Extent of the Acidification by N7-Coordinated cis-Diammine-Platinum(II) on the Acidic Sites of Guanine Derivatives

Coordination of two monoprotonated 2'-deoxyguanosine 5'-monophosphate species, H(dGMP)−, via N7 to cis-(NH2)2Pt2+ gives the complex cis-(NH2)2Pt(H·dGMP)2 which is a four-protonic acid. The corresponding acidity constants were measured by potentiometric pH titrations (25℃; I = 0.1 M, NaNO3). The first two protons are released from the two -P(O)2(OH)− groups (PKa/1= 5.57; PKa/2 = 6.29) and the next two protons are from the H(N1) sites of the guanine residues (PKa/3 = 8.73; PKa/4 = 9.48). The micro acidity constants of the various sites are also evaluated. Comparison of these data with those determined for the three-protonic H2(dGMP)± (PKa/1 = 2.69 for the H+(N7) site; PKa/2 = 6.29 for -P(O)2(OH)− ;PKa/3 = 9.56 for H(N1)) shows that on average the N-7-coordinated Pt2+ acidifies the phosphate protons by Δ pKa = 0.36 and the H(N1) sites by Δ pKa = 0.46. These results are further compared with those obtained previously for cis-(NH2)2Pt(L)2, where L = 9-ethylguanine or monoprotonated 2'-deoxycytidine 5'-monophosphate. Conclusions regarding platinated DNA are also presented.

Coordination of the two dGMPs via N7 to cis-(NH3)2Pt 2+ is confirmed by the acidbase properties of 1 described in this study; these are in accord with a pt2+-N7 coordination in cis-(NH3)2Pt(H.dGMP)2 only.

Materials and Apparatus for the Titration Experiments
The disodium salt of 2'-deoxyguanosine 5'-monophosphate was purchased from Sigma Chemical Co., St. Louis, MO, USA. Potassium hydrogen phthalate, NaNO 3, HNO3 and NaOH (Titrisol) (all pro analysO were obtained from Merck AG, Darmstadt, Germany. The disodium salt of Pt(dGMP)" was prepared as described in Section 2.1. For all solutions distilled CO2-free water was used.
The titer of the NaOH used for the titrations was determined with potassium hydrogen phthalate. The stock solutions of dGMP 2" and cis-(NH3)2Pt(dGMP) " were freshly prepared daily, and the pH was adjusted to about 8.4 and 7.6, respectively; the exact concentrations of these solutions (titrated in the presence of an excess of HNO3; see Section 2.3) were measured by titrations with NaOH.
The potentiometric pH titrations were carded out with a Metrohm E536 potentiograph equipped with an E665 dosimat and a 6.0202.100(NB) combined macro glass electrode. The buffer solutions (pH 4.64, 7.00, 9.00; based on the NIST scale) used for calibration were also from Metrohm AG, Hedsau, Switzerland. The direct pH meter readings were used in the calculations of the acidity constants; i.e., these constants are so-called practical, mixed or Brnsted constants. [] Their negative logarithms given for aqueous solutions at I 0.1 M and 25C may be converted into the corresponding concentration constants by subtracting 0.02 from the listed pK a values. [9'1]

Potentiometric pH Titrations
The acidity constants KHH2(dGMP), K(dGMP and /IGM P of H2(dGMP) + were determined by titrating 50 mL of aqueous 1.08 mM HNO3 (25C; I = 0.1 M, NaNO3) in the presence and absence of 0.3 mM or 0.4 mM of dGMP under N 2 with 3 mL 0.03 M or 2 rrL 0.045 M NaOH, respectively, and by using the differences in NaOH consumption between two such titrations for the calculations. The constants were calculated with an IBM compatible computer with an 80486 processor (connected with a Brother M1509 printer and a Hewlett-Packard 7475A plotter) by a curve-fit procedure using a Newton-Gauss non-linear least-squares program within the pH range 3.1 to 10.3, corresponding to about 72% neutralisation for the equilibrium H2(dGMP)+/H(dGMP) and about 85% neutralisation for the equilibrium dGMp2/(dGMP-H)3. The results listed in Table 1 Table 1 are the averages of 4 independent pairs of titrations.
From the results summarized in Table 1 it is immediately evident that deprotonation of the -P(O)2(OH)residues occurs in all species relatively close to pH 6, whereas the proton from the H(N1) site of the guanine moieties is released in the pH range of about 9. Of some surprise may appear the fact that the H+(N7) site of the positively charged H(9-EtG) + re-  H2(dGMP) +species the release occurs already with pK a 2.69 (Table 1). This lower basicity of N7 in H(dGMP)" compared with that of 9-EtG is clearly attributable to the sugar residue, and is thus probably a solvation effect because H(guanosine) + and H(2'-deoxyguanosine) + are H+(N7)-deprotonated with pK a 2.11 + 0.04 [11] and pK a 2.30:1: 0.04, [12] respectively. Comparison of the values for H(dGuo) + (pK a 2.30) and H(d.GMP)-(pK a 2.69) shows that the expected charge effect is now operating.  It becomes thus evident that all these pK a values are close to the statistical expectation; they are on average only about 0.1 pK units larger than z PKaJst 0.6. In other words, the mutual influence that the two corresponding acidic sites in these complexes exert on each other is quite small, which indicates that the distances between these sites (at least in the protonated forms) must be relatively large.

Micro Acidity Constants for the cis-(NH3)2Pt(L). 2 Species and Acidifying
Effect of the N7-Coordinated Pt 2+ The negative logarithms of the two acidity constants, e.g., pKHpt(H.dCMP)2 and P/(dCMP)(H.dCMP), are only slightly more apart from each other than the statistically expected 0.6 pK units; this means, the buffer regions of the two species, Pt(H.dCMP) 2 and Pt(dCMP)(H.dCMP)-, are strongly overlapping. The same also applies to the other Pt 2+ complexes considered here; i.e., those formed between cis-(NH3)2P + and 9-EtG, H(dGMP)-, or dGMP 2 (see Table 1). Therefore, for a clean quantification of the acidity of the various sites it is necessary to calculate the micro acidity constants for the individual sites. Following known routes [13'14] we have summarized in Figure 3, as an example, the equilibrium scheme for cis-(NH3)2Pt(H'dCMP) 2 defining the micro acidity constants (k) and giving their interrelation with the macro acidity constants (K). There are three independent equations (a), (b), and (c), but four unknown constants; [3] however, by taking into account the statistical considerations of Section 3.2 the matter becomes simple in the present case because P/(H.dCMP) + log 2 5.73 + 0.3 6.03 pk Pkl; the analogues reasoning provides pk 2, etc.
The values for the micro acidity constants, pk pk and pk 2 pk2, for the four cis-(NH3)2Pt(L)2 complexes appearing in Table 1 are summarized in columns 2 and 3 of Table  2, respectively. Columns 4 and 5 provide the differences between the pKa values of the free ligands, like 9-EtG or H(dGMP)-, and the values given for pk pk and pk 2 pk2. Thus, these values quantify the acidifying effect of Pt 2+ on the individual sites. For a general comparison, however, we feel it is more appropriate to take the average  PKa/av pK H (pKHR(H.L)2 + pKHpt(L)(H.L)) (9) H(L) -' In equation (9) the difference is taken between the pK a value of the free ligand and the average of PKa/1 and PKa/2 for the complex formed with two such ligands by their coordination to cis-(NH3)2Pt 2+. This latter method is identical with the one we have applied before. [79] In the footnotes to Table 2 some detailed examples for the calculation procedures indicated above are given.
From the z PKa/av value in row 1 and column 6 of Table 2 it is evident that the acidifying effect on the two H(N1) sites of Pt 2+ coordinated to the N7 sites in the guanine residues of cis-(NH3)2Pt(9-EtG)22 + is quite significant (L PKa/av 1.23). The corresponding Table 2. Micro Acidity Constants for cis-(NH3)2Pt(L)2 Species (defined in analogy to pk pk b pk 2 pk 2 b L pk c,d L pk 2 d L PKa/av (z PKa/av 0.14 :t: 0.03), where it is N3-bound, is by about 0.2 pK units lower than in cis-(NH3)2Pt(H.dGMP)2 (z PKa/av 0.36 + 0.04) despite the fact that in both instances the two-P(O)2(OH)" groups are acidified, is not clear. Maybe the spatial orientation of the pt2+-coordinated nucleotides is different.

CONCLUSIONS
It is evident from the present study that a nucleobase-coordinated ci$-(NH3)2Pt 2+ affects only little the basicity of phosphate residues of nucleoside 5'-monophosphates; the same may be surmised for the phosphate groups in the backbone of DNA. As the basicity of phosphate groups is only slightly lowered, one may assume that the metal ion affinity of these groups is still quite pronounced; in fact, for cis-(NH3)2Pt(dCMP) " this has already been proven. [9] Consequently, one may suggest that, e.g., Mg 2+ binding to the phosphate backbone of platinated DNA is not much inhibited by the nucleobase-bound Pt2+.
A further interesting observation is the rather significant acidification of the H(N1) sites of guanine residues by N7-coordinated Pt2+. This suggests, and evidence pointing into this direction has already been found, [7] that in this way the H(N1) site is transformed into an even better H donor suitable for hydrogen bonding than it is the case in the uncomplexed guanine residue.