The Roles of the PSEUDO-RESPONSE REGULATORs in Circadian Clock and Flowering Time in Medicago truncatula

PSEUDO-RESPONSE REGULATORs (PRRs) play key roles in the circadian rhythms and flowering in plants. Here, we identified the four members of the PRR family in Medicago truncatula, including MtPRR9a, MtPRR9b, MtPRR7 and MtPRR5, and isolated their Tnt1 retrotransposon-tagged mutants. They were expressed in different organs and were nuclear-localized. The four MtPRRs genes played important roles in normal clock rhythmicity maintenance by negatively regulating the expression of MtGI and MtLHY. Surprisingly, the four MtPRRs functioned redundantly in regulating flowering time under long-day conditions, and the quadruple mutant flowered earlier. Moreover, MtPRR can recruit the MtTPL/MtTPR corepressors and the other MtPRRs to form heterodimers to constitute the core mechanism of the circadian oscillator.


Introduction
Plants have evolved an internal circadian biological clock system with a period of about 24-h to regulate important activities at the right time of day [1,2].Most studies on the plant circadian clock have been investigated in Arabidopsis thaliana and included many processes, such as flowering time regulation, hormone synthesis, signal transduction, stress response pathways, and interactions between the plant and pathogens [3][4][5][6].The traditional circadian clock consists of three major components: the central oscillator, clock input pathways, and clock output pathways.The central oscillator is made up of several highly interconnected transcriptional and post-transcriptional feedback loops [1,[7][8][9].The CIRCADIAN CLOCK ASSOCIATED1 (CCA1), LATE ELONGATED HYPOCOTYL (LHY), LUX ARRHYTHMO (LUX), EARLY FLOWERING 3 (ELF3), EARLY FLOWERING 4 (ELF4) and PSEUDO-RESPONSE REGULATOR (PRR) genes are major members of the central oscillator [10].CCA1 and LHY function redundantly to repress the expression of TIMING OF CAB EXPRESSION1 (TOC1), LUX, ELF3, and ELF4, but to promote the expression of PRR9 and PRR7, while PRR9 and PRR7 could also suppress the expression of CCA1 and LHY in a feedback loop in the morning [11][12][13][14].LUX, ELF3, and ELF4 form a trimeric complex named as the evening complex (EC), which represses the expression of GIGANTEA (GI), PRR9, and PRR7 in the evening [15].
The PRR proteins have an N-terminal pseudo-receiver (PR) domain and a C-terminal CTT (CONSTANS, CONSTANS-LIKE, TOC1) domain [16].They are highly conserved in the circadian rhythm in Arabidopsis thaliana [17,18].PRR9, PRR7, PRR5, PRR3, and TOC1/PRR1 are members of the Arabidopsis PRR family, and these genes are regulated by the circadian clock in a temporal order [17].Mutation in TOC1 or PRR5 results in a short period, while mutations in PRR9 or PRR7 lead to an extended period [19][20][21].Arrhythmia is only observed in the prr5 prr7 prr9 triple mutant in continuous light conditions, and the triple mutant also exhibits less sensitivity to photoperiodicity and photomorphogenic responses [22].ZEITLUPE (ZTL), which encodes an F-box protein, recognizes the PR domains of TOC1 and PRR5 to promote their protein degradation, and TOC1 could form heterodimers with PRR3 and PRR5 by the PR domains to influence the protein stability of TOC1 [23,24].These findings suggest that the PR domains are critical to protein-protein interactions.TOC1 could bind to the promoters of CCA1 and LHY to suppress the gene expression, while mutation or deletion of the CCT domain in TOC1 prevents this repression, and PRR5 also binds to the CCA1 promoter region through its CCT domain, indicating that the CCT domain of PRRs is necessary for DNA binding [16,25].
Flowering time regulation is extremely important in the regional climatic adaptation of elite germplasm, and floral transition determines the production of crops [26].Medicago truncatula, which functions as a model plant in legume forage, is a temperate species.The flowering time of M. truncatula is promoted by exposure to a prolonged period of cold conditions, so it is cultivated in a relatively limited geographic range [27,28].Exploring the genes involved in flowering time regulation will be helpful to understand the climate adaptation of M. truncatula, and some important genes have been reported so far.For example, FTa1, one of the FLOWERING LOCUS T (FT) orthologs in M. truncatula, is a key regulator of flowering time, and the loss of function of FTa1 exhibits late flowering, while overexpression of MtFTa1 accelerates flowering time [29].A legume-specific gene in Glycine max was named E1, and the mutation of the homolog of E1 causes a delay in flowering in M. truncatula [30].In addition, SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1 (SOC1), PHYTOCHROME A and FRUITFULL-LIKE genes promote the flowering in M. truncatula [31][32][33].But the functions of MtPRRs in flowering time regulation are still unknown.The PRR genes have been reported to regulate flowering time through the expression of FT in Arabidopsis [34].TOC1 was reported as a modulator of flowering time, and mutation of TOC1 exhibits early flowering, but toc1 enhances the late flowering phenotype of prr5, indicating a special function of TOC1 in flowering time control [35,36].PRR5 protein directly represses the expression of CDF (CYCLING DOF FACTOR) genes, which are involved in the clock output pathways and flowering time regulation [16].Furthermore, the mutant of prr7 shows late flowering time, while PRR7 overexpression plants exhibit early flowering, indicating PRR7 functions as a positive factor in floral transition [19,37].In addition, Photoperiod-H1, which belongs to the PRR family in barley (Hordeum vulgare), negatively controls the expression of HvFT, and the mutant is late flowering [38].On the contrary, in short-day (SD) crops, such as rice (Oryza sativa) and sorghum (Sorghum bicolor), PRR family genes OsPRR37 and SbPRR37 regulate late flowering [39,40].Interestingly, overexpressing the AtPRR5 in rice causes the late flowering phenotype [41].Furthermore, Time of Flowering 11 (Tof11) and Tof12, two homologous genes of AtPRR3 in Glycine max, act via GmLHY to promote the transcription of E1 and delay flowering time under long-day conditions, and also improve the adaptability of soybean to high latitudes during domestication [42][43][44].
In Medicago truncatula, a few clock genes have been reported, including the MtLHY gene, which orchestrates the endogenous circadian rhythm in nodules and positively regulates the salt stress response [45,46], MtLUX controls the expression of clock genes, nodule formation, leaf movement, and flowering time [47].Circadian clock genes form a complicated regulatory network, and the regulation mechanism remains unclear in M. truncatula.Here, we characterized the Tnt1 retrotransposon-tagged mtprr9a, 9b, 7, and 5 mutants in M. truncatula.We demonstrated that the four MtPRRs play vital roles in maintaining clock gene expression and regulating flowering.In addition, MtPRR could form complexes with MtTPL/MtTPR corepressors and another MtPRR, indicating the post-transcription regulation of the circadian clock in M. truncatula.

Identification of PRRs and Their Loss-of-Function Mutants in M. truncatula
To identify the putative orthologs of PRRs in M. truncatula, the protein sequences of PRRs in Arabidopsis thaliana were used as a query in BLAST searches against the protein sequence database of the M. truncatula genome in the National Center for Biotechnology and the Phytozome database.Based on the homology analysis, seven M. truncatula PRR proteins, namely MtPRR9a, MtPRR9b, MtPRR7, MtPRR5, MtPRR3, MtTOC1a, and MtTOC1b were isolated (Figure 1A).The phylogenetic analysis also showed that MtPRR9a and MtPRR9b were more closely related to MtPRR5 (PRR5/9 clade) and MtPRR3 was close to MtPRR7 (PRR3/7 clade), whereas MtTOC1a and MtTOC1b were grouped into the TOC1 clade and separated from the other members of MtPRRs (Figure 1A).Subsequently, the amino acid sequence alignments of the PRR putative orthologs were carried out, and the result showed two types of conserved domains, including the PR domain and the CCT domain (Supplementary Figure S1), which are characteristic of the PRR proteins.
proteins, namely MtPRR9a, MtPRR9b, MtPRR7, MtPRR5, MtPRR3, MtTOC1a, and MtTOC1b were isolated (Figure 1A).The phylogenetic analysis also showed that MtPRR9a and MtPRR9b were more closely related to MtPRR5 (PRR5/9 clade) and MtPRR3 was close to MtPRR7 (PRR3/7 clade), whereas MtTOC1a and MtTOC1b were grouped into the TOC1 clade and separated from the other members of MtPRRs (Figure 1A).Subsequently, the amino acid sequence alignments of the PRR putative orthologs were carried out, and the result showed two types of conserved domains, including the PR domain and the CCT domain (Supplementary Figure S1), which are characteristic of the PRR proteins.
To determine the functions of MtPRRs in M. truncatula, loss-of-function mutants of MtPRR9a, 9b, 7, and 5 were isolated by a reverse genetic screening in the Tnt1 retrotransposon-tagged mutant collection of M. truncatula, but the mutant of MtPRR3 failed to be isolated [48].We focused on MtPRR9a, 9b, 7, and 5 in the subsequent study.Sequence analysis showed that a single Tnt1 was inserted in the first exon of MtPRR9a and MtPRR9b in mtprr9a-1 and mtprr9b-1, the third exon of MtPRR7 in mtprr7-1, and the sixth exon of MtPRR5 in mtprr5-1, respectively (Figure 1B-E).Reverse transcription PCR (RT-PCR) data showed that the transcripts of MtPRR9a, 9b, 7, and 5 were not detected in the mtprr9a-1, mtprr9b-1, mtprr7-1, and mtprr5-1 mutants, indicating that the transcriptions of the four MtPRRs were disrupted by Tnt1 insertion (Figure 1F-I).To determine the functions of MtPRRs in M. truncatula, loss-of-function mutants of Mt-PRR9a, 9b, 7, and 5 were isolated by a reverse genetic screening in the Tnt1 retrotransposontagged mutant collection of M. truncatula, but the mutant of MtPRR3 failed to be isolated [48].We focused on MtPRR9a, 9b, 7, and 5 in the subsequent study.Sequence analysis showed that a single Tnt1 was inserted in the first exon of MtPRR9a and MtPRR9b in mtprr9a-1 and mtprr9b-1, the third exon of MtPRR7 in mtprr7-1, and the sixth exon of MtPRR5 in mtprr5-1, respectively (Figure 1B-E).Reverse transcription PCR (RT-PCR) data showed that the transcripts of MtPRR9a, 9b, 7, and 5 were not detected in the mtprr9a-1, mtprr9b-1, mtprr7-1, and mtprr5-1 mutants, indicating that the transcriptions of the four MtPRRs were disrupted by Tnt1 insertion (Figure 1F-I).

Subcellular Localizations of MtPRR9, 7, and 5 and Expression Patterns
To study the cellular localization of MtPRR9a, 9b, 7, and 5, the proteins of MtPRR9a, 9b, 7, and 5 were fused with green fluorescent protein (GFP) under the control of the Cauliflower mosaic virus (CaMV) 35S promoter and transformed into tobacco (Nicotiana benthamiana) leaf cells.Using fluorescence microscopy observation, the 35S:GFP was localized in both the cytoplasm and nucleus of epidermal cells, while the MtPRR9a, 9b, 7, and 5 fusion proteins were exclusively localized to the nucleus, which further supported their functions as transcription factors (Figure 2A).and exons are represented by boxes.(F-I) RT-PCR shows the transcript abundance of MtPRR9a, 9b, 7, and 5 in the leaf of wild type and relative mutants.MtACTIN was used as the control.

Subcellular Localizations of MtPRR9, 7, and 5 and Expression Patterns
To study the cellular localization of MtPRR9a, 9b, 7, and 5, the proteins of MtPRR9a, 9b, 7, and 5 were fused with green fluorescent protein (GFP) under the control of the Cauliflower mosaic virus (CaMV) 35S promoter and transformed into tobacco (Nicotiana benthamiana) leaf cells.Using fluorescence microscopy observation, the 35S:GFP was localized in both the cytoplasm and nucleus of epidermal cells, while the MtPRR9a, 9b, 7, and 5 fusion proteins were exclusively localized to the nucleus, which further supported their functions as transcription factors (Figure 2A).
To analyze the expression patterns of MtPRR9a, 9b, 7, and 5, quantitative real-time PCR (qRT-PCR) was used to examine the expression levels of the four MtPRRs in different organs, including root, stem, leaf, flower, pod, petiole, and vegetative bud.The results showed that MtPRR9a and MtPRR7 were highly expressed in leaf, petiole, and pod (Figure 2B,D), the expression levels of MtPRR9b were similar in all issues (Figure 2C), and MtPRR5 showed the highest expression level in the leaf (Figure 2E).In order to detect whether MtPRR9a, 9b, 7, and 5 are all subjected to a circadian rhythm at the level of transcription, we examined their expression patterns in plants grown in constant light conditions.Results showed that MtPRR9a, 9b, 7, and 5 transcripts varied considerably in an oscillatory manner during the given 24-h period (Figure 3A-D).Results also showed that both MtPRR9a and MtPRR9b had very similar expression patterns, and their expression levels reached their maximum after releasing into constant light at approximately 6 h (Figure 3A,B).The expression levels of MtPRR7 and MtPRR5 reached the peak after releasing into constant light at approximately 12 h, and the abundance of MtPRR5 was significantly decreased at 15 h, while MtPRR7 decreased slowly (Figure 3C,D).To analyze the expression patterns of MtPRR9a, 9b, 7, and 5, quantitative real-time PCR (qRT-PCR) was used to examine the expression levels of the four MtPRRs in different organs, including root, stem, leaf, flower, pod, petiole, and vegetative bud.The results showed that MtPRR9a and MtPRR7 were highly expressed in leaf, petiole, and pod (Figure 2B,D), the expression levels of MtPRR9b were similar in all issues (Figure 2C), and MtPRR5 showed the highest expression level in the leaf (Figure 2E).
In order to detect whether MtPRR9a, 9b, 7, and 5 are all subjected to a circadian rhythm at the level of transcription, we examined their expression patterns in plants grown in constant light conditions.Results showed that MtPRR9a, 9b, 7, and 5 transcripts varied considerably in an oscillatory manner during the given 24-h period (Figure 3A-D).Results also showed that both MtPRR9a and MtPRR9b had very similar expression patterns, and their expression levels reached their maximum after releasing into constant light at approximately 6 h (Figure 3A,B).The expression levels of MtPRR7 and MtPRR5 reached the peak after releasing into constant light at approximately 12 h, and the abundance of MtPRR5 was significantly decreased at 15 h, while MtPRR7 decreased slowly (Figure 3C,D).

The Expression Patterns of Genes Associated with the Circadian Clock Are Altered in mtprr Mutants
The expression patterns of MtPRR9a, 9b, 7, and 5 exhibited a diurnal and circadian rhythm, implying that MtPRRs may play important roles in maintaining normal clock rhythmicity.To confirm this hypothesis, two clock genes MtLHY and MtGI were analyzed in the wild type and mtprr single or multiple mutants under constant light conditions for two days (Figure 4).The results showed that the expression of MtGI maintained the robust rhythmic cycles in the mtprr single or multiple mutants compared with the wild type during the two days, and the expression level was higher in the mtprr9a-1 mtprr9b-1 mtprr7-1 mtprr5-1 mutant than that in the wild type at most time points (Figure 4A,B).In addition, the expression of MtLHY was increased throughout the daytime, particularly at the peak time in mtprr7-1 and mtprr9a-1 mtprr9b-1 mtprr7-1 mtprr5-1 mutants (Figure 4C,D).These results indicated that the expression levels of circadian clock-related genes are severely compromised in the mtprr mutants under constant light conditions, suggesting the potential roles of MtPRR9a, 9b, 7, and 5 in regulating the clock rhythmicity.

The Expression Pa erns of Genes Associated with the Circadian Clock Are Altered in mtprr Mutants
The expression pa erns of MtPRR9a, 9b, 7, and 5 exhibited a diurnal and circadian rhythm, implying that MtPRRs may play important roles in maintaining normal clock rhythmicity.To confirm this hypothesis, two clock genes MtLHY and MtGI were analyzed in the wild type and mtprr single or multiple mutants under constant light conditions for two days (Figure 4).The results showed that the expression of MtGI maintained the robust rhythmic cycles in the mtprr single or multiple mutants compared with the wild type during the two days, and the expression level was higher in the mtprr9a-1 mtprr9b-1 mtprr7-1 mtprr5-1 mutant than that in the wild type at most time points (Figure 4A,B).In addition, the expression of MtLHY was increased throughout the daytime, particularly at the peak time in mtprr7-1 and mtprr9a-1 mtprr9b-1 mtprr7-1 mtprr5-1 mutants (Figure 4C,D).These results indicated that the expression levels of circadian clock-related genes are severely compromised in the mtprr mutants under constant light conditions, suggesting the potential roles of MtPRR9a, 9b, 7, and 5 in regulating the clock rhythmicity.

MtPRR9, 7, and 5 Physically Interact with MtTPL/MtTPR Proteins
It has been shown that PRR9, PRR7, and PRR5 act as transcriptional repressors and interact with the TOPLESS/TOPLESS RELATED PROTEINs (TPL/TPRs) family to repress the transcription of CCA1 and LHY [49,50].The TPL/TPR proteins were also reported to function as corepressors in M. truncatula [51].To understand the potential regulatory mechanism of MtPRR9, 7, and 5, we performed yeast two-hybrid (Y2H) experiments to examine the interaction between MtPRRs and MtTPL/MtTPRs.Results showed that MtPRR9a, MtPRR9b, MtPRR7, and MtPRR5 could interact with MtTPL in yeast cells (Figure 6A-C).Furthermore, MtPRR9a was selected to detect the interactions between MtPRR and MtTPRs, and our results showed that MtPRR9a could interact with MtTPR1-MtTPR5 in the yeast (Supplementary Figure S2).These data suggest that MtPRR9, 7, and 5 may recruit the MtTPL/MtTPRs to form the complex for rhythmic regulation.

MtPRR9, 7, and 5 Can Form Heterodimers
It has been reported that TOC1 forms heterodimers with PRR3 and PRR5 in the regulation of TOC1 nuclear accumulation, phosphorylation, and protein stability [24,52,53].To test the possibility of the interactions among four MtPRR proteins in M. truncatula, Y2H experiments were carried out.Results showed that MtPRR9a could interact with MtPRR9b, MtPRR7, and MtPRR5 in yeast cells and that MtPRR5 could also interact with MtPRR9b and MtPRR7 (Figure 6D-F).These data suggest that MtPRR9, 7, and 5 proteins can form heterodimers in the yeast.These interactions and homodimerization between MtPRR9a and MtPRR9b, MtPRR7, MtPRR5 were verified in tobacco by bimolecular fluorescence complementation (BiFC) assays.Yellow fluorescence was observed when MtPRR9a was fused to the C-terminal of YFP (YC), and MtPRR9b, MtPRR7, or MtPRR5 was fused to the N-terminal of YFP (YN) (Supplementary Figure S3).Taken together, these data suggest that MtPRR9, 7, and 5 physically interact to potentially form complexes.

MtPRR9, 7, and 5 Physically Interact with MtTPL/MtTPR Proteins
It has been shown that PRR9, PRR7, and PRR5 act as transcriptional repressors and interact with the TOPLESS/TOPLESS RELATED PROTEINs (TPL/TPRs) family to repress the transcription of CCA1 and LHY [49,50].The TPL/TPR proteins were also reported to function as corepressors in M. truncatula [51].To understand the potential regulatory mechanism of MtPRR9, 7, and 5, we performed yeast two-hybrid (Y2H) experiments to examine the interaction between MtPRRs and MtTPL/MtTPRs.Results showed that MtPRR9a, MtPRR9b, MtPRR7, and MtPRR5 could interact with MtTPL in yeast cells (Figure 6A-C).Furthermore, MtPRR9a was selected to detect the interactions between MtPRR and MtTPRs, and our results showed that MtPRR9a could interact with MtTPR1-MtTPR5 in the yeast (Supplementary Figure S2).These data suggest that MtPRR9, 7, and 5 may recruit the MtTPL/MtTPRs to form the complex for rhythmic regulation.Yeast colonies that were able to grow on TDO (SD/-His/-Leu/-Trp) with X-α-Gal and displayed blue coloration confirmed the protein-protein interaction.

MtPRR9, 7, and 5 Can Form Heterodimers
It has been reported that TOC1 forms heterodimers with PRR3 and PRR5 in the regulation of TOC1 nuclear accumulation, phosphorylation, and protein stability [24,52,53].To test the possibility of the interactions among four MtPRR proteins in M. truncatula, Y2H experiments were carried out.Results showed that MtPRR9a could interact with

Discussion
The PRR gene family is a major component of the circadian clock, regulating various plant physiological processes, such as photomorphogenesis, maintenance of mitochondrial homeostasis, stress responses, and flowering time regulation [10].The studies of the clock system in plants mainly focused on Arabidopsis; however, the circadian clock is little explored in the legume species.Here, we identified the orthologs of PRR in the legume model species M. truncatula.Our study showed that MtPRRs can be grouped into the PRR5/9 clade, PRR3/7 clade, and TOC1 clade, which is highly conserved in many species, such as Arabidopsis, rice, Brassica rapa, and rose [18,54,55].According to the previous study, there are five members of clock PRR genes in Arabidopsis thaliana, Sorghum bicolour, and Oryza sativa, all including one TOC1/PRR1 member gene, two genes in the PRR5/9 clade, and two genes in the PRR3/7 clade [41,56].However, seven PRR genes were shown in the Medicago truncatula genome, including two genes in the TOC1/PRR1 clade, three members in the PRR5/9 clade, and two PRR3/7 members.This is mainly because PRRs were expanded in M. truncatula during evolution, and the duplication event tends to happen in the TOC1/PRR1 clade and PRR5/9 clade, which is also seen in the PRR family of Populus trichocarpa and Vitis vinifera [56].Like the PRR proteins in Arabidopsis, rice, sorghum, and other plants, the PRRs in M. truncatula are also nuclear-localized with an N-terminal PR domain followed by the C-terminal CCT motif, and the structure and location similarity imply their potential functional conservation in monocot and dicot species.
The MtPRR9a, 9b, 7, and 5 genes are all under circadian control and are expressed in a sequential wave that differs from the Arabidopsis PRRs [19], MtPRR9a, and MtPRR9b peak firstly followed by peaks in MtPRR7 and MtPRR5 expression.These results suggest that there may be a change in the PRR-associated circadian rhythm regulatory mechanism of Arabidopsis and M. truncatula.In the mtprr single or multiple mutants, the expressions of MtLHY and MtGI are upregulated, implying that they may act in a linear pathway.This result is consistent with the roles of PRR5, PRR7, and PRR9 binding to the promoters of CCA1 and LHY to repress their transcription in Arabidopsis [49].In addition, to fully understand the relationship between MtPRRs and MtLHY or MtGI, the CRISPR/Cas9 genome-editing system needs to be applied to generate the mtprr9a mtprr9b mtprr7 mtprr5 mtprr3 quintuple mutant in the future.
According to the previous study, PRRs also act by forming protein complexes [49,54,56,57].To verify the conservation of regulatory mechanism, protein interactions among MtPRRs were performed.MtPRR9a, 9b, 7, and 5 could form dimers with other members, indicating the MtPRRs may be included in the post-transcriptional regulation of the circadian clock, such as the phosphorylation, subnuclear localization, and protein stability of MtPRRs.In Arabidopsis, the PRR5, 7, and 9 proteins contain a conserved EAR (ethylene-responsive element binding factor-associated amphiphilic repression) motif, specifically interacting with TPL/TPR to repress the CCA1 and LHY expression [49]; similarly, MtPRR9a, 9b, 7, and 5 formed complexes with MtTPL/MtTPR.Combining the transcriptional regulatory relationship between MtPRR and MtGI or MtLHY in clock regulation, we put forward a plausible hypothesis that MtPRR9, 7, and 5 may recruit the MtTPL/MtTPRs to form complexes to suppress the expression of MtLHY and MtGI to regulate plant development (Figure 7).independent of the CO-FT module, and they may regulate flower by repressing the MtGI and MtLHY at a transcriptional level (Figure 7).Their orthologs are required for normal circadian rhythms and photoperiodic flowering in Arabidopsis, and LHY also represses the GI expression [63,64].In conclusion, studying the functions of the PRR genes will be helpful to understand the mechanism of circadian rhythm in Medicago truncatula, and provides useful information to improve the agricultural traits by the circadian clock genes.

Plant Materials and Growth Conditions
Medicago truncatula ecotype R108 was used as the wild-type accession for all the experiments described in this study.Mutants of mtprr9a-1, mtprr9b-1, mtprr7-1, and mtprr5-1 were identified from the Tnt1 retrotransposon-tagged mutant collection of M. truncatula [65].For flowering time measurement, the plants were grown in the greenhouse at 22 °C, with a long-day (16 h light and 8 h dark) photoperiod with a light intensity of 150 μmol/m −2/ s −1 and 70-80% relative humidity.For the circadian clock genes expression analyses, plants were grown at 22 °C in a light incubator with 70 and 80% relative humidity, and a light intensity of 90 μmol/m −2 /s −1 .Plants were grown under a long-day photoperiod for three weeks, and then transferred to a short-day photoperiod (12 h light and 12 h dark) for one week, followed by two days of constant light for sampling.For diurnal rhythmic analyses of MtPRRs, the wild type was grown in a short-day period for four weeks, and the leaves were collected for further analysis.Modern agriculture needs to precisely control the flowering time against the emerging patterns of climate change [58].Simultaneous disruption of MtPRR9a, 9b, 7, and 5 induced early flowering in M. truncatula, providing genetic evidence for the function of the circadian clock in PRRs flowering regulation.This result is consistent with the previously described role of PRRs as negative flowering time regulators in rice, sorghum, and soybean, but is inconsistent with the function of PRRs in Arabidopsis [19,34,39,41,43].Importantly, the flowering time in M. truncatula is advanced in the long-day and prolonged cold conditions, so the negative function of MtPRRs in flowering time regulation may be helpful to improve the adaptability to high latitudes and increase the yield of the leguminous forage.According to the reports above, the circadian clock PRRs may be positive regulators in flowering time regulation in long-day plants, and negative regulators in short-day plants.However, the PRRs negatively control the flowering time in the long-day species Medicago truncatula, indicating a special regulatory mechanism.In Arabidopsis, the GI-CONSTANS (CO)-FT module is the major photoperiod pathway for floral regulation, and PRRs can regulate the transcription of CO or stabilize the CO protein to enhance the transcription of FT in promoting flowering time [59].In rice, the photoperiod pathway is conserved, but the OsPRRs inhibit the expression of OsFT under long-day conditions [60]; similarly, SbPRR37 also represses the expression SbFT in sorghum [39].These results indicate that the clock PRRs function divergently in the long-day and short-day plants, and the flowering regulation appears to act in a CO-FT dependent manner.However, three CO-like genes, named MtCOLa, MtCOLb, and MtCOLc, are not involved in the photoperiodic flowering in M. truncatula [61].Furthermore, overexpression of the MtCDF (CYCLING DOF FACTOR) gene results in delayed flowering time, and the expression levels of the MtCOL genes are unchanged [62].Moreover, FTa1 plays a major role in flowering time regulation but is not involved in the photoperiodic induction of flowering [29].Based on these results, we assume that MtPRRs regulating flowering time in M. truncatula is independent of the CO-FT module, and they may regulate flower by repressing the MtGI and MtLHY at a transcriptional level (Figure 7).Their orthologs are required for normal circadian rhythms and photoperiodic flowering in Arabidopsis, and LHY also represses the GI expression [63,64].In conclusion, studying the functions of the PRR genes will be helpful to understand the mechanism of circadian rhythm in Medicago truncatula, and provides useful information to improve the agricultural traits by the circadian clock genes.

Plant Materials and Growth Conditions
Medicago truncatula ecotype R108 was used as the wild-type accession for all the experiments described in this study.Mutants of mtprr9a-1, mtprr9b-1, mtprr7-1, and mtprr5-1 were identified from the Tnt1 retrotransposon-tagged mutant collection of M. truncatula [65].For flowering time measurement, the plants were grown in the greenhouse at 22 • C, with a long-day (16 h light and 8 h dark) photoperiod with a light intensity of 150 µmol/m −2 /s −1 and 70-80% relative humidity.For the circadian clock genes expression analyses, plants were grown at 22 • C in a light incubator with 70 and 80% relative humidity, and a light intensity of 90 µmol/m −2 /s −1 .Plants were grown under a long-day photoperiod for three weeks, and then transferred to a short-day photoperiod (12 h light and 12 h dark) for one week, followed by two days of constant light for sampling.For diurnal rhythmic analyses of MtPRRs, the wild type was grown in a short-day period for four weeks, and the leaves were collected for further analysis.

Identification and Phylogenetic Analysis of MtPRRs
The sequences of the AtPRRs were obtained from the Arabidopsis Information Resource (TAIR) database (http://www.arabidopsis.org/,accessed on 1 March 2020).AtPRR proteins were used to perform a BLASTP search against the sequence database of the Medicago truncatula in Phytozome (https://phytozome-next.jgi.doe.gov/,accessed on 20 March 2020).
To study the phylogenetic relationships between PRRs in M. truncatula and Arabidopsis, seven identified MtPRR proteins in M. truncatula and five Arabidopsis PRR proteins were used to generate the phylogenetic tree.Multiple sequence alignments were executed using CLUSTALW online (http://www.genome.jp/tools-bin/clustalw,accessed on 27 March 2020).Then, the phylogenetic tree was generated with the neighbor-joining method and 1000 bootstrap replications using the MEGA7.1 program in the p-distance model.The scale bar was shown to indicate the genetic distance based on branch length.

Conserved Domains Analysis
The PRR proteins were aligned using Clustal X2, and the GeneDoc 2.7 software was used for homology shading [66,67].

RNA Extraction, RT-PCR, qRT-PCR, and Statistical Analysis
The leaves and the other tissues were collected for total RNA isolation using the Trizol-RT Reagent (Invitrogen).For RT-PCR analysis, RNA was extracted from vegetative buds of wild type and mutant lines.RNA extraction, cDNA synthesis, RT-PCR, and qRT-PCR analyses were performed as described previously [46].The primers used for RT-PCR and RT-qPCR analysis are listed in Supplementary Table S1.The t-test was used to compare the means of different populations.

Y2H and BiFC Assays
The Y2H assay was performed using the Matchmaker Gold System (Clontech, Shiga, Japan).To detect the interactions between MtTPL and MtPRRs, the CDS of MtTPL were cloned into the pGADT7 (AD) or pGBKT7 (BD) vector, and the CDS of MtPRRs were cloned into the pGBKT7 or pGADT7 vector, correspondingly.To detect the interactions between MtTPRs and MtPRR9a, the CDS of MtTPRs were cloned into pGADT7, while the MtPRR9a CDS was cloned into pGBKT7.To check the interactions among MtPRRs, the CDS of MtPRRs were cloned and introduced into pGADT7 or pGBKT7.Protein-protein interactions were performed as described previously according to the manufacturer's protocol [68].
For the BiFC assays, MtPRR9a was cloned to the pEarleyGate202-YC vector, while MtPRR9b, MtPRR7, and MtPRR5 were cloned into pEarleyGate201-YN using the Gateway system (Invitrogen).All the constructs were introduced into the Agrobacterium tumefaciens EHA105 strain.Various combinations of transformed cells were simultaneously infiltrated into tobacco leaves.After infiltration for 48 h, the yellow fluorescent protein (YFP) signals were observed under a confocal laser scanning microscope (Zeiss).

Conclusions
In this study, we identified the PRR genes in M. truncatula.The MtPRR9a, 9b, 7, and 5 gene expression profiles were characterized in different tissues, and the proteins were located in the nucleus.The diurnal expression of MtPRR9a, 9b, 7, and 5 also showed a clear circadian rhythm, suggesting that they are important components of the circadian clock.Further investigations showed that MtPRR9a, 9b, 7, and 5 function redundantly in flowering time.At last, MtPRR9a, 9b, 7, 5 could form heterodimers with MtTPL/MtTPRs or with each other.Further study is also needed to elucidate the functions of all the MtPRRs that are involved in the circadian rhythm and plant development.

Figure 1 .
Figure 1.Molecular characterization of MtPRR9a, 9b, 7, and 5 in M. truncatula.(A) Phylogenetic tree analysis of MtPRR putative orthologs in Arabidopsis and M. truncatula.The PRR5/9, PRR3/7, and TOC1 clade are shown.The scale bar indicates the genetic distance based on branch length.(B-E) Schematic representations of the gene structures of MtPRR9a, 9b, 7, 5 showing the Tnt1 insertion sites in relative mutants.The positions of the ATG start and TAA/TGA stop codons are shown.Vertical arrows mark the location of Tnt1 retrotransposons in mutants.Introns are represented by lines

Figure 1 .
Figure 1.Molecular characterization of MtPRR9a, 9b, 7, and 5 in M. truncatula.(A) Phylogenetic tree analysis of MtPRR putative orthologs in Arabidopsis and M. truncatula.The PRR5/9, PRR3/7, and TOC1 clade are shown.The scale bar indicates the genetic distance based on branch length.(B-E) Schematic representations of the gene structures of MtPRR9a, 9b, 7, 5 showing the Tnt1 insertion sites in relative mutants.The positions of the ATG start and TAA/TGA stop codons are shown.Vertical arrows mark the location of Tnt1 retrotransposons in mutants.Introns are represented by lines and exons are represented by boxes.(F-I) RT-PCR shows the transcript abundance of MtPRR9a, 9b, 7, and 5 in the leaf of wild type and relative mutants.MtACTIN was used as the control.

Figure 2 .
Figure 2. Subcellular localization and expression patterns of MtPRR9a, 9b, 7, 5 in M. truncatula.(A) The subcellular localization of MtPRR-GFP fusion proteins.Free GFP was used as the control.(B-E) The expression levels of MtPRR9a, 9b, 7, and 5 in different organs.MtUBIQUITIN was used as the internal control.Values are shown as means ± SD of three biological replicates.

Figure 2 .
Figure 2. Subcellular localization and expression patterns of MtPRR9a, 9b, 7, 5 in M. truncatula.(A) The subcellular localization of MtPRR-GFP fusion proteins.Free GFP was used as the control.(B-E) The expression levels of MtPRR9a, 9b, 7, and 5 in different organs.MtUBIQUITIN was used as the internal control.Values are shown as means ± SD of three biological replicates.

Figure 3 .
Figure 3. MtPRR9a (A), 9b (B), 7 (C), and 5 (D) transcript levels in the long-day conditions of the 4week-old wild type plant.MtUBIQUITIN was used as the internal control.Values are shown as means ± SD of three biological replicates.White bars at the bo om indicate periods of constant light, and ZT means the zeitgeber time.

Figure 3 .
Figure 3. MtPRR9a (A), 9b (B), 7 (C), and 5 (D) transcript levels in the long-day conditions of the 4-week-old wild type plant.MtUBIQUITIN was used as the internal control.Values are shown as means ± SD of three biological replicates.White bars at the bottom indicate periods of constant light, and ZT means the zeitgeber time.

Figure 4 .
Figure 4. Regulation of MtPRR9a, 9b, 7, 5 on the circadian expression of clock genes in continuous light.The transcriptional behaviors of MtGI (A,B) and MtLHY (C,D) in the leaves of 4-week-old wild type and mtprr single, double, triple, and quadruple mutants grown under constant light conditions.Values are shown as means ± SD of three biological replicates.White bars at the bo om indicate periods of constant light, and ZT means the zeitgeber time.

Figure 4 .
Figure 4. Regulation of MtPRR9a, 9b, 7, 5 on the circadian expression of clock genes in continuous light.The transcriptional behaviors of MtGI (A,B) and MtLHY (C,D) in the leaves of 4-week-old wild type and mtprr single, double, triple, and quadruple mutants grown under constant light conditions.Values are shown as means ± SD of three biological replicates.White bars at the bottom indicate periods of constant light, and ZT means the zeitgeber time.

Figure 4 .
Figure 4. Regulation of MtPRR9a, 9b, 7, 5 on the circadian expression of clock genes in continuous light.The transcriptional behaviors of MtGI (A,B) and MtLHY (C,D) in the leaves of 4-week-old wild type and mtprr single, double, triple, and quadruple mutants grown under constant light conditions.Values are shown as means ± SD of three biological replicates.White bars at the bo om indicate periods of constant light, and ZT means the zeitgeber time.

Figure 7 .
Figure 7. Model for the proposed roles of MtPRR9, 7, and 5 in M. truncatula.The complexes between MtPRR9, 7, and 5 and MtTPL/MtTPRs are responsible for the downregulation of MtLHY and MtGI.The MtPRR9-, 7-, and 5-associated circadian clock is involved in the flowering time regulation.

Figure 7 .
Figure 7. Model for the proposed roles of MtPRR9, 7, and 5 in M. truncatula.The complexes between MtPRR9, 7, and 5 and MtTPL/MtTPRs are responsible for the downregulation of MtLHY and MtGI.The MtPRR9-, 7-, and 5-associated circadian clock is involved in the flowering time regulation.