Photochemistry and Photobiology, 1998, 68(6): 864-868
Electron Transfer FIuorescence Quenching of Blepharisma japonicum Photoreceptor Pigments Nicola Angelinil,*,Annamaria Quarantal, Giovanni Checc~ccil.~, Pill-Soon S ~ n gand ~ . Francesco ~ Lenci*l 'Istituto di Biofisica, CNR, Pisa, Italy; 'Dottorato Fotobiologia, Universita di Sassari, Sassari, Italy; 3DipartimentoEcologia, Etologia Evoluzione, Universita di Pisa, Pisa, Italy; 4Departmentof Chemistry, University of Nebraska, Lincoln, NE and 5K~mho Life and Environmental Science Laboratory, Kwangju, Republic of Korea Received 31 July 1998; accepted 18 September 1998
ABSTRACT
INTRODUCTION
The hypericin analogs blepharismin (BP), oxyblepharismin (OxyBP) and stentorin (ST), the photosensing chromophores responsible for photomotile reactions in the ciliates Blepharisma japonicum (red and blue cells) and Stentor coeruleus, represent a new class of photoreceptor pigments whose chemical structures have recently been determined. In the case of ST it has been shown that the first excited singlet state can be deactivated by donation of an electron to an appropriate acceptor molecule (e.g. a quinone molecule). This charge transfer can be considered a possible mechanism for the primary photoprocess for the photomotile responses in S. coeruleus. To determine whether an electron transfer process also occurs in the deactivation of excited blepharismin, we studied the fluorescence quenching of OxyBP in dimethyl-sulfoxide (DMSO) and in ethanol using electron acceptors with different reduction potentials. Under our experimental conditions ground state and excited state complexes (like fluorescent exciplexes) are not formed between the fluorophore and the quenchers. In DMSO the bimolecular quenching constant values (k,) calculated on the basis of the best fitting procedures clearly show that the quenching efficiency decreases with the quencher negative reduction potential, E". The k , (M-l s l ) and Eo (V) values are, respectively, 7.8 X lo9 and -0.134 for 1,4-benzoquinone, 8.9 X lo9 and -0.309 for 1,4-naphthoquinone, 2.4 X 10' and -0.8 for nitrobenzene, 0.009 X 10' and - 1.022 for azobenzene and 0 and -1.448 for benzophenone. These findings point to the conclusion that upon formation of the encounter complex between OxyBP and the quencher, an electron is released from excited OxyBP to the quencher, similar to what happens in ST. It is suggested that in the pigment granules such a light-induced charge transfer from excited blepharismin to a suitable electron acceptor triggers sensory transduction processes in B. japonicum.
Stentorin (ST)? and blepharismin (BP), the photosensing chromophores responsible for photomotile reactions in the ciliates Stentor coeruleus and Blepharisma japonicum (1-3 and references therein), have been identified as a naphthodianthrone (4) and a benzodianthrone (5,6) derivative, respectively. Very recently the structure of oxyblepharismin (OxyBP), the photoreceptor pigment mediating step-up photophobic responses in B. japonicum blue cells (7-9), has also been determined (10). The proposed molecular structure of oxyblepharismin, resulting from a photoinduced rearrangement and irreversible dehydrogenation of blepharismin, adequately accounts for the resemblance between its absorption spectrum and those of hypericin (HYP) and ST (10) (Fig. 1). In the landscape of photoreceptor pigments for photomotile responses of microorganisms (e.g. rhodopsins, flavins, pterins, p-coumaric acid [2,31), the HYP analogs ST, BP and OxyBP constitute a new class of photopigments (Fig. 1) to which other as yet undiscovered light-detecting chromophores may belong. A HYP-type chromophore was also suggested to be present, possibly together with a rhodopsinlike pigment, in the photoreceptor system of the salt water ciliate Fabrea salina (1 1). As far as the functional properties of these photoreceptor pigments are concerned, several experimental lines of evidence have indicated that a light-driven proton translocation process is involved in the early steps of the photosensory transduction chain in both B. japonicum and S. coeruleus (7,12-14 and references therein). These findings suggest that protons could be released from the first excited singlet state of the chromophore (14,15), but this hypothesis had to be modified as the results of time resolved fluorescence measurements did not support it (16,17). A light-induced electron transfer from the excited pigment was subsequently suggested as the very first step of the photoreaction (18). In the case of ST it has been shown that the first excited singlet state can be deactivated by electron
*To whom correspondence should be addressed at: Francesco Lenci, Istituto di Biofisica, CNR, Via San Lorenzo 26, 56127 Pisa, Italy. Fax: +39-050-553501; e-mail:
[email protected] $5.00+0.00 C 1998 American Society for Photobiology 003 1 -8655/98
tAhhreviations: BP, blepharismin; DMSO, dimethyl-sulfoxide; EtOH, ethanol; HYP, hypericin; OxyBP, oxyblepharismin; ST. stentorin.
864
Photochemistry and Photobiology, 1998, 68(6) 865 transfer to an acceptor molecule, e.g. to a disulfide acceptor (19,20). This light-driven charge transfer, which plays a major role in the fast deactivation of the excited pigment, has been suggested to be a possible mechanism for the primary photoprocess for photomotile responses in S. coeruleus (20). Furthermore, in the physiological molecular environment the photoinduced electron transfer could generate a transient intracellular pH gradient, which in turn could cause opening of Ca channels. CaA+influx, finally, would trigger the halt and subsequent reversal of ciliary beating, i.e. the photomotile response (3 and references therein). This paper reports the results of a spectroscopic study aimed at ascertaining the occurrence of an electron transfer process for the fast deactivation of the first excited singlet state of OxyBP and discussing whether such a photoinduced electron transfer can be an early step of the photosensory perception and transduction chain in B. japonicurn. +
OH
0
OH
OH
0
,
0
OH
, OH
..
OH
0
OH
1) hypericin
OH
2) blepharismin
OH
0
OH
OH
0
OH
OH
0
OH
OH
0
OH
MATERIALS AND METHODS 3) stentorin
Because of its relative instability, BP samples contained trace contaminants dctectable by fluorescence measurements. Preliminary results, moreover, indicated that in the presence of an efficient electron acceptor like 1,4-benzoquinone, BP underwent partial oxidation even if it was stored in the dark. To avoid misleading results, OxyBP was used throughout all of our experiments. Chemic.czZ.r. HPLC grade methanol (Sigma-Aldrich, Steinheim, Germany), trifluoroacetic acid (Fluka, Buchs, Switzerland) and ethylacetaate (Sigma-Aldrich, Steinheim, Germany) and spectroscopic grade dimethyl-sulfoxide (DMSO, Fluka, Buchs, Switzerland) and ethanol (EtOH, Baker, Deventer, The Netherlands) were used without further purification. All fluorescence quenchers (1,4-benzoquinone, 1,4-naphthoquinone, nitrobenzene, azobenzene and benzophenone) were purchased from Sigma-Aldrich. Preparation ($ .samples. Cells were grown and pigment crude extracts were prepared as previously described (2 1,22). The BP fraction was purified by HPLC [Nucleosil C,,] reversed phase column (Alltech, Deerfield, IL USA) as reported by Checcucci et al. ( 5 ) by using an LKB (Amerham Pharmacia Biotech, Uppsala, Sweden) Bromma 2152 HPLC equipped with a Shirnadzu (Tokyo, Japan) RF-I0 AXI. fluorometric detector (excitation A = 330 nm, observation A = 605 nm) and a Hewlett-Packard (Palo Alto, CA USA) 3396 Series 11 integrator. Purified BP was dried under vacuum and photoconverted to OxyBP in aerobic conditions using cool white light (8 W m 2 , for about 72 h. Every 12 h the sample was dissolved in ethanol and again vacuum dried to obtain homogeneous photoconversion of the whole sample. OxyBP was finally repurified by HPLC. The effects of increasing amounts of fluorescence quenchers were studied by adding microliters of 0.2 M quencher stock solutions to the solutions of OxyBP in DMSO or in EtOH. All measurements were performed at 25°C 2 1°C using a Pharmacia Biotech MultiTemp 111 thermostat. Absorption and jluorescence meusurements. Absorption spectra were recorded with a JASCO 7850 spectrophotometer. Fluorescence emission and excitation spectra were recorded with a Perkin Elmer LS SOB spectrofluorometer. As the long-wavelength absorption tails of some. of the quenchers extend to over 600 nm (Fig. 2A), OxyBP fluorescence emission spectra were corrected for fluorescence reabsorption effects and the fluorescence intensities were normalized to the total sample absorption at the excitation wavelength, taking into account the relative contributions of both OxyBP and the quencher. As the cuvette compartment of the Perkin Elmer LS SOB is equipped with two mirrors that double the optical path of the excitation beam as well as that of the emitted fluorescence, it is necessary to take into account such instrumental peculiarities and use the following formula for fluorescence intensity F :
400
4) oxyblepharismin
4.50
SO0
550
600
6.50
700
Wavelength (nm)
Figure 1. Absorption spectra and structural formulas of ( I j HYP; (2) BP (from Checcucci et ul. [ S ] ) ,(3) ST (from Tao et ul. [4]) and (4) OxyBP (from Spitzner et UZ. [lo]). The ST absorption spectrum is similar to spectra I and 4.
I,
F = (1
(0
10-Z A f h e x c J ). A o n ~ R p ( h e n c ) A(A,,,)
where AOnyRP(Acxc)and A(A,,,) are the absorbances of OxyBP (calculated) and of the sample (measured) at the excitation wavelength and I , is the integral area below the fluorescence spectrum If(ACnJ, point-by-point corrected according to
where A(A,,,,,) is the total sample absorbances at the emission wavelength A,,, and Imra,(AemJis the measured fluorescence spectrum. Fluorescence lifetimes were measured with a single photon counting apparatus previously described (23) using an Argon ion laser (Coherent CR- 18) operating in the mode-locked regime (A,,, = 476 nm) with an overall time resolution of 100 ps. The fluorescence decay waveforms were analyzed by a nonlinear least-squares fitting procedure and the quality of the fitting was evaluated on the basis of the weighted residuals and their autocorrelation function (24). Multiple measurements on the same sample resulted in fluctuations smaller than 5% in the lifetime values and their rclativc amplitudes.
866 Nicola Angelini et a/.
A
0.025
1
0.ooo-I 500
, /A\ '
I
550
'
I
600
'
8
650
700
0.75
750
I 0
Wavelength (MI)
10
20
30
40
50
[Quencher] (mM)
B
1
A
* B Z Q A N P Q mNTB -AZB x B Z P
BZQ
2.50
NTB
x
B
BZP
i
/
2.00 -
B
1.50 -
1.00 .I 500
550
600
650
700
750
Wavelength (MI)
Figure 2. A: Absorption spectra in DMSO of (1) OxyBP, (2) OxyBP in the presence of 30 mM 1,4-benzoquinone and (3) 30 mM I ,4-benzoquinone. Spectrum 4 is the difference between spectrum 2 and spectrum 3 and was calculated taking into account the different volumes of the samples. B: Fluorescence emission spectra in DMSO (A,, = 555 nm) of (I) OxyBP and (2) OxyBP in the presence of 30 mM 1,4-benzoquinone.
RESULTS The HPLC elution profile of OxyBP consisted of a single band. Exactly the same elution profile was obtained by injecting OxyBP 24 h after the purification, indicating that no OxyBP degradation occurred. Fluorescence lifetime (7) measurements of OxyBP in DMSO and in EtOH revealed the presence in both solvents of two emitting species: a fast-decaying component largely predominant in DMSO as well as in EtOH (DMSO: T~ = 1.3 ns, relative amplitude = 98-99%; EtOH: 7" = 1.6 ns, relative amplitude = 97%) and a slow-decaying one (about 4-5 ns in both solvents) with a small relative amplitude (at most 3% in EtOH). Under these circumstances, steady-state fluorescence emission with and without quenchers can be ascribed to the short-living species and any contribution from the trace amounts of the long-living species can be disregarded. In both DMSO (Fig. 2A) and EtOH (data not shown) the absorption spectra of OxyBP were not affected by the presence of any of the quenchers used. This implies that no significant ground state interaction occurs between the OxyBP chromophore and the quencher and that the diffusion reac-
0
10
20
30
40
50
[Quencher] (mM)
Figure 3. Stern-Volrner plots of OxyBP fluorescence quenching by 1,4-benzoquinone (BZQ), 1,4-naphthoquinone (NPQ), nitrobenzene (NTB), azobenzene (AZB) and benzophenone (BZP) in (A) DMSO (A,,, = 555 nm) and (B) EtOH (A,, = 550 nm). Experimental data points are fitted with Eq. 3 and, in the case of BZQ, with Eq. 4. xz values are
