A double SORLIP1 element is required for high light induction of ELIP genes in Arabidopsis thaliana

Abstract

Promoter elements that contribute to high light (HL) induction of the Arabidopsis ELIP1 gene were defined using a transgenic promoter-reporter system. Two adjacent SORLIP1 elements (double SORLIP1, dSL) were found to be essential for HL induction of a GUS reporter gene. The dSL element was also found to be essential for HL induction conferred by the ELIP2 promoter. SORLIP1 elements were enriched in ELIP promoters throughout the plant kingdom, and showed a clade-specific pattern of gain or loss that suggested functionality. In addition, two G-box elements were found to redundantly contribute to HL induction conferred by the ELIP1 promoter.

Fig. 1

ELIP1 and ELIP2 promoter sequences. Sequence characteristics of a ELIP1- At3g22840 promoter (from −984 to +97 bp) and b ELIP2-At4g14690 promoter (from −883 to +71 bp). Upper case letters correspond to the promoter region and lower case letters correspond to the 5′-UTR. The start of translation (ATG) is underlined and in bold. Below the cis-regions that have been subjected to site-directed mutagenesis in this study (in bold and underlined) are their names and position relative to the start of transcription

Fig. 2

GUS activity in ELIP1p-GUS transgenic lines. A 984 bp region of ELIP1p was fused to the GUS reporter gene (ELIP1 WT). Site-directed mutants of the 984 bp region were generated and also fused to the GUS reporter gene. Transgenic lines were generated and tested for GUS activity in LL and after 4 h of HL to produce the HL/LL fold induction. 20–25 transgenic lines were analyzed for each construct and error bars indicate the interquartile range (IQR). The location and sequence of promoter elements are shown in Fig. 1 while the nucleotide changes for each element are shown in Table 1. The SL1 and SL2 mutants change one of the two SORLIP1 elements in the dSL region. Pairwise comparisons between ELIP1 WT and each mutant construct were evaluated for statistical significance using the Mann–Whitney test. *p value 0.0500–0.0100, **p value 0.009–0.001, ***p value 0.0009 or below, but greater than 0

Fig. 3

GUS mRNA levels in ELIP1p-GUS transgenic lines. ELIP1p-GUS lines were tested for RNA expression by real-time qPCR harvested from the same tissue used for GUS activity measurements. Both GUS and native ELIP1 mRNA levels were quantified and samples that displayed a less than twofold induction of native ELIP1 were removed from the analysis. GUS mRNA levels with a significantly lower HL/LL fold induction than ELIP1p WT are shown in white bars. The number of transgenic lines analyzed varied among constructs: ELIP1 WT (15), G-box-UpG-box (14), SL1 (8), SL2 (20), dSL (26), dSL-G-box (19), dSL-G-box-UpG-box (16). Error bars represent IQR. Statistical analysis and p values are as indicated for Fig. 2

Fig. 4

The dSL element of ELIP2 is required for HL induction. ELIP2 WT (n = 14) and dSL mutant (n = 15) transgenic lines were tested for native ELIP2 and GUS mRNA expression by real-time qPCR. Samples that displayed a less than twofold induction of native ELIP2 were removed from the analysis. Error bars represent IQR. Statistical analysis and p values are as indicated for Fig. 2

Fig. 5

SORLIP1 elements in ELIP promoters throughout the plant kingdom. 37 ELIP promoters throughout the plant kingdom were identified and 1,500 bp upstream from the start of translation was scanned for SORLIP1 elements (GCCAC). The number of promoters (y-axis) with different numbers of SORLIP1 elements (x-axis) is shown. SORLIP1 elements were enriched 1.5 fold in ELIP promoter regions compared to random occurrence

Fig. 6

SORLIP1 element gain/loss in ELIP promoter regions throughout the plant kingdom. ELIP promoters were scanned for SORLIP1 elements (GCCAC) and a phylogenetic tree was generated using Mesquite. Black signifies presence of at least one SORLIP1 element. The probability of a common ancestor having one or more SORLIP1 elements is shown by the proportion of black in common ancestor circles. The numbers represent maximum likelihood values supporting the ancestral proportions