**2. Photosynthesis under stress: How a vital process copes with a permanently changing environment**

Photosynthesis constitutes a highly integrated process involving four multi-subunit membrane-protein complexes: photosystem II (PSII), photosystem I (PSI), cytochrome *b6f*  and F-ATPase (Nelson & Yocum, 2006) that is exquisitely designed to funnel the electron current initiated by photon absorption to the reducing reactions that generate NADPH from NADP+. The process of electron transport also generates an electrochemical proton gradient that powers ATP synthesis. The reductive power and ATP generated are then employed to reduce inorganic sources of carbon, nitrogen and sulfur needed for the synthesis of carbohydrates, amino acids and proteins used to maintain cellular homeostasis and growth respiration for cell division and expansion (Paul & Foyer, 2001). The need to balance the light energy absorbed by the photosystems with the energy consumed by metabolic sinks of the plant renders the photosynthetic process highly sensitive to any disturbing change in environmental conditions, such as fluctuating illumination, limitation of CO2 fixation by low temperatures, salinity or low nutrient or water availability, and biotic stress. Several molecular short-term and long-term acclimation mechanisms are deployed by photosynthetic organisms (predominantly green algae and land plants) to maintain or restore photosynthetic efficiency under adverse conditions and counteract stresses (Öquist & Huner, 2003; Ensminger et al., 2006). On a time scale of minutes, organisms can reduce the efficiency of energy transfer to PSII either by redistributing light energy to PSI at the expense of PSII through state transitions or by dissipating excess energy as heat by nonphotochemical quenching associated with the light-harvesting complex (LHC) antenna (Hüner et al., 1998; Szyszka et al., 2007). Alternatively, it has also been proposed that quenching of excess energy may occur at the reaction centers, in addition to zeaxanthindependent antenna quenching (Krause & Weis, 1991; Bukhov et al., 2001; Matsubara & Chow, 2004; Ivanov et al., 2006). Long-term acclimation responses include alterations in light harvesting antenna size and adjustments of PSI: PSII stoichiometry that balance the excitation light energy absorbed by the two photosystems (Yamazaki et al., 2005; Ozaki et al., 2007, Solanke & Sharma, 2008; Muramatsu et al., 2009). A significant body of experimental evidence indicates that redox signals from photosynthetic electron transport and reactive oxygen species (ROS) or ROS-scavenging molecules play a central role in the regulation of acclimation and stress responses (Foyer & Noctor, 2009; Pfannschmidt et al., 2009; Mubarakshina et al., 2010).

## **2.1 Short-term acclimation mechanisms 2.1.1 State transitions**

156 Artificial Photosynthesis

demanding defense responses, although the concept of a down-regulated photosynthetic apparatus as a protective measure against oxidative damage has also been proposed. Jasmonic acid (JA), frequently interacting with ethylene, is recognized as the primary regulator of the defense response against defoliating insects and is strongly associated with the down-regulation of photosynthesis genes. Conversely, the mechanism(s) responsible for the compensatory photosynthetic activity that allows increased plant growth or fitness after insect herbivory are not well understood. Experimental evidence gathered to date suggests that the onset of compensatory photosynthesis may be dependent on many factors. The timing of herbivory injury, which also influences the plant's source/sink relationships along its phenology, is important. Also influential are the type of tissue damaged, the type and extension of damage, which in turn may be influenced by highly specific factors such as the composition of the herbivore´s saliva and the type of endosimbiotic bacteria colonizing the insect´s gut, the development stage of the herbivore, its feeding guild and even differences within guilds. Finally, the tolerance capacity of the plant to injury, the environmental conditions surrounding the plant, and the type of defense produced, if any, in response to herbivory, will also define whether a given assault on the plant leads to compensatory responses. The suggested role for phytochrome as a regulator of resource allocation between plant growth and anti-herbivore defense also implies its participation as a signaling element in herbivory-dependent changes in photosynthesis. Moreover, emerging data suggest that stress acclimation and chloroplast-to-nucleus signaling is mediated by phytohormones acting via AP2/EREBP transcription factors, which are believed to play a major and

The overall perspective on the varying factors that influence the effect that insect herbivory may have on photosynthesis is that of complexity. This chapter will also concentrate on the description of these multiple factors, including emerging data that have shed light on the poorly understood mechanisms that regulate photosynthesis-related gene expression or that define how herbivore damage is interpreted by the plant, either as a "friendly" jolt to increase photosynthesis and stimulate growth and promote fitness or as an "act of war"

Photosynthesis constitutes a highly integrated process involving four multi-subunit membrane-protein complexes: photosystem II (PSII), photosystem I (PSI), cytochrome *b6f*  and F-ATPase (Nelson & Yocum, 2006) that is exquisitely designed to funnel the electron current initiated by photon absorption to the reducing reactions that generate NADPH from NADP+. The process of electron transport also generates an electrochemical proton gradient that powers ATP synthesis. The reductive power and ATP generated are then employed to reduce inorganic sources of carbon, nitrogen and sulfur needed for the synthesis of carbohydrates, amino acids and proteins used to maintain cellular homeostasis and growth respiration for cell division and expansion (Paul & Foyer, 2001). The need to balance the light energy absorbed by the photosystems with the energy consumed by metabolic sinks of the plant renders the photosynthetic process highly sensitive to any disturbing change in environmental conditions, such as fluctuating illumination, limitation of CO2 fixation by low temperatures, salinity or low nutrient or water availability, and biotic stress. Several

diversified role in environmental signal integration.

leading to austerity measures to privilege a defense effort.

**permanently changing environment** 

**2. Photosynthesis under stress: How a vital process copes with a** 

State transitions represent a short-term response, occurring within a time-frame of seconds to minutes, required to balance the light excitation energy between the antennae systems of PSII and PSI which preferentially absorb 650 and 700 nm light, respectively. Because of these differences in light absorption properties, changes in light conditions, such as those happening under shaded or light-limiting conditions, or as a consequence of shifts in the spectral filtering properties of leaf canopies, can lead to unequal excitation of the two photosystems (Allen & Forsberg, 2001; Haldrup et al., 2001; Wollman, 2001; Dietzel et al., 2008; Pesaresi et al., 2010). Preferential excitation of PSI leads to the oxidation of the plastoquinone (PQ) pool and to state 1. In state 1 the mobile light-harvesting antenna is bound to PSII and the photosynthetic electron transport chain acts mostly in a linear mode generating NADPH and ATP. Preferential excitation of PSII relative to PSI leads to a reduced state of the PQ pool and thus to the docking of plastoquinol to the Qo site of the cytochrome *b6f* complex. It is the PQ redox state only, acting independently of photoreceptors (Fey et al., 2005), that then activates the so-called redox-sensitive thylakoid LHCII kinase needed to phosphorylate the peripheral LHCII "mobile pool", which then migrates laterally, as a consequence of charge repulsion, from PSII to PSI (state 2) (Lunde et al., 2000). Chemical cross-linking and RNA interference approaches performed in Arabidopsis plants maintained in state 2 have provided evidence for an association of LHCII polypeptides to a specific PSI docking domain composed of subunits PsaL, PsaH and PsaO (Lunde et al., 2000; Zhang et al., 2004; Pesaresi et al., 2010), whereas a gentle mechanical fractionation of the thylakoid membranes showed that the lateral movement of phosphorylated LHCII might be confined to a very limited portion of the thylakoid membranes, more precisely to the grana margins (Tikkanen et al., 2008). Regardless of the mechanism, the re-distribution of the light-harvesting chlorophyll to PSI at the expense of

Friend or Foe? Exploring the Factors that Determine the Difference

Lemeille et al., 2009)

independent phosphorylation sites.

Between Positive and Negative Effects on Photosynthesis in Response to Insect Herbivory 159

The recent characterization of Stt7 in *C. reinhardtii* revealed a structural organization in which a transmembrane helix separates its stroma-exposed catalytic domain from its lumenlocated N-terminal end. This organization permits the co-localization of the catalytic site with the target sites on the LHCII proteins. It also identified two conserved cysteine residues that are critical for its activity (Lemeille et al., 2009). In addition, co-immunoprecipitation assays have shown that Stt7 interacts with Cyt *b6f*, PSI and LHCII, suggesting that all these protein complexes might be clustered together, possibly in very restricted areas of thylakoid membranes, such as the grana margins (Tikkanen et al., 2008; Lemeille et al., 2009; see above). The curious fact that the Qo site of the cytochrome *b6f* complex, which is critical for the activation of the kinase, is on the lumen side necessarily implies that a signal for kinase activation needs to be transported across the membrane. A mobile Rieske protein (Zhang et al., 1998; Breyton, 2000) and subunit V (also called PetO) from the cytochrome *b6f* complex, which is the only protein of the complex capable of under-going reversible phosphorylation during state transitions (Hamel et al., 2000), have been proposed as possible signal transducing candidates in a pathway model that suggests that sensing of the structural changes of the Rieske protein by the lumenal domain of PetO is transmitted through its trans-membrane region to its stromal domain in order to allow its interaction with the kinase. Another interesting feature found in Stt7 and STN7 involves the presence of two conserved Cys residues near the N-terminal end which could be the targets of thioredoxin (Rintamaki et al., 2000). The loss of state transitions and LHCII phosphorylation as the result of site-directed mutagenesis of either of the conserved Cys residues in both Stt7 and STN7 strongly suggests that these residues play an important role in the activation of the kinase. It is also likely that the high-light-induced reduction of this bond may occur through a transthylakoid thiol-reducing pathway driven by the ferredoxin-thioredoxin system which is also required for cytochrome *b6f* assembly and heme biogenesis (Lemeille et al., 2009). Therefore, it appears likely that STN7 kinase activity is regulated not by PQ alone, but by a complex network involving co-operative redox control by PQ and the Cyt *b6f* complex, as well as by the ferredoxin/ thioredoxin system in the stroma of the chloroplasts (Rintamaki et al., 2000;

Although it is clear that the Stt7/STN7 kinase is required for LHCII phosphorylation and for state transitions, it is not yet known whether it acts in a kinase cascade or recognizes LHCII as its direct substrate. Some of these uncertainties were dispelled by the findings of a recent study that compared the thylakoid phosphoproteome of the wild-type strain and the *stt7*  mutant of *C. reinhardtii* under state 1 and state 2 conditions (Lemeille et al., 2010). The study revealed that under state 2 conditions several Stt7-dependent phosphorylations occur in the Lhcbm1/Lhcbm10, Lhcbm4/Lhcbm6/Lhcbm8/Lhcbm9, Lhcbm3, Lhcbm5, and CP29 proteins located at the interface between PSII and its light-harvesting system. One of the two Stt7-dependent phosphorylation sites detected specifically in CP29 under state 2 was proposed to play a crucial role in the dissociation of CP29 from PSII and/or in its association to PSI where it serves as a docking site for LHCII in state 2. Moreover, the Stt7-dependent phosphorylation of the thylakoid protein kinase Stl1 under state 2 conditions, suggested the existence of a thylakoid protein kinase cascade. Curiously, the auto-phosphorylation of Stt7 in state 2, was found not to be required for state transitions. Additional findings included the identification of redox (or state 2)-dependent but Stt7-independent, and redox-

PSII results in a balanced excitation of PSII and PSI to ensure optimal quantum efficiency for photosynthetic electron transport. Under PQ oxidizing conditions the LHCII kinase is inactive, LHCII is or becomes dephosphorylated and is relocated to PSII (state I). The identity of the redox-independent and constitutively active protein phosphatase that presumably dephosphorylates LHCII during the transition from State 2 to 1 was the subject of intense research and remained unknown until recently. The search finally yielded fruit with the identification of a LHCII-specific phosphatase, called PPH1/TAP38. This enzyme dephosphorylates LHCII upon a transition from state 2 to state 1 (Pribil et al., 2010; Shapiguzov et al., 2010), by specifically dephosphorylating the major trimeric Lhcb1 and Lhcb2 proteins. It is a chloroplast protein that is mainly associated with the stromal membranes of the thylakoid membranes and belongs to the family of monomeric PP2C type phosphatases. Its regulatory role was demonstrated in experiments where the loss of PPHI/TAP38 gave rise to an increase in the antenna size of PSI and strongly impaired state transitions.

### **2.1.2 The nature of the redox-sensitive thylakoid LHCII kinase**

LHCII kinase activity was first reported in 1977 (Bennett, 1977, 1979), a finding that also triggered an intensive search for its identity. A first approximation of its nature came from a screening for proteins capable of interacting with the N-terminal region of the lightharvesting proteins known to contain the amino-acid targets for phosphorylation during states 1–2 transition. This approach led to the identification of a small family of three kinases, called TAK kinases (for thylakoid associated kinases) in *Arabidopsis thaliana* (Snyders & Kohorn, 1999), whose exact role in state transitions regulation remains undefined until now, notwithstanding biochemical and genetic experimental evidence showing that TAKs do indeed participate in LHCII phosphorylation (Snyders & Kohorn, 2001). Moreover, the failure to identify TAK orthologs in the green motile unicellular alga *Chlamydomonas reinhardtii* either suggests that the TAK kinases perform a role which is specific to land plants or that a considerable diversion of these kinases happened as a result of evolutionary divergence between green algae and plants.

Later, advantage was taken of the large chlorophyll fluorescence changes that occur during a transition from states 1 to 2 in *C*. *reinhardtii* (that can be efficiently measured with a fluorescence video imaging system) to screen insertional mutants that could lead to the identification of the LHCII kinase and other factors of the signal transduction pathway leading to state transitions (Fleischmann et al., 1999; Kruse et al., 1999). This strategy permitted the isolation of the *state transition-deficient mutant 7* (*stt7*), which was found to encode a thylakoid-associated Ser-Thr protein kinase, and of another protein kinase of unknown function but related to Stt7, called Stl1. Orthologs of these two proteins, called STN7 and STN8, respectively, were subsequently found in Arabidopsis, rice and in marine algae (Depége et al., 2003). Utilization of Arabidopsis T-DNA insertion lines with disruptions in the *STN7* or *STN8* genes helped elucidate the function of these proteins. Thus, STN7 was found to be required for state transitions and for the specific phosphorylation, under state 2 conditions, of several LHCII proteins that did not include the major thylakoid proteins CP43, D1 and D2 (Bellafiore et al., 2005). An additional site directed mutagenesis approach provided conclusive evidence demonstrating that the kinase activity of STN7 is essential for state transitions. It is now known that Stt7 and STN7 are both structurally and functionally related.

PSII results in a balanced excitation of PSII and PSI to ensure optimal quantum efficiency for photosynthetic electron transport. Under PQ oxidizing conditions the LHCII kinase is inactive, LHCII is or becomes dephosphorylated and is relocated to PSII (state I). The identity of the redox-independent and constitutively active protein phosphatase that presumably dephosphorylates LHCII during the transition from State 2 to 1 was the subject of intense research and remained unknown until recently. The search finally yielded fruit with the identification of a LHCII-specific phosphatase, called PPH1/TAP38. This enzyme dephosphorylates LHCII upon a transition from state 2 to state 1 (Pribil et al., 2010; Shapiguzov et al., 2010), by specifically dephosphorylating the major trimeric Lhcb1 and Lhcb2 proteins. It is a chloroplast protein that is mainly associated with the stromal membranes of the thylakoid membranes and belongs to the family of monomeric PP2C type phosphatases. Its regulatory role was demonstrated in experiments where the loss of PPHI/TAP38 gave rise to an increase in the antenna size of PSI and strongly impaired state

LHCII kinase activity was first reported in 1977 (Bennett, 1977, 1979), a finding that also triggered an intensive search for its identity. A first approximation of its nature came from a screening for proteins capable of interacting with the N-terminal region of the lightharvesting proteins known to contain the amino-acid targets for phosphorylation during states 1–2 transition. This approach led to the identification of a small family of three kinases, called TAK kinases (for thylakoid associated kinases) in *Arabidopsis thaliana* (Snyders & Kohorn, 1999), whose exact role in state transitions regulation remains undefined until now, notwithstanding biochemical and genetic experimental evidence showing that TAKs do indeed participate in LHCII phosphorylation (Snyders & Kohorn, 2001). Moreover, the failure to identify TAK orthologs in the green motile unicellular alga *Chlamydomonas reinhardtii* either suggests that the TAK kinases perform a role which is specific to land plants or that a considerable diversion of these kinases happened as a result

Later, advantage was taken of the large chlorophyll fluorescence changes that occur during a transition from states 1 to 2 in *C*. *reinhardtii* (that can be efficiently measured with a fluorescence video imaging system) to screen insertional mutants that could lead to the identification of the LHCII kinase and other factors of the signal transduction pathway leading to state transitions (Fleischmann et al., 1999; Kruse et al., 1999). This strategy permitted the isolation of the *state transition-deficient mutant 7* (*stt7*), which was found to encode a thylakoid-associated Ser-Thr protein kinase, and of another protein kinase of unknown function but related to Stt7, called Stl1. Orthologs of these two proteins, called STN7 and STN8, respectively, were subsequently found in Arabidopsis, rice and in marine algae (Depége et al., 2003). Utilization of Arabidopsis T-DNA insertion lines with disruptions in the *STN7* or *STN8* genes helped elucidate the function of these proteins. Thus, STN7 was found to be required for state transitions and for the specific phosphorylation, under state 2 conditions, of several LHCII proteins that did not include the major thylakoid proteins CP43, D1 and D2 (Bellafiore et al., 2005). An additional site directed mutagenesis approach provided conclusive evidence demonstrating that the kinase activity of STN7 is essential for state transitions. It is now known that Stt7 and STN7 are both structurally and

**2.1.2 The nature of the redox-sensitive thylakoid LHCII kinase** 

of evolutionary divergence between green algae and plants.

transitions.

functionally related.

The recent characterization of Stt7 in *C. reinhardtii* revealed a structural organization in which a transmembrane helix separates its stroma-exposed catalytic domain from its lumenlocated N-terminal end. This organization permits the co-localization of the catalytic site with the target sites on the LHCII proteins. It also identified two conserved cysteine residues that are critical for its activity (Lemeille et al., 2009). In addition, co-immunoprecipitation assays have shown that Stt7 interacts with Cyt *b6f*, PSI and LHCII, suggesting that all these protein complexes might be clustered together, possibly in very restricted areas of thylakoid membranes, such as the grana margins (Tikkanen et al., 2008; Lemeille et al., 2009; see above). The curious fact that the Qo site of the cytochrome *b6f* complex, which is critical for the activation of the kinase, is on the lumen side necessarily implies that a signal for kinase activation needs to be transported across the membrane. A mobile Rieske protein (Zhang et al., 1998; Breyton, 2000) and subunit V (also called PetO) from the cytochrome *b6f* complex, which is the only protein of the complex capable of under-going reversible phosphorylation during state transitions (Hamel et al., 2000), have been proposed as possible signal transducing candidates in a pathway model that suggests that sensing of the structural changes of the Rieske protein by the lumenal domain of PetO is transmitted through its trans-membrane region to its stromal domain in order to allow its interaction with the kinase. Another interesting feature found in Stt7 and STN7 involves the presence of two conserved Cys residues near the N-terminal end which could be the targets of thioredoxin (Rintamaki et al., 2000). The loss of state transitions and LHCII phosphorylation as the result of site-directed mutagenesis of either of the conserved Cys residues in both Stt7 and STN7 strongly suggests that these residues play an important role in the activation of the kinase. It is also likely that the high-light-induced reduction of this bond may occur through a transthylakoid thiol-reducing pathway driven by the ferredoxin-thioredoxin system which is also required for cytochrome *b6f* assembly and heme biogenesis (Lemeille et al., 2009). Therefore, it appears likely that STN7 kinase activity is regulated not by PQ alone, but by a complex network involving co-operative redox control by PQ and the Cyt *b6f* complex, as well as by the ferredoxin/ thioredoxin system in the stroma of the chloroplasts (Rintamaki et al., 2000; Lemeille et al., 2009)

Although it is clear that the Stt7/STN7 kinase is required for LHCII phosphorylation and for state transitions, it is not yet known whether it acts in a kinase cascade or recognizes LHCII as its direct substrate. Some of these uncertainties were dispelled by the findings of a recent study that compared the thylakoid phosphoproteome of the wild-type strain and the *stt7*  mutant of *C. reinhardtii* under state 1 and state 2 conditions (Lemeille et al., 2010). The study revealed that under state 2 conditions several Stt7-dependent phosphorylations occur in the Lhcbm1/Lhcbm10, Lhcbm4/Lhcbm6/Lhcbm8/Lhcbm9, Lhcbm3, Lhcbm5, and CP29 proteins located at the interface between PSII and its light-harvesting system. One of the two Stt7-dependent phosphorylation sites detected specifically in CP29 under state 2 was proposed to play a crucial role in the dissociation of CP29 from PSII and/or in its association to PSI where it serves as a docking site for LHCII in state 2. Moreover, the Stt7-dependent phosphorylation of the thylakoid protein kinase Stl1 under state 2 conditions, suggested the existence of a thylakoid protein kinase cascade. Curiously, the auto-phosphorylation of Stt7 in state 2, was found not to be required for state transitions. Additional findings included the identification of redox (or state 2)-dependent but Stt7-independent, and redoxindependent phosphorylation sites.

Friend or Foe? Exploring the Factors that Determine the Difference

(Pesaresi et al., 2010).

(Fuhrmann et al., 2009).

photosystem stoichiometry, respectively).

Between Positive and Negative Effects on Photosynthesis in Response to Insect Herbivory 161

dissociation from the thylakoid membrane upon phosphorylation (Pesaresi et al., 2009), showed that neither LHCII phosphorylation, nor the conformational changes in the thylakoid associated with state transitions themselves, appear to play any role in LTR (Carlberg et al., 2003; Pesaresi et al., 2009). This argued against the possibility that the signal pathways leading to state transitions and LTR were part of a hierarchically organized signaling cascade, with changes in PQ redox state first triggering state transitions and then LTR, via a STN7-dependent phosphorylation cascade. In order to conciliate the above data, an alternative hypothesis proposing that the PQ redox state must reach a still undefined threshold value to be able to induce the specific, and reversible, STN7-dependent phosphorylation steps that trigger the signaling events leading to LTR was considered

In most species investigated, the re-adjustment of photosystem stoichiometry involves an enhanced expression of the PSI reaction-center genes *psaA* and *psaB* (which encode the P700 apoproteins) upon active reduction of the PQ pool or repression of its oxidation. LTRs are also known to involve the regulated expression of the PSII reaction-center gene *psbA* (encoding the D1 protein) (Pfannschmidt, 2003) and changes in several other physiological and molecular parameters, including the chlorophyll *a/b* ratio, steady state chlorophyll fluorescence and structural modifications of the thylakoid membrane system (Bonardi et al., 2005; Tikkanen et al., 2006). Several proteins were recognized as possible regulators of photosystem stoichiometry in the cyanobacteria *Synechocystis* sp., including photomixotrophic growth-related and CO2-concentrating-mechanism proteins, a probable esterase, cytochrome c oxidase subunits II and III and a hypothetical protein with a von Willebrand factor type A domain. The latter suggested a role for protein-protein interactions in the regulation of photosystem stoichiometry in these organisms (Ozaki et al., 2007). In addition, the depletion of the vesicle inducing protein in plastids 1 (Vipp1), believed to be essential for thylakoid membrane formation in Arabidopsis and cyanobacteria, was found to negatively affect photosystem stoichiometry in *Synechocystis* sp. This effect was associated with a concerted decrease in the number of thylakoid layers and associated photosystem I (PSI) complexes in individual cyanobacterial cells, and an enrichment of PSI monomeric species resulting from of PSI trimer destabilization

More recently, photosystem stoichiometry adjustment in plants and algae, was found to be governed by a modified two-component sensor kinase of cyanobacterial origin, known as chloroplast sensor kinase (CSK) (Puthiyaveetil et al., 2011), acting together with chloroplast sigma factor 1 (SIG1) and a plastid transcription kinase (PTK). These findings confirmed previous data implicating CSK as a control of chloroplast gene expression (Puthiyaveetil et al., 2008), via its role as a sensor of the PQ redox state. Moreover, they confirmed the concept assigning different signaling pathways to state transitions and photosystem stoichiometry adjustments, with the two pathways sensing PQ redox state independently of each other (i.e the reduced and oxidized forms of the quinone recognized for state transitions and

The LTR is also accompanied by dynamic changes in metabolite pools that depend to the prevailing illumination (Bräutigam et al., 2009). For instance, the propagation of plants under PSI-specific light is known to cause a lower accumulation of transitory starch. Moreover, contrasting light conditions have been observed to exert different co-regulation effects on biosynthetic pathways for organic acids and several amino acids linked to

The existence of the conserved STN7/ STN8 and Stt7/Stl1 kinase couples in Arabidopsis and *Chlamydomonas* also suggests a possible functional interaction between STN7/Stt7 and STN8/Stl1. This is in accordance with data proposing that these proteins appear to act synergistically, since the de-phosphorylation phenotype of LHCII and PSII core proteins in the double mutant *stn7*/*stn8* is more pronounced than than those observed in the two single mutants (Bonardi et al., 2005; Vainonen et al., 2005). Moreover, field tests revealed that fitness, as measured by seed production, was significantly decreased in the double mutant whereas it was decreased to a smaller extent in *stn7* and not significantly affected in *stn8* mutants, respectively (Frenkel et al., 2007).

#### **2.1.3 The importance of state transitions in flowering plants**

The magnitude of state transitions is much larger in *C. reinhardtii* than in flowering plants, where displacements have been reported to involve up to 85% of the LHCII antenna from PSII in State 2. In contrast, only 20-30% of the total LHCII is mobile in green plants. Moreover, state transitions in green algae represent a unique adaptive mechanism that allows the organism to switch between linear (State 1) and cyclic (State 2) electron flow through PSI (Finazzi et al., 2001), whereas *C. reinhardtii* mutants unable to undergo state transitions, such as *stt7*, exhibit altered photosynthetic performance and a marked decrease in growth rate (Depége et al., 2003). Conversely, plant development and fitness under laboratory and field conditions have been found to be only marginally affected in Arabidopsis mutants impaired in state transitions (Lunde et al., 2000; Bonardi et al., 2005; Bellafiore et al., 2005; Frenkel et al., 2007). However, a marked decrease in growth rate relative to the parental single mutants, which was accompanied by a consistent drop in the effective quantum yield of PSII and an increase in the reduction state of the PQ pool, was detected in double Arabidopsis mutants affected both in the linear electron transport leading to an increased pool of reduced PQ (i.e. *psad1-1* and *psae1-3*), and state transitions (i.e. *stn7-1 or psal-1*) (Lunde et al., 2000; Pesaresi et al., 2009). This behavior implied that state transitions become critical for plant performance when linear electron flow is perturbed. Further spectroscopic analyses performed on the different genotypes led to the conclusion that, in flowering plants as in green algae, state transitions play an important role in balancing energy distribution between photosystems.

#### **2.2 Long Term acclimation Responses (LTR)**

#### **2.2.1 LTR mechanisms**

Besides inducing short term acclimation processes such as state transitions, changes in light conditions are known to lead to long term responses (LTR) characterized by changes in the amounts of the antenna proteins of PSII and PSI and in photosystem stoichiometry. These changes are implemented over periods lasting hours or days (Dietzel et al., 2008). This process is achieved through a signaling network involving coordinate gene expression in the nucleus and chloroplast (Pfannschmidt, 2003; Pfannschmidt et al., 2009). Most experimental evidence gathered to date indicates that STN7 is also required for triggering LTR (Allen & Pfannschmidt, 2000; Bonardi et al., 2005; Tikkanen et al., 2006), suggesting a dual role for STN7, acting as a common redox sensor and/or signal transducer for both state transitions and LTR responses. However, experimental evidence obtained with mutant or silenced Arabidopsis lines affected in various components required for state transitions (including the novel TSP9 protein, suggested to function in the signaling pathway due to its partial

The existence of the conserved STN7/ STN8 and Stt7/Stl1 kinase couples in Arabidopsis and *Chlamydomonas* also suggests a possible functional interaction between STN7/Stt7 and STN8/Stl1. This is in accordance with data proposing that these proteins appear to act synergistically, since the de-phosphorylation phenotype of LHCII and PSII core proteins in the double mutant *stn7*/*stn8* is more pronounced than than those observed in the two single mutants (Bonardi et al., 2005; Vainonen et al., 2005). Moreover, field tests revealed that fitness, as measured by seed production, was significantly decreased in the double mutant whereas it was decreased to a smaller extent in *stn7* and not significantly affected in *stn8*

The magnitude of state transitions is much larger in *C. reinhardtii* than in flowering plants, where displacements have been reported to involve up to 85% of the LHCII antenna from PSII in State 2. In contrast, only 20-30% of the total LHCII is mobile in green plants. Moreover, state transitions in green algae represent a unique adaptive mechanism that allows the organism to switch between linear (State 1) and cyclic (State 2) electron flow through PSI (Finazzi et al., 2001), whereas *C. reinhardtii* mutants unable to undergo state transitions, such as *stt7*, exhibit altered photosynthetic performance and a marked decrease in growth rate (Depége et al., 2003). Conversely, plant development and fitness under laboratory and field conditions have been found to be only marginally affected in Arabidopsis mutants impaired in state transitions (Lunde et al., 2000; Bonardi et al., 2005; Bellafiore et al., 2005; Frenkel et al., 2007). However, a marked decrease in growth rate relative to the parental single mutants, which was accompanied by a consistent drop in the effective quantum yield of PSII and an increase in the reduction state of the PQ pool, was detected in double Arabidopsis mutants affected both in the linear electron transport leading to an increased pool of reduced PQ (i.e. *psad1-1* and *psae1-3*), and state transitions (i.e. *stn7-1 or psal-1*) (Lunde et al., 2000; Pesaresi et al., 2009). This behavior implied that state transitions become critical for plant performance when linear electron flow is perturbed. Further spectroscopic analyses performed on the different genotypes led to the conclusion that, in flowering plants as in green algae, state transitions play an important role in

Besides inducing short term acclimation processes such as state transitions, changes in light conditions are known to lead to long term responses (LTR) characterized by changes in the amounts of the antenna proteins of PSII and PSI and in photosystem stoichiometry. These changes are implemented over periods lasting hours or days (Dietzel et al., 2008). This process is achieved through a signaling network involving coordinate gene expression in the nucleus and chloroplast (Pfannschmidt, 2003; Pfannschmidt et al., 2009). Most experimental evidence gathered to date indicates that STN7 is also required for triggering LTR (Allen & Pfannschmidt, 2000; Bonardi et al., 2005; Tikkanen et al., 2006), suggesting a dual role for STN7, acting as a common redox sensor and/or signal transducer for both state transitions and LTR responses. However, experimental evidence obtained with mutant or silenced Arabidopsis lines affected in various components required for state transitions (including the novel TSP9 protein, suggested to function in the signaling pathway due to its partial

mutants, respectively (Frenkel et al., 2007).

**2.1.3 The importance of state transitions in flowering plants**

balancing energy distribution between photosystems.

**2.2 Long Term acclimation Responses (LTR)** 

**2.2.1 LTR mechanisms** 

dissociation from the thylakoid membrane upon phosphorylation (Pesaresi et al., 2009), showed that neither LHCII phosphorylation, nor the conformational changes in the thylakoid associated with state transitions themselves, appear to play any role in LTR (Carlberg et al., 2003; Pesaresi et al., 2009). This argued against the possibility that the signal pathways leading to state transitions and LTR were part of a hierarchically organized signaling cascade, with changes in PQ redox state first triggering state transitions and then LTR, via a STN7-dependent phosphorylation cascade. In order to conciliate the above data, an alternative hypothesis proposing that the PQ redox state must reach a still undefined threshold value to be able to induce the specific, and reversible, STN7-dependent phosphorylation steps that trigger the signaling events leading to LTR was considered (Pesaresi et al., 2010).

In most species investigated, the re-adjustment of photosystem stoichiometry involves an enhanced expression of the PSI reaction-center genes *psaA* and *psaB* (which encode the P700 apoproteins) upon active reduction of the PQ pool or repression of its oxidation. LTRs are also known to involve the regulated expression of the PSII reaction-center gene *psbA* (encoding the D1 protein) (Pfannschmidt, 2003) and changes in several other physiological and molecular parameters, including the chlorophyll *a/b* ratio, steady state chlorophyll fluorescence and structural modifications of the thylakoid membrane system (Bonardi et al., 2005; Tikkanen et al., 2006). Several proteins were recognized as possible regulators of photosystem stoichiometry in the cyanobacteria *Synechocystis* sp., including photomixotrophic growth-related and CO2-concentrating-mechanism proteins, a probable esterase, cytochrome c oxidase subunits II and III and a hypothetical protein with a von Willebrand factor type A domain. The latter suggested a role for protein-protein interactions in the regulation of photosystem stoichiometry in these organisms (Ozaki et al., 2007). In addition, the depletion of the vesicle inducing protein in plastids 1 (Vipp1), believed to be essential for thylakoid membrane formation in Arabidopsis and cyanobacteria, was found to negatively affect photosystem stoichiometry in *Synechocystis* sp. This effect was associated with a concerted decrease in the number of thylakoid layers and associated photosystem I (PSI) complexes in individual cyanobacterial cells, and an enrichment of PSI monomeric species resulting from of PSI trimer destabilization (Fuhrmann et al., 2009).

More recently, photosystem stoichiometry adjustment in plants and algae, was found to be governed by a modified two-component sensor kinase of cyanobacterial origin, known as chloroplast sensor kinase (CSK) (Puthiyaveetil et al., 2011), acting together with chloroplast sigma factor 1 (SIG1) and a plastid transcription kinase (PTK). These findings confirmed previous data implicating CSK as a control of chloroplast gene expression (Puthiyaveetil et al., 2008), via its role as a sensor of the PQ redox state. Moreover, they confirmed the concept assigning different signaling pathways to state transitions and photosystem stoichiometry adjustments, with the two pathways sensing PQ redox state independently of each other (i.e the reduced and oxidized forms of the quinone recognized for state transitions and photosystem stoichiometry, respectively).

The LTR is also accompanied by dynamic changes in metabolite pools that depend to the prevailing illumination (Bräutigam et al., 2009). For instance, the propagation of plants under PSI-specific light is known to cause a lower accumulation of transitory starch. Moreover, contrasting light conditions have been observed to exert different co-regulation effects on biosynthetic pathways for organic acids and several amino acids linked to

Friend or Foe? Exploring the Factors that Determine the Difference

repair capacity is surpassed (Melis, 1999; Andersson & Aro, 2001).

is a light regulated process (Hihara et al., 2001; Singh et al., 2005).

**2.3.3 Other stress-response programs**

antenna, as recently reported (Johnson et al., 2009).

**2.3.2 D1 repair cycle** 

Between Positive and Negative Effects on Photosynthesis in Response to Insect Herbivory 163

(PSII) centers (Ivanov et al., 2006, 2008) or through conformational changes within the PSII

In addition to the D1 and D2 proteins that conform its core reaction center, PSII contains *α* and *β* subunits of cytochrome *b*559, the *psb*I gene product and a few low molecular weight polypeptides. The D1-D2 heterodimer within PSII binds all the electron carriers and cofactors necessary for electron transport (Nanba & Satoh, 1987; Mattoo et al., 1989). The reaction center protein D1 of PSII is also the primary target of photo-inhibition (Mattoo et al., 1984; Prasil et al., 1992). Due to its intrinsic vulnerability, the short-lived D1 protein must be constantly replaced by new copies via a complicated and evolutionary conserved process known as the PSII or D1 repair cycle, whose significance remains elusive (Mattoo et al., 1989; Andersson & Aro, 2001; Baena-Gonzales & Aro, 2002; Yokthongwattana & Melis, 2006; Edelman & Mattoo, 2008). Nevertheless, the process has an undoubted physiological importance considering that an accumulation of photo-inactivated PSII centers, leading to a decreased photochemical efficiency and the consequent photo-damage, occurs whenever its

D1 is a target of at least five post-translational modifications during its life cycle, including N-acetylation, palmitoylation and phosphorylation (Edelman & Mattoo, 2008). One or more of these post-translational modifications could potentially alter protein degradation kinetics, although the use of nitric oxide donors to inhibit *in vivo* phosphorylation of the D1 protein suggested that redox-dependent phosphorylation and D1 degradation in plants are not linked events (Booij-James et al., 2009). The DegP and FtsH proteases have been shown to be involved in D1 degradation *in vitro* (Haussuhl et al., 2001; Kanervo et al., 2003; Lindahl et al., 2000). The physiological significance of these specific proteases was demonstrated in the Arabidopsis *var2* (for *yellow variegated2*) or *var1* mutants, lacking the FtsH2 or FtsH5 membrane-bound metalloproteases, respectively, and the *fu-gaeri1* (*fug1*) mutant that suppresses interfering leaf variegation in *var1* and *var2*, all of which led to an inefficient degradation of the D1 protein and a concomitant increase in ROS levels that was connected to an enhanced susceptibility to photoinhibition (Bailey et al., 2002; Kato et al., 2009). Similar results were reported in cyanobacteria, where impaired D1 protein turnover was detected in an FtsH inactivation mutant (Silva et al., 2003). Moreover, experiments in which an increased transcription of two FtsH-coding genes and of FtsH protease activity was found to be induced upon transfer of cyanobacteria to high light, demonstrated that FtsH proteolysis

The accumulation of excitation energy produced when the rate of absorption of photons exceeds the rate of utilization of excitation energy in photosynthetic electron transport leads to an accumulation of reduced electron acceptors that eventually produce excited states of chlorophyll (i.e. triplet state). This process is presumed to predominantly occur in the PSII reaction center where quenching by carotenoids is less effective. Triplet state chlorophyll readily reacts with oxygen to give rise to singlet oxygen, a highly destructive excited oxygen species causing photo-oxidations (Triantaphylidés & Havaux, 2009) (Figure 1). Superoxide, H2O2 (produced via reduction or dismutation of superoxide) and hydroxyl radicals, all of which are more reactive than ground state triplet O2, can also be produced by numerous

secondary metabolism in plants. Thus, the LTR appears to contribute also to the adaptation of plant primary productivity to environmental conditions (Pesaresi et al., 2010).

All evidence gathered to data indicates that short- and long-term photosynthetic acclimation responses are triggered by changes in the redox state of the PQ pool and require the modulated activity of the kinase STN7. Due to its dual regulatory role, STN7 initiates a phosphorylation cascade that induces state transitions by phosphorylating LHCII and promotes the LTR process via the phosphorylation of as yet unknown chloroplast proteins. Beyond this point, the LTR signaling pathway is divided into two main branches: one is responsible for transcriptional regulation of chloroplast gene expression, while the other controls the expression of nuclear photosynthesis-related genes at transcriptional and posttranscriptional levels (see Figure 1) (Pesaresi et al., 2010).

#### **2.3 Acclimation responses under high or excess excitation pressure**

State transitions and LTR are acclimation responses that typically occur under low-light conditions and are controlled via redox signals. Under conditions resulting in high or excess excitation pressure other acclimation responses are activated, such as non-photochemical dependent antenna quenching, the D1 repair cycle or various other stress-response programs. These responses are also controlled via redox signals originating from the photosynthetic process (i.e. the PQ redox state and signals from the PSI acceptor side), but may also involve the participation of ROS such as hydrogen peroxide (H2O2) or singlet oxygen (Pfannschmidt et al., 2009).

#### **2.3.1 Non-photochemical quenching**

Non-photochemical quenching (NPQ) is a rapid de-excitation photo-protective quenching mechanism (qE) that involves dissipation of excess energy occurring upon short-term high light exposure. In addition, NPQ is considered to act as a ''light intensity counter,'' providing the photosynthetic membrane with a ''memory'' of the light-exposure history of the leaf (Horton et al., 2008; Foyer & Noctor, 2009). qE is induced by a low thylakoid lumen pH (i.e. a high ΔpH) generated by photosynthetic electron transport in excess light and involves the harmless thermal dissipation of excess energy in the chlorophyll (Chl) singlet excited states (1Chl\*) in photosystem II (PSII) of green plants and algae. qE is designed to minimize alternative reaction pathways that generate toxic photo-oxidative intermediates. It functions by activating a lumen-localized violaxanthin de-epoxidase enzyme that catalyses the conversion of violaxanthin to zeaxanthin via the intermediate antheraxanthin, in what is known as the xanthophyll cycle (which is completed by the conversion of zeaxanthin back to violaxanthin by means of a zeaxanthin epoxidase activated under limiting light conditions). The process also requires the protonation of PsbS, a PSII subunit that plays a role in the regulation of photosynthetic light harvesting and is also necessary for qE *in vivo*  (possibly by establishing a binding site for zeaxanthin that facilitates the de-excitation of singlet excited chlorophyll via energy or electron transfer). Experimental evidence has shown that energy transfer from chlorophyll molecules to a chlorophyll-zeaxanthin heterodimer that undergoes charge separation is the main mechanism for excess energy dissipation during feedback de-excitation (Horton et al., 1999; Külheim et al., 2002; Niyogi et al., 2005; Holt et al., 2004, 2005). However, quenching of excess energy can also occur independently of zeaxanthin via a reversible inactivation of a fraction of photosystem II (PSII) centers (Ivanov et al., 2006, 2008) or through conformational changes within the PSII antenna, as recently reported (Johnson et al., 2009).
