hairy roots. The results showed that upregulation of GmNINa, miR172 and GmENOD40‐1 was impaired in amiSPL9d roots after inoculation with rhizobia (Fig. 7j–l). Since GmNINa, miR172c and GmENOD40‐1 are important components in the nodulation signalling pathway, these results suggest that the miR156b/GmSPL9d module may function upstream of these genes in nodulation signalling during soybean nodulation.

GmSPL9d is coexpressed with GmNINa and GmENOD40‐1 during nodulation

To confirm the relationship between GmSPL9d and GmNINa and GmENOD40‐1 during soybean nodulation, we analysed the spatial expression patterns of these genes using in situ hybridization. At 3 DAI, GmSPL9d, GmNINa and GmENOD40‐1 mRNAs were clearly detected in the outer dividing cortex cells below the infected epidemic cells, cortex cells and pericycle cells, although there was a weak GmSPL9d signal in the pericycle cells (Figs 8a–c, S9a–c). During nodule primordia formation at 5 and 7 DAI, GmSPL9d, GmNINa and GmENOD40‐1 were highly expressed in nodule primordia, pericycle cells and dividing cell clusters of the cortex (Figs 8d–i, S9d–i). The pattern of GmENOD40‐1 expression detected here was the same as that detected in previous research (Yang et al.,1993). In young nodules at 14 DAI, GmSPL9d, GmNINa and GmENOD40‐1 were predominantly expressed in the boundary cell layer of the infected zone and the vascular bundles (Figs 8j–l, S9j–l), and this particular expression pattern of GmSPL9d, GmNINa and GmENOD40‐1 was maintained up to nodule maturation at 28 DAI (Figs 8m–o, S9m–o). Notably, weaker expression of these three genes was also observed in the infected zone of developing and mature nodules. Coexpression of GmSPL9d and GmNINa and GmENOD40‐1 during early nodulation suggests that these genes act together in rhizobial infection, nodule organogenesis and functionality of mature nodules in soybean.

Fig. 8.

Fig. 8

Open in a new tab

In situ hybridization of GmSPL9d, GmNINa and GmENOD40‐1 transcripts during soybean nodulation. Positive hybridization signals are visible as dark blue to violet colour development. (a–c) Transverse sections of root segments at 3 d after seed sowing and inoculation (DAI). Hybridization signals are visible at the pericycle of the root stele (indicated by red arrowhead) and the infected site (indicated by black arrowhead). p, pericycle; is, infected site; oc, outer cortex; ic, inner cortex. Bar, 200 μm. (d–f) Transverse sections of root segments at 5 DAI. Signals are detected on the pericycle, nodule meristem (indicated by blue arrowhead), and dividing cell clusters of the inner cortex between the pericycle and nodule meristem. nm, nodule meristem. Bar, 200 μm. (g–i) Transverse sections of roots at 7 DAI, at which nodule primordia formed in the outer cortex. Signals can be detected clearly in nodule primordia (indicated by yellow arrowhead) and in dividing cell clusters of the inner cortex between pericycle and nodule primordia and GmENOD40‐1 signal on the peripheral cells of the nodule meristem. np, nodule primordia. Bar, 200 μm. (j–l) Transverse sections of young nodules at 14 DAI. Hybridization signals were clearly detected in the boundary layer of the infected zone and in the vascular bundles of endodermis (indicated by black arrow). iz, infected zone; b, boundary cell layer of infected zone; e, vascular bundles of endodermis. Bar, 200 μm. (m–o) Transverse sections of mature nodules at 28 DAI. Hybridization signals are similarly detected as in young nodules. The GmSPL9d transcript was also partially localized in the infected zone (indicated by the yellow arrow). Blank arrowheads indicate vascular bundles of endodermis. Bar, 200 μm. The experiments were repeated at least three times, and representative images are presented.

GmSPL9d directly binds to the promoters of GmNINa, miR172c and GmENOD40‐1

Previously, we showed that GmNINa activates miR172c, which then regulates GmENOD40‐1 expression through NNC1 (Wang et al., 2014, 2019). Based on the observation that miR156b and GmSPL9d affect GmNINa, miR172c and GmENOD40‐1 expression, we hypothesized that the miR156b‐GmSPL9d module may act upstream of GmNINa to regulate expression of miR172c and GmENOD40‐1 during nodulation. To test this hypothesis, we first analysed the promoters of the GmNINa gene to determine whether there are any cis elements (GTACs) for SPL transcription factors. The results showed that the GmNINa promoter (2 kb in length) contains seven typical cis‐regulatory elements for the binding of SPL transcription factors (Fig. S10a), which suggests that GmSPL9d may directly bind to the promoter of GmNINa. To prove this hypothesis, we conducted an EMSA to detect GmSPL9d and GmNINa promoter interactions using a GST‐GmSPL9d recombinant protein. Two DNA fragments containing GTAC cis‐elements located nearest to the start code of GmNINa were chosen as the specific probes for EMSA. Indeed, specific shifted bands were detected when the DNA probes were incubated with the GST‐GmSPL9d fusion protein (Figs 9a, S10b). By contrast, no shifted bands were observed when the DNA probes were incubated with the GST protein only. The shifted bands disappeared when GmSPL9d was incubated with competitors (Fig. 9a), confirming that GmSPL9d directly and specifically binds to the GmNINa promoter. To confirm the direct binding of GmSPL9d to the GmNINa promoter, we also conducted a ChIP assay and found that GmSPL9d was enriched in the regions containing GTAC cis elements (Fig. 9d). These results indicate that the GmSPL9d protein could associate with the GmNINa promoter in vivo and in vitro.

Fig. 9.

Fig. 9

Open in a new tab

GmSPL9d protein binds to the promoters of GmNINa, miR172c and GmENOD40‐1 in vitro and in vivo. (a–c) EMSA to analyse the binding of GmSPL9d to specific DNA probes containing the SPL binding sites from the GmNINa (a), miR172c (b) and GmENOD40‐1 (c) promoters. Competitions for GmSPL9d binding were performed using 200× competitive and 400× competitive probes, and GST protein only was used as a negative control. The red arrowhead points to shifted bands. (d–f) ChIP assays showing the specific enrichment of GmSPL9d in the promoters of GmNINa (d), miR172c (e) and GmENOD40‐1 (f). DNA fragments corresponding to the regions indicated were analysed by qPCR. The DNA fragments were normalized to the input data. GmELF1B was employed as an internal control. All experiments were repeated three times. Student’s t‐test was performed. Asterisks indicate significant differences from the empty vector control: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant.

It is well known that miR156 regulates miR172 through its SPL target genes (Wang, 2014). This led us to explore whether GmSPL9d can also bind to the promoter of miR172c. There are two GTAC cis‐elements for GmSPL9d binding within a 2 kb promoter region of miR172c (Fig. S10a), thus suggesting that GmSPL9d may bind to the promoter of miR172c. To test this hypothesis, we performed EMSA, and shifted bands were clearly detected when the DNA probes containing GTAC in the miR172c promoter were incubated with the GmSPL9d protein, whereas no shifted bands were observed when the probes were incubated with the GST protein only (Fig. 9b). The direct binding of GmSPL9d to the two regions containing GmSPL9d binding sites of the miR172c promoter was also confirmed by the ChIP assay (Fig. 9e).

In silico analysis revealed that the GmENOD40‐1 promoter also contains four cis elements for GmSPL9d binding (Fig. S10a), indicating that GmSPL9d may also directly target the GmENOD40‐1 gene. To test this possibility, we performed EMSA and ChIP assays as described above. Indeed, GmSPL9d showed a specific interaction with the GTAC‐containing probe nearest to the start codon of GmENOD40‐1 but not the adjacent probe containing the cis element (Figs 9c, S10b). This result confirmed the specific binding of GmSPL9d to the GmENOD40‐1 promoter and revealed the crucial effects of the GTAC flanking sequences on GmSPL9d binding to the GmENOD40‐1 promoter. The ChIP assay results supported the binding of GmSPL9d to the specific site of the GmENOD40‐1 promoter (Fig. 9f). Together, these results proved that GmSPL9d could directly associate with both the promoters of GmNINa, miR172c and GmENOD40‐1 promoters.

Discussion

Symbiotic nodulation is a complex process that involves the dynamics of NF receptors and the downstream components of signalling pathways in various cells. The nodulation signalling pathway that mediates rhizobial infection and nodule organogenesis has been intensively studied, and many core components of the nodulation signalling pathway have been identified (Gamas et al., 2017; Ferguson et al., 2019). However, how symbiotic signals are transduced from NF receptors to downstream signal components and how the regulatory networks that control various aspects of nodulation are coordinated remain poorly understood. The results presented here provide new insight into the existence of the mechanism by which miR156 regulates multiple nodulation genes through its target GmSPL9d to spatially and temporally coordinate the expression of these nodulation genes.

A previous study showed that miR156 negatively regulates soybean nodulation (Yan et al., 2013). As an evolutionarily conserved miRNA gene family, the miR156 family usually has multiple members. However, there is still a lack of understanding of exactly how miR156 family genes are involved in soybean nodulation. In this study, we provided several lines of evidence to show that miR156b is a key family member that negatively regulates soybean nodulation by repressing its target gene GmSPL9d. First, we showed that 25 miR156 family genes were differentially expressed during soybean nodulation (Fig. 1). Intriguingly, all of the miR156s were responsive to rhizobial infection. Only miR156b displayed a continuous decline in expression with infection progression (Fig. 1d), which is opposite to the continuous upregulation of miR172c (Wang et al., 2014). Second, knockout and overexpression of miR156b resulted in an increased or reduced number of nodules, respectively. Third, miR156b regulates nodulation mainly by repressing its target gene GmSPL9d. GmSPL9d expression was downregulated or upregulated by miR156b overexpression or knockdown, respectively (Fig. 4a). Expression of GmSPL9d was greatly induced by rhizobial infection in soybean roots (Fig. 4g). Moreover, overexpression or silencing of GmSPL9d significantly increased or reduced the number of nodules, respectively (Fig. 5). Thus, GmSPL9d acts as a main target gene of miR156b in soybean nodulation. Since several other target genes, such as GmSPL6c, GmSPL6e and GmSPL9b, were also affected by knockout and overexpression of miR156b (Fig. 4), it cannot be rule out that other GmSPL members could act synergically in earlier or later stages of root nodule symbiosis.

miR156a was first identified as a central regulator that controls the phase transition from vegetative to reproductive stages (J. W. Wang et al., 2009; Wu et al., 2009). miR156a is continuously downregulated during plant growth and maturation and sequentially activates miR172 through SPL, which positively switches vegetation to the flowering stage (Fornara & Coupland, 2009; Wu et al., 2009; Wang, 2014; Wang & Wang, 2015). Since miR156 and miR172 antogonistically regulate soybean nodulation (Yan et al., 2013; Wang et al., 2014), it is generally assumed that miR156 may exert its function by regulating miR172, but no evidence has been provided to support the relationship between miR156 and miR172 in soybean nodulation. Previously, we showed that GmSPL9d is a typical SPL family transcription factor that interacts with other transcription factors in the nucleus to transcriptionally regulate downstream gene expression, resulting in shoot branching (Sun et al., 2019). Here, we observed a similar inhibitory effect of miR156b on miR172c expression (Fig. 7e). Importantly, we proved that GmSPL9d is a nuclear protein (Fig. S7) that can directly bind to the promoter of miR172c (Fig. 9b,e). Thus, our results confirm that miR156 and miR172 indeed regulate the progression of nodulation in soybean. In our previous work, we showed that GmNNC1 inhibits miR172c, but GmNINa activates miR172c, which regulates nodule formation by inducing the expression of ENOD40s (Wang et al., 2014, 2019). Thus, it is conceivable that miR172 expression is regulated by both GmNNC1, GmNINa and the miR156‐GmSPL module during soybean nodulation.

Intriguingly, we found that miR156b‐GmSPL9d also directly targeted more nodulation genes in soybean. These genes include the upstream central regulator GmNINa and downstream key effector GmENOD40‐1 in soybean (Yang et al., 1993; Wang et al., 2019). We found that overexpression and knockdown of miR156b resulted in up‐ and downregulation of GmNINa and GmENOD40‐1, respectively (Fig. 7d,f), and knockdown of GmSPL9d downregulated expression of GmNINa and GmENOD40‐1 (Fig. 7j,l). These results suggest that miR156b negatively regulates GmNINa and GmENOD40‐1 through GmSPL9d. Furthermore, our in situ hybridization results revealed that GmSPL9d was largely coexpressed with GmNINa and GmENOD40‐1 during rhizobial infection, nodule formation and nodule development (Fig. 8). Moreover, both the GmNINa and the GmENOD40‐1 promoters contain several cis elements for binding the GmSPL9d protein (Fig. S10a) and, most importantly, GmSPL9d can directly bind to both the GmNINa and the GmENOD40‐1 promoters (Figs 9a,c,d,f, S10). Thus, we conclude that the miR156b‐GmSPL9d module exerts its regulatory function in soybean nodulation by directly activating expression of multiple core genes in the nodulation signalling pathway. Given that GmSPL9d was expressed during early nodulation and nodule development (Fig. 8), it is conceivable that miR156 dynamically controls the process of nodulation through transcription regulatory networks regulated by GmSPL transcription factors in soybean. The observations that several key upstream regulatory genes, including the NIN and ENOD40 genes, were downregulated by LjmiR156a overexpression in L. japonicus (Wang et al., 2015) support the notion that the central regulatory role of miR156‐SPL modules on symbiotic nodulation may be conserved in leguminous plants, such as soybean and L. japonicus.

We have shown that upon rhizobial inoculation, GmNINa activates miR172c, which then regulates the expression of GmENOD40(s) and leads to nodule formation (Wang et al., 2014, 2019). This study not only identified miR156b‐GmSPL9d as the key module but also established the sequential relationship between miR156b and miR172c in soybean nodulation. Notably, our findings reveal a novel regulatory network by which miR156b acts as a master regulator to coordinate nodulation by regulating multiple key nodulation genes, such as GmNINa, miR172c and GmENOD40‐1, in soybean (Fig. 10). Because miR156, NIN, miR172 and ENOD40 are highly conserved in legume nodulation, our results will provide novel insight into mechanisms for genetic regulation of nodulation in legumes. Because L. japonicus miR156a overexpression also affects other early nodulation genes, such as NFR1, SymRK, POLLUX, CYCLOPS and NSP1, miR156‐SPL may target more genes involved in various aspects of rhizobial infection and nodule organogenesis (Wang et al., 2015). Since the miR156b‐GmSPL9d module plays a crucial role in soybean branching (Sun et al., 2019), we hypothesized that it may coordinately regulate nodulation and plant growth. Further research will help to dissect the complex regulatory networks that coordinate all processes in symbiotic nodulation and plant development of legumes.

Fig. 10.

Fig. 10

Open in a new tab

A schematic model for miR156b‐GmSPL9d module‐coordinated soybean nodulation. In response to rhizobial infection, miR156b expression is reduced, resulting in increased GmSPL9d and enhanced activation of GmNINa, miR172c and GmENOD40‐1, which positively regulate nodulation in soybean. Arrows indicate activation. Blunt ended arrows indicate inhibition.

Author contributions

XL, JY and ZS designed the research; JY, ZS, QJ, YW, CW, FZ and YL performed the experiments and data analyses; XL, ZS and JY wrote the manuscript. JY and ZS are joint first authors.

Supporting information

Fig. S1 Identification of the reverse primers and quantitative primers for miR156 family genes.

Fig. S2 The precursor sequence alignment of soybean miR156 family genes.

Fig. S3 The precursor sequence alignment of soybean miR156b and miR156f.

Fig. S4 Diagram of STTM156b/f‐88 structure showing the design strategy.

Fig. S5 Other members of the miR156 family are downregulated in STTM156‐88 hairy roots.

Fig. S6 Sequencing results of pre‐miR156b and pre‐miR156f in the roots of pre‐miR156b/f‐CRISPR (mir156b/f) mutants.

Fig. S7 GmSPL9b and GmSPL9d gene structure, GmSPL9d peptide sequence and subcellular localization of GmSPL9d protein.

Fig. S8 Sequencing results of GmSPL9d/CRISPR mutant roots.

Fig. S9 In situ hybridization of GmSPL9d, GmNINa and GmENOD40‐1 negative controls during nodulation.

Fig. S10 GmSPL9d binds to the GTAC motifs in the miR172c, GmENOD40‐1 and GmNINa promoters.

Table S1 The primers for reverse transcription and qPCR of miR156 family members.

Table S2 The primers used for qPCR analysis of miR156b target genes.

Table S3 The primers used for the vector constructions.

Table S4 The primers used for in situ hybridization.

Table S5 Probes for electrophoretic mobility shift assay experiments.

Table S6 The primers used for ChIP‐PCR assays.

Please note: Wiley Blackwell are not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

Click here for additional data file. (2.1MB, pdf)

Acknowledgements

We thank Dr Kan Wang (Iowa State University) for providing pTF101.1 vector, Dr Xigang Liu (Institute of Genetics and Developmental Biology of the Chinese Academy of Sciences) for technical assistance in the in situ hybridization experiments and Dr Wenxin Chen at China Agricultural University for kindly providing the B. diazoefficiens strain USDA 110. This programme was supported by the National Natural Science Foundation of China (31730066, 91540112 and 31961133029), the Ministry of Agriculture of the People’s Republic of China (2018ZX0800955B) and Huazhong Agricultural University’s Scientific and Technological Self‐innovation Foundation (2015RC014).

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. Aung B, Gao R, Gruber MY, Yuan Z, Sumarah M, Hannoufa A. 2017. MsmiR156 affects global gene expression and promotes root regenerative capacity and nitrogen fixation activity in alfalfa. Transgenic Research 26: 541–557. [DOI] [PubMed] [Google Scholar]
  2. Aung B, Gruber MY, Amyot L, Omari K, Bertrand A, Hannoufa A. 2015. MicroRNA156 as a promising tool for alfalfa improvement. Plant Biotechnology Journal 13: 779–790. [DOI] [PubMed] [Google Scholar]
  3. Broghammer A, Krusell L, Blaise M, Sauer J, Sullivan JT, Maolanon N, Vinther M, Lorentzen A, Madsen EB, Jensen KJ et al. 2012. Legume receptors perceive the rhizobial lipochitin oligosaccharide signal molecules by direct binding. Proceedings of the National Academy of Sciences, USA 109: 13859–13864. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Campalans A, Kondorosi A, Crespi M. 2004. Enod40, a short open reading frame‐containing mRNA, induces cytoplasmic localization of a nuclear RNA binding protein in Medicago truncatula . Plant Cell 16: 1047–1059. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Charon C, Sousa C, Crespi M, Kondorosi A. 1999. Alteration of ENOD40 expression modifies Medicago truncatula root nodule development induced by Sinorhizobium meliloti . Plant Cell 11: 1953–1965. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chen C, Ridzon DA, Broomer AJ, Zhou Z, Lee DH, Nguyen JT, Barbisin M, Xu N, Mahuvakar VR, Andersen MR et al. 2005. Real‐time quantification of microRNAs by stem‐loop RT‐PCR. Nucleic Acids Research 33: e179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Desbrosses GJ, Stougaard J. 2011. Root nodulation: a paradigm for how plant‐microbe symbiosis influences host developmental pathways. Cell Host & Microbe 10: 348–358. [DOI] [PubMed] [Google Scholar]
  8. Dong Z, Shi L, Wang Y, Chen L, Cai Z, Wang Y, Jin J, Li X. 2013. Identification and dynamic regulation of microRNAs involved in salt stress responses in functional soybean nodules by high‐throughput sequencing. International Journal of Molecular Sciences 14: 2717–2738. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Downie JA. 2014. Legume nodulation. Current Biology 24: R184–R190. [DOI] [PubMed] [Google Scholar]
  10. Fan F, Zhang F, Song Y, Sun J, Bao X, Guo T, Li L. 2006. Nitrogen fixation of faba bean (Vicia faba L.) interacting with a non‐legume in two contrasting intercropping systems. Plant and Soil 283: 275–286. [Google Scholar]
  11. Ferguson BJ, Indrasumunar A, Hayashi S, Lin MH, Lin YH, Reid DE, Gresshoff PM. 2010. Molecular analysis of legume nodule development and autoregulation. Journal of Integrative Plant Biology 52: 61–76. [DOI] [PubMed] [Google Scholar]
  12. Ferguson BJ, Mens C, Hastwell AH, Zhang M, Su H, Jones CH, Chu X, Gresshoff PM. 2019. Legume nodulation: the host controls the party. Plant, Cell & Environment 42: 41–51. [DOI] [PubMed] [Google Scholar]
  13. Fornara F, Coupland G. 2009. Plant phase transitions make a SPLash. Cell 138: 625–627. [DOI] [PubMed] [Google Scholar]
  14. Gamas P, Brault M, Jardinaud MF, Frugier F. 2017. Cytokinins in symbiotic nodulation: when, where, what for? Trends in Plant Science 22: 792–802. [DOI] [PubMed] [Google Scholar]
  15. Griesmann M, Chang Y, Liu X, Song Y, Haberer G, Crook MB, Billault‐Penneteau B, Lauressergues D, Keller J, Imanishi L et al. 2018. Phylogenomics reveals multiple losses of nitrogen‐fixing root nodule symbiosis. Science 361: eaat1743. [DOI] [PubMed] [Google Scholar]
  16. Guo C, Xu Y, Shi M, Lai Y, Wu X, Wang H, Zhu Z, Poethig RS, Wu G. 2017. Repression of miR156 by miR159 regulates the timing of the juvenile‐to‐adult transition in Arabidopsis. Plant Cell 29: 1293–1304. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Ha M, Kim VN. 2014. Regulation of microRNA biogenesis. Nature Reviews Molecular Cell Biology 15: 509. [DOI] [PubMed] [Google Scholar]
  18. Hardin SC, Tang G, Scholz A, Holtgraewe D, Winter H, Huber SC. 2003. Phosphorylation of sucrose synthase at serine 170: occurrence and possible role as a signal for proteolysis. The Plant Journal 35: 588–603. [DOI] [PubMed] [Google Scholar]
  19. Hirsch S, Kim J, Munoz A, Heckmann AB, Downie JA, Oldroyd GE. 2009. GRAS proteins form a DNA binding complex to induce gene expression during nodulation signaling in Medicago truncatula . Plant Cell 21: 545–557. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hofferek V, Mendrinna A, Gaude N, Krajinski F, Devers EA. 2014. MiR171h restricts root symbioses and shows like its target NSP2a complex transcriptional regulation in Medicago truncatula . BMC Plant Biology 14: 199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Holt DB, Gupta V, Meyer D, Abel NB, Andersen SU, Stougaard J, Markmann K. 2015. micro RNA 172 (miR172) signals epidermal infection and is expressed in cells primed for bacterial invasion in Lotus japonicus roots and nodules. New Phytologist 208: 241–256. [DOI] [PubMed] [Google Scholar]
  22. Íñiguez LP, Nova‐Franco B, Hernández G. 2015. Novel players in the AP2‐miR172 regulatory network for common bean nodulation. Plant Signaling and Behavior 10: e1062957. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jian B, Hou W, Wu C, Liu B, Liu W, Song S, Bi Y, Han T. 2009. Agrobacterium rhizogenes‐mediated transformation of Superroot‐derived Lotus corniculatus plants: a valuable tool for functional genomics. BMC Plant Biology 9: 78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kawaharada Y, James EK, Kelly S, Sandal N, Stougaard J. 2017. The ethylene responsive factor required for nodulation 1 (ERN1) transcription factor is required for infection‐thread formation in Lotus japonicus . Molecular Plant–Microbe Interactions 30: 194–204. [DOI] [PubMed] [Google Scholar]
  25. Kereszt A, Li D, Indrasumunar A, Nguyen CD, Nontachaiyapoom S, Kinkema M, Gresshoff PM. 2007. Agrobacterium rhizogenes‐mediated transformation of soybean to study root biology. Nature Protocols 2: 948–952. [DOI] [PubMed] [Google Scholar]
  26. Klein J, Saedler H, Huijser P. 1996. A new family of DNA binding proteins includes putative transcriptional regulators of the Antirrhinum majus floral meristem identity gene SQUAMOSA . Molecular and General Genetics 250: 7–16. [DOI] [PubMed] [Google Scholar]
  27. Kouchi H, Hata S. 1993. Isolation and characterization of novel nodulin cDNAs representing genes expressed at early stages of soybean nodule development. Molecular and General Genetics 238: 106–119. [DOI] [PubMed] [Google Scholar]
  28. Kumagai H, Kinoshita E, Ridge RW, Kouchi H. 2006. RNAi knock‐down of ENOD40s leads to significant suppression of nodule formation in Lotus japonicus . Plant and Cell Physiology 47: 1102–1111. [DOI] [PubMed] [Google Scholar]
  29. Liu CW, Breakspear A, Guan D, Cerri MR, Jackson K, Jiang S, Robson F, Radhakrishnan GV, Roy S, Bone C et al. 2019. NIN acts as a network hub controlling a growth module required for rhizobial infection. Plant Physiology 179: 1704–1722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Liu J, Rutten L, Limpens E, van der Molen T, van Velzen R, Chen R, Chen Y, Geurts R, Kohlen W, Kulikova O et al. 2019. A remote cis‐regulatory region is required for NIN expression in the pericycle to initiate nodule primordium formation in Medicago truncatula . Plant Cell 31: 68–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Liu M, Soyano T, Yano K, Hayashi M, Kawaguchi M. 2019. ERN1 and CYCLOPS coordinately activate NIN signaling to promote infection thread formation in Lotus japonicus . Journal of Plant Research 132: 641–653. [DOI] [PubMed] [Google Scholar]
  32. Li Y, Yu C, Cheng X, Li C, Sun J, Zhang F, Lambers H, Li L. 2009. Intercropping alleviates the inhibitory effect of N fertilization on nodulation and symbiotic N2 fixation of faba bean. Plant and Soil 323: 295–308. [Google Scholar]
  33. Llave C, Xie Z, Kasschau KD, Carrington JC. 2002. Cleavage of Scarecrow‐like mRNA targets directed by a class of Arabidopsis miRNA. Science 297: 2053–2056. [DOI] [PubMed] [Google Scholar]
  34. Long JA, Moan EI, Medford JI, Barton MK. 1996. A member of the KNOTTED class of homeodomain proteins encoded by the STM gene of Arabidopsis. Nature 379: 66. [DOI] [PubMed] [Google Scholar]
  35. Ma X, Zhang Q, Zhu Q, Liu W, Chen Y, Qiu R, Wang B, Yang Z, Li H, Lin Y et al. 2015. A robust CRISPR/Cas9 system for convenient, high‐efficiency multiplex genome editing in monocot and dicot plants. Molecular Plant 8: 1274–1284. [DOI] [PubMed] [Google Scholar]
  36. Minami E, Kouchi H, Cohn JR, Ogawa T, Stacey G. 1996. Expression of the early nodulin, ENOD40, in soybean roots in response to various lipo‐chitin signal molecules. The Plant Journal 10: 23–32. [DOI] [PubMed] [Google Scholar]
  37. Moling S, Pietraszewska‐Bogiel A, Postma M, Fedorova E, Hink MA, Limpens E, Gadella TW, Bisseling T. 2014. Nod factor receptors form heteromeric complexes and are essential for intracellular infection in Medicago nodules. Plant Cell 26: 4188–4199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Nicolas C, Monzon JP, Specht JE, Grassini P. 2017. Is soybean yield limited by nitrogen supply? Field Crops Research 213: 204–212. [Google Scholar]
  39. Nova‐Franco B, Íñiguez LP, Valdés‐López O, Alvarado‐Affantranger X, Leija A, Fuentes SI, Ramírez M, Paul S, Reyes JL, Girard L et al. 2015. The micro‐RNA172c‐APETALA2‐1 node as a key regulator of the common bean‐Rhizobium etli nitrogen fixation symbiosis. Plant Physiology 168: 273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Pueppke SG, Bolaños‐Vásquez MC, Werner D, Bec‐Ferté MP, Promé JC, Krishnan HB. 1998. Release of flavonoids by the soybean cultivars McCall and Peking and their perception as signals by the nitrogen‐fixing symbiont Sinorhizobium fredii . Plant Physiology 117: 599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Reynoso MA, Blanco FA, Bailey‐Serres J, Crespi M, Zanetti ME. 2013. Selective recruitment of mRNAs and miRNAs to polyribosomes in response to rhizobia infection in Medicago truncatula . The Plant Journal 73: 289–301. [DOI] [PubMed] [Google Scholar]
  42. Schauser L, Roussis A, Stiller J, Stougaard J. 1999. A plant regulator controlling development of symbiotic root nodules. Nature 402: 191. [DOI] [PubMed] [Google Scholar]
  43. Sousa C, Johansson C, Charon C, Manyani H, Sautter C, Kondorosi A, Crespi M. 2001. Translational and structural requirements of the early nodulin gene ENOD40, a short‐open reading frame‐containing RNA, for elicitation of a cell‐specific growth response in the alfalfa root cortex. Molecular Biology of the Cell 21: 354–366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Soyano T, Hirakawa H, Sato S, Hayashi M, Kawaguchi M. 2014. Nodule inception creates a long‐distance negative feedback loop involved in homeostatic regulation of nodule organ production. Proceedings of the National Academy of Sciences, USA 111: 14607–14612. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Soyano T, Kouchi H, Hirota A, Hayashi M. 2013. Nodule inception directly targets NF‐Y subunit genes to regulate essential processes of root nodule development in Lotus japonicus . PLoS Genetics 9: e1003352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Subramanian S, Fu Y, Sunkar R, Barbazuk WB, Zhu J, Yu O. 2008. Novel and nodulation‐regulated microRNAs in soybean roots. BMC Genomics 9: 160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Sun Z, Su C, Yun J, Jiang Q, Wang L, Wang Y, Cao D, Zhao F, Zhao Q, Zhang M et al. 2019. Genetic improvement of the shoot architecture and yield in soya bean plants via the manipulation of GmmiR156b . Plant Biotechnology Journal 17: 50–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Svistoonoff S, Sy MO, Diagne N, Barker DG, Bogusz D, Franche C. 2010. Infection‐specific activation of the Medicago truncatula ENOD11 early nodulin gene promoter during actinorhizal root nodulation. Molecular Plant–Microbe Interactions 23: 740–747. [DOI] [PubMed] [Google Scholar]
  49. Tang T, Yu X, Yang H, Gao Q, Ji H, Wang Y, Yan G, Peng Y, Luo H, Liu K et al. 2018. Development and validation of an effective CRISPR/Cas9 vector for efficiently isolating positive transformants and transgene‐free mutants in a wide range of plant species. Frontiers in Plant Science 9: 1533. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Treiber T, Treiber N, Meister G. 2019. Regulation of microRNA biogenesis and its crosstalk with other cellular pathways. Nature Reviews Molecular Cell Biology 20: 5–20. [DOI] [PubMed] [Google Scholar]
  51. Velzen R, Holmer R, Bu F, Rutten L, van Zeijl A, Liu W, Santuari L, Cao Q, Sharma T, Shen D et al. 2018. Comparative genomics of the nonlegume Parasponia reveals insights into evolution of nitrogen‐fixing rhizobium symbioses. Proceedings of the National Academy of Sciences, USA 115: E4700–E4709. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Vernie T, Kim J, Frances L, Ding Y, Sun J, Guan D, Niebel A, Gifford ML, de Carvalho‐Niebel F, Oldroyd GE. 2015. The NIN transcription factor coordinates diverse nodulation programs in different tissues of the Medicago truncatula root. Plant Cell 27: 3410–3424. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wan X, Hontelez J, Lillo A, Guarnerio C, van de Peut D, Fedorova E, Bisseling T, Franssen H. 2007. Medicago truncatula ENOD40‐1 and ENOD40‐2 are both involved in nodule initiation and bacteroid development. Journal of Experimental Botany 58: 2033–2041. [DOI] [PubMed] [Google Scholar]
  54. Wang H, Wang H. 2015. The miR156/SPL module, a regulatory hub and versatile toolbox, gears up crops for enhanced agronomic traits. Molecular Plant 8: 677–688. [DOI] [PubMed] [Google Scholar]
  55. Wang JW. 2014. Regulation of flowering time by the miR156‐mediated age pathway. Journal of Experimental Botany 65: 4723–4730. [DOI] [PubMed] [Google Scholar]
  56. Wang JW, Czech B, Weigel D. 2009. miR156‐regulated SPL transcription factors define an endogenous flowering pathway in Arabidopsis thaliana . Cell 138: 738–749. [DOI] [PubMed] [Google Scholar]
  57. Wang L, Sun Z, Su C, Wang Y, Yan Q, Chen J, Ott T, Li X. 2019. A GmNINa‐miR172c‐NNC1 regulatory network coordinates the nodulation and autoregulation of nodulation pathways in soybean. Molecular Plant 12: 1211–1226. [DOI] [PubMed] [Google Scholar]
  58. Wang Y, Li P, Cao X, Wang X, Zhang A, Li X. 2009. Identification and expression analysis of miRNAs from nitrogen‐fixing soybean nodules. Biochemical and Biophysical Research Communications 378: 799–803. [DOI] [PubMed] [Google Scholar]
  59. Wang Y, Wang L, Zou Y, Chen L, Cai Z, Zhang S, Zhao F, Tian Y, Jiang Q, Ferguson BJ. 2014. Soybean miR172c targets the repressive AP2 transcription factor NNC1 to activate ENOD40 expression and regulate nodule initiation. Plant Cell 26: 4782–4801. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Wang Y, Wang Z, Amyot L, Tian L, Xu Z, Gruber MY, Hannoufa A. 2015. Ectopic expression of miR156 represses nodulation and causes morphological and developmental changes in Lotus japonicus . Molecular Genetics and Genomics 290: 471–484. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Wu G, Park MY, Conway SR, Wang J, Weigel D, Poethig RS. 2009. The sequential action of miR156 and miR172 regulates developmental timing in Arabidopsis. Cell 138: 750–759. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Xie F, Murray JD, Kim J, Heckmann AB, Edwards A, Oldroyd GED, Downie JA. 2012. Legume pectate lyase required for root infection by rhizobia. Proceedings of the National Academy of Sciences, USA 109: 633–638. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Yan J, Gu Y, Jia X, Kang W, Pan S, Tang X, Chen X, Tang G. 2012. Effective small RNA destruction by the expression of a short tandem target mimic in Arabidopsis. Plant Cell 24: 415–427. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Yan Z, Hossain MS, Wang J, Valdes‐Lopez O, Liang Y, Libault M, Qiu L, Stacey G. 2013. miR172 regulates soybean nodulation. Molecular Plant–Microbe Interactions 26: 1371–1377. [DOI] [PubMed] [Google Scholar]
  65. Yang W, Katinakis P, Hendriks P, Smolders A, de Vries F, Spee J, van Kammen A, Bisseling T, Franssen H. 1993. Characterization of GmENOD40, a gene showing novel patterns of cell‐specific expression during soybean nodule development. The Plant Journal 3: 573–585. [DOI] [PubMed] [Google Scholar]
  66. Yoo CY, Pence HE, Jin JB, Miura K, Gosney MJ, Hasegawa PM, Mickelbart MV. 2010. The Arabidopsis GTL1 transcription factor regulates water use efficiency and drought tolerance by modulating stomatal density via transrepression of SDD1 . Plant Cell 22: 4128–4141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Yu Y, Jia T, Chen X. 2017. The ‘how’ and ‘where’ of plant microRNAs. New Phytologist 216: 1002–1017. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Fig. S1 Identification of the reverse primers and quantitative primers for miR156 family genes.

Fig. S2 The precursor sequence alignment of soybean miR156 family genes.

Fig. S3 The precursor sequence alignment of soybean miR156b and miR156f.

Fig. S4 Diagram of STTM156b/f‐88 structure showing the design strategy.

Fig. S5 Other members of the miR156 family are downregulated in STTM156‐88 hairy roots.

Fig. S6 Sequencing results of pre‐miR156b and pre‐miR156f in the roots of pre‐miR156b/f‐CRISPR (mir156b/f) mutants.

Fig. S7 GmSPL9b and GmSPL9d gene structure, GmSPL9d peptide sequence and subcellular localization of GmSPL9d protein.

Fig. S8 Sequencing results of GmSPL9d/CRISPR mutant roots.

Fig. S9 In situ hybridization of GmSPL9d, GmNINa and GmENOD40‐1 negative controls during nodulation.

Fig. S10 GmSPL9d binds to the GTAC motifs in the miR172c, GmENOD40‐1 and GmNINa promoters.

Table S1 The primers for reverse transcription and qPCR of miR156 family members.

Table S2 The primers used for qPCR analysis of miR156b target genes.

Table S3 The primers used for the vector constructions.

Table S4 The primers used for in situ hybridization.

Table S5 Probes for electrophoretic mobility shift assay experiments.

Table S6 The primers used for ChIP‐PCR assays.

Please note: Wiley Blackwell are not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

Click here for additional data file. (2.1MB, pdf)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from The New Phytologist are provided here courtesy of Wiley

RESOURCES