SEARCH

SEARCH BY CITATION

Keywords:

  • Arabidopsis;
  • codon usage;
  • DRONPA;
  • export;
  • fluorescence recovery after selective photoswitching;
  • nuclear import;
  • reversible photoswitchable fluorescent protein;
  • RNA-binding protein

Abstract

  1. Top of page
  2. Abstract
  3. Results and Discussion
  4. Materials and Methods
  5. Acknowledgments
  6. References
  7. Supporting Information

Fluorescent reporter proteins that allow repeated switching between a fluorescent and a non-fluorescent state are novel tools for monitoring intracellular protein trafficking. A codon-optimized variant of the reversibly photoswitchable fluorescent protein DRONPA was designed for the use in transgenic Arabidopsis plants. Its codon usage is also well adapted to the mammalian codon usage. The synthetic protein, DRONPA-s, shows photochemical properties and switching behavior comparable to that of the original DRONPA from Pectiniidae both in vitro and in vivo. DRONPA-s fused to the RNA-binding protein AtGRP7 (Arabidopsis thaliana glycine-rich RNA-binding protein 7) under control of the endogenous AtGRP7 promoter localizes to cytoplasm, nucleoplasm and nucleolus of transgenic Arabidopsis plants. To monitor the intracellular transport dynamics of AtGRP7-DRONPA-s, we set up a single-molecule sensitive confocal fluorescence microscope. Fluorescence recovery after selective photoswitching experiments revealed that AtGRP7-DRONPA-s reaches the nucleus by carrier-mediated transport. Furthermore, photoactivation experiments showed that AtGRP7-DRONPA-s is exported from the nucleus. Thus, AtGRP7 is a nucleocytoplasmic shuttling protein. Our results show that the fluorescent marker DRONPA-s is a versatile tool to track protein transport dynamics in stably transformed plants.

Upon transcription and throughout their life, mRNAs are bound by a suite of proteins. Their dynamic association with RNA defines pre-mRNA processing, lifetime, export from the nucleus and the rate at which a specific mRNA is translated (1). AtGRP7 (Arabidopsis thaliana glycine-rich RNA-binding protein 7) is a representative of a class of small glycine-rich RNA-binding proteins with a single RNA recognition motif (RRM) at its N-terminus. AtGRP7 is part of the endogenous timing system that controls the daily life of the plant. Further, it promotes the transition to flowering and participates in pathogen defense in Arabidopsis presumably via the regulation of target transcripts at the post-transcriptional level (2–5). The AtGRP7 transcript itself undergoes so-called circadian, i.e. 24-h, oscillations in steady-state abundance with a peak at the end of the daily light phase. AtGRP7 impacts the oscillations of its own transcript by negative autoregulation at the post-transcriptional level. It binds to its own pre-mRNA, causing the production of an alternative splice form with a premature termination codon that is rapidly degraded through nonsense-mediated decay (6–8). The influence of AtGRP7 on splice-site selection of its pre-mRNA obviously occurs in the nucleus. Presently, it is not known whether AtGRP7 impacts additional processing steps such as RNA degradation or translation. However, its effect on the steady-state abundance of a suite of transcripts suggests it also has a cytoplasmic function (4,6).

Immunogold labeling has first shown the presence of the orthologous protein SaGRP (Sinapis alba glycine-rich protein) in the nucleus of mustard plants (9). To reach the nuclear compartment, proteins translated on ribosomes in the cytoplasm need to cross the nuclear envelope. This occurs through the nuclear pore complexes (NPCs) embedded in the nuclear envelope (10,11). NPCs contain a central aqueous channel with an estimated radius of about 2.6 nm, as measured in HeLa cells (12). Proteins with a molecular mass above around 40 kDa require facilitated, energy-dependent transport, whereas smaller proteins can passively diffuse through the channel (11,13). However, simple diffusion does not allow proteins to accumulate against a concentration gradient and some smaller proteins including histones pass the nuclear envelope also by carrier-mediated transport. One advantage of such a carrier-dependent passage is the possibility to restrict the access as an additional layer to control, e.g. gene expression.

Nuclear import is mediated by peptide motifs in the cargo proteins that are recognized by receptor proteins (11,13). Conventional nuclear localization signals (NLSs) consist of one or two stretches of basic amino acid residues, in particular lysines. AtGRP7 has a different nuclear import signal, an M9 domain originally found in some mammalian heterogeneous nuclear ribonucleoproteins (hnRNPs) implicated in many aspects of RNA processing (14). M9 domains have been reported to mediate transport both into and out of the nucleus (15). The AtGRP7 M9 domain is recognized by AtTRN1, the Arabidopsis orthologue of the import receptor transportin (also known as karyopherin β2). AtGRP7 is imported into the nucleus of tobacco BY2 cells (16). Notably, it can also be imported into the nuclei of permeabilized HeLa cells by both Arabidopsis AtTRN1 and human transportin (16). In Arabidopsis, an AtGRP7-GFP (green fluorescent protein) fusion protein driven by the strong constitutive Cauliflower Mosaic Virus (CaMV) promoter is found in both the nucleus and the cytoplasm (5,17). The molecular mechanisms of the nuclear uptake have not been investigated and it is not known whether AtGRP7 is re-exported from the nucleus to the cytoplasm.

Fluorescence recovery after photobleaching (FRAP) and photoactivation of fluorescent reporter proteins provide insights into carrier-mediated, directed transport and diffusion in the cell. However, photobleaching and photoactivation are irreversible processes in conventional fluorescent markers such as eGFP (enhanced GFP) or pa-GFP (photoactivatable GFP).

To overcome these limitations, Ando et al. have cloned a fluorescent protein from a Pectiniidae that can be reversibly photoswitched between a bright and a dark state, and thus allows repeated measurements or even different kinetic measurements on a single cell (18). It has been named ‘DRONPA’, after ‘dron’, a ninja designation for ‘vanishing’, and ‘pa’ for ‘photoactivation’. When irradiated with 488-nm light, DRONPA emits bright green fluorescence. With increasing laser power, DRONPA is photoswitched to a non-fluorescent off-state. Irradiation at 400 nm, however, recovers the fluorescent on-state. Moreover, photoswitching of DRONPA to the off-state requires many fewer photons (quantum yield φ = 3.2 × 10−4) than photobleaching of eGFP (quantum yield φ = 8 × 10−6), allowing fluorescence recovery after selective photoswitching measurements to be performed at much lower excitation intensities with less cell damage (18,19).

Recently, Martin et al. have used DRONPA as a reporter in a tobacco (Nicotiana benthamiana) transient expression system (20). In another study, a DRONPA fusion to secretory carrier membrane protein 2 (SCAMP2) monitored the movement of secretory vesicles through the endomembrane system in transiently transformed tobacco BY-2 cells (21). Often, however, intact plants are preferred systems because faithful monitoring of intracellular movement requires the macromolecule's native environment to allow for specific interactions or post-translational modifications to occur. This requires the use of endogenous promoters, whereas transient expression commonly relies on the strong constitutive CaMV promoter. In the case of some GFP-tagged RNA-binding proteins, this has been found to alter their dynamic properties (22–25).

To avoid potential incongruence of the Pectiniidae sequence with the mRNA processing and translation apparatus in plants, we designed a synthetic gene, DRONPA-s (synthetic DRONPA), by adapting the codons to the codon bias of the model plant A. thaliana. This approach was prompted by the previous observation that an 84-nucleotide sequence within GFP was mistaken as a cryptic intron in planta, preventing proper GFP expression in transgenic plants (26). This was remedied by modifications of the GFP codons that removed the sequences erroneously taken as splice sites (26). The newly designed DRONPA variant DRONPA-s is also better adapted to the codon usage of mammals than the original DRONPA sequence cloned from the coral Pectiniidae. High-level expression of DRONPA-s was achieved in transgenic Arabidopsis plants, in a tobacco (N. benthamiana) transient expression system as well as in transfected HeLa and COS-7 cells.

To monitor intracellular transport dynamics, we set up a confocal fluorescence microscope with enhanced sensitivity that allows the use of a low laser power and thus reduces off-switching of DRONPA-s fusion proteins during imaging.

Fluorescence recovery after selective photoswitching experiments in transgenic Arabidopsis plants expressing the AtGRP7-DRONPA-s fusion protein revealed that AtGRP7 is imported into the nucleus via carrier-mediated transport. Furthermore, photoactivation experiments in the nucleus provided evidence that AtGRP7-DRONPA-s is exported out of the nucleus. This shows that AtGRP7 undergoes bidirectional nucleocytoplasmic trafficking.

Results and Discussion

  1. Top of page
  2. Abstract
  3. Results and Discussion
  4. Materials and Methods
  5. Acknowledgments
  6. References
  7. Supporting Information

Photochemical properties of synthetic DRONPA-s

A variant of the reversibly photoswitchable DRONPA protein was engineered to adapt the codon usage for expression in A. thaliana. In addition, fortuitous regulatory motifs including cryptic splice signals, TATA boxes or ribosomal entry sites were avoided (Figures S1 and S2). To characterize in detail the photochemical properties of the synthetic DRONPA-s protein, we purified recombinantly expressed DRONPA-s from Escherichia coli. The absorption spectrum of the synthetic protein has a maximum at 503 nm (Figure S3A). The emission spectrum, recorded at an excitation wavelength of 488 nm, peaks at 516 nm. Thus, the codon-optimized variant resembles the original DRONPA protein cloned from Pectiniidae(18).

The photoswitching performance of the synthetic DRONPA-s was tested in standard quartz cuvettes. Irradiation at 404 nm quantitatively transfers DRONPA-s to the on-state. Subsequently, the solution was exposed to 488-nm laser light for 2 min applying an excitation intensity of ∼20 mW/cm2 in order to generate the non-fluorescent off-state. The procedure was repeated several times. Between each round of photoswitching, the fluorescence emission (500–580 nm) was recorded. Plotting the integrated fluorescence against irradiation time revealed a photoswitching half-time of 253 seconds (Figure S3B). To compare this to the literature, we calculated the photoswitching half-time of the original DRONPA molecule based on the data by Ando et al. to be 12 seconds at 400 mW/cm2(18). Taken together, these data indicate an anti-proportional dependence of photoswitching half-time and laser intensity.

Subsequently, we determined the minimal region of a sample that could be photoswitched with our confocal setup using immobilized DRONPA-s protein. A spin-coated DRONPA-s layer was embedded in a Mowiol polymer matrix. Using a confocal microscope, a single spot was irradiated at 488 nm with an excitation power of 1 µW (corresponding to an excitation intensity of 300 W/cm2) for 3 seconds to transfer DRONPA-s to the off-state (Figure 1A). The intensity profile of an area of 10 µm × 10 µm surrounding the photoswitched spot was approximated by a Gaussian function, revealing an effective focus diameter of 1.15 µm in lateral direction (Figure 1B). This implies that a minimal region of ∼1-µm diameter can be selectively transferred to the off-state upon laser irradiation. Selective switching such a small area is of advantage to avoid bleaching of the cytoplasm surrounding plant nuclei that are localized at the periphery of the cell. The area could afterwards be switched to the on-state several times without significant loss in fluorescence intensity (not shown). Thus, our synthetic DRONPA-s is ideally suited for reversible photoswitching and fluorescence tracking experiments in living cells.

image

Figure 1. Switching behavior of immobilized DRONPA-s. Recombinant DRONPA-s embedded in a Mowiol matrix is switched off by exposure to 488 nm. A) Dark spot of DRONPA-s switched to dim state in the region of the laser focus. Fluorescence intensity is indicated by the quantitative color code: yellow corresponds to the highest fluorescence intensity. B) Calculation of the spot size with the full width at half maximum (FWHM) convention. Shown are the fluorescence intensities along the indicated line in (A).

Download figure to PowerPoint

High-level transient expression of DRONPA-s in planta and in mammalian cells

Next, we compared the expression characteristics of the codon-optimized DRONPA-s to that of the Pectiniidae sequence in three different systems. For Agrobacterium-mediated transient expression in plants, both DRONPA-s and the open reading frame with the original codon usage present in the commercial vector DRONPA-GREEN pDG-S1 (MLB Group Company) were placed under control of the strong constitutive CaMV promoter. Both constructs were introduced into N. benthamiana leaves via agroinfiltration. After 4 days, fluorescence was recorded in native protein extracts at 480-nm excitation wavelength (Figure 2A). Extracts from leaves expressing DRONPA-s showed a higher fluorescence intensity than leaves expressing DRONPA-GREEN. Furthermore, an immunoblot analysis showed that DRONPA-s protein was present at higher levels than DRONPA-GREEN (Figure 2B). We note that DRONPA-GREEN runs at a slightly higher apparent molecular weight. This may be attributed to an unknown post-translational modification.

image

Figure 2. Comparison of transiently expressed DRONPA-s and DRONPA-GREEN.Nicotiana benthamiana leaves were infiltrated with agrobacteria harboring DRONPA-s (Ds) or DRONPA-GREEN under control of the CaMV promoter. HeLa and COS-7 cells were transfected with pcDNA3.1+ harboring DRONPA-s or DRONPA-GREEN (DG) under control of the CMV promoter. A) Fluorescence was recorded in native protein extracts at 5 µg/mL dilution at 480-nm excitation and peak emission at 516 nm was expressed as means ± SD (n = 10). Student's t-tests showed that DRONPA-s fluorescence is significantly different from DRONPA-GREEN fluorescence for N. benthamiana (p = 0.000094), COS-7 (p = 0.000212) and HeLa (p = 0.006170) cells. B) The immunoblot of the same protein extracts was probed with an antibody against DRONPA. For tobacco leaves, the blot was reprobed with a light-harvesting chlorophyll-binding protein (LHCP) antibody to confirm equal loading (3). For the transfected HeLa and COS-7 cells, the blot was probed with an antibody against neomycin phosphotransferase encoded on the vector pcDNA3.1+. The positions of the molecular mass markers are indicated.

Download figure to PowerPoint

The synthetic sequence is also closer to the codon usage of mammals than the original nucleotide sequence. Thus, we compared DRONPA-s and DRONPA-GREEN expression under control of the strong constitutive cytomegalovirus (CMV) promoter in cell culture. Two days after transfection of COS-7 and HeLa cells, respectively, DRONPA fluorescence recorded in native protein extracts at 480-nm excitation wavelength was higher for cells expressing DRONPA-s than for cells expressing DRONPA-GREEN (Figure 2A). Furthermore, DRONPA-s protein was present at higher levels in both transfected COS-7 and HeLa cells compared to DRONPA-GREEN (Figure 2B).

These data show DRONPA-s is transiently expressed at higher levels in both plants and mammalian cell lines and suggest DRONPA-s will likely yield high expression levels in transgenic plants.

Expression of DRONPA-s in stably transformed plants

Transgenic Arabidopsis plants expressing DRONPA-s fused to the AtGRP7 coding region under control of the AtGRP7-regulatory regions, including 1.4 kb of the promoter, intron and 5′ and 3′ untranslated regions (UTRs), were generated (AtGRP7:AtGRP7-DRONPA-s). As a control, DRONPA-s alone was expressed using the same regulatory elements (AtGRP7:DRONPA-s).

We monitored DRONPA fluorescence predominantly in root cells of transgenic plants because of the high AtGRP7 expression in roots (Figure 3A). To confirm that the signal is not because of autofluorescence (e.g. of lignin in the cell walls), we took advantage of the photoswitching capability of DRONPA-s. After imaging under moderate excitation conditions, the intensity was increased and DRONPA-s was switched off. Upon subsequent photoactivation with 404 nm, only the DRONPA-s fluorescence returns because endogenous autofluorescence is not photoactivatable (not shown). Thus, the synthetic DRONPA-s could be stably expressed in transgenic plants. In a recent study, stable expression of proteins tagged with the irreversibly photoconvertable fluorescent protein Dendra2 in transgenic Arabidopsis plants was not achieved (27).

image

Figure 3. Localization of AtGRP7-DRONPA-s and DRONPA-s in transgenic Arabidopsis plants. A) Root of transgenic AtGRP7:AtGRP7-DRONPA-s plants. The scale bar corresponds to 50 µm. B) Intracellular distribution of fluorescence signal in AtGRP7:DRONPA-s-expressing plants. The arrowhead denotes the nucleolus. The scale bar corresponds to 10 µm. C) Intracellular distribution of fluorescence signal in AtGRP7:AtGRP7-DRONPA-s plants. The scale bar corresponds to 10 µm. The arrowhead denotes the nucleolus. The arrows denote the nucleus (N) and the cytoplasm (Cp), respectively. V, vacuole.

Download figure to PowerPoint

In transgenic plants expressing DRONPA-s alone under control of the AtGRP7 promoter, the fluorescence signal was distributed between the nucleus and the cytoplasm (Figure 3B). The AtGRP7:AtGRP7-DRONPA-s plants also showed fluorescence in both the cytoplasm and the nucleus and an additional particularly strong signal in the nucleolus (Figure 3C). Whereas nucleolar localization has not been reported for AtGRP7-GFP driven by the CaMV promoter (5,28), we find AtGRP7-GFP driven by its authentic promoter and regulatory elements also in the nucleolus of transgenic plants (not shown). Nucleolar localization of the hnRNP-like AtGRP7 protein with its role in splicing regulation is corroborated by the fact that the nucleolus in addition to its traditional role in rRNA biogenesis is engaged in RNP assembly, mRNA export and nonsense-mediated decay (29).

The AtGRP7-DRONPA-s fusion protein retains biological activity

To test whether AtGRP7 remains functional when fused to DRONPA-s, the chimeric AtGRP7:AtGRP7-DRONPA-s construct was transformed into the T-DNA insertion line atgrp7-1(5). In this mutant, the loss of AtGRP7 expression leads to an elevated level of AtGRP8, a closely related RNA-binding protein that is under negative control by AtGRP7, due to relief of repression by AtGRP7 (3). Furthermore, the atgrp7-1 mutant develops more leaves than wild-type (wt) plants until it starts to flower (3). If AtGRP7-DRONPA-s can compensate for the lack of endogenous AtGRP7 in this mutant, AtGRP8 protein should return to wt levels. In independent T2 plants that showed a strong DRONPA-s fluorescence, the intact fusion protein as well as a small amount of degraded protein were detected with an anti-DRONPA antibody (Figure S4A). Whereas we cannot rule out that a small amount of free DRONPA-s is also present in planta, most of it presumably is generated during protein extraction. The lack of endogenous AtGRP7 in the atgrp7-1 background was confirmed using an anti-peptide antibody specific for AtGRP7 (Figure S4A). Probing the blot with an anti-peptide antibody specific for AtGRP8 revealed that the strong upregulation of AtGRP8 in the atgrp7-1 mutant was reduced by expressing AtGRP7:AtGRP7-DRONPA-s. Furthermore, whereas the atgrp7-1 mutant flowers with 67 ± 2.98 leaves under short-day conditions compared to wt plants that flower with 57 ± 2.02 leaves, the atgrp7-1 AtGRP7:AtGRP7-DRONPA-s plants flower with 60 ± 2.71 leaves (Figure S4B). Thus, AtGRP7-DRONPA-s can compensate for endogenous AtGRP7 in the loss-of-function atgrp7-1 mutant. We conclude that DRONPA-s as a tag does not negatively influence the in vivo function of AtGRP7 and the fusion protein will reliably report subcellular localization and transport of AtGRP7.

Switching behavior of DRONPA-s in planta

Next, we investigated the switching behavior of DRONPA-s in vivo (Figure 4A). When the fluorescence was switched off in a small region of interest within the nucleus of AtGRP7-DRONPA-s-expressing plants for 10 seconds with a power of 10 µW (corresponding to a laser intensity of 3 kW/cm2), fluorescence was lost inside the entire nucleus (Figure 4B). We attribute this to diffusion of the protein inside the nucleus because in root cells fixed with paraformaldehyde (where diffusion is impossible) the same treatment resulted in a bleached spot of a size similar to that observed for DRONPA-s immobilized in Mowiol (Figure 4C).

image

Figure 4. Selectively switching off fluorescence in the nucleus. Shown is a cell in roots of transgenic AtGRP7-DRONPA-s plants before (A) and after (B) exposing a spot of 1.15-µm diameter to 488-nm irradiation. N, nucleus; V, vacuole; Cp, cytoplasm. C) In a fixed root cell, a small spot of AtGRP7-DRONPA-s in the nucleus was switched to the dim state. The arrowhead denotes the irradiated spot. The scale bars correspond to 10 µm.

Download figure to PowerPoint

Recording of nuclear import by fluorescence recovery after selective photoswitching

The AtGRP7-DRONPA-s fusion protein localizes to both the cytoplasm and the nucleus (Figure 3C). To visualize the nuclear import, we performed fluorescence recovery after selective photoswitching experiments in transgenic Arabidopsis plants. The confocal fluorescence microscope used (see Materials and Methods) operates with single-photon-counting avalanche photodiodes instead of photomultiplier tubes commonly used. The enhanced sensitivity enables us to reduce the off-switching of DRONPA-s during imaging by using a laser power of less than 1 µW (corresponding to 300 W/cm2 laser intensity).

AtGRP7-DRONPA-s fluorescence inside the nucleus of a root cell (Figure 5A) was switched off by irradiating a spot of 1.15 µm in diameter at 488 nm for 10 seconds. At that point in time, fluorescence was lost in the entire nucleus due to diffusion inside the nucleus (Figure 5B). Nuclear fluorescence recovered within 20 min (Figure 5C–H). A typical recovery curve is shown in Figure 5I. Saturation of the relative fluorescence intensity was reached at 70% of the initial fluorescence intensity in the nucleus. This would point to a fraction of the protein being immobile. However, the initial fluorescence in the nucleus is stronger than the initial fluorescence in the cytoplasm (Figure 5A). Therefore, we consider it is more likely that a limited pool of fluorescent proteins in the cytoplasm precludes full recovery in the nucleus (30). In fact, when we normalized the fluorescence intensities in the nucleus to the initial fluorescence in the cytoplasm in control experiments, the recovery reached 80–90% (data not shown), indicating that there is only a small immobile protein fraction.

image

Figure 5. Time series of a fluorescence recovery after selective photoswitching experiment in transgenic root cells. Shown are images taken before switching off (A), the first image after switching off (t = 0) (B), and images taken at 2-minute intervals during the recovery phase (C–H). I) Recovery curve of AtGRP7-DRONPA-s (circles) and of DRONPA-s (triangles). Fluorescence intensity was normalized to the initial fluorescence. The scale bar corresponds to 10 µm.

Download figure to PowerPoint

Full recovery is also masked by unwanted photoswitching as corroborated by control experiments with DRONPA-s only. Upon fluorescence recovery after selective photoswitching on nuclei of plants expressing DRONPA-s lacking an NLS, steady-state levels of the DRONPA-s fluorescence are reached much faster but at a strongly reduced level (Figure 5I). This suggests that DRONPA-s by itself enters the nucleus by diffusion but does not accumulate there. After reaching maximal recovery, the curve shows a gradual linear drop of fluorescence. We attribute this to unintended photoswitching to the off-state during image acquisition. For AtGRP7-DRONPA-s, this effect presumably is partially compensated through continued import. Nevertheless, the effect has to be considered to adequately describe fluorescence recovery curves measured in the nucleus (Figure 5I).

Taken together, these data indicate that in contrast to DRONPA-s alone the AtGRP7-DRONPA-s fusion protein reaches the nucleus by carrier-mediated transport. This is congruent with our previous finding that AtGRP7 interacts with the nuclear import receptor AtTRN1. Furthermore, when added to permeabilized HeLa cells, AtGRP7 does not enter the nucleus in the absence of transportin (16). Although the 17-kDa AtGRP7 protein is below the size-exclusion limit of the NPC, nuclear import obviously does not rely solely on diffusion. Passive diffusion through the NPC presumably occurs at a reasonable speed only for small proteins and does not allow the protein to accumulate against a concentration gradient (11). Moreover, carrier-mediated transport allows for an additional possibility to regulate trafficking.

Nuclear import at different times of the day

AtGRP7 influences the circadian oscillation of its own transcript by binding to its pre-mRNA, causing a shift to an alternative splice form with a premature termination codon that is rapidly degraded via nonsense-mediated decay (31). This impact on splice-site selection obviously is a nuclear event. Therefore, we wanted to determine whether the nuclear uptake of AtGRP7-DRONPA-s correlates with daytime.

To simultaneously monitor fluorescence recovery after selective photoswitching at different times of the day, plants were grown in an antiparallel day/night regime. For fluorescence recovery after selective photoswitching curves recorded around the time of the circadian maximum of the AtGRP7 oscillation (zt11, zeitgeber time 11 = 11 h after lights on), the mean half-time t1/2 was 88 ± 8 seconds. Around the time of the circadian minimum, at zt4, t1/2 was 150 ± 17 seconds (Figure 6A). For DRONPA-s alone, the fluorescence recovery after selective photoswitching curves were nearly identical at the two time-points (Figure 6B). Thus, our data imply that the nuclear import of AtGRP7 appears to be slower around the circadian minimum. Presently, the molecular basis for this behavior is not known. One may speculate that the higher import rate at the time of maximal AtGRP7 mRNA concentration (zt11) may relate to an accordingly higher translation rate. Also, newly synthesized protein may bear post-translational modifications that impact import rates.

image

Figure 6. Import at different times of the day. A) Shown are the means of t1/2 values for import of AtGRP7-DRONPA-s at zt4 and zt11, the standard error Sx with the number of experiments n and the p value obtained by a Student's t-test at a significance level of 0.05. B) Recovery curves of selective fluorescence recovery after selective photoswitching experiments at zt11 (diamonds) and zt4 (triangles) for DRONPA-s. For reasons discussed in the text, the recovery curve cannot be fitted with a reasonable model and thus t1/2 was not determined.

Download figure to PowerPoint

Whether the slightly different import rates impact AtGRP7 function, e.g. the regulation of other oscillating downstream target transcripts (4), remains to be determined. In mammals, the RNA-binding protein hnRNP I that shuttles between the cytoplasm and the nucleus contributes to oscillations of the mRNA encoding the circadian clock protein Per2 (Period2). hnRNP I causes degradation of the Per2 mRNA through interaction with its 3′ UTR (32). At times of the day when accumulation of hnRNP I in the cytosol is highest, Per2 transcript oscillations are in the declining phase, suggesting a contribution of hnRNP I trafficking in shaping the Per2 oscillations (for review, see Ref. 33). To further explore our findings, the import kinetics could be compared to those of other M9 domain-containing plant proteins, and the measurements have to be extended to a full 24-h circadian cycle. We are working toward a microscope system allowing continuous monitoring over such a long time span without damage to the plant. Furthermore, the expression of AtGRP7-DRONPA-s in mutants with a defective circadian clock will be revealing.

Nuclear export of AtGRP7-DRONPA-s

Recombinant AtGRP7 is imported into the nucleus of BY-2 cells (16) and transgenic Arabidopsis plants (Figure 5). To determine whether AtGRP7-DRONPA-s is also exported from the nucleus to the cytoplasm, photoswitching to the on-state was performed. On DRONPA-s immobilized in Mowiol, the minimal region that can be selectively switched off had a diameter of 1.15 µm (Figure 1B). Using the same experimental design, we switched off a 20 × 20-µm region of recombinant DRONPA-s embedded in Mowiol by exposure to 488 nm with an excitation power of 10 µW (corresponding to 3 kW/cm2 laser intensity) and an exposure time of 5 ms/pixel. The minimal region that can be selectively switched back to the on-state was determined to have a diameter of 2.7 µm (Figure S5). This assures that during photoactivation experiments only fusion proteins inside the nucleus are photoswitched. Using these laser settings, the fluorescence in the entire cell was switched off (Figure 7B). Subsequently, a small population of the AtGRP7-DRONPA-s fusion protein was photoactivated inside the nucleus (Figure 7C). The fluorescence loss in the nucleus and the concomitant accumulation of the fluorescence signal in the cytoplasm indicate that AtGRP7 is exported from the nucleus (Figure 7D–J).

image

Figure 7. Export of AtGRP7-DRONPA-s from the nucleus of a root cell. Shown are images taken before switching off (A) and the first image after switching off the AtGRP7-DRONPA-s fluorescence in the entire cell (B). After photoactivation of AtGRP7-DRONPA-s in the nucleus, images were taken at 1-min intervals (C–J). Six independent experiments showing this behavior have been performed. The scale bar corresponds to 10 µm.

Download figure to PowerPoint

Differential subcellular localization is widespread for mammalian hnRNP proteins. While essentially nuclear at steady state, several hnRNP proteins continuously shuttle between the nucleus and cytoplasm. hnRNP A1 is particularly well studied and shuttles through the action of its M9 domain (15).

In plants, bidirectional movement between the nucleus and the cytoplasm has been described for serine/arginine (SR)-rich proteins, an important class of splicing regulators with N-terminal RRMs and C-terminal domains rich in Arg and Ser. Using fluorescence loss in photobleaching (FLIP) of GFP-tagged RSZp22 expressed under the CaMV promoter, nucleocytoplasmic shuttling was shown in transient expression (23). AtGRP7 is the first hnRNP-like protein in plants shown to undergo bidirectional trafficking between the nucleus and the cytoplasm. AtGRP7 does not harbor a classical leucine-rich nuclear export sequence (NES) that mediates export by CRM1/XPO1/exportin1 and thus the components mediating AtGRP7 export are not known (34). In mammals, the M9 domain is thought to mediate both import and export and in fact, site-specific mutagenesis failed to separate import and export activity (15). Whether the M9 domain mediates protein export from the nucleus in plants is presently unknown.

Conclusion

The codon-optimized synthetic DRONPA-s can be stably expressed at high levels in transgenic plants. It retains the photoswitching properties of the original DRONPA molecule from Pectiniidae(18). Thus, DRONPA-s allows combining fluorescence recovery after selective photoswitching and photoactivation experiments in the same cell. By repeating FRAP experiments in the same cell, statistically significant data can be obtained on the level of a single cell. This is particularly important in plants because young meristematic cells with higher metabolic activity are located close to older, differentiated cells.

The synthetic DRONPA-s is also better adapted to the codon usage of mammals than the original DRONPA sequence cloned from the coral Pectiniidae and indeed high-level expression was obtained in HeLa and COS-7 cells. Therefore, the synthetic variant can be used advantageously for various applications.

The single-molecule sensitive confocal fluorescence microscope we set up to monitor intracellular protein dynamics allowed the use of a low laser intensity and thus reduced off-switching of DRONPA-s during imaging. Fluorescence recovery after selective photoswitching measurements revealed that AtGRP7-DRONPA-s expressed from the native promoter undergoes carrier-mediated transport into the nucleus. Using photoactivation, we showed that AtGRP7-DRONPA-s is also exported from the nucleus into the cytoplasm. Thus, AtGRP7 is the first hnRNP-like protein in plants shown to undergo bidirectional nucleocytoplasmic trafficking.

Our data suggest that this transport is faster at zt11 when the transcript levels are at their daily maximum than at zt4 when the transcript levels are at their daily minimum. Whether this impacts AtGRP7 function, e.g. the regulation of other oscillating downstream target transcripts (4, 7), is subject of further investigations.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Results and Discussion
  4. Materials and Methods
  5. Acknowledgments
  6. References
  7. Supporting Information

Recombinant glutathione S-transferase DRONPA-s fusion protein

DRONPA-s was created by adaptation of the codon usage to the bias of A. thaliana using codon adaptation index (CAI) performed by GeneArt using their in-house proprietary software GeneOptimizer® (Figure S1). Furthermore, cryptic splice signals, TATA boxes, ribosomal entry sites, repeat sequences and extensive RNA secondary structure were avoided and a Kozak sequence surrounds the start codon (Figure S2). The synthetic gene is available from the authors upon request.

DRONPA-s was fused in frame to glutathione S-transferase in the vector pGEX-6P1 (GE Healthcare). Recombinant GST-DRONPA-s was expressed in E. coli BL21 (DE5) cells and purified from crude lysate by batch absorption on GST agarose as described (7). The DRONPA-s moiety was recovered by cleavage of the fusion protein with PreScission protease (GE Healthcare) on the column according to the manufacturer's recommendation.

Experimental setup—single-photon confocal microscope

A home-made confocal laser scanning microscope with single photon sensitivity was used for fluorescence imaging. Briefly, the setup consists of an inverted microscope (Axiovert 200M; Zeiss) equipped with a piezo stage (PI-509; Physik Instrumente). The piezo stage that moves the specimen through the focus is controlled by an analog output card (PCI-6713; National Instruments). An argon-ion laser (Ion Laser Technology) emitting at 488 nm and a laser diode (Vioflame, Coherent) emitting at 404 nm are used for excitation. Both laser sources are controlled by a shutter and a variable attenuator. The 488-nm laser light beam was extended using a telescope, overlaid with the 404-nm laser beam by the use of a dichroic mirror and coupled into the microscope where it was directed into the objective (Plan Apo, 60×/1.4 oil; Olympus). The light beam was focused onto the sample and fluorescence light was collected by the same objective and spectrally separated using a dichroic beamsplitter (z 405/488; Chroma Technology). The parallel light beam was first focused onto a 100-µm pinhole and then focused onto the active area of an avalanche photodiode (AQR-16; Perkin Elmer) passing an emission filter (HQ 500 LP; Chroma Technology). The electric signal was then processed by a time-correlated single-photon counting device (SPC-630; Becker & Hickl).

The whole set-up is controlled by a software based on matlab (The MathWorks) that controls the movement of the piezo scan stage and attributes photon information obtained by the photon-counting device to the appropriate x- and y-positions on the sample.

Fluorescence spectra of recombinant DRONPA-s in solution

Purified DRONPA-s protein was diluted with PBS to 10−5m in a standard quartz cuvette. The emission spectra of DRONPA (≥500 nm) were measured with a fluorescence spectrometer (Cary Eclipse, Varian) at 488-nm excitation wavelength.

Immobilization of recombinant DRONPA in the polymer Mowiol

Mowiol polyvinyl alcohol (PVA) (Calbiochem) was heated to 37°C and mixed with a solution of recombinantly expressed DRONPA-s (0.5 mg/mL in PBS). Microscopy coverslips were immediately coated using a home-made spin coater resulting in a film only a few micrometers thick. Measurements were performed on the setup described above.

Construction of the AtGRP7-DRONPA-s fusion

To allow in-frame fusion of DRONPA-s to the C-terminus of the AtGRP7 coding sequence (including the intron), the AtGRP7 stop codon was replaced by a BamHI site. The AtGRP7-DRONPA-s fusion protein was expressed under control of 1.4 kb of the AtGRP7 promoter, the AtGRP7 5′ UTR and 3′ UTR (construct AtGRP7:AtGRP7-DRONPA-s). As a control, DRONPA-s was placed under control of the same regulatory elements (35). The cassettes were inserted into HindIII-XbaI-cut pUC19 polylinker of the binary vector pHPT1 (31).

Plant transformation and growth

Agrobacterium-mediated stable transformation of A. thaliana Col was performed using vacuum infiltration (36). Seeds were germinated on one-half strength MS plates (37) containing 0.5% sucrose supplemented with the appropriate antibiotic and grown in 16-h light/8-h dark cycles at 20°C in a Percival growth chamber.

Transient expression in Nicotiana benthamiana

Both, DRONPA-s and DRONPA from the commercial vector DRONPA-GREEN pDG-S1 (MLB Group Company), were placed under control of the CaMV promoter in the vector pRT104 (38). The cassettes were inserted into the binary vector pCAMBIA 3000 and the constructs were introduced into Agrobacterium tumefaciens GV3101 (pMP90) (39).

Agrobacterium cultures were independently centrifuged at 8000 ×g for 15 min, washed once with and resuspended in infiltration media (10 mm MES–KOH pH 5.7, 10 mm MgCl2, 100 µm acetosyringone) and adjusted to an optical density (OD)600 of 0.8. Cultures expressing DRONPA-s or DRONPA-GREEN were mixed with an equal volume of agrobacteria expressing the viral silencing suppressor p19. Multiple individual leaves (4–5 leaf stage, ca. 10 cm high) were mechanically infused by pressing the tip of the syringe against the lower surface of the leaf and applying gentle pressure to the plunger.

Four days after infiltration, leaves were harvested and quick-frozen in liquid N2. For determination of the DRONPA fluorescence, native protein extracts were prepared and diluted to a concentration of 5 µg/mL with PBS. After excitation at 480 nm, the fluorescence emission spectrum was recorded (500–600 nm). Peak fluorescence intensity at 516 nm was expressed as means ± SD (n = 10). Student's t-test was used to determine whether the differences were statistically significant.

Transfection of HeLa and COS-7 cells

Both, DRONPA-s and DRONPA from the commercial vector DRONPA-GREEN pDGS1 (MLB Group Company), respectively, were inserted into the vector pcDNA3.1+ under control of the CMV promoter. Transfection of confluent HeLa or COS-7 cells was performed with TurboFect (Fermentas) according to the manufacturer's instructions. Twenty-four hours after transfection, proteins were extracted using the mammalian protein extraction reagent (M-PER) mammalian protein extraction reagent (Thermo Scientific) according to the manufacturer's instructions and DRONPA fluorescence was determined as described for tobacco leaves.

Fluorescence recovery after selective photoswitching

For fluorescence recovery after selective photoswitching measurements, the selected area of the sample was moved to the laser focus by means of the piezo stage. A preswitch image was acquired with the setup described above. Then the laser power was increased to 10 µW for 10 seconds through a variable attenuator in order to switch DRONPA-s to the off-state. The switched region corresponds to the dimensions of the laser focus. The cell was scanned directly after the end of the switching process. Subsequently, images were acquired at 2-min intervals. The mean gray values in the nucleus were calculated for each image and plotted versus the time. A mono-exponential function f(t) = A× (1 − ekt), where t is the time, A the apparent saturation and k the recovery constant, was fitted to the data points to extract the t1/2 value. The images were stored at 16-bit resolution and analyzed using ImageJ (U.S. National Institutes of Health; http://www.rsbweb.nih.gov/ij/). A Student's t-test was used to evaluate the differences in recovery half-times at zt4 and zt11, respectively. A p value < 0.05 was considered significant.

Immunoblot analysis

Total protein extracts from Arabidopsis were prepared as previously described (40). Transfected tobacco leaves were homogenized in 50 mm Tris–HCl, pH 7.5, 150 mm NaCl, 0.1% (v/v) Tween-20, 0.1% (v/v) β-mercaptoethanol (41). Protein extraction from HeLa and COS-7 cells was performed with M-PER mammalian protein extraction reagent (Thermo Scientific) according to the manufacturer's instruction. Incubation of protein gel blots with antipeptide antibodies raised against AtGRP8 and AtGRP7 followed by chemiluminescence detection was performed as previously described (3,7). The DRONPA antibody and neomycin phosphotransferase antibody were obtained from MBL International and Millipore, respectively, and used according to the manufacturers' instructions.

Determination of flowering time

Plants were grown in a randomized manner on soil in short days (8-h light–16-h darkness) at a constant temperature of 20°C in Percival AR66-L3 incubators (CLF Laboratories). The flowering time was determined by counting rosette leaves once the bolt was 0.5 cm tall. Normal distribution was proven by Kolmogorov–Smirnov test. Mean value and standard deviation were calculated using statistica 6.0 (Stat. Soft, Inc., 2001). Significance was determined using anova and the p value was calculated by post hoc Dunnett's test (3).

Acknowledgments

  1. Top of page
  2. Abstract
  3. Results and Discussion
  4. Materials and Methods
  5. Acknowledgments
  6. References
  7. Supporting Information

We thank Klaus Harter, University of Tübingen, for valuable suggestions on N. benthamiana transient expression and fluorescence determination. This work was supported by the DFG (SFB 613 and STA 653/2).

References

  1. Top of page
  2. Abstract
  3. Results and Discussion
  4. Materials and Methods
  5. Acknowledgments
  6. References
  7. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Results and Discussion
  4. Materials and Methods
  5. Acknowledgments
  6. References
  7. Supporting Information

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Figure S1: Alignment of the DRONPA nucleotide sequence of the commercial vector DRONPA-GREEN pDG-S1 (MLB Group Company) and the codon-optimized DRONPA-s. Changes in DRONPA-s are highlighted in black. The deduced amino acid sequence is shown below the nucleotide sequences.

Figure S2: Photoswitching performance of DRONPA-s. A) CAI for the original DRONPA and the codon-optimized DRONPA-s, determined by GeneArt using their in-house proprietary software GENEOPTIMIZER®. The histograms show the percentage of sequence codons, which fall into a quality class (defined by the frequency of usage). The quality of the most often used codon for a given amino acid is set to 100, the remaining codons are scaled accordingly (42). B) Details on sequence motifs removed during the optimization process.

Figure S3: Photoswitching performance of DRONPA-s. A) Absorption spectrum (blue, broken line) and emission spectrum (green, solid line) of the synthetic DRONPA-s in solution. B) Decay in fluorescence intensity after irradiation with 488 nm (20 mW/cm2). Each data point corresponds to the integral of the emission spectrum between 500 and 580 nm.

Figure S4: Functional complementation of the atgrp7-1 T-DNA insertion line with the AtGRP7-DRONPA-s fusion protein. A) An immunoblot of total protein extracts from atgrp7-1, Col-0 wt and two independent transformants of AtGRP7-DRONPA-s in atgrp7-1 was probed with an antibody against DRONPA to detect the fusion protein (the asterisk indicates a degradation product) (top) and reprobed with a light-harvesting chlorophyll-binding protein (LHCP) antibody to confirm equal loading. In parallel, a blot was probed with an antibody against AtGRP7 (middle). The part corresponding to endogenous AtGRP7 protein is shown. The blot was reprobed with the LHCP antibody to confirm equal loading. On the blot probed with an antibody against AtGRP8 (bottom), unspecific cross-reaction with the large subunit of ribulose-1,5-bisphosphate carboxylase (RbcL) confirms equal loading. B) The flowering time of Col-0 wt, atgrp7-1 and AtGRP7-DRONPA-s in atgrp7-1 grown in short days was compared by determining total leaf number. Data are based on two independent experiments with n ≥ 20 each. The leaf number in atgrp7-1 and the complemented line is significantly different (p value = 0.000385).

Figure S5: Switching behavior of immobilized DRONPA-s. A) A 20 × 20-µm region of recombinant DRONPA-s embedded in a Mowiol matrix is switched off by exposure to 488 nm. Subsequently, a small spot in the region of the laser focus is switched back to the fluorescent state by irradiation with 404-nm light. Fluorescence intensity is indicated by the quantitative color code: yellow corresponds to the highest fluorescence intensity. B) Calculation of the spot size with the full width at half maximum (FWHM) convention. Shown are the fluorescence intensities along the indicated line in (A).

FilenameFormatSizeDescription
TRA_1180_sm_figure_S1.jpg956KSupporting info item
TRA_1180_sm_figure_S2.jpg310KSupporting info item
TRA_1180_sm_figure_S3.jpg498KSupporting info item
TRA_1180_sm_figure_S4.jpg201KSupporting info item
TRA_1180_sm_figure_S5.jpg251KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.