Genome-wide investigation of SQUAMOSA promoter binding protein-like genes in Liriodendron and functional characterization of LcSPL2

Abstract

The plant-specific SQUAMOSA promoter-binding protein-like (SPL) transcription factors play a pivotal role in various developmental processes, including leaf morphogenesis and vegetative to reproductive phase transition. Liriodendron chinense and Liriodendron tulipifera are widely used in landscaping due to their tulip-like flowers and peculiar leaves. However, the SPL gene family in Liriodendron has not been identified and systematically characterized. We systematically identified and characterized the SPL family members in Liriodendron, including phylogeny, gene structure and syntenic analyses. Subsequently, we quantified the expression patterns of LcSPLs across various tissue sites through transcription-quantitative polymerase chain reaction (RT-qPCR) assays and identified the target gene, LcSPL2. Finally, we characterized the functions of LcSPL2 via ectopic transformation. Altogether, 17 LcSPL and 18 LtSPL genes were genome-widely identified in L. chinense and L. tulipifera, respectively. All the 35 SPLs were grouped into 9 clades. Both species had three SPL gene pairs arising from segmental duplication events, and the LcSPLs displayed high collinearity with the L. tulipifera genome. RT-qPCR assays showed that SPL genes were differentially expressed in different tissues, especially. Because LcSPL2 is highly expressed in pistils and leaves, it was selected to describe the SPL gene family of L. chinense by ectopic expression. We showed that overexpression of LcSPL2 in Arabidopsis thaliana resulted in earlier flowering and fewer rosette leaves. Moreover, we observed that overexpression of LcSPL2 in A. thaliana up-regulated the expression levels of four genes related to flower development. This study identified SPL genes in Liriodendron and characterized the function of LcSPL2 in advancing flower development.

Keywords: Flowering time; Liriodendron chinense; SPL transcription factors family; genetic transformation.

Figure 1.

Phylogenetic tree of SPL proteins in L. chinense, L. tulipifera, M. truncatula, P. trichocarpa, A. thaliana and O. sativa. The full-length amino acid sequences of 17 LcSPLs, 18 LtSPLs, 23 MtSPLs, 28 PtSPLs, 16 AtSPLs and 19 OsSPLs were used to construct a phylogenetic tree with 1000 bootstrap replicates by MEGA11.

Figure 2.

Phylogenetic relationships and architecture of the conserved motifs in SPL proteins from Liriodendron. (A) The phylogenetic tree was constructed based on the full-length sequences of LcSPL and LtSPL proteins. (B) Conserved domains structure of LcSPL and LtSPL genes. Black lines indicate relative protein lengths, boxes represent conserved domains and domains are colour-coded. (C) Amino acid motifs in the LcSPL and LtSPL proteins are represented by coloured boxes. Black lines indicate relative protein lengths.

Figure 3.

Multiple sequence alignment of SBP domain between LcSPLs and LtSPLs proteins. C3H and C2HC were two zinc finger structures, NLS was bidirectional nuclear localization signal, which were corresponding to the sequence logos of conserved domains.

Figure 4.

Intraspecific synteny analysis of LcSPLs and LtSPL. (A) Intraspecific synteny analysis of SPL genes in L. tulipifera. The chromatic line indicates there is a collinearity between two given LtSPL genes. (B) Intraspecific synteny analysis of SPL genes in L. chinense. The chromatic  line indicates there is a collinearity between two given LcSPL genes.

Figure 5.

Synteny analysis of SPL genes from LcSPLs and LtSPLs. Grey lines in the background indicate collinear blocks between L. chinense and L. tulipifera, whereas the chromatic lines highlight syntenic SPL gene pairs.

Figure 6.

Synteny analysis of SPL genes among L. chinense, L. tulipifera, A. thaliana and V. vinifera. Grey lines in the background indicate the collinear blocks within different plant genomes, while colour lines highlight the syntenic SPL gene pairs.

Figure 7.

Cis-element regions of SPL genes promoters were analysed by PlantCARE in L. chinense and L. tulipifera. The numbers in the box represent the number of cis-element.

Figure 8.

Expression analysis of LcSPLs in root, stem, leaf, calyx, floral bud, petal, pistil and stamen of L. chinense. The expression was determined by quantitative RT-PCR. LcActin97 was used as an internal control. Means with different letters were significantly different. The error bars indicated ± SD of three technical replicates. Analysis of variance (ANOVA) with a subsequent Duncan’s test was performed (P < 0.05).

Figure 9.

The phenotypes and relative expression levels of LcSPL2 in the transgenic overexpression lines and WT A. thaliana. (A) The rosette leaf of WT A. thaliana and 35S::LcSPL2-1 A. thaliana, bar = 1 cm. (B) The right three A. thaliana plants represented three transgenic lines, LcSPL2-1, LcSPL2-3 and LcSPL2-4, bar = 1 cm. (C) Comparison of the number of rosette leaf at flowering between each 35S::LcSPL2 line and WT A. thaliana. (D) Comparison of the flowering time in each 35S::LcSPL2 transgenic line and WT A. thaliana. The first flowering duration was recorded in days from seeding to the occurrence of the first flower, unit in days. The error bars indicate the standard deviation (SD) of three biological replicates consisting of independent samples (no less than 10 plants per replicate). ANOVA with a subsequent Duncan’s test was performed (double asterisks, P < 0.01).

Figure 10.

RT-qPCR analysis of four genes in WT A. thaliana and three 35S::LcSPL2 overexpression lines under LD condition. (A)–(D) RT-qPCR analysis of four flowering-related genes, AtSOC1, AtFT, AtFUL and AtAP1, in WT and transgenic A. thaliana. Samples were collected on the days to the initial flowering of 35S::LcSPL2 and WT A. thaliana. AtActin2 was used as an internal control. The error bars indicate ± SE of three technical replicates. Means with different letters were significantly different. ANOVA with a subsequent Duncan’s test was performed (P < 0.05).