Background

Soil contamination with toxic metals such as cadmium (Cd), lead (Pb) and mercury (Hg) is a global environmental problem [1]. Cd is one of the naturally occurring toxic heavy metal present in the Earth’s crust and negatively affects the plant growth and development. Cd is readily absorbed by plants. Cd excess in plants can disrupt nutrient homeostasis, change physiological processes, and induce massively toxic reactive oxygen species (ROS) [2,3,4], all of which are associated with the damage of proteins, nucleic acids and lipids of plasma membrane [5,6,7]. More seriously, Cd accumulation in crops, particularly in edible parts, risks crop production and food safety [8].

Many plant species have developed diverse mechanisms to deal with the issues by Cd efflux, chelation, sequestration or detoxification [9, 10]. For example, most of brassicaceae species like India mustard, Brasisca napus or Arabidopsis generate phytochelatins, when challenged to environmental Cd [5, 11]. PCs are small cysteine-rich peptides with general structures of (γ-Glu-Cys)n-Gly (n = 2–7). The PCs biosynthesis is catalyzed by phytochelatin synthase (PCS, EC 2.3.2.15) with glutathione as a substrate, and has been long considered as metal detoxified chelatins because Cd can be compounded by the free thiol groups [12]. The PCs-Cd complex (2500–3600 Da) is across the tonoplasts and sequestered in vacuoles for Cd detoxification [13]. In Arabidopsis, PCS1 and PCS2 are two important components for the metal-dependent transpeptidation and mainly contribute plant tolerance to Cd toxicity [14, 15]. To date, the PCs biosynthetic pathway is thought of one of the most important mechanisms for plant tolerance to Cd stress. Some other Cd-inducible genes were also found to be involved in PCs-dependent pathway, such as HsfA4a [16], HMT1 [17], PAD2–1 and VTC2–1 [18], ZAT6 [19]. For example, Chen et al. [20] reported a Cd-inducible gene MAN3 (also called XCD1 or XVE system-induced cadmium-tolerance 1), which is coding for an endo-β-mannanase. The Cd-induced MAN3 increased the mannanase activity and the cell wall mannose content as well. Interestingly, MAN3 overexpression enhanced Cd accumulation and tolerance through the GSH-dependent PCs synthetic pathway [20]. In addition, an ABC-type transporter AtABCC3 was reported involving the Cd-responsive phytochelatin-mediated tolerant pathway [21].

Ferrochelatase-1 (FC1, EC4.99.1.1) is the terminal enzyme of heme biosynthesis, catalyzing insertion of ferrous iron into protoporphyrin IX (Proto IX) to generate protoheme [22,23,24]. Proto IX is the branch point of the tetrapyrrole pathway of heme and chlorophyll biosynthesis. While chlorophyll is known as an important pigment for light absorption, transfer and photosynthesis in plants, heme serves as a cofactor of many important proteins for diverse biological processes such as electron transfer, oxygen metabolism, and secondary metabolism [25]. FC1-produced heme is required for up-regulation of ROS scavenger genes for plant tolerance to abiotic stress [26,27,28,29,30]. FC-1 is highly conserved in green algae, plants and mammals, although the different number of the FC gene isoforms exists [31,32,33]. Soybean (Glycine max), poplar (Populus trichocarpa), and apple (Malus domestica) contain five or more FC gene members, whereas Arabidopsis have only two (FC1, AT5G26030) and (FC2, AT2G30390). In Arabidopsis, AtFC1 and AtFC2 have a distinct gene expression model. While AtFC1 is expressed at all plant tissues including stems, flowers, leaves, and roots, AtFC2 mainly expresses in photosynthetic tissues like shoots [22,23,24, 33, 34]. This suggests that AtFC1 and AtFC2 have different roles in various tissues. The differentially expressed AtFC1 and AtFC2 can be associated with the distinct genetic phenotypes in their functionally defective mutants. For example, fc2 mutants display reduced chlorophyll and carotenoid abnormal phenotypes, while fc1 mutants are hardly detected [33]. Promoter analyses show that AtFC1 can be activated by norflurazon (an inhibitor of carotenoid biosynthesis) and viral infection, whereas AtFC2 promoter activity was repressed under the same condition [26]. Furthermore, AtFC1 is induced by a variety of biotic and environmental stresses such as drought, wounding and oxidative stress; in contrast, no such observations were found in AtFC2 [33]. These findings indicate that AtFC1 and AtFC2 differently contribute to the biological functions.

To date, the majority of studies are focusing on identification of the essential role of FC in regulating plant growth, development and metabolism [33,34,35], and only a few of them concern FC in plant responses to abiotic stress [27, 36]. No reports are available for FC functional regulation of plant adaption to metal stress such as Cd-induced toxicity in plants. In this study, we reported a previously unknown function of AtFC1 in response to Cd stress. We showed that AtFC1 could be induced by Cd. Arabidopsis plants overexpressing AtFC1 manifested enhanced Cd tolerance, while mutation of AtFC1 led to more sensitivity of plants to Cd. AtFC1 was found to activate genes involved in PCs production, suggesting that Cd-induced AtFC1 was able to positively regulate Cd tolerance through the PCs synthesis-mediated pathway.

Results

Cd stress induced expression of AtFC1 in Arabidopsis

To ascertain whether AtFC1 responses to Cd stress, the expression pattern of AtFC1 in Arabidopsis wild-type (WT) was monitored under Cd stress by qRT-PCR. Expression of AtFC1 was significantly induced by 50–200 μM Cd, while the maximal enhancement of AtFC1 expression was found at 100 μM Cd (Fig. 1a). Cd-induced AtFC1 expression was also found in a time-dependent manner, in which treatment with Cd at 100 μM enhanced 5.9-fold higher expression at 9 h than the control (Fig. 1b). In contrast, AtFC2 failed to be induced by Cd exposure, even though the Cd concentration was up to 200 μM (Fig. 1c). This result is consistent with the conclusion that FC2 appears not to respond to environmental stress [33, 34].

Fig. 1
figure 1

Expression pattern of FC-1 and FC-2 in Arabidopsis seedlings exposed to Cd. Three weed-old seedlings (wide-type) cultured in half-strength Hoagland solution were exposed to 0, 50, 100 and 200 μM Cd for 9 h (a, c) or to 100 μM Cd for 0, 1, 3, 6, 9 and 12 h (b). Following treatment, total RNA was extracted. Transcript levels were analyzed by qRT-PCR. Vertical bars represent standard deviation. Asterisks indicate that the mean values are significantly different between the Cd treatment (+Cd) and control (−Cd) (p < 0.05)

Full size image

Overexpression of AtFC1 enhanced Cd tolerance in Arabidopsis

To identify the role of AtFC1 in regulating plant response to Cd stress, we generated AtFC1-overexpressing transgenic lines of Arabidopsis. Full length of AtFC1 sequence was fused to the cauliflower mosaic virus 35S promoter and transformed into the Col-0 ecotype background. The 35S::FC1 transgenic lines used in this study were screened and identified. The transgenic plants carrying 35S::AtFC1 showed 13.5 to 22.8-fold higher transcripts of AtFC1 than the wild-type plants (Additional file 1: Data S1). We further used a T-DNA insertion mutant of AtFC1 (SALK-150001.42.45×, fc1) from the Arabidopsis Biological Resource Center. The fc1 mutant was verified by diagnostic PCR using gene-specific primers and was found with a T-DNA insertion in the 5′-untranslated regions; qRT-PCR analysis showed that the AtFC1 transcript levels in fc1 mutant plants was only 29% of those in wild type (Additional file 1: Data S2).

Seeds of WT, 35S::FC1, and fc1 were germinated and grown on the half strength MS agar plates with Cd (0, 50, 100 and 150 μM) for 12 d. In non-Cd medium, no significant differences of biomass within the WT, 35S::FC1, and fc1 mutant plants were observed (Fig. 2a). When seedlings were exposed to 50–150 μM Cd, 35S::FC1 plants had increased fresh biomass, whereas fc1 mutants had reduced biomass, as compared to WT (Fig. 2b). A similar result was observed for dried biomass (Additional file 1: Data S3). We then examined the primary root elongation of seedlings under Cd stress. 35S::FC1 and WT seeds were placed on the solid 1/2 MS medium supplemented with 50 μM Cd for 12 d. 35S::FC1 plants showed strong root growth over the wild-type (Fig. 2c). Under Cd exposure, the elongation of 35S::FC1 roots was 1.2–1.4 fold higher than that of wild-type; in contrast, the root elongation of fc1 mutants was weak compared to the wild-type (Fig. 2d). Chlorophyll is sensitive to heavy metals and often used as a biomarker of metal stress [37]. Compared to WT, the levels of Chl-a and Chl-b were not affected in 35S::FC1 or fc1 mutant in non-Cd medium (Fig. 2e, F). Under 50 μM Cd stress, the concentrations of Chl-a and Chl-b increased in 35S::FC1 plants, but decreased in fc1 mutants. These results showed that AtFC1 was able to improve the plant tolerance to Cd stress.

Fig. 2
figure 2

Analysis of cadmium (Cd) tolerance in the wild-type (WT), AtFC1 transgenic lines (35S::FC1) and fc1 mutants of Arabidopsis. a, b: Phenotype and fresh weigh of seedlings grown on 1/2 MS medium with or without 50, 100 or 150 μM Cd for 12 d. c, d: Root length of AtFC1 transgenic seedlings and fc1 mutants grown on 1/2 MS medium with or without 50 μM Cd for 12 d. E and F: The chlorophyll-a and chlorophyll-b contents of AtFC1 transgenic seedlings and fc1 mutants. Seedlings were grown hydroponically for 21 d and transferred to the same culture solution with and without 50 μM Cd for 5 d. Vertical bars represent standard deviation. Asterisks indicate that the mean values are significantly different between the transgenic plants/mutants and wild-type plants (p < 0.05)

Full size image

Overexpression of AtFC1 improved antioxidation capacity in plants

It is well known that Cd can induce oxidative stress in plants [6, 38]. To examine whether the transgenic plants have an antioxidative capability, we performed the histochemical staining with nitroblue tetrazolium (NBT), by which O2 -. And H2O2 were visualized in a dark blue by insoluble formazan compound and in a brown by 3,3-diaminobenzidine (DAB) reagent, respectively. Compared to WT, plant exposure to 50 μM Cd led to dark staining with NBT (Fig. 3a) and DAB (Fig. 3b) in the leaves of fc1 mutant, while in 35S::FC1 leaves, the staining was relatively light. The AtFC1-regulated decrease of H2O2 in plant tissues was proved by the quantitative analysis (Fig. 3c, d).

Fig. 3
figure 3

Accumulation of O2 −, H2O2 and other oxidants in the wild-type (WT), AtFC1 transgenic lines (35S::FC1) and fc1 mutants of Arabidopsis under Cd stress. Seedlings were grown in the half-strength Hoagland solution for 21 d and transferred to the same culture solution with 50 μM Cd for 1 d. a: Histochemical detection of O2 •− on leaves by NBT staining. b: Histochemical detection of H2O2 on leaves by DAB staining. c, d: H2O2 contents in roots and shoots. e: DCFH-DA fluorescent imaging for endogenous ROS. f: Quantitative analysis of fluorescent density of image of (e). g, h: TBARS contents in roots and shoots. Vertical bars represent standard deviation. Asterisks indicate that the mean values are significantly different between the transgenic plants/mutants and WT (p < 0.05)

Full size image

Both O2 -. and H2O2 as ROS can damage the plasma membrane and other biomolecules, An in situ detection of endogenous ROS in roots was further performed by the specific fluorescent probe DCFH-DA (2′, 7′-dichlorofluoresce indiacetate) [39]. Treatment with 50 μM Cd for 30 min increased the DCF (dichlorofluoresce) fluorescent density in WT (Fig. 3e, f). A relatively lower level of DCF fluorescence was detected in the Cd-exposed 35S::FC1 plants. Conversely, the fc1 mutant generated much more DCF fluorescence under the same condition. TBARS (thiobarbituric acid reactive substances) as a biomarker represents a degree of damage of cellular plasma membrane [39]. All these oxidative products were quantified in plants under Cd stress. Addition of 50 μM Cd into media led to a significant increase in TBARS content in both shoot and root of WT plants (Fig. 3g, h). Transgenic plants overexpressing AtFC1 showed a lower level of TBARS than WT plants. In fc1 mutant plants the TBARS content was always higher. These results suggest that overexpression of AtFC1 contributed to the antioxidation capacity by repressing the generation of O2 -. and H2O2 in plants.

Overexpression of FC1 improved non-protein thoils production and Cd accumulation in plants

The non-protein thoils (NPT) compounds, consisting of several acid-soluble sulfhydryl components such as cysteine, γ-glutamyl-cysteine, GSH and PC, are able to chelate toxic metals for plant detoxification [9]. Under the normal condition, there were no differences of NPT concentrations in WT, 35S::FC1 or fc1 mutant plants; when challenged to 50 μM Cd, 35S::FC1 plants accumulated more NPT in roots and shoots, whereas a lower level of NPT in fc1 mutants was detected (Fig. 4a, b). Similarly, 35S::FC1 plants accumulated more Cd than WT, while fc1 mutant plants accumulated less Cd, compared to WT (Fig. 4c, d).

Fig. 4
figure 4

NPT and Cd concentrations in shoots and roots of WT, (35S::FC1) and fc1 mutant plants. Seedlings were grown in the half-strength Hoagland solution for 21 d and transferred to the same culture solution with 50 μM Cd for 5 d. a, b: NPT contents. c, d: Cd contents. Vertical bars represent standard deviation. Asterisks indicate that the mean values are significantly different between the transgenic plants/mutants and WT (p < 0.05)

Full size image

AtFC1 improved Cd tolerance through PCs-dependent pathway

Buthionine sulfoximine (BSO) is a GSH synthetic inhibitor and was employed to investigate the involvement of GSH-dependent PCs pathway in AtFC1-regulated Cd response in Arabidopsis. The 35S::FC1 plants showed enhanced root growth and biomass under Cd stress compared to WT, but the enhanced-root growth could be blocked by BSO (Fig. 5a-d), suggesting that AtFC1-improved root growth was mediated by PCs-dependent detoxifying pathway. We then used a heme (the product of AtFC1) substitute, hematin (protoheme, with the same function and high stability) [28, 40], to examine the effect of hematin in fc1 mutants. In the absence of Cd, no difference of root elongation was observed between WT and fc1 plants; treatment with 50 μM Cd reduced the root elongation of fc1 mutants relative to WT (Fig. 5e-g). Supplying 10 μM hematin could largely rescue the root elongation and biomass of fc1 mutants under Cd stress (Fig. 5e-h). Supplying hematin also enhanced the GSH, PCs and NPT levels in the roots and shoots of wild-type under Cd stress (Additional file 1: Data S4). A combinational effect of BSO and hematin on the root growth was monitored. Under the normal condition, the roots of WT and fc1 mutant seedlings showed no significant difference in the presence of BSO or in combination with hematin (Fig. 6a). Under Cd exposure, however, the positive effect of hematin on root elongation of WT and fc1 mutants was diminished by BSO (Fig. 6b, c). A similar result was found for the fresh weight of the seedlings under the same condition (Additional file 1: Data S5).

Fig. 5
figure 5

Effects of buthionine sulfoximine (BSO) with combination of hematin on the growth of seedlings of the wild-type (WT), 35S::FC1 lines and fc1 mutants under Cd stress. Seedlings were grown on 1/2 MS media with 50 μM Cd, 100 μM BSO or 10 μM hematin for 12 d. (A, B): Phenotypes of seedlings growing with BSO under -Cd (A, 1/2 MS, control) and +Cd (B) stress. (C, D): Root elongation and fresh weight of seedlings corresponding to (a) and (b), respectively. e, f, g: Root elongation. h: Fresh weight corresponding to (e) and (g), respectively. Vertical bars represent standard deviation. Asterisks indicate that the mean values are significantly different between 35S::FC1 lines/fc1 mutants and WT (p < 0.05)

Full size image
Fig. 6
figure 6

Effect of buthionine sulfoximine (BSO) with combination of hematin on growth of seedlings of the wild-type (WT) and fc1 mutants under Cd stress. Seedlings were grown on 1/2 MS media with 50 μM Cd, 100 μM BSO or 10 μM hematin for 12 d. (A, B): Phenotypes of WT and fc1 seedlings growing with BSO and/or Hematin under -Cd (A, 1/2 MS, control) and –Cd stress (B). (C): Quantitative analysis of root elongation corresponding to (a) and (b). Vertical bars represent standard deviation. Asterisks indicate that the mean values are significantly different between the fc1 and WT plants (p < 0.05)

Full size image

AtFC1-regulated genes were related to the GSH-dependent PC synthesis pathway

Because overexpression or knock-down of AtFC1 led to contrasting phenotypes of plants under Cd stress, we assumed that many Cd-responsive genes including Cd tolerance and detoxifying genes were likely involved in the process. To address the question, we created four libraries with fc1 mutant and wild-type plants under –Cd and +Cd stresses, and employed the Illumina RNA-sequencing platform (HiSeq 2000), which allows to sequence transcripts in a high-throughput manner, to identify the global transcriptome of fc1 mutant and wild-type plants. A total of 29.9–31.6 million clean reads were generated from four libraries (Additional file 1: Data S6). Mapping the reads to Arabidopsis genome led to identification of 3317 (2026 up and 1291 down) and 2579 (1809 up and 770 down) unique genes in wild-type and fc1 mutant plants (> two fold change, p< 0.05) under Cd stress, respectively (Additional file 1: Data S7–10). The distribution of differentially expressed genes was shown by color dots (blue, down, and red, up) (Fig. 7a). The fc1 mutants under Cd stress displayed more red dots (differentially expressed genes, DEGs up-regulated) and less blue dots (DEGs down-regulated) than the wild-type (Col-0) plants. Also, under the control (−Cd) and Cd stress (+Cd), more red dots than blue dots were found in the fcl/Col-0 group, indicating that more genes were induced in the fc1 mutants than those in the wide-type (Fig. 7a). These results were also presented by the heat-map graph (Fig. 7b). The number of the DEGs was further detailed by plotting Venn diagrams. There were 1130 genes that were specifically upregulated and 1043 downregulated in wild-type (Col-0) under Cd stress; also, 913 specifically upregulated genes and 522 repressed genes were found in the fc1 mutant plants (Fig. 7c). Comparative analysis of DEGs between fc1 mutant and wild-type plants revealed that 1210 genes were specifically induced and 976 genes were repressed under the normal condition, whereas 227 genes were specifically induced and 148 genes were repressed under Cd stress (Fig. 7d).

Fig. 7
figure 7

Differentially expressed genes in the wild-type (Col-0) and fc1 mutants of Arabidopsis under Cd stress. Two week-old Arabidopsis seedlings were exposed to 0 (−Cd) and 50 μM Cd (+Cd) in the half-strength Hoagland solution. a: Distribution and expression levels of DEGs in WT and fc1 mutant seedlings with or without Cd exposure. The x axis represents the log2 fold change under the mean normalized expression of all transcripts (y axis). Green dots indicate the down-regulated genes, and red dots show the up-regulated genes. b: Heatmap of a one-dimensional hierarchical clustering of differential gene expression as determined by mRNA-seq for the Cd-exposed (+Cd) seedlings (WT and fc1) relative to the control (−Cd). c: Venn diagram showing up- and down-regulated genes in Col-0 and fc1 mutant seedlings under Cd stress. d Venn diagram showing up- and down-regulated genes in fc1 seedlings relative to Col-0 grown under the control (–Cd) and Cd-exposed (+Cd) condition.

Full size image

The differentially expressed genes from fc1/Col-0 category following Cd exposure could be functionally classified using Gene Ontology. Based on the functional annotation, all genes can be divided into three major groups including biological process, cellular component and molecular function (Fig. 8a). Several pathways such as metabolic process, response to stimulus, transcription regulator activity and transporter activity were examined. As an example, genes involved in oxidative stress and heavy metal transport/detoxification were presented. The differentially expressed genes from the two combined library groups including Col-0 (+Cd)/Col-0 (−Cd) and fc1 (+Cd)/fc1 (−Cd) were selected for analysis (Fig. 8b, c). For oxidative stress-responsive DEGs, there were 65 genes from each library group. Of these, 49 genes overlapping in the two library groups showed a similar expression pattern. The remaining 16 genes expressed differentially between the two library groups. As shown in Fig. 8d, genes in the Col-0 (+Cd)/Col-0 (−Cd) group were upregulated, whereas those in the fc1 (+Cd)/fc1 (−Cd) group were downregulated under Cd exposure, indicating that the expression of the genes was impaired in fc1 mutant plants. Several genes encoding glutathione peroxidase, glutathione S-transferases, NAD(P)H dehydrogenase and lipid transport superfamily protein were examined. Some of the genes such as glutamate synthase, glutathione receptor and glutathione peroxidase are related to GSH synthesis, while others like glutathione S-transferases are responsible for glutathione transport and PCs synthesis 12 [12]. Similarly, 14 genes encoding metal transporters were found to be repressed in the fc1 (+Cd)/fc1 (−Cd) group under Cd exposure (Fig. 8c). Most of the genes such as those encoding ABC transporter C, heavy metal transport/detoxification protein, metal tolerance protein and copper transporter are responsible for functional metal transport and detoxification (Fig. 8e). These results indicated that disruption of AtFC1 expression altered the transcription of the genes responsible for Cd detoxification.

Fig. 8
figure 8

The mRNA-seq based differentially expressed genes for the Cd-exposed fc1 mutants relative to the Cd-exposed Col-0 wild-type. Two week-old Arabidopsis seedlings were exposed to 0 (−Cd) and 50 μM Cd (+Cd) in the half-strength Hoagland solution. a: GO analysis of specific genes from the 227 up-regulated and 148 down-regulated genes (Figure 8d) in fc1 mutants relative to the wild-type under Cd stress. Genes most related to Cd tolerance are noted with red arrows. b, c: Venn diagrams displaying the genes differentially expressed in Col-0 (+Cd)/Col-0 (−Cd) and fcl (+Cd)/fcl (−Cd) under Cd stress. The presented genes are functionally involved in Cd-induced oxidative stress (b) and Cd transport or detoxification (c). The small ellipses in the big ellipses represent the overlap of the genes between Col-0 and fc1 mutant plants. d, e: Heatmap represents the gene expression levels of individual reactive oxygen species-related genes (d) and heavy metal transport and detoxification genes (e). Some of the genes most related to Cd tolerance were highlighted by red tick off

Full size image

Because AtFC1-produced heme contributed to the Cd detoxification in plants, we further examined and confirmed the expression of genes involved in GSH and PCs synthesis in wild-type plants under Cd stress with or without Hematin by qRT-PCR. GSH1 encodes a γ-glutamyl-Cys synthetase (γ-ECS), which is the key enzyme limiting the synthesis of GSH in plants, and GSH2 encodes a glutathione synthetase [12, 41]. Both PCS1 and PCS2 encode a phytochelatin synthase in Arabidopsis [9]. Cd exposure significantly induced the expression of GSH1, GSH2, PCS1, PCS2 and GR1 (Fig. 9). Compared to the Cd treatment alone, supplying hematin enhanced the expression of the four genes. Glutathione reductase-1 (GR1) catalyzes the reduction of oxidized GSH and is one of the key enzymes involved in modulation of redox homeostasis in Cd-tress plants [42, 43]. In this study, we also found that expression GR1 was induced by Cd + hematin over the Cd treatment alone (Fig. 9).

Fig. 9
figure 9

Transcript levels of genes related to PC synthesis in the wild-type under Cd exposure in the absence or presence of hematin. Seedlings were grown in the half-strength Hoagland solution for 21 d and transferred to the same culture solution without or with 50 μM Cd in the presence of 10 μM Hematin for 1 d. Vertical bars represent standard deviation. Asterisks indicate that mean values are significantly different between the treatments and control (p < 0.05)

Full size image

Discussion

Although several genes that mediate Cd accumulation and detoxification through glutathione and phytochelatins pathway have been identified [4, 20, 44, 45], the mechanisms for the molecular regulation and genetic basis for the physiological process are not fully understood. The present study characterized a role of FC1 in Arabidopsis by exploring the effects of gain- and loss-of-function of the gene on the plant response to Cd exposure. The transcription of AtFC1 was sufficiently induced by Cd. This observation is consistent to the previous results that AtFC1 could be induced by other abiotic stresses [26, 27, 29]. Our result also supports the observation that AtFC1 is responsible for heme and hemeproteins involved in the defense responses [26, 27, 29].

To illustrate that AtFC1 was able to regulate plant response to Cd stress, transgenic Arabidopsis overexpressing AtFC1 was generated. The AtFC1 loss of function mutant fc1 plants was also identified in the study. The 35S::FC1 plants displayed high tolerance to Cd, with improved root elongation, biomass of seedlings, chlorophyll accumulation, and non-protein thiol compounds, but showed a low degree of oxidation with reduced ROS (e.g. hydrogen peroxide and superoxide anion) and lipid peroxides. Conversely, fc1 mutants displayed a phenotype sensitive to Cd stress. These data indicated that AtFC1 was able to mediate the Cd stress response in Arabidopsis. To get an insight into the mechanism leading to AtFC1-regulated detoxification of Cd in plants, we profiled transcriptome data from fc1 mutants under Cd stress. A large number of specific genes (522 genes) were found to be repressed in the fc1 mutants, suggesting that the requirement for AtFC1 is at least part of the Cd stress response. The Gene Ontology analysis revealed several pathways closely related to Cd stress response. Of these, genes involved in oxidative stress response and metal transport and detoxification should be paid more attention. The Cd-induced oxidative stress responsive genes contained several genes involved in GSH and PCs synthesis [22].

The GSH-dependent PCs synthetic pathway has been so far one of the most important mechanisms for metal tolerance in higher plants because PCs are prone to chelating and detoxifying toxic heavy metals like cadmium [4, 9]. Under Cd stress, some plant species can form Cd-GSH and Cd-PCs complexes to sequester Cd within vacuoles [3, 46] or disperse Cd through xylem and phloem vessels effectively [11]. Overexpression of GSH in Indian mustard and poplar enhanced Cd tolerance owning to the higher capacity of GSH and PCs synthesis [47,48,49]. Coexpression of PCS1 and GSH1 in Arabidopsis showed coordinated regulation of Cd tolerance and PCs production [44]. Thus, PCs play a vital role in heavy metal detoxification in plants [5]. The present study showed that AtFC1 overexpression increased accumulation of total NPTs (including GSH and PCs), and the increased accumulation of NPTs was associated with enhanced Cd accumulation and tolerance in the transgenic plants. This process was correlated with the activation of genes involved in PCs synthesis, suggesting that AtFC1 was one of the important regulators in the GSH-dependent PCs synthetic pathway contributing to Cd accumulation and tolerance. While the increased Cd accumulation was displayed in AtFC1 transgenic plants, a lower level of Cd in fc1 mutant plants was observed. This could be the result of weak generation of heme in fc1 mutants. AtFC1-produced heme is one of the most important tetrapyrroles in plant cells because it is a cofactor essential for many important enzymes, proteins and transporters involved in many biological activities including oxygen metabolism, metabolite transport, electron transfer, biotic and abiotic stress responses [25]. In this regard, a shortage of heme due to the loss of function of AtFC1 might directly or indirectly to affect some metal transporters in a negative way, which most likely impairs the uptake of Cd into plants.

Our analysis showed that pre-treatment with heme conferred plant tolerance to Cd stress. This observation can be implicated in the mechanisms for heme regulating Cd detoxification through GSH-dependent PCs synthetic pathway. Firstly, application of hematin increased the concentrations of GSH and PCs and as a consequence, the Cd tolerance in WT and fc1 mutant plants was enhanced. Secondly, the hematin-promoted accumulation of GSH and PCs coincided with the upregulation of the expression of genes (e.g. GSH1, PCS1, and PCS2) involved in GSH and PCs synthesis. Finally, the positive effect of hematin was cancelled by the GSH synthesis inhibitor BSO. The site-directed mutagenesis study suggests that Thr 49 in PCS1 can be phosphorylated by casein kinase 2, which in turn increases the activity of PCS1 [50]. Casein kinase 2 was suggested to involve heme-regulated protein synthesis in mammalian cells [51]. In this case, heme may be directly or indirectly involved in promoting PC synthesis by enhance PCS1 activity through modulating its phosphorylation state. Whether heme is involved in past-modification of PCS1 activity needs further investigation.

Oxidative stress is triggered by over-generation of ROS, an observation that occurs in plants under Cd stress [45]. Plant cells possess two intrinsic systems to remove excessive ROS as a result of toxic metal exposure. The first one is anti-oxidative enzymatic system including superoxide dismutase, ascorbate peroxidase, glutathione reductase and catalase; and the second antioxidant system includes GSH and ascorbate [37]. The protective role of FC1 against photodynamically induced oxidative stress has been proposed in Arabidopsis [27]. Therefore, the anti-oxidative role of AtFC1 in detoxifying Cd would at least partially attribute to the regulation of both anti-oxidants and anti-oxidative enzymes, which could be supported by several lines of evidence from this study and others: (1): 35S::FC1 plants accumulated less ROS, reflecting a high capacity of AtFC1 transgenic plants, in which the Cd-induced ROS generation was effectively scavenged by anti-oxidants and anti-oxidative enzymes; (2) AtFC1-generated heme positively regulated the GSH synthesis by coordinately activating the expression of GSH1. GR is an important enzyme involved in the regulation of GSH-AsA cycle for maintaining redox homeostasis in plants upon metal exposure [52]. Hematin was shown to induce the expression of GR1 in Cd-treated Arabidopsis, suggesting that heme may maintain the redox homeostasis by regulating GR1 in plants in response to Cd stress; and (3) heme can be utilized as a redox cofactor to enhance the activity of many so-called heme-proteins. Several heme-proteins such as ascorbate peroxidase and catalase are involved in response to oxidative stress [53]. Ectopic expression of a Bradyrhizobium japonicum FC in rice protected rice from photodynamically induced oxidative stress by enhancing the activity of heme-CAT and heme-APX [36]. These results suggest that FC1 would contribute to plant tolerance to Cd through antioxidative mechanism.

Conclusion

Our physiological and genetic evidence showed that AtFC1 was transcriptionally induced by Cd exposure. AtFC1 overexpression conferred plant tolerance to Cd stress by enhancing primary root growth, biomass and chlorophyll concentration, but reducing accumulation of reactive oxygen species such as H2O2, O·2-, and TBARS in tissues. The Cd-induced AtFC1 and AtFC1-generated heme were able to contribute to the plant tolerance to Cd by activating the genes responsible for the GSH-dependent phytochelatin synthetic pathway. Thus, our study would broaden our understanding of the role of FC1 in mediating plant response to Cd stress and provide a basis for further exploration of its downstream genes.

Methods

Plant growth condition and treatments

Wild-type seeds of Arabidopsis thaliana (ecotype Col) and the fc1 mutant (SALK_15000.142.45.X) were ordered from the Arabidopsis Biological Resource Center, Ohio, USA. Agrobacterium (EHa105) were ordered from BaiSi Biotechnology company (Hangzhou, China). Wild-type seeds, the fc1 mutant (AT5G26030) and transgenic seeds (T4, homozygote, Col back-ground) were surface-sterilized and plated on half-strength Murashige and Skoog (MS) solid media supplemented with 1% sucrose and 0.8% agar (pH 5.8). The plates were stored for 3 d in darkness at 4 °C and transferred to growth chamber with the conditions of 100 μE m−2 s−1, 16/8 h light/dark cycle, and 22 °C. For Cd treatment experiments, seeds of WT, the fc1 mutants or transgenic seed were germinated and grown on half-strength Hoagland nutrient solution with Cd (pH 6.0), buthionine sulfoximine (BSO; Sigma) or other reagents. Concentrations of Cd were set as 0, 50, 100 and 200 μM for dose-response experiment or 100 μM for a time-dependent (0, 1, 3, 6, 9 and 12 h) experiment. The seedlings were grown in a growth chamber with the same condition described above. The growth solution was changed every two days. After treatments plants were sampled and immediately frozen by liquid nitrogen for following experiments.

Transcript analysis by RT-PCR

Total RNA was extracted from tissues using TRIzol (Invitrogen) and 1.0 μg of RNA was used as a template for cDNA synthesis. A 1% agarose gel, stained by ethidium bromide, was run to check the integrity of the RNA. All RNA samples were quantified and examined for protein contamination (A260 nm/A280 nm ratios) and reagent contamination (A260 nm/A230 nm ratios) by a Nanodrop ND 1000 spectrophotometer.

The first strand cDNA was synthesized from 1.0 μg total RNA by Moloney Murine Leukemia Virus Reverse Transcriptase (Promega) using oligo(dT) primers. The quantitative RT-PCR (qRT-PCR) was performed with a MyiQ Single Color Real-time PCR system (Bio-Rad) in a final volume of 20 μL containing 2 μL of a 1/10 dilution of cDNA in water, 10 μL of the 2 × SYBR Premix Ex Taq (TaKaRa) and 200 nM of forward and reverse primers. qRT-PCR was conducted using a CFX96 Real Time PCR Detection System (Bio-Rad). The thermal cycling conditions were 40 cycles of 95 °C for 5 s for denaturation and 60 °C for 30 s for annealing and extension. All reactions were run in triplicate by monitoring the dissociation curve to control the dimers. PCR efficiency was determined by a series of 2-fold dilutions of cDNAs. The calculated efficiency of all primer pairs was 0.9 to 1.0. Gene Actin 2 was used as a reference and relative expression levels of genes were presented by 2-△CT. All primers used for qRT-PCR are presented in Additional file 1: Data S11.

Histochemical detection of H2O2, O2 •− and intracellular ROS

Hydrogen peroxide (H2O2) and superoxide radicals (O2 •−) were stained using diaminobenzidine (DAB) and nitroblue tetrazolium (NBT), respectively, according to the method described by Romero-Puertas et al. [38]. Seedlings were grown hydroponically for 21 d and transferred to the same culture solution containing 50 μM Cd for 24 h. When H2O2 was detected in tissue, Arabidopsis seedlings were incubated in DAB solution (pH 5.8, 1 mg mL−1) at room temperature for 24 h in the absence of light until the appearance of brown spots. For O2 •− detection, the treated plants were immersed in NBT solution (1 mg mL−1) at room temperature, and illuminated until the appearance of blue spots characteristic of blue formaz anprecipitate. After staining, leaves were transferred to 95% (v/v) ethanol to remove chlorophyll, and images were captured with a Nikon SMZ 1000 stereomicroscope.

To perform in situ detection of endogenous ROS, specific fluorescent probe DCFH-DA (2′, 7′ - dichlorofluoresce indiacetate) was used according to Foreman et al. [39]. Wild-type, fc1 mutants and transgenic seeds were germinated and vertically grown on the half-strength MS media supplemented with 1% sucrose and 1% agar (pH 5.8). The seedlings were then transferred to 1/4 Hoagland nutrient solution moist filter paper with 50 μM Cd for 30 min. Roots of seedlings were incubated in 10 mM of DCFH-DA at 25 °C for 10 min and rinsed with distilled water for three times. After that, the samples were visualized with a fluorescence microscope (excitation 488 nm and emission 525 nm, ECLIPSE, TE2000-S, Nikon). Image-Pro Plus 6.0 (Media Cybernetics, Inc) was used to analyze The relative fluorescent density of the fluorescent images.

Determination of O2 − and H2O2 and lipid peroxidation

Seedlings were grown hydroponically for 21 d and transferred to the same culture solution containing 50 μM Cd for 1 d. The H2O2 content was quantified as described previously with minor modification [54]. One gram of Arabidopsis shoots and roots was extracted in 1 mL of 80% ethanol. One hundred μL plant extracts were incubated for 30 min with 1 mL solution containing 90% methanol (v/v), 25 mM H2SO4 (v/v), 250 μM ferrous ammonium sulfate hexahydrate and 100 μM xylenol orange. The absorbance of the homogenate was measured at 560 nm. To calculate H2O2 concentrations, standard-curves ranging from 0 to 200 μM. Lipid peroxide, in terms of thiobarbituric acid reactive substances (TBARS), was determined by the method described previously [37].

Analysis of non-protein thiol compounds

Seedlings were grown hydroponically for 21 d and transferred to the same culture solution without or with 50 μM Cd for 1 d. The treated seedlings were harvested and used for analysis of total GSH and total NPT. Normally, 100 mg samples were ground in liquid nitrogen with a mortar and pestle. The ground samples were mixed with 300 μL solution containing 1 M NaOH and 1 mg L−1 NaBH4. The homogenates were centrifuged at 13,000 g at 4 °C for 5 min. A 300 μL supernatant was acidified by adding 50 μL of 37% (w/v) HCl. Part of the solution was collected for measurement of GSH and NPT. Total NPT was quantified as described previously [55]. Ten μL extract indicated above was mixed with 500 μL of 6 mM 5, 5-dithiobis (2-nitrobenzoic acid) in stock buffer (143 mM sodium phosphate and 6.3 mM Na2-EDTA, pH 7.5), and was incubated at 30 °C for 2 min. The content of NPT was determined at 412 nm. Analysis of total GSH content was conducted using the glutathione reductase recycling assay as described by Anderson [56]. The concentration of PCs was calculated by subtracting the amount of GSH from the amount of total acid-soluble sulfhydryl compounds [20].

Dry weight and measurement of cadmium content

Wild-type, fc1 mutant and AtFC1 transgenic seedlings were grown hydroponically for 21 d and transferred to the same culture solution containing 50 μM Cd for 5 d. When harvested, plants were thoroughly washed with 5 mM CaCl2 at 4 °C for 1 h and rinsed with sterile water. Shoots and roots from treated plants were separately sampled and dried at 75 °C for 2 d. After that, the samples were weighted. The dried samples were weighted and digested with nitric acid and hydrogen peroxide (HNO3: H2O2 = 1: 1, v/v). Total Cd contents from dried samples were determined using inductively coupled plasma-atomic emission spectrometry (ICP-AES) (Optimal 2100DV, Perkin Elmer Instruments, Waltham, MA, USA) [57].

Chlorophyll quantification

Seedlings were grown hydroponically for 21 d and transferred to the same culture solution containing 50 μM Cd for 5 d. Chl of plant leaves (0.1 g FW) was extracted with 1 mL of 80% (v/v) acetone until complete bleaching was achieved. Chlorophyll a, Chlorophyll b and total Chlorophyll were quantified by reading the absorption at A663 and A645, and its concentration was calculated by the method described previously [58].

Transformation of AtFC1 in Arabidopsis

pCAMBIA1304 was employed as an expression vector with CaMV35S as a promoter and NOS terminator as a transcriptional termination sequence [59]. The AtFC1 genomic sequences were PCR-amplified using primers with restriction enzyme sites at the 5′-end of forward and reverse primers, respectively. PCR amplified component sequences were first cloned to a T/A vector (pMD19, Takara), sequenced and digested with enzymes. The digested segments were cloned into pCAMBIA1304 and the cloning was confirmed by sequencing and restriction analysis. The confirmed clones were transformed into Agrobacterium tumefaciens strain LBA4404 by the floral dip method [30]. Positive transgenic lines were selected on the 1/2 MS medium with 50 mg L−1 kanamycin. Twenty independent 35S::AtFC1 transgenic lines (T4 homozygous lines) were obtained. Four lines were randomly selected for transcript analysis by qRT-PCR, and two of them were used for functional characterization.

Preparation of total RNA libraries and mRNA sequencing

Two week-old Arabidopsis seedlings were exposed to 0 and 50 μM Cd and sampled at 6, 12 and 24 h, respectively. Total RNA from Cd-exposed and Cd-unexposed seedlings was isolated using the TRIzol Reagent (Invitrogen, USA) and pooled for RNA sequencing. The extracted RNA was treated with DNaseI (Qiagen, USA) at 25 °C for 30 min and confirmed in quality. mRNA was purified with oligo (dT)-rich magnetic beads and broken into short fragments. The first and second strand cDNAs were synthesized. The cDNAs were end-repaired and phosphorylated using T4 DNA polymerase and Klenow DNA polymerase. The Illumina paired-end solexa adaptors were ligated to these cDNA fragments. The ligated products were purified on a 2% agarose gel. Four libraries (Col-0-Cd, fc1-Cd, Col-0 + Cd and fc1 + Cd) were sequenced using an Illumina hiseq2500.

The original image data generated by the sequence providers were transferred into nucleotide sequences data by base calling, defined as raw reads and saved as ‘fastq’ files. All subsequent analyses were performed on the high-quality clean read datasets according to the bioinformatics analysis approach summarized in Supplementary Data S6. A rigorous algorithm was used to identify differentially expressed genes (DEGs) between the samples. The expression level for each transcript was calculated as FPKM (fragments per kilobase of exon per million fragments mapped)-derived read counts based on the number of uniquely mapped reads that overlapped with exonic regions. The false discovery rate (FDR) was used to determine the threshold of the p-value in multiple tests, which corresponded to the differential gene expression test. In this study, FDR ≤ 0.001 and the absolute value of Log2Ratio > 1 were used as a threshold to judge the significant differences of gene expression. The resulting RNA-Seq data were deposited in the NCBI SRA database (Accession No. SRR5829541).

Gene ontology analysis

The Gene Ontology (GO) category of the DEGs with functional significance was subject to the ultra-geometric test with Benjamini-Hochberg correction (http://www.geneontology.org/). GO terms with corrected p-value (p ≤ 0.05) were regarded as significant enrichment for the DEGs compared to the genome background.

Statistical analyses

Each result shown in the figures was the mean of three biological replicates and each treatment contained at least 15–20 seedlings. The significant differences between treatments were statistically evaluated by standard deviation and analysis of variance (ANOVA). The data between differently treated groups were compared statistically by one-way ANOVA followed by the least significant difference test if the ANOVA result is significant at p < 0.05. The statistical analyses were performed with SPSS 12.0.