The Tyrosyl-DNA Phosphodiesterase 1β (Tdp1β) Gene Discloses an Early Response to Abiotic Stresses

Tyrosyl-DNA phosphodiesterase 1 (Tdp1) is involved in DNA repair pathways as it mends the topoisomerase I—DNA covalent complexes. In plants, a small Tdp1 gene family, composed by Tdp1α and Tdp1β genes, was identified, but the roles of these genes in abiotic stress responses are not fully understood. To investigate their specific stress response patterns, the present study made use of bioinformatic and molecular tools to look into the Tdp1β gene function, so far described only in the plant kingdom, and compare it with Tdp1α gene coding for the canonical, highly conserved α isoform. The expression profiles of Tdp1α and Tdp1β genes were examined under abiotic stress conditions (cold, heat, high osmolarity, salt, and UV-B) in two model species, Arabidopsis thaliana and Medicago truncatula. The two isoforms of topoisomerase I (TOP1α and TOP1β) were also taken into consideration in view of their known roles in DNA metabolism and cell proliferation. Data relative to gene expression in Arabidopsis were retrieved from the AtGenExpress microarray dataset, while quantitative Real-Time PCR was carried out to evaluate the stress response in M. truncatula cell cultures. These analyses revealed that Tdp1β gene expression was enhanced during the first hour of treatment, whereas Tdp1α enhanced expression succeeded at subsequent timepoints. In agreement with the gene-specific responses to abiotic stress conditions, the promoter regions of Tdp1α and Tdp1β genes are well equipped with stress-related cis-elements. An in-depth bioinformatic characterization of the HIRAN motif, a distinctive feature of the Tdp1β protein, showed its wide distribution in chromatin remodeling and DNA repair proteins. The reported data suggests that Tdp1β functions in the early response to abiotic stresses.


Introduction
Tyrosyl-DNA phosphodiesterase 1 (Tdp1, EC: 3.1.4.-) breaks the covalent 3 -phosphotyrosyl bond between the DNA termini and the catalytic tyrosine residue of DNA topoisomerase I (topo I), removing the highly cytotoxic stabilized topoisomerase I-DNA covalent complexes that impair DNA replication and transcription [1]. Differently from animals, two distinct Tdp1 isoforms are found in plants, as reported by Macovei et al. [2] in the model legume Medicago truncatula. The MtTdp1α amino acid sequence shows similarity with the animal Tdp1 enzyme, while the MtTdp1β protein contains a HIRAN (HIP116 Rad5p N-terminal) domain whose function is still poorly understood [2]. The two isoforms are encoded by MtTdp1α and MtTdp1β genes, which are upregulated in M. truncatula plants in response to heavy metal and osmotic stresses as well as during seed imbibition [2][3][4]. Additional work provided further information on the possible involvement of the Tdp1β gene in the plant stress response. Santos et al. [5] showed the upregulation of Tdp1β gene in Medicago sativa suspension cultures exposed to genotoxic doses of CdSe/ZnS quantum dots, highlighting a strong correlation with DNA damage accumulation/repair kinetics. An RNA-Seq analysis carried out in MtTdp1α-depleted M. truncatula plants revealed that the MtTdp1β gene does not play a redundant function since the β isoform was not able to compensate for the lack of MtTdp1α gene in planta [6]. Even though the Tdp1β gene was upregulated in the depleted lines, the overall Tdp1(α + β) transcript was reduced, leading to the dwarf phenotype resulted from the Tdp1α depletion [6].
The specific role of the Tdp1β gene in plants remains unclear. Here, we propose to investigate the expression profiles of the Tdp1 gene family in response to multiple abiotic stress agents in two model plants, Arabidopsis thaliana (thale cress) and Medicago truncatula (barrel medic). The two different model species, belonging to Brassicaceae and Leguminosae families, respectively, were used to evaluate the conservative response among species. The promoter regions of Tdp1α and Tdp1β genes were also examined with bioinformatic tools. Additionally, the presence and distribution of the HIRAN domain in the plant kingdom were investigated.

Plant Material and Treatments
Suspension cultures of M. truncatula cv. Jemalong, M9-10a genotype were propagated in liquid Murashige-Skoog (MS) medium [7] containing 3% (w/v) sucrose, 0.5 mg L −1 2,4-dichlorophenoxyacetic acid (2,4-D, Micropoli, Milan, Italy), and 0.5 mg L −1 kinetin (Micropoli), at pH 5.7. The suspension cultures were sub-cultured every eight days and maintained at 23 • C, in agitation (80 rpm) under dark conditions. Highly proliferating four-day-old suspension cultures were used for further treatments. To perform the stress treatments, 10 mL suspension culture was transferred to 50 mL Falcon tubes, maintained in agitation. The following treatments were imposed to the cell suspensions: salt (NaCl, 250 mM), high osmolarity (PEG6000, 150 g L −1 ), cold (4 • C), heat (42 • C), and UV-B (15 min exposure to 280-315 nm). For the NaCl and PEG treatments, the solutions were added to the liquid MS medium. For the cold and heat treatments, the tubes were incubated at the indicated temperatures. Exposure to UV-B was performed in opened Petri dishes, containing the same volume of suspension, which was circularly moved every 2 min during the time of exposure. For all treatments, samples were collected at 0 h, 0.5 h (30 min), 1 h, 3 h, and 6 h following exposure.

Detection of Cell Death
Loss of plasma membrane integrity was assessed through Evans Blue staining, as previously described [8]. Proliferating M. truncatula cell cultures subjected to the abovementioned treatments for 6 h were used for this analysis. Briefly, Evans Blue (Sigma-Aldrich, Milan, Italy) stock solution (10 mg/mL) was added to cell suspension samples (1 mL) to a final concentration of 0.025% (v/v). After 10 min of incubation at room temperature, the culture was extensively washed with distilled water to remove excess and unbound dye. Subsequently, Evans Blue bound to dead cells was extracted using 50% (v/v) methanol with 1% (w/v) SDS at 60 • C for 30 min and the absorbance was read at 600 nm using a Jasco 7800 UV/Vis Spectrophotometer (JASCO, Easton, MD, USA). Non-treated samples and cells subjected to heat-shock (HS, 65 • C for 10 min) were used and positive and negative controls, respectively. Images were recorded using an Olympus SZX9 Stereomicroscope (Olympus Italia S.R.L., Milan, Italy). For each treatment combination, three independent and two biological replicas were used. Results are presented as % of cell death, where absorbance of heat-shock (HS) treated cells is considered as 100%. The results were statistically evaluated by t-test (*, p < 0.05), with a non-treated control taken as a reference.
For both Arabidopsis and Medicago collected data, the selected timepoints per treatment were 0 h, 0.5 h (30 min), 1 h, 3 h, and 6 h after treatment. The data is presented as fold change to control (untreated samples), with each timepoint being normalized to its corresponding control. The data not normalized to the untreated control is provided in Supplementary Tables S1 and S2, respectively. MeV (Multiple Experiment Viewer) software (http://mev.tm4.org) was used to represent the data.

Promoter Analysis
The PlantCARE database [13] was used for the in silico analysis of promoter sequences. Sequences corresponding to 1500 bp upstream of the start codon were retrieved from NCBI for each gene. The percentage of stress-related cis-elements was calculated based on the total number of elements identified for each promoter region. A Venn diagram (http://bioinformatics.psb.ugent.be/ webtools/Venn/) analysis was used to assess the number of common vs. specific elements for each promoter sequence.

Phylogenetic Tree Analysis
To investigate the occurrence of HIRAN domain in the plant kingdom, the Tdp1β amino acid sequences were compared using the ExPaSy SIB BLAST Network Service (http://www.expasy.ch/ tools/blast//) and the EMBL-EBI Clustal W2 Multiple Sequence Alignment (http://www.ebi.ac.uk/ tools/msa/clustalw2) tools. The phylogenetic tree was built using the Plaza 2.5 tool (bioinformatics. psb.ugent.be/plaza/) [14].

Effect of Treatments on Cell Viability
M. truncatula cell cultures treated with cold, heat, PEG, NaCl, UV-B, and respective positive (CTRL NT) and negative (CTRL HS) controls, were subjected to Evans Blue staining to evaluate cellular mortality ( Figure 1). Measurements were taken after 6 h of treatment, because this is indicated as the minimum threshold required for the induction of programmed cell death [15]. The treatments that most affected cell viability were heat, NaCl, and UV-B, while PEG and cold treatments did not show significant changes compared to the non-treated control (CTRL NT) (Figure 1a). M. truncatula M9-10a suspension cultures are composed mainly of microcalli (Figure 1b), in accordance with a previous description defining them as highly embryogenic cell aggregates of small spherical cells [16]. as the minimum threshold required for the induction of programmed cell death [15]. The treatments that most affected cell viability were heat, NaCl, and UV-B, while PEG and cold treatments did not show significant changes compared to the non-treated control (CTRL NT) (Figure 1a). M. truncatula M9-10a suspension cultures are composed mainly of microcalli (Figure 1b), in accordance with a previous description defining them as highly embryogenic cell aggregates of small spherical cells [16].

The Expression Profiles of Both Tdp1β and Tdp1α Genes Change during Abiotic Stress Treatments
The expression levels of Tdp1α and Tdp1β genes were evaluated in response to various abiotic stress conditions in two different model plants, A. thaliana and M. truncatula ( Figure 2). The analysis of Arabidopsis gene expression in shoots shows that the strongest expression (green colour) of AtTdp1α corresponds to 6 h following exposure to heat and high osmolarity, and 3-6 h after treatment with salt (Figure 2a, AtTdp1α; Supplementary Table S1). Conversely, the lowest expression (red colour) is observed at 3 h following exposure to heat and UV-B. When considering the expression profiles of AtTdp1β, the strongest expression is evident mostly within the first hours of treatment (0.5 h high osmolarity, salt, cold, UV-B, and 1 h salt stress). The lowest gene expression is associated with exposure to heat stress (at 0.5 h, 1 h, and 3 h), as well as at later timepoints after cold ( Table S1).
To assess whether the early response to abiotic stress observed in Arabidopsis is a conserved feature of the plant Tdp1β gene, the expression profiles of the Tdp1 gene family were analyzed in a different system, namely actively proliferating M. truncatula in vitro cell cultures. Also in this case, the MtTdp1β gene expression is strongest soon after the beginning of the treatment (0.5 and 1 h) with salt, high osmolarity agent, and UV-B, while heat stress resulted in the lowest expression (Figure 2b Table S2).
The early (30 min to 1 h) expression profiles of the Tdp1β gene are thus maintained in both A. thaliana and M. truncatula model systems. Differences in the expression patterns in response to time and treatments could be due to the use of different methods to quantify gene expression (microarray in Arabidopsis and qRT-PCR in barrel medic) and different plant material (shoots vs. cell suspensions).

The Expression Profiles of Both Tdp1β and Tdp1α Genes Change during Abiotic Stress Treatments
The expression levels of Tdp1α and Tdp1β genes were evaluated in response to various abiotic stress conditions in two different model plants, A. thaliana and M. truncatula ( Figure 2). The analysis of Arabidopsis gene expression in shoots shows that the strongest expression (green colour) of AtTdp1α corresponds to 6 h following exposure to heat and high osmolarity, and 3-6 h after treatment with salt (Figure 2a, AtTdp1α; Supplementary Table S1). Conversely, the lowest expression (red colour) is observed at 3 h following exposure to heat and UV-B. When considering the expression profiles of AtTdp1β, the strongest expression is evident mostly within the first hours of treatment (0.5 h high osmolarity, salt, cold, UV-B, and 1 h salt stress). The lowest gene expression is associated with exposure to heat stress (at 0.5 h, 1 h, and 3 h), as well as at later timepoints after cold ( Table S1).
To assess whether the early response to abiotic stress observed in Arabidopsis is a conserved feature of the plant Tdp1β gene, the expression profiles of the Tdp1 gene family were analyzed in a different system, namely actively proliferating M. truncatula in vitro cell cultures. Also in this case, the MtTdp1β gene expression is strongest soon after the beginning of the treatment (0.5 and 1 h) with salt, high osmolarity agent, and UV-B, while heat stress resulted in the lowest expression (Figure 2b Table S2).
The early (30 min to 1 h) expression profiles of the Tdp1β gene are thus maintained in both A. thaliana and M. truncatula model systems. Differences in the expression patterns in response to time and treatments could be due to the use of different methods to quantify gene expression (microarray in Arabidopsis and qRT-PCR in barrel medic) and different plant material (shoots vs. cell suspensions).
To assure that the chosen treatments affect DNA metabolism and cell proliferation, the expression of topoisomerase I α and β isoforms was also evaluated. In the Arabidopsis system, AtTOP1α is mostly downregulated, while upregulation is evident only after 6 h of treatment with heat, NaCl, and UV-B (Figure 2a, AtTOP1α; Supplementary Table S1). AtTOP1β is highly expressed both at 3 and 6 h after treatment with PEG, NaCl, and UV-B (Figure 2a, AtTOP1β; Supplementary  Table S1). When considering the cell suspension system, both MtTOP1α and MtTOP1β were highly expressed after 1 h of treatment, and mostly downregulated following the 1 h time point (Figure 2b, Supplementary Table S2). In is important to note that the MtTdp1 and MtTOP1 genes followed similar patterns of expression in M. truncatula cell cultures (e.g., upregulation during the first hours of treatment vs. downregulation at later timepoints). Another point is that all genes were downregulated during treatment with heat, which also resulted in high cell mortality rates, indicating that high temperature is the most repressive affliction in our cell culture system. To assure that the chosen treatments affect DNA metabolism and cell proliferation, the expression of topoisomerase I α and β isoforms was also evaluated. In the Arabidopsis system, AtTOP1α is mostly downregulated, while upregulation is evident only after 6 h of treatment with heat, NaCl, and UV-B (Figure 2a, AtTOP1α; Supplementary Table S1). AtTOP1β is highly expressed both at 3 and 6 h after treatment with PEG, NaCl, and UV-B (Figure 2a, AtTOP1β; Supplementary  Table S1). When considering the cell suspension system, both MtTOP1α and MtTOP1β were highly expressed after 1 h of treatment, and mostly downregulated following the 1 h time point (Figure 2b, Supplementary Table S2). In is important to note that the MtTdp1 and MtTOP1 genes followed similar patterns of expression in M. truncatula cell cultures (e.g., upregulation during the first hours of treatment vs. downregulation at later timepoints). Another point is that all genes were downregulated during treatment with heat, which also resulted in high cell mortality rates, indicating that high temperature is the most repressive affliction in our cell culture system.

In Silico Analysis of Tdp1β and Tdp1α Promoter Sequences Reveals Abundant Stress-Related cis-Elements
The cis-elements found in the four promoter regions (1500 bp upstream from the start codon) of AtTdp1α, AtTdp1β, MtTdp1α, and MtTdp1β genes are quite diverse. Nonetheless, all four promoters are well equipped with stress-related cis-elements (Figure 3). Among these, the light-responsive and hormone-responsive elements are the most abundant. Another well-represented class belongs to the defense and stress response, accounting for 15.4% and 13.4% of the overall cis-elements in AtTdp1α and MtTdp1β (Figure 3a). To assess the number of common and different cis-elements per promoter, a Venn diagram was constructed. The analysis showed that 10 cis-elements are common to all four promoters (Figure 3b). These include cis-elements involved in anaerobic induction (ARE), light response (Box 4, Box I, G boxes), heat stress (HSE), salicylic acid response (TCA-element), endospermspecific expression (Snk1_motif), as well as cis-regulatory elements which act as enhancer (CAATbox) and transcription factors binding sites (TATA-box) (Supplementary Table S3). When considering the elements specific to each promoter region, the MtTdp1α has the highest number of specific elements (10) when compared with the other promoter regions. These include elements involved in auxin response (TGA-element, AuxRR-core), meristem-specific expression (CAT-box), low temperature (LTR) and light response (3-AF1 binding site, GATA-motif, GTGGC-motif, I-box), the binding site of AT-rich DNA binding protein (ATBP-1) (AT-rich element), and the cis-acting

In Silico Analysis of Tdp1β and Tdp1α Promoter Sequences Reveals Abundant Stress-Related cis-Elements
The cis-elements found in the four promoter regions (1500 bp upstream from the start codon) of AtTdp1α, AtTdp1β, MtTdp1α, and MtTdp1β genes are quite diverse. Nonetheless, all four promoters are well equipped with stress-related cis-elements (Figure 3). Among these, the light-responsive and hormone-responsive elements are the most abundant. Another well-represented class belongs to the defense and stress response, accounting for 15.4% and 13.4% of the overall cis-elements in AtTdp1α and MtTdp1β (Figure 3a). To assess the number of common and different cis-elements per promoter, a Venn diagram was constructed. The analysis showed that 10 cis-elements are common to all four promoters (Figure 3b). These include cis-elements involved in anaerobic induction (ARE), light response (Box 4, Box I, G boxes), heat stress (HSE), salicylic acid response (TCA-element), endosperm-specific expression (Snk1_motif), as well as cis-regulatory elements which act as enhancer (CAAT-box) and transcription factors binding sites (TATA-box) (Supplementary Table S3). When considering the elements specific to each promoter region, the MtTdp1α has the highest number of specific elements (10) when compared with the other promoter regions. These include elements involved in auxin response (TGA-element, AuxRR-core), meristem-specific expression (CAT-box), low temperature (LTR) and light response (3-AF1 binding site, GATA-motif, GTGGC-motif, I-box), the binding site of AT-rich DNA binding protein (ATBP-1) (AT-rich element), and the cis-acting regulatory element, the OBP-1 site. In the AtTdp1α promoter region, the six specific elements are involved in abscisic acid (ABA) and viviparous1 (VP1) responsiveness (CE3), lignin biosynthesis (AC I, AC II), and light response (GA-motif, Gap-box, Box II). The detected elements specific only to AtTdp1β include cis-elements involved in gibberellin response (GARE-motif), differentiation of the palisade mesophyll cells (HD-Zip I), control of leaf morphology (HD-Zip 2), and light response (4cl-CMA2b, CATT-motif, MRE). Lastly, the specific cis-elements encountered in the MtTdp1β promoter regions are involved in gibberellin (P-box) and light response (chs-CMA2a, GAG-motif, ATCT-motif, AT1-motif).
Overall, the promoter region of the four genes investigated are both similar and diverse, considering the type and number of cis-elements present, respectively. regulatory element, the OBP-1 site. In the AtTdp1α promoter region, the six specific elements are involved in abscisic acid (ABA) and viviparous1 (VP1) responsiveness (CE3), lignin biosynthesis (AC I, AC II), and light response (GA-motif, Gap-box, Box II). The detected elements specific only to AtTdp1β include cis-elements involved in gibberellin response (GARE-motif), differentiation of the palisade mesophyll cells (HD-Zip I), control of leaf morphology (HD-Zip 2), and light response (4cl-CMA2b, CATT-motif, MRE). Lastly, the specific cis-elements encountered in the MtTdp1β promoter regions are involved in gibberellin (P-box) and light response (chs-CMA2a, GAG-motif, ATCT-motif, AT1-motif).
Overall, the promoter region of the four genes investigated are both similar and diverse, considering the type and number of cis-elements present, respectively.   (Figure 4b). In the HIRAN-HKD structure, present in the plant Tdp1β protein, the HIRAN motif is flanked by two interacting HKD motifs, producing an active site for the phosphodiesterase activity [1]. The HIRAN-HKD organization includes also an FHA motif, present in several eukaryotic nuclear proteins, which mediates phosphorylation-dependent protein-protein interactions [17]. The FHA domain recognizes phosphopeptides arising from hyperphosphorylation mediated by cell-cycle checkpoint kinases in response to DNA damage [17]. The HIRAN-SMARCA domain organization is typical for the A-SMARC (SWI/SNF-Related, Matrix-associated, Actin-dependent Regulator Chromatin) group of proteins, represented by DNA-dependent ATPases able to modify histone-DNA interactions and modulate chromatin organization [18]. It also includes other domains, e.g., the SNF2 involved in transcription regulation, recombination, and chromatin remodeling [19], and the C3HC4 type zincfinger motif with functions in the ubiquitination pathway [20]. Additionally, the HIRAN-SMARCA   (Figure 4b). In the HIRAN-HKD structure, present in the plant Tdp1β protein, the HIRAN motif is flanked by two interacting HKD motifs, producing an active site for the phosphodiesterase activity [1]. The HIRAN-HKD organization includes also an FHA motif, present in several eukaryotic nuclear proteins, which mediates phosphorylation-dependent protein-protein interactions [17]. The FHA domain recognizes phosphopeptides arising from hyperphosphorylation mediated by cell-cycle checkpoint kinases in response to DNA damage [17]. The HIRAN-SMARCA domain organization is typical for the A-SMARC (SWI/SNF-Related, Matrix-associated, Actin-dependent Regulator Chromatin) group of proteins, represented by DNA-dependent ATPases able to modify histone-DNA interactions and modulate chromatin organization [18]. It also includes other domains, e.g., the SNF2 involved in transcription regulation, recombination, and chromatin remodeling [19], and the C3HC4 type zinc-finger motif with functions in the ubiquitination pathway [20]. Additionally, the HIRAN-SMARCA domain organization is found in the Helicase C-terminal motif, typical for the SF1 and SF2 superfamilies [21]. The HIRAN-VRR-NUC domain, poorly characterized in plants, is found in human cells within the highly conserved protein KIAA1018/FAN1 (Fanconi anemia-Associated Nuclease 1), a DNA repair nuclease recruited to damaged DNA [22] which localizes to stalled replication forks coordinating S-phase arrest and DNA repair [23]. In plants, this domain is present in the KIAA1018-like protein in association with an N-terminal Rad18 zinc finger region and C-terminal tetratricopeptide repeats (TPRs) [24].

The HIRAN Domain is Ubiquitously Distributed in Plants
The phylogenetic analysis shows that the HIRAN-HKD domain organization is detected in Dicots, such as the model plant A. thaliana, the legumes M. truncatula and Glycine max, and some tropical species (Figure 4c). The HIRAN-HKD domain organization is also present in monocots but absent in the moss Physcomitrella patens. Both the HIRAN-SMARCA and HIRAN-VRR-NUC domain organizations are widely distributed in the plant kingdom. Indeed, the most widespread domain structure involving the HIRAN motif in eukaryotes is found in the N-terminus of the SWI2/SNF2 ATPases required for the activation of cell-cycle checkpoints [25]. domain organization is found in the Helicase C-terminal motif, typical for the SF1 and SF2 superfamilies [21]. The HIRAN-VRR-NUC domain, poorly characterized in plants, is found in human cells within the highly conserved protein KIAA1018/FAN1 (Fanconi anemia-Associated Nuclease 1), a DNA repair nuclease recruited to damaged DNA [22] which localizes to stalled replication forks coordinating S-phase arrest and DNA repair [23]. In plants, this domain is present in the KIAA1018like protein in association with an N-terminal Rad18 zinc finger region and C-terminal tetratricopeptide repeats (TPRs) [24]. The phylogenetic analysis shows that the HIRAN-HKD domain organization is detected in Dicots, such as the model plant A. thaliana, the legumes M. truncatula and Glycine max, and some tropical species (Figure 4c). The HIRAN-HKD domain organization is also present in monocots but absent in the moss Physcomitrella patens. Both the HIRAN-SMARCA and HIRAN-VRR-NUC domain organizations are widely distributed in the plant kingdom. Indeed, the most widespread domain structure involving the HIRAN motif in eukaryotes is found in the N-terminus of the SWI2/SNF2 ATPases required for the activation of cell-cycle checkpoints [25].

Discussion
Plant Tdp genes are far less studied when compared with their human counterparts. Moreover, the presence of multiple Tdp1 genes in plants [2] further expands the need to design dedicated studies to investigate their implication in plant development and stress response. In human cells, Tdp1 is strongly linked with complex functions in DNA repair [26], whereas mutations in its catalytic sites are associated with serious diseases (e.g., spinocerebellar ataxia) [27]. Studies conducted so far in plants have proven the involvement of the Tdp1α gene in DNA damage repair and stress response

Discussion
Plant Tdp genes are far less studied when compared with their human counterparts. Moreover, the presence of multiple Tdp1 genes in plants [2] further expands the need to design dedicated studies to investigate their implication in plant development and stress response. In human cells, Tdp1 is strongly linked with complex functions in DNA repair [26], whereas mutations in its catalytic sites are associated with serious diseases (e.g., spinocerebellar ataxia) [27]. Studies conducted so far in plants have proven the involvement of the Tdp1α gene in DNA damage repair and stress response [2,6,28]. Moreover, the lack of the canonical Tdp1 isoform (Tdp1α) was associated with phenotypic defects (e.g., dwarfism) in both Arabidopsis [28] and M. truncatula [6]. On the other side, the Tdp1β gene is far less characterized. In M. truncatula, it was shown to be ubiquitously expressed in all plant tissues and developmental stages [2,3]. As the MtTdp1β gene was not able to compensate for the deficiency of MtTdp1α gene in M. truncatula transgenics [6], we hypothesize that the two genes might not have an overlapping function, although both of them play certain roles in response to abiotic stresses [2]. To further assess this hypothesis, here we investigated the expression profiles of the Tdp1β gene, alongside the canonical Tdp1α gene, in two model species, A. thaliana and M. truncatula. Several abiotic stress conditions (cold, heat, salinity, osmotic stress, and UV-B) were imposed. Our analysis showed that while Tdp1β is strongly expressed at the earliest timepoints (0.5-1 h) following exposure to stress in both Arabidopsis and M. truncatula (Figure 2), the expression of the Tdp1α gene is more variegated between the two species. The early stress response pattern of Tdp1β gene is conserved between the two species in spite of the different methods used to quantify gene expression (microarray in Arabidopsis and qRT-PCR in M. truncatula) and different plant material (shoots vs. cell suspensions generated from aerial parts). The use of microarrays to evaluate gene expression levels is quite widespread, while this often requires validation by qRT-PCR. Nonetheless, most studies using both methods agree on a high level of equivalency among them [29][30][31]. When considering the different type of material used, a recent study showed the occurrence of similar mitotic indexes and gene expression profiles in M. truncatula leaves grown in a greenhouse and calli-cultured in vitro [32]. Likewise, our work using M. truncatula overexpressing the MtTdp2α gene revealed a high degree of similarity in the expression pattern of several DNA repair genes in both plants grown in vitro and cell suspension cultures [10,33]. Hence, the differences encountered in the expression levels of Tdp1α and Tdp1β among Arabidopsis and M. truncatula are most likely species-specific.
The percentage of cellular mortality and expression of TOP1 isoforms were assessed to gain more insight into cell behaviour during treatments. As DNA topoisomerase I is involved in solving the conformational changes in DNA topology, it plays essential roles in several cellular processes (e.g., replication, transcription, recombination) [34] as well as in the response to stress agents [35][36][37]. It can also act as a damage sensor and cofactor in DNA repair pathways [38,39]. Moreover, the topo I-DNA covalent complexes represent the substrate for the activity of Tdp1 [26,40]. Due to their central role in DNA metabolism, decreased activity of the TOP1 gene negatively influences cell culture growth and vitality [15]. The results presented here corroborate with this finding well, as enhanced cellular mortality (Figure 1) corresponded to a decrease in MtTOP1α and MtTOP1β gene expression ( Figure 2) at 6 h following treatments. Downregulation of TOP1 could be associated with a temporary block of cell-cycle progression to allow more time for DNA repair. In M. truncatula cell cultures, higher expression of TOP1 genes after 1 h of treatment was observed in parallel with higher expression of Tdp1 genes, suggesting for an early and coordinated activation of these genes under stress conditions. It is reasonable to hypothesize that these may be acting as damage sensors rather than repair activities, as further exposure to stress (e.g., subsequent timepoints) resulted in decreased gene expression.
The activation or suppression of gene expression is tightly regulated at the transcriptional level through the activity of gene promoters and related cis-acting elements. As a consequence, the regulation of gene transcription revolves around the type, number, position, and combination of regulatory elements present inside and around the promoter [41]. As we detected different expression patterns in Tdp1α and Tdp1β genes in response to stress, we investigated their promoter regions in the two model species. In agreement with the gene-specific responses to abiotic stress conditions, the promoter regions of all four genes are adequately equipped with stress-related cis-elements (Figure 3). Common as well as divergent cis-elements are encountered in the promoter regions of the four genes. Focusing on the different types of cis-elements, the hormone-responsive elements are worth specifying because abiotic stress responses are strongly correlated with hormonal signaling [42]. Our analysis showed that ABA-responsive elements are present only in AtTdp1α, and auxin-responsive elements are present only in MtTdp1α, while gibberellin-responsive elements are common to both AtTdp1β and MtTdp1β promoter sequences. On the other hand, salicylic acid-responsive cis-elements are encountered in all four promoter regions. It is well established that phytohormones and their corresponding cross-talk pathways play pivotal roles during plant development and stress response [43]. Another point to be raised is the presence of cis-elements with roles in lignin biosynthesis in the AtTdp1α promoter region. This is important because lignin, as the main structural component of the cell wall, is involved in the overall plant stress management [44]. Moreover, we showed that in M. truncatula, MtTdp1α-depleted plants have a reduced permeability of the cell wall and ticker cuticula than their wild-type counterparts, and this was associated with altered expression of defense genes and high susceptibility to stress [45]. A recent study showed the presence of common metabolites in the formation of lignin and cuticular biopolymers in mosses, suggesting that the pre-lignin pathway may be crucial for the formation of cuticular elements [46]. However, despite the fact that the cis-element promoter analysis represents an informative tool to figure out differences in gene expression profiles, future promoter validation studies are required to corroborate the in silico investigation.
To further unfold the reasons of Tdp1β early stress response, we investigated the presence and distribution of the HIRAN domain (conserved in many DNA processing proteins) in plants. This analysis showed that the HIRAN-HKD domain, specific to the Tdp1β protein, along with the HIRAN-SMARCA and HIRAN-VRR-NUC conformations, are ubiquitously found in plants ( Figure 4). The structure of the HIRAN domain was recently reported to be composed of six β-strands and two α-helices, forming an OB (oligonucleotide/oligosaccharide binding)-fold structure commonly found in single strand DNA (ssDNA) binding proteins [47], while the DNA binding site of the free domain displayed high degrees of conformational heterogeneity [48]. It has been hypothesized that the HIRAN motif might act in the recruitment of repair/remodeling enzymes to specific DNA sites, playing a role in cell cycle-checkpoints arising from stalled replication forks and post-replication damage [49]. In human cells, the filling-in of gaps in damaged DNA during replication, carried out by the HLTF (helicase-like transcription factor) function, is dependent on the HIRAN domain [50], which also promotes the HLTF-dependent fork reversal, playing roles in DNA damage tolerance [51]. One of the main tasks of the DNA damage tolerance pathway is to minimize fork stalling, pushing for the bypass of replication blocks [52]. In plants, the HIRAN domain was investigated in relation to AtRAD5A, a DNA translocase that catalyzes the fork regression, and was shown to be able to bind to branched DNA structures and promote DNA repair [53]. However, all the cited literature refers to the HIRAN-SMARCA type of architecture, while no information on the HIRAN-HKD is available in plants. Based on the assumption that the HIRAN domain could act as a sensor to initiate the repair processes at damaged DNA checkpoints [49], we hypothesize that the presence of this motif in the Tdp1β protein might sustain the early response of the gene to abiotic stress conditions. Nonetheless, further experimental studies, such as targeted modifications of the Tdp1β HIRAN domain, are needed to prove this theory.
Supplementary Materials: The following are available online at www.mdpi.com/2073-4425/8/11/305/s1, Table S1: Arabidopsis thaliana AtTdp1α, AtTdp1β, and AtTOP1α, AtTOP1β gene expression profiles retrieved from the AtGenExpress microarray abiotic stress dataset. Data are presented as the mean of three replicates ± standard deviation, Table S2: Medicago truncatula MtTdp1α, MtTdp1β, and MtTOP1α, MtTOP1β gene expression profiles obtained from qRT-PCR analysis. Data are presented as the mean of three replicates ± standard deviation, Table S3: Promoter cis-elements identified using the PlantCARE online tool. The function of each cis-element is specified, alongside with the frequency of presence in the promoter regions of AtTdp1α, AtTdp1β, MtTdp1α, and MtTdp1β.