Medicago truncatula SOC1 Genes Are Up-regulated by Environmental Cues That Promote Flowering

Abstract

Like Arabidopsis thaliana, the flowering of the legume Medicago truncatula is promoted by long day (LD) photoperiod and vernalization. However, there are differences in the molecular mechanisms involved, with orthologs of two key Arabidopsis thaliana regulators, FLOWERING LOCUS C (FLC) and CONSTANS (CO), being absent or not having a role in flowering time function in Medicago. In Arabidopsis, the MADS-box transcription factor gene, SUPPRESSOR OF OVEREXPRESSION OF CONSTANS 1 (AtSOC1), plays a key role in integrating the photoperiodic and vernalization pathways. In this study, we set out to investigate whether the Medicago SOC1 genes play a role in regulating flowering time. Three Medicago SOC1 genes were identified and characterized (MtSOC1a-MtSOC1c). All three MtSOC1 genes, when heterologously expressed, were able to promote earlier flowering of the late-flowering Arabidopsis soc1-2 mutant. The three MtSOC1 genes have different patterns of expression. However, consistent with a potential role in flowering time regulation, all three MtSOC1 genes are expressed in the shoot apex and are up-regulated in the shoot apex of plants in response to LD photoperiods and vernalization. The up-regulation of MtSOC1 genes was reduced in Medicago fta1-1 mutants, indicating that they are downstream of MtFTa1. Insertion mutant alleles of Medicago soc1b do not flower late, suggestive of functional redundancy among Medicago SOC1 genes in promoting flowering.

Keywords: Medicago; flowering time; genome evolution; legume; photoperiod; vernalization.

FIGURE 1

Phylogenetic relationships among SOC1 proteins. (A) Phylogeny of Rosid SOC1 proteins showing two distinct clades of Fabaceae SOC1, based upon a neighbor-joining tree from a MUSCLE alignment of residues 61–170, comprising a region spanning the Intervening and Keratin domains. Related A. thaliana AGAMOUS-LIKE sequences were used as outgroups. Species abbreviations: At: Arabidopsis thaliana, Bj: Brassica juncea, Bn: Brassica napus, Bp: Betula platyphylla, Br: Brassica rapa, Ca: Cicer arietinum, Cf: Cardamine flexuosa, Cs: Citrus sinensis, Fv: Fragaria vesca, Glyma: Glycine max, Hb: Hevea brasiliensis, Jc: Jatropha curcas, Md: Malus x domestica, Mt: Medicago truncatula, Phvul: Phaseolus vulgaris, Pm: Prunus mume, Pp: Pyrus pyrifolia, Py: Prunus x yedoensis, Ps: Pisum sativum, Sa: Sinapis alba, Tp: Trifolium pratense. (B) Alignment of full-length Arabidopsis thaliana (Col-0) and Medicago (R108 accession) SOC1.

FIGURE 2

Complementation of late-flowering Arabidopsis soc1-2 (Col) mutants by Medicago SOC1 genes. Flowering time phenotypes of genetically independent transgenic plant lines ectopically expressing SOC1 transgenes in the soc1-2 mutant background. (A) Flowering time of homozygous T₃ transgenic lines harboring 35S:MtSOC1a or 35S:MtSOC1b cassettes. (B) Flowering time of homozygous T₃ transgenic lines harboring a 35S:MtSOC1c cassette. (C) Flowering time of T₂ transgenic lines harboring a 35S:AtSOC1 cassette. Data are mean ± SD of at least 12 plants grown under 16 h illumination at ∼120 μE.m².s^-1 light intensity at 22°C day time/18°C night time. Statistically significant differences between means of soc1-2 versus each other genotype were determined by one-way ANOVA. Asterisks, where annotated, denote P-values. ^∗P ≤ 0.05, ^∗∗P ≤ 0.01, ^∗∗∗P ≤ 0.001, ^∗∗∗∗P ≤ 0.0001.

FIGURE 3

Expression pattern of MtSOC1 genes. (A) MtSOC1a, (B) MtSOC1b, and (C) MtSOC1c. RNA was isolated from various tissues of WT c.v. Jester plants grown under inductive conditions. Vegetative tissues were harvested 15 days after sowing and reproductive tissues 35 days after sowing in soil. Data are the mean ± SE of three biological replicates and transcripts were normalized to PDF2. All tissues were harvested at ZT 2.

FIGURE 4

Time course of MtSOC1 expression during development. (A–E) Expression of indicated genes in wild-type R108 and (F–I) in fta1-1 (R108 background). Wild-type plants were grown under LD and SD photoperiod, with and without vernalization. fta1-1 plants were grown under LD alone, with and without vernalization. All treatments were sown into soil at the same time. Aerial tissues of plants excluding cotyledons were harvested for days 5–15 and apical nodes only for days 20 and 25. Data are the mean ± SE of 2–4 biological replicates and transcripts were normalized to PDF2. All samples were harvested at ZT 2.

FIGURE 5

Regulation of MtSOC1 in apical nodes by changes in photoperiod. Expression of (A) MtSOC1a, (B) MtSOC1b, (C) MtSOC1c, and (D) MtPIM and MtFULc. Vernalized WT (c.v. R108, black) and fta1-1 (R108, gray) plants were grown under SD photoperiod until the monofoliate leaf had appeared (∼6 days). Plants were sampled [SD-BS refers to SD sample collected before shifting (BS) into LDs] and then transferred to LD photoperiod for sampling for 3 days (1 LD, 2 LD, and 3 LD). The remaining plants were then returned to SD for 3 days and sampled each day (1 SD, 2 SD, and 3 SD). LD control plants were grown permanently under LD and harvested at the conclusion of the experiment. Samples are the mean ± SE of 3 biological replicates and transcripts were normalized to PDF2. All tissues were harvested at ZT 4 in both light regimes.

FIGURE 6

The soc1b mutant does not flower late. Flowering time of vernalized wild-type R108 and insertion mutant line NF1789 grown under LD photoperiod. Individual plants of a segregating NF1789 line were genotyped to determine if they were wild-type, heterozygous or homozygous mutants. Data are mean ± SE of 8 or more plants. There is no statistically significant difference between the flowering times of the lines.