New insights into the post-translational modification of multiple phosphoenolpyruvate carboxylase isoenzymes by phosphorylation and monoubiquitination during sorghum seed development and germination

Journal of Experimental Botany, Vol. 67, No. 11 pp. 3523–3536, 2016 doi:10.1093/jxb/erw186 Advance Access publication 18 May 2016 This paper is available online free of all access charges (see http://jxb.oxfordjournals.org/open_access.html for further details) RESEARCH PAPER New insights into the post-translational modification of multiple phosphoenolpyruvate carboxylase isoenzymes by phosphorylation and monoubiquitination during sorghum seed development and germination IsabelRuiz-Ballesta1, GuillermoBaena1, JacintoGandullo 1, Liqun Wang 2,3, Yi-MinShe2, William CharlesPlaxton4 and CristinaEchevarría1,* 1 Departamento de Biología Vegetal, Facultad de Biología, Universidad de Sevilla, Avda Reina Mercedes nº 6, 41012 Sevilla, Spain 2 Shanghai Center for Plant Stress Biology, Chinese Academy of Sciences, 3888 Chenhua Road, Shanghai 201602, China 3 State Key Laboratory of Crop Genetics and Germplasm Enhancement, Soybean Research Institute, Nanjing Agricultural University, Nanjing, Jiangsu 210095, China 4 Department of Biology, Queen’s University, Kingston, Ontario K7L 3N6, Canada * Correspondence: [email protected] Received 23 December 2015; Accepted 11 April 2016 Editor: Gerhard Leubner, Royal Holloway, University of London Abstract Phosphoenolpyruvate carboxylase (PEPC; E.C. 4.1.1.31) was characterized in developing and germinating sorghum seeds, focusing on the transcript and polypeptide abundance of multiple plant-type phosphoenolpyruvate carboxylase (PTPC) genes, and the post-translational modification of each isoenzyme by phosphorylation versus monoubiquitination during germination. We observed high levels of SbPPC4 (Sb07g014960) transcripts during early development (stage I), and extensive transcript abundance of SbPPC2 (Sb02g021090) and SbPPC3 (Sb04g008720) throughout the entire life cycle of the seed. Although tandem mass spectrometry (MS) analysis of immunopurified PTPC indicated that four different PTPC isoenzymes were expressed in the developing and germinating seeds, SbPPC3 was the most abundant isozyme of the developing seed, and of the embryo and the aleurone layer of germinating seeds. In vivo phosphorylation of the different PTPC isoenzymes at their conserved N-terminal seryl phosphorylation site during germination was also established by MS/MS analysis. Furthermore, three of the four isoenzymes were partially monoubiquitinated, with MS/MS pinpointing SbPPC2 and SbPPC3 monoubiquitination at the conserved Lys-630 and Lys-624 residues, respectively. Our results demonstrate that monoubiquitination and phosphorylation simultaneously occur in vivo with different PTPC isozymes during seed germination. In addition, we show that PTPC monoubiquitination in germinating sorghum seeds always increases at stage II (emergence of the radicle), is maintained during the aerobic period of rapid cell division and reserve mobilization, and remains relatively constant until stage IV–V when coleoptiles initiate the formation of the photosynthetic tissues. Key words: Germination, monoubiquitination, phosphorylation, phosphoenolpyruvate carboxylase, post-translational modifications, seed development, seeds, Sorghum bicolor, tissue-specific gene expression. © The Author 2016. Published by Oxford University Press on behalf of the Society for Experimental Biology. Abbreviations: BTPC, bacterial-type phosphoenolpyruvate carboxylase; COS, castor oil seed; DTT, dithiothreitol; Glc-6-P, glucose-6-phosphate; MALDI TOF MS, matrix-assisted laser desorption ionization time-of-flight mass spectrometry; p110 and p107, 110 and 107-kDa polypeptides, respectively; PEP, phosphoenolpyruvate; PEPC, PEP carboxylase; PPCK, PEPC-kinase; PTPC, plant-type phosphoenolpyruvate carboxylase; PTM, post-translational modification. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. 3524 | Ruiz-Ballesta etal. Introduction Phosphoenolpyruvate (PEP) carboxylase (PEPC; EC 4.1.1.31) is an important, tighly regulated, cytosolic metabolic enzyme that plays a pivotal photosynthetic role in primary CO2 fixation by C4 and Crassulacean acid metabolism plants, but also has a variety of additional functions, including roles in seed development and germination (O’Leary etal., 2011a). It catalyses the irreversible β-carboxylation of PEP to form oxaloacetate (OAA) and orthophosphate (Pi) (Chollet et al., 2006). The plant-type PEPC (PTPC) genes encode closely related 100–110 kDa polypeptides that typically exist as Class-1 PEPC tetramers that are post-translationally regulated by a combination of several allosteric effectors, including Glc-6-P (activation) and L-malate (inhibition), reversible phosphorylation (Nimmo, 2003; O’Leary et al., 2011a), and monoubiquitination (Uhrig etal., 2008, Shane etal., 2013; Ruiz-Ballesta etal., 2014). The metabolite control is integrated with the enzyme’s reversible phosphorylation at a highly conserved serine residue located near the N-terminus of its PTPC subunits (Chollet et al., 1996; Echevarría and Vidal, 2003; O’Leary et al., 2011a). Phosphorylation activates the enzyme by making it less sensitive to feedback inhibition by malate. Phosphorylation is catalysed by a PTPC-kinase (more commonly referred to as PEPC-kinase; PPCK), whereas dephosphorylation is catalysed by a protein phosphatase type-2A (Nimmo, 2000; Izui et al., 2004; O’Leary et al., 2011a). Both controls are interactive and thus allow a fine modulation of the enzyme’s kinetic properties as required by the physiological context (Echevarría and Vidal, 2003). Regulatory PTPC monoubiquitination has been demonstrated in the endosperm of germinating castor oil seeds (COS) (Uhrig etal., 2008), in developing proteoid roots of P-deficient harsh hakea (Shane etal., 2013), and in lily pollen (Igawa etal., 2010). Arecent study by our group has shown monoubiquitination of SbPPC3 in sorghum, a cereal with starch-storing seeds (Ruiz-Ballesta etal., 2014). Monoubiquitination is inhibitory as it results in increased Km (PEP) values and enhanced sensitivity to allosteric inhibitors (Uhrig etal., 2008; Shane etal., 2013; Ruiz-Ballesta etal., 2014). In COS, PTPC monoubiquitination and phosphorylation appear to be mutually exclusive, while PTPC polypeptides are alternatively phosphorylated or monoubiquitinated in developing proteoid roots of harsh hakea (Shane et al., 2013). In addition, the elimination of photosynthate supply caused by depodding in COS triggers in vivo dephosphorylation and subsequent partial monoubiquitination of PTPC polypeptides (Uhrig et al., 2008; O’Leary et al., 2011b). Previous studies have indicated that PTPC polypeptides become in vivo phosphorylated in de-embryonated sorghum seed during imbibition (i.e. on the basis of a significant increase in the enzyme’s IC50 for L-malate and in vivo 32P incorporation) (Nhiri etal., 2000). In other starch-storing cereal seeds such as barley and wheat, the phosphorylation of both PTPC polypeptides during germination has also been confirmed by immunoprecipitation, followed by SDSPAGE and autoradiography of the in vivo 32P-labeled PEPC (Osuna etal., 1999; Feria etal., 2008). As far as we know, no previous study has focused on how these post-translational modifications (PTMs) are combined in the different subunits and isoenzymes of a PEPC’s protein family. Deciphering how monoubiquitination and phosphorylation influence each PTPC isoenzyme during the life cycle of sorghum seeds was a major objective of the currentstudy. Bacterial-type PEPC (BTPC) genes encode larger 116– 118 kDa polypeptides exhibiting low (<40%) sequence identity with PTPCs. BTPC lacks the conserved N-terminal seryl phosphorylation motif characteristic of PTPCs, and in vascular plants it appears to only exist as catalytic and regulatory subunits of novel Class-2 PEPC heteromeric complexes composed of tightly associated PTPC and BTPC subunits (Igawa etal., 2010; O’Leary etal., 2011a; Park etal., 2012). Using western blot analysis and specific anti-BTPC-IgGs in crude extract of barley seeds, Class-2 PEPC complexes have not been identified (Feria etal., 2008). This is an interesting finding as the BTPC and Class-2 PEPC appear to play an important role during COS development (Uhrig etal., 2008). Assessing, by MS/MS and qRT-PCR, the possible occurrence of BTPC (SbPPC6) and/or Class-2 PEPC during the course of the life cycle of sorghum seeds was also an important objective of thiswork. Sorghum (Sorghum bicolor L.Moench) is the world’s fifth most important cereal crop (www.fao.org) and provides food, feed, fiber, fuel, and chemical/biofuel feedstocks across a range of environments and production systems (Kresovich et al., 2005). Sorghum encodes five PTPC genes (Paterson etal., 2009): Sb02g021090 and Sb04g008720 encode housekeeping and inducible SbPPC2 and SbPPC3, respectively (Lepiniec etal., 1993), whilst Sb07g014960 and Sb03g035090 encode PTPC isoenzymes SbPPC4 and SbPPC5 but these have unknown functions. All of these are C3-type PTPCs. Sb10g021330 encodes C4 photosynthetic SbPPC1 subunits that are highly expressed in sorghum leaves. In addition, Sb03g008410 (SbPPC6) encodes a single BTPC (Table1). In this study, the biochemical and immunological properties of PTPCs during development and germination of starch-storing sorghum seeds were further characterized. We also elucidated the transcript abundance of PTPC genes, and determined subunit composition, relative abundance, and PTMs of each PTPC isoenzyme during germination. As far as we know, this represents the most detailed study to date that examines both phosphorylation and monoubiquitination of a PTPC protein family in any plant species. Furthermore, we show that enhanced PTPC monoubiquitination always accompanies radical emergence during seed germination, and that both processes are accelerated in parallel under aerobic conditions or elevated temperature. Materials and methods Plant material Mature sorghum [Sorghum bicolor (L.) Moench, var. PR87G57 and var. PR88Y20, Pioneer Hi-Bred, Spain] seeds were sterilized and germinated for up to 96 h at 25 or 35ºC as previously reported (Nhiri PTMs of sorghum seed PEPC isozymes | 3525 etal., 2000). For immunological studies or to analyze PEPC activity, the aleurone layers, starchy endosperm (aleurone-endosperm), and embryos were carefully dissected at the indicated times post-imbibi- tion. Sterilized sorghum seeds were also submerged in distilled water (hypoxic condition) or placed on moistened filter paper (normoxic condition) at 25 oC for 48 and 72 h. Harvested tissues were frozen in liquid nitrogen and stored at –80ºC until used. Developing seeds were obtained from sorghum plants cultivated in a greenhouse under a 14 h day (25ºC)/10-h night (18ºC) cycle. Seeds were harvested at various stages of development, ground to a powder under liquid nitrogen, lyophilized, and stored at 4ºC for later analyses. Antibodies Polyclonal antibodies against native C4-photosynthetic SbPPC1 from sorghum leaves (anti-C4 PTPC) were prepared as described in Pacquit etal. (1995). Anti-C19 and anti-N24 were raised, respectively, against synthetic peptides corresponding to the C-terminal [(Y) EDTLILTMK GIAAGMQNTG] and the dephosphorylated N-terminal [ERHHSIDAQLRALAPGKVSEE24(YG)] ends of SbPPC1 (i.e., sorghum C4-photosynthetic PTPC) as previously described (Ruiz- Ballesta et al., 2014). Anti-COS PTPC was raised against native Class-1 PEPC purified from the endosperm of developing castor oil seeds as described in Tripodi etal. (2005). Immunohistochemistry Seeds germinated at 24 h post-imbibition were fixed in 4% (w/v) paraformaldehyde and in 0.25% (v/v) glutaraldehyde dehydrated in a graded series of aqueous ethanol solutions, and embedded in Paraplast Plus (Panreac) as described by González et al. (1998). Sections (15 µm thick) were prepared using a microtome (model RM2165, Leica) and placed on poly-L-Lys-coated microscope slides. After deparaffinizing in xylol and rehydrating in decreasing concentrations of ethanol, sections were blocked for 30 min in TBS buffer containing 3% (w/v) BSA. Then, 500µl of the same solution containing affinity-purified anti-C4 PTPC (3µg of protein) (Pacquit etal., 1995) or pre-immune serum was placed on the samples and incubated overnight at 4°C. Unbound antibodies were removed by three 10-min washes in TBS. Tissue sections were then incubated with alkaline phosphatase-conjugated goat anti-(rabbit IgG)-IgG (Bio-Rad) for 2 h at 37°C. The reaction of alkaline phosphatase was developed colorimetrically with a BCIP/NBT Liquid Substrate System (Sigma). Protein extraction Frozen powder (0.4 g) from whole developing or germinated sorghum seeds was ground in a chilled mortar with sand in 1 ml of an ice-cold buffer that contained 100 mM Tris-HCl (pH 7.5), 10 mM MgCl2, 1 mM EDTA, 20% (v/v) glycerol, 14 mM β-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, 10µg ml−1 chymostatin, 10µM leupeptin, 50 mM KF, 1 mM Na2MoO4, 1 mM Na3VO4−, and 50 nM microcystin-LR. Homogenates were centrifuged at 17 000 g for 7 min and the supernatants (0.5 ml) desalted through 3-ml Sephadex G-25 spin columns equilibrated in extraction buffer (lacking PVPP and PVP) prior to enzymatic analysis. Electrophoresis and immunoblotting SDS-PAGE, non-denaturing PAGE and in-gel PEPC activity staining, immunoblotting, and visualization of antigenic polypeptides using an alkaline-phosphatase-conjugated secondary antibody with chromogenic detection were performed as previously described (Rivoal etal., 2002; Tripodi etal., 2005). Alternatively, immunoreactive polypeptides were detected using a chemiluminescence system (SuperSignal West Pico Rabbit IgGs; Thermo Scientific) according to the manufacturer’s instructions in a Fujifilm LAS3000 mini-sys- tem. The immunoreactive polypeptides were quantified via analysis of the scanned blots using Multi-Gauge V3.0 software (Fujifilm). Determination of PEPC activity and protein concentration PEPC activity was assayed at 340 nm using a Molecular Devices Spectramax Kinetics Microplate reader. Optimized assay conditions were: 50 mM HEPES-KOH (pH 8.0), 2.5 mM PEP, 5 mM NaHCO3, 5 mM MgCl2, 0.15 mM NADH, 10% (v/v) glycerol, 1 mM DTT, and 5 units ml−1 porcine heart malate dehydrogenase (0.2 ml final volume). One unit of activity is defined as the amount of PEPC catalysing the production of 1µmol of oxaloacetate min−1 at 25ºC. Protein concentrations were determined by the Coomassie Blue G-250 dye-binding method using bovine γ-globulin as the standard (Tripodi etal., 2005). Immunoprecipitation and immunoaffinity chromatography Immunoprecipitation of PTPC polypeptides from clarified seed extracts was performed using anti-C4 PTPC as previously described (Osuna etal., 1999). For immunoaffinity chromatography, frozen whole 48-h germinated sorghum seeds (5 g) were homogenized (1:2.5; w/v) using a Polytron PT-3100 homogenizer in ice-cold buffer Athat contained 100 mM HEPES-KOH (pH 7.5), 1 mM EDTA, 1 mM EGTA, 15% (v/v) glycerol, 5 mM MgCl2,100 mM KCl, 10 mM NaCl, 25 mM NaF, 1mM Na3VO4−, 0.1% (w/v) polyvinylpolypyrrolidone, 0.1% (v/v) Triton X-100, 2 mM 2,2´-dipyridyl disulfide, and 10 µl ml−1 ProteCEASE-100 (G-BioSciences). Homogenates were centrifuged at 4ºC for 10 min at 48 500 g, and the supernatant fluid recentrifuged for 5 min and filtered through a layer of Miracloth (Calbiochem). Clarified extracts were pre-cleared at 25ºC by eluting at 0.5 ml min−1 through an AminoLink column (1 × 4 cm) that had been pre-saturated with 1 M Tris-HCl (pH 7.4). Unbound proteins were immediately absorbed at 1 ml min−1 onto an anti-COS PTPC Table1. The sorghum PEPC gene family. Names used in this work for the different PEPC genes, the equivalence with other names, and assigned functions described by Lepiniec et al. (1993, 1994), and orthologs established by Paterson etal. (2009). Sb, Sorghum bicolor; Sv, Sorghum vulgare; Zm, Zea mays; Os Oryza sativa. Name of gene Accession numbera Named by Lepiniec etal. (1993) and Ruiz-Ballesta etal. (2014) Known function Orthologs (Paterson etal., 2009) SbPPC1 Sb10g021330 CP46 C4 PTPC, photosynthetic C4 gene X63756 (SvPEPC1); NM001111948 (ZmPEPC1) SbPPC2 Sb02g021090 CP28 C3 PTPC, housekeeping gene X59925 (SvPEPC4); NM001111968 (ZmPEPC2) SbPPC3 Sb04g008720 CP21 C3 PTPC, root-inducible gene X65137 (SvPEPC3); NM001112033 (ZmPEPC3 SbPPC4 Sb07g014960 - C3 PTPC, unknown so far Os08g0366000 SbPPC5 Sb03g035090 - C3 PTPC, unknown so far Os01g0758300 SbPPC6 Sb03g008410 - BTPC, as Class-2 PEPC Os01g0110700 a NCBI database 3526 | Ruiz-Ballesta etal. immunoaffinity column (1 × 2 cm; prepared as described by Uhrig etal., 2008) that had been pre-equilibrated with PBS. Flow-through fractions were collected by eluting the column at 1 ml min−1 with PBS until A280 decreased to baseline, and bound proteins were eluted at 0.5 ml min−1 with 100 mM Gly-HCl (pH 2.8), and immediately neutralized (1-ml fractions collected into 0.1 ml of unbuffered 1 M Tris). A280 absorbing fractions were concentrated to ~1 mg ml−1 and analyzed for their polypeptide composition by SDS-PAGE and immunoblotting. Mass spectrometry In the initial MALDI-TOF MS analysis, PTPCs at the different stages of seed development and germination were extracted and purified by immunoprecipitation from the clarified extracts, followed by SDS-PAGE of the solubilized immunoprecipitates (Osuna etal., 1999). Protein-staining bands of interest were manually excised from micro-preparative gels using biopsy punches, then reduced, alkylated, and digested with trypsin according to Sechi and Chait (1998). Briefly, gel bands were washed twice with water, shrunk for 15 min with 100% acetonitrile, and dried in a Savant SpeedVac for 30 min. Then the samples were reduced with 10 mM dithioerythritol in 25 mM ammonium bicarbonate (NH4HCO3) for 30 min at 56ºC and subsequently alkylated with 55 mM iodoacetamide in 25 mM NH4HCO3 for 15 min in the dark. Finally, samples were digested with 12.5 ng µl−1 sequencing grade trypsin (Roche Molecular Biochemicals) in 25 mM NH4HCO3 (pH 7.6) overnight at 37ºC. After digestion, 1µl of each sample was spotted onto a MALDI plate and allowed to air-dry at room temperature. Then, 0.4µl of 3 mg ml−1 α-cyano-4-hydroxy-cinnamic acid as the matrix (Sigma) in 50% acetonitrile was added onto the spots and they were again allowed to air-dry at room temperature. Sample analyses were performed on a 4800 Plus Proteomics Analyzer MALDI-TOF/TOF mass spectrometer (AB Sciex, Toronto, Canada). The instrument was operated in the positive reflector mode with an accelerating voltage of 20 000 V. Mass spectra were calibrated internally using the peptides of trypsin autolysis. Mass spectrometric measurements yielded a list of monoisotopic masses of the observed peptides at a signal-to-noise ratio greater than 10, which were used for peptide mapping. Peptides of interest were further analyzed by MS/MS in order to assess their sequences. Suitable precursor ions were selected from the MS spectra and broken down using collision induced dissociation (CID) with atmospheric air as the collision gas, at a collision energy of 1 kV and in the positive ion reflector mode. The precursor ions were selected within the mass windows of ±4 Da. The nonredundant NCBI database (downloaded on 24/2/2011; 13 135 398 sequences) and in-house databases with the sorghum entries from NCBInr (71 862 sequences) and the sorghum PEPC sequence only were used for protein identification, and the mass spectrometric data were searched using the MASCOT 2.3 server through the Global Protein Server v3.6 (AB Sciex). The search parameters were set up to no limitation in taxonomy, trypsin as the cleavage enzyme, carbamidomethylation on cysteine as the fixed modification, and the oxidation of methionine and ubiquitination (Gly-Gly adducts) of lysine, phosphorylation of serine or threonine as variable modifications. Mass tolerances of single MS and MS/MS were set up to 80 ppm and 0.3 Da, respectively. For trypsin digestion, one missed cleavage was allowed. In such analyses, the probability scores greater than the score fixed by mascot were considered to be significant at a P-value of <0.05. De novo sequencing of the MS/MS spectra of the peptides was performed using the DeNovo tool software (AB Sciex), the retrieved peptide sequences were manually checked and validated. Asubsequent homology search of these sequences was obtained by BLAST (www.ncbi.nlm.nih.gov/BLAST). PTPC PTMs were analyzed by a high-accuracy and sensitivity LC MS/MS. Coomassie blue-stained polypeptides were excised from SDS-PAGE gels, in-gel reduced, alkylated, and digested with trypsin using standard protocols (Uhrig etal., 2008). Tryptic peptides were extracted using acetonitrile/0.1% trifluoroacetic acid (TFA) (v/v, 60:40), and dried using a CentriVap refrigerated centrifugal concentrator (Labconco Corp.). Peptides were reconstituted in 4µl of 0.1% (v/v) formic acid (FA) and identified using two instrumentation methods: nanoAcquity ultra performance LC systems (Waters, Milford, MA) coupled with a TripleTOF 5600+ quadrupole time-of-flight mass spectrometer (AB Sciex, Concord, ON, Canada) at low-energy CID; and an Orbitrap Fusion Tribrid MS system (Thermo Fisher Scientific Inc.Watham, MA) in higher energy collision induced dissociation (HCD) mode. The sample was trapped by a 2G-V/MT Trap symmetry C18 column (5μm particles, 180μm id × 20 mm length) at a flow rate of 5μl min–1, and separated on a BEH130 C18 analytical column (1.7μm particles, 100µm id ×100 mm length) at 300 nl min−1 for 60 min. The mobile phase was set up to a linear gradient from 5–30% solvent B (0.1% FA in acetonitrile), followed by 85% solvent B over 10 min for peptide elution. TripleTOF MS and MS/MS scans were acquired with a high resolution of 30.000, and the m/z regions of the MS survey scan and MS/MS were selected from m/z 350–1600 and m/z 100–1250, respectively, and by data-dependent scanning of the top twenty ions at multiply charged states of 2+, 3+ and 4+. Dynamic exclusion was set to a period of time of 30 s. For comparison purposes, similar analyses were performed on the gel-separated proteins using the Orbitrap Fusion Tribrid LC-MS system under high-accuracy MS survey scan at a resolution of 60.000 followed by sequential HCD scans at the twenty top ions. MS/MS data from the two instruments were searched against the viridiplantae (green plants) protein sequences in the NCBI database using the Mascot Server (version 2.4.0, Matrix Science, London, UK). The search parameters were restricted to tryptic peptides at a maximum of two missed cleavages. Cysteine carbamidomethylation was designated as a fixed modification, and deamidation of asparagine and glutamine, oxidation of methionine, phosphorylation of serine/threonine/tyrosine, ubiquitination of lysine were considered as variable modifications. Mass tolerances were set up to 30 ppm for TripleTOF 5600 MS ions and 0.05 Da for MS/MS fragment ions, 20 ppm for Orbitrap MS and 0.8 Da for the MS/MS fragments. Peptide assignments were filtered by an ion score cut off at 15, and the identified MS/MS spectra were also verified manually. RNA Extraction and cDNA synthesis Total RNA was isolated from 100 mg of frozen, powdered halfembryonated germinating seeds or whole developing seeds using the IQeasyTM Plant RNA Extraction (Intron Biotechnology). Extracted nucleic acids were DNase treated to eliminate genomic DNA. RNA was quantified using a NanoDrop 2000 spectrophotometer (Thermo Scientific). Reverse-transcription reactions were performed using 1µg of purified total RNA, 1µl ImProm-IITM Reverse Transcriptase (Promega) and a reaction buffer containing 0.5 mM dNTP, 6 mM MgCl2, 20 units recombinant RNasin® ribonuclease inhibitor, and 0.5µg oligo(dt)15. qRT-PCR Quantitative polymerase chain reaction (qRT-PCR) was performed in a final volume of 20µl consisting of 1µl of the cDNA, 10 µl of SensiFAST SYBR No-ROX kit (Roche) and 15 µM of the gene-specific primers pairs as follows: SbPPC3 (forward 5´– TGTTGAACAGTTTCTGGAACCTCTT-3´, reverse 5´-GCTTCA CAAGGGCAAGCCCAAAG-3´), SbPPC2 (forward 5´-CCGCCT CGCAACACCTGAAACA-3´, reverse 5´-ACCGGGAGGTGGAA CCGTGT-3´), SbPPC4 (forward 5´-TGAGCTTCGGGCACAAGC AGATG-3´, reverse 5´-GCTCCAAAGGCTCTAAGAACTGCT C-3´), and 18 S rRNA (forward 5´-GGGGAAACTTACCA GGTCCA-3´, reverse 5´-GGATGGCTCCGCATAGCTA-3´). PCR was conducted on the MiniOpticonTM Real-Time PCR Detection System (Biorad), and the threshold cycles (Ct) were determined using Opticon MonitorTM analysis software for all treatments. To PTMs of sorghum seed PEPC isozymes | 3527 normalize the obtained values, 18S ribosomal RNA was used as internal control in each sample. This gene displays a steady RNA level across the experimental conditions. Statistical analysis Statistical analysis was performed using SigmaStat (Systat Software Inc., San José, CA, USA). Data were analyzed using the Student’s t-test or with the Mann-Whitney U test. Means were considered to be significantly different at P<0.05. Results and discussion PEPC activity, and PTPC integrity and cellular localization PEPC was characterized throughout the development, maturation, and germination of sorghum seeds, focusing on the enzyme’s activity, integrity, subunit composition, and immunolocalization. Stages of sorghum seed development and germination are described in Supplementary Fig. S1 (at JXB online); among these, stage IV of development marks the beginning of seed desiccation, whilst stage II of germination marks the protrusion of the radicle and the completion of germination (Bewley etal., 2013). During seed development, PEPC activity of desalted extracts increased by about 3-fold to reach a maximum level at stage II–III and then decreased (Fig.1). During germination, PEPC activity in the embryo increased to a maximum at 72 to 96 h post-imbibition, while in the aleurone/endosperm it did not change significantly and was 10–15-fold lower than that of embryos (Fig.1). In germinating COS, the p110 and p107 PTPC polypeptides appear to be in vivo truncated by 19 amino acids such that their N-terminal phosphorylation domain is absent (Uhrig et al., 2008). In contrast, non-monoubiquitinated p107 and monoubiquitinated p110 PTPCs from germinating sorghum seeds at 48 h post-imbibition were shown to be intact and to therefore contain the phosphorylation motif (Ruiz-Ballesta etal., 2014). In this study, we demonstrate the integrity of PTPC polypeptides throughout the life cycle of sorghum seeds since the anti-C19 and anti-N24 both crossreacted with the p110 and p107 (see Supplementary Fig. S2). PTPC was also immunolocalized in germinating sorghum seed tissues characterized by high metabolic activities (epicotyl, radicle, aleurone layer, scutellum, scutellum’s epithelium) (Fig.2). Immunohistochemistry studies have localized PTPC in the protein bodies of wheat grains, where it was suggested to contribute to amino acid and protein biosynthesis during grain development (Araus etal., 1993). PTPC has also been localized in tissues of developing and germinating wheat grains characterized by high metabolic activity (González etal., 1998). It is noteworthy that PEPC was detected as an abundant protein in the aleurone layer using immuno-locali- zation (Fig.2E). Consistent with this result, the role of PEPC in malate production by the barley aleurone layer has been described by Macnicol and Raymond, (1998). Collectively, these results indicated the stability and consistent functional relevance of PTPC throughout the life cycle of sorghum seeds. Expression of PEPC genes in developing and germinatingseeds Sorghum PEPC is encoded by a small multigene family consisting of six genes: SbPPC2, SbPPC3, SbPPC4, SbPPC5 (all C3 PTPCs), and SbPPC1 (C4 photosynthetic PTPC) encode closely related PTPCs, whilst SbPPC6 encodes the distantly related BTPC (Table 1) (Paterson etal., 2009). In this present study, we used qRT-PCR to analyze the pattern of their transcript abundance. SbPPC2, SbPPC3, and SbPPC4 transcripts were present in developing and germinating sorghum seeds (Fig. 3). Transcripts of SbPPC2, the housekeeping PTPC (Lepiniec etal., 1994), were detected during seed development and its abundance increased slightly during the maturation phase (Fig.3A). In the germinating seeds, SbPPC2 showed maximal transcript abundance at 24–48 h post-imbibition and then decreased (Fig.3B). By contrast, SbPPC3 transcript levels were maximum at the beginning of seed development, during the period of cellularization (Bewley et al., 2013); this was followed by a decrease of SbPPC3 transcripts until stage III and then an increase during the desiccation period with a peak at stage V (Fig.3C). SbPPC3 transcripts were also detected during germination with two important peaks at 24 and 72 h post-imbibition (Fig.3D). Finally, SbPPC4 transcripts were quite abundant at the beginning of development, suggesting a possible specific role for SbPPC4 during the phase of cellularization (Fig.3E). Cellularization is the massive production of new cells. Malate is a ubiquitous molecule that provides carbon skeletons and reducing power and is needed for the synthesis of storage end-products by developing seeds. SbPPC4 transcripts were also very abundant at 24–72 h post-imbibition in germinating seeds (Fig.3F). By contrast, transcripts of SbPPC5 and SbPPC6 (BTPC) were very low during sorghum seed development and germination and, as expected, no SbPPC1 transcripts were detected (results not shown). These results show a novel pattern of the steady-state of transcript abundance of PTPC isogenes during the life cycle of sorghum seeds, with an important presence of SbPPC3 and also an important contribution of the SbPPC4 isogene to the cellularization stage of development (stage I) and during germination. Different PTPC isoenzymes co-exist in developing and germinatingseeds Peptide mass fingerprinting by MALDI-TOF MS of the p110 and p107 polypeptides immunoprecipitated using anti- C4 PTPC revealed that different PTPC isoenzymes co-exist in developing and germinating sorghum seeds (Table2; asterisks denote confirmation of the identification via LC MS/MS sequencing of tryptic peptides). SbPPC3 was the most abundant during the course of the life cycle of the seeds in both p107 and p110. SbPPC3 was highly abundant in the embryo (Table2) and less abundant in the aleurone/endosperm of the germinating seed; however, SbPPC3 was the only isoenzyme detected in this tissue (Table2). The housekeeping SbPPC2 (Lepiniec etal., 1993, 1994) was present in early development only as a p107 (i.e. probably non-monoubiquitinated) subunit 3528 | Ruiz-Ballesta etal. (Table2, stages Iand III) and was identified as a p110 and p107 subunit in stage VI (Table2). SbPPC2 was also present as a monoubiquitinated p110 subunit in the embryo of dry and 48-h post-imbibition seeds, but was not identified in the aleurone (Table2). Finally, the SbPPC4 isogene was shown to be translated into the corresponding p107 PTPC subunits in early development (Table2, stages Iand III), in agreement with the transcript profiling results of Fig. 3E. In addition, SbPPC4 was also detected in p110, indicating the probable monoubiquitination of this PTPC isoenzyme. Although the precise physiological function of SbPPC4 remains unclear, our results support a role for this isoenzyme during the early stages of seed development. During the phase of cellularization SbPPC4 is hypothesized to generate the C-skeletons needed as precursors for synthesis of new components such as amino acids/proteins, and membrane or storage lipids. Finally, SbPPC5 was detected only as p110 polypeptides in stage I(Table2). SbPPC6 polypeptides were never detected by MS/MS analysis, in agreement with the absence of SbPPC6 expression during sorghum seed development and germination. Nor was a high molecular mass Class-2 PEPC hetero-octamer evident when clarified extracts from different stages of development and germination were subjected to non-denaturing PAGE followed by in-gel PEPC activity staining (see Supplementary Fig. S3). However, tetrameric and dimeric Class-1 PEPC oligomers were observed, Fig.1. Time course of PEPC activity in developing and germinating sorghum seeds. PEPC activity was assayed at optimal pH (8.0), 2.5 mM PEP, and 30ºC in clarified extracts from: (A) whole seeds during development; and (B) embryos or (C) aleurone-endosperm of germinating seeds. (A) PEPC activity is expressed on a per seed basis (mU seeds−1) or on a protein basis (U mg−1 protein). Results are means ±SE of three independent experiments. Mean values that are not significantly different (P<0.05, Student’s t-test) are indicated with the same letters. Data for aleurone-endosperm are not significantly different. PTMs of sorghum seed PEPC isozymes | 3529 with the dimeric form being the most abundant in vitro (Supplementary Fig. S3A). This dimeric form was not altered by de-ubiquitination of the enzyme after incubation with 2µM of the catalytic subunit of ubiquitin specific protease-2 (USP-2c) (Supplementary Fig. S3B). A number of important points emerge from these results, as follows. (i) Multiple C3 PTPC isoenzymes are present during the development and the germination of sorghum seeds. SbPPC3 appears to have an essential role as it is the isoenzyme most consistently present during the course of the life cycle of the seeds , and was the only one detected in the aleurone/endosperm. (ii) The SbPPC4 isozyme appears to play a major role during the very early stage of seed development, known as the phase of cellularization (Bewley etal., 2013).(iii) BTPC is not significantly expressed throughout the life cycle of the seeds, which is an important difference relative to developing COS, a non-green oil seed, in which p118 BTPC polypeptides, and thus Class-2 PEPC complexes are highly expressed (O’Leary etal., 2011b). Post-translational modifications of the different Class-1 PEPC isoenzymes during germination In the endosperm of germinating COS, p110 PTPC subunits are monoubiquitinated but not phosphorylated (Uhrig etal., 2008; O’Leary etal., 2011b). On the other hand, simultaneous p110 monoubiquitination and p107 phosphorylation have been described as an alternative PTM pattern for Class-1 PEPC of immature proteoid (cluster) roots of harsh hakea (Shane et al., 2013). However, the coincident occurrence Fig.2. Immunological localization of PTPC in germinating sorghum seeds. Grains were imbibed for 24 h. Sections (15µm) were probed with: (A, B), preimmune serum, or (C–F) polyclonal anti-C4 PTPC (SbPPC1). Abbreviations: al, aleurone layer; ep, epicotyl; ept, epithelium; sc, scutellum; se, starchy endosperm; r, radicle; p, pericarp. 3530 | Ruiz-Ballesta etal. of both PTMs in sorghum seed p110 and p107 PTPC (i.e. SbPPC3) subunits has only been shown in vitro (Ruiz-Ballesta etal., 2014). In this study, we show both PTMs also occur in vivo with three of the four PTPC isoenzymes present in germinating sorghum seeds. To this end, PTPC polypeptides of clarified extracts from 48-h post-imbibition germinating seeds were isolated by immunoaffinity chromatography with anti-COS PEPC (see Supplementary Fig. S4A, B). Coomassie blue R-250-stained p110 and p107 were excised from SDS gels of the immunopurified PTPC and individually subjected to in-gel tryptic digestion and detailed analysis by LC MS/MS. AMascot database search of p110 and p107 datasets derived from nano HPLC TripleTOF MS of their tryptic peptides (Table3) revealed the presence of SbPPC2, SbPPC3, and SbPPC4 PTPCs in p110 and p107, as well as SbPPC5 in p107. Consistent with our earlier results (Ruiz- Ballesta et al., 2014), p107 N-terminal tryptic peptides of SbPPC3 containing non-phosphorylated Ser7 were identified by Orbitrap LC MS/MS. However, the monoubiquitinated SbPPC3 p110 subunits were also determined to be phosphorylated at Ser7 (Table3). Monoubiquitinated SbPPC2 and SbPPC4 (i.e. p110 subunits), and deubiquitinated SbPPC5 (i.e. p107 subunits) were also shown to be phosphorylated at their conserved N-terminal seryl phosphorylation sites; i.e. Ser13, Ser10, and Ser11 for SbPPC2, SbPPC4, and SbPPC5, respectively (Table3). To the best of our knowledge, this is the first evidence that in vivo phosphorylation of different PTPC isoenzymes can simultaneously occur in any plant tissue. MS/ MS also confirmed the monoubiquitination of SbPPC3 p110 subunits at Lys624 (Table3), as previously reported (Ruiz- Ballesta etal., 2014). In addition, TripleTOF MS/MS analysis of the quadruply charged SbPPC2 p110 peptide ion of m/z 455.7462 (LC retention time 16.61 min) and the triply charged ion of m/z 507.9268 (LC retention time 18.01 min), corresponding to residues 623–637 (VAKDFGVKLTMFHGR) and 626–637 (DFGVKLTMFHGR), showed that the masses of the C-terminal fragments up to y7 matched the predicted values, but the mass shift of the fragment y8 by 114 Da indicated a ubiquitination site localized at Lys-630, where a characteristic mass increment of 114 Da indicates a Gly-Gly motif attachment of ubiquitin at the side chain (Supplementary Fig. S5A, B). LC MS/MS measurements of the doubly and Fig.3. qRT-PCR analysis of PTPC transcript levels in developing and germinating sorghum seeds. To normalize the values, 18S RNA was used as internal control in each sample. Data were normalized to stage VI for developing seeds (A, C, E), or 14 h post-imbibition germinating seeds (B, D, F). Results are means ±SE of at least three independent experiments. Mean values that are not significantly different (P<0.05, Student’s t-test) are indicated with identical letters. PTMs of sorghum seed PEPC isozymes | 3531 triply charged ions of m/z 769.3827 and m/z 513.2607 (LC retention time 23.22 min) using a high-resolution Orbitrap Fusion Tribrid instrument identified the ubiquitination of the oxidized peptide at residues 626–637 (Supplementary Fig. S5C, D). Under high-resolution HCD fragmentation, the mass shift of 114 Da on both the N-terminal b5 ion and the C-terminal y8 ion unambiguously confirmed Lys-630 as p110’s monoubiquitination site in SbPPC2. This residue precisely aligns with the Lys-624 and Lys-628 (a highly conserved PTPC residue that is immediately proximal to catalytic residues involved in PEP binding) monoubiquitination sites, respectively determined for p110 subunits of sorghum seed SbPPC3 PTPC (Table3 and Ruiz-Ballesta etal., 2014) and PTPC (RcPPC3) from germinated COS (Uhrig etal., 2008). Collectively, these results establish for the first time several important points. (i) Different PTPC isoenzymes are co-expressed during sorghum seed germination and all of them except the p107 subunits of SbPPC3 are phosphorylated during germination. (ii) Monoubiquitination of the corresponding conserved Lys was confirmed by MS/MS analysis for SbPPC3, and determined in this study for SbPPC2, demonstrating that this PTM occurs with multiple PTPC isoenzymes at a conserved lysine residue within a single plant species. (iii) Monoubiquitination and phosphorylation can simultaneously occur on the same PTPC polypeptide and are therefore not mutually exclusive PTMs, as appears to be the case with PTPCs from developing proteoid roots of harsh hakea (Shane etal., 2013), as well as various green and non-green tissues of the castor oil plant (Uhrig etal., 2008; O’Leary etal., 2011b). Is PTPC monoubiquitination linked to specific stages and/or environmental conditions during sorghum seed germination? Because of its recent discovery (Uhrig etal., 2008), there are few studies about the influence of environmental factors or biotic/abiotic stresses on PTPC monoubiquitination during seed development and germination. Elimination of photosynthate translocation (a type of abiotic stress) from the shoot to the seed by depodding (fruit excision) of developing COS triggered rapid in vivo p107 dephosphorylation, followed by monoubiquitination of 50% of the p107 subunits to form p110 characteristic of germinating COS (O’Leary etal., 2011b). Sorghum is a cereal cultivated in warm environments and seeds germinate well within a wide range of temperature from 25–35ºC, with an optimum of over 30ºC (Kanemasu etal., 1975). This allowed us to correlate the effect of temperature with germination and the pattern of PTPC monoubiquitination (i.e. appearance of p110 polypeptides on PTPC immunoblots). To this end, seeds from two sorghum varieties PR88Y20 Table2. Identification of p107 and p110 PTPC isoforms from developing and germinating sorghum seeds. Proteins were identified via MALDI-TOF/TOF MS peptide mass fingerprinting of tryptic peptides derived from immunoprecipitated samples of representative stages. Asterisks indicate identification via MS/MS sequencing of tryptic peptides and BLAST analysis. DPA, days post-anthesis. STAGE Subunit Protein name Accession numberaScore Sequence coverage (%) No. of Matching Peptides I (7–12 DPA) p110 PEPC SbPPC3* gi|241931686 248 35 31 PEPC SbPPC4* gi|241940554 118 24 22 PEPC SbPPC5 gi|241930496 98 22 21 p107 PEPC SbPPC3 gi|241931686 325 45 39 PEPC SbPPC4 gi|241940554 114 26 21 PEPC SbPPC2 gi|241925556 107 23 20 III (16–20 DPA) p110 PEPC SbPPC3 gi|241931686 103 24 21 p107 PEPC SbPPC3 gi|241931686 206 33 29 PEPC SbPPC4 gi|241940554 66 19 16 PEPC SbPPC2 gi|241925556 61 17 15 VI (31–40 DPA) p110 PEPC SbPPC3 gi|241931686 60 10 13 PEPC SbPPC2* gi|241925556 36 9 8 p107 PEPC SbPPC3 gi|241931686 121 15 18 PEPC SbPPC2* gi|241925556 76 8 10 Dryseed Aleurone/endosperm p110 PEPC SbPPC3 gi|241931686 30 13 9 p107 PEPC SbPPC3 gi|241931686 72 20 15 Dryseed Embryo p110 PEPC SbPPC3* gi|241931686 310 45 39 PEPC SbPPC2 gi|241925556 91 21 18 p107 PEPCSb PPC3 gi|241931686 283 42 48 PEPC SbPPC2 gi|241925556 63 16 14 48 h post-imbibition Aleurone/endosperm p110 PEPC SbPPC3* gi|241931686 95 24 16 p107 PEPC SbPPC3 gi|241931686 67 16 15 48 h post-imbibition Embryo p110 PEPC SbPPC3 gi|241931686 305 44 38 PEPC SbPPC2 gi|241925556 54 15 14 p107 PEPC SbPPC3* gi|241931686 164 29 38 PEPC SbPPC2 gi|241925556 55 15 13 aNCBI database