`
`Oxford, Exh. 1009, p. 1
`
`Oxford, Exh. 1009, p. 1
`
`

`

`
`
`Molecular
`Cloning
`
`SEDRATTeS
`
`Oxford, Exh. 1009, p. 2
`
`Oxford, Exh. 1009, p. 2
`
`
`
`

`

`
`
`Oxford, Exh. 1009, p. 3
`
`Oxford, Exh. 1009, p. 3
`
`

`

`
`
`Oxford, Exh. 1009, p. 4
`
`Oxford, Exh. 1009, p. 4
`
`

`

`Molecular
`Cloning
`A LABORATORY MANUAL
`SECOND EDITION
`
`All rights reserved
`© 1989 by Cold Spring Harbor Laboratory Press
`Printed in the United States of America
`
`98765
`
`Book and cover design by Emily Harste
`
`Cover: The electron micrograph of bacteriophage A particles
`stained with uranyl acetate was digitized and assigned false color
`by computer. (Thomas R. Broker, Louise T. Chow, and JamesI.
`Garrels)
`
`1-800-843-4388, In New York (516) 367-8423. FAX: (516) 367-8432.
`
`Cataloging in Publications data
`Sambrook, Joseph
`Molecular cloning : a laboratory manual / E.F.
`Fritsch, T. Maniatis—2nded.
`p.
`cm.
`Bibliography:p.
`Includes index.
`ISBN 0-87969-309-6
`1. Molecular cloning—Laboratory manuals. 2. Eukaryotic cells-
`-Laboratery manuals. I. Fritsch, Edward F. IL Maniatis, Thomas
`TIT. Title.
`QH442.2.M26 1987
`574.8773224—dc19
`
`
`
`87-35464
`
`Researchers using the procedures of this manualdo so at their own risk. Cold Spring Harbor
`Laboratory makes no representations or warranties with respect to the material set forth in
`this manual and has noliability in connection with the use of these materials.
`
`Authorization to photocopy items for internal or personaluse, or the internalor personal use of
`specific clients, is granted by Cold Spring Harbor Laboratory Press for libraries and other
`users registered with the Copyright Clearance Center (CCC) Transactional Reporting Service,
`provided that the base fee of $0.10 per page is paid directly to CCC, 21 Congress St., Salem MA
`01970, [0-87969-309-6/89 $00 + $0.10] This consent does not extend to other kinds of copying,
`such as copying for general distribution, for advertising or promotional purposes, for creating
`new collective works, or for resale.
`
`All Cold Spring Harbor Laboratory Press publications may be ordered directly from Cold
`Spring Harbor Laboratory Press, 10 Skyline Drive, Plainview, New York 11803. Phone:
`
`Oxford, Exh. 1009, p. 5
`
`Oxford, Exh. 1009, p. 5
`
`

`

`13
`
`
`rrrnainniimamnrammmnmnnsnaammmmmmmmmmmmmmammmaaassssmnanammmammmmminamaainaneanniita
`DNA Sequencing
`
`
`
`Sequencing Techniques and Strategies 13.3
`SANGER METHOD OF DIDEOXY-MEDIATED CHAIN TERMINATION 13.6
`Reagents Used in the Sanger Method of DNA Sequencing 13.6
`PRIMERS 13.6
`TEMPLATES 13.7
`DNA POLYMERASES 13.7
`RADIOLABELED dNTPs 13.9
`ANALOGS OF dNTPs
`13.10
`
`MAXAM-GILBERT CHEMICAL DEGRADATION OF DNA METHOD 13.11
`
`SEQUENCING STRATEGIES 13.14
`Confirmatory Sequencing 13.14
`De Novo Sequencing 13.14
`FACTORS AFFECTING THE CHOICE BETWEEN RANDOM AND DIRECTED
`STRATEGIES 13.18
`
`Random Sequencing 13.21
`
`GENERATION OF A LIBRARY OF RANDOMLY OVERLAPPING
`CLONES 13.24
`Purification and Ligation of the Target DNA 13.24
`Fragmentation of the Target DNA 13.26
`SONICATION 13.26
`DIGESTION WITH DNAase I IN THE PRESENCE OF MANGANESE IONS 13.28
`Repair and Size Selection of DNA 13.30
`Preparation of Vector DNA 13.31
`Ligation to Vector DNA 13.33
`
`Directed Sequencing 13.34
`
`GENERATION OF NESTED SETS OF DELETION MUTANTS 13.34
`Generation of Nested Sets of Deletions with Exonuclease HI 13.39
`
`Sequencing by the Sanger Dideoxy-mediated Chain-termination
`Method 13.42
`
`SETTING UP DIDEOXY-MEDIATED SEQUENCING REACTIONS 13.42
`Preparation of Single-stranded DNA 13.42
`Preparation of Primers
`13.42
`Microtiter Plates
`13.42
`Chain-extension/Chain-termination Reaction Mixtures
`STOCK SOLUTIONS OF dNTPs AND ddNTPs
`13.44
`
`138.43
`
`DENATURING POLYACRYLAMIDE GELS 13.45
`13.47
`Preparation of Buffer-gradient Polyacrylamide Gels
`Loading and Running Gradient Sequencing Gels
`13.54
`Autoradiography of Sequencing Gels
`13.56
`Reading the Sequence
`13.58 ©
`
`
`SnSarraa
`
`
`
`EXEE Contents
`
`
`
`Oxford, Exh. 1009, p. 6
`
`Oxford, Exh. 1009, p. 6
`
`

`

`
`
`
`
`
`
`
`
`
`
`
`DIDEOXY-MEDIATED SEQUENCING REACTIONS USING THE KLENOW
`FRAGMENTOF E. coli DNA POLYMERASEI 13.59
`Preparation 13.59
`PREPARATION OF WORKING SOLUTIONS OF dNTPs
`PREPARATION OF WORKING SOLUTIONS OF ddNTPs
`Sequencing Reactions
`13.61
`
`13.60
`13.60
`
`DIDEOXY-MEDIATED SEQUENCING REACTIONS USING
`SEQUENASES 13.65
`Preparation 13.65
`Sequencing Reactions 13.67
`
`Lithium Chloride
`
`13.72
`
`PROBLEMS THAT ARISE WITH DIDEOXY-MEDIATED SEQUENCING 13.73
`Template-specific Problems
`13.73
`Systematic Problems
`13.73
`Problems with Polyacrylamide Gels
`
`13.74
`
`Sequencing by the Maxam-Gilbert Method 13.78
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`SEQUENCING DENATURED DOUBLE-STRANDED DNA TEMPLATES 13.70
`Sequencing of Plasmid DNAs Purified by Equilibrium Centrifugation in CsCl—
`Ethidium Bromide Gradients
`13.71
`
`
`Removal of RNA from Minipreparations of Plasmid DNA by Precipitation with
`
`
`
`
`
`14
`
`In Vitro Amplification of DNA by the
`Polymerase Chain Reaction
`
`APPLICATIONS OF PCR AMPLIFICATION 14.5
`Generation of Specific Sequences of Cloned Double-stranded DNAfor Use as
`Probes
`14.6
`
` xxiii
`Contents
`
`Oxford, Exh. 1009, p. 7
`
`Asymmetric Labeling of Target DNA 13.78
`Preparation of Target DNA for Maxam-Gilbert Sequencing 13.83
`Reagents, Solutions, and Apparatuses
`13.83
`
`THE TRADITIONAL METHOD OF MAXAM-GILBERT SEQUENCING 13.88
`Cleavage at G Residues
`13.88
`13.90
`Cleavage at Purine Residues (A+G)
`Cleavage at Pyrimidine Residues (C+T)
`13.91
`Cleavage at C Residues
`13.92
`Cleavage at A and C Residues (A>C)
`Treatment of Samples with Piperidine
`
`13.93
`13.94
`
`ALTERNATIVE METHODS OF MAXAM-GILBERT SEQUENCING 13.95
`
`TROUBLESHOOTING GUIDE FOR MAXAM-GILBERT SEQUENCING 13.95
`Reading Sequencing Gels
`13.95
`Problems Commonly Encountered 13.95
`
`References 13.102
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Oxford, Exh. 1009, p. 7
`
`

`

`
`
`Oxford, Exh. 1009, p. 8
`
`Oxford, Exh. 1009, p. 8
`
`

`

`In the mid-1970s, when molecular cloning techniques in general were rapidly
`improving, simple methods were also developed to determine the nucleatide
`sequence of DNA. These advances laid the foundation for the detailed
`analysis of the structure and function of large numbers of genes. The first
`attempts to sequence DNA mirrored techniques developed in the 1960s to
`sequence RNA(see Sangeret al. 1965; Brownlee et al. 1968; Brownlee 1972).
`These involved (1) specific cleavage of the DNA into smaller fragments by
`enzymatic digestion (endonuclease IV [Robertson et al. 1973; Ziff et al. 1973])
`or chemical digestion (pyrimidine tract analysis [Robertson et al. 1973; Ziff et
`al. 1973]), (2) nearest neighbor analysis (Wu and Taylor 1971), and (3) the
`wandering spot method (Sanger et al. 1973; Tu and Wu 1980).
`Indeed, in
`some studies the DNA wastranscribed into RNA with Escherichia coli RNA
`polymerase and then, sequenced as RNA (Gilbert and Maxam 1973).
`It is a
`testimony to the success of DNA sequencing that today most protein se-
`quences are deduced from the nucleotide sequences of genes or cDNAs.
`
`
`
`S
`|
`:
`
`
`
`
`
`13.2 DNA Sequencing
`
`
`
`Oxford, Exh. 1009, p. 9
`
`Oxford, Exh. 1009, p. 9
`
`

`

`
`
`
`Sequencing Techniques and Strategies
`
`The two rapid sequencing techniques in current use are the enzymatic
`method of Sanger et al. (1977) and the chemical degradation method of
`Maxam and Gilbert (1977). Although very different in principle, these two
`methods both generate separate populations of radiolabeled oligonucleotides
`that begin from a fixed point and terminate randomly at a fixed residue or
`combination of residues. Because every base in the DNA has an equal chance
`of being the variable terminus, each population consists of a mixture of
`oligonucleotides whose lengths are determined by the location of a particular
`base along the length of the original DNA. These populations of oligonu-
`cleotides are then resolved by electrophoresis under conditions that can
`discriminate between individual DNAsthat differ in length by aslittle as one
`nucleotide. Whenthe populations are loaded into adjacent
`lanes of a
`sequencing gel, the order of nucleotides along the DNA canbe read directly
`from an autoradiographic image of the gel (see, e.g., Figure 13.1).
`
`
`
`DNA Sequencing 13.3
`
`Oxford, Exh. 1009, p. 10
`
`Oxford, Exh. 1009, p. 10
`
`

`

`
`
`snvdVn.nnn
`FVIIIrnIIIyyw)nnn
`JewWudzapyosjonuobljoQS&~ee,YNG.ae}dwa}pepuesys-e/6uls‘S%
`
`
`
`
`a———_wvIIA)
`
`
`diVPpdil9ppdopedilpp
`
`
`
`
`
`GLNPJeuouay)joeoejdUlpayesodiooulsiGLNPP&UBYMeyeulUe}SuTeEYOpaziseylUASAjMauUOu,
`ryLS!
`
`
`sisaioydosyoaieAqYNpeleqejoiperjosjusuiGe1yayesedespueoinyeueg
`
`
`
`
`Burpnioul!)Sd.NP QasesewAjodYNG.&(aLNPldze]ino}=3=wh
`
`
`
`
`
`
`
`apyoajonuAxoapipBuneuiusa}
`
`
`
`
`
`
`
`
`VNpeziseujudsAjmou\Jeunid
`
`
`
`
`
`<=
`
`Oxford, Exh. 1009, p. 11
`
`Oxford, Exh. 1009, p. 11
`
`
`

`

`
`
`
`
`
`
`DNA Sequencing
`
`13.5
`
`j'
`
`Oxford, Exh. 1009, p. 12
`
`UOBUIUIII}
`
`ureyopoyerpeur
`
`
`
`
`
`-Axoapip1esuegey}AqSurouonbog
`
`Oxford, Exh. 1009, p. 12
`
`
`
`
`

`

`SANGER METHOD OF DIDEOXY-MEDIATED CHAIN TERMINATION
`
`4
`
`Reagents Used in the Sanger Method of DNA Sequencing
`PRIMERS
`~
`
`The current chain-termination method evolved from the +/— sequencing
`technique (Sanger and Coulson 1975), which first described (1) the use of a
`specific primer for extension by DNA polymerase, (2) base-specific chain
`termination, and (3) the use of polyacrylamide gels to discriminate between
`single-stranded DNA chains differing in length by a single nucleotide. De-
`spite these advances, the +/ — method wastoo inaccurate and clumsyto gain
`general acceptance, and it was not until the introduction of chain-terminating
`dideoxynucleoside triphosphates (ddNTPs) (Sanger et al. 1977) that en-
`zymatic methods of DNA sequencing were used extensively.
`'2'3' ddNTPsdiffer from conventional dNTPs in that they lack a hydroxyl
`residue at the 3’ position of deoxyribose. They can be incorporated by DNA
`polymerases into a growing DNAchain through their 5’ triphosphate groups.
`However,
`the absence of a 3'-hydroxyl residue prevents formation of a
`phosphodiester bond with the succeeding dNTP. Further extension of the
`growing DNAchain is therefore impossible. Thus, when a small amount of
`one ddNTP is included with the four conventional dNTPs in a_ reaction
`mixture for DNA synthesis, there is competition between extension of the
`chain and infrequent, but specific, termination. The products of the reaction
`are a series of oligonucleotide chains whose lengths are determined by the
`distance between the terminus of the primer used to initiate DNA synthesis
`andthe sites of premature termination. By using the four different ddNTPs
`in four separate enzymatic reactions, populations of oligonucleotides are
`generated that terminate at positions occupied by every A, C, G, or T in the
`template strand (see Figure 13.1, pages 13.4—13.5).
`
`13.6 DNA Sequencing
`
`In enzymatic sequencing reactions, priming of DNA synthesis is achieved by
`the use of a synthetic oligonucleotide complementary to a specific sequence on
`the template strand.
`In many cases, this template is obtained as a single-
`stranded DNA molecule by cloning the target DNA fragment into a bac-
`teriophage M13 or phagemid vector. However, it is also possible to use the
`Sanger method to sequence denatured double-stranded DNA templates(e.g.,
`denatured plasmid DNA)
`(see pages 13.70-13.72).
`In either. case,
`the
`problem of obtaining primers that are complementary to an unknown se-
`quence of DNA is then solved by using a “universal” primer that anneals to
`vector sequences that flank the target DNA. Universal primers used for the
`sequencing of bacteriophage M13 recombinant clones are typically 15-29
`nucleotides in length and annealto the sequences immediately adjacent to (1)
`the HindIII site in the polycloning region of bacteriophage M13mp18 and(2)
`the EcoRI site in the polycloning. region of bacteriophage M13mp19. These
`primers, which can also be used for “double-stranded” sequencing of DNAs
`cloned into pUC plasmids, are available from a large number of commercial
`suppliers.
`In addition, several companies sell primers that have been
`designed to allow sequencing of target DNAscloned into a variety of restric-
`tion sites in different plasmids.
`
`
`
`Oxford, Exh. 1009, p. 13
`
`Oxford, Exh. 1009, p. 13
`
`

`

`TEMPLATES
`
`DNA POLYMERASES
`
`Several different enzymes are commonly used for dideoxy-mediated sequenc-
`ing, including the Klenow fragment of E. coli DNA polymerase I (Sanger et
`al. 1977), reverse transcriptase (see, e.g., Mierendorf and Pfeffer 1987),
`bacteriophage T7 DNA polymerases that have been modified to eliminate
`3’-> 5’ exonuclease activity (Sequenase and Sequenase version 2.0) (Tabor
`and Richardson 1987), and the thermostable DNA polymerase isolated from
`Thermus aquaticus (Taq DNA polymerase). The properties of these DNA
`polymerases (see Table 18.1) differ greatly in ways that can considerably
`affect the quantity and quality of the DNA sequence obtained from chain-
`termination reactions.
`
`Klenow fragment of E. coli DNA polymerase I
`
`two types of DNA can be used as templates in the
`As mentioned above,
`Sanger method of sequencing: pure single-stranded DNA and double-strand-
`ed DNA that has been denatured by heat or alkali. The best results are
`obtained from single-stranded DNA templates, which are usually isolated
`from recombinant bacteriophage M13 particles. When care is taken to
`optimize the ratio of single-stranded template to primer, it is possible to
`obtain several hundred nucleotides of sequence from each set of chain-
`termination reactions. Results of this quality are more difficult to obtain
`when denatured double-stranded DNA is used as a template. Despite the
`apparent simplicity and convenience of the method (Chen and Seeburg 1985),
`it has only recently been improved to the point where it can reliably yield
`unambiguous results from double-stranded DNA templates. Two factors are
`critical: the quality of the template DNA and the type of DNA polymerase
`that is used (see below). Minipreparations of plasmid DNA are always
`contaminated by small oligodeoxyribonucleotides and ribonucleotides, which
`serve as random primers, and by inhibitors of DNA polymerases. As a
`consequence, sequencing gels are frequently obscured by a variety of “ghost”
`bands, strong stops, and other artifacts. We therefore recommend that
`minipreparations of plasmid DNAs should not be used to determine the
`sequence of cloned segments of unknown DNA. However, such DNAs are
`often adequate templates for confirming a sequence that has already been
`determined by another method. Plasmid DNA that has been purified by
`equilibrium centrifugation in CsCl—ethidium bromide gradients yields far
`better results, although the labor and expense of preparing plasmid DNA in
`this way are considerable.
`
`DNA Sequencing 13.7
`
`This enzyme wasoriginally used to develop the Sanger method andisstill
`used extensively for DNA sequencing. Two problems frequently arise:
`
`* The low processivity of the enzyme causes the Klenow fragmentto generate
`a high background of fragments that terminate, not by incorporation of a
`ddNTP, but by the random dissociation of the polymerase from the tem-
`plate. The inability of the enzyme to travel long distances along the
`template limits the length of sequence that can be obtained from standard
`
`
`
`Oxford, Exh. 1009, p. 14
`
`Oxford, Exh. 1009, p. 14
`
`

`

`sequencing reactions using this enzyme. Typically, such reactions generate
`approximately 250 to 350 nucleotides of sequence. The amount of sequence
`can be doubled by carrying out the reaction in two steps—an initial labeling
`step containing low concentrations of dNTPs,followed by a chain-extension/
`chain-termination reaction containing ddNTPs and a high concentration of
`dNTPs (Johnston-Dow et al. 1987; Stambaugh and Blakesley 1988). How-
`ever, even with these improvements, the Klenow enzymedoes not routinely
`yield as much sequence as the more processive Sequenase enzymes (see
`below).
`¢ The enzyme will not efficiently copy homopolymeric tracts or other regions
`of high secondary structure in the template. This problem can beallevi-
`ated, but not completely solved, by increasing the temperature of the
`polymerization reaction to 55°C (Gomer and Firtel 1985). dNTP analogs
`(e.g., dITP or 7-deaza-dGTP[see page 13.10]), which are sometimes used to
`obtain sequence information at regions of the template that form stable
`secondary structures, are less effective with the Klenow enzyme than with
`Sequenases, perhaps because they decrease still further the already low
`processivity of the enzyme.
`In summary, the Klenow fragment of £. coli DNA polymerase I is the
`enzyme of choice for determining the sequence of tracts of DNA that lie
`within 250 bases of the 5’ terminus of the primer. It is not recommendedfor
`sequencing longer segments of DNA or DNAs with dyad symmetry and/or
`homopolymeric tracts.
`
`13.8 DNA Sequencing
`
`Reverse transcriptase
`Although not widely used for routine sequencing, this enzymeis occasionally
`employed to resolve problems caused by the presence of homopolymeric
`regions of A/T or G/C in the template DNA. Reverse transcriptases from
`both avian and murine sources appearto beslightly better in this respect
`than the Klenow enzyme (Karanthanasis 1982; Graham et al. 1986), al-
`though perhaps not as good as the Sequenases (Cameron-Mills 1988; Revak
`et al. 1988).
`
`TABLE 13.1 Properties of DNA Polymerases Used in DNA
`Sequencing Reactions
`
`BeeeSetEnzyme Processivity* Rate of polymerization’
`
`A
`
`\
`
`Klenow fragment
`of E. coli DNA
`45
`10-50
`polymeraseI
`Reverse transcriptase
`5
`n.d.
`(AMV)
`~2000 -
`Sequenase and
`300
`~3000
`Sequenase version 2.0
`35-100
`> 7600
`Tag DNApolymerase and AmpliTaq’
`“ Processivity is expressed as the average number of nucleotides synthesized before the enzyme
`dissociates from the template; n.d. indicates not determined.
`> Rate of polymerization in nucleotides/ second,
`° Taq DNA polymerase, a highly processive DNA polymerase, is useful for determining the
`sequence of DNA templates that form stable secondary structures.
`
`
`Oxford, Exh. 1009, p. 15
`
`Oxford, Exh. 1009, p. 15
`
`

`

`Sequenases
`Sequenase™™ is a form of bacteriophage T7 DNA polymerase that has been
`chemically modified to eliminate much of the enzyme’s powerful 3’—5’
`exonuclease activity. Sequenase version 2.0 is a genetically engineered form
`of Sequenase that entirely lacks 3'—5' exonuclease activity, is extremely
`stable, and has a threefold higher specific activity than the chemically
`modified enzyme. Sequenases are the enzymes of choice for determining the
`sequences of long tracts of DNA because of their very high processivity, their
`high rate of polymerization, and their wide tolerance for nucleotide analogs
`such as dITP and 7-deaza-dGTP that are used to resolve regions of compres-
`sion in sequencing gels. Sequenases travel such long distances along the
`template that several hundred nucleotides of DNA sequence can often be
`determined from a single set of reactions.
`In fact, the amount of sequenceis
`limited more by the resolving power of polyacrylamide gels than by the
`properties of the enzyme.
`To take full advantage of the high processivity of Sequenases, a two-step
`sequencing reaction is set up. In thefirst stage, low concentrations of dNTPs
`and low temperature are used to limit the extent of synthesis and to ensure
`efficient incorporation of a radiolabeled dNTP. The products of this reaction
`are primers that have been extended by only 20-30 bases. Thefirst reaction
`is then divided into the standard set of four reactions, each of which contains
`high concentrations of dNTPs and a single ddNTP. Polymerization then
`continues until a chain-terminating nucleotide is incorporated into the grow-
`ing chain.
`
`DNA Sequencing 13.9
`
`°
`
`Taq DNA polymerase
`Taq DNA polymerase is useful for determining the sequence of single-
`stranded DNA templates that form extensive stable secondary structures at
`37°C. This is because Tag DNA polymerase worksefficiently at 70—75°C, a
`temperature that precludes formation of secondary structure even in tem-
`plates that are rich in G+C. When used as described by Innis etal. (1988),
`sequencing ladders produced by Tag DNA polymerase demonstrate a high
`uniformity of band intensity for several hundred nucleotides, suggesting that
`the enzyme has a high degree of processivity.
`
`RADIOLABELED dNTPs
`
`Until a few years ago, virtually all DNA sequencing was carried out with
`[a-"’P]ANTPs. However, the strong B particles emitted by *’P created two
`problems. First, because of scattering, the bands on the autoradiograph were
`far larger and more diffuse than the bands of DNAin the gel. This affected
`the ability to read a sequence correctly (particularly from the upper part of
`the autoradiograph) and limited the numberof nucleotides that could be read
`from a single gel. Second, decay of **P caused radiolysis of the DNA in the
`sample. Sequencing reactions radiolabeled with *?P could therefore be stored
`for only 1 or 2 days before the DNA wasso badly damaged that it generated
`indecipherable sequencing gels.
`The introduction of [*°S]dGATP (Biggin et al. 1983) greatly alleviated both of
`these problems. Because of the decreased scatter of the weaker f particles
`
`
`Oxford, Exh. 1009, p. 16
`
`Oxford, Exh. 1009, p. 16
`
`

`

`produced by decayof *’S, thereis little loss of resolutién between the gel and
`the autoradiograph.
`This allows unambiguous determination of several
`hundred nucleotides of DNA sequence from a single reaction set. Further-
`more, the lower energy of *°S producesless radiolysis, allowing sequencing
`reactions to be stored for up to 1 week at —20°C without noticeable loss of
`resolution. Thus, if technical problems arise with a polyacrylamide gel, the
`sequencing reactions can simply be reanalyzed.
`
`ANALOGS OF dNTPs
`
`13.10 DNA Sequencing
`
`Regions of DNA with dyad symmetry (especially those with a high G+C
`content) can form intrastrand secondary structures that are not fully dena-
`tured during electrophoresis. This can cause an anomalous pattern of
`migration in which adjacent bands of DNA become compressed to the point
`where they are difficult to read. Compression is entirely dependent on the
`presence of secondary structures in DNAandcannotbe alleviated by chang-
`ing the type of DNA polymerase used in the sequencing reaction. However,
`compressed regions of gels can usually be resolved by using a nucleotide
`analog such as dITP (2'-deoxyinosine-5'-triphosphate) or 7-deaza-dGTP (7-
`deaza-2'-deoxyguanosine-5’-triphosphate). These analogs pair weakly with
`conventional bases and are good substrates for DNA polymerases such as the
`Sequenases and Tag DNA polymerase (Gough and Murray 1983; Mizusawa
`et al. 1986; Innis et al. 1988).
`Some compressions are not resolved by
`7-deaza-dGTP; others (particularly those occurring in GC-rich regions) are
`not resolved by dITP.
`If it is necessary to use analogs, try dITP first (see
`pages 13.74—13.75). This analog, in contrast to 7-deaza-dGTP, does not affect
`the sharpness of the DNA bandsin the sequencing gel. Any compression that
`is not resolved by either dITP or 7-deaza-dGTP can almost always be cleared
`up by determining the sequence of both strands of the DNA.
`As discussed above, both forms of Sequenase and Taq DNA polymerase
`tolerate nucleotide analogs better than does the Klenow fragmentof EF. coli
`DNA polymerase I. In addition, the manufacturer claims that Sequenase
`version 2.0 is superior to the original enzyme when sequencing templates
`with strong secondary structure. Version 2.0 is more processive than Sequen-
`ase, having less tendency to pause,
`thereby eliminating “ghost” bands.
`Furthermore, version 2.0 appears to tolerate nucleotide analogs such as dITP
`better than does the original version.
`
`
`
`Oxford, Exh. 1009, p. 17
`
`Oxford, Exh. 1009, p. 17
`
`

`

`
`
`MAXAM-GILBERT CHEMICAL DEGRADATION OFDNA METHOD
`the
`Unlike the chain-termination technique, which involves synthesis,
`Maxam-Gilbert method involves chemical degradation of the original DNA.
`This method grew out of studies of the interaction between the lac repressor
`and the Jac operator in vitro.
`To this day,
`the ability to probe DNA
`conformations and protein-DNAinteractions remains a unique feature of the
`Maxam-Gilbert method.
`In this procedure (Maxam and Gilbert 1980), a fragment of DNA
`radiolabeled at one end is partially cleaved in five separate chemical reac-
`tions, each of which is specific for a particular base or type of base. This
`generates five populations of radiolabeled molecules that extend from a
`common point (the radiolabeled terminus) to the site of chemical cleavage.
`Each population consists of a mixture of molecules whose lengths are de-
`termined by the locations of a particular base along the length of the original
`DNA.
`These populations are then resolved by electrophoresis through
`polyacrylamide gels, and the end-labeled molecules are detected by au-
`toradiography (see Figure 13.2).
`The Maxam-Gilbert method has remained relatively unchangedsince its
`initial development. Although additional chemical cleavage reactions have
`been devised (for review, see Ambrose and Pless 1987), these are generally
`used to supplement the reactions originally described by Maxam and Gilbert
`(1977, 1980). The success of the method dependsentirely on the specificity of
`these cleavage reactions, which are carried out in two stages.
`In the first
`stage, specific bases (or types of bases) undergo chemical modification; in the
`second stage, the modified base is removed from its sugar and the phos-
`phodiester bonds 5’ and 3’ to the modified base are cleaved (see Table 13.2).
`In every case,
`these reactions are carried out under carefully controlled
`conditions to ensure that on average only one of the target bases in each DNA
`molecule is modified. Subsequent cleavage by piperidine at the 5’ and 3’
`sides of the modified bases yields a set of end-labeled molecules whose
`lengths range from one to several hundred nucleotides. The DNA sequence
`can then be read from an autoradiograph of a sequencing gel by comparing
`the G, A+G, C+T, C, and A>C tracks. For a numberof reasons (e.g., the
`use of 2’P as a radiolabel,
`the specific activity of end-labeled DNA, the
`statistical distribution of cleavage
`sites, and the limitations of gel
`technology), the range of the Maxam-Gilbert method is less than that of the
`Sanger method; the Maxam-Gilbert method works optimally for DNAse-
`quences that lie less than 250 nucleotides from the radiolabeled end.
`When the Maxam-Gilbert and Sanger methods werefirst developed in the
`1970s, sequencing by chemical degradation was both more reproducible and
`more accessible to the average worker. The Sanger method required single-
`stranded templates, specific oligonucleotide primers, and access to high-
`quality preparations of the Klenow fragment of E. coli DNA polymerase I.
`The Maxam-Gilbert method used simple chemical reagents that were avail-
`able to everyone. However, with the developmentof bacteriophage M13 and
`phagemid vectors, the ready availability of synthetic primers, and improve-
`ments to the sequencing reactions, the dideoxy-mediated chain-termination
`method is now used much more extensively than the Maxam-Gilbert method.
`Nevertheless, the chemical degradation approach has one clear advantage
`
`DNA Sequencing 13.11
`
`Oxford, Exh. 1009, p. 18
`
`
`
`Oxford, Exh. 1009, p. 18
`
`

`

`over the chain-termination method: Sequence is obtained from the original
`DNA molecule and not from an enzymatic copy. Therefore, with the Maxam-
`Gilbert method, one can sequence synthetic oligonucleotides, analyze DNA
`modifications such as methylation, and study both DNA secondary structure
`and the interaction of proteins with DNA by either chemical protection or
`modification interference experiments. However, because of its ease and
`rapidity, the Sanger technique is now the best choice for simple determina-
`tion of DNA sequence. In fact, most of the current sequencing strategies have
`been designed for use with this method.
`
` 13.12 DNA Sequencing
`
`
`
`32:
`
`G
`
`(AtG)
`
`C
`
`(C+T) A>C
`
`ACACTGAACGTTCATGTCGA....
`
`32
`
`me
`PACACTGAACGTTCATGTCGA....
`
`32p
`
`me
`
`ACACTGAACGTTCATGTCGA....
`
`32
`
`me
`PACACTGAACGTTCATGTCGA see
`
`me
`
`;P
`
`32p
`
`32p
`
`ACACT
`
`ACACTGAAC
`
`ACACTGAACGTTCAT
`
`P
`
`ACACTGAACGTTCATGTC
`
`
`
`Radiolabel target DNA
`at only one end
`
`Carry out five base-specific
`cleavage reactions
`
`For example,
`partial modification
`of G residues by
`dimethyl sulfate
`
`Release of modified G
`residues and hydrolysis
`of sugar-phosphate back-
`bone by piperidine
`
`Separate fragments of radiolabeled
`DNAbygel electrophoresis
`
`FIGURE 13.2
`Sequencing by the Maxam-Gilbert chemical degradation of DNA method.
`
`Oxford, Exh. 1009, p. 19
`
`Oxford, Exh. 1009, p. 19
`
`

`

`A+G
`
`C+T
`
`C
`
`TABLE 13.2 Chemical Modifications Used in the Maxam-Gilbert
`MethodSeell
`Base
`Specific modification®
`
`G
`Methylation of N, with dimethyl sulfate at pH 8.0 makes the
`C,—C, bond specifically susceptible to cleavage by base
`Piperidine formate at pH 2.0 weakens the glycosidic bond of
`adenine and guanine by protonating nitrogen atoms in the
`purine rings resulting in depurination
`Hydrazine opens pyrimidine rings, which recyclize in a five-
`membered form that is susceptible to removal
`In the presence of 1.5 M NaCl, only cytosine reacts appreciably
`with hydrazine
`1.2~ NaOH at 90°C results in strong cleavage at A and weaker
`A>C
`cleavage at CeenceDIN
`* Hot (90°C)piperidine (1 M in H,O) is used to cleave the sugar-phosphate chain of DNAat the
`sites of chemical modifications.
`
`DNA Sequencing 13.13
`
`
`
`Oxford, Exh. 1009, p. 20
`
`Oxford, Exh. 1009, p. 20
`
`

`

`SEQUENCING STRATEGIES
`Before beginning to sequence, it is important to develop an overall strategy
`that takes into accountthesize of the region to be sequenced, the accuracy of
`the sequence required, and thefacilities that are available. Only a minor
`proportion of projects involve the de novo accumulation of large tracts of
`virgin sequence. More often, sequencing is used to map and identify
`mutations (e.g., point mutations and deletions) and to verify the orientation
`and structure of recombinant DNAconstructs. The approaches used for these
`two purposes are very different.
`
`De Novo Sequencing
`The aim of de novo sequencingis to provide the accurate nucleotide sequence
`of a virgin segment of DNA that may be many kilobases in length. This task
`requires careful planning because the maximum length of target DNA that
`can be sequenced accurately in a single set of sequencing reactions is
`approximately 400 bases. Target DNAsof this size can be sequenced by
`cloning in opposite orientations in each of two bacteriophage M13 vectors
`(e.g., M13mp18 and M13mp19). The entire sequence of each strand can then
`be determined in a single reaction set using a universal sequencing primer.
`To sequence longer target DNAs(e.g., several kilobases in length), one of two
`general strategies can be used:
`
`Confirmatory Sequencing
`Confirmatory sequencing (e.g., sequencing of mutants generated by oligonu-
`cleotide-mediated mutagenesis) often requires no more than one set of
`reactions that generates the nucleotide sequence of a local region of one of the
`two strands of DNA.
`This can usually be achieved by sequencing an
`appropriate restriction fragment
`that has been subcloned into a bac-
`teriophage M13 or phagemid vector. In many cases, the region of interest
`will lie within the sequencing range of a universal primer; if not, the best
`strategy is to synthesize a priming oligonucleotide 17-19 nucleotides long
`that is complementary to sequences located approximately 50-100 nu-
`cleotides from the region of interest. Wheneverpossible, the sequence of the
`homologous region of the wild-type gene should be determined at the same
`time as that of the mutant. Direct comparison of the sequences on the same
`autoradiograph greatly facilitates confirmation of the sequence of the altered
`region and clearly reveals any unexpected, additional differences between
`mutant and wild-type genes.
`
`13.14 DNA Sequencing
`
`¢ A random approach (or shotgun sequencing), in which sequence data are
`collected from subclones containing random segments of the target DNA.
`No attempt is made to determine where these subclones mapin the target
`DNAor which strand of DNAis being sequenced. Instead, the accumulated
`data are stored andfinally arranged in order by a computer (Staden 1986).
`This approach, which was pioneered by the M.R.C. Laboratory in Cam-
`bridge, has been used successfully to determine the sequences of human
`mitochondrial DNA (Andersonet al. 1981), human adenovirus DNA (Gin-
`
`
`
`Oxford, Exh. 1009, p. 21
`
`Oxford, Exh. 1009, p. 21
`
`

`

`geras et al. 1982; Roberts et al. 1986), bacteriophage A DNA (Sangeret al.
`1982), and Epstein-Barr virus DNA (Baeret all. 1984).
`* Directed approaches, in which sequences of the target DNAareobtained in
`a systematic fashion. For example, the entire sequence of the target DNA
`might be obtained by sequencing a nested set of deletion mutants that
`begin