Subfunctionalization of phytochrome B1/B2 leads to differential auxin and photosynthetic responses

Abstract

Gene duplication and polyploidization are genetic mechanisms that instantly add genetic material to an organism's genome. Subsequent modification of the duplicated material leads to the evolution of neofunctionalization (new genetic functions), subfunctionalization (differential retention of genetic functions), redundancy, or a decay of duplicated genes to pseudogenes. Phytochromes are light receptors that play a large role in plant development. They are encoded by a small gene family that in tomato is comprised of five members: PHYA, PHYB1, PHYB2, PHYE, and PHYF. The most recent gene duplication within this family was in the ancestral PHYB gene. Using transcriptome profiling, co-expression network analysis, and physiological and molecular experimentation, we show that tomato SlPHYB1 and SlPHYB2 exhibit both common and non-redundant functions. Specifically, PHYB1 appears to be the major integrator of light and auxin responses, such as gravitropism and phototropism, while PHYB1 and PHYB2 regulate aspects of photosynthesis antagonistically to each other, suggesting that the genes have subfunctionalized since their duplication.

Keywords: auxin; gene duplication; photosynthesis; phototropism; phytochrome; subfunctionalization.

Figure 1

phyB1 and phyB2 regulate expression of genes involved in different biological processes. We identified 121 phyB1‐regulated genes and 73 phyB2‐regulated genes. Gene ontology functional enrichment analysis of these gene groups identified biological processes specifically regulated by phyB1 and phyB2. For all significant GO category enrichments, the black bars represent the number of genes with that annotation in that group (Significant) and the gray bars represent the expected number of genes with that annotation if representation was random (Expected)

Figure 2

Co‐expression modules show phyB1 and phyB2 differently regulate gene networks involved in auxin and photosynthesis related biological processes among others. (a) For each co‐expression module (indicated by color) and the genes that did not fall into a co‐expression module (gray), the average expression vector (eigenvector) across conditions and genotypes was correlated to condition (dark = 0, 60 min R exposure = 1) and genotype (phyB1 column: WT and phyB2 = 0, phyB1 = 1; phyB2 column: WT and phyB1 = 0, phyB2 = 1). R ² values from the Pearson correlations are indicated in the heatmap by color according to scale on the right as well as by their printed value in the grid with p‐values below in parentheses. (b) Gene ontology functional enrichment analysis identified biological processes central to each co‐expression module. Displayed here are four enriched GO biological processes for the brown, green, and blue modules. The black bars represent the number of genes with that annotation in that group (Significant) and the gray bars represent the expected number of genes with that annotation if random (Expected)

Figure 3

In white light, phyB1 mutants show significantly faster phototropism than wild type or phyB2 mutants. The average degree to which 5‐day‐old dark‐grown seedlings bent toward unidirectional white light (bend angle) over 3 hr is shown. Error bars represent standard error. Combined data from three biological replicates are shown, n = 5 seedlings per genotype per time point per biological replicate. A two‐way ANOVA with time and genotype was performed followed by Tukey's post hoc test using the software R. Shared letters represent no statistically significant difference

Figure 4

In R, phyB1 mutants show significantly faster gravitropism than wild type or phyB2 mutants. The average degree to which 5‐day‐old dark‐grown seedlings bent toward the negative gravity vector (i.e., upwards) after gravistimulation over 24 hr is shown. Seedlings were either gravistimulated in the dark (left), or with 135 µmol photons m⁻² s⁻¹ of R. Error bars represent standard error. The dark and R plots each contain data from three biological replicates. N = 20 per genotype per time point per biological replicate. A three‐way ANOVA with time, genotype, and light condition was performed followed by Tukey's post hoc test in R. Shared letters represent no statistically significant difference

Figure 5

Photosynthetic activity is enhanced by phyB2 and repressed by phyB1 independent of light intensity. Photosynthetic activity was measured under varying light intensities in 6‐week‐old WT, phyB1, and phyB2 mutants grown at 25°C (16 hr day/8 hr night) using a LiCOR 6400XT. Three biological replicates were performed with 10 plants per genotype per replicate. Data were normalized in two different ways either by leaf area (a and c) or by leaf area and fresh weight of the leaf tissue that was used for photosynthetic rate measurement. Data were statistically analyzed with a one‐way ANOVA followed by a Tukey post hoc test using the software R. In each panel, data points not connected by a shared letter are statistically significantly different