1. Introduction 110

2. The occurrence of helical junctions in nucleic acids 111

2.1 The four-way DNA junction and genetic recombination 111

2.2 Helical junctions in RNA 112

2.3 Homology and branch migration of four-way junctions 112

2.4 Forms of helical junctions available for study 113

3. The structure of the four-way DNA junction 114

3.1 The global structure of the junction 114

3.2 The stacked X-structure 114

3.3 The junction has antiparallel character 115

3.4 The stereochemistry of the four-way DNA junction 116

3.4.1 Molecular modelling 116

3.4.2 NMR studies 116

3.4.3 Crystallography 116

4. Role of metal ions in the folding of the four-way DNA junction 122

4.1 An extended structure of the four-way junction at low salt concentrations 122

4.2 Structural interconversion between the extended and stacked X-structures 122

4.3 Location of structural metal ions in the four-way junction 124

5. Conformational variation in the four-way junction 125

5.1 Formation of alternative stacking conformers and sequence-dependent bias 125

5.1.1 Demonstration of alternative stacking conformers 125

5.1.2 Simultaneous presence of both stacking conformers 126

5.1.3 Exchange between stacking conformers 128

5.1.4 Longer-range sequence dependence 129

5.2 Variability in the interhelical angle 129

5.2.1 The interhelical angle 129

5.2.2 Variation, flexibility and malleability of the interhelical angle 130

5.3 Perturbation of the junction structure 130

6. Branch migration 131

6.1 Strand exchange between homologous sequences, and branch migration 131

6.2 The rate of branch migration 131

6.3 The effect of magnesium ions on branch migration rates 132

6.4 Branch migration and the structure of the DNA junction 132

7. Three-way DNA junctions 133

7.1 The perfectly basepaired three-way junction 133

7.2 The effect of unpaired bases; the bulged three-way junction 134

7.3 Two inequivalent stacking conformers 134

7.4 The stereochemistry of the bulged three-way junction 136

8. Helical junctions in RNA 136

8.1 The four-way junction in RNA 136

8.2 Some important four-way junctions in functional RNA species 137

8.2.1 The U1 snRNA junction 137

8.2.2 The four-way junction of the hairpin ribozyme 138

8.3 Three-way helical junctions in RNA 138

9. Recognition and distortion of four-way DNA junctions by proteins 139

9.1 Junction-resolving enzymes 139

9.1.1 Occurrence of junction-resolving enzymes 140

9.1.2 Cleavage of DNA junctions by resolving enzymes 140

9.1.3 Structure-selective binding of resolving enzymes to four-way junctions 143

9.1.4 Distortion of the structure of junctions by resolving enzymes 143

9.1.5 Relationship between distortion and cleavage of DNA junctions 144

9.2 Branch migration proteins 145

9.3 Site-specific recombinases 146

9.4 Other proteins 149

10. Summary and conclusions 149

11. Acknowledgements 151

12. References 151

Helical junctions in nucleic acids are important in biology. In DNA, their main significance is as intermediates in both homologous and site-specific recombination events. In RNA they are important architectural elements.

Helical junctions may be defined as branchpoints where double-helical segments intersect with axial discontinuities, such that strands are exchanged between the different helical sections. Thus the integrity of junctions is maintained by the covalent continuity of the component strands. Junctions can be perfectly basepaired, such that every base is paired with its Watson–Crick complement, or they can contain mismatches or unpaired bases; the latter can have significant effects on the folding of the structures. A systematic nomenclature exists for the unambiguous description of different junctions (Lilley et al. 1995) and some examples are illustrated in Fig. 1.

The purpose of the present article is to review what is known about the structures of helical junctions, and their recognition by proteins. The recent presentation of crystal structures of four-way junctions (Nowakowski et al. 1999; Ortiz-Lombardía et al. 1999; Eichman et al. 2000) provides a good opportunity to examine the current state of knowledge. We can also ask whether there are general principles behind the folding of branched nucleic acid species. Two possible principles emerge.

1. Introduction 1

1.1 The ‘band gap’ in silicon 2

2. Principles of CCD detector operation 3

2.1 Direct detection 3

2.2 Electron energy conversion into light 4

2.3 Optical coupling: lens or fibre optics? 6

2.4 Readout speed and comparison with film 8

3. Practical considerations for electron microscopic applications 9

3.1 Sources of noise 9

3.1.1 Dark current noise 9

3.1.2 Readout noise 9

3.1.3 Spurious events due to X-rays or cosmic rays 10

3.2 Efficiency of detection 11

3.3 Spatial resolution and modulation transfer function 12

3.4 Interface to electron microscope 14

3.5 Electron diffraction applications 15

4. Prospects for high-resolution imaging with CCD detectors 18

5. Alternative technologies for electronic detection 23

5.1 Image plates 23

5.2 Hybrid pixel detectors 24

6. References 26

During the past decade charge-coupled device (CCD) detectors have increasingly become the preferred choice of medium for recording data in the electron microscope. The CCD detector itself can be likened to a new type of television camera with superior properties, which makes it an ideal detector for recording very low exposure images. The success of CCD detectors for electron microscopy, however, also relies on a number of other factors, which include its fast response, low noise electronics, the ease of interfacing them to the electron microscope, and the improvements in computing that have made possible the storage and processing of large images.

CCD detectors have already begun to be routinely used in a number of important biological applications such as tomography of cellular organelles (reviewed by Baumeister, 1999), where the resolution requirements are relatively modest. However, in most high- resolution microscopic applications, especially where the goal of the microscopy is to obtain structural information at near-atomic resolution, photographic film has continued to remain the medium of choice. With the increasing interest and demand for high-throughput structure determination of important macromolecular assemblies, it is clearly important to have tools for electronic data collection that bypass the slow and tedious process of processing images recorded on photographic film.

In this review, we present an analysis of the potential of CCD-based detectors to fully replace photographic film for high-resolution electron crystallographic applications.

1. Introduction 308

2. Electron microscopy 311

2.1 Specimen preparation 311

2.2 The electron microscope 311

2.3 Acceleration voltage, defocus, and the electron gun 312

2.4 Magnification and data collection 313

3. Digitisation and CTF correction 317

3.1 The patchwork densitometer 318

3.2 Particle selection 320

3.3 Position dependent CTF correction 321

3.4 Precision of CTF determination 321

4. Single particles and angular reconstitution 323

4.1 Preliminary filtering and centring of data 323

4.2 Alignments using correlation functions 324

4.3 Choice of first reference images 324

4.4 Multi-reference alignment of data 325

4.5 MSA eigenvector/eigenvalue data compression 328

4.6 MSA classification 330

4.7 Euler angle determination (‘angular reconstitution’) 332

4.8 Sinograms and sinogram correlation functions 332

4.9 Exploiting symmetry 335

4.10 Three-dimensional reconstruction 337

4.11 Euler angles using anchor sets 339

4.12 Iterative refinements 339

5. Computational hardware/software aspects 341

5.1 The (IMAGIC) image processing workstation 342

5.2 Operating systems and GUIs 342

5.3 Computational logistics 344

5.4 Shared memory machines 344

5.5 Farming on loosely coupled computers 346

5.6 Implementation using MPI protocol 347

5.7 Software is what it's all about 347

6. Interpretation of results 348

6.1 Assessing resolution: the Fourier Shell Correlation 348

6.2 Influence of filtering 351

6.3 Rendering 351

6.4 Searching for known sub-structures 352

6.5 Interpretation 353

7. Examples 353

7.1 Icosahedral symmetry: TBSV at 5·9 Å resolution 354

7.2 The D6 symmetrical worm hemoglobin at 13 Å resolution 356

7.3 Functional states of the 70S E. coli ribosome 357

7.4 The 50S E. coli ribosomal subunit at 7·5 Å resolution 359

8. Perspectives 361

9. Acknowledgements 364

10. References 364

In the past few years, electron microscopy (EM) has established itself as an important – still upcoming – technique for studying the structures of large biological macromolecules. EM is a very direct method of structure determination that complements the well-established techniques of X-ray crystallography and NMR spectroscopy. Electron micrographs record images of the object and not just their diffraction patterns and thus the classical ‘phase’ problem of X-ray crystallography does not exist in EM. Modern microscopes may reach resolution levels better than ∼ 1·5 Å, which is more than sufficient to elucidate the polypeptide backbone in proteins directly. X-ray structures at such resolution levels are considered ‘excellent’. The fundamental problem in biological EM is not so much the instrumental resolution of the microscopes, but rather the radiation sensitivity of the biological material one wants to investigate. Information about the specimen is collected in the photographic emulsion with the arrival of individual electrons that have (elastically) interacted with the specimen. However, many electrons will damage the specimen by non-elastic interactions. By the time enough electrons have passed through the object to produce a single good signal-to-noise (SNR) image, the biological sample will have been reduced to ashes. In contrast, stable inorganic specimens in material science often show interpretable details down to the highest possible instrumental resolution.

1. Background to RNA structure 200

1.1 Types of RNA 200

1.1.1 Transfer RNA (tRNA) 200

1.1.2 Messenger RNA (mRNA) 201

1.1.3 Ribosomal RNA (rRNA) 201

1.1.4 Other ribonucleoprotein particles 202

1.1.5 Viruses and viroids 202

1.1.6 Ribozymes 202

1.2 Elements of RNA secondary structure 203

1.3 Secondary structure versus tertiary structure 205

2. Theoretical and computational methods for RNA secondary structure determination 208

2.1 Dynamic programming algorithms 208

2.2 Kinetic folding algorithms 210

2.3 Genetic algorithms 212

2.4 Comparative methods 213

3. RNA thermodynamics and folding mechanisms 216

3.1 The reliability of minimum free energy structure prediction 216

3.2 The relevance of RNA folding kinetics 218

3.3 Examples of RNA folding kinetics simulations 221

3.4 RNA as a disordered system 227

4. Aspects of RNA evolution 233

4.1 The relevance of RNA for studies of molecular evolution 233

4.1.1 Molecular phylogenetics 234

4.1.2 tRNAs and the genetic code 234

4.1.3 Viruses and quasispecies 235

4.1.4 Fitness landscapes 235

4.2 The interaction between thermodynamics and sequence evolution 236

4.3 Theory of compensatory substitutions in RNA helices 238

4.4 Rates of compensatory substitutions obtained from sequence analysis 240

5. Conclusions 246

6. Acknowledgements 246

7. References 246

This article takes an inter-disciplinary approach to the study of RNA secondary structure, linking together aspects of structural biology, thermodynamics and statistical physics, bioinformatics, and molecular evolution. Since the intended audience for this review is diverse, this section gives a brief elementary level discussion of the chemistry and structure of RNA, and a rapid overview of the many types of RNA molecule known. It is intended primarily for those not already familiar with molecular biology and biochemistry.

Ribonucleic acid consists of a linear polymer with a backbone of ribose sugar rings linked by phosphate groups. Each sugar has one of the four ‘bases’ adenine, cytosine, guanine and uracil (A, C, G, and U) linked to it as a side group. The structure and function of an RNA molecule is specific to the sequence of bases. The phosphate groups link the 5′ carbon of one ribose to the 3′ carbon of the next. This imposes a directionality on the backbone. The two ends are referred to as 5′ and 3′ ends, since one end has an unlinked 5′ carbon and one has an unlinked 3′ carbon. The chemical differences between RNA and DNA (deoxyribonucleic acid) are fairly small: one of the OH groups in ribose is replaced by an H in deoxyribose, and DNA contains thymine (T) bases instead of U. However, RNA structure is very different from DNA structure. In the familiar double helical structure of DNA the two strands are perfectly complementary in sequence. RNA usually occurs as single strands, and base pairs are formed intra-molecularly, leading to a complex arrangement of short helices which is the basis of the secondary structure. Some RNA molecules have well-defined tertiary structures. In this sense, RNA structures are more akin to globular protein structures than to DNA.

The role of proteins as biochemical catalysts and the role of DNA in storage of genetic information have long been recognised. RNA has sometimes been considered as merely an intermediary between DNA and proteins. However, an increasing number of functions of RNA are now becoming apparent, and RNA is coming to be seen as an important and versatile molecule in its own right.

1. Introduction 255

1.1 General thermodynamics 256

2. Nucleic acid thermodynamics 260

2.1 DNA duplexes 261

2.2 RNA duplexes 263

2.3 Hybrid DNA–RNA duplexes 264

2.4 Hydration 267

2.5 Conformational flexibility 269

2.6 Thermodynamics 272

3. Nucleic acid–ligand interactions 277

3.1 Minor groove binders 278

3.2 DNA intercalators 284

3.3 Triple-helical systems 288

3.3.1 Structures 288

3.3.2 Hydration 291

3.3.3 Thermodynamics 291

4. Conclusions 295

5. Acknowledgements 298

6. References 298

In recent years the availability of large quantities of pure synthetic DNA and RNA has revolutionised the study of nucleic acids, such that it is now possible to study their conformations, dynamics and large-scale properties, and their interactions with small ligands, proteins and other nucleic acids in unprecedented detail. This has led to the (re)discovery of higher order structures such as triple helices and quartets, and also the catalytic activity of RNA contingent on three-dimensional folding, and the extraordinary specificity possible with DNA and RNA aptamers.

Nucleic acids are quite different from proteins, even though they are both linear polymers formed from a small number of monomeric units. The major difference reflects the nature of the linkage between the monomers. The 5′–3′ phosphodiester linkage in nucleic acids carries a permanent negative charge, and affords a relatively large number of degrees of freedom, whereas the essentially rigid planar peptide linkage in proteins is neutral and provides only two degrees of torsional freedom per backbone residue. These two properties conspire to make nucleic acids relatively flexible and less likely to form extensive folded structures. Even when true 3D folded structures are formed from nucleic acids, the topology remains simple, with the anionic phosphates forming the surface of the molecule. Nevertheless, nucleic acids do occur in a variety of structures that includes single strands and high-order duplex, triplex or tetraplex (‘quadruplex’) forms. The principles of biological recognition and the related problem of understanding the forces that stabilise such folded structures are in some respects more straightforward than for proteins, making them attractive model systems for understanding general biophysical problems. This view is aided by the relatively facile chemical synthesis of pure nucleic acids of any desired size and defined sequence, and the ease of incorporation of a wide spectrum of chemically modified bases, sugars and backbone linkers. Such modifications are considerably more difficult to achieve with oligopeptides or proteins.

1. Transverse relaxation and the molecular size limit in liquid state NMR 161

2. TROSY: how does it work? 163

2.1 Transverse relaxation in coupled spin systems 163

2.2 The TROSY effect, relaxation due to remote protons and 2H isotope labeling 165

3. Direct heteronuclear chemical shift correlations 168

3.1 Single-Quantum [15N,1H]-TROSY 168

3.2 Zero-Quantum [15N,1H]-TROSY 171

3.3 Single-Quantum TROSY with aromatic 13C–1H moieties 176

4. Resonance assignment and NOE spectroscopy of large biomolecules 180

4.1 TROSY-based triple resonance experiments for 13C, 15N and 1HN backbone resonance assignment in uniformly 2H, 13C, 15N labeled proteins 180

4.2 TROSY-type NOE spectroscopy 186

5. Scalar coupling across hydrogen bonds observed by TROSY 187

6. The use of TROSY for measurements of residual dipolar coupling constants 190

7. Conclusions 191

8. Acknowledgements 191

9. References 191

The application of nuclear magnetic resonance (NMR) spectroscopy for structure determination of proteins and nucleic acids (Wüthrich, 1986) with molecular mass exceeding 30 kDa is largely constrained by two factors, fast transverse relaxation of spins of interest and complexity of NMR spectra, both of which increase with increasing molecular size (Wagner, 1993b; Clore & Gronenborn, 1997, 1998b; Kay & Gardner, 1997). The good news is that neither of these factors represent a fundamental limit for the application of NMR techniques to protein structure determination in solution (Clore & Gronenborn, 1998a; Wüthrich, 1998; Wider & Wüthrich, 1999). In fact, in the past few years the size limitations imposed by these factors have been pushed up to 50–70 kDa by the use of 13C, 15N and 2H isotope labeling combined with selective reprotonation of individual chemical groups in conjunction with the use of triple-resonance experiments (Bax, 1994; Gardner et al. 1997; Gardner & Kay, 1998) and heteronuclear-resolved NMR (Fesik & Zuiderweg, 1988; Marion et al. 1989a; Otting & Wüthrich, 1990). Among the largest biomolecules whose 3D structure was solved by NMR are the 44 kDa trimeric ectodomain of simian immunodeficiency virus (SIV) gp41 (Caffrey et al. 1998) and 40–60 kDa particles of the elongation initiation factor 4E solubilized in CHAPS micelles (Matsuo et al. 1997; McGuire et al. 1998).

1. Introduction 29

2. Landmarks in NMR of macromolecules 32

2.1 Protein structures and methods development 32

2.1.1 Sequential assignment method 32

2.1.2 Triple-resonance experiments 34

2.1.3 Structures of large proteins 36

2.2 Protein–nucleic acid complexes 37

2.3 RNA structures 38

2.4 Membrane-bound systems 39

3. NMR spectroscopy today 40

3.1 State-of-the-art structure determination 41

3.2 New methods 44

3.2.1 Residual dipolar couplings 44

3.2.2 Direct detection of hydrogen bonds 44

3.2.3 Spin labeling 45

3.2.4 Segmental labeling 46

3.3 Protein complexes 47

3.4 Mobility studies 50

3.5 Determination of time-dependent structures 52

3.6 Drug discovery 53

4. The future of NMR 54

4.1 The ease of structure determination 54

4.2 The ease of making recombinant protein 55

4.3 Post-translationally modified proteins 55

4.4 Approaches to large and/or membrane-bound proteins 56

4.5 NMR in structural genomics 56

4.6 Synergy of NMR and crystallography in protein structure determination 56

5. Conclusion 57

6. Acknowledgements 57

7. References 57

Since the publication of the first complete solution structure of a protein in 1985 (Williamson et al. 1985), tremendous technological advances have brought nuclear magnetic resonance spectroscopy to the forefront of structural biology. Innovations in magnet design, electronics, pulse sequences, data analysis, and computational methods have combined to make NMR an extremely powerful technique for studying biological macromolecules at atomic resolution (Clore & Gronenborn, 1998). Most recently, new labeling and pulse techniques have been developed that push the fundamental line-width limit for resolution in NMR spectroscopy, making it possible to obtain high-field spectra with better resolution than ever before (Dötsch & Wagner, 1998). These methods are facilitating the study of systems of ever-increasing complexity and molecular weight.

1. Introduction 372

1.1 Residual dipolar couplings as a route to structure and dynamics 372

1.2 A brief history of oriented phase high resolution NMR 374

2. Theoretical treatment of dipolar interactions 376

2.1 Anisotropic interactions as probes of macromolecular structure and dynamics 376

2.1.1 The dipolar interaction 376

2.1.2 Averaging in the solution state 377

2.2 Ordering of a rigid body 377

2.2.1 The Saupe order tensor 378

2.2.2 Orientational probability distribution function 380

2.2.3 The generalized degree of order 380

2.3 Molecular structure and internal dynamics 381

3. Inducing molecular order in high resolution NMR 383

3.1 Tensorial interactions between the magnetic field and anisotropic magnetic susceptibilities 383

3.2 Dilute liquid crystal media: a tunable source of order 384

3.2.1 Bicelles : from membrane mimics to aligning media 385

3.2.2 Filamentous phage 387

3.2.3 Transfer of alignment from ordered media to macromolecules 388

3.3 Magnetic field alignment 389

3.3.1 Paramagnetic assisted alignment 389

3.3.2 Advantages of using magnetic alignment 389

4. The measurement of residual dipolar couplings 391

4.1 Introduction 391

4.2 Frequency based methods 392

4.2.1 Coupling enhanced pulse schemes 392

4.2.2 In phase anti-phase methods (IPAP): 1DNH couplings in proteins 393

4.2.3 Exclusive correlated spectroscopy (E-COSY): 1DNH, 1DNC′ and 2DHNC′ 395

4.2.4 Extraction of splitting values from the frequency domain 396

4.3 Intensity based experiments 397

4.3.1 J-Modulated experiments: the measurement of 1DCαHα in proteins 397

4.3.2 Phase modulated methods 399

4.3.3 Constant time COSY – the measurement of DHH couplings 399

4.3.4 Systematic errors in intensity based experiments 400

5. Interpretation of residual dipolar coupling data 401

5.1 Structure determination protocols utilizing orientational constraints 401

5.1.1 The simulated annealing approach 401

5.1.2 Order matrix analysis of dipolar couplings 402

5.1.3 A discussion of the two approaches 402

5.2 Reducing orientational degeneracy 403

5.2.1 Multiple alignment media in the simulated annealing approach 404

5.2.2 Multiple alignment media in the order matrix approach 405

5.3 Simplifying effects arising due to molecular symmetry 406

5.4 Database approaches for determining protein structure 407

6. Applications to the characterization of macromolecular systems 408

6.1 Protein structure refinement 408

6.2 Protein domain orientation 409

6.3 Oligosaccharides 413

6.4 Biomolecular complexes 415

6.5 Exchanging systems 416

7. Acknowledgements 418

8. References 419

Within its relatively short history, nuclear magnetic resonance (NMR) spectroscopy has managed to play an important role in the characterization of biomolecular structure. However, the methods on which most of this characterization has been based, Nuclear Overhauser Effect (NOE) measurements for short-range distance constraints and scalar couplings measurements for torsional constraints, have limitations (Wüthrich, 1986). For extended structures, such as DNA helices, for example, propagation of errors in the short distance constraints derived from NOEs leaves the relative orientation of remote parts of the structures poorly defined. Also, the low density of observable protons in contact regions of molecules held together by factors other than hydrophobic packing, leads to poorly defined structures. This is especially true in carbohydrate containing complexes where hydrogen bonds often mediate contacts, and in multi-domain proteins where the area involved in domain–domain contact can also be small. Moreover, most NMR based structural applications are concerned with the characterization of a single, rigid conformer for the final structure. This can leave out important mechanistic information that depends on dynamic aspects and, when motion is present, this can lead to incorrect structural representations. This review focuses on one approach to alleviating some of the existing limitations in NMR based structure determination: the use of constraints derived from the measurement of residual dipolar couplings (D).

1. Introduction 68

2. Ferredoxin reduction by photosystem I 72

3. Ferredoxins 73

4. Ferredoxin[ratio ]thioredoxin reductase 73

4.1 Spectroscopic investigations of FTR 76

4.2 The three-dimensional structure of FTR from the cyanobacterium Synechocystis sp. PCC6803 77

4.2.1 The variable subunit 77

4.2.2 The catalytic subunit 81

4.2.3 The iron–sulfur center and active site disulfide bridge 82

4.2.4 The dimer 84

4.3 Thioredoxin f and m 85

4.4 Ferredoxin and thioredoxin interactions 86

4.5 Mechanism of action 88

4.6 Comparison with other chloroplast FTRs 92

5. Target enzymes 95

5.1 NADP-dependent malate dehydrogenase 95

5.1.1 Regulatory role of the N-terminal extension 97

5.1.2 Regulatory role of the C-terminal extension 99

5.1.3 Thioredoxin interactions 101

5.2 Fructose-1,6-bisphosphatase 101

5.3 Redox regulation of chloroplast target enzymes 103

6. Conclusion 103

7. Acknowledgements 104

8. References 104

A pre-requisite for life on earth is the conversion of solar energy into chemical energy by photosynthetic organisms. Plants and photosynthetic oxygenic microorganisms trap the energy from sunlight with their photosynthetic machinery and use it to produce reducing equivalents, NADPH, and ATP, both necessary for the reduction of carbon dioxide to carbohydrates, which are then further used in the cellular metabolism as building blocks and energy source. Thus, plants can satisfy their energy needs directly via the light reactions of photosynthesis during light periods. The situation is quite different in the dark, when these organisms must use normal catabolic processes like non-photosynthetic organisms to obtain the necessary energy by degrading carbohydrates, like starch, accumulated in the chloroplasts during daylight. The chloroplast stroma contains both assimilatory enzymes of the Calvin cycle and dissimilatory enzymes of the pentose phosphate cycle and glycolysis. This necessitates a strict, light-sensitive control that switches between assimilatory and dissimilatory pathways to avoid futile cycling (Buchanan, 1980, 1991; Buchanan et al. 1994; Jacquot et al. 1997; Schürmann & Buchanan, 2000).