These authors contributed equally to this work.
NITRILASE1 regulates the exit from proliferation, genome stability and plant development
Article first published online: 25 FEB 2013
© 2013 The Authors. New Phytologist © 2013 New Phytologist Trust
Volume 198, Issue 3, pages 685–698, May 2013
How to Cite
Doskočilová, A., Kohoutová, L., Volc, J., Kourová, H., Benada, O., Chumová, J., Plíhal, O., Petrovská, B., Halada, P., Bögre, L. and Binarová, P. (2013), NITRILASE1 regulates the exit from proliferation, genome stability and plant development. New Phytologist, 198: 685–698. doi: 10.1111/nph.12185
- Issue published online: 12 APR 2013
- Article first published online: 25 FEB 2013
- Manuscript Accepted: 9 JAN 2013
- Manuscript Received: 20 NOV 2012
- Centre of the Region Haná for Biotechnological and Agricultural Research. Grant Numbers: P501/12/2333, 204/09/H084, IAA500200719
- Grants Agency of the Czech Republic. Grant Numbers: P501/12/2333, 204/09/H084
- Grant Agency of the ASCR. Grant Numbers: IAA500200719, NCZ.1.05/2.1.00/01.0007
- cell cycle;
- NITRILASE 1;
- programmed cell death
- Top of page
- Material and Methods
- Supporting Information
- Nitrilases are highly conserved proteins with catabolic activity but much less understood functions in cell division and apoptosis. To elucidate the biological functions of Arabidopsis NITRILASE1, we characterized its molecular forms, cellular localization and involvement in cell proliferation and plant development.
- We performed biochemical and mass spectrometry analyses of NITRILASE1 complexes, electron microscopy of nitrilase polymers, imaging of developmental and cellular distribution, silencing and overexpression of nitrilases to study their functions.
- We found that NITRILASE1 has an intrinsic ability to form filaments. GFP-NITRILASE1 was abundant in proliferating cells, distributed in cytoplasm, in the perinuclear area and associated with microtubules. As cells exited proliferation and entered differentiation, GFP-NITRILASE1 became predominantly nuclear. Nitrilase silencing dose-dependently compromised plant growth, led to loss of tissue organization and sustained proliferation. Cytokinesis was frequently aborted, leading to enlarged polyploid cells. In reverse, independently transformed cell lines overexpressing GFP-NITRILASE1 showed slow growth and increased rate of programmed cell death.
- Altogether, our data suggest that NITRILASE1 homologues regulate the exit from cell cycle and entry into differentiation and simultaneously are required for cytokinesis. These functions are essential to maintain normal ploidy, genome stability and tissue organization.
- Top of page
- Material and Methods
- Supporting Information
There is a rapid expansion of sequenced genomes spanning all the kingdoms, yet hundreds of conserved proteins exist for which general biochemical activities are known but biological functions are only predicted. Nitrilases, members of the C-N hydrolase superfamily catalyzing hydrolysis of nitriles into carboxylic acids and ammonia, are on the top of this list of ‘conserved hypotheticals’ (Galperin & Koonin, 2004). In Arabidopsis, there are four nitrilases NITRILASE1–4, of which NITRILASE1 (NIT1) shows the highest homology to animal and bacterial nitrilases. Following the plant nomenclature, animal orthologues are named NITRILASE1 (NIT1).
In the family of Brassicaceae, NIT1 has an enzymatic function in cyanide detoxification and catabolism of cyanogenic glycosides (Vorwerk et al., 2001). NIT1 was also connected to the tryptophan-mediated alternative pathway of auxin biosynthesis from indole-3-acetonitrile (Michalczuk et al., 1992). However, more recently it has been shown that NIT1 homologues have only very low activity on an indole-3-acetonitrile substrate (Agerbirk et al., 2008). Therefore, the most accepted enzymatic functions for nitrilases in plants are in detoxification and nitrogen recycling, while their importance for auxin biosynthesis seems to be minor (Piotrowski, 2008).
There is accumulating evidence for a biological role of nitrilases beyond their enzymatic functions (Huebner et al., 2011). Bacterial and fungal nitrilases have an ability to assemble into filaments in vitro (Thuku et al., 2007; Kaplan et al., 2011). In Arabidopsis leaves, the GFP fusion of NIT1 was observed as aggregated foci in the vicinity of the nuclei after wounding or herbicide treatment during the early steps of plant programmed cell death (PCD; Cutler & Somerville, 2005). NIT1 was also implicated in PCD in animal cells (Semba et al., 2006; Sun et al., 2009). In addition, NIT1 in animal cells is a tumour suppressor gene; it was shown to negatively regulate T-cell proliferation (Zhang et al., 2009). Curiously, in some organisms, in Drosophila melanogaster and in Caenorhabditis elegans, NIT1 forms a translational fusion with the FRAGIL HISTIDIN TRIAD (FHIT) protein, suggesting that these proteins function together (Pekarsky et al., 1998). FHIT is a well-characterized tumour suppressor gene inducing apoptosis and inhibiting tumorogenesis (Wali, 2010). In other organisms, for example in mouse, NIT1 and FHIT are encoded by separate genes, but the expression and function of NIT1 and FHIT overlap. Moreover, the silencing of either FHIT, or NIT1 in mouse, resulted in increased susceptibility to forestomach tumour and decreased apoptosis, induced by DNA damage (Semba et al., 2006). While tumour suppressor functions of FHIT are well understood in animal cells, the mode of action and the biological function of NIT1, either in cooperation with or independent of FHIT, is much less clear (Huebner et al., 2011).
Here, we provide evidence that in Arabidopsis NIT1 is present in high-molecular-mass polymers and forms filamentous structures similar to bacterial and fungal nitrilases. In agreement with the report of Cutler & Somerville (2005), we confirmed the aggregation of NIT1 at perinuclear locations upon induction of PCD and, in addition, we found that overproduction of NIT1 resulted in a reduced cell growth rate and higher levels of PCD. Downregulation of NIT1 homologues, NIT1–3 with RNAi silencing showed that these Arabidopsis nitrilases are important for cell cycle exit during differentiation, for normal cytokinesis and for maintaining regular ploidy levels and genome integrity. Nitrilases belong to a growing number of enzymes with function in biological processes distinct from their metabolic function. As shown by our data, Arabidopsis NIT1–3 are implicated in processes regulating cell proliferation and differentiation.
Material and Methods
- Top of page
- Material and Methods
- Supporting Information
Plant material and cell cultures
Arabidopsis thaliana cell suspension cultures of Columbia (Col) and Landsberg erecta (Ler) were grown under continuous darkness at 25°C as described (Drykova et al., 2003). Seedlings of Arabidopsis thaliana (L.) Heynh. ecotype Columbia – Columbia lines expressing PM GFP marker (Cutler et al., 2000) – were grown on half-strength MS agar plates as described (Binarova et al., 2006).
The complete coding region of Arabidopsis NIT1 was cloned using GatewayR technology. Vectors for 35S CaMV promoter driven expression for N- and C-terminal GFP fusions and for Multisite Gateway cloning of N-terminal GFP fusion of NIT1 under its native promoter were purchased from Ghent University (Ghent, Belgium). (For details, see Supporting Information Methods S1.) The constructs were transformed into Agrobacterium tumefaciens strain GV3101 which has been used for several independent transformations of Arabidopsis cultured cells and plants (Clough & Bent, 1998; Davis et al., 2009). Seedlings were selected on kanamycin (50 μg ml−1) medium. T1 transformants carrying GFP constructs had no discernible abnormalities, and we obtained T2 and T3 generations.
In order to produce a silencing RNAi construct, we cloned the 335 bp fragment of the NIT1 sequence of the central conserved part of the NIT1 molecule into the pHANNIBAL vector to produce an inverted repeat and then this NIT-RNAi cassette was transferred into the binary vector pART27 (CSIRO Plant Industry, Clayton South, Australia). (For details and primers, see Methods S1.) With the NIT-RNAi construct we performed 15 independent transformations with Columbia, GFP-MAP4 and PM GFP Columbia plants. After 7 d of kanamycin selection resistant seedlings were transferred onto nonselective media. Phenotype was analysed in 6- to 19-d-old seedlings; transformants with the empty pART27 vector were used as the controls. The NIT-RNAi plants with phenotypes could not reach maturity and produce seeds. Therefore, we characterized > 200 of the independent transformants with phenotype in the T1 generation.
Protein extracts were prepared from cultured cells and seedlings as described before (Drykova et al., 2003). (For details of differential centrifugation, see Methods S1.)
Custom polyclonal rabbit antibody (GenScript) for the NIT1 was raised against the C-terminal sequence DIARAKLYFDSVGHYSRPDV (aa 299–318) of the NIT1 molecule, affinity purified on peptide and used in dilution of 1 : 100 to 1 : 1000.
Gel filtrations were carried out from re-solubilized P100 on Superose 6 10/300 GL column (GE Healthcare Bio-Sciences, Uppsala, Sweden) according to Doskocilova et al. (2011).
Co-immunoprecipitation was performed using GFP-Trap_A (ChromoTek, Planegg-Martinsried, Germany) according to the manufacturer's instructions.
Protein identifications by MALDI mass spectrometry were performed from excised silver-stained protein bands as described (Halada et al., 2001). Mass spectra were acquired on Ultraflex III instrument (Bruker Daltonics, Billerica, MA, USA) and searched against SwissProt 2011_04 database using the in-house Mascot search engine.
The microtubule spin-down experiments were performed as described previously (Binarova et al., 2006). Three independent experiments were conducted; representative Western blots are shown.
For immunogold labelling and electron microscopy, 5-μl samples were applied directly onto glow-discharged activated Formvar/carbon coated nickel grids for 30 s and fixed in 4% PFA and 0.1% glutaraldehyde in PBS (pH 7.4) for 10 min. (For details of the immunogold labelling and negative staining, see Methods S1). The grids were examined using a Philips CM100 electron microscope at 80 kV and a magnification of ×46 000. Digital images were recorded using a MegaViewII (Olympus, Tokyo, Japan) slow scan camera at a magnification of ×64 000 giving the digital resolution of c. 1 nm per pixel.
Immunofluorescence labelling was performed as described (Drykova et al., 2003) using anti-NIT1 (1 : 100), anti-α-tubulin (1 : 2000) and anti-γ-tubulin TU-32 antibodies (1 : 6). Drugs were diluted in DMSO and used in working concentrations of: 5 μM for amiprophosmethyl (APM, A0185, Duchefa Biochemie, Haarlem, Netherlands), 10 μM for taxol (Paclitaxel, Sigma T7402), and 10 μg ml−1 for aphidicolin (Sigma, A0781) and applied as indicated.
Fluorescence microscopy and live cell imaging was performed using the Olympus IX-81 inverted microscope with Olympus DSU – Disc Scanning Confocal Unit and Olympus IX-81 FV-1000 confocal microscope (Olympus); images were taken and analysed as described previously (Doskocilova et al., 2011).
Cell cycle analysis
EdU staining of replicating cells was performed using Click-iT EdU Alexa Fluor 488 HCS Assay (Molecular Probes, Eugene, OR, USA) based on methods modified from Vanstraelen et al. (2009). Mitotic index was determined after Carnoy's fixation and 0.45% lacto-propionic-orcein staining. To determine the growth curve of the Arabidopsis cell suspension, 5 ml of cultured cells were sedimented and cells were weighed.
For TUNEL assay, cells were fixed by 3.7% paraformaldehyde, attached to poly-l-lysine slides, and permeabilised (0.1% Triton X-100 in 0.1% sodium citrate for 10 min). The labelling reaction was carried out according to the manufacturer's protocol (Roche 12156792910) and the DNA was stained by DAPI. For PCD induction, 2 mM acetylsalicylic acid (ASA, Sigma A6810) treatment for 3–4 h was applied.
For chromocentra counting cells were fixed in an ethanol:acetic acid mixture (v/v 3 : 1) at room temperature for 1 h then macerated in 0.1 M HCl for 10 min at 37°C, rinsed in water, squeezed in 45% acetic acid at 45°C, and stained with 2 μg ml−1 DAPI for 10 min in the dark.
Flow cytometry sample preparation was performed using Partec CyStain UV precise P kit (Partec, Munster, Germany). Measurements were performed on a BD LSRII (BD Biosciences, Franklin Lakes, NJ, USA) with a solid state laser (excitation 405 nm) and 450/50 band pass filter. Forward scatter and side scatter were taken at 488 nm. Data were evaluated in FlowJo (Tree Star, Ashland, OR, USA). (For full details of the materials and methods, see Methods S1.) Whenever possible, at least three independent experiments were performed.
- Top of page
- Material and Methods
- Supporting Information
NIT1 is present in high-molecular-mass forms that were identified by EM as filamentous structures
Nitrilases are self-polymerizing proteins, as shown for recombinantly expressed bacterial protein in vitro (Thuku et al., 2007). Aggregation of Arabidopsis NIT1-GFP was observed in early stages of PCD by Cutler & Somerville (2005), but high-molecular-mass forms of plant nitrilases were not characterized. There are four genes encoding nitrilases in the Arabidopsis thaliana genome (Fig. S1a). NIT1, NIT2 and NIT3 (referred to as NIT1 homologues) code for proteins sharing high amino acid similarities (84–90%), while their similarity with NIT4 is lower (66–68%). There are orthologues of NIT1 in bacteria, fungi, and animals (Fig. S1b). In order to study the biochemical properties of the protein and its cellular distribution, we prepared antibody against a peptide derived from the C-terminus of NIT1. On Western blots of crude extracts of Arabidopsis cultured cells and seedlings, the affinity purified antibody detected a band of c. 38 kDa, corresponding to the calculated NIT1 molecular mass (Fig. S2). We also prepared NIT1 C- and N-terminal GFP fusion constructs under the control of the CaMV 35S promoter (p35S::NIT1-GFP and p35S::GFP-NIT1) as well as a NIT1 N-terminal GFP fusion under the control of NIT1 promoter (pNIT1::GFP-NIT1). We transformed Arabidopsis cultured cells and plants using these constructs. The NIT1 GFP fusion proteins had the expected 65 kDa molecular mass detected on Western blots as a single band (Figs 1b, S2).
In order to learn about NIT1 molecular forms, we used differential centrifugation. Differential centrifugation from crude cellular extracts showed that endogenous NIT1 remained in the supernatant (S100) and a portion of it was pelleted to the microsomal fraction at 100 000 g (P100). At 160 000 g a further fraction of NIT1 was pelleted from the S100 (P160). The fact that NIT1 was sedimented to high-speed pellets also when differential centrifugation was performed with the nonionic detergent NP-40 suggested that higher molecular forms of NIT1 were not membrane-bound (Fig. 1a). Differential centrifugation performed from extracts of cells transformed with pNIT1::GFP-NIT1 construct showed a similar distribution of GFP-NIT1 to the endogenous NIT1 (Fig. 1b).
The presence of NIT1 in large molecular assemblies was also indicated by the broad range of molecular masses of the immunopurified GFP-NIT1 separated under native conditions by blue native PAGE (Fig. 1c). Similarly, gel filtration chromatography of the re-solubilized microsomal P100 pellets also showed a broad size distribution of NIT1 with no significant maxima, ranging from the column void volume at > 2 MDa through to fractions of c. 440-kDa-oligomeric forms (Fig. 1d). We conclude that NIT1 exists both as soluble and high-molecular-mass forms.
In order to analyse the composition of the high-molecular-mass forms of NIT1, we immunopurified the complex through the GFP tag and identified the proteins from SDS PAGE silver stained gels by MALDI-MS (Fig. S3). Together with GFP-NIT1, we found untagged endogenous NIT1 and NIT2 in almost equimolar amounts, suggesting that the large molecular mass forms are assemblies of NIT1 and NIT2.
Next we analysed the high-molecular-mass fractions from gel filtration enriched for NIT1, based on Western blot detection (Fig. 1d, asterisks), by electron microscopy (EM). Negative stain EM revealed highly ordered linear filaments of variable lengths (Fig. 2a). These polymers, of up to 200 nm or occasionally longer, exhibited helical substructures, strongly resembling the earlier described filaments of recombinant bacterial nitrilase assembled in vitro (Thuku et al., 2007). Closer inspection and measurements of several independent preparations demonstrated that the Arabidopsis putative nitrilase filaments have a uniform diameter of 12 ± 2 nm (mean ± SD, n = 356). To confirm the identity of filaments, we performed immunogold labelling of the samples from gel filtration with anti-NIT1 antibody. As shown in Fig. 2(b), the 12 ± 2 nm-diameter filamentous structures were specifically decorated by the anti-NIT1 antibody and secondary antibody coupled with gold particles. NIT1 labelling was preferentially associated with more open parts of the filaments (Fig. 2b, inset III), probably due to better accessibility of the epitope. Taken together, our biochemical analysis revealed the high-molecular-mass assemblies of Arabidopsis nitrilases that were visualized by EM as filamentous assemblies.
NIT1 is present in cytoplasm, with microtubular arrays and in the area of cell plate
We confirmed the stress-induced aggregation of NIT1 in the vicinity of nuclei described by Cutler & Somerville (2005). However, we also noticed microtubular localization in untreated proliferating cells, which we characterized further in Arabidopsis cultured cells by double-immunofluorescence labelling with affinity purified antibody against NIT1 and anti-α-tubulin antibody (Fig. 3a), or in cells transformed with pNIT1::GFP-NIT1 construct (Fig. 3b). NIT1 was mostly present in the cytoplasm, but in mitotic cells it localized with microtubules. A weak signal was associated with the preprophase band of cortical microtubules; in metaphase, NIT1 was enriched in the spindle area, and during cytokinesis NIT1 was present in the phragmoplast area and in the area of cell plate formation (Fig. 3a).
GFP-NIT1, expressed under its own promoter in several stable cultured Arabidopsis cell lines showed similar localization patterns in vivo as observed for endogenous NIT1 protein visualized by immunofluorescence (Fig. 3b). In interphase, the signals for GFP-NIT1 were present in the cytoplasm, in the vicinity of the nuclei and weakly in the nuclei. A portion of GFP-NIT1 signal was localized with the mitotic spindle and was enriched with the phragmoplast during cytokinesis. When expression was driven from the 35S CaMV promoter, the GFP signal was stronger, particularly in the cytoplasm (Videos S1, S2). However, the enrichment of 35SGFP-NIT1 with mitotic spindle and cytokinetic apparatus was comparable to that observed for GFP-NIT1 expressed from its own promoter.
The microtubule associated NIT1 signal was enriched in taxol-treated cells (Fig. 4a). To confirm the association of NIT1 with microtubules, we performed a co-sedimentation assay with taxol-induced polymerization of microtubules from high-speed S70 supernatant (Drykova et al., 2003). As shown in Fig. 4(b), NIT1 was detected with pelleted plant microtubules in the presence of taxol, but not in negative controls without taxol. However, unlike γ-tubulin that became further enriched in the pellet by the addition of taxol-stabilized brain microtubules (Drykova et al., 2003), the portion of NIT1 pelleted in the presence or absence of taxol-stabilized brain microtubules remained approximately the same. Next we depolymerized microtubules with amiprophos methyl (APM), which led to the dispersion of NIT1 in the cytosol (Fig. 4c). As we showed previously, the remnants of kinetochore microtubules left after APM treatment are abundantly decorated with γ-tubulin (Binarova et al., 2000), but hardly any NIT1 signal could be detected with the kinetochore microtubular stubs. Our data showed that a portion of NIT1 protein was associated with microtubules but not with sites of microtubule nucleation (Drykova et al., 2003).
NIT1 is abundant in proliferating cells
Signal for GFP-NIT1 under the control of its native promoter was observed predominantly in the meristematic zone of primary roots and in developing young leaves (Fig. 5a). In proliferating cells the GFP-NIT1 signal was diffusely present in the cytoplasm (Figs 5c, S4a). As in cultured cells (Fig. 3a,b) it was somewhat enriched with mitotic microtubules (Fig. 5b). As cells exit proliferation, the nuclear signal became stronger in cells at the expansion and differentiation zones of roots (Figs 5c,d, S4b).
Silencing NITRILASE1–3 expression leads to compromised differentiation and continued cell proliferation
The nit1–3 mutant allele was originally identified based on its resistance to auxin (Normanly et al., 1997), but later it was shown that IAA synthesis rates and concentrations were similar to wild-type (Ljung et al., 2005). NIT1–3 are members of a small gene family with overlapping expression pattern and possibly redundant functions. To gain insights into nitrilase function, we downregulated NIT1–3 expression in Arabidopsis plants using a dsRNA interference construct composed of a hairpin with inverted repeats, corresponding to a 335-bp fragment of the central part of the NIT1 molecule with significant sequence similarities to NIT1–3 homologues (NIT-RNAi). Several independent transformations were performed with the NIT-RNAi construct alongside an empty vector pART27 using the Arabidopsis ecotype Columbia, an Arabidopsis line expressing the GFP marker for tubulin (GFP-MAP4; Marc et al., 1998) and GFP marker for plasmatic membrane (Cutler et al., 2000). Independent of the genetic backgrounds, and the transformation events, the T1-independent transformants with NIT-RNAi showed similar phenotypes ranging from milder to severe reduction of leaf and root growth (Fig. 6a,b). The phenotypes were not observed in plants transformed with empty vector. When the first true leaves of NIT-RNAi plants developed they were small and thick (arrows in Fig. 6c), abaxial polarity and tissue organization was gradually lost and leaves turned into disorganized structures (asterisk in Fig. 6c). In the strongest cases, callus-like structures were formed at the shoot apex instead of true leaves from the very beginning (arrowhead in Fig. 6c). NIT-RNAi plants with reduced levels of NIT1–3 expression could not reach maturity and produce seeds and therefore we could only analyse seedlings in the T1 generation. To link the developmental abnormalities to the transformation with the NIT-RNAi construct, we counted the frequency of T1-independent transformants showing the phenotype obtained from independent transformations carried out at different times. We found that the phenotype always occurred with a frequency ranging from 30% to 60%, while with the empty vector this was 0%.
We determined by quantitative RT-PCR the reduction of NIT1–4 expression in c. 60 independent NIT-RNAi T1 transformants showing the characteristic phenotypes. There was c. 80% reduction in three of the nitrilases, NIT1–3 in the NIT-RNAi transformants compared to the controls (Fig. S5a). To correlate the strength of phenotypes with the reduction of NIT1 expression, we analysed independent T1 NIT-RNAi transformants with phenotypes ranging from milder to severe abnormalities in true leaf differentiation and compared the NIT1 expression levels to that of wild-type controls. The reduction of NIT1 expression in the NIT-RNAi lines correlated with the severity of the observed phenotypes (Fig. S5b). Furthermore, similar developmental defects were not found in plants transformed with other unrelated dsRNA constructs, such as NodGS (Doskocilova et al., 2011) or γ-tubulin (Binarova et al., 2006) suggesting that the phenotypes we observed are specific for the NIT-RNAi construct and are the result of the silencing of NIT1–3 expression.
Exit from cell proliferation is impaired when NITRILASE1–3 are silenced and cells became highly polyploid
In order to further analyse the leaf developmental defects, we microscopically inspected the first true leaves of 6–19-d-old seedlings from at least 100 independent NIT-RNAi seedlings with developmental abnormalities that were labelled with DAPI to stain nuclei. All of these plants showed similar type of defects at the cellular level; typical examples are shown in Fig. 7. Cells of young leaves of NIT-RNAi plants became large with grossly enlarged nuclei in comparison to the control. Also the regular line of marginal cells was disrupted and replaced by protrusions from highly enlarged cells (Fig. 7a). The enlarged cells of young malformated leaves of NIT-RNAi plants retained a high nuclei to cytoplasm volume ratio, typical for cells of meristem and leaf primordial, and did not become vacuolated. Correspondingly, cells with giant nuclei remained mitotically active. We observed all stages of mitosis from prophase to anaphase with a highly increased number of chromosomes, ranging from 20 to 60, instead of the regular diploid number of 10 in wild type (Fig. 7b,c). This suggested that the increased nuclei size did not arise through the process of endoreduplication (repeated S-phases) but likely through abnormal cell division.
In order to confirm the increased ploidy, we analysed DNA content by flow cytometry. We collected newly emerging leaves 3–4 and older leaves 1–2 from seedlings 16 d post germination from WT plants and the malformed leaves from NIT-RNAi seedlings at the same age. In wild-type control seedlings, cells of leaves 1–2 have entered endoreduplication and increased DNA content up to 8C while cells of the newly emerging leaves 3–4 were still diploid with 2C and 4C DNA content. By contrast, as shown in Fig. 7(d), for the malformed leaves of 16-d-old NIT-RNAi a grossly increased DNA content up to 32C and an almost complete lack of 2C DNA level were observed. To prove that the increased DNA content measured by flow cytometry in NIT-RNAi leaves is caused by polyploidy and not by increased level of endoreduplication, which naturally occurs in plant cells during development, we counted chromocentra. The chromocentra are formed by heterochromatin aggregates corresponding to mitotic centromeres (Fransz et al., 2002). The average number of chromocentra in NIT-RNAi plants was 18.1 ± 8.0 (n = 287) compared to 8.2 ± 1.4 (n = 224) in control plants. In addition, in cells of NIT-RNAi leaves there is a broad variation in chromocentra number (6–79; Fig. 7e,f), which also corresponds with the lack of distinct peaks in the flow cytometry measurements, and indicates high genome instability in the NIT-RNAi plants. The mitotic figures and the counting of chromocentra firmly established that increased DNA content in the NIT-RNAi cells is not the consequence of elevated levels of endoreduplication.
Taken together, our data show that nitrilases are required to repress the entry into proliferation and in their absence cells become largely polyploid.
NITRILASE1–3 is required for cytokinesis
As polyploid cells often show cytokinetic failure, we searched for cytokinetic defects. Although the cotyledons of NIT-RNAi seedlings had no morphological abnormalities (Fig. 6b), closer inspection revealed that 74% of stomata were aberrant (Fig. 8a). Cytokinetic defects of stomata in cotyledons were early signs of the NIT-RNAi phenotype in T1-independent transformants (Fig. 8b). Round-shaped undivided mother guard cells, single guard cells, cells with aborted cell wall stubs – all indicating defective cytokinesis – were observed in cotyledons of NIT-RNAi seedlings. Abortive cytokinesis of stomata was also present in hypocotyls of NIT-RNAi seedlings (Fig. 8b). In true leaves with milder phenotypes, stomata were present, but both the stomata and pavement cells had extremely frequent cytokinetic defects (Fig. 8c). Hallmarks of aberrant cytokinesis, such as cell wall stubs, enlarged nuclei, fusion of nuclei or binuclear cells, were all abundant (Fig. 8c–e). Binuclear cells remained in the cell cycle and were capable to enter into S-phase, as revealed by positive labelling of replicating nuclei after a 3 h pulse in EdU assay (Fig. 8d). Although c. 100 independent NIT-RNAi T1 transformants were evaluated, we did not observe clustering of stomata, indicating that the cytokinesis but not cell lineage patterning during stomata development was affected in NIT-RNAi plants.
Although developmental abnormalities were more prominent in aerial parts, root development was also affected by silenced NIT1–3 expression (Fig. S6). At the elongation zone, where normally cells exit the cell cycle, in NIT-RNAi, a large proportion of cells still remained in proliferation. The abnormally enlarged nuclei were detected by EdU labelling to go through DNA synthesis (Fig. S6a). In comparison to WT roots, the meristematic zone of primary roots of NIT-RNAi seedlings was markedly shortened and malformed; there were irregular cell files with cells having enlarged nuclei or two nuclei, suggesting defects in cytokinesis also in root cells (arrowheads in Fig. S6b).
Overexpression of NIT1 results in the suppression of cell growth and increased rate of PCD
We aimed to study the effect of overexpression of GFP-NIT1 driven by the 35S promoter, but among 20 independent transformants we could not obtain lines with significant overexpression, and plants expressing GFP-NIT1 at comparable levels to wild-type had no discernible phenotypic effects (Fig. S2). This indicates that the expression of NIT1 in plants is tightly regulated. Contrary to moderate expressions found in the plants, we could obtain lines with high GFP-NIT1 expression in cultured Arabidopsis cells transformed with the p35::GFP-NIT1 construct (Fig. S2).
The effect of GFP-NIT1 overproduction on growth and division was analysed in three independent transformed cell lines. To determine whether the entry into S-phase is affected by higher NIT1 levels, we applied a 3 h pulse labelling of EdU immediately after subculture of stationary grown cells. As shown in Fig. 9(a), the proportion of cells labelled for S-phase during the first 3 h after subculture was reduced two- to three-fold in the GFP-NIT1 overexpression line as compared to the control. Reduced entry and pass through S-phase in GFP-NIT1 overexpressing lines was confirmed also by flow cytometry. We found a lower proportion of cells with G2 DNA content 24 h after subculture in 35S::GFP-NIT1 overexpression lines as compared to control (Fig. 9b). The lower number of cells progressing through the cell cycle within 24 h after subculture corresponded to the lower mitotic index (Fig. 9c). Furthermore, the growth rate of the overexpression lines determined by measuring fresh cell mass was lower compared to control (Fig. 9d).
NIT1 is implicated in PCD both in animal and plant cells (Cutler & Somerville, 2005; Zhang et al., 2009). Clustering of GFP-NIT1 in early PCD of Arabidopsis was described by Cutler & Somerville (2005). To confirm whether GFP-NIT1 behaves similarly in our experiments, we used acetylsalicylic acid (ASA). ASA-induced PCD in Arabidopsis cell cultures as shown by Garcia-Heredia et al. (2008) and by us (Fig. S7). Immunofluorescence labelling of endogenous protein or confocal microscopy imaging of GFP-NIT1 in vivo showed a patchy signal for NIT1 accumulated in the perinuclear area or even penetrated into the nuclei in ASA-treated cells. Accumulation of GFP-NIT1 in perinuclear area and in nuclei was observed also after aphidicolin induced replication stress (Fig. S8).
We further addressed whether the overexpression of GFP-NIT1 alters the frequency of PCD measured by the DNA double-strand breaks through the TUNEL assay. As shown in Fig. 9(e), the number of TUNEL-positive cells was higher in 35S::GFP-NIT1 lines compared to wild-type. The difference between cells entering PCD became more prominent with time after subculture, being 7.4% and 20.9% for the 35S::GFP-NIT1 overexpression cultures, compared to 2.3% and 3.7% for control cultures on the first and fifth day after subculture, respectively.
Altogether, our data suggest that in Arabidopsis overproduction of GFP-NIT1 led to the reduction in cell proliferation and an increased level of PCD.
FHIT expression parallels NIT1 levels in NIT-RNAi and 35S::GFP-NIT1 overexpression lines
Because in animals NIT1 functions closely together with the FHIT tumour suppressor gene (Semba et al., 2006), we searched for the Arabidopsis orthologue of FHIT and found a gene At5 g58240 with 50% identity within the protein coding region (Fig. S9). The Arabidopsis gene is annotated as nucleoside phosphoramidase and adenylylsulfatase (TAIR; http://www.arabidopsis.org). In agreement, the animal FHIT is also involved in purine metabolism. FHIT is co-expressed with NIT1 in animal cells (Sun et al., 2009). Therefore, we tested FHIT expression together with expression of NIT1 in silenced NIT-RNAi seedlings. As shown in Fig. S10, FHIT expression was reduced in NIT-RNAi seedlings while in cell culture with GFP-NIT1 overexpression it was elevated compared to the control. These data indicate that as in animal cells, Arabidopsis NIT1 might function together with FHIT.
In order to analyse further the deregulated cell proliferation we also determined the expression of CDKA;1 and CDKB1;1 genes by Q-RT-PCR and found both to be upregulated in the NIT-RNAi plants compared to wild-type (Fig. S10c).
- Top of page
- Material and Methods
- Supporting Information
Although NIT1 homologues are described as metabolic enzymes involved in glucosinolate catabolism in Arabidopsis or in Brassica rapa (Janowitz et al., 2009), a function in early PCD was suggested by Cutler & Somerville (2005). We demonstrated that in Arabidopsis thaliana extracts NIT1 assembled into filaments forming a broad range of molecular masses and lengths. Similar filaments are formed in vitro from recombinantly expressed bacterial and fungal nitrilases. However, for this to happen, the Rhodococcus rhodochrous and Aspergillus niger nitrilases have to be posttranslationally modified; C-terminally cleaved to enable stable arrangement of helices (Thuku et al., 2007; Kaplan et al., 2011). The corresponding C-terminal part is missing in the Arabidopsis NIT1-3 sequences (Fig. S11), suggesting that in Arabidopsis these protein molecules have an intrinsic property to assemble into filamentous polymers. The evolutionarily conserved filament-forming capacity assigns the NIT1 proteins to the expanding group of metabolic filament-forming enzymes that evolved other functions besides their catalytic activities (Ingerson-Mahar et al., 2010; Noree et al., 2010). The functions of these NIT1 filaments are as yet unknown. NIT1 aggregation precedes the occurrence of PCD hallmarks in Arabidopsis (Cutler & Somerville, 2005). We confirmed these findings of Cutler and Somerville with our GFP-NIT1 fusion in cell culture treated with a PCD-inducing drug, and moreover, we found that overexpression of GFP-NIT1 increased the frequency of PCD. Overexpression of NIT1 in animal cells also elevates apoptosis and suppresses cell division (Semba et al., 2006). It would be interesting to find out whether animal NIT1 homologue preserves the filament-forming capability of bacterial and plant nitrilases and whether translocation and agregation of NIT1 occurs during animal apoptosis.
We characterized the developmental regulation of NIT1 expression and its subcellular localization by imaging pNIT1::GFP-NIT1 transformants and by immunofluorescence with NIT1-specific antibodies that we produced. We found that in proliferating cells of root meristem, young leaves or cultured cells, NIT1 is mostly cytoplasmic, but a small portion appears to be associated with microtubular structures. We further confirmed microtubular association with spin down of taxol-stabilized plant microtubules. However, several lines of evidence suggest that the role of NIT1 in proliferation and differentiation is unlikely to be executed via the nucleation and organization of microtubules: (1) unlike γ-tubulin, NIT1 is not associated with microtubule stubs after APM-induced MT depolymerization, (2) aberrant metaphase chromosome alignment and separation of chromatids in anaphase are typical when microtubule function is disturbed; however, these were not evident in dividing polyploid cells with NIT-RNAi silencing, suggesting the presence of functional microtubular arrays. It is more likely that microtubules might provide a signalling platform for transiently associated portion of NIT1 cellular pool.
An important clue on possible NIT1 functions in animal cells came from the realization that in Drosophila melanogaster and in Caenorhabditis elegans, NIT1 is fused to FHIT, an important tumour suppressor gene (Pekarsky et al., 1998). However, in vertebrates, these two proteins are encoded by two separate genes. The NIT1 and FHIT genes are also at separate genome locations in Arabidopsis, where FHIT was characterized for its dinucleoside triphosphate hydrolase and phosphodiesterase activities (Guranowski et al., 2008). The involvement of NIT1 and FHIT in a common pathway is indicated by their overlapping expression and their common functions in the regulation of cell proliferation and apoptosis in animal cells (Semba et al., 2006; Sun et al., 2009; Zhang et al., 2009). Furthermore, in animals, FHIT plays a role in oxidative stress and DNA damage response; its inactivation contributes to the accumulation of abnormal checkpoint phenotypes during cancer development. In Arabidopsis cells FHIT might also operate in a common pathway with NIT1, as suggested by our finding of their coordinated expression. Our proteomics analysis of NIT1 interactors did not identify FHIT; this is consistent with a failure to show physical interaction of NIT1 and FHIT in human cells (Huebner & Croce, 2001). Interestingly, in human cells FHIT is associated in vitro with microtubules, and the association is required for the tumour suppressor activity of FHIT but most likely does not regulate microtubular nucleation (Chaudhuri et al., 1999). Thus, similar to what we proposed for NIT1, microtubules might provide a signalling platform also for FHIT.
Animal cells with defective nuclear division, or polyploidy, withdraw permanently from the cell cycle due to strong checkpoint controls, and enter apoptosis. NIT1 either alone or together with the tumour suppressor FHIT is suggested to be involved in these checkpoints to function as a repressor of the cell cycle and as a switch to apoptosis (Zhang et al., 2009). In agreement, the level of NIT1 expression is significantly reduced in tumours (Sun et al., 2009), and kidney cells deficient for NIT1 show hyperproliferation (Semba et al., 2006). In our experiments, the checkpoint control was abrogated in NIT-RNAi plants that failed to exit proliferation during differentiation, carried on cycling and became highly polyploid. Although the checkpoint control of the cell division in plant meristematic cells was suggested to be weaker in comparison to animals (Castellano & Sablowski, 2008), our data suggest that the role of plant NIT1 in the cell cycle checkpoints is conserved. Meristematic cells respond to DNA damage by ATM/ATR dependent PCD, despite the absence of plant homologues of key transducers and executioners of the ATM/ATR activated PCD pathway (Fulcher & Sablowski, 2009). Both NIT1 and FHIT are suggested to play a role in DNA damage-induced apoptosis in animals (Zhang et al., 2009). It would be interesting to know whether NIT1 and possibly also FHIT play a conserved role in the process also in plants, as suggested by accumulation of NIT1 in perinuclear area during early phases of apoptosis and during aphidicolin-induced replication stress. Progression through S-phase was shown to be specifically sensitive to levels of NIT1 in our experiments being elevated or reduced upon dowregulation and overproduction of NIT1.
The Arabidopsis NIT1 protein was predicted to be ubiquitylated (Maor et al., 2007), which was confirmed later using several experimental approaches (Igawa et al., 2009; Saracco et al., 2009). Interaction of human NIT1 with RAD23 protein was found with yeast two hybrid assay (BioGrid, Stark et al., 2006). Rad23 is involved both in animals and plants in DNA repair and proteasomal protein degradation. In Arabidopsis, RAD23 interacts with the 26S receptor, RPN10, functioning in the delivery of ubiquitin-conjugating enzyme substrates to 26S proteasome (Farmer et al., 2010). Interestingly, in human cells FHIT interacts with the ubiquitin-conjugating enzyme, hUBC9 (Shi et al., 2000). Significance of these interactions for possible function of NIT1/FHIT in 26S proteasome degradation pathway or in proteasome-independent ubiquitylation driven signalling in processes such as DNA repair, replication, signal transduction and cell division, needs to be determined.
Cell division and differentiation are coordinated in distinct zones of the meristems to maintain meristem function. This appears to be disturbed by the silencing of NIT1–3, leading to severe developmental defects both in the root and aerial parts of the seedlings, including sustained proliferation and failure to exit into differentiation, as indicated also by the S-phase EdU labelling at the transition zone of root meristem to elongation growth. However, overexpression of GFP-NIT1 reduced the entry into S-phase and cell proliferation. Particularly leaf development was affected by NIT1–3 silencing, which was in severe cases manifested as a complete lack of cellular differentiation and tissue organization. Expression of CDKA;1 and CDKB1;2 was elevated in NIT-RNAi seedlings, where proliferating unorganized calli-like structures emerged at the flanks of the meristem instead of leaves. The A and B2-type CDKs were shown to be required not only for cell cycle progression, but also for reprogramming development and meristem organization (Andersen et al., 2008; Gaamouche et al., 2010). Further research is required to address whether and how NIT1–3 function in negative regulation of cell proliferation interplays with the cell cycle regulators during development.
For cells to become polyploid, the sustained proliferation has to be accompanied by abortive cytokinesis. Severe cytokinetic defects were unusually frequent in the NIT-RNAi plants in all organs we have studied and our data pointed to an important but so far uncharacterized function of nitrilases in cytokinesis. Similar to NIT1 silencing, mutation of the destruction box in mitotic B1-type cyclins and its compromised degradation also leads to cytokinetic defects and polyploidization (Weingartner et al., 2004). NIT1 was shown to be ubiquitylated (Igawa et al., 2009; Saracco et al., 2009). As discussed above, association of NIT1 with microtubules and accumulation in the area of cell plate formation is likely to reflect a signalling function rather than function in microtubule nucleation. In plants, a cytokinetic MAPK signalling pathway that associates with mitotic microtubules was characterized (Bogre et al., 1999). Interestingly, this cytokinesis-regulating pathway is multifunctional and just as NIT1, also responds and regulates stress responses.
In summary, we have shown that Arabidopsis NIT1 belongs to growing number of metabolic enzymes with filament-forming capacity. It is likely that different developmental, cellular and stress signals such as wounding, might modulate aggregation of high molecular forms of nitrilases, which in turn could provide platforms for molecular interactions in regulation of cellular processes. Furthermore, we show that NIT1 has a role in repression of proliferation both in conjunction with developmental cues and possibly in response to cell cycle checkpoints. Thus, NIT1 might regulate cell division and differentiation in balance with apoptosis. Another independent function of NIT1 is to ensure correct cytokinesis; compromising NIT1 function resulted to continued proliferation and aberrant cytokinesis leading to immensely polyploid and aneuploid cells. Formation of polyploids and aneuploids are major evolutionary driving forces within the plant kingdom (Doyle et al., 2008). Thus, NIT1 is a largely important multifunctional protein to safeguard plant development, genome integrity and could have important impacts on plant evolution.
- Top of page
- Material and Methods
- Supporting Information
This work was supported by grant from the Grant Agency of the Czech Republic (P501/12/2333). B.P. was supported by the Centre of the Region Haná for Biotechnological and Agricultural Research (NCZ.1.05/2.1.00/01.0007). We thank Gabriela Kočárová and Jan Svoboda for technical assistance, the CSIRO Plant Industry, Australia for vectors pHANNIBAL and pART27, and Richard Cyr (Pennsylvania State University, USA) for the GFP-MAP4 vector.
- Top of page
- Material and Methods
- Supporting Information
- 2008. Sinapis phylogeny and evolution of glucosinolates and specific nitrile degrading enzymes. Phytochemistry 69: 2937–2949. , , , .
- 2008. Requirement of B2-type cyclin-dependent kinases for meristem integrity in Arabidopsis thaliana. Plant Cell 20: 88–100. , , , , , , , .
- 2000. Nuclear gamma-tubulin during acentriolar plant mitosis. Plant Cell 12: 433–442. , , , , , , , .
- 2006. Gamma-tubulin is essential for acentrosomal microtubule nucleation and coordination of late mitotic events in Arabidopsis. Plant Cell 18: 1199–1212. , , , , , , .
- 1999. A MAP kinase is activated late in plant mitosis and becomes localized to the plane of cell division. Plant Cell 11: 101–113. , , , , , , , , , et al.
- 2009. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clinical Chemistry 55: 611–622. , , , , , , , , , et al.
- 2008. Phosducin-Like Protein 3 is required for microtubule-dependent steps of cell division but not for meristem growth in Arabidopsis. Plant Cell 20: 969–981. , .
- 1999. The tumor suppressor protein Fhit. A novel interaction with tubulin. Journal of Biological Chemistry 274: 24 378–24 382. , , , , , .
- 1998. Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant Journal 16: 735–743. , .
- 2000. Random GFP:cDNA fusions enable visualization of subcellular structures in cells of Arabidopsis at a high frequency. Proceedings of the National Academy of Sciences, USA 97: 3718–3723. , , , .
- 2005. Imaging plant cell death: GFP-Nit1 aggregation marks an early step of wound and herbicide induced cell death. BMC Plant Biology 5: 4. , .
- 2009. Protocol: streamlined sub-protocols for floral-dip transformation and selection of transformants in Arabidopsis thaliana. Plant Methods 5: 3. , , , , .
- 2011. A nodulin/glutamine synthetase-like fusion protein is implicated in the regulation of root morphogenesis and in signalling triggered by flagellin. Planta 234: 459–476. , , , , , , , .
- 2008. Evolutionary genetics of genome merger and doubling in plants. Annual Review of Genetics 42: 443–461. , , , , , , .
- 2003. Plant gamma-tubulin interacts with alphabeta-tubulin dimers and forms membrane-associated complexes. Plant Cell 15: 465–480. , , , , , .
- 2010. The RAD23 family provides an essential connection between the 26S proteasome and ubiquitylated proteins in Arabidopsis. Plant Cell 22: 124–142. , , , , , .
- 2002. Interphase chromosomes in Arabidopsis are organized as well defined chromocenters from which euchromatin loops emanate. Proceedings of the National Academy of Sciences, USA 99: 14 584–14 589. , , , , .
- 2009. Hypersensitivity to DNA damage in plant stem cell niches. Proceedings of the National Academy of Sciences, USA 106: 20 984–20 988. , .
- 2010. Cyclin-dependent kinase activity maintains the shoot apical meristem cells in an undifferentiated state. Plant Journal 64: 26–37. , , , , , , , , , et al.
- 2004. ‘Conserved hypothetical’ proteins: prioritization of targets for experimental study. Nucleic Acids Research 32: 5452–5463. , .
- 2008. Acetylsalicylic acid induces programmed cell death in Arabidopsis cell cultures. Planta 228: 89–97. , , , .
- 2008. Fhit proteins can also recognize substrates other than dinucleoside polyphosphates. FEBS Letters 582: 3152–3158. , , , , , , , , .
- 2001. Identification of HL60 proteins affected by 5-aminolevulinic acid-based photodynamic therapy using mass spectrometric approach. Collection of Czechoslovak Chemical Communications 66: 1720–1728. , , , , .
- 2001. FRA3B and other common fragile sites: the weakest links. Nature Reviews Cancer 1: 214–221. , .
- 2011. Hits, Fhits and Nits: beyond enzymatic function. Advances in Enzyme Regulation 51: 208–217. , , , , .
- 2009. Isolation and identification of ubiquitin-related proteins from Arabidopsis seedlings. Journal of Experimental Botany 60: 3067–3073. , , , , , , , , .
- 2010. The metabolic enzyme CTP synthase forms cytoskeletal filaments. Nature Cell Biology 12: 739–746. , , , , .
- 2009. Evolution of nitrilases in glucosinolate-containing plants. Phytochemistry 70: 1680–1686. , , .
- 2011. Heterologous expression, purification and characterization of nitrilase from Aspergillus niger K10. BMC Biotechnology 11: 2. , , , , , , , , , et al.
- 2005. Sites and regulation of auxin biosynthesis in Arabidopsis root. Plant Cell 17: 1090–1104. , , , , , , .
- 2007. Multidimensional protein identification technology (MudPIT) analysis of ubiquitinated proteins in plants. Molecular and Cellular Proteomics 6: 601–610. , , , , , .
- 1998. A GFP-MAP4 reporter gene for visualizing cortical microtubule rearrangements in living epidermal cells. Plant Cell 10: 1927–1940. , , , , , , .
- 1992. Regulation of indole-3-acetic acid biosynthetic pathways in carrot cell cultures. Plant Physiology 100: 1346–1353. , , , .
- 2010. Identification of novel filament-forming proteins in Saccharomyces cerevisiae and Drosophila melanogaster. Journal of Cell Biology 190: 541–551. , , , .
- 1997. Arabidopsis mutants resistant to the auxin effects of indole-3-acetonitrile are defective in the nitrilase encoded by the NIT1 gene. Plant Cell 9: 1781–1790. , , , .
- 1998. Nitrilase and Fhit homologs are encoded as fusion proteins in Drosophila melanogaster and Caenorhabditis elegans. Proceedings of the National Academy of Sciences, USA 95: 8744–8749. , , , , , , , , , et al.
- 2012. Plant Aurora kinases play a role in maintenance of primary meristems and control of endoreduplication. New Phytologist 193: 590–604. , , , , , , , .
- 2008. Primary or secondary? Versatile nitrilases in plant metabolism. Phytochemistry 69: 2655–2667. .
- 2009. Tandem affinity purification and mass spectrometric analysis of ubiquitylated proteins in Arabidopsis. Plant Journal 59: 344–358. , , , , , .
- 2006. Biological functions of mammalian Nit1, the counterpart of the invertebrate NitFhit Rosetta stone protein, a possible tumor suppressor. Journal of Biological Chemistry 281: 28 244–28 253. , , , , , , , , , .
- 2000. Association of FHIT (fragile histidine triad), a candidate tumour suppressor gene, with the ubiquitin-conjugating enzyme hUBC9. Biochem Journal 352(Pt 2): 443–448. , , , .
- 2006. BioGRID: a general repository for interaction datasets. Nucleic Acids Research 34: D535–D539. , , , , , .
- 2009. Nit1 and Fhit tumor suppressor activities are additive. Journal of Cellular Biochemistry 107: 1097–1106. , , , , , , .
- 2007. Post-translational cleavage of recombinantly expressed nitrilase from Rhodococcus rhodochrous J1 yields a stable, active helical form. FEBS Journal 274: 2099–2108. , , , .
- 2009. APC/C-CCS52A complexes control meristem maintenance in the Arabidopsis root. Proceedings of the National Academy of Sciences, USA 106: 11 806–11 811. , , , , , , , , , .
- 2001. Enzymatic characterization of the recombinant Arabidopsis thaliana nitrilase subfamily encoded by the NIT2/NIT1/NIT3-gene cluster. Planta 212: 508–516. , , , , , , .
- 2010. FHIT: doubts are clear now. ScientificWorldJournal 10: 1142–1151. .
- 2004. Expression of a nondegradable cyclin B1 affects plant development and leads to endomitosis by inhibiting the formation of a phragmoplast. Plant Cell 16: 643–657. , , , , , , , , , .
- 2009. Mammalian nitrilase 1 homologue Nit1 is a negative regulator in T cells. International Immunology 21: 691–703. , , , , , .
- Top of page
- Material and Methods
- Supporting Information
Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.
Fig. S1 Multiple sequence alignments of Arabidopsis NIT1–4 and of nitrilases from various organisms.
Fig. S2 Overproduction of GFP-NIT1 in cultured cells and seedlings.
Fig. S3 Silver stained SDS-PAGE and MALDI-MS data of proteins co-immunopurified with GFP-NIT1 from re-suspended P100 fraction.
Fig. S4 Expression of GFP-NIT1 driven by its own promoter in Arabidopsis seedling root.
Fig. S5 Transcript level of NIT1, 2, 3 and 4 in NIT-RNAi plants; correlation of transcript levels with strength of the phenotype.
Fig. S6 Root phenotypes of plants with NIT1–3 downregulated by RNAi.
Fig. S7 Induction of programmed cell death (PCD) in Arabidopsis cells by acetylsalicylic acid.
Fig. S8 NIT1 localization during induction of PCD by acetylsalicylic acid and during aphidicolin-induced replication stress.
Fig. S9 Sequence alignment of FHIT proteins from Arabidopsis thaliana and Homo sapiens.
Fig. S10 Co-expression of NIT1 and FHIT in NIT-RNAi plants and cultured cells overexpressing GFP-NIT1 and expression of CDKA;1 and CDKB1;2 in NIT-RNAi plants.
Fig. S11 Sequence alignment of Rhodococcus rhodochrous J1 nitrilase and Arabidopsis thaliana NIT1.
Methods S1 Detailed materials and methods.
|nph12185-sup-0002-VideoS1.avi||video/avi||471K||Video S1 Mitosis in Arabidopsis cultured cell expressing GFP-NIT1 under 35S CaMV promoter.|
|nph12185-sup-0003-VideoS2.avi||video/avi||473K||Video S2 Cytokinesis in Arabidopsis cultured cell expressing GFP-NIT1 under 35S CaMV promoter.|