Allosteric Conversion of Z DNA to an Intercalated Right-handed

Document Sample
Allosteric Conversion of Z DNA to an Intercalated Right-handed Powered By Docstoc
					THE JOURNAL OF BIOLOGICAL    CHEMISTRY                                                                Vol . 261,No.   Issue of July 5, pp. 8899-8907 1986
0 1986 by The American Swiety of Biological Chemists, Inc                                                                                Printed in L~.s.A.

Allosteric Conversion of Z DNA to an Intercalated Right-handed
Conformation by Daunomycin*
                                                                                                 (Received for publication, December 9, 1985)

                 Jonathan B. Chaires
                 From the Department of Biochemistry, The University of Mississippi Medical Center, Jackson, Mississippi 39216-4505

  Absorbance and fluorescence methods were used to                        minichromosome (5). Nuclear proteins that interact specifi-
measure the binding the anticancer drug
                       of                      daunomy-                   cally with Z DNA havebeen isolated from a variety of sources
cin to poly(dGdC) under ionic conditions that initially                   (6-8). The precise role of Z DNA in gene expression, however,
favor the left-handed Z conformation of the polymer.                      remains undefined. Recent reviews summarize the current
Drug binding was cooperative under these       conditions                 understanding of Z DNA structure, dynamics, and function
and may be fully accounted for by an allosteric model                     (9-11).
in which the drugbinds preferentially (but exclu-                            A number of small molecules, among them the intercalators,
sively) to the right-handed B conformation and shifts                     may have profound effects on the B to Z transition (12-17).
the polymer from theZ to an intercalated right-handed                     Intercalators, for example, inhibit the rateof B to Z transition
conformation. Quantitative analysis of binding iso-                       (13). In addition, intercalators may act as allosteric effectors
therms in terms of the allosteric model allowed for                       and convert Z DNA to an intercalated right-handed form
estimation of the equilibrium constants for conver-
sion of a base pair at a B-Z interface from the to the
                                                  2                       under solution conditions that would otherwise favor the Z
B conformation and for the formationof a base pair in                     conformation (12, 14, 16,17). Theexact mechanism by which
the B conformation within a stretch of helix in the Z                     intercalators act as allosteric effectors on the B to Z equilib-
conformation. The free energy the Z to B conversion
                                 of                                       rium is poorly defined and is of interest not only in terms of
of a base pair was  calculated from thisdata and ranges                   the molecular mechanism of intercalator action, but in addi-
from +0.03 to +0.3 kcal/mol over the NaCl range of                        tion in terms of the structure, stability, and dynamics of Z
2.4-3.5 M. The free energy for the formation a B-Z
                                                 of                       DNA. Pohl and co-workers (12) first demonstrated striking
junction was nearly constant +4.0 kcal/mol over the
                               at                                         allosteric effects of ethidium on Z form poly(dGdC) and used
same range of NaCl concentrations. The salt depend-                       the allosteric model of Monod et al. (18) to analyze their
ence of the free energy of the Z to B transition indicates                cooperative binding data. While this model is certainly a
preferential Na+ binding to the Z form and that there                     conceptually correct way to account for the observed cooper-
is a net release of Na+ upon conversion of a base pair                    ative interaction of an intercalator with Z DNA, strict appli-
from the Z to the B conformation. The energetically                       cation of this model neglects important aspects of the inter-
unfavorable Z to B transition was found by this anal-                     calator-DNA interaction and of the B to Z equilibrium. In
ysis to be driven by coupling to the energetically fa-                    particular, the phenomenon of neighbor exclusion (19, 20) is
vorable interactionof daunomycin with B form DNA.                         not included in the model used by Pohl et al. (12) to analyze
In 3.5 M NaCl, for example, the free    energy changefor                  their data, nor is the B to Z equilibrium treated in a manner
                     (Z                   +
the overall reaction DNA base pairs) (daunomycin)                         appropriate for the structural transition of a DNA lattice.
a (right-handed complex) is -7.0 kcal/mol, nearly all                     The allosteric model of Dattagupta et al. (21) is an extension
of which is contributed by the binding of drug to B
DNA. Analysis using the allosteric model also shows                       of the basic model of Monod et al. (18) that appropriately
that the number of base pairs converted from theZ to                      treatsthe DNA structuraltransitionand           which includes
the B conformation per bound drug molecule is salt                        neighbor exclusion effects. The model of Dattagupta et al.
dependent and provides evidence that drug molecules                       (21) is applied here to explain and analyze the cooperative
partition into regions of the polymer in the right-                       interaction of the potent anticancer drug daunomycin with Z
handed conformation.                                                      DNA.
                                                                             Previous results from this laboratory showed that dauno-
                                                                          mycin would effectively inhibit the rate of the B to Z transi-
                                                                          tion and would shift the midpoint of the NaC1-induced tran-
  The chemistryand biology of Z DNA, the left-handed                      sition of poly(dGdC) from the B to the Z form to higher salt
helical form of DNA first observed in solution in 1972 byPohl             concentrations (14). Further, under some conditions, the drug
and Jovin (l), of intense current interest, in part because
                 is                                                       would actively convert Z DNA back to an intercalated right-
of the possible role of Z DNA in gene expression. Z DNA                   handed form (14, 17). The allosteric conversion of Z DNA to
exists, or is inducible, in chromosomes from Drosophila (2,3)             the intercalated right-handedform wasdemonstrated directly
and Chironomus (4), and sequences that may adopt the Z                    over a range of ionic conditions by circular dichroism, sedi-
conformation are found in the enhancer regions of the SV40                mentation velocity, and susceptibility to nuclease attack for
                                                                          both poly(dGdC) and poly(dGm'dC) (17). The results de-
  * This work wassupported by United States Public Health Service         scribed here extend these previous findings. The allosteric
Grant CA35635 awarded by the National Cancer Institute, Depart-           model of Dattagupta et al. (21) is applied here to account for
ment of Health and Human Services and by National Science Foun-
dation Grant DMB-8421185. The costs of publication of this article        the cooperative binding of daunomycin to DNA initially in
were defrayed in part by the payment of page charges. This article        the Z conformation and provides considerable insight into the
must therefore be hereby marked "advertisement" in accordance with        mechanism bywhich daunomycin converts Z DNA to an
18 U.S.C. Section 1734 solely to indicate this fact.                      intercalated right-handed form. Further, application of the
8900                                               Allosteric Conversion of Z DNA
model provides a description of the equilibriumof the Z t o B          nz). Six parameters thus define the model. No attempt was made to
transition over a range of NaCl concentrations, permitting             use nonlinear regression to estimate these parameters since the six
the estimation of the free energies of the formation of a €3-Z         independent parameters preclude the usefulness of such an effort
                                                                       (21). The following algorithm was used to estimate values for the
junction and for the conversion of a base pair at an interface         parameters used in the allosteric model. For r values greater than the
from the Z to the B conformation. The salt dependence of               r value corresponding to themaximum in r/C, the polymer is entirely
these transitions is also estimated. The model   thus provides                                                     fit
                                                                       in the right-handed form, and data may be to the      standard neighbor
fundamental infor.mation o n the B t o Z transition in addition        exclusion model as described above to estimate KB and ne. Kz may
to providing a useful qualitative and quantitative theoretical         be estimated from the intercept in the r/C axis, obtained by linear
basis for understandinghow intercalators allosterically affect         regression of low r data extrapolated to theintercept. KB, nB,and Kz
                                                                       may thus be estimated and constrained in the subsequent analysis.
DNA conformation.                                                      From the statistics of the fits by this procedure KB, nB,and Kz are
                                                                       estimated to be accurate to 10-15%. The remaining parameters, nZ,
                     MATERIALS ANDMETHODS                              s, and u were estimated by iterative adjustment.The value chosen for
   Poly(dGdC)-Poly(dGdC) (lot 317910)was purchased from P-L n~ had little effect on            calculated isotherms, arising from the fact that
Biochemicals. The sample was dissolved in a buffer containing 6m    M there is little binding of the drug to the Z form. A value of nz = 2.0
Na2HP04,2 m NaH2P04, 1 m disodium EDTA, 0.185 M NaCl, was arbitrarily chosen, but increasing nz to 3.0 had no discernible
               M                  M
pH 7.0. The dissolved sample was fractionated over Sepharose 4B, effect on the subsequent estimation of s and U . The parameters s and
and thecentral fractionswere pooledand characterized as previously u are crucial in describing the low r region of the binding isotherm
described (22). The sample was further characterized by sedimenta- with positive slope. The parameters s and KB/KZ determine the r
tion velocity at 36,000 rpm in a Spinco model E analytical ultracen- value at the midpoint of this rising portion of the isotherm, and u
trifuge equipped wi a photoelectric scanner system and electronic determines the steepness of the intraconversion of Z form to B form
speed control. The G ) method developed by van Holde was used to DNA. Values for s were found to be constrained to a rather narrow
characterize the polydispersity of the sample (23). The sedimentation  range in order to describe the experimental data. The nucleation
rate at the midpoint of the boundary was 9.64 f 0.2 S; 90% of the parameter, u, was found to have a larger latitude. For example, data
boundary sedimented within their range 9-10.4 S. These sedimenta- obtained a t 3.0 M NaCl could be described equally well by (a = 0.01,
tion rates correspond to an average size of 1300 f 300 bp for the s = 0.75) or (u = 0.001, s = 0.73) with the remaining values fixed.
sample (24). An extinction coefficient of 16,800 M” (bp’) cm” at 255 Our estimates for u thus are probably good to anorder of magnitude,
nm was used for concentration determination (25).                      but values for s are defined to a much narrower range, probably within
   Measurement of the B to 2 Transition-The B to Z transition was 10%. For physical reasons, to be discussed later, u = 0.001 seems to
measured using an individual sample for each NaCl concentration. A be the most reasonable choice.
1-ml sample (35.7 PM bp) was brought to the desired NaCl concen-          Care was taken to test the validity of estimated values for the
tration by the addition of solid NaCl and then dialyzed against a parameters for the allosteric model by additional experiments in lieu
buffered NaCl solution of the desired concentration for a t least 24 h of rigorous statistical analysis. Thus, independent experiments were
at room temperature. Followingdialysis, the complete UV absorbance conducted that yielded an estimate for KB/Kz,a prediction for the
spectrum and the UV derivative absorbance spectrum were recorded salt concentration when s = 1.0, and the binding ratio required for
for each sample a t ambient temperature using a Cary 219 spectro- the complete conversion of the polymer from the Z to theB form. All
photometer. The ratio of absorbance at 260 nm to that at 295 nm        of these were in excellent quantitative agreement with calculations
was used as a measure of the B to Z transition (1).                    using the allosteric model with the parameters derived by the methods
   Binding Studies-Daunomycin (Sigma) binding to poly(dGdC) was described above. Without question, the parameters settled upon rep-
measured using visible absorbance and fluorescence methods previ- resent one physically meaningful set.
ously described in detail (14, 17, 22, 26). Poly(dGdC) samples for        CompetitionDialysis-Competition        dialysis measurements used
binding studies were brought to the desired NaCl concentrations by the procedure of Mueller and Crothers (27). Samples of poly(dGdC)
the addition of solid NaCl, followed by dialysis for 24-48 h against and calf thymus DNA were equilibrated to 3.5 M NaCl by dialysis
buffered NaCl solutions of the desired concentrations. Titration and brought to an identical concentration of 80 pM (bp) by dilution
experiments required 45-60-min equilibration times following the with dialysate. Under these conditions, the poly(dGdC) adopts the Z
addition of drug because of the slow equilibration of the Z to B conformation, while the calf thymus DNA remains in the right-
equilibrium. Following the addition of drug to poly(dGdC) in the Z handed conformation. Samples of poly(dGdC) and of calf thymus
form, the absorbance or fluorescence was continuously monitored at DNA of identical volume and concentration were then dialyzed
an appropriate fixed wavelength to ensure complete equilibration of against a common solution containing daunomycin and 3.5 M NaCl
the binding reaction.                                                  until dialysis equilibrium was reached. The amount of free dauno-
   Data Analysis-Binding data were cast into the   form of a Scatchard mycin in the dialysate and the amount of total daunomycin in the
plot of r versus r/C, where r is the ratio of bound drug to the total respective polymer samples were then determined by fluorescence
base pair concentration and C is the concentration of free drug. For intensity measurements (Aex = 480 nm; Lm= 555 nm), with reference
concave binding isotherms, data beyond the r value corresponding to to a standard curve of daunomycin alone, following the addition of
the maximum in r/C were fit to theneighbor exclusion model             dimethyl sulfoxide to a concentration of 50% (v/v) to dissociate the
                                                                       drug-DNA complex (28).
            r/C = K(l - nr)((l - nr)/(l - (n - l)r))(”-l)
                                                                                             RESULTS AND ANALYSIS~
where K is the binding constant to an isolated site and n is the
neighbor exclusion parameter (19, 20). Anonlinearleast squares             Cooperative Binding of Daunomycin to Z Form
fitting routinebased on the Marquart-Levenberg algorithm available       poly(dGdC)-Binding    isotherms forthe interaction of dauno-
through the National Institutes of Health PROPHET Computer               mycin with poly(dGdC) under ionic conditions that initially
Resource was used to fit the data.                                       favortheleft-handed     Z conformation of the polymer are
   Binding isotherms were “fit”to theallosteric model of Dattagupta
et ~ l(21) by an iterative process using a computer program developed
       .                                                                 shown in Fig. 1. In all cases shown, the plots are concave-
by Prof. Donald Crothers and kindly made available by him. Briefly,      down, indicative of a cooperative binding process. The allo-
the model postulates that DNA is in equilibrium between two confor-      steric binding model of Dattagupta et al. (21) may be used to
mational states, which for the case a t hand may be unambiguously
assigned to the B and Z conformations. An equilibrium constant, s,           Portions of this paper (including part of ‘Results” and Figs. 1M-
is assigned for the conversion of a base pair a t a B-Z interface from   4M) arepresented in miniprint a t the end of this paper. Miniprint is
the Z to the conformation. An equilibrium constant, g s , is assigned
             B                                                           easily read with the aid of a standard magnifying glass. Full size
for the formation of a base pair in the B conformation within a          photocopies are available from the Journal of Biological Chemistry,
stretch of helix in the Z conformation. Drug molecules may bind to       9650 Rockville Pike, Bethesda, MD 20814. Request Document No.
either the B or the Z form with characteristic binding constants (KB     85M-3992, cite the authors, and include a check or money order for
and Kz, respectively) and characteristic neighbor exclusion (ng and      $2.80 per set of photocopies. Full size photocopies are also included
                                                                         in the microfilm edition of the Journal that available from Waverly
    The abbreviation used is: bp, base pair(s).                          Press.
                                                        Allosteric Conversion of Z DNA                                              8901
analyze the binding data shown in Fig. 1. Table I shows the             = -0.91, indicates that sodium ions are released as base pairs
results of the analysis of the data in Fig. 1 in terms of the           are converted from the Z to theB conformation.
allosteric model. Calculated binding isotherms are indicated               The data of Fig. 2 may be extrapolated to the point where
in Fig. 1 by the solid lines. Excellent agreement between the           s = l(ln s = 0) to obtain a corresponding sodium activity that
experimental data and the allosteric model may be obtained              is equivalent to [NaCl] = 2.3 M. This should correspond to
with the parameters in Table I.                                         the midpoint of the B to Z transition curve. Fig. 3 shows that
   Does Daunomycin Associate with Z DNA?-The methods                    this is in factthe case. The same poly(dGdC) sample used for
used to obtain binding data assume that the    observed changes         the binding studies was used to obtain the dataof Fig. 3, and
in the optical properties of daunomycin arise primarily from            the resultant midpoint is in excellent agreement with that
its intercalation into B form DNA. In light of recent results           predicted by the extrapolation of the data of Fig. 2, as indi-
that conclude that ethidium may intercalate intoZ form DNA              cated by the solid symbol. This provides further independent
with optical changes similar to those observed for its inter-           verification of the reliability to theparameters estimated for
calation into B form DNA (29), this assumption was critically           the allosteric model.
evaluated in a series of experiments that are described fully              Sult Dependence of Ks and KZ-The salt dependence of the
in the appended Miniprint. These experiments validate the               intrinsic binding constants for the interaction of daunomycin
assumption that theoptical changes observed in our binding              with B and Z form DNA is shown in Fig. 4 The salt depend-
studies arise primarily from intercalation of the drug into B           ence of daunomycin binding to Z form DNA is seen to be
form DNA and indicate that theweak binding of the drug to               greater than that seen for the interaction of the drug with B
Z form DNA is accompanied by little change in the    absorbance         form DNA.For the interaction of the drug with B form DNA,
and fluorescence properties of the drug.                                the slope of the line in Fig. 4 is -0.67. The slope of the line
   The relative affinity of daunomycin for B and Z form DNA             for the interaction of the drug with 2 form DNA is nearly
was tested directly by the method of competition dialysis (27).         twice that value, -1.5. The differences in the slope must
The results of the experiment are shown in Table 11. Dauno-             reflect differences in the number of counterions associated
mycin binds preferentially to the right-handed DNA, as is               per phosphates between the B and Z conformations, as will
evident from the higher total drug concentration that parti-            be discussed in detail in a later section, suggesting greater ion
tions into the calf thymus DNA samples. The ratio of bound              binding to theZ conformation.
daunomycin to the right- and left-handed DNA samples is                    The Number of Base Pairs Converted from the Z to the B
calculated to be 40-50. In the limit as r approaches zero, that         Conformation per Bound Drug Molecule-The allosteric
ratio represents the ratio of the binding affinity of the drug          model enables one to calculate the fraction of the polymer
to right-and left-handed DNA (27). The value of 40-50                   converted from the Z to the B conformation for each bound
obtained in these experimentsis in excellent agreement with             drug molecule. Fig. 5 shows the results of these calculations
that calculated from the binding experiments in Fig. 1 (Table           using the parameters shown in Table I. Circular dichroism
I) and provides independent verification of the results ob-             experiments (described in the Miniprint) at 2.4 and 3.5 M
tained from the fits to the  allosteric model.                          NaCl provided an independent approach to estimate the frac-
   Salt Dependence of the B to Z Transition-The data of                 tion of helix in theright-handed form as a function of bound
Table I provides fundamental informationon the saltdepend-              daunomycin. The agreement with the predictions from the
ence of the Z to B transition. Fig. 2 shows the saltdependence          allosteric model is excellent, lending confidence to the validity
of the equilibrium constant for the Z to B transition as a                                               in
                                                                        of those estimated parameters Table I. Total conversion to
double logarithmic plot. The slope of the line, (d In s/d In a)         the (intercalated) B form occurs at rcrit, which corresponds to


                                                                                               ;     6.0

   FIG. 1. Interaction of daunomycin
with poly(dGdC) underionic condi-                                                                     2.0
tions that initially favor the Z con-
formation. Binding isotherms  are
shown for the binding of daunomycin to
poly(dGdC) in solutions containing 3.5
M ( A ) ,3.0 M (B),2.8 M (c), 2.4 M (D)
NaCl. Binding data were obtained by
absorbance and fluorescence methods.
The solid lines are calculated isotherms
using the allosteric model described in                                                              18.0
the text and the parameters listed in
Table I. In panel D,the solid line is the                                                            14.0
                                                 6.0                                           ?
least-squares fit of the data beyond r =                                                        P
0.05 to the neighbor exclusion model,                                                           la   10.0
while the dashed line is the isotherm
                                                 4.0                                           c
calculated using the allosteric model.                                                                8.0

                                                                                                            0   0.1   0.2     0.3     0.4

       8902                                                               Allosteric Conversion of Z DNA
                                      TABLE   I
           Summary of the parametersdescribing the allosteric binding of
        daunomycin to poly(dGdC) under ianie conditions favorable to the 2
          The parameters shown here were used to calculate the binding
       isotherms shown in Fig. 1. KB is the binding constant for the inter-
       action of a drugmolecule with an isolated site in the B conformation;
       n is the corresponding neighbor exclusion parameter. K z is the binding
       constant for the interaction of a drug molecule with an isolated site                                          -0     1 0 2.0 3.0 4.0
       in the Z conformation. The corresponding neighbor exclusion param-
       eter, nZ, was set to 2.0 for all salt concentrations. The parameters                                                      NaC1, M
       describing the Z to B conformational transition are s, the equilibrium
       constant for the conversion of a base pair at a pre-existing B-Z                          FIG. 3. Salt-induced transition of poly(dGdC). The ratio of
       interface from the 2 to the B conformation, and u, the nucleation                      absorbance at 260 nm to that at 295 nm is shown as a function of
       parameter for the conversion of a base pair within a stretch of Z helix                NaCl concentration. Poly(dGdC) at a concentration of 35.7 p~ (bp)
       to the B Conformation.                                                                 was dialyzed at ambient temperature against a buffer containing 6
               NaCl          KB        n        Kz     KdKz      s       0
                                                                                                M                M
                                                                                              m Na2HP04,2 m NaH2P04,1 m Na2EDTA, pH 7.0, and NaCl
                                                                                              as indicated. Complete UV spectra and UV derivative spectra were
               M              XIO-' M"       bp     XIO"   M"                                 recorded for each sample at 25 "C following dialysis. The symbols are
         Daunomycin                                                                           the average of three determinations, and theerror bars represent the
              2.4                  3.2      2.8       1.8           17.5       0.95   0.001   range of the threereplicate determinations.
              2.8                  2.45     2.5       0.98          25.0       0.81   0.001
              3.0                  2.4      2.0       0.8           30.0       0.73   0.001
              3.5                  2.25     2.0       0.5           45.0       0.59   0.001
               2.8                 8.8      2.0       1.06          88.0       0.85 0.001

                                     TABLE  I1
                    Summary of competition dialysis experiment
                                                 M               M
          Solution conditions were BP buffer (6 m Na2HP04,2 m H2P04,
       pH 7.0) 3.5 M NaC1. Total polymer concentration was 80 FM (bp).
       Samples were dialyzed for 48 h. Dialysis equilibrium was complete as
       judged by control samples in which calf thymus DNA solutions were
       dialyzed against identicalcalf thymus DNA solutions. The totaldrug
       concentration within the dialysis sack was, in these cases, identical
       to within +5%.
                                            Bound daunomycin
                                                                                                                       0.4     0.6     0.8     1.0
                                                                                                                                     I at
                                                                                                 FIG. 4. Ionic strengthdependence of KBand Kz. the nat-  A,
                                                                                              ural logarithm of K g , the binding constant for the interaction of
                   PM                                PM
                                                                                              daunomycin with an isolated base pair in the B conformation, is
                 0.08                     1.78 40.5              0.044                        plotted as a function of the logarithm of the mean molal activity of
3.38             0.12                          52.8              0.064                        sodium, following Record et a . (36). The line is the linearleast
                                                                                              squares fit to the data, yielding a slope of -0.67 and a correlation
                                                                                              coefficient of 0.903. B, the natural logarithm of Kz, the binding
                                                                                              constant for the interaction of daunomycin with an isolated base pair
                                                                                              in the Z conformation, as a function of the natural logarithm of the
                                                                                              mean molal activity of sodium. The line is the linear least-squares fit
                               0                                                              to the data, yielding a slope of -1.5 with a correlation coefficient of

                            -0.2                                                              the r value where the maximum in r/C occurs. The number
                                                                                              of base pairs converted to theB conformation/drug molecule
                                                                                              is 1/rC+ Table I11 summarizes the saltdependence of the rcrit,
                            -0.4                                                              and it is reciprocal. The data further emphasize the long range
                                                                                              allosteric effects exerted by daunomycin on theDNA confor-
                                                                                              mation. Nearly 19 bp are converted from the Z to the B
                                                                                              conformation by the binding of a single drug molecule in 2.4
                                    0.4       0.8          0.8           1.0                  M NaC1, corresponding to nearly two turns of the helix. These
                                            In Q
                                                                                              effects are well beyond the immediate binding site of the drug,
                                                                                              which spans 3-4 bp.
          FIG. 2. Ionic strengthdependence of the Z to B equilibrium                             Partitioning of Drug Molecules into Right-handed Regwm-
       constant (e). The natural logarithm of s, the equilibrium constant
       for the conversion of a base pair at a jundion from the 2 to the B                     The allosteric model may also be used to calculate the distri-
       conformation, is plottedas a function of the naturallogarithm of the                   bution of bound drug between the regions of the polymer in
       mean molal sodium activity. The open symbols are data taken from                       the B and Z conformations over the course of the titration.
       Table I. The solid line is the linear least-squares fit to the data,                   Fig. 6 shows the results of these calculations that provide
       yielding a slope of -0.91 (correlation coefficient = 0.9983). The solid                evidence that drug molecules partition into the regions of the
       symbol is the experimentally determined midpoint of the B to Z                         polymer in the right-handed conformation. The plot shows
       transition (Fig. 3), at which point Ins = 0. The dashed line is the
       extrapolation of the fitted line and indicates that the midpoint is                    the binding ratio of drug within the regions of the polymer in
       correctly predicted by the fit to the data,
                                                 independently verifying the                  the right-handed conformation as a function of the overall
       estimates for s i Table I.
                        n                                                                     binding ratio. The r value of drug in regions of the polymer
                                                    Allosteric Conversion of Z DNA                                         8903
                   I          I           I                          in the right-handed conformation depends upon the total
                                                                     NaCl concentration andis nearly constant over a wide range
                                                                     of total bound drug. In 3.5 M NaC1, the binding ratio is near
                                                                     0.3 mol of drug/molofbp,which         correspondsto 1 drug
                                                                     molecule/3 bp in these regions, near the saturation level for
                                                                     the drug. This indicates that most the drug is bound in the
                                                                     right-handed regions. A similar phenomenon occurs at the
                                                                     lower saltconcentrations,but     witha lower bindingratio
                                                                     within the regions in the right-handed conformation, arising
                             0.1         0.2       0.3               from the fact that each bound drug molecule converts a longer
                                                                     stretch of helix to the right-handed conformation at these
                                 r bound                             ionic strengths.
     FIG. 5. Fraction of polymer in the €3 form as a function of        Comparison with Adriamycin-The interaction of adria-
bound daunomycin. The allosteric model was usedto calculate the mycin with poly(dGdC) in 2.8 M NaC1 was studied by the
fraction of the polymer in the right-handed B form as a function of same methods as described above for daunomycin. Adriamy-
the bindingratio r (= mol of daunomycin bound/molof bp) using the cin bindscooperatively to thepolymer under these conditions
parameters in Table I. The lines refer, from left to right, to 2.4 M but converts poly(dGdC) to the right-handed   form more effi-
(-),     2.8 M (---), 3.0 M (-- - -), and 3.5 M (-----) NaC1. The ciently than daunomycin (data not shown).       Adriamycin con-
symbols referto experimentally determined values 2.4 M NaCl (0)
and 3.5 M NaCl (m) obtained by circular dichroism as described in verts 7.5 bp from the to theB conformation per bound drug
the Miniprint.                                                       molecule, compared to 6.3 for daunomycin (Table 111). Anal-
                                                                     ysis of the adriamycin-poly(dGdC) interaction in terms of the
                                                                     allosteric model yields the parameters shown in the last line
                                                                     inTable I. Thesedataindicatethatadriamycin            is more
                             TABLE   111
                                                                     efficient than daunomycin solely because of its higher affinity
                                                                     for B form DNA.
 Number of base pairs conuerted from the 2 to the B conformation/
                       bound daunomycin molecule
   Values for rest, the binding ratio at which the polymer is converted
entirely to the right-handed conformation, were calculated usingthe      Daunomycin binds   cooperatively to poly(dGdC) under     ionic
allosteric model as showninFig. 5. The reciprocal, l/rcfiL, the  is   conditions thatfavor the left-handed 2 form of the polymer.
number of base pairs converted per bound      drug molecule from the ZPrevious physical studies from this laboratory demonstrated
 to the B conformation. DM, daunomycin.
 -                                                                    that daunomycin binding under these conditions converts the
                                            rhl           vrctit      polymer from the left-handed form to an intercalated right-
                            Nacl          (DM/bp)       (bp/DM)
                                                                      handed form (17). Striking long-range allosteric effects are
                                                                      apparent under certain conditions, with as little as 1 bound
        Daunomycin            18.5
                             2.4           0.054                      drug molecule/20 bp driving the left to right conversion to
                            6.3            0.159
4.3           0.233          3.0                                      completion (17). The allosteric  model of Dattagupta et al. (21)
                           3.3             0.303                                                       to
                                                                      is used in the present report analyze the cooperative inter-
                                                                      action of daunomycin with Z DNA. The model may be used
        Adriamycin           2.8           0.133            7.5       to calculate binding isotherms that are in excellent agreement
                                                                      with the experimental  results. The analysisprovides estimates
                                                                      for the equilibrium constants for the Z to B conversion over
                                                                      a range of ionic strengths in addition to estimates binding
                                                                      constants and neighbor exclusion parameters describing the
                                                                      interaction of daunomycin with B and Z form DNA. This
                                                                      analysis provides fundamental information on the      mechanism
                                                                      by which ligands affect the stabilityof Z DNA and the     origins
                                                                      of the long-range allosteric effectspreviouslyreported.        In
                                                                      addition, fundamental information on the Z to B transition
                                                                      is obtained by the analysis,over a range of Na' concentration
                                                                      where more direct physical measurements would be hampered
                                                                      by the fact that the   B to Z equilibrium isshifted largely
                                                                      toward the Z conformation, and the    average physical proper-
                                                                      ties would reflect mostly the Z conformation.
                                                                         FreeEnergy of the Z to B Transition-Analysis of the
                                                                      binding data shown in Fig. 1 using the allosteric model pro-
                                0.1         0.2       0.3             vides estimates for the equilibrium constant (s) for the con-
                                 r bound (TOW)                        version of a base pair at a pre-existing interface from the Z
    FIG. 6. Partitioning of bound daunomycin molecules. The
                                                                      to the B conformation and for the equilibrium constant for
 binding ratio of daunomycinwithinregionsin          the right-handed the formation of a base pair in the B conformation with a
 conformation as a function of the overall binding ratio is shown. stretch of base pairs in the   Z conformation ( 2 s ) . These values
 Curves were calculated usingthe allosteric model and the parameters are listed in Table I and may be used t o calculate the free
 in Table I for 3.5 M (-1,   3.0 M (- - -), and 2.8 M (- - - -) NaCI. energy for the transition of a base pair from the Z to the B
 Over a large range of total bound drug, the binding ratio within the conformation (AGO = RTln(s))and for the nucleationprocess
 right-handed regions is constant, indicating that the drug molecules
 are clustered. In all cases, binding ratio of drug within regions in (AGO = -RZ'ln(u2s)). T h e calculated values are listedin Table
 the Z conformation is well over an order of magnitude lower than the IV. Over the NaCl concentrations studied, nucleation, with
 binding ratio of drug withinthe right-handed regions.                the concomitant formationof two B-Z junctions, is energeti-
8904                                             Allosteric Conversion of Z DNA
                              TABLE   IV                                binding of the drug to B form. The difference between the
                  Free energy of the Z to B transition                  free energy of drug binding to the Z and the B form is -2.3
   The free energy for conversion of a basepair at aB-Z interface
                       the                                              kcal/mol and is the coupling free energy of the system. The
from the 2 to the B conformationwas calculatedfrom the equilibrium      negative sign indicates a favorable coupling free energy and
constant s shown in Table I according to the equation A G h =           indicates that the conversion of a base pair to the B confor-
“‘Ins. The free energy for nucleation, the formation of a base pair
in theB conformation within stretch of helix in the 2 conformation,
                                 a                                      mation facilitates drug binding.
is AGOnuelsstion -RTln$s. Two B-2 interfaces are created per nuclea-
               =                                                           Salt Dependence of the Z to B Tramition-The salt de-
tion event.                                                             pendence of the equilibrium constant for the conversion of a
                                                                        base pair from the Z to the B conformation is shown in Fig.
                                                                        2. The salt dependence arises from the linkage between so-
            M                                 kcallmol
           2.4             f0.03                +8.2                    dium binding and the Z to B transition. The negative slope
           2.8             +0.12                +8.3                    seen in Fig. 2 indicates that the Z conformation is stabilized
           3.0             +0.19                +8.4                                                                is
                                                                        by increased sodium activity and that there a net release of
           3.5             +0.31                +8.5                    sodium ion upon conversion of the polymer to the B form.
                                                                        The saltdependence of the equilibrium constant s is governed
                                                                        by the equation
                             P0B. POD
                                                                                           (d In s/d In a,) = 2 A r Z
                                                                        where AI’& is the stoichiometrically weighted difference in
                                                                        preferential interaction parameters of the product and reac-
                                                                        tant macromolecular components (35). For the B to Z tran-
                                                                        sition, A G 2probably contains contributions  from electrostatic
                                                       = -7.3           and hydration effects, and itis, therefore, premature to assign
                                                                        detailed molecular meaning to the value of -0.91 found for
                                                                        the slope of the line in Fig. 2. The negative slope indicates
                                                                        unambiguously preferential sodium binding to Z form DNA.
                                                                           Further evidence for greater sodium ion binding to Z form
                                                                        DNA comes from the saltdependence of daunomycin binding
   FIG. 7. Free energy diagram for the interaction of dauno-            to Z and B form DNA. Record and co-workers (36) have
mycin with B and Z form DNA. The free energies for the inter-           shown that for the interaction of a charged ligand with DNA,
action of daunomycin with B and Z form DNA are shown schemati-
cally following the suggestion Weber (34). The chemical potential
                               of                                                         (dlnK/dln a*) = -23     +k
of a base pair in the Z or B conformation is denoted as pz and p ~ ,
respectively. The chemical potential the freedrug is denoted as p ~ .
                                     of                                 where Z is the charge on the ligand, \k is the fraction of
The complexesofdaunomycinwith           Z and B DNA basepairare         counterions associated with each phosphate, and k is the net
denoted p z and ~ B D .In this figure, the values for the equilibrium
                                                                        amount of ion release from the ligand upon complex formation
constants for the various steps were taken from the 3.5 M NaCl data
in Table I and used to calculate the freeenergyaccording to the         (36). In Fig. 3, the salt dependence of Kz is seen to be greater
standard relationAG = -RTlnK.                                           than the salt dependence of K g , with slopes of -1.5 and -0.67,
                                                                        respectively. Since the charge on daunomycin is expected to
cally unfavorable, with a positive free energy of about 8 kcal/         be constant at +1 and there is no evidence for an appreciable
mol. The free energy for the formation of a single B-Z junction         contribution from k, this difference must arise from differ-
is thus about +4 kcal/mol. The conversion of a base pair from           ences in \k. Therefore, Z DNA must have a greater fraction
the Z to the B conformation is characterized by a small                 of counterions associated per phosphate,       or     significant
positive free energy. The signs and magnitudes of these free            changes in the hydration of Z DNA occur upon ligand binding.
energies are similar to those found for the B to Z transition             Electrostatic calculations by Matthew and Richards (37)
of poly(dGdC) inserts driven by supercoiling in low ionic               show that Z DNA indeed binds more sodium ion and that Z
strength (30-32). The dataof Table IV are ingood agreement              DNA may contain specific sites for sodium binding in addition
with the free energy estimates made by Pohl(33) for the B to            to the more general electrostatic condensed ion-binding mode.
Z transition using alternating GC oligonucleotides of defined           A recent theoretical study by Soumpasis (38) predicts a salt-
length.                                                                 dependent free energy for the B to Z transition that is in good
    Energetics of the Allosteric Conversion of Z DNAto the              agreement with that observed in Fig. 2.
Intercalated Right-handed Form-Under the ionic conditions                  Origins of the Long-range Allosteric Effects-The extreme
used in these studies, the conversion of poly(dGdC) from the            cooperativity of the B to Z transition and the        long-range
Z to the B conformation is energetically unfavorable. The               allosteric effects produced by daunomycin binding may be
driving force for the conformational transition in the  polymer         understood as resulting from “incompatible” helix structures,
i s provided by the coupling of the transition to theenergeti-          as proposed as a general mechanism for the transmission of
cally favorable binding of daunomycin to the B form of the              conformational changes along the DNA helix by Crothers and
polymer. This is shown schematically inthe free energy                  Fried (39). The free energy differences between the 2 and the
diagram (34) in Fig.   7. The free energy for daunomycin                B form of a base pair is slight (Table IV), but the formation
binding to the B and Z forms is calculated from the binding             of a boundary between the two forms is energetically unfa-
constants in Table I using thestandard relation AG =                    vorable, with a free energy of over +4.0 kcal/mol. Because of
-RTlnK. The free energy for the overall reaction                        the high free energy of the boundary, intermediates in which
                                                                        the two helical forms are intermixed will be avoided, and the
(Z DNA base pair)     + daunomycin e (daunomycin                        transition between the two helical conformations willbe
                                                                        sharp. Binding of a ligand to one of the helical forms, such as
                                            - B DNA base pair)          the binding of daunomycin to the    right-handed form, will lock
is a favorable -7.0 kcal/mol, most of which comes from the              that region of the helix into itsparticular conformation. This
                                              Allosteric Conversion of Z DNA                                                        8905
altered structure will be propagated into adjacent regions of      excess of Z DNA are nearly identical to those of the free drug.
the helix in order to avoid the energetically unfavorable          Since intercalationresultsina        pronounced bathochromic
boundary between the two helical forms. Ligands will tend to       shift in the absorbance spectrum of the drug and in nearly
partition into the regions containing itspreferred binding site,   total quenching of the drug’s intrinsic fluorescence, the ab-
a phenomenon indicated by Fig. 6.                                  sence of pronounced optical changes suggest nonintercalative
   Comparisonwith Other Zntercalators-Walker et al. (16)           binding under these conditions. Fluorescence quenching ex-
have used the allosteric model to analyze binding data for the     periments show that in the presence of excess Z form DNA,
cooperative interaction of ethidium and actinomycin and ac-        daunomycin is freely accessible to added quencher, again
tinomone (16) with Z form poly(dGdC) in 4.4 M NaCl. Their          suggestive of a nonintercalated complex. Second, the meas-
results may be compared with those obtained in this study.         ured binding constant for the interaction of daunomycin with
The allosteric model provided an excellent fit to their exper-     Z form DNA is lower in magnitude than what would be
imental data. Ethidiumwas found to convert 3-4 bp from the         expected for the formation of an intercalated complex. The
Z to the B form per bound molecule, whereas actinomycin            values for Kz in Table I are of comparable magnitude to the
converted 5 bp. Under low salt conditions, in which 40 ~ L M       equilibrium constant found for the first stepalong the reaction
hexamine cobalt was used to convert the polymer to the Z           pathway leading to the  formation of a right-hand  intercalation
form, both ethidium and actinomycin converted 25 bp per            complex (43). Since this step  was interpreted asthe formation
bound drug molecule to the right-handed form. These values         of an “outside” bound form of the drug prior to the actual
are of comparable magnitude to those reported here for dau-        intercalation event, a similar type of complex is proposed here
nomycin. The ratio of KB/Kz was found to be 300 for ethidium       for the daunomycin-Z DNA interaction. The daunomycin-Z
and 1000 for actinomycin. These appearlarger than the    values    DNA complex, then, is suggested to be a nonintercalated,
found here for daunomycin and adriamycin but refer to a            perhaps largely electrostatic, “outside” bound complex. Such
higher sodium concentration. Extrapolation of the data of                            be
                                                                   a complex would freely accessible to added quencher, would
Fig. 4 to 4.4 M NaCl predicts a KB/Kz ratio for daunomycin         be mobile, and would be characterized by a low binding free
of comparable magnitude tothat reported for ethidium.              energy, all of which are in accord with the experimental
Walker et al. (16) found that actinomone would not convert         evidence reported here. A recent NMR study concludes that
poly(dGdC) from the Z to theright-hand form in 4.4 M NaCl          daunomycin binds exclusively to B form DNA (44). The data
but would convert 11bp per bound drug molecule to the     right-   presented here show that daunomycin strongly prefers B form
handed form under low salt conditions (16). This result is         DNA but will form a weak complex with Z DNA.
entirely consistent with studies on the B anomer of adria-            Origin of Preferential Daunomycin Binding to B Form
mycin (40), which show the same type of behavior.                  DNA-Daunomycin binds preferentially to right-handed
   Walker et al. (16) chose a value of a higher than used here     DNA, as reflected by the KB/Kz ratios in Table I. The struc-
to model their data. At 4.4 M NaC1, a = 0.013 and s = 0.56         turalbasis for this preference maybe inferred from the
were used to generate an isotherm that matched their ethid-        crystallographic data of Quigley et al. (45), in which the
ium-binding data compared to a = 0.001, s = 0.59 used here         strxture of a daunomycin-DNA complex was determined to
to match data at 3.5 M NaCl. These two estimates for a are         atomic resolution. Daunomycin was found to intercalate with
probably within the tolerance of the iterative method used to      the long axis of its anthraquinone ring system at right angles
model the binding data, as discussed previously. There are,        to the long axis of the DNA base pairs. The daunosamine
however, significant physical reasons for believing that a is      moiety and the acetyl group on the daunomycin A ring ex-
of the order of       First the free energy for the formation of   tended in opposite directions to occupy the minor groove,
two B-Z junctions is A O = -RTln($s). With a = 0.001,
                        GJ                                         providing anatural fit into the right-hand helix. Such a
is 4.25 kcal mol”/junction at 3.5 M NaCl, a value that agrees      favorable geometric fit would not bepossible in the left-
with previous estimates of 3-5 kcal mol” (30, 31) and 5 kcal       handed Z DNA.
mol (32). The dataof Walker et al. (16) suggest a value of AGO
                                                                      Acknowledgment-I thank Prof. Donald Crothers for providing a
= +2.7/junction, an estimate lower than those previously
                                                                   Fortran program for calculation of binding isotherms based on the
reported. Second, a may be used to estimate thelength of the       allosteric model and for his interest and insight into the problem.
cooperative unit, No, at the midpoint of the B-Z transition
according to the relation No = 1 + a+. For a = 0.001, No =                                    REFERENCES
33 bp, while for a = 0.013, No= 10 bp. The cooperative length       1. Pohl, F. M., and Jovin, T . M. (1972) J. Mol. Bwl. 67,375-396
of the B-Z transition was experimentally estimated to be -25        2. Nordheim, A., Pardue, M. L., Lafer, E. M., Moller, A., Stollar, B.
bp, corresponding to a = 0.0017 (41). Thus, a =        certainly         D., and Rich, A. (1981) Nature 294,417-422
represents a physically reasonable estimate consistent with         3. Lancillotti, F., Lopez, M. C., Alonso, C., and Stollar, D. (1985) J.
several previous experimental observations.                              Cell Bwl. 100,1759-1766
                                                                    4. Robert-Nicoud, M., Arndt-Jovin, D. J., Zarling, D. A., and Jovin,
   Nature of the Daunomycin-Z DNA Complex-The possibil-                  T. M. (1984) EMBO J. 3.721-731
ity of the intercalation of drugs into the Z DNA helix was          5. Nordheim, A., and Rich, A.-(1983)Proc. Natl. Acad. Sci. U. S. A .
suggested on the basis of a stereochemical analysis by Gupta             80,1821-1825
et al. (42). Shaferand co-workers (29) subsequently proposed        6. Nordheim, A., Teaser, P., Azorin, F., Kwon, Y. H., Moller, A.,
that ethidium in fact intercalates inform poly(dGdC), based
                                          Z                              and Rich, A. (1982) Proc. Natl. Acad. Sci. U.S. A . 79, 7729-
on experiments that indicated changes in the optical proper-        7. Azorin, F., and Rich, A. (1985) Cell 41,365-374
ties of ethidium in the presence of excess Z DNA that were          8. Lafer, E. M., Sousa, R., Rosen, B., Hsu, A., and Rich, A. (1985)
similar to those observed upon its intercalation into right-             Biochemistry 24,5070-5076
handed DNA. The proposal that ethidium intercalates intoZ           9. Rich, A., Nordheim, A., and Wang, A.H-J. (1984) Annu. Rev.
DNA is contradicted by the early work of Pohl et al. (12) and            Biochem. 53,791-846
the more recent work of Walker et al. (16). From the results       10. Leng, M. (1985) Biochirn. Biophys. Acta 826,339-344
                                                                   11. Wells, R. D.(1985) in Progress in Clinical and Biological Research,
of the studies described here, the formation of a daunomycin-            Molecular Basis of Cancer (Rein, R., ed) Vol. 172A, pp. 47-53,
Z DNA intercalation complex seems unlikely. First of all, the            Alan R. Lias, Inc., New York
optical properties of daunomycin in the presence of a large        12. Pohl, F.M., Jovin, T. M., Baehr, W., and Holbrook, J. J. (1972)
8906                                                                                                             Allosteric Conversion of 2 DNA
      Proc. Natl. Acad. Sci. U.S. A . 69, 3805-3809                                                                                                           Acad. Sci. U. s. A. 80,2447-2451
13. Mirau, P. A., and Kearns, D.R. (1983) Nucleic Acids Res. 11,                                                                                        31. Klysik, J., Stirdivant, S. M., Singleton, C. K., Zacharias, W., and
      1931-1991                                                                                                                                               Wells, R. D. (1983) J. Mol. Biol. 168, 51-71
14. Chaires, J. B. (1983) Nucleic Acids Res. 11,8485-8494                                                                                               32. Peck, L. J., and Wang, J. C. (1983) Proc. Natl. Acad. Sci. U.S.
15. Chen, C., Knop, R. H., and Cohen, J. S. (1983) Biochemistry 22,                                                                                           A . 80,6206-6210
      5468-5471                                                                                                                                         33. Pohl, F. M. (1983) Cold Spring Harbor Symp. Quunt. Biol. 47,
16. Walker, G. T., Stone, M. P., and Krugh, T. R.(1985) Biochemistry                                                                                          113-117
      24,7462-7479                                                                                                                                      34. Weber, G. (1975) Adu. Protein Chem.29, 1-83
17. Chaires, J. B. (1985) Biochemistry 24,7479-7486                                                                                                     35. Anderson, C. F., and Record, M. T. (1982) Annu. Rev. Phys.
18. Monod, J., Wyman, J., and Changeux, J-P. (1965) J. Mol. Biol.                                                                                             Chem. 33,191-222
      12,88-118                                                                                                                                         36. Record, M. T., Jr., Anderson, C. F., and Lohman, T. M. (1978)
19. Crothers, D. M. (1968) Biopolymers 6, 575-584                                                                                                             Q. Reu. Biophys. 11,103-178
20. McGhee, J. D.,   and von Hippel, P. H. (1974) J. Mol. Biol. 86,                                                                                     37. Matthew, J. B., and Richards, F. M. (1984) Biopolymers 23,
      469-489                                                                                                                                                 2743-2759
21. Dattagupta, N., Hogan, M., and Crothers, D. M. (1980) Bioehem-                                                                                      38. Soumpasis, D-M. (1984) Proc. Natl. Acad. Sci.U. A. 81,5116-
      istry 19,5998-6005                                                                                                                                      5120
22. Chaires, J. B. (1983) Biochemistry 22, 4204-4211                                                                                                    39. Crothers, D. M., and Fried, M. (1983) Cold Spring Harbor Symp.
23. van Holde, K. E., and Weischet, W. 0.(1978) Biopolymers 17,                                                                                               Quunt. Biol. 47, 263-269
      1387-1403                                                                                                                                         40. Britt, M., Zunino, F., and Chaires, J. B. (1986) Mol. Pharmacol.
24. Kovacic, R. T.,  and van Holde, K. E. (1977) Biochemistry 16,                                                                                             29,74-80
      1490-1498                                                                                                                                         41. Ivanov, V. I., and Minyat, E. E. (1981) Nucleic Acids Res. 9,
25. Wells, R. D., Larson, J. E., Grant, R.C., Shortle, B.E., and                                                                                              4783-4798
      Cantor, C. R. (1970) J. Mol. Biol. 54,465-486                                                                                                     42. Gupta, G., Dhingra, M. M., and Sarma, R. H. (1983) J. Biomol.
26, Chaires, J. B., Dattagupta, N., and Crothers, D. M. (1982) Bio-                                                                                           Struct. Dyn. 1,97-113
      chemistry 2 1,3933-3940                                                                                                                           43. Chaires, J. B., Dattagupta, N., and Crothers, D. M. (1985) Bio-
27. Muller, W., and Crothers, D.M. (1975) Eur. J. Biochem. 54,                                                                                                chemistry 24,260-267
      267-277                                                                                                                                           44. Neumann, 3. M., Cavailles, J. A., Herve, M., Tran-Dinh, S.,
28. Graves, D. E., and Krugh, T. R. (1983) Biochemistry 22, 3941-                                                                                             Langlois d'Estaintot, B., Huynh-Dinh, T., and Igolen, J. (1985)
      3947                                                                                                                                                    FEBS Lett. 182,360-364
29. Shafer, R. H., Brown, S. C., Delbarre, A., and Wade, D. (1984)                                                                                      45. Quigley, G. J., Wang, A. H-J., Ughetto, G., van der Marel, G.,
      Nucleic Acids Res. 12,4679-4690                                                                                                                         van Boom, J. H., and Rich, A. (1980) Proc. Natl. Acad. Sei. U.
30. Singleton, C. K., Klysik, J., and Wells, R. D. (1983) Proc. Natl.                                                                                         S. A. 77, 7204-7208

                                                         Supplemental Material to
                                  The Allosteric Conversion                of 2 DNA to an I n r e r e s l e t e d .
                                                    Right-Handed      Helix by Daunomycin

                                                             Jonathan 8 . C h a i r e s

                                             PROPERTIES THE
                                                      OF DAUNOIFICIN-Z                                             DNA COWPLEX                                   Yhile t h e r e is l i t t l e change in t h e relative f l u o r e s c e n c ei n t e n s i t y between daunomy-
                                                                                                                                                        cin f r e e in Bolution and i n the
                                                                                                                                                                                          presence                                        of exeese 2 f o n DNA.                           t h e r e i8   11   measurable
                                                                                                                                 and t h a t the
         A recent r e p o r t suggests t h a te t h i d i u m              may intercalate i n t o L-form DNA.                                          d i f f e r e n c e in t h ef l u o r t a e e n c ep o l a r i z a t i o n .I n              3.5 M NaC1, the f r e ed r u g                  shows       B   polari-
int.irc(I1afianevent                   results i n o p t i c a l chengerr in ethidium similar                     t o rhoeoobservrdfor                  ration Of 0.133 + f - 0.004. while in t h e pre~ence of excess (DMfbp                                0.012S) poly (dCdC).
its b i n d i n gt o         B-form DNA ( 2 9 ) .        This o b s e m a t i o n 10 in c o n t r a d i c t i o n t o t h e earlier work                t h e value increase* s l i g h t l y t o 0.160 +/- 0.003.            Under t h e s e conditions. P value f o r t h e
of P o h l e t d .           (12).         and the mare r e c e n t wcvk of Walker             st. (16).             Since theabeorbance                d r u g i n t e r c a l a t e d into B f o m DNA cannot be d e t e m i n e d , aince t h e f l u o r e s c e n c e of t h e drug
and f l u a r e e e e n c e methadauead              in t h i a r e p o r t 10 a n a l y z e daunomycin binding                          t
                                                                                                                             . I I B U ~ ~h a t   The   i. nearlycompletely                   quenched upon b i n d i n g to B fom DNA.                                    The p o l a l i r l l t i o n of a dnunomy-
                      i n Che drug arise from its interaction w i t h B form DNA, t h i s aSBumPtiOn
o p t i c a lc h a n g e s                                                                                                                              d o - p o l y (dAdT) complex.
                                                                                                                                                                                    which                       retains a p p r e c i a b l ef l u o r e a c e n c e               ( 2 2 ) . was previaualyfound
requires e x p e r i m e n t a l e r i f i c a t i o n .
                               v                             Such data are preeenrsd
                                                                                   here.  The resulcs d e s c r i b e d                                 t o be      0.6 for a polymer of approximatelythe                                  8816         size as rhepoly                        uaed
                                                                                                                                                                                                                                                                                           (dMC)               hare ( 2 2 ) .
h e r e show that t h e n t e r a c r i o n
                          i                              of daunomycin with 2 form DNA results i n little or no                                         In presence
                                                                                                                                                         the                           of exeella poly (dCdC). daunomycin t h u s rereins Lppreclsble
change         in che abeorbance or f l u o r e s c e n c e o p e r t i e s
                                                        pr                                         of    the Further.
                                                                                                          drug.   iodide                                freedom. inconsistent w i t h what vauldheexpected                                            for     a drugintercslarad                   into DNA.
quenchingexperiments                        show t h a t in the presence         of a l a r g e molar excess of 2 DNA, daunomy-

cin 18 f r e e l y s c c e a r i b l e t o         added queneher. end is t h e r e f o r e probablY nor i n t e r c a l a t e d .                               Figure M3 ahows a Stem-volmer p l o tf o tt h eq u e n c h i n g                                           of daunomycin i n 3 . 5              M NaCl in
                                                                                                                                                        Lhr prestlxe of e x ~ e s a (DM/bp                          0 . 0 1 2 5 ) Z form poly (dCdC).                          The f r e ed r u g         Y~II   fovnd co

         The extinction c o e f f i c i e n t of daunomycio is u n a l t e r e d i n the

           The apparent extinetion c o e f f i c i e n t at 480 nm of daunomycin in 3 . 5 M NaCl in t h e
                       exsees p o l y (dCdC) in t h e 2 farm or c a l f t h m s DNA in t h e B form i e *horn in
                                                                                                                                        excetls 2       have a Stern-Volmet quenching Constant.
                                                                                                                                                        e e l l y are i n a e e o s s i b l s to added
                                                                                                                                                        previously f a r daunomycin (22).
                                                                                                                                                                                                                     The s l o p e
                                                                                                                                                                                                                                  KBv, of

                                                                                                                                                                                                                                          Of        l
                                                                                                                                                                                                                                                     13.4 U-'

                                                                                                                                                                                                                                                 tho ine

                                                                                                                                                                                                                                               t h e r e f o r e .h a v e
                                                                                                                                                                                                                                                in f i g u r e 13 p r o v i d e s K
                                                                                                                                                                                                                                                                                           0 ms indeed
                                                                                                                                                                                                                                                                                                                   -   found

f i g u r e MI.         Addition Qf low molar ratios of
                                                      drug                              to B form DNA eeaulra i n am apparent                           +/- 0.3 M'
                                                                                                                                                                 -.                  daunomycin
                                                                                                                                                                                 Thus.                           is f r e e l y acceaaible t o addedquencher        in t h e presence Of
e x t i n c t i o nc o e f f i c i e n t     near 7000 M-lcm'l,        B   v a l u ei d e n t i c a l   t o thatpreviouslydetermined                    e x c e s ~poly (dGdC). Contrary LO                      whatwould be expected i f ic vert i n t e r c a l s r e d .
to b e h a r a c t e r i s t i c
         c                                                the
                                    of t h e bound f o m of drug        (26).    ID Contrast, a d d i t i o n of IOU
mler ratios of drug co Z form p e l y (dCdC) r e e u l ~ ein little L ~no change in thenppsrcnr

extinction c o e f f i c i e n t .       A t a molar ratio of drug/bp   of               e            t
                                                                               0.01, t h a p p a r e ne x t i n c t i o n
c o e f f i c i e n t is w i t h i n 5X of t h e value seen for t h e r u g
                                                                     d        alone. These conditions ehould
favor t h e fornation of even L weak complex, and t h e data suggest that t h e formation of B
daunomycin 2 DNA c ~ m p l c xresulra in l i t t l e alteration of t h e absorbance of thedrug.                           This
result contraat8 completely rho           with reported        result of  Shsfer            .
                                                                                       1 (29) f o r ethidium.
and 1s consintent with the behaviot of ethidium reporred                                          by Walker     If.g.

          Pluoreseoce p r o p e r t i e s o f          the daunomycin 2 DNA              CODP~PX.        F i g u r e M 2 ahows t h e r e l a t i v e
f l u o r e s c e n c e intensity Of damomyein                  LIB   a f u n c t i o n of total drug concentretion alone and
i n the
      preaencc                   of        exeees poly    (dGdC) (L-form)            and c a l f        thymus DNA ( B form).
presence                e x c e m B form DNA,                                                   daunomycin
                                                         the intrineic f l u o r e s c e n c e of                             is nearly con-
                                                                            Allosteric Conversionof                    ZDNA
                                                                                                       CIRCULAR DICHROISM EXPERIHEWTS                          TO DETERMINE THE FRACTION OF                   RIGHT-HANDEU HELIX AS A
                11w.      1                                                                            N H C T I O N OF BO!lND DAIRIOHYCIN.

                                                                                                              Circular diehroiem                    (CO) was usad t o monitor the conformation                      of poly ( d M C )       1111   a
       R                                                                                               function       of bound                daunomycinunder            B formconditione(BPEbuffer)                      end    LVO    2    form
       w                                                                                               conditions       (2.4 W NaCl                    and   3.5   I4    NaCl).     CD   titration experiments v c w                  perfomed
       T                                                                                               essentially      (18    described in (16).                   CD    spectra   were recorded in           B   1 cm p e t h l m g t h eel1
                                                                                                       fhermoetsted et 2 . .
                                                                                                                        5C                         A Jasco 5-500 apectropolarimeter interfaced t o an 1BH PC computer
       X                                                                                               wa8   uaed for all                 I L L ~ U ~ ~ ~ C ~ C Spectra
                                                                                                                                                                B .            yere recorded from         350 to 200 n at 0 2 nm
                                                                                                                                                                                                                      m    .
       T                                                                                               intervals, and four s p e c t r a were recorded and averaged for each sdditlonof drug.                             Wolar
       N                                                                                               ellipticity      "88       calculated, a f t e r baseline correction. according LO the equation,
                                                                                                                              [el         -    (LOO
                                                                                                       vhere C 18 the molar base p a i r concantrarion. 1 1 s the parhleogrh and 0 is the measured
       N          6BBB.
                                                                                                       ellipticlLy.           An Optical titration                  vas performed        in parallel t o determine the binding
                                                                                                       r a t i o for each addition of daunomycin.
                       0.      .I   .2 . .4 .5 .6 .7
                                        3                                      .8     0    I.
                          <-        O M
                                     H -   8 . P. W>
                                                                                                            Figure H shows the results of the experiments.       B form DNA ehovs an approximate
                                                                                                       inoellipric point at 295 nm (panel A ) . 2 f O r m DNA shows an inirially strang negative
                                                                                                       ellipticity ST this wavelength char increases with bovnd drug .e the polymer revert8 t o
                                                                                                       the right-handed                conformation.           At 250 nm (panel E) 2 DNA shows                     poairive ellipticity.
                                                                                                       while B form DNA shows II strong negative ellipticity.                                T e fraction of polymer (a) in the
                                                                                                                                                                  Figure I44 from the equation.
                                                                                                       right-handed conformation nag be calculated from the data of

                                                                                                                              a       -   1   - ([e~,-rs~,)/(lel~-~e~~)
                                                                                                       vhere le], and             [elg are            the molar ellipticity          of 2 and B form poly              (dCdC) a t      a    given
                                                                                                       wavelength and le1 is the appnrtnr ellipticity at a particular binding ratio r .                                                Figure 5
                                                                            0                          in che main t e x t shows the fraction o f right-handed hell,: as s function of t h e binding
                                                                                                       r a t i o r.

                                                                       0    0

                                          8                                                                            M
                                                                            A                                          A
                                                                       A                                               R

     Figure W2.    Fluorescence emission of daunomycin in the presence of B end 2 form DNA.
The relative fluorescence intensity of dauamyycin in 3.5 II NaCl ia sham as P function of
totel dlug eaneencration for the drug alone ( a         ).    in the presence of 80 YW    (bp) poly
(dcdc) ( 0 ) . and in the presence of 80 YW (bp) calf r h p u DNA       (A).    Tho fluorescence o f
the drug in the preeenee of 2 form poly ( d M C ) is unchanged at low eoneentrationa of added
drug, but is nearly completely quenched
the fluorescence measurements, l e x

                  2.5+      ...
                                       - nm.
                                               in the presence of B form calf thymus DNA.
                                                  hem   I   555   n.
                                    : . . . t . . . r . . . : . - . t
                                                                                                                                          2.       k
            /                                                                                                           I                                                                                 A
            I                                                                                                           P                                                                             0
                                                                                                                        T                                                                         0
                                                                                                                        I                                                                                      A     A

                                                                                                                                              ~~                                                                      .          .
                                                                                                                                               e.                       .1                   .2                      .3         .34

                                                                                                              Figure    4
                                                                                                                       W .        I

                                                                                                       for the i n t e r a c t i o n of daunomycin with poly (dGdC).                     The symbole refer t o differcnt buffer
                                                                                                       eondiriana:      (0)           BPE (B form);          ( 4 2.4 H NaCl ( 2 farm); ( b ) 3.5          n   NaC1 ( 2 form).         The data
                                                                                                       in panel A vas obtained at 295                        W. while        the datain panel B vas obtained LL 250                  m

Shared By: