MTX1071
`ModernaTX, Inc. v. CureVac AG
`IPR2017-02194
`
`1
`
`

`

`To:
`
`Our parents, who encouraged us,
`Our teachers, who enabled us, and
`
`Our children, who put up with us.
`
`Cover Art: One of a series of color studies of horse heart
`
`cytochrome c designed to show the influence of amino
`acid side chains on the protein’s three-dimensional fold-
`ing pattern. We have selected this study to symbolize
`the discipline of biochemistry: Both are beautiful but
`still
`in process and hence have numerous ”rough
`edges." Drawing by Irving Geis in collaboration with
`Richard E. Dickerson.
`
`Cover and part opening illustrations
`copyrighted by Irving Geis.
`
`Cover Designer: Madelyn Lesure
`
`Photo Research: John Schultz, Eloise Marion
`
`Photo Research Manager: Stella Kupferberg
`
`Illustration Coordinator: Edward Starr
`
`Copy Editor: Jeannette Stiefel
`
`Production Manager: Lucille Buonocore
`
`Senior Production Supervisor: Linda Muriello
`
`Copyright © 1990, by John Wiley 8: Sons, Inc.
`
`All rights reserved. Published simultaneously in Canada.
`
`Reproduction or translation of any part of
`this work beyond that permitted by Sections
`107 and 108 of the 1976 United States Copyright
`Act without the permission of the copyright
`owner is unlawful. Requests for permission
`or further information should be addressed to
`the Permissions Department, John Wiley & Sons.
`
`Library of Congress Cataloging in Publication Data:
`Voet, Donald.
`Biochemistry / by Donald Voet and Judith G. Voet.
`p.
`cm.
`Includes bibliographical references.
`ISBN 0-471—61769-5
`
`1. Biochemistry. I. Voet, Judith C. [1. Title.
`QP514.2.V64
`1990
`574.19'2—dc20
`
`Printed in the United States of America
`
`10987654321
`
`89—16727
`CIP
`
`
`
`2
`
`

`

`
`NUCLEIC
`
`— C
`
`hapter 28
`
`
`
`his
`
`lac
`
`
`
`
`
`
`
`
`
`
`
`
`1. Chemical Structure and Base Composition
`_
`2. Double Helical Structures
`
`A. The Watson—Crick Structure: B-DNA
`B. Other Nucleic Acid Helices
`
`C. The Size of DNA
`
`3. Forces Stabilizing Nucleic Acid Structures
`A. Denaturation and Renaturation
`
`B. Sugar— Phosphate Chain Conformations
`C. Base Pairing
`0- Base Stacking and Hydrophobic Interactions
`E. Ionic Interactions
`
`
`
`C. Chain-Terminator Method
`D. RNA Sequencing
`
`7. Chemical Synthesis of Oligonucleotides
`
`8. Molecular Cloning
`
`Cloning Vectors
`
`Gene Splicing
`Genomic Libraries
`
`DNA Amplification by the Polymerase Chain Reaction
`Production of Proteins
`
`Social Considerations
`
`TWPQFDP"
`
`
`
`
`
`
`
`
`There are two classes of nucleic acids, deoxyribonucleic
`acid (DNA) and ribonucleic acid (RNA). DNA is the
`
`hereditary molecule in all cellular life forms, as well as in
`many viruses. It has but two functions:
`_
`1. To direct its own replication during cell division.
`.
`.
`.
`2. To direct the transcrlptlon of complementary mole-
`cules of RNA.
`
`RNA, in contrast, has more vaned biological functions:
`1. The RNA transcripts of DNA sequences that specify
`polypeptides, messenger RNA (mRNA), direct the
`ribosomal synthesis of these polypeptides in a pro-
`cess known as translation.
`
`4- Hacieic Acid Fractionation
`A. Solution Methods
`B. Chromatography
`C. Electrophoresis
`'
`0- Ultracen '
`trifugation
`
`Sulitercoiled D
`NA
`A S'Uperhelix Topology
`B-Measurements of Supercoiling
`'TOPOisomerases
`
`NA
`
`uCleic Acid Sequencing
`A-R
`BCeStrlction Endonucleases
`hemical Cleavage Method
`
`
`
`3
`
`

`

`
`
`792 Section 28-1. Chemical Structure and Base Composition
`
`2. The RNAs of ribosomes, which are about two-thirds
`RNA and one-third protein, probably have func-
`tional as well as structural roles.
`
`3. During protein synthesis, amino acids are delivered
`to the ribosome by molecules of transfer RNA
`(tRNA).
`
`4. Certain RNAs are associated with specific proteins to
`form ribonucleoproteins that participate in the
`post-transcriptional processing of other RNAs.
`
`5. In many viruses, RNA, not DNA, is the carrier of
`hereditary information.
`
`In this chapter we examine the structures of nucleic
`acids with emphasis on DNA (the structure of RNA is
`detailed in Section 30-2A), and discuss methods of puri-
`fying, sequencing, and chemically synthesizing nucleic
`acids. We end by outlining how recombinant DNA tech-
`nology, which has revolutionized the study of biochem-
`istry, is used to manipulate, synthesize, and express
`DNA.
`
`1. CHEMICAL STRUCTURE AND
`
`
`BASE COMPOSITION
`
`The chemical structures of the nucleic acids were elu-
`
`cidated by the early 19508 largely through the efforts of
`Phoebus Levine followed by those of Alexander Todd.
`Nucleic acids are, with few exceptions, linear polymers of
`nucleotides whose phosphates bridge the 3’ and 5’ positions
`of successive sugar residues (e.g., Fig. 28—1). The phos—
`phates of these polynucleotides, the phosphodiester
`groups, are acidic so that, at physiological pH ’5, nucleic
`acids are polyanions.
`
`Figure 28-1
`(a) The tetranucleotide adenyl-3’,5’-uridyl-3',5’-cytidyl-3’,5’-
`guannyI-3’-phosphate. The sugar atom numbers are primed
`to distinguish them from the atomic positions of the bases.
`By convention, polynucleotide sequences are written with
`their 5’ end at the left and their 3’ end to the right. Thus,
`reading left to right, the phosphodiester bond links
`neighboring ribose residues in the 5’ -—> 3’ direction. The
`above sequence may be abbreviated ApUpCpGp or just
`AUCGp (where a “p” to the left and/or right of a nucleoside
`symbol indicates a 5’ and/or a 3’ phosphoryl bond,
`respectively; see Table 26-1 for other symbol definitions).
`The correspending deoxytetranucleotide is abbreviated
`d(ApUpCpGp) or d(AUCGp). (b) A schematic representation
`of AUCGp. Here a vertical line denotes a ribose residue, its
`attached base is indicated by the corresponding one letter
`abbreviation and a diagonal line flanking an optional “p”
`represents a phosphodiester bond. The atomic numbering of
`the ribose residues, which is indicated here, is usually
`omitted. The equivalent representation of
`deoxypolynucleotides differ only by the essence of the
`2’-OH groups.
`
`
`
`
`
`
`
`4
`
`

`

`
`
`
`NA'S Base Composition Is Governed by
`.
`.
`Dhargaff’s Rules
`.
`DNA has equal numbers ofadenine and thymine resuiues
`, T) and equal numbers of guanine and cytosine resi-
`(A e5 (G s C). These relationships, known as Chargafi’s
`dules, were discovered in the late 19405 by Erwin Char-
`rflff who first devised reliable quantitative methods for
`51:6 Separation (by paper chromatography) and analysis
`t f DNA hydrolysates. Chargaff also found that the base
`composition of DNA from a given organism is charac-
`feristic of that organism; that is, it is independent of the
`tissue from which the DNA is taken as well as the age of
`the organism, its nutritional state or any other environ—
`mental factor. The structural basis of Chargaff’s rules
`derives from DNA’s double—stranded character (Section
`28-2A)-
`DNA'5 base composition varies widely among differ-
`ent organisms. It ranges from ~ 25 to 75% G + C in
`different species of bacteria. It is, however, more or less
`constant among related species; for example, in mam-
`mals G + C ranges from 39 to 46%.
`RNA, which usually occurs as a single—stranded mole-
`cule, has no apparent constraints on its base composi-
`tion. However, double-stranded RNA, which comprises
`the genetic material of several viruses, obeys Chargaff’s
`rules. Conversely, single-stranded DNA, which occurs
`in certain viruses, does not obey Chargaff’s rules. Upon
`entering its host organism, however, such DNA is repli—
`cated to form a double-stranded molecule, which then
`
`LJ
`
`1H
`
`obeys Chargaff’s rules.
`
`Nucleic Acid Bases May Be Modified
`Some DNAs contain bases that are chemical deriva-
`
`tives of the standard set. For example, dA and dC in the
`DNAs of many organisms are partially replaced by N‘-
`methyl-dA and 5-methyl-dC, respectively.
`
`H
`
`\ N /
`
`CH3
`
`‘\ / N
`
`N
`
`|
`dR
`
`NH
`
`2
`
`/ /CH‘3
`
`0
`
`N
`
`|
`dR
`
`Chapter 28. Nucleic Acid Structures and Manipulation 793
`
`B
`
`9-H
`
`Bn+1
`
`2'
`0—H)
`3. o\
`\P//0
`‘0/ Y0}
`
`. . 5‘
`
`RNA
`
`o-~-
`
`Bn
`
`Bn + 1
`
`OH
`
`O
`
`O
`
`O
`
`\P//
`\ _
`0 /
`O
`
`+
`
`HO
`
`2',3'-Cyclic nucleotide
`
`\QH20
`
`OH
`
`or
`
`0P0?
`
`2'-Nucleotide
`
`3'-Nucleotide
`
`Figure 28-2
`The mechanism of base-catalyzed RNA hydrolysis. The
`base-induced deprotonation of the 2’-OH group faciEtates its
`nucleOphilic attack on the adjacent phosphorus atom
`thereby cleaving the RNA backbone. The resultant 2’,3’-
`cyclic phosehate group subsequently hydrolyzes to either
`the 2' or the 3’ phosphate. Note that the RNase-catalyzed
`hydrolysis of RNA follows a nearly identical reaction
`sequence (Section 14-1 A).
`
`N s-Methyl-dA
`5-Methyl-dC
`The altered bases are generated by the sequence specific
`enzymatic: modification of normal DNA (Sections
`28‘6A and 31-7). The modified DNAs obey Chargaff’s
`Eles if the derivatized bases are taken as equivalent to
`e“ Parent bases. Likewise, many bases in RNA and, in
`Particlllar, in tRNA (Section 30-2), are derivatized.
`
`ENA but Not DNA Is Susceptible to
`ase‘Catalyzed Hydrolysis
`SisiNA is highly susceptible to base-catalyzed hydroly-
`So
`3’ the reaction mechanism diagrammed in Fig. 28—2
`as t0 yield a mixture of 2’ and 3’ nucleotides. In
`
`contrast, DNA, which lacks 2’-OH groups, is resistant to
`base catalyzed hydrolysis and is therefore much more
`chemically stable than RNA. This is probably why DNA
`rather than RNA evolved to be the cellular genetic ar—
`chive.
`
`2. DGUBLE HELICAL
`
`
`STRUCTURES
`
`The determination of the structure of DNA by James
`Watson and Francis Crick in 1953 is often said to mark
`
`
`
`5
`
`

`

`
`
`
`
`
`
`794 Section 28-2. Double Helical Structures
`
`(
`a)
`
`O
`I H
`H\N/
`OkN H
`
`l
`R
`
`H
`
`\O
`l H
`N/
`okN H
`
`I
`R
`
`Uracil
`(keto or lactam form)
`
`Uracil
`(enol or lactim form)
`
`
`
`Guanine
`(keto or lactam form)
`
`H
`
`\o
`./
`N/
`
`N\
`
`R
`
`Guanine
`(enol or lactim form)
`
`Figure 28-3
`Some possible tautomeric conversions for (a) uracil and (b)
`guanine residues. Cytosine and adenine residues can
`undergo similar proton shifts.
`
`the birth of modern molecular biology. The Watson-
`Crick structure of DNA is of such importance because,
`in addition to providing the structure of what is argu-
`ably the central molecule of life, it suggested the molec—
`ular mechanism of heredity.'Watson and Crick’s accom—
`plishment, which is ranked as one of science’s major
`intellectual achievements, tied together the less than
`universally accepted results of several diverse studies:
`
`1. Chargaff’s rules. At the time, these relationships
`were quite obscure because their significance was not
`apparent. In fact, even Chargaff did not emphasize
`them.
`
`2. The correct tautomeric forms of the bases. X-ray,
`NMR, and spectroscopic investigations have firmly
`established that the nucleic acid bases are over-
`whelmingly in the keto tautomeric forms shown in
`Fig. 28-1. In 1953, however, this was not generally
`appreciated. Indeed, guanine and uracil were widely
`believed to be in their enol forms (Fig. 28—3) because
`it was thought that the resonance stability of these
`aromatic molecules would thereby be maximized.
`Knowledge of the dominant tautomeric forms, which
`was prerequisite for the prediction of the correct hy-
`drogen bonding associations of the bases, was pro-
`vided by Jerry Donohue, an office mate of Watson
`and Crick and an expert on the X—ray structures of
`small organic molecules.
`
`3. Information that DNA is a helical molecule. This was
`provided by an X-ray diffraction photograph of a
`
`DNA fiber taken by Rosalind Franklin (Fi
`DNA, being a threadlike molecule, does nOtg. 28‘4;
`lize but, rather, can be drawn out in fibers C0 Crystal-
`of parallel bundles of molecules; Section 7-231813%
`scription of the photograph enabled Crick, an A de-
`crystallographer by training who had Earlier d Xiray
`the equations describing diffraction by heliCalenved
`cules, to deduce that DNA is (a) a helical mOITehole-
`and (b) that its planar aromatic bases form a S Cule
`parallel rings that is parallel to the fiber axis' tack of
`
`This information only provided a few crude landm
`that guided the elucidation of the DNA structurzlk's
`mostly sprang from Watson and Crick’s imaginatié 1t
`through model building studies. Once the Watso ns
`Crick model had been published, however, its basic Sill:
`plicity combined with its obvious biological releVanC~
`led to its rapid acceptance. Later investigations hav:
`confirmed the essential correctness of the Watsoni
`Crick model although its details have been modified.
`It is now realized that double helical DNA and RNA
`can assume several distinct structures that vary With
`such factors as the humidity and the identities of the
`cations present, as well as with base sequence. In this
`section, we describe these various structures.
`
`large'y
`
`.
`_ 7
`an
`if: :-
`
`+
`Figure 28-4
`An X-ray diffraction photograph of a vertically oriented Na t
`DNA fiber in the B conformation. This is the photograph ”“3
`provided key information for the elucidation of the W350"
`Crick structure. The central X-shaped pattern of spots IS
`indicative of a helix, whereas the heavy black arcs 0“ the top
`and bottom of the diffraction pattern correSpond to a
`distance of 3.4 A and indicate that the DNA structure
`repeats every 3.4 A along the fiber axis. [Courtesy Of
`Maurice Wilkins, King’s College, London.]
`
`6
`
`

`

` Table 28'1
`
`Chapter 28. Nucleic Acid Structures and Manipulation 795
`
`structural Features of Ideal A, B, and Z-DNA
`
`
`A
`Helical sense
`Right handed
`Right handed
`Left handed
`
`fer
`~26A
`~20A
`~18A
`
`Base pairs per
`11
`10
`12 (6 dimers)
`helical turn
`Helical lWiSt per
`
`base pair
`Helix pitch (use
`
`per turn)
`Helix rise per
`
`base pair
`
`Base tilt normal
`to the helix axis
`Flat
`Wide and Deep
`Narrow and deep
`Major groove
`
`Narrow and deep
`Narrow and deep
`Wide and shallow
`Minor groove
`
`C(2')-endo for pyrimidines; C(3’)—endo for purines
`C(2’)—endo
`C(3’)—endo
`Sugar pucker
`
`Anti for pyrimidines; syn for purines
`Anti
`Anti
`clycosidic bond
`
`
`
`B
`
`Z
`
`60° (per dimer)
`a
`45 A
`.
`3.7 A
`
`7°
`
`33°
`0
`28 A
`fl
`2.6 A
`
`20°
`
`36°
`a
`34 A
`.,
`3.4 A
`
`6°
`
`
`A. The Watson— Crick Structure: B-DNA
`.
`,
`Figs $2311:Aflassuriw thedig-callied B conformatilpn,
`
`as“: 0 cate . y.
`en 113?},
`trlac 0;: pagerps, v; in
`
`relagvuglenégitls $1119;0/61 glimpse asd :1 at;
`t e
`e um y is
`o.
`is regar e as t e na-
`-
`
`tive form because its X-ray pattern resembles that of the
`
`DNA in intact sperm heads.
`
`The Watson—Crick structure of B-DNA has the fol—
`
`
`lowing major features (Table 28-1):
`
`1. It consists of two polynucleotide strands that wind about
`a common axis with a right-handed twist to form an
`
`~20A in diameter double helix (Fig. 28-5). The two
`
`strands are antiparallel (run in opposite directions) and
`
`wrap around each other such that they cannot be
`
`separated without unwinding the helix (a phenome-
`
`non known as plectonemic coiling). The bases oc-
`
`Cupy the core of the helix while its sugar—phosphate
`
`Chains are coiled about its periphery thereby mini—
`
`mizing the repulsions between charged phosphate
`
`groups,
`
`
`' The planes of the bases are nearly perpendicular to
`
`the helix axis. Each base is hydrogen bonded to a base on
`
`the opposite strand to form a planar base pair (Fig.
`
`2&5). It is these hydrogen bonding interactions, a
`
`Phenomenon known as complementary base pair-
`
`111g, that result in the specific association of the two
`
`chains of the double helix.
`
`3- The ”ideal” B-DNA helix has 10 base Pairs (bp) Per
`turn (a helical twist of 36° per bp) and, since the
`aromatic bases have van der Waals thicknesses of
`3.4 A and are partiain stacked on each other (base
`stacking; Fig. g8—5b),
`the helix has a pitch (rise
`per turn) of 34 A.
`
`The most remarkable feature of the Watson—Crick
`structure is that it can accommodate only two types of base
`pairs: Each adenine residue must pair with a thymine resi—
`due and vice versa, and each guanine residue must pair with
`a cytosine residue and vice versa. The geometries of these
`AT and G - C base pairs, the so—called Watson-Crick
`base pairs, are shown in Fig. 28—6. It can be seen that
`both of these base pairs are interchangeable in that they can
`replace each other in the double helix without altering the
`positions of the sugar—phosphate backbone’s C(1’) atoms.
`Likewise, the double helix is undisturbed by exchanging the
`partners of a Watson — Crick base pair, that is, by changing a
`G-C to a CC or a A-T to a T-A. In contrast, any other
`combination of bases would significantly distort the
`double helix since the formation of a non—Watson—
`Crick base pair would require considerable reorientation
`of the sugar—phosphate chain.
`The two deep grooves that wind about the outside of
`B—DNA between the sugar— phosphate chains are of un-
`equal size (Fig. 28—5a) because: (1) the top edge of each
`base pair, as drawn in Fig. 28-6, is structurally distinct
`from the bottom edge; and (2) the deoxyribose residues
`are asymmetric. The minor groove is that in which the
`
`
`
`
`
`a
`
`«L
`
`W
`
`“of
`
`“Nit
`
`“M _
`"'
`m:
`
`(Mi:
`'
`
`-
`
`7
`
`

`

`796 Section 28-2. Double Helical Structures
`
`
`
`Figure 28-5
`The structure of B-DNA as represented by ball-and-stick
`drawings and the corresponding computer-generated
`space-félling models. The repeating helix is based on the
`X-ray structure of the self-complementary dodecamer
`d(CGCGAATTCGCG) determined by flichard Dickerson and
`Horace Drew. (a) View perpendicular to the helix axis. In the
`drawing, the sugar—phosphate backbones, which wind
`about the periphery of the molecule, are blue, and the
`bases, which occupy its core, are red. In the space-filling
`
`,
`model, C, N, O, and P atoms are white, blue, red, and .
`green, respectively. H atoms have been omitted for clarity I"
`both draw's'sgs. Note that the two sugar—phOSphate chainsl.x
`run in opposite directions. (b) (opposite) View along the “‘39.
`axis. In the drawing, the ribose ring 0 atoms are red and
`nearest base pair is white. Note that the helix axis Passes
`through the base pairs so that the helix has a solid cor9- hics
`[Drawings copyrighted © by Irving Gels. Computer grap
`courtesy of Robert Stodola, Fox Chase Cancer Center-1
`
`
`
`8
`
`

`

`
`
`Chapter 28. Nucleic Acid Structures and Manipulation 797
`
`
`
`
`
`Major groove
`
`Figure 28-5 (b)
`
`Minor groove
`
`Major groove
`
`
`
`
`1,
`
`Minor groove
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Sm
`:liX
`the
`'
`
`cs
`
`
`
`
`
`
`
`
`C(1’)-helix axis-C(l’) angle is < 180° (opening towards
`the bottom in Fig. 28-6; the helix axis passes through the
`middle of each base pair in B—DNA), whereas the majer
`groove opens towards the opposite edge of each base
`pair (Fig. 28-6).
`The Watson—Crick structure can accommodate any
`sequence of bases on one polynucleotide strand if the
`opposite strand has the complementary base sequence.
`This immediately accounts for Chargaff’s rules. More
`importantly, it suggests that hereditary information is en-
`coded in the sequence of bases on either strand.
`
`Real DNA Deviates from the Ideal
`Watson - Crick Structure
`
`a By the late 19705, advances in nucleic acid chemistry
`permitted the synthesis and crystallization of ever
`longer oligonucleotides of defined sequences (Section
`28-7). Consequently, some 25 years after the Watson—
`Crick structure had been formulated. the X-ray crystal
`Structures of DNA fragments wereclearly visualized for
`the first time (fiber diffraction studies provide only
`Crude low resolution images in which the base pair elec—
`
`Figure 28-6
`The Watson—Crick base pairs. The line joining the C(1’)
`at0ms is the same length in both base pairs and makes equal
`anQles with the glycosidic bonds to the bases. This gives
`DNA a series of pseudo-twofold symmetry axes (often
`referred to as dyad axes) that pass through the center of
`839*! base pair (red line) and are perpendicular to the helix
`:X'S- Note that A - T base pairs associate via two hydrogen
`hondS, whereas C-G base pairs are joined by three
`ydrogen bonds. [After Arnott, 3., Dover, s. D., and
`Onacott, A. J., Acta Cryst. B25, 2196 (1969).]
`
`
`
`9
`
`

`

`798 Section 28-2. Double Helical Structures
`
`tron density is the average electron density of all the
`base pairs in the fiber). Richard Dickerson and Horace
`Drew have shown that the self-complementary dodeca-
`mer d(CGCGAATTCGCG) crystallizes in the B—confor-
`mation. The molecule has an average rise per residue of
`3.4 A and has 10.}. bp per turn (a helical
`twist of
`35.6°per bp), which is nearly equal to that of ideal
`B-DNA. Nevertheless, individual residues significantly
`depart from this average conformation in a manner that
`appears to be sequence dependent (Fig. 28-5). For example,
`the helical twist per base pair in this dodecamer ranges
`from 28 to 42°. Each base pair further deviates from its
`ideal conformation by such distora'ons as propeller
`twisting (the opposite rotation of paired bases about the
`base pair’s long axis; in the above dodecamer these
`values range from 10 to 20°) and base pair roll (the
`tilting of a base pair as a whole about its long axis).
`Indeed, rapidly accumulating X-ray and NMR studies of
`other double helical DNA oligomers have amply dem—
`onstrated that the structure of DNA is surprisingly irregu-
`lar in a sequence-specific manner. This phenomenon, as we
`shall see (Sections 29-3C and E) is important for the se-
`quence-specific binding to DNA of proteins that process
`genetic information.
`
`DNA Is Semiconservatively Replicated
`The Watson— Crick structure also suggests how DNA
`can direct ig own replication. Each polynucleotide
`strand can act as a template for the formation of its
`complementary strand through base pairing interac-
`tions. The two strands of the parent molecule must
`therefore separate so that a complementary daughter
`strand may be enzymatically synthesized on the surface
`of each parent strand. This results in two molecules of
`duplex (double stranded) DNA, each consisting of one
`polynucleotide strand from the parent molecule and a
`newly synthesized complementary strand (Fig. 1-16).
`Such a mode of replication is termed semiconservative
`in contrast with conservative replication which, if it
`occurred, would result in a newly synthesized duplex
`copy of the original DNA molecule with the parent
`DNA molecule remaining intact. The mechanism of
`DNA replication is the main subject of Chapter 31.
`The semiconservative nature of DNA replication was
`elegantly demonstrated in 1958 by Matthew Meselson
`and Franklin Stahl. The density of DNA was increased
`by labeling it with 15N, a heavy isotope of nitrogen (“N
`is the naturally abundant isotope). This was accom-
`plished by growing E. coli for 14 generations in a me-
`dium that contained 15NH4C1 as its only nitrogen source.
`The labeled bacteria were then abruptly transferred to
`an 1‘N-containing medium and the density of their
`DNA was monitored as a function of bacterial growth
`by equilibrium density gradient ultracentrifugaa'on
`(Section 5-53; a technique Meselson, Stahl, and Jerome
`Vinograd had developed for the purpose of distinguish-
`ing 15N-labeled DNA from unlabeled DNA).
`
`The results of the Meselson—Stahl experiment
`displayed in Fig. 28-7. After one generation (dbllblin are
`the cell population), all of the DNA had a density eXa30f
`halfway between the densities of fully 15N—labe1ed 13le
`and unlabeled {)NA. This DNA must therefore Conta'A
`equal amounts of 14N and 15N as is expected after 0 In
`generation of semiconservative replication. Congewne
`tive DNA replication, in contrast, would result in» tE.
`preservation of the parental DNA, so that it maintaines
`its original density, and the generation of an equal
`amount of unlabeled DNA. After two generatiOns, One
`half of the DNA molecules were unlabeled and the re-
`mainder were 1“N—“N hybrids. This is also in accord
`with the predictions of the semiconservative replication
`model and in disagreement with the conservative repli-
`cation model. In succeeding generations, the amount of
`unlabeled DNA increased relative to the amount of by.
`brid DNA although the hybrid never totally disap-
`peared. This is again in harmony with semiconservatiVe
`replication but at odds with conservative replication,
`which predicts that the fully labeied parental DNA Will
`always be present and that hybrid DNA never forms,
`Meselson and Stahl also demonstrated that DNA is
`double stranded. DNA from 15N-labeled E. coli that
`. were grown for one generation in an 14N medium was
`heat denatured at 100°C (which causes strand separa-
`tion; Section 28—3A) and then subjected to density gra-
`dient ultracentrifugation. Two bands were observed;
`one at the density of fully 15N-labeled DNA and the
`other at the density of unlabeled DNA. Moreover the
`molecular masses of the DNA in these bands, as esti-
`
`mated from their peak shapes, was one half that of
`undenatured DNA (the peak width varies with molecu-
`lar mass). Native DNA must therefore be composed of
`two equal-sized strands that separate upon heat dena-
`turation.
`
`B. Other Nucleic Acid Helices
`
`Double-stranded DNA is a conformationally variahle
`molecule. In the following subsections we discuss Its
`major conformational states besides B-DNA and also
`those of double—stranded RNA.
`
`A-DNA’s Base Pairs Are Inclined to the Helix AXiOS
`When the relative humidity is reduced to 75 /°’
`B—DNA undergoes a reversible conformational Change
`to the so-called A form. Fiber X—ray studies indiCa.te that
`A-DNA forms a wider and flatter right—handed helix than .
`does B—DNA (Fig. 28-8; Table 28-1). A-DNA has 31 h?
`per turn and a pitch of 28 A which gives A-DNA at;
`axial hole (Fig. 28-817). The most striking featlfre 0e
`A-DNA, however, is that the planes of its base Palrs a;
`tilted 20° with respect to the helix axis. A-DNA theta
`fore has a deep major groove and a very shallOW mm d
`8Toove; it can be described as a flat ribeIl Woun1f“
`around a 6 A in diameter cylindrical hole. MO5t 5e
`
`
`
`
`LMMIIM.”
`
`l
`
`J .
`
`i in §.|:.ll.l”;‘jl tilt”
`I
`
`Wm
`|
`
`.
`Plkl'hl*,m.
`i
`
`
`
`10
`
`

`

`Chapter 28. Nucleic Acid Structures and Manipulation 799
`
`——+
`
`15N DNA (heavy)
`
`‘L
`
`2;.
`
`0.3
`
`0.7
`
`1.0
`
`——>
`
`‘L
`
`i
`+
`
`% é
`& &y
`
`HbrIdDNA
`
`1.1
`
`1.5‘
`
`1.9
`
`2.5
`
`3.0
`
`4.1
`
`9+?4?%Z?Z?g 14NDNA(light)
`i
`
`“L
`‘—F 3
`
`‘L
`1
`
`k
`
`-
`
`in?
`a.
`i
`at+at
`7%.
`7%.
`
`
`
`
`
`
`
`
`2 \x
`14N
`DNA
`
`_
`Hybrld
`DNA
`
`15N
`DNA
`
` .
`
`
`
`
`
`Figure 28-7
`r e_der_nonstration of the semiconservative nature of DNA
`eFJIICatIon in E. coli. DNA in a CsCl solution of density 1.71
`9 ' cm~a
`was subjected to equilibrium density gradient
`lJltracen
`trifugation at 140,000 g in an analytical
`Ultrt-Jcen
`trifuge (a device in which the spinning sample can be
`C’Dticau
`y observed). The enormous centrifugal acceleration
`ca“186d the CsCl to form a density gradient in which DNA
`migrated to its. position of buoyant density. The left panels
`are UV
`absorption photographs of ultracentrifuge cells (DNA
`StFOngl
`Y absorbs UV light) and are arranged such that
`re9km
`3 Of equal density have the same horizontal positions.
`The
`r middle panels are microdensitometer traces of the
`res[Bonding photographs in which the vertical
`
`displacement is proportional to the DNA concentration. The
`buoyant density of DNA increases with its 15N content. The
`bands furthest to the right (greatest radius and density) arise
`from DNA that is fully 15N labeled, whereas unlabeled DNA,
`which is 0.014 g - cm‘3 less dense, forms the leftmost
`bands. The bands in the intermediate position result from
`duplex DNA in which one strand is “N labeled and the other
`strand is unlaEeled. The accompanying interpretive drawings
`(right) indicate the relative numbers of DNA strands at each
`generation donated by the original parents (blue, 15N labeled)
`and synthesized by succeeding generations (red, unlabeled).
`[From Meselson, M. and Stahl, F. W., Proc. Natl. Acad. Sci.
`44, 674 (1958).]
`
`Density ———> Density —> Generations
`
`
`
`
`
`\
`15N
`DNA
`
`“N
`DNA
`
`Hybrid
`DNA
`
`11
`
`

`

`800 Section 28—2. Double Helical Structures
`
`
`
`”WWW“
`
`Milli “h”
`
`I.<I'}i}uii Ml
`
`Figure 28-8
`Ball-and-stick drawings and the corresponding space—filling
`models of A-DNA as viewed (a) perpendicular to the helix
`axis, and (b) (opposite) along the helix axis. The color codes
`are given in Fig. 28-5. The repeating helix was generated by
`Richard Dickerson based on the X-ray structure of the
`self-complementary octamer d(GGTATACC) determined by
`
`Olga Kennard, Dov Rabinovitch, Zippora Shakked, and -
`Mysore Viswamitra. Note that the base pairs are inclined to
`Compare
`the helix axis and that the helix has a hollow core.
`Irving
`this figure with Fig. 28-5. [Drawings copyrighted © by
`Fox
`Geis. Computer graehics courtesy of Robert Stodolai
`Chase Cancer Center.]
`
`12
`
`
`
`12
`
`

`

` Figure 28.8 (b)
`
`complementary oligonucleotides of < 10 base pairs; for
`example, d(GGCCGGCC) and d(GGTATACC), crys—
`tallize in the A-DNA conformation. Like B—DNA, these
`molecules exhibit considerable sequence-specific con—
`formational variation. It has not been established that
`
`A-DNA exists in vivo although a few experimental ob—
`servations suggest that certain DNA segments normally
`assume the A conformation.
`
`Z-DNA Forms a Left-Handed Helix
`
`Occasionally, a seemingly well understood or at least
`familiar system exhibits quite unexpected properties.
`Over 25 years after the discovery of the Watson—Crick
`Structure,
`the
`crystal
`structure determination of
`C1(CGCGCG) by Andrew Wang and Alexander Rich re—
`Vealed, quite surprisingly, a left-handed double helix (Pig.
`28-9; Table 28-1). A similar helix is
`formed by
`élCGCATGCG). This helix, which has been dubbed
`Z‘DNA, has 12 Watson—Crick base pairs per turn, a pitch
`if45A and, in contrast to A-DNA, a deep minorgroove and
`a:dZSCernable major groove. Z-DNA therefore resembles
`Z‘gf’hl'landed drill bit in appearance. The base pairs in
`38—1NA are flipped 380 ° relative to those in B-DNA (Fig.
`SectiO) through conformational changes discussed in
`Z‘D13111 28—33. As a consequence, the repeating unit of
`and A18 a dinucleotide, d(Xpr), rather than a single
`‘tSuaeiftlde as it is in the other DNA helices. Here, X is
`resid y a pyrimidine residue/and Y is usually a purine
`£0rmue- because the purine nucleotide assumes a con-
`pyfinigén that would be sterically unfavorable in the
`9hos ldlne nucleotide. The line ioining successive
`Phate groups on a polynucleotide strand of Z-DNA
`th
`erefOre follows a zi za
`ath aro nd the helix (Fi
`.
`g g P
`8
`
`Chapter 28. Nucleic Acid Structures and Manipulation 801
`
`
`
`28—9a; hence the name Z-DNA) rather than a smooth
`curve as it does in A- and B—DNAs (Figs. 28—5a and
`28-8a).
`Fiber diffraction and NM? studies have shown that
`
`complementary polynucleotides with alternating pu-
`rines and pyrimidines, such as poly d(GC) - poly d(GC)
`or poly d(AC) - poly d(GT), take up the Z—DNA confor-
`mation at high salt concentrations. Evidently,
`the
`Z—DNA conformation is most readily assumed by DNA seg-
`ments with alternating purine— pyrimidine base sequences
`(for structural reasons explained in Section 28-38). A high
`salt concentration stabilizes Z-DNA relative to B-DNA
`
`by reducing the otherwise increased electrostatic repul-
`sions between closest approaching phosphate groups
`on opposite strands (8 A in Z-DNA vs 12 Ain B—DNA).
`The methylation of cytosine residues at C(5), a common
`biological modification (Section 31-7), also promotes
`Z-DNA formation since a hydrophobic methyl group in
`this position is less exposed to solvent in Z-DNA than it
`is in B-DNA.
`
`Does Z-DNA have any biological significance? Rich
`has proposed that the reversible conversion of specific
`segments of B—DNA to Z-DNA under appropriate cir-
`cumstances acts as a kind of switch in regulating genetic
`expression. Yet, the in vivo existence of Z-DNA has been
`difficult to prove. A major problem is demonstrating
`that a particular probe for detecting Z—DNA, a Z—DNA-
`specific antibody, for example, does not in itself cause
`what would otherwise be B-DNA to assume the Z
`
`conformation—a kind of biological uncertainty princi—
`ple (the act of measurement inevitably disturbs the sys-
`tem being measured). Recently, however, Z-DNA has
`been shown to be present in E. coli by employing an
`E. coli enzyme that methylates a specific base sequence
`
`
`
`13
`
`

`

`802 Section 28-2. Double Helical Structures
`
`Figure 28-9
`Ball-and-stick drawings ane the corresponding space-filling
`models of Z-DNA as viewed (a) perpenécular to the helix
`axis and (b) (opposite) along the helix axis. The color codes
`are given in Fig. 28-5. The repeating helix was generated by
`Richard Dickerson based on the X-ray structure of the
`self-complementary hexamer d(CG
`G) determined by
`
`14
`
` .
`
`eft
`
`'
`.
`Andrew Wang and Alexander Rich. Note that the helix '5'
`handed and that the sugar—phosphate chains follow iradli
`zigzag course (alternate ribose residues lie at differen
`in Part D) indicating that the Z-DNA’s reaeating m0tif i358,
`dinucleotide. Compare this figure with Figs. 28-5 and raphics
`[Drawings copyrighted © by Irving Geis. Compu'fer 9
`courtesy of Robert Stodola, Fox Chase Cancer center-l
`
`14
`
`

`

`
`
`Chapter 28. Nucleic Acid Structures and Manipulation 803
`
`
`
`Figure 28-9 (b)
`
`in vitro when the DNA is in the B form but not when it is
`
`in the Z form. The in viva methylation of this base se-
`quence is inhibited when it is cloned in E. coli (by