PhotosynthesisResearch 42: 157-166, 1994. (~) 1994KluwerAcademicPublishers.Printedin the Netherlands. Regular paper

Triplet energy transfer between bacteriochlorophyll and carotenoids in B850 light-harvesting complexes of Rhodobacter sphaeroides R-26.1 R o y a F a r h o o s h 1, V e e r a d e j C h y n w a t 1, R o n a l d G e b h a r d 2, J o h a n L u g t e n b u r g 2 & H a r r y A. F r a n k 1,*

1Department of Chemistry, 215 Glenbrook Road, University of Connecticut, Storrs, CT06269-3060, USA 2Department of Chemistry, Gorleous Laboratories, Leiden University, 2300 RA Leiden, The Netherlands; *Author for correspondence Received 22 April 1994;accepted in revised form 13 August 1994

Key words: bacteriochlorophyll, carotenoid, flash absorbance spectroscopy, light-harvesting complex, Rhodobacter sphaeroides, triplet state

Abstract

The build-up and decay of bacteriochlorophyll (BChl) and carotenoid triplet states were studied by flash absorption spectroscopy in (a) the B800-850 antenna complex of Rhodobacter (Rb.) sphaeroides wild type strain 2.4.1, (b) the Rb. sphaeroides R-26.1 B850 light-harvesting complex incorporated with spheroidene, (c) the B850 complex incorporated with 3,4-dihydrospheroidene, (d) the B850 complex incorporated with 3,4,5,6-tetrahydrospheroidene and (e) the Rb. sphaeroides R-26.1 B850 complex lacking carotenoids. Steady state absorption and circular dichroism spectroscopy were used to evaluate the structural integrity of the complexes. The transient data were fit according to either single or double exponential rate expressions. The triplet lifetimes of the carotenoids were observed to be 7.0 q- 0.1 /zs for the B800-850 complex, 14 + 2 #s for the B850 complex incorporated with spheroidene, and 19 + 2 #s for the B850 complex incorporated with 3,4-dihydrospheroidene. The BChl triplet lifetime in the B850 complex was 80 -4- 5 #s. No quenching of BChl triplet states was seen in the B850 complex incorporated with 3,4,5,6-tetrahydrospheroidene. For the B850 complex incorporated with spheroidene and with 3,4-dihydrospheroidene, the percentage of BChl quenched by carotenoids was found to be related to the percentage of carotenoid incorporation. The triplet energy transfer efficiencies are compared to the values for singlet energy transfer measured previously (Frank et al. (1993) Photochem. Photobiol. 57: 49-55) on the same samples. These studies provide a systematic approach to exploring the effects of state energies and lifetimes on energy transfer between BChls and carotenoids in vivo.

Abbreviations: BChl - bacteriochlorophyll; Car - carotenoid; CD - circular dichroism Introduction

In photosynthetic organisms, carotenoids act as lightharvesting pigments by transferring excited singlet state energy to bacteriochlorophyll (BChl) in antenna pigment-protein complexes (Duyens 1952; Goedheer 1959, 1969; Siefermann-Harms 1985; Frank and Cogdell 1992a). Carotenoids also protect the photosynthetic apparatus by quenching BChl triplet states which prevents the BChl-sensitized formation of sin-

glet state oxygen, a major oxidizing agent (Foote 1968; Renger and Wolff 1977; Boucher et al. 1977; Frank and Cogdell 1992a). Studies on triplet energy transfer in vivo have utilized flash-transient optical spectroscopic methods (Mathis 1969; Witt 1971; Monger et al. 1976; Schenck et al. 1984) and electron paramagnetic resonance techniques (Frank et al. 1980; Frank et al. 1987; Frank 1992b) to probe the rates and efficiencies of energy transfer between BChl and carotenoids in native,

158 OCl-I~ 3.4,5,6-tetr ahydrospheroidene OCH 3

3,4-dihydrospheroidene OCH 3

sl:Y3eroidene Fig. 1. The structures of the carotenoids used in the present work.

carotenoid-containing complexes and in proteins isolated from carotenoidless mutant where carotenoids have been incorporated. Here we present an investigation of the energy transfer between BChl and carotenoids focusing on three specific carotenoids, spheroidene, 3,4-dihydrospheroidene and 3,4,5,6tetrahydrospheroidene. These molecules, which have extents of 7r-electron conjugation ranging from 8 to 10 carbon-carbon double bonds, have been systematically incorporated into the B850 light-harvesting complex of the carotenoidless mutant Rhodobacter sphaeroides R26.1. Apart from the differences in conjugation, these three carotenoids are structurally identical. (See Fig. 1). In the present experiments, flash absorption spectroscopy is used to monitor the triplet energy transfer from the BChl to the carotenoid. Incorporation of these three carotenoids into the B850 complex of Rb. sphaeroides R-26.1 makes it possible to study the effects of varying the triplet state energies and dynamics of carotenoids on the triplet energy transfer from BChl to carotenoids.

Materials and methods

Sample preparation The B800-850 antenna complex of Rb. sphaeroides wild type 2.4.1 was prepared as previously described (Frank et al. 1987). The B850 light-harvesting complex from Rb. sphaeroides R-26.1 was prepared as follows: The Rb. sphaeroides R-26.1 cells were diluted with 15 mM Tris buffer, pH 8.0, in order to obtain an absorbance of 40-50 at 850 nm. A very small amount of solid MgCIa (,-~10 mg per 250 mL) and DNAase (,-~5 mg per 250 mL) was added to the cells. The cells

were then passed through a French pressure cell three times at 20 000 psi. During the procedure both broken and unbroken cells were kept on ice in the dark. The broken cells were then centrifuged in a Sorvall SS34 rotor at 27000 × g at 4 °C for 10 min. The supernatant was decanted into a flask and kept on ice. The pellet-free supernatant was centrifuged at 250 000 x g in a 55.2 Ti rotor at 4 °C for 90 min. The yellow supernatant was discarded. At this stage the pellet had a bright, shiny, blue-green color and was suspended in 15 mM Tris buffer, containing 150 mM NaCI, pH 8.0. The final absorbance was 40-50 at 860 nm. While stirring in low light, N,N-dimethyldodecyl-amine-Noxide (LDAO) was added dropwise to a final concentration of 0.6% v/v. The mixture was then stirred for 30 min at room temperature in the dark, and then centrifuged at 250 000 x g for 90 min at 4 °C. The supernatant contained mostly reaction centers, and the pellet contained the B850 light-harvesting complex. The B850 antenna complex was purified using a discontinuous sucrose density gradient consisting of 0.3 M, 0.6 M and 1.2 M sucrose in 20 mM Tris buffer, 0.1% lithium dodecyl sulfate (LDS), pH 8.0. The pellet was added to the top of the 0.3 M solution and centrifuged at 150 000 x g at 4 °C for 18 h. The purified B850 light-harvesting complex was located and removed from the 1.2 M sucrose layer. Spheroidene and 3,4-dihydrospheroidene were extracted from the whole cells of Rb. sphaeroides wild type strain 2.4.1 and Rhodobacter capsulatus M T l l 3 1 as previously described (DeCoster et al. 1992). The synthesis and purification of 3,4,5,6-tetrahydrospheroidene has been previously described (Gebhard 1991a; Gebhard et al. 1991b). The incorporation of the carotenoids into the lightharvesting complex was done similarly to the procedure described by Noguchi et al. (1990) with some modifications. The light-harvesting complex was dialyzed against 15 mM Tris buffer containing no detergent, pH 8.0, overnight. Then, 10% sodium deoxycholate (m/v) in 15 mM Tris buffer, pH 8.0, was added to obtain a final solution of 2% of sodium deoxycholate. A 20-fold molar excess of carotenoids relative to BChl was added in petroleum ether (35-60 °C b.p.) onto the surface of the light-harvesting complex. The petroleum ether was evaporated with a steam of nitrogen. The resulting mixture was sonicated for 30--45 min at 4 °C in the dark. The solution was dialyzed against 0.02% sodium deoxycholate, 15 mM Tris buffer, pH 8.0, overnight. The solution was layered on a sucrose density gradient using 2 M, 1.5 M and 0.75 M in 15

159 mM Tris buffer, pH 8.0. Centrifugation at 150 000 × g at 4 °C for 18 h eliminated the excess of carotenoids. The B850 light-harvesting complex with carotenoids incorporated was found at the interface of 1.5 M and 2 M sucrose solutions. The sample concentration was adjusted with 15 mM Tris buffer containing 0.02% sodium deoxycholate to an absorbance of 1.1-1.2 at 850 nm for the transient experiments.

Steady state spectral analysis The absorption spectra were recorded at room temperature using an SLM Aminco 3000 single diode anay spectrometer. The circular dichroism (CD) spectra were obtained using a Jasco-710 spectropolarimeter at room temperature. A quartz microcell with a 1 cm path were used for the CD measurements. The concentrations of the samples used in these experiments were based on the absorption of the complex at 850 nm using the extinction coefficient 0.184 # m - l c m - 1 at this wavelength (Clayton and Clayton 1981). The concentrations were: 5.5 x 10 - 6 M for the B800-850 complex from Rb. sphaeroides wild type 2.4.1, 7.0 × 10 - 6 M for the B850 complex from Rb. sphaeroides R-26.1 incorporated with spheroidene, 4.6 × 10 -6 M for the B850 complex incorporated with 3,4-dihydrospheroidene, 4.3 x 10 - 6 M for the B850 complex with 3,4,5,6-tetrahydrospheroidene, and 6.1 x 10 -6 M for the B850 complex lacking carotenoids.

Transient absorption measurements Laser-induced optical absorption changes were measured at room temperature in the following way: Light from a 450 W Xe arc lamp was filtered through 5 cm of water in a glass bottle and focused through a pinhole aperture onto a 1 cm path sample cuvette after passing through a 450 nm cut-off filter. The light transmitted through the sample was detected through an Instrument SA model L H 2 0 1200 g/mm monochromator and focused with a lens onto a photodiode detector (EG & G type-539). The output was then amplified using an Evans Associates model 4163 amplifier and fed to a Nicolet LAS 12/70 transient digitizer operating at a dwell time of 2 #s per point (total trace 256 points). This provided the best manner in which to resolve the ,-~100 #s lifetime of the BChl transient simultaneously on the same dynamics curve with the ,,~10/~s lifetimes of the carotenoids. The intensity of the measuring beam was kept below the saturation level for the transient signals. Also, to prevent any photoxidation of

the light harvesting complex during the measurements, a small amount of sodium dithionite was added to the samples to render them anaerobic. The flash-induced photochemistry was initiated by a Quanta-Ray DCR3/PDL-2 Nd:YAG-pumped dye laser (pulse duration: 7 ns; A = 580 nm) using the dye Rhodamine 610 in pure spectroscopic grade methanol. The pulse repetition rate was 20 Hz. The laser beam was focused onto the sample at a right angle to the measuring beam. The laser intensity was kept in the region where the amplitudes of the transients increased linearly with flash intensity.

Results The steady state absorption spectra allow a calculation of the extent of carotenoid incorporation relative to the B850 BChl in each sample (Frank et al. 1993a). This was done using the extinction coefficients of the carotenoids at maximum absorption to calculate their concentrations (Cogdell et al. 1976). The concentrations were then normalized to the absorbance of BChl at 850 nm. The extents of carotenoid incorporation were then determined by comparing these numbers with the same value for the Rb. sphaeroides wild type B800-850 complex in which 100% incorporation of spheroidene is assumed. It was also assumed that the 850 nm extinction coefficient for BChl in the B850 complex is the same as in the B800-850 complex. The extents of incorporation of carotenoids into the B850 complex were calculated to be 35 + 3% for spheroidene, 80 + 3% for 3,4-dihydrospheroidene and 48 + 5% for 3,4,5,6-tetrahydrospheroidene. Despite the fact that all of the carotenoids were treated similarly, the extents of incorporation were quite different for the different carotenoids. This can be attributed to the somewhat different solubilities these molecules have in the detergent solution, their different propensities to aggregate and their disparate chemical stabilities. Fluorescence excitation spectroscopy was used to verify that the carotenoids are actually bound in the light-harvesting complexes and functioning as light-harvesting pigments (Frank et al. 1993a). The molecules were excited into the carotenoid region and the BChl emission band at 850 nm was monitored to confirm singlet energy transfer between the carotenoid and the BChl. Carotenoid-to-BChl singlet state energy transfer efficiencies for these samples were reported in previous paper (Frank et al. 1993a). The values were found to be 56 4- 3% for spheroidene, 71 -4- 6% for 3,4-dihydrospheroidene and

160

a)

b)

B800-850

B850

+ spheroidene

5 0 ps

1 O 0 I~s F ;i

F-

H-

v I

I

I

I

I

I

I

c)

B850

I

I

Time

Time

d)

+ 3,4-dihydrospheroidene

B850

+ 3,4,5,6-tetrahydrospheroidene

1 O 0 I~s I~~I F-

F-

I

I

I

I

I

I

I

I

I

Time

Time

e)

B850

!oo

z,,w

I

I

I

I

I

Time Fig. 2.

Rise and decay of the triplet-triplet transient transmittance changes excited at 580 nm and probed at (a) 535 nm for the B800-850 complex; (b) 535 nm for spberoidene incorporated into the B850 complex; (c) 505 nm for 3,4-dihydrospheroidene incorporated into the B850 complex; (d) 495 nm for 3,4,5,6-tetrahydrospheroidene incorporated into the B850 complex; and (e) 490 nm for the B850 complex lacking carotenoids. The probe wavelengths correspond to the maximum signal for each sample. The smooth line represents the calculated values for the best fit to a single (for curves (a), (d) and (c)) and a double (for curves (b) and (c)) exponential rate expression. All experiments were done at room temperature.

161 [3BChl]o + eSar [1Car]o + eSchl[lBChl]o (2) Where eTa~ and eBChl T are the triplet extinction coefficients of the carotenoid and BChl, [3Car]o and [3BChl]o are the triplet state concentrations immediately after the laser flash, [1Car]o and [lBChl]o are the concentrations of the ground and excited singlet states of the carotenoid and BChl immediately after the laser flash (i.e. at time t = 0), and I' is the intensity of transmitted light immediately after the laser flash. Subtracting Eq. (1) from Eq. (2) and using

['car] = ['Car]o + [ Car]o AI

0

tI

[1BChl] = ['BChl]o + [3BChl]o I t = I - AII

Io Time

A~BChl ~

Fig. 3. Transient schematic that defines the notation in the model evaluating the triplet decay parameters. Io, measuring light intensity; I, intensity of light transmitted through the sample; V, intensity of light transmitted through the sample immediately after the laser flash; I ' , intensity of light transmitted through the sample after laser flash at a time tl; AIt = 1 - I ~ and A I ' = I - I".

66 + 4% for 3,4,5,6-tetrahydrospheroidene incorporated into the B850 complex. Figure 2 shows representative kinetic traces of the flash-induced transmittance changes for (a) the B 800-850 antenna complex of Rb. sphaeroides wild type strain 2.4.1; the B850 light-harvesting complex from Rb. sphaeroides R-26.1 incorporated with (b) spheroidene; (c) 3,4-dihydrospheroidene; (d) 3,4,5,6tetrahydrospheroidene; and (e) the Rb. sphaeroides R-26.1 B850 complex lacking carrotenoids. The transients were analyzed according to either single or double exponential rate expressions according to the following kinetic model: Referring to Fig. 3, before the laser flash, the absorbance of the sample is A -- l o g ~ = eScar[1Car] + escht[XBChl]

(1)

where ~BChl s and 6BChl s are the singlet extinction coefficients of the cartenoid and BChl, [1Car] and [1BChl] are the singlet state concentrations before the laser flash, Io is the intensity of the measuring light, and I is the intensity of the measuring light transmitted through the sample. Immediately after the laser flash (in the present experiment this is within 2 #s of the flash) the absorbance becomes Io T Ao = 1og~7 = e~ar [3Car]o + eBCh 1

A6car =

T S 6 B c h l - 6BChl 6Tar -- 'Sar

one obtains for any time, t, after the laser flash: I

l o g i _ A i t = Aecar [3Car]t q- AeBchl[3BChl]t

(3)

At a time h that is long compared to both the carotenoid triplet lifetime (,-~10 #s) and the BChl triplet lifetime (,,,80 #s), it can be assumed that all carotenoid triplets have decayed to the ground state while some residual BChl triplet signal remains. Under these conditions [3Car]tl = 0. Therefore, at time h Eq. (3) reduces to I

log I - AI" = AeBchl[3BChl]h

(4)

From Eq. (4) the concentration of BChl triplets at time h, [3BChl]h, can be calculated. Then using the BChl triplet lifetime of 80 ps measured for the B850 complex, one can obtain the concentration of BChl triplets immediately following the flash; i.e. at time t = 0. This is done using the relationship [3aChl]o = e ktl [3BChl]h

(5)

Where k =1/~- and r is the BChl triplet lifetime. These [3BChl]o values represent the amount of BChl that is not quenched by the carotenoid. They can then be used in Eq. (3) in which AeCa~ and Aeachl are known from the literature (Borland et al. 1987, 1989) to obtain the triplet state concentrations of the carotenoids, [3Car]o, immediately after the laser flash. The Aeachl value was determined as follows. The 25,000 M-1 c m - l value o f A~BChl at 360 nm and the triplet spectrum from Borland et al. (1987) were used to obtain a value of 9000

162

v

I

460

I

480

500

I

520

I

I

540

560

580

Wavelength/nm Fig. 4. Spectra of flash-induced transmittancechanges. The plot displays the spectraof the single exponential, 80 #s decaytransient of the B850 complex(A), the 14/zs componentof the double exponential decay transient from the B850 complex incorporated with spheroidene (o), and the 19 #s componentof the doubleexponential decaytransient from the B850 complexincorporatedwith 3,4-dihydrospheroidene (o). The changes in the transmittance (AT) were divided by the transmittance (T) at each wavelengthfor normalization purposes.

M - 1 c m - l for AeaChl in the 505-535 nm region which is appropriate to the present experiments. These triplet extinction coefficients were derived from the spectrum of BChl in pyridine (Borland et al. 1987). This is justified here because in that solvent BChl is a pentacoordinated monomer (Ballschmiter and Katz 1968; Shipman et al. 1976) as is thought to be the case in the light-harvesting complexes where they are probably ligated to histadine residues (Zuber 1986). The [3Car]o concentrations are numerically identical to the concentrations of BChl that have been quenched. Therefore, the following equation can be used to calculate the percentage of BChl quenched. % of BChl quenched

=

[3Car]o/([3Car]o + [3BChl]o) However, in order to compare these numbers to the efficiencies of singlet energy transfer which are measured relative to the B850 BChl, the values for the percentage of BChl triplets quenched must also be normalized. This is done by dividing the percentages by the fraction of carotenoid incorporation relative to B850 BChl. The normalized values are presented in Table 1. The transient profiles of the B850 complex incorporated with spheroidene and 3,4-dihydrospheroidene (Figs. 2b and c) were best fit by double exponential kinetic rate expressions. This is because in both cases, the incorporation of the carotenoids into B850 com-

plex was not 100%. Therefore, not all the BChl triplets are quenched by the carotenoids, and the transient profiles correspond to both BChl and carotenoid triplet decay. The transient profiles of B 800°850, B 850 incorporated with 3,4,5,6-tetrahydrospheroidene and B850 complexes (Figs. 2a, d and e) were best fit by single exponential decay expressions. This is because for the B8000850 complex essentially all the BChl triplets are quenched by carotenoids triplets, and the transient corresponds to a single species decay. Also, for the B850 complex which lacks carotenoids, the transient corresponds only to the decay of BChl triplet states and is best fitted by a single exponential. For the B800-850 complex, the lifetime of the decaying species (7.0 + 0.1 #s) corresponds well to previously measured carotenoid triplet decay values of 2-8 #s obtained by Monger et al. (1976) and the ,-,12 #s value obtained by Bensasson et al. (1976) in vitro. The transient species observed after excitation of the B850 complex displayed a lifetime of 80 + 5 #s and is assigned to the BChl triplet based on the previously observed value of 70 #s measured by Monger et al. (1976). Based on the lifetimes measured in these two control samples, the faster kinetic component observed in the carotenoid-incorporated samples could be attributed to the carotenoid triplet decay, and the longer-lived component could be assigned to the BChl. To confirm these assignments, the transient amplitudes of the B850 complexes incorporated with spheroidene and 3,4-dihydrospheroidene were plotted versus wavelength. (See Fig. 4). This figure shows that the spectral maximum of the fast component of the biexponential fit to the decay traces is different for the two complexes. In fact, the spectral maximum shifts to shorter wavelength as the carotenoid chain is shortened. This is very typical of triplet signal observed from carotenoids and reported in the literature (Mathis and Kleo 1973; Truscott et al. 1973). Furthermore, no fast (,,- 10/is) component nor spectral maximum in this wavelength range is observed from the sample of B850 complex lacking carotenoids (Fig. 4). Instead, a slower, 80 #s transient is found with a gradual sloping spectral profile in this wavelength region. This is the same behavior observed by Borland et al. (1987) for BChl in solution. These dynamics and spectral traces confirm our assignments of these transients to carotenoids and bacteriochlorophyll. For the B850 complex incorporated with 3,4,5,6tetrahydrospheroidene, the best fit of the data was to a single exponential expression with a lifetime of 85 + 7/zs. This indcates that only BChl contributes to the

163 Table 1. The singlet energy transfer efficiencies, percentages of BChl quenched by triplet energy transfer, and extents of carotenoid incorporation. The percentages of BChl quenched were obtained using the formula [3Car]o/([3BChl]o+[3BChl]o) and were normalized for the extent of carotenoid incorporation relative to B850 BChl as described in the text. The data on the efficiency of Car-to-BChl singlet energy transfer was taken from a previous paper by Frank et al. (1993a). All values are expressed as percents (%) Complex

Efficiency of Car to BChl singlet energy transfer

Normalized percentages of BChl quenched

Extent of Car incorporation relative to B850 BChl

564-3

504-8

354-3

714-6

594-13

804-3

664-4

0

484-5

B850 incorporated with spheroidene B 850 incorporated with 3,4-dihydrospheroidene B850 incorporated with 3,4,5,6-tetrahydrospheroidene

Table 2. The lifetimes of the transients observed at 535 nm for the B800-850 complex from Rb. sphaeroides wild type 2.4.1,490 nm for the B850 complex from Rb. sphaeroides R-2.1,535 nm for the B850 complex incorporated with spheroidene, 505 nm for the B850 complex incorporated with 3,4-dihydrospheroidene, and 495 nm for the B850 complex incorporate with 3,4,5,6-tetrahydrospheroidene. The transients were analyzed according to the equation, y = A + B exp (-t/~-i) + C exp (-t/~-2), where y is the intensity of the transmitted light, A is the baseline, B and C are preexponential factors, t is the time, ~'1 and T2 are the lifetimes in microseconds (#s) of the shorter and longer decaying transients, nd means not determined

Table 3. Singlet (2lAg) and triplet state energies for BChl and the carotenoids used in the present experiments. The values for the triplet energies of the carotenoids were obtained assuming that they are approximately one-half of the energies of their lowest lying 21Ag singlet states (Hudson et. al 1982). The 2lAg energies were reported previously (DeCoster et al. 1992). The errors in the numbers were derived from the uncertainties in the assignments of the spectral origins of the 2lAg state fluorescence from the carotenoids. The triplet energy of BChl was taken from Takiff and Boxer (1988) for pentacoordinated BChl in solution na means not applicable Molecule

Complex B800-850 B850 B850 incorporated with spheroidene B850 incorporated with 3,4-dihydrospheroidene B850 incorporated with 3.4,5,6-tetrahydrosnheroidene

T1

T2

B/C

7.04-0.1 nd 144-2

nd 804-5 1304-30

nd nd 104-0.5

194-2

1204-20

184-0.5

nd

854-7

nd

triplet decay in this sample despite its large (~48%) extent of carotenoid incorporation. Table 2 summarizes the lifetimes of the triplet states measured here. In order to examine whether any major changes in the carotenoid or BChl structures resulted from incorporation of these carotenoids into the B850 complex, CD spectra were taken. As can be seen from Fig. 5, the carotenoid bands between 400 to 550 nm and the BChl bands between 300 to 400 nm and 550 to 650 nm for the incorporated complexes are very similar to the native B800-850 complex. The l o w intensities of the carotenoid bands in the B850

BChl spheroidene 3,4-dihydrospheroidene 3,4,5,6-tetrahydrospheroidene

31Ag energy ( c m - 1) T1 energy ( c m - 1) na 14,1004-100 15, 3004-200 16,700-1-150

8,2404-20 7,0504-50 7,6504-100 8,3504-75

complex incorporated with spheroidene or 3,4,5,6tetrahydrospheroidene is expected from the low percentage (relative to the 3,4-dihydrospheroidene) of incorporation of those molecule. No major structural differences in the bound carotenoids or bacteriochlorophylls are evident from the CD of these samples.

Discussion T h e B850 complex incorporated with 3,4,5,6tetrahydrospheroidene shows no evidence of carotenoid triplet state formation. This is despite the fact that there is a 48% extent of incorporation of this carotenoid relative to BChl. Moreover, the similarities in the CD spectra of this complex with the B850 complexes

164

b)

a) B 8 0 0 - 8 5 0 5O

8

E~ (b -0

-(3

\

E

-16

-20 400

B850

+

500 600 700 Wavelength/nm

i

d) B850

3,4-dihydrospheroidene

i

400

800

i

I

500 600 700 Wavelength/nm

+

800

3,4,5,6-tetrahydrospheroidene

5

6

E~

E~ (D -(3

-(3

E

E

- 3

+ spheroidene

O~

E

C)

B850

i

400

i

i

-3

i

500 600 700 Wavelength/nm

e)

800

400

500 600 700 Wavelength/nm

800

B850

6

400

500 600 700 Wavelength/rim

800

Fig. 5. Circular dichroism spectra of: (a) the B800-850 complex from Rb. sphaeroides wild type 2.4.1 ; (b) the Rb. sphaeroides R-26.1 B850 complex incorporated with spheroidene; (c) the B850 complex incorporated with 3,4-dihydrospheroidene; (d) the B850 complex incorporated with 3,4,5,6-tetrahydrospheroidene; and (e) the B850 complex lacking carotenoids. 0 is the ellipticity expressed in units of millidegrees.

165 incorporated with other carotenoids, where carotenoid triplet signals are observed, suggests that the binding of the carotenoid is appropriate for triplet transfer to occur. The absence of carotenoid triplet signals in the B850 complex incorporated with 3,4,5,6tetrahydrospheroidene is most likely due to the fact that this particular carotenoid, which has 8 carboncarbon double bonds, and has the shortest length of conjugation for the series presented here (see Fig. 1), has a triplet energy too high to quench the BChl triplet states that are formed. Although the exact triplet energy for any carotenoid molecule has never been directly determined (e.g. by phosphorescence) flash absorption measurements using triplet sensitizers of known triplet state energies (Land et al. 1971; Mathis and Kleo 1973; Bensasson et al. 1976) and theoretical studies on model polyenes (Hudson et al. 1982) suggest that their values should correspond to approximately onehalf the energies of their lowest lying singlet (2lAg) states. (See the review of Frank and Cogdell (1992a) for a discussion of this topic.) The 2lAg energies of several low lying singlet states of carotenoids have recently been described in the literature (DeCoster et al. 1992; Frank et al. 1993b). Based on these studies, the triplet state energies of the three carotenoids studied here are suggested to be 7050 cm -1, 7650 cm - l and 8350 cm -1 for spheroidene, 3,4-dihydrospheroidene and 3,4,5,6-tetrahydrospheroidene respectively (see Table 3). Only 3,4,5,6-tetrahydrospheroidene has a triplet energy that is higher than the pentacoordinated BChl triplet state energy of 8240 c m - l observed via phosphorescence techniques (Takiff and Boxer 1988). Based on phosphorescence measurements of the primary donor, dimeric, BChl in bacterial reaction centers, also carried out by Takiff and Boxer (1988), the triplet energy of 8240 cm-1 for monomeric BChl is probably slightly higher than expected for the dimeric BChl thought to comprise the B850 chromophore. However, this would not affect the conclusion that carotenoids having less than 9 carbon-carbon double bonds are incapable of trapping triplet energy in the light-harvesting complexes because their triplet energies lie above that of BChl. Precisely the same observation has been made for triplet quenching in reaction centers (Frank and Farhoosh, unpublished results). The other carotenoids are capable of acting as efficient triplet quenchers because their triplet states lie at energies below that of BChl. It is known that for triplet energy transfer to be efficient, very close proximity, essentially van der Waals contact, must be achieved between pigments

(Gust and Moore 1991). Owing to its spin forbiddenness, triplet energy transfer must proceed via electron exchange (Dexter 1953) as opposed to the dipole mechanism advanced by F6rster (1948, 1965). Singlet energy transfer to BChl via the exchange mechanism has also been postulated to occur from the low lying 2lAg states of carotenoids (See Frank and Cogdell (1992a) for a review of this topic). The exchange mechanism is appropriate to describe singlet energy transfer from states that involve transitions having vanishingly small oscillator strength owing to the forbiddenness of their transitions with the ground state. The data in Table 1 show that the efficiencies of carotenoid-to-BChl singlet energy transfer and the normalized values for the percentage of BChl quenched are very similar. This suggests that both singlet and triplet energy transfer may be occurring by the same mechanism. As discussed above, this mechanism is most likely the exchange mechanism.

Acknowledgements The authors wish to thank Dr Paul Mathis for useful discussions. This work was supported by grants to H.A.E from the National Institutes of Health (GM30353) and the University of Connecticut Research Foundation, and to J.L. and R.G. from the Netherlands Foundation of Chemical Research (SON), which is financed by the Netherlands Organization for the Advancement of Pure Research (NWO).

References Bensasson R, Land EJ and Maudinas B (1976) Triplet states of carotenoids from photosyntheticbacteria studied by nanosecond ultraviolet and electron pulse irradiation. Photochem Photobiol 23:189-193 Borland CF, McGarvey DJ, Truscott TG, Cogdell RJ and Land EJ (1987) Photophysicalstudies of Bacteriochlorophylla and bacteriopheophytin a - Singlet oxygen generation. J Photochem Photobiol B 1:93-101 Borland CE Cogdell RJ, Land EJ and Truscott TG (1989) Bacteriochlorophyll a triplet state and its interactions with bacterial carotenoids and oxygen.J PhotochemPhotobiol 3:237-245 Boucher E van der Rest M and Gingras G (1977) Structure and function of carotenoids in the photoreaction center from Rhodospirillium rubrum. Biochim Biophys Acta 461:339-357 Clayton RK and Clayton BJ (1981) B850 pigment-protein complex ofRhodopseudonas sphaeroides: Extinctioncoefficients,circular dichroism, and the reversible bindingof bacteriochlorophyll.Proc Natl Acad Sci 78:5583-5587 Cogdell RJ, Parson WW and Kerr MA (1976) The type. amount, location and energy transfer properties of the carotenoid in reaction centers from Rhodopseudomonas sphaeroides. BiochimBiophys Acta 430:83-93

166 Cogdell RJ and Frank HA (1987) How carotenoids function in photosynthetic bacteria. Biochim Biophys Acta 895:63-79 DeCoster B, Christensen RL, Gebhard R, Lugtenburg J, Farhoosh J and Frank HA (1992) Low lying electronic states of carotenoids. Biochim Biophys Acta 1102:107-114 Dexter DL (1953) A theory of sensitised luminescence in solids. J Chem Plays 21: 836-860 Duyens LNM (1952) Transfer of excitation energy in photosynthesis. Thesis, University of Utrecht, The Netherlands Foote CS (1968) Mechanisms of photosynthesized oxidation. Science 162:963-970 Ftirster Th (1948) Intermolecular energy transfer and fluorescence. Ann Phys 2:55-75 FOrster Th (1965) Action of light and organic crystals. In: Sinanoglu O (ed) Modem Quantum Chemistry Part III, pp 93-137. Academic Press, New York Frank HA (1992b) Electron paramagnetic resonance studies of carotenoids. In: Packer L (ed) Methods in Enzymology. Carotenoids, Part A. Chemistry, Separation, Quantitation and Antioxidation, Vo1213, pp 305-312. Academic Press, New York Frank HA and Cogdell RJ (1992a) Photochemistry of carotenoids. In: Young A and Britton G (eds) Carotenoids in Photosynthesis, pp 252-326. Springer-Verlag, London Frank HA, Bolt JD, De B Costa SM and Saner K (1980) Electron paramagnetic resonance detection of carotenoid triplet states. J Am Soc 102:4893-4898 Frank HA, Chadwick BA, Oh JJ, Gust D, Moore TA, Liddell PA, Moore AL, Makings LR and CogdeU RJ (1987) Triplet-triplet energy transfer in B800-850 light-harvesting complexes of photosynthetic bacteria and synthetic carotenoporphyrin molecules investigated by electron spin resonance. Biochim Biophys Acta 892:253-263 Frank HA, Farhoosh R, Aldema ML, DeCoster B, Christensen RL, Gebhard R and Lugtenburg J (1993a) Carotenoidto-bacteriochlorophyll singlet energy transfer in carotenoidincorporated B850 light-harvesting complexes of Rhodobacter sphaeroides R-26.1. Photochem Photobiol 57:49-55 Frank HA, Farhoosh R, Gebhard R, Lugtenburg J, Gosztola D and Wasielewski MR (1993b) The dynamics of the $1 excited states of carotenoids. Chem Phys Lett 207:88-92 Gebhard R (1991a) Synthesis and Application of Isotopically and Chemically Modified Spheroidene, Phi) Thesis, University of Leiden, The Netherlands Gebhard R, Van der Hoef K, Violette CA, De Groot HJM, Frank HA and Lugtenburg J (1991b) 13C MAS NMR evidence for a 15, 15r-Z configuration of the spheroidene chromophore in the Rhodobacter sphaeroides reaction center. Pure Appl Chem 63: 115-122 Goedheer JC (1959) Energy transfer between carotenoids and bacteriochlorophyll in chromatophores of purple bacteria. Biochim Biophys Acta 35:1-8 Goedheer JC (1969) Energy transfer from carotenoids to chlorophyll in blue-green, red and green algae and greening bean leaves. Biochim Biophys Acta 172:252-265

Gust D and Moore TA (1991) Mimicking photosynthetic electron and energy transfer. In: Volam D, Hammond G and Neckers D (eds) Advances in Photochemistry, Vol 16, pp 1-63. John Wiley & Sons, New York Hudson BS, Kohler BE and Shulten K (1982) Linear polyene electronic structure and potential surfaces. In: Lim EC (ed) Excited States, Vol 6, pp 22-95. Academic Press, New York Land EJ, Sykes A and Truscott TG (1971) The in vitro photochemistry of biological molecules-II. The triplet states of ~-carotene and lycopene excited by pulse radiolysis. Photochem Photobiol 13:311-320 Mathis P (1969) Triplet-triplet energy transfer from chlorophyll a to carotenoids in solutions and in chloroplast. In: Metzner H (ed) Progress in Photosynthesis Research, Vol 2, pp 881-822. TUbingen, Germany Mathis P and Kleo J (1973) The triplet state of t-carotene and of analog polyenes of different lengths. Photochem Photobiol 18: 343-346 Monger TG, Cogdell RJ and Parson WW (1976) Triplet states of bacteriochlorophyll and carotenoids in chromatophores of photosynthetic bacteria. Biochim Biophys Acta 449:136-153 Noguchi T, Hayashl H and Tasumi M (1990) Factors controlling the efficiency of energy transfer from carotenoids to bacteriochlorophyll in purple photosynthetic bacteria. Biochim Biophys Acta 1017:280-290 Renger G and Wolff C (1977) Further evidence for dissipative energy migration via triplet states in photosynthesis. The protective mechanism of carotenoids in Rhodopseudomonas sphaeroides chromatophores. Biochim Biophys Acta 460:47-57 Schenck CC, Mathis P and Lutz M (1984) Triplet formation and triplet decay in reaction centers from the photosynthetic bacterium Rhodopseudomonas sphaeroides. Photochem Photobiol 39: 407-417 Shreve AP, Trautman JK, Frank HA, Owens TG and Albrecht AC (1991) Femtosecond energy-transfer processes in the B800850 light-harvesting complex of Rhodobacter sphaeroides 2.4.1. Biochim Biophys Acta 1058:280-288 Shipman LL, Cotton TM, Norris JR and Katz JJ (1976) An analysis of the visible absorption spectrum of chlorophyll a monomer, dimer, and oligomers in solution. J Am Chem Soc 98:8222-8230 Siefermann-Harms D (1985) Carotenoids in photosynthesis. I. Location in photosynthetic membranes and light-harvesting function. Biochim Biophys Acta 811:325-355 Takiff L and Boxer SG (1988) Phosphorescence spectra of bacteriochlorophyUs. J Am Chem Soc 110:4425--4426 Truscott TG, Land EJ and Sykes A (1973) The in vitro photochemistry of biological molecules - I I I . Absorption spectra, lifetimes and rates of oxygen quenching of the triplet states of t-carotene, retinal and related polyenes. Photochem Photobiol 17:43-51 Witt HT (1971) Coupling of quanta, electrons, felds, ions and phosphorylation in the functional membrane of photosynthesis. Q Rev Biophys 4:365-477 Zuber H (1986) Structure of light-harvesting antenna complexes of photosynthetic bacteria cyanobacteria and red algae. Trends Biol S c i l l : 414-419