Xinyang
Liao
,
Ben
Zhang‡
,
Michael R.
Blatt
and
Gareth I.
Jenkins
*
Institute of Molecular, Cell and Systems Biology, College of Medical, Veterinary and Life Sciences, Bower Building, University of Glasgow, Glasgow G12 8QQ, UK. E-mail: Gareth.Jenkins@Glasgow.ac.uk
First published on 4th December 2018
The photoreceptor UVR8 has a pivotal role in mediating plant responses to UV-B wavelengths. Dimeric UVR8 dissociates into monomers following UV-B photoreception, and there is evidence that this process is accompanied by conformational changes that may facilitate interaction of UVR8 with other proteins to initiate signaling. Hence monitoring UVR8 dimer/monomer status and conformation is key to understanding UVR8 action. Here we have used Fluorescence Resonance Energy Transfer (FRET) to study these processes in both wild-type and mutant UVR8 proteins in vivo. UVR8 was fused to GFP and mCherry at the C- and N-termini, respectively and both the FRET efficiency and loss of GFP fluorescence after photobleaching were measured. In addition, measurements were made for UVR8 fused to either GFP or mCherry to eliminate intra-molecular FRET signals. The results indicate that dissociation of UVR8 dimer to monomer principally accounts for the loss of FRET signal for wild-type UVR8 and there is little evidence of a contribution from conformational change in vivo. Examination of plants expressing UVR8W285F and UVR8D96N,D107N are consistent with these mutant proteins being constitutively dimeric and monomeric, respectively. The methods employed here will be valuable for monitoring UVR8 dimer/monomer status in vivo in relation to signaling, and will facilitate characterization of dimer/monomer status and conformation of further UVR8 mutants.
UVR8 is a 7-bladed β-propeller protein that exists as a homodimer in the absence of UV-B.3,4 UV-B exposure induces dissociation of the dimer to produce monomers that initiate downstream responses through interaction with the protein CONSTITUTIVELY PHOTOMORPHOGENIC 1 (COP1)5 and with specific transcription factors.6,7 Further proteins, termed REPRESSOR OF UV PHOTOMORPHOGENESIS (RUP) proteins, interact with UVR8 to displace COP1 and promote re-association of monomers to form the dimer.8 Under photoperiodic illumination with light containing UV-B both monomer formation and re-dimerization occur, resulting in establishment of a photo-equilibrium of UVR8 dimer and monomer.9
There is evidence that UV-B photoreception by UVR8, with the resulting monomerization results in conformational changes to the protein.5,10–13 However, since the N- and C-termini do not appear in the crystal structures of UVR8,3,4,12 detailed information about their location and conformational changes following monomerization are lacking. It is proposed that conformational changes are likely to increase accessibility of the C-terminus of monomeric UVR8 to COP1. The interaction between UVR8 and COP1 is UV-B-dependent14 and involves binding to a specific region within the C-terminus.15,16 A recent study13 indicates that the UVR8 monomer can adopt multiple conformations with an extended C-terminus.
Fluorescence Resonance Energy Transfer (FRET) is a commonly used method in confocal microscopy to monitor interactions between two proteins. It involves the use of two different fluorescent molecules, such that excitation of one results in radiationless excitation energy transfer that activates the other, leading to a fluorescent signal that can be measured. FRET will only occur if the donor and acceptor molecules are in sufficiently close proximity. For example, in plant photobiology FRET has been used to show an interaction between phytochrome B and cryptochrome 2 photoreceptors.17 However, FRET can also be adapted to monitor the proximity of two regions within the same molecule; such intramolecular or ‘single-molecule’ FRET involves introducing two fluorophores into the molecule, often at the N- and C-termini. For example, this approach has been used elegantly to study the structural dynamics of potassium channels18,19 and to monitor the voltage-driven conformational changes associated with channel voltage-sensor domain interactions with vesicle trafficking proteins.20
FRET has the potential to provide valuable insights into UVR8 molecular function. The extent of monomer formation is key to UVR8 function, because the monomer is the form that initiates signaling. As reported here, FRET enables the dimer/monomer status of UVR8 to be monitored in vivo, both in relation to environmental stimuli and also in plants expressing specific mutant UVR8 proteins. Various mutations have been found to alter the strength of the UVR8 dimer, and in some cases constitutively dimeric and monomeric mutant proteins have been identified.3–5,21,22 Moreover, intra-molecular FRET has the potential to monitor conformational changes in the UVR8 protein, either as a result of UV-B exposure or in mutant proteins. In this study we developed FRET-based methods for monitoring the relative positions of the N- and C-termini of UVR8 to gain insights into UVR8 dimer/monomer status and conformation. In addition, the methods employed here will facilitate characterization of further UVR8 mutants in vivo.
Four to six-week-old Nicotiana tabacum plants were used for leaf infiltration for transient transformation. Plants were well watered on the day before infiltration. The lower epidermis of the chosen leaf was grazed by a blade or a needle. Then, the prepared Agrobacterium suspension was injected into the wound on the abaxial leaf epidermis by a needle-less 1 ml syringe. Plants were moved back to the growth chamber and the expression was observed after 2–3 days by confocal microscopy.
Where indicated, leaves were exposed to broadband UVB-313 fluorescent tubes (Q-Panel Co., USA; Fig. 2-1C) covered by cellulose acetate film (FLM400110/2925, West Design Products) to filter out UV-C. The spectrum of the UV-B source is published in Cloix et al. (2012). A Skye Spectrosense 1 meter (Skye Instruments) with a SKU 430 sensor (Skye Instruments) was used to measure the UV-B fluence rates.
As shown in Fig. 2A, the expected fluorescence is seen when the negative control proteins UVR8-GFP and mCherry-UVR8 are separately excited at 488 nm and 552 nm respectively, but no FRET signal was seen following excitation at 488 nm, since this would require the presence of UVR8 fused to both GFP and mCherry. FRET images for the UVR8 FRET-pair constructs were captured under the same microscopy settings (Fig. 2B) and the FRET efficiency of these constructs was quantified. The FRET efficiency was calculated as the ratio of mean fluorescence intensity between FRET signal [mCherry (488)/GFP (488)] and GFP signal in randomly selected areas (Fig. 2C). The data for wild-type UVR8 show a relatively high FRET signal prior to UV-B exposure, when the protein is in the dimeric form, and a decrease in FRET efficiency following UV-B exposure, when dimers dissociate into monomers. For the UVR8 dimer, the FRET signal could potentially come from two components (Fig. 1C): (i) energy transfer from the C-terminus to the N-terminus of the same monomer molecule; (ii) energy transfer between the two monomer molecules in the dimer. When UVR8 is monomerized by UV-B, or because of mutations, the second source of the FRET signal is lost, but the first type of FRET signal is still possible.
Examination of the UVR8 mutant fusions shows a correlation between FRET efficiency and dimer/monomer status (Fig. 2C). The FRET value of the constitutively dimeric UVR8W285F is relatively high whereas that for constitutively monomeric UVR8D96N,D107N is relatively low. In both mutants the FRET efficiency is unaffected by UV-B treatment. Moreover, results for UVR8D96N,D107N demonstrate that the distance between the N-terminus and C-terminus of UVR8 is close enough for FRET, indicating that this method is suitable to investigate any conformational change of UVR8. However, there is no evidence for any conformational change for UVR8D96N,D107N following UV-B exposure. Since there is a large change in FRET efficiency between the dimeric and monomeric mutants, it is likely that the decrease in FRET signal in wild-type UVR8 following UV-B exposure is principally due to dissociation of the dimer to produce monomers.
Fig. 4 Photobleaching results with the pFRET-NcCg-DEST vector. GFP fluorescence intensity was analysed before and after photobleaching for mCherry-UVR8-GFP, mCherry-UVR8D96N,D107N-GFP and mCherry-UVR8W285F-GFP expressed from the pFRET-NcCg-DEST vector (fusions abbreviated in the figure to UVR8, UVR8D96N,D107N and UVR8W285F). Plants were exposed (+UV-B) or not (−UV-B) to 3 μmol m−2 s−1 broadband UV-B for 1 hour before being used for experiments. Primary data obtained as described in Fig. 3 are shown in Fig. S3,† together with the negative control results. The percentage increase in GFP fluorescence intensity after photobleaching for each construct was calculated by [(IPb − IPre)/IPre] × 100, where IPre and IPb indicate GFP intensities in the bleaching area before and after photobleaching, respectively. Means with S.E. are shown (n > 5). Significance at P < 0.01 is indicated by letters. |
The primary photobleaching data for UVR8, UVR8D96N,D107N and UVR8W285F are shown in Fig. S3A† and the corresponding negative controls are shown in Fig. S3B.† For the negative controls, there is no significant difference of GFP intensity between pre-bleaching and post-bleaching for all samples. For wild-type UVR8, the percentage increase in GFP fluorescence after photobleaching is greater before than after UV-B exposure (Fig. 4), indicating that UV-B causes a decrease in FRET. The photobleaching of UVR8W285F leads to a strong increase of GFP fluorescence intensity, similar to that for wild-type UVR8 without UV-B, whereas UVR8D96N,D107N shows a smaller change, similar to wild-type UVR8 after UV-B treatment. The fluorescence signals for both mutants are unaffected by UV-B treatment (Fig. 4). These results are consistent with the FRET efficiency data (Fig. 2C) and indicate that FRET is greater in dimeric UVR8 than in the monomeric proteins.
The 2in1 constructs with UVR8, UVR8D96N,D107N and UVR8W285F were transiently expressed in tobacco and checked for protein expression and dimer/monomer status by immunodetection with the appropriate antibodies (Fig. S2†). FRET experiments were performed as described previously and the fluorescence images are shown in Fig. 6A. Quantification of the FRET efficiencies (Fig. 6B) shows that all values are lower than those obtained with the pFRET-NcCg-DEST vector (Fig. 2), possibly because of the loss of intramolecular FRET signals. However, the overall trend for all UVR8 variants is similar to that seen with the pFRET-NcCg-DEST vector. UVR8 and UVR8W285F have a relatively strong FRET signal in the absence of UV-B, whereas UVR8D96N,D107N shows a very weak FRET signal. Wild-type UVR8 shows a FRET signal after UV-B exposure, most likely because of incomplete monomerization and the reversion of some monomers to dimers. Monomeric UVR8D96N,D107N still has a detectable FRET signal, possibly because of weak transient interactions between monomers in vivo, which were observed in a previous study.22
The photobleaching method was also carried out with these constructs and the primary data are shown in Fig. S4.† The percentage increase in GFP fluorescence after photobleaching was calculated as described previously and results are shown in Fig. 7. The overall trend in the results is similar to that for the FRET efficiency (Fig. 6); UVR8 monomer has a reduced signal relative to the dimer, and for UVR8D96N,D107N there is no significant increase in GFP intensity after photobleaching (Fig. 7), suggesting that essentially no FRET happens for this constitutive monomer mutant.
Fig. 7 Photobleaching results with the pFRET-2in1-DEST vector. Analysis of GFP fluorescence intensity before and after photobleaching of mCherry-UVR8-mCherry/GFP-UVR8-GFP, mCherry-UVR8D96N,D107N-mCherry/GFP-UVR8D96N,D107N-GFP and mCherry-UVR8W285F-mCherry/GFP-UVR8W285F-GFP expressed from the pFRET-2in1-DEST vector (fusions abbreviated in the figure to UVR8, UVR8D96N,D107N and UVR8W285F). Plants were exposed (+UV-B) or not (−UV-B) to 3 μmol m−2 s−1 broadband UV-B for 1 hour before being used for experiments. Primary data (shown in Fig. S4†) were obtained as described in Fig. 3. The percentage increase in GFP fluorescence intensity after photobleaching was calculated by [(IPb − IPre)/IPre] × 100, where IPre and IPb indicate GFP intensities in the bleaching area before and after photobleaching, respectively. Means with S.E. are shown (n > 5). Significance at P < 0.01 is indicated by letters. |
To measure the FRET efficiency, two different approaches were used. The first involves calculation of the fluorescence intensity ratio of acceptor to donor (Fig. 2 and 6). For this method, there are several potential sources of error. First of all, the protein expression level can have a large impact on the results. Overexpression causes the signal to be too strong, and low expression may cause false negative results. Second, because of the spectral overlap of the donor and acceptor, the spectral bleed-through of exciting light may cause a false positive.24,25 This could explain why UVR8D96N,D107N expressed from the pFRET-2in1-DEST vector still has some FRET signals (Fig. 6). Moreover, when GFP is activated, it excites mCherry almost at the same time, so the GFP fluorescence intensity obtained here is actually the signal after FRET, suggesting that the actual GFP intensity should be higher than the observed value and the ratio of mCherry to GFP should be lower. The second approach employed was acceptor photobleaching, which measures the energy loss of the donor during FRET and therefore avoids the complications associated with measuring the ratio of signals. On the other hand, a potential disadvantage of this method is that it may cause a false negative if the intensity change of the donor is too low to be detected. In this study, both methods were shown to be feasible and to produce similar results. However, the photobleaching method may be more accurate for the reasons stated above.
The two different vectors used in this study enable different molecular properties to be characterized. The pFRET-NcCg-DEST vector is designed to test the conformational change of a single molecular protein and is therefore ideal for the monomeric form of UVR8. However, UVR8 exists in a dimer/monomer photo-equilibrium in light-grown plants and unfortunately the UVR8 dimer cannot be completely removed in vivo because monomers continually re-associate to form dimers,9 which complicates interpretation. For the UVR8 dimer, as discussed above, there are two different sources of the FRET signal, the intermolecular signal between two monomers and the intra-monomer signal. It is difficult to calculate the relative contribution of these components to the FRET signal, but use of the pFRET-2in1-DEST vector23 can overcome this problem because only intermolecular FRET is possible. A major advantage of this vector is that it can express two different gene fusions in a 1:1 ratio, ensuring both proteins are expressed at very similar levels. In this study, it was used to assay FRET signals between two differently labeled UVR8 monomers. However, the dimer/monomer photo-equilibrium still probably affects the result. The differently labeled monomers will randomly combine to form three kinds of dimers, but only dimers of GFP-UVR8-GFP with mCherry-UVR8-mCherry will generate FRET. Dimers with the same tags, GFP-UVR8-GFP with GFP-UVR8-GFP or mCherry-UVR8-mCherry with mCherry-UVR8-mCherry, do not contribute to the FRET signal, but their fluorescence will interfere with the FRET efficiency calculation.
The results obtained for UVR8 and the two mutants are largely consistent with previous reports but also raise some interesting points. UV-B induced monomerization of wild-type UVR8 strongly reduces the FRET signal in all of the experiments shown here. It is interesting that the results with the pFRET-NcCg-DEST and pFRET-2in1-DEST vectors are similar, because this suggests that there is relatively little, if any, change in the intramolecular FRET signal, and hence little UV-B induced change in UVR8 conformation. Several previous studies using a number of techniques have provided evidence for conformational change of UVR8 associated with UV-B photoreception.5,10–12 In addition, a recent report shows that the monomer can adopt multiple conformations, including partial unraveling of the β-propeller.13 Production of the most extended conformation may be required to facilitate interaction with other proteins, in particular at the C-terminal region, to initiate signaling. However, the above studies of conformational changes were undertaken with purified protein, whereas the present experiments were with UVR8 expressed in vivo, the major difference being that UVR8 is in a cellular environment and can bind to proteins, such as the RUPs, that may stabilize the structure. An additional point to consider is that detection of a change in the intramolecular FRET signal depends on a change in distance between the N- and C-termini of the protein, and it is quite possible for conformational changes to occur without much change in their relative positions. It will therefore be valuable to undertake further studies of UVR8 conformational change in vivo using additional approaches.
The FRET results for UVR8W285F are consistent with it being a constitutive dimer that is unresponsive to UV-B treatment, as reported previously both for the purified protein and in vivo.3–5,21,26 UVR8D96N,D107N is a constitutively monomeric mutant,22 and is therefore not expected to show an intermolecular FRET signal. The very low signal seen using photobleaching with the pFRET-2in1-DEST vector is likely to be due to either transient, weak interaction between monomers22 and/or a false positive caused by spectral bleed-through of the exciting light, as mentioned above. Since intermolecular FRET is very low, the pFRET-NcCg-DEST vector is suitable to investigate any conformational change of UVR8D96N,D107N. The results (Fig. 2C and 4) show that UV-B exposure does not change the FRET efficiency, indicating that there is no conformational change, or at least no change that significantly alters the relative positions of the N- and C-termini. UVR8D96N,D107N is functional in vivo in initiating responses to UV-B. However, in contrast to wild-type UVR8, UVR8D96N,D107N can constitutively interact with COP1, although binding does increase after UV-B exposure. Hence, the mutation may produce a protein conformation that is normally only produced in wild-type UVR8 by UV-B photoreception. One possibility is that the UVR8D96N,D107N monomer adopts one of the extended forms described by Camacho et al. (2018),13 enabling it to interact with COP1, and that UV-B photoreception further activates the protein to generate the signaling-active state. Research to date therefore suggests that UV-B plays at least two roles in UVR8 signaling: inducing the monomerization of UVR8 homodimer, which is necessary for UVR8 to interact with COP1, and activating UVR8 monomer to initiate signaling.
Footnotes |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8pp00489g |
‡ Present address: School of Life Science, Shanxi University, Taiyuan, 030006, China. |
This journal is © The Royal Society of Chemistry and Owner Societies 2019 |