Protein Crystallography A Concise Guide

Visit the link below to download the full version of this book:

https://medidownload.com/product/protein-crystallography-a-concise-guide/

Click Download Now

Protein Crystallography
A Concise Guide

Eaton E. Lattman and Patrick J. Loll

The Johns Hopkins University Press

Baltimore
∫ 2008 The Johns Hopkins University Press
All rights reserved. Published 2008
Printed in the United States of America on acid-free paper
987654321

The John Hopkins University Press

2715 North Charles Street
Baltimore, Maryland 21218-4363
www.press.jhu.edu

Lattman, Eaton.
Protein crystallography : a concise guide / by Eaton E. Lattman and Patrick J. Loll.
p. ; cm.
Includes bibliographical references and index.
ISBN-13: 978-0-8018-8806-9 (hardcover : alk. paper)
ISBN-10: 0-8018-8806-9 (hardcover : alk. paper)
ISBN-13: 978-0-8018-8808-3 (pbk. : alk. paper)
ISBN-10: 0-8018-8808-5 (pbk. : alk. paper)
1. Proteins—Structure. 2. X-ray crystallography. I. Loll, Patrick. II. Title.
[DNLM: 1. Proteins—chemistry. 2. Crystallography, X-Ray—methods. 3. Protein Conformation.
4. Proteomics—methods. QU 55 L364p 2008]
QP551.L345 2008
612%.01575—dc22 2007036099

A catalog record for this book is available from the British Library.

Special discounts are available for bulk purchases of this book. For more information, please contact
Special Sales at 410-516-6936 or [email protected].

The Johns Hopkins University Press uses environmentally friendly book materials, including
recycled text paper that is composed of at least 30 percent post-consumer waste, whenever possible.
All of our book papers are acid-free, and our jackets and covers are printed on paper with recycled
content.
To David Sayre
This page intentionally left blank
 Contents

Preface ix

Chapter 1. Introduction 1
1.1 What Is X-ray Crystallography? 1

1.2 A Quick Look at Protein Crystals 12

1.3 Noncrystalline Specimens 15

1.4 Summary 17

Further Reading 17

Chapter 2. A Physical Understanding of Di√raction 19

2.1 What Is Di√raction? 19

2.2 Di√raction from One-Dimensional Crystals 25

2.3 Reconstructing Images from Di√raction Patterns 29

2.4 Summary 35

Further Reading 36

Chapter 3. Di√raction from Three-Dimensional Crystals 37

3.1 The Electron Density Function in Three Dimensions 37

3.2 Calculating the Di√raction Pattern from a Known Structure 49
3.3 Summary 53
Further Reading 54
viii Contents

Chapter 4. Phase Determination by Isomorphous Replacement 55

4.1 Measuring the Phases 55

4.2 MAD Phasing 61

4.3 Fitting Models to Experimental Electron Density Maps 68
4.4 Summary 70

Further Reading 70

Chapter 5. The Patterson Function 72

5.1 Definition of the Patterson Function 72

5.2 Using the Patterson Function to Locate Atoms 76

5.3 Summary 81
Further Reading 82

Chapter 6. Phasing with Partially Known Structures 83

6.1 Di√erence Fourier Maps 83

6.2 Molecular Replacement 87

6.3 Summary 101

Further Reading 101

Chapter 7. Crystallographic Refinement 102

7.1 Refinement Improves the Model 102

7.2 Least-Squares Refinement 103

7.3 Summary 111

Further Reading 112

Glossary 113
Index 131
 Preface

Why another book on protein crystallography? Because we believe that there is a

demand for a concise, accessible introduction to this powerful method. Proteins,
as the agents that carry out the functions of cells and tissues, are assuming a
paramount role in research areas ranging from drug development to cell biology.
The atomic structures of proteins, as revealed by crystallography, are increasingly
being used to provide detailed insights into function and mechanism. Indeed,
dealing with protein structure is becoming unavoidable. Thus, the number of
scientists who desire a basic grasp of crystallography is growing, but there has not
been a volume available that o√ers an appropriate introduction.
We have designed our book to meet this need. First, it is quite short, less than
half the length of most other books on this topic. Second, we have worked hard to
make it accessible. It has lots of illustrations, and equations are introduced by
using intuitive arguments. Third, it focuses tightly on crystallography as an
imaging or microscopical process, deemphasizing the practical details. Finally, it
contains a glossary that defines many crystallographic terms. The book is aimed
at anyone who would like a trip through the essentials of crystallography. It will
not turn you into a crystallographer; but we hope reading it will make it easier to
converse with one.

The authors are grateful to Alexander McPherson and Jason McLellan for the gift
of figures and to Michaelis Hadjithomas for creating and drawing many of the
figures in the book. They thank each other for support during periods of slow
progress. Many figures in this book were prepared with the help of PyMOL (De-
Lano, W. L. The PyMOL Molecular Graphics System [2002]; www.pymol.org).
This page intentionally left blank
Protein Crystallography
This page intentionally left blank
 1
Introduction

1.1 What Is X-ray Crystallography?

In 1912 Walter Friedrich and Paul Knipping, following up on Max von Laue’s
prediction, illuminated a crystal of copper sulfate with X rays and obtained the
first X-ray di√raction pattern. This simple experiment has evolved into a power-
ful technique that has enabled us to decipher the structure of DNA and of protein
molecules. This technique has resulted in at least eight Nobel prizes, and re-
searchers in the field anticipate more to come.
X-ray crystallography is a powerful form of microscopy that allows us
to visualize atoms and molecules. Almost all that we know about the three-
dimensional structures of proteins and nucleic acids, we have learned from the
use of X-ray crystallography. This method provides images of molecular struc-
tures that are far more detailed than any provided by light microscopy because
the very short wavelengths of X rays allow them to ‘‘sense’’ structural variation at
the atomic level.
X-ray crystallography di√ers from the more familiar light microscopy in
several important ways. The key di√erences are:

1. Crystallography uses X rays instead of visible light. The X rays used in crys-
tallography have wavelengths of about 1 Å = 10–10 meters, while visible light
has a wavelength of &5000 Å.
2. Unlike microscopy, X-ray crystallography is lensless. We are familiar with the
idea that X rays pass straight through objects—like people!—without being
1
2 Protein Crystallography

deflected. This suggests why there are no substances that make good lenses
for X rays: it is the job of a lens to bend or refract light, and X rays are so
energetic that they resist bending.
3. The specimens used in crystallography are crystals, which contain many
perfectly repeated copies of the molecules we wish to see. In contrast, micro-
scopists frequently examine single objects.

The very short wavelength of X rays is what makes them useful for studying
the structure of matter at the atomic level. The ability of a microscope to resolve

Figure 1.1. Summation of two circular waves emanating from two sources near to
each other. Such waves might be produced by light interacting with two small holes in
an optical mask, or by two pebbles dropped simultaneously into a pool of water.
Bright areas represent the wave crests, and dark areas correspond to the wave troughs.
The two individual waves are shown on the left, and their sum is shown on the right.
In the upper panels, the waves are emanating from points that are separated by a
distance d1 ⬇ twice the wavelength. The resultant wave shows clear interference
patterns, which would reveal to an observer some distance away that this resultant
wave pattern contains contributions from two waves. In other words, the observer
can resolve the two sources. In the lower panels, the waves emanate from points
separated by a distance d2, which is less than the wavelength. In this case, the resultant
wave pattern appears almost identical with a wave coming from a single source. It is
impossible to determine whether this wave pattern was generated from one pebble or
two, and an observer would therefore be unable to resolve the two sources. In this
manner, the wavelength of the scattered radiation limits our ability to distinguish two
closely spaced points.
Introduction 3

fine details is limited by the wavelength of the radiation used. For example, two
carbon atoms joined by a single bond are about 1.5 Å apart. To resolve these two
atoms as separate objects, we need to use a wavelength of light that is (roughly)
no larger than the distance between the atoms; light with such a short wavelength
falls in the X-ray region of the spectrum.
We give the name resolution to the ability to distinguish or resolve two nearby
points. The e√ect of wavelength upon resolution is illustrated in Figure 1.1. The
upper panel shows waves emanating from two well-separated points; the lower
panel shows what happens when the two points are separated by less than one
wavelength. In the latter case, the two wave patterns blend together and can
hardly be distinguished from the waves emanating from a single point. Thus, it is
impossible to know whether the pattern of scattered waves results from two
closely spaced objects or a single larger object.
We have described X rays as electromagnetic waves with short wavelengths. It
is one of the counterintuitive findings of quantum mechanics that light can be
equally well described as a stream of particles called photons. The energy of a
photon of light is inversely proportional to its wavelength. Because the X-ray
photons used in crystallography have short wavelengths, they have high energies
(&10,000 electron volts). The highly energetic character of X rays complicates the
imaging process, as described below.
In an ordinary light microscope, light waves scattered by the specimen are
collected by a high-quality lens (or system of lenses) and focused to form an
image. This is illustrated in Figure 1.2a. Because X rays have much greater pene-
trating power than visible light, it is more di≈cult to fabricate lenses that are able
to focus X rays. The best X-ray lenses currently known are of poor quality
compared with optical lenses, and they cannot produce high-resolution images.
Thus, as shown in Figure 1.2b, we carry out crystallographic experiments with-
out lenses, collecting the scattered radiation directly on film or some other
detector. The pattern that the scattered radiation makes on the detector is the
di√raction pattern. We use this measured di√raction to perform calculations on a
computer that mimic the action of the lens. These calculations allow us to
recombine the scattered radiation to form an image.
Unfortunately, calculating the image is not quite as simple as we have sug-
gested. Experimental limitations preclude our extracting all the information
contained in the di√raction pattern. Specifically, a complex number is required
to completely describe the scattered waves that make up the di√raction pattern;
Figure 1.2. (a) Schematic of a conventional imaging experiment with visible light.
Here, the specimen is represented by a cat. Radiation scattered by the cat is collected
by the lens, and an image is created at the image plane of the lens. For each point in
the specimen, the lens collects light scattered in many di√erent directions and re-
directs it to form an image of that point. (b, c) Schematics of imaging experiments
involving proteins. In (b) a single molecule is shown scattering X rays that are
collected by a detector with no intervening lens. The pattern on the detector will look
nothing like the object itself. The computer and monitor symbolize the process by
which calculations that mimic the function of a lens are used to create an image. In (c)
a protein crystal is shown as the specimen, along with the punctate di√raction pattern
characteristic of crystals. In (b) and (c) the horizontal line entering from the left
represents the incident X-ray beam and the diagonal lines represent di√racted rays.
Adapted from Crystallography Made Crystal Clear by Gale Rhodes.
Introduction 5

however, we are able to measure only the squared modulus of this complex
number. We need to know the correct complex square root of this number to
create the image. In practical terms, we can measure the amplitude of the dif-
fracted rays but not their relative phases. The problem of recovering the correct
square roots is the phase problem and is considered in more detail in Chapter 3.
Crystallographic experiments require crystals, as shown in Figure 1.2c. Why
do we use crystals? In principle we could do an X-ray experiment to image a
single molecule, but there are two practical obstacles to this. First, it would be
impossible to measure the di√raction pattern from a single molecule because it
would be too weak and drowned in noise from scattering by other elements of
the system. The second obstacle is specimen damage; a single molecule would be
burned up by the X rays before it could give rise to a useful di√raction pattern.
Crystals help us with both of these di≈culties.
Crystalline specimens greatly increase signal-to-noise in the measured dif-
fraction pattern. The protein* crystals we use contain 1012 or more molecules,
and so di√ract at least 1012 times as much radiation as a single molecule. For
reasons that we discuss later, the di√raction pattern from a crystal is confined to
rays or beams emanating in certain directions, which form ‘‘spots’’ on the film,
making the pattern much easier to measure. These spots are often called reflec-
tions because the incoming X-ray beam appears to be reflected by the crystal to
form the outgoing ray.
A typical example of a crystallographic di√raction pattern is shown in Figure
1.3. The complete di√raction pattern comprises many pictures such as this one,
taken with the crystal in various orientations. The number of independent reflec-
tions in the di√raction pattern of a protein crystal is very large, typically tens or
even hundreds of thousands. This is a tremendous amount of information,
which should not be surprising—the di√raction pattern specifies the detailed
image of a molecule that may contain thousands of atoms.
Crystalline specimens also reduce specimen damage. The X rays that create
the di√raction pattern are scattered from all the molecules in the crystal lattice, so
each individual molecule receives a much smaller dose. For example, if 1012
photons are required to create a high-resolution di√raction pattern and a crystal
contains 1012 molecules, then on average each molecule is hit by only one photon.

*This book describes the use of crystallography to determine protein structure. The reader
should be aware, however, that most of the techniques we discuss are equally applicable to other
macromolecules such as DNA.
6 Protein Crystallography

Figure 1.3. This figure shows the di√raction pattern created when a crystal of the
protein ferritin was rotated through a small angle (0.5\) while illuminated by the
X-ray beam. The white feature extending from the left is the shadow of the beam stop,
a small piece of metal positioned to capture the undi√racted X rays that pass through
the crystal. The reasons behind the punctate character of the di√raction pattern will
be discussed later in the book.

Crystallography is an imaging technique. Any image maps a particular char-

acteristic of an object—color, reflectivity, and so on. X rays are scattered by
electrons, and so our image is a record of the spatial distribution of the mole-
cule’s electrons. This image is the electron density function. The maximum resolu-
tion that can be achieved in the electron density image is determined by the
wavelength of the light used. However, factors such as disorder in the crystals
degrade the resolution of the image, and as a result it is generally not possible to