Clarification of the dispensability of PDX1.2 for Arabidopsis viability using CRISPR/Cas9
PDX1.2 has recently been shown to be a regulator of vitamin B6 biosynthesis in plants and is implicated in biotic and abiotic stress resistance. PDX1.2 expression is strongly and rapidly induced by heat stress. Interestingly, PDX1.2 is restricted to eudicota, wherein it behaves as a non-catalytic pseudoenzyme and is suggested to provide an adaptive advantage to this clade. A first report on an Arabidopsis insertion mutant claims that PDX1.2 is indispensable for viability, being essential for embryogenesis. However, a later study using an independent insertion allele suggests that knockout mutants of pdx1.2 are viable. Therefore, the essentiality of PDX1.2 for Arabidopsis viability is a matter of debate. Given the important implications of PDX1.2 in stress responses, it is imperative to clarify if it is essential for plant viability.
We have studied the previously reported insertion alleles of PDX1.2, one of which is claimed to be essential for embryogenesis (pdx1.2–1), whereas the other is viable (pdx1.2–2). Our study shows that pdx1.2–1 carries multiple T-DNA insertions, but the T-DNA insertion in PDX1.2 is not responsible for the loss of embryogenesis. By contrast, the pdx1.2–2 allele is an overexpressor of PDX1.2 under standard growth conditions and not a null allele as previously reported. Nonetheless, upregulation of PDX1.2 expression under heat stress is impaired in this mutant line. In wild type Arabidopsis, studies of PDX1.2-YFP fusion proteins show that the protein is enhanced under heat stress conditions. To clarify if PDX1.2 is essential for Arabidopsis viability, we generated several independent mutant lines using the CRISPR-Cas9 gene editing technology. All of these lines are viable and behave similar to wild type under standard growth conditions. Reciprocal crosses of a subset of the CRISPR lines with pdx1.2–1 recovers viability of the latter line and demonstrates that knocking out the functionality of PDX1.2 does not impair embryogenesis.
Gene editing reveals that PDX1.2 is dispensable for Arabidopsis viability and resolves conflicting reports in the literature on its function.
KeywordsVitamin B6 CRISPR-Cas9 T-DNA insertion Mutant alleles Heat-stress Heat shock element Arabidopsis thaliana PDX1.2 Viability
Clustered Regularly Interspersed Short Palindromic Repeats
real-time quantitative reverse transcription PCR
transcriptional start site
Yellow fluorescent protein
The coenzyme pyridoxal 5′-phosphate (PLP) is essential for all organisms due to its involvement in numerous core metabolic enzyme reactions. PLP is biosynthesized de novo in plants, whereas animals must take it in their diet as vitamin B6. The pathway for biosynthesis of PLP in plants comprises just two enzymes PDX1 (PYRIDOXINE SYNTHASE1) and PDX2 (PYRIDOXINE SYNTHASE 2) [26, 27, 28]. These two enzymes form a complex that functions as a glutamine amidotransferase that utilizes ribose 5-phosphate, glyceraldehyde 3-phosphate and glutamine as substrates to facilitate PLP biosynthesis. There are three homologs of PDX1 in Arabidopsis (PDX1.1, PDX1.2 and PDX1.3), while there is only one homolog of PDX2 . Of the PDX1s, only PDX1.1 and PDX1.3 are catalytic enzymes, whereas PDX1.2 is non-catalytic and is regarded as a pseudoenzyme. The term pseudoenzyme refers to the burgeoning family of proteins that are very similar to catalytic counterparts but do not function as enzymes . In the case of PDX1.2, its classification as a pseudoenzyme refers to the fact that while it is expressed, key active site residues required for coordinating PLP biosynthesis are not conserved, rendering the protein catalytically non-functional . The role of PDX1.2 appears to be as a positive regulator of PLP biosynthesis under environmental stress conditions, particularly under heat stress [4, 17]. Indeed, known homologs of PDX1.2 from various plant species have a heat-shock element (HSE) in the region upstream of the translational start codon that binds heat shock transcription factors of the A1 family (HSFA1) . Upregulation of PDX1.2 abundance under heat stress appears to enhance stability of the catalytic PDX1s and thus sustain PLP levels under these conditions [4, 17], but the precise mechanistic details remain to be deciphered. In an attempt to provide insight into the nature of the interaction of PDX1.2 with its catalytic counterparts, we recently solved the X-ray crystal structures of complexes of PDX1.2 with PDX1.3 . While the structure of the heteromeric PDX1.2-PDX1.3 complex is very similar to the structure of the PDX1.3 homomeric complex , we were unable to decipher key features that contribute to enhancement of PLP biosynthesis by PDX1.2, due to statistical disorder resulting from the near structural identity of both proteins. Nonetheless, recent studies of PDX1.2 have revealed a number of interesting features that include its restriction to eudicota and its important contribution to plant fitness . Studies in which expression of PDX1.2 was knocked down by RNA interference have rendered the plants more susceptible to disease, e.g. in tomato  and in Arabidopsis ; or more sensitive to abiotic stress, e.g. in Arabidopsis . Interestingly, one study has reported that PDX1.2 is essential for embryogenesis in Arabidopsis , based on the analysis of a T-DNA insertion mutant line of the SAIL collection . However, a more recent study indicated that a null allele of PDX1.2 was viable . The study of additional mutant alleles of PDX1.2 would help to clarify the effect of knocking out PDX1.2 functionality on Arabidopsis growth and development and resolve these contradictory reports. Furthermore, the study of pseudoenzymes has been neglected for a long time, due to their consideration as non-functional genes. However, pseudoenzymes are tightly conserved in sequence and although they have lost their catalytic capabilities, it is suggested that they have gained new functions . A majority of those identified new functions are fulfilled by an allosteric control of bona fide enzymes (usually their catalytic homologs), e.g. in the case of kinases by either acting as a scaffold within a signaling cascade or by modulating the activity of gene expression modulators, such as transcription factors [10, 19]. Thus, although our knowledge of pseudoenzymes is still limited, they appear as regulators that can modulate physiological responses. In this context, it would be unexpected to find that a null mutation is embryonic lethal in a pseudoenzyme like PDX1.2, which is only induced in stress conditions. It is therefore important to clarify if PDX1.2 is essential for Arabidopsis viability.
While the large collections of Arabidopsis T-DNA insertion and EMS mutants have played a crucial role in investigations of gene function, the newly developed RNA-guided endonuclease-mediated targeted mutagenesis with the Clustered Regularly Interspersed Short Palindromic Repeats (CRISPR)/Cas9 methodology  facilitates the isolation of additional independent alleles. Having multiple mutant alleles to rigorously test hypotheses for gene functionality has been limiting in some previous studies. Indeed, the relative ease with which independent alleles can be generated with CRISPR-Cas9 technology is already proving to be important in the revision of original mechanistic hypothesis, e.g. the molecular function of the RIN protein in tomato . Tomato rin mutants were isolated over half a century ago and were distinguished by an inability to ripen, lack of red pigmentation, as well as inability to soften or induce an ethylene burst , as is normally observed in climacteric fruit ripening . Thus, RIN was assumed to be an activator, indispensable for induction of fruit ripening. Moreover, the discovery of rin led to this allele being integrated into food used for global consumption for decades. The recent isolation of CRISPR-Cas rin mutant alleles that ripen in the absence of RIN and associated studies demonstrate that the original rin mutant isolated is a gain of function mutant (not a loss of function mutant as previously assumed), which produces a protein that actively represses ripening, rather than activating it . The study by Ito and colleagues  provides an excellent example of how the modern approach of gene editing technology can be used to improve our understanding of gene function and refine strategies for application of the findings.
Here we report on the use of CRISPR-Cas9 to clarify the notion that PDX1.2 is essential for Arabidopsis viability. We studied both of the previously described T-DNA insertion mutants and generated several additional mutants of PDX1.2 using CRISPR-Cas9. We report on the phenotype of loss of function mutants of PDX1.2, which clarify the dispensability of this gene under standard growth conditions and provide insight into PDX1.2 transcriptional regulation. We also show that the PDX1.2 protein accumulates under heat stress conditions. These findings are important in the possible deployment of PDX1.2 during applied studies in conferring tolerance to environmental stresses.
Examination of PDX1.2 mutant lines available from collections
A second T-DNA insertion line, CS872273, which we have annotated here as pdx1.2–2 was also obtained from the European Arabidopsis Stock Center. Our molecular analysis revealed that the T-DNA insertion was located in the promoter region of PDX1.2 at position − 202 bp before the ATG translational start codon (Fig. 1a). Seeds homozygous for the T-DNA insertion could be isolated from segregating progeny. This mutant line has previously been reported as a null mutant allele for PDX1.2 expression but in stark contrast to pdx1.2–1 is reported to be viable and indistinguishable from wild type under standard growth conditions . Indeed in our hands, growth of pdx1.2–2 under a 16-h photoperiod at 22 °C, 8 h of darkness at 18 °C, (ambient CO2, 60% relative humidity) did not distinguish it from wild type plants grown under the same conditions (hereafter referred to as standard conditions). However, in contrast to the study of Zhang and colleagues , our quantitative real-time RT-PCR (qPCR) analysis of pdx1.2–2 grown under standard conditions indicated increased expression of PDX1.2 in this line compared to wild type (Fig. 1b, left panel). Expression of PDX1.2 has recently been shown to be transcriptionally upregulated by heat stress  and is mediated by the HSFA1 transcription factor family . Interestingly, here we observed that induction of PDX1.2 by heat stress was considerably attenuated in pdx1.2–2 compared to wild type under the same conditions (Fig. 1b, right panel).
In the context of the above observations, we noted that the T-DNA insertion in pdx1.2–2 (at − 202 bp relative to the ATG translational start site) is just upstream of the HSE (at − 165 to − 153 bp relative to the ATG translational start site) in the promoter region of PDX1.2 (Fig. 1a). It has been shown previously that there are two transcriptional start sites (TSS1 and TSS2) in the region immediately upstream of the translational start site in PDX1.2  at bp − 156 and bp − 83, respectively (Fig. 1a). TSS1 is within the HSE to which HSFA1 binds to mediate induction of PDX1.2 expression under heat stress conditions. Consequently, TSS2 that is downstream of the HSE is used as an additional or alternative transcriptional start site under these conditions . To provide more insight into the observations with the pdx1.2–2 mutant, we quantified the level of transcripts starting from TSS1 (using a primer pair that anneals either side of TSS2 (Fig. 1a)) in the absence and presence of heat stress. PDX1.2 transcript levels as a function of TSS1 were similar under both conditions in wild type (Fig. 1c), i.e. increased transcript abundance under heat stress is predominantly derived from TSS2 . By contrast, enhanced expression observed in the absence of heat stress in pdx1.2–2 was considerably attenuated in the presence of heat-stress and in fact similar to wild-type transcript levels as a function of TSS1 (Fig. 1c). In this context, it is interesting to know that HSFB1 and HSFB2b are suppressors of heat stress induced genes  and it was previously noted that PDX1.2 transcript levels are enhanced in a global transcriptome analysis of the hsfb1 hsfb2b double mutant under standard growth conditions [4, 8]. Indeed, we could confirm increased expression of PDX1.2 in the absence of heat stress and show that overall expression of PDX1.2 was enhanced in the presence of heat stress in hsfb1 hsfb2b compared to wild type (Fig. 1d). Moreover, quantification of transcripts derived from the use of TSS1 in PDX1.2 in the presence and absence of heat stress, show that while they remain the same in the wild type, they are increased under heat stress in the hsfb1 hsfb2b mutant (Fig. 1e). This observation is consistent with the hypothesis that HSFA1 and HBFB1/2b regulate expression of PDX1.2 probably through the HSE. We postulate that the close proximity of the T-DNA insertion to the HSE in pdx1.2–2 interferes with the binding of HSFB1/2b and HSFA1 and therefore contributes to the transcriptional increase under standard growth conditions and to the transcriptional attenuation observed under heat stress conditions and could be studied in more elaborative studies of the heat stress response in the future.
Taken together, we conclude that pdx1.2–2 can be considered as an overexpression mutant under standard growth conditions and an underexpression mutant under heat stress conditions.
PDX1.2 protein levels are induced by heat stress
Generation of PDX1.2 mutant lines by CRISPR-Cas9 activity
Phenotypic analyses of PDX1.2 mutant lines
PDX1.2 has been reported to be critical for embryo development based on an analysis of a single T-DNA insertion mutant line SAIL_640_D11 (pdx1.2–1) . In this report, it was not clear if the observed phenotype was a combination of gametophytic and embryogenesis impairment or solely the latter due to non-conforming segregation ratios. Given the results of our Southern blot analysis, we decided to further probe SAIL_640_D11 as a function of PDX1.2. In the first instance, based on the assumption that the mutant pdx1.2–3 generated in this study is a loss of function mutant (i.e. truncated protein missing the N terminus and unlikely to fold properly), we performed reciprocal crosses of heterozygous pdx1.2–1 (PDX1.2 pdx1.2–1) with pdx1.2–3 (CRISPR1). If PDX1.2 is indeed critical for embryo development as reported and pdx1.2–3 is non-functional, then we would not expect to find pdx1.2–1 pdx1.2–3 in the F1 generation. On the other hand, if pdx1.2–3 can overcome the mutation in pdx1.2–1 then the progeny, analyzed at the seedling stage, should segregate 1:1 for pdx1.2–1 pdx1.2–3 and PDX1.2 pdx1.2–3. A segregation analysis of PDX1.2 gene and T-DNA insertion specific PCR of F1 seedlings derived from this cross demonstrated that approximately half of the progeny carry the T-DNA insertion of pdx1.2–1 (Fig. 4b). Thus, we can conclude that pdx1.2–3 can overcome the mutation in pdx1.2–1. If pdx1.2–3 is indeed non-functional, then loss of PDX1.2 is not critical for embryogenesis.
To probe this further, we performed a similar analysis using another of the CRISPR mutants generated, pdx1.2–5. The premature stop codon in the pdx1.2–5 mutant leads to a protein that is estimated to be only half the size (19.2 kDa if translated from the first ATG and/or 13.6 kDa if a product is translated from an ATG after the premature stop codon) of the mature protein (33.8 kDa) and cannot be functional based on the necessity to fold as a β/α8 barrel [17, 23]. Reciprocal crosses of heterozygous pdx1.2–1 (PDX1.2 pdx1.2–1) with pdx1.2–5 (CRISPRB) and a segregation analysis of PDX1.2 gene and T-DNA insertion specific PCR of F1 seedlings derived from this cross showed that approximately half of the progeny carry the T-DNA insertion of pdx1.2–1 (Fig. 4c). Therefore, pdx1.2–5 (as for pdx1.2–3) can overcome the mutation in pdx1.2–1.
Taken together, we conclude that loss of PDX1.2 is not critical for embryogenesis in Arabidopsis.
The CRISPR-Cas system that originated from prokaryotes as an adaptive immunity tool has been exploited extensively in the last few years as a precision tool to achieve genome editing. The technology is powerful in that several independent mutations can be generated relatively quickly in most organisms. Here we used the system to generate mutants of PDX1.2, as previous reports on phenotypes of T-DNA insertion mutant lines were inconsistent. The first T-DNA insertion mutant line (pdx1.2–1) studied, led to the claim that PDX1.2 is necessary for embryogenesis, as the pdx1.2–1 mutant was unviable . However, from our study here, it is clear that there are several T-DNA insertions in this line, which would explain the previously reported incongruities and inexplicable segregation outcomes with respect to PDX1.2 . Moreover, we have shown here that crossing the pdx1.2–1 line with independent CRISPR mutants of non-functional PDX1.2, allows us to isolate mutants that carry both the T-DNA insertion from pdx1.2–1 and the PDX1.2 CRISPR mutations. Therefore, these crossed lines effectively rescue the embryogenic defect and show that PDX1.2 is dispensable for growth under the conditions used (i.e. standard laboratory conditions). On the other hand, the T-DNA insertion mutant line, pdx1.2–2, was previously described as a loss of function line or null allele . The latter study did not report the location of the T-DNA insertion. However, our analyses indicate that the insertion is at − 202 bp upstream of the ATG translational start site. This places the T-DNA insertion just upstream of the validated HSE (at − 165 to − 153 bp relative to the ATG translational start site) in the PDX1.2 promoter region . This is relevant because although a strong induction of PDX1.2 expression is observed under heat stress in wild type plants and is under control of the HSFA1 transcription factor family, we previously hypothesized that expression may be actively repressed under ambient conditions . The latter hypothesis stemmed from the observation that PDX1.2 is among the set of genes induced under ambient conditions in a global transcriptome analysis of the hsfb1 hsfb2b double mutant . Here, we provide supporting evidence for this hypothesis by specifically examining the response of PDX1.2 expression in the absence and presence of heat stress in hsfb1 hsfb2b compared to wild type (Fig. 1d). The HSFB1 and HSFB2b heat shock factors suppress expression of heat-shock inducible genes under ambient conditions and are thought to mediate their function through the HSE-like consensus sequences . Indeed, the hsfb1 hsfb2b mutant is considered to be in a constitutive moderate heat stress state, with several heat shock response genes induced in this mutant . Thus, given the location of the T-DNA insertion in pdx1.2–2, i.e. relatively close to the HSE, and the enhanced expression of PDX1.2 in pdx1.2–2 under ambient conditions, we are tempted to speculate that the suppression of expression is impaired in this mutant, i.e. explaining enhanced expression of PDX1.2 under ambient conditions in this mutant. In addition, the T-DNA insertion may interfere with HSFA1 binding to the HSE under heat stress conditions, which would account for the observed attenuation of the response in pdx1.2–2 compared to wild type under these conditions.
The analysis of PDX1.2-YFP fusion proteins corroborate the previous hypothesis that PDX1.2 is upregulated by heat stress and may serve to stabilize its catalytic counterparts, i.e. PDX1.1 or PDX1.3 in Arabidopsis . Indeed, we have recently been able to solve the crystal structure of the complex of PDX1.2 with PDX1.3 . Unfortunately, statistical disorder prevented us from distinguishing PDX1.2 from PDX1.3 in the complex. Nevertheless, we could conclude that although PDX1.2 has very subtle effects on the conformation of PDX1.3, it likely serves to prime key catalytic regions for functionality in vitamin B6 biosynthesis. It is interesting that PDX1.2 protein accumulation is observed in the guard cells under heat stress conditions. It is well established that one of the first physiological responses of Arabidopsis (and many other plants) to heat stress takes place at the guard cells. Thus, this may be the site where PDX1.2 action is required most during such stress conditions.
In this study, we have examined several mutant alleles of PDX1.2 with the aim of defining the dispensability of this gene under standard laboratory growth conditions for Arabidopsis viability. Specifically, we characterized the previously reported null alleles of PDX1.2 and generated additional null alleles using the CRISPR-Cas9 technology. Our studies show that PDX1.2 is dispensable for growth under normal conditions and its absence does not induce embryo lethality as previously reported. Moreover, one of the previously reported pdx1.2 null alleles is in fact enhanced in PDX1.2 expression under standard conditions but attenuated under heat stress conditions, likely due to misregulation in the vicinity of the HSE involving the transcription factors HSFA1 and HSFB1/2b. Our data reinforces the fact that definition of gene functionality requires the rigorous analysis of multiple alleles and that consistent outcomes as well as conclusions can be readily achieved in this era by making use of current gene editing techniques. Indeed, this study provides a good example of the use of CRISPR-Cas to resolve issues related to conflicting reports on the functionality of particular genes.
Plant material and growth conditions
Arabidopsis thaliana (Columbia-0 ecotype) was used throughout. The T-DNA insertion mutant lines SAIL_640_D11  and CS872273  annotated as pdx1.2–1 in  and pdx1.2–2 in this study, respectively, were obtained from the European Arabidopsis Stock Center (NASC). The hsfb1–1 hsfb2b-1 (hsfb1 hsfb2b) seeds were a generous gift from Masaru Ohme-Takagi, National Institute of Advanced Industrial Science and Technology, Japan. Seeds cultivated in sterile culture were surface-sterilized in 70% ethanol (v/v) and dried prior to plating on half-strength MS medium without vitamins  (Duchefa, http://www.duchefa.com) containing 0.8% agar (w/v) in Petri dishes. Seeds cultivated under non-sterile conditions were sown on soil (Einheitserde, Classic Ton Kokos). Seeds were stratified for two to four days at 4 °C in the dark before transfer to a growth incubator (CLF Climatics CU-22 L for sterile cultures; CLF Climatics AR-66 for soil grown cultures). Plants were grown either under long-day (16 h) or short day (8 h) photoperiods (100 to 150 μmol photons m− 2 s− 1generated by fluorescent lamps [Philips Master T-D Super 80 18 W/180]) at 22 °C and 60% relative humidity, followed by either 8 h or 16 h of darkness at 18 °C, respectively, all at ambient CO2. Eight-day-old seedlings grown under long-day photoperiods were used for heat stress experiments. Heat stress at 37 °C was achieved by transferring the seedlings to an incubator (CLF Climatics I-30Bl4/D) at the defined temperature with the remaining conditions as above (100–150 μmol photons m2 s− 1, 60% relative humidity and ambient CO2) for 1–3 h, as indicated. Plant lines carrying the pdx1.2–1 and pdx1.2–2 T-DNA insertions were verified by PCR analysis of genomic DNA (see Additional file 2: Table S1 for oligonucleotides used). The level of expression of PDX1.2 in the respective lines was verified by qPCR (see below). Plants homozygous for either pdx1.2–3 (CRISPR1) or pdx1.2–5 (CRISPRB) were crossed with those heterozygous for pdx1.2–1 (PDX1.2/pdx1.2–1). The F1 progeny were analyzed, which are heterozygous for either pdx1.2–3 or pdx1.2–5 and PDX1.2/pdx1.2–1.
Southern blot analysis was carried out using genomic DNA isolated from leaves of heterozygous lines of pdx1.2–1 (PDX1.2 pdx1.2–1) and the corresponding wild type (Col-0) plants, grown under long-day conditions. DNA samples (15 mg) were digested overnight at 37 °C with either the HindIII, EcoR1, SacI or NcoI (150 U) restriction enzyme, as indicated, in a final volume of 75 μl followed by electrophoresis on a 0.7% agarose gel. The digested samples were probed with an anti-digoxigenin (DIG) labeled fragment matching a portion of the BAR gene, which forms part of the T-DNA construct used to produce the SAIL collection . The probe was prepared via PCR amplification from the pDAP101 plasmid using primers CGAAATAAACGACCAAATTAGTAGAA and ACCCTATAAGAACCCTAATTCCCTTAT (Additional file 2: Table S1). The probe DNA was labeled using the PCR DIG probe synthesis mix (Sigma-Aldrich), hybridized overnight in DIG EasyHyb buffer containing DIG-AP Fab fragments (Roche) against neutrally-charged nylon membranes (0.45 μm pore size, Nytran). Membranes were prepared for imaging with the DIG luminescence detection kit (Sigma-Aldrich), before exposure against Super RX X-ray film (Fujifilm).
For gene expression analyses by real-time quantitative reverse transcription PCR (qPCR), tissue samples were collected from 8-day-old seedlings grown under long-day conditions. RNA was extracted using the PureLink RNA Mini kit (Ambion) according to the manufacturer’s instructions. DNA was removed by an on-column DNase digest during the RNA extraction. Reverse transcription was performed using 0.5 μg total RNA as a template, Superscript II reverse transcriptase (200 U) and oligo (dT)20 primers (500 ng) (Thermo Fisher Scientific). qPCR was performed in 384 well plates on a 7900HT Fast Real-Time PCR system (Applied Biosystems) using Power SYBR Green master mix (Applied Biosystems) and the following amplification program: 10 min denaturation at 95 °C followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min. The data were analyzed using the comparative cycle threshold method (2−ΔCT) normalized to the Arabidopsis reference gene GAPDH (At1g13440). Each experiment was performed with three biological and three technical replicates. In all cases, primer pairs used are given in Additional file 2: Table S1.
Generation of PDX1.2-YFP lines, molecular analysis and confocal microscopy
For expression of PDX1.2-YFP, PDX1.2 without its stop codon was amplified from cDNA of 8-days-old seedlings using Phusion proofreading polymerase (ThermoFisher) and specific primer pairs (PDX1.2-YFP F, GGGGACAAGTTTGTACAAAAAAGCAGGCTATGGCGGATCAAGCTATGAC and PDX1.2-YFP R, GGGGACCACTTTGTACAAGAAAGCTGGGTAACACTGCCTTGGCCAAAGTC). The amplified products were purified and cloned into the pDONR221 vector by the BP recombination reaction using BP Clonase™ II mix (ThermoFisher) according to the manufacturer’s instructions to generate pDONR221:PDX1.2-YFP, sequenced, and subsequently cloned into the destination vector pB7YWG2  by an LR reaction using the LR Clonase™ mix II (ThermoFisher) to generate pB7YWG2::PDX1.2-YFP. Afterwards, the region comprising bp − 1 to − 1495 upstream of the ATG translational start codon of PDX1.2 was amplified from genomic DNA of 8-days-old seedlings using the primer pair pPDX1.2::PDX1.2-YFP F, AATATGAGCTCTTAATTATCTCTCTCAATGAG and pPDX1.2::PDX1.2-YFP R, ATATTAACTAGTTTTTAGGTTCTGTGAGTTTTTAGTAACAG, where the regions in italics denote implemented SacI and SpeI restriction sites, respectively. The amplicon was digested with SacI and SpeI, purified and ligated into similarly digested and purified pB7YWG2 to replace the CaMV 35S promoter to generate pPDX1.2::pB7YWG2. Subsequently, pDONR221::PDX1.2-YFP and pPDX1.2::pB7YWG2 were recombined by an LR reaction using the LR Clonase™ mix II (ThermoFisher) to generate pB7YWG2::pPDX1.2::PDX1.2-YFP. In this case, PDX1.2 will be expressed as a fusion protein with YFP at the C terminus. The construct and the empty vector (pB7YWG2) as a control were introduced into Agrobacterium tumefaciens strain C58 and used to transform wild type (Col-0) Arabidopsis plants by the floral dip method . As the respective constructs contain the BAR gene, transformants were selected by resistance to BASTA™. Resistant plants were allowed to self-fertilize, and homozygous lines were selected from the T3 generation according to their segregation ratio for BASTA™ resistance. Eight-days-old transgenic seedlings grown in sterile culture under long-day conditions, as described above were used for confocal microscopy. The heat stress treatment was performed by exposing seedlings to 37 °C for 3 h before microscopic analysis. In each case, seedlings were mounted in water between the slide and the coverslip with a double-sided Scotch tape spacer. They were imaged with an SP5 confocal laser-scanning microscope (Leica) equipped with a resonant scanner and a × 40 Oil, numerical aperture 1.25 PlanApo lens. The pinhole was set at 1 Airy unit and the zoom was set so that the pixel size was between 140 and 150 nm. YFP was excited at 514 nm and a HyD detector collected its emission between 519 nm and 570 nm. The software Fiji  was used to process images and to mount selected z slices, which were colorized with the “Fire” look-up table. Col-0 and PDX1.2-YFP samples were all imaged and contrasted with identical parameters; 35S-driven YFP samples, displaying a significantly higher expression, were imaged with lower laser power.
Fluorescence intensity quantifications were performed using Fiji  as follows: All measurements were performed on epidermal cells and only on the top-most slices of each z stack, so that signal loss due to tissue depth was negligible. On each image, a Gaussian blur (radius 0.6 pixel) was applied to reduce noise, and the average background (shot noise and detector offset) was subtracted. A line was drawn across the interface between two adjacent cells (two rhizodermal cells in roots, or a guard cell and a pavement cell in cotyledons) and the maximum intensity value encountered along the line was retained. In these thin-walled cells, the resolution of our imaging setup did not allow to discriminate the two thin cytoplasm strands of adjacent cells, which were then seen as a single line: the value measured thus gave a proxy of the average cytosolic intensity of the cell pair considered. To increase the robustness of the measurement and to reduce the uncertainty due to cytosolic fluctuations and to the presence of sub-resolution organelles, for each cell pair 5 such measurements were made, and their average value was calculated. At least two plants were analyzed for each genotype and condition, and the number of cell pairs measured in case were the following: pPDX1.2::PDX1.2-YFP L1 (nCotyledons -HS = 61; nCotyledons + HS = 63; nRoots -HS = 44; nRoots + HS = 37); pPDX1.2::PDX1.2-YFP L3 (nCotyledons -HS = 19; nCotyledons + HS = 30; nRoots -HS = 21; nRoots + HS = 24); Col-0 (nCotyledons -HS = 12; nCotyledons + HS = 8; nRoots -HS = 23; nRoots + HS = 41); 35S-YFP (nCotyledons -HS = 43; nCotyledons + HS = 67; nRoots -HS = 54; nRoots + HS = 33).
For immunochemical analysis, total protein was extracted from 8-days-old whole seedlings using the same protocol as described in  employing 50 mM sodium phosphate buffer, pH 7.0, containing 5 mM β-mercaptoethanol, 10 mM EDTA, 0.5% Triton X-100 [v/v], 0.1 mM PMSF, and 1% [v/v] complete plant protease inhibitor cocktail (Sigma-Aldrich) and subsequent centrifugation (16,000 g). The supernatant was decanted, and 30 μg of total protein (measured using the Bradford assay kit (Bio-Rad ;) was separated on a 12% SDS-polyacrylamide gel. Immunodetection was carried out employing primary antibodies against GFP (SC8334, Santa Cruz Biotechnology, Inc.) and Actin (A0480, Sigma-Aldrich, as a loading control), both at 1:5000 dilution, the corresponding secondary antibody (peroxidase conjugated goat anti-rabbit, Bio-Rad) at 1:10000 dilution, the iBlot system (Invitrogen) and the SNAP i.d. 2.0 system (Millipore) as described by Colinas and colleagues . Chemiluminescence was detected using western Bright ECL (Advansta) and captured using an ImageQuant LAS 4000 system (GE Healthcare).
Generation of PDX1.2 mutants by CRISPR-Cas9
Mutations were introduced into Arabidopsis PDX1.2 through the Non-Homologous End Joining (NHEJ) method of RNA-guided endonuclease-mediated targeted mutagenesis with the Clustered Regularly Interspersed Short Palindromic Repeats (CRISPR)-Cas9 system [15, 16], employing the plasmids pEN-Chimera and pDE-Cas9  in the first instance, which were a kind gift from Prof. Holger Puchta (Karlsruhe Institute for Technology, Germany). Briefly, the specific guide RNA (sgRNA) was designed using an online tool http://crispr.mit.edu/. See Additional file 2: Table S1 for the sequence used. Two self-annealing oligonucleotides were designed with a 22 bp guide sequence targeting from + 84 in the PDX1.2 coding sequence. For annealing, the primers were mixed in equimolar concentrations (2 μM each), denatured at 95 °C for 5 mins, cooled to ambient temperature followed by ligation (T4 DNA ligase) into BbsI digested gel-purified pEN-Chimera (1 μg) to generate pEN-Chimera-sgPDX1.2. The Arabidopsis U6–26 promoter and PDX1.2 sgRNA (using 150 ng pEN-Chimera-sgPDX1.2) were then transferred into the destination vector pDE-Cas9 (150 ng) by the Gateway LR recombination reaction using the LR Clonase™ II enzyme to generate pDE-Cas9-sgPDX1.2. The construct was introduced into Agrobacterium tumefaciens strain C58 and used to transform wild type Arabidopsis plants (Col-0) by the floral dip method . As the pDE-Cas9-sgPDX1.2 construct contains the BlpR gene, transformants were selected by resistance to phosphinothricin. Resistant plants (24 in C1) were allowed to self-fertilize. Plants of the C2 generation were checked for mutations by amplifying the region of PDX1.2 (see Additional file 2: Table S1 for oligonucleotides used) expected to harbor Cas9 mutagenesis and identifying DNA heteroduplexes (http://www.crisprflydesign.org/t7-endo-i-assay/) by T7 endonuclease (Biolabs). Digestion of the amplicon by the T7 endonuclease indicated a Cas9 driven mutation and was verified by sequencing (Microsynth AG, Switzerland). Two independent transgenic lines with single nucleotide insertions in PDX1.2 leading to a premature stop codon (CRISPR1, A inserted at + 89 bp; CRISPR2, C inserted at + 84 bp) were identified in this generation, however, only one (CRISPR1) could be isolated to homozygosity in the C3 generation and was annotated pdx1.2–3. Seeds harboring a single integration event were subsequently screened for loss of the Cas9 expression cassette (see Additional file 2: Table S1 for oligonucleotides used).
Mutations were also introduced into Arabidopsis PDX1.2 with the NHEJ method of RNA-guided endonuclease-mediated targeted mutagenesis CRISPR-Cas9 system of Wang and colleagues , employing the plasmids pHEE2A-TRI and pHEE-401E, which were a kind gift from Prof. Qi-Jun Chen (China Agricultural University, PRC). The vectors for targeting PDX1.2 were constructed as follows: The sequences between target 1 (+ 182 to + 204 for CRISPRA or + 436 to + 458 for CRISPRB in PDX1.2) and target 2 (+ 225 to + 247 for CRISPRA or + 487 to + 509 for CRISPRB in PDX1.2) in pHEE2A-TRI was amplified with two pairs of primers: either DT1-F0A and DT2-R0A or DT1-F0B and DT2-R0B (see Additional file 2: Table S1). The two pairs of PCR products were purified and then re-amplified with the following primers: either DT1-BsFA and DT2-BsRA or DT1-BsFB and DT2-BsRB (see Additional file 2: Table S1). The products were purified and used for cloning into the pHEE-401E plasmid using the BsaI restriction site to generate pHEE401E-sgPDX1.2A and pHEE401E-sgPDX1.2B followed by transformation into E. coli DH5α. Bacterial transformants were selected first on kanamycin, propagated and rechecked for kanamycin and streptomycin resistance. Colonies that were resistant only to kanamycin were chosen. Constructs were verified by sequencing (Microsynth AG, Switzerland) and introduced into Agrobacterium tumefaciens strain C58 and used to transform wild type Arabidopsis plants (Col-0) by the floral dip method . Plant transformants were selected by resistance to hygromycin (7–8 per construct). The region of PDX1.2 expected to harbor Cas9 mutagenesis was amplified in resistant plants of the C1 generation (see Additional file 2: Table S1 for oligonucleotides used) and screened by sequencing (Microsynth AG, Switzerland). Two independent transgenic lines with single nucleotide insertions in PDX1.2 leading to a premature stop codon were identified in the C1 generation, allowed to self-fertilize and isolated to homozygosity. Seeds harbouring a single integration event were subsequently screened for loss of the Cas9 expression cassette. The lines isolated to homozygosity were annotated pdx1.2–4 (CRISPRA, T inserted at bp + 187) and pdx1.2–5 (CRISPRB, C inserted at bp + 502), respectively. Potential off-targets were examined using the CRISPRdirect (https://crispr.dbcls.jp/)  (see Additional file 3: Table S2).
We acknowledge the technical assistance of Mireille de Meyer-Fague and Céline Roux (both University of Geneva) for parts of this study. We thank the European Arabidopsis Stock Centre (NASC) for seeds of SAIL-640-D11 and CS872273. We thank Masaru Ohme-Takagi, National Institute of Advanced Industrial Science and Technology, Japan for generously providing the hsfb1-1 hsfb2b-1 seeds. We are grateful to Prof. Holger Puchta (Karlsruhe Institute for Technology, Germany) for providing pEN-Chimera and pDE-Cas9 and Prof. Qi-Jun Chen (China Agricultural University, PRC) for providing pHEE2A-TRI and pHEE-401E.
T.B.F. and E.D. designed the research; E.D. performed most of the experimental work and contributed research tools; I.D. performed molecular plant work, crosses and phenotype analyses; S.L. performed the confocal microscopy; T.B.F., I.D., S.B. and E.D. analyzed data; T.B.F. wrote the article with input from all authors. All authors read and approved the final manuscript.
This work was supported by the Swiss National Science Foundation (grants PPOOA_1191186 and 31003A-141117/1 to T.B.F.) and by the University of Geneva. The funding bodies did not play any role in the design of the study nor in the collection, analysis and interpretation of data, nor in the writing of the manuscript.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
- 23.Robinson GC, Kaufmann M, Roux C, Martinez-Font J, Hothorn M, Thore S, Fitzpatrick TB. Crystal structure of the pseudoenzyme PDX1.2 in complex with its cognate enzyme PDX1.3: a total eclipse. Acta Crystallogr. D Struct. Biol. 2019;75:400–15.Google Scholar
- 24.Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012;9:676–82.CrossRefGoogle Scholar
- 25.Sessions A, Burke E, Presting G, Aux G, McElver J, Patton D, Dietrich B, Ho P, Bacwaden J, Ko C, Clarke JD, Cotton D, Bullis D, Snell J, Miguel T, Hutchison D, Kimmerly B, Mitzel T, Katagiri F, Glazebrook J, Law M, Goff SA. A high-throughput Arabidopsis reverse genetics system. Plant Cell. 2002;14:2985–94.CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.