Akademisk avhandling
som med vederbörligt tillstånd av rektorsämbetet vid Umeå Universitet, för erhållande av filosofie doktorsexamen i ämnet Växters cell- och molekylärbiologi, offentligen kommer att försvaras fredagen den 28 februari
2003 klockan 10:00 i hörsal KB3A9, KBC, Umeå Universitet.
Fakultetsopponent är Professor Henrik Vibe Scheller, Institut for Plantebiologi, KVL, Frederiksberg (Köpenhamn), Danmark.
Dissecting the photosystem II light-harvesting antenna
Dissecting the photosystem II light-harvesting antenna
Dissecting the photosystem II light-harvesting antenna
Dissecting the photosystem II light-harvesting antenna
Dissecting the photosystem II light-harvesting antenna
Jenny Andersson
Jenny Andersson (2003)
Dissecting the photosystem II light-harvesting antenna
Dissecting the photosystem II light-harvesting antenna
Dissecting the photosystem II light-harvesting antenna
Dissecting the photosystem II light-harvesting antenna
Dissecting the photosystem II light-harvesting antenna
UPSC, Department of Plant Physiology, Umeå University, SE-901 87 Umeå, Sweden Doctoral dissertation ISBN 91-7305-387-2
Dissertation abstract: In photosynthesis, sunlight is converted into chemical energy that is stored mainly as carbohydrates and supplies basically all life on Earth with energy. In order to efficiently absorb the light energy, plants have developed the outer light harvesting antenna, which is composed of ten different protein subunits (LHC) that bind chlorophyll a and b as well as different carotenoids. In addition to the light harvesting function, the antenna has the capacity to dissipate excess energy as heat (feedback de-excitation or qE), which is crucial to avoid oxidative damage under conditions of high excitation pressure. Another regulatory function in the antenna is the state transitions in which the distribution of the trimeric LHC II between photosystem I (PS I) and II is controlled. The same ten antenna proteins are conserved in all higher plants and based on evolutionary arguments this has led to the suggestion that each protein has a specific function.
I have investigated the functions of individual antenna proteins of PS II (Lhcb proteins) by antisense inhibition in the model plant Arabidopsis thaliana. Four antisense lines were obtained, in which the target proteins were reduced, in some cases beyond detection level, in other cases small amounts remained.
The results show that CP29 has a unique function as organising the antenna. CP26 can form trimers that substitute for Lhcb1 and Lhcb2 in the antenna structure, but the trimers that accumulate as a response to the lack of Lhcb1 and Lhcb2 cannot take over the LHC II function in state transitions. It has been argued that LHC II is essential for grana stacking, but antisense plants without Lhcb1 and Lhcb2 do form grana. Furthermore, LHC II is necessary to maintain growth rates in very low light.
Numerous biochemical evidences have suggested that CP29 and/or CP26 were crucial for feedback de-excitation. Analysis of two antisense lines each lacking one of these proteins clearly shows that there is no direct involvement of either CP29 or CP26 in this process. Investigation of the other antisense lines shows that no Lhcb protein is indispensable for qE. A model for feedback de-excitation is presented in which PsbS plays a major role.
The positions of the minor antenna proteins in the PS II supercomplex were established by comparisons of transmission electron micrographs of supercomplexes from the wild type and antisense plants.
A fitness experiment was conducted where the antisense plants were grown in the field and seed production was used to estimate the fitness of the different genotypes. Based on the results from this experiment it is concluded that each Lhcb protein is important, because all antisense lines show reduced fitness in the field.
Key words: antisense, Arabidopsis thaliana, chlorophyll, carotenoid, feedback de-excitation, fitness, LHC, NPQ, photosynthesis, state transitions, xanthophyll
Dissecting the photosystem II light harvesting antenna
Dissecting the photosystem II light harvesting antenna
Dissecting the photosystem II light harvesting antenna
Dissecting the photosystem II light harvesting antenna
Dissecting the photosystem II light harvesting antenna
Umeå Plant Science Centre, Department of Plant Physiology, Umeå University, Sweden Dissertation Umeå 2003
© Jenny Andersson, 2003 Umeå Plant Science Centre Department of Plant Physiology Umeå University
SE-901 87 UMEÅ Sweden
ISBN 91-7305-387-2 Printed by nra
List of papers
List of papers
List of papers
List of papers
List of papers
This thesis is based on the following papers, which will be referred to in the text by their Roman numerals.
J. Andersson, R. G. Walters, P. Horton, and S. Jansson (2001). Antisense inhibition of the photosynthetic antenna proteins CP29 and CP26: implications for the mechanism of protective energy dissipation. Plant Cell, 13:1193-1204
A. E. Yakushevska, W. Keegstra, E. J. Boekema, J. P. Dekker, J. Andersson, S. Jansson, A. V. Ruban and P. Horton (2003). The structure of photosystem II in Arabidopsis: localization of the CP26 and CP29 antenna complexes. Biochemistry 42:608-613 J. Andersson, M. Wentworth, R. G. Walters, C. A. Howard, A. V. Ruban, P. Horton and S. Jansson (2003). Absence of the main light-harvesting complex of photosystem II affects photosynthetic function. Provisionally accepted for publication in Plant J. A. V. Ruban*, M. Wentworth*, A. E. Yakushevska*, J. Andersson*, P. J. Lee, W. Keegstra, J. P. Dekker, E. J. Boekema, S. Jansson and P. Horton (*equally contributing authors) (2003). Plants lacking the main light harvesting complex retain photosystem II macro-organization. Nature in press
J. Andersson and S. Jansson (2003). Loss of Lhcb1 and Lhcb2 decreases growth in extreme low light. Manuscript
Papers I, II and IV are reprinted by kind permission of the publishers.
Not included in this thesis: S. Jansson, J. Andersson, S.-J. Kim and G. Jackowski (2000). An
Arabidopsis thaliana protein homologous to cyanobacterial high-light-inducible proteins. Plant
Mol.Biol. 42: 345-351.
I
II
III
IV
V
Table of contents
Table of contents
Table of contents
Table of contents
Table of contents
List of papers
List of papers
List of papers
List of papers
List of papers...
...
...
...
... 33333
Abbreviations
Abbreviations
Abbreviations
Abbreviations
Abbreviations ...
...
...
...
... 66666
Introduction: photosynthesis – the antenna perspective
Introduction: photosynthesis – the antenna perspective
Introduction: photosynthesis – the antenna perspective
Introduction: photosynthesis – the antenna perspective
Introduction: photosynthesis – the antenna perspective ...
...
...
... 88888
...
Organization of the thylakoid membrane ... 10
The light harvesting antenna and the Lhc super gene family ... 11
Protein composition of the photosystems and their antennae ... 11
LHC structure ... 13
Pigment composition ... 15
Does PsbS bind pigments? ... 15
Supermolecular organization of PS II and its antenna ... 16
Different LHC II trimer binding sites ... 16
Diverse polypeptide composition in LHC II trimers ... 16
Association of monomeric Lhcb proteins (the minor antenna) ... 16
LHC II not directly bound to PS II/peripheral LHC II ... 17
Association of LHC-related proteins ... 17
Supercomplex arrays ... 18
Regulatory mechanisms ... 18
The destructive power of light and oxygen ... 18
Avoiding over excitation ... 20
Feedback de-excitation ... 20
State transitions ... 21
Acclimation of the light harvesting antenna to different light conditions ... 23
Methods
Methods
Methods
Methods
Methods ...
...
...
... 25
...
25
25
25
25
Model species: Arabidopsis thaliana ... 25Antisense inhibition ... 26
T-DNA tagged mutants ... 27
Chlorophyll fluorescence ... 27
Fv/Fm ... 27
Non-photochemical quenching (NPQ) ... 27
Evaluation of fitness – field experiment ... 29
Discussion
Discussion
Discussion
Discussion
Discussion...
...
...
...
... 31
31
31
31
31
Transgenic lines ... 31Why did Flachmann and Kühlbrandt fail? ... 33
Feedback de-excitation in the antisense plants ... 34
Speculative model for feedback de-excitation ... 35
PS II antenna organization ... 37
Conservation of supercomplex ultra-structure ... 37
Photosynthesis in the absence of antisense inhibited antenna proteins ... 39
Grana ... 40
State transitions ... 40
Is each Lhcb necessary? ... 41
Functions of the Lhcb proteins ... 41
CP29 ... 41 CP26 ... 41 CP24 ... 42 Lhcb1/Lhcb2 ... 42 Lhcb3 ... 42
Future perspectives
Future perspectives
Future perspectives
Future perspectives
Future perspectives...
...
...
...
... 43
43
43
43
43
Den ljusskördande antennen hos fotosystem II
Den ljusskördande antennen hos fotosystem II
Den ljusskördande antennen hos fotosystem II
Den ljusskördande antennen hos fotosystem II
Den ljusskördande antennen hos fotosystem II...
...
...
... 45
...
45
45
45
45
Acknowledgements
Acknowledgements
Acknowledgements
Acknowledgements
Acknowledgements ...
...
...
...
... 48
48
48
48
48
References
References
References
References
References ...
...
...
... 49
...
49
49
49
49
Abbreviations
Abbreviations
Abbreviations
Abbreviations
Abbreviations
Arabidopsis Arabidopsis thaliana
ATP; ADP adenosine triphosphate; adenosine diphosphate DCCD dicyclohexylkcarbodiimide
Fm, Fo, Fv maximal, minimal, variable fluorescence
LHC/Lhc light harvesting complex/gene encoding light harvesting complex LHC I light harvesting complex I (composed of Lhca1-4)
LHC II light harvesting complex II (trimers composed of Lhcb1-3 in any combination) Lhca/Lhca light-harvesting proteins of PS I/corresponding genes
Lhcb/Lhcb light-harvesting proteins of PS II/corresponding genes NADPH nicotine adenine dinucleotide phosphate, reduced (oxidised) NPQ non-photochemical quenching of chlorophyll fluorescence P680 the photoactive chlorophyll a molecule in PS II
P700 the photoactive chlorophyll a molecule in PS I PCR polymerase chain reaction
PS I, PS II photosystem I, photosystem II RC reaction centre
T1 (T2, T3) the first (second, third) generation from a transformed plant line TEM transmission electron microscopy
T-DNA the transferable part of the Ti plasmid of Agrobacterium tumefaciens qE feedback de-excitation = ∆pH dependent NPQ
qI irreversible NPQ (photoinhibition)
qT fluorescence reduction due to state transitions VDE; ZE violaxanthin de-epoxidase; zeaxanthin epoxidase
Aim
Aim
Aim
Aim
Aim
Photosystem II antenna proteins have been studied for more than 30 years, since light-harvesting complex (LHC) II was first isolated by Ogawa et al. (1966) and Thornber et al. (1967), but many questions regarding for example the organization of the antenna complexes and the individual functions of each protein remain. Because the LHC protein family is conserved in all higher plants it has been hypothesised that every LHC has a separate function. The first aim of my work was to construct a set of transgenic Arabidopsis lines, each lacking one of the Lhcb proteins. Secondly, by comparing these lines to the wild type I wanted to uncover details about the potentially different functions of each protein, and their positions in the antenna. Specifically we were interested in the mechanism of protective energy dissipation, that has been assumed to take place in the PS II antenna, and of the acclimative response to low growht ligh. We also wanted to evaluate the evolutionary argument that claims importance for each LHC.
This thesis deals mainly with the chlorophyll a/b binding photosynthetic light-harvesting antenna proteins of PS II. It describes the results from successful antisense inhibition of five of the proteins (CP29; CP26; CP24; Lhcb1/Lhcb2), which give insight into the function of these proteins. The impact of the loss of these proteins on plant growth, photosynthesis and fitness was assessed.
Introduction: photosynthesis – the antenna perspective
Introduction: photosynthesis – the antenna perspective
Introduction: photosynthesis – the antenna perspective
Introduction: photosynthesis – the antenna perspective
Introduction: photosynthesis – the antenna perspective
Sunlight is the primary source of energy for life on Earth as we know it. Plants, animals, insects and bacteria are all directly or indirectly dependent on photosynthesis (except for a few lithotrophic organisms living in extreme environments such as in deep ocean hot springs). Photosynthetic organisms utilise light energy to synthesise macromolecules (carbohydrates, amino acids and fatty acids) that are in turn used by other organisms as raw material and fuel for metabolism. The first steps, by which solar energy is converted into high-energy molecules (ATP) and reducing power (NADPH), take place in macromolecular pigment/protein complexes embedded in the thylakoid membrane which, in plants, is situated in the chloroplast (Figures 1 and 2). The enzymes involved in carbon fixation are soluble and located in the chloroplast stroma.
In order to convert the light energy into chemical energy, the first step for the photosynthetic cell is to absorb the photons and safely trap the energy in a more long-lived form. The most common pigment used for energy absorption in higher plants is chlorophyll, with the well-known absorption spectrum that gives our planet its green colour (Figure 3). The photosynthetic pigments are highly co-ordinated by protein complexes in the photosynthetic reaction centres and in the light-harvesting antennae. This enables them to efficiently absorb light energy and transfer it to the reaction centres. Moreover, the light-harvesting antenna has the ability to rapidly switch to an energy-dissipating mode, which is essential to protect the photosynthetic apparatus from over-excitation (Horton et al., 1996; Müller et al., 2001; Niyogi, 1999). In addition to the chlorophylls, the antenna of higher plants also include a number of carotenoids, which are involved both in light harvesting and, perhaps more importantly, in the defence against over excitation (Cogdell and Frank, 1987; Demmig-Adams, 1990; Havaux and Niyogi, 1999). When light hits a chlorophyll molecule in the light-harvesting antenna, absorption of the energy of the photon brings one of the unpaired electrons (Π-electrons) in the conjugated porphyrin
(A)
(B) (C)
(D)
Figure 1. The chloroplast. (A) Transmission electron
micrograph showing the thylakoid membrane ultrastructure,
(B) schematic drawing, (C) schematic grana stack, (D)
STROMA
LUMEN Buchanan et al., 2000, ASPB, printed with permission
Figure 2. The photosynthetic electron transport chain. Excitation of photosystem II (PS
II) by light (hυ) causes charge separation in the reaction centre where the primary radical pair (P680+Pheo-) is formed. P680+ withdraws one electron from a tyrosine residue in the D1
protein which in turn is re-reduced by electrons from the manganese cluster which oxidises water and release protons (H+) and O
2 into the lumen. Pheo- reduces a bound quinone (QA)
which passes the electron on to a second quinone (QB) to form a semiquinol (QB-). In a second
turn, QB is fully reduced to quinol and acquires two H+ from the stroma and diffuses from its
binding site as plastoquinol (PQH2). PQH2 is oxidised in the Q-cycle by the cytochrome b6f
complex (Cyt b6f) that reduces plastocyanin (PC) and release protons into the lumen. In
photosystem I (PS I) light absorption leads to charge separation between P700 and the primary electron acceptor A0 (a chlorophyll). The electron is passed on via phylloquinone (A1) and a number of Fe-S centres (FX, FA and FB) to the soluble Fe-S protein ferredoxin (Fdx). The Fdx-NADP+ reductase (FNR) reduces NADP+ to NADPH with electrons from Fdx and a H+ from the
stroma. P700+ is re-reduced with electrons from PC. The translocation of H+ from the stroma
to the lumen generates a proton motive force that drives phosphorylation of ADP to ATP by the ATP synthase (CF0 CF1). As a summary, solar energy is used to oxidise water to protons, electrons and molecular oxygen. The electrons are converted into reducing power in the form of NADPH. The H+ from water oxidation and the Q-cycle are used to synthesise the high
energy molecule ATP. The next part of photosynthesis is the consumption of NADPH and ATP for the assimilation of CO2 into carbohydrates in the Calvin-Benson cycle (not shown).
ring to an excited energy state (provided the photon is of an appropriate wavelength, see Figure 3). Chlorophyll has two major absorption bands in the visible region, one in the blue and one in the red part of the spectrum. Absorption of red wavelengths excites the electron to the first excited level, Qy. Absorption of blue wavelengths excites the electron to a higher energy level, the Soret transition, which rapidly relaxes to the Qy level dissipating the spare energy as heat. The excitation is passed on to neighbouring chlorophyll by resonance transfer with a high efficiency, and this is repeated until
the excitation reaches the reaction centre. In the reaction centre, charge separation occurs resulting in the transfer of electrons, rather than the transfer of excitation energy, and this is the beginning of the photosynthetic electron transport chain. The chemical environment around the pigment, in this case amino acid residues in the apoprotein that binds the pigment, influences the energy level of the excited states. This is demonstrated in Figure 3 by the differences between the absorption spectra of isolated pigments in solution and the thylakoid membrane preparation. The excitation energy of chlorophyll a is lower than that of chlorophyll b and in the antenna, the excitation energy of pigment molecules decreases slightly with decreasing distance to the reaction centre. This directs the energy towards the reaction centre, regardless where in the antenna the photon was absorbed. However, the energy difference is small enough to allow substantial probability for the excitation to travel in the opposite direction, away from the reaction centre (Schatz et al., 1988).
Organization of the thylakoid membrane
In most higher plants and some green algae, the thylakoid membrane is organized into appressed domains called grana stacks that are interconnected via stroma exposed thylakoids (Figure 1). Grana stacking is dynamic. Rozak et al. (2002) show that the size and number of grana stacks change within minutes of transition between different light conditions. The distribution of photosynthetic complexes in the thylakoid membrane shows a high degree of lateral heterogeneity (Figure 1) which was observed already in 1980 (Andersson and Anderson, 1980). PS II is mainly localised to the appressed membranes (PS II α) and grana margins (PS II β) while PS I is mainly localized to the stroma thylakoids and grana margins. The ATPase show
Wavelength (nm)
Violet Blue Green Yellow Orange Red
Absorption 400 500 600 700 Chlorophyll a Chlorophyll b Carotenoid Thylakoid
Figure 3. Absorption spectra of pigments and pigment-protein complexes. Absorption spectra of
chlorophyll a, chlorophyll b, and carotenoid in nonpolar s o l v e n t s , a n d a t h y l a k o i d p r e pa r a t i o n w h e r e chlorophylls and carotenoids are protein bound.
the same distribution pattern as PS I, and the cytochrome b6f complex is found in all membrane
fractions.
There are several suggestions to explain the occurrence of lateral heterogeneity. Separation of the photosystems in different membrane domains prevent energy drain from PS II to PS I (Anderson, 1999). PS I contains some chlorophylls (the so called red chlorophylls) with lower energy excited state than any of the chlorophylls in PS II. Hence, if the antenna systems of both photosystems were connected, the excitation energy would end up in PS I. Also, energy trapping by PS II is three fold slower than in PS I (Anderson, 1999) which further increases the need for separation of the photosystems. Another advantage of stacked membranes is that it allows close packing of pigment proteins (Anderson, 1999), which is important in light limited conditions where a large antenna system is necessary. It has been observed that low light grown plants have more grana than high light grown plants. Horton (1999) suggests that grana allow interactions to occur between PS II and the PS II antenna both within a thylakoid and between adjacent membranes in the grana stack. This is hypothesised to prevent aggregation of the antenna complexes which could induce a highly dissipative state due to high chlorophyll concentration (Horton et al., 1991). Examination of the structures of the two photosystems gives a steric explanation for their distribution (Allen and Forsberg, 2001). PS I has large stroma exposed subunits which would not fit into the narrow space between adjacent thylakoids in a granum while PS II protrudes only 10 Å at the stroma side of the membrane (Zouni et al., 2001). Also, PS I donates electrons at the stroma side of the membrane to the soluble ferredoxin, which may not have free access to grana stacks while PS II donates electrons to plastoquinone which diffuses in the membrane.
The light harvesting antenna and the Lhc super gene family
Protein composition of the photosystems and their antennae
PS II is a complex structure composed of more than 20 protein subunits (Barber et al., 1997; Hankamer et al., 1997a; Wollman et al., 1999), including the reaction centre (RC; the D1 and D2 proteins with cofactors and the cytochrome b-559) that perform the primary charge separation, the oxygen evolving complex that splits water, the core and outer antennae, that enhances light harvesting and provides most of the excitation energy that powers the other reactions. RC, the oxygen evolving complex and the core antenna form the PS II core complex, which is dimeric in vivo (Wollman et al., 1999) review.
All higher plants investigated so far, have light harvesting antennae composed of the same set of light harvesting proteins. The PS II core antenna consists of chloroplast-encoded polypeptides (CP43 and CP47) binding chlorophyll a and β-carotene. In PS I, the core antenna is located in the same polypeptides as the reaction centre, PSI-A and PSI-B, which are also chloroplast
encoded and bind chlorophyll a and β-carotene. The outer antenna consists of ten different chlorophyll a/b and carotenoid binding (LHC) proteins encoded by a nuclear gene family (Dunsmuir, 1985; Jansson, 1994; Jansson, 1999). The outer antenna of PS I is referred to as LHC I and is composed of hetero dimers of Lhca1 and Lhca4 (LHCI-730) and hetero or homo dimers of Lhca2 and Lhca3 (LHCI-680; Jansson et al., 1996; Schmid et al., 1997). In PS II the outer antenna is composed of the monomeric, minor LHCs CP29, CP26 and CP24 and the trimeric LHC II, which is composed of Lhcb1, Lhcb2 and Lhcb3 (described more thoroughly below). LHC II also associates with PS I and the distribution of LHC II may contribute to balance the excitation level between the two photosystems (Allen, 1992) and is regulated by the so called state transitions (see below). Electron microscopy indicates that LHC I binds to one side of PS I, and LHC II to the other (Boekema et al., 2001), notably to the PSA-H subunit (Lunde et al., 2000).
Tabell 1 The members of the Lhc super-gene family of Arabidopsis thaliana
Gene Protein TAIR EST clone GenBank mRNA Size MSH Ref.
Lhca1 LHCI-730 At3g54890 93I7T7 M85150 15 197 3 1
Lhca2 LHCI-680 At3g61470 32F4T7 AF134120 15 213 3 2
Lhca3 LHCI-680 At1g61520 40G8Y7 U01103 30 232 3 3
Lhca4 LHCI-730 At3g47470 91O23T7 M63931 15 199 3 4
Lhca5 ? At1g45474 177O9T7 AF134121 1 211 ? 2
Lhca6 ? At1g19150 E1H6T7 U03395 1 220 ? 4
Lhcb1.1 Lhcb1 At1g29920 36H7T7 X03907 5 232 3 5 Lhcb1.2 Lhcb1 At1g29910 98N12T7 X03908 5 232 3 5 Lhcb1.3 Lhcb1 At1g29930 138O13T7 X03909 80 232 3 5 Lhcb1.4 Lhcb1 At2g34430 39E1T7 X64459 25 231 3 6 Lhcb1.5 Lhcb1 At2g34420 35F5T7 X64460 40 232 3 6 Lhcb2.1 Lhcb2 At2g05100 31F8T7 AF134122 6 (+1) 228 3 2 Lhcb2.2 Lhcb2 At2g05070 167L10T7 AF134123 8 228 3 2 Lhcb2.3 Lhcb2 At3g27690 227K7T7 AF134125 1 228 3 2 Lhcb3 Lhcb3 At5g54270 98K5T7 AF134126 10 223 3 2 Lhcb4.1 CP29 At5g01530 103O22T7 X71878 20 258 3 7 Lhcb4.2 CP29 At3g08940 20D3T7 AF134127 15 256 3 2 Lhcb4.3 CP29 At2g40100 149G3T7 AF134128 1 244 3 2
Lhcb5 CP26 At4g10340 37A1T7 AF134129 30 243 3 2
Lhcb6 CP24 At1g15820 23A1T7 AF134130 20 211 3 2
PsbS PsbS At1g44575 137M5T7 AF134131 15 205 4 2
Lil1.1 ELIP1 At3g22840 127N22T7 U89014 4 149 3 2
Lil1.2 ELIP2 At4g14690 VCVCD09 AF134132 1 151 3 2
Lil2 HLIP At5g02120 105P6T7 AF054617 1 69 1 2
Lil3.1 ? At4g17600 114M20T7 AF134133 2 ? ? 2
Lil3.2 ? At5g47110 187G12T7 - 1 ? ? 2
Lil4 SEP1 At4g34190 235A5T7 AF133716 1 103 2 8
Lil5 SEP2 At2g21970 212I19T7 AF133717 1 181 2 8
FC Fe-chelatase At2g30390 122F6T7 Y13156 1 ? 1 9
Gene: the common name of the gene; Protein: the common name of the protein; TAIR: accession number in the Arabidopsis genomic database; EST clone: accession number to an EST clone; GenBank: accession number to the full length coding region; mRNA: the number of EST’s found (Jansson, 1999); Size: the size of the deduced mature protein in amino acid residues. 1 Jensen et al., 1992; 2 Jansson et al., 1999; 3 Wang et al., 1994; 4 Zhang et al., 1991; 5 Leutweiler et al., 1986; 6 McGrath et al., 1992; 7 Green et al., 1993; 8 Heddad et al., 2000; 9 Chow et al., 1998.
LHC structure
The atomic structure of Lhcb1/2 has been determined (Figure 4) based on a three-dimensional map at 3.4 Å resolution, obtained by electron microscopy on two-dimensional crystals of a mixture of Lhcb1 and Lhcb2 (Kühlbrandt et al., 1994). The structure of CP26 (Green and Kühlbrandt, 1995) and CP29 (Sandonà et al., 1998) were modelled with the Lhcb1/2 map as a template and it may be assumed that all LHCs share the same basic features of that structure, due to the high conservation of amino acid sequence in membrane spanning regions (Figure 5; Pichersky et al., 1991. The structure comprises three membrane-spanning α-helices in which the first and third helix are held together by ion pairs, and a short, amphiphatic α-helix at the lumenal side of the membrane (not present in CP24).
Proteins with sequence similarity to the LHCs
In addition to the antenna proteins, there are several other proteins, with more or less obscure functions but with sequence similarity to the LHCs, that are grouped together in the Lhc super gene family (Table 1; Jansson, 1999). PsbS has recently been shown to play a major role in protective energy dissipation (Li et al., 2000). The ELIPs are transiently induced during greening of etiolated plants (Meyer, 1984), during desiccation (Bartels 1992) and high light stress (Adamska et al., 1992), and are speculated to be involved in high light protection. HLIP has one membrane spanning helix and share its highest sequence similarity with the cyanobacterial HLIP (Dolganov et al., 1995), and its expression is up-regulated in high light (Jansson et al., 2000). Lil3.1 and Lil3.2 have been found in EST databases (Jansson, 1999) and in the Arabidopsis genome, but the corresponding proteins are unknown. Lil4 and Lil5, encoding
3
1
4
2
(A)
(B)
Buchanan et al., 2000, ASPB, printed with permission
Figure 4. LHC protein structure. (A) The protein structure
of an LHC II monomer indicating the three membrane spanning helices (1-3) and the amphiphatic helix (4) (cf Figure 5). The location of the two central carotenoids and the twelve known chlorophylls are shown. (B) Trimeric LHC II.
Lhcb1.3 MAASTMALSSPAFAGKAVKLSPAASEVLGSGRVTMRKTVAKPKGPSGSPWYGSDRVKYL––GPFS–GESPSYLTGEFPGDYGWDTAGLSAD––––––––––––––– ––– 88 Lhcb2.1 MATSAIQQSSFAGQTALKPSNELLRKVGVSGGGRVTMRRTVKS–––TPQSIWYGPDRPKYL––GPFS–ENTPSYLTGEYPGDYGWDTAGLSAD––––––––––––––– ––– 87 Lhcb3 MASTFTSSSSVLTPTTFLGQTKASSFNPLRDVVSLGSPKYTM––––––––G––NDLWYGPDRVKYL––GPFS–VQTPSYLTGEFPGDYGWDTAGLSAD––––––––––––––– ––– 85 Lhcb4.1 MAATSAAAAAASSIMGTRVAPGIHPGSGRFTAVFGFGKKKAAPKKSAKKTVTTD–RPLWYPGAIS–––––––––––––PDWLDGSLVGDYGFDPFGLGKPAEYLQFDIDSLDQNL AKN 104 Lhcb5 MASLGVSEMLGTPLNFRAVSRSSAPLASSPSTFKTVALFSKKKPAPAKSKAVSETSDELAKWYGPDRRIFLPDGLLDRSEIPEYLNGEVAGDYGYDPFGLGKK––––––––––––––– ––– 103 Lhcb6 MAMAVSGAVLSGLGSSFLTGGKRGATALASGVGTGAQRVGRKTLIVAAAAAQPKKSWIPAVKGGGN–––––––FLDPEWLDGSLPGDFGFDPLGLGKD––––––––––––––– ––– 91 Lhcb1.3 –––––––––––––––––––––––––PETFARNREL E VI H SRWAMLGALGCVFPELLARNGVKFG–EAVWFKAGSQIFSDGGLDYLGNPSLVHAQSILAIWAT Q VILMGAV E GY R VAGNGPLG 184 Lhcb2.1 –––––––––––––––––––––––––PETFAKNREL E VI H SRWAMLGALGCTFPEILSKNGVKFG–EAVWFKAGSQIFSEGGLDYLGNPNLIHAQSILAIWAV Q VVLMGFI E GY R IGG–GPLG 182 Lhcb3 –––––––––––––––––––––––––PEAFAKNRAL E VI H GRWAMLGAFGCITPEVLQKWVRVDFKEPVWFKAGSQIFSEGGLDYLGNPNLVHAQSILAVLGF Q VILMGLV E GF R INGLDGVG 182 Lhcb4.1 LAGDVIGTRTEAADAKSTPFQPYSEVFGIQRFRECELIHGRWAMLATLGALSVEWLTGVT––––––––WQDAGKVELVDGS–SYLGQPLPF–––SISTLIWIEVLVIGYIEFQRNAELD SEK 214 Lhcb5 –––––––––––––––––––––––––PENFAKYQAFELIHARWAMLGAAGFIIPEALNKYGANCGPEAVWFKTGALLLDGNTLNYFGKNIPI–––NLVLAVVAEVVLLGGAEYYRITNGL DFE 197 Lhcb6 –––––––––––––––––––––––––PAFLKWYREAELIHGRWAMAAVLGIFVGQAWSG––––––––VAWFEAGAQPDAIAPF––––––––––––SFGSLLGTQLLLMGWVESKRWVDFF NPD 158 Lhcb1.3 EAEDL–––––––––––––––––––LYPGG–SFDPLGLAT–––––––––DPEAFAELKVK E LK N G R LAMFSMFGFFV Q AIVTGKG–––––––PIENLAD H LADPVNNNAWAFATNFVPGK 267 Lhcb2.1 EGLDP–––––––––––––––––––LYPGG–AFDPLNLAE–––––––––DPEAFSELKVK E LK N G R LAMFSMFGFFV Q AIVTGKG–––––––PIENLFD H LADPVANNAWSYATNFVPGN 265 Lhcb3 EGND––––––––––––––––––––LYPGGQYFDPLGLAD–––––––––DPVTFAELKVK E IK N G R LAMFSMFGFFV Q AIVTGKG–––––––PLENLLD H LDNPVANNAWAFATKFAPGA 265 Lhcb4.1 R–––––––––––––––––––––––LYPGGKFFDPLGLAA–––––––––DPEKTAQLQLAEIKHARLAMVAFLGFAVQAAATGKG–––––––PLNNWATHLSDPLHTTIIDTFSSS 290 Lhcb5 DK––––––––––––––––––––––LHPGG–PFDPLGLAK–––––––––DPEQGALLKVKEIKNGRLAMFAMLGFFIQAYVTGEG–––––––PVENLAKHLSDPFGNNLLTVIAGTAERA PTL 280 Lhcb6 SQSVEWATPWSKTAENFANYTGDQGYPGGRFFDPLGLAGKNRDGVYEPDFEKLERLKLAEIKHSRLAMVAMLIFYFEAGQ–GKTPLGALGL 258 Helix 1 (B) Helix 2 (C) Helix 3 (A) Helix 4 (D) T ransit peptide Figure 5. Alignment of the Arabidop sis Lhcb proteins.
The deduced amino acid sequences of one represent
ative (the most highly expressed
as judged from the occurrence in the EST
database,
cf
T
able 1) of each of the six Lhcb proteins were aligned with the ClustalW program and
thereaf
ter manually adjusted. Four boxes indicate the membrane spanning (1-3) and amphiphatic (4) helices (according to Kühlbra
nd
t et al.,
1994).
The helices are sometimes denoted with the letters
A-D which is also shown. The N-terminal transit peptide is boxed acco
rding to the
putative cleavage site. Bold letters indicate the known chlorophyll ligands in LHC II (Kühlbrand
t et al., 1994).
T
riangles abov
e the alignment
SEP1 and SEP2 respectively, are expressed in response to stress, but the functions are unknown (Heddad and Adamska, 2000). Finally, one of the two Arabidopsis ferrochelatases is included in the family, due to LHC sequence similarity in a region believed to form a membrane spanning helix (Jansson, 1999).
Pigment composition
It is estimated that PS II, including the LHC antenna, binds 150-200 chlorophyll molecules, of which approximately 40 associate with the PS II core. From these numbers it is concluded that 70-80 % of the PS II light energy absorption occurs in the LHC antenna. All LHC proteins bind both chlorophyll a and b but the ratio between them varies and in addition to chlorophyll, the LHC proteins bind various carotenoids (Table 2).
Does PsbS bind pigments?
The data regarding the pigment binding properties of PsbS are ambiguous. Funk et al. (1994 and 1995) showed that it does bind both chlorophyll (mainly chlorophyll a) and carotenoids (including violaxanthin). On the other hand, Dominici et al. (2002) found no pigments on the native protein and also showed that PsbS could be reconstituted without pigments under conditions where LHC proteins recruit pigments. Dominici et al. conclude that either PsbS is not a pigment binding protein, or its pigment binding mechanism is very different from the LHC proteins.
Table 2 Pigment binding stochiometry of the light-harvesting proteins
Protein Chl a Chl b Chl a/b Lutein Violax. Neox. β-car Ref.
nLHC IIa nLHC IIa nLHC IIa 7 6.6 - 5 5.0 - 1.4 1.32 1.4 1.68 1.9 1.8 0.32 0.2 0.2 1.05 1.0 1 0 0 0 1 2 3 nCP29 rCP29 6.8 6 2.0 2 3.40 3.0 0.89 0.9 0.64 1.2 0.47 0.6 0 0 2 4 nCP26 7.5 3.0 2.5 1.2 0.9 1.0 0 2 nCP24 2.7 2.3 1.2 0.53 0.47 <0.01 0 5 rLhca1 rLhca1 5.47 - 1.57 - 3.48 4.0 1 1.81 0.17 1.05 0.20 0.12 0.03 - 6 7 rLhca2 4.99 2.19 2.28 1 0.13 0.07 0.03 6 rLhca3 4.63 0.76 6.14 1 0.14 0.16 0.09 6 rLhca4 rLhca4 5.56 - 2.15 - 2.59 2.3 1 1.5 0.13 0.5 0.09 0 0.03 - 6 7
Average pigment binding of the Lhcb complexes
Protein Chl a Chl b Lutein Violax. Neox.
LHC II trimer 21 15 6 1 3
CP29 7 2 1 1 0 or 1
CP26 7 3 1 1 1
CP24 3 2 0 or 1 0 or 1 0
aper monomer; n – native protein; r – recombinant protein reconstituted with pigments 1 Remelli et al., (1999); 2 Ruban et al., (1999); 3 Croce et al., (1999); 4 Clinque et al., (2000); 5 Pagano et al., (1998); 6 Schmid et al., (2002); 7 Croce et al., (2002)
Supermolecular organization of PS II and its antenna
Different LHC II trimer binding sites
When thylakoid membranes are solubilized with
n-dodecyl-α,D-maltoside, purified on a gel filtration column or by sucrose gradient centrifugation and subjected to transmission electron microscopy (TEM), supercomplexes with outer antennae of varying size can be obtained (Boekema et al., 1995). Three different binding sites for LHC II trimers have been defined and are named after their apparent binding strength to the PS II core: strongly (S), moderately (M) and loosely (L) bound (Boekema et al., 1998; 1999a; 1999b). In each supercomplex six binding possibilities exists for LHC II, two S, two M and two L sites (Figure 6), although it needs to be pointed out that a supercomplex with a full set of six LHC II has never been observed.
Diverse polypeptide composition in LHC II trimers
LHC II itself consists of three distinct proteins (Lhcb1, Lhcb2 and Lhcb3), which in different combinations form the trimers. Lhcb1 and Lhcb2 are encoded by multi-gene families in most species (see Table 1 for Arabidopsis). Several different trimeric complexes have been characterised in terms of their polypeptide composition in maize (De Luca et al., 1999) carnation (Jackowski and Jansson, 1998), Arabidopsis (Jackowski et al., 2001) and spinach (Jackowski and Pielucha, 2001). S-LHC II trimers are composed of Lhcb1 and Lhcb2 in a 2:1 ratio (Hankamer et al., 1997b), and M-LHC II trimers of Lhcb1 and Lhcb3 in a 2:1 ratio (Boekema et al., 1999b). Lhcb1/Lhcb2/Lhcb3 heterotrimers have also been observed (Jackowski et al., 2001).
Association of monomeric Lhcb proteins (the minor antenna)
Several methods have been used to assign the specific locations of the minor antenna proteins in relation to the core complex and LHC II. Cross linking studies in Marchantia polymorpha (Harrer et al., 1998), which is similar to higher plants in terms of PS II, was used to identify the masses seen on TEM and yielded the tentative model shown in Figure 6. In Paper II we provide evidence based on analysis of antisense plants lacking CP26 or CP29, which show that this model is correct. CP29 is most intimately associated with the core complex and seems to be essential for the stability of the supercomplex, since it is present in all preparations (Boekema
Figure 6. Hypothetical supercomplex. Four
LHC II binding sites per PS II have been found. In this figure all of them are shown, although such a supercomplex has never been observed. (Boekema et al., 1999b; printed with permission.)
et al., 1999b) and supercomplexes cannot be isolated from antisense plants lacking this protein (Paper II, see also the Discussion section). CP29 is located in direct contact with CP47 of the inner antenna. CP24 is, as shown previously, believed to be in contact with CP29, but more peripheral from the core complex. CP24 seems loosely attached to the supercomplex, and for stable association, M-LHC II seems necessary (Boekema et al., 1999b). CP26 is located in contact with CP43 in the “corner” of the supercomplex. In order for CP26 to be stable within the supercomplex, binding of S-LHC II is required (Boekema et al., 1999b).
LHC II not directly bound to PS II/peripheral LHC II
The supercomplex described above has six putative LHC II binding sites, although it seems exceptional that all of them are filled. In Arabidopsis, the PS II dimer is associated with no more than four LHC II (Yakushevska et al., 2001). On the other hand, biochemical analyses indicate a stochiometry of up to eight LHC II per core dimer (Dainese and Bassi, 1991; Jansson, 1994; Peter and Thornber, 1991). This leaves a large proportion of LHC II unaccounted for. A peripheral population of LHC II exist, that is not as closely associated with PS II. Several studies show that this subpopulation can be phosphorylated, and participates in the state transition process (further discussed below). Peter and Thornber (1991) saw a multimeric complex of LHC II that was not directly associated with PS II, and more recently (Dekker et al., 1999) observed structures composed of seven LHC II trimers that did not interact with PS II (Figure 7). Furthermore, (Boekema et al., 2000) observed LHC II in membrane domains deficient in PS II.
Association of LHC-related proteins
There is not much information of the locations of the other members of the LHC family that are assumed to belong to PS II. Supercomplexes containing PsbS has been isolated (Eshaghi et al., 1999), which indicates that this protein associates closely with the PS II core. However, in the
M. polymorpha cross linking study mentioned above, supercomplex preparations were depleted
in PsbS compared to grana membranes (Harrer et al., 1998). Moreover, supercomplexes isolated from a mutant (npq4-1) that does not synthesise PsbS appear identical on TEM (Boekema, unpublished), that is, no apparent mass is missing. This indicates that PsbS is not located close to the reaction centre or the inner antenna of PS II.
Figure 7. Heptameric LHC II. A complex
of seven LHC II trimers (that is 21 protein s u b u n i t s ) . T h r e e o f t h e t r i m e r s a r e indicated with tripods. (Dekker et al., 1999)
Supercomplex arrays
In freeze-etch and freeze-fracture studies several authors have observed crystalline areas in the thylakoid membrane. Boekema et al. (2000) isolated paired grana membranes free of stroma membranes and grana margins from spinach. On electron micrographs they found that most, but not all, membranes contained semi-crystalline domains that appeared as rows spaced by 26.3 nm and were shown to consist of supercomplex arrays (for an example see Figure 4 in Paper II). The repeating unit was suggested to be C2S2M supercomplexes, possibly intermixed with some C2S2M2 supercomplexes causing some disorder of the crystal lattice. There was space for one CP24 per PS II dimer in this type of lattice. Some membranes (~1%) had rows spaced by 23 nm and were concluded to consist of arrays of C2S2 supercomplexes. Furthermore, it was found that PS II centres in one membrane domain frequently faced a domain containing exclusively LHC II in the adjacent membrane, and it is suggested that energy transfer may occur between adjacent membranes. In a similar study in Arabidopsis (Yakushevska et al., 2001), the arrays were spaced by 25.6 nm and were concluded to consist of repetitions of C2S2M2. In Papers II and IV (see also Discussion), it is shown that antisense plants lacking certain antenna protein complexes assemble in lattices with some differences from the wild type.
It should be noted that all PS II does not exist in ordered lattices. There is a high degree of heterogeneity in antenna size and composition within one membrane.
Regulatory mechanisms
Under conditions when light energy is present in excess, which means that all energy that is absorbed cannot be converted into electron transport and proton translocation, several detrimental processes may take place (Niyogi, 1999). This occurs for example at increased irradiances, at decreased temperatures or under insufficient CO2 concentrations, which may be induced by stomata closure upon water deficiency. Figure 8 shows the natural variations in light intensity a normal summer day, when clouds and trees occasionally shade the sun. In order to compete successfully with other plants, as much energy as possible must go into photosynthesis, without unnecessary loss, and surplus energy must quickly be dissipated during periods of over excitation. The light-harvesting antenna has the dual capacity for efficient light capture and energy dissipation and this is strictly regulated to avoid destruction by over excitation and still maximise energy transfer during more favourable conditions.
The destructive power of light and oxygen
Singlet oxygen (1O
2), superoxide radicals (O2-•), hydrogen peroxide (H2O2) and hydroxyl radicals
(•OH), collectively called reactive oxygen species, are chemically aggressive molecules that can cause oxidative damage to proteins, pigments and membrane lipids. In addition, the PS II
primary electron donor itself in its oxidised form, P680+, is a powerful oxidising agent (Anderson
et al., 1998). 1O
2 may be formed by interaction of molecular oxygen, which is a triplet in the
ground state, with triplet chlorophyll (3Chl), which can be formed via recombination of P680+
and Pheo- that may occur when the forward electron transport is blocked, or by intersystem
crossing from singlet excited chlorophyll. Another pathway of oxidative damage may occur at very low light intensities. When the time between the first and second charge separation event is long, QA- or Q
B- may recombine with the PS II donor side causing the formation of reactive
oxygen species (Keren et al., 1997).
O2-• is formed in the Mehler reaction, in which PS I uses O
2 as the terminal electron acceptor,
for example when NADP+ is not available. O
2-• can reduce metal ions, such as Fe3+ and Cu2+,
which in turn may react with H2O2 and produce •OH. O2-• is converted to water via H 2O2 by
superoxide dismutatse and ascorbate peroxidase in the water-water cycle (Asada, 1999). Carotenoids protect against 1O
2 in several ways (Cogdell and Frank, 1987). Firstly carotenoids
prevent 1O
2 formation by accepting energy from 3Chl and dissipate it as heat. Secondly by
“scavenging” where carotenoids interact directly with 1O
2 and dissipate the energy as heat.
Carotenoids are also involved in the regulation of feedback de-excitation which is discussed throughout this thesis, which prevents the formation of 3Chl by allowing 1Chl to relax via heat
dissipation. 0 200 400 600 800 1000 1200 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (h) P AR ( µ mol photons m -2s -1) Shady Sunny F i g u r e 8 . Va r i a t i o n s i n n a t u r a l s u n l i g h t . T h e i n t e n s i t y o f t h e s u n l i g h t w a s
recorded during two consecutive 24 h periods at the two locations where our field experiment is conducted. The sunny location is not shaded by any vegetation while in the shady location, the sunlight is filtered through trees and bushes.
Avoiding over excitation
In addition to the protection against reactive oxygen, plants may prevent excitation damage by mechanisms that decrease the excitation pressure. Several regulatory mechanisms have developed to ensure a proper balance between the amount of excitation energy that reaches the RC, and the amount of energy that may be utilised. There are two ways to lower the excitation pressure: either decrease the amount of energy that comes in or increase the amount of energy that goes out. In order to decrease the amount of light intercepted by the antenna, many plant species respond physiologically by adjusting the angle of the leaf towards the sun and relocating the chloroplasts. Plants may also synthesise antocyanins, water-soluble pigments that act as a sunscreen, filtering out wavelengths otherwise absorbed by chlorophyll b (Gould et al., 1995). The long term response by the antenna to high light is to reduce its size (Anderson, 1986; Bailey et al., 2001; Mäenpää and Andersson, 1989). Alternatively, plants can increase the energy usage, either by increased electron transport rate and turnover of ATP and NADPH, or by dissipation of excess energy by the phenomenon known as the qE type of non-photochemical quenching (NPQ; Horton et al., 1996; Niyogi, 1999), which we now prefer to call feedback de-excitation, in which excitation energy is dissipated as heat. Ways to sustain a high electron transport rate during excess light may involve terminal electron acceptors other than CO2, for example O2 in the Mehler reaction, O2 in photorespiration or N and S in amino acid metabolism.
Feedback de-excitation
When the enzymatic reactions that consume ATP cannot keep pace with the light reactions, the trans-thylakoid ∆pH increases because protons that are transported into lumen along with electron transport cannot be released through the ATP synthase in the absence of ADP. The low lumenal pH activates Violaxanthin de-epoxidase (VDE; Rockholm and Yamamoto, 1996), which converts violaxanthin via anteraxanthin to zeaxanthin in the so-called xanthophyll cycle (Figure 9; Demmig-Adams, 1990), and leads to protonation of several antenna polypeptides. Zeaxanthin formation and antenna protonation trigger the feed back de-excitation (qE), thus preventing over excitation and its destructive consequences. The Arabidopsis npq1 mutant that is deficient in VDE and hence has a non-functioning xanthophyll cycle is incapable of inducing qE (Niyogi et al., 1998). It has also been shown that the PsbS protein is essential for this process, because the Arabidopsis npq4 mutant lacking PsbS has no feed back de-excitation (Li et al., 2000). The mechanism of this kind of energy dissipation is not clear, but a tentative model will be presented in the Discussion.
Violaxanthin has 9 conjugated carbon double bonds whereas zeaxanthin has 11. This led to the hypothesis that violaxanthin had a higher S1 state that could not take energy from singlet excited chlorophyll, but zeaxanthin, with its larger conjugated system would have an S1 state low enough to quench chlorophyll excitation (Frank et al., 1994). Although it is tempting to believe
that zeaxanthin acts as a direct dissipater of chlorophyll excitation, while violaxanthin would not have this capacity, it is not likely to be the case. Determination of the energy level of the S1 state for violaxanthin and zeaxanthin shows that both are lower than the Qy state of chlorophyll a (Polívka et al., 1999). This indicates no difference in their capacity for quenching the singlet excited state of chlorophyll a.
Another explanation is that zeaxanthin might bring about a conformational change that induces dissipation (Crofts and Yerkes, 1994; Horton et al., 1996). The shapes of violaxanthin and zeaxanthin are different and may induce different conformations of the protein to which the pigment bind. The end groups of the zeaxanthin molecule are in the plane with the polyene chain whereas the epoxides in violaxanthin
twist the end groups out of the plane (Ruban et al., 1998b). Evidence supporting the modulating role of zeaxanthin comes from in vitro experiments using the non-native xanthophyll auraxanthin (Ruban et al., 1998b). Auraxantin is similar in size to zeaxanthin, but has the shape of violaxanthin and a smaller system of conjugated carbon double bonds (7 double bonds). Ruban et al. show that auraxanthin has the capacity to induce quenching in isolated LHC II similarly to zeaxanthin. This strongly indicates that it is the molecular structure of zeaxanthin that is important for qE, and not the energy level of the S1 state.
State transitions
Several studies have shown the presence of two pools of LHC II, which are often referred to as the inner (or tightly bound) and the peripheral pools (Larsson and Andersson, 1985; Larsson et al., 1987b). The inner pool appears to be more tightly associated with PS II (perhaps corresponding to S, M and L-LHC II). Peripheral LHC II is more loosely associated with PS II and participates in the state transitions (Larsson et al., 1987b). Peripheral LHC II may correspond to the trimers not associated directly with PS II, as described above.
Zeaxanthin
Antheraxanthin
Violaxanthin
High light
Low light
βββββ
-Carotene
VDE
VDE
ZE
ZE
β
-hydroxylase
Figure 9. The xanthophyll cycle. Violaxanthin is
synthesised from β-carotene by β-hydroxylase and zeaxanthin epoxidase (ZE). Under conditions that lead to lumen acidification (indicated to the left by ”High light”), violaxanthin de-epoxidase (VDE) is activated and converts violaxanthin to antheraxanthin, which has one epoxide group, and zeaxanthin, which has no epoxide. When the lumen pH raises (indicated to the right by ”Low light”), VDE is deactivated and ZE convert zeaxanthin and antheraxanthin back to violaxanthin.
PS I and PS II have different absorption spectra, with PS I being red-shifted compared to PS II and even capable of using far-red wavelengths. These discrepancies mean that different light qualities excite the two photosystems unequally. In order for photosynthesis to function optimally, electron transfer between PS II and PS I must be coordinated, which requires regulation of the excitation balance (Allen, 1992). Under conditions that excites PS I more, the mobile pool of LHC II transfer energy to PS II (state 1). Light that excite PS II more, leads to the migration of LHC II to PS I (state 2). One can speculate that state 2 can be advantageous in other conditions as well. For example when the cellular need for ATP is increased, cyclic electron transport around PS I may benefit from increased PS I absorption.
A widely accepted model for the regulation of state transitions is based on phosphorylation of LHC II, which leads to migration from PS II and attachment to PS I (Allen, 1992). In vivo experiment (Andrews et al., 1993) using light enriched in either PS I or PS II wavelengths showed phosphorylation of LHC II in PS II light. Peripheral LHC II is phosphorylated by an LHC II kinase that is activated under conditions that cause a reduced plastoquinone to remain bound to the Qo site of reduced cytochrome b6f complex (Vener et al., 1997), that is for example
when PS I is less excited than PS II. However, regulation of the LHC II kinase is more complex (Hou et al., 2002). (Rintamäki et al., 2000) suggest that the kinase can be inactivated by thioredoxin in high light.
The mechanism that cause the migration has been explained as electrostatic repulsions between the negatively charged phosphate groups or to depend on molecular recognition between the altered conformation of phosphorylated LHC II and PS I. However, other factors besides phosphorylation are likely to be necessary to induce LHC II migration. Antisense plants lacking the PSA-H subunit of PS I (Lunde et al., 2000), that are deficient in state transitions, show a high level of LHC II phosphorylation, although LHC II remains to excite PS II, showing that phosphorylation itself does not cause LHC II to migrate away from PS II.
Nevertheless, antisense plants with reduced activity of a kinase that is shown to phosphorylate LHC II in vitro, has reduced capacity for state transitions (Snyders and Kohorn, 2001), and a
Chlamydomonas state transition mutant was shown to lack LHC II phosphorylation
(Fleischmann et al., 1999; Kruse et al., 1999), indicating that phosphorylation is a prerequisite for state transitions.
Phosphorylation of LHC II in vivo has been shown to be most pronounced in low light (half the growth light intensity) and very low in high light (Pursiheimo et al., 1998; Rintamäki et al., 1997). This may be explained by the emphasised need for ATP in very low light, where the plant does not assimilate carbon so reducing power (NADPH) is not needed, but steady state cellular activities require ATP. Cyclic electron transport generates a pH gradient that drives
ATP formation without NADP+ reduction. Since PS I, but not PS II, is involved in this process,
it is favourable to induce state 2. In addition to the migration of LHC II, state transitions are accompanied with a redistribution of the cytochrome b6f complex (Vallon et al., 1991) with a
larger proportion of cytochrome b6f in the stroma-exposed thylakoids (where PS I is located)
in state 2. This also enhances the capacity for cyclic electron transport. However, there are no direct evidences for enhanced cyclic electron transport in low light. Besides state transitions, phosphorylation of LHC II may have other functions such as regulating the structure of LHC II (Nilsson et al., 1997; Zer et al., 2002).
State transitions cause a small decrease in PS II fluorescence, qT, because the antenna connected to PS II is smaller in state 2 than in state 1, leading to a decrease in absorbed energy in state 2. However, the state transition mechanism is not a high light protection, as for example shown by the low phosphorylation levels of LHC II in high light, but function as a modulation of the relative efficiencies of PS I and PS II.
Acclimation of the light harvesting antenna to different light conditions
In contrast to animal development, which is predominantly genetically determined, plant development is to a significant extent governed by environmental factors. Being sessile, plants need to be able to respond and acclimate to a broad spectrum of growth climates and stress factors that vary at timescales from seasons to seconds. Light has a profound impact at most levels of plant existence, for example leaf size and thickness, stem elongation, mesophyll structure, the number of chloroplasts per cell, the onset of flowering and senescence, as well as net biomass production. Both light quality, intensity and diurnal rhythm are important and are perceived through various signal transduction pathways mediated by for example phytochrome (Neff et al., 2000), cryptochromes (Cashmore et al., 1999) and chloroplast redox potential (Pfannschmidt, 2003).
A comparison of the antenna protein composition in plants from the same species grown in high or low light intensity show that the main difference lies in the amount of the subunits of the peripheral LHC II namely Lhcb1 and Lhcb2 (Anderson, 1986; Bailey et al., 2001; Larsson et al., 1987a), which increase with decreasing intensity of growth light. In addition, plants grown in extremely low light increase the amounts of CP26 and Lhca4 (Bailey et al., 2001), but not to the same extent as LHC II. These discrepancies in protein composition are regulated via altered gene expression. Lhc transcription is induced by phytochromes and cryptochromes in a circadian mode (Anderson and Kay, 1995; Hamazoto et al., 1997; Mazzella et al., 2001). The transcription rate is modulated by signals from the chloroplast but the source and nature of these signals are poorly described (Surpin et al., 2002). Some evidence indicates that the redox state of the PQ pool regulates Lhc expression. Manipulation of the redox state by DCMU that oxidises PQ (mimics low excitation pressure) and DBMIB that reduces PQ (mimics high excitation pressure) have been used to investigate this. In Dunaliella tertiolecta the expression
of Lhc is high in the presence of DCMU, and low in the presence of DBMIB (Escoubas et al., 1995). The cue-1 (cab under-expressed) mutant, which is deficient in the chloroplast envelope phosphoenolpyruvate/phosphate translocator is hypothesised to under-express Lhc as a consequence of a small PQ pool (Streatfield et al., 1999) - the synthesis of PQ is dependent on the shikimate and isopentenyl pathways, to which phosphoenolpyruvate is a substrate. Chlorophyll intermediates have been suggested to act as signals from the chloroplast and it was recently shown that accumulation of Mg-protoporphyrin IX, represses Lhc expression along with a range of other nucleus encoded photosynthesis genes (Strand et al., 2003). Other hypotheses involve reactive oxygen species in the regulation of nuclear photosynthesis genes (Mullineaux and Karpinski, 2002; Rodermel, 2001).
The difference in antenna size and composition influences the chlorophyll content of the leaf both quantitatively and qualitatively. In high light the chlorophyll content per leaf area is higher and because the outer antenna is smaller the chlorophyll a/b ratio is higher (chlorophyll b is only present in the outer antenna).
Plants growing in high light plants invest energy into synthesising more of the electron transport chain subunits and enzymes involved in carbon assimilation. These are rate limiting steps in saturating light that determine the maximum photosynthetic rate (Pmax), hence Pmax is higher for plants grown at high light. Surprisingly, however, it may be difficult to detect differences in the quantum yield of O2 evolution or CO2 uptake in limiting light (Anderson, 1986; Lee et al., 1999).
Methods
Methods
Methods
Methods
Methods
I would like to elaborate on some of the methods that were used, because they were important to the work and may be less well known to the reader.
Model species: Arabidopsis thaliana
All work presented in this thesis was done in the model plant Arabidopsis thaliana (Figure 10). Arabidopsis is naturally found over a large area on earth, and different ecotypes have developed (Figure 10A). A few of them are in use in laboratories over the world, for example Columbia (which I used), Landsberg erecta and Wassilevskija. This annual plant has an extraordinarily short generation time. Depending on growth conditions (mainly day length) it can be as short as six weeks from seed to seed. For photosynthesis experiments, wide leaves and large biomass are beneficial which may be achieved by a short day length (8 h) and moderate light intensity (150 µmol photons m-2 s-1), and under these conditions generation time increases to at least 10
weeks. Arabidopsis is self fertile, often pollinating already before the flower opens, but may be cross-pollinated by hand. Seeds are small, facilitating screening, and keeping, many thousands of them. Of superior importance for my work was the possibility to transform Arabidopsis with exogenous DNA, which is easily done by routine protocols using Agrobacterium tumefaciens.
(A)
(B)
(C)
Figure 10. A r a b i d o p s i s thaliana. In (A) the
distribution of different ecotypes is shown. For my experiments, the plant material usually looked as in (B) which shows a
seven weeks old plant grown at 150 µmol photons m-2s-1 in 8 h photoperiod, 23/18°C day/night temperature. (C) shows a flowering plant and an enlargement of the f l o w e r s ...
Recently the complete genome of Arabidopsis (Columbia) was sequenced (Arabidopsis Genome Initiative, 2000). One significant reason for the choice of Arabidopsis to be fully sequenced was that the genome size is the smallest of all known plant genomes, 7x107 bp, which can be
compared to the largest known plant genome, Fritillaria, 1x1011 bp, and the human genome,
3x109 bp (Ferl and Paul, 2000). The small size is due to little non-gene DNA, but also that
many proteins are encoded by singe copy genes. An advantage given by the compact genome is that it is feasible to saturate the genome with insertion inactivation tags. Determination of the DNA flanking the tag, gives, together with the complete genome sequence, the exact position of the insert and the potentially inactivated gene. In addition to the genomic sequence, a large number of expressed sequence tags (ESTs) have been obtained from different tissues and different growth and stress conditions, which gives information about expressed genes. Knowledge of gene sequences, families and expression patterns gives the opportunity to construct antisense and over-expression lines.
Antisense inhibition
Mutants and transgenic organisms are useful tools to understand biological processes on a molecular level. For a long time, scientists have gone from phenotype to genotype: an aberrant individual was discovered, it was found out what was wrong in it and after tedious crossings and marker analyses the mutated gene could be identified. When gene sequence information started to build up, many genes with unknown function became known. This raised numerous questions about the structure and function of the gene products. With the possibility to create individuals that differ from the wild type only in one known gene, it is often possible to draw some conclusions about the gene function. In bacteria, it is very easy to disrupt gene function by introducing a deletion or insertion via homologous recombination. However, this is not possible in plants, since they lack an efficient homologous recombination process. Instead, a new copy of the gene of interest, in antisense direction, is inserted into the genome. The antisense gene will be transcribed and diminish, and sometimes abolish, translation of the original gene. The gene function may also be disrupted by co-suppression, in which a sense copy of the gene of interest is introduced. For the integration into the genome there are a few methods in use, for example bombarding plant material with DNA coated particles, or Agrobacterium tumefaciens mediated transformation.
Antisense inhibition functions by a mechanism in which double stranded RNA (dsRNA) molecules silence similar genes (for example Hannon, 2002; Hutvagner and Zamore, 2002). In this process, dsRNA is cleaved into ~22 bp fragments (Bernstein et al., 2001) that are incorporated into an enzyme complex and used as a template for the degradation of mRNAs that contain the 22 nt sequence in question. This suggests that efficient antisense inhibition of heterologous genes requires absolute sequence identity in a ~22 nt sequence.
T-DNA tagged mutants
If the gene sequence is known, another way to retrieve a plant with disrupted gene expression is to screen for a favourable T-DNA insertion among the numerous collections of randomly mutated Arabidopsis plants that are available. Either, the screen is performed with gene specific primers that are designed in a way that they generate a PCR product if one gene specific primer is used in combination with a T-DNA specific primer on a template of genomic DNA provided that a T-DNA is inserted into the gene of interest. A PCR fragment is not produced if there is no T-DNA in or close to the gene. Another way is to determine the genomic DNA sequence flanking the insert. In the case of Arabidopsis, where the entire genome has been sequenced, this will tell exactly in which gene the insert is found.
Chlorophyll fluorescence
When a chlorophyll molecule is excited, it has four ways to return to the ground state. 1) transfer of the energy to another chlorophyll molecule, 2) charge separation in the reaction centre leading to electron transport (photochemistry), 3) dissipation of the absorbed energy as heat, or 4) fluorescence, in which the energy is emitted as light, which is possible to detect and quantify. Since all the excitation energy in a leaf will meet one of these four fates, it is possible to deduce a lot of information about photochemistry and quenching from the amount of fluorescence that is emitted when known amounts of light is applied to a leaf (or to a thylakoid or chloroplast suspension) (Maxwell and Johnson, 2000) .
Fv/Fm
A commonly used parameter is Fv/Fm, which is the ratio between the variable and the maximum fluorescence that reflects the optimal photochemical efficiency of PS II. This is recorded on dark-adapted leaves in order to avoid any effects of energy dissipation mechanisms, and the light intensity given to induce maximum fluorescence should be applied as a short (<1s) saturating (that is reducing all PS II centres) pulse.
Non-photochemical quenching (NPQ)
A dark-adapted leaf is neither ready to do photochemistry or energy dissipation, therefore a high amount of fluorescence is obtained when light is applied to such a leaf. When photosynthesis starts, which occurs within a few minutes upon light exposure and reflects the light dependent activation of carbon metabolism enzymes, the amount of fluorescence decreases because more of the excitation energy can be used in photochemistry. The decrease in fluorescence that is caused by photochemistry is termed photochemical quenching. In addition to the induction of photosynthesis, the antenna begins to dissipate energy, especially in excess light, which also decreases the amount of fluorescence. This is commonly denoted non-photochemical quenching of chlorophyll fluorescence, or non-photochemical quenching (NPQ), or sometimes just
quenching. Non-photochemical quenching may be split up into at least three separate components:
NPQ = qT + qE + qI
qT is the reduction in fluorescence due to detachment of LHC II from PS II in the State Transition process. qI is the sustained reduction in quenching that is caused by non-functional PS II and
Table 3 Commonly used fluorescence parameters (See also Figure 11)
φPSII Quantum yield of PSII (Fm´-Ft)/Fm´)
qP Proportion of open PSII (Fm´-Ft)/(Fm´-Fo´)
Fv/Fm Maximum quantum yield of PSII (Fm-Fo)/Fm
NPQ Non-photochemical quenching (Fm-Fm´)/Fm´
NPQS Slowly relaxing NPQ (Fm- Fmr)/ Fmr
NPQF (qE) Rapidly relaxing NPQ (NPQ-NPQS) (Fm/Fm´)-(Fm/Fmr)
Figure 11. Fluorescence trace. This figure shows a chlorophyll fluorescence experiment performed
on an attached leaf. At MB the measuring beam is turned on and the Fo level is recorded as indicated. At SP a short (~800 ms) pulse of saturating light intensity is given and the Fm level is recorded as indicated. At AL on the actinic light, which drives photosynthesis, is turned on. Saturating pulses are given (not indicated) and give the Fm’ value, as indicated. When the actinic light is turned off (AL off) far red light is given to oxidise the electron transport chain and obtain the minimal fluorescence ”in the light”, Fo’. After relaxation in the dark, Fmr is recorded.
Fm... Fo... F o’... ...Fm’ ...Ft MBSP AL on AL off ...Fmr 1 min
