Identification of two CiGADs from Caragana intermedia and their transcriptional responses to abiotic stresses and exogenous abscisic acid

Plant materials and treatments

Seeds of C. intermedia were collected from a desert habitat in Inner Mongolia, China. Seedlings were cultivated as described by Shi et al. (2010). Briefly, seeds were washed and then germinated in sand. Three-day-old seedlings (radical length, 30–40 mm) were selected and cultivated with water in pots in a growth chamber under a 14-h light/10-h dark photoperiod and a day/night temperature of 23/18 °C. The seedlings were grown for 2–3 weeks, and the water was refreshed every 2 days. Subsequently, the seedlings were subjected to the following stress treatments: (1) NaCl (200 mM), ZnSO₄ (100  µM), CdCl₂ (500  µM), high temperature (40 °C), low temperature (4 °C), or dehydration (for 3 h); (2) NaCl (200 mM) for different durations (0, 3, 6, 12, and 24 h); and (3) exogenous ABA (0, 0.5, 1, 10, 100, and 200  µM) combined with the 200 mM NaCl treatment for 0, 3, and 24 h. Roots and leaves were harvested, frozen immediately in liquid nitrogen, and then stored at −80 °C. Also, different tissues including roots, stems, leaves, seeds, testae, and bark were collected from mature individuals growing at the Chinese Academy of Forestry, Beijing, frozen immediately in liquid nitrogen, and then stored at −80 °C. Every experimental treatment had three biological replicates.

Cloning, sequencing, and bioinformatic analysis

Total RNA was extracted from pooled roots and leaves with Trizol reagent (Invitrogen, Carlsbad, CA, USA), and complementary DNA (cDNA) was generated with the SMARTer™ RACE cDNA amplification kit (Clontech, Palo Alto, CA, USA) according to the manufacturer’s instructions. Then, 5′ and 3′ RACE were performed using nested gene-specific primers (Table S1). The obtained amplification products were cloned into the pMD19-D vector (Takara, Dalian, China) and sequenced. Then, the two full-length sequences were assembled by DNAMAN 6.0. CiGAD1 and CiGAD2 have been deposited in GenBank under the accession numbers KU586714 and KU586715, respectively.

The two full-length cDNA sequences of CiGAD1 and CiGAD2 were analyzed with ORF Finder (https://www.ncbi.nlm.nih.gov/orffinder/), and the isoelectric point (pI) and molecular weight (MW) of the predicted protein were estimated using ExPASy (http://web.expasy.org/compute_pi/). Sequences for the protein homology analysis were obtained from phytozome (https://phytozome.jgi.doe.gov/pz/portal.html), and aligned using DNAMAN 8.0. The phylogenetic tree was constructed using MEGA 6.06 software by the maximum-likelihood algorithm with 1,000 bootstrap replicates.

Southern blot analysis

Genomic DNA was isolated from leaves of C. intermedia using CTAB method (Porebski, Bailey & Baum, 1997), and 20  µg DNAs were digested with BamH I and EcoR V, respectively. The obtaining fragments were separated by electrophoresis through a 1.0% agarose, then the DIG-labeled DNA probes of CiGAD1 and CiGAD2 were hybridized followed the protocol of DIG-High Prime DNA Labeling and Detection Starter Kit I (Roche Applied Science, Mannheim, Germany). Briefly, the DIG-labeled DNA probe was prehybridized at 60 °C for 2 h and then hybridized 37 °C for 12 h. Following hybridization, the blot was washed twice in 2 × SSC containing 0.1% SDS at 25 °C for 5 min and was washed twice in 0.1 × SSC containing 0.1% SDS at 50 °C for 15 min. The immunodetection of the DIG-labeled probe was performed with 1:10,000 anti DIG-AP. The blot was exposed to X-ray film (AGFA, Germany). DNA probes were amplified from genomic DNA by using the primers designed in the sequenced DNA fragments based on the cDNA sequences of CiGAD1 and CiGAD2, while the probe for CiGAD1 was mainly in the region from 1,076–1,370 bp in the full-text cDNA, and the probe for CiGAD2 was mainly in the region from 639–1,031 bp in the full-text cDNA. Primers are listed in Table S1 and probe sequences are listed in Supplemental Information 3.

Assay of GAD activity

About 0.1 g sample (root, stem, leaf, seed, or bark) was ground into a powder in liquid nitrogen and transferred to a 2-mL centrifuge tube. Then, 1 mL pre-cooled extraction buffer (150 mM potassium phosphate, 5 mM EDTA, 1 mM magnesium chloride, 0.5% (w/v) polyvinylpyrrolidone, 3 mM 2-mercaptoethanol, 0.2 mM PLP (pyridoxal 5′-phosphate), 10% (v/v) glycerol, 1 mM phenylmethylsulfonyl fluoride, pH 5.8) was added, and the homogenate was mixed well. After 15 min incubation on ice, the homogenate was centrifuged at 10,000 g for 15 min at 4 °C. The supernatant was collected into a 2-mL volumetric flask and the volume was adjusted to 1 mL. This was used as the crude enzyme solution (CES; Liu et al., 2014).

To measure GAD activity, the assay mixture (1 mL) contained 0.4 mL 0.1% sodium glutamate, 0.2 mL 0.25 mM PLP, and 0.4 mL CES. After mixing well, the reaction mixture was incubated at 40 °C for 5 h and then boiled for 5 min to end the reaction. Sodium glutamate was replaced by 0.4 mL distilled water in the blank control. After cooling, the absorbance of the mixture was measured at 640 nm using a Multi-Mode Detection Platform (SpectraMax Paradigm; Molecular Devices Co., Sunnyvale, CA, USA). A GABA standard curve was constructed and used to calculate the GABA content in the samples based on absorption value, as described by Yang, Hui & Gu (2016). And protein contents were tested by Bradford method (Kruger, 1994) with bovine serum albumin as standard. One unit (U) was defined as the amount (µmol) of GABA produced by the action of one mg protein per hour. All determinations were representative of three biological experiments and four technical replicates.

Determination of GABA

The plant tissues were ground to a fine powder in liquid nitrogen, and about 0.1 g of the frozen homogenate was extracted using methanol and lanthanum chloride, and then centrifuged at 13,000g for 5 min, and 0.8 ml of the supernatant was transferred to a new Eppendorf tube. The 160  µl of 1 M KOH was added, and centrifugation as before. The 100  µl resulting supernatant was used in the spectrophotometric GABA determination (Zhang & Bown, 1997). The assay mixture (300 µl) also contained 75 mM potassium pyrophosphate (pH 8.6), 3.3 mM 2-mercaptoethanol, 1.25 mM β-NADP⁺, 5 mM 2-ketoglutarate and 0.02 units of GABase (Sigma). The increase of OD340 nm was recorded using 96-well microplate reader. The amount of GABA was calculated according to external calibration curve of GABA (Renault et al., 2010). Values shown were representative of three biological experiments and four technical replicates.

qRT-PCR

Total RNAs were extracted from a pool of roots and leaves with Trizol reagent (Invitrogen), and then cDNAs were synthesized with 0.5 µg RNAs using the Prime Script™ RT reagent kit (Perfect Real Time; Takara, Dalian, China) according to the manufacturer’s instructions. Specific RT-PCR primers (Table S2) were designed to have melting temperatures of 60 °C and amplicon lengths of 150–200 bp using Primer3 software (http://primer3.ut.ee/). Real-time qRT-PCR was performed in quadruplicate using the SYBR Premix Ex Taq™ II kit (Takara) on a Roche lightCycler 480 (Roche Applied Sciences, Penzberg, Germany) according to the manufacturer’s instructions. The 2^−ΔCt method was adopted to analyze the qRT-PCR result, and ΔCt = Ct_target-Ct _EF1α. Values shown were representative of three biological experiments.

Statistical analysis

Data were evaluated by Duncan’s multiple test in SPSS v.19. Differences were considered significant at P < 0.05.

Isolation of CiGADs and bioinformatics analyses

Two full-length cDNAs of glutamate decarboxylase (GAD) genes were isolated from C. intermedia by 5′- and 3′-RACE, as shown in Fig. S1, and designated as CiGAD1 (GenBank ID: KU586714) and CiGAD2 (GenBank ID: KU586715). Bioinformatics analyses showed that CiGAD1 had a 1922-bp full-length mRNA sequence containing a 1521-bp open reading frame (ORF), a 107-bp 5′ puntranslated region (UTR), and a 294-bp 3′-UTR; and that CiGAD2 had a 1797-bp full-length mRNA sequence containing a 1797-bp ORF, a 60-bp 5′-UTR, and a 237-bp 3′-UTR. The alignment showed that the nucleotide identity between CiGAD1 and CiGAD2 was 69.0%. Theoretically, CiGAD1 encodes a 57.3-kDa peptide consisting of 506 amino acids with the theoretical isoelectric point (pI) of 5.5, and CiGAD2 encodes a 56.1-kDa peptide consisting of 499 amino acids with the theoretical isoelectric point (pI) of 5.6.

The deduced amino acid sequences of the two CiGADs were aligned with GAD family members from three typical model plants; Arabidopsis, Glycine, and Populus (Fig. S2). The amino acid identity between CiGAD1 and CiGAD2 was 71.4%. CiGAD1 shared 88.5%, 79.9% and 79.4% amino acid identity with Glycine GAD5, Arabidopsis GAD1, and Populus GAD1, respectively. CsGAD2 shared 83.1%, 75.0%, and 72.1% amino acid identity with Glycine GAD2, Populus GAD5, and Arabidopsis GAD5, respectively. Both CiGAD proteins contained two conserved domains: a PLP-binding domain in the middle region and a CaM-binding domain at the carboxyl terminus (Fig. 1). The ML tree showed that CiGAD1 was located close to GmGAD5, GmGAD1, and GmGAD3; while CiGAD2 was in a completely different clade, close to GmGAD2, GmGAD4, PtGAD5, and AtGAD5 (Fig. 2). These findings indicated that CiGAD proteins from C. intermedia, a salt- and drought-resistant desert legume shrub, had higher homologies with GAD homologs in soybean, another well-known legume.

We also performed a southern blot analysis to estimate the copy number of two GAD genes in C. intermedia. The result showed one band in BamH I or EcoR I restriction the digest lane of CiGAD1 (Fig. 3A) and two bands in the BamH I or EcoRI restriction digest lane of CiGAD2 (Fig. 3B), respectively. This confirmed that CiGAD1 had one copy, and CiGAD2-related genes were present as two copies in the genome of C. intermedia.

Tissue-specific expression of two CiGADs

The expression of two CiGADs showed different patterns in various issues including roots, stems, leaves, seeds, testae, and bark from mature individuals (Fig. 4A). The transcript levels of CiGAD1 were lower than those of CiGAD2 in roots, seeds, testae, and bark, but the opposite pattern was observed in other tissues. The two genes showed contrasting trends in mRNA abundance in most tissues. The highest transcript level of CiGAD1 was in testae, followed by leaves and stems, while it was almost undetectable in roots and bark. In contrast, CiGAD2 showed the highest transcript level in the bark (37.9-fold that of CiGAD1 in the bark), followed by testae and roots (1.9 and 21.3 fold higher, respectively, than that of CiGAD1). The GAD activity was highest in testaes (9.9 U), followed by seeds (3.2 U). But its activity was lower, and similar, among the other tested tissues (1.1–1.9 U) (Fig. 4B).

Stress-specific expression of two CiGADs

To investigate whether the two CiGADs were expressed differently in response to various stress conditions, total RNA was extracted from roots and leaves of young seedlings under several different stress treatments. Both CiGADs showed dramatic increases in their transcript levels in roots and leaves in response to stress (Figs. 5A, 5B ). Under Zn stress, the transcript level of CiGAD1 increased by 74.3-fold and that of CiGAD2 increased by 6.3-fold, compared with their respective levels in control roots. CiGAD1 showed 3.4–11.3 fold increases in its transcript levels in roots under dehydration, Cd, Na, and low/high temperature stresses. CiGAD2 transcript levels showed 1.0–5.5 fold increases in roots under all of the stresses except Cd stress (Fig. 5A). In leaves, the transcript levels of CiGAD1 and CiGAD2 strongly increased in response to high temperature by 218.1-fold and 114.2-fold, respectively, compared with their respective levels in the control leaves (Fig. 5B). The transcript levels of CiGAD1 and CiGAD2 increased by 20.6-fold and 36.0-fold, respectively, under NaCl stress, and by 19.7-fold and 21.8-fold, respectively, under low temperature stress. Under Zn, Cd, and dehydration stress, CiGAD1 expression increased by 4.4–8.5 fold, and CiGAD2 expression increased by 1.5–5.2 fold. These results indicated that the transcript levels of CiGAD1 and CiGAD2 increased to varying degrees in response to various stresses.

To analyze the expression of the two CiGADs under stress in more detail, seedlings were treated with NaCl for different times, as shown in Fig. 6. The transcript levels of CiGAD1 and CiGAD2 significantly increased as the duration of the NaCl treatment extended, although there were some fluctuations in their transcript levels in leaves at 12 h and in roots at 6 h (Figs. 6A and 6B). In roots, the transcript levels of CiGAD1 and CiGAD2 had increased by 20.8-fold and 16.4-fold, respectively, after 6 h of salt stress, and by 41.2-fold and 12.0-fold, respectively, after 24 h of salt stress, compared with their respective levels at 0 h (Fig. 6A). However, in the leaves, CiGAD1 and CiGAD2 transcript levels remained relatively stable except at 6 h of salt stress, when they were increased by 20.6–23.7 fold, and 11.5–13.8 fold, respectively, compared with their respective levels at 0 h (Fig. 6B). Further analyses showed that CiGAD1 was more strongly induced than was CiGAD2 as the NaCl treatment extended. Corresponding to the expression of CiGADs, there was a considerable increase in endogenous GABA with increasing duration of the NaCl treatment (Fig. 6C). The GABA content in roots increased rapidly, to 5.9-fold its initial level after 3 h of NaCl stress, and to more than 10-fold its initial level at 12 and 24 h of NaCl stress. In the leaves, the GABA content slightly increased after 3 h of NaCl stress, and markedly increased to 4.7–5.6 fold its initial level by 6 h of NaCl stress. These findings suggested that there was a positive correlation between the expression of CiGAD genes and GABA accumulation during NaCl stress.

ABA regulated CiGADs expression under NaCl stress

Treatment of NaCl-stressed seedlings with exogenous ABA significantly affected the expression of both CiGADs, and the changes in expression differed between the two genes (Fig. 7). In the roots, both CiGADs showed gradually increasing transcript levels with increasing concentrations of exogenous ABA at 3 h. However, CiGAD1 showed a 3.8-fold increase in its transcript levels in response to 0.5 µM ABA, similar to the increase in response to 200 µM ABA. The two genes showed almost opposite trends in expression in the roots at 24 h of treatment with ABA; the transcript levels of CiGAD1 decreased with increasing ABA concentrations, but were still much higher than the transcript levels of CiGAD2 (Fig. 7A). In the leaves, however, after 3 h of ABA treatment, low concentrations of exogenous ABA (0.5 and 1.0 µM) inhibited the expression of both CiGADs, but higher concentrations of ABA induced the expression of both genes. After 24 h of treatment with ABA, 0.5 µM ABA considerably increased the expression levels of both genes to more than 5.6-fold that in the control (0 µM ABA), but ABA at other concentrations drastically reduced the transcript levels of both genes, except for CiGAD1 at 10 µM ABA (Fig. 7B). These results demonstrated that, even at micromolar concentrations, ABA participates in regulating CiGAD1 and CiGAD2 transcription during salt stress.