Beta-subunits of the SnRK1 complexes share a common ancestral function together with expression and function specificities; physical interaction with nitrate reductase specifically occurs via AKINbeta1-subunit

Abstract

The SNF1/AMPK/SnRK1 kinases are evolutionary conserved kinases involved in yeast, mammals, and plants in the control of energy balance. These heterotrimeric enzymes are composed of one alpha-type catalytic subunit and two gamma- and beta-type regulatory subunits. In yeast it has been proposed that the beta-type subunits regulate both the localization of the kinase complexes within the cell and the interaction of the kinases with their targets. In this work, we demonstrate that the three beta-type subunits of Arabidopsis (Arabidopsis thaliana; AKINbeta1, AKINbeta2, and AKINbeta3) restore the growth phenotype of the yeast sip1Deltasip2Deltagal83Delta triple mutant, thus suggesting the conservation of an ancestral function. Expression analyses, using AKINbeta promoterbeta-glucuronidase transgenic lines, reveal different and specific patterns of expression for each subunit according to organs, developmental stages, and environmental conditions. Finally, our results show that the beta-type subunits are involved in the specificity of interaction of the kinase with the cytosolic nitrate reductase. Together with previous cell-free phosphorylation data, they strongly support the proposal that nitrate reductase is a real target of SnRK1 in the physiological context. Altogether our data suggest the conservation of ancestral basic function(s) together with specialized functions for each beta-type subunit in plants.

Figure 1.

Complementation of the yeast sip1Δsip2Δgal83Δ triple mutant by AKINβ1, AKINβ2, and AKINβ3. The yeast sip1Δsip2Δgal83Δ mutant was transformed with Arabidopsis AKINβ1, AKINβ2, or AKINβ3 coding sequence. Two serial dilutions were plated on YPD plates supplemented with Glc 2% (Glc) or glycerol-ethanol 3% to 2% (Gly/ethanol) as carbon sources. The cells were grown at 30°C for 4 d. The wild-type (WT) strain was used as a control. sip1Δsip2Δgal83Δ is the untransformed triple mutant strain. [See online article for color version of this figure.]

Figure 2.

Positions of the regulation boxes identified in the promoters of AKINβ genes. The figure shows the positions of the regulatory sequences identified among those described in the literature. All positions are referenced to the transcription start site (t.s.). Exons are boxed, hatched boxes represent 5′ UTR, and bars indicate promoter regions except for those labeled I₁ and I_2, representing leader introns. Braces show the sequences used for the promoter∷GUS constructs. Pollen-specific sequences are represented by squares; full squares are for LAT56 (TGTGGTT; Twell et al., 1991) and gray squares are for AAATGA box (AAATGA; Weterings et al., 1995). Dark induction or light repression boxes are represented by triangles; full triangles are for DE1 dark-inducible element (phytochrome dependent; GGATTTTACAGT; Inaba et al., 2000) and gray triangles are for GT2 sequence (GGTAATT; Dehesh et al., 1990). Sugar repression sequences are represented by circles; full circles are for sugar-repressive element (TTATCC; Lu et al., 1998), gray circles are for α-amylase (TATCCAT; Morita et al., 1998), and white circles are for pyrimidine box (CCTTTT; Morita et al., 1998). Sugar induction SUSIBA2 binding motif is in bold circles (TGACT; Sun et al., 2003). Auxin response elements are represented by diamond shapes (TGTCTC; Ulmasov et al., 1997). The Arabidopsis Information Resource accession numbers for the genes encoding for AKINβ1, AKINβ2, and AKINβ3, respectively, are At5g21170, At4g16360, and At2g28060.

Figure 3.

Histochemical localization of GUS activity in transgenic Arabidopsis plants containing AKINβ-GUS fusions in different organs and during development. A, Seedling development of the AKINβ-GUS Arabidopsis transformants. a, Germination; b, whole seedling; b', zoom on the shoot apical meristem area. B, Flower bud development. c, Inflorescence; d and e, buds artificially opened for the photographs; f, opened flowers; the insets show GUS staining in the pollen. C (g), Flowers and siliques, cleared by chloralhydrate treatment. D, Summary of GUS staining data observed in different organs and during development of Arabidopsis. Each color represents the expression of one, two, or the three AKINβ genes, blue for AKINβ1, yellow for AKINβ2, red for AKINβ3, green for AKINβ1 and AKINβ2, orange for AKINβ2 and AKINβ3, and dark blue for AKINβ1, AKINβ2, and AKINβ3. 1 (a–g), AKINβ1; 2 (a–g), AKINβ2; 3 (a–g), AKINβ3.

Figure 4.

Levels of AKINβ transcripts during senescence. A, Expression pattern of AKINβ2∷GUS in the pedicel. Histochemical localization of GUS activity. B, Position and developmental stages of the leaves used for northern analyses. Arabidopsis seeds, ecotype Columbia (C0), were grown in soil at 21°C under 16-h-light/8-h-dark cycle (long day). Leaves were taken from different developmental stages. a and b, Leaves of, respectively, 8/10-leaf or 13/14-leaf rosettes. c to e, 1-, 2-, and 4-cm-long leaves, respectively, from a 10-cm-diameter rosette. f and g, Leaves from a plant at the beginning of the flowering period. h, Oldest rosette leaves from a flowering plant; and i, oldest leaves at the end of the flowering period. C, Northern blots experiments. Full-length AKIN and EF1α were used to probe the RNA blots. 28S RNA are visualized under UV light illumination of the agarose gel stained with ethidium bromide. rRNA and expression level of EF1α attest to the level of degradation of the RNA and thus the senescent condition of the leaves analyzed.

Figure 5.

Dark/light regulation of AKINβ gene expression. A, B, and C, GUS activities of the AKINβ∷GUS, AKINβ2∷GUS, and AKINβ3∷GUS transformants, respectively, in the middle of the photoperiod (light) or after 2.5 d of dark. In vitro-cultured plants were grown on half-strength Murashige and Skoog media supplemented with 20 g/L Suc at 20°C under a 16-h-light/8-h-dark regime for 3 weeks. Plants were then maintained under this regime for 2.5 d or transferred to dark for 2.5 d (darkness) and collected in the middle of the light period (light). Data represent the mean ± sd of two independent experiments and triplicate measurements. Asterisks indicate significant differences between expression in dark and light conditions (**, <0.01 and *, <0.05, Student's t test).

Figure 6.

Regulation of AKINβ1 gene expression by dark and sugar. Effects of Suc on GUS expression from AKINβ1∷GUS transformants during the photoperiod (A) and after 2.4 d in the dark (B). Arabidopsis seeds were grown, in vitro, on half-strength Murashige and Skoog media diluted twice supplemented with 5, 20, or 40 g/L Suc at 20°C under a 16-h-light/8-h-dark regime during 3 weeks. Plants were then either transferred in the dark for 2.5 d (darkness) or kept under this regime for 2.5 d and collected in the middle of the light period (light). Data represent the mean ± sd of two independent experiments and triplicate measurements. Asterisks indicate significant differences between expression of plants grown on 20 or 40 g/L Suc compared to plants grown on 5 g/L Suc (*, <0.05, Student's t test).

Figure 7.

Interaction of NR with members of the AKIN kinase complex. Interaction between NR2 and members of the AKIN kinase complex was analyzed using in vitro cell-free interaction (A–C) and two-hybrid (D) approaches. For the in vitro cell-free interaction assay, NR2 was bound to ProteinA-Sepharose. Interaction with a partner was considered positive when the binding of the putative partner was higher on the NR2-ProteinA-Sepharose than on the ProteinA-Sepharose alone. Each assay has been performed at least three times. On the NR2 lane, the + and − indicate, respectively, presence and absence of NR2 on the ProteinA-Sepharose; + cold indicates that NR2 was not radiolabeled; since NR2 in vitro translation products present an extra protein of the same size as AKINα2 (approximately 58 kD), only AKINα2 has been radiolabeled in the assay NR2 versus AKINα2. φ, Control lane containing NR2 alone. A star indicates the position of the AKIN subunit or subdomain tested. For each interaction tested, results are summarized at the bottom of tables (+ for a positive interaction and − for no interaction). Since AKINβ3 is constituted only by a KIS and an ASC domain, full-length (FL) and the [KIS/ASC] fragment correspond to the same sequences and thus the assay versus Nter (N terminus) does not exist. C, Control for cell-free interaction assays using luciferase, a protein which does not interact with NR2. On the right, an aliquot of luciferase in vitro translation assay (Tra) is used as a control of translation efficiency. D, Two-hybrid results are representative of eight clones obtained from two independent experiments. Clones are plated on SD − LTH + 3AT 5 mm, at 30°C. The interaction between AKINβ3 and AKINβγ is a positive growth control. Due to the autoactivation of BD-β2KIS and AD-β2KIS constructs, the interaction between NR2 and β2KIS has not been studied by this approach (ND, not determined). The first part of the section corresponds to the interaction between the different subdomains of the AKINβ-subunits (FL, full length, KIS/ASC, KIS, ASC) fused to the AD of GAL4 and NR2 fused to the BD of GAL4. Since AKINβ3[KIS/ASC] corresponds to AKINβ3FL, yeast clones have been plated only once (=FL, same as full length). The second part of the section corresponds to the autoactivation controls of these different subdomains against the empty pGBKT7 vector.