肌醇磷酸合成酶

Planta (2010) 231:1211–1227 DOI 10.1007/s00425-010-1127-8 ORIGINAL ARTICLE IdentiWcation and organization of chloroplastic and cytosolic L-myo-inositol 1-phosphate synthase coding gene(s) in Oryza sativa: comparison with the wild halophytic rice, Porteresia coarctata Sudipta Ray · Barunava Patra · Aparajita Das-Chatterjee · Arnab Ganguli · Arun Lahiri Majumder Received: 28 November 2009 / Accepted: 11 February 2010 / Published online: 6 March 2010 © Springer-Verlag 2010 Abstract The gene coding for rice chloroplastic L-myo- inositol-1-phosphate synthase (MIPS; EC 5.5.1.4) has been identiWed by matrix-assisted laser desorption time-of-Xight mass spectrometry analysis of the puriWed and immunolog- ically cross-reactive »60 kDa chloroplastic protein follow- ing two-dimensional polyacrylamide gel electrophoresis, which exhibited sequence identity with the cytosolic MIPS coded by OsINO1-1 gene. A possible chloroplastic transit peptide sequence was identiWed upstream of the OsINO1-1 gene upon analysis of rice genome. RT-PCR and confocal microscope studies conWrmed transcription, eVective trans- lation and its functioning as a chloroplast transit peptide. Bioinformatic analysis mapped the chloroplastic MIPS (OsINO1-1) gene on chromosome 3, and a second MIPS gene (OsINO1-2) on chromosome 10 which lacks conven- tional chloroplast transit peptide sequence as in OsINO1-1. Two new PcINO1 genes, with characteristic promoter activ- ity and upstream cis-elements were identiWed and cloned, but whether these proteins can be translocated to the chloroplast or not is yet to be ascertained. Electrophoretic mobility shift assay carried out with nuclear extract of Porteresia coarctata leaves grown under both control and stressed condition shows binding of nuclear proteins with the upstream ele- ments. Nucleotide divergence among the diVerent Oryza and Porteresia INO1 genes were calculated and compared. Keywords Chloroplastic L-myo-inositol-1-phosphate synthase · MALDI-TOF MS · Transit peptide · GFP Xuorescence · Genomewalking · Porteresia coarctata Abbreviations MIPS L-myo-Inositol-1-phosphate synthase MALDI-TOF MS Matrix-assisted laser desorption ioniza- tion time-of-Xight mass spectrometry CaMV CauliXower mosaic virus GFP Green Xuorescent protein 2D PAGE Two-dimensional polyacrylamide gel electrophoresis ORF Open reading frame GUS �-Glucuronidase Introduction L-myo-Inositol-1-phosphate synthase (MIPS, EC 5.5.1.4) catalyzes the enzymatic conversion from glucose 6-phosphate S. Ray · B. Patra · A. Das-Chatterjee · A. Ganguli · A. L. Majumder (&) Plant Molecular and Cellular Genetics, Bose Institute (Centenary Campus), P1/12 CIT Scheme VII M, Kolkata 700054, India e-mail: lahiri@bic.boseinst.ernet.in Present Address: S. Ray Department of Botany, Centre of Advanced Studies, University of Calcutta, 35, Ballygunge Circular Road, Kolkata 700019, India Present Address: B. Patra Kentucky Tobacco Research and Development Centre, University of Kentucky, Lexington, KY, USA Present Address: A. Das-Chatterjee Molecular and Cell Biology Department, Goldman School of Dental Medicine, Boston University Medical Center, Boston, MA 02118, USA Present Address: A. Ganguli Chembiotek, Block BN, Plot 7, Sector V, Salt lake Electronic Complex, Kolkata 700091, India 123 1212 Planta (2010) 231:1211–1227 to L-myo-inositol 1-phosphate, the immediate precursor of free inositol. The reaction consists of a coupled oxidation and reduction catalyzed by L-myo-inositol-1-phosphate synthase (MIPS) (Sherman et al. 1969; Loewus and Loewus 1983). The occurrences of MIPS in diverse organ- isms suggest that the pathway is of ancient evolutionary origin (Majumder et al. 1997, 2003; Bachhawat and Mande 2000). MIPS was shown to be present both in the cytosolic frac- tion and in organelles (Lackey et al. 2003). The organellar form was earlier shown to be present in the chloroplasts of Pea (ImhoV and Bourdu 1973), Vigna radiata, Euglena gracilis (Adhikari et al. 1987), and Oryza sativa (Ray Chaudhuri et al. 1997; Hait et al. 2002). The chloroplastic form of the enzyme was found to be the similar to the cyto- solic form with respect to biochemical and immunological properties and shown to be regulated by light and salt (Ray Chaudhuri and Majumder 1996; Ray Chaudhuri et al. 1997; Hait et al. 2002). The chloroplastic MIPS protein/gene may have complex evolutionary history as MIPS has been reported from higher plants and also from cyanobacteria, e.g. Spirulina sp, the pre- sumed chloroplast progenitor (Ray Chaudhuri et al. 1997). The native protein from Spirulina showed similar enzymatic and immunological properties to other MIPS proteins. However, no unique structural gene coding for MIPS from Spirulina or any other cyanobacterium was reported until, two unidentiWed open reading frames (ORFs), sll1722 and sll1981 from Synechocystis sp PCC 6803 were reported as MIPS coding genes (Chatterjee et al. 2004, 2006). Out of the 65 genes present in the chloroplast genome of O. sativa (Accession no. NC_008155), no coding gene for chloroplastic MIPS could be determined by a comprehen- sive bioinformatics analysis suggesting that the chloroplas- tic MIPS might be nuclear encoded. Moreover, the MIPS coding ORFs sll1722 and sll1981 from Synechocystis sp PCC 6803, a member of the presumptive progenitor group of higher plant chloroplasts, showed no similarity with any of the genes encoded by the chloroplast genome. Hence it was presumed that the chloroplastic MIPS protein might be coded by the nuclear genome and post-translationally tar- geted to the chloroplast by a chloroplast-speciWc transit peptide (Jarvis and Soll 2001). Though such transit pep- tides possess characteristic features, there is so far no known motif of these and the degeneracy of the sequence precludes a PCR-based approach to identify chloroplast- localized proteins, hence a proteomic approach becomes the method of choice for identiWcation of the chloroplastic MIPS coding gene. In the present work, MALDI-TOF analysis of the puri- Wed chloroplastic MIPS from rice was carried out and MIPS gene(s) were analyzed for target signals. Reverse transcriptase (RT) PCR assay and green Xuorescent protein (GFP) tagged transit peptides were used to validate the transcription and translation of putative transit peptide. In Porteresia coarctata (Roxb.) Tateoka, the halophytic wild rice, a novel salt-tolerant L-myo-inositol-1-phosphate synthase (PcINO1) was reported earlier from the present laboratory (Majee et al. 2004). In order to Wnd a similar chloroplastic MIPS coding gene from P. coarctata, upstream sequence for MIPS gene(s) from P. coarctata were isolated. Thus two new INO1 genes from this plant were identiWed although no gene coding for the chloroplas- tic MIPS in Porteresia, as in Oryza, could be identiWed. In addition, such investigation showed the presence of regula- tory elements in the upstream of INO1 genes and also showed characteristic promoter activity. The cloning and sequencing of the diVerent MIPS coding genes in O. sativa and P. coarctata enabled us to draw the phylogenetic rela- tionship between the two species in terms of their INO1 gene organization. Materials and methods Plant material Rice seeds were washed with distilled water, spread on water-soaked cotton bed and allowed to germinate in dark- ness for 3 days at 30°C. The seedlings were grown for 4 days in plant growth chamber under 16/8 h photoperiod at 30°C. Tobacco (Nicotiana tabacum) plants were also grown in a plant growth chamber at 22°C under 16/8 h pho- toperiod. The wild-rice plants, P. coarctata, were collected from the saline river banks of the Sunderbans. Isolation and puriWcation of chloroplastic MIPS Isolation and puriWcation of the chloroplastic MIPS was carried out following essentially the procedure outlined by Ray Chaudhuri et al. (1997). Chloroplasts were isolated from 7-day-old rice leaves following the method of Rath- nam and Edwards (1976) with modiWcations. Leaves were homogenized with Wve times their fresh weight in ice cold chloroplast isolation buVer (0.33 M sucrose, 50 mM Tris– HCl, pH 7.5, 2 mM EDTA, 1 mM each of MgCl2 and MnCl2) in a prechilled blender. The homogenate was sieved through cheese cloth and centrifuged for 10 min at 4°C. The supernatant was centrifuged for 30 min at 4°C to obtain crude chloroplast pellet. The chloroplast pellet was resuspended in chloroplast isolation buVer and puriWed through Percoll gradient according to Guillot-Salomon et al. (1987). Plastids were homogenized in three times the volume of homogenization buVer. The MIPS proteins from rice chlo- roplasts were partially puriWed using a two-step puriWcation 123 Planta (2010) 231:1211–1227 1213 process by gel Wltration chromatography using Superose12 followed by DEAE Sephacel (Amersham Pharmacia, Upp- sala, Sweden) column chromatography (Ray Chaudhuri et al. 1997). MIPS assay was performed as described earlier (Ray Chaudhuri et al. 1997) and the enzymatically active fractions were pooled and loaded on 2D PAGE. Two-dimensional SDS-PAGE and western blot analysis Analytical SDS-PAGE of partially puriWed chloroplastic MIPS was performed according to Laemmli (1970). 2D- PAGE was performed according to manufacturer’s protocol in the Mini-PROTEAN 2D apparatus (Bio-Rad, USA). Pro- tein spots were detected by Coomassie brilliant blue R-250 according to Laemmli (1970). Western blot was performed as described earlier (Ray Chaudhuri et al. 1997; Hait et al. 2002). After immunodetection using anti-rice-MIPS anti- body, corresponding protein spot was cut out of the SDS gel for proteolytic digestion. MALDI-TOF MS analysis of the 60 kDa form of plastidial MIPS The in-gel digestion of the proteins was performed accord- ing to the method described by Dihazi et al. (2001). The extracted peptides were dried in vacuum centrifuge. Desalt- ing of the samples was carried out with ZipTip C18 (Milli- pore, Schwalbach, Germany). The pellet of the protein digest was dissolved in 0.1% triXuoroacetic acid with 50% acetonitrile to a Wnal concentration of 20–40 ng/�l. 2 �l of the solution was mixed with the same volume of matrix solution. 1 �l aliquot of sample-matrix solution was depos- ited onto a stainless steel 384 sample target and allowed to dry at room temperature, resulting in a uniform layer of Wne granular matrix crystals. All mass spectra were obtained on a Bruker BiXex III mass spectrometer (Bruker Daltonik GmbH, Bremen, Germany) (Clauser et al. 1999) equipped with a nitrogen laser (337 nm, 5 ns pulse) in the reXector-mode operation. The instrument was calibrated with signals of the positive MH+ ion of angiotensin II (mass 1,045.535 Da), adrenocorticotropic hormone (mass 2,434.191 Da) and somatostatin (mass 3,147 Da). Reverse transcriptase (RT)-PCR for isolation of the transit peptide Total RNA was isolated from leaves of O. sativa by Trizol method (Invitrogen Carlsbad, CA) according to manufac- turer’s protocol. Total RNA treated with DNaseI was used for Wrst strand cDNA synthesis using Superscript II reverse transcriptase (Stratagene), following the manufacturer’s instruction. cDNA thus synthesized was used to amplify the transit peptide sequence predicted by TargetP (Nielsen et al. 1997). Forward primer (5�atgatactcctcgcctcgccg cttgcctc3�) was designed from the 5� end of the putative transit peptide. The reverse primer (5�tctccacgcggaagctctc- gat3�) was designed from complementary chain within the MIPS coding sequence (Wrst exon). A 294-bp fragment was ampliWed and cloned in pGEM-T easy vector (Promega, USA) according to manufacturer’s instruction. Cloning of MIPS gene(s) PcINO1-1 and PcINO1-2 from P. coarctata Genomic DNA was isolated from P. coarctata following the method of Dellaporta et al. (1983). Genome-walking PCR was performed using a Genome Walker kit (BD Bio- sciences) according to the manufacturer’s manual. The primary PCR was conducted using an adapter and gene- speciWc primer (5�taccggtactccgactcgatctc3�). Secondary PCR was done using the nested adapter and gene-speciWc primer (5�tctccacgcggaagctctcgat3�) that was designed complementary to the 5� region of MIPS cDNA. The DNA ampliWed by PCR was cloned into TOPO XL PCR cloning vector (Invitrogen Carlsbad, CA) and sequenced. To deter- mine the PcINO1-1 gene sequence, forward primer (5�tccgatggaagggtaattcggatttttcc3�) based on the diVerence between the upstream sequences obtained by genome walk- ing and reverse primer (5�caacgcaggccctcatgatgttctcga3�) designed from the 3� end of the MIPS cDNA sequence was used to amplify gene sequence of PcINO1-1, while forward and reverse primers with sequence (5�tcctctgtagcgcaggc- tatcgac3�) and (5�gctcgagcttgttactccaggatcatgtt3�), respec- tively was used to amplify the PcINO1-2 gene. A nested PCR reaction was carried out with forward primer (5�acata tgttcatcgagagcttccg3�) and reverse primer (5�cttgttactcc aggatcatgtt3�) designed from 5� and 3� end of the MIPS ORF, ended up with a 1.8 kb fragment. Based on this PCR fragment, forward primer (5�ttgtgttcgggggctgggacattag3�) along with the same reverse primer (5�cttgttactccaggatcatgtt3�) was used to amplify the remaining part of the gene. Thus, full length PcINO1-2 gene sequence was simulated from the two partial clones obtained. PCR products were cloned into pGEM-T easy vector (Promega, USA) according to manufacturer’s instruction and sequenced. Bioinformatic analysis of DNA sequences The upstream sequences obtained for MIPS coding genes in Oryza and Porteresia were subjected to comprehensive bioinformatic analysis. The programs Translate tool (http:// au.expasy.org/tools/dna.html), PSORT (http://psort.ims. u-tokyo.ac.jp/), ChloroP (http://www.cbs.dtu.dk/services/ ChloroP/), TargetP (http://www.cbs.dtu.dk/services/TargetP/), SignalP (http://www.cbs.dtu.dk/services/SignalP/) and Predo- tar (http://urgi.versailles.inra.fr/predotar/predotar.html) were 123 1214 Planta (2010) 231:1211–1227 used for predicting organelle targeting and transit peptide proteolysis sites. The NCBI blastN (http://www.ncbi. nlm.nih.gov/BLAST/) program and Mascot (http:// www.matrixscience.com) were used to identify the chloro- plast gene in O. sativa. The sequences obtained upstream of the MIPS gene from Porteresia were compared with each other by bl2seq program of NCBI (Altschul et al. 1990). Plant-speciWc promoter elements in the upstream sequence of the MIPS coding gene(s) from Porteresia and Oryza were identiWed using both PLACE database (http:// www.dna.affrc.go.jp/htdocs/PLACE/) (Higo et al. 1999) and PlantCARE (http://bioinformatics.psb.ugent.be/webt- ools/plantcare/html/). Predictions of eukaryotic promoter and transcription initiation sites were performed using the TSSP/Prediction of PLANT Promoters (Using RegSite Plant DB, Softberry Inc.). The structure of MIPS gene was analyzed using genewise (http://www.ebi.ac.uk/Wise2/ index.html) that compares a protein sequence to a genomic DNA sequence. Making chimeric gene cassette and tobacco transformation The upstream sequence of MIPS gene from Porteresia (PcINO1-1) was PCR ampliWed by end-to-end primers (5�gatggatccatgaggttaacacc3�) and (5�ctcgatgaccatggcgcc cgc3�), while PcINO1-2 was PCR ampliWed by (5�gatggatc ccgacggcccgggctggtaaaatag3�) and (5�gacccatggatcgcccgc ttcgtcgatagcc3�) with BamHI and NcoI sites and cloned in promoter reporter vector pBT2GUS (obtained as a gift). The cauliXower mosaic virus (CaMV) 35S promoter along with Nos terminator was cloned in HindIII–BamHI and EcoRI-SacI sites, respectively in pCAMBIA1301. This construct has been used to transform tobacco plants. The modiWed GFP sequence (mgfp5; Accession no. U87973.1; Siemering et al. 1996) was cloned at SmaI site of pCAMBIA1301 harboring CaMV35S promoter and Nos terminator. The putative transit peptide sequence, cloned in pGEM-T easy vector, was PCR ampliWed by primers (5�tct agaatgatactcctcgcctcgccgcttgcctc3�) and (5�ggatccactcgcccc gctccgcggaagctagc3�) with XbaI and BamHI restriction sites at 5� and 3� ends, respectively and subsequently cloned at XbaI and BamHI sites of pCAMBIA1301. pCAMBIA1301- Tp-mgfp5 construct contains a hygromycin resistance gene for selection of transformed shoots, a chloroplast transit peptide sequence and a modiWed GFP gene for monitoring the in vivo expression of GFP in the chloroplasts. The con- struct was mobilized into Agrobacterium tumefaciens LBA 4404 by freeze–thaw method (Nishiguchi et al. 1987). Tobacco leaf discs were infected with A. tumefaciens containing the above construct. After 3 days of co-cultiva- tion the leaf discs were transferred to the regeneration medium supplemented with cefotaxim (250 mg/l) and hygromycin (15 mg/l). Cultures were maintained at 26°C and under 16/8 h photoperiod. The transformed shootlets were either cryostat-sectioned or used to isolate the protop- lasts. Leaf sections or protoplasts were observed in LSM 510 Meta confocal microscope (Carl Zeiss, MicroImaging Inc, USA). Protoplast isolation and transformation Transformed tobacco plants were propagated on hormone- free MS medium (Murashige and Skoog 1962). Leaves were sliced with razor blades and Xoated for 15 h on 1% cellulase R10, 0.25% macerozyme R10 in solution contain- ing 0.4 M mannitol, 15 mM CaCl2·2H2O (pH 5.6). Intact protoplasts were puriWed by suspending in 20% sucrose solution and separated by centrifugation and washed with 0.4 M mannitol, 15 mM CaCl2·2H2O (pH 5.6). Protoplasts were resuspended in solution containing 5 mM MES–KOH (pH 5.6), 0.4 M mannitol, 15 mM MgCl2. Three hundred microliters of protoplast containing approximately 100,000 protoplasts were incubated with 10 �g of puriWed DNA at room temperature for 30 min. 300 �l of PEG (mol. wt 6,000) was added slowly and mixed thoroughly, after 45 min of incubation at room tem- perature. 1.4 ml of solution containing 5 mM MES–KOH (pH 5.6), 0.4 M mannitol, 15 mM MgCl2 was added to it and kept at 25°C for 18 h. Determination of �-glucuronidase (GUS) activity Protoplasts were taken in 1.5 ml Eppendorf tube and to it 200 �l GUS extraction buVer (50 mM NaHPO4 pH 7.0, 10 mM 2-mercaptoethanol, 10 mM Na2EDTA, 0.1% sodium lauryl sarcosine, 0.1% Triton X-100) was added. Protein was extracted by freeze–thaw method followed by centrifugation for 10 min 14,000 rpm at 4°C. Protein con- centration was determined with the Bradford reagent (Bio- Rad, USA). Fifty microliters of supernatant was mixed with equal amount of GUS assay solution (2 mM 4-methylumbellife- ryl-D-glucuronide in extraction buVer) and incubated at 37°C for 1, 3 and 24 h. On completion of the incubation period 900 �l of stop buVer (0.2 M sodium carbonate) was added to each tube and GUS activity was determined by measuring the MUG degradation by UV–VIS spectropho- tometer at 316 nm. GUS activity was also determined by measuring the amount of MU formed. Amount of MU was determined by Xuorometric method (excitation wavelength 365 nm and emission wavelength 455 nm). Electrophoretic mobility shift assay The 156 bp fragment was PCR ampliWed from the upstream sequence of the PcINO1-1 gene by forward primer (5�catgg 123 Planta (2010) 231:1211–1227 1215 atccctgctcccacc3�) and reverse primer (5�ggagatatcggacttat agc3�). The fragment was endlabeled with �32P dATP with T4 Polynucleotide kinase (Promega). Nuclear extracts from leaves were prepared from Port- eresia grown under control condition and from leaves treated with 100 mM salt for 5 days by the following proce- dure. All steps were performed at 4°C as described by Roy Choudhury et al. (2008). BrieXy, »100 g of fresh tissue was homogenized to a Wne powder in liquid nitrogen. After thawing for 45 min at 4°C, the powder was homogenized in 300 ml of ice cold nuclei isolation buVer (25 mM Tris–Cl pH 7.5, 1 M sucrose, 10 mM MgCl2, and 10 mM 2-mercap- toethanol) and kept for 1 h with mild shaking at regular intervals. It was then Wltered through muslin cloth presoa