Time-Resolved Infrared Spectroscopy on Plant Cryptochrome

Article Cite This: J. Phys. Chem. A XXXX, XXX, XXX−XXX

pubs.acs.org/JPCA

Time-Resolved Infrared Spectroscopy on Plant CryptochromeRelevance of Proton Transfer and ATP Binding for Signaling Published as part of The Journal of Physical Chemistry virtual special issue “Time-Resolved Vibrational Spectroscopy”. Lea Schroeder, Sabine Oldemeyer, and Tilman Kottke* Physical and Biophysical Chemistry, Department of Chemistry, Bielefeld University, Universitätsstraße 25, 33615 Bielefeld, Germany S Supporting Information *

ABSTRACT: Plant cryptochromes are light receptors in land plants and algae with very diverse functions such as circadian timing and lifecycle progression. The receptor consists of a photolyase homology region (PHR) binding the flavin chromophore and a C-terminal extension (CCT) responsible for signaling. The reputed signaling state, the flavin neutral radical, is formed by a femtosecond electron transfer and microsecond proton transfer to the excited, oxidized flavin. Subsequently, a 500 μs loss of β-sheet structure ∼25 Å away from flavin was resolved and suggested to be part of the signal conduction to the CCT. Here, we performed time-resolved, stepscan Fourier transform IR spectroscopy on the PHR of the plant cryptochrome pCRY (formerly CPH1) from Chlamydomonas reinhardtii. In a mutant lacking the proton donor aspartic acid 396 only the flavin anion radical is formed, but we observed the loss of β-sheet structure with a time constant of 1.3 ms, similar to the 500 μs of the wild type. This finding implies that the anion radical may be considered signaling-competent. In the steady state, a variation of external pH up to 8.3 did not have any effect on the difference spectra including the protonated state of Asp396. However, we detected the prominent loss of β-sheet structure by illumination only in the presence of adenosine triphosphate (ATP). We conclude that the bound ATP stabilizes these lightinduced changes in secondary structure to ensure a physiological lifetime compatible with signaling by plant cryptochrome.

■

INTRODUCTION

reinhardtii. pCRY was formerly referred to as CPH1 (Chlamydomonas photolyase homologue 1)18,19 and acts as a photoreceptor in both circadian timing and lifecycle progression of the alga.20 The shared PHR domain results in a common photochemistry, which is initiated through absorption of blue light by FAD (Figure 1). Electron transfer to the excited FAD occurs with a time constant of 400 fs21 similar to that of a distant cryptochrome PHR from diatoms.22 The electron is transferred to the FAD by a nearby tryptophan, which is part of highly conserved tryptophan triad,16,23 albeit alternate electron transfer pathways exist that can be opened by nucleotide binding.24 Hole transport in the triad proceeds with a time constant of 30 ps concomitant with vibrational cooling.21 Formation of the FAD neutral radical from the anion radical takes place with a time constant of 2 μs separate from electron transfer.25−27 The proton donor has been identified as the nearby Asp396 (in AtCry1, Asp393 in pCRY, AtCRY1

The flavoprotein family of photolyases and cryptochromes can be found through all kingdoms of life.1 Whereas photolyases repair DNA lesions using ultraviolet A (UVA) and blue light,2 cryptochromes perform additional diverse functions.1 They are responsible for blue light sensing in plants,3−5 algae,6 fungi,7 bacteria,8 and insects.9 Even a sensitivity to red light has been demonstrated.10 Moreover, cryptochromes act light-independently as part of the internal clock in insects11 and mammals12 and are involved in magnetoreception in flies and cockroaches.13,14 Plant cryptochromes contain a photolyase homology region (PHR), which binds the flavin adenine dinucleotide (FAD) cofactor noncovalently.15 The PHR domain consists of ∼500 amino acids forming an α-helical subunit, where FAD is bound, and an α/β-subunit.16 The C-terminal extension of variable sequence and length is responsible for signal transduction.17 Physicochemical characterization has been performed on three homologous plant cryptochromes, namely, cryptochromes 1 and 2 from Arabidopsis thaliana (AtCRY1 and AtCRY2) and pCRY from the green alga Chlamydomonas © XXXX American Chemical Society

Received: October 16, 2017 Revised: December 4, 2017 Published: December 6, 2017 A

DOI: 10.1021/acs.jpca.7b10249 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

CRYβ can be formed without proton transfer. Previous results from steady-state experiments indicate that at least a change in turn elements is preserved in the D396C mutant. 28 Accordingly, the roles of proton transfer and ATP binding in signaling need to be clarified. Here, we investigated pCRY-PHR by time-resolved, stepscan FT-IR spectroscopy. We show that Asp396 is protonated in the dark irrespective of bound ATP and from pH 7.1 to 8.3. ATP binding stabilizes strongly not only the FAD radical as the photoproduct but also the CRYβ intermediate. Even in the D396C mutant, CRYβ formation is observed with τ = 1.3 ms, decoupling the conformational changes from the proton transfer.

■

MATERIALS AND METHODS Protein Expression and Purification. Wild-type pCRYPHR and pCRY-PHR-D396C were expressed and purified as described previously.30,38 The protein was finally obtained in 50 mM sodium phosphate buffer, pH 7.8, 100 mM NaCl, 20% (v/ v) glycerol after dialysis. UV/Vis Experiments. For UV/vis spectra of pCRY-PHR, the buffer was exchanged to 50 mM phosphate buffer, 100 mM NaCl, 20% glycerol, pH 7.2, 7.8, and 8.2, respectively, by concentrating the sample using a Vivaspin 500 filter device (Sartorius) with a 50 kDa cutoff and adding new buffer. UV/vis spectra were recorded using a Shimadzu UV-2450 spectrometer and a quartz cuvette (Suprasil, Helma) with a 1 cm path length. To investigate the influence of ATP, UV/vis spectra were recorded 15 times and averaged using 400 μL of the protein solution at pH 7.2, 7.8, and 8.2, respectively. ATP (30 mM) in 40 μL buffer with a corresponding pH were then added. The spectra were recorded again, averaged, and corrected for the dilution. Sample Preparation for FT-IR Experiments. For FT-IR spectra, the wild type and the D396C mutant were washed three times with 20 mM phosphate buffer, 100 mM NaCl, 1% glycerol using a Vivaspin 500 filter device (Sartorius) with a 50 kDa cutoff to result in pH 7.1, 7.8, and 8.3, respectively. The influence of ATP was investigated by adding ATP to 5 mM to the washing buffer. Preparation of samples with pH 7.8 and 8.3 with 5 mM ATP was performed under complete exclusion of light. Samples were concentrated to an OD450 of 10−12, equivalent to a concentration of protein-bound flavin of 0.9− 1.1 mM, shock-frozen in liquid nitrogen, and stored at −80 °C. Samples at pH 8.3 were not stored but used directly, because the protein is less stable at such high pH. Steady-State FT-IR Experiments. FT-IR samples were prepared by placing 2 μL of the protein sample on a BaF2 window (20 mm diameter) and sealing it with a second window. Samples were obtained with an amide I/water absorbance at 1650 cm−1 of 0.9−1.1 and an absorbance ratio of amide I/water to amide II at 1550 cm−1 of 2.5−2.9, which provides evidence for a full hydration of the protein. FT-IR spectroscopy was performed on Bruker IFS 66v and 66v/S spectrometers with a spectral resolution of 2 cm−1. The temperature was maintained at 10 °C. Difference spectra were recorded below 2256 cm−1 using a long wave pass filter. 1024 scans were recorded in the dark, and then the sample was illuminated for 4 s with a light-emitting diode (LED; Luxeon Star, Philips Lumileds) with a maximum emission at 455 nm and an intensity of ∼23 mW/cm−2. Directly after illumination, another 1024 scans were taken to produce a difference spectrum.

Figure 1. Photoreaction of FAD in plant cryptochromes. Arrows in green describe the complete photoreaction of the wild-type cryptochrome with the proton donor D396 present. Red arrows represent the photoreaction of the D396C mutant. The structure on the left shows the binding pocket of FAD with D396 and an ATP analogue (PDB code: 1U3D).

numbering is used in the following for clarity).28−30 The FAD neutral radical is considered to be the signaling state in vivo.31−33 Conformational changes in α-helices and turn elements have been observed by time-resolved Fourier transform infrared (FT-IR) spectroscopy concomitant with neutral radical formation (termed CRYα) and a subsequent destabilization of the β-sheet in the α/β subdomain, 25 Å apart from flavin, with τ = 500 μs (termed CRYβ).30 Finally, the Cterminal extension responds by unfolding,34 assigned by transient grating spectroscopy to τ = 400 ms.35 The role and protonation state of Asp396 have been under investigation. In pCRY, the D396C mutation inhibits the proton transfer and greatly accelerates the decay of the radical species from the minute time scale to apparent time constants of 9 and 400 μs.28 Accordingly, an intrinsic role of proton transfer for stabilization of the signaling state has been postulated.28 In AtCRY1, a pH-induced shift in the UV/vis spectrum and a severe loss in the yield of neutral radical have been attributed to the deprotonation of Asp396 in the dark with a pKa of 7.4.36 A competitive ultrafast proton transfer has been shown to be operative to some extent in AtCRY1.36 However, a significant contribution of this reaction pathway is at odds with results from time-resolved UV/vis and FT-IR spectroscopy on AtCRY1 and pCRY.21,25,26,30 Addition of adenosine triphosphate (ATP) has a strong and diverse influence on the photocycle. ATP binding induces a conformational change,37 and it strongly prolongs the lifetime of the light-induced neutral radical38 as well as anion radical (in the D396C mutant).28 It increases the reaction yield up to 14% and prevents any ultrafast proton transfer.36 For AtCRY1, the pKa of Asp396 has been deduced to shift to more than 9 upon addition of ATP.36 Furthermore, it has been observed that the cryptochrome PHR exhibits autophosphorylation activity upon ATP binding.38,39 Three pressing questions arise from these findings. First, the pKa of 7.4 attributed to Asp396 in AtCRY1 is inconsistent with many findings on pCRY performed at a pH of 7.8 in the absence of ATP. Second, the effect of ATP on light-induced conformational changes as observed in CRYα and CRYβ intermediates is unclear. Third, it remains to be shown whether B


Article

The Journal of Physical Chemistry A Time-Resolved Step-Scan FT-IR Experiments. To 10 μL of pCRY-PHR-D396C in 20 mM phosphate buffer, pH 7.8, with 100 mM NaCl and 1% glycerol, 1 μL of 100 mM ferricyanide in the same buffer was added. A droplet of 5 μL was placed on a CaF2 window (20 mm diameter), kept for 3−4 min at 500 mbar to gently reduce the water content, and then sealed with a second window. The obtained samples had an amide I/water absorbance of 0.9−1.1 and an absorbance ratio of amide I/water to amide II of 2.3−2.7. The temperature of the sample was maintained at 10 °C. Experiments were performed using an IFS 66v spectrometer (Bruker) and a photoconductive mercury cadmium telluride (MCT) detector with a spectral resolution of 8 cm−1. By using an infrared band-pass filter (Laser Components), the spectral range was restricted to 1974−988 cm−1. The sample was excited using a 10 ns light pulse with an energy of 17 mJ at 450 nm. The pulse was generated by a tunable optical parametric oscillator (Opta, Bensheim) pumped by the 355 nm third harmonic of a Nd:YAG laser (Quanta-Ray, Spectra Physics) at 10 Hz. The final 2 Hz excitation rate was generated using an optical shutter (LSTXY, nm Laser Products). The infrared intensity was monitored at 277 mirror positions with 2 to 4 coadditions in 1000 time slices of 5 μs at time points from 3.5 to 4.6 ms. Samples were used for up to 14 coadditions in total corresponding to 4000 excitations per sample before replacement. 83 experiments were averaged with 320 coadditions in total. Data Analysis. The time-resolved spectra were analyzed using MATLAB (The Mathworks). Step-scan data were averaged on a logarithmic time scale to obtain the data matrix A, and contribution of water at ∼1650 cm−1 was subtracted using time-resolved difference spectra of water as described previously.30 A weighted global fit was performed with weighting factors calculated from the variance of the reference intensity at the different wavenumbers. As a reaction model, sequential first-order reactions with three intermediates were postulated. Fewer or greater numbers of intermediates did not account better for the measured data. The species-associated difference spectra D were calculated using the concentration profiles C from the global fit via matrix division according to D = AC+ (with C+ being the pseudoinverse of C).

Figure 2. Effect of ATP on the UV/vis spectrum of pCRY-PHR. (A) At pH 7.8, a small but significant red-shift of the band at ∼370 nm can be observed upon the addition of ATP. (B) The UV/vis difference spectra induced by addition of ATP to 3 mM are similar at pH 7.2, 7.8, and 8.2.

PHR, however, the direction of the shift and thereby the sign of the difference spectrum by ATP addition is inverted, and similar difference spectra are observed over the whole pH range from 7.2 to 8.2. These findings indicate an effect of ATP binding independent of the protonation state of Asp396. Moreover, the fine structure of the 370 nm band in the absorption spectrum of pCRY without ATP is characterized by two maxima of similar absorbance (Figure 2A), which has been observed for photolyase only after substrate binding.41 Effect of pH and Bound ATP on the FT-IR Difference Spectra of pCRY. The concentration of the protein sample was increased strongly by reducing the volume at different pH values by filtration. The water content was kept high enough to ensure full solvatization and thereby prevent any loss in response of the protein as observed in the flavin-binding receptor light-oxygen-voltage (LOV) domain.42 Illumination with blue light resulted in light-minus-dark difference spectra at pH 7.1, 7.8, and 8.3 (Figure 3A). The difference spectrum of pCRY-PHR at pH 7.8 is in full agreement with those presented and assigned previously.28,43,44 The prominent negative band at 1733 cm−1 has been assigned to the deprotonation of Asp396.28,29 The same signal is present in the difference spectra at pH 7.1 and 8.3. The relative size of the signal with respect to other bands originating from FAD does not change with pH, which indicates that at all pH values the

■

RESULTS Effect of Bound ATP on the UV/Vis Absorption Spectrum of pCRY. For AtCRY1-PHR, a shift of bands in the UV/vis spectrum was observed by a change in pH with a pKa = 7.4 and by binding of ATP.36 To analyze the effect of ATP on pCRY-PHR, the protein was expressed in Escherichia coli, purified, and the buffer exchanged to pH 7.2, 7.8, and 8.2 in a range limited by the stability of the protein. UV/Vis spectra of the samples were taken before and after the addition of ATP (Figure 2A). The difference spectra caused by the addition of ATP to 3 mM to pCRY-PHR at pH 7.2, 7.8, and 8.2 show distinct positive and negative signals that are in agreement with each other (Figure 2B). The most prominent difference signal at 390 nm represents a red shift of the FAD band at ∼370 nm (Figure 2A). This shift is a characteristic indication for an increase in the polarity of the surrounding of the FAD or an increase in H-bonding to FAD.40 For AtCRY1-PHR, a similar shift has been interpreted previously to be caused by the change in protonation state of Asp396 close to FAD (Figure 1).36 Similar effects, but more pronounced, have been observed in photolyase upon binding of the DNA substrate.41 In pCRYC


Article


Supporting Information, Figure S1). Accordingly, these bands are attributed to a loss of β-sheet structure in the protein that has been observed previously in the CRYβ intermediate, rising with τ = 500 μs in time-resolved experiments.30 Therefore, the addition of ATP seems to stabilize a transient state that would otherwise decay to a large extent within the millisecond time range. It has been demonstrated that the addition of ATP favors the formation and inhibits the decay of the radical.28,38 Accordingly, a significantly increased infrared difference signal was observed here that remained constant over a long period of time (Figure 3B). Quantum yields of radical formation of up to 14% were reported in the presence of ATP and external reductant.23,36 We determined the quantum yield by comparing the loss of oxidized flavin directly after 10 s of illumination in pCRY-PHR with that of the LOV1 domain from phototropin45,46 under identical conditions by UV/vis spectroscopy (see Supporting Information, Figure S2). As a result, the quantum yield of the flavin neutral radical before the final decay process (τ = 6100 s)28 is ∼2% in the presence of ATP and the absence of external reductant. Taking into account the contribution of a loss of population in the microsecond time region of ∼40%,30 the initial quantum yield of the neutral radical is estimated to 3− 4%. Time-Resolved Step-Scan FT-IR Experiments on the D396C Mutant Lacking Proton Transfer. The mutant D396C was expressed, purified, and prepared in the same manner as the wild type to obtain equally well-hydrated samples. Ferricyanide was added to accelerate the back-reaction and enable measurements with a repetition rate of 2 Hz.30 Time-resolved experiments were performed covering the time range from 3.5 μs to 4.6 ms. A global analysis of the timeresolved data set using a model of sequential reactions yielded three time constants with τ1′ = 55 μs, τ1 = 664 μs, and τ2 > 5 ms attributed to the decay of intermediates I1′, I1, and I2, respectively (see Supporting Information Figure S3). The species-associated difference spectrum (SADS) of I1′ is highly similar to that of I1 but overlapped by broad spectral features (see Supporting Information Figure S4). Similar observations on the wild type with precisely the same time constant were assigned to heat artifacts.30,47 The intermediate I1 exhibits a decay with an apparent τ1 = 664 μs into I2, but a decrease in yield in the reaction to I2 indicates a parallel decay to the dark state. In support of this finding, such loss of flavin anion radical was observed previously by UV/vis spectroscopy with τ = 400 μs.28 When this information is included in the analysis, the formation of the intermediate I2 takes place with τf,1 = 1.3 ms in competition to the decay to the dark state with τb,1 = 1.4 ms. I2 decays after the detection window of the experiment. The resulting SADS of I1 and I2 of the D396C mutant (Figure 4A) are compared to the same experiments performed on the wild type (Figure 4B).30 There are notable differences in the spectra of the mutant lacking the proton donor Asp396. The deprotonation signal at 1733 cm−1 is missing as well as contributions of the FAD neutral radical at 1530 and 1641 cm−1. Instead, marker bands of the FAD anion radical in pCRYPHR were detected at 1490, 1518, and 1625 cm−1 in agreement with previous steady-state recordings on the mutant.28 Other spectral features are strikingly similar in both samples. Negative contributions are observed in I2 of the mutant at 1693, 1658, 1636, and 1543 cm−1. To distinguish this loss in absorbance from changes in the chromophore, the kinetics of the negative bands at 1636 and 1632 cm−1 were extracted from the raw data

Figure 3. Effects of pH and ATP binding on the light-induced FT-IR difference spectra of pCRY-PHR. Negative bands show contributions of the dark form, while positive bands represent the photoproduct. (A) Comparison of difference spectra obtained at pH 7.1, 7.8, and 8.3. No significant changes in the relative amplitude of the ν(CO)OH vibration of Asp396 at 1733 cm−1 were observed. Spectra at pH 7.1 and 8.3 were scaled to the spectrum at pH 7.8 as indicated using the ν(CO) vibration of flavin at 1708 cm−1. The sample at pH 8.3 was less stable leading to signal fluctuations in the water region (1683−1600 cm−1), which was omitted for clarity. (B) Difference spectra of pCRYPHR in the presence and the absence of ATP at pH 7.8. A significant effect by bound ATP was observed in the β-sheet region at ∼1634 cm−1, but no influence on the protonation state of Asp396 was found. The spectrum was scaled as indicated using the band at 1545 cm−1.

deprotonation is taking place to the same extent. Additionally, the amplitude of all difference signals does not decrease with the change in pH from 7.1 to 8.3 but increases slightly instead. Thereby the possibility is eliminated that only proteins with a protonated Asp396 take part in the photocycle and are selectively detected here. It can be concluded that the protonation state of Asp396 is not affected by the pH of the bulk solvent at pH 7.1 to 8.3. However, by lowering the pH to 7.1, an additional signal at 1422 cm−1 and a compensation of intensity at 1577 cm−1 were observed. The same procedure was followed to obtain FT-IR difference spectra in the presence of ATP (Figure 3B). The bound ATP significantly changed the spectral pattern. Small changes in the spectral signature are located at 1422 and 1577 cm−1 similar to those observed at pH 7.1 and have not been assigned yet. The most prominent changes are the additional negative contributions at ∼1634 cm−1. A comparison to the absorption spectrum of ATP does not reveal a signal at this frequency (see D


Article


and compared to the recovery of the oxidized state of FAD with a marker band of the CO stretch at 1705 cm−1 (Figure 4C). The kinetics of the loss show a maximal decrease at ∼1 ms. In the wild type, very similar changes at 1693, 1632, and 1545 cm−1 have been assigned previously to the CRYβ intermediate characterized by a loss of β-sheet structure in the α/β subdomain ∼25 Å away from flavin.30 The time constant of formation of CRYβ is slowed in the mutant from 500 μs to 1.3 ms. The loss of β-sheet structure has been proposed to be part of the signal transduction mechanism to the CCT. The finding that a similar reorganization takes place in the mutant indicates that the formation of CRYβ is not dependent on the proton transfer to flavin.

■

DISCUSSION Cryptochromes perform diverse functions in all kingdoms of life, and therefore the way they translate outside stimuli such as light into biochemical signals is of high interest. Here we investigated the influence of the pH and ATP on the formation of the signaling state in the plant cryptochrome pCRY. Furthermore, we characterized the photocycle in the absence of the proton donor D396 to evaluate the importance of the proton transfer for the subsequent changes in secondary structure. Stabilizing Effect of Bound ATP on Conformational Changes of Plant CRY. ATP has been demonstrated to have diverse influences on plant cryptochromes but to our knowledge not on members of other subfamilies.48 It is assumed to bind to the protein in a cleft that is a remnant of the DNA binding pocket of photolyases16 (Figure 5A) with a dissociation constant of 20 μM.39 How this binding influences the photocycle on a molecular basis is unclear yet. Previous studies have shown an increase in yield and lifetime of the flavin (anion and neutral) radical. Here, we observed that ATP stabilizes in addition conformational changes characteristic for intermediates of the photocycle. The steady-state FT-IR difference spectra of wild-type pCRY show that only in the presence of ATP strong negative contributions by the loss of βsheet structure of the CRYβ intermediate are detectable (Figure 5B). In contrast, the corresponding changes in secondary structure in the D396C mutant were limited to turn elements at ∼1670 cm−1 characteristic for CRYγ (Figure 5B).28 This difference might be interpreted as the result of a more effective deceleration of the photocycle by ATP in the wild type. In conclusion, ATP binding seems to strongly stabilize the CRYβ intermediate in the wild type. Its lifetime is prolonged from the millisecond to the minute time scale, which is the physiological lifetime of signaling of plant cryptochromes in vivo.33 Such stability is in agreement with the postulated physiological role of the CRYβ intermediate in signaling by initiating the dissociation of the CCT.30 Differences and Similarities between Homologues of Plant Cryptochrome. We did not obtain evidence for the pronounced increase in apparent pKa of Asp396 by ATP binding in pCRY-PHR as it was described for AtCRY1-PHR.36 While we witnessed a shift in the UV/vis spectrum upon the addition of ATP, this shift shows an inverse direction compared to the one observed for AtCRY1-PHR. Furthermore, we could not detect an influence of the pH on the signal of D396 deprotonation in the FT-IR spectra between pH 7.1 and 8.3. As a possible explanation for this discrepancy there might be a difference between the homologues pCRY and AtCRY1 concerning the immediate environment of the aspartic acid

Figure 4. Time-resolved step-scan FT-IR spectroscopy on the D396C mutant of pCRY-PHR. (A) SADS were obtained from the global fit using a model of sequential reactions. A negative contribution is formed in the intermediate I2 at ∼1636 cm−1 with τ = 1.3 ms (highlighted in gray). Bands are broadened because of the low spectral resolution of 8 cm−1. (B) The comparison to the SADS of the wild type (taken from Thöing et al.30) reveals a high similarity of I2 to the intermediate CRYβ characterized by a profound loss of β-sheet structure and marker bands at 1693, 1632, and 1545 cm−1. The spectral resolution of the wild-type experiments was 4.5 cm−1. (C) Kinetics of marker bands for the loss of β-sheet structure at 1632/1636 cm−1 in comparison to a marker band for recovery of the oxidized state of FAD at 1705 cm−1. The loss of β-sheet structure shows a maximal decrease at ∼1 ms. E


Article


Figure 5. Influence of proton transfer and ATP binding on the photocycle of pCRY-PHR. (A) The secondary structure of pCRY-PHR is shown with bound FAD chromophore and ATP analogue. The β-sheet of the α/β subdomain is highlighted in red (structure modeled on that of AtCRY1-PHR, PDB code: 1U3D). (B) Photocycles of D396C mutant and wild type are given with time constants from UV/vis spectroscopy marked in blue and those from FT-IR spectroscopy marked in red. Bound ATP stabilizes transient intermediates into the time range of minutes (green). In both samples, the CRYβ intermediate is formed with similar time constants (1.3 ms and 500 μs, respectively). Therefore, the characteristic loss of β-sheet proceeds independently of proton transfer to flavin.

Asp396 might start signal propagation43 similar to flavinbinding LOV domains49 and blue light sensors using flavin (BLUF).50,51 Alternatively, electrostatic effects have been proposed to drive signaling either by the repulsion of bound ATP from the charged Asp39652 or by the buried charge on the flavin anion radical/Asp396 acting as the origin of conformational changes.28,30 In the D396C mutant, the hydrogenbonding network around flavin is most likely corrupted, and a charge on Asp396 is not available anymore. Therefore, our findings of pronounced light-induced conformational changes in the mutant point to an electrostatic epicenter model53 in plant cryptochromes originating from the charge on the flavin anion radical (and subsequently from charged Asp396 in the wild type). Similarly, electrostatic interactions have been suggested to initiate signaling in the insect cryptochrome dCRY,54 despite the low homology to plant cryptochromes.

residue, which results in a shift of pKa. However, all residues in 5 Å distance around the Asp396 carboxylic group in AtCRY1 are conserved with similar or higher polarity in the sequence of pCRY. Instead, the accessibility of the solvent to Asp396 might be higher in AtCRY1, which would also explain the effect of ATP binding on the apparent pKa by blocking such access. A less likely possibility is that the changes observed in AtCRY1 by UV/vis spectroscopy originate from the influence of a different amino acid than Asp396, which might be clarified by applying FT-IR spectroscopy. Note that both studies agree on the most relevant aspect, that is, that Asp396 is fully protonated in the dark in plant cryptochromes under physiological conditions. Irrelevance of Proton Transfer to Flavin for Signal Progression. It has been demonstrated previously that the proton transfer to flavin is inhibited in the D396C mutant, but a structural response of the protein prevails in the form of a change in turn elements upon illumination. Accordingly, the anion radical of flavin has been suggested to be sufficient for driving signal progression.28 The time-resolved experiments on the D396C mutant strengthen this hypothesis. The photocycles of pCRY in the presence and the absence of a proton donor (Figure 5B) include the same sequence of intermediates CRYα, CRYβ, and CRYγ, albeit CRYα is difficult to recognize in the mutant spectra because of overlapping signals by the flavin anion radical. Fundamental differences were observed only in the population of the different intermediates, because an additional decay channel is present in the mutant with an apparent time constant of 9 μs, which leads to a loss of anion radical of ∼40%.28 This decay pathway is responsible for the significantly lower signal in the time-resolved FT-IR experiments as compared to those of the wild type. Together with the subsequent loss in population in the sub-millisecond time range (see Supporting Information, Figure S3), almost all radical species decay in the mutant before they could drive the process of CCT dissociation (τ = 400 ms) required for signaling. Models for the Origin of Signaling in Plant Cryptochromes. The signaling state in plant cryptochromes contains the flavin neutral radical.31−33 However, neither the charge nor the geometry of the flavin changes from oxidized flavin to the neutral radical, raising the question of how the signaling process is initiated. It has been proposed that the change in the hydrogen-bonding network by deprotonation of

■

CONCLUSIONS The key findings in this study are that the β-sheet conformational change in the α/β subdomain of plant cryptochrome, the CRYβ intermediate, was observed in the absence of proton transfer to flavin and that this reorganization can be fully stabilized into the time range of minutes by addition of ATP. Time-resolved infrared spectroscopic experiments on plant cryptochrome in the presence of ATP would be required to finally link these two independent findings, which is challenging because of the resulting very long lifetime of the flavin radical. As a consequence, the hypothesis is strengthened that the conformational change necessary for signal transduction is induced by the formation of a flavin anion radical as buried charge. Although this anion radical is considered signalingcompetent in the D396C mutant, the lifetime of CRYβ is only prolonged sufficiently by ATP binding in the wild type, underlining the additional importance of the proton transfer by Asp396.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpca.7b10249. FT-IR spectra of ATP and adenosine monophosphate; determination of the quantum yield by UV/vis spectrosF


Article


■

(13) Gegear, R. J.; Casselman, A.; Waddell, S.; Reppert, S. M. Cryptochrome mediates light-dependent magnetosensitivity in Drosophila. Nature 2008, 454, 1014−1018. (14) Bazalova, O.; Kvicalova, M.; Valkova, T.; Slaby, P.; Bartos, P.; Netusil, R.; Tomanova, K.; Braeunig, P.; Lee, H. J.; Sauman, I.; et al. Cryptochrome 2 mediates directional magnetoreception in cockroaches. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 1660−5. (15) Lin, C.; Robertson, D. E.; Ahmad, M.; Raibekas, A. A.; Jorns, M. S.; Dutton, P. L.; Cashmore, A. R. Association of flavin adenine dinucleotide with the Arabidopsis blue light receptor CRY1. Science 1995, 269, 968−970. (16) Brautigam, C. A.; Smith, B. S.; Ma, Z.; Palnitkar, M.; Tomchick, D. R.; Machius, M.; Deisenhofer, J. Structure of the photolyase-like domain of cryptochrome 1 from Arabidopsis thaliana. Proc. Natl. Acad. Sci. U. S. A. 2004, 101, 12142−12147. (17) Yang, H. Q.; Wu, Y. J.; Tang, R. H.; Liu, D.; Liu, Y.; Cashmore, A. R. The C termini of Arabidopsis cryptochromes mediate a constitutive light response. Cell 2000, 103, 815−827. (18) Small, G. D.; Min, B.; Lefebvre, P. A. Characterization of a Chlamydomonas reinhardtii gene encoding a protein of the DNA photolyase/blue light photoreceptor family. Plant Mol. Biol. 1995, 28, 443−454. (19) Reisdorph, N. A.; Small, G. D. The CPH1 gene of Chlamydomonas reinhardtii encodes two forms of cryptochrome whose levels are controlled by light-induced proteolysis. Plant Physiol. 2004, 134, 1546−1554. (20) Müller, N.; Wenzel, S.; Zou, Y.; Kunzel, S.; Sasso, S.; Weiss, D.; Prager, K.; Grossman, A.; Kottke, T.; Mittag, M. A plant cryptochrome controls key features of the Chlamydomonas circadian clock and its life cycle. Plant Physiol. 2017, 174, 185−201. (21) Immeln, D.; Weigel, A.; Kottke, T.; Pérez Lustres, J. L. Primary events in the blue light sensor plant cryptochrome: intraprotein electron and proton transfer revealed by femtosecond spectroscopy. J. Am. Chem. Soc. 2012, 134, 12536−12546. (22) Brazard, J.; Usman, A.; Lacombat, F.; Ley, C.; Martin, M. M.; Plaza, P.; Mony, L.; Heijde, M.; Zabulon, G.; Bowler, C. Spectrotemporal characterization of the photoactivation mechanism of two new oxidized cryptochrome/photolyase photoreceptors. J. Am. Chem. Soc. 2010, 132, 4935−4945. (23) Giovani, B.; Byrdin, M.; Ahmad, M.; Brettel, K. Light-induced electron transfer in a cryptochrome blue-light photoreceptor. Nat. Struct. Mol. Biol. 2003, 10, 489−490. (24) Engelhard, C.; Wang, X.; Robles, D.; Moldt, J.; Essen, L. O.; Batschauer, A.; Bittl, R.; Ahmad, M. Cellular metabolites enhance the light sensitivity of Arabidopsis cryptochrome through alternate electron transfer pathways. Plant Cell 2014, 26, 4519−4531. (25) Langenbacher, T.; Immeln, D.; Dick, B.; Kottke, T. Microsecond light-induced proton transfer to flavin in the blue light sensor plant cryptochrome. J. Am. Chem. Soc. 2009, 131, 14274−14280. (26) Maeda, K.; Robinson, A. J.; Henbest, K. B.; Hogben, H. J.; Biskup, T.; Ahmad, M.; Schleicher, E.; Weber, S.; Timmel, C. R.; Hore, P. J. Magnetically sensitive light-induced reactions in cryptochrome are consistent with its proposed role as a magnetoreceptor. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 4774−4779. (27) Lüdemann, G.; Solov’yov, I. A.; Kubar, T.; Elstner, M. Solvent driving force ensures fast formation of a persistent and well-separated radical pair in plant cryptochrome. J. Am. Chem. Soc. 2015, 137, 1147− 1156. (28) Hense, A.; Herman, E.; Oldemeyer, S.; Kottke, T. Proton transfer to flavin stabilizes the signaling state of the blue light receptor plant cryptochrome. J. Biol. Chem. 2015, 290, 1743−1751. (29) Kottke, T.; Batschauer, A.; Ahmad, M.; Heberle, J. Blue-lightinduced changes in Arabidopsis cryptochrome 1 probed by FTIR difference spectroscopy. Biochemistry 2006, 45, 2472−2479. (30) Thö i ng, C.; Oldemeyer, S.; Kottke, T. Microsecond deprotonation of aspartic acid and response of the α/β subdomain precede C-terminal signaling in the blue light sensor plant cryptochrome. J. Am. Chem. Soc. 2015, 137, 5990−5999.

copy using a LOV domain as standard; concentration profiles of intermediates from the global fit; speciesassociated difference spectra of intermediates I1′ and I1 (PDF)

AUTHOR INFORMATION

Corresponding Author

*Phone: +49-521-106-2062. Fax: +49-521-106-2981. E-mail: [email protected]. ORCID

Tilman Kottke: 0000-0001-8080-9579 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by a fellowship of the Studienstiftung des Deutschen Volkes to L.S. and by a Heisenberg Fellowship of the German Research Foundation to T.K. (KO3580/4-1). We thank T. Hellweg for generous support.

■

REFERENCES

(1) Chaves, I.; Pokorny, R.; Byrdin, M.; Hoang, N.; Ritz, T.; Brettel, K.; Essen, L. O.; van der Horst, G. T.; Batschauer, A.; Ahmad, M. The cryptochromes: blue light photoreceptors in plants and animals. Annu. Rev. Plant Biol. 2011, 62, 335−364. (2) Sancar, A. Structure and function of photolyase and in vivo enzymology: 50th anniversary. J. Biol. Chem. 2008, 283, 32153−32157. (3) Lin, C.; Shalitin, D. Cryptochrome structure and signal transduction. Annu. Rev. Plant Biol. 2003, 54, 469−496. (4) Somers, D. E.; Devlin, P. F.; Kay, S. A. Phytochromes and cryptochromes in the entrainment of the Arabidopsis circadian clock. Science 1998, 282, 1488−1490. (5) Ahmad, M.; Cashmore, A. R. HY4 gene of A. thaliana encodes a protein with characteristics of a blue-light photoreceptor. Nature 1993, 366, 162−166. (6) Coesel, S.; Mangogna, M.; Ishikawa, T.; Heijde, M.; Rogato, A.; Finazzi, G.; Todo, T.; Bowler, C.; Falciatore, A. Diatom PtCPF1 is a new cryptochrome/photolyase family member with DNA repair and transcription regulation activity. EMBO Rep. 2009, 10, 655−661. (7) Bayram, O.; Biesemann, C.; Krappmann, S.; Galland, P.; Braus, G. H. More than a repair enzyme: Aspergillus nidulans photolyase-like CryA is a regulator of sexual development. Mol. Biol. Cell 2008, 19, 3254−3262. (8) Hendrischk, A. K.; Frühwirth, S. W.; Moldt, J.; Pokorny, R.; Metz, S.; Kaiser, G.; Jäger, A.; Batschauer, A.; Klug, G. A cryptochrome-like protein is involved in the regulation of photosynthesis genes in Rhodobacter sphaeroides. Mol. Microbiol. 2009, 74, 990− 1003. (9) Stanewsky, R.; Kaneko, M.; Emery, P.; Beretta, B.; Wager-Smith, K.; Kay, S. A.; Rosbash, M.; Hall, J. C. The cryb mutation identifies cryptochrome as a circadian photoreceptor in Drosophila. Cell 1998, 95, 681−692. (10) Beel, B.; Prager, K.; Spexard, M.; Sasso, S.; Weiss, D.; Müller, N.; Heinnickel, M.; Dewez, D.; Ikoma, D.; Grossman, A. R.; et al. A flavin binding cryptochrome photoreceptor responds to both blue and red light in Chlamydomonas reinhardtii. Plant Cell 2012, 24, 2992− 3008. (11) Zhu, H.; Sauman, I.; Yuan, Q.; Casselman, A.; Emery-Le, M.; Emery, P.; Reppert, S. M. Cryptochromes define a novel circadian clock mechanism in monarch butterflies that may underlie sun compass navigation. PLoS Biol. 2008, 6, e4. (12) Griffin, E. A., Jr.; Staknis, D.; Weitz, C. J. Light-independent role of CRY1 and CRY2 in the mammalian circadian clock. Science 1999, 286, 768−771. G


Article

The Journal of Physical Chemistry A (31) Banerjee, R.; Schleicher, E.; Meier, S.; Viana, R. M.; Pokorny, R.; Ahmad, M.; Bittl, R.; Batschauer, A. The signaling state of Arabidopsis cryptochrome 2 contains flavin semiquinone. J. Biol. Chem. 2007, 282, 14916−14922. (32) Bouly, J. P.; Schleicher, E.; Dionisio-Sese, M.; Vandenbussche, F.; Van Der Straeten, D.; Bakrim, N.; Meier, S.; Batschauer, A.; Galland, P.; Bittl, R.; et al. Cryptochrome blue light photoreceptors are activated through interconversion of flavin redox states. J. Biol. Chem. 2007, 282, 9383−9391. (33) Herbel, V.; Orth, C.; Wenzel, R.; Ahmad, M.; Bittl, R.; Batschauer, A. Lifetimes of Arabidopsis cryptochrome signaling states in vivo. Plant J. 2013, 74, 583−592. (34) Partch, C. L.; Clarkson, M. W.; Ö zgür, S.; Lee, A. L.; Sancar, A. Role of structural plasticity in signal transduction by the cryptochrome blue-light photoreceptor. Biochemistry 2005, 44, 3795−3805. (35) Kondoh, M.; Shiraishi, C.; Müller, P.; Ahmad, M.; Hitomi, K.; Getzoff, E. D.; Terazima, M. Light-induced conformational changes in full-length Arabidopsis thaliana cryptochrome. J. Mol. Biol. 2011, 413, 128−137. (36) Müller, P.; Bouly, J. P.; Hitomi, K.; Balland, V.; Getzoff, E. D.; Ritz, T.; Brettel, K. ATP binding turns plant cryptochrome into an efficient natural photoswitch. Sci. Rep. 2015, 4, 5175. (37) Burney, S.; Hoang, N.; Caruso, M.; Dudkin, E. A.; Ahmad, M.; Bouly, J. P. Conformational change induced by ATP binding correlates with enhanced biological function of Arabidopsis cryptochrome. FEBS Lett. 2009, 583, 1427−1433. (38) Immeln, D.; Schlesinger, R.; Heberle, J.; Kottke, T. Blue light induces radical formation and autophosphorylation in the lightsensitive domain of Chlamydomonas cryptochrome. J. Biol. Chem. 2007, 282, 21720−21728. (39) Bouly, J. P.; Giovani, B.; Djamei, A.; Mueller, M.; Zeugner, A.; Dudkin, E. A.; Batschauer, A.; Ahmad, M. Novel ATP-binding and autophosphorylation activity associated with Arabidopsis and human cryptochrome-1. Eur. J. Biochem. 2003, 270, 2921−2928. (40) Kotaki, A.; Naoi, M.; Yagi, K. Effect of proton donors on the absorption spectrum of flavin compounds in apolar media. J. Biochem. 1970, 68, 287−292. (41) Jorns, M. S.; Wang, B. Y.; Jordan, S. P.; Chanderkar, L. P. Chromophore function and interaction in Escherichia coli DNA photolyase: reconstitution of the apoenzyme with pterin and/or flavin derivatives. Biochemistry 1990, 29, 552−561. (42) Iwata, T.; Yamamoto, A.; Tokutomi, S.; Kandori, H. Hydration and temperature similarly affect light-induced protein structural changes in the chromophoric domain of phototropin. Biochemistry 2007, 46, 7016−7021. (43) Immeln, D.; Pokorny, R.; Herman, E.; Moldt, J.; Batschauer, A.; Kottke, T. Photoreaction of plant and DASH cryptochromes probed by infrared spectroscopy: The neutral radical state of flavoproteins. J. Phys. Chem. B 2010, 114, 17155−17161. (44) Sommer, C.; Dietz, M. S.; Patschkowski, T.; Mathes, T.; Kottke, T. Light-induced conformational changes in the plant cryptochrome photolyase homology region resolved by selective isotope labeling and infrared spectroscopy. Photochem. Photobiol. 2017, 93, 881−887. (45) Losi, A.; Kottke, T.; Hegemann, P. Recording of blue lightinduced energy and volume changes within the wild-type and mutated phot-LOV1 domain from Chlamydomonas reinhardtii. Biophys. J. 2004, 86, 1051−1060. (46) Holzer, W.; Penzkofer, A.; Hegemann, P. Photopysical and photochemical excitation and relaxation dynamics of LOV domains of phot from Chlamydomonas reinhardtii. J. Lumin. 2005, 112, 444−448. (47) Lórenz-Fonfría, V. A.; Furutani, Y.; Kandori, H. Active internal waters in the bacteriorhodopsin photocycle. A comparative study of the L and M intermediates at room and cryogenic temperatures by infrared spectroscopy. Biochemistry 2008, 47, 4071−4081. (48) Spexard, M.; Thöing, C.; Beel, B.; Mittag, M.; Kottke, T. Response of the sensory animal-like cryptochrome aCRY to blue and red light as revealed by infrared difference spectroscopy. Biochemistry 2014, 53, 1041−1050.

(49) Yee, E. F.; Diensthuber, R. P.; Vaidya, A. T.; Borbat, P. P.; Engelhard, C.; Freed, J. H.; Bittl, R.; Möglich, A.; Crane, B. R. Signal transduction in light-oxygen-voltage receptors lacking the adductforming cysteine residue. Nat. Commun. 2015, 6, 10079. (50) Stelling, A. L.; Ronayne, K. L.; Nappa, J.; Tonge, P. J.; Meech, S. R. Ultrafast structural dynamics in BLUF domains: transient infrared spectroscopy of AppA and its mutants. J. Am. Chem. Soc. 2007, 129, 15556−15564. (51) Domratcheva, T.; Hartmann, E.; Schlichting, I.; Kottke, T. Evidence for tautomerisation of glutamine in BLUF blue light receptors by vibrational spectroscopy and computational chemistry. Sci. Rep. 2016, 6, 22669. (52) Müller, P.; Bouly, J. P. Searching for the mechanism of signalling by plant photoreceptor cryptochrome. FEBS Lett. 2015, 589, 189− 192. (53) Xie, A.; Kelemen, L.; Hendriks, J.; White, B. J.; Hellingwerf, K. J.; Hoff, W. D. Formation of a new buried charge drives a largeamplitude protein quake in photoreceptor activation. Biochemistry 2001, 40, 1510−1517. (54) Ganguly, A.; Manahan, C. C.; Top, D.; Yee, E. F.; Lin, C.; Young, M. W.; Thiel, W.; Crane, B. R. Changes in active site histidine hydrogen bonding trigger cryptochrome activation. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 10073−10078.

H


Time-Resolved Infrared Spectroscopy on Plant Cryptochrome

Recommend Documents