Retention of inhibitory potency of an ACE inhibitor conjugated with Rh(III) and Pd(II) (iminophosphorano)phosphines. Synthesis and X-ray structural investigations

Article abstract of DOI:10.1021/ja9802403

Succinimido, amido, and ester functionalized tetrafluoroaryl azides selectively oxidize bisdiphenylphosphinomethane at one of the P(III) centers giving (iminophosphorano)phosphines 6, 7, and 8 respectively in high yields. Succinimido functionalized perflu

Full text of DOI:10.1021/ja9802403

11364
J. Am. Chem. Soc. 1998, 120, 11364-11373
Retention of Inhibitory Potency of an ACE Inhibitor Conjugated with
Rh(III) and Pd(II) (Iminophosphorano)phosphines. Synthesis and
X-ray Structural Investigations
Raghoottama S. Pandurangi,*,†,‡ Kattesh V. Katti,§ Loreen Stillwell,| and
Charles L. Barnes‡
Contribution from the Department of Internal Medicine, Department of Chemistry, and Department of
Radiology and Missouri UniVersity Research Reactor, UniVersity of Missouri, Columbia, Missouri 65211,
and Monsanto, Chesterfield Pkw. St. Louis, Missouri 64321
ReceiVed January 21, 1998
Abstract: Succinimido, amido, and ester functionalized tetrafluoroaryl azides selectively oxidize bisdiphe-
nylphosphinomethane at one of the P(III) centers giving (iminophosphorano)phosphines 6, 7, and 8 respectively
in high yields. Succinimido functionalized perfluoroaryl azido (iminophosphorano)phosphine is attached to
angiotensin converting enzyme (ACE) inhibitor, lisinopril at one end, leaving the other end for chelation to
Rh(III) and Pd(II) precursors including radioactive analogues establishing the hetero-bifunctionality for potential
in vivo tracking of the radiotracer. The measurement of inhibitory potency of lisinopril-metal conjugates
(Rh and Pd), modified through the primary amine reveals an increase in inhibitory potency, although small,
retaining the target potential of native lisinopril toward specific biological sites. However, direct complexation
utilizing the carboxylic groups of lisinopril with a Cu precursor resulted in the reduction of inhibitory potency
from nM to µM levels rendering it less useful for applications as an ACE inhibitor. Single-crystal X-ray
structural investigation of the Rh(III) perfluoroaryl (iminophosphorano)phosphine complex 12 shows a distorted
mer octahedral configuration with two ligands per metal center and only one of the phosphiniminato nitrogen
atom coordinating to the metal. Pd(II) complex 18 reveals that the metal is bound to the iminato nitrogen
atom and the P(III) center via cis disposition to form a five-membered ring. X-ray data for 12: triclinic, P-1,
a ) 11.570 (6) Å b ) 13.668 Å(7) c ) 20.709 (10) Å, R ) 86.068 (10), â ) 83.774 (10), γ ) 83.503 (10)
V ) 3229.6 (3) Å3, Z ) 2, R ) 0.028, Rw ) 0.050. X-ray data for 18: triclinic, P-1, a ) 11.457 (3) Å, b )
12.223 (3) Å, c ) 13.219 (4) Å, R ) 89.98 (20), â ) 73.710 (20), γ ) 69.980 (20), V ) 1665.7 (8) Å3, Z )
2, R ) 0.024, Rw ) 0.031.
Introduction
binding of ACE with fibrous tissue formation in the failing
human heart. An impressive array of ACE inhibitors (e.g.,
lisinopril, captopril, etc.) and their 125I analogues are shown to
inhibit the overexpression of ACE in vivo and to suppress
fibrosis in animal models.6 This observation indicates that ACE
inhibitors have a high target potential toward specific biological
sites rich in ACE. Functionalization of ACE inhibitors by
bifunctional chelating agents (BFCAs) carrying radiometals
(e.g., 99mTc or 105Rh) adds diagnostic potential to the already
established target potential of ACE inhibitors and may open a
new possibility of tracking them to specific biological sites in
vivo. Lisinopril, a lysine analogue of enalapril ([s]-Na-[1-
carboxy-3-phenylpropyl]-lys-prolin), is a clinically proven drug
for the treatment of hypertension7-9 and hence, attachment of
a clinically useful transition metal radiotracer permits non-
invasive monitoring of ACE-related activity in human beings
since 125I-lisinopril is limited to only animal studies.
Angiotensin converting enzyme (ACE, EC 3.4.15.1) is a zinc
metalloprotease enzyme which catalyzes the hydrolysis of
carboxy-terminal dipeptides from oligopeptide substrates.1-3
ACE is an important component of the renin-angiotensin
system (RAS) associated with the biological conversion of
angiotensin I (AI) to angiotensin II (AII), a vasoconstrictor
shown to be partly responsible for hypertension.4 Recent
studies5 have revealed the anatomic concordance of high-density
* To whom the correspondence to be addressed.
† Department of Internal Medicine, University of Missouri.
‡ Department of Chemistry, University of Missouri.
§ Department of Radiology and Missouri, University Research Reactor.
| Monsanto.
(1) (a) Johnson, C. I. Drugs 1990, 39, 21. (b) Campbell, D. J. J. Clin.
InVest. 1987, 79, 1. (c) Ehlers, M. R. W; Riordan J. F. Biochemistry 1990,
28, 5311. (d) Ehlers, M. R. W.; Riordan J. F. In Hypertension: Patho-
physiology, Diagnosis and Management; Laragh, J. H., Brenner, B. M.,
Eds.; Raven Press: New York, 1989. (e) Campbell, D. J. J. Clin. InVest.
1987, 79, 1. (f) Erdos, E. G. J. Lab. InVest. 1987, 56, 345 (g) Ondetti, M.
A.; Cushman D. W. Annu. ReV. Biochem. 1982, 51, 283. (h) Bunning, P.;
Holmquist; Riordan, J. F. Biochemistry 1983, 22, 103.
(2) (g) Soffer, R. L. Annu. ReV. Biochem. 1976, 45, 73. (h) Soffer, R. L.
In Biochemical Regulation of Blood Pressure; Soffer, R. L., Ed.; Wiley:
New York, 1981; pp 123-164.
(3) Ondetti, M. A.; Cushman, D. W. J. Med. Chem. 1981, 24, 355.
(4) Elisiva, Y. E.; Orekhovich, V. N.; Pavlikhina, L. V.; Alecsenco, L.
P. Clin. Chim. Acta 1971, 31, 413.
(5) Weber, K. T. Inflammation and Healing in the heart . In Wound
Healing in CardioVascular Disease; Weber, K. T., Ed.; Futura Publishing
Co. Inc: Armonk, NY, 1995; pp 1-21.
(6) (a) Sun, Y.; Cleutjens, J. P. M.; Diaz-Arias, A. A.; Weber K. T.
CardioVasc. Res. 1994, 28, 1423. (b) Sun, Y.; Ratajska, A.; Zhou, G.; Weber
K. T. J. Lab. Clin. Med. 1993, 122, 395.(c) Sun, Y.; Ratajska, A.; Zhou,
G.; Weber K. T. J. Lab. Clin. Med. 1995, 126, 95. (d) Sun, Y.; Weber K.
T. J. Mol. Cell Cardiol. 1996, 28, 851. (e) Sun, Y.; Weber K. T. J. Lab.
Clin. Med. 1996, 127, 94. (f) Weber, K. T.; Sun Y.; Katwa, L. C. and
Cleutjens, J. P. M. Basic Res. Cardiol. 1997, 92, 75.
10.1021/ja9802403 CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/22/1998

Retention of Inhibitory Potency of ACE Inhibitors
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11365
Cavel et al.10 have established the combination of phosphine
and iminophosphorane, produced by a selective oxidation of
one of the diphosphines, as versatile chelating agents to several
transition metals including Rh(I). The recent studies11 on the
extension of the coordination chemistry of iminophosphoranes
to the diagnostically useful transition metal 99mTc and their
stability in aqueous medium indicates that these conjugates are
potential in vivo tracking agents, provided they are conjugated
to a pharmacophore of interest. Since 99mTc has a short half-
life12 (6.5 h), we have used Rh- and Pd-based BFCAs for the
present studies. 105Rh has a low abundance of γ-rays (Eγ )
306 keV, 5%, 319 keV, 19%) which would allow in vivo
tracking of drugs13 at specific biological sites. However, the
major requirements for radiometalated ACE inhibitors to be
diagnostic agents include (a) the establishment of the coordina-
tion chemistry of Rh(III) including the geometrical configuration
of the chelate ring with respect to the binding sites of lisinopril
and (b) the confirmation of the retention of the inhibitory
potency of ACE inhibitors, functionalized by relatively bulky
and electron-rich metalated chelating agents. In our earlier
studies, we have shown that functionalization of lisinopril at
the primary and secondary amine groups by tetrafluoroaryl
azides has improved their binding potency, assessed by auto-
radiography studies,14 competitive radioligand binding assays,14
molecular dynamics calculations,15 and IC50 measurements using
amydolytic assays.15 Since attachment of perfluoroaryl azides
to lisinopril shows promising inhibitory potency, further func-
tionalization of azides by phosphines provides a needed lisi-
nopril-BFCA conjugate ready to incorporate transition metals
and analogous radiometals. Selective oxidation of one of the
phosphines of diphosphines provides asymmetrically substituted
(iminophosphorano)phosphine combination which is expected
to improve the stability of the complex through bidentate
coordination to metal.16,17
Our interest here is to identify the best site on lisinopril for
further functionalization and to test the effect of metalated
chelating frameworks at different sites of lisinopril (primary
amine group, secondary amine group, carboxylic groups, etc.)
on the inhibitory potency of lisinopril conjugates in the post
modification. Although Rh(III) complexes with phosphines are
known16,17 and Rh(I)-(iminophosphorano)phosphine complexes
are quite common,18 surprisingly, Rh(III)-(iminophosphorano)-
phosphines are uncommon. Since 105Rh is available in a
virtually “no carrier added” version as Rh(III)Cl3‚xH2O, the
ultimate success of the ACE-inhibitor avid radiopharmaceutical
as an in vivo tracking agent depends on the precise structural
determination of the nonradioactive Rh(III) complex in solid
form and its conformation in solution. Hence, as a part of the
ongoing research on BFCAs19 and bifunctional photolabile
chelating agents (BFPCAs),20 we wish to elucidate the X-ray
structural investigations on representative Rh(III) and Pd(II)
perfluoroaryl functionalized (iminophosphorano)phosphines,
supplemented by multinuclear and 2D NMR of functionalized
lisinopril derivatives. We also would like to establish the
retention of the inhibitory potency at nM level for primary amine
functionalized Rh(III) and Pd(II) conjugates, in contrast to
functionalization at carboxylic groups with a Cu complex which
resulted in the significant loss of activity. These findings
constitute a first report on the biological activity of lisinopril,
tailored by metalated chelating agents.
(7) (a) Bohacek, R.; Lombaert, S. D. McMartin, C.; Priestle, J.; Grutter,
M. J. Am. Chem. Soc. 1996, 118, 8231. (b) Wyvratt, M. J.; Patchette, A.
A. Med. Res. ReV. 1985, 5, 483. (c) Opie, L. H. Angiotensin ConVerting
Enzyme Inhibitors; Wiley-Liss: New York, 1992. (d) Squire, I. B.; Lees,
K. R. ReV. Contemp. Pharmacother. 1994, 4, 45. (e) Lawton, G.; Pacioreck,
P. M.; Waterfall, J. F. The design and biological profile of ACE inhibitors.
In AdVances in Drug Research; Jovanovich, H. B., Ed.; Academic Press:
New York, 1992; Vol. 23, pp 161-220.
(8) (a) Edwards, C. R. W.; Padfield, P. L. Lancet 1985, 30. (b) Johnston,
C. I. Med. J. Aust. 1988, 148, 488. (c) Patchett, A. A. Enalapril and
Lisinopril. In Chronicles of Drug DiscoVery; Lednicer, D., Ed.; ACS
Professional Reference Book, American Chemical Society: Washington,
DC, 1993; Vol. 3, pp 125-162. (d) Goa, K. L.; Balfour, J. A.; Zuanetti, G.
Drugs 1996, 52, 564.
(9) (a) Mendelsohn, F. A. O. Clin. Exp. Pharmacol. Physiol. 1984, 11,
431. (b) Reid, J. L. J. CardioVasc. Pharmacol. 1993, 22 (suppl), s 41.
(10) (a) Reed, R. W.; Santarsiero, B.; Cavell, R. G. Inorg. Chem. 1996,
35, 4292. (b) Li, J.; McDonald, R.; Cavell R. G. Organometallics 1996,
15, 1033. (c) Balakrishna, M. S.; Klein, R.; Uhlenbrock, S.; Pinkerton, A.;
Cavell R. G. Inorg. Chem. 1993, 32, 5676. (d) Li, J.; Pinkerton, A. A.;
Finnen, D. C.; Kumar, M.; Martin, A.; Weisemann, F.; Cavel, R. G. Inorg.
Chem. 1996, 35, 5684.
Experimental Section
All synthetic procedures were conducted in a dry nitrogen atmosphere
using standard Schlenk tube techniques and prepurified solvents.
Reactions involving the synthesis of azides were carried out in a
subdued light by wrapping the flasks with aluminum foil. Nuclear
magnetic resonance spectra were recorded in CDCl3 on a Bruker WH-
300 spectrometer, and chemical shifts are reported in ppm downfield
1
from SiMe4 for H NMR. The 31P NMR chemical shifts are reported
with respect to 85% H3PO4 as an external standard, and positive shifts
lie downfield from the standard. 19F NMR chemical shifts are reported
(11) (a) Katti, K. V. Singh, P. R. Katti, K. K.; Barnes, C. L.; Kopica,
K.; Ketring, A. R.; Volkert, W. A. Phosphorus, Sulfur Silicon Relat. Elem.
1993, 55. (b) Jurisson S.; Eble, E.; Berning, D.; Barnes, C. L.; Katti, K. V.
Phosphine Complexes of Technetium (VII). In Chemistry and Nuclear
Medicine; Nicolini, M., Bandole, G.. Mazzi, U., Eds.; SGE: Padova, 1995;
pp 201-203. (c) Katti, K. V.; Singh, P. R.; Katti, K. K. Kopica, K.; Volkert,
W. A.; Ketring, A. R. J. Labelled Compd. Radiopharm. 1993, 22, 407.
(12) Volkert, W. A.; Jurisson, S. Technetium-99m chelates as radio-
pharmaceuticals. In Technetium and Rhenium their chemistry and it’s
applications; Topics in Current Chemistry; Ed. by Yoshihara. K., Omori,
T., Eds.; Springer Verlag: Berlin, Heidelberg, 1996; Chapter 4, pp 123-
148. (b) Volkert, W. A. Ligand systems useful in designing high specific
activity 99mTc or 186Re, 188Re radiopharmaceuticals. In Technetium and
Rhenium in Chemistry and nuclear Medicine, 4th ed.; Nicolini, M., Mazzi,
U., Eds.; Cortina International: Verona, Italy, 1995; pp 239-242. (c)
Jurisson, S.; Berning, D.; Jia, W.; Ma, D. Chem. ReV. 1993, 93, 1137. (e)
Meares, C. F. Nucl. Med. Biol. 1986, 13, 311. (e) Schwochau, K. Angew.
Chem., Int. Ed. Engl. 1994, 33, 2258.
(16) (a) Jardine, F. H.; Sheridan, P. S. In ComprehensiVe Coordination
Chemistry; Eds. Wilkinson, G., Gillard, R. D., McCleverty, J. A., Eds.;
Pergamon: Oxford, 1987; Vol. 4, Chapter 48.
(17) Mayer, H. A.; Kaska, W. C. Chem. ReV. 1994, 94, 1239.
(18) (a) Bader, A.; Lindner, E. Cord. Chem. ReV. 1991, 108, 27. (b)
Katti, K. V.; Cavell, R. G. Comm. Inorg. Chem. 1990, 10, 53. (c) Katti, K.
V.; Cavell, R. G. Organometallics 1988, 7, 2236. (d) Katti, K.V; Cavell,
R. G. Inorg. Chem. 1989, 28, 413. (e) Katti, K.V.; Cavell, R. G. Inorg.
Chem. 1989, 28, 3033. (f) Katti, K. V.; Cavell, R. G. Organometallics 1989,
8, 2147. (g) Katti, K. V.; Batchelor, R. J.; Einstein, F. W. B.; Cavell, R. G.
Inorg.Chem. 1989, 29, 808. (h) Liu, C. Y.; Chen, D. Y.; Cheng, M. C.;
Peng, S. M.; Liu, S. T. Organometallics 1995, 14, 1983. (i) Weight A.;
Bischoff, S. Phosphorus, Sulfur Silicon Relat. Elem. 1995, 102, 91. (j) Katti,
K. V.; Santarseiro, B. D.; Pinkerton, A. A.; Cavell R. G. Inorg. Chem.
1993, 32, 5919. (k) Law, D. J.; Bigam, G.; Cavel R. G. Can. J. Chem. ReV.
Can. De Chimie. 1995, 73, 635.
(19) (a) Pandurangi, R. S.; Karra, S. R.; Kuntz, R. R.; Volkert, W. A.
Photochem. Photobiol. 1997, 65, 101. (b) Pandurangi, R. S.; Kuntz, R. R.;
Volkert, W. A.; Barnes, C. L.; Katti, K. V. J. Chem. Soc., Dalton Trans.,
1995, 565. (c) Pandurangi, R. S.; Karra, S. R.; Kuntz, R. R.; Volkert, W.
A. Bioconjugate Chem. 1995, 6, 630. (d) Pandurangi, R. S.; Kuntz, R. R.;
Volkert, W. A. Appl. Radiat. Isot. 1995, 46, 233. (e) Pandurangi, R. S.;
Karra, S. R.; Katti, K. V.; Volkert, W. A.; Kuntz, R. R. J. Org. Chem.
1997, 62, 2587. (f) Pandurangi, R. S.; Karra, S. R.; Kuntz, R. R.; Volkert,
W. A. Photochem. Photobiol. 1996, 64, 100. (g) Pandurangi, R. S.; Katti,
K. V.; Volkert, W. A.; Kuntz, R. R. Inorganic Chemistry 1996, 35, 3716.
(13) Ehrhardt, G.J; Ketring, A. R.; Volkert, W. A. A production of
isotopes at nuclear reactors. In Synthesis and Applications of Isotopically
Labeled Compounds; Buncel, E., Kabalka, G. W., Eds.; Elsevier Science:
The Netherlands, 1991; pp 159-164. (b) Goswami, N.; Alberto, R.; Barnes,
C. L.; Jurisson, S. Inorg. Chem. 1996, 35, 7546.
(14) Pandurangi, R. S.; Kuntz, R. R.; Sun, Y.; Weber K. T. Bioorg. Chem.
1997, 57, 75.
(15) Pandurangi, R. S.; Rao, S. N.; Stillwell, L; Barnes, C. L.; Kuntz R.
R. J. Am. Chem. Soc., submitted.

11366 J. Am. Chem. Soc., Vol. 120, No. 44, 1998
Pandurangi et al.
Table 1. Summary of the Crystallographic Data for Rh Complex
12 and Pd Complex 18
from ref 21(a-g). All hydrogen atoms in both of the structures were
located in difference Fourier maps and refined with fixed isotropic
thermal parameters. Listing of full experimental details, coordinates,
temperature factors, and anisotropic temperature factors are deposited
as Supporting Information.
formula
MW
C34H27N3OF4P2Cl2Pd C66H50N2O4F8P4Cl3Rh‚CH3CN
808.85
1461.32
0.30 × 0.35 × 0.35
triclinic
crystal size (mm) 0.15 0.15 × 0.35
Synthesis of Perfluoro Functionalized Phosphinimines 6-10. A
solution of N-succinimidyl 4-azido 2,3,5,6-tetrafluorobenzoate (0.488
g, 1.47 mmol) in chloroform was added dropwise to a solution of
diphenylphosphinomethane (0.566 g, 1.47 mmol) cooled to 0 °C under
dry nitrogen atmosphere. The solution turned yellow, and the mixture
was stirred for 6 h before the solvent was removed in a vacuum to
give a pale yellow crystalline material. Recrystallization was carried
out by dissolving the crude material in hot acetonitrile and cooled at 0
°C to get analytically pure material 6 (87.1% yield). 31P NMR (CDCl3)
δ 11.3 (dt, 2JPP ) 54.1 Hz, 4JPF ) 5.6 Hz), -30.5 (d, 2JPP ) 54.6 Hz).
1H NMR (CDCl3) δ 2.86 (s, 4H), 3.39 (dd, 2H, 2JPH ) 12.2 Hz), 7.1-
7.9 (m, 20H). 19F NMR (CDCl3) δ -138.81 (m, 2F), -154.05 (m,
2F). Anal. Calcd for C36H26N2F4P2O4: C, 62.78; H, 3.81; N, 4.07.
Found: C, 62.12; H, 3.87; N, 4.15.
crystal system
space group
a (Å)
triclinic
P-1
P-1
11.457(3)
12.223(3)
13.219(4)
87.980(20)
73.710(20)
69.980(20)
1665.7(8)
810.85
11.5702(6)
13.6685(7)
20.7090(10)
86.0680(10)
83.7740(10)
83.5030(10)
3229.6(3)
1483.5
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
U (Å3)
F(000)
Z
2
2
D
calc (g/cm3)
2θmax (deg)
µ (cm-1
1.613
1.503
45.9
54.0
)
8.5
5.5
hkl range
-11 to 12, 0 to 13,
-14 to 14
4911
4636
0.024
-14, 14, 0, 17,
-26, 26
19282
13299
0.028
A solution of 4-azido 2,3,5,6-tetrafluoromethylbenzoate (2.5 g, 10.03
mmol) in chloroform was added dropwise to a solution of diphe-
nylphosphinomethane (3.86 g, 10.01 mmol) cooled to 0 °C under dry
nitrogen atmosphere. The solution turned yellow, and the mixture was
stirred for 6 h before the solvent was removed in a vacuum to obtain
a pale yellow solid which was recrystallized by dissolving it in hot
acetonitrile followed by slow cooling at room temperature to obtain 7
total data
unique data
Ra
Rb
0.031
0.050
2
a R ) ∑(|Fo| - |Fc|)/∑|Fo|. b R′ ) [∑w(|Fo| - |Fc|)2/∑|Fo| )+, w-1
) [σ|Fo| + 0.0005(Fo)2].
2
4
(82.4% yield). 31P NMR (CDCl3) δ 11.8 (dt, JPP ) 53.5 Hz, JPF
)
6.6 Hz). -30.5 (d, JPP ) 53.6 Hz). 1H NMR (CDCl3) δ 3.36 (dd,
2
with respect to CFCl3 as an external standard. Elemental analysis for
the new compounds were done by Oneida Research Services, Inc., New
York. Syntheses of functionalized perfluoroaryl azides (1-5) were
accomplished by the methods reported earlier.10,19 In general, organic
azides are potentially explosive, and caution must be taken while
preparing them.10,17 Metallic precursors, RhCl3xH2O and PdCl2-
(PhCN)2, were obtained from Aldrich. Lisinopril was a gift from Merck
and was used as such without further purification.
ACE Binding Assays. Estimation of the inhibitory potency of all
lisinopril derivatives was measured using amydolyic assays.1i About
10 µL of 0.25 µg/mL ACE (from rabbit lung) in 30 mL of buffer (50
mM HEPES containing 300 mM NaCl at pH 7.5) was mixed with
various concentrations of lisinopril derivatives (10-12 M -10-6 M) and
the mixture was incubated at room temperature for 10 min. The
cleaving reaction was initiated with the addition of 40 µL of substrate
(1.25 mM N-(3-[2-furyl}acryloyl)-Phe-Gly-Gly) and monitored continu-
ously for 8 min at 340 nm using a Molecular Devices ThermoMAX
plate reader. Cleavage of the substrate by the enzyme results in a
product which has an absorbance at 340 nm. Binding of new lisinopril
derivatives to ACE changes the cleaving potential of ACE which is
reflected in the reduction of absorbance at 340 nm. The relative
integration of the areas under the curve is used for computing the
relative IC50 values of new lisinopril derivatives.
2H, 2J
)12.3 Hz), 3.87 (s, 3H), 7.1-7.9 (m, 20H). 19F NMR
PH
(CDCl3) δ -144.00 (m, 2F), -154.80 (m, 2F). Anal. Calcd for C33H25-
NF4P2O2: C, 65.45, H, 4.13; N, 2.31. Found: C, 66.15; H, 4.15; N,
2.11.
A solution of 4-azido 2,3,5,6-tetrafluorobenzamide (2.32 g, 10.99
mmol) in toluene was slowly added to diphenylphosphinomethane
(dppm, 4.22 g, 10.97 mmol) in tolune at 0 °C. The reaction mixture
was stirred for 3 h and monitored by 31P NMR spectroscopy. The
solvent was removed in a vacuum to obtain a crystalline solid which
was dissolved in hot acetonitrile and recrystallized in the presence of
hexane to obtain a pure solid 8 (78.5% yield). 31P NMR (CDCl3) δ
11.3 (dt, 2JPP ) 54.1 Hz, 4JPF ) 5.6 Hz), -30.5 (d, 2JPP ) 54.6 Hz). 1H
2
NMR (CDCl3) δ 6.61, 6.56 (s, 2H), 3.39 (dd, 2H, JPH ) 12.2 Hz),
7.1-7.9 (m, 20H). 19F NMR (CDCl3) δ -138.81 (m, 2F), -154.05
(m, 2F). Anal. Calcd for C32H24N2F4P2O: C, 65.12; H, 4.06; N, 4.74.
Found: C, 65.55, H, 4.15; N, 4.51.
Analytical Data for 9: (92.4% yield). 31P NMR (CDCl3) δ 11.6
2
4
2
(dt, JPP ) 53.1 Hz, JPF ) 4.8 Hz). -30.2 (d, JPP ) 53.7 Hz). 1H
NMR (CDCl3) δ 3.28 (dd, 2H, 2JPH ) 12.8 Hz), 7.0-7.8 (m, 20H). 19
NMR (CDCl3) δ -56.8 (t, 3F, JFF ) 21.6 Hz), - 144.23 (m, 2F),
-155.21 (m, 2F). Anal. Calcd for C32H22NF7P2: C, 62.43; H, 3.60;
N, 2.28. Found: C, 62.01; H, 3.67; N, 2.15.
F
4
X-ray Data Collection and Processing. The crystal data and the
details of data collection for crystals of 12 and 18 are listed in Table
1. X-ray data were collected on an Enraf-Nonius CAD-4 diffracto-
meter20,21 with Mo KR radiation and a graphite monochromator at 22
(1) °C. The cell dimensions were obtained from a least-squares fit to
setting angles of 25 reflections with 2θ in the range 20.0-30.0°. The
crystal exhibited no significant decay under X-ray irradiation.
The structures were solved by direct methods and were subsequently
refined by full matrix least-squares method which minimizes Σw/(|F0|
- |Fc|)2 where w-1 ) [σ(counting) + (0.008(F0)2)/4F0]. Atomic
scattering factors including anomalous scattering contributions were
A solution of 4-azido 2,3,5,6-tetrafluorobenzamido lisinopril (25.95
mg, 0.032 mmol) in DMF was slowly added to diphenylphos-
phinomethane (12.4 mg, 0.032 mmol) in dry DMF at 0 °C. The reaction
mixture was stirred for 3 h and monitored by 31P NMR spectroscopy.
The solvent was removed in a vacuum to obtain a pasty solid which
was dissolved in CH2Cl2 and recrystallized to obtain solid 10.
Analytical Data for 10: (76.5% yield). 31P NMR (CDCl3) δ 11.4
2
4
2
(dt, JPP ) 45.3 Hz, JPF ) 6.1 Hz), -29.2 (d, JPP ) 45.8 Hz). 1H
2
NMR (CDCl3) δ 3.39 (dd, 2H, JPH ) 12.1 Hz), 7.1-7.9 (m, 20H).
All other resonances coincide with native lisinopril. 19F NMR (CDCl3)
δ -143.61 (m, 2F), -155.82 (m, 2F). Anal. Calcd for C53H52N4-
F4P2O6: C, 65.01; H, 5.36; N, 5.73 Found: C, 65.55, H, 5.23; N, 5.61.
Synthesis of Rh(III) Complexes 11-15. To RhCl3 in chloroform
(0.148 g, 0.71 mmol) were added a few drops of alcohol. The mixture
was heated until it dissolved. The hot RhCl3 solution was added slowly
to a solution of 6 (0.487 g, 0.71 mmol) in chloroform, refluxed for 3
h, and monitored by 31P NMR spectroscopy until the disappearance of
phosphorus signals due to ligand 6. The solvent was removed by a
flash evaporator, and the orange-red solid was dissolved in a mixture
of dichloromethane and hexane (8:2) and cooled at 0 °C to obtain a
crystalline material 11 (78.5% yield). 31P NMR (CDCl3) δ 7.43 (dt,
(20) Pandurangi, R. S.; Lusiak, P.; Kuntz, R. R.; Volkert, W. A.;
Rogowski, J.; Platz, M. S. J. Org. Chem. 1998, in press.
(21) The following references are relevant to the NRCVAX system: (a)
Gabe, E. G.; Page, Y. L.; Charland, J. L.; Lee, F. L.; White, P. S. J. Appl.
Crystallogr. 22, 384, 1989. (b) Flack, L. Acta Crystallogr. Sec. A 39, 876
1983. (c) Johnson C. K.; ORTEP- A Fortran Thermal Ellipsoid Plot
Program, Technical Report ORNL-5138, Oak Ridge, 1976. (d) Larson A.
C. Crystallographic Computing; Munksgaard: Copenhagen, 1970; p 293.
(e) Page, Y. L. J. Appl. Crystallogr. 1988, 21, 983. (f) Page, Y. L.; Gabe
E. J. J. Appl. Crystallogr. 1979, 12, 464. (g) Rogers, D. Acta Crystallogr.
Sect. A 1981, 37, 7.

Retention of Inhibitory Potency of ACE Inhibitors
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11367
4
1
2
2JPP ) 30.5 Hz, JPF ) 5.3 Hz), 16.5 (m, JP-Rh )115.8 Hz, JPP
)
)
NMR (CDCl3) δ -141.10 (m, 2F), -152.80 (m, 2F). Anal. Calcd
for C33H25NF4P2O2 PdCl2: C, 50.71; H, 3.23; N, 1.79. Found: C,
50.23; H, 3.29; N, 1.82.
Synthesis of 18. A solution of PdCl2 (PhCN)2 (0.348 g, 90.7 mmol)
in dichloromethane (50 mL) was slowly added to a solution of ligand
8 (0.520 g, 91.6 mmol) also in dichloromethane at 25 °C. After stirring
for an hour, the solution was partially evaporated in a vacuum and
treated with n-hexane to precipitate the Pd complex 18. The solution
was filtered, and the precipitate was washed (3 × 20 mL) with n-hexane
and dissolved in hot acetonitrile. After slow evaporation of the solvent,
24.2 Hz), 24.2 (m, 1JP-Rh )115.4 Hz, 2JPP ) 24.6 Hz), 44.1 (dt, 2JPP
15.2 Hz), 1H NMR (CDCl3) δ 2.81 (s, 4H), 4.18 (dd, 2H, 2JPH ) 11.6
Hz), 5.20 (dd, 2H, JPH ) 10.4 Hz), 6.6-7.6 (m, 40H). 19F NMR
2
(CDCl3) δ -132.2 (m, 2F), -140.7 (m, 2F), -141.4 (m, 2F) -154.1
(m, 2F). 13C NMR (CDCl3) 28.4 (m, JC-P ) 26.3 Hz) and 37.1 (m,
1
1JC-P ) 28.3 Hz). Anal. Calcd for C72H52N4F8P4O8 RhCl3: C, 54.54;
H, 3.31; N, 3.54. Found: C, 55.11; H, 3.65; N, 3.61.
RhCl3 (0.069 g, 0.329 mmol) was added to chloroform along with
a few drops of alcohol and heated till it dissolved. The hot solution of
RhCl3 was added slowly to a solution of 7 (0.200 g, 0.329 mmol) in
chloroform, refluxed for 3 h, and monitored by 31P NMR spectroscopy
until the disappearance of phosphorus signals due to ligand 7. The
solvent was removed by a flash evaporator to obtain an orange-red
solid 12. To get X-ray quality crystals, the solid was redissolved in
hot CHCl3 and slowly evaporated over a layer of hexane at room
temperature (86.5% yield). 31P NMR (CDCl3) δ 6.15 (dt, 2JPP ) 30.3
Hz, 4JPF ) 5.3 Hz), 16.7 (m, 1JP-Rh )109.1 Hz, 2JPP ) 27.1 Hz), 24.4
orange-red crystals were obtained for further X-ray studies. (78.1%
2
yield), (PdN) ) 1245 cm-1
.
31P NMR (CD2Cl2) δ 27.7 (m, JPP
)
27.8 Hz), -53.2 (b, 2JPP ) 27.7 Hz). 1H NMR (CD2Cl2) δ 5.61, 6.22
(s, 2H), 7.1-8.0 (m, 20H), 3.38 (dd, 2H, JPH ) 12.3 Hz), 19F NMR
2
(CD2Cl2) δ -141.2 (m, 2F), -153.15 (m, 2F). Anal. Calcd for
C32H24N2F4P2PdCl2‚CH3CN: C, 50.56; H, 3.37; N, 5.21. Found: C,
50.38, H, 3.28; N, 5.36.
Analytical Data for 19: 31P NMR (CD2Cl2) δ 26.7 (m, 2JPP ) 25.8
1
2
2
4
2
Hz), -56.3 (b, JPP ) 25.2 Hz). 1H NMR (CD2Cl2) δ 3.56 (dd, 2H,
(m, JP-Rh )111.9 Hz, JPP ) 23.5 Hz), 42.3 (dt, JPP ) 15.8 Hz, JPF
) 4.85 Hz). 1H NMR (CDCl3) δ 3.89 (s, 3H), 3.97 (s, 3H), 4.01 (m,
2JPH ) 12.8 Hz), 7.1-7.9 (m, 20H). 19F NMR (CD2Cl2) δ -55.9 (t,
3F 4JFF ) 19.8 Hz), - 140.23 (m, 2F), -151.42 (m, 2F). Anal. Calcd
for C32H22NF7P2 PdCl2: C, 48.55; H, 2.81; N, 1.77. Found: C, 48.48;
2H), 5.18 (dd, 2H, JPH ) 12.3 Hz), 6.5-7.8 (m, 40H). 19F NMR
2
(CDCl3) δ -132.9 (m, 2F), -140.9 (m, 2F), -143.8 (m, 2F) -153.9
(m, 2F). 13C NMR (CDCl3) 27.4 (m, JC-P ) 24.8 Hz) and 36.1 (m,
1
H, 2.79; N, 1.83.
1JC-P ) 29.1 Hz). Anal. Calcd for C68H53N3F8P4O4 RhCl3. CH3CN:
C, 55.93; H, 3.66; N, 2.88. Found: C, 55.32; H, 3.51; N, 2.81.
Analytical data for 13: (84.5% yield). 31P NMR (CDCl3) δ 6.83
2
Analytical Data for 20: 31P NMR (CD2Cl2) δ 28.1.7 (m, JPP
)
25.8 Hz), -54.3 (b, JPP ) 27.2 Hz). 1H NMR (CD2Cl2) δ 3.46 (dd,
2H, 2JPH ) 11.8 Hz), 7.1-7.9 (m, 20H). 19F NMR (CD2Cl2) δ -140.61
(m, 2F), -152.32 (m, 2F). Anal. Calcd for C53H52N4F4P2O6PdCl2:
C, 55.10; H, 4.54; N, 4.85. Found: C, 55.16, H, 4.60; N, 4.78.
2
(dt, 2JPP ) 32.5 Hz, 4JPF ) 5.6 Hz), 17.5 (m, 1JP-Rh )118.7 Hz, 2JPP
23.8 Hz), 23.4 (m, 1JP-Rh )131.5 Hz, 2JPP ) 27.1 Hz), 43.1 (dt, 2JPP
)
)
14.2 Hz, 4JPF ) 5.1 Hz). 1H NMR (CDCl3) δ 6.76, 6.56 (s, 2H), 4.12
(dd, 2H, 2JPH ) 11.3 Hz), 5.28 (dd, 2H, 2JPH ) 10.8 Hz), 6.6-7.6 (m,
40H). 19F NMR (CDCl3) δ -131.9 (m, 2F), -140.4 (m, 2F), -142.6
Results and Discussion
(m, 2F) -155.5 (m, 2F). 13C NMR (CDCl3) 28.9 (m, JC-P ) 22.5
1
I. Synthesis, Spectroscopy, and Coordination Chemistry
of Rh (III) and Pd (II) Perfluoroaryl (iminophosphorano)-
phosphines. Iminophosphoranes are known to display versatile
bonding interactions in both coordination and organometallic
chemistry.17 The incorporation of an iminophosphorane moiety
with the other donor atoms to form a bidentate chelating
framework or selective oxidation of one of the arms of
diphosphines leading to asymmetric (iminophosphorano)phos-
phine introduces mixed ligating centers P(III) and P(V) and has
been a subject of extensive research in the last 7 years.17-22
For example, reactions of functionalized perfluoroaryl azides
(R-C6F4N3, R ) succinimide, COOMe, CONH2, and CF3) with
diphenylphosphinomethane (dppm) give the corresponding
monosubstituted (iminophosphorano)phosphine in high yields
(> 90%, Scheme 1). Generally, these reactions are selective
in oxidizing only one P(III) center in dppm. In fact, some of
the functionalized aryl azides (R ) CONH2, CF3, COOMe etc.)
when used 2-3-fold excess over dppm produce mono-oxidized
derivatives in >85% yield. Succinimido functionalized per-
fluoroaryl azides are useful precursors of BFCAs since they can
be conjugated to a biomolecule or pharmacophore at one end
and can be derivatized by a chelating framework at the other
end. For example, succinimido perfluoroaryl azide 1 is
conjugated to lisinopril at one end, and at the other end, N3 is
oxidized selectively by one of the phosphorus of dppm to yield
10 in good yields (>75%). All of the iminophosphoranes are
characterized by multinuclear NMR (31P, 1H, 13C, and 19F), IR,
and elemental analysis.
Hz) and 35.8 (m, 1JC-P ) 29.1 Hz). Anal. Calcd for C64H48N4F8P4O2-
RhCl3: C, 55.33; H, 3.49; N, 4.04. Found: C, 55.28, H, 3.54; N, 4.09.
Analytical data for 14: (77.5% yield). 31P NMR (CDCl3) δ 6.89
(dt, 2JPP ) 29.5 Hz, 4JPF ) 5.7 Hz), 15.8 (m, 1JP-Rh )118.5 Hz, 2JPP
23.9 Hz), 24.8 (m, 1JP-Rh )133.2 Hz, 2JPP ) 28.6 Hz), 42.1 (dt, 2JPP
)
)
)
16.2 Hz, JPF ) 4.9 Hz). 1H NMR (CDCl3) δ 4.26 (dd, 2H, JPH
4
2
11.3 Hz), 5.13 (dd, 2H, 2JPH ) 11.8 Hz), 6.6-7.6 (m, 40H). 19F NMR
(CDCl3) δ -56.5 (t, 3H, 4JFF ) 18.3 Hz), -131.9 (m, 2F), -141.1 (m,
2F), -142.4 (m, 2F) -155.8 (m, 2F). 13C NMR (CDCl3) 26.4 (m,
1JC-P ) 25.3 Hz) and 35.8 (m, JC-P ) 28.0 Hz). Anal. Calcd for
1
C64H44N2F14P4RhCl3: C, 53.41; H, 3.08; N, 1.95. Found: C, 53.35;
H, 3.13; N, 1.88.
Analytical data for 15: (72.5% yield). 31P NMR (CDCl3) δ 6.53
(dt, 2JPP ) 31.3 Hz, 4JPF ) 4.8 Hz), 15.8 (m, 1JP-Rh )113.7 Hz, 2JPP
24.8 Hz), 24.7 (m, 1JP-Rh )132.3 Hz, 2JPP ) 26.6 Hz), 42.9 (dt, 2JPP
)
)
)
15.8 Hz, JPF ) 4.9 Hz). 1H NMR (CDCl3) δ 4.32 (m, 2H, JPH
4
2
12.6 Hz), 5.10 (m, 2H, 2JPH ) 11.6 Hz), 6.6-7.6 (m, 40H). All other
resonances coincide with native lisinopril. 19F NMR (CDCl3) δ -133.4
(m, 2F), -141.2 (m, 2F), -142.4 (m, 2F) -155.8 (m, 2F). 13C NMR
1
1
(CDCl3) 28.1 (m, JC-P ) 23.7 Hz) and 37.9 (m, JC-P ) 29.9 Hz).
Anal. Calcd for C106H104N8F8P4O12: C, 58.77; H, 4.84; N, 5.18.
Found: C, 58.68; H, 4.79; N, 5.24.
Synthesis of Pd(II) Complexes 16-20. A solution of PdCl2
(PhCN)2 (0.139 g, 36 mmol) in dichloromethane (50 mL) was slowly
added to a solution of 6 (0.250 g, 36 mmol) also in dichloromethane
at 25 °C. After stirring for an hour, the solution was partially
evaporated in a vacuum and treated with n-hexane to precipitate the
Pd complex 16. The solution was filtered, and the precipitate was
washed (3 × 20 mL) with n-hexane and dissolved in hot acetonitrile.
After slow evaporation of the solvent, orange-red crystalline material
was obtained (78.6% yield). 31P NMR (CD2Cl2) δ 25.7 (m, JPP
)
2
II. X-ray Structural Investigations. Single-crystal X-ray
structural investigations on the Rh(III) complex 12 and Pd(II)
complex 18 are determined to confirm the conclusions drawn
27.8 Hz), -56.3 (b, JPP ) 27.3 Hz). 1H NMR (CDCl3) δ 2.89 (s,
4H), 3.42 (dd, 2H, 2JPH ) 13.2 Hz), 7.1-7.9 (m, 20H). 19F NMR (CD2-
Cl2) δ -139.91 (m, 2F), -152.05 (m, 2F). Anal. Calcd for
C36H26N2F4P2O4PdCl2: C, 50.00; H, 3.03; N, 3.24. Found: C, 50.09;
H, 3.07; N, 3.28.
2
(22) (a) Miles, J. S.; Visscher, M. O.; Carlton, K. G. Inorg. Chem. 1974,
13, 1632. (b) Imhoff, P.; Elsevier: C. J.; Stam C. H. Inorg. Chim. Acta
1990, 175, 209. (c) Imhoff, P.; Asselt, R. V.; Elsevier: C. J.; Zoutberg, M.
C.; Stam C. H. Inorg. Chim. Acta 1991, 184, 73. (d) Imhoff, P.; Elsevier,
C. J. J. Organomet. Chem. 1989, 369, c61. (e) Imhoff, P.; Sylvia, C. A.;
Elsevier, C. J. Organometallics 1991, 10, 1421.
Analytical Data for 17: (82.1% yield). 31P NMR (CD2Cl2) δ 23.7
(m, 2JPP ) 27.8 Hz), -58.3 (b, 2JPP ) 27.3 Hz). 1H NMR (CD2Cl2) δ
3.91 (s, 3H), 3.36 (dd, 2H, JPH ) 13.8 Hz), 7.1-7.9 (m, 20H). 19F
2

11368 J. Am. Chem. Soc., Vol. 120, No. 44, 1998
Pandurangi et al.
Scheme 1
[133.36° (8)] on the noncoordinated arm of the other imino-
phosphorane. The bond length for PdN involved in chelation
[(1.606 Å(17)] is slightly elongated compared to free PdN
[(1.578 Å (18)]. Similarly, the bond length N(2)-C(59) [1.411
Å (3)] on the coordinating part is also longer than the bond
length N(1)-C(26) [1.352 Å(3)] on the noncoordinate part of
iminophosphorane. In the uncomplexed iminophosphorane arm,
the lone pair of electrons on nitrogen may delocalize through
the perfluoroaryl group which may be responsible for the larger
angle for P(2)-N(1)-C(26) [133.36° (8)] and shorter bond
length for N(1)-C(26) [1.352 Å(3)], leaning toward linear
geometry. However, the trigonal planar geometry around
nitrogen center on the complex arm [P(4)-N(2)-C(59) [120.56°
(13)] indicates the directional donation of the lone pair of sp2
nitrogen to Rh metal. The similar lengthening of PdN bond
lengths and reduction in bond angles is well-reported for Rh(I)
complexes.18,20 The Rh-Cl and Rh-P bond lengths in complex
12 can be discussed in relation to the trans influence of the
chlorine and the iminophosphorane PdN group. For example,
the P(3)-Rh bond [(2.305 Å (5)], being trans to chlorine is
slightly longer compared to the P(1)-Rh bond [(2.289 Å (5)],
reflecting the small, yet noticeable relative trans influence of
the chlorine and PdN moiety, respectively. The Rh-Cl(2) bond
trans to phosphorus [2.384 Å (5)] is longer than Rh-Cl trans
to chloride (e.g., Rh-Cl3, 2.349 Å or Rh-Cl1, 2.365 Å)
indicating that a larger trans influence for Rh-Cl(2) to
phosphorus may be due to Rh-P π bonding. The elongation
of Rh-Cl(2) compared to that of Rh-Cl(1) and Rh-Cl(3) is
not unusual and in fact, reported for a number of dmpe
complexes.24f The terminal chloride ligands Cl(1), Cl(2), and
Cl(3) have meridional geometry common to many Rh(III)
complexes with phosphorus-oxygen ligands.26 Table 1 shows
the summary of the crystallographic data on crystals 12 and
18. Selected bond lengths and bond angles are given in Tables
2 and 3 for Rh and Pd complexes, respectively.
from the solution data (vide below) including the geometry and
stereochemical features. ORTEP diagrams of the Rh(III)
complex 12 and Pd(II) complex 18 with their atomic numbering
schemes are shown in Figures 1 and 2 respectively. The
structure of the Rh complex involves two ligands per metal
center with one of the iminophosphoranes not involving in
coordination to the metal and the other one forming a five
membered ring with P(III) phosphorus. The availability of two
different P(V) centers on the same molecule may be helpful in
internal structural correlations and in addressing the hemilabile
behavior of ligands, chelating effect, and stability factors.24
The Rh center has slightly distorted mer octahedral geometry
with trans chloride ligands and two iminophosphorane units,
one being away from the metal center such that the two P(III)
phosphorus atoms are in cis arrangement. This observation from
the crystal structure is also derived from the 31P COSY NMR
spectrum of 12 where P1 and P3 couple with a coupling constant
of 30 Hz and is in the general range for cis coupled Rh
complexes24 (in fact, trans coupling is around 630 Hz, vide
below). Compression of P(3)-Rh-N(2) [82.0° (4)] compared
to P(3)-Rh-P(1) [98.8° (18)] indicates the ring strain which
is the expected value for the related phosphorus-nitrogen
systems.25 A relative reduction of the angle P(4)-N(2)-C(59)
[120.56° (13)] on the coordinated arm of the iminophosphorane
is noted, compared to the bond angle P(2)-N(1)-C(26)
(23) (a) Pregosin, P. S.; Kuntz, R. W. 31P and 13C NMR of Transition
Metal Complexes; Springer: Berlin, Heidelburg, New York, 1979. (b)
Appleton, T. G.; Clark, H. C.; Manzer, L. E. Coord. Chem. ReV. 1973, 10,
335 (c) Goggin, P. L.; Knight J. R. J. Chem. Soc. Dalton 1972, 704.
(24) (a) Garrou, P. E. Chem. ReV. 1981, 81, 829. (b) Chadwell, S. J.;
Coles, S. J.; Edwards, P. G.; Hursthouse, M. B. J. Chem. Soc. Dalton 1995,
3551. (c) Braunstein, P.; Matt, D.; Mathey, F.; Thaward, D. J. Chem. Res.
(S) 1978, 232. (d) Verkhade, J. G. Coord. Chem. ReV. 1972, 9. (e) Garrou,
P. E. Inorg. Chem. 1975, 14, 1435 (f) Suzuki, T.; Isobe, K.; Kashiwabara,
K. J. Chem. Soc. Dalton 1995, 3609.
(25) (a) Al-Soudani, A. R. H.; Batsanov, A. S.; Edwards, P. G.; Howard,
J. A. K. J. Chem. Soc. Dalton 1994, 987. (b) Knight, D. A.; Cole-Hamilton,
D. J.; Cupertino, D. C.; Harman, M.; Hursthouse, M. B. Polyhedron 1992,
11, 1987. (c) Braunstein, P.; Matt, D.; Mathey, F.; Thaward, D. J. Chem.
Res. (M) 1978, 3041. (d) Skapski, A. C.; Stephans, F. A. J. Chem. Soc.
Dalton 1973, 1789.
The ORTEP plot of 18 confirm the proposed five-membered
metallacyclic structure with Pd(II) in square planar geometry.
Pd metal is bound to the iminato nitrogen and P(III) center via
cis disposition to form a five-membered chelate. The P-N bond
length in 18 (1.60 Å) is in the normal range expected for P-N
double bonds found in many reported complexes.17 Similarly,
the Pd-P bond length (2.224 Å) is also in the range normally
expected for similar Rh and Pd complexes.
III. NMR Investigations. Rh(III) complexes are produced
by the slow addition of a hot solution of RhCl3‚xH2O in CHCl3
to the solution of ligands 6-10 in CHCl3 in the presence of a
few drops of ethyl alcohol, followed by refluxing it for 4 h
(Scheme 2). Rh complex 12 is sparingly soluble in methanol
and ethanol and is characterized by multinuclear NMR (31P,
1H, 19F, and 2D COSY 31P NMR), IR, and elemental analysis.
31P NMR of 12 shows multiplets at δ 6.15, 16.7, 24.4, and 42.3
(Figure 3) with equal intensity (1:1:1:1). The multiplets at δ
16.7 and 24.4 appear to be similar with coupling constants 109.1
and 111.1 Hz, respectively. These coupling constants are typical
1
of J(P-Rh) coupling reported for a number of Rh complexes.23
On the basis of the coupling constant range, multiplets at δ 16.7
and 24.4 are assigned to P(III) phosphorus atoms. The two
peaks also show characteristic large downfield shifts (∆s ) 47.2
and 54.9, respectively) compared to that of ligand 7, which is
reminiscent of the shifts exhibited by P(III) coordination in
several metal complexes.22 The peak at δ 6.5 is a doublet and
not shifted substantially compared to that of the P(V) phosphorus
(26) Braunstein, P.; Chauvin, Y.; Nahring, Y.; Decian, A. and Fischer,
J. J. Chem. Soc., Dalton 1995, 863.

Retention of Inhibitory Potency of ACE Inhibitors
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11369
Figure 1. ORTEP diagram of Rh complex 12.
Table 2. Bond Lengths and Bond Angles for Complex 12
Rh1-Cl1
Rh1-Cl2
Rh1-Cl3
Rh1-P1
Rh1-P3
Rh1-N2
P2-N1
2.3658(5)
2.3840(5)
2.3496(5)
2.2894(5)
2.3054(5)
2.2316(16)
1.5786(18)
1.6064(17)
1.352(3)
Cl1-Rh1-Cl2
Cl1-Rh1-Cl3
Cl1-Rh1-P1
Cl1-Rh1-P3
Cl1-Rh1-N2
Cl2-Rh1-Cl3
Cl2-Rh1-P1
Cl2-Rh1-P3
Cl2-Rh1-N2
Cl3-Rh1-P1
Cl3-Rh1-P3
Cl3-Rh1-N2
P1-Rh1-P3
P1-Rh1-N2
P3-Rh1-N2
Rh1-P1-C1
Rh1-P1-C2
Rh1-P1-C8
N2-P4-C34
P2-N1-C26
Rh1-N2-P4
Rh1-N2-C59
P4-N2-C59
84.755(18)
172.004(17)
86.958(17)
104.17(17)
88.65(4)
87.249(18)
88.709(17)
168.529(18)
91.14(4)
92.765(17)
83.777(17)
91.63(4)
98.826(18)
175.59(4)
82.01(4)
108.66(7)
113.65(6)
115.84(7)
103.32(9)
133.36(15)
123.11(9)
116.32(12)
120.56(13)
P4-N2
N1-C26
N2-C59
1.411(3)
CH2 groups of the two separate ligands, the latter peak being
shifted further downfield which can be assigned to CH2 of the
P-C-P backbone in which both P(III) and P(V) are involved
in coordination. 1H NMR also shows two closely spaced peaks
(3.89 and 3.91) assigned to two types of OCH3 groups.
Similarly, 19F NMR shows four signals at δ -132.9, -140.9,
-143.8, and -153.9 instead of the expected two signals of the
AA′ XX′ pattern for the perfluoro core, as for example in ligand
7 (δ -144.0 and -154.0.8). On closer examination, it appears
that two sets of fluorine signals (i.e., δ -143.8 and -153.9)
are similar to the fluorine signals in ligand 7 and, hence, can
be assigned to perfluoroiminophosphorane in which PdN is not
coordinated to Rh. Further characterization of the Rh(III)
complex is provided by 13C NMR corroborating well with the
inference of two ligands per metal center by showing two sets
of multiplets (doublet of doublet) at δ 28.4 (1JC-P ) 26.4 Hz)
and 37.4 (1JC-P ) 32.6 Hz), respectively for two different CH2
groups on each of the P-C-P backbones of two ligands.
Although peaks due to P(V) phosphorus could be assigned,
it is difficult to assign peaks at δ 16.7 and 24.4 as to which
Figure 2. ORTEP diagram of Pd complex 18.
in ligand 7 (δ 11.8) and, hence, indicates that it may not be
involved in chelation to Rh via the nitrogen atom. On the other
hand, the peak at δ 42.3 is shifted downfield (∆s ) 30.5) with
a value similar to the reported values for P(V) phosphorus in
Rh(I) iminophosphorane complexes,17 indicating that it may be
involved in coordination to the metal center. Both peaks exhibit
doublets with coupling constants 30.3 and 15.8 Hz (2JP-P),
respectively. However, it is to be noted that the broad P(V)
phosphorus signals are not well-resolved, masking additional
information on long-range fluorine and Rh coupling. On the
basis of the above observations, it can be safely assumed that
two ligands are involved in Rh complex formation.
The inference of two ligands per metal center also comes
1
from the investigations on H, 19F, and 13C NMR of 12. For
1
example, H NMR of 12 shows two distinct peaks (doublet of
doublet, 2JP-H ) 12.3 Hz) at δ 4.01 and 5.18 (compared to the
2
peak at δ 3.36, JPH ) 12.3 Hz in ligand 7) assigned to two

11370 J. Am. Chem. Soc., Vol. 120, No. 44, 1998
Pandurangi et al.
Scheme 2
Table 3. Bond Lengths and Bond Angles for Complex 18
rationalized on the basis of ring formation through which A
couples to B. The shape of both B and C peaks appear
complicated owing to the coupling with phosphorus of the
second ligand through the ring in addition to the existing
coupling with P(V) of the same arm. The coupling constants
Pd-C11
Pd-C12
Pd-P2
Pd-N1
P1-N1
P1-C8
P1-C14
P1-C20
P2-C20
P2-C21
P2-C27
N1-C1
2.2927(10)
2.3577(9)
2.2248(9)
2.0798(24)
1.6029(24)
1.787(3)
1.792(3)
1.795(3)
1.834(3)
1.808(3)
1.812(3)
1.399(3)
C11-Pd-C12
C11-Pd-P2
C11-Pd-N1
C12-Pd-P2
C12-Pd-N1
P2-Pd-N1
N1-P1-C8
N1-P1-C14
N1-P1-C20
C8-P1-C14
C8-P1-C20
C14-P1-C20
Pd-P2-C20
Pd-P2-C21
Pd-P2-C27
C20-P2-C21
C20-P2-C27
C21-P2-C27
Pd-N1-P1
93.71(4)
86.79(3)
174.30(7)
174.51(3)
90.57(7)
2
around 23-28 Hz can be attributed to JPP coupling. The
89.32(7)
assignment of C to phosphorus P3 (Figure 1) involved in the
ring formation with Rh is also based on the normal upfield shift
due to effect often called “ring contribution” proposed by Garrou
et al.24 It has been shown that the phosphorus atom in a five-
membered chelate ring gives a lower field resonance compared
to a phosphorus atom not involved in the ring.24b
109.22(13)
115.68(12)
106.14(13)
108.66(13)
107.21(13)
109.60(13)
105.81(9)
115.14(11)
117.06(10)
104.54(13)
102.67(13)
109.93(14)
111.98(12)
118.65(18)
127.23(19)
121.52(54)
122.86(24)
Functionalized Rh complexes (11, 13-15) show similar NMR
spectra including the one which is attached to lisinopril at the
position para to perfluoroaryl azide group. On the basis of the
similarity of the spectroscopic properties and X-ray structural
investigations of 12 (vide infra), it is reasonable to assume that
the lisinopril-conjugated Rh complex has also two ligands per
metal center, which is also verified by conventional element
analysis.
Pd-N1-C1
P1-N1-C1
N1-C1-C2
N1-C1-C6
Unlike Rh complexes, Pd complexes of the functionalized
perfluoroaryl derivatives have a relatively simple NMR pattern.
For example, compound 18 shows downfield shifts of 56 ppm
and 45 ppm for P(III) and P(V), respectively, after coordination.
belongs to which phosphorus using 1D NMR spectrum alone.
To understand the complicated coupling pattern and confirm
the assignments to different phosphorus P(III) signals, a 2D 31
P
2
The magnitude of JPP coupling decreases to 25 Hz upon
COSY experiment was conducted. Figure 4 shows the 2D NMR
spectrum for complex 12. For simplicity, phosphorus peaks
are numbered A, B, C, and D. To start with, A and D are
already assigned to coordinated P(V) and noncoordinated P(V),
respectively. Coupling of D with B, but not with C, indicates
that D is connected to B on the same arm of iminophosphorane
which is not coordinated to Rh. Hence, peak B can be assigned
to P1(vide infra). Coupling of A with both C and B can be
complexation. The complexation with Pd II with 8 is expected
to withdraw electron density away from the N-P-C-P
skeleton and presumably cause a decrease in the s character
within the P-C-P framework. The chemical constitution of
18 was confirmed by the single-crystal X-ray investigations.
IV. IC50 Measurements Using ACE Amidolytic Assay.
ACE from rabbit lung is incubated with lisinopril derivatives

Retention of Inhibitory Potency of ACE Inhibitors
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11371
Figure 3. 31P NMR spectrum of Rh complex.
In the absence of the availability of the crystal structure of
ACE, studies on the determination of the binding sites of ACE
revolve around the measurement of IC50 values for ACE
inhibitors, functionalized by a variety of groups.27,28 However,
ACE is a metalloprotease enzyme similar to carboxypeptidase
A29 or thermolysin,30 and the availability of the crystal structure
of thermolysin and its binding studies with different substrates
can be extended to understand the binding modes of ACE with
ACE inhibitors. Of special interest was the bidentate coordina-
tion of the carboxylate oxygens next to the secondary amine
group on lisinopril (5 in Scheme 1). Along with the hydro-
phobic binding pockets with phenyl S1, lysyl part S1 , and
1
leucine S21 these are found to be responsible for the tight binding
of the ACE inhibitor to ACE. The importance of the carboxylic
group of the N-carboxy alkyl fragment of ACE inhibitors is
revealed by the marked decrease in inhibitor potency from 1.2
nM to 5.8 µM after replacing the carboxylic group with
hydrogen (Table 9 in ref 8c). Our studies on the esterification
of both carboxylic groups on lisinopril also resulted in the
reduction of inhibitory potency from nM to µM levels.14,15 The
biological activity of ACE has long been known to be abolished
by chelating agents such as EDTA, and in fact, the EDTA effect
has led to the conclusion that ACE is a metalloenzyme.32 The
functional role of Zn in ACE-mediated hydrolysis of oligopep-
tides is shown to be similar to that of Zn in carboxypeptidase
A29 or thermolysin.30,31 It is possible that ACE inhibitors may
also bind to substrates other than ACE (e.g., in the deactivation
of bradykinin5). However, the concordance of high-density
Figure 4. 2D 31P COSY NMR spectrum of Rh complex 12.
or metal conjugates at various concentrations in a mixture of
methanol and buffer. The inhibition induced by modified
lisinopril derivatives reduces the cleaving potency of ACE which
results in the decreasing absorbance of the cleaved product of
the substrate at 340 nm. The change in the percentage of
inhibition is calculated by measuring the relative integration of
peak areas. Lisinopril forms a positive control. Figure 5 shows
the concentration dependence of lisinopril conjugates (5, 10,
15, and 20) on their inhibitory potency compared to that of
native lisinopril. IC50 values calculated from the graph for 12
and 18 indicates that functionalization at the primary amine site
sustains the inhibitory potency at nM level, and hence, these
derivatives should be useful for further radiochemical studies.
IC50 values for the other derivatives are shown in Table 4.
(27) Mendelsohn, F. A. O. Clin. Exp. Pharmacol. Physiol. 1984, 11,
431.
(28) Bernstein, K. E.; Martin, B. M.; Edwards, A. S.; Bernstein, E. A.
J. Biol. Chem. 1989, 264, 11945.
(29) (a) Valle, B. L.; Glades, D. S.; Auld; Riordan J. F. In Metal ions in
Biology: Zinc Enzymes; Spiro, T. G., Ed.; Wiley: New York, 1983; Vol.
5, pp 25-75. (b) Lipscomb, W. N.; Hartsuck, J. A.; Quiocho, F. A.; Reeke,
G. N. Proc. Natl. Acad. Sci. U.S.A. 1969, 64, 28. (c) Vallee, B. L.; Rupley,
J. A.; Coombs, T. L.; Neurath, H. J. Biol. Chem. 1960, 235, 64.
(30) (a) Holmquist, B. and Vallee, B. L. J. Biol. Chem. 1974, 249, 4601.
(b) Kester, W. R.; Matthews, B. W. Biochemistry 1977, 16, 2506.
(31) Colman, P. M.; Jansonius J. N.; Matthews, B. W. J. J. Mol. Biol.
1972, 70, 701.
(32) Gonzalez, E. B.; Farkas, E.; Soudi, A. A.; Tan, T.; Yanovski, A. I.;
Nolan, K. B. J. Chem. Soc., Dalton Trans. 1997, 2377.

11372 J. Am. Chem. Soc., Vol. 120, No. 44, 1998
Pandurangi et al.
Figure 5. Concentration dependence of functionalized lisinopril derivatives 5, 10, 15, and 20.
Table 4. IC50 Values for Lisinopril Conjugates
obscure the contrast between normal and fibrotic sites. The
combination of two phosphines and one iminophosphorane may
provide a much needed stability for a Rh(III) metal complex in
biological conditions.
compd
IC50 (nM)
ref
lisinopril
I-125 351A
11.2
2.6
7c, d, 8c, 14, 15
6c, 8c, 27
14, 15
primary amine modified lisinopril 5
secondary amine modified lisinopril
esterified lisinopril
0.62
30.2
201
26
14, 15
14, 15
V. Conclusions
compound 10
this work
this work
this work
this work
Functionalization of ACE inhibitors by Rh- and Pd-based
BFCAs opens a possibilty of introducing clinically useful
radionuclides to pharmacophores for in vivo tracking applica-
tions in humans. Measurement of IC50 values of lisinopril
conjugates functionalized at various positions on lisinopril enable
us to introduce relatively bulky chelating frameworks far away
from the binding sites retaining the inhibitory potency in the
postmodification stage. Selective oxidation of one arm of
diphosphines by functionalized perfluoroaryl azides R-C6F4N3
(R ) COOMe, succinimide, CF3, and lisinopril) lead to a
combination of a phosphine-iminophosphorane framework
suitable for introducing Rh(III) and Pd(II) precursors. Tailoring
lisinopril through the primary amine functionalized perfluoroaryl
azides with or without metal exhibited high inhibitory potency,
retaining the character of native lisinopril. On the other hand,
direct complexation with carboxylic oxygens of lisinopril by a
metal abolishes the ACE activity, exemplified by the low IC50
value. Structural characterization of the Rh(III) complex
perfluoroaryl (iminophosphorano)phosphine 12 by multiprobe
NMR and single crystal X-ray crystallography revealed the
presence of two ligands per metal center with two phosphines
and only one of the iminophosphorane participating in coordina-
tion to Rh. These studies also confirmed that the electron-rich
centers in BFCA-lisinopril conjugates are far away from the
binding sites of lisinopril, verifying the retention of their
inhibitory potency. These new findings on the retention of
inhibitory potency of Rh and Pd functionalized lisinopril form
the basis for introduction of 105Rh and 109Pd radiomarkers. The
Rh complex 12
5.2
8.9
267
Pd complex with 18
lisinopril-Cu complexa
a Reference 32.
ACE binding with fibrotic tissues and the absence of carboxy-
peptidase or thermolysin at such sites makes ACE inhibitor
scintigraphy feasible. Lisinopril conjugates bind ACE strongly
(IC50 values are in nM range) which may be attributed to the
availability of additional binding sites in the form of hydro-
phobic interactions and H-bonded interactions similar to N-
carboxy alkyl dipeptides.15 Recently, Gonzalez et al. have
reported on a Cu(II) dimeric complex with lisinopril in which
oxygen atoms of carboxylate and carbonyl groups as well as
the secondary amine nitrogen of one of the lisinopril ligands
occupy three positions in the basal plane of Cu, whereas the
fourth position is occupied by the prolyl carboxylate of a second
lisinopril ligand.32 Our IC50 values, measured on a Cu-
lisinopril conjugate, show a significant reduction in inhibitory
potency to µM levels (Figure 5) indicating the sensitivity of
substituents to the biological activity of lisinopril. In contrast,
introduction of Rh and Pd on the long-chain end of the aliphatic
primary amine function did not affect the inhibitory potency
after functionalization. Such modified lisinopril derivatives may
be amenable to noninvasive imaging of biological sites in vivo.
However, the ultimate diagnostic applications demand high
stability of the metal complex in vivo since demetalation would

Retention of Inhibitory Potency of ACE Inhibitors
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11373
synthesis and radiochemical characterization of 105Rh and
109Pd functionalized lisinopril derivatives and their biodistribu-
tion studies in animals are in progress and will be published
elsewhere.
acknowledges the useful discussions with Professor Karl T.
Weber of Internal Medicine.
Supporting Information Available: Crystal structure data
on 12 and 18, atomic coordinates and equivalent isotropic
displacement parameters, selected interatomic distances, tor-
sional angles, displacement parameters for hydrogen atoms, and
intensity data (49 pages, print/PDF). See any current masthead
page for ordering information and Web access instructions.
Acknowledgment. The work was supported by funds
provided by DOE Grant DEFG0289ER 60875. NMR data were
collected on instruments purchased with funds from NSF Grants
8908304 and 9221835. We appreciate the help of Dr. Guo and
Mr. Gao of the NMR facility for help with the decoupling
experiments. We are grateful to Professors Wynn Volkert and
Robert Kuntz for helpful discussions and support. R.S.P.
JA9802403