Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Genome-Wide Analysis of the MADS-Box Gene Family in Brachypodium distachyon

  • Bo Wei,

    Affiliation State Key Laboratory of Plant Cell and Chromosome Engineering, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing, PR China

  • Rong-Zhi Zhang,

    Affiliation National Key Facility for Crop Gene Resources and Genetic Improvement and Institute of Crop Sciences, Chinese Academy of Agricultural Sciences, Beijing, PR China

  • Juan-Juan Guo,

    Affiliation National Key Facility for Crop Gene Resources and Genetic Improvement and Institute of Crop Sciences, Chinese Academy of Agricultural Sciences, Beijing, PR China

  • Dan-Mei Liu,

    Affiliation National Key Facility for Crop Gene Resources and Genetic Improvement and Institute of Crop Sciences, Chinese Academy of Agricultural Sciences, Beijing, PR China

  • Ai-Li Li,

    Affiliation National Key Facility for Crop Gene Resources and Genetic Improvement and Institute of Crop Sciences, Chinese Academy of Agricultural Sciences, Beijing, PR China

  • Ren-Chun Fan,

    Affiliation State Key Laboratory of Plant Cell and Chromosome Engineering, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing, PR China

  • Long Mao , (LM); (XZ)

    Affiliation National Key Facility for Crop Gene Resources and Genetic Improvement and Institute of Crop Sciences, Chinese Academy of Agricultural Sciences, Beijing, PR China

  • Xiang-Qi Zhang (LM); (XZ)

    Affiliation State Key Laboratory of Plant Cell and Chromosome Engineering, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing, PR China


MADS-box genes are important transcription factors for plant development, especially floral organogenesis. Brachypodium distachyon is a model for biofuel plants and temperate grasses such as wheat and barley, but a comprehensive analysis of MADS-box family proteins in Brachypodium is still missing. We report here a genome-wide analysis of the MADS-box gene family in Brachypodium distachyon. We identified 57 MADS-box genes and classified them into 32 MIKCc-type, 7 MIKC*-type, 9 Mα, 7 Mβ and 2 Mγ MADS-box genes according to their phylogenetic relationships to the Arabidopsis and rice MADS-box genes. Detailed gene structure and motif distribution were then studied. Investigation of their chromosomal localizations revealed that Brachypodium MADS-box genes distributed evenly across five chromosomes. In addition, five pairs of type II MADS-box genes were found on synteny blocks derived from whole genome duplication blocks. We then performed a systematic expression analysis of Brachypodium MADS-box genes in various tissues, particular floral organs. Further detection under salt, drought, and low-temperature conditions showed that some MADS-box genes may also be involved in abiotic stress responses, including type I genes. Comparative studies of MADS-box genes among Brachypodium, rice and Arabidopsis showed that Brachypodium had fewer gene duplication events. Taken together, this work provides useful data for further functional studies of MADS-box genes in Brachypodium distachyon.


MADS-box genes are important transcription factors for plant development, especially floral organ identities [1][2]. According to their roles in flower development, MADS-box genes are classified as having A, B, C, D and E functions [3][4], or are the basis for the so-called “ABCDE model” that summarizes the interactive functions of MADS-box proteins during this process [3], [5][7]. Generally, A- and C-lineage genes are respectively involved in sepal and carpel development. A- and B-lineage genes together are required for petal development, whereas B- and C-lineage genes are both needed for stamen development. D-lineage genes function in ovule development [5], [8][11], while E-lineage proteins are required for the development of all floral organs by forming MADS-box protein complexes with proteins of other lineages [11][12]. Evolutionarily, plant MADS-box genes are divided into type I and type II. Type II comprises most well-known MADS-box genes and can be further classified into MIKCc- and MIKC*-types due to differences in gene structures [13]. In general, type II proteins are composed of the most conserved MADS (M) domain for DNA binding, the keratin (K) domain for protein-protein interaction, the intervening (I) domain located between the M and K domain, and the C-terminus (C) domain that is mainly responsible for transcription activation [14]. On the other hand, MIKC*-type proteins constitute a subgroup of type II MADS-box proteins with longer I domains and less-conserved K domains when compared with MIKCc-type proteins [15][16]. Unlike type II MADS-box proteins, the structure of type I proteins is simpler which lacks the K domain. In fact, type I MADS-box genes are shared by plants and animals and thus represent a class of more ancient MADS-box genes [1], [12]. These genes are further divided into Mα, Mβ and Mγ subgroups [17][18]. In most plants, type I MADS-box genes experienced a faster pace of birth-and-death process than type II genes due to higher frequency of segmental gene duplications and weaker purifying selection [19].

Genome-wide analyses of MADS-box genes have been reported for Arabidopsis, poplar, rice, apple, cucumber, and soybean [18], [20][24]. Extensive work shows that, in addition to their functions in floral organ development, MADS-box genes may also be involved in abiotic responses. In wheat, for instance, TaMADS2 was up-regulated in response to the infection of stripe rust fungus [25], while in rice, OsMADS26, an AGL12 class gene, was also reported to be involved in stress tolerances [26]. Recently, flowering related MADS-box genes, such as rice SHORT VEGETATIVE PHASE (SVP)-like genes OsMADS22 and OsMADS55 and Arabidopsis SUPPRESSOR OF CONSTANS 1 (SOC1) are found to be involved in stress tolerances [26][28], revealing novel roles for MADS-box genes.

Brachypodium distachyon, a member of the Pooideae subfamily, is a wild annual grass endemic to the Mediterranean and Middle East and belongs to the “core pooid” genera that boast the majority of important temperate cereals and forage grasses such as wheat and switchgrass [29]. It is considered a model plant with amenities of short life cycle, undemanding growing condition, and highly efficient genetic transformation system. The available of its whole genome sequence makes it a promising model for functional genomic studies of biofuel plants and temperate grasses [29], [30]. Despite these, a genome-wide investigation of Brachypodium MADS-box genes is still lacking. Recently, we studied the functional divergence between two duplicated D-lineage MADS-box genes, BdMADS2 and BdMADS4 in Brachypodium [31]. Here we report a more systematic study of this gene family. A total of 57 MADS-box genes were identified. A detailed phylogenetic, gene structure and conservation domain analyses were performed. In addition, we also studied the expression patterns of Brachypodium MADS-box genes under normal and abiotic stress conditions. We found that Brachypodium type I MADS-box genes are not only expressed in floral organs, but respond to abiotic stresses. Our work provides useful information on the functions of this important family of transcription factors in Brachypodium distachyon.

Materials and Methods

Plant materials

The Brachypodium distachyon accession BD21-3 was kindly provided by Dr. David Garvin, USDA. Plants were grown as previously described [25]. Mature root, stem, and leaf were harvested from the same plants. Flower tissues were collected at Brachypodium flowering time. Immature seeds were collected when the glumes were half-filled. The dissected organs were frozen immediately in liquid nitrogen and stored at -70°C till RNA extraction. For stress treatments, seeds were germinated in Hoagland solution in a petri dish on top of a Whatman 3-mm paper and vernalized at 4°C for 1–2 weeks. Plants were then transferred to a light chamber (22°C for 16 h light/18°C for 8 h dark) until emergence of the fourth leaf. For salt stress, seedlings were transferred to a fresh petri dish containing 200 mM NaCl. For cold stress, seedlings were cultured in a chamber at 4°C. To mimic drought stress, seedlings were grown in a fresh petri dish containing 18% PEG 6000. Plants were then harvested at 0, 0.5 and 1 h after abiotic treatments mentioned above. All experiments were repeated three times.

Sequence collection and chromosome mapping

Brachypodium MADS-box genes were obtained by BlastP search using rice MADS-box proteins as queries against the Brachypodium 21-3 8X release dataset ( ClustalX software was used to perform the multiple sequence alignment and remove redundancies [26]. Eventually, 57 MADS-box genes were identified and were named BdMADS1-57, including BdMADS2 and BdMADS4 that have been reported previously [27]. Each MADS-box gene was mapped to the Brachypodium genome according to their coordinates on the Brachypodium genome. Editseq (DNASTAR Lasergene 7.1) was employed to predict the molecular weight and isoelectric point (IP) of each protein [28].

Phylogenetic analysis

Brachypodium MADS-box proteins were aligned using ClustalX with those of rice and Arabidopsis. An un-rooted neighbor-joining (NJ) tree was constructed using the MEGA5 package [29]. The tree nodes were evaluated by bootstrap analysis for 100 replicates [30]. Braches with less than 50% bootstrap values were collapsed. Information of all MADS-box genes including their accessions were listed in Table S2 in File S1.

Gene structure and conserved motif analysis

Gene structures were obtained by comparisons of open reading frames (ORFs) and genomic sequences, and displayed using a gene structure display server [31]. MEME program was used to predict conserved motifs with the following parameters: number of repetitions - any, maximum number of motifs - 20, optimum motif width set to ≥ 6 and ≤ 200 [32]. The identified motifs were annotated using SMART (Simple Modular Architecture Research Tool) protein analyzing software [33]-[34].

Calculation of Ka/Ks values and divergence times estimation

Protein sequences and ORFs of the gene pairs were aligned by DNASTAR MegAlign software, respectively. The synonymous substitution (Ks) and non-synonymous substitution (Ka) rates were calculated using the PAL2NAL web server ( [35], which used the CODEML program of PAML [36]. The divergence times (T) of the gene pairs were estimated using the formula: T  =  Ks/2λ [37], with the divergence rate λ  =  6.5×10– 9 [38].

Quantitative RT-PCR

Total RNA was prepared with TRIZOL reagent following the manufacturer’s instructions (Invitrogen, CA). Reverse transcription was performed using 2.5 µg total RNA treated with RNase-free DNase I and used for first strand cDNA synthesis in a 20 µl reaction containing 10 mM DTT, 0.5 µM dNTP, 40 U RNA inhibitor, 200 U M-MLV reverse transcriptase (Promega) and 0.5 µM oligonucleotide T15. The reaction was performed at 42°C for 60 min with 5 min denaturation at 90°C. For gene expression quantification, specific primers were designed for each MADS-box gene (Table S3 in File S1). PCR was carried out using a Veriti 96-well Thermal Cycler (Applied Biosystems) with following programs: initial denaturation at 95°C for 3 min; 30 cycles at 95°C for 15 s, 55°C for 15 s, 72°C for 30 s, and a final extension at 72°C for 10 min or adjusted accordingly to give the best results. The expression level of BdACT7 was used as loading control [39]. RT-PCR products were sequenced to ensure that they were derived from the desired target genes. Three independent biological replicates were performed.


Identification and phylogenetic analysis of Brachypodium MADS-box genes

Using both type I and type II rice MADS-box domain sequences iteratively as queries to search the Brachypodium protein sequence dataset, a total of 57 non-redundant MADS-box proteins were identified and serially named as BdMADS1 through BdMADS57, including BdMADS2 and BdMADS4 that have been reported before [27] (Table S1 and S2 in File S1). To determine the evolutionary relationship between Brachypodium MADS-box proteins and known MADS-box proteins from other species, we performed multiple sequence alignment and generated a Neighbor-Joining phylogenetic tree for MADS-box proteins from Brachypodium, Arabidopsis and rice. They were classified into functional groups according to Arabidopsis and rice MADS-box genes that have been extensively studied (Fig. 1) [40]. In total, 39 genes were identified as type II MADS-box genes including 32 MIKCc-type and 7 MIKC*-type. MIKCc-type MADS-box genes were divided into 9 classic clades, each of them comprised of close paralogs from rice, Arabidopsis, and Brachypodium (Fig. 1). Interestingly, OsMADS32 and BdMADS19 represented a novel monocot specific clade [41]. Two Brachypodium MADS-box genes BdMADS41 and BdMADS43 belong to the MIKC*-type gene group and were grouped with Arabidopsis AGL30, 65, 66, 94 and 104 and rice OsMADS62, 63 and 68. The second MIKC* clade consisted BdMADS22, 35, 39, 40 and 42, together with rice OsMADS37, 50 and 65, with no Arabidopsis members, indicating that this group is monocot specific (Fig. 1). No Brachypodium and rice MADS-box gene fell in the FLC-clade which appeared to be Arabidopsis specific. On the other hand, a total of 18 MADS-box genes were identified as type I, which were further classified into Mα, Mβ and Mγ that contained 9, 7 and 2 members respectively (Fig. 1 and Table S1 in File S1). Among them, Brachypodium BdMADS54 and 55 and rice OsMADS90 and 91 formed a monocot-specific Mβ group.

Figure 1. Phylogenetic analysis of MADS-box proteins in Brachypodium, rice and Arabidopsis.

A total of 57 MADS-box proteins in Brachypodium, 75 in rice and 98 in Arabidopsis were used to construct the NJ tree. The MADS-box proteins in Brachypodium were marked by solid triangles. Branches with less than 50% bootstrap support were collapsed.

Gene structure and conserved motif distribution analysis

To better understand the structural diversity of MADS-box genes, intron/exon arrangements and conserved motifs were compared according to their phylogenetic relationships. We obtained each gene structure by comparing their ORFs with their genomic sequences. As shown in Fig. 2, closely related genes were generally more similar in gene structures, differing only in intron and exon lengths. But some close gene pairs were indeed distinct in intron/exon arrangements. For example, BdMADS56 consisted 12 exons, whereas its close paralogues BdMADS9 and 15 had only one and two respectively, despite that their phylogenetic relationship was supported by a 99% bootstrap value. We then used the MEME program to analyze conserved motifs in MADS-box proteins which were then subject to SMART annotation. A total of 20 conserved motifs were identified (Fig. 3 and Table S4 in File S1). Motif 1 was represented by the typical MADS-box domain of ∼57 amino acids (aa). All of type II proteins, Mα and Mγ proteins contained motif 1. The Mβ MADS-box proteins BdMADS54 and 55 harbored motif 6 which contained MADS-box domain with longer amino acids, 129 aa. In this study, motifs 2 and 3 were two fragments of the K domain, which were the second most conserved domain and essential for protein-protein interactions among MADS-box transcriptional factors [42][44]. Nearly all type II proteins (except for BdMDS19 and 37) contained motifs 2 and/or 3, and except for BdMADS38, all contained motif 4 or 20 that correspond to the two fragments of the I domain. For MIKC*-type proteins, BdMADS22, 39 and 40 harbored motif 4, whereas BdMADS35, 41, 42 and 43 did not. Motif 8, a coiled coil motif, was present in two recently duplicated proteins, BdMADS8 and 21. In contrast, only two type I proteins (BdMADS44 and 50) contained motif 4. Unknown motifs (motifs 5, 7, 9–19) were largely located in C-terminals, the most diverse domain of MADS-box proteins [14] and appeared mainly in recently duplicated genes. For example, nearly all Mα group proteins contained motif 5 (except for BdMADS47), suggesting that these proteins may be derived from a common ancestor.

Figure 2. Phylogenetic relationship and gene structure analysis of MADS-box genes in Brachypodium.

Un-rooted neighbor-joining tree was constructed from the alignment of full-length amino acid sequences using the MEGA5 package. Branches with less than 50% bootstrap values were collapsed. Lengths of exons and introns of each MADS-box gene were displayed proportionally. The green solid boxes represent exons; black lines represent introns.

Figure 3. Conserved motif analysis of Brachypodium MADS-box proteins according to the phylogenetic relationship.

Each motif is represented by a number in a colored box. Motif 1 is the MADS-box domain; 2 and 3 are two different components of the K domain; 4 and 20 are I domains; 6 is a 129 amino acid MADS-box domain; 8 is a structure; 5, 7 and 9–19 are unidentified regions. Box length corresponds to motif length. Specific lengths, locations and p-values of each motif can be found in Table S4 in File S1.

Genomic distribution of MADS-box genes in Brachypodium

To analyze genomic locations of MADS-box genes and study their evolution in the context of whole genome duplication, we mapped each gene on the Brachypodium genome according to their mapping coordinates. The distribution of 57 MADS-box genes on the Brachypodium genome appeared to be random and proportional chromosome lengths (Fig. 4). There were 20, 12, 14, 7 and 4 MADS-box genes on chromosomes 1 to 5 respectively (Fig. 4). Of these, MIKCc-type genes (32) were scattered across all five chromosomes, while the seven MIKC*-type genes were observed on chromosomes 2, 3 and 4. For type I genes, Mα group genes were found on chromosomes 1, 2, 3 and 4, whereas Mβ group genes were localized on chromosomes 1, 2 and 3. Two Mγ group genes were located on chromosomes 1 and 5 (Fig. 4). According to the genome duplication information and phylogenetic relationship, five pairs of type II genes were found to be located on segmental or tandem duplicated genome blocks (Fig. 4).

Figure 4. Genomic locations of MADS-box genes and duplicated gene pairs in the Brachypodium genome.

Gene pairs located in the segmental duplicated chromosome region are linked using red lines.

Expression patterns of Brachypodium MADS-box genes

To analyze expression patterns of MADS-box genes in Brachypodium tissues, we harvested tissues from root, stem, leaf, and floral organs including those of lodicule, lemma, palea, stamen, carpel, and young seed. Specific primer sets were designed for each MADS-box gene and their expression patterns were detected by semi-quantitative RT-PCR (Fig. 5 and Table S3 in File S1). In general, paralogous genes of the same clade conferred similar expression patterns, although some close gene pairs did display distinct expression profiles, suggesting possible functional divergences.

Figure 5. Expression patterns of Brachypodium MADS-box genes in vegetative and reproductive organs.

Sources of the samples are as follow: root (R), stem (S), leaf (L), lodicule (Lo), lemma (Le), palea (Pa), stamen (St), carpel (Ca) and young seed (YS).

A-clade MADS-box genes. There were four genes in the A function clade. Three of them, BdMADS3, 10 and 33, were mainly expressed in reproductive organs such as lodicule, lemma, palea, stamen, and young seed. The forth one, BdMADS31, was only weakly expressed in leaf, a pattern similar to those of their orthologues in rice and Arabidopsis [18], . In addition, the weak expression signal of BdMADS33 (but not BdMADS10) in young seed may indicate functional divergence of these two genes after their arise by the last genome duplication (Fig. 1, 4 and 5).

B-clade MADS-box genes. There were six genes in this clade. BdMADS5 contained AP3 domain and was mainly detectable in lodicule and stamen, similar to Arabidopsis AP3 and rice OsMADS16. Two PISTILLATA (PI) homologues, BdMADS16 and 20, were strongly expressed in lodicule and stamen and weakly expressed in palea, with BdMADS16 additionally expressing in carpel and young seed, indicating that BdMADS16 may also be involved in the fourth whorl development. In other words, BdMADS16 may have expanded functions in all four floral whorls. Two other B-clade genes BdMADS17 and 23 had low transcript levels in palea and young seed, whereas BdMADS38 was weakly detected only in stamen. The diversified expression patterns for the B-clade MADS-box genes suggest diverged functions among these closely related genes (Fig. 5).

C- and D-clade MADS-box genes. This clade has four genes, with BdMADS14 and 18 fallen into the C-lineage, whereas BdMADS2 and 4 were of the D-lineage. BdMADS18 was clearly expressed in stamen, carpel, young seed, lodicule, and palea, whereas BdMADS14 was expressed preferentially in stamen and young seed. Similar expression patterns were observed for their homologs in rice [45]. For the two D-lineage MADS-box genes, as reported before [22], expression of BdMADS2 in the carpel was much stronger than that of BdMADS4, indicating subfunctionalization between the two genes.

E-clade MADS-box genes. The E clade contains six SEP-like genes, including BdMADS1, 7, 11, 26, 28 and 32. Expressions of these were detectable in reproductive organs, however, each had a unique expression pattern. Particularly, BdMADS11 and BdMADS26 were expressed in all of tested reproductive tissues. In comparison, BdMADS7 was observed in lemma, palea, stamen, and carpel. Compared to BdMADS26, its close paralogue BdMADS32 was absent from the first whorl (lemma and palea), indicating functional divergence between these two genes. In addition, BdMADS28 was weakly expressed in stamen and lodicule.

Other MIKCc-type MADS-box genes. Of the three ANR1-like clade genes, BdMADS29 and 37 were specifically expressed in root, similar to their Arabidopsis homolog ANR1 (AGL44) [46]. For SVP-clade MADS-box genes, BdMADS6 and BdMADS12 were preferentially expressed in root, stem, and leaf. A third SVP homolog BdMADS30 was detectable not only in vegetative tissues but also in lodicule, stamen, carpel, and young seed. Both SOC1-clade MADS-box genes BdMADS13 and 25 were observed to be expressed in all tested tissues. Interestingly, genes in the AGL12-clade displayed most diversified expression patterns. BdMADS34 was detected in stamen and young seed, whereas BdMADS24 was expressed specifically in root. BdMADS36 was expressed not only in root and leaf, but also in lodicule and lemma. BdMADS19 expression was restricted to palea and stamen, suggesting a role for development in these tissues.

MIKC*-type MADS-box genes. The five MIKC*-type genes were divergent in expression profiles. Of them, BdMADS22 were expressed weakly in palea and young seed, whereas transcripts of BdMADS39 and 40 were observed in nearly all organs assayed, except for root and palea (Fig. 5). Additionally, BdMADS42 was expressed only in leaf, lodicule, lemma, and palea, while BdMADS43 expression was observed in young seed specifically (Fig. 5).

Type I MADS-box genes. There are 18 type I MADS-box genes. Nine of them belong to the Mα group and only two BdMADS44 and BdMADS46 were found to be expressed, with BdMADS44 expressing in palea, stamen and young seed and BdMADS46 in young seed only (Fig. 5). In contrast, all seven genes of the Mβ group (BdMADS8, 9, 15, 21, 54, 55 and 57) were expressed in young seed (Fig. 5). Additionally, BdMADS55 and 57 also were observed in palea and stamen, whereas BdMADS9 and 15 were expressed only in palea and BdMADS54 was expressed only in stamen (Fig. 5). Two genes BdMADS8 and 21 were a pair of recently duplicated genes, but with slightly different patterns. They were expressed with comparable expression patterns in root, lemma, palea, stamen, and young seed, but BdMADS21 also expressed weakly in leaf and lodicule (Fig. 5). The Mγ group contained two genes. BdMADS53 was strongly and preferentially expressed in root, stem, leaf, lodicule, lemma and palea, while BdMADS56 was hardly detectable in any tissues.

In total, 12 out of the 57 Brachypodium MADS-box genes were found not expressed in any of the organs detected. The type I Mα group has the fewest genes (7 out of 9) expressed while the remaining groups have one or two (BdMADS1 and 27 in MIKCc-type, BdMADS35 and BdMADS41 in MIKC*-type, and BdMADS56 in Mγ).

Expression analysis MADS-box genes under abiotic stress conditions

The accumulative reports on MADS-box genes involving stress tolerance triggered us to investigate the responses of MADS-box genes to abiotic stresses in Brachypodium. As shown in Figure 6, we applied seedlings to NaCl, PEG 6000, and low-temperature treatment to mimic salt, drought, and cold stresses, respectively. The results showed that 12 MADS-box genes (BdMADS8, 9, 12, 2123, 28, 31, 33, 44, 54 and 55) were up-regulated when seedlings were treated with 200 mM NaCl (Fig. 6) and only one gene BdMADS30 was down-regulated. Similarly, 15 genes (BdMADS4, 5, 8, 11, 12, 17, 23, 24, 33, 34, 43, 50, 54, 55 and 57) were up-regulated, and two BdMADS30 and BdMADS39 were down-regulated under drought stress by PEG 6000, indicating that MADS-box genes are active in response to stress conditions. The expression pattern for the cold treatment was clear but less intensive. Four genes (BdMADS23, 33, 55 and 57) were up-regulated, whereas three BdMADS10, 13 and 30 were down-regulated. BdMADS30, a SVP-clade gene, was also down-regulated significantly by salt and drought (Fig. 6). Thus, as suggested before, MADS-box genes in Brachypodium may also be involved in stress response as an escaping strategy [28].

Figure 6. Differential expressions of MADS-box genes in response to the salt, drought, and cold stress.

The Mα, Mβ and MIKC*-type genes were highlighted with green, blue and red background respectively.

Duplication and functional divergence of MADS-box gene pairs in Brachypodium

Gene duplication events contributed significantly to the proliferation of MADS-box genes in the plant kingdom [1], [47]. Duplicated genes often evolved to partition existing functions (sub-functionalization) or obtain new functions (neo-functionalization), enhancing plants’ adaptability [48][49]. Using a divergence rate of 6.5×10-9 mutations per synonymous site per year [38], we calculated the divergence time between 18 closely related gene pairs as shown by the phylogenetic tree (Fig. 1). Table 1 shows that the divergence periods for most gene pairs were at ∼70 MYA (Million Years Ago), with a standard deviation of 20 MYA, indicating that most gene duplication events occurred before the divergence of grass species [50]. Two pairs of genes, BdMADS39/40 and BdMADS46/48, were estimated to divergence at about 10 and 17 MYA and may represent two newly duplicated gene pairs, long after the Brachypodium diverged from other grass species. On the contrary, the divergence times of five pairs of genes (BdMADS14/18, 27/29, 24/36, 23/38 and 41/43) well surpassed the divergence time of grass species (56–73 MYA) [51][52] and hence it is difficult to infer their actual divergence time.

Table 1. Estimated divergence period of MADS-box gene pairs in Brachypodium.

To study the selection pressures among duplicated MADS-box genes, the substitution ratio of non-synonymous (Ka) to synonymous (Ks) mutations (Ka/Ks) were calculated for the 18 gene pairs. Ka/Ks values of all the gene pairs were less than 1, suggesting that these duplicated MADS-box gene pairs evolved under purifying selection in Brachypodium. In spite of this, some closely related gene pairs displayed different expression patterns indicating that have experienced subtle functional divergences. For example, among three SVP-like genes, BdMADS6 and 12 were expressed dominantly in vegetative tissues, whereas BdMADS30 was expressed in reproductive tissues, particularly in stamen, and their expression patterns were very similar to their rice homologs (Fig. 5) [53][54]. In addition, BdMADS30 was down-regulated significantly by salt, drought and low-temperature stresses, however, expression of BdMADS12 was up-regulated by salt and drought, and BdMADS6 did not respond to abiotic stress (Fig. 6). Similarly, as two AP1-like genes, BdMADS10, but not BdMADS33, was down-regulated by cold treatment. Additionally, BdMADS26 and 32 are close SEP-like paralogues, and their expression signals were consistently distributed in lodicule, stamen, carpel and young seed uniformly, while BdMADS26 was also expressed in palea (Fig. 5). Expression patterns of other duplicated genes, such as BdMADS7/11, BdMADS14/18, BdMADS46/48, BdMADS54/55 and BdMADS44/47 were also diverged significantly. These data indicate that Brachypodium MADS-box genes are in the process of divergence under the purifying pressure by natural selection.


The slow birth and death rate for Brachypodium MADS-box genes

Extensive studies have shown that MADS-box gene families expand by gene duplication events from whole genome duplication [52], [55][56]. In Brachypodium, there were 32 MIKCc-type genes, similar to those in rice (38), Arabidopsis (39), cucumber (29), grape (38), and poplar (55) (Table 2). Arabidopsis boasts the unique FLC clade, for which there was no counterpart in Brachypodium and rice. In Arabidopsis and Brassicaceae, FLC was identified to control flowering time and vernalization responses [57][59]. In temperate grass plants, VRN1 (BdMADS33) was a homolog of Arabidopsis AP1 which was up-regulated by low-temperature. However, VRN1 responds to vernalization through a different pathway from those in Arabidopsis and Brassicaceae [60][61]. The fact that salt and drought stresses induced BdMADS33 expression may suggest that VRN1-like genes be an important factor for abiotic stress response as well, which deserves further investigation. For Brachypodium type I MADS-box genes, we found that Brachypodium had a comparable number of Mα genes, but significantly fewer of Mβ, Mγ genes than rice and Arabidopsis, suggesting that rice and Arabidopsis genomes underwent more gene duplication events than the Brachypodium genome [52], [62]. In other words the Brachypodium genome may have experienced less gene duplication events after it diverged from rice about 40–53 MYA and hence slower birth and death rate [52].

Table 2. MADS-box genes in Brachypodium, Arabidopsis, rice, poplar, apple, cucumber, grapevine and soybean genomes.

Differential expression of type II MADS-box genes in Brachypodium, rice and Arabidopsis

Comparing expression patterns of MADS-box gene homologous, we found that close counterparts had different expression profiles, offering clues about possible functional divergence beyond initial divergence of Brachypodium and rice 40–53 MYA [52]. Three AP1-like genes BdMADS3, 10 and 33 were uniformly expressed in reproductive organs, similar to what occurs in Arabidopsis [63]. However, rice homologues were also expressed in leaf (OsMADS14 and 15) and root (OsMADS18), besides in panicle and seed [23], [64]. Similar to expression patterns of PI in Arabidopsis, homologues in Brachypodium (BdMADS16 and 20) and rice (OsMADS2 and 4) also were obviously expressed in lodicule and stamen [65][66]. Furthermore, the subtle differences in expression patterns of each of the two PI-like paralogous genes in rice and Brachypodium likely reflected functional divergences. In addition, both OsMADS2 and 4 were expressed in lodicule and stamen, and OsMADS2 also was expressed in root and leaf [23]. In Brachypodium, BdMADS16, but not BdMADS20 had weak expression signals in carpel and young seed (Fig. 5). Furthermore, AP3 and its Brachypodium homolog BdMADS5 had similar expression patterns in the second and third flower whorl, however, the rice AP3-like gene OsMADS16 was expressed in the leaf and root [67][68]. Along with rice OsMADS29, 30 and 31, BdMADS17, 23 and 38 formed the B-function subclade that is specific in monocot plants (Fig. 1). OsMADS29 is interesting because it expresses specifically during S1-S4 stage of seed development, positively regulating embryo and endosperm development by affecting hormone homeostasis [23], [69]. But, no such functions were observed for the two paralogues (OsMADS30 and 31) [23]. The expression of the Brachypodium orthologue of OsMADS29 (BdMADS23) can be detected in both palea and young seed, indicating that it may play a similar role, while BdMADS38 can only be detected in stamen. These data suggest that the B-function clade genes in monocots may have significantly diversified. The OsMADS32-like clade was identified as a novel group specific for monocots [23], [41], [70]. In this study, BdMADS19 transcripts were mainly detectable in palea and stamen, and this expression pattern was similar to that of OsMADS32 [41]. However, the close homolog TaAGL14 and 15 in wheat, was expressed in the leaf and root of most plant developmental stages other than panicle and developing seed [70]. Expression patterns of OsMADS32-like genes are strikingly different among monocot plants, suggesting that this young clade is undergoing dynamic functional divergence.

Functions of MIKC*-type MADS-box genes are less clearer than MIKCc-type genes. In Arabidopsis, the heterodimers of MIKC*-type proteins were required for the pollen maturation and tube growth [15]. Rice MIKC*-type genes displayed low level transcription, except for OsMADS65 which had strong and constitutive signals [23]. In this study, MIKC*-type genes also had weak transcription signals, suggesting a need for further investigation.

The potential functions of type I MADS-box genes

In contrast to type II MADS-box genes, not much is known about type I plant genes [71]. Recent studies suggest that type I genes are important for plant reproduction and developmental stages, especially for determining female gametophyte, embryo, and endosperm development in Arabidopsis [72][75]. As similar in Arabidopsis, expression of some type I genes was too weak to detect using the RT-PCR method in Brachypodium, and was difficult to detect using microarray techniques in rice [23], [76][78]. Of the nine Mα group genes, only BdMADS44 and 46 were detectable in our studies. BdMADS44 was mainly expressed in palea, stamen, and young seed, similar to its homologs OsMADS77, 78 and 79 in rice. BdMADS46 and its rice counterpart OsMADS75 was preferentially expressed in young seed [23]. All Mβ genes were detectable in reproductive organs. The recent duplicates BdMADS8 and 21 were also weakly expressed in root. The expression of type I MADS-box genes in reproductive organs manifested their functions in plant reproduction. On the other hand, our work on the response of type I genes to abiotic stresses demonstrate that these ancient genes are crucial to survival of the plants under restricted growth conditions. Therefore, type I genes warrant further investigations for their roles in specifying floral organs as well as response to stressing environments.

Supporting Information

File S1.

Table S1 MADS-box genes in Brachypodium and their characteristics. Table S2 Accession numbers for MADS-box genes in Brachypodium. Table S3 Primer sets used for the semi-RT PCRs. Table S4 Conserved motifs predicted by MEME program.



We thank Dr. Ruibo Hu, Institute of bioenergy and bioprocess technology, CAS for critical reading of this manuscript.

Author Contributions

Conceived and designed the experiments: BW LM XZ. Performed the experiments: BW JG DL. Analyzed the data: BW RZ AL RF. Contributed reagents/materials/analysis tools: BW RZ. Wrote the paper: BW LM.


  1. 1. Alvarez-Buylla ER, Pelaz S, Liljegren SJ, Gold SE, Burgeff C, et al. (2000) An ancestral MADS-box gene duplication occurred before the divergence of plants and animals. Proceedings of the National Academy of Sciences 97: 5328–5333.
  2. 2. Causier B, Kieffer M, Davies B (2002) MADS-box genes reach maturity. Science 296: 275–276.
  3. 3. Pelaz S, Ditta GS, Baumann E, Wisman E, Yanofsky MF (2000) B and C floral organ identity functions require SEPALLATA MADS-box genes. Nature 405: 200–203.
  4. 4. Theissen G, Becker A, Di Rosa A, Kanno A, Kim JT, et al. (2000) A short history of MADS-box genes in plants. Plant Molecular Biology 42: 115–149.
  5. 5. Bowman JL, Drews GN, Meyerowitz EM (1991) Expression of the Arabidopsis floral homeotic gene AGAMOUS is restricted to specific cell types late in flower development. Plant Cell 3: 749–758.
  6. 6. Coen ES, Meyerowitz EM (1991) The war of the whorls: genetic interactions controlling flower development. Nature 353: 31–37.
  7. 7. Angenent GC, Colombo L (1996) Molecular control of ovule development. Trends in Plant Science 1: 228–232.
  8. 8. Ma H, dePamphilis C (2000) The ABCs of floral evolution. Cell 101: 5–8.
  9. 9. Colombo L, Franken J, Koetje E, van Went J, Dons H, et al. (1995) The Petunia MADS box gene FBP11 determines ovule identity. Plant Cell 7: 1859–1868.
  10. 10. Yanofsky MF, Ma H, Bowman JL, Drews GN, Feldmann KA, et al. (1990) The protein encoded by the Arabidopsis homeotic gene agamous resembles transcription factors. Nature 346: 35–39.
  11. 11. Weigel D, Meyerowitz EM (1994) The ABCs of floral homeotic genes. Cell 78: 203–209.
  12. 12. Kaufmann K, Melzer R, Theiβen G (2005) MIKC-type MADS-domain proteins: structural modularity, protein interactions and network evolution in land plants. Gene 347: 183–198.
  13. 13. Hartmann U, Höhmann S, Nettesheim K, Wisman E, Saedler H, et al. (2000) Molecular cloning of SVP: a negative regulator of the floral transition in Arabidopsis. The Plant Journal 21: 351–360.
  14. 14. Kramer EM, Dorit RL, Irish VF (1998) Molecular evolution of genes controlling petal and stamen development: duplication and divergence within the APETALA3 and PISTILLATA MADS-box gene lineages. Genetics 149: 765–783.
  15. 15. Adamczyk BJ, Fernandez DE (2009) MIKC* MADS domain heterodimers are required for pollen maturation and tube growth in Arabidopsis. Plant Physiology 149: 1713–1723.
  16. 16. Henschel K, Kofuji R, Hasebe M, Saedler H, Münster T, et al. (2002) Two ancient classes of MIKC-type MADS-box genes are present in the moss Physcomitrella patens. Molecular Biology and Evolution 19: 801–814.
  17. 17. De Bodt S, Raes J, Florquin K, Rombauts S, Rouzé P, et al. (2003) Genomewide structural annotation and evolutionary analysis of the type I MADS-box genes in plants. Journal of Molecular Evolution 56: 573–586.
  18. 18. Pařenicová L, de Folter S, Kieffer M, Horner DS, Favalli C, et al. (2003) Molecular and phylogenetic analyses of the complete MADS-box transcription factor family in Arabidopsis: new openings to the MADS world. Plant Cell 15: 1538–1551.
  19. 19. Nam J, Kim J, Lee S, An G, Ma H, et al. (2004) Type I MADS-box genes have experienced faster birth-and-death evolution than type II MADS-box genes in angiosperms. Proceedings of the National Academy of Sciences 101: 1910–1915.
  20. 20. Fan CM, Wang X, Wang YW, Hu RB, Zhang XM, et al. (2013) Genome-wide expression analysis of soybean MADS genes showing potential function in the seed development. PLoS ONE 8: e62288.
  21. 21. Hu L, Liu S (2012) Genome-wide analysis of the MADS-box gene family in cucumber. Genome 55: 245–256.
  22. 22. Velasco R, Zharkikh A, Affourtit J, Dhingra A, Cestaro A, et al. (2010) The genome of the domesticated apple (Malus x domestica Borkh.). Nature Genetics 42: 833–839.
  23. 23. Arora R, Agarwal P, Ray S, Singh AK, Singh VP, et al. (2007) MADS-box gene family in rice: genome-wide identification, organization and expression profiling during reproductive development and stress. BMC Genomics 8: 242.
  24. 24. Leseberg CH, Li A, Kang H, Duvall M, Mao L (2006) Genome-wide analysis of the MADS-box gene family in Populus trichocarpa. Gene 378: 84–94.
  25. 25. Wei B, Cai T, Zhang R, Li A, Huo N, et al. (2009) Novel microRNAs uncovered by deep sequencing of small RNA transcriptomes in bread wheat (Triticum aestivum L.) and Brachypodium distachyon (L.) Beauv. Functional & Integrative Genomics 9: 499–511.
  26. 26. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 25: 4876–4882.
  27. 27. Wei B, Liu D, Guo J, Leseberg CH, Zhang X, et al. (2013) Functional divergence of two duplicated D-lineage MADS-box genes BdMADS2 and BdMADS4 from Brachypodium distachyon. Journal of Plant Physiology 170: 424–431.
  28. 28. Burland T (1999) Bioinformatics methods and protocols. Humana Press 71–91 p.
  29. 29. Kumar S, Nei M, Dudley J, Tamura K (2008) MEGA: A biologist-centric software for evolutionary analysis of DNA and protein sequences. Briefings in Bioinformatics 9: 299–306.
  30. 30. Saitou N, Nei M (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4: 406–425.
  31. 31. Guo A, Zhu Q, Chen X, Luo J (2007) GSDS: a gene structure display server. Yi Chuan 29: 1023–1026.
  32. 32. Bailey TL, Boden M, Buske FA, Frith M, Grant CE, et al. (2009) MEME SUITE: tools for motif discovery and searching. Nucleic Acids Research 37: W202–208.
  33. 33. Schultz J, Milpetz F, Bork P, Ponting CP (1998) SMART, a simple modular architecture research tool: Identification of signaling domains. Proceedings of the National Academy of Sciences 95: 5857–5864.
  34. 34. Letunic I, Doerks T, Bork P (2012) SMART 7: recent updates to the protein domain annotation resource. Nucleic Acids Research 40: D302–305.
  35. 35. Suyama M, Torrents D, Bork P (2006) PAL2NAL: robust conversion of protein sequence alignments into the corresponding codon alignments. Nucleic Acids Research 34: W609–612.
  36. 36. Yang Z (2007) PAML 4: phylogenetic analysis by maximum likelihood. Molecular Biology and Evolution 24: 1586–1591.
  37. 37. Lynch M, Conery JS (2000) The evolutionary fate and consequences of duplicate genes. Science 290: 1151–1155.
  38. 38. Gaut BS, Morton BR, McCaig BC, Clegg MT (1996) Substitution rate comparisons between grasses and palms: synonymous rate differences at the nuclear gene Adh parallel rate differences at the plastid gene rbcL.. Proceedings of the National Academy of Sciences 93: 10274–10279.
  39. 39. Hong S, Seo P, Yang M, Xiang F, Park C (2008) Exploring valid reference genes for gene expression studies in Brachypodium distachyon by real-time PCR. BMC Plant Biology 8: 1–11.
  40. 40. Liu C, Teo ZWN, Bi Y, Song S, Xi W, et al. (2013) A conserved genetic pathway determines inflorescence architecture in Arabidopsis and rice. Developmental cell 24: 612–622.
  41. 41. Sang X, Li Y, Luo Z, Ren D, Fang L, et al. (2012) CHIMERIC FLORAL ORGANS1, encoding a monocot-specific MADS box protein, regulates floral organ identity in rice. Plant Physiology 160: 788–807.
  42. 42. Ma H, Yanofsky MF, Meyerowitz EM (1991) AGL1-AGL6, an Arabidopsis gene family with similarity to floral homeotic and transcription factor genes. Genes & Development 5: 484–495.
  43. 43. Fischer A, Baum N, Saedler H, Theiβen G (1995) Chromosomal mapping of the MADS-box multigene family in Zea mays reveals dispersed distribution of allelic genes as well as transposed copies. Nucleic Acids Research 23: 1901–1911.
  44. 44. Davies B, Egea-Cortines M, de Andrade Silva E, Saedler H, Sommer H (1996) Multiple interactions amongst floral homeotic MADS box proteins. The EMBO journal 15: 4330–4343.
  45. 45. Yamaguchi T, Lee DY, Miyao A, Hirochika H, An G, et al. (2006) Functional diversification of the two C-class MADS box genes OsMADS3 and OsMADS58 in Oryza sativa.. Plant Cell 18: 15–28.
  46. 46. Zhang H, Forde BG (1998) An Arabidopsis MADS box gene that controls nutrient-induced changes in root architecture. Science 279: 407–409.
  47. 47. Shan H, Zahn L, Guindon S, Wall PK, Kong H, et al. (2009) Evolution of plant MADS box transcription factors: evidence for shifts in selection associated with early angiosperm diversification and concerted gene duplications. Molecular Biology and Evolution 26: 2229–2244.
  48. 48. Airoldi CA, Davies B (2012) Gene duplication and the evolution of plant MADS-box transcription factors. Journal of Genetics and Genomics 39: 157–165.
  49. 49. Lee HL, Irish VF (2011) Gene duplication and loss in a MADS box gene transcription factor circuit. Molecular Biology and Evolution 28: 3367–3380.
  50. 50. Eckardt NA (2008) Grass genome evolution. The Plant Cell Online 20: 3–4.
  51. 51. Gaut BS (2002) Evolutionary dynamics of grass genomes. New Phytologist 154: 15–28.
  52. 52. Initiative TIB (2010) Genome sequencing and analysis of the model grass Brachypodium distachyon. Nature 463: 763–768.
  53. 53. Lee JH, Park SH, Ahn JH (2012) Functional conservation and diversification between rice OsMADS22/OsMADS55 and Arabidopsis SVP proteins. Plant Science 185–186: 97–104.
  54. 54. Lee S, Choi SC, An G (2008) Rice SVP-group MADS-box proteins, OsMADS22 and OsMADS55, are negative regulators of brassinosteroid responses. The Plant Journal 54: 93–105.
  55. 55. Paterson AH, Bowers JE, Chapman BA (2004) Ancient polyploidization predating divergence of the cereals, and its consequences for comparative genomics. Proceedings of the National Academy of Sciences 101: 9903–9908.
  56. 56. Yu J, Wang J, Lin W, Li S, Li H, et al. (2005) The genomes of Oryza sativa: a history of duplications. PLoS Biol 3: e38.
  57. 57. Sheldon CC, Rouse DT, Finnegan EJ, Peacock WJ, Dennis ES (2000) The molecular basis of vernalization: The central role of FLOWERING LOCUS C (FLC). Proceedings of the National Academy of Sciences 97: 3753–3758.
  58. 58. Michaels SD, Amasino RM (1999) FLOWERING LOCUS C encodes a novel MADS domain protein that acts as a repressor of flowering. Plant Cell 11: 949–956.
  59. 59. Becker A, Theiβen G (2003) The major clades of MADS-box genes and their role in the development and evolution of flowering plants. Molecular Phylogenetics and Evolution 29: 464–489.
  60. 60. Trevaskis B, Bagnall DJ, Ellis MH, Peacock WJ, Dennis ES (2003) MADS box genes control vernalization-induced flowering in cereals. Proceedings of the National Academy of Sciences 100: 13099–13104.
  61. 61. Yan L, Loukoianov A, Tranquilli G, Helguera M, Fahima T, et al. (2003) Positional cloning of the wheat vernalization gene VRN1.. Proceedings of the National Academy of Sciences 100: 6263–6268.
  62. 62. Guo X, Xu G, Zhang Y, Hu W, Fan L (2004) Small-scale duplications play a significant role in rice genome evolution. Rice Sinence 12: 173–178.
  63. 63. Alejandra Mandel M, Gustafson-Brown C, Savidge B, Yanofsky MF (1992) Molecular characterization of the Arabidopsis floral homeotic gene APETALA1.. Nature 360: 273–277.
  64. 64. Fornara F, Parenicova L, Falasca G, Pelucchi N, Masiero S, et al. (2004) Functional characterization of OsMADS18, a member of the AP1/SQUA subfamily of MADS box genes. Plant Physiology 135: 2207–2219.
  65. 65. Goto K, Meyerowitz EM (1994) Function and regulation of the Arabidopsis floral homeotic gene PISTILLATA. Genes & Development 8: 1548–1560.
  66. 66. Yao S, Ohmori S, Kimizu M, Yoshida H (2008) Unequal genetic redundancy of rice PISTILLATA orthologs, OsMADS2 and OsMADS4, in lodicule and stamen development. Plant and Cell Physiology 49: 853–857.
  67. 67. Jack T, Brockman LL, Meyerowitz EM (1992) The homeotic gene APETALA3 of Arabidopsis thaliana encodes a MADS box and is expressed in petals and stamens. Cell 68: 683–697.
  68. 68. Xiao H, Wang Y, Liu D, Wang W, Li X, et al. (2003) Functional analysis of the rice AP3 homologue OsMADS16 by RNA interference. Plant Molecular Biology 52: 957–966.
  69. 69. Nayar S, Sharma R, Tyagi AK, Kapoor S (2013) Functional delineation of rice MADS29 reveals its role in embryo and endosperm development by affecting hormone homeostasis. Journal of Experimental Botany doi:10.1093/jxb/ert231.
  70. 70. Zhao T, Ni Z, Dai Y, Yao Y, Nie X, et al. (2006) Characterization and expression of 42 MADS-box genes in wheat (Triticum aestivum L.). Molecular Genetics and Genomics 276: 334–350.
  71. 71. Masiero S, Colombo L, Grini PE, Schnittger A, Kater MM (2011) The emerging importance of type I MADS box transcription factors for plant reproduction. Plant Cell 23: 865–872.
  72. 72. Bouyer D, Roudier F, Heese M, Andersen ED, Gey D, et al. (2011) Polycomb repressive complex 2 controls the embryo-to-seedling phase transition. PLoS Genet 7: e1002014.
  73. 73. Kang IH, Steffen JG, Portereiko MF, Lloyd A, Drews GN (2008) The AGL62 MADS domain protein regulates cellularization during endosperm development in Arabidopsis. Plant Cell 20: 635–647.
  74. 74. Bemer M, Wolters-Arts M, Grossniklaus U, Angenent GC (2008) The MADS domain protein DIANA acts together with AGAMOUS-LIKE80 to specify the central cell in Arabidopsis ovules. Plant Cell 20: 2088–2101.
  75. 75. Portereiko MF, Lloyd A, Steffen JG, Punwani JA, Otsuga D, et al. (2006) AGL80 is required for central cell and endosperm development in Arabidopsis. Plant Cell 18: 1862–1872.
  76. 76. Wuest SE, Vijverberg K, Schmidt A, Weiss M, Gheyselinck J, et al. (2010) Arabidopsis female gametophyte gene expression map reveals similarities between plant and animal gametes. Current biology 20: 506–512.
  77. 77. Tiwari S, Spielman M, Schulz R, Oakey R, Kelsey G, et al. (2010) Transcriptional profiles underlying parent-of-origin effects in seeds of Arabidopsis thaliana.. BMC Plant Biology 10: 1–22.
  78. 78. Day RC, Herridge RP, Ambrose BA, Macknight RC (2008) Transcriptome analysis of proliferating Arabidopsis endosperm reveals biological implications for the control of syncytial division, cytokinin signaling, and gene expression regulation. Plant Physiology 148: 1964–1984.
  79. 79. Díaz-Riquelme J, Lijavetzky D, Martínez-Zapater JM, Carmona MJ (2009) Genome-wide analysis of MIKCc-type MADS box genes in Grapevine. Plant Physiology 149: 354–369.