Photosynthesis Research 31: 75-87, 1992. (~) 1992 Kluwer Academic Publishers. Printed in the Netherlands.

Regular paper

Temperature dependence and polarization of fluorescence from Photosystem I in the cyanobacterium Synechocystis sp. PCC 6803 Bruce P. Wittmershaus 1'3, Vincent M. W o o l f 1'3 & Wim F.J. Vermaas 2'3

~Department of Physics and Astronomy; 2Department of Botany; and 3The Center for the Study of Early Events in Photosynthesis, Arizona State University, Tempe, AZ 85287, USA Received 27 March 1991; accepted in revised form 23 September 1991

Key words: cyanobacteria, energy transfer, fluorescence, Photosystem I, polarization Abstract

To determine the fluorescence properties of cyanobacterial Photosystem I (PS I) in relatively intact systems, fluorescence emission from 20 to 295 K and polarization at 77 K have been measured from phycobilisomes-less thylakoids of Synechocystis sp. PCC 6803 and a mutant strain lacking Photosystem II (PS II). At 295 K, the fluorescence maxima are 686 nm in the wild type from PSI and PS II and at 688 nm from PSI in the mutant. This emission is characteristic of bulk antenna chlorophylls (Chls). The 690-nm fluorescence component of PSI is temperature independent. For wild-type and mutant, 725-nm fluorescence increases by a factor of at least 40 from 295 to 20 K. We model this temperature dependence assuming a small number of Chls within PS I, emitting at 725 nm, with an energy level below that of the reaction center, P700. Their excitation transfer rate to P700 decreases with decreasing temperature increasing the yield of 725-nm fluorescence. Fluorescence excitation spectra of polarized emission from low-energy Chls were measured at 77 and 295 K on the mutant lacking PS II. At excitation wavelengths longer than 715 nm, 760-nm emission is highly polarized indicating either direct excitation of the emitting Chls with no participation in excitation transfer or total alignment of the chromophores. Fluorescence at 760 nm is unpolarized for excitation wavelengths shorter than 690 nm, inferring excitation transfer between Chls before 760-nm fluorescence Occurs.

Our measurements illustrate that: 1) a single group of low-energy Chls (F725) of the core-like PSI complex in cyanobacteria shows a strongly temperature-dependent fluorescence and, when directly excited, nearly complete fluorescence polarization, 2) these properties are not the result of detergentinduced artifacts as we are examining intact PSI within the thylakoid membrane of S. 6803, and 3) the activation energy for excitation transfer from F725 Chls to P700 is less than that of F735 Chls in green plants; F725 Chls may act as a sink to locate excitations near P700 in PS I.

Abbreviations: Chl-chlorophyll; BChl-bacteriochlorophyll; PS-Photosystem; S. 6803-Synechocyst& sp. PCC 6803; P G P - potassium glycerol phosphate Introduction

Since their discovery (Butler 1961), the purpose of low-energy antenna chlorophylls (Chls) and bacteriochlorophylls (BChls) with first excited singlet-state energy levels at or below that of the

primary trap (the reaction center primary donor) has been puzzling (Butler 1978, Satoh and Butler 1978a, b, Butler et al. 1979, Karukstis and Sauer 1983, Golbeck 1987, Scheer and Schneider 1988). Our work is aimed at characterizing the fluorescence properties of antenna Chls from

76 Photosystem I (PS I) of cyanobacteria, especially the low-energy Chls. Low-energy antenna Chls in PS I of green plants have been discovered with fluorescence maxima at 720 and 735 nm (Butler 1961, Goedheer 1964, Butler 1978, Satoh and Butler 1978a,b, Butler et al. 1979, Mullet et al. 1980), while in algae and cyanobacteria, such Chl species have fluorescence maxima at 708720 nm (Goedheer 1964, Wollman and Bennoun 1982, Garnier et al. 1986, 1987), and at 720730nm (Goedheer 1964, Mohanty et al. 1972, Rijgersberg and Amesz 1980, Takahashi et al. 1982, Huang and Berns 1983, Vermaas et al. 1987a, Wittmershaus 1987a), respectively. The first excited singlet-state energy levels of these Chls are all close to or below that of the reaction center of PS I, P700, which has an absorption maximum at 700 nm. In the photosynthetic bacteria Rhodopseudomonas viridis (Thornber et al. 1978), Rhodobacter sphaeroides (Kramer et al. 1984, Van Grondelle et al. 1988), Rhodospirillum rubrum (Kramer et al. 1984), and Heliobacterium chlorum (Trost and Blankenship 1989) antenna BChls with energy levels close to or below that of the reaction center have also been observed. At first glance it would appear that these Chls and BChls should compete with the reaction center as traps for excitations coming from the bulk antenna pigments. This would present a hindrance to the efficiency of photoconversion. It is believed that a possible role of Chls and BChls with low-energy excited states may be to trap and localize excitations from the bulk antenna close to the reaction center (Butler et al. 1979, Tusov et al. 1980, Wittmershaus 1987a, Van Grondelle and Sundstrom 1988, Van Grondelle et al. 1988). Sufficient thermal energy available at room temperature (0.0257eV at 298 K) would allow excitations in energy levels below the reaction center to then be transferred back to the reaction center. If this hypothesis is correct, decreasing the temperature is expected to lead to an increase in fluorescence lifetime of low-energy Chls resulting in an increase in fluorescence yield. This relation between the temperature dependence of fluorescence yield and lifetime has been observed (Butler et al. 1979, Tusov et al. 1980). More recent work implies that the change in fluorescence lifetime probably is not the only factor contributing to the tern-

perature dependence of the fluorescence yield (Tabbutt 1987, Mukerji and Sauer 1989). No consensus has been reached yet. A number of models (some inconsistent with each other) have been proposed describing the organization and connection of antenna and reaction centers in PSI and based primarily on measurements on PS I in eukaryotes (Bassi and Simpson 1987, Tabbutt 1987, Wittmershaus 1987a, Searle et al. 1988, Mukerji and Sauer 1989). Typical evidence for the presence of low-energy Chls (or BChls) is a strongly temperaturedependent, highly polarized long-wavelength fluorescence upon excitation at wavelengths greater than the absorption maximum of the reaction center (Wollman and Bennoun 1982, Kramer et al. 1984). This temperature-dependent emission is not a property of monomeric Chl in solution (Goedheer 1964). In PS I from green plants, many studies have examined a population of Chls with a fluorescence peak at 735 nm (F735 Chls) which increases in yield by about a factor of 30 from 295 to 70 K (Butler et al. 1979, Tusov et al. 1980, Mukerji and Sauer 1989). Through the use of detergents, PS I from green plants can be separated into two parts, LHC I (a light-harvesting Chl-protein complex) and PS 1-100 or the PSI core complex which contains approximately 100 antenna Chl a molecules per P700 (Lam et al. 1984, Malkin et al. 1985, Vainstein et al. 1989). The Chl a/b protein LHC Ib containing the F735 Chls has been isolated and identified as part of LHC I (Haworth et al. 1983, Lam et al. 1984, Vainstein et al. 1989). PS 1-100 also contains a population of low energy Chls (F720 Chls) exhibiting a strong temperature dependence in yield and lifetime of emission at 720 nm (Malkin et al. 1985, Tabbutt 1987). Many fluorescence measurements on the core antenna system of PSI have been performed on PSI particles prepared with detergents. To avoid the possibility of fluorescence artifacts resulting from shifts of absorption and fluorescence spectra (Nechushtai et al. 1986) and disconnection of the antenna pigments and the reaction center upon extraction, results obtained on intact systems are preferable. Thus, we utilized the wild type and a PS II-less mutant of the cyanobacterium Synechocystis sp. PCC 6803 (S. 6803)

77 (Vermaas et al. 1988, Yu and Vermaas 1990) to measure temperature-dependent emission from PS I. Cyanobacteria are attractive organisms for such studies. Their antenna system is simpler because PS I from cyanobacteria contains no Chl b or an equivalent to the LHC I antenna complex of green plants (Vermaas et al. 1987a, Golbeck 1987). The mutant of S. 6803 lacking PS II permits measurements on P S I without the additional fluorescence components from Chls associated with PSII. In cyanobacteria, there are about 120 Chl a per P700 in P S I with only a single group of low-energy Chls characterized by an emission peak at 725 nm at low temperature. The low-energy Chls of S. 6803 are closer energetically to P700 than the 735-nm emitting Chls in LHC I and functionally similar to the F720 Chls found in the PS 1-100 particles prepared from green plants (Malkin et al. 1985). The two apoproteins making up the P S I core complex of the cyanobacterium Synechococcus sp. PCC 7002 are more than 95% homologous to those of higher plants and algae indicating a strong conservation of structure and function (Cantrell and Bryant 1987). Our aim is to characterize the temperaturedependent fluorescence of a core-like complex of PS I without contributions of PS II emission or possible artifacts caused by detergent isolation. The PS II-less mutant of the cyanobacterium S. 6803 allows this. Observing the fluorescence characteristics of the low-energy Chls of the core complex of P S I is very difficult in other undisturbed systems, such as in chloroplasts of green plants where fluorescence from F735 Chls dominates so that emission from the F720 Chls is difficult to resolve. If the temperature-dependent fluorescence of low-energy Chls in P S I from both cyanobacteria (F725 Chls) and LHC I in green plants (F735 Chls) can be attributed to changes in excitation transfer to P700, then the difference in energy levels of these two Chl components relative to P700 should cause the temperature dependence of their fluorescence to differ appropriately. Another aim of our study is to investigate whether F725 Chls have highly polarized emission which has been observed in low-energy BChls (Kramer et al. 1984) and Chls (Mukerji and Sauer 1990). Polarization studies can provide information regarding excitation

transfer among the low-energy Chls of P S I from S. 6803.

Materials and methods

The wild-type and mutant lacking PS II of S. 6803 were cultivated as described previously (Vermaas et al. 1987a). The mutant lacking PS II was created by deletion of psbDI and psbDII, the genes coding for the reaction center proteins D2, and ofpsbC, the gene coding for CP43. This mutant has been characterized as containing no D1, D2 and CP43 and only a trace of CP47 (Vermaas et al. 1987a,b, Vermaas et al. 1988, Yu and Vermaas 1990). Thylakoids were isolated from the cells as described (Vermaas et al. 1990). Remaining phycobilisomes were removed by adding dodecylmaltoside to give a detergent/Chl ratio of 1:2 (w/w) and then collecting the thylakoids by centrifugation at 14 000 rpm for 15 min. This step was repeated twice followed by a wash in buffer to remove any remaining detergent. Thylakoids were finally suspended in a buffer solution of 50mM H E P E S / N a O H ( p H = 7 , 23 °C), 5 mM MgCI 2, 50mM CaC12, 5% (v/v) glycerol and 0.5% (v/v) dimethylsulfoxide, immediately frozen in liquid nitrogen, and stored at - 7 0 °C until use. Two separate cultures of both the wild-type and mutant were used in the experiments to account for possible growth-related biological variations. For temperature-dependence measurements, thylakoids were thawed, diluted in the buffer described above, and resuspended to a final solution ( p H = 7 , 23°C) of 10, 30 and 60% by volume of buffer/thylakoids, glycerol and potassium glycerol phosphate (PGP) (75% v/v), respectively. The final Chl concentration was approximately 1.6/xg m1-1. The sample was placed within an O-ring between two glass slides held together with a copper ring sample holder to give a sample thickness of about 1 mm with an optical density, corrected for scattering, of less than 0.13 at 682 nm. The sample holder was attached to a helium refrigerator with quartz optical windows (Air Products Model DE202 head and 1R04WSL compressor) capable of temperatures ranging from 15 to 300K (---1%) when used with a temperature controller (Scientific Instr. Inc.,

78 Model 5500-5). Few cracks occurred in the sample upon freezing. Data were taken in steps of 5 to 10 K from 20 to 60 K, steps of 10 to 20 K from 60 to 180 K, and steps of 30 to 40 K from 180 to 290 K. Fluorescence measurements were conducted on a fluorimeter (SPEX Inc., Model F-102) with a cooled R-928 (Hamamatsu) photomultiplier tube as the detector. Fluorescence was collected at an angle of 90° to the excitation beam. Corrections for the response of the photomultiplier tube were made. Corrections were also made for intensity variation in the excitation light during and between measurements using a reference photodiode (UDT, Model Pin-6DP). The exciting light was at 440 nm (5.6-nm bandwidth) and emission collected from 650 to 780 nm (1.9-nm bandwidth). When excited at 440 nm, fluorescence from PGP peaks at 410 nm with a weak, smooth and gradually decreasing long-wavelength tail. All sample fluorescence spectra were corrected for PGP emission by subtracting the fluorescence spectrum of a solution of 10, 30 and 60% by volume of buffer, glycerol and PGP (75% v/v), respectively, taken under identical experimental conditions. Excitation polarization measurements were taken on thylakoids of the PS II-less mutant of S. 6803 suspended in a buffer/sample, glycerol, PGP solution with the same volume ratio used in the temperature-dependence measurements. The optical density, corrected for scattering, was less than 0.2 at 682 nm. The solution was placed in a glass tube of 4-mm internal diameter and cooled by immersing the tube in a liquid nitrogen dewar designed for optical spectroscopy. Dry air was blown on the outside surface of the dewar to prevent the accumulation of condensation. PGP in the solution allowed us to obtain a frozen glass with no cracks. The absence of cracks is essential to limit scattering. Scattered excitation light is nearly 100% polarized and can introduce large errors into a polarization measurement. A set of four measurements, one for each combination of the entrance and exit polarizers (i.e., HH, HV, VV and VH) is required for each polarization curve. The polarization, P, is calculated using the equation (Parker 1968) P = [ V V - (HV*VH/HH)]/ [VV + ( H V * V H / H H ) ] .

This equation corrects for any polarization dependence of the optics in the fluorimeter. Spectra were measured with exciting light from 670 to 725nm (5.6-nm bandwidth) and an emission wavelength of 760 nm (5.6-nm bandwidth) with a 750-nm long-pass cut-off filter used in the emission monochrometer to filter out any scattered excitation light.

Results

Figure 1 illustrates the basic spectral differences between room and low temperature emission from the wild type and PS II-less mutant of S. 6803. At 295 K, a prominent peak at 686 and 688 nm is present for the wild-type and the mutant lacking PS II, respectively, with the latter showing a proportionately larger shoulder at wavelengths longer than 700 nm in comparison to the wild-type. At 20K, each spectrum is dominated by a strong 725-nm emission attributed to the F725 low-energy Chls of PS I. This reflects the large increase in yield of 725-nm fluorescence at low temperatures relative to emission in the 690-nm region in agreement with earlier observations (Goedheer 1964, Mohanty et al. 1972, Rijgersberg and Amesz 1980). The wild-type has additional prominent bands at 685 and 695 nm attributed to emission from PS II (Goedheer 1964, Mohanty et al. 1972, Rijgersberg and Amesz 1980). In Fig. 2, results are presented to further illustrate the temperature dependence of emission from the mutant lacking PS II of S. 6803. It shows a broad, strongly temperature-dependent emission band centered near 725 nm at low temperatures. Its amplitude increases by at least a factor of 40 from 295 to 20 K. The shoulder at 690 nm, attributed to fluorescence from the bulk of the antenna Chls of PS I, shows little or no increase in yield upon lowering the temperature. In the wild type (Fig. 3), the temperature dependence of fluorescence from F725 Chls is observed to be like that of the mutant lacking PS II. The additional emission bands at 685 and 695 nm increase their fluorescence yield slightly with decreasing temperature (Goedheer 1964, Rijgersberg and Amesz 1980). The fluorescence at 690nm from PSI in the wild type is weak in comparison to the 685 and 695-nm emission from

79

1.0

I

I

-

t

I

I

I

!

I

I

I

es, (20 K) I I I

I

I

I

I

I

710

730

Wavelength

(nm)

/'

0.8

e-

I

)

0.6

.E \

t~ C

0.4

2,,

v J

/ _ _

~--~l

o.o 650

670

~(2Q) S

~ I 690

750

770

Fig. 1. Normalized fluorescence emission spectra from phycobilisome-depleted thylakoids of the wild-type (WT) and the mutant lacking PS II (PS I I - ) of Synechocystis sp. PCC 6803 at 295 and 20 K ( : mutant, 20 K; --: wild type, 20 K; - - - - : mutant, 295 K; . . . . : wild type, 295 K). Spectra were normalized to the intensity at 685 nm (295 K) or 725 nm (20 K). Excitation wavelength was 440 nm.

PS II at low temperatures as can be seen in Fig. 1 or by comparing Figs. 2a and 3a. All Chls have a weak long-wavelength vibronic emission band just to the red of their fluorescence maximum. Therefore, Chls within the wild-type and mutant lacking PS II with fluorescence maxima around 685 to 695 nm contribute some fraction of the measured emission in the 700 to 780-nm region. At high temperatures, this contribution is very significant since emission from F725 Chls is so weak (Figs. 1-3) and the fluorescence bands are broader. At lower temperatures, the yield of emission from F725 Chls is so large that the emission around 725 nm from other Chls is relatively very small. To assess the temperature dependence of the fluorescence contribution from the F725 Chls in PS I, the contribution of the F690 Chls in the mutant lacking PS II and the F685 and F695 Chls in the wildtype to the measured fluorescence spectra were estimated and subtracted. The 295 K spectrum was directly taken as an estimate of the qualitative emission spectrum from the F690 component at all temperatures. This assumes the contribution of F725 Chls to the 295 K emission is small and can be neglected as an approximation. The

best F690 fit was obtained for the mutant lacking PS II when the amplitude of the 295 K fluorescence spectrum (used for subtraction) remained constant within -+5% for emission spectra taken from 240 to 20 K, the range where 725-nm emission measurably increases compared to the 295 K spectrum. We conclude from this that the yield of fluorescence from F690 Chls in PS I is virtually independent of temperature. The only adjustments needed were to shift the 295 K spectrum to the red by approximately 2 nm at 220 K, 3 nm at 185 K and 4 nm for data taken from 170 K to 20 K. This shift presumably is caused by a liquidto-solid phase transition the sample goes through in the 220 to 180 K temperature region. In the case of the wild-type spectra, correction for the slightlY temperature-dependent PS II emission would be needed. Our approach of subtracting the 295 K spectra was, therefore, only a partial correction at best. The contribution of F685, F690 and F695 Chls to emission around 725 nm is small at temperatures below 180K. The temperature dependence of the 725-nm emission from the wild-type and mutant lacking PS II were observed to be the same within the error of the measurements at temperatures below 180 K.

80

I--~

1.0

I

25 K r \

t

I

)

1 t i

0.8

t

t

m e-

0.6

77 K

m t-

0.4

/

\

/

o

\

, \

m

/ o

\

0.2

120

0.0

i

670

650

K

, \

\

I

690

710

Wavelength (a)

--

730

750

770

(nm)

1.0

0.8 ta t-

t-

0.6

eq}

0.4

m

?. o w

i,

0.2

0.0

~'-650

670

690

710 Wavelength

730

750

770

(nm)

(b)

Fig. 2. Fluorescence emission spectra (a) and normalized fluorescence spectra (h) at selected temperatures from phycobilisomedepleted thylakoids of the mutant lacking PS II of Synechocystis sp. PCC 6803 ( : 220 K, - - - - : 120 K; - - - - : 77 K; ....

: 25 K). For (b), the spectra in (a) were normalized at the maximum of each spectrum. Excitation wavelength was 440 nm.

81 I

1.0

I

I

I

I

I

I

I

I

I

I

25 K

0.8 ° i

C

a)

0.6

C ,i

90 K

U e=

\

0.4

, \

¢J

j o

,

120 K

\

~\"

0.2

m

200 K

-r~l

0.0 650

i 670

~

i

r

6 90

i

710 Wavelength

i

i----r--~--

730

750

770

(nm)

(a)

I

I

I

I 200

I

I

...

I

/,

I

I

I

Ii)

/ \

0.6

I

K

/ \ i

I

0.4 O

2

,,

_~

0.2

J/f.' o.o ~--r --'--I650

670

I

I 690

I

I

I

710

730

Wavelength

(nm)

I

I 750

I

I 770

(b) Fig. 3. Fluorescence emission spectra (a) and normalized fluorescence spectra (b) at selected temperatures from phycobilisomedepleted thylakoids of the wild-type of Synechocystis sp. PCC 6803 ( : 200 K; - - - - : 120 K; - - - - : 90 K; . . . . : 25 K). For (b), the spectra in (a) were normalized at the maximum of each spectrum. Excitation wavelength was 440 nm.

82 The spectra resulting from subtraction of the adjusted 295 K emission data from the measured fluorescence spectra of the PS II-less mutant were taken as representative of the fluorescence spectra of F725 Chls vs. temperature. Each of these was fit at their central peak to a single gaussian. The area of each gaussian was taken as a measure of the relative yield of fluorescence as a function of temperature for F725 Chls as plotted in Fig. 4. The peak and width of these gaussians varied from 722 and 34 nm at 240 K to 726 and 22 nm at 20 K, respectively. The shift in peak location was characterized by a transition from 723 nm at 220 K to 726 nm at 180 K. This is also the temperature region where the sample undergoes a solid-to-liquid phase transition as indicated by the loss of cracks in the sample around 200 K. The narrowing of the 725-nm emission band with decreasing temperature was more gradual with the bandwidth remaining at about 22 nm after the sample reached approximately 60 K. The temperature dependence of the fluorescence yield at 725 nm was fit using the Arrhenius equation for temperature dependence of a rate

r Q

.

.

.

.

.

constant (Fig. 4) (Tusov et al. 1980, Tabbutt 1987): fluorescence yield at 725 nm (T) = (A + B e-e/ kr) -1 , where A and B are constants, T is temperature in kelvins, k is Boltzmann's constant and E is the activation energy. Assuming the model of lowenergy Chls transferring their excitations to the reaction center in PS I, the average result for the activation energy between the F725 Chls to P700 is E = 0.036 -+ 0.005 eV. This simple expression for fitting gave good results for all temperatures except the extremes. To examine the excitation transfer interactions between F725 Chls, the polarization of fluorescence at 760 nm was measured as a function of excitation wavelength for the mutant lacking PS II of S. 6803 at 77 K (Fig. 5). As excitations are transferred from chromophore to chromophore, the initial polarization of the exciting light gets lost, assuming the transition moments of the chromophores are not aligned. Fluorescence

I .

.

I

I

.

o o

G) O C O

E @ ,7

o 0

I

I

I

I

50

1 O0

150

200

Temperature

5 .....

i ........ 250

6 300

(K)

Fig. 4. Yield of fluorescence emission from F725 Chls as a function of t e m p e r a t u r e from phycobilisome-depleted thylakoids of the m u t a n t lacking PS II of Synechocystis sp. P C C 6803. Excitation wavelength was 440 nm. T h e equation used for fitting the data ( ) is described in the text. Values for this data set are: A = 0.2059, B = 14.59, E = 0.0291 eV. Also plotted for comparison is a fit of the t e m p e r a t u r e dependence of 735-nm fluorescence in spinach ( . . . . ) where E = 0.08 eV (taken from T u s o v et al. 1980).

83

E p.

0.4

I

I

I

I

I

I

O ¢O

0.3 ¢B p. u)

P O

0.2

,T

to

0.1

N ..,-

"6 a.

0.0 670.0

680.0

690.0 Excitation

I 700.0 Wavelength

710.0

720.0

Inm)

Fig. 5. Polarization of fluorescence at 760 nm as a function of excitation wavelength from phycobilisome-depletedthylakoids of the mutant lacking PS II of Synechocystis sp. PCC 6803 at 77 K. The solid line is an interpolated curve.

from F725 Chls has a polarization close to 0 when excited at wavelengths less than 690nm, and a polarization of about 0.4 when excited at wavelengths longer than 715 nm. A polarization value of 0.4 from a sample of randomly oriented molecules indicates that the fluorescence is almost completely polarized (Parker 1968). Therefore, the absorption dipole m o m e n t and emitting dipole are essentially of the same alignment. Measuring at emission wavelengths longer than 760 nm caused no changes in the polarization curve, demonstrating that the high polarization is not an artifact caused by scattered excitation light. Preliminary measurements at 295 K also indicate that when excited at wavelengths longer t h a n 715 nm, the emission from F725 Chls is highly polarized; thus, this polarization is not an artifact of low temperature measurements.

Discussion A dramatic t e m p e r a t u r e - d e p e n d e n t fluorescence yield from Chls in P S I with first excited singletstate energy levels near or below that of P700 is characteristic of all P S I complexes in green plants, algae and cyanobacteria (Karukstis and

Sauer 1983, Golbeck 1987, Scheer and Schneider 1988). Similar temperature dependence of fluorescence from low-energy BChls has also been observed in numerous photosynthetic bacteria ( K r a m e r et al. 1984, Van Grondelle et al. 1988, Trost and Blankenship 1989). This temperaturedependent emission is not a property of individual Chl molecules. The relative peak heights of the two fluorescence bands of Chl a in solution are observed to have little temperature d e p e n d e n c e ( G o e d h e e r 1964). The most studied low-energy Chl in P S I is in the Chl a / b pigment-protein complex L H C I responsible for the 735-nm emission from P S I in green plants ( H a w o r t h et al. 1983, Lam et al. 1984, Vainstein et al. 1989). Yet it is clear that neither L H C I nor Chl b is a necessary element to a photosystem that produces long-wavelength emission. Cyanobacteria do not have L H C I or Chl b, but contain a long-wavelength fluorescence band attributed to low-energy Chls ( G o e d h e e r 1964, Mohanty et al. 1972, Rijgersberg and Amesz 1980). Even in green plants, if L H C I is removed from PS I, antenna Chl a with t e m p e r a t u r e - d e p e n d e n t emission around 720 nm remains (Tabbutt 1987, Searle et al. 1988). W h a t e v e r its role, there appears to be a strong

84 evolutionary pressure to preserve a group of antenna Chls (or BChls) with energy levels near or just below that of the reaction center. Our observations support a model is which the low-energy antenna Chls of PSI in S. 6803 can transfer their excitations to P700 efficiently at 295 K. This is in keeping with similar models addressing the role of F735 and F720 Chls in PS I from green plants (Butler 1961, Satoh and Butler 1978b, Butler et al. 1979, Tusov et al. 1980). As the temperature is lowered, the population in the higher vibrational energy levels of the first excited state of F725 Chls decreases, diminishing the probability of excitation transfer from an excited F725 Chl to P700. In cyanobacterial PSI particles denatured down to a size of 65 Chls per P700 a similar temperature-dependence of 720nm emission is observed (Takahashi et al. 1982). This suggests that the low-energy Chls of cyanobacteria may indeed be located close to the reaction center to facilitate excitation transfer. The difference in the temperature dependence of long-wavelength fluorescence yield in S. 6803 (F725 Chls) and in green plant PSI (F735) (Fig. 4) is attributed to the difference in energy levels relative to P700. The activation energy for excitation transfer from F725 Chls to P700 in S. 6803 is 0.036eV in comparison to 0.08 eV for F735 Chls (Tusov et al. 1980). F725 Chls transfer excitations to P700 at lower temperatures because their energy level is closer to P700 than the F735 Chls. This may also explain why the lifetime of 720-nm emission from isolated cyanobacterial PSI was observed to be about 310 ps at 77 K (Wittmershaus 1987a) instead of 2 to 3 ns as observed for 735-nm emission from green plant PSI (Karukstis and Sauer 1983). The model of low-energy Chls transferring excitations to P700 is not without question. There is evidence that a temperature-dependent change in excitation transfer between low-energy Chls and P700 may account for only part of the change in fluorescence yield with temperature (Tabbutt 1987, Mukerji and Sauer 1989). Observation of temperature-dependent emission from C705 in isolated L H C I (Mukerji and Sauer 1990), where P700 is absent, also raises the question of whether low-energy Chls may transfer their excitations back to the other (high energy level) antenna Chls nearby instead of to

P700. While our data are consistent with the model of low-energy Chls transferring excitations (directly or indirectly) to P700, our results do not prove the existence of this transfer pathway. Figure 1 illustrates a proportionately larger shoulder at long wavelengths in the emission spectrum of the mutant lacking PS II in comparisons to the wild type. This difference is even more dramatic when comparing emission from green plant chloroplasts to isolated PS 1-200 particles (Mullet 1980, Mukerji and Sauer 1989). The relative increase in long wavelength emission at 295 K is probably a combination of the broader and slightly red-shifted emission from the F690 Chls of PS I and the emission from the F725 Chls. We have not yet been able to unambiguously separate the relative fluorescence contributions from F690 and F725 Chls to the overall emission spectrum in the region greater than 700 nm at 295 K. The lack of significant change in the yield of 690-nm fluorescence with temperature from the PS II-less mutant of S. 6803 (Fig. 3) is consistent with similar observations on fluorescence around 690 nm from PS 1-200 (Tusov et al. 1980, Mukerji and Sauer 1989) and C P I (Tabbutt 1987, Searle et al. 1988) preparations from green plants. In agreement with these results, timeresolved fluorescence measurements on isolated PSI particles from green plants (Tusov et al. 1980, Tabbutt 1987, Wittmershaus 1987a, Mukerji and Sauer 1989) and cyanobacteria (Wittmershaus 1987a, Wittmershaus et al. 1987b) at 295 and 77 K have shown little change in the lifetime of 690-nm fluorescence. Concurrent with the low fluorescence yield at room temperature, the 690-nm fluorescence lifetime of PSI is between 20 and 90ps (Karukstis and Sauer 1983, Holzwarth 1986, Van Grondelle and Sundstrom 1988). This fluorescence is characteristic of the bulk of antenna Chls in PSI (Mullet et al. 1980, Karukstis and Sauer 1983, Holzwarth 1986, Van Grondelle and Sundstrom 1988) and dominates the room temperature emission (Fig. 2). Their rapid loss of excitation by energy transfer is reflected in the short lifetime and low yield of fluorescence. This apparently is temperature independent from 295 to 20 K. Thus, temperature does not appear to radically alter the organization or energy levels of the Chls in-

85 volved in energy transfer. The red-shift of Chl energy levels and the narrowing of spectral linewidths observed with decreasing temperature therefore is not expected to dramatically change the rate of energy transfer out of the F690 Chls. The 4-nm shift of the emission peaks from 688 and 722 nm at 240 K to 692 and 726 nm below 180 K is attributed to an effect of the liquid-tosolid phase transition in the sample. In P S I from green plants, the slight increase in fluorescence lifetime of F690 Chls with decreasing temperature might be attributed to similar changes in linewidths or shifting energy levels (Mukerji and Sauer 1989, Wittmershaus 1987a). The highly polarized 760-nm fluorescence resulting from excitation at wavelengths longer than 715 nm demonstrates the existence of the long-wavelength absorbing and fluorescing Chls (F725) in S. 6803. A t these excitation wavelengths, excitation of F725 Chls occurs mainly by direct absorption of the exciting light. The high degree of polarization of fluorescence from F725 Chls indicates one or more of the following: 1. These Chls are aligned with each other. 2. T h e r e is a low probability of excitation transfer among the F725 Chls themselves. 3. T h e r e is only one F725 Chl per PS I. All of these possibilities are consistent with a small n u m b e r of low-energy Chls in P S I as indicated by the lack of any feature in the longwavelength tail of the absorption spectrum of the mutant lacking PS II of S. 6803 or PS 1-200 (Butler 1961, Satoh and Butler 1978b, Mullet et al. 1980, Mukerji and Sauer 1989), PSI-100 (Malkin et al. 1985, Tabbutt 1987), or L H C I (Mukerji and Sauer 1990) from green plants. The low polarization of 760-nm fluorescence resuiting from excitation at wavelengths shorter than 690 nm is attributed to diffusion of the initial polarization through extensive excitation transfer among the F690 Chls and finally, transfer to the F725 Chls. Our polarization results are qualitatively similar to those from measurements on low-energy BChls ( K r a m e r et al. 1984) or F735 Chls in higher plants ( G a r a b et al. 1976, Mukerji and Sauer 1990). In isolated L H C I at 278 K from green plants, a polarization of up to 0.3 was observed for 735-nm fluorescence from F735 Chls excited above 700 nm (Mukerji and Sauer

1990). Highly polarized emission from F735 Chls was also observed on 'measurements of oriented spinach chloroplasts (Garab et al. 1976) indicating that no excitation transfer occurs between these Chls a n d / o r that they are highly oriented. It therefore appears that another c o m m o n property of all low-energy Chls and BChls is a highly polarized emission when directly excited.

Acknowledgements Our special thanks to D r T o m G r o y for use of his H e refrigerator, to D r Bob Blankenship for use of his lab in preparing samples, and to Ms Cathy Madsen for cultivating the cells. This is publication #88 from the Arizona State University Center for the Study of Early Events in Photosynthesis. The Center is funded by U.S. Department of Energy grant #DE-FG88ER13969 as a part of the U S D A / D O E / N S F Plant Science Centers Program. This work was supported in part by the Center for the Study of Early Events in Photosynthesis and the Arizona State University Laser Facility.

References Bassi R and Simpson D (1987) Chlorophyll-protein complexes of barley Photosystem I. Eur J Biochem 163:221-231 Butler WL (1961) A far-red absorbing form of chlorophyll, in vivo. Archiv Biochem Biophys 93:413-422 Butler WL (1978) Energy distribution in the photochemical apparatus of photosynthesis. Ann Rev Plant Physiol 29: 345-378 Butler WL, Tredwell CJ, Malkin R and Barber J (1979) The relationship between the lifetime and yield of the 735 nm fluorescence of chloroplasts at low temperatures. Biochim Biophys Acta 545:309-315 Cantrell A and Bryant DA (1987) Molecular cloning and nucleotide sequence of the psaA and psaB genes of the cyanobacterium Synechococcus sp. PCC 7002. Plant Mol Biol 9:453-468 Garab GyI and Breton J (1976) Polarized light spectroscopy on oriented spinach chloroplasts fluorescence emission at low temperature. Biochem Biophys Res Com 71: 10951102 Garnier J, Maroc J and Guyon D (1986) Low-temperature fluorescence emission spectra and chlorophyll-proteincomplexes in mutants of Chlamydomonas reinhardtii: Evidence for a new chlorophyll-a-protein related to Photosystem I. Biochim Biophys Acta 851:395-406 Garnier J, Maroc J and Guyon D (1987) Characterization of

86 new strains of photosynthesis mutants of Chlamydomonas reinhardtii IV impaired excitation transfer in three low fluorescent mutants. Plant Cell Physiol 28:1117-1131 Goedheer JC (1964) Fluorescence bands and chlorophyll a forms. Biochem Biophys Acta 88:304-317 Golbeck JH (1987) Structure, function and organization of the Photosystem I reaction center complex. Biochim Biophys Acta 895:167-204 Haworth P, Watson JL and Arntzen CJ (1983) The detection, isolation and characterization of a light-harvesting complex which is specifically associated with Photosystem I. Biochim Biophys Acta 724:151-158 Holzwarth A R (1986) Excited state kinetics of chlorophyll antenna pigments. In: Staehelin LA and Arntzen CJ (eds) Encyclopedia of Plant Physiology, Vol 19, pp 299-309. Springer-Verlag, New York Huang C and Berns DS (1983) Partial characterization of six chlorophyll a-protein complexes isolated from a blue-green alga by a nondetergent method. Arch Biochem Biophys 220:145-154 Karukstis KK and Sauer K (1983) Fluorescence decay kinetics of chlorophyll in photosynthetic membranes. J Cell Biochem 23:131-158 Kramer HJM, Pennoyer JD, Van Grondelle R, Westerhuis WHJ, Niederman R A and Amesz J (1984) Low-temperature optical properties and pigment organization of the B875 light-harvesting bacteriochlorophyll-protein complex of purple photosynthetic bacteria. Biochim Biophys Acta 767:335-344 Lam E, Ortiz W, Mayfield S and Malkin R (1984) Isolation and characterization of a light-harvesting chlorophyll a/bprotein complex associated with Photosystem I. Plant Physiol 74:650-655 Malkin R, Ortiz W, Lam R and Bonnerjea J (1985) Structural organization of thylakoid membrane electron transfer complexes: Photosystem I. Physiol V6g 23:619-625 Mohanty P, Zilinskas-Braun B, Govindjee and Thornber JP (1972) Chlorophyll fluorescence characteristics of system I chlorophyll a-protein complex and system II particles at room and liquid nitrogen temperatures. Plant Cell Physiol 13:81-91 Mukerji I and Sauer K (1989) Temperature-dependent steady-state and picosecond kinetic fluorescence measurements of a Photosystem I preparation from spinach. In: Briggs WH (ed) Photosynthesis, pp 105-122. A.R. Liss, New York Mukerji I and Sauer K (1990) A spectroscopic study of a Photosystem I antenna complex. In: Baltscheffsky M (ed) Current Research in Photosynthesis, Vol 2, pp 321-324. Kluwer Academic Publishers, Dordrecht Mullet JE, Burke JJ and Arntzen CJ (1980) Chlorophyll proteins of Photosystem I. Plant Physiol 65:814-822 Nechushtai R, Nourizadeh SD and Thornber JP (1986) A reevaluation of the fluorescence of the core chlorophylls of Photosystem I. Biochim Biophys Acta 848:193-200 Parker CA (1968) Photoluminescence of Solutions. Elsevier Publishing Co, New York Rijgersberg CP and Amesz J (1980) Fluorescence and energy transfer in phycobiliprotein-containing algae at low temperature. Biochim Biophys Acta 593:261-271

Satoh K and Butler WL (1978a) Competition between the 735 nm fluorescence and the photochemistry of Photosystem I in chloroplasts at low temperature. Biochim Biophys Acta 502:103-110 Satoh K and Butler WL (1978b) Low temperature spectral properties of subchloroplast fractions purified from spinach. Plant Physiol 61:373-379 Scheer H and Schneider S (eds) (1988) Photosynthetic Light Harvesting Systems. W. de Gruyter, Berlin Searle GFW, Tamkivi R, van Hoek A and Schaafsma TJ (1988) Temperature dependence of antennae chlorophyll fluorescence kinetics in Photosystem I reaction centre protein. J Chem Soc Faraday Trans 2, 84:315-327 Tabbutt S (1987) Spectroscopic studies of energy transfer in photosynthetic reaction centers of higher plants. Thesis-U. of California, Berkeley, CA Takahashi T, Hiroyuki K and Katoh S (1982) Multiple forms of chlorophyll-protein complexes from a thermophilic cyanobacterium Synechococcus sp. Arch Biochem Biophys 219:209-218 Thornber JP, Trosper TL and Strouse CE (1978) Bacteriochlorophyll in vivo: Relationship of spectral forms to specific membrane components. In: Clayton RK and Sistrom WR (eds) The Photosynthetic Bacteria, pp 133-160. Plenum, New York Trost JT and Blankenship RE (1989) Isolation of a photoactive photosynthetic reaction center-core antenna complex from Heliobacillus mobilis. Biochemistry 28:9898-9904 Tusov VB, Korvatovskii BN, Pashchenko V Z and Rubin LB (1980) Nature of 735-nm fluorescence of chloroplasts at room and low temperatures. Doklady-biophysics [Eng. trans.] 252:112-115 Vainstein A, Peterson CC and Thornber JP (1989) Lightharvesting pigment-proteins of Photosystem I in maize. J Biol Chem 264:4058-4063 Van Grondelle R and Sundstrom V (1988) Excitation energy transfer in photosynthesis. In: Scheer H and Schneider S (eds) Photosynthetic Light Harvesting Systems, pp 403438. W. de Gruyter, Berlin Van Grondelle R, Bergstrom H, Sundstrom V, Van Dorssen RJ, Vos M and Hunter CN (1988) Excitation energy transfer in the light-harvesting antenna of photosynthetic purple bacteria: The role of the long-wavelength absorbing pigment B896. In: Scheer H and Schneider S (eds) Photosynthetic Light Harvesing Systems, pp 519-530. W. De Gruyter, Berlin Vermaas WFJ, Charit6 J and Shen G (1990) Glu-69 of the D2 protein in Photosystem II is a potential ligand to Mn involved in photosynthetic oxygen evolution. Biochemistry 29:5325-5332 Vermaas WFJ, Ikeuchi M and Inoue Y (1988) Protein composition of the Photosystem II core complex in genetically engineered mutants of the cyanobacterium Synechocystis sp. PCC 6803. Photosynth Res 17:97-113 Vermaas WFJ, Williams JGK and Arntzen CJ (1987a) Sitedirected mutations of two histidine residues in the D2 protein inactivate and destabilize Photosystem II in the cyanobacterium Synechocystis 6803. Z Naturforsch 42c: 762-768 Vermaas WFJ, Williams JGK, Chrisholm DA and Arntzen C

87 (1987b) Site-directed mutagenesis in the Photosystem II gene psbD, encoding the D2 protein. In: Biggins J (ed) Progress in Photosynthesis Research, Vol. 4, pp 805-808. Martinus Nijhoff/Dr W. Junk Publishers, Dordrecht Wittmershaus BP (1987a) Measurements and kinetic modeling of picosecond time-resolved fluorescence from Photosystem I and chloroplasts. In: Biggins J (ed) Progress in Photosynthesis Research, Vol 1, pp 75-82. Martinus Nijhoff/Dr W. Junk Publishers, Dordrecht Wittmershaus BP, Berns DS and Huang C (1987b) Pico-

second time-resolved fluorescence from detergent-free Photosystem I particles. Biophys J 52:829-836 Wollman F-A and Bennoun P (1982) A new chlorophyllprotein complex related to Photosystem I in Chlamydomonas reinhardii. Biochim Biophys Acta 680:352-360 Yu J and Vermaas WFJ (1990) Transcript levels and synthesis of Photosystem II components in cyanobacterial mutants with inactivated Photosystem II genes. Plant Cell 2: 315322