- Email: [email protected]
Differentiation between chelated and non-chelated DNA - Pt complexes by atomic absorption spectrophotometry
142 Bioehimiea et Biophysiea Aeta, 442 (1976) 142--146 © El~evie~ScientificPublishingCompany,Amsterdam-- Printed in The Nethedands
BBA98666 DIFFERENTIATION BETWEEN CHELATED AND NON-CHELATED DNA" Pt COMPLEXES BY ATOMIC ABSORPTION SPECTROPHOTOMETRY J.P. MACQUET* and T. THEOPHANIDES** Department of Chemistry, University of Montreal, Montreal H3C 3VI, Quebec (Canada)
(Received January 6th, 1976) Summary As a continuation of our study of the properties of the cis-trans platinum ~'eries we have investigated the DNA. Pt complexes of [Pt(dien)Cl]Cl, cis.Pt (en)C12, ¢/e.Pt (NH3hCI~, trans-Pt (NH3hCI2 and K~[PtCI4] by atomic absorption spectrophotometw. The DNA. Pt complexes cotTespond to the saturation of the N~-(Guanine) site~. It has been t~ound that the chelate complexes obtained with e/s-Pt(en)C12, c/e-Pt(NH~hCI2 and K2[PtCI4] show the same absorbance. The [Pt(dien)Cl]Cl and trane.Pt (NH3hCI2 which axe bound to the N| (guanine) site only, show an ab~rbance greater than the chelate complexes by a factor of two. In addition, it ha~ been possible in the case of the trans-Pt (NH3hCI~ complex to follow the fixation of platinum to DNA by atomic absorption spectrophotomettT. The result is similar to the ionic chlorine liberation procedure reported previously. Introduction Recently, we have shown [1] the possibility of determining the ci~trans stereochemistry of platinum-nucleoside complexes obtained with IT4[PtCI~]and c/e-Pt (NH~hCI~ complexes when they react with the same nueleoside, from their abeorbances. The in vitro platinum interaction [2] with DNA werestudied in order to differentiate the ci~ and trans-platinum compounds in their reactions with DNA, since only the cis compounds have antitumour properties. We have been able to identify the DNA. Pt complexes obtained with [Pt(dien)Cl] * P r e x n t ~ldref~: L~horatolze de Chtmle de Coo~dinat|on, C.N,lg~., B,P. 4142, SlOSO- T o o l o u ~ C.4~dex, Fi~lnfe, ** I ' o whom eo~¢sl~ndenee should be ~dd~e~d, A b b r ~ l a t l o ~ ; en s ethylened~mlne, H2N-CH243H2.NH2; d|en m blJ(2~mlnosthyl)amlne, H2N. CH2-NH'CH2~H2oNH2; NT(Gu~t) s nitrogen a t o m o n positinrJ 7 o f the gtumine molecule in DNA; 06(Otut) ~ o~ygen atom on carbon 6 of the gtumine molecule In DNA.
143
CI~/s-Pt(NH3)2CI2 and trans-Pt(NH3)2C12 from their specific reaction with the N~ (Gua) sites [2], the chlorine liberation data [3] during that reaction and from the nitrogen and oxygen binding energies obtained from X-ray photoelectron spectroscopy [4] in the solid state. The DNA-Pt complexes have the formulae: [Pt(dien)NT(Gua)]Cl2, c/s.[Pt(NH3)2N~(Gua).0~(Gua)]CI~ and trans-[Ft(NH3)2 N~(Gua)CI]CI. The cis-Pt(NH3)2C12 binds with two sites (NT(Gua), 06), whereas the trar~-Pt(NH3)~Cl2 is bound to one site only (N~(Gua)). The reaction of K2[PtC~4] with the DNA gave a complex of the formula Ft. (DNA)CI2 [5,6]. Analytical and spectroscopic data indicate this general formula which implies one PtCI2 unit per ( A . T; G • C) planes. The complex seems to have a cis-geometry as in the case of the reaction of c/s-Pt (NH3)2C12 with DNA. The purpose of the present work is the continuation of the study of the properties of the cis-trans-Pt(NH~)2Cl~ isomers and their reaction with the nucleic acids and their constituents. Experimental The apparatus, the operating parameters, the reagents and the standard solutions were the same as previously described [1,2]. The reaction of DNA as a function of time with trans-Pt(NH3)2Cl2 was carried out as follows. Each point in Fig. 1 (") corresponds to a single reaction and was repeated at least twice. The absorbances are the average of 300 measurements and the accuracy is 0.001. The DNA. Pt complexes were directly aspirated into the flame without any interference inhibitors. The platinum content was checked as reported [2]. Aspiration was kept constant with a 50 #g/ml Pt solution of (NH4)2[PtCI4] in 0.6 M HCI in the presence of 20 000/~g/ml lan. thanum (LaCI3). Zero time was considered the moment after the addition of trans-Pt(NH3)2Cl2 to DNA in which no platinum was fixed. The absorbance corresponded to the atomization of trans-Pt(NH~)2Cl~ in the presence of DNA (acting as an interference). Results and Discussion In a previous, study [2] we have determined the platinum content in DNA. Pt complexes without decomposing them by aspirating a solution containing 30 000/~g/ml PO~- (K3PO4). One cannot determine the platinum content in DNA. Pt complexes without an interference inhibitor (K3PO4). There is no linear relationship between absorbance and platinum concentrstion if no interference inhibitor is used. In the present work the absorbances of the DNA. Pt complexes depended on the nature of the platimum compound used and on the mode of fixation [1]o In Table I we see that the absorbances of the cis chelate complexes are not significantly different. The trans- and (dien) complexes have twice the absorbance of the cis- chelate complexes. It seems possible from atomic absorption measurements to differentiate a chelate DNA. Pt complex from a non.chelate complex, i.e., where platinum is bound only to one site, N~(Gua). Since the geometry of the complex obtained from DNA and K~[PtCi~] was not known we checked and compared its absorbance to that of cis-Pt(NH~)~CI2 and [Pt(dien)Cl]Cl and their DNA complexes. From Table I, it
144 TABLE I /~TOMIC ABSORPTION OF DIFFERENT DNA • Pt COMPLEXES Complexes obtefned with: 1. cts.Pt(NH3)2CI 2, ~. c/J-Pt(en)Ci2, 3. K2[PtCI4], 4. (Pt(dien)C1]C! and S.
tran#-.Pt(NH3)2Cl 2 . DNA. ~:. complexes *
C[~-[Pt(HN3)2N7(Gms).O6(Gtm)] CI2 cfs.[P~(en)NT(Gt~)-O6(Gua)] CI2 cis.Pt(NT(Gua).O6(Gua))Cl2 ** [Pt(dlen)NT(Gua)] CI2 tro~-[Pt(NH3)2Nq(Oua)Cl] C!
Abso~b/mces H20
S0 000 #s/ml P O ~ (K3P04)
0.012 ~.014 0.GII 0.028 0.030
0.170 0.180 0,170 0.180 0.181
* 50 #g/ml of platinum. ** This complex was dlalysed in order to remove the KCI liberated during the reaction.
is shown that the absorbance of this DNA complex is similar to that of cischelate complexes. This result is in agreement with the formula, ¢is-Pt(NT(GUa)0~(Gua))Cl2. The low absorbances in atomic absorption spectrometry of the DNA • Pt complexes are due to the interi'erence of DNA. We have already shown [1] that in Pt • nucleoside complexes the nucleotides do interfere. In DNA the interferences come from the nucleoside and in particular from the phosphates which are known to depress platinum absorbances in atomic absorption when in low concentrations [7,8]. The addition of an interference inhibitor (30 000/~g/ml PO~-, as K3PO4) eliminates the interferences. We checked the platinum absorbances just after the addition o f the platinum compound to DNA and found that in the case of trans.Pt(NH3)~Cl2there was a significant difference between its absorbance at the beginning o f the addition to DNA (time zero) and the absorbance of the DNA • Pt complex, when all the
TABLE n DECREASE OF THE ABSORBANCE OF TRANS.Pt(NH3)2CI 2 (S0 #g/ml) DURING ITS REACTION WITH DNA AND COMPARISON WITH THE IONIC CHLORINE LIBERATION REACTION STUDIED PREVIOUSLY Time (h)
Ab~orbsnces *
Number of Pt atoms fixed
Number of ionic chlorines liberated **
0 0.5 1 2 3 4 6 12 24
0.060 0.052 0.049 0.045 0.041 0.037 0,033 0.030 0.030
0.00 0.26 0.36 0.50 0.63 0.76 0.90 1,00 1.00
0.00 0.12 0.26
Accuracy of 0,001 which corresponds to 0.04 Pt atom. ** Accuracy of 0.04 Pt atom, There values are taken from tel. 3.
0.44 0.63 0.76 0.87 0,97 1.00
145
°IMECHOURS) Fig. 1. Comparison between the ionic chlorine liberation (o) and the number of platinum a toms fixed ( s ) during the reaction between D N A and Irans-Pt(NH3):zCl2. The value of 1 Pt a t o m corresponds to the saturation of all the N-/(Gua) sites (200% specificity, i.e., 0 . 8 2 Pt a t o m per (Ade • Thy; O ua • CYt) in the case of a DNA extracted from ~ l m o n sperm containing 41% Gua • Cyt).
platinum is bound to D N A . The absorbance of 50/~g/ml Pt of t~ans-Pt(NH3)2 C12 in H 2 0 is 0.178. In the presence of 0.5 mg/ml D N A this value becomes 0.060 at the beginning of the reaction. W h e n the reaction is completed, the final absorbance is 0.030 (see Table If),i.e.,a factor of two in the absorbance between the beginning and the end of the reaction. This reaction allows us to follow the fixation of trans.Pt(NH3)2Cl2 to D N A as a function of time and to compare it with the chlorine displacement reaction [3]. The results are given in Table II and in Fig. 1. The concentration of fixed platinum was calculated from the decrease of absorbance during the reaction. After the complete fixation of platinum the total difference in absorbance was known and corresponded to one Pt atom. In Fig. 1 is shown the similarity between the reaction of platinum fixation and that of ionic chlorine liberation. This surprising result is obtained under two different conditions and may be related to the affinity of platinum for DNA which is common in both reactions of platinum fixation to DNA. This is an extremely interesting result and will be checked with other systems as well. In the case of the other platinum compounds (cis-Pt(NH3)2CI2, cis-Pt(en)Cl2, [Pt(dien)Cl]Cl and K2[PtCI4]) the difference in absorbance at the beginning of the reaction (time zero) and at the end is not significant enough to follow the reaction. This is explained by the difference in absorbance between the transPt(NH~)2CI2 complex and the other platinum compounds in water [1]. This is the first time to our knowledge that a reaction (fixation of a metal to a ligand) could be followed using atomic absorption spectrophotometry.
Conclusion This work shows two interesting results. First the ability to differentiate the mode of fixation of platinum compounds to DNA from absorbances of DNA. Pt complexes obtained by atomic absorption spectrophotometry, found to
146
differ by a factor of at least two between a chelate complex (2 fixation sites) and a non.chelate complex (1 fixation site). Furthermore, it has been possible to follow directly the kinetics of platinum fixation in the reaction with DNA in the case of trans-Pt (NH~)2CI2. The surprising result is, however, that this plati~ num fixation is in agreement with the ionic chlorine liberation and platinum fixa.'.ion observed in the solution reactions [3], considering the completely different conditions under which the reaction takes place in the two cases. Acknowledgement The financial support of ~he National Research Council of Canada and the 'Minist~re de l'Education du Quebec' is greatly appreciated. One of us (J.P,M.) gratefully acknowledges an N.R.C.C. Fellowship. References 1 2 3 4 5 6
Macquet, J.P. and Theophanides, T. (1975) At. Absorpt. Newsl. 14, 23--25 Macquet. J.P. and Theophanides, T. (1975) Biopolymer~ 14 781--799 Maequet, J.P. and Theophanldes. T. (1975) Bioinorg. Chem. 5, 59--66 MilIard, M.M., Maequet, J.Po and Theophanldes, T. (1975) Bionh/m. Blopt~ys. Aeta 4 0 2 , 1 6 6 - - 1 7 0 Maequet, J.P, and Theophanldes, T. (1976) Inorg. Chim. Aeta, 1 8 , 1 8 9 - - 1 9 4 Maequet, J.P. (1975) Ph.D, Thes~s, University of Montreal order No. 75-28, 046, Univ. Mlcrofflms, Ann Arbor. Mich.. U.S.A. 7 Macquet, J.P., Hubert. J. and Theoph,mides, T. (1974) Anal. Chim. Aeta 72, 251~259 8 Pitts, A.E., Van Loon, J.C. and BeL,nish, F.E. (1970) A n d . Chlm. Acta 5 0 , 1 8 1 - - 1 9 4 9 Macquet, J.P. and Theophancs, T. (1974) Spectroehim. Acta 29B, 241--247
BBA98666 DIFFERENTIATION BETWEEN CHELATED AND NON-CHELATED DNA" Pt COMPLEXES BY ATOMIC ABSORPTION SPECTROPHOTOMETRY J.P. MACQUET* and T. THEOPHANIDES** Department of Chemistry, University of Montreal, Montreal H3C 3VI, Quebec (Canada)
(Received January 6th, 1976) Summary As a continuation of our study of the properties of the cis-trans platinum ~'eries we have investigated the DNA. Pt complexes of [Pt(dien)Cl]Cl, cis.Pt (en)C12, ¢/e.Pt (NH3hCI~, trans-Pt (NH3hCI2 and K~[PtCI4] by atomic absorption spectrophotometw. The DNA. Pt complexes cotTespond to the saturation of the N~-(Guanine) site~. It has been t~ound that the chelate complexes obtained with e/s-Pt(en)C12, c/e-Pt(NH~hCI2 and K2[PtCI4] show the same absorbance. The [Pt(dien)Cl]Cl and trane.Pt (NH3hCI2 which axe bound to the N| (guanine) site only, show an ab~rbance greater than the chelate complexes by a factor of two. In addition, it ha~ been possible in the case of the trans-Pt (NH3hCI~ complex to follow the fixation of platinum to DNA by atomic absorption spectrophotomettT. The result is similar to the ionic chlorine liberation procedure reported previously. Introduction Recently, we have shown [1] the possibility of determining the ci~trans stereochemistry of platinum-nucleoside complexes obtained with IT4[PtCI~]and c/e-Pt (NH~hCI~ complexes when they react with the same nueleoside, from their abeorbances. The in vitro platinum interaction [2] with DNA werestudied in order to differentiate the ci~ and trans-platinum compounds in their reactions with DNA, since only the cis compounds have antitumour properties. We have been able to identify the DNA. Pt complexes obtained with [Pt(dien)Cl] * P r e x n t ~ldref~: L~horatolze de Chtmle de Coo~dinat|on, C.N,lg~., B,P. 4142, SlOSO- T o o l o u ~ C.4~dex, Fi~lnfe, ** I ' o whom eo~¢sl~ndenee should be ~dd~e~d, A b b r ~ l a t l o ~ ; en s ethylened~mlne, H2N-CH243H2.NH2; d|en m blJ(2~mlnosthyl)amlne, H2N. CH2-NH'CH2~H2oNH2; NT(Gu~t) s nitrogen a t o m o n positinrJ 7 o f the gtumine molecule in DNA; 06(Otut) ~ o~ygen atom on carbon 6 of the gtumine molecule In DNA.
143
CI~/s-Pt(NH3)2CI2 and trans-Pt(NH3)2C12 from their specific reaction with the N~ (Gua) sites [2], the chlorine liberation data [3] during that reaction and from the nitrogen and oxygen binding energies obtained from X-ray photoelectron spectroscopy [4] in the solid state. The DNA-Pt complexes have the formulae: [Pt(dien)NT(Gua)]Cl2, c/s.[Pt(NH3)2N~(Gua).0~(Gua)]CI~ and trans-[Ft(NH3)2 N~(Gua)CI]CI. The cis-Pt(NH3)2C12 binds with two sites (NT(Gua), 06), whereas the trar~-Pt(NH3)~Cl2 is bound to one site only (N~(Gua)). The reaction of K2[PtC~4] with the DNA gave a complex of the formula Ft. (DNA)CI2 [5,6]. Analytical and spectroscopic data indicate this general formula which implies one PtCI2 unit per ( A . T; G • C) planes. The complex seems to have a cis-geometry as in the case of the reaction of c/s-Pt (NH3)2C12 with DNA. The purpose of the present work is the continuation of the study of the properties of the cis-trans-Pt(NH~)2Cl~ isomers and their reaction with the nucleic acids and their constituents. Experimental The apparatus, the operating parameters, the reagents and the standard solutions were the same as previously described [1,2]. The reaction of DNA as a function of time with trans-Pt(NH3)2Cl2 was carried out as follows. Each point in Fig. 1 (") corresponds to a single reaction and was repeated at least twice. The absorbances are the average of 300 measurements and the accuracy is 0.001. The DNA. Pt complexes were directly aspirated into the flame without any interference inhibitors. The platinum content was checked as reported [2]. Aspiration was kept constant with a 50 #g/ml Pt solution of (NH4)2[PtCI4] in 0.6 M HCI in the presence of 20 000/~g/ml lan. thanum (LaCI3). Zero time was considered the moment after the addition of trans-Pt(NH3)2Cl2 to DNA in which no platinum was fixed. The absorbance corresponded to the atomization of trans-Pt(NH~)2Cl~ in the presence of DNA (acting as an interference). Results and Discussion In a previous, study [2] we have determined the platinum content in DNA. Pt complexes without decomposing them by aspirating a solution containing 30 000/~g/ml PO~- (K3PO4). One cannot determine the platinum content in DNA. Pt complexes without an interference inhibitor (K3PO4). There is no linear relationship between absorbance and platinum concentrstion if no interference inhibitor is used. In the present work the absorbances of the DNA. Pt complexes depended on the nature of the platimum compound used and on the mode of fixation [1]o In Table I we see that the absorbances of the cis chelate complexes are not significantly different. The trans- and (dien) complexes have twice the absorbance of the cis- chelate complexes. It seems possible from atomic absorption measurements to differentiate a chelate DNA. Pt complex from a non.chelate complex, i.e., where platinum is bound only to one site, N~(Gua). Since the geometry of the complex obtained from DNA and K~[PtCi~] was not known we checked and compared its absorbance to that of cis-Pt(NH~)~CI2 and [Pt(dien)Cl]Cl and their DNA complexes. From Table I, it
144 TABLE I /~TOMIC ABSORPTION OF DIFFERENT DNA • Pt COMPLEXES Complexes obtefned with: 1. cts.Pt(NH3)2CI 2, ~. c/J-Pt(en)Ci2, 3. K2[PtCI4], 4. (Pt(dien)C1]C! and S.
tran#-.Pt(NH3)2Cl 2 . DNA. ~:. complexes *
C[~-[Pt(HN3)2N7(Gms).O6(Gtm)] CI2 cfs.[P~(en)NT(Gt~)-O6(Gua)] CI2 cis.Pt(NT(Gua).O6(Gua))Cl2 ** [Pt(dlen)NT(Gua)] CI2 tro~-[Pt(NH3)2Nq(Oua)Cl] C!
Abso~b/mces H20
S0 000 #s/ml P O ~ (K3P04)
0.012 ~.014 0.GII 0.028 0.030
0.170 0.180 0,170 0.180 0.181
* 50 #g/ml of platinum. ** This complex was dlalysed in order to remove the KCI liberated during the reaction.
is shown that the absorbance of this DNA complex is similar to that of cischelate complexes. This result is in agreement with the formula, ¢is-Pt(NT(GUa)0~(Gua))Cl2. The low absorbances in atomic absorption spectrometry of the DNA • Pt complexes are due to the interi'erence of DNA. We have already shown [1] that in Pt • nucleoside complexes the nucleotides do interfere. In DNA the interferences come from the nucleoside and in particular from the phosphates which are known to depress platinum absorbances in atomic absorption when in low concentrations [7,8]. The addition of an interference inhibitor (30 000/~g/ml PO~-, as K3PO4) eliminates the interferences. We checked the platinum absorbances just after the addition o f the platinum compound to DNA and found that in the case of trans.Pt(NH3)~Cl2there was a significant difference between its absorbance at the beginning o f the addition to DNA (time zero) and the absorbance of the DNA • Pt complex, when all the
TABLE n DECREASE OF THE ABSORBANCE OF TRANS.Pt(NH3)2CI 2 (S0 #g/ml) DURING ITS REACTION WITH DNA AND COMPARISON WITH THE IONIC CHLORINE LIBERATION REACTION STUDIED PREVIOUSLY Time (h)
Ab~orbsnces *
Number of Pt atoms fixed
Number of ionic chlorines liberated **
0 0.5 1 2 3 4 6 12 24
0.060 0.052 0.049 0.045 0.041 0.037 0,033 0.030 0.030
0.00 0.26 0.36 0.50 0.63 0.76 0.90 1,00 1.00
0.00 0.12 0.26
Accuracy of 0,001 which corresponds to 0.04 Pt atom. ** Accuracy of 0.04 Pt atom, There values are taken from tel. 3.
0.44 0.63 0.76 0.87 0,97 1.00
145
°IMECHOURS) Fig. 1. Comparison between the ionic chlorine liberation (o) and the number of platinum a toms fixed ( s ) during the reaction between D N A and Irans-Pt(NH3):zCl2. The value of 1 Pt a t o m corresponds to the saturation of all the N-/(Gua) sites (200% specificity, i.e., 0 . 8 2 Pt a t o m per (Ade • Thy; O ua • CYt) in the case of a DNA extracted from ~ l m o n sperm containing 41% Gua • Cyt).
platinum is bound to D N A . The absorbance of 50/~g/ml Pt of t~ans-Pt(NH3)2 C12 in H 2 0 is 0.178. In the presence of 0.5 mg/ml D N A this value becomes 0.060 at the beginning of the reaction. W h e n the reaction is completed, the final absorbance is 0.030 (see Table If),i.e.,a factor of two in the absorbance between the beginning and the end of the reaction. This reaction allows us to follow the fixation of trans.Pt(NH3)2Cl2 to D N A as a function of time and to compare it with the chlorine displacement reaction [3]. The results are given in Table II and in Fig. 1. The concentration of fixed platinum was calculated from the decrease of absorbance during the reaction. After the complete fixation of platinum the total difference in absorbance was known and corresponded to one Pt atom. In Fig. 1 is shown the similarity between the reaction of platinum fixation and that of ionic chlorine liberation. This surprising result is obtained under two different conditions and may be related to the affinity of platinum for DNA which is common in both reactions of platinum fixation to DNA. This is an extremely interesting result and will be checked with other systems as well. In the case of the other platinum compounds (cis-Pt(NH3)2CI2, cis-Pt(en)Cl2, [Pt(dien)Cl]Cl and K2[PtCI4]) the difference in absorbance at the beginning of the reaction (time zero) and at the end is not significant enough to follow the reaction. This is explained by the difference in absorbance between the transPt(NH~)2CI2 complex and the other platinum compounds in water [1]. This is the first time to our knowledge that a reaction (fixation of a metal to a ligand) could be followed using atomic absorption spectrophotometry.
Conclusion This work shows two interesting results. First the ability to differentiate the mode of fixation of platinum compounds to DNA from absorbances of DNA. Pt complexes obtained by atomic absorption spectrophotometry, found to
146
differ by a factor of at least two between a chelate complex (2 fixation sites) and a non.chelate complex (1 fixation site). Furthermore, it has been possible to follow directly the kinetics of platinum fixation in the reaction with DNA in the case of trans-Pt (NH~)2CI2. The surprising result is, however, that this plati~ num fixation is in agreement with the ionic chlorine liberation and platinum fixa.'.ion observed in the solution reactions [3], considering the completely different conditions under which the reaction takes place in the two cases. Acknowledgement The financial support of ~he National Research Council of Canada and the 'Minist~re de l'Education du Quebec' is greatly appreciated. One of us (J.P,M.) gratefully acknowledges an N.R.C.C. Fellowship. References 1 2 3 4 5 6
Macquet, J.P. and Theophanides, T. (1975) At. Absorpt. Newsl. 14, 23--25 Macquet. J.P. and Theophanides, T. (1975) Biopolymer~ 14 781--799 Maequet, J.P. and Theophanldes. T. (1975) Bioinorg. Chem. 5, 59--66 MilIard, M.M., Maequet, J.Po and Theophanldes, T. (1975) Bionh/m. Blopt~ys. Aeta 4 0 2 , 1 6 6 - - 1 7 0 Maequet, J.P, and Theophanldes, T. (1976) Inorg. Chim. Aeta, 1 8 , 1 8 9 - - 1 9 4 Maequet, J.P. (1975) Ph.D, Thes~s, University of Montreal order No. 75-28, 046, Univ. Mlcrofflms, Ann Arbor. Mich.. U.S.A. 7 Macquet, J.P., Hubert. J. and Theoph,mides, T. (1974) Anal. Chim. Aeta 72, 251~259 8 Pitts, A.E., Van Loon, J.C. and BeL,nish, F.E. (1970) A n d . Chlm. Acta 5 0 , 1 8 1 - - 1 9 4 9 Macquet, J.P. and Theophancs, T. (1974) Spectroehim. Acta 29B, 241--247