The Arabidopsis thylakoid ADP/ATP carrier TAAC has an additional role in supplying plastidic phosphoadenosine 5'-phosphosulfate to the cytosol

Abstract

3'-Phosphoadenosine 5'-phosphosulfate (PAPS) is the high-energy sulfate donor for sulfation reactions. Plants produce some PAPS in the cytosol, but it is predominantly produced in plastids. Accordingly, PAPS has to be provided by plastids to serve as a substrate for sulfotransferase reactions in the cytosol and the Golgi apparatus. We present several lines of evidence that the recently described Arabidopsis thaliana thylakoid ADP/ATP carrier TAAC transports PAPS across the plastid envelope and thus fulfills an additional function of high physiological relevance. Transport studies using the recombinant protein revealed that it favors PAPS, 3'-phosphoadenosine 5'-phosphate, and ATP as substrates; thus, we named it PAPST1. The protein could be detected both in the plastid envelope membrane and in thylakoids, and it is present in plastids of autotrophic and heterotrophic tissues. TAAC/PAPST1 belongs to the mitochondrial carrier family in contrast with the known animal PAPS transporters, which are members of the nucleotide-sugar transporter family. The expression of the PAPST1 gene is regulated by the same MYB transcription factors also regulating the biosynthesis of sulfated secondary metabolites, glucosinolates. Molecular and physiological analyses of papst1 mutant plants indicate that PAPST1 is involved in several aspects of sulfur metabolism, including the biosynthesis of thiols, glucosinolates, and phytosulfokines.

Figure 1.

Biosynthesis and Transport of PAPS in Plant Cells. PAPS is synthesized either via ATPS1-4 and APK1/2/4 in plastids or via ATPS and APK3 in the cytosol. Sulfation catalyzed by cytosolic sulfotransferases results in the formation of PAP, which is redirected to plastids and converted into AMP and Pi by the phosphatase FRY1/SAL1. APS is metabolized to sulfite by APS reductase (APR). Sulfite is further reduced to sulfide, which is subsequently incorporated into Cys, a precursor for the synthesis of GSH. A putative PAPS/PAP antiporter exports PAPS into the cytosol and imports PAP into chloroplasts. APR, APS reductase; FRY1/SAL1, FIERY/Increased Salt Tolerance (At5g63890); SOT, sulfotransferase. For details, see text.

Figure 2.

TAAC Is Regulated by MYB Factors Controlling Biosynthesis of GS. (A) Transcript levels of putative PAPS transporters in rosette leaves of 5-week-old MYB51 and MYB28 overexpression plants were determined by quantitative RT-PCR. Relative gene expression values are given compared with the wild type (=1). Data show means ± sd (n = 3). (B) Trans-activation of promoters of putative PAPS transporters by MYB51 and MYB28. Cotransformation assays for the determination of target gene specificity of MYB51 and MYB28 effectors using two different promoter fragments of At5g01500/TAAC [ProAt5g01500/TAAC-(2):GUS and ProAt5g01500/TAAC-(3):GUS], At3g51870, KVAG1, and KVAG2. The promoter of the BAT5 gene was used as a positive control for the activation by MYB28. Analysis of trans-activation by MYB4 was used as a negative control. This transcription factor is not involved in the biosynthesis of secondary sulfated compounds. An assay with no transcription factor (no TF) provided an additional negative control. Cultured Arabidopsis cells were inoculated with the supervirulent Agrobacterium strain LBA4404.pBBR1MCS.virGN54D containing either only the reporter construct (TargetPromoter:GUS:pGWB3i) or the reporter construct and, in addition, the MYB effector. GUS staining indicates trans-activation of a given promoter by an “effector.”

Figure 3.

TAAC-Mediated Uptake of [α-³²P]ATP into Preloaded Liposomes. TAAC was purified from E. coli inclusion bodies and reconstituted into liposomes preloaded with 5 mM of the indicated test substrates or a buffer only control (-). Uptake was conducted at a concentration of 50 µM [α-³²P]ATP and stopped at 2.5 min. The horizontal line displays the uptake value of nonpreloaded control liposomes. Homoexchange (ATP/ATP transport) was set to 100%, and transport into the remaining differently preloaded liposomes was calculated accordingly. Data are means of three independent experiments; standard errors are given.

Figure 4.

Subcellular Localization of PAPST1 Revealed by Confocal Fluorescence Microscopy. Transient expression of the full-length PAPST1 coding region fused to GFP under control of the endogenous PAPST1 promoter ([A], [D], and [G] to [I]) or the 35S CaMV promoter ([B] and [E]). (A) to (F) Subcellular localization studies of PAPST1 in cultured Arabidopsis root cells ([A] to [C]) and in Arabidopsis mesophyll protoplasts ([D] to [F]). The TPT-GFP transiently expressed in cultured Arabidopsis root cells (C), TPT-RFP in Arabidopsis mesophyll protoplasts (F), and TPT-GFP in Arabidopsis mesophyll cells. For TPT-RFP, the autofluorescence is shown in green and RFP in red (F). For PAPST1-GFP and TPT-GFP constructs, the autofluorescence is shown in red and GFP in green ([D] and [E]). (E) Inset is a Z-scan through the cell showing GFP-labeled chloroplast envelopes (indicated by white arrows) and thylakoids (indicated by yellow arrows). (G) to (I) Expression of PAPST1-GFP driven by its own promoter. (J) to (L) Expression of TPT-GFP driven by the CaMV 35S promoter in mesophyll cells of Arabidopsis leaves transiently expressing corresponding fusion proteins. (J) to (L) The positive control for IEM localization. (D), (F), (I), and (L) are overlays of GFP and chlorophyll autofluorescence signals. Bars = 10 μm.

Figure 5.

Histochemical GUS Staining in Tissues of ProPAPST1:GUS Plants. This reporter gene fusion construct comprises the PAPST1 promoter, the full-length PAPST1 genomic sequence, and its 3′-UTR region. (A) Meristematic tissue in shoot of 7- to 10-d-old seedlings. Bar = 500 μm. (B) Meristematic tissue in roots of 7- to 10-d-old seedlings. Bar = 500 μm. (C) Three-week-old plant. Bar = 0.5 cm. (D) Adult plant. Bar = 3 cm. (E) Transition to flowering. Bar = 3 cm. (F) Flowers and meristem. Bar = 0.3 cm. (G) Inflorescences. Bar = 0.5 cm. (H) Roots of an adult plant. Bar = 500 μm. (I) GUS induction at cut site of inflorescences. Bar = 1.25 cm. (J) Developing seeds. Bar = 200 μm. (K) Seed with embryo. Bar = 100 μm. (L) Embryo. Bar = 200 μm.

Figure 6.

T-DNA Insertion Lines of papst1 Knockout Plants. (A) The positions of two independent T-DNA insertion alleles of PAPST1 in Col-0 (SALK_039194) and Ws-4 (FLAG 443D03). The insertion in FLAG 443D03 has been previously described by Thuswaldner et al. (2007). Boxes indicate exons; lines indicate introns. (B) Phenotypes of the two papst1 mutant lines grown for 5 weeks in soil. Bar = 3 cm.

Figure 7.

PAPST1 Is Involved in GS Biosynthesis. (A) Contents of GS (nmol/mg dry weight) in papst1-1 mutant plants relative to wild-type plants (Col-0). AG, aliphatic GS; IG, indolic GS. (B) Contents of GS in papst1-2 relative to wild-type plants (Ws-4). (C) Content of desulfo-GS (nmol/mg dry weight) in papst1-1 mutant plants relative to wild-type plants (Col-0). (D) Content of desulfo-GS (nmol/mg dry weight) in papst1-2 mutant plants relative to wild-type plants (Ws-4). (E) Transcript levels of GS/PAPS pathway genes in rosette leaves of 5-week-old papst1-1 were determined by qRT-PCR. Relative gene expression values are given compared with wild-type Col-0 (=1). Data show means ± sd (n = 3). Asterisks indicate significant differences in comparison to wild-type Col-0 (Student’s t test, P < 0.05).

Figure 8.

Sulfate Assimilation Is Affected in papst1-1 Mutant Plants. Levels of thiols were determined as described in Methods. Means ± sd (n = 3). Asterisk indicates significant difference in comparison to wild-type Col-0 (Student’s t test, P < 0.05). γ-EC, γ-glutamylcysteine; FW, fresh weight.

Figure 9.

Effect of PAPST1 Mutation on PSK Precursor Genes. Transcript levels of PSK2 and PSK4 precursor genes in rosette leaves of 5-week-old papst1-1, apk1 apk2 apk4, and tpst mutants were determined by qRT-PCR. Relative gene expression values are given compared with wild-type Col-0 (=1). Means ± sd (n = 3). Asterisks indicate significant differences in comparison to wild-type Col-0 (Student’s t test, P < 0.05).