Download Structural and Functional Studies of the Light-Dependent Protochlorophyllide Oxidoreductase ... PDF

TitleStructural and Functional Studies of the Light-Dependent Protochlorophyllide Oxidoreductase ...
File Size19.4 MB
Total Pages202
Table of Contents
	List of Abbreviations
	1. Introduction
		1.1 Oxygenic photosynthesis
		1.2 Photosynthetic Bacteria
			1.2.1 Cyanobacteria
		1.3 Structure of chlorophylls
		1.4 Common steps in tetrapyrrole biosynthesis
			1.4.1 Formation of δ-aminolaevulinic acid
			1.4.2 δ-aminolaevulinic acid to porphobilinogen
			1.4.3 Porphobilinogen to hydroxymethylbilane
			1.4.4 Hydroxymethylbilane to uroporphyrinogen III
			1.4.5 Uroporphyrinogen III to coproporphyrinogen III
			1.4.6 Coproporphyrinogen III to protoporphyrinogen IX
			1.4.7 Protoporphyrinogen IX to protoporphyrin IX
		1.5 The haem and chlorophyll branch point
			1.5.1 Ferrochelatase
		1.6 Protoporphyrin IX to Mg protoporphyrin IX
		1.7 Mg protoporphyrin IX to Mg protoporphyrin IX monomethyl ester
		1.8 Formation of the isocyclic ring
		1.9 Protochlorophyllide to Chlorophyllide
			1.9.1 Light-independent protochlorophyllide oxidoreductase (DPOR) DPOR subunits and homology to nitrogenase Structure of L-protein Structure of NB-protein Redox function of DPOR
			1.9.2 Light-dependent protochlorophyllide oxidoreductase
			1.9.3 POR in plants
			1.9.4 SDR Superfamily of enzymes Rossmann Fold Catalysis Oligomerisation
			1.9.5 POR homology model
			1.9.6 Biochemistry of POR Substrate binding Hydride transfer Proton transfer Excited state processes Product release Role of conserved cysteine residues Quantum tunnelling
			1.9.7 Comparing LPOR and DPOR
		1.10 Reduction of the C8-vinyl group
		1.11 Addition of the phytol tail
		1.12 NMR of large proteins
			1.12.1 Isotopic labelling
			1.12.2 TROSY experiments
			1.12.3 Structural restraints
			1.12.4 Non-uniform sampling (NUS)
	Chapter 2
	2. Materials and Methods
		2.1 Materials
		2.2 Standard buffers, reagents and media
		2.3 Bacterial strains and plasmids
		2.4 Production of competent E. coli cells
		2.5 Transformation of E. coli cells
		2.6 Over-expression of pET9His T.POR in E. coli without induction
			2.6.1 Growth of starter cultures
			2.6.2 Large Scale Culture Growths
		2.7 Over-expression of pET9His T.POR in E. coli by IPTG induction
			2.7.1 Growth of starter cultures
			2.7.2 Large Scale Culture Growths
			2.7.3 Measuring the Growth of E. Coli
			2.7.4 Protein Induction
		2.8 Harvesting E. coli cells
		2.9 Fractionation of E. coli cells
		2.10 Protein purification
			2.10.1 Ammonium Sulphate precipitation
			2.10.2 Purification of His-tagged POR on a Nickel column
			2.10.3 Purification of His-tagged POR on an SP Sepharose column
		2.11 Estimating the Concentration of His6-POR
		2.12 Site-directed mutagenesis
		2.13 Small-scale preparation of plasmid DNA (mini-prep)
		2.14 DNA sequencing
		2.15 Pigment preparation from R. sphaeroides ΔbchJ
			2.15.1 Growth of R. sphaeroides starter cultures
			2.15.2 Large Scale Growth
			2.15.3 Pigment Extraction
			2.15.4 Pigment Purification by high performance liquid chromatography
		2.16 POR Assays
			2.16.1 Concentration of Pchlide
			2.16.2 Concentration of NADPH
			2.16.3 POR Assay
		2.17 Formation of a POR-Pchlide-NADP(H) Ternary Complex
		2.18 SDS-polyacrylamide gel electrophoresis (SDS-PAGE)
		2.20 Labelling POR with MTSL
		2.21 NMR Experiments
		2.22 Asstools
		2.23 Analytic Ultracentrifugation
		2.24 Electron Paramagnetic Relaxation (EPR)
		2.25 Electron Microscopy (EM)
		2.26 Atomic Force Microscopy (AFM)
		2.27 Dynamic Light Scattering (DLS)
		2.28 Bioinformatics
	3. Bioinformatic analysis of protochlorophyllide oxidoreductase
		3.1  Introduction
		3.2  Structure and disorder predictions
			3.2.1  Structural Predictions for POR and the loop region
			3.2.2  Disorder Predictions for POR
		3.3  Homology modelling
		3.4  Database searching and Python scripting
			3.4.1  ClustalW alignments
		3.5  Structural Alignment
		3.6  Loop Structure Analysis
		3.7  Refinement of the POR Structural Model
	3.8  Discussion
	4. Preparation and structure determination of chlorophyll precursors
		4.1 Introduction
		4.2 Initial Protochlorophyllide Purification
			4.2.1 Initial Method for Pigment Expression and Bung Extraction
			4.2.2 Purification of Pigment on CM-Sepharose Column
			4.2.3 NMR of Pchlide
		4.3 Optimisation of Protochlorophyllide Preparation
			4.3.1 Protochlorophyllide Expression and Solvent Extraction
			4.3.2 HPLC Purification Using Reverse-phase Column
		4.4 NMR of Pure Protochlorophyllide
			4.4.1 One-Dimensional Protochlorophyllide Spectra
			4.4.2 13C-HSQC Protochlorophyllide Spectrum
			4.4.3 Selective NOE Experiments
			4.4.4 Final Pchlide Preparation
	The sample prepared in 4.4 can be produced in relatively large quantities by a straightforward protocol; the final Pchlide preparation has been shown to be highly pure by NMR.  This preparation has therefore been used for the production of complexes of POR, as described later in the chapter.  The increased purity of the pigment seems to have a dramatic effect on the complex formation, suggesting that the previous purification was inadequate for the production of Pchlide suitable for structural studies.  This new preparation should, therefore, be considered as the preferred method for production of Pchlide for use in future experiments.
		4.5 NMR and Structure Determination of A433 Pigment
			4.5.1 NMR of Mg-protoporphyrin monomethylester IX One-Dimensional Mg-protoporphyrin monomethylester IX Spectrum Selective NOE Experiments
			4.5.2 One-Dimensional A433 Spectrum
			4.5.3 13C-HSQC, TOCSY and Selective NOE experiments on A433 pigment
		4.5 Discussion
	Chapter 5
	5. Formation and Macromolecular Structure of POR Ternary Complex
		5.1 Introduction
		5.2  Optimising protein purification of POR
			5.2.1  Ammonium sulphate precipitation
			5.2.2  Nickel affinity chromatography
			5.2.3  Concentration of POR
			5.2.4  Anion Exchange Chromatography
		5.3 Ternary Complex Preparation using Solvent Solubilised Pchlide
			5.3.1 Formation of Ternary Complex
			5.3.2 Purification on SP Sepharose Column
			5.3.3 Native Protein Gels
			5.3.4 Dynamic Light Scattering (DLS)
			5.3.5 Analytical Ultracentrifugation (AUC)
		5.4 Ternary Complex Preparation using Detergent Solubilised Pchlide
			5.4.1 Formation of Ternary Complex
			5.4.2 Purification on SP Sepharose Column
			5.4.5 Electron Microscopy (EM) Aggregated Ternary Complex Sample Non-aggregated Ternary Complex sample
	The reason for the difference in structures observed between the two samples is unclear.  The breakdown of the initial large aggregates in the first sample is clearly lightdependent, and therefore must be a functionally related phenomenon; these aggregates subsequently break down into similar structures as observed in the second sample.  It is feasible that these smaller complexes are components of the larger structures, which in the case of the second sample, have not formed into these aggregates.  This could be due to a slight increase in detergent concentration, which prevents association into the larger aggregates; alternatively, a slightly different concentration of Pchlide may have been present and this may not have permitted the association of these complexes.
		5.4.6 Atomic Force Microscopy (AFM)
		5.4.7 Dynamic Light Scattering (DLS)
		5.4.8 Analytical Ultracentrifugation (AUC)
		5.5 Discussion
	6. Mutational analysis of the POR loop region
		6.1  Introduction
		6.2  Point Mutations of possible loop ‘hinge’ residues
		6.4  Enzymatic Assays of Point Mutants
		6.5  EPR of the POR loop mutants
			6.5.1  EPR of C37S/C89S/D164C/C198S mutants
			6.5.2  EPR of C37S/C89S/C198S mutants
			6.5.3  Activity of C226S mutant
			6.5.4  EPR of C37S/C89S/D164C/C198S/C226S mutants
		6.6  AUC
			6.6.1  AUC of apo protein samples
			6.6.2  AUC of ternary complex samples
			6.6.3  AUC of irradiated complex samples
		6.7  EM
		6.8  Discussion
	7. Preparation and NMR of 2H, 13C, 15N-labelled POR
		7.1 Introduction
		7.2 Optimisation of NMR pH
	Previous studies on POR involved the optimisation of the NMR conditions in order for production of the best possible quality spectra (Proudfoot, 2011).  The decision to perform experiments at pH 5.5 has since been questioned, due to the suggestion that broad signals, which are observed in the spectra, may be due to the presence of a molten globule state of POR at this pH.  In order to determine the optimal pH for NMR of POR, it was decided to carry out both 1D 1H and 2D 1H, 15N-HSQC experiments at a range of pH values.
	A 15N-labelled POR sample was prepared for NMR (detailed 2.7  2.10) and transferred to NMR buffer (detailed A3.6) with the pH adjusted to 7.5.  A 500 μl sample at 150 μM was placed in a NMR tube and sealed; the tube was then transferred to a Bruker Avance I 800 MHz NMR machine set to a temperature of 318 K (45 °C); both 1D 1H experiments and 2D 1H, 15N-HSQC experiments were performed on the sample.  After completion of these experiments, the sample was removed, the pH adjusted to 6.5 and the sample returned to the machine; the experiments were then repeated at this pH, and again at pH 5.5.
	The HSQC spectra at pH 5.5, 6.5 and 7.5 are shown in figure 7.1A, 7.1B and 7.1C, respectively.  The majority of the peaks in the three spectra overlay well, however it is clear that there is an increase of broadening in signals in the central region of the pH 5.5 and 6.5 spectra.  There are also, however, a number of extra peaks present in the pH 5.5 spectrum, compared to the pH 6.5 and 7.5 spectra, indicating that many signals that were not identifiable at higher pH, are observable at pH 5.5.  This suggests that the broadening in the centre of the pH 5.5 spectrum has arisen due to the presence of previously unidentified peaks with broad signals.
	The one-dimensional 1H experiments (Figure 7.1D) overlay extremely well, indicating no clear overall change in protein structure.  The upfield shifted proton peaks, at chemical shifts around or below 0 ppm, are representative of protons packed against the faces of aromatic side chains.  These peaks are, therefore, good indicators of any disruption in protein structure as this would lead to changes in the proximity of these aromatic groups and thus would cause these peaks to be shifted downfield; it is clear from figure 7.1D that these peaks are not shifted upon changes in pH.
	Based on these results, it is evident that the global structure of the protein is not affected by pH. POR does not exist fully in a molten globule state at pH 5.5.  NMR experiments performed at pH 5.5 undeniably give an increase in observable signals; some of these are broad peaks and thus contribute to the lack of clarity of the HSQC spectrum.  These broad signals are indicative of mobile structures in intermediate exchange, thus are likely to be from unstructured regions of POR; this makes these regions very interesting for studying the dynamic properties of the protein.  It was therefore decided to continue to use POR samples at pH 5.5 for any future experiments to ensure that the maximum number of signals could be observed.
		7.3 Production a of 2H, 13C, 15N-labelled POR sample
			7.3.1 Growth of E. coli in deuterated minimal media
			7.3.2 Optimisation of POR induction
			7.3.3 Optimisation of POR backbone amide exchange
			7.3.4 Comparison of sample with 15N-labelled POR
		7.4 Backbone assignment experiments on 2H, 13C, 15N-labelled POR
			7.4.1 Initial backbone assignment experiments at 600 MHz
			7.4.2 Backbone assignment experiments at 900 MHz
			7.4.3 Peak Picking
			7.4.3 Assignment
			7.4.4 Relaxation experiments
			7.4.5 TALOS-N Prediction of Protein Torsion Angles
			7.4.6 Application of NMR data to the POR Structural Model
		7.5 Discussion
	8. Conclusions and Future Work
		8.1 Conclusions
		8.2 Future work
	Appendix A Media and Buffer Recipes
Listpars and Pulse Programs Final
Document Text Contents
Page 1

Structural and Functional Studies of the

Light-Dependent Protochlorophyllide

Oxidoreductase Enzyme

David Robert Armstrong

Department of Molecular Biology and Biotechnology

A thesis submitted for the degree of Doctor of Philosophy

September 2014

Page 2

Page | i


The light-dependent enzyme protochlorophyllide oxidoreductase (POR) is a key enzyme in

the chlorophyll biosynthesis pathway, catalysing the reduction of the C17 - C18 bond in

protochlorophyllide (Pchlide) to form chlorophyllide (Chlide). This reaction involves the light-

induced transfer of a hydride from the nicotinamide adenine dinucleotide phosphate

(NADPH) cofactor, followed by proton transfer from a catalytic tyrosine residue. Much work

has been done to elucidate the catalytic mechanism of POR, however little is known about

the protein structure. POR isoforms in plants are also notable as components of the

prolamellar bodies (PLBs), large paracrystalline structures that are precursors to the thylakoid

membranes in mature chloroplasts.

Bioinformatics studies have identified a number of proteins, related to POR, which contain

similar structural features, leading to the production of a structural model for POR. A unique

loop region of POR was shown by EPR to be mobile, with point mutations within this region

causing a reduction in enzymatic activity. Production of a 2H, 13C, 15N-labelled sample of POR

for NMR studies has enabled significant advancement in the understanding of the protein

structure. This includes the calculation of backbone torsion angles for the majority of the

protein, in addition to the identification of multiple dynamic regions of the protein.

The protocol for purification of Pchlide, the substrate for POR, has been significantly

improved, providing high quality pigment for study of the POR ternary complex. Various

biophysical techniques have been used to study the macromolecular structure of these

complexes, indicating the formation of large aggregates of the cyanobacterial enzyme

induced by substrate binding, similar to PLBs. This has also led to the identification of ring

structures, composed of 5 and 6 monomers of POR, which are likely to be the primary

components of the cyanobacterial POR structures.

Page 101

Page | 82

5.2.2 Nickel affinity chromatography

The optimisation of the POR purification was, ultimately, to produce the best possible sample

of deuterated protein for NMR triple resonance experiments. Expression of protein grown in

deuterated medium was expected to be low therefore the purification protocol was required

to be optimised for low concentrations of POR. The nickel affinity resin chosen was Protino

Ni-TED resin, selected due to its high purification levels of polyhistidine tagged proteins and

high specificity. The protein was purified under gravity (detailed 2.10.3), using LEW Binding

buffer, LEW Wash buffer and LEW Elution buffer (detailed A3.1 – 3.3). Protein was eluted

directly into sample tubes containing DTT (to final concentration of 1 mM) and Roche

complete protease inhibitor.

Figure 5.2 SDS PAGE gel monitoring levels of POR during the purification. Marker lane is at the far left with
marker protein masses labelled on the left. POR expression prior to nickel affinity purification is in lane 2. Protein
that did not bind the nickel column is in lane 3. Nickel column elution, subsequently loaded to the SP sepharose
column, is in lane 4. Lanes 5 - 9 contain protein samples taken from the SP sepharose column, with lanes 5 and 6
from peak A and lanes 7 - 9 from peak B. The sample eluted from the nickel column clearly contains highly pure
POR, with very little protein lost in the flow-through. The SP sepharose column removes a number of further
contaminants, in lanes 5 and 6, giving a final sample with even higher purity.

Page 102

Page | 83

Samples from the nickel column addition, flow through and elution were run on SDS-PAGE

(Fig. 5.1) in lanes 2, 3 and 4 respectively. The addition in lane 2 shows that the expression of

POR was good, with it clearly being the most expressed protein with the largest band in the

gel. In the flow through sample in lane 3, very little, if any, POR can be seen indicating that it

had bound very efficiently to the resin. In the elution sample, which is also the sample that

was loaded to the SP sepharose column, there is a very large band for POR with only very

weak bands of contaminants therefore the purity after the nickel column was good.

5.2.3 Concentration of POR

After elution, the samples containing protein were pooled, transferred to a spin concentrator

and centrifuged at 3,000 rpm at 4 °C until reaching the desired volume. Previously, the use of

a spin concentrator had been replaced with an Amicon stirred ultrafiltration cell due to the

issue of protein precipitation at the bottom. However it was found that if the concentration

step was performed immediately after elution, with the relatively high salt concentration

from the buffer, then the protein would remain in solution throughout the concentration

process. This was the preferred method of concentration due to the increased control

possible, with the volume of the sample significantly more accurate to determine.

5.2.4 Anion Exchange Chromatography

In the original protocol, the anion exchange chromatography step was carried out using a SP

sepharose column, run using Tris pH 7.4 buffers. Due to the requirement for the protein to

be in phosphate buffer for NMR, this meant that the protein would have to be thoroughly

buffer exchanged before the sample is ready for NMR. It was decided to change the buffer

for the SP sepharose column to a phosphate pH 6.5 buffer for two reasons. Firstly, this pH is

further from the theoretical pI of 9.4 for POR therefore the ionic association with the column

is likely to be stronger; secondly, the pH is close to the pH required in NMR experiments,

therefore very minimal adjustments must be made to the sample following elution from the

SP sepharose column.

The SP sepharose column was run as detailed in 2.10.3, with the elution profile exhibited in

Fig. 5.3. The UV absorbance at 280 nm (blue line) was monitored from the point of protein

injection onto the column to follow the protein elution from the column. The pink line

displays the change in salt concentration throughout the run, with the values displayed on

the right hand axis. Protein that did not bind to the column was eluted at around 20 min

Similer Documents