New Insights into the Mechanism of Uphill Excitation Energy Transfer

2 days ago - The uphill excitation energy transfer (EET) from the core antenna (LH1) to the reaction center (RC) of purple photosynthetic bacteria was...
1 downloads 0 Views 1MB Size
Subscriber access provided by YORK UNIV

Biophysical Chemistry, Biomolecules, and Biomaterials; Surfactants and Membranes

New Insights into the Mechanism of Uphill Excitation Energy Transfer from Core Antenna to Reaction Center in Purple Photosynthetic Bacteria Liming Tan, Jie Yu, Tomoaki Kawakami, Masayuki Kobayashi, Peng Wang, Zheng-Yu Wang-Otomo, and Jian-Ping Zhang J. Phys. Chem. Lett., Just Accepted Manuscript • Publication Date (Web): 04 Jun 2018 Downloaded from http://pubs.acs.org on June 4, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

New Insights into the Mechanism of Uphill Excitation Energy Transfer from Core Antenna to Reaction Center in Purple Photosynthetic Bacteria Li-Ming Tan,† Jie Yu,† Tomoaki Kawakami,‡ Masayuki Kobayashi,§ Peng Wang,† Zheng-Yu Wang-Otomo,‡ Jian-Ping Zhang†,* †

Department of Chemistry, Renmin University of China, Beijing 1000872, P. R. China ‡

§

Faculty of Science, Ibaraki University, Mito 310-8512, Japan

Institute of National Colleges of Technology, Ariake College, Omuta, Fukuoka 836-8585, Japan

AUTHOR INFORMATION Corresponding Author *To whom correspondence should be addressed. (J.-P. Zhang) E-mail: [email protected] Tel: +86-10-62516604; Fax: +86-10-62516444

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 28

ABSTRACT

The uphill excitation energy transfer (EET) from the core antenna (LH1) to the reaction center (RC) of purple photosynthetic bacteria was investigated at room temperature by comparing the native LH1-RC from Thermochromatium (Tch.) tepidum with the hybrid LH1-RC from a mutant strain of Rhodobacter (Rba.) sphaeroides. The latter protein with chimeric Tch-LH1 and Rba-RC exhibits a substantially larger RC-to-LH1 energy difference (∆E = 630 cm−1) of 3-fold thermal energy (3kBT). The spectroscopic and kinetics results are discussed on the basis of the newly reported high-resolution structures of Tch. tepidum LH1-RC, which allow us to propose the existence of a passage formed by LH1 BChls that facilitates the LH1→RC EET. The semilogarithmic plot of the EET rate against ∆E was found to be linear over a broad range of ∆E, which consolidates the mechanism of thermal activation as promoted by the spectral overlap between the LH1 fluorescence and the special pair absorption of RC.

TOC GRAPHICS

KEYWORDS: Excitation energy transfer, Kinetics, Light-harvesting, Photosynthesis, Thermal activation, Time resolved spectroscopy.

ACS Paragon Plus Environment

2

Page 3 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

The photosynthesis of purple bacteria begins with light harvesting via specialized pigmentprotein complexes embedded in the intracytoplasmic membranes (ICMs), which can be categorized into two major groups: The light-harvesting complex 1 (LH1) encircling the reaction center (RC), which is collectively called after the core complex (LH1-RC), and the peripheral light-harvesting complex 2 (LH2), which may have spectral variants LH3 and/or LH4 in minority.1-3 Upon light absorption, LHs transfer the electronic excitation in a cascaded manner, i.e. LH2→LH1→RC, to initiate the primary electrochemical reaction in the RC. Among the processes of excitation energy transfer (EET), the step LH1→RC with a timescale of (30−65) ps is known to be rate limiting, because all of the other intra- and inter-complex EET processes are at least one order of magnitude faster.4-6 At room temperature, the rate of LH1→RC EET is (45 ps)−1 for the core complexes of Rhodobacter (Rba.) sphaeroides,6 despite an uphill energy barrier of ∆E = 200 cm−1 as defined by the excited state energy of the special pair bacteriochlorophylls (P-BChl2) of RC minus that of the LH1-BChls. The EET reaction proceeding at a few tens of picoseconds can be explained, within the theoretical framework of Förster resonance energy transfer, by thermal activation in view of the comparable size of thermal energy (kBT = 210 cm−1).7-8 However, for LH1-RCs from various bacterial species with a ∆E larger than thermal energy, the LH1→RC EET rate does not drop significantly, e.g., it is (55 ps)−1 for Roseospirillum (Rss.) parvum with ∆E = 290 cm−1,9 (65 ps)−1 for Thermochromatium (Tch.) tepidum with ∆E = 370 cm−1,10-11 and (65 ps)−1 for the strain 970 with ∆E = 430 cm−1.12 Extensive experimental and theoretical efforts have been directed to understand the so-called LH1→RC uphill EET processes, among which a recent work examined the temperature dependence of the EET rate for the native and chemically modified LH1-RCs

ACS Paragon Plus Environment

3

The Journal of Physical Chemistry Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 28

from Tch. tepidum, and evaluated the activation and reorganization energy of the EET reaction.13 For the LH1-RC with ∆E = 350 cm−1 from Blastochloris (Blc.) viridis at cryogenic temperature, a theoretical work proposed the superexchange mechanism, i.e. P* virtually mediated the uphill EET reaction without being populated.8,14 Despite the efforts, the detailed mechanism of the uphill EET reaction remains to be further elucidated. The crystallographic structures of Tch. tepidum LH1-RC have been recently resolved to the resolutions of 3.0 Å15 and 1.9 Å16. These structures reveal that the LH1 complex is assembled by 16 α/β-heterodimers forming a complete ring, and that the 32 BChls bound to the α/β-subunits are 37.4–49.0 Å from the central P-BChl2. Meanwhile, the structure of Blc. viridis LH1-RC has been recently resolved at 2.9 Å resolution, showing also a complete ring assembly with 17 α/βheterodimers binding 34 BChl b molecules and additional 16 γ-polypeptides.17 The additional structural information on bacterial LH1-RCs will promote the mechanistic understandings towards the LH1→RC uphill EET reaction. On the other hand, a hybrid core complex consisting of the Tch. tepidum LH1 and Rba. sphaeroides RC has been recently prepared using a heterologous expression system of Rba. sphaeroides18 (hereafter denoted as hybrid LH1-RC). In addition, it has been noted that the full-length Tch. tepidum LH1 was also heterologously expressed using a similar system, but the product appeared to be in a LH1-only form.19 Among others, a striking feature of the hybrid LH1-RC is the sizable ∆E (630 cm−1), i.e. about three folds of room-temperature thermal energy (3kBT). Thus, on the basis of updated structural information and new EET kinetics results, it is of considerable interest to explore the mechanism underlying the LH1→RC EET reaction with an uphill barrier substantially larger than those of the previously reported native strains.

ACS Paragon Plus Environment

4

Page 5 of 28

The present work has comparatively investigated, by the use of femtosecond transient absorption spectroscopy and at room temperature (∼296 K), the LH1→RC EET processes of the native LH1-RCs from Tch. tepidum and the hybrid LH1-RCs from the genetically engineered strain of Rba. sphaeroides in ‘open’ (reduced) state. The LH1-only preparation from the mutant strain was used as the reference for deriving the rate and efficiency of LH1→RC EET reaction.20,21 Details of the femtosecond time-resolved absorption spectrometer were described elsewhere.11,22 Further details of sample preparation and spectroscopic experiments can be found

1.0

Qy

915

Absorption LH1-only Hybrid LH1-RC Native LH1-RC Rba. sphearoides RC Tch. tepidum RC Flurorescence LH1-only Hybrid LH1-RC Native LH1-RC

0.8 0.6 0.4 Qx

0.2 865 885

0.0 600

700

800

900

933

Flurorescence intensity (a.u.)

in Supporting Information.

Absorbance (a.u.)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

1000

Wavelength (nm)

Figure 1. Near-infrared absorption (solid) and fluorescence (dashed) spectra of LH1-only (black), native LH1-RC (blue) and hybrid LH1-RC (red) complexes at room temperature. Absorption spectra are normalized at the Qy maxima. Fluorescence spectra were photoexcited at 595 nm with same A595. Dash dot curves are the special pair (P) absorption spectra of Rba. sphaeroides RC (red) and Tch. tepidum RC (blue) shown for reference. Arrows along abscissa axis indicate the photoexcitation wavelengths for time-resolved measurements. As seen in Figure 1, the LH1-only, native LH1-RC and hybrid LH1-RC complexes exhibit nearly identical LH-Qy absorption (fluorescence) peaking at 915 nm (933 nm) with a half bandwidth of 46 nm (46 nm). (See Figure S1 for the absorption spectra extending to visible spectral region.) The spectral similarity suggests that these proteins have structurally similar circular aggregates of LH1-BChls. However, the hybrid and the native LH1-RCs differ in the RC

ACS Paragon Plus Environment

5

The Journal of Physical Chemistry Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 28

absorption spectra, the former exhibits significant blue shifted P absorption (865 vs 885 nm), and slight red shifted accessary BChl (B) absorption (805 vs 800 nm). These differences are in accord with the documented RC absorption spectra of wild-type Tch. tepidum23 and Rba. sphaeroides 2.4.124. Nevertheless, the significant difference of P energy (260 cm−1) allows us to examine, without varying the LH1 energy, the influence of RC-to-LH1 energy barrier (∆E) on the dynamics of LH1→RC EET (also referred to as RC-trapping). Under similar actinic conditions, the LH1-only, native LH1-RC and hybrid LH1-RC complexes show distinctly different LH1-fluorescence intensity (Figure 1): With reference to the LH1-only complexes, the native and the hybrid LH1-RCs are quenched by 75% and 57%, respectively, which is considered to be mainly due to the RC-trapping. The precise rate and efficiency of LH1→RC EET, however, need to be determined by the use of kinetics parameters, for which the intrinsic Qy-state lifetime of LH1-BChls (τ0) without involving RC-trapping is indispensable. In the present work, the LH1-only complex is of the same kind as those of the pair of LH1-RCs, which would allow us to derive a τ0 pertinent to the evaluation of the EET rate and efficiency.

ACS Paragon Plus Environment

6

Page 7 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

Figure 2. Transient absorption spectra at selected delay times for (A, B) LH1-only, (C, D) native LH1-RC and (E, F) hybrid LH1-RC complexes in ‘open’ state under photoexcitation at 880 nm or 920 nm. For clarity, all spectra are shifted vertically, and spectra at 500 ps and 1800 ps are magnified by a factor of 3 (light grey). In time resolved spectroscopic measurements, the pigment-protein complexes were excited at the blue (880 nm) and the red (920 nm) edges of the LH1-Qy band, corresponding to a photonenergy separation of 490 cm−1. In view of the P-absorption maximum/bandwidth of 885 nm/70 nm and 865 nm/60 nm for the native and the hybrid RCs, respectively, the red-edge excitation applied mainly to LH1-BChls, while the blue-edge excitation could also ignite the P-BChl2 with a P/LH1 extinction ratio estimated to be 5%. The difference absorption spectra in Figure 2(A–F) at selected delay time (∆t) can be characterized as followings. In each panel of Figure 2, the positive excited state absorption (ESA) appearing to the shorter wavelength side are known as the excitonic absorption of LH1-BChls, and the accompanied negative bands are originated from the bleaching of ground state absorption (BLC). For each pigment-protein preparation, both ESA and BLC maxima upon 880 nm excitation are ∼15 nm shorter than those upon 920 nm excitation as verified by comparing the transients in the initial phase (e.g. ∆t = 0.00 ps). The λEx-dependence of ESA and BLC can be ascribed to the inhomogeneous broadening of LH1-Qy absorption, reflecting the variation of site energy of the LH1-BChls. At ∆t = 500 ps, the ESA and BLC features of the pair of LH1-RCs decayed almost completely (Figure 2C−F), whereas those of LH1 remained (Figure 2A, B). The shorter-lived LH1-excitations of LH1-RCs are resultant from RC-trapping.

ACS Paragon Plus Environment

7

The Journal of Physical Chemistry Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 28

Figure 3. Dynamic shift of the zero-crossing wavelengths (λISO) for (A) LH1-only, (B) native LH1-RC and (C) hybrid LH1-RC complexes in ‘open’ state under excitation wavelengths (λEx) of 880 nm and 920 nm. In Figure 2, each set of transient spectra shows ultrafast spectral shift in the early phase as highlighted by a pair of spectral curves in red. Such dynamic shift can be characterized by the time evolution of the zero-crossing wavelength jointing the ESA and BLC features,30 which will be referred to as the isosbestic wavelength (λISO). Figure 3 shows the λISO∼∆t kinetics, which can be characterized below. (i) At ∆t = 0.00 ps, each of the proteins shows ~20 nm shorter λISO under 880 nm excitation than that under 920 nm excitation. Afterwards (>0.1 ps), the pair of λISO∼∆t kinetics in each panel shift in opposite directions, but they never converge completely. The decay time constants of all of the λISO∼∆t kinetics are in the range of 0.1∼0.2 ps, which is more than 2 orders of magnitude shorter than the typical time constant of RC-trapping (∼50 ps)6,9-12. (ii)

ACS Paragon Plus Environment

8

Page 9 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

Under photoexcitation at 880 nm, we see a distinct difference between the pair of LH1-RCs, i.e. the λISO of hybrid LH1-RC stabilized to a wavelength (902 nm, Figure 3C) slightly shorter than the native LH1-RC (907 nm, Figure 3B). Finally, we note that the dynamic spectral shift can also be characterized by plotting the maximal wavelength of BLC (λBLC) against ∆t, and the λBLC∼∆t kinetics evolved similarly as the λISO∼∆t kinetics for all of the pigment-protein complexes (cf. Figure S2). Nevertheless, the subpicosecond spectral equilibration, characterized by the λISO ∼∆t and λBLC∼∆t kinetics, is to be explained in terms of excitation equilibration among LH1-BChls. Table 1. Decay time constants (τi, i = 1–3) derived by fitting the kinetics curves of different complexes to a tri-exponential model function (cf. Figure S3). At an excitation wavelength (λEx), the pair of kinetics curves probed at different wavelengths (λPr) were simultaneously fitted with τ1 fixed at 0.2 ps. See text for more details for curve fitting and for the estimation of RC-trapping time constant (τtr) and efficiency (η). Complexes LH1-only

λEx (nm) 880 920

Native LH1-RC

880 920

Hybrid LH1-RC

880 920

λPr (nm) 890/920 890/920 (Ave) 890/920 890/920 (Ave) 890/920 890/920 (Ave)

τ1 (ps) 0.15 0.15 0.15 0.15 0.15 0.15 0.15 0.15 0.15

τ2 (ps) τ3 (ps) τtr (ps) 1.3±0.0 720±17 1.6±0.0 691±14 1.5±0.0 705±16 1.5±0.0 67±5 74±6 1.3±0.0 67±5 74±6 1.4±0.0 67±5 74±6 2.5±0.0 140±7 175±12 1.9±0.0 126±8 153±11 2.2±0.0 133±7 164±12

η (%)

91 91 91 80 82 81

We plotted, from the time-resolved spectra as shown in Figure 2, the kinetics curves at λPr = 890 nm (ESA) and 920 nm (BLC) for three of the pigment-protein complexes (cf. Figure S3). Fitting the kinetics to a tri-exponential model function yielded the parameters as listed in Table 1. In view of the timescale of dynamic shifts of λBLC and λISO (∼0.15 ps; Figure S2, Figure 3), we fixed τ1 at 0.15 ps in the global fitting of the kinetics curves at 890 nm and 920 nm, because such ultrafast spectral shift was probed at a fixed probing wavelength. Here, we note that τ1 at 0.1–0.2

ACS Paragon Plus Environment

9

The Journal of Physical Chemistry Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 28

ps did not cause substantial difference in the resultant fitting parameters. The second exponential component, decayed with an average time constant of τ2 of ∼2 ps, is due to the rapid depopulation of LH1-excitation owing to the excitation annihilation, because this component also presented in the case of LH1-only without involving the processes of RC trapping and/or detrapping. The long lived third component (τ3) of the LH1-only complex is substantially shortened in the cases of LH1-RC, which is ascribed to the RC-trapping effect. Within experimental errors, the decay time constants, τ3, are nearly independent on λEx. To derive the rate and efficiency of LH1→RC EET, the decay time constant τ3 of LH1-only complex is taken as τ0 (∼705 ps), which is in agreement with the values reported for the RC-less LH1s from Rba. sphaeroides (630–800 ps)25-28 and that of the chromatophores of Rba. sphaeroides strain PM-8 dpl (1000 ps)29. On the other hand, the decay time constants τ3 of the pair of LH1-RCs are taken as τobs. Table 1 lists the LH1→RC EET time constants (τtr) derived ିଵ ିଵ with ߬୲୰ = ߬୭ୠୱ − ߬଴ିଵ . It is seen that, from the native to the hybrid LH1-RCs, the τtr extends ିଵ ିଵ from 74±6 ps to 164±12 ps. Accordingly, the RC-trapping efficiency (ߟ୲୰ = ߬୲୰ /߬୭ୠୱ ) decreases

from 91% to 81%. Interestingly, the RC-trapping process of the hybrid LH1-RC is substantially retarded with reference to the native LH-RC. In addition, for each LH1-RC complex, the 490 cm−1 difference of excitation photon energy made rather small difference in the EET rates and efficiency. We now come to discuss the issue of uphill LH1→RC EET of bacterial photosynthesis. A recent work examined the temperature dependence of the EET rates for the LH1-RCs having same P energy but different LH1 energy from Tch. tepidum, and found 2–3 fold increased reorganization energy for the chemically modified LH1-RCs with reference to the native one.13

ACS Paragon Plus Environment

10

Page 11 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

In the present work the hybrid LH1-RC binds Ca2+ similar as the native LH1-RC does.18 This together with the nearly identical LH1-Qy bands indicate similar protein rigidity and BChl organization, and hence similar LH1 reorganization energy of the two different types of LH1RCs. Most importantly, the present work compares the EET processes at room temperature for the native and the hybrid LH1-RCs differed in the ∆E by 260 cm−1. Note that the hybrid LH1RCs bear a ∆E as large as 630 cm−1, which is to our knowledge the largest ever reported. From the native to the hybrid LH1-RCs with increasing ∆E from 370 cm−1 to 630 cm−1, the trapping time constant τtr prolongs substantially from 74 ps to 164 ps. However, for each type of LH1-RC, the blue- and red-edge photoexcitation yielded nearly negligible difference in the trapping rate and efficiency (Table 1). In view of the rather weak photon-energy influence and the distinct effect of ∆E on τtr, we envisage the existence of a passage formed by LH1-BChls, capable of preferentially transferring LH1-excitation to P-BChl2. A similar vision was previously suggested in refs 7 and 15.

Figure 4. (A) Schematic orientation of BChl cofactors of Tch. tepidum LH1-RC (PDB ID: 3WMM).15 Green bars represent the Qy transition dipole moments of BChls along with Mg atoms (red dots). (B) Quadratic geometric factor (κ2) as a function of the serial number of LH1BChls obtained in the present work, and transition dipole-transition dipole coupling strengths (VDA) between individual LH1-BChls and P-BChl2 taken from ref. 7. The shaded regions in (A) highlight the LH1-BChls having relatively large κ2.

ACS Paragon Plus Environment

11

The Journal of Physical Chemistry Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 28

Figure 4A shows the relative orientation of the Qy transition dipole moments of BChls (ࣆ) in the LH1-RC complex of Tch. tepidum. It is seen that the LH1-BChls are not oriented equivalently with reference to the P-BChl2. According to Förster’s EET mechanism,31 the strength of transition dipole-transition dipole interaction (VDA) is proportional to the mutual orientation between the donor (D) and the acceptor (A) BChls as measured by the geometric ෡ ൯൫ࣆ ෡ ൯. On the basis of the LH1-RC structure,15 we calculated factor ߢ = ࣆ ෝ୅ ∙ ࣆ ෝ ୈ − 3൫ࣆ ෝୈ ∙ ࡾ ෝ୅ ∙ ࡾ

κ for individual LH1-BChls, and plot the κ2 against the serial number of LH1-BChls (Figure 4B). It is seen that BChl10∼14 and BChl26∼30 bound to (α/β)5∼7 and (α/β)13∼15, respectively, exhibit relatively large κ2 than the other LH1-BChls, implying that these groups of BChls are relatively more efficient donors for the LH1→RC EET. As illustrated in Figure 4B, the variation of κ2 over ଶ individual LH1-BChls is in good agreement with that of the ܸୈ୅ , the strength of individual LH1-

BChls coupling with the P-BChl2 calculated by taking into account the excitonic nature of PBChl2.7 Here, we note that the calculation in ref. 7 was based on a hexadecameric model of LH1RC, where 32 BChls were assumed to be organized in a circular rather than an elliptical ring as in the case of Tch. tepidum LH1-RC. Therefore, the orientational preference for EET found for Tch. tepidum LH1-RC should hold for other types of LH-RCs such as those from Blc. viridis17 and Rhodopseudomonas (Rps.) palustris32 (cf. Figure S4). The presence of preferential BChls for the LH1→RC EET can be related to the subpicosecond excitation equilibration among LH1-BChls: Following the primary exciton relaxation with a time constant of ∼150 fs, excitation hopping in a LH1 ring takes place on an even shorter timescale of