BIOPOLYMERS
`
`VOL. 13, 669-675 (1974)
`
`The Effect of Thymine Dimers on
`DNA: DNA Hybridization
`
`MICHAEL KAHN, Department of Biological Sciences, Stanford University,
`Stanford, California 94S05
`
`synopsis
`DNA from bacteriophage T7 was irradiated at long ultraviolet wavelengths in the pres-
`ence of silver ions. Such treatment leads to selective production of thymine: thymine
`dimen in DNA. The DNA was melted and the renaturation rate was determined as a
`function of thymine dimer content and renaturation temperature. Under “normal”
`hybIidieation conditions little change in the renaturation rate was observed even when
`30% of the thymine was dimerised. This result is consistent with the view that up to a
`15% change in the primary sequence of DNA does not appreciably change the renatura-
`tion rate.
`
`The kinetics of nucleic acid hybridization have been shown to be depen-
`dent on a variety of factors such as temperature, ionic strength, and gua-
`nine-cytosine c0ntent.l A long-standing question is to what extent a
`mismatch of bases (i.e., unconventional pairing of bases) will affect the
`renaturation rate.’ Since the measurable parameter in the Wetmur and
`Davidson formulation2 of renaturation kinetics K , contains both a com-
`plexity correction factor and a putative mismatch correction factor, the
`estimation of complexity based solely on K N might be questioned in cases
`in which mismatch is known to be p r e ~ e n t . ~ . ~
`As an example of the effect of mismatching, reannealed hybrids of the
`rapidly renaturing fraction of eukaryotic DNA show a decreased melting
`temperature after renaturation, indicating that the hybrid contains mis-
`matched bases.5 Direct sequence analysis of the guinea pig a satellite
`DNA suggests that mutational change in the primary DNA sequence is
`common and will lead to mismatch upon hybridization. Sutton and
`RlcCallum4 reannealed mouse satellite DNA and subsequently separated
`the duplexes into four classes on the basis of differing melting temperature
`T,. The rate of renaturation of these four classes was strongly dependent
`on their respective T,’s-an
`observation that led Southern to propose3 that
`the lowered thermal stability was the result of mutational changes in some
`homogeneous primal sequence and that these changes reduced the rate of
`hybrid formation.
`The system described in this paper was designed to determine whether a
`correction for mismatch should be applied to kinetic hybridization data.
`In this system the intrastrand thymine dimer was used to model a muta-
`669
`
`@ 1974 by John Wiley & Sons, Inc.
`
`Enzo Exhibit 2015
`Hologic, Inc. v. Enzo Life Sciences, Inc.
`Case IPR2016-00820
`
`
`Exhibit 2015 Page 1
`
`

`
`670
`
`KAHN
`
`tional mismatch. The choice was justified by the knowledge that a dimer
`decreases the stability of the DNA duplex and that its action is at the level
`of primary structure. These are the minimum defining properties of mis-
`match. Brunk6 and Shafranovkaya ct al.7 concluded by diff erent methods
`that at low dimer concentrations the distribution of dimers in DNA is not
`random. Brunk showed that the Iongcr pyrimidine tracts contained a
`greater than expected percentage of thymine as dimers. At a high level
`of dimerization the dimers are as random as the pyrimidine t,racts in which
`they occur.6
`The advantages of the thymine dimcr as a specific lesion in DNA are
`numerous. Using silver ions, high yiclds of dimers are possible with
`negligible contribution from other photoproducts.8 Doublc-stranded
`DNA can be used and therefore melting temperatures will show no effects
`due to self-sorting of altered strands. Introduction of the photoproduct
`is rapid and convenient, and the photoproducts have been well charac-
`terized and are fairly simple to assay.
`Similar model systems have been used in nttempts to determine the
`effect of mismatch on renaturation rate. Deamination has been uscd in
`order to model transition-type mutation^.^-^^ Transversions have been
`modeled by using glyoxal to completo a third ring on guanosine residues"
`and by using chloroacetaldehyde to modify adenine and cytosine.l2 In all
`of these studies renaturation rate was not found to be very dependent on
`the presence of the alteration; the maximum rate depression reported is 80%
`for heavily glyoxylated DNA with a melting temperature 24OC lower than
`that of native DNA."
`
`MATERIALS AND METHODS
`
`DNA
`T7 wild-type phage was obtained from M. Chamberlain; E . coli strain
`B/r thy- was obtained from H. Nakayama. Unlabeled T7 DNA was
`prepared in a manner similar to that of England.13 Tritium labeled DNA
`was prepared by growing E . coli B/r thy- in 0.2% glucose, 0.1% casamino
`acids (Difco) to an OD of 0.8 in a medium containing 3H (methyl) thymine
`(1 pg/ml, 2 pm Ci/ml NEN) and infecting with T7 at a multiplicity of 0.1.
`Phage were prepared following Thomas and Abe1s0n.l~ All phage were
`sedimented in CsCl step gradients and banded at equilibrium. The DNA
`was extracted with phenol, dialyzed against 0.06 M NaH2P04, 0.06 14
`N&HPO,, (0.12 M Nap), and stored over chloroform. Concentration
`was measured by determining A260.
`Formation of Dimers
`DNA was irradiated at 7 or 70 pg DNA/ml for various lengths of time
`with a high-pressure mercury lamp (PEEZ 10010) through a 1.6-mm glass
`filter. The glass had a transmittance of less than 2% at 300 mp and 47%
`
`
`Exhibit 2015 Page 2
`
`

`
`DNA: DNA HYBRIDIZATION
`
`67 1
`
`at 310 mp. Silver nitrate was addcd to a Ag:DNA-phosphatc ratio of
`greater than 1. Water-saturated nitrogen was bubblcd through the DNA-
`silver solution prior to and during irradiation. An aliquot of thc irradiated
`DNA was acid hydrolyzed and analyzcd for photoproducts on Dowcx
`formate columns and by paper chromatography.
`Hybridization
`The irradiated DNA4 was dialyzed oncc against 0.02 M NaCN and then
`twice against 0.12 M NaP to remove Ag+. Itclease of silver was com-
`plcte as judged by disappearance of Agllom in control cxpcrimcnts. The
`DNA was then sonicated and its molecular wcight determined by scdi-
`mentation through a 5-20% alkaline sucrosc gradicnt using a method
`similar to that of Abelson and Thomas16 (Spinco 50.1 rotor, 46,000 rpm,
`Nlolecular weights wcrcin thc rangc of 3-5 X lo5 daltons.
`6 hr, 20°C).
`DNA samples were sonicatcd (Bronson LS-75 sonificr) and mclted at
`100°C for 10 min, then quick cooled in icc watcr. The solutions wcre
`transferred to shell vials mountcd in a floating shell vial holder and placed
`in a constant tempcraturc bath ; 0.1-ml samples were talien at prcdcter-
`mined timcs and placed in 1 ml of icc-cold 0.12 M NsP. Thcsc were stored
`cold until analyzcd for duplex by a hydroxylapatite-centrifuge method
`similar to that of Brenner et a1.17 The rate constant K was determined and
`corrected for length and concentration cff
`Melting Temperature
`Sonicatcd DNA was dialyzed against 0.01 M NaCl, 0.001 M NaHZEDTA,
`0.003 M NaP pH = 7, and placed in quartz cuvettes. Tcmperature was
`raised using a temperature-controllcd cuvettc holder connected to a Haake
`circulating water bath. Absorbancc at 260 nm was measured with a Zeiss
`PMQ I1 spectrophotometcr. Tempcraturc rcgulation was accurate to
`0.2"C (R. Baldwin, personal communication) and DNA standards wcrc
`mclted at the same time. Change in melting temperature AT, was takcn
`to be the temperature diffcrencc betwecn thc midpoints of thc hyper-
`chromicity curve of thc sample and that of the standard.
`
`RESULTS AND DISCUSSION
`The spccificity of the silver-sensitization tcchniyue for dimer formation
`was examined. Acid stable nondimer photoproducts contributcd less than
`0.1% of total label as detcrmincd by column chromat~graphy.'~ The
`contribution of cytosinc dimers (CT) was less than 1% of thc total label,
`under conditions in which 30% of thc label was identified as dimer as-
`sociated. Strand breakage was obscrvcd but only with unsonicated DNA
`and times of irradiation longer than those used in this study. Sonication
`after irradiation eliminated strand breakages as a factor in the actual
`renaturation. No evidence for an incrcascd extent of strand breakage was
`observcd in dimcrized DNA at higher temperatures.
`
`
`Exhibit 2015 Page 3
`
`

`
`672
`
`KAHN
`
`Figure 1 is a plot of total dimers (mostly thymine dimers) versus ATm.
`Regression of the points leads to the empirical formula
`ATm = KIFFJ/T
`where K = 0.6"C per dimer per 100 base pairs ("C/%). Such a figure is
`lower than the value of K = l.Z°C/% obtained by Hutton and Wetmur"
`for glyoxylated DNA and 1.6"C/% obtained by Ullman and McCarthy for
`deaminated DNA.'* A single dimer is thus approximately half as destabi-
`lizing as these other alterations.
`
`Irradiated DNA hydro-
`Fig. 1. Effect of thymine dimers on melting temperature.
`lyzed by heating at 121OC for 3 hr in equal volume of HC10,. Solution then neutralized
`with KOH and solid KC104 removed. Sample applied to column of AG 1-X8 resin
`(formate form, BioRad), which had been pre-equilibrated with 0.02 M NH4OH. Dimers
`eluted with 0.02 M NHIOH, 0.02 M HCOOH pH = 8, monomers with 0.02 M HCOOH.
`Plotted are dimer counts per total counts. AT,,, determined as described in "Methods."
`
`A
`
`I .o
`
`0.8
`
`0.6
`
`A-A,,
`Amax-*o o.4
`
`0.2
`
`Fig. 2. Melting profiles of irradiated DNA. Dimer concentration 29.0% (A), 10% (0),
`6.8% (01, or 0% (A).
`
`
`Exhibit 2015 Page 4
`
`

`
`DNA: DNA HYBRIDIZATION
`
`673
`
`The width of the melting curves indicates a sharp transition (Figure 2).
`This contrasts with the broad melt of DNA irradiated at X = 254 mp but is
`consistent with results obtained by irradiating at X = 310 mp in the pres-
`ence of acetophen~ne.'~ It is probable that the former results are due to a
`variety of photoproducts other than dimers, which have great,er destabiliz
`ing effects than dimers. These photoproducts are absent in the DNA
`used in the present work.
`
`Renaturation Rate
`Sonicated T7 DNA was reannealed as described in "Methods" and the
`Cot curves plotted in Figure 3. The minimum rate observed at 65°C is
`62% of that for unaltered DNA. This change, with DNA containing 29%
`of its thymine label as dimers and having a AT, of 10.2"C indicates that in a
`stability region comparable to reannealed eukaryotic DNA, the rate con-
`stant is not altered by even a factor of 2. The extent of dimerization is not
`extensive enough to test the hypothesis of Hutton and Wetmur" concern-
`ing the precise form of the dependence of rate constant on AT, but is
`consistent with their data in the range explored in this study.
`In order to see how the rate constant varies as the conditions of hybridiza-
`tion were made more stringent, the incubation temperature was raised.
`Results are plotted in Figure 4. The dimer-dependent decrease in renatura-
`tion rate becomes more pronounced as temperature rises.
`
`0
`
`Fig. 3. Reassociation of irradiated DNA. Samples obtained as described in text were
`added to 0.5 g hydroxylapatite (DNA grade, BioRad) suspended in 10 ml of 0.12 ill Nap;
`after equilibration at 65°C samples centrifuged in Sorvall desk-top centrifuge. Single-
`stranded DNA in supernatant precipitated in 5% cold CbCCOOH. Pellet resuspended
`in 0.4 M Nap, centrifuged, and supernatant doublestranded DNA precipitated. Sam-
`ples filtered through Millipore filters, and counted in toluenePPO-POPOP. Plotted is
`ratio SS/(SF + DS). Dimer content 29.0% (A), 22.6% (m), 10.0% (0) and 0% (A).
`
`
`Exhibit 2015 Page 5
`
`

`
`674
`
`KAHN
`
`K
`'K.
`
`ae
`
`0.6
`
`04
`
`0
`
`\
`
`0.3
`
`I
`0.4
`
`0.1
`
`0.2
`elT
`
`(K/Ko) is rate
`Fig. 4. Effect of annealing temperature on rate of reassociation.
`constant at given dimer concentration divided by rate constant of unirradiated DNA.
`Renaturations performed at 65°C (+), 75°C (0), 80°C (A) and 85°C (0). 29% dimer-
`ized sample melts at 84.5"C in this buffer (0.12 Nap) as determined by hydroxyapatite
`binding.
`
`A correction to the renaturation constant is needed but this correction
`is small-3.50Jo/"C
`of AT, at 65°C. Such a factor is not large enough to
`account for the results of Sutton and h/lcCallum.* This study therefore
`supports the hypothesis of Hutton and Wetmur," that the rate difference
`is due to classes of differing complexities and thermal stabilities, rather
`than Southern's random mutation hypothesis.3
`
`The author thanks Dr. Philip Hanawalt for timely advice and encouragement. This
`work was supported by Atomic Energy Commission Grant AT(O43)326-7 to Philip
`Hanawalt and a National Institutes of Health predoctoral traineeship to the author.
`
`References
`Walker, P. M. B. (1969) Progr. Nucleic Acid Res. Mol. Bwl. 9,301326,
`1.
`Wetmur, J. G. & Davidson, N. (1968) J . MoZ. Biol., 31, 349-370.
`2.
`Southern, E. M. (1971) Nature New Bwl., 232,82-83.
`3.
`Sutton, W. D. & McCallum, M. (1971) Nature New Biol., 232,8345.
`4.
`Britten, R. J. & Kohne, D. E. (1968) Science 161,529-540.
`5.
`Brunk, C. F. (1973) Nature New Biol. 241,74-76.
`6.
`7.
`Shafranovkaya, N. N., Trifonov, E. N., Lazurkin, Yu. S. & Frank-Kamenetskii,
`M. D. (1973) Nature New Biol. 241 58-60.
`8. Rahn, R. 0. & Landry, L. C. Photockem. Photobiol. (submitted).
`9. McCarthy, B. J. & Farquhar, M. N. (1972) in Evolution of Genetic Systems, H. H.
`Smith, Ed., Brookhaven, Symp., 23 pp. 1-43.
`10. Bonner, T., Brenner, D. & Britten, R. (1971) Camegie Inst. Washington Yearbook,
`71,287-289.
`11. Hutton, J. R. & Wetmur, J. G. (1973) Biochemistry 12,558-563.
`12. Lee, C. H. & Wetmur, J. G. (1973) Bwchem. Bwphys. Res. Commun. 80, 879-
`885.
`13. England, P. T. (1972) J . MoZ. Biol. 66,209-224.
`14. Thomas, C. A. & Abelson, J. (1966) in Procedures in Nucleic Acid Research, G. L.
`Cantoni & D. R. Davies, Eds., V, 1,553-568.
`
`
`Exhibit 2015 Page 6
`
`

`
`DNA: DNA HYBRIDIZATION
`
`675
`
`15. Lamola, A. A. (1969) Photochem. Photobiol. 9,291-294.
`16. Abelson, J. &Thomas, C. A. (1966) J . MoZ. Biol. 18,262-291.
`17. Brenner, D. J., Fanning, G. R., Rake, S. V. & Johnson, K.E. (1969) Anal. Bio-
`chm. 28,447-459.
`18. Ullman, J. & McCarthy, B. J. (1973) Biochim. Biophys. Acta 294,405-424.
`19. Charlier, M., Helene, C. & Carrier, W. L. (1972) Photochem. Pholobiol. 15, 527-
`536.
`Received September 24,1973
`Revised January 2,1974
`
`
`Exhibit 2015 Page 7