This protocol details the reconstitution of light-harvesting complexes in vitro. These integral membrane proteins coordinate chlorophylls and carotenoids and are responsible for harvesting light in higher plants and green algae.
In plants and green algae, light is captured by the light-harvesting complexes (LHCs), a family of integral membrane proteins that coordinate chlorophylls and carotenoids. In vivo, these proteins are folded with pigments to form complexes which are inserted in the thylakoid membrane of the chloroplast. The high similarity in the chemical and physical properties of the members of the family, together with the fact that they can easily lose pigments during isolation, makes their purification in a native state challenging. An alternative approach to obtain homogeneous preparations of LHCs was developed by Plumley and Schmidt in 19871, who showed that it was possible to reconstitute these complexes in vitro starting from purified pigments and unfolded apoproteins, resulting in complexes with properties very similar to that of native complexes. This opened the way to the use of bacterial expressed recombinant proteins for in vitro reconstitution. The reconstitution method is powerful for various reasons: (1) pure preparations of individual complexes can be obtained, (2) pigment composition can be controlled to assess their contribution to structure and function, (3) recombinant proteins can be mutated to study the functional role of the individual residues (e.g., pigment binding sites) or protein domain (e.g., protein-protein interaction, folding). This method has been optimized in several laboratories and applied to most of the light-harvesting complexes. The protocol described here details the method of reconstituting light-harvesting complexes in vitro currently used in our laboratory, and examples describing applications of the method are provided.
The photosynthetic apparatus of plants and algae include integral membrane proteins that bind chlorophyll a (chl a), b (chl b) and carotenoids (car). These pigment-protein complexes are active in harvesting light energy and transferring that excitation energy to the reaction centers, where it is used to promote charge separation2. They are also involved in regulatory feedback mechanisms that protect the photosynthetic apparatus from high light damage3,4. The light harvesting complexes (LHCs) are comprised of a large family of related proteins in plants and algae5.
The homogeneous purification of each member of the family has been complicated by the highly similar chemical and physical properties of the complexes. In addition, purification procedures often result in loss of pigments or other potential cofactors such as lipids. In vitro reconstitution represents a powerful method to overcome these problems. The LHC associated with Photosystem II (LHC-II) was first reconstituted in vitro by Plumley and Schmidt in 19871. The researchers extracted delipidated protein and pigments separately from plant chloroplasts, and then combined the heat denatured protein with pigments in the presence of Lithium Dodecyl Sulfate (LDS), followed by three cycles of freezing and thawing1. They showed that the spectral properties of the reconstituted LHC complexes were very similar to complexes purified from plants. The ease of reconstituting LHC pigment-protein complexes, likely due to some inherent self-assembly feature, along with the difficulty in isolating purified complexes from organisms, led to the quick adoption of the method by other researchers. The reconstitution of photosynthetic proteins overexpressed in Escherichia coli (E. coli) was achieved by Paulsen and colleagues in 19906. In E. coli, overexpressed membrane proteins are typically contained in inclusion bodies, which facilities their purification. Reconstitution is achieved through heat denaturation of the inclusion bodies containing recombinant protein in the presence of LDS, followed by the addition of pigments which initiates the protein folding. Folding of the LHCII complex is a two-step process: first, chlorophyll a is bound in less than 1 min; second, chlorophyll b is bound and stabilized over several minutes7.
In addition to providing insight into the folding dynamics, in vitro reconstitution combined with site-directed mutagenesis has allowed the identification of specific amino acids important for stability (e.g., 8,9) or pigment coordination (e.g., 10). Manipulation of refolding conditions by adjusting parameters such as pigment composition or detergents have also identified elements critical for proper folding, such as the requirement of Xanthophylls for the LHCII complex (e.g., 1,11). In addition, investigation of the properties of individual pigments bound to the complexes has been possible using complexes reconstituted in vivo (e.g., 10).
The method described here begins with isolation of pigments (chlorophylls and carotenoid) from spinach and the green alga Chlamydomonas reinhardtii. The expression and purification of a LHC protein from E. coli in the form of inclusion bodies is then detailed, followed by the reconstitution of LHC and subsequent purification by Ni affinity column. In the final step, the reconstituted complexes are further purified by sucrose gradient centrifugation to remove free pigments and unfolded apoprotein. This protocol represents an optimized procedure incorporating several modifications that have been introduced by different laboratories over time1,6,10,12–14.
1. Total Pigment Extraction from Spinach Leaves
2. Extraction of Carotenoids from Spinach
3. Total Pigment and Carotenoid Extraction from Chlamydomonas reinhardtii
4. Purification of Inclusion Bodies
5. Reconstitution
This protocol typically yields 1-2 ml of reconstituted protein with an OD of 4 when absorbance is measured in the Qy region (600-750 nm). Quantity can be adjusted as desired, although care should be taken to maintain the proper ratios during the procedure.
6. Nickel Column Purification
7. Sucrose Gradient Centrifugation
This protocol details a method to reconstitute chorophyll a/b binding proteins in vitro. This technique permits the folding of these pigment-protein complexes in vitro starting from the apoprotein, which can be obtained by overexpression in a heterologous system, and pigments extracted from plant or algae. After reconstitution, the refolded pigment-protein complex is purified from the excess of pigments and the unfolded apoprotein in two steps. The first step (Figure 1 A-B) is based on the presence of His-tag at the C-terminal of the protein, which permits the removal of large part of the unbound pigments. The second purification step utilizes sucrose density gradient centrifugation, (Figure 2) where the unfolded protein usually migrates slower than the green band containing the reconstituted protein. The goal of the reconstitution in vitro is to obtain complexes with the same properties as the native ones. To illustrate this outcome, the spectroscopic properties of an in vivo light-harvesting complex is compared with the same LHC complex reconstituted in vitro13,20,21. The absorption spectrum of the LHCs in the visible range (350 nm and 750 nm) depends on the pigment composition of the complex, as well as on the pigment’s environment (which includes the protein) and it is thus a sensitive tool to check the quality of the reconstitution. In Figure 3, the absorption spectrum of CP24, a chlorophyll a/b binding protein from Arabidopsis thaliana, reconstituted in vitro, is compared with the spectrum of the same complex purified from Arabidopsis thylakoids21. In the spectra, it is possible to recognize the Qy and the Soret transition of Chl a (peaks at 671/439 nm) and Chl b (peaks at 649/466 nm). The native and reconstituted complexes show identical absorption spectra, indicating a virtually identical pigment composition and organization. Fluorescence spectroscopy can be used to assess the quality of the reconstituted complex. The fluorescence emission spectra is measured upon excitation at different wavelengths, which excite preferentially different pigments: Chl a at 440 nm, Chl b at 475 nm, and Xanthophylls at 500 nm. In a properly folded protein-pigment complex, Chl b and Xanthophylls transfer their excitation energy primarily to Chl a within a few picoseconds, and the fluorescence originates from a thermally equilibrated system resulting in a single peak with the same shape and maxima at all three excitation wavelengths (Figure 4A–B). The presence of Chl b not coordinated to the protein can be recognized by an additional peak or shoulder around 650 nm upon 475 nm excitation (Figure 4C). The presence of free Chl a instead leads to additional emission around 675 nm, which is mainly present upon 440 nm excitation. The fluorescence emission spectra upon 475 nm excitation of both reconstituted and the native CP24 complexes (Figure 4D) show a single peak at 681 nm, indicating that reconstituted complex is correctly folded. An additional confirmation that the pigment-protein complex is correctly reconstituted comes from circular dichroism (CD) measurements. The CD signal in the visible region depends on the excitonic interactions between pigments and it is thus very sensitive to even small changes in the organization of the chromophores22. Figure 5 shows the CD spectra of reconstituted and native CP24, with the typical fingerprint peaks at 681 nm, 650 nm and 481 nm. In conclusion, the high similarity between the spectroscopic properties of native and the reconstituted CP24 confirms that the reconstitution procedure yields native-like complexes suitable for in vitro study of light-harvesting proteins.
Figure 1. Representation of the purification of recombinant LHC proteins with a His tag using a nickel column. (A) During the purification, His-tagged protein, comprised of both reconstituted complexes (green hexagon) and un-reconstituted/aggregated protein (orange hexagon) are bound to the surface of the Ni-Sepharose (blue spot), while unbound pigments (small colored spots) flow through. (B) When the column is washed with the elution buffer containing imidazole, the reconstituted and un-reconstituted proteins are collected in the flow through.
Figure 2. Sucrose gradient of reconstituted LHCII after purification by nickel column. The reconstituted complexes are separated from the free pigment by the density gradient. The dark green band represents reconstituted LHCII and the pale green background is composed of free pigments.
Figure 3. Absorption spectra of reconstituted protein CP24 (rCP24, red line) and the native one (nCP24, black line) isolated from Arabidopsis thaliana. In both spectra, it is possible to recognize the Qy and the Soret transition of Chl a (peaks at 671/439 nm) and Chl b (peaks at 649/466 nm). This figure has been modified from Passarini et al. 201421.
Figure 4. Fluorescence emission spectra. The fluorescence emission spectra of reconstituted CP24 wildtype complex (A) and normalized to the maximum (B) showing efficient energy transfer from Chl b and Xanthophyls to Chl a. (C) Fluorescence emission spectra of reconstituted CP24 (rCP24) and the native complex (nCP24) isolated from Arabidopsis thaliana. The spectra are normalized to the maximum of the peak (D). Please click here to view a larger version of this figure.
Figure 5. Circular Dichroism Spectra. Reconstituted CP24 (rCP24, red line) and the native complex (nCP24, black line) isolated from Arabidopsis thaliana shows very similar spectra.
Figure 6. Absorption spectra of CP29 wild type (CP29_WT) and mutated CP29 (CP29_A2). The green line shows the differences between the two plots.
All the buffers can be stored at 4 °C. | |||
Components | Final Concentration | Additional notes | |
Grinding Buffer | Sorbitol | 0.4 M | |
Tricine | 0.1 M | pH 7.8 | |
NaCl | 10 mM | ||
MgCl2 | 5 mM | ||
Milk Powder | 0.5% w/v | ||
Wash Buffer | Sorbitol | 50 mM | |
Tricine | 5 mM | pH 7.8 | |
EDTA | 10 mM | pH 8 | |
Lysis Buffer | Tris | 50 mM | pH 8 |
Sucrose | 2.5% w/v | ||
EDTA | 1 mM | pH 8 | |
Detergent buffer | NaCl | 200 mM NaCl | |
Deoxycholic acid | 1% w/v | ||
NONIDET P-40 | 1% w/v | ||
Tris | 20 mM | pH 7.5 | |
EDTA | 2 mM | pH 8 | |
beta-mercaptoethanol | 10 mM | ||
Triton Buffer | Triton X-100 | 0.5% w/v | |
Tris | 20 mM | pH 7.5 | |
beta-mercaptoethanol | 1 mM | ||
Buffer TE | Tris | 50 mM | pH 8 |
EDTA | 1 mM | pH 8 | |
Reconstitution Buffer | HEPES | 200 mM | |
Sucrose | 5% w/v | ||
Lithiumdodecylsulfate (LDS) | 4% w/v | ||
Benzamidine | 2 mM | ||
Aminocaproic Acid | 10 mM | ||
OG Buffer | Octylglucoside | 1% w/v | |
Sucrose | 12.5% w/v | ||
NaCl | 0.2 M | ||
HEPES | 20 mM | ||
Imidazole | 10 mM | ||
OG Rinse Buffer | n-Dodecyl-beta-D-Maltoside (β-DM) | 0.06% w/v | |
HEPES | 40 mM | pH 7.5-9 | |
NaCl | 0.2 M | ||
Elution Buffer | Imidazole | 0.5 M | |
n-Dodecyl-beta-D-Maltoside (β-DM) | 0.06% w/v | ||
HEPES | 40 mM | pH 8 | |
NaCl | 0.2 M | ||
Sucrose Solution | Sucrose | 20% w/v | |
n-Dodecyl-beta-D-Maltoside (β-DM) | 0.06% w/v | ||
HEPES | 0.01 M | pH 7.6 | |
Acetone 80% buffered with Sodium Carbonate | Acetone | 80% v/v | |
Sodium Carbonate | 1 M | ||
Ethanol 96% buffered with Sodium Carbonate | Ethanol | 96% v/v | |
Sodium Carbonate | 1 M |
Table 1. List of buffers and solutions used in this protocol.
Chla a/b mix | Chla a/b | Chl a | Chl b | Neo | Viola | Lute | Chl tot | Chl/Car | |
nCP26 | – | 2.2±0.05 | 6.2 | 2.8 | 0.61 | 0.38 | 1.02 | 9 | 4.5±0.1 |
rCP26 | 8 | 2.71±0.05 | 6.57 | 2.43 | 0.72 | 0.32 | 0.97 | 9 | 3.9±0.04 |
rCP26 | 5.5 | 2.25±0.05 | 6.23 | 2.77 | 0.77 | 0.3 | 0.96 | 9 | 4.0±0.1 |
rCP26 | 3 | 2.08±0.04 | 6.08 | 2.92 | 0.76 | 0.3 | 1.04 | 9 | 4.1±0.1 |
rCP26 | 1 | 1.7±0.05 | 5.7 | 3.3 | 0.7 | 0.3 | 0.9 | 9 | 4.3±0.05 |
rCP26 | 0.3 | 1.11±0.04 | 4.7 | 4.28 | 0.7 | 0.3 | 0.9 | 9 | 4.2±0.2 |
rCP26 | 0.05 | 0.23±0.01 | 1.4 | 5.6 | 0.58 | 0.24 | 1.11 | 7 | 3.1±0.06 |
rCP26 | <0.01 | 0.11±0.01 | 0.7 | 6.3 | 0.64 | 0.3 | 1.08 | 7 | 3.06±0.06 |
Table 2. Pigment content of CP26 native complex compared to reconstituted protein-pigment complexes with different Chl a/b Ratios39.
Membrane proteins are not so easy to study. Isolation of native membrane proteins is complicated by the need to solubilize the lipid bilayer with detergents, which can damage the protein and remove essential cofactors. These proteins might also be present at low levels in biological membranes, or be mixed with closely related proteins, as in the case of the light harvesting complexes, that makes purification of single complexes difficult. Heterologous protein expression in E. coli and in vitro reconstitution offers the possibility to avoid these problems. In vitro reconstitution and purification of folded proteins results in complexes that possess characteristics very similar to those of the native complexes20,21,23 and thus can be used to study complexes that cannot be purified to homogeneity 24–27.
This method uses spinach, which is easily attainable year-round, as a source for the total pigment and carotenoid preparations. For some reconstitutions of proteins native to algae, use of pigments purified from algae is preferred due to different pigment compositions. The Chl a/b ratio and Chl/car ratio remains the same regardless of pigment source.
It is important to realize that the efficiency of the reconstitution is usually around 35%28. Thus it is necessary to remove the non-bound pigments and the unfolded apoprotein from the solution after the reconstitution. A two-step purification protocol is presented in this protocol (see also results). However, it should be noted that the sucrose gradient step does not allow the complete separation of apo- and holo-protein. For most analyses this is not a problem, as the apoprotein does not contain pigments and thus does not interfere with the functional measurements. However, in case it is necessary to fully remove the apoprotein from the fraction containing the reconstituted complex (for example, to calculate the pigment to protein stoichiometry), an anionic exchange column can be used (see Passarini et al. 200929 for details).
The capacity to refold recombinant light harvesting proteins with isolated pigments in vitro provides an opportunity to “manipulate” the complexes by modifying the reconstitution “environment” in various ways, thereby changing the characteristics of the resulting complex. For example, changing the pigment composition during reconstitution can result in a complex with altered pigment composition. This feature can be utilized to study the influence various pigments have on the structure and stability of the complex. Usually the pigment preparation obtained from spinach has a Chl a/b ratio of 3:1 and a Chl/car ratio of 2.9:1. This ratio typically produces a reconstituted protein with the same properties as the native one. However, adjustment of the Chl a/b ratio by the addition of purified Chl a or b can influence the binding of different pigments due to varying selectivity of the binding sites30–33. This is possible because most of the pigment binding sites are not completely selective for Chl a or Chl b, but can accommodate both, although with different affinity10,30,34. In a similar way, the carotenoid binding sites were also shown to be able to accommodate more than one Xanthophyll species8,35–38. Different reconstitutions of CP26, another pigment-protein complex of higher plants, using various pigment compositions are shown in Table 2 39. These reconstitutions were used to assess the affinity of binding sites for particular pigments39. It is interesting to note that in order to obtain a complex with the same pigment composition as the native one, the Chl a/b ratio of pigment mix must be 3:1. This seems to be the case for all LHC complexes of higher plants20,40.
The combination of molecular biology with the reconstitution technique allows the properties of a Chl-binding complex to be studied in more detail. The importance of different protein domains on the stability and folding of the complexes, or their involvement in the protein-protein interactions, have been determined by truncating the apoprotein or performing random mutagenesis8,41–44. Single amino acid residues important for the coordination of different pigments can be altered through site-directed mutagenesis in order to analyze the properties of individual pigments or assess their contribution to the function and stability of the complex10,28,29,45–52. Figure 6 shows reconstituted Lhcb4 (CP29) with a mutation of the histidine at position 21653. A comparison of the pigment composition of wildtype and mutant complexes shows that the mutation induces the loss of one Chl a molecule, indicating that the targeted site accommodates a Chl a in the WT complex. The differences of the absorption spectra of WT and mutant, upon normalization to the pigment content, also shows the absorption properties of the lost pigment. In this case, the difference can be seen in the main peak at 680 nm, indicating that the Chl a coordinated by His216 absorbs at this wavelength (for more details about this mutant and the spectroscopic properties see Mozzo et al. 200853). Mutation analysis can also be used to determine the effect of the environment on the spectroscopic properties of the pigments54.
In conclusion, light harvesting proteins can readily be reconstituted in vitro resulting in pigment-protein complexes with very similar properties to native complexes. In this way, the difficulties of isolating native proteins are eliminated, while also delivering protein preparation with high yield and purity for further study. The importance of a 3:1 Chl a/b ratio in producing an authentic complex is emphasized, and examples of reconstituted wildtype and mutant LHCs are provided to illustrate applications of the technique.
The authors have nothing to disclose.
This work was supported by the European research council by a ERC starting/consolidator grant to RC and by the Dutch Foundation for research on matter (FOM) via a FOM program (10TM01).
Name of Material/ Equipment | Company | Catalog Number | Comments/Description |
HisTrap HP | GE Healthcare | 17-5247-01 | |
Nylon cloth | 20 μm pores | ||
Soft artists paint brush | |||
NONIDET P-40 | Sigma | 74385 | |
Beta-DM | Sigma | D4641 | |
DNAase | ThermoScientific | EN0525 | |
Milk Powders | |||
RNAase | ThermoScientific | EN0531 | |
Sonicator | |||
Octyl β-D-glucopyranoside | Sigma | O8001 | |
Ultracentrifuge XL | Beckman-Coulter | ||
TAP medium | see reference 17 | ||
LB medium | see reference 19 |