Introduction

Heavy metal (HM) toxicity is considered a major threat to living organisms. HM-polluted soils derived from increasing geologic and anthropogenic activities have significantly impacted the production of high-quality agricultural crops in certain regions of the world. Plants growing on these soils exhibit reduced growth, photosynthetic performance, and yield1. As a naturally occurring HM pollutant, Cd exposure has been documented in most organisms, particularly plants and humans2.

World fertilizer consumption is increasing and will eventually reach a point where the drawbacks outweigh the benefits3. The same mechanisms that drive improved plant productivity often create side effects such as environmental contamination. Furthermore, some components of fertilizers, especially Cd, accumulate in both body and food chain, where they remain for an extended period and causes adverse health effects4. Therefore, fertilizers containing very high levels of Cd (417 mg/kg) threaten human health by accumulation in important crop such as maize5. Maize, as one of the most popular cereal grain, is widely cultivated across the world. Maize is also produced at an industrial scale as a key input in various products such as syrups, soft drinks and charcoals6,7. Therefore, there is a strong incentive to minimize the toxic effects of Cd in maize.

The mechanism of Cd toxicity in plants is still under investigation though it has been proposed to disturb photosynthetic function via inhibition of oxygen evolution and reduced chlorophyll content8,9,10. Plants exposed to high Cd concentrations exhibit reduced photosynthesis and root cellular metabolism11,12,13,14. Investigation on PSII function during the photo-activation process confirms that Cd possesses an inhibitory effect through active competition with Ca2+ ions in the catalytic center15.

Figure 1
figure 1

Effects of exogenous GABA (0, 25 and 50 µM) applications on shoot and root fresh (A,C) and dry weights (B,D) of maize seedlings exposed to 0 (control) and 250 µM cadmium (Cd).Values are the means of six replicates and bars indicate means ± SEM.

Figure 2
figure 2

Effects of exogenous GABA (0, 25 and 50 µM) applications on shoot to root fresh [(FW), A] and dry weight [(DW), B] ratios in maize seedlings exposed to 0 (control) and 250 µM cadmium (Cd). Values are the means of six replicates and bars indicate means ± SEM.

During evolution plants have evolved diverse adaptive strategies to alleviate the negative effects of high Cd concentrations. These adaptations are associated with the production of Cd-sequestering compounds such as phytochelatins or with active Cd export from the cell16. Recent studies have revealed a link between low molecular weight thiol (–SH) compound in most plant and Cd toxicity17. In different studies, increase in glutathione (GSH) content, as one of the main thiol compound; have been reported after Cd exposure; as it is the precursor for the synthesis of phytochelatins18. Detoxification is another general mechanism developed in plants to cope with highly toxic metals and to maintain physiological tolerance. Under stress condition endogenous regulating networks set in motion to maintain a delicate balance between reactive oxygen species (ROS) production and ROS-scavenging pathways. Antioxidants enzymes comprising superoxide dismutase (SOD), ascorbate peroxidase (APX), catalase (CAT) and glutathione peroxidase are commonly present in all intracellular regions and mainly responsible for oxidative burst. Alternatively, chelation of metals by ligands is a possible strategy by which compartmentalization of the ligand-metal into a limited area prevents the free circulation of ions in the cytosol12. Notably, in response to HM stresses, plants accumulate signaling molecules that maintain their capacity for growth and development19,39,40,41. GABA accumulation has been reported as a result of exposure to different stress conditions such as high temperature42, osmotic pressure19,43, cold shock44, salinity stress39, cold tolerance in peach fruits (Amygdalus persica)45, salt tolerance in Nicotiana sylvestris46 and Lactuca sativa41 and drought tolerance in perennial ryegrass (Lolium perenne)47 have been previously reported. The beneficial role of GABA against stress-induced oxidative damage in various plant species has been often reported41,48,39. Earlier studies reported the increased photosynthetic capacity by exogenous application of PAs in different plant species under various stress conditions28,73,74. Previous study conducted in maize and rice plants have revealed that cd-responsive orthologous are conserved in both plants. Among which ZmGAD1 gene, responsible for glutamate decarboxylase (GAD) enzyme (a key enzyme in GABA biosynthesis) has been introduced as a novel Cd-responsive gene and successfully conferred with Cd tolerance in yeast75. Here we have focused on GABA modulatory effect on PAs as different line of the tolerance pathway in maize plants. The biological link between GABA and PAs is established by the fact that GABA is interconverted through PAs catabolism34. In our study, IBP levels remained constant when both concentrations of GABA were applied to non-stressed plants (Fig. 9B). This can imply that GABA inter-relation with Put formation is mainly due to the stress-induced signals in plant76. Higher GABA concentrations (50 µM) reduced final IBP level. These results are indicative of higher GABA concentration having a negative feedback effect on IBP biosynthesis. A similar trend was also obtained for FP in Cd-exposed plants (Fig. 9A). Results obtained from SCP measurement indicated a sharp induction from 25 µM GABA application on Cd-exposed plants (Fig. 9C). This result is consistent with a study from Liu and colleagues (2004) reporting higher FP level in drought-sensitive wheat plants versus induced SCP form in tolerant cultivars77. In light of these results, a possible hypothesis could be the conversion of FP to SCP forms modulated by 25 µM exogenous GABA application. Enhanced level of SCP has also been correlated with improved drought tolerance in rice plants78,79. However, reduced SCP content has been detected by Yong and colleagues in GABA + PEG-treated white clover plants80. This conflict can be explained by dynamic strategies employed by different plant species in response to various stressors. Drought stress mainly refers to curtailed water status in plants while in Cd-stressed conditions plants have the challenge of actively removing excess Cd from the cytoplasm and compartmentalizing it to the vacuoles. In both distinct strategies, elevated Put level has been addressed80,81. Moreover, elevated expression of ORDC gene by GABA in this study (Fig. 12A) substantiates the notion of GABA-induced Put accumulation. However, there are limited studies establishing Put as a major cause of such effects. This led us to speculate that plants have potential to address an array of variable responses by using the same metabolic resources. However, there is a gap of sufficient information as how the various forms of similar PAs (Put) are being modulated under different stress-related conditions. GABA-treated plants, particularly at the 25 µM concentration, exhibited an observable enhancement in both FS and SCS contents under Cd stress (Fig. 10A,C), while IBS showed dramatic increase at 50 µM GABA and no significant differences at exposure to 25 µM GABA (Fig. 10B). These data indicate that higher FS and SCS levels are not required to cope with Cd stress. PA biosynthesis and degradation by exogenous GABA has been reported in other plant species under stressful conditions82,83. However, in our study increased levels of SCS were observed in both Cd-exposed and control plants. In Cd-exposed plants, SPDS gene expression was elevated (Fig. 12C) and more interestingly, PAO gene expression was inhibited by GABA application (Fig. 12B). This tempts us to speculate that the two effects of GABA–induction of Spd formation and inhibition of self-degradation–are different sides of the same coin. In an opposite sense, reduced Spm content by GABA application could be attributed to the degradation of Spm to produce Spd. Increased level of Spm in GABA-treated plants can be explained by inhibited degradation of Spm by reduced PAO gene expression. These results imply a multifunctional role for GABA under dynamic plant cell responses.

Several lines of evidence concluded that Cd exposure results in accumulation of thiol compounds in living organisms. Sulphydryl groups of thiols (e.g. GSH) potentially bind to the HMs like Cd and lead to efficient metal sequestration84. However our finding is in disagreement with the assumption that Cd tolerance involves an increase of GSH. Our data showed that in maize plants GSH content does not change by Cd exposure and GABA also manifested null effect on GSH level. Previous findings have also shown that the activity of γ-glutamyl-cysteine synthase, a rate limiting enzyme in GSH metabolism, in tolerant plants is lower than its activity in susceptible ones18. In a same fashion, transgenic Arabidopsis plants harboring bacterial γ-glutamyl-cysteine synthetase resulted in Cd sensitivity85. These findings combined with our result let us to speculate that Cd tolerance in different plants may rely on various mechanisms. GSH can “transiently” increase by abiotic stresses such as Cd shock86. Alternatively, plants employ largely variable strategy to commit excessive Cd level. One example can be the mechanism that inhibits Cd absorption or translocation in plants. At the cellular level, Cd enters plants mostly through root epidermal cells. This entrance is associated with three well-known steps including, (i) Cd exchange with H2CO3-released H+, (ii) recruitment of the Fe2+, Zn2+ (belong to ZIP family transporters) such as Yellow Stripe 1-Lik (YSL) protein and Ca2+ channels and iii) formation of metal ligand complexes with mugineic acids (MA) in soil and being authorized for entrance1 (Fig. 13). Upon entry, long distance translocation of Cd is governed by the xylem loading process mediated by HM P1B-type ATPases enzymes87,88. In our current study, Cd uptake by maize plants was significantly reduced by application of GABA, particularly at the 50 µM concentration (Fig. 4). GABA alleviated the Cd level in two district steps comprising uptake from soil and transfer to the shoot. Although, it is hard to determine specific function for GABA in Cd uptake and translocation processes. In this regard, one possible mechanism could be GABA-mediated Cd chelation, which would distance the Cd from the root system. Alternatively, GABA may change the osmotic pressure by regulation of cytosolic pH50,89,90, leading to lower Cd absorption at cellular level, and ultimately reduced Cd concentration in the cell. GABA may regulate enzymatic activities to inhibit intra-plant Cd transfer, thereby limited concentration in shoot. This is consistent with our finding as evident from shoot to root ratio that implies higher inhibitory effect of Cd in root growth. To date, knowledge about GABA interaction with certain receptors has been limited to the plasma membrane localized aluminum-activated malate transporter, ALMT, which functions as a GABA receptor51. Although this study suggested an inhibitory effect of GABA on Cd uptake by maize plant, a more defined mechanism/s of Cd uptake regulated by GABA still remains to be discovered.

Our results indicate that from early stage, Cd in the rhizosphere can negatively influence growth and development. However, positive effects of GABA on growth parameters in both control and Cd-treated plants are conclusive. Current results manifested a generic role for GABA in improvement of the Cd-borne defects in plants ranging from early growth to cell metabolism ultimately resulting in the protection of photosynthetic functionality. Multifaceted role of GABA was defined by concerted regulation upon ROS production and Cd uptake. Along with Cd stress, GABA conferred positive effects on PA biosynthesis, conversion and gene expression, though in this respect the intermediary role of GABA or its direct effect are still under debate. In light of GABA exposure, stress-related metabolite, ion hemostasis and maintenance of plant cell membrane integrity addressed dynamic responses induced by GABA and imply an impressive function for that (Fig. 1). However, our results illustrate a wide range of roles, which strongly implies a central role for GABA, in plant programmed responses to Cd toxicity. A broader view will be provided by analyzing the role of GABA associated with Cd perception, reverse genetic studies on GABA related pathways, fundamental studies on GABA/phyohormone crosstalk and a closer examination on the effect of GABA on cell membrane ion channels.

Methods

Plant materials and growth conditions

Maize (Zea mays) seeds were sterilized with 0.5% sodium hypochlorite for 5 minutes and soaked for 6 hours in distilled water at room temperature. Seeds were sown in pots (20 × 30 cm) filled with equal volumes of perlite and cocopite and grown under controlled conditions (light/dark regime of 16/8 hour at 25/20 °C of day/night cycles and relative humidity of 50%). Uniformly germinated seeds were selected for chemical treatment in the greenhouse. Seven day-old seedlings were irrigated with half strength Hoagland solution containing 0, 250 µM Cd; and 0, 25 and 50 µM GABA for three weeks. Irrigation was performed two times per week; once with solution contained Cd and GABA the rest with the half strength Hoagland solution. Every other week, root medium was washed off to preclude any additive effects of Cd accumulation.

Determination of plant growth parameters

To determinate fresh weight, root and shoot samples were washed off with water to remove soil and blotted gently with soft paper towel to remove any free surface moisture. Fresh weights were determined immediately and dry weights were measured after drying in an oven at 60 °C for 72 h.

Maximum quantum efficiency of photosystem II (Fv/Fm)

Youngest fully developed leaves were used for measuring Fv/Fm in a chlorophyll fluorescence imaging system (Handy FluorCam FC 1000-H; Photon System Instruments, Brno, Czech Republic.(Intact, attached leaves to the plants were dark-adapted for 20 minutes and immediately were used to measure Fv/Fm. Fv/Fm was calculated using a custom-made protocol91,92,93. Leaf samples were exposed to short flashes in darkness followed by a saturating light pulse (3900 µmol m−2 s−1) causing reduction of primary quinone acceptor of photosystem II. The fluorescence data obtained during short flashes in darkness (F0) and during the saturating light flash (Fm) were used to achieve the variable fluorescence (Fv = Fm − F0). The Fv/Fm value was calculated using version 7 of FluorCam software.

Determination of antioxidant enzyme activity

To extract and measure the activity of antioxidant enzymes, the leaf tissues were powdered by liquid nitrogen and enzyme extraction was performed according to Sariam et al.94. To measure CAT enzyme activity, 0.5 g of leaf powder was mixed in 10 ml of 0.1 M phosphate buffer (pH:7.5) containing 0.5 ml of EDTA. To extract APX, 0.5 g of powdered leaf was mixed in 10 ml of 0.1 M phosphate buffer (pH:7) containing 0.5 mg of ascorbic acid. The mixture was filtered using soft cloth and the solution was strained and transferred to the centrifuge tubes. The solutions were centrifuged for 15 minutes at 4 °C with 20,000 g. Transparent extractions (supernatant) were used to evaluate the enzyme activity. The activity of CAT enzyme was measured according to the modified method described by Díaz-Vivancos et al.95. The reaction was started by adding 100 μl of enzyme extraction and decrease in the absorbance of H2O2 during 1 minute at 240 nm. One CAT unit was considered as the amount of enzyme needed to oxidize 1 mM H2O2 per minute.

APX enzyme activity was measured according to Nakano et al.96. APX reaction consisted of 50 mg phosphate buffer (pH:7), 0.5 mM ascorbic acid, 0.1 mM H2O2 and 100 μl enzyme extraction. APX activity was calculated based on the reduction of ascorbic acid absorption per minute at 290 nm wavelength. One unit of activity of APX was considered as the amount of enzyme necessary for the oxidation of 1 ml of ascorbic acid per minute. The activity of the enzymes was expressed as a specific activity by mg of enzyme unit/fresh leaf weight.

The activity of the SOD enzyme was measured by Giannopolitis et al.97. The reaction solution contained 13 mM methionine, 75 mM Nitroblue-tetrazolium (NBT), 2 mM riboflavin, 50 mM phosphate buffer (7.8 ppm). The solution was placed under a 15-watt fluorescent lamp with a light intensity of 1000 lux and reaction started by the turning on the fluorescent lamp and left for 10 minutes. The reaction was terminated by turning off the lamps. The reaction solution was coated with black cloth to measure absorbance. The absorbance was measured at 560 nm. One of the samples was not exposed to light and considered as a control. SOD activity was considered according to the amount of enzyme required to 50% inhibition of the photochemical recovery of nitroblute tetrazolium chloride and was calculated based on Asado et al.98.

Measurement of EL percentage

EL percentage was measured by using 0.5 cm leaf discs. Leaf discs placed in tubes containing 10 ml of deionized water and incubated for 24 hours at 25 °C on a rotary shaker; first electrical conductivity of the solution was recorded as C1. Second electrical conductivity was determined in 20 minutes autoclaved samples after equilibration at 25 °C and labeled as C2. The EL was calculated based on the ratio between C1and C299.

Determination of H2O2 level

Leaf samples homogenized in the reaction mixture containing 0.5 ml of TCA (0.1%), 0.5 ml of K-phosphate buffer (100 mM) and 2 ml reagent (1 M KI w/v in fresh double-distilled water). The standard reaction was prepared by the same solution in the absence of leaf sample. The reaction was developed for 1 hour in darkness and the absorbance was measured at 390 nm. H2O2 level was determined according to the standard curve prepared with known concentrations of H2O2 levels base on the method described by Patterson et al.100.

Measurement of superoxide anion and lipid peroxidation

Superoxide anion (\({{{\rm{O}}}_{2}}^{-\cdot }\)) content was measured following the method of Elstner et al.101. To measure the lipid peroxidation, a testosterone (TBA) test, which identifies MDA as the final product of lipid peroxidation was conducted according to Heath et al.102. To do so, 0.1 g of fresh leaf tissue was homogenized in 2 ml TCA 0.1% (w/v). The resulting solution was centrifuged at 12000 g for 15 min, 0.5 ml of supernatant were added to 1 ml of 20% TCA solution containing 0.5% TBA. The solution was placed in boiling water for 30 minutes and the reaction was stopped by placing the samples in the ice. The absorbance of the resulting solution was measured at 532 nm and the absorption of other non-specific pigments was also read at a wavelength of 600 nm. Determination of MDA concentration from a shutter ratio of 155 nM−1cm−1 in terms of nanomol per gram of fresh tissue was calculated using the following equation.

$${\rm{MDA}}=\frac{(\text{A532}-\text{A600})}{155}\times 1000$$

Measurement of proline content

Proline content was measured according to Bates et al.103. In this method, a ninhydrin solution was prepared and 1.25 g of ninhydrin were weighed and combined with 30 ml of glacialacetic acid and 20 ml of phosphoric acid 6 M. After dissolution, it incubated at 4 °C for 24 hours. To measure proline, 0.1 g fresh weight sample was weighed and centrifuged in 2 ml of sulfosalicylic acid 3%. Samples were centrifuged for 10 min with 8000 g. 400 microliters of ninhydrin was added to equal volume of the extract and ultimately combined with 400 μl of glacial acetic acid (100%). The solutions were incubated at 100 °C for 1 hour and were transferred to the ice. After cooling down, 800 μl of toluene was added to the sample and vortex vigorously to separate the distinct phases. Finally supernatant (red solution) was removed and read on a spectrophotometer at a wavelength of 520 nm. The amount of proline was calculated according to the following equation:

$$\mu \text{mole}\,{\rm{proline}}/{\rm{g}}=[(\mu {\rm{g}}\,{\rm{proline}}/{\rm{ml}}\times {\rm{ml}}\,{\rm{toluene}})/115.5\,\mu {\rm{g}}/\mu \text{mol}]/[({\rm{g}}\,\text{sample}/5)]$$

Measurement of GHS content

Fresh leaf samples (0.5 g) were homogenized in 2 ml of 5% sulphosalicylic acid at 4 °C. The homogenate was centrifuged at 10,000 × g for 10 min. 600 µl of 100 mM Phosphate buffer (pH 7.0) and 40 µl of 5′5′ dithiobis-2-nitrobenzoic acid (DTNB) were added to 500 µl of supernatant. The absorbance was read after 2 minutes at 412 nm17.

HPLC analysis of Pas

Extraction of PAs was carried out with a few modifications according to Hassannejad et al.104. All steps were carried for three replicates and three samples per replicate. 0.2 g of the fresh root tissue was homogenized with 1 μl of perchloric acid (PCA) at 5% volumetric concentration and centrifuged for 35 min at 4 °C with 15000 g. Precipitates were used for insoluble bound PAs extraction and supernatants were separated for extraction of soluble conjugated and free PAs. To retrieve all soluble conjugated PAs, the precipitates were washed three times with PCA. Final sediment were supplemented by 1000 μl of PCA and centrifuged for 35 minutes with 15000 g. Since bound PAs are insoluble in PCA, 1 ml of NaOH (1N) was added to the residual sediment and centrifuged with 21000 g for 30 minutes. Subsequently, 100 μl of the last solution (for analyzing the insoluble bound PAs) and 100 μl of the primary supernatant (for the analysis of the conjugated PAs) was placed into the separate test tubes and 200 μl of HCl (6 Normal) were added to start hydrolysis process. Tubes were closed by a flame and placed in an oven at 110 °C for 18 hours. In order to evaporate the acid, the flamed lids were broken and again placed in an oven for 18 hours at 80 °C. Further, the same procedure was applied to extract all types of PAs. To do so, 100 μl of each solutions containing free, soluble conjugated and the insoluble bound PAs was placed in a separate test tubes and covered with foil. 200 μl Na2CO saturated solution was added to each tube. 400 μl of dansyl chloride (concentration of 5 mg/ml) was added to all tubes and incubated for 90 minutes in 60 °C. Subsequently dansyl chloride was removed from the solution and 200 μl of proline (concentration of 0.1 g/ml) was added to each tube and left at 60 °C for 45 minutes. Later, 500 μl of toluene was added to each tube and vortex for 30 seconds. After few minutes different phases performed and the PAs phase was slowly removed and placed in −20 °C. The concentration of all PAs was measured by HPTLC (CAMAG, Switzerland) and winCATS software 1.2.2. The blotting was done on the Merck (60 F254 TLC) papers covered with silica gel. Length of 4 mm and a gap of 10 mm were considered for each band. Ultimately measurement was performed using phase containing cyclohexane and ethyl acetate (V/V 5:4; in which 4 units of cyclohexane with 5 units of ethyl acetate was mixed). Chromatogram scanning was done at two wavelengths of 245 nm and 366 nm.

RNA extraction and quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was isolated from maize roots and treated with DNAse I using Qiagen plant RNAeasy kit (Qiagen; according to the manufacture’s instruction). RNA concentration was quantified by NanoDrop spectrophotometer. cDNA was synthesized from 1 μg of total RNA using an i-script cDNA synthesis kit (Bio-Rad, Hercules, USA) as described in the manufacturer’s protocol. Quantitative RT-PCR was performed using SYBR green based detection (Bio-Rad, Hercules, USA). Experimental setup and execution was conducted using a MyIQ optical cycler, according to protocol provided by the manufacturer (Biorad, Hercules, USA). Data analysis was performed using BioRad iQ5 software (BioRad). Baselines were set at 100 RFU to calculate Ct values. Refrence primer sequnece was obtained from membrane protein PB1A10.07c acording to Manoli et al.105. Primer sequences are listed in Table 1.

Table 1 Primers used in this experiment.
Figure 11
figure 11

Effects of exogenous GABA (0, 25 and 50 µM) applications on different forms of spermine (Spm) [free (A) insoluble bound (B) and soluble conjugated (C)] contents in maize plants exposed to 0 (control) and 250 µM cadmium (Cd). Values are the means of three biological and three technical replicates and bars indicate means ± SEM.

Figure 12
figure 12

Relative expression levels of ornithine decarboxylase (ORDC) (A), PA oxidase (PAO) (B) and Spd synthase (SPDS) (C) genes in the root of maize plants exposed to different concentrations of GABA (0, 25 and 50 µM) and cadmium (Cd) [0 (control) and 250 µM]. Expression levels are based on the average of three biological replicates that included three technical replicates in maize roots.

Figure 13
figure 13

Schematic representation of GABA regulatory role in plant cells under Cd stress. Cd induces PAs and GABA induction in the cell and initiates downstream signaling cues. GABA is involved in different metabolic and molecular processes (black arrows and lines) and in return GABA content in the cell is influenced by diverse pathways (red arrows). PA levels in this scheme have been drawn based on the results obtained from currents experiment in which blue, red and purple boxes representing free, soluble conjugated insoluble bound forms, respectively. Question marks indicate unknown processes. ALTM: Aluminum-activated malate transporter, Arg: Arginine, Cd: Cadmium, CM: Cell membrane, GABA: Gama amino butyric acid, MA: Mugineic acids, ORDC: Ornithine decarboxylase, PAO: Polyamine oxidase, Put: Putrescine, Spd: Spermidine, Spm: Spermine, SPDS: Spermidine synthase, YSL: Yellow Stripe 1-Lik protein.

Measurement of Cd concentration

Dried root and shoot samples were weighed and overheated by 700–500 °C for 3 hours and digested in 5 ml of nitric acid and left for 18 hours. The solutions were filtered by Whatman filter paper and diluted with distilled water up to volume of 10 ml. The Cd absorption was measured by Schimadzu atomic absorption/flame emission. In every sample Cd concentration was calculated based on the standard curve creating by a range of defined Cd concentrations. The following equation was used to calculate the Cd content (mg/kg dry weight) of each sample106.

$$\begin{array}{c}{\rm{Cd}}\,{\rm{content}}\,{\rm{in}}\,{\rm{root}}\,{\rm{or}}\,{\rm{aerial}}\,{\rm{part}}\,({\rm{mg}}/{\rm{kg}})\\ =\,[{\rm{Cd}}\,{\rm{concentration}}\,{\rm{of}}\,{\rm{tissue}}\,({\rm{mg}}/{\rm{L}})\times {\rm{final}}\,{\rm{volume}}\,{\rm{of}}\,{\rm{sample}}\,({\rm{ml}})]/{\rm{Dry}}\,{\rm{weight}}\,{\rm{of}}\,{\rm{tissue}}\,({\rm{g}}).\end{array}$$

Cd translocation factor was calculated according to Rezvani et al.107.