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

`

`To:
`Ourparents, who encouraged us,
`Our teachers, who enabled us, and
`Ourchildren, who put up with us.
`
`Cover Art: Oneofa series of color studies of horse heart
`cytochromec designed to show theinfluence of amino
`acid side chains onthe protein’s three-dimensionalfold-
`ing pattern. We haveselected 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 openingillustrations
`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 & Sons, Inc.
`
`All rights reserved. Published simultaneously in Canada.
`
`Reproductionor 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
`owneris 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.
`Pp.
`cm.
`Includes bibliographical references.
`ISBN 0-471-61769-5
`
`1. Biochemistry. I. Voet, Judith G. II. Title.
`QP514.2.V64
`1990
`574.192 —dc20
`
`Printed in the United States of America
`
`10987654321
`
`89-16727
`CIP
`
`
`
`2
`
`

`

`
`
`| C
`
`hapter 28
`NUCLEIC A¢
`
`S$ AND
`
`
`ON
`
`. Chemica! Structure and Base Composition
`
`. Double Helical Structures
`
`C. Chain-Terminator Method
`
`D. RNA Sequencing
`
`A. The Watson-—Crick Structure: B-DNA
`
`7. Chemical Synthesis of Oligonucleotides
`
`
`
`B. Other Nucleic Acid Helices
`
`C. The Size of DNA
`
`. Forces Stabilizing Nucleic Acid Structures
`A. Denaturation and Renaturation
`B. Sugar-Phosphate Chain Conformations
`C. Base Pairing
`D. Base Stacking and Hydrophobic Interactions
`E.
`lonic Interactions
`
`8. Molecular Cloning
`. Cloning Vectors
`. Gene Splicing
`. Genomic Libraries
`
`nm™moOWDYP
`
`. DNA Amplification by the Polymerase Chain Reaction
`. Production of Proteins
`
`. Social Considerations
`
`n Color
`: Seno.
`dren of
`8enetic
`te with
`ness,
`
`of bac-
`
`lication
`“Tecipi-
`rder of
`ight, in
`tic map
`
`me
`
`ao
`
`- Nucleic Acid Fractionation
`A. Solution Methods
`B. Chromatography
`C, Electrophoresis
`D. Ultracentrifugation
` Supercoiled DNA
`“. Superhelix Topology
`- Measurements of Supercoiling
`: Topoisomerases
`Nucte:
`ileic Acid Sequencing
`* Restriction Endonucleases
`* Chemical Cleavage Method
`
`There are two classes of nucleic acids, deoxyribonucleic
`acid (DNA) and ribonucleic acid (RNA). DNA is the
`hereditary molecule in all cellularlife forms, as well as in
`many viruses. It has but two functions:
`
`1. To direct its own replication duringcell division.
`
`2. To direct the transcription of complementary mole-
`cules of RNA.
`
`RNA,in contrast, has more varied biological functions:
`
`1. The RNAtranscripts of DNA sequences that specify
`polypeptides, messenger RNA (mRNA), direct the
`ribosomal synthesis of these polypeptides in a pro-
`cess knownastranslation.
`
`
`
`3
`
`

`

`(a)
`
`NH,
`
`ey a
`le aeN
`
`N
`
`5
`5'end—*> HOCH, 9
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`792 Section 28-1. Chemical Structure and Base Composition
`
`2. The RNAsof ribosomes, which are about two-thirds
`RNAand one-third protein, probably have func-
`tional as well as structuralroles.
`
`3. During protein synthesis, amino acids are delivered
`to the ribosome by molecules of transfer RNA
`(tRNA).
`
`4. Certain RNAsareassociated with specific proteins to
`form ribonucleoproteins that participate in the
`post-transcriptional processing of other RNAs.
`gi. In many viruses, RNA, not DNA,is the carrier of
`hereditary information.
`
`In this chapter we examinethe structures of nucleic
`acids with emphasis on DNA(the structure of RNAis
`detailed in Section 30-2A), and discuss methodsof puri-
`fying, sequencing, and chemically synthesizing nucleic
`acids. We end byoutlining how recombinant DNAtech-
`nology, whichhasrevolutionized 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 1950s 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’s, nucleic
`acids are polyanions.
`
`
`
`. |
`
`i) sad
`
`ne
`
`Figure 28-1
`(a) The tetranucleotide adenyl-3',5’-uridyl-3’,5‘-cytidyl-3',5’-
`guanylyl-3’-phosphate. The sugar atom numbersare primed
`to distinguish them from the atomic positions of the bases.
`By convention, polynucleotide sequencesare written with
`their 5’ end at the left and their 3’ end to the right. Thus,
`readingleft to right, the phosphodiester bond links
`neighboring ribose residuesin the 5‘ — 3’ direction. The
`above sequence may be abbreviated ApDUpCpGporjust
`AUCGp(wherea ‘‘p” to theleft and/orright of a nucleoside
`symbolindicates a 5’ and/or a 3’ phosphoryl bond,
`respectively; see Table 26-1 for other symbol definitions).
`The correspending deoxytetranucieotide is abbreviated
`d(ApUpCpGp)or d(AUCGp). (b) A schematic representation
`of AUCGp. Here a vertical line denotes a ribose residue, its
`attached baseis indicated by the corresponding oneletter
`abbreviation and a diagonalline flanking an optional ‘‘p’’
`represents a phosphodiester bond. The atomic numbering of
`the ribose residues, whichis indicated here, is usually
`omitted. The equivalent representation of
`deoxypolynucleotides differ only by the absence of the
`2’-OH groups.
`
`
`
`
`
`
`
`
`
`
`
`(6)
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`4
`
`

`

`Chapter 28. Nucleic Acid Structures and Manipulation 793
`oe
`
`nr
`
`B
`
`oH
`5
`OH
`
`, o\
`NV
`0}
`
`oD
`
`Brat
`
`-- OH
`
`oe
`
`RNA
`
`B, +1
`
`OH
`
`Or
`
`x
`
`HO
`
`B,
`
`oO
`
`0
`
`0
`
`a\
`
`oO c
`
`2',3'-Cyclic nucleotide
`
`wy
`
`B,
`
`eC
`
`B, +1
`
`O—POF
`
`OH
`
`or
`
`OH
`
`OPO;
`
`2'-Nucleotide
`
`3'-Nucleotide
`
`Figure 28-2
`The mechanism of base-catalyzed RNA hydrolysis. The
`base-induced deprotonation of the 2’-OH groupfacilitates its
`nucleophilic attack on the adjacent phosphorus atom
`thereby cleaving the RNA backbone. The resultant 2',3’-
`cyclic phosphate 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 14-1A).
`
`contrast, DNA,which lacks 2’-OHgroups,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 thecellular genetic ar-
`chive.
`
`2. DOUBLE HELICAL
`
`STRUCTURES
`
`
`The determination of the structure of DNA by James
`Watsonand Francis Crick in 1953 is often said to mark
`
`
`
`
`'g Base Composition Is Governed by
`3
`chargaff’s Rules
`DNA has equal numbers ofadenine and thymineresidues
`21) and equal numbers of guanine and cytosine resi-
`(A os (G = C). These relationships, known as Chargaff’s
`er were discoveredin the late 1940s by Erwin Char-
`OA whofirst devised reliable quantitative methods for
`fie separation (by paper chromatography)and analysis
`: fDNA hydrolysates. Chargaff also foundthat the base
`°mposition of DNAfrom a given organism is charac-
`reristic of that organism;thatis, it is independentof the
`yssue from which the DNAis takenas well as the age of
`she 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’s base composition varies widely amongdiffer-
`ent organisms. It ranges from ~25 to 75% G+C in
`different species of bacteria. It is, however, moreorless
`constant amongrelated species; for example, in mam-
`mals G + C ranges from 39 to 46%.
`RNA,whichusually occursas 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 DNAisrepli-
`cated to form a double-stranded molecule, which then
`obeys Chargaff’s rules.
`
`Nucleic Acid Bases May Be Modified
`Some DNAscontain bases that are chemical deriva-
`tives of the standard set. For example, dA and dC in the
`DNAsof manyorganismsare partially replaced by N®-
`methyl-dA and 5-methyl-dC,respectively.
`
`HL 7s
`
`NH,
`
`AN Ais
`
`HH
`
`Ul YN
`
`N
`
`O°
`
`SN
`
`|
`|
`dR
`dR
`N &Methyl-dA
`5-Methyl-dC
`The altered bases are generated by the sequencespecific
`‘Nzymatic modification of normal DNA (Sections
`8-64 and 31-7). The modified DNAs obey Chargaff’s
`rules if the derivatized bases are taken as equivalent to
`“it parent bases. Likewise, many bases in RNA and,in
`Particular, in tRNA (Section 30-2), are derivatized.
`RNA but Not DNAIs Susceptible to
`‘Se-Catalyzed Hydrolysis
`Sis } Ais highly susceptible to base-catalyzed hydroly-
`y the reaction mechanism diagrammedin Fig. 28-2
`a8 to yield a mixture of 2’ and 3’ nucleotides. In
`
`
`
`5
`
`

`

`iargely
`
`DNA fiber taken by Rosalind Franklin Fi
`8-4.
`DNA,being a threadlike molecule, does not :
`lize but, rather, can be drawn outin fibers co “TYsta}.
`of parallel bundles of molecules; Section 7-0)8 ing
`scription of the photograph enabled Crick, an A de.
`crystallographer by training whohadearlier q, *-
`the equations describing diffraction by helical..’®
`cules, to deduce that DNAis (a) a helical mole
`and(b)that its planar aromatic bases form a stack«0
`parallel rings thatis parallel to the fiber axis.
`This information only provided a few crude landm,
`arks
`that guided the elucidation of the DNAstructure.;
`mostly sprang from Watson and Crick’s imaginat: it
`through model building studies. Once the Wats .
`Crick model had been published, however,itsbasic sim
`plicity combined with its obvious biological relevance
`led to its rapid acceptance. Later investigations have
`confirmed the essential correctness of the Watson—
`Crick model althoughits details have been modified.
`It is now realized that double helical DNA and RNA
`can assumeseveral 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.
`
`ee
`
`ti
`
`ih
`
`«3
`
`Figure 28-4
`7
`An X-ray diffraction photograph ofa vertically oriented Na :
`DNAfiberin the B conformation. This is the photograph th@
`provided key information for the elucidation of the Watson
`Crick structure. The central X-shaped pattern of spots |S
`indicative of a helix, whereas the heavy black arcs on the toP
`and bottom ofthe diffraction pattern correspond to 4
`distance of 3.4 A and indicate that the DNA structure
`repeats every 3.4 A alongthefiber axis. [Courtesy of
`Maurice Wilkins, King’s College, London.]
`
`794 Section 28-2. Double Helical Structures
`
`(a)
`°
`
`0
`ANN | H
`O°
`“N~
`~H
`
`R
`
`
`
`H
`
`~o
`nZ ) H
`a H
`
`|
`R
`
`Uracil
`(keto or lactam form)
`
`Uracil
`(enol or lactim form)
`
`\ R
`
` A
`
`~o
`Ps
`N~
`
`Guanine
`(keto or lactam form)
`
`Guanine
`(enol or lactim form)
`
`babi
`
`siahit
`Wik
`
`
`BS| |
`
`
`
`Figure 28-3
`Somepossible tautomeric conversions for (a) uracil and (6)
`guanineresidues. Cytosine and adenine residues can
`undergosimilar proton shifts.
`
`the birth of modern molecular biology. The Watson -
`Crick structure of DNAis of such importance because,
`in addition to providing the structure of whatis argu-
`ably the central moleculeof 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
`werequite obscure becausetheir significance wasnot
`apparent. In fact, even Chargaff did not emphasize
`them.
`
`2. The correct tautomeric forms of the bases. X-ray,
`NMR,andspectroscopic investigations have firmly
`established that the nucleic acid bases are over-
`whelmingly in the keto tautomeric forms shownin
`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 thoughtthat the resonancestability of these
`aromatic molecules would thereby be maximized.
`Knowledge of the dominanttautomeric forms, which
`wasprerequisite for the prediction of the correct hy-
`drogen bondingassociations 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 DNAis a helical molecule. This was
`provided by an X-ray diffraction photograph of a
`
`6
`
`

`

`
`
`Chapter 28. Nucleic Acid Structures and Manipulation 795
`
`gable 28-1
`
`gtructural Features of Ideal A, B, and Z-DNA
`
`A
`B
`Zz
`
`
`
`Helical sense
`Righthanded
`Righthanded
`Left handed
`piametet
`~26A
`~20A
`~18A
`pase pairs per
`11
`10
`12 (6 dimers)
`
`helical turn
`Helical twist per
`33°
`36°
`60° (per dimer)
`
`
`base pair
`:
`
`Helix pitch (rise
`28 A
`34A
`45A
`
`per turn)
`.
`Helix rise per
`2.6A
`3.4A
`3.7A
`
`
`base pair
`
`pasetilt 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’)-endofor purines
`C(2’)-endo
`C(3’)-endo
`Sugar pucker
`
`
`
`
`Anti AntiGlycosidic bond Anti for pyrimidines; syn for purines
`
`
`20°
`
`6°
`
`7°
`
`
`
`Marg
`
`(Aig
`
`{tt ig
`
`(th aq:
`
`(emi
`
`3. The “ideal” B-DNAhelix has 10 base pairs (bp) per
`A. The Watson-Crick Structure: B-DNA
`turn (a helical twist of 36° per bp) and, since the
`aromatic bases have van der Waals thicknesses of
`Fibers of DNA assumethe so-called B conformation,
`
`as indicated by their X-ray diffraction patterns, when
`3.4 A andare partially stacked on each other (base
`
`the counterion is an alkali metal such as Na* and the
`stacking; Fig. 28-5b),
`the helix has a pitch (rise
`
`relative humidity is 92%. B-DNAis regarded as the na-
`per turn) of 34 A.
`
`tive form because its X-ray pattern resembles that of the
`The most remarkable feature of the Watson—Crick
`
`DNAin intact sperm heads.
`
`structureis that it can accommodate only twotypesof base
`The Watson—Crick structure of B-DNA hasthe fol-
`
`pairs: Each adenine residue must pair with a thymine resi-
`lowing major features (Table 28-1):
`
`due and vice versa, and each guanineresidue mustpair with
`a cytosine residue and vice versa. The geometries of these
`1. It consists of two polynucleotide strands that wind about
`A-T and G:C basepairs, the so-called Watson - Crick
`a common axis with a right-handed twist to form an
`base pairs, are shownin Fig. 28-6. It can be seen that
`
`~20A in diameter double helix (Fig. 28-5). The two
`both of these base pairs are interchangeablein that they can
`strands are antiparallel (run in opposite directions) and
`replace each other in the double helix without altering the
`
`wrap around each other such that they cannot be
`positions of the sugar- phosphate backbone’s C(1’) atoms.
`
`separated without unwindingthe helix (a phenome-
`Likewise, the double helix is undisturbed by exchanging the
`
`non knownas plectonemic coiling). The bases oc-
`partners of a Watson —Crick basepair, that is, by changing a
`
`cupythe core of the helix while its sugar— phosphate
`G:CtoaC-GoraA:T toaT-A. In contrast, any other
`
`chainsare coiled about its periphery thereby mini-
`combination of bases would significantly distort the
`
`mizing the repulsions between charged phosphate
`double helix since the formation of a non-Watson—
`
`Crick basepair would require considerable reorientation
`Stoups.
`
`of the sugar—phosphate chain.
`.
`
`- The planes of the bases are nearly perpendicular to
`
`The two deepgroovesthat wind aboutthe outside of
`the helix axis. Each base is hydrogen bondedto a base on
`
`B-DNAbetweenthe sugar- phosphate chainsareof un-
`the opposite strand to form a planar base pair (Fig.
`
`equalsize (Fig. 28-5a) because:(1) the top edge of each
`28-5). It is these hydrogen bonding interactions, a
`
`base pair, as drawnin Fig. 28-6,is structurally distinct
`Phenomenon known as complementary basepair-
`
`from the bottom edge; and (2) the deoxyribose residues
`Ing, that result in the specific association of the two
`
`are asymmetric. The minor grooveis that in which the
`Chains of the double helix.
`
`
`
`
`
`7
`
`

`

`796 Section 28-2. Double Helical Structures
`
`oeh
`
`ahny
`
`iP
`
`ahanag
`
`courtesy of Robert Stodola, Fox Chase Cancer Center‘J
`
`
`Figure 28-5
`The structure of B-DNA as represented byball-and-stick
`drawings and the corresponding computer-generated
`space-filling models. Thé repeating helix is based on the
`X-ray structure of the self-complementary dodecamer
`d(CGCGAATTCGCG) determined by Richard Dickerson and
`Horace Drew.(a) View perpendicular to the helix axis. In the
`drawing, the sugar-— phosphate backbones, which wind
`aboutthe periphery of the molecule, are blue, and the
`bases, which occupyits core, are red. In the space-filling
`
`model, C, N, O, and P atomsare white, blue, red, and.
`green, respectively. H atoms have been omitted for clarity in
`both drawings. Notethat the two sugar-phosphate chains..
`run in opposite directions. (b) (opposite) View along the ee
`axis. In the drawing,the ribose ring O atomsare red and
`nearest basepair is white. Note that the helix axis passes
`through the basepairs so that the helix has a solid core- hics
`[Drawings copyrighted © byIrving Geis. Computer grap
`
`8
`
`

`

`
`
`Figure 28-5 (b)
`
`C(1’)-helix axis-C(1’) 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 informationis en-
`coded in the sequenceof bases on either strand.
`
`Real DNA Deviates from the Ideal
`Watson - Crick Structure
`. By the late 1970s, advancesin 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 whichthe basepair elec-
`
`Minor groove
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Chapter 28. Nucleic Acid Structures and Manipulation 797
` Major groove
`Major groove
`
`
`
`Figure 28-6
`The Watson-Crick base pairs. The line joining the C(1’)
`atoms is the same length in both base pairs and makes equal
`&ngles with the glycosidic bonds to the bases. This gives
`NAaseries of pseudo-twofold symmetry axes (often
`rb
`referred to as dyad axes) that pass throughthe centerof
`cs
`Minor groove
`€ach basepair (red line) and are perpendicular to the helix
`axis. Note that A:T base pairs associate via two hydrogen
`‘Onds, whereas C-G base pairs are joined by three
`drogen bonds. [After Arnott, S., 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 ofall the
`basepairs in thefiber). Richard Dickerson and Horace
`Drew haveshownthatthe self-complementary dodeca-
`mer d(CGCGAATTCGCG)crystallizes in the B-confor-
`mation. The molecule has an averagerise per residue of
`3.4A and has 10.1 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 distortions as propeller
`twisting (the opposite rotation of paired bases aboutthe
`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 aboutits long axis).
`Indeed, rapidly accumulating X-ray and NMRstudies of
`other double helical DNA oligomers have amply dem-
`onstrated thatthe structure of DNAis surprisingly irregu-
`lar in a sequence-specific manner. This phenomenon, as we
`shall see (Sections 29-3C and E) is importantfor the se-
`quence-specific binding to DNA of proteins that process
`genetic information.
`
`The results of the Meselson-Stahl experimen;
`displayed in Fig. 28-7. After one generation (doubjin ate
`thecell population),all of the DNAhad a density exact]
`halfway betweenthedensities offully “N-labeled yy y
`and unlabeled DNA. This DNA must therefore contay.
`equal amounts of '*N and '°Nasis expected after n
`generation of semiconservative replication. Consery,.
`tive DNAreplication, in contrast, would resultjne
`preservationof the parental DNA,sothatit maintained
`its original density, and the generation of an e ual
`amountof unlabeled DNA.After two generations, one
`half of the DNA molecules were unlabeled andthe ye.
`mainder were 4N—°N hybrids. Thisis also in accord
`with the predictions of the semiconservative replication
`modeland in disagreementwith the conservative repli-
`cation model. In succeeding generations, the amountof
`unlabeled DNAincreasedrelative to the amountof hy.
`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 labeled parental DNA will
`always be present and that hybrid DNA neverforms,
`Meselson and Stahl also demonstrated that DNAis
`double stranded. DNA from 'N-labeled E. coli that
`_were grownfor one generation in an “N 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 *N-labeled DNA and the
`other at the density of unlabeled DNA. Moreoverthe
`molecular masses of the DNA in these bands,asesti-
`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 variable
`molecule. In the following subsections we discuss its
`major conformational states besides B-DNA and also
`those of double-stranded RNA.
`
`im4| Mb
`
`/
`
`|
`ati pawlJaki
`
`|
`
`DNAIs Semiconservatively Replicated
`The Watson - Crick structure also suggests how DNA
`can direct
`its own replication. Each polynucleotide
`strand can act as a template for the formation ofits
`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 parentstrand. 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 modeofreplication 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
`DNAreplication is the main subject of Chapter 31.
`The semiconservative nature of DNA replication was
`A-DNA’s Base Pairs Are Inclined to the Helix Axis
`elegantly demonstrated in 1958 by Matthew Meselson
`When the relative humidity is reduced to 75°01
`and Franklin Stahl. The density of DNA wasincreased
`B-DNA undergoesa reversible conformational change
`by labeling it with 45N, a heavyisotope ofnitrogen (4N
`to the so-called A form.Fiber X-ray studies indicate that
`is the naturally abundant isotope). This was accom-
`A-DNAforms a widerand flatter right-handed helix tha
`plished by growing E. coli for 14 generations in a me-
`does B-DNA(Fig. 28-8; Table 28-1), A-DNA has11 PP
`dium that contained *NH,Clasits only nitrogen source.
`per turn andapitch of 28 A which gives A-DNA oe
`The labeled bacteria were then abruptly transferred to
`axial hole (Fig. 28-8b). The most striking feature -
`an 14N-containing medium and the density of their
`A-DNA,however,is that the planesofits base pa" *
`DNAwas monitored as a function of bacterial growth
`tilted 20° with respect to the helix axis. A-DNA nie
`by equilibrium density gradient ultracentrifugation
`fore has a deep major groove and a very shallow mn a
`(Section 5-5B; a technique Meselson,Stahl, and Jerome
`groove; it can be described asaflat ribbon wou”1f-
`Vinograd had developedfor the purposeof distinguish-
`around a 6 A in diameter cylindrical hole. Most 5°
`ing *N-labeled DNA from unlabeled DNA).
`
`
`
`10
`
`

`

`ive pensity ————_> Density —————>-
`
`
`
`
`
`
`
`
`
`DNA
`
`
`
`
`
`14N
`DNA
`
`Chapter 28. Nucleic Acid Structures and Manipulation 799
`
`C
`a 15N DNA (heavy)
`feonnm
`
`Hybrid DNA
`
`—%42bse 14NDNA(light)
`
`:
`
`—
`
`G %,
`\ als
`
`Generations
`
`0.3
`
`0.7
`
`41.0
`
`oe
`
`1.5
`
`1 i)
`
`2.5
`
`3.0
`
`A.A
`
`0 and 1.9
`mixed
`
`O and 4.1
`mixed
`
`displacementis proportional to the DNA concentration. The
`buoyant density of DNA increases with its 5N content. The
`bandsfurthest to the right (greatest radius and density) arise
`from DNAthatis fully *N labeled, whereas unlabeled DNA,
`whichis 0.014 g-cm~$ fess dense, forms the leftmost
`bands. The bandsin the intermediate position result from
`duplex DNAin which onestrandis 1=N labeled and the other
`strand is unlabeled. The accompanying interpretive drawings
`(right) indicate the relative numbers of DNA strands at each
`generation donated bytheoriginal parents (blue, 9N labeled)
`and synthesized by succeeding generations (red, unlabeled).
`[From Meselson, M. and Stahl, F. W., Proc. Natl. Acad. Sci.
`44, 674 (1958).]
`
`
`
`
`~DNA
`
`DNAs
`Hybrid
`DNA
`
`Hybrid
`DNA
`Figure 28-7
`; e demonstration of the semiconservative nature of DNA
`“plication in E. coli. DNA in a CsClsolution of density 1.71
`a was subjected to equilibrium density gradient
`acentrifugation at 140,000 g in an analytical
`0 racentrifuge (a device in which the spinning sample can be
`ically observed). The enormouscentrifugal acceleration
`‘sed the CsCl to form a density gradient in which DNA
`areated to its position of buoyant density. The left panels
`stro V absorption photographsof ultracentrifuge cells (DNA
`re oly absorbs UV light) and are arranged such that
`T Ss of equal density have the same horizontal positions.
`; Middle panels are microdensitometertraces of the
`"sponding photographsin which the vertical
`
`
`
`11
`
`

`

`
`
`
`wihthy
`
`amaa
`
`“hi
`
`800 Section 28-2. Double Helical Structures
`
`
`
`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 to
`Mysore Viswamitra. Note that the base pairs are inclined F
`the helix axis and that the helix has a hollow core. Compr
`this figure with Fig. 28-5. [Drawings copyrighted © byIM
`Geis. Computer graphics courtesy of Robert Stodola, FOX
`Chase CancerCenter.]
`
`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-
`fermational variation. It has not been established that
`A-DNAexists in vivo although a few experimental ob-
`servations suggest that certain DNA segments normally
`assume the A conformation.
`
`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
`28-82).
`Fiber diffraction and NMRstudies have shownthat
`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-DNAconfor-
`mation at high salt concentrations. Evidently,
`the
`Z-DNA conformation is most readily assumed by DNA seg-
`ments with alternating purine —pyrimidine base sequences
`Z-DNA Formsa Left-Handed Helix
`(for structural reasons explained in Section 28-3B). A high
`Occasionally, a seemingly well understoodorat least
`salt concentration stabilizes Z-DNArelative to B-DNA
`familiar system exhibits quite unexpected properties.
`by reducing the otherwise increasedelectrostatic repul-
`Over 25 years after the discovery of the Watson-—Crick
`sions between closest approaching phosphate groups
`on opposite strands (8 Ain Z-DNA vs 12 Ain B-DNA).
`Stucture,
`the
`crystal
`structure determination of
`d(CGCGCG) by Andrew Wang and AlexanderRich re-
`The methylation of cytosine residues at C(5), a common
`Yealed, quite surprisingly,a left-handed doublehelix(Fig.
`biological modification (Section 31-7), also promotes
`48-9; Table 28-1). A similar helix is
`formed by
`Z-DNAformation since a hydrophobic methyl group in
`€(CGCATGCG).
`This helix, which has been dubbed
`this position is less exposed to solvent in Z-DNAthanit
`“DNA,has 12 Watson -Crick base pairs per turn, a pitch
`is in B-DNA.
`Y 5A and, in contrast toA-DNA, a deep minorgroove and
`Does Z-DNA have any biological significance? Rich
`; iscernable majorgroove. Z-DNAtherefore resembles
`has proposed that the reversible conversion of specific
`eft-handed drill bit in appearance. The base pairs in
`segments of B-DNA to Z-DNA under appropriatecir-
`Bi are flipped 180°relative to thoseinB-DNA(Fig.
`cumstancesacts as a kind of switch in regulating genetic
`.. through conformational changes discussed in
`expression. Yet, the in vivo existence of Z-DNA has been
`Da 28-3B. As a consequence, the repeating unit of
`difficult to prove. A major problem is demonstrating
`mic] A 1s a dinucleotide, d(XpYp), rather than a single
`1 to
`that a particular probe for detecting Z-DNA, a Z-DNA-
`aoe as it is in the other DNA helices. Here, X is
`pare
`specific antibody, for example, does not in itself cause
`ving
`tesig Y a pyrimidine residue and Y is usually a purine
`what would otherwise be B-DNA to assume the Z
`Ox
`oe because the purine nucleotide assumes a con-
`conformation —akind of biological uncertainty princi-
`Pytingee that would besterically unfavorable in the
`ple (the act of measurementinevitably disturbs the sys-
`a Idine nucleotide. The line joining successive
`tem being measured). Recently, however, Z-DNA has
`thereate groups on a polynucleotide strand ofZ-DNA
`been shown to be present in E. coli by employing an
`°re follows a zigzag path arogind the helix (Fig.
`E. coli enzymethat methylates a specific base sequence
`
`
`13
`
`

`

`
`
`
`
`. a
`
`pe4 a
`
`binty
`
`802 Section 28-2. Double Helical Structures
`
`
`
`J
`
`Figure 28-9
`Ball-and-stick drawings and the corresponding space-filling
`models of Z-DNA as viewed (a) perpendicular to the helix
`axis and (6) (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
`
`a.
`Andrew Wang and AlexanderRich. Note that the helix Is |
`handed and that the sugar~ phosphate chains follow 4 adi
`zigzag course(alternate ribose residueslie at different -
`in Part b) indicating that the Z-DNA’s repeating motif 06-8:
`dinucleotide. Compare this figure with Figs. 28-5 an
`aphics
`[Drawings copyrighted © byIrving Geis. Computer gr
`courtesy of Robert Stodola, Fox Chase Cancer Genter.
`
`14
`
`14
`
`

`

`
`
`Figure 28-9 (b)
`
`Chapter 28. Nucleic Acid Structures and Manipulation 803
`
`in vitro when the DNAis in the B form but not whenitis
`in the Z form. The in vivo methylation of this basese-
`quence is inhibited whenit is clonedin E. coli (by tech-
`niques discussed in Section 28-8) within or adjacent toa
`DNA segmentthat can form Z-DNA. Moreover, there is
`a balance betweenthe in vivo B and Z forms of these
`DNAsthatis thoughtto be influenced by environmental
`factors such as salt concentration andprotein binding.
`Nevertheless, the biological function of Z-DNA,if any,
`rema