Genome-wide identification of MADS-box gene family in sacred lotus (Nelumbo nucifera) identifies a SEPALLATA homolog gene involved in floral development

Abstract

Background: Sacred lotus (Nelumbo nucifera) is a vital perennial aquatic ornamental plant. Its flower shape determines the horticultural and ornamental values. However, the mechanisms underlying lotus flower development are still elusive. MADS-box transcription factors are crucial in various features of plant development, especially in floral organogenesis and specification. It is still unknown how the MADS-box transcription factors regulate the floral organogenesis in lotus.

Results: To obtain a comprehensive insight into the functions of MADS-box genes in sacred lotus flower development, we systematically characterized members of this gene family based on the available genome information. A total of 44 MADS-box genes were identified, of which 16 type I and 28 type II genes were categorized based on the phylogenetic analysis. Furthermore, the structure of MADS-box genes and their expressional patterns were also systematically analyzed. Additionally, subcellular localization analysis showed that they are mainly localized in the nucleus, of which a SEPALLATA3 (SEP3) homolog NnMADS14 was proven to be involved in the floral organogenesis.

Conclusion: These results provide some fundamental information about the MADS-box gene family and their functions, which might be helpful in not only understanding the mechanisms of floral organogenesis but also breeding of high ornamental value cultivars in lotus.

Keywords: Floral organogenesis; Genome-wide analysis; Lotus; MADS-box; SEP3.

Fig. 1

Classification and phylogenetic analyses of MADS-box family genes among different plant species. a The number of MADS-box family genes and their classification in 35 different plant species. The data for lotus, Arabidopsis, and rice are marked in green, red and blue, respectively, which were used for the following phylogenetic analysis. b Phylogenetic analysis of the MADS-box proteins from lotus, Arabidopsis and rice. The phylogenetic tree was constructed with the full-length protein sequences by the neighbor-joining method using MEGA software version 5.2. The digital number on each branch represents bootstrap scores for 1000 replicate analyses. A branch with a bootstrap score below 50 was considered unreliable and cut off. The text out of the blue and black lines represent different types or subtypes of MADS-box proteins

Fig. 2

Phylogenetic relationship, gene structure, and motif analyses of NnMADS genes. a A neighbor-joining tree was constructed based on the alignment of full-length amino acid sequences of NnMADS protein. b The gene structures of each NnMADS gene without the untranslated regions (UTRs). Analysis was carried out using the Gene Structure Display Server tool (http://gsds.cbi.pku.edu.cn/). The size scale is indicated at the bottom. c Putative motifs in NnMADS proteins were analyzed by MEME. The boxes with different colors represent different putative motifs, and are marked by the numbers at the bottom. The scale bar represents the length of proteins in terms of amino acids

Fig. 3

Conservative analysis on MADS-box domain among different NnMADS proteins. a MADS-box MEF like alignment. b MADS-box SRF like alignment. HsMEF2A, HsMEF2B, HsMEF2C, HsMEF2D and SRF sequences were retrieved from Homo sapiens database as reference. Protein sequences correspond to the conserved regions on all NnMADS. In the consensus line, uppercase letters represent identity in more than 50% of sequences and X represents less than 50% identity. The red box marked letters indicate the specific residues involved in a conserved feature (referred to MEF and SRF domain from NCBI)

Fig. 4

The expressional patterns of NnMADS genes in different tissues of N. nucifera. a The left panel shows the MIKC^c sub-type members of type II MADS box genes; the upper panel of right side shows MIKC* sub-type members of type II MADS box genes, and the lower panel of the right side shows those of the type I. The FPKM values were collected from the online data as mentioned in M&M, and used for drawing the heatmap. b qRT-PCR analysis on the relative expressions of 14 representative NnMADS genes in different tissues of N. nucifera. Transcript level of each NnMADS were first normalized to those of the housekeeping gene NnActin and then compared to the expression level of each gene in leaf. Data are means ±SD (n = 3). L: Leaf; Pe: Petiole; three parts of the rhizome, such as Rt: rhizome tip, Rez: Rhizome elongation zone, Ri: Rhizome internode; R: Root; Fb: Flower bud; P: Petal; Being collected stamen, carpel, receptacle before pollination named as immature stamen (iSt), immature carpel (iC), immature receptacle (iRe); Being collected stamen, carpel, receptacle after pollination named as mature stamen (mSt), mature carpel (mC), mature receptacle (mRe); Sc: Seed coat; Co: Cotyledon

Fig. 5

Subcellular localization of six representative NnMADSs fused with green fluorescent protein. NnMADSs were constructed into plasmid with fusions of GFP and driven by the CaMV35S promoter. DAPI was used to mark nuclei. Panels from left to right refer to GFP, Bright, DAPI, and Merged images, respectively. Protein fusions co-localized with DAPI markers are marked with red arrows. The boxes are enlarged images for cytomembranes. Scale bars = 25 μm

Fig. 6

Phenotyping of 35S:NnSEP3 transgenic Arabidopsis. a Early flowering of two transgenic Arabidopsis lines. The inserted panel shows the overexpression of NnSEP3 in the two lines (OE2 and OE8). b Cauline leaf of wide type Arabidopsis. c Transition of cauline leaf to flower in 35S:NnSEP3 transgenic Arabidopsis. The inserted panel shows the transitional flower. d Normal flower in wide type Arabidopsis. e Flower with two pistils in the OE2 line. f Transition of single flower into double flowers with two pistils in the OE8 line. g Normal silique in wide type Arabidopsis. (h, i) Aberrant silique of OE2 (h) and OE8 (i) lines. The inserted panels show the enlarged images of the area in the white rectangles. Bars are 500 μm. j Expression analysis of the endogenous flowering and leaf development-related genes in Arabidopsis. WT: wide type, OE: 35S::NnSEP3. LFY, AP1, SEP3, AG, and FT were analyzed by qRT-PCR in leaves

Fig. 7

Expression analysis of NnSEP3 gene in the flower bud of different lotus strains. a Flower of the longer flowering lotus in late autumn. The flowering time expand from June to November. b Flowers of different lotus species. c Expression levels of NnSEP3 gene among different strains. I, longer flowering lotus; II, ordinary lotus; III, strain with stamen petaloid; IV, strain with ovary petaloid; V, all-double-petalled-flower strain; VI, strain with multi- all-double-petalled-flowers in one flower bud