Spermine is a potent modulator of proton transport through LHCII

Accepted Manuscript Title: Spermine is a potent modulator of proton transport through LHCII Author: Theodoros Tsiavos Nikolaos E. Ioannidis Achilleas Tsortos Electra Gizeli Kiriakos Kotzabasis PII: DOI: Reference:

S0176-1617(15)00016-4 http://dx.doi.org/doi:10.1016/j.jplph.2015.01.010 JPLPH 52115

To appear in: Received date: Revised date: Accepted date:

13-9-2014 18-1-2015 20-1-2015

Please cite this article as: http://dx.doi.org/10.1016/j.jplph.2015.01.010 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Spermine is a potent modulator of proton transport through LHCII

Theodoros Tsiavos¶, Nikolaos E. Ioannidis¶, Achilleas Tsortos§, Electra Gizeli¶§ and

¶

ip t

Kiriakos Kotzabasis¶* Department of Biology, University of Crete, Voutes University Campus, GR-70013

§

cr

Heraklion, Crete, Greece.

Institute of Molecular Biology and Biotechnology, FORTH, GR-70013, Heraklion, Crete,

an

us

Greece

*Corresponding author: Prof. Kiriakos Kotzabasis, Department of Biology, University of Crete, Voutes University Campus, GR-70013 Heraklion, Crete, Greece.

M

e-mail: [email protected] Tel: +30 2810 394059

te

d

Fax: +30 2810 394408

Ac ce p

Abbreviations: LHCII, light-harvesting complex II; Chl, chlorophyll; PAs, polyamines; Spm, spermine; Spd, spermidine; Put, putrescine; ΔpH, thylakoid proton gradient; kH+, proton permeability; qE, ΔpH-dependent quenching of chlorophyll fluorescence; β-DM, dodecyl-βD-maltoside; PTS, 8-hydroxypyrene-1,3,6-trisulfonic acid; Val, valinomycin; H+, protons; PC, phosphatidylcholine.

Keywords: liposomes; light harvesting complex II; aggregation; spermine; proton permeability

1 Page 1 of 21

Abstract The effect of spermine on proton transport across large unilamellar liposomes containing incorporated complexes of the PSII antenna has been studied with the application of a pHsensitive dye entrapped inside the vesicles. Both monomeric LHCbs and trimeric LHCII

ip t

increased the permeability of proteoliposomes to protons when in a partly aggregated state within the lipid membrane. We have previously shown that a spermine-induced

cr

conformational change in LHCII results in its aggregation and ultimately in the enhancement of excitation energy as heat (qE). In this paper, spermine-induced aggregation of LHCII was

us

found to facilitate proton transport across the proteoliposomes, indicating that a second protective mechanism (other than qE) might exist and might be regulated in vivo by

Ac ce p

te

d

M

an

polyamines when photosynthesis is saturated in excess light.

2 Page 2 of 21

1. Introduction The chloroplast thylakoid membrane of photosynthetic organisms is the site where light energy is primarily converted. An extensive system of membrane-associated light-harvesting pigment-protein complexes (LHCs) serve as the antenna that absorbs and delivers solar

ip t

energy to the reaction centres of photosystems II (PSII) and I (PSI), where primary charge separation takes place. The two photosystems energize the electron transfer chain (ETC) from

cr

water to NADP+ which is coupled with a proton flow across the photosynthetic membrane. The accumulation of these protons in the inter-thylakoid membrane space (lumen) generates a

us

transmembrane gradient of protons, called the proton motive force (pmf) which comprises two components, the proton concentration gradient (ΔpH) and the membrane potential (Δψ). Pmf is converted into ATP through ATP synthase according to the chemiosmotic hypothesis

an

(Mitchell, 1966; Ioannidis and Kotzabasis, 2014). In addition to its role in driving ATP synthesis, the ΔpH component of pmf acts as a key signal in regulating energy dependent quenching (qE) in the photosynthetic antenna (Wraight and Crofts, 1970; Briantais et al.,

M

1979; Muller et al., 2001).

qE is the major component of non-photochemical chlorophyll fluorescence quenching (NPQ),

d

a photoprotective mechanism that dissipates the excess absorbed light energy as heat within

te

the PSII antenna (LHCII), preventing photoinhibition. Under light-saturating conditions, overaccumulation of protons in the lumen occurs and a high ΔpH is formed. Low lumen pH

Ac ce p

changes LHCII antenna conformation/organization activating the quenching pigment(s). qErelated spectroscopic signatures in vivo have been correlated with those observed when the purified LHCII adopt quenched states upon aggregation in vitro (Horton et al., 1991; Ruban et al., 1991; 1993a; b; 1998; Miloslavina et al., 2008; Ballottari et al., 2010). In addition to the protective energy dissipation, a physiological role of LHCII aggregation in controlling the ion fluxes across the thylakoid membrane under high-light conditions has also been suggested (Wardak et al., 2000; Iwaszko et al., 2004). It is known that proton permeability of LHCIIcontaining membranes is twice as high compared to control pure lipid vesicles (Iwaszko et al., 2004). A number of studies from our group have implicated polyamines (PAs) in the stimulation of qE (Ioannidis and Kotzabasis, 2007; Ioannidis et al., 2009; 2011; Tsiavos et al., 2012). The main PAs [putrescine (Put), spermidine (Spd) and spermine (Spm)] are found in vivo in chloroplast and bound to LHCII (Navakoudis et al., 2007), while upon illumination they 3 Page 3 of 21

accumulate in the lumen (Ioannidis et al., 2012). Based in previous work with amines it is anticipated that amines could reach 90 mM in the lumen for a concentration of about 2 mM in the stroma (Gaensslen and McCarty, 1971). Upon shuttering of actinic light it is anticipated that ΔpH between stroma and lumen will be minimized and polyamines will “return” to the

ip t

stroma. Recently, PAs were found to stimulate, at physiological pH, fluorescence quenching of both trimeric LHCII and monomeric LHCb complexes in vitro, mimicking to a great extent the action of protons (Tsiavos et al., 2012). Spm was the most potent quencher and

cr

induced aggregation of LHCII trimers, due to its highly cationic character. In the present work, we studied the effect of the aggregation state of LHCII on non-specific proton

us

permeation across lipid membranes of liposomes. Based in the results obtained, a new

an

protective role of Spm is discussed.

2. Materials and methods

M

2.1 Materials

Egg phosphatidylcholine (egg-PC) and cholesterol of the highest purity were purchased from

d

Avanti (Avanti Polar Lipids Inc., Alabaster, Alabama). The non-ionic surfactant n-dodecyl-βD-maltoside (β-DM), spermine, valinomycin, 8-hydroxypyrene-1,3,6-trisulfonic acid (PTS)

te

and sephadex G-100 were obtained from Sigma (Sigma-Aldrich, St. Louis, MO). SM2 BioBeads were purchased from Bio-Rad (Bio-Rad Laboratories Inc., Hercules, CA) and

Ac ce p

polycarbonate filters from Avestin (Avestin Inc., Ottawa, Canada). All other reagents were of analytical grade.

2.2 Preparation of LHCII complexes Thylakoids from spinach leaves were isolated according to Bassi et al. (1985). The LHCII was isolated as described in Tsiavos et al. (2012). The Chl a/b ratio of the trimeric LHCII preparation was 1.36, indicating that each monomer contained about 7 Chl a and 5 Chl b molecules (Kühlbrandt et al., 1994; Ruban et al., 1999). The Chl a/b ratio of LHCbs was 1.8. 2.3 Preparation of vesicles The production of unilamellar liposomes and proteoliposomes with a low ionic permeability is illustrated in Figure 1 and was carried out in four stages according to Rigaud and Lévy (2003): (1) preparation of preformed large, unilamellar liposomes, (2) solubilization of liposomes, (3) membrane protein reconstitution and (4) proteoliposome formation upon detergent removal. 4 Page 4 of 21

2.3.1 Liposome preparation At first, large homogeneous and unilamellar liposomes were prepared from egg-PC and cholesterol in a molar ratio of 9:1. Lipids were dried from chloroform in a stream of nitrogen gas to give a thin film in a glass flask. To this thin film a solution of 10mM Tricine (pH 7.5),

ip t

50mM KCl containing 4x10-4M PTS, a fluorescent pH-sensitive probe, was added. The solution was mixed in a vortex mixer to form multilamellar vesicles. The suspension was then passed through an extruder (21 times) using a 0.2 μm pore size polycarbonate filter in a

cr

LiposoFast extrusion system to form large unilamellar vesicles of 180-200 nm diameter

2.3.2 Solubilization of liposomes by dodecyl maltoside

us

(Mayer et al., 1986; Reimhult et al., 2003).

The higher β-DM concentration needed to be added to the liposome suspension in order to

an

reach the onset of solubilization, was calculated by the equation given by Lambert et al. (1998):

, in which

is the lipid concentration,

detergent to be added, detergent concentration and

is the aqueous monomeric

is the detergent-to-lipid ratio in detergent-saturated

M

liposomes. Given that

is the concentration of the

of β-DM in the presence of lipids is 0.3 mM (or 0.15 mg/ml),

d

is 1 mol β-DM/mol lipid (or 0.625% w/w) and the lipid concentration 1.25 mg/ml, the

te

final concentration of β-DM added was 0.65 mM; this is far below the critical concentration for lipid-detergent micelles formation. The liposomal suspension was stirred for 1 h at room

Ac ce p

temperature before protein addition.

2.3.3 Incorporation of LHCII proteins into liposomes To test the effect of LHCII aggregation on proton permeability, we prepared proteoliposomes containing preincubated and non preincubated LHCII samples with SM2 BioBeads or Spm (Fig. 1). More particularly, in order to obtain proteoliposomes with integrated quenched LHCII, we preincubated LHCII samples for 1 h with SM2 BioBeads or for 15 min with 100 μM Spm. Unquenched or quenched LHCII (after the removal of the polystyrene beads) was added to the lipid-detergent suspension at a lipid-to-protein ratio of 80 (w/w) and was incubated for 1 h at room temperature before detergent removal and proteoliposome reconstitution. Note that an equal volume of buffer containing 14 mM Hepes (pH 7.5) without LHCII, was also added to the liposomal suspension before liposome reconstitution for control measurements. 2.3.4 Liposome and proteoliposome reconstitution

5 Page 5 of 21

Liposome and proteoliposome reconstitution was performed by detergent removal using adsorption onto polystyrene BioBeads SM2. 0.05 g SM2 BioBeads were added per 0.5 ml of suspension and stirred for 1 h at room temperature. 2.3.5 Removal of external probe

ip t

In the end, gel chromatography was applied to separate the PTS that had not been incorporated into liposomes and proteoliposomes. The used Sephadex G-100 packed column, had an internal diameter of 0.8 cm and 10.5 cm length. A solution of 10 mM Tricine (pH

cr

7.5), 50 mM KCl was used as a mobile phase while the flow rate was adjusted to 0.27 fractions of liposomal and proteoliposomal suspensions. 2.4 Kinetic measurements

us

ml/min. Combined detection of UV-Vis absorption spectra was applied to identify the

spectrophotometer

luminometer

an

Proton permeability across lipid membranes was monitored fluorometrically using a LS-50B of

Perkin

Elmer

(Perkin

Elmer,

Waltham

Massachusetts). The PTS-containing liposomes and proteoliposomes were subjected to an

M

external acidic pulse and changes in internal fluorescence were monitored as a function of time. The emission was detected at 513 nm and the excitation was set at 402 and 452nm (at the main maxima of the absorption bands of protonated and non-protonated forms of PTS

d

respectively). The fluorescence intensity of PTS at 402 nm (F402) and 452 nm (F452) was used

te

to monitor the acidification kinetics of the liposome and proteoliposome interior. The buildup of a potential counteracting H+ release in the liposome interior was prevented by addition

Ac ce p

of valinomycin (a K+ ion carrier, in an ethanolic solution) 40 s after the beginning of the recording at a final concentration of 3 μM. The proton gradient across the membranes was generated 2 min after the valinomycin addition by injection of a small volume of 0.67 N HCl into the cuvette. The suspension was acidified to a pH level of 5.7 (versus pH 7.3 inside the vesicles). The pH changes inside liposomes and proteoliposomes were calculated on the basis of a calibration curve (Fig. 2). The fluorescence intensity of the main band of PTS at 402 nm was found to reach its maximum at pH 6.45 and therefore could not be used for the estimation of the exact pH values inside vesicles (Fig. 2). On the contrary, the fluorescence intensity of the non-protonated form of PTS at 452 nm showed a higher sensitivity to acidic pH values and it was selected to monitor the acidification kinetics presented in this paper. First order fit on the exponential decay curves of the fluorescence intensity at 452 nm was used to estimate the rate constants kH+= 1/τ (min-1) for proton permeability. One representative measurement of at least three independent experiments is presented for each treatment. 6 Page 6 of 21

3. Results 3.1 The effect of the incorporation of LHCII and LHCbs on proton permeation In order to achieve a robust and reproducible assay for producing large unilamellar

ip t

proteoliposomes, we tested the permeability of liposomes composed of egg-PC with or without admixture of cholesterol (chol). The presence of chol, in a PC-Chol molar ratio of 9:1, reduced proton leakage of liposomes by 30% (in terms of permeability) and therefore it

cr

was chosen for the experiments described below. To probe proton movement across the membrane, PTS, a pH-dependent fluorescent and membrane-impermeable hydrophilic probe,

us

was trapped within the liposome lumen (Kano and Fendler, 1978; Clement and Gould, 1981; Lévy et al., 1990). The fluorescence at 513 nm was recorded with excitation at 452 nm (main

an

absorption band of the non-protonated PTS). The fluorescence intensity at 452 nm before and after the acidification of the medium probed the pH increase in the lumen of the liposome. Addition of valinomycin (val) into the cuvette prior to acidification enabled free K+ diffusion

M

across the lipid membrane and diminished residual Δψ due to ions trapped in the lumen. These gradients could constrain proton permeation and therefore we diminished them before the HCl addition. Figure 3 presents the fluorescence changes of PTS entrapped inside

d

liposomes (control) and proteoliposomes containing either LHCII trimers

(+LHCII) or

te

monomeric LHCbs (+LHCbs). Injection of HCl (H+) at 160 s caused a rapid decrease of the fluorescence intensity. Both LHCII trimers and monomeric LHCbs increased proton

Ac ce p

permeability of liposomes as can be seen from the larger decrease in the internal pH of the reconstituted proteoliposomes after acidification. The inner pH of liposomes and proteoliposomes was found to be stabilized at 7.3 (F452=1) before the addition of HCl (at 160 s). All kinetic measurements presented in this paper have been normalized at the fluorescence values right before acid addition (160 s). According to the pH-metric curve of Figure 2, liposomes in Figure 3 reached a proton transport equilibrium within ca. 100 s after acidification, at a level of pH 6.1 (F452= 0.141). Despite the proton gradient across the membrane (pH 5.7 outside versus 6.1 inside the liposomes), no further ion transport was observed (i.e., vesicles can sustain a ΔpH of 0.4). On the contrary, proteoliposomes containing LHCII trimers and proteoliposomes containing LHCbs reached equilibrium at a level of pH 5.95 (F452=0.113 and 0.11 respectively) (i.e., vesicles can sustain a ΔpH of 0.25). The response of the membrane proton flux to changes in ΔpH was obtained from the kinetic

7 Page 7 of 21

traces of the first 50 s. The presence of LHCII increased 27 % and of LHCbs 44 % the proton permeability of liposomes (for kH+ values see Fig.3). 3.2 The effect of the aggregation state of LHCII on proton permeation

ip t

Protein aggregation levels were estimated indirectly via fluorescence spectroscopy by assuming that low aggregation leads to low level of Chl a fluorescence quenching and that high level is indicative of high level of aggregation. Fluorescence quenching of LHCII caused

cr

by detergent removal or by Spm addition at room temperature is shown in Figure 4A. It is clear that under the selected experimental conditions there are at least two distinct states of

us

quenching/aggregation. The first is low level of quenching (using well solubilised LHCII) while the second is intermediate level of quenching (using Spm or Biobeads). To test the

an

effect of LHCII aggregation on proton permeability, we prepared proteoliposomes containing preincubated and non preincubated LHCII samples with SM2 BioBeads or Spm. Kinetic measurements of these treatments are presented in Figure 4. The effect of the trimeric

M

unquenched LHCII (-BioBeads/-Spm) on increasing proton permeability of liposomes was only marginal (Fig. 4B). On the contrary, the most significant effect was recorded when partly aggregated LHCII trimers, due to preincubation with SM2 Biobeads (+BioBeads/-

d

Spm) or with 100 μM Spm for 15 min (-BioBeads/+Spm), were reconstituted into

te

proteoliposomes (Fig. 4C and D respectively). LHCII containing proteoliposomes have a kH+ of ~ 9.5 min-1 and the pre-incubation of LHCII with Spm and BioBeads increase their kH+ to

Ac ce p

~ 10.9 min-1 (+14.7%) and ~ 11.8 min-1 (+24 %) respectively. Spermine’s highly cationic character induces LHCII aggregation and consequently quenching of fluorescence in buffer solution, which leads to an enhancement of proton permeability. In order to verify if Spm is active on LHCII already integrated in membranes and search for evidence that might support direct, specific effects of Spm on LHCII properties, Spm was also added after the proteoliposome reconstitution, during the kinetic measurements. Injection of Spm into the cuvette (100 μM final concentration), 20 s after the beginning of the recording, did alter the rate of proton permeability of proteoliposomes containing especially unquenched LHCII. Proteoliposomes containing quenched (due to preincubation with SM2 Biobeads) and unquenched LHCII have a kH+ of ~ 11.8 min-1 and 9.5 min-1 respectively. Exogenously added Spm increased their kH+ to 12 min-1 (1.7 %) and 10.7 min-1 (12.6 %) respectively.

8 Page 8 of 21

4. Discussion This study corroborates similar findings by Gruszecki’s group (Wardak et al., 2000; Iwaszko et al., 2004) on the effect of the incorporation of LHCII on proton transport across liposome membranes made by egg-PC As monitored fluorometrically, with the application of pH-

ip t

sensitive PTS entrapped inside vesicles, both types of proteoliposomes containing LHCII or LHCb complexes showed increased proton permeability in comparison to liposomes (Fig. 3).

cr

It has been found that the LHCII-related increase in the membrane conductivity depends on the aggregation state of LHCII (Wardak et al., 2000). In these earlier studies, cation-induced

us

formation of large aggregates of LHCII was applied in its isolation from thylakoid membranes, according to the method of Krupa et al. (1987). Our LHCII samples were solubilised at 200 μM β-DM after their isolation and had only a marginal effect on proton

an

permeability of liposomes (Fig. 4B). Aggregation of isolated major LHCII complex is well known to be induced under low detergent conditions in vitro (Ruban and Horton, 1992;

M

Ruban et al., 1994; Tsiavos et al., 2012). Incubation of LHCII samples with SM2 BioBeads prior to their insertion in liposomes, led to the adsorption of β-DM. The removal of the detergent caused the aggregation of the complexes, which in turn caused a significant

d

decrease in their fluorescence at 681 nm (Fig. 4A). When these aggregated LHCII complexes

te

were inserted in unilamellar liposomes, they induced a large increase (24%) in proton permeability across the lipid membranes (Fig. 3 and 4C). After acidification, an internal pH

Ac ce p

change of 1.2 and 1.35 units (according to Figure 2) was obtained for liposomes and proteoliposomes containing LHCII respectively. In agreement with our results, Iwaszko et al. (2004) reported that acidification of both liposome and proteoliposome suspensions caused a decrease of their internal pH from 6.6 to 5.5 (1.1 units) and 5.3 (1.3 units) respectively. Proteoliposomes containing aggregated LHCb complexes showed similar permeability levels to protons (F452 = 0.11), but higher rates than proteoliposomes containing aggregated trimeric LHCII (Fig. 3). This could be explained in terms of different pigment-protein composition of the two bands harvested from the sucrose gradient. The band corresponding to LHCb proteins contained the minor CP29, CP26 and CP24 as well as monomers of LHCII (Lhcb1, 2 and 3) (Tsiavos et al., 2012). Thus, under conditions that induce their aggregation, structures favouring proton transport within the minor and the major trimeric complexes might be formed.

9 Page 9 of 21

Aggregation of LHCII in vivo was suggested to occur under excess light conditions (Li et al., 2009). Build-up of transthylakoid membrane proton gradient (ΔpH) triggers the protonation of LHCII proteins (Walters et al., 1994; 1996) which leads to their dissociation from PSII and their aggregation within the membrane (Ruban et al., 1996; Betterle et al., 2009). This

ip t

process is fundamental in the photoprotective dissipation of excess excitation energy as heat within the LHCII that is monitored by the non-photochemical quenching (NPQ) of Chl fluorescence (Johnson et al., 2011). The concept (based on liposome results of Wardak et al.,

cr

2004 and present work) that upon aggregation LHCII increases the permeability of lipid membranes to protons in vitro, indicates that a second protective mechanism (other than qE)

us

could be activated in vivo. Overacidification of lumen results in detachment of oxygen evolving complex, hindering of LEF at the level of Cytb6f and puts at risk photosynthetic

an

apparatus (Krieger and Weis, 1993; Kramer et al., 1999; Rott et al., 2011). Thus, safety valves that relief this extremely high concentration of protons from lumen back to stroma might play a protective role.There is evidence that the tetramine Spm, which is normally

M

found in the chloroplast and bound to LHCbs (Ioannidis et al., 2011), is involved in the quenching mechanism. Spm supply increased qE in leaf discs at low light (Ioannidis and Kotzabasis, 2007) and increased qF of isolated LHCII from algae (Ioannidis et al., 2011) and

d

plants (Tsiavos et al., 2012). Spm was found to shift the relation between qE and ΔpH so that

te

the quenching could be activated even at low light conditions (lower ΔpH, i.e., higher lumen pH) in thylakoids and at physiological pH values in LHCII aggregates. Raman studies of

Ac ce p

Spm-treated LHCII revealed the same spectroscopic signatures of LHCII aggregation that have been correlated with the induction of qE (Tsiavos et al., 2012). Here, Spm was also shown to affect indirectly (via aggregation formation) the proton release into the liposome lumen through LHCII in vitro. Preincubation of LHCII with 100 μM Spm resulted in the aggregation of the complexes, which caused a significant decrease in their fluorescence at 681 nm (Fig. 4A). When these Spm-induced aggregated LHCII complexes were reconstituted into proteoliposomes, they increased proton permeability across the lipid membranes, similar to the effect of SM2 BioBeads (Fig. 4C and D). Although the mode of protein-protein interaction in aggregation phenomena is not fully characterised we assume here that Spminduced and BioBeads-induced aggregation is of similar nature. This mode of action of Spm could be of importance for a second photoprotective mechanism of the photosynthetic membrane in vivo, that should be activated only at high level of energization. ΔpH is normally established and increased upon light intensity increase. The 10 Page 10 of 21

aggregation of LHC after a certain level causes some leakage in the thylakoids, which hinders further increase of ΔpH. At this point qE is still high (for the sake of the argument we assume qE values of about 1.5-2) because aggregation of LHC is intense and ΔpH is high, but slow leakage via LHC aggregates may happen (either within LHCII or in the interface of the

ip t

aggregate with lipids of the membrane). The fact that exogenously added Spm altered the rate of proton permeability of

cr

proteoliposomes containing unquenched LHCII might be indicative of a specific effect of Spm. There is a lumen-exposed pocket in the trimeric LHCII, which could allow bulk lumen

us

protons to reach deep in the trimer center and is large enough to accommodate one or more Spm molecules (Ioannidis et al., 2011). Further studies need to be done with different cations in a range of concentrations and light conditions. Recent studies on the structural flexibility

an

of lipid-LHCII macroassemblies have revealed a light-dependent mode of action of cations (such as Mg2+ and spermidine) in LHCII aggregation and the stacking of LHCII-containing membranes (Hind et al., 2014). These findings are in line with former studies of our group

M

that showed Spd is more effective than monovalent and divalent inorganic cations in restoring Fv/Fm (via LHCII aggregation and membrane stacking), but less effective than Spm

d

(Ioannidis and Kotzabasis 2007). We must note here, that phosphatidylcholine is not a thylakoid lipid component and thus the in vivo situation might be different to our in vitro

te

system. Future work will check whether this is the case also for native membranes.

Ac ce p

In conclusion, proton permeation across liposomal membranes containing LHCII was found to be dependent on the degree of the aggregation of the complexes. Spm-induced aggregation of LHCII increased the membrane permeability to protons (Fig. 5).

References

Ballottari M, Girardon J, Betterle N, Morosinotto T, Bassi R. Identification of the chromophores involved in aggregation-dependent energy quenching of the monomeric photosystem II antenna protein Lhcb5. J Biol Chem 2010; 285: 28309-21. Bassi R, Machold O, Simpson D. Chlorophyll-proteins of two photoststem I preparations from maize. Carlsberg Res Commun 1985; 50: 145-62. Betterle N, Ballottari M, Zorzan S, de Bianchi S, Cazzaniga S, Dall'Osto L, et al. Light-induced dissociation of an antenna hetero-oligomer is needed for non-photochemical quenching induction. J Biol Chem 2009; 284: 15255-66. Briantais JM, Vernotte C, Picaud M, Krause GH. A quantitative study of the slow decline of chlorophyll a fluorescence in isolated chloroplasts. Biochim Biophys Acta 1979; 548: 128-38. 11 Page 11 of 21

Ac ce p

te

d

M

an

us

cr

ip t

Clement NR, Gould JM. Pyranine (8-hydroxy-1,3,6-pyrenetrisulfonate) as a probe of internal aqueous hydrogen ion concentration in phospholipid vesicles. Biochemistry 1981; 20: 1534-8. Gaensslen RE, McCarty RE. Amine uptake in chloroplasts. Arch Biochem Biophys 1971; 147: 55-65. Hind G, Wall JS, Várkonyi Z, Istokovics A, Lambrev PH, Garab G. Membrane crystals of plant lightharvesting complex II disassemble reversibly in light. Plant Cell Physiol 2014; 55(7):1296303. Horton P, Ruban AV, Rees D, Pascal AA, Noctor G, Young AJ. Control of the light-harvesting function of chloroplast membranes by aggregation of the LHCII chlorophyll-protein complex. FEBS lett 1991a; 292: 1-4. Ioannidis NE, Kotzabasis K. Effects of polyamines on the functionality of photosynthetic membrane in vivo and in vitro. Biochim Biophys Acta 2007; 1767: 1372-82. Ioannidis N, Kotzabasis K. Polyamines in Chemiosmosis in vivo: a cunning mechanism for the regulation of ATP synthesis during growth and stress. Front Plant Sci 2014; doi: 10.3389/fpls.2014.00071. Ioannidis NE, Ortigosa SM, Veramendi J, Pintó-Marijuan M, Fleck I, Carvajal P, et al. Remodeling of tobacco thylakoids by over-expression of maize plastidial transglutaminase. Biochim Biophys Acta 2009; 1787: 1215-22. Ioannidis N, Sfichi-Duke L, Kotzabasis K. Polyamines stimulate non-photochemical quenching of chlorophyll fluorescence in Scenedesmus obliquus. Photosynth Res 2011; 107: 169-75. Ioannidis NE, Cruz JA, Kotzabasis K, Kramer DM. Evidence that putrescine modulates the higher plant photosynthetic proton circuit. PLoS ONE 2012; 7: e29864. Iwaszko E, Wardak A, Krupa Z, Gruszecki WI. Ion transport across model lipid membranes containing light-harvesting complex II: an effect of light. J Photoch Photobio B 2004; 74: 1321. Johnson MP, Goral TK, Duffy CDP, Brain APR, Mullineaux CW, Ruban AV. Photoprotective energy dissipation involves the reorganization of photosystem II light-harvesting complexes in the grana membranes of spinach chloroplasts. Plant Cell 2011; 23: 1468-79. Kano K, Fendler JH. Pyranine as a sensitive pH probe for liposome interiors and surfaces. pH gradients across phospholipid vesicles. Biochim Biophys Acta 1978; 509: 289-99. Kramer DM, Sacksteder CA, Cruz JA. How acidic is the lumen? Photosynth Res 1999; 60: 151–63. Krieger A, Weis E. The role of calcium in the pH-dependent control of photosystem II. Photosynth Res 1993; 37: 117–30.Krupa Z, Huner NPA, Williams JP, Maissan E, James DR. Development at cold-hardening temperatures: the structure and composition of purified rye light harvesting complex II. Plant Physiol 1987; 84: 19-24. Kühlbrandt W, Wang DN, Fujiyoshi Y. Atomic model of plant light-harvesting complex by electron crystallography. Nature 1994; 367: 614– 621. Lambert O, Lévy D, Ranck J-L, Leblanc G, Rigaud J-L. A new gel-like phase in dodecyl maltoside lipid mixtures: implications in solubilization and reconstitution studies. Biophys J 1998; 74: 918-30. Lévy D, Seigneuret M, Bluzat A, Rigaud JL, Evidence for proton countertransport by the sarcoplasmic reticulum Ca2(+)-ATPase during calcium transport in reconstituted proteoliposomes with low ionic permeability. J Biol Chem 1990; 265: 19524-34. Li Z, Wakao S, Fischer BB, Niyogi KK. Sensing and responding to excess light. Annu Rev Plant Biol 2009; 60: 239-60. Mayer LD, Hope MJ, Cullis PR. Vesicles of variable sizes produced by a rapid extrusion procedure. Biochim Biophys Acta 1986; 858: 161-8. Miloslavina Y, Wehner A, Lambrev PH, Wientjes E, Reus M, Garab G, et al. Far-red fluorescence: A direct spectroscopic marker for LHCII oligomer formation in non-photochemical quenching. FEBS lett 2008; 582: 3625-31. Mitchell P. Chemiosmotic coupling in oxidative and photosynthetic phosphorylation. Biol Rev 1966; 41: 445-501. Muller P, Li XP, Niyogi KK. Non-photochemical quenching. A response to excess light energy. Plant Physiol 2001; 125: 1558-66.

12 Page 12 of 21

Ac ce p

te

d

M

an

us

cr

ip t

Navakoudis E, Vrentzou K, Kotzabasis K. A polyamine- and LHCII protease activity-based mechanism regulates the plasticity and adaptation status of the photosynthetic apparatus. Biochim Biophys Acta 2007; 1767: 261-71. Reimhult E, Höök F, Kasemo B. Intact vesicle adsorption and supported biomembrane formation from vesicles in solution: influence of surface chemistry, vesicle size, temperature and osmotic pressure. Langmuir 2003; 19: 1681–91. Rigaud J-L, Lévy D. Reconstitution of membrane proteins into liposomes. In: Nejat D, Editor. Methods in Enzymology. Academic Press, 2003. p 65-86. Rott M, Martins NF, Thiele W, Lein W, Bock R, Kramer DM, et al. ATP synthase repression in tobacco restricts photosynthetic electron transport, CO2 assimilation, and plant growth by overacidification of the thylakoid lumen. Plant Cell 2011; 23: 304–21. Ruban AV, Horton P. Mechanism of ΔpH-dependent dissipation of absorbed excitation energy by photosynthetic membranes. I. Spectroscopic analysis of isolated light-harvesting complexes. Biochim Biophys Acta 1992; 1102: 30-8. Ruban AV, Rees D, Noctor GD, Young A, Horton P. Long-wavelength chlorophyll species are associated with amplification of high-energy-state excitation quenching in higher plants. Biochim Biophys Acta 1991; 1059: 355-60. Ruban AV, Horton P, Young AJ. Aggregation of higher plant xanthophylls: Differences in absorption spectra and in the dependency on solvent polarity. J Photoch Photobio B 1993a; 21: 229-34. Ruban AV, Young AJ, Horton P. Induction of nonphotochemical energy dissipation and absorbance changes in leaves (evidence for changes in the state of the light-harvesting system of photosystem II in vivo). Plant Physiol 1993b; 102: 741-50. Ruban AV, Young A, Horton P. Modulation of chlorophyll fluorescence quenching in isolated light harvesting complex of Photosystem II. Biochim Biophys Acta 1994; 1186: 123-127. Ruban AV, Young AJ, Horton P. Dynamic properties of the minor chlorophyll a/b binding proteins of photosystem II, an in vitro model for photoprotective energy dissipation in the photosynthetic membrane of green plants. Biochemistry 1996; 35: 674-8. Ruban AV, Pesaresi P, Wacker U, Irrgang KDJ, Bassi R, Horton P. The relationship between the binding of dicyclohexylcarbodiimide and quenching of chlorophyll fluorescence in the lightharvesting proteins of photosystem II. Biochemistry 1998; 37: 11586-91. Ruban AV, Lee PJ, Wentworth M, Young AJ, Horton P. Determination of the stoichiometry and strength of binding of xantophylls to the photosystem II light harvesting complexes. J Biol Chem 1999; 274: 10458–10465 Tsiavos T, Ioannidis NE, Kotzabasis K. Polyamines induce aggregation of LHC II and quenching of fluorescence in vitro. Biochim Biophys Acta 2012; 1817: 735-43. Walters RG, Ruban AV, Horton P. Higher plant light-harvesting complexes LHCIIa and LHCIIc are bound by dicyclohexylcarbodiimide during inhibition of energy dissipation. Eur J Biochem 1994; 226: 1063-9. Walters RG, Ruban AV, Horton P. Identification of proton-active residues in a higher plant lightharvesting complex. Proc Natl Acad Sci USA 1996; 93: 14204-9. Wardak A, Brodowski R, Krupa Z, Gruszecki WI. Effect of light-harvesting complex II on ion transport across model lipid membranes. J Photoch Photobio B 2000; 56: 12-8. Wraight CA, Crofts AR. Energy-dependent quenching of chlorophyll a fluorescence in isolated chloroplasts. Eur J Biochem 1970; 17: 319-27.

13 Page 13 of 21

Figure legends Figure 1. A schematic representation of the procedure for producing large unilamellar proteoliposomes in four stages according to Rigaud and Lévy (2003): (1) preparation of large, unilamellar liposomes, (2) solubilization of liposomes, (3) membrane protein reconstitution

ip t

and (4) proteoliposome formation upon detergent removal. Two different pathways were used for the production of proteoliposomes containing unquenched (path A) or quenched (path B)

cr

LHCII. In pathway B, LHCII samples were preincubated for 1 h with SM2 BioBeads or 100μM Spm. LHCII samples were then added to the lipid-detergent suspension for

us

proteoliposome reconstitution. The external probe was removed with gel chromatography before the kinetic measurements. To test the effect of exogenously added Spm, 100μM Spm

an

was added into the cuvette during these measurements.

Figure 2. pH titration curve of PTS. The change in fluorescence intensity upon excitation at 402 nm and 452 nm is recorded at different pH values. The fluorescence emission of 0.2 μM

M

PTS was detected at 513 nm. Due to the internal pH of vesicles before the acidification, the curve was normalized at pH 7.3

d

Figure 3. Example kinetics of fluorescence intensity changes of PTS entrapped inside unilamellar liposomes and unilamellar proteoliposomes with integrated trimeric LHCII

te

(+LHCII) or monomeric LHCb (+LHCbs) complexes (both in partly quenched states). The arrows indicate the time point of valinomycin (Val) and HCl (H+) addition into the cuvette.

Ac ce p

The samples were excited at 452 nm, and the fluorescence emission was detected at 513 nm. Initial pH inside the vesicles was 7.3. Single-exponential decay curves of the first 50 s after the acidification were used to calculate the proton permeability rate of the membranes (kH+). kH+ for liposomes, LHCII-proteoliposomes (+LHCII) and LHCb-proteoliposomes (+LHCbs) was 9.3 ± 0.19, 11.8 ± 0.28 and 13.4 ± 2.15 min-1 respectively. The standard error was obtained from 3-4 measurements of different preparations. Figure 4. A. Chl fluorescence intensity at 681 nm of trimeric LHCII, fully solubilized in 200 μM β-DM and partly aggregated after its incubation with SM2 BioBeads for 1 h (LHCII+BioBeads) or with 100 μM Spm for 15 min (LHCII+Spm). The Chl concentration was 6.4 μg/ml. After incubation samples of 1 mL were diluted in 6 ml of 10 mM Tricine (pH 7.5), 50 mM KCl buffer to obtain the same final concentration as in proteoliposome kinetic measurements. B, C and D. Example kinetics of fluorescence changes of PTS entrapped inside unilamellar liposomes (-LHCII) and unilamellar proteoliposomes (liposomes with 14 Page 14 of 21

integrated LHCII trimers, +LHCII), monitoring proton transport across the liposome membranes. B. Proteoliposomes with integrated unquenched LHCII (-BioBeads/-Spm). C. Proteoliposomes with integrated partly quenched LHCII due to pre-incubation of LHCII trimers with SM2 BioBeads (+BioBeads/-Spm). D. Proteoliposomes with integrated partly

ip t

quenched LHCII due to pre-incubation of LHCII trimers with 0.1mM Spm (BioBeads/+Spm). Liposomes in panels B and C have also been incubated with BioBeads and Spm respectively at the same step of the protocol. The samples were excited at 452nm, and

cr

the fluorescence emission was detected at 513nm. kH+ values for proteoliposomes in panels

us

B, C and D are shown.

Figure 5. A simplified scheme for the indirect effect of Spm on the proton permeation across LHCII-containing liposomal membranes. Preincubation of LHCII trimers with Spm induces

Ac ce p

te

d

M

an

LHCII aggregation that increases membrane permeability to protons.

15 Page 15 of 21

an

us

cr

ip t

Figure 1

Figure 1. A schematic representation of the procedure for producing large unilamellar

M

proteoliposomes in four stages according to Rigaud and Lévy (2003): (1) preparation of large, unilamellar liposomes, (2) solubilization of liposomes, (3) membrane protein reconstitution

ed

and (4) proteoliposome formation upon detergent removal. Two different pathways were used for the production of proteoliposomes containing unquenched (path A) or quenched (path B)

ce pt

LHCII. In pathway B, LHCII samples were preincubated for 1 h with SM2 BioBeads or 100μM Spm. LHCII samples were then added to the lipid-detergent suspension for proteoliposome reconstitution. The external probe was removed with gel chromatography before the kinetic measurements. To test the effect of exogenously added Spm, 100μM Spm

Ac

was added into the cuvette during these measurements.

Page 16 of 21

Figure 2

2

F402

1.8

F452

1.4 1.2 1

ip t

Fluorescence intensity

1.6

0.8

cr

0.6

0.2 0 5.5

6

6.5

7

7.5

8

8.5

an

5

us

0.4

M

pH

Figure 2. pH titration curve of PTS. The change in fluorescence intensity upon excitation at

ed

402 nm and 452 nm is recorded at different pH values. The fluorescence emission of 0.2 μM PTS was detected at 513 nm. Due to the internal pH of vesicles before the acidification, the

Ac

ce pt

curve was normalized at pH 7.3.

Page 17 of 21

Figure 3

+

H 1

LHCII-Proteoliposomes (+LHCII) LHCbs-Proteoliposomes (+LHCbs)

0.8

Val

0.4

kH+ = 13.4 min

0.2

kH+ = 9.3 min

-1

kH+ = 11.8 min

-1

0 100

200

300

400

an

0

-1

cr

ip t

0.6

us

Fluorescence intensity at 452nm

Liposomes

500

M

Time (sec)

Figure 3. Example kinetics of fluorescence intensity changes of PTS entrapped inside unilamellar liposomes and unilamellar proteoliposomes with integrated trimeric LHCII

ed

(+LHCII) or monomeric LHCb (+LHCbs) complexes (both in partly quenched states). The arrows indicate the time points of valinomycin (Val) and HCl (H+) addition into the cuvette.

ce pt

The samples were excited at 452 nm, and the fluorescence emission was detected at 513 nm. Initial pH inside the vesicles was 7.3. Single-exponential decay curves of the first 50 s after the acidification were used to calculate the proton permeability rate of the membranes (kH+). kH+ for liposomes, LHCII-proteoliposomes (+LHCII) and LHCb-proteoliposomes (+LHCbs)

Ac

was 9.3 ± 0.19, 11.8 ± 0.28 and 13.4 ± 2.15 min-1 respectively. The standard error was obtained from 3-4 measurements of different preparations.

Page 18 of 21

Figure 4

0.3

LHCII + BioBeads 100

50

0

0.25

0.2

0.15

0.1 640

660

680

700

720

- LHCII

kH+ = 11.8 min 100

200

0.3

300

400

500

-BioBeads/+Spm

0.25

an

0.25

Fluorescence intensity at 452nm

D

kH+ = 9.3 min

-1

-LHCII

0.15

kH+ = 9.5 min

-1

+LHCII

0.1 100

200

300

400

ce pt

Time (sec)

0.2

M

0.2

ed

Fluorescence intensity at 452nm

+LHCII

Time (sec)

-BioBeads/-Spm

B

-1

us

λ (nm) 0.3

ip t

LHCII + Spm

150

+Biobeads/-Spm

C

LHCII

Fluorescence intensity at 452nm

Fluorescence intensity (rel.units)

200

cr

A

500

0.15

0.1

-LHCII kH+ = 10.9 min 100

200

-1

+LHCII 300

400

500

Time (sec)

Figure 4. A. Chl fluorescence intensity at 681 nm of trimeric LHCII, fully solubilized in 200 μM β-DM and partly aggregated after its incubation with SM2 BioBeads for 1 h

Ac

(LHCII+BioBeads) or with 100 μM Spm for 15 min (LHCII+Spm). The Chl concentration was 6.4 μg/ml. After incubation samples of 1 mL were diluted in 6 ml of 10 mM Tricine (pH 7.5), 50 mM KCl buffer to obtain the same final concentration as in proteoliposome kinetic measurements. B, C and D. Example kinetics of fluorescence changes of PTS entrapped inside unilamellar liposomes (-LHCII) and unilamellar proteoliposomes (liposomes with integrated LHCII trimers, +LHCII), monitoring proton transport across the liposome membranes. B. Proteoliposomes with integrated unquenched LHCII (-BioBeads/-Spm). C. Proteoliposomes with integrated partly quenched LHCII due to pre-incubation of LHCII trimers with SM2 BioBeads (+BioBeads/-Spm). D. Proteoliposomes with integrated partly quenched LHCII due to pre-incubation of LHCII trimers with 0.1mM Spm (-

Page 19 of 21

BioBeads/+Spm). Liposomes in panels B and C have also been incubated with BioBeads and Spm respectively at the same step of the protocol. The samples were excited at 452nm, and the fluorescence emission was detected at 513nm. kH+ values for proteoliposomes in panels

Ac

ce pt

ed

M

an

us

cr

ip t

B, C and D are shown.

Page 20 of 21

ip t

Figure 5

cr

Figure 5. A simplified scheme for the indirect effect of Spm on the proton permeation across LHCII-containing liposomal membranes. Preincubation of LHCII trimers with

Ac

ce pt

ed

M

an

us

Spm induces LHCII aggregation that increases membrane permeability to protons.

Page 21 of 21