| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Physics and Nanotechnology, NanoBiotechnology Section, UltrafastBioSpectroscopy Group, Aalborg University, Aalborg, Denmark; Institute of Molecular Pathology and Immunology of the University of Porto, 4200 Porto, Portugal; Biological Engineering Research Group, Instituto Superior Técnico, 1049-001 Lisbon, Portugal; and Center of Structural and Molecular Biomedicine, Universidade do Algarve, Campus de Gambelas, 8005-139 Faro, Portugal
Correspondence: Address reprint requests to Maria Teresa Neves-Petersen, E-mail: tnp{at}nanobio.aau.dk.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUDING REMARKSACKNOWLEDGEMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUDING REMARKSACKNOWLEDGEMENTSREFERENCES |
|---|
). During catalysis in plant peroxidases a porphyrin cation 
radical is formed (2), and in yeast cytochrome c peroxidase a tryptophan radical plays the equivalent role (3,4). In both cases in the first reaction step the iron ion becomes hexacoordinated to an oxygen atom (FeIV=O). In classical plant peroxidases tryptophan is uniquely represented and is highly conserved, located in a surface loop connecting two helices. The indole ring is directed toward the core of the protein lying above the plane defined by the heme group at an average distance of 16–18 Å (5–8) (Fig. 1 A). Trp fluorescence in classical plant peroxidases is highly quenched in the native state due to energy transfer from the excited state of tryptophan to the heme group (5,9,10). Quantum yields are 100 orders of magnitude lower than that of tryptophan free in solution. Ferriprotoporphyrin also displays a visible absorption spectrum.
|
). It is known that tryptophan and tyrosine radicals can be induced by ultraviolet (UV) illumination (12–14). Heme groups are also excellent UV absorbers. UV illumination might in some cases induce enzyme inactivation (15–18). The mechanisms of inactivation are believed to be due to ionization of aromatic residues associated with an electron transfer mechanism and radical formation together with disruption of disulphide bridges (19–23). Earlier studies confirm that the major reaction after UV excitation of enzymes is photoionization (24). The excited tryptophan after photoionization gives TrpH+, and electron is ejected from the molecule. In heme proteins, the TrpH+ might reduce the ferrous heme (25). Porphyrins are a major source of free radicals and singlet oxygen and can induce photobleaching in fluorescent molecules (26). Several studies are reported on photoinactivation of catalase, a tetrameric heme-containing enzyme. Catalase degrades H2O2 to oxygen and water and represents an important part of the antioxidative system in cells. Blue light (380–500 nm) illumination of sunflower catalase causes partial heme destruction, and oxidation of histidine, present in the catalytic center of catalase, perhaps is an early event in photoinactivation (27). UV irradiation (365 nm) of bovine liver catalase leads to the formation of catalytically inactive compounds III (oxyferrous catalase) and II (catalase FeIV) (28). UV C irradiation of flavocytochrome b2, which houses a heme prosthetic group, results in enzyme inactivation, where Trp and Tyr residues as well as the heme group are modified (25). Oxygen presence offers protection of tryptophan against UV-induced damage due to quenching (11), whereas the heme group can act as a photosensitizer thus generating harmful singlet oxygen (29). This work presents a detailed spectroscopic analysis of UV-induced changes in secondary structure, enzymatic activity, fluorescence emission, fluorescence lifetime distribution, fluorescence mean lifetime, and heme absorption of horseradish peroxidase A2 (HRPA2). The catalytic activity, conformational changes at the secondary structure level, protein fluorescence, and heme absorption were monitored after Trp 296-nm UV illumination. A streak camera study on the effects of UV illumination time on the distribution of the two shortest fluorescence lifetimes of the single endogenous aromatic residue Trp in HRPA2, at pH 4, is presented. We also highlight the outstanding value of streak camera technology in following ultrafast processes and show how streak camera data can be correlated with protein structural changes. Also, we hereby present the effects of continuous illumination of the heme group at 403 nm on Trp fluorescence emission and on heme absorption bands.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUDING REMARKSACKNOWLEDGEMENTSREFERENCES |
|---|
) and data reported by Hiner et al. (31). The degree of purity was checked by sodium dodecylsulfate-polyacrylamide gel electrophoresis, and a single band was observed when stained with Coomassie. HRPA2 concentrations were determined considering the molar extinction coefficient 
403nm = 100 mM–1cm–1 (32). Protein solutions used for spectroscopic studies were dissolved in buffer made with Milli-Q water. All salts were of analytical grade. Acetate buffer was used for pH 4. Buffer concentration was always 25 mM.
Extinction coefficient as a function of wavelength of HRPA2 and hematin
The absorbance from 250 nm to 700 nm of a horseradish peroxidase solution prepared as mentioned above was measured on a Thermo Electron (Waltham, MA) UV1 spectrophotometer and compared with a study performed by Du et al. (33) on the absorption of hematin in acetic acid.
Tryptophan irradiation-steady-state fluorescence
The enzyme solution (3 mL of a 5 µM protein solution) was continuously irradiated for different time periods at 296 nm using a 75 W Xenon arc lamp from an RTC 2000 PTI (Photon Technology International, Birmingham, NJ) spectrometer provided with a monochromator. Excitation and emission slit widths were set to 6 nm. Tryptophan fluorescence was monitored at 350 nm (excitation spectra) and excited at 296 nm (emission spectra). Temperature in the cell, a quartz cuvette (1-cm path length) was controlled using a Peltier element. The sample was continuously stirred at 650 rpm to maintain the homogeneity of the solution, and the temperature was kept constant at 20°C. Line voltage was controlled and maintained at 4 V, thus avoiding fluctuations deriving from the power coming from the electrical outlet.
The fluorescence emission curve as a function of illumination time (displayed in Fig. 3 A) was fitted to a biexponential function like the one described by Eq. 1,
 | (1) |
|
Activity measurements
HRPA2 activity was measured at room temperature using 50 mM guaiacol in 25 mM acetate buffer (pH 4) and 4.4 mM H2O2. The reaction was followed for 1 min by reading the increase in absorbance at 470 nm. The extinction coefficient of the oxidation product, 470nm = 26.6 mM–1cm–1, was used to calculate initial velocities.
Far-UV CD measurements
CD measurements were carried out using a Jasco (Tokyo, Japan) spectropolarimeter, model J-715. The ellipticity values were obtained in mdegrees directly from the instrument and converted to the mean residue ellipticity 
MRW as previously stated (8). The far-UV CD spectra were measured using a rectangular quartz cell of 1-mm path length. Each spectrum was an average of six scans between 300 and 200 nm. The resultant ellipticities of the HRPA2 solutions were calculated by subtracting the ellipticity of the buffer solution. The wavelength of 223 nm was used to monitor thermal denaturation in the far-UV CD range. Temperature scans were carried out in the temperature range 293–358 K using a Peltier element (irradiated samples) or a thermostated cuvette by means of a circulating water bath, and a temperature probe was immersed in the protein (dark control samples). The experimental parameters were as follows: 1-nm bandwidth, 0.2-K step resolution, 2-s response time, and scanning rates 1.5 (irradiated samples) and 2.6 K min–1 (control).
HRPA2 heme irradiation studies
Irradiation and steady-state fluorescence
A total of 3 mL of a 4 µM protein solution was continuously irradiated for 30 h at 403 nm using a 75-W Xenon arc lamp from a RTC 2000 PTI (Photon Technology International) spectrometer provided with a monochromator (slits width 6 nm). Temperature in the cell, a quartz cuvette (1-cm path length) was controlled using a Peltier element and was kept constant at 298 K. The sample was continuously stirred at 650 rpm to maintain the homogeneity of the solution. Line voltage was controlled and maintained at 4 V, thus avoiding fluctuations deriving from the power coming from the electrical outlet. Before 403 nm irradiation and at specified times during the irradiation, tryptophan fluorescence emission intensity at 350 nm upon 296-nm excitation was acquired.
Irradiation and absorption measurements
In another experiment, absorption by the protein solution was monitored before and after irradiation with 403-nm light for 26 h. The sample was irradiated using the same conditions as above. Measurements were performed on a Thermo Electron UV1 spectrophotometer.
Time-resolved lifetime measurements—a streak camera study
Time-resolved measurements were carried out upon exciting the samples with ultrashort UV laser pulses at 280 nm and 290 nm (for selective excitation of Trp residue). The 280-nm excitation light was generated by sending the output from a Spectra Physics (Mountain View, CA) Tsunami laser (<100-fs pulse duration, 12-nm fullwidth at half-maximum, 80-MHz repetition rate, 
= 840 nm, Tsunami 3960, Spectra Physics pumped by a high power (5 W at 532 nm) Millennia V solid state laser, Spectra Physics) through a pulse picker, which decreased the pulse repetition rate to 8 MHz. The fundamental pulse (840 nm) was mixed with its second harmonic (420 nm) in a frequency doubler/tripler unit (GWU-Lasertechnik, Erfstadt, Germany) to generate pulses at 280 nm. The power at 280 nm was 0.290 mW. The 290-nm excitation light was generated by tuning the Spectra Physics Tsunami laser to 870 nm. After the pulse picker, the fundamental pulse (870 nm) was mixed with its second harmonic (435 nm) in a frequency doubler/tripler unit (GWU-Lasertechnik) to generate pulses at 290 nm. The power at 290 nm was 0.216 mW. The sample was placed in a 1400-µL quartz cuvette and excited along the 1-cm light path. The cuvette was placed in front of the input slit (100 µm) of a spectrograph (Oriel, Darmsdadt, Germany, MS257, with a grating blazed at 400 nm with 600 lines/mm), after which it was focused into the slit of the input optics (100-µm slit) of the streak camera (Optronis, Kehl, Germany).
The sample was continuously irradiated with the 280-nm and 290-nm excitation light. Sample concentration was 49 µM. The different illumination times were 0, 34, 52, 67, 84, 106, 122, 137, and 152 min for the 280-nm series and from 0 to 170 min every 10 min for the 290-nm series. Fluorescence emission was followed in a time window of 
1.65 ns after excitation. The fluorescence lifetimes of horseradish peroxidase (HRP) were assumed to remain the same independent of illumination time, whereas the preexponential factors were assumed to change, thus enabling a global analysis approach (see note below regarding global analyses). The governing equations for the time-resolved intensity decay data were assumed to be a sum of discrete exponentials as in
 | (2) |
i is the amplitude (preexponential factor), 
i is the fluorescence lifetime of the ith discrete component, and 
i = 1.0. The fractional intensity, fi, of each decay time is given by
 | (3) |
 | (4) |
Note on global analyses
To model excited state processes or to unravel heterogeneity in the distribution of fluorophores, experiments under a variety of conditions can be performed. One can change experimental parameters such as excitation and emission wavelengths, pH, quencher concentration, timescale, temperature, orientation of excitation, and emission polarizers. Finally, a multi-dimensional fluorescence decay surface is obtained. From the separate analyses of the individual decay traces, a model can be deduced. The appropriateness of the model is checked by verifying the consistency of the parameter values obtained from each decay curve analysis. However, the parameter estimates resulting from the various single decay curve analyses may suffer from a large uncertainty so that the model building becomes difficult. It has to be realized that many parameters appear in a nonlinear way in the model function and that in most cases the functions within the model are nonorthogonal. It has been suggested to perform a simultaneous analysis of related decay traces, i.e., of the fluorescence decay surface, by linking the common parameters. The merits of this global analysis approach have been emphasized and used broadly within the scientific community (34). Global analysis of fluorescence lifetime data can be used to obtain an accurate fit of multi-exponential fluorescence decays. Global analysis algorithms simultaneously fit multiple measurements acquired under different experimental conditions to achieve higher accuracy (35).
In our studies, the experimental parameter that changes was the preillumination time of the sample, before acquiring the fluorescence decays.
Analysis of denaturation curves
The temperature values at the midpoint of the denaturation curve (Tm where formally 50% of the protein molecules are native and 50% are unfolded) were determined assuming that during transitions two distinct and populated states were present, the native (N) and the unfolded state (U) (8). The observed value of the spectroscopic signal, y, at any point will be y = yN fN + yU fU, where yN and yU are the values of y characteristic of the native and unfolded states, respectively, under the conditions where y is being measured. fN + fU = 1, with fN and fU representing the fraction of protein in the native and unfolded conformations. The equilibrium constant, K, and the free energy change, 
G, can be calculated using K = (yN – y)/(yN – yU), and G = –RTlnK, where R is the gas constant and T the absolute temperature. Tm values were calculated from a plot of G versus temperature to G = 0.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUDING REMARKSACKNOWLEDGEMENTSREFERENCES |
|---|
|
Before irradiation, an excitation spectrum (emission at 350 nm) of fresh peroxidase was recorded. Afterward, excitation spectra were recorded after different times of irradiation (22 min, 44 min, 76 min, 116 min, 6 h and 16 h, emission at 350 nm), as displayed in Fig. 3 B. It can be observed that an increase in the fluorescence intensity in Fig. 3 A is correlated with an increase in the excitation fluorescence intensity displayed in Fig. 3 B. Moreover it can be observed that the excitation spectra become increasingly blue shifted the longer the irradiation time (Fig. 3 B). Interestingly, when observing the ratio between the normalized excitation spectra (Fig. 3 C), being the ratios defined as the intensity of each excitation spectrum divided by the excitation spectrum at time zero, it can be seen that illumination of the peroxidase sample at 296 nm induces spectral changes between 250 and 270 nm, where both Tyr and Trp absorb and fluoresce at 350 nm (Phe does not fluoresce at 350 nm) and above 290 nm, where only Trp absorbs. No spectral changes were observed from 278 to 290 mn.
Activity and Reinheitzahl value changes in HRPA2 on UV-light irradiation
In peroxidases the catalytic activity is dependent on the presence and correct conformation of the heme group and the residues forming the catalytic pocket. Conformational changes on the protein likely affect enzyme activity. Fig. 4 A shows the change of enzyme activity upon different UV illumination times compared to a nonilluminated HRPA2 sample. The change of the Reinheitzahl value (Rz), defined by the ratio between the Soret band absorbance and the absorbance at 280 nm, upon different UV illumination times compared to a nonilluminated HRPA2 sample is also displayed. Activity decays exponentially with a rate constant, kact, 0.00377 min–1 (0.23 h–1) for the irradiated sample and 0.00042 min–1 (0.03 h–1) for the dark control. Fitting to a double exponential model did not improve the quality of the fit (no improvement on 
2 and larger errors associated with the fitted parameters). After 16 h of UV illumination the enzyme displayed 2% of its initial activity, whereas the nonirradiated enzyme displayed 65% of its initial activity. The activity of the enzyme when the intensity of the 350-nm emission is at its maximum (6 h of irradiation at 296 nm) is 30% of the initial, whereas the nonirradiated sample maintained 92% of the initial activity value after the same period of time.
|
3.70 to 1.18 and 1.16, respectively, 31% of the initial value. Over the same period, the Rz values of the nonilluminated sample decreased from 3.39 to 2.96 and 2.94 after 6 h and 16 h at room temperature, 87% of the initial value. The Rz values as a function of illumination time could be described by a single exponential decay with rate constant kRz of 0.0038 min–1 (0.23 h–1). Apparently the process is faster for the nonirradiated control sample with a rate constant kRz, 0.0065 min–1 (0.39 h–1). Fitting to a double exponential model did not improve the quality of the fit. The data point for the Rz value observed after 6 h of illumination was more than 10 standard deviations from the model obtained by fitting to the rest of the data points and was thus not considered.
Absorbance spectral changes in HRPA2 on UV-light irradiation
Spectral properties of the HRPA2 heme group as a function of irradiation time are presented in Fig. 4 B. Increasing irradiation time at 296 nm lead to a decrease of the Soret absorption band as well as the disappearance of the charge transfer transitions (CT1, 498 nm; CT2, 643 nm) characteristic of a pentacoordinate state of the iron.
UV irradiation-induced changes in secondary structure of HRPA2
Complementary analysis of protein conformational alterations was investigated by circular dichroism (CD) spectroscopy. Thermal stability of 5 µM HRPA2 samples irradiated with light at 296 nm for 6 and 16 h were investigated. The respective control measurements were performed. Temperature dependence of the ellipticity at 223 nm of native HRPA2 and samples irradiated for different periods of time are shown in Fig. 5. Data reveal that the longer the irradiation period the lower the melting temperature, Tm, of HRPA2 at pH 4, as displayed in Table 1. After 6-h UV illumination, the Tm of the protein dropped 2.25 K and its ellipticity at 295 K dropped from –16 degcm2dmol–1 to –11 degcm2dmol–1. After a 16-h UV illumination, the protein still showed the presence of secondary structure elements (ellipticity had decreased from –16 degcm2dmol–1 to a residual value of –8 x 10–3 degcm2dmol–1 at 295 K) although no transition was detected in the temperature scans. Fig. 5 clearly shows that irradiation leads to protein denaturation. Concerning the nonirradiated sample, a 16-h period waiting time at room temperature after the sample had been freshly prepared lead to a decrease in the Tm value of 1.4 K (Table 1). No significant helical loss monitored by CD was observed (data not shown).
|
|
|
|
|
Fluorescence and absorption spectra changes induced by illumination at 403 nm
Excitation of HRPA2 at 403 nm, where the heme group displayed maximum absorbance, also had an effect on the protein's Trp fluorescence. During the experiment where HRPA2 was continuously illuminated at 403 nm, the protein's Trp fluorescence emission intensity at 350 nm upon 296-nm excitation was measured at different times over a period of more than 30 h. The emission intensity was observed to increase more the longer the illumination time at 403 nm, reaching approximately twice the initial Trp fluorescence intensity after 30 h of exposure to the 403-nm light, as displayed in Fig. 7 A.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUDING REMARKSACKNOWLEDGEMENTSREFERENCES |
|---|
Fluorescence quantum yield
The results show that excitation of the protein at all mentioned wavelengths lead to an increase in the fluorescence quantum yield of the protein, observation correlated with loss of absorption of the heme group (Figs. 4 B and 7 B). Fluorescence intensity of the single tryptophan residue of HRPA2 in its native state is highly quenched because of the presence of an adjacent prosthetic group, Fe-PPIX. The efficiency of energy transfer (E) between the Trp donor and the heme acceptor is almost maximum, e.g., in horseradish peroxidase (HRPC) E is 0.93 (10) and in soybean peroxidase E is 0.97 (7). Tryptophan quantum yield is known to be 0.001 in classic plant peroxidases due to energy transfer to the heme group. This value increases 20 and 10 times for apoproteins of horseradish peroxidase C and soybean peroxidase, respectively (7,9). However, as displayed in Figs. 3 A and 7 A, excitation of horseradish peroxidase at 296 nm and 403 nm increased its fluorescence quantum yield. The increase might be due to light-induced separation of the Trp residue from the quencher heme group and/or photoinduced chemical changes in the prosthetic group of HRPA2. The first hypothesis is supported by the work done by Gryczynski and Bucci (36) on another heme-containing protein, horse heart myoglobin, and by the work of Grotjohann et al. (27) on photoinactivation of a plant catalase, also a heme protein. Grotjohann et al. showed that blue light excitation of catalase would induce heme dissociation and heme photochemistry. Grotjohann et al. also report that the velocity of heme destruction is lower than the rate of enzyme photoinactivation, suggesting that other photoinduced reactions lead to enzyme inactivation. Gryczynski and Bucci observed that the steady-state fluorescence emission intensity was reversibly increased with decreasing pH, almost doubling at pH 4.5, and was correlated with an increase in heme dissociability at lower pH, consistent with the titration of the proximal and distal histidines inside the heme pocket. Their analyses revealed the presence of three species originating from heme-protein interactions with three distinct fluorescence lifetimes: the native form of crystalline protein (30–114 ps), the conformation with disordered hemes (240–1400 ps), and the reversibly dissociated heme-free myoglobin (3100–5100 ps).
This work has been done at pH 4. Acidic pH values will favor the protonation of the distal and proximal His residues in HRPA2, which pKa values have been reported to be 2.5 for the distal His (1) and below 3 for the proximal His (37), for horseradish peroxidase C. Wolff et al. recently estimated, using NMR, the pKa of the His-32 that binds heme in the protein HasASM as being below 4.8 (38). These pKa values are considerably lower than the pKa of His when in the model tripeptide Gly-His-Gly (pKmodel = 6.4) (39). Once protonated, the positively charged His will electrostatically repulse the nearby positively charged heme group and catalytic Arg, which is positively charged at acidic pHs (Fig. 1 B). Therefore, acidic pH values will induce conformational changes in the active side pocket of peroxidases that might induce heme release (40). Low pH is actually used to induce release of heme from the protein to create the apoprotein (41). Ejection of the heme group has been reported to lead to an irreversible thermal unfolding of HRPA2 and HRPA1 at pH 4, whereas at pH 7 and 10 refolding was observed (8) since the protein recaptures the heme group upon cooling. This is correlated with the electrostatic repulsion between the heme group and the heme moiety at acidic pH values. The presence of positively charged groups renders the pKa value of both the distal and proximal His considerably lower than its pKmodel since once neutral, there will be no electrostatic repulsions between the His and the positively charged groups. Also, since the distal His will the hydrogen-bond acceptor upon binding H2O2, it will be a better acceptor if neutral. These are two likely reasons that contribute to the low pKa value of proximal and distal His observed in peroxidases.
An increase in fluorescence emission intensity upon UV excitation of the single Trp residue in other proteins, like cutinase, has been observed by Weisenborn et al. (42), Prompers et al. (20), and Neves-Petersen et al. (21). The reason for the observed increase in fluorescence emission intensity was the disruption of the disulphide bridge, an excellent protein fluorescence quencher, located near the Trp residue upon Trp UV excitation (20,21,23). The study reported in this work points at a similar possible mechanism explaining the increase in fluorescence intensity of HRPA2 upon UV illumination: putative light-induced dissociation and/or photoinduced damage of the strong fluorescence quencher in horseradish peroxidase, the prosthetic group ferriprotoporphyrin IX, and the heme moiety. In the Protein Data Bank structure of the highly homologous Arabidopsis thaliana peroxidase A2 (Fig. 1 A), this group is located within 8 Å (closest distance) from the Trp residue. The closer a quencher is from the fluorescence donor and the higher its extinction coefficient, the stronger the quenching will be. The disulphide bridges in HRPA2 are located farther away from the Trp residue, which means that in terms of distance the prosthetic group is a better fluorescence quencher.
Furthermore, compared to disulphide bridges, the prosthetic ferriprotoporphyrin IX group has a much larger extinction coefficient in the wavelength range where Trp emits (43). Therefore it is the prosthetic group in peroxidases that is mainly responsible for the proteins' fluorescence emission intensity due to its quencher properties. The changes reported here in the absorption spectrum of ferriprotoporphyrin IX group upon UV illumination of horseradish peroxidase (both 296 nm and 403 nm; Figs. 4 B and 7 A) suggest that the prosthetic group putatively dissociates from the protein and/or suffers photodamage upon UV illumination. Irradiation of Trp in HRPA2 can induce alterations in the protein moiety through electron transfer from triplet states of Trp (3Trp) to suitable acceptors such as cystines (13). The UV light-induced opening of disulphide bridges upon excitation of the protein's aromatic residues was observed for many other proteins, such as hydrolytic enzymes (lipases/esterases, lysozyme), proteases (human plasminogen), alkaline phosphatase, immunoglobulins' Fab fragment, major histocompability complex class I protein (23), and -lactalbumin (22). Detection of free thiols groups was observed after photoinduced disruption of the disulphide bridges. HRPA2 contains eight cysteine residues involved in four disulphide bridges conserved in all members of the classical plant peroxidases class. The nearest cystine in the A. thaliana peroxidase A2 (ATPA2) crystal structure (Fig. 1 A) is located 12 Å from the single Trp residue. Photoinduced disruption of disulphide bridges located as far as 10 Å from the aromatic residue has been reported for -lactalbumin (22).
Molecular photobleaching
Fig. 3 A also shows that prolonged UV illumination (after 7-h illumination at 296 nm) leads to molecular photobleaching, leading to a decrease in the fluorescence intensity. A possible UV-induced reaction in proteins leads to, e.g., electron transfer, ionization, and radical formation (12–14,25,44–56, S. Klitgaard, M. T. Neves-Petersen, V. Sundström, T. Polivka, A. Yartsev, T. Pascher, and S. B. Petersen, unpublished). These photoinduced reactions lead to photochemical changes of the fluorophores, with concomitant conversion of the indole chromophore into a nonfluorescing group (42). Photobleaching originates from the triplet excited state, which is created from the singlet state (S1) via intersystem crossing.
Protein stability, heme absorption, and Reinheitzahl values
CD data clearly show that prolonged UV irradiation leads to protein denaturation. After 16-h UV illumination no transition is detected in the temperature scans, as displayed in Fig. 5. This observation is correlated with data presented in Fig. 3 A, which indicate that after 16 h of UV illumination the protein has already suffered photophysical- and photochemical-induced reactions leading to photobleaching. These reactions will lead to structural damage of proteins that will ultimately impair their activity. As can be seen in Table 1 and Fig. 4 A, after 6 h and 16 h of UV illumination the protein's residual activity was 30% and 2%, respectively.
It was also observed in our study that illumination of HRPA2 with 296 nm and 403 nm leads to loss in the Soret absorbance band, as displayed in Figs. 4 B and 7 B, respectively. The intensity of the visible bands at 500 nm and 643 nm also decrease with longer irradiation periods of time. Heme release has been reported to be the first event occurring during unfolding of HRPA2 (8). These observations do correlate with the hypothesis that light induces heme dissociation and/or chemical modification of the heme group.
UV illumination has also affected the protein's Rz and its thermal stability. Residual ellipticity was observed after denaturation of horseradish peroxidase (Fig. 5). The observed decrease in the HRPA2 Soret band absorbance upon UV illumination lead to a decrease in the Rz value of the protein and was accompanied by activity loss (Fig. 4). Both Rz and activity values decay exponentially with identical first order decay constants (kact and kRz), which might indicate that light-induced changes in the heme moiety and the aromatic pool moiety do lead to changes in the structure and therefore in the activity of the protein. Irradiation damage was observed to be immediate. The loss of secondary structural elements upon UV irradiation (Fig. 5) reveals that photophysical and photochemical processes initiated in the Trp residue and probably on the adjacent heme group are damaging the overall structure of HRPA2. The lower melting temperature, Tm, of HRPA2 upon irradiation (Table 1) as well as the decrease in the far-UV ellipticity after 6-h irradiation (Fig. 5) is likely to be caused by light-induced structural changes in the heme environment. The proximal His is covalently bound to the heme iron and responsible for the heme architecture and consequently the enzyme activity (2). Conformation changes in the heme moiety will affect the active site architecture.
Photochemistry, reactive oxygen species
Oxidation of histidine, present in the catalytic center of catalases is perhaps an early event in catalase photoinactivation as it is an oxygen-dependent process (27). Likewise, oxidation of proximal His in HRPA2 might be responsible for the enzyme activity loss and heme spectra alterations. Indirect photooxidation of proteins is known to occur via the formation and subsequent reactions of singlet oxygen. Heme moieties have, in general, the capability to generate reactive oxygen (29). Other amino acid residues in the protein are susceptible to reacting with 1O2, namely excited Trp and Tyr, Met, or Cys, leading to chemical changes of these residues. It is known that N-formylkynurenine, a tryptophan photoproduct, can act as an endogenous photosensitizer that can generate singlet oxygen and superoxide. Tyr radicals are also reported to be formed after excitation of Trp residues (49). Trp can transfer energy to Tyr residues via Förster-type singlet energy transfer (57). The process of HRPA2 photoinactivation is likely to be induced also by formation of 1O2 upon heme excitation and since the heme group is a sensitizer that once in the excited state can undergo intersystem crossing, enter a triplet state, and react with molecular oxygen generating singlet oxygen, one of the most reactive oxygen species.
Both illumination of aromatic residues and of the heme group leads to changes in heme absorption and fluorescence quantum yield of horseradish peroxidase, indicative of heme leaving the molecule and/or photoinduced chemical changes in the heme moiety. Direct illumination of the heme group can induce photochemical reactions such as electron ejection from the porphyrin ring, associated to cation and radical formation, and singlet oxygen formation that can change the conformation of the heme pocket. The excited state aromatic residues in HRPA2 will deliver a great part of their excitation energy to the prosthetic ferriprotoporphyrin IX group, since this is an effective fluorescence quencher. Furthermore, since UV illumination of aromatic residues leads to electron ejection from their side chains (see references above), it is likely that the prosthetic group also acts as an electron acceptor, further destabilizing the heme pocket.
Spectral changes of aromatic residues
As displayed in Fig. 3, B and C, prolonged UV illumination of HRPA2 lead to spectral shifts in the excitation spectra of the protein. This is indicative of chemical changes suffered by the Trp or other aromatic residues in the protein. Trp has been selectively excited at 296 nm. We can see from Fig. 3 C that this lead to a change in the excitation spectra above 292 nm, where Trp absorbs. If Tyr residues were also excited due to Förster-type singlet energy transfer from Trp residues, phenolate anions (PheO–) could have been created. The pKa of the OH group in UV-excited Tyr residues is expected to be around 4 (32), and thus much lower than the pKa around 10 for ground state tyrosine (39). PheO– is known to have larger extinction coefficients above 290 and fluorescence emission is centered around 345 nm (11,39). Fig. 3 C shows that changes are also observed from 250 nm to 275 nm. No changes are observed from 275 nm to 290 nm. Since these spectra have been obtained with emission set to 350 nm, the excitation spectra report changes only on Trp and Tyr residues, since Phe do not emit at 350 nm.
Fluorescence lifetime studies
Peroxidases are usually characterized by three fluorescence lifetimes: in HRPA1, pH 7, the three lifetimes are 55 ps, 130 ps, and 1.2 ns (58) and in HRPC the observed lifetimes were 45–76 ps, 1.1–1.5 ns, and 3.9–4.9 ns, depending on pH (10). In this study, with the detection system used (streak camera), we have only observed picosecond lifetime components since the streak camera has a temporal window of 1.8 ns. The longest nanosecond fluorescent lifetime component was not observed. In our study we are interested in following changes in the short lifetime components. The shortest picosecond lifetime components are known to be associated to the heme group here. Changes in the heme moiety will most likely induce changes in the fluorescence lifetime distribution.
As can be seen from Fig. 6, B and C, and Table 2, the two fastest relaxation pathways of horseradish peroxidase are characterized by lifetimes of 100 ± 1 ps and 746 ± 34 ps (when using 280-nm excitation), respectively, and 83 ± 0.8 ps and 819 ± 37 ps (when using 290-nm excitation), respectively. The observed change in the preexponential factors associated with each of the above mentioned lifetimes, as a function of UV illumination time, leads to an increase in the protein's mean lifetime. This observation supports the hypothesis that UV light induced dissociation of the strong protein fluorescence quencher group, the ferriprotoporphyrin IX group, and/or chemical changes in the heme moiety.
The effect of preillumination of HRP at 403 nm, where heme absorbs, on the dynamics of HRP fluorescence emission, upon 280 nm and 290 nm excitation, has also been studied (data not shown). Preillumination of the heme group leads to an increase of the fluorescence mean lifetime of horseradish peroxidase.
|
CONCLUDING REMARKS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUDING REMARKSACKNOWLEDGEMENTSREFERENCES |
|---|
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUDING REMARKSACKNOWLEDGEMENTSREFERENCES |
|---|
M.T.N.-P. acknowledges the support from Novi Invest and Licfond and from the Danish Research Agency, Novo Nordisk A/S, Novozymes A/S.
Submitted on August 21, 2006; accepted for publication November 27, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUDING REMARKSACKNOWLEDGEMENTSREFERENCES |
|---|
2. Veitch, N. C., and A. T. Smith. 2001. Horseradish peroxidase. Adv. Inorg. Chem. 51:107–162.
3. Mauro, J. M., L. A. Fishel, J. T. Hazzard, T. E. Meyer, G. Tollin, M. A. Cusanovich, and J. Kraut. 1988. Tryptophan-191—phenylalanine, a proximal-side mutation in yeast cytochrome c peroxidase that strongly affects the kinetics of ferrocytochrome c oxidation. Biochemistry. 27:6243–6256.[CrossRef][Medline]
4. Sivaraja, M., D. B. Goodin, M. Smith, and B. M. Hoffman. 1989. Identification by ENDOR of Trp191 as the free-radical site in cytochrome c peroxidase compound ES. Science. 245:738–740.
5. Brunet, J. E., G. A. Gonzalez, and C. P. Sotomayor. 1983. Intramolecular tryptophan heme energy transfer in horseradish peroxidase. Photochem. Photobiol. 38:253–254.[Medline]
6. Pappa, H. S., and A. E. G. Cass. 1993. A step towards understanding the folding mechanism of horseradish peroxidase. Tryptophan fluorescence and circular dichroism equilibrium studies. Eur. J. Biochem. 212:227–235.[Medline]
7. Kamal, J. K., and D. V. Behere. 2001. Steady-state and picosecond time-resolved fluorescence studies on native and apo seed coat soybean peroxidase. Biochem. Biophys. Res. Commun. 289:427–433.[CrossRef][Medline]
8. Carvalho, A. S., E. P. Melo, B. S. Ferreira, M. T. Neves-Petersen, S. B. Petersen, and M. R. Aires-Barros. 2003. Heme and pH-dependent stability of an anionic horseradish peroxidase. Arch. Biochem. Biophys. 415:257–267.[CrossRef][Medline]
9. Das, T., and S. Mazumdar. 1995. Conformational substates of apoprotein of horseradish peroxidase in aqueous solution: a fluorescence dynamics study. J. Phys. Chem. 99:13283–13290.[CrossRef]
10. Das, T., and S. Mazumdar. 1995. pH-induced conformational perturbation in horseradish peroxidase. Picosecond tryptophan fluorescence studies on native and cyanide-modified enzymes. Eur. J. Biochem. 227:823–828.[Medline]
11. Lakowicz, J. R. 1999. Principles of Fluorescence Spectroscopy, 2nd ed. Kluwer Academic/Plenum Publishers, New York.
12. Bent, D. V., and E. Hayon. 1975. Excited state chemistry of aromatic amino acids and related peptides. I. Tyrosine. J. Am. Chem. Soc. 97:2599–2606.[CrossRef][Medline]
13. Bent, D. V., and E. Hayon. 1975. Excited state chemistry of aromatic amino acids and related peptides. III. Tryptophan. J. Am. Chem. Soc. 97:2612–2619.[CrossRef][Medline]
14. Creed, D. 1983. The photophysics and photochemistry of the near-UV absorbing amino acids-II. Tyrosine and its simple derivatives. Photochem. Photobiol. 39:563–575.
15. Augenstein, L., and P. Riley. 1964. The inactivation of enzymes by ultraviolet light, IV: the nature and involvement of cystine disruption. Photochem. Photobiol. 3:353–367.
16. Igelman, J. M., T. C. Rotte, E. Schecter, and D. J. Blaney. 1965. Exposure of enzymes to laser radiation. Ann. N. Y. Acad. Sci. 122:790–801.[CrossRef][Medline]
17. Berns, M. W., S. I. Matsui, R. S. Olson, and D. E. Rounds. 1970. Enzyme inactivation with ultraviolet laser energy (2650 Angstroms). Science. 169:1215–1217.
18. Saha, A. 1997. Photo-induced inactivation of dihydroorotate dihydrogenase in dilute aqueous solution. Int. J. Radiat. Biol. 72:55–61.[CrossRef][Medline]
19. Vladimirov, Y. A., D. I. Roshchupkin, and E. E. Fesenko. 1970. Photochemical reactions in amino acid residues and inactivation of enzymes during U.V.-irradiation. A review. Photochem. Photobiol. 11:227–246.[Medline]
20. Prompers, J. J., C. W. Hilbers, and H. A. M. Pepermans. 1999. Tryptophan mediated photoreduction of disulfide bond causes unusual fluorescence behaviour of Fusarium solini pisi cutinase. FEBS Lett. 456:409–416.[CrossRef][Medline]
21. Neves-Petersen, M. T., Z. Gryczynski, J. Lakowicz, P. Fojan, S. Pedersen, E. Petersen, and S. B. Petersen. 2002. High probability of disrupting a disulphide bridge mediated by an endogenous excited tryptophan residue. Protein Sci. 11:588–600.
22. Vanhooren, A., B. Devreese, K. Vanhee, J. van Beumen, and I. Hanssens. 2002. Photoexcitation of tryptophan groups induces reduction of two disulfide bonds in goat -lactalbumin. Biochemistry. 41:11035–11043.[CrossRef][Medline]
23. Neves-Petersen, M. T., T. Snabe, S. Klitgaard, M. Duroux, and S. B. petersen. 2006. Photonic biosensors: UV light induced molecular switch allows sterically oriented immobilisation of biomolecules and the creation of protein nanoarrays. Protein Sci. 11:343–351.
24. Aeschbach, R., R. Amado, and H. Neukom. 1976. Formation of dityrosine cross-links in proteins by oxidation of tyrosine residues. Biochim. Biophys. Acta. 439:292–301.[Medline]
25. Bhattacharya, D., S. Basu, and P. C. Mandal. 2000. UV radiation effects on flavocytochrome b2 in dilute aqueous solution. J. Photochem. Photobiol. B. 59:54–63.[CrossRef][Medline]
26. Greenbaum, L., C. Rothmann, R. Lavie, and Z. Malik. 2000. Green fluorescent protein photobleaching: a model for protein damage by endogenous and exogenous singlet oxygen. Biol. Chem. 381:1251–1258.[CrossRef][Medline]
27. Grotjohann, N., A. Janning, and R. Eising. 1997. In vitro photoinactivation of catalase isoforms from cotyledons of sunflower (Helianthus annuus L.). Arch. Biochem. Biophys. 346:208–218.[CrossRef][Medline]
28. Aubailly, M., J. Haigle, A. Giordani, P. Morliere, and R. Santus. 2000. UV photolysis of catalase revisited: a spectral study of photolytic intermediates. J. Photochem. Photobiol. B. 56:61–67.[CrossRef][Medline]
29. Aft, R. L., and G. C. Mueller. 1984. Hemin-mediated oxidative degradation of proteins. J. Biol. Chem. 259:301–305.
30. Shannon, L. M., E. Kay, and J. Y. Lew. 1966. Peroxidase isozymes from horseradish roots. I. Isolation and physical properties. J. Biol. Chem. 241:2166–2172.
31. Hiner, A. N., J. Hernández-Ruíz, M. B. Arnao, F. García-Cánovas, and M. Acosta. 1996. A comparative study of the purity, enzyme activity, and inactivation by hydrogen peroxide of commercially available horseradish peroxidase isoenzymes A and C. Biotechnol. Bioeng. 50:655–662.[CrossRef]
32. Strickland, E. H., E. Kay, L. M. Shannon, and J. Horwitz. 1968. Peroxidase isoenzymes from horseradish roots. 3. Circular dichroism of isoenzymes and apoisoenzymes. J. Biol. Chem. 243:3560–3565.
33. Du, H., R. A. Fuh, J. Li, A. Corkan, and J. S. Lindsey. 1998. PhotochemCAD: A computer-aided design and research tool in photochemistry. Photochem. Photobiol. 68:141–142.[CrossRef]
34. Janssens, L. D., N. Boens, M. Ameloot, and F. C. De Schryver. 1990. A systematic study of the global analysis of multlexponential fluorescence decay surfaces using reference convolution. J. Phys. Chem. 94:3564–3576.[CrossRef]
35. Verveer, P. J., A. Squire, and P. I. Bastiaens. 2000. Global analysis of fluorescence lifetime imaging microscopy data. Biophys J. 78:2127–2137.
36. Gryczynski, Z., and E. Bucci. 1998. Time resolved emissions in the picosecond range of single tryptophan recombinant myoglobins reveal the presence of long range heme protein interactions. Biophys. Chem. 74:187–196.[CrossRef][Medline]
37. Ascenzi, P., M. Brunori, M. Coletta, and A. Desideri. 1989. pH effects on the haem iron co-ordination state in the nitric oxide and deoxy derivatives of ferrous horseradish peroxidase and cytochrome c peroxidase. Biochem. J. 258:473–478.[Medline]
38. Wolff, N., C. Deniau, S. Letoffe, C. Simenel, V. Kumar, I. Stojiljkovic, C. Wandersman, M. Delepierre, and A. Lecroisey. 2002. Histidine pKa shifts and changes of tautomeric states induced by the binding of gallium-protoporphyrin IX in the hemophore HasASM. Protein Sci. 11:757–765.
39. Creighton, T. E. 1993. Proteins: Structure and Molecular Properties, 2nd ed. W. H. Freeman & Company, New York.
40. Smith, A., and R. T. Hunt. 1990. Hemopexin joins transferrin as representative members of a distinct class of receptor-mediated endocytic transport systems. Eur. J. Cell Biol. 53:234–245.[Medline]
41. Schulz, H., H. Hennecke, and L. Thöny-Meyer. 1998. Prototype of a heme chaperone essential for cytochrome c maturation. Science. 281:1197–1200.
42. Weisenborn, P. C. M., H. Meder, M. R. Egmond, T. J. W. G. Visser, and A. van Hoek. 1996. Photophysics of the single tryptophan residue in Fusarium solani Cutinase: evidence for the occurrence of conformational substates with unusual fluorescence behaviour. Biophys. Chem. 58:281–288.[CrossRef][Medline]
43. Klitgaard, S., M. T. Neves-Petersen, and S. B. Petersen. 2006. Quenchers induce wavelength dependence on protein fluorescence lifetimes. J. Fluoresc. 16:595–609.[CrossRef][Medline]
44. Grossweiner, L. I., G. W. Swenson, and E. F. Zwicker. 1963. Photochemical generation of the hydrated electron. Science. 141:805–806.
45. Bent, D. V., and E. Hayon. 1975. Excited state chemistry of aromatic amino acids and related peptides. II. Phenylalanine. J. Am. Chem. Soc. 97:2606–2612.[CrossRef][Medline]
46. Hoffman, M. Z., and E. Hayon. 1972. One-electron reduction of the disulfide linkage in aqueous solution. Formation, protonation, and decay kinetics of the RSSR– radical. J. Am. Chem. Soc. 94:7950–7957.[CrossRef]
47. Robbins, R. J., G. R. Fleming, G. S. Beddard, G. W. Robinson, P. J. Thistlethwaite, and G. J. Woolfe. 1980. Photophysics of aqueous tryptophan: pH and temperature effects. J. Am. Chem. Soc. 102:6271–6279.[CrossRef]
48. Mialocq, J. C., E. Amouyal, A. Barnas, and D. Grand. 1982. Picosecond laser photolysis of aqueous indole and tryptophan. J. Phys. Chem. 86:3173–3177.[CrossRef]
49. Grossweiner, L. I. 1984. Photochemistry of proteins: a review. Curr. Eye Res. 3:137–143.[Medline]
50. Kandori, H., R. F. Borkman, and K. Yoshihara. 1993. Picosecond transient absorption of aqueous tryptophan. J. Phys. Chem. 97:9664–9667.[CrossRef]
51. Stevenson, K. L., G. A. Papadantonakis, and P. R. LeBreton. 2000. Nanosecond UV laser photoionization of aqueous tryptophan: temperature dependence of quantum yield, mechanism, and kinetics of hydrated electron decay. J. Photochem. Photobiol. Chem. 133:159–167.[CrossRef]
52. Tsentalovich, Y. P., J. J. Lopez, P. J. Hore, and R. Z. Sagdeev. 2002. Mechanisms of reactions of flavin mononucleotide triplet with aromatic amino acids. Spectrochim. Acta A. Mol. Biomol. Spectrosc. 58:2043–2050.[CrossRef][Medline]
53. Borman, C. D., C. Wright, M. B. Twitchett, G. A. Salmon, and A. G. Sykes. 2002. Pulse radiolysis studies
) 22 min; (
) 76 min; (
) 184 min; (x) 6 h; and (
) 16 h and the spectrum for nonirradiated HRPA2.
