вход по аккаунту


Binuclear Lanthanide Complexes as Catalysts for the Hydrolysis of Bis(p-nitrophenyl)-phosphate and Double-Stranded DNA.

код для вставкиСкачать
Binuclear Lanthanide Complexes as Catalysts
for the Hydrolysis of Bis(p-nitropheny1)phosphate and Double-Stranded DNA**
known to form particularly stable complexcs with lanthanides."'] This might be one explanation for the lower. yet
still remarkable rate enhancements for BNPP hydrolysis obtained with Pr3+ complexes (Table 1, Scheme 1 I . The rate en-
Kaliappa G. R a g u n a t h a n and Hans-Jorg Schneider*
The catalysis of phosphate ester hydrolysis is of particular
interest from two diametric points of view. The formation and
cleavage of these ester linkages in nucleic acids are among the
most important molecular functions in living systems. At the
same time. as chemical warfare agents phosphate esters are the
most disastrous inventions of synthetic chemistry; however, it is
often forgotten that as insecticides the related aryl phosphates
have saved the lives of millions of people. The present communication addresses the hydrolysis of both kinds of esters catalyzed by binuclear metal complexes.
Recent work in particular by Bredow,"' Komiyama,[21Morrow,131Chin,'41 and their collaborators has shown that highly
charged cations, in particular of lanthanides, can greatly accelerate phosphate hydrolysis.[51Our own efforts are directed primarily towards the development of ligdnds that form sufficiently stable complexes and enhance the catalytic activity of these
lanthanide complexes.I6] The presence of several metal ions,
both in enzymes as well as in synthetic analogs, is known to
enhance rates of hydrolysis substantially.['] In very recent studies with carefully designed binuclear zinc complexes Chapman
and Breslow[81have demonstrated how the catalytic efficiency
depends critically on the distance between the metal centers and
the conformation of the spacer. However, the rate of hydrolysis['] for UpU as a DNA model was enhanced only by a factor
of 2, and for his( p-nitropheny1)phosphate (BNPP) the rate enhancement factor was around 12.[81In other recent reports the
application of' binuclear complexes having peroxo ligands,["]
zinc complexes for the hydrolysis of dinucleotides,[' and intramolecular metal complexes" 2 ] for the hydrolysis of phosphonates have been described. With binuclear complexes based on
imidazoles, Trogler et al.1131found a ligand rate factor['] of only
2 for BNPP hydrolysis; in studies with double-stranded D N A
binding but no cleavage was observed. We describe here binuclear lanthanide systems that lead to ligand rate factors['] of up to
72 for the hydrolysis of BNPP and to dramatic enhancements
for the cleavage of plasmid DNA.
Ligands 1-4 were synthesized according to literature procedures1i41
and treated with twoequivalents of Eu3+ or Pr3+ salts
in anhydrous methanol. The complexes with nitrate ions were
insoluble in methanol and precipitated immediately from the
reaction mixture; the complexes with chloride ions were soluble
in methanol and were precipitated by slow addition of acetonitrile. The washed precipitates were generally analytically pure.
The elemental analyses and titrations of the complexes with
EDTA (the complexes were dissolved in water at pH 2; the pH
was then increased to 5.2) indicate that all of the solid complexes
contain two metal ions. Unfortunately, the complexes with ligands 1 and 2. containing the more lipophilic benzene rings, were
water-soluble only as nitrate salts, which makes comparison to
the more reactive chloride salts with 4 difficult. Nitrates are
[*] Prof Dr. H.-J. Schneider, Dr. K. G. Ragunathan
FB Organische Chemie der Universiidt des Saarlandes
D-66041 Saarbrhcken (Germany)
Fax. Int. code +(681)302 4105
e-mail: CHl2hsvr(r<SBUSOL. RZ. UNI-SB. D E
I**] Supramolecular Chemistry. Part 58. This work was supported by the Deutsche
Forschungsgemeinschaft and the Fonds der Chemischen Industrie. K. R.
thanks the Alexander von Humboldt Foundation for a fellowship. Part 57:
H -J. Schneider. F. Eblingen. J. Sartorius, J. Rammo. J. Mol. Reeogn.. i n press.
,411geii..C ' l w i i i . Iirr E d Cigl. 1996, 35. N o . I f
Table 1. Summary of the results of phosphate ester and DNA hhdrolysis catalyzed
by lanthanide complexes.
F for BNPP [a]
DNA cleavage [ O h R F Ill [b]
1 [dl
1 Id1
[a] Efficiency of binuclear complexes for the hydrolysis of BNPP is given as
F = XM2Jk, from rates at 1 mM [M,L] and 2 mM [MI: [BNPP] = 3 35 x lo-' M;
T = 50 C; N-(2-hydroxyethyl)piperazine-N'-propanesulfonic acid (EPPS) buffer
(0.01 M),pH 7.0. The hydrolysis was followed by UV-Vis spectroscopy by measuring the increase in absorbance a t 400 nm due to the release ofp-nitrophenol during
hydrolysis. [b] Cleavage of double-stranded DNA is given as the percent of the
open circular form R F I1 after correction for the amount of this form in the starting
sample of DNA; incubation for 2 h at 37 'C with plasmid DNA (pBR322; concen-
buffer (0.01 M), pH 7.0. The solutions of the metal complexes were prepared by
dissolving the freshly prepared complexes in dimethyl sulfoxide ( 5 % DMSO in the
final solution), adding buffer, and adjusting the pH. The macrocyclic polyamines
inhibit the migration of DNA in the gel electrophoresis; this was overcome by using
ion exchange resins as previously reported [6b]. Densitometry after electrophoresis
showed, in double determinations. average deviations of i2.5 % [c] Without ligands. [d] Relative rate = 1 ; absolute rate k with EuCI, is 6.0 x lo-'. with PrCI,
Scheme 1. Structures of ligands 1-4
hancement factor of 72 observed with Pr:' . 4 indicates that the
metal centers should have a certain separation; complexes with
ligands 1-3 have a smaller distance between the metal binding
sites and lead to only moderate rate enhancements at best. In
line with the report of Chapman and BreslowI8] such a separation seems to be necessary for simultaneous electrostatic stdbilization of the bound phosphate group at one end of the complex and activation of water at the other.['61
The analysis of the reaction kinetics (van't Hoff method, second-order kinetics) indicates that the catalysis arises from the
action of two metal ions in the transition state. A plot (Fig. 1) of
Igu versus Ig[M] at constant ligand concentration, where u is the
initial rate (measured at < 10% conversion) and [MI is the metal
Cl VCH Verlugsgesellschufi inhH. 0-69451
Weinheim, 1596
B f5.00+ . 3 1 0
2 75
Fig. 2. Electrophoresis gel demonstrating cleavage of double-stranded DNA after
2 h. For conditions see the footnote [b] in Table 1. Lane 1. DNA alone: lane 2:
DNA and Eu"; lane 3 . DNA and Eu:+-4: lane4: DNA and Eu:--Z; lane 5:
DNA and Eu:' 1: lane 6: DNA and Pr": lane 7: DNA and P r i + - 4 ;lane 8: DNA
and Pr:+.Z; and lane 9: DNA and Pr:+.l.
-2 5 0
k[MlFig. I . Plot of Igc (from initial rate r of BNPP hydrolysis at < 1 O ' X conversion.
[BNPP] = 7 x
M) versus Ig[M]. where [MI = [Pr3-] at constant concentration
of ligand 4 ( I mM). The concentration of the metal was varied between I and 2 mM.
Observed slope = 2.3k0.2. For details on the preparation of the metal complexes
see text, footnotes in Table 1, and ref. [17].
concentration (between 1 and 2 mM), gives a line with a slope of
2 within experimental error. This treatment neglects the (small)
contributions of free metal ions and mononuclear complexes.
The van? Hoff plot (Fig. 1) proves that two metal ions are involved in the rate-determining step. Spectroscopic or other
methods that indicate a 2 : l complex in the solid state or in
solution are not conclusive, as the catalytically active complex
could be thermodynamically less favorable. Alternative explanations for the observed van't Hoff slope of 2 could be the
participation either of two free cations or-even more unlikely-of a single ion complex ML and a free M cation. The first
alternative is excluded by the clean first-order reaction observed
with uncomplexed lanthanide ions@'] and by the 72-times slower reaction with the metal cation alone. The second alternative
is excluded by the Fact that the reaction catalyzed by a 1 : 1
complex in the presence of an equimolar amount of free Ln3+
ions is five times slower than that catalyzed by the 2: 1 complex.
The 1 : l complex was prepared from equimolar amounts of
Ln3+ and the ligand in water instead of methanol. The complexes with varying metal content were prepared by mixing L and M
in methanol (dried over magnesium methanolate), which was
then was removed at 4 0 T , and then dissolving the residue in
water. The slow rates of complex formation require special conditions in all cases.[171When EDTA is added (1 mM in the final
solution) the reaction rate decreases by about 70% within minutes; with 2 mM EDTA no reaction is observed. This indicates
that the complexes are not stable to exchange with EDTA, in
line with similar findings by Morrow et al.[lsl
Surprisingly, the activity of the complexes of 1-4 with plasmid D N A follows the same sequence as that with BNPP, although the nitrophenolate leaving group could lead to a change
of mechanism.['91 After the reaction of plasmid D N A with
Pr:' .4, electrophoresis and densitometry indicates that single
cleavage of form R F I yields 80% of the open circular form
R F I1 (after correction for the R F I1 form present in the starting
material). Visual inspection of the gels shows only form R F I1
as the cleavage product (Fig. 2). This is to our knowledge the
highest proportion of hydrolytic cleavage yet reported for plasmid DNA. In contrast to radical cleavage processes no further
cleavage products such as the linear form R F 111 are apparent in
the agarose gels (Fig. 2).[201That the cleavage occurs by hydrolysis like that with enzymes is furthermore in line with the recent
observations of Chin et al.[lO1and Komiyama et a1.r2b1
redox-active Ce4+ ions act as hydrolytic catalysts. The fact that
the rate of D N A cleavage is not accelerated when hydrogen
peroxide is added[6b1also speaks against oxidative radical-based
We have
that the cleavage of plasmid D N A by
lanthanide ions after electrophoretic separation and densitometry is characterized by clean pseudo-first-order rate constants,
which then may be used to derive Michaelis-Menten-type equilibrium and rate constants K, and k,,,, respectively, from saturation kinetics. We applied the same procedures to study the
cleavage of double-stranded D N A with the binuclear complexes
prepared from ligand 4. The observed rates with Eu2+.4 and
Pr: . 4 showed the expected increase with catalyst concentration, but was limited by solubility problems at higher concentrations. The corresponding saturation curve for Pr;' . 4 (Fig. 3)
0 002
0 000
Fig. 3 Saturation kinetics of the cleavage of plasmid DNA by Pr:+ 4. The theoretmin-'
ical curve corresponds to K , = 3 3 x l W 4 M and k,,, = 1 7 x
with K , around 3 x
M and k,,, around 1.5 x lo-' min-'
agree with values found with other lanthanide systems.'6b' The
rate constant with Pr2+.4 at 37°C is approximately 2 x lo6
times greater than that of the uncatalyzed hydrolysis of doublestranded D N A at 25' and pH 7.0, which was previously determined only very approximately.[21]
In conclusion, binuclear lanthanide complexes can accelerate
the cleavage of double-stranded D N A by almost two orders of
magnitude over that with metal ions alone. The structure of the
ligands, in particular the distance between the metal centers,
plays an important role. Unexpectedly, the trends observed with
the different ligands are similar for hydrolysis of both activated
nitrophenyl phosphates and plasmid DNA. The half-life of
BNPP is reduced from approximately 75 years[’*] to as little as
8 min (with Eu:+ - 4 a t 50 “C and p H 7.0); that ofdouble-stranded DNA from approximately 2 0 0 0 y e a r ~to~ 50min
~ ~ ~ (with
Pr:’.4 at 37 ‘ C and pH 7.0).
Received: September 25, 1995
Revised version: December 29, 1995 [Z8424IE]
German version: Angel$. Chem. 1996. 108. 1314-1316
Keywords: DNA cleavage . hydrolyses . phosphorus compounds
supramolecular chemistry . lanthanide compounds
[ l ] R. Breslow. D -L. Hudng, PI-oc. N;fl. Acud Sci. U S A 1991,88, 4080.
[?I a) M. Komiyama. K. MatSdmUrd, Y. Matsumoto, J. Chem. Sor. Chem. Comnnm 1992, 640: b) M. Komiyama, N. Takeda, Y Takahashi. H. Uchida, T.
Shiiba. T. Kodama, M Yashiro, J. Chixm. Sue. Perkm Truns. 2 1995,269, and
references therein
[3] J. 1.Morrow. L A Buttrey. V. M. Shelton. K. A. Berback. J. Am C / ~ e mSoc.
1992. 114. 1903.
[4] D. Wahnon. R. C. Hynes, J. Chin, J. Chem. Soc. Chem. Commun. 1994, 1441.
[j] Lanthanides were used to catalyze the hydrolysis of phosphate esters decades
ago without recognition of their full potential. For a review see E. Bamann, H.
Trapman, A h , . En;ymol. Relul. Suhj Biochrm. 1959. 21, 169.
161 a ) H -J. Schneider. J. Rammo. R. Hettich. Angew Chem. 1993. 105. 1773;
A n p i ’ Chi,m. Inf. Ed. Engl. 1993.32.1716: b) J. Rammo, R. Hettich, A. Roigk,
H.-J. Schneider. J Cheni. Soc. Chem. Commun. 1996. 105-107.
[7] For a review on the use of binucledr metal complexes as catalysts for different
hydrolysis reactions see. M. W. Gobel, Angew. Chern. 1994, 106, 1201. Angew.
Chenr.I n f . Ed. E I I ~ I1994,
33, 1141
[8] W. H. Chapman. Jr.. R. Breslow, J. Am. Chem. Sue. 1995. 117, 5851.
[9] As the factor F‘chardcterizing the efficiency of a ligand we use the ratio kM2,/kM
for the reaction with the complex [M,L] and with the corresponding metal M
alone at the same concentration as with the ligand L.
[lo] 8. K. Takasaki. J Chin. J. Am. Chem. Soc.1995, 117, 8582.
[I I ] M Yashiro. A Ishikubo, M. Komiyama. J. Chem. Soc. Chem. Commun. 1995,
[12] A. Tsubouchi. T. C. Bruice. J. Am. Chem. Soc.1995, 117, 7399.
[13] E.A Kesicki. M. A. DeRosch, L. H. Freeman, C. L. Wdlton, D. F. Harvey.
W. C. Trogler, lnorg. Chim. 1993, 32, 5851
[14] a) D. Chen. A. E. Martell, Tefruhedron 1991.47,6895 for 1 and 3; b) R Menif.
A. E Martell. P. J. Squattrito, A. Clearfield. Inorg. Chem. 1990,29,4723 for 2 ;
c) R. Menif. D. Chen, A. E. Martell, [hid. 1989, 26, 4633 for 4
[lS] R. B King. R Heckley, J. An7. Chem. So?. 1974. 96, 3118.
1991. 24, 145.
[16] J. Chin. A<? < ‘ I r ~ ~ i r iRc~s.
[17] The rate constant observed for the 1 : 1 complex of ligand 4 and P r 3 + prepared
in buffer solution by equilibriating 1 mM 4 and 1 mM Pr3+ for 8 h at 25°C and
30 min at 50 C wds only 4.1 x lo-’ min-l, and that for the complex prepared
in methanol (dried over Mg(OMe),) by mlxing the ligdnd and the metal in 1 :1
ratio and warming the solution to 40’C to remove methanol was
9 1 x lo-’ min
The van? Hoff plot for the hydrolysis of BNPP by the complexes ofligdnd 4 and various amounts of Pr”, prepared directly in the buffer
solution. gave a slope of only 1.2. When the ligdnd 4 and Pr3’ in the ratio 2: 1
were warmed in anhydrous methanol, catalysis with the resulting complex was
even slower than with free Pr” alone. This could be due to formation of the
2: 1 sandwich complex. in which the metal is shielded against reaction with the
ester and/or u ater.
[I81 J. R. Morrou. L. A. Buttrey, V. M. Shelton, K. A. Berback, J. Am. Chem. Sor.
1992, 114. 1903.
1191 A. J. Kirby. M. Younas. J. Chem. Soc. B 1970, 510.
1201 JLL. Sagripanti. K. H. Kraemer, J. M u / . Biol. 1989, 264, 1729.
1211 .I.Eigner. H. Boedtker, G. Michaels. 5joch1m. Bioph.vs. Arm. 1961, Sl, 165.
[22j J. Chin. M Banaszczyk. V. Jubian. X. 20%J. Am. Chem. Soc. 1989, 111, 186.
Efficient Synthesis of New
2-Cycloalk(en)ylpropanoic Acid DerivativesMedium and Large Rings as Bioisosteres
of Alkylphenyl Moieties?**
Bjorn Greve, Peter Imming,* and Stefan Laufer
Dedicated to Professor Gunther Seitz
on the occasion of his 60th birthday
Ring systems with 9-12 (or 7-14) members are classed as
medium-sized rings. They occur less frequently and are more
difficult to prepare than five- and six-membered and large rings,
and their applicability is restricted.“] In an evaluation of two
standard encyclopedias of chemistry and pharmacyi2]we found
only 132 compounds with a n eight-, 23 with a nine-, 26 with a
ten-, 12 with an eleven-, 28 with a twelve-, and 6 with a thirteenmembered ring. Of these, 81 % are natural products and 73 %
are heterocycles. In medicinal chemistry particular 9- 12-membered lactams are of current interest as mimics of fi-turn~.[~I
this paper we introduce a new approach to the use of medium
and large ring systems.
Medium-sized carbocycles can be seen as lipophilic “clusters”
and should be bioisosteric to alkylphenyl moieties. In comparison to these they will bind more weakly to electrophilic protein
side chains, but because of their conformative flexibility
they should display better adaptation to binding domains of
protein^.'^] Medium-sized carbocycles also provide a far more
versatile framework than phenyl and “normal” rings (five- and
six-membered rings) for the regio- and stereodefined attachment
of substituents.
To test our concept we prepared compounds that are structurally derived from both arachidonic acid (1) (the physiological
substrate of cyclo- (COX) and lipoxygenases (LOX)) and inhibitors of these enzymes[51like 2-(alkylphenyl)propanoic acids
(for example ibuprofen (2)). In the active site of the enzyme[61
and in solution[71arachidonic acid adopts a bent conformation
as shown. A substrate with a medium-sized ring should at least
partly mimic the conformation of 1.
We have developed a synthetic route to 2-cycloalk(en)ylpropanoic acids of the general formula 3 (MR = medio-/macro-
[*] Priv.-Doz. Dr. P. Imming, DiplLChem. B. Greve
Institut fur Pharmazeutische Chemie der Universitdt
D-35032 Marburg (Germany)
Fax: h i . code +(6421)287052
e-mail . imming(a
Anges. Clremr. Int Ed Engl
1996, 35, No. 1I
Dr. S . Laufer
Merckle GmbH
Postfach 1161. D-89135 Blaubeuren (Germany)
This work was supported by the Fonds der Chemischen Industrie and the
Bundesministerium fur Forschung und Technologie. We thank Hiils AG
(Marl) and Firmenich S A . (Genf) for financial and material support.
8 V C H Verlu~sgesell.~chu~i
m h H . 0-69451
Weinhelm, 1996
OS70-0833/96~3Sl1-1221$ 1S.OOf . 3 ~ 0
Без категории
Размер файла
436 Кб
binucleata, stranded, phosphate, dna, nitrophenyl, double, bis, complexes, lanthanides, catalyst, hydrolysis
Пожаловаться на содержимое документа