**2. Differential FTIR spectroscopy and data analysis**

#### **2.1 Time-resolved infrared spectroscopy**

To gain information about the dynamics of the system time-resolved infrared spectroscopy has to be used. Characterization of unstable and short-lived reaction intermediates is required to understand the photo-chemical reactions of light harvesting systems. To investigate the dynamics of such systems time-resolved pump-probe spectroscopy is required. Two different time-resolved infrared (trIR) techniques are commonly used namely ultrafast midIR and step scan FTIR spectroscopy.

In these cases the light-induced difference of infrared absorption (ΔOD) of a sample is measured as a function of time. Then, the ΔOD signal at a given wavenumber and at a given delay time between the pump and the probe, ΔOD(υ,t), is given by:

$$
\Delta OD(\nu, t) = OD(\nu, t)\_{Light} - OD(\nu)\_{Dark} = -\log \frac{I(\nu, t)\_{Light}}{I(\nu)\_{Dark}} \tag{1}
$$

TrIR can provide many molecular details of the reaction mechanism of protein, associated with their specific lifetimes, which are complementary to X-ray and NMR structure analysis, such as:


Ultrafast midIR and step scan FTIR spectroscopy are complementary. On the subnanosecond time scale most of the reaction is confined to the chromophore and the neighboring amino acid residues in the binding pocket and can be monitored using ultrafast IR spectroscopy. Long lived excited states such as triplet state, side chain (de)protonation and large scale protein motion involving changes in secondary-tertiary structure of the protein generally take place on the nanosecond to millisecond time scale and can be monitored using step scan FTIR spectroscopy. In this chapter step scan spectroscopy will be described in detail.

#### **2.2 Step scan-Time resolved FTIR spectroscopy**

Step scan-Time resolved FTIR spectroscopy allows monitoring molecular reaction mechanism of proteins at longer time-scales than ultrafast IR spectroscopy. The absorbance changes can be monitored with time resolutions down to nanoseconds and followed for time periods ranging over nine orders of magnitude. The technique has already been successfully applied to the light-driven proton pump bacteriorhodopsin (Kotting and Gerwert 2005), the photosynthetic reaction center and the GTPase Ras (Kotting 2005), PYP (Brudler, Rammelsberg et al. 2001) and Appa (Majerus, Kottke et al. 2007). In contrast to ultrafast mid-IR which uses a spectrograph containing a grating to disperse the probe beam on a 32-element MCT detector, Fourier-Transform Infrared (FTIR) spectroscopy is an interferometric method. The FTIR spectrometer consists of an infrared source, an interferometer, the sample, and the infrared detector. The interferometer is the heart of the spectrometer and consists in its simplest form of a beam splitter, a fixed mirror, and a moving mirror scanning back and forth. Therefore, the spectrum is not directly measured but its interferogram, i.e. the IR intensity reaching the detector as a function of the mirror position. The spectrum is subsequently obtained by Fourier transformation of the interferogram. The major advantages of FTIR spectroscopy, as compared to conventional dispersive IR spectroscopy, are the so-called multiplexing advantage (Felgett advantage) and the high energy flux reaching the detector (Jacquinot advantage), allowing rapid spectrum acquisition at a high signal to noise ratio. In the step-scan mode, the interferometer moving mirror may be visualized as being held stationary at the interferogram mirror position xn. The protein activity is initiated, by a laser flash, and the time dependence of the intensity change at this interferogram position xn is measured. Then the interferometer "steps" to the next interferogram data position xn+1, and the reaction is repeated and measured again. This process is continued at each sampling position of the interferogram.

#### **2.3 Data analysis for time-resolved infrared spectra. Global and target analysis**

TrIR experiments result in a 3-dimensional dataset as the changes in intensity (*ΔOD*(υ,t)) are measured as a function of time and wavelength. A typical dataset results in ~20 000 data points. To analyze such large amounts of data a Global analysis procedure is required (van Stokkum, Larsen et al. 2004). Global analysis implies a simultaneous analysis of the entire 3 dimensional dataset which gives a correlation between different wavelength-regions and time scales. The total dataset is a superposition of contributions from different species (components) having their own time constants (lifetime). Description of such a total dataset should be obtained with a minimum amount (i) of time constants (ki) that result in a good quality fit. Each component starts with a given concentration (ci) that decays in time and possesses its own specific time-independent spectrum (εi). Mathematically, the observed signal S at any given time (t) or wavenumber (υ) can be described with:

$$S(\boldsymbol{\nu}, t) = \sum\_{i=1}^{n} c\_i(t) \, \* \, \boldsymbol{\varepsilon}\_i(\boldsymbol{\nu}) \tag{2}$$

with ( ) *<sup>i</sup> k t <sup>i</sup> c e*

232 Infrared Spectroscopy – Life and Biomedical Sciences

To gain information about the dynamics of the system time-resolved infrared spectroscopy has to be used. Characterization of unstable and short-lived reaction intermediates is required to understand the photo-chemical reactions of light harvesting systems. To investigate the dynamics of such systems time-resolved pump-probe spectroscopy is required. Two different time-resolved infrared (trIR) techniques are commonly used namely

In these cases the light-induced difference of infrared absorption (ΔOD) of a sample is measured as a function of time. Then, the ΔOD signal at a given wavenumber and at a given

( ,) ( , ) ( , ) ( ) log ( )

TrIR can provide many molecular details of the reaction mechanism of protein, associated with their specific lifetimes, which are complementary to X-ray and NMR structure analysis,




Ultrafast midIR and step scan FTIR spectroscopy are complementary. On the subnanosecond time scale most of the reaction is confined to the chromophore and the neighboring amino acid residues in the binding pocket and can be monitored using ultrafast IR spectroscopy. Long lived excited states such as triplet state, side chain (de)protonation and large scale protein motion involving changes in secondary-tertiary structure of the protein generally take place on the nanosecond to millisecond time scale and can be monitored using step scan FTIR spectroscopy. In this chapter step scan spectroscopy will be

Step scan-Time resolved FTIR spectroscopy allows monitoring molecular reaction mechanism of proteins at longer time-scales than ultrafast IR spectroscopy. The absorbance changes can be monitored with time resolutions down to nanoseconds and followed for time periods ranging over nine orders of magnitude. The technique has already been successfully applied to the light-driven proton pump bacteriorhodopsin (Kotting and Gerwert 2005), the photosynthetic reaction center and the GTPase Ras (Kotting 2005), PYP (Brudler, Rammelsberg et al. 2001) and Appa (Majerus, Kottke et al. 2007). In contrast to


*Light Dark*

 

(1)

*Light*

*Dark*

*I t*

*I* 

**2. Differential FTIR spectroscopy and data analysis** 

delay time between the pump and the probe, ΔOD(υ,t), is given by:

*OD t OD t OD*

**2.1 Time-resolved infrared spectroscopy** 

ultrafast midIR and step scan FTIR spectroscopy.

propagation of structural changes.

**2.2 Step scan-Time resolved FTIR spectroscopy** 

such as:

residues.

described in detail.

The resulting spectra εi may represent mixture of known physico-chemical species and then do not contain directly physically relevant information. In such case decisions have to be made about a model that not only describes the raw data, but also generates relevant and meaningful spectra. Data previously obtained using different experimental techniques can be helpful in such context. The simplest model describing the measured data has to be chosen. The simplest model templates are the sequential and parallel models (Figures 1A and 1B) where one compartment flows directly into the next compartment with increasing lifetimes or to the ground state, respectively. A compartment represents a spectroscopic distinct state or physico-chemical species and symbolizes a component of the reaction with

Time-Resolved FTIR Difference Spectroscopy Reveals the Structure and Dynamics

addition, carotenoids are also able to scavenge singlet oxygen.

Singlet oxygen can be formed by the quenching of chlorophyll triplet states:

3Chl-*a*\* + Car Chl-*a* + 3Car\* (chlorophyll triplet quenching)

1O2\* + Car 3O2 + 3Car (singlet oxygen scavenging)

To prevent this reaction, carotenoids with their low-lying triplet states quench long-lived chlorophyll triplets (Nagae, Kakitani et al. 1993; Frank and Cogdell 1996; Krueger, Scholes et al. 1998; Krueger, Scholes et al. 1998; Damjanovic, Ritz et al. 1999; Damjanovic, Ritz et al.

In PCP, the triplet states of Chl-*a* are formed *via* intersystem crossing (ISC) from Chl-*a* singlet states. The latter are formed after direct Chl-*a* excitation or after excitation energy transfer (EET) which follows direct excitation of Per (Bautista, Hiller et al. 1999; Krueger, Lampoura et al. 2001; Zigmantas, Hiller et al. 2002). The Per triplet states are populated by triplet excitation energy transfer (TEET), which is governed by an electron-exchange interaction (Dexter mechanism) (Dexter 1953; Cogdell and Frank 1987). The PCP visible T-S spectra (Carbonera, Giacometti et al. 1996; Kleima, Wendling et al. 2000) exhibit a Qy Chl-*a* differential signal assigned to Chl-*a*-carotenoid interaction and is similar to carotenoid T-S spectra found in bacterial and higher plant antennae (Van der Vos, Carbonera et al. 1991; Angerhofer, Bornhauser et al. 1995; Peterman, Dukker et al. 1995). This interaction signal shows concerted dynamics with the carotenoid triplet (Peterman, Dukker et al. 1995; Kleima, Wendling et al. 2000) and has been discussed in terms of a Stark effect, but its exact nature remains unclear. It is then of interest to investigate such triplet dynamic using microsecond time-resolved FTIR spectroscopy to measure triplet formation in PCP by monitoring excitation-induced variations in the vibrational modes of the chromophores. For

3Chl-*a*\* + 3O2 Chl-*a* + 1O2\* (singlet oxygen formation)

2000) and scavenge singlet oxygen (Cogdell, Howard et al. 2000):

of Carotenoid and Chlorophyll Triplets in Photosynthetic Light-Harvesting Complexes 235

life on Earth, ultimately depends on the energy provided by the Sun. Photosynthetic organisms use antenna pigment-proteins to harvest light. Absorbed solar energy is transferred to the reaction center (RC) where it is converted into an electrochemical gradient, which is used to synthesize ATP, powering the cell (van Grondelle, Dekker et al. 1994). Together with (Bacterio)Chlorophyll ((B)Chl) carotenoids are the main pigments of photosynthesis. In addition to their structural involvement in the antenna architecture, carotenoids have a dual function namely light harvesting and photoprotection (Frank and Cogdell 1996). They harvest light in the blue –green spectral region where (B)Chl absorbs weakly thus increasing the absorption cross-section of the light harvesting system where sunlight is optimal. Carotenoid excitation is followed by ultrafast energy transfer to (B)Chl with a high efficiency (Shreve, Trautman et al. 1991; Ritz, Damjanovic et al. 2000; Walla, Linden et al. 2000; Zhang, Fujii et al. 2000; Papagiannakis, Kennis et al. 2002; Holt, Kennis et al. 2003; Zigmantas, Hiller et al. 2004). (B)Chl intersystem crossing (ISC) may lead to triplet formation, the efficiency of which depends on the singlet excited state lifetime. The longlived (ms timescale) (B)Chl triplet reacts with ground state oxygen (which is a triplet state) to produce the highly reactive oxygen singlet (1O2) (Krieger-Liszkay 2005). In light harvesting complexes (LHC), this process competes with fast (ns timescale) (B)Chl triplet excitation energy transfer to the carotenoid thereby (Krieger-Liszkay 2005) avoiding formation of 1O2 and ultimately the destruction of the light harvesting apparatus. In

Fig. 1. Schematic view of a sequential model (A), a parallel model (B) and a branched scheme (C).

its spectrum (εi) and associated lifetime (ki). Spectra that result from the application of a sequential and parallel model are called evolution associated difference spectra (EADS) and decay associated difference spectra (DADS), respectively. EADS estimated by a sequential model may represent mixtures of coexisting molecular states, since the possibility of branched or parallel dynamics are not taken into account. In order to disentangle the contributions a more complicated model must be used such as a branched model (Figure 1C), where one compartment can populate two other compartments. If one data set is analyzed with the three described models, then the resulting spectra for each compartment are different, except for compartment 1. Using a model different from the sequential or the parallel model is called a target analysis and can only be done if the data allows for it and/or if additional information is available from other techniques. In this latter case the spectra are no longer called DADS or EADS but species associated difference spectra (SADS).

### **3. Carotenoid and chlorophyll triplet state dynamic**

#### **3.1 Photoprotection mechanisms in Peridinin chlorophyll proteins. The chlorophyll triplet quenching by peridinin**

Photosynthesis is the process by which some terrestrial and marine organisms acquire energy from sunlight and transform it into chemical energy (organic compounds) (Gest 2002). Life on earth began roughly ~3.8 billion years ago (Schopf 1992), with some cyanobacterial species being the first photosynthetic organisms to appear. Accumulation of oxygen in our atmosphere has thus been possible due to the presence of these organisms, capable of performing oxygenic photosynthesis. Later the appearance of higher plants led to the production of organic compounds which constitute the basic source of energy for most living organisms. The photosynthetic process and, therefore the abundance and diversity of life on Earth, ultimately depends on the energy provided by the Sun. Photosynthetic organisms use antenna pigment-proteins to harvest light. Absorbed solar energy is transferred to the reaction center (RC) where it is converted into an electrochemical gradient, which is used to synthesize ATP, powering the cell (van Grondelle, Dekker et al. 1994). Together with (Bacterio)Chlorophyll ((B)Chl) carotenoids are the main pigments of photosynthesis. In addition to their structural involvement in the antenna architecture, carotenoids have a dual function namely light harvesting and photoprotection (Frank and Cogdell 1996). They harvest light in the blue –green spectral region where (B)Chl absorbs weakly thus increasing the absorption cross-section of the light harvesting system where sunlight is optimal. Carotenoid excitation is followed by ultrafast energy transfer to (B)Chl with a high efficiency (Shreve, Trautman et al. 1991; Ritz, Damjanovic et al. 2000; Walla, Linden et al. 2000; Zhang, Fujii et al. 2000; Papagiannakis, Kennis et al. 2002; Holt, Kennis et al. 2003; Zigmantas, Hiller et al. 2004). (B)Chl intersystem crossing (ISC) may lead to triplet formation, the efficiency of which depends on the singlet excited state lifetime. The longlived (ms timescale) (B)Chl triplet reacts with ground state oxygen (which is a triplet state) to produce the highly reactive oxygen singlet (1O2) (Krieger-Liszkay 2005). In light harvesting complexes (LHC), this process competes with fast (ns timescale) (B)Chl triplet excitation energy transfer to the carotenoid thereby (Krieger-Liszkay 2005) avoiding formation of 1O2 and ultimately the destruction of the light harvesting apparatus. In addition, carotenoids are also able to scavenge singlet oxygen.

Singlet oxygen can be formed by the quenching of chlorophyll triplet states:

3Chl-*a*\* + 3O2 Chl-*a* + 1O2

234 Infrared Spectroscopy – Life and Biomedical Sciences

Fig. 1. Schematic view of a sequential model (A), a parallel model (B) and a branched

**3. Carotenoid and chlorophyll triplet state dynamic** 

its spectrum (εi) and associated lifetime (ki). Spectra that result from the application of a sequential and parallel model are called evolution associated difference spectra (EADS) and decay associated difference spectra (DADS), respectively. EADS estimated by a sequential model may represent mixtures of coexisting molecular states, since the possibility of branched or parallel dynamics are not taken into account. In order to disentangle the contributions a more complicated model must be used such as a branched model (Figure 1C), where one compartment can populate two other compartments. If one data set is analyzed with the three described models, then the resulting spectra for each compartment are different, except for compartment 1. Using a model different from the sequential or the parallel model is called a target analysis and can only be done if the data allows for it and/or if additional information is available from other techniques. In this latter case the spectra are no longer called DADS or EADS but species associated difference spectra

**3.1 Photoprotection mechanisms in Peridinin chlorophyll proteins. The chlorophyll** 

Photosynthesis is the process by which some terrestrial and marine organisms acquire energy from sunlight and transform it into chemical energy (organic compounds) (Gest 2002). Life on earth began roughly ~3.8 billion years ago (Schopf 1992), with some cyanobacterial species being the first photosynthetic organisms to appear. Accumulation of oxygen in our atmosphere has thus been possible due to the presence of these organisms, capable of performing oxygenic photosynthesis. Later the appearance of higher plants led to the production of organic compounds which constitute the basic source of energy for most living organisms. The photosynthetic process and, therefore the abundance and diversity of

scheme (C).

(SADS).

**triplet quenching by peridinin** 

\* (singlet oxygen formation)

To prevent this reaction, carotenoids with their low-lying triplet states quench long-lived chlorophyll triplets (Nagae, Kakitani et al. 1993; Frank and Cogdell 1996; Krueger, Scholes et al. 1998; Krueger, Scholes et al. 1998; Damjanovic, Ritz et al. 1999; Damjanovic, Ritz et al. 2000) and scavenge singlet oxygen (Cogdell, Howard et al. 2000):


In PCP, the triplet states of Chl-*a* are formed *via* intersystem crossing (ISC) from Chl-*a* singlet states. The latter are formed after direct Chl-*a* excitation or after excitation energy transfer (EET) which follows direct excitation of Per (Bautista, Hiller et al. 1999; Krueger, Lampoura et al. 2001; Zigmantas, Hiller et al. 2002). The Per triplet states are populated by triplet excitation energy transfer (TEET), which is governed by an electron-exchange interaction (Dexter mechanism) (Dexter 1953; Cogdell and Frank 1987). The PCP visible T-S spectra (Carbonera, Giacometti et al. 1996; Kleima, Wendling et al. 2000) exhibit a Qy Chl-*a* differential signal assigned to Chl-*a*-carotenoid interaction and is similar to carotenoid T-S spectra found in bacterial and higher plant antennae (Van der Vos, Carbonera et al. 1991; Angerhofer, Bornhauser et al. 1995; Peterman, Dukker et al. 1995). This interaction signal shows concerted dynamics with the carotenoid triplet (Peterman, Dukker et al. 1995; Kleima, Wendling et al. 2000) and has been discussed in terms of a Stark effect, but its exact nature remains unclear. It is then of interest to investigate such triplet dynamic using microsecond time-resolved FTIR spectroscopy to measure triplet formation in PCP by monitoring excitation-induced variations in the vibrational modes of the chromophores. For

Time-Resolved FTIR Difference Spectroscopy Reveals the Structure and Dynamics

Papagiannakis et al. 2005).

Giacometti et al. 1996).

**3.2 Peridinin and chlorophyll triplet state in solvent** 

of Carotenoid and Chlorophyll Triplets in Photosynthetic Light-Harvesting Complexes 237

molecular structure constituted of an allene moiety and a lactone ring in conjugation with the π-electron system of the carotenoid backbone, an epoxy group with a secondary alcohol group on one beta-ring and an ester group located on the opposite beta-ring with a tertiary alcohol group (Figure 2B). The structural differences between Per and other carotenoids are most likely related to its specific function in PCP. In contrast to most photosynthetic LHCs, the carotenoid (Per), and not Chl-*a*, is the main light-absorbing pigment in PCP. Per's unusual structure and especially its conjugated lactone carbonyl confers a high plasticity of absorptivity and reactivity to this carotenoid, for instance by mixing CT states with the lower excited states (Bautista, Connors et al. 1999; Zigmantas, Hiller et al. 2002; Vaswani, Hsu et al. 2003; Ilagan, Shima et al. 2004; Zigmantas, Hiller et al. 2004; Premvardhan,

A different dinoflagellate species *Heterocapsa pygmaea* uses a LHC closely related to A-PCP, H-PCP. H-PCP has the same pigment stoichiometry as A-PCP but its peptide unit is about half the size of that of A-PCP and contains only half the number of pigments (Song, Koka et al. 1976; Norris and Miller 1994; Hiller, Crossley et al. 2001). The overall identity of the H-PCP monomer with that of the N and C terminal domains of A-PCP is about 70% (Hiller, Crossley et al. 2001). 3-D Modeling of the H-PCP sequence on the high-resolution x-ray structure of A-PCP has shown only major differences in the trimer interface (Hiller, Crossley et al. 2001). In nature, H-PCP is found as a dimer. Thus, the A-PCP monomer can be considered as the covalent equivalent of the H-PCP dimer and the pigment conformation in both systems should be very similar. This view is supported by the high similarity of the spectroscopic properties (Abs, CD, T-S spectra) of the two complexes (Carbonera,

It is important to firmly establish our spectral assignment of Per and Chl-*a* modes in the 3Per and 3Chl-*a* states using an artificial system where Per and Chl-*a* are not strongly coupled. To this end, we performed step-scan time resolved FTIR measurements on Per mixed with Chl*a* in organic solvent to observe TEET on the µs timescale and extract their individual spectra using global and target analysis. The two chromophores were mixed in THF with a stoichiometric ratio Per: Chl-*a* of 1:8 with a Chl-*a* OD (670 nm) of about 100 cm-1. Timeresolved FTIR data has been obtained by direct excitation of Chl-*a* at 625 nm and the ensuing spectral evolution was analyzed using global and target analysis. Global analysis in terms of a sequential kinetic scheme shows that three components are required to fit the data. In order to determine the origin of the third component, we used target analysis. The best fit was obtained with a three level scheme, in which the first component decays in 3.5 µs in parallel to the second and the third component, which decay to the ground state in 7 µs and 3 ms, respectively. The kinetic model is displayed in Fig.3. The first component decays into the second and third components with an estimated yield of 60% and 12% respectively. About 28% of the first component amplitude is lost *via* triplet-triplet annihilation and decays to the ground state. Experiments with Chl-*a* only in THF confirmed that the first component decays into a component having similar lifetime and spectral features as compared to the third component observed in the mixed Chl-*a*/Per sample. We note that Per is unable to quench the third component which suggests that it can be assigned to a radical state of Chl-*a*. The three SADS that result from the target analysis of the mixed Chl-*a*/Per data are shown

infrared spectroscopy a good model to study this photoprotection mechanism is the water soluble LHC from the dinoflagellate *Amphidinium carterae*, A-PCP. Indeed it binds only one type of carotenoid and chlorophyll, Per and Chl a, respectively. In addition, Per and Chl-*a* possess conjugated carbonyl groups which are sensitive to the protein environment and the electronic state of the pigment. In other words, the carbonyl modes are the molecular probes for the pigments: the 9-keto mode of the Chl-*a* and the carbonyl group of the lactone ring of Per. The keto modes are generally expected at lower energies (≤ 1700 cm-1) compared to lactone modes (> 1700 cm-1) which makes them easily distinguishable.

The 2.0 Å crystal structure of A-PCP reveals a trimeric arrangement. The protein is mainly composed of α-helices structure and surrounds a hydrophobic cavity in which besides the pigments also two lipids are bound. The A-PCP monomer contains 2 Chlorophyll-*a* and 8 Peridinins. The molecular structures of Chlorophyll-*a* (Chl-*a*) and peridinin (Per) are shown Figure 2A and 2B. In each half of the PCP monomer one Chl-*a* is closely surrounded by 4 Per (Figure 2C) (Hofmann, Wrench et al. 1996; Hofmann 1999) leading to high efficiency for both light harvesting and photoprotection (Kleima 2000; Alexandre, Luhrs et al. 2007). The N- and C-terminal halves of the polypeptide to which the two clusters of pigments are bound form almost identical domains related by a pseudo two-fold symmetry axis. PCP most likely functions as a LHC that transfers its energy mainly to the PSII RC complex (Knoetzel and Rensing 1990; Mimuro, Tamai et al. 1990). Since the latter complex is located in the photosynthetic membrane, transfer has to occur from a water-soluble complex to a membrane bound complex. As no specific binding site has been identified so far, PCP could in principle either transfer its energy directly to the PSII RC or indirectly *via* the membrane bound Chl-a/c complex, which is an analog of LHCII of green plants. Per has a unique

Fig. 2. Molecular structure of Chl-*a* (A) and Per (B). Pigments organization, one Chl-*a* (black) surrounded by 4 Per (grey) (C).

infrared spectroscopy a good model to study this photoprotection mechanism is the water soluble LHC from the dinoflagellate *Amphidinium carterae*, A-PCP. Indeed it binds only one type of carotenoid and chlorophyll, Per and Chl a, respectively. In addition, Per and Chl-*a* possess conjugated carbonyl groups which are sensitive to the protein environment and the electronic state of the pigment. In other words, the carbonyl modes are the molecular probes for the pigments: the 9-keto mode of the Chl-*a* and the carbonyl group of the lactone ring of Per. The keto modes are generally expected at lower energies (≤ 1700 cm-1) compared to

The 2.0 Å crystal structure of A-PCP reveals a trimeric arrangement. The protein is mainly composed of α-helices structure and surrounds a hydrophobic cavity in which besides the pigments also two lipids are bound. The A-PCP monomer contains 2 Chlorophyll-*a* and 8 Peridinins. The molecular structures of Chlorophyll-*a* (Chl-*a*) and peridinin (Per) are shown Figure 2A and 2B. In each half of the PCP monomer one Chl-*a* is closely surrounded by 4 Per (Figure 2C) (Hofmann, Wrench et al. 1996; Hofmann 1999) leading to high efficiency for both light harvesting and photoprotection (Kleima 2000; Alexandre, Luhrs et al. 2007). The N- and C-terminal halves of the polypeptide to which the two clusters of pigments are bound form almost identical domains related by a pseudo two-fold symmetry axis. PCP most likely functions as a LHC that transfers its energy mainly to the PSII RC complex (Knoetzel and Rensing 1990; Mimuro, Tamai et al. 1990). Since the latter complex is located in the photosynthetic membrane, transfer has to occur from a water-soluble complex to a membrane bound complex. As no specific binding site has been identified so far, PCP could in principle either transfer its energy directly to the PSII RC or indirectly *via* the membrane bound Chl-a/c complex, which is an analog of LHCII of green plants. Per has a unique

**Per 612**

Fig. 2. Molecular structure of Chl-*a* (A) and Per (B). Pigments organization, one Chl-*a* (black)

**Per 611**

**Per 613**

**Chla 601**

**Per 614**

lactone modes (> 1700 cm-1) which makes them easily distinguishable.

**B**

**A**

surrounded by 4 Per (grey) (C).

**C**

molecular structure constituted of an allene moiety and a lactone ring in conjugation with the π-electron system of the carotenoid backbone, an epoxy group with a secondary alcohol group on one beta-ring and an ester group located on the opposite beta-ring with a tertiary alcohol group (Figure 2B). The structural differences between Per and other carotenoids are most likely related to its specific function in PCP. In contrast to most photosynthetic LHCs, the carotenoid (Per), and not Chl-*a*, is the main light-absorbing pigment in PCP. Per's unusual structure and especially its conjugated lactone carbonyl confers a high plasticity of absorptivity and reactivity to this carotenoid, for instance by mixing CT states with the lower excited states (Bautista, Connors et al. 1999; Zigmantas, Hiller et al. 2002; Vaswani, Hsu et al. 2003; Ilagan, Shima et al. 2004; Zigmantas, Hiller et al. 2004; Premvardhan, Papagiannakis et al. 2005).

A different dinoflagellate species *Heterocapsa pygmaea* uses a LHC closely related to A-PCP, H-PCP. H-PCP has the same pigment stoichiometry as A-PCP but its peptide unit is about half the size of that of A-PCP and contains only half the number of pigments (Song, Koka et al. 1976; Norris and Miller 1994; Hiller, Crossley et al. 2001). The overall identity of the H-PCP monomer with that of the N and C terminal domains of A-PCP is about 70% (Hiller, Crossley et al. 2001). 3-D Modeling of the H-PCP sequence on the high-resolution x-ray structure of A-PCP has shown only major differences in the trimer interface (Hiller, Crossley et al. 2001). In nature, H-PCP is found as a dimer. Thus, the A-PCP monomer can be considered as the covalent equivalent of the H-PCP dimer and the pigment conformation in both systems should be very similar. This view is supported by the high similarity of the spectroscopic properties (Abs, CD, T-S spectra) of the two complexes (Carbonera, Giacometti et al. 1996).
