Plant materials and growth conditions
Seeds from the JIC Pisum germplasm collection and TILLING lines (http://www-urgv.versailles.inra.fr/tilling/index.htm; Dalmais et al. 2008) were grown in JIC glasshouses for plant material and trait analysis, with supplementary heat and light in winter months. A set of diverse lines, comprising Pisum germplasm accessions and cultivars (19 lines: JI 2822, JI 185, JI 73, JI 1294, JI 813, JI 2775, JI 281, JI 399, JI 3129, JI 1201, JI 1194, JI 2202, JI 15, cv. Cameor, cv. Brutus, cv. Birte, cv. Kahuna, cv. Princess and cv. Enigma), was used to screen for genetic diversity. Nicotiana benthamiana plants for transient expression assays were grown in a JIC containment glasshouse for 5–6 weeks before use. TILLING lines were grown to select homozygous mutant genotypes and multiplied to the M5 generation to provide enough seeds for comparative purposes. For comparisons of the pea sgrL
3421 (W197STOP) mutant and control (cv. Cameor) plants, two independent experiments were performed using either 20 (plant trait measurements) or 50 (plant trait and biochemical measurements) replicate plants of each line. Plants were grown in 9 cm2 pots and watered normally for 27 days. Then half of each group was maintained under either well-watered or drought-stress conditions, following procedures previously described (Charlton et al. 2008), with modifications for the latter treatment. At the onset of the drought treatment, plants were not watered for nine days and, thereafter, these plants were given 20 ml of water per day throughout the recovery period until plants senesced. Biochemical data were collected on the final day on which water was withheld and 7 and 13 days later (Recovery Day (RD) 0, RD7 and RD13). Climate data for RD0, 7 and 13 were collected on the John Innes Centre site using a T200 Horticultural Computer (a TomTech weather station).
Library construction and sequence identification
RNA preparations from well-watered or drought-stressed plants (all as described in Charlton et al. 2008) were used for the construction of libraries and suppression subtractive hybridisation (SSH) carried out to identify those transcripts that were drought-responsive. Individual RNA batches were tested for induction of dehydrin, as a control drought-stress responsive sequence (Charlton et al. 2008), where expression of a His-Asp phosphorelay gene (GenBank AJ831475.1) was used as a control (Supplementary Fig. S1). A pea SSH cDNA library, enriched for drought-responsive leaf cDNAs, was constructed (Clontech PCR-Select™ cDNA Subtraction Kit) with checks throughout on the quality of the polyA+ RNA, cDNA synthesis, ligation of the adapter primers and the subtractive hybridisations. Subtracted cDNA was cloned into the plasmid vector pCR2.1 (Invitrogen) and transformed into a super-competent E. coli strain (One Shot TOP 10, Invitrogen). Sequence analysis of clones identified 557 unigenes (database accessions EBI AM161647-AM162203), and these were classified into groups according to likely function (Supplementary Fig. S1). Representative clones were chosen to examine the ratio of transcript level in drought-stressed compared with well-watered plants across four independent experiments. Amplified PCR products from 96 clones (1 µl from 10 µl purified product spotted on duplicate nylon filters) were hybridized with digoxygenin-labelled total cDNA from control and stressed plants. Quantification of the relative signal from images of the filters provided a conservative estimate that over 15 % of the products were up-regulated in the drought-stressed cDNA.
Genomic analysis
DNA was prepared from pea leaves using a manual extraction method (Welham and Domoney 2000). Genomic DNA amplification was performed using TaKaRa Ex Taq (Clontech-Takara Bio Europe), according to the manufacturers’ instructions, with the following PCR conditions: 98 °C for 10 s, 58 °C for 30 s, 72 °C for 1 min per kb amplified fragment, final extension at 72 °C for 5 min, hold at 4 °C. Sequencing of genomic amplicons was performed by TGAC (tgac.ac.uk) and Eurofins (eurofinsgenomics.eu) sequencing services. The primers used for genomic and cDNA amplification and sequencing are available in Supplementary Table 1.
Genetic mapping of SGRL
Mapping in the JI 281 × JI 399 population was carried out using 91 progeny lines and exploiting a Cleaved Amplified Polymorphic Sequence (CAPS) marker, based on the enzyme SspI to digest an amplicon, generated by the primers MK_C15F and MK_C15rev (Supplementary Table 1). Recombination frequencies and marker associations were estimated using THREaD Mapper (Cheema et al. 2010) and Haldane functions (Haldane 1919). Mapping in the Princess × JI 185 population was carried out using 152 progeny lines and performed by sequence analysis of a single nucleotide polymorphism (SNP) in an amplicon generated by the primers, SGRL-F8 and SGRL-R1, and sequenced using the primer SGRL-R3 (Supplementary Table 1); data were analysed using JoinMap (Stam 1993).
cDNA and qRT-PCR analysis
Total RNA was isolated from tissues which were frozen in liquid nitrogen, and stored at −80 °C. Tissues were either powdered directly or after freeze-drying (for high water content tissues) and ground in 700 µl RNA extraction buffer (1 M Tris–HCl pH 9.0, 1 % SDS, 10 mM EDTA), extracted twice with 350 µl phenol and 350 µl chloroform/IAA (24:1) and centrifuged at 14,000 rpm for 5 min. Following addition of 50 μl 3 M sodium acetate and 1 ml 100 % ethanol to 500 μl of supernatant, RNA was precipitated at −80 °C for 1 h, recovered by centrifugation at 14,000 rpm for 5 min, dried and dissolved in 200 µl RNAse-free water, and precipitated by addition of 200 µl 4 M LiCl overnight at 4 °C. RNA pellets were washed with 900 µl 2 M LiCl, twice with 900 µl 100 % ethanol, centrifuged, dried and dissolved in 25 µl RNase-free water.
RNA samples were DNase-treated (Qiagen RNeasy Mini Kit) prior to first strand cDNA synthesis, which was carried out using 1–3 µg RNA and 10 pmol primer A236 (poly A adaptor) in 11 µl H2O which was heated to 70 °C for 10 min, immediately cooled on ice and centrifuged briefly. Following addition of 4 µl enzyme reaction buffer, 2 µl 0.1 M DTT, 1 µl RNase inhibitor cocktail (Invitrogen), 1 µl dNTP (10 mM) and 1 µl Superscript II reverse transcriptase (Invitrogen), reactions were incubated at 37 °C for 2 h. To obtain the 5′ untranslated SGRL sequence, a 5′ RACE system (Invitrogen) was used, according to the manufacturer’s instructions, with A236 primer as GSP1, and using SGRL R4 and R7 primers (Supplementary Table 1) in the nested PCR step with the AAP and AUAP kit primers, respectively.
Quantitative real-time PCR (qPCR) amplification of first strand cDNA templates from a variety of plant organs and the subsequent quantification of SGR and SGRL products was carried out, essentially under conditions described (Hellens et al. 2010; Chinoy et al. 2011), and using pea actin as a reference gene (Cooper et al. 2005). In each analysis the data were calibrated against the lowest expressing tissue (assigned a value of 1). The specificity of the primers was confirmed by gel analysis of products and melting curve analysis, and the authenticity of the products was verified by sequencing. The mean relative gene expression levels presented were determined from three independent experiments. Analysis of RNA from developing seeds at 10, 20, 30 days after flowering provided contrasting developmental stages (Vigeolas et al. 2008). For comparisons of gene expression during leaf development, leaves were sampled according to phenotype. Very young leaves were close to the growing apex within 1–2 days of unfurling. Young to mature leaves were fully expanded leaves near the apex or mid-way along the length of fully-grown plants. Old and very old leaves were those showing some loss of colour (wild-type SGR) and/or structure (mutant sgr).
Suppression PCR and gene walking
Using a method adapted from that of Siebert et al. (1995), 15 restriction digests of pea genomic DNA (BamHI, BclI, BglII, BstYI, ClaI, MspI, TaqI, DraI, HpaI, EcoRV, NaeI, ScaI, PvuII, SspI and StuI) were performed (5 μl DNA, 1 μl restriction enzyme, 8 μl 5× RL buffer (50 mM Tris acetate pH 7.5, 50 mM Mg acetate, 250 mM K acetate, 25 mM dithiothreitol, 250 ng/μl bovine serum albumin, 26 μl H2O)). Digests were incubated overnight at the appropriate temperature for each restriction enzyme. The digests were ligated by adding 2 μl 5× RL buffer, 1 μl T4 ligase (Gibco BRL Life Technologies), 5 μl 10 mM ATP, 1 μl adaptor primer appropriate to the enzyme (Supplementary Table 1) and 1 μl H2O, and incubation for 6 h at room temperature. Ligated restriction digests were diluted 1:1 with T0.1E (10 mM Tris–HCl, 0.1 mM EDTA) pH 8.0 before PCR amplification. Nested PCR was based on two gene-specific primers, the adaptor primers APX1A and APX1B (Supplementary Table 1) and the standard PCR master mix. The PCR conditions were: 94 °C for 2 min, 20 cycles of: 94 °C for 1 min, 65 °C for 30 s and 72 °C for 2 min, followed by a final 5 min extension at 72 °C and held at 4 °C for the first PCR; the second PCR conditions were: 94 °C for 2 min, 5 cycles of: 94 °C for 1 min, 65 °C for 30 s and 72 °C for 2 min, followed by 40 cycles of: 94 °C for 1 min, 62 °C for 1 min and 72 °C for 2 min, a final 5 min extension at 72 °C and held at 4 °C.
GATEWAY BP and LR cloning
Infiltration of leaves of Nicotiana benthamiana (adapted from Sainsbury et al. 2009) was exploited as a system to transiently express genes of interest and monitor phenotype. Coding sequences were amplified from first strand cDNA samples, using attB adaptor primers (Supplementary Table 1) in high fidelity PCR (Phusion polymerase, New England BioLabs), according to the manufacturer’s instructions, and the following conditions: 98 °C for 1 min, 5 cycles of: 98 °C for 10 s, 56 °C for 30 s, 72 °C for 1 min per 1 kb to be amplified, followed by 30 cycles of: 98 °C for 10 s and 72 °C for 1 min per 1 kb to be amplified, and a final elongation at 72 °C for twice the elongation time in the previous cycles, before being held at 4 °C. PCR products were cleaned using Wizard® SV Gel and PCR Clean-up System (Promega), according to the manufacturer’s instructions, before cloning. Coding sequences were cloned by GATEWAY via two reactions into pEAQ-HT Dest vectors (GenBank GQ497237.1) and subsequent transformation into Agrobacterium.
For the BP reaction, 1 µl of BP Clonase (Invitrogen BP Clonase kit) was added to 1 µl of Phusion PCR product, 1 µl pDONR207 (Invitrogen) and 2 µl TE buffer and left at 25 °C overnight. The enzyme was inactivated by adding 0.5 µl Proteinase K with incubation at 37 °C for 10 min. An aliquot (1 µl) of BP (or LR) reaction was added to 50 μl competent DH5α E. coli (Invitrogen) or 25 µl One Shot® TOP10 Chemically Competent E. coli (Invitrogen) cells and left on ice for 30 min. Cells were heat-shocked at 42 °C for 45 s and immediately cooled on ice. Recovery was in 900 or 450 µl S.O.C. medium (2 % tryptone, 0.5 % yeast extract, 10 mM NaCl, 2.5 mM KCl, 10 mM MgCl2, 10 mM MgSO4, 20 mM glucose) for the different cells, respectively (37 °C, shaking for 2 h). Aliquots of 20 and 50 µl were spread onto LB agar plates (tryptone 10 g L−1, yeast extract 5 g L−1, NaCl 10 g L−1, 1.1 % agar (pH 7.0)), supplemented with the appropriate antibiotic (gentamicin for BP cultures; kanamycin for LR cultures) and grown overnight at 37 °C.
PCR was performed on individual colonies, which were sampled first into PCR mix and then into 50 µl LB (as above without agar), supplemented with the appropriate antibiotic(s) (gentamicin for BP colonies; kanamycin for LR colonies; rifampicin, kanamycin and tetracycline for Agrobacterium colonies (see later)). Colony PCR conditions were 94 °C for 2 min, 35 cycles of: 94 °C for 15 s, 56 °C for 30 s, 70 °C for 1 min per 1 kb DNA amplified, followed by 4 °C hold. PCR amplicons were verified by sequencing and validated clones grown overnight at 37 °C with shaking (28 °C for Agrobacterium). Plasmids were extracted using the Wizard® PLUS SV Minipreps DNA purification system (Promega), according to the manufacturer’s instructions.
For the LR reactions, 1 µl of LR Clonase (Invitrogen LR Clonase kit) was added to 1 µl of BP plasmid DNA, 1 µl of pEAQ-HT DEST 1 or DEST 3 (pEAQ-HT vector and modifications for GATEWAY compatibility, Sainsbury et al. 2009) and 2 µl TE buffer, and incubated at 25 °C overnight. The enzyme was inactivated by adding 0.5 µl Proteinase K (Invitrogen) and incubation at 37 °C for 10 min, and plasmids cloned and selected as described above.
Agrobacterium transformation and agro-infiltration
A 2 µl aliquot of LR plasmid DNA was added to 50 µl electro-competent Agrobacterium cells (C58C1) and electroporation performed at 2.5 V for 4.5–4.8 s. Cells were recovered in 1 ml LB broth with shaking at 28 °C for 2 h. Aliquots (50 and 100 µl) were plated onto LB agar supplemented with rifampicin, kanamycin and tetracycline and grown at 28 °C for 2 days before colony selection as above.
Agrobacterium cells were recovered by centrifugation, re-suspended in MMA solution (1 ml 1 M MES, 1 ml 1 M MgCl2, 150 µl 0.1 M acetosyringone, in 100 ml H2O), diluted to OD600 0.4 ± 0.025 (unless otherwise stated) and shaken at room temperature for 3 h. Cells were infiltrated into leaves of Nicotiana benthamiana plants (5–6 weeks old) by piercing the back of the leaf with a needle and injecting the re-suspended cells into the pierced hole using a syringe. Up to four samples were infiltrated into a single leaf, with between six and nine replicates for every construct tested per experiment. For double infiltration experiments, a gene encoding Green Fluorescent Protein (GFP) was used as a control that was not related directly to the chlorophyll metabolic pathway. For dark incubation experiments, leaves were covered with a cardboard box and an outer black bag immediately following infiltration.
Evaluation of leaf phenotypes
A Photosynthesis Efficiency Analyser (PEA) (Hansatech, UK) was used to analyse photosynthetic activity of leaves, using a number of parameters provided by the analyser; Fv/Fm gave a measure of photosystem II efficiency. Chlorophyll assays were carried out, using discs of 5 mm diameter from leaf samples, which were freeze-dried, ground with acid-washed sand and extracted in 500 μl of 10 mM Tris–HCl buffered, pH 8.0, 80 % acetone for 30 min at 0 °C (Nicotiana benthamiana) or in 100 % acetone overnight at −20 °C (pea). Samples were centrifuged for 2 min at 14,000 rpm and absorbance at 664 and 647 nm measured, using 450 μl of supernatant. Chlorophyll concentrations were determined (Porra 2002), where:
$${\text{Chl}}a\left( {\upmu{\text{g}}/{\text{ml}}} \right) = \left( {12.25 \times A664} \right){-}\left( {2.55 \times A647} \right)$$
$${\text{Chl}}b\left( {\upmu{\text{g}}/{\text{ml}}} \right) = \left( {20.31 \times A647} \right){-}\left( {4.91 \times A664} \right)$$
$${\text{Chl}}a + b\left( {\upmu{\text{g}}/{\text{ml}}} \right) = \left( {17.76 \times A647} \right) + \left( {7.34 \times A664} \right)$$
Evaluation of transient gene expression using His-tagged proteins
All the constructs assembled for transient expression in Nicotiana benthamiana leaves were designed to encode proteins with or without carboxy-terminal His-tags; for these alternatives, the same vector was generally used, with or without the natural stop codon of the gene in question. Where the stop codon was removed, an alternative codon (Y) ensured read-through to an additional stretch of 16 amino acids, culminating in six histidine residues (Sainsbury et al. 2009). Leaf samples that had been infiltrated with constructs predicting a His-tagged protein were freeze-dried, ground and extracted in LDS sample buffer (100 μl/mg) (Invitrogen) containing 0.05 M DTT. Samples were analysed on 4–12 % Bis–Tris gels (Novex, Life Technologies), alongside SeeBlue Plus2 markers (Invitrogen). Following electrophoresis, proteins were blotted onto nylon membranes using the Lightning Blot system (Perkin Elmer), membranes were blocked with 3 % BSA in PBST (150 mM NaCl, 10 mM NaPO4, pH 7.2 containing 0.5 % Tween 20) for 3 h, and incubated with Anti-6X His tag® antibody [AD1.1.10] (Alkaline Phosphatase) (Abcam, UK) diluted 7:15,000 in 1 % BSA in PBST. Blots were washed three times in PBST, developed using pre-mixed BCIP®/NBT solution (SIGMA Aldrich) for up to 15 min, and washed in distilled water to stop development. Duplicate gels were stained for protein using InstantBlue stain (Expedeon) to ensure even loading of samples.
Evaluation of photosystem complexes following transient expression
Thylakoids were prepared as described by Jarvi et al. (2011), using 5 × 5 mm leaf discs of infiltrated leaf areas per sample. Pelleted thylakoids were suspended in 40 μl native PAGE buffer (Invitrogen) with 1 % n-dodecyl-β-d-maltoside (Invitrogen) on ice for 15 min, centrifuged for 15 min at 4 °C, and the supernatants removed and stored at −80 °C. Prior to loading gels, 0.75 μl 5 % G-250 sample additive was added to 15 μl aliquots of thylakoid protein extracts. Electrophoresis was carried out using NativePAGE™ 4–16 % Bis–Tris gels (Invitrogen) at 150 V. For two-dimensional analysis, lanes from native PAGE were excised and proteins denatured in LDS sample buffer containing 0.05 M DTT for 30 min. Proteins in gel slices were analysed by electrophoresis using 12 % SDS Bis–Tris two-dimensional well gels (Novex, Life Technologies). Identification of bands was based on comparisons with earlier NativePAGE analytical data (Liu and Last 2015).
Statistical analysis
Bonferroni tests were used to compare multiple means in datasets, using GenStat 17th Edition, and a significance threshold of p < 0.05. Pairwise t tests were performed using Excel (*p < 0.05; **p < 0.01; ***p < 0.001).
Accession of sequence data
Sequence data from this article can be found in the EMBL/GenBank data libraries under accession numbers LN810021 (SGRL genomic, cv. Cameor) and LN810020 (SGRL mRNA, cv. Cameor).