Molecular recognition of a monoclonal antibody (AC1106) cross-reactive for derivatives of Ru(bpy)32+a and Ru(phen)32+

Article abstract of DOI:10.1021/ja952014o

The characterization of a monoclonal antibody (AC1106) elicited via immunization with a co(dmbpy)(bpy)₂³⁺-methyl viologen hapten (1) is described. AC1106 was found cross-reactive for a variety of luminescent ruthenium(II) metal complexes which served as useful probes to investigate the molecular recognition properties of this antibody. AC1106 was found to be specific for methylated derivatives of Ru(bpy)₃²⁺ and Ru(phen)₃²⁺ in the order of Ru(dmbpy)₃²⁺ > Ru(dmbpy)(bpy)₂²⁺ > Ru(dmphen)₃²⁺ > Ru(bpy)₃²⁺ >> Ru(phen)₃²⁺. The affinities of AC1106 for these metal complexes were found to range from ≥ 5 x 10⁷ to ≤ 1 x 10³ M^-1. When bound (> 98%) by AC1106, the luminescence decay traces for the racemic Ru(dmbpy)₃²⁺ and Ru(dmbpy)(bpy)₂²⁺ gave a satisfactory fit to a single-exponential decay process. Furthermore, D₂O/H₂O experiments with Ru(dmbpy)₃²⁺ indicate that AC1106 protects approximately 70% of the antibody-bound Ru(dmbpy)₃²⁺ from excited state deactivation by the solvent. Competition ELISA data indicate that both the metal center and the methyl viologen moiety present in a Ru(bpy)₃²⁺-methyl viologen conjugate ([Ru(mv²⁺-bpy)(bpy)₂]⁴⁺) are important recognition elements for AC1106. Despite the apparent affinity of AC1106 for methyl viologen, no evidence for simultaneous binding of methyl viologen and Ru(dmbpy)(bpy)₂²⁺ inside the binding pocket of AC1106 could be found. Rather, the addition of methyl viologen was found to result in the displacement of AC1106-bound Ru(dmbpy)(bpy)₂²⁺ from the antibody binding site. The characterization of a monoclonal antibody (AC1106) elicited via immunization with a Co(dmbpy)-(bpy)₂³⁺-methyl viologen hapten (1) is described. AC1106 was found cross-reactive for a variety of luminescent ruthenium (II) metal complexes which served as useful probes to investigate the molecular recognition properties of this antibody. AC1106 was found to be specific for methylated derivatives of Ru(bby)₃²⁺ and Ru(phen)₃²⁺. Furthermore, D₂O/H₂O experiments with Ru(dmbpy)₃²⁺ indicate that AC1106 competition ELISA data indicate that both the metal center and the methyl viologen moiety present in a Ru(bpy)₃²⁺-methyl viologen conjugate ([Ru(mv²⁺-bpy)(bpy)₂]⁴⁺) are important recognition elements for AC1106. Despite the apparent affinity of AC1106 for methyl viologen, no evidence for simultaneous binding of methyl viologen and Ru(dmbpy)(bpy)₂²⁺ inside the binding pocket of AC1106 could be found. Rather, the addition of methyl viologen was found to result in the displacement of AC1106-bound Ru(dmbpy)(bpy)₂²⁺ from the antibody binding site.

Full text of DOI:10.1021/ja952014o

3192
J. Am. Chem. Soc. 1996, 118, 3192-3201
Molecular Recognition of a Monoclonal Antibody (AC1106)
2+
2+
Cross-Reactive for Derivatives of Ru(bpy)3 and Ru(phen)3
Kevin Shreder,† Anthony Harriman,‡ and Brent L. Iverson*,†
Contribution from the Department of Chemistry and Biochemistry, UniVersity of Texas,
Austin, Texas 78712, and Faculte´ de Chimie, UniVersite´ Louis Pasteur, 1, rue Blaise Pascal,
67008 Strasbourg, France
ReceiVed June 21, 1995X
Abstract: The characterization of a monoclonal antibody (AC1106) elicited Via immunization with a Co(dmbpy)-
(bpy)23+-methyl viologen hapten (1) is described. AC1106 was found cross-reactive for a variety of luminescent
ruthenium(II) metal complexes which served as useful probes to investigate the molecular recognition properties of
2+
2+
this antibody. AC1106 was found to be specific for methylated derivatives of Ru(bpy)3 and Ru(phen)3 in the
order of Ru(dmbpy)3 > Ru(dmbpy)(bpy)2 > Ru(dmphen)3 > Ru(bpy)3 . Ru(phen)32+. The affinities of
AC1106 for these metal complexes were found to range from g 5 × 107 to e 1 × 103 M-1. When bound (>98%)
by AC1106, the luminescence decay traces for the racemic Ru(dmbpy)32+ and Ru(dmbpy)(bpy)22+ gave a satisfactory
fit to a single-exponential decay process. Furthermore, D2O/H2O experiments with Ru(dmbpy)32+ indicate that AC1106
2+
2+
2+
2+
2+
protects approximately 70% of the antibody-bound Ru(dmbpy)3 from excited state deactivation by the solvent.
2+
Competition ELISA data indicate that both the metal center and the methyl viologen moiety present in a Ru(bpy)3
-
methyl viologen conjugate ([Ru(mv2+-bpy)(bpy)2]4+) are important recognition elements for AC1106. Despite the
apparent affinity of AC1106 for methyl viologen, no evidence for simultaneous binding of methyl viologen and
2+
Ru(dmbpy)(bpy)2 inside the binding pocket of AC1106 could be found. Rather, the addition of methyl viologen
was found to result in the displacement of AC1106-bound Ru(dmbpy)(bpy)2 from the antibody binding site.
2+
Introduction
nescent probes. Both polyclonal and monoclonal antibodies
specific for, or elicited by, luminescent molecules such as
adriamycin,7 derivatives of dansyl chloride,8 9-(2-carboxy-2-
cyanovinyl)julolidine,9 and fluorescein10 have been reported. By
far, the most extensively studied set of data exists for fluorescein.
In general, antibody binding of fluorescein results in fluores-
cence quenching.11 Use of this spectroscopic handle, pioneered
by the research group of Voss,12 has made possible a variety of
elegant studies that include investigations of hapten dissociation
and association rates13 and kinetic affinity maturation.14
One class of luminescent molecules that has gone relatively
unexplored as antibody probes is that of ruthenium(II) metal
complexes, such as Ru(bpy)3 and Ru(phen)3 (Figure 1).
Several qualities of these metal complexes facilitate their use
as probes of antibody response and recognition, including their
stability in water,15 synthetic accessibility to a variety of metal
complexes of a well-defined shape,16 and long-lived excited
Antibodies with high affinity and exquisite specificity can
be elicited by repeated immunization with a molecule of interest.
Antibodies1 can be considered a unique type of media in that
they can be “tailor-made” to bind a desired ligand2 Via
immunization with an appropriately chosen hapten. The as-
sociation between antibody and antigen is the result of im-
munologically tuned, complementary interactions (e.g., charge-
charge, dipole-dipole, and π-stacking) between an antibody
ligand and amino acid residues in the binding pocket. Large
association constants for antibody ligands (up to 1012 M-1) and
a high degree of selectivity are common.3,4 Thus, antibodies
are useful in the design of both catalytic5 and recognition based
systems6 where high affinity and specificity are desirable
elements.
2+
2+
Of considerable importance in the design of antibody-based
systems is a fundamental understanding of the binding properties
associated with antibody-antigen interactions. One approach
to understanding these properties is through the use of lumi-
(5) (a) Lerner, R. A.; Benkovic, S. J. Chemtracts 1990, 3, 1-36. (b)
Lerner, R. A.; Benkovic, S. J.; Schultz, P. G. Science 1991, 252, 659-667.
(c) Schultz, P. G.; Lerner, R. A. Acc. Chem. Res. 1993, 26, 391-395.
(6) (a) Hohsaka, T.; Kawashima, K.; Sisido, M. J. Am. Chem. Soc. 1994,
116, 413-414. (b) Stewart, J. D.; Roberts, V. A.; Crowder, M. W.; Getzhoff,
E. D.; Benkovic, S. J. J. Am. Chem. Soc. 1994, 116, 415-416. (c) Willner,
I.; Blonder, R.; Dagan, A. J. J. Am. Chem. Soc. 1994, 116, 9365-9366.
(7) Chien, Y.-H.; Levine, L. Immunochemistry 1975, 12, 291-296.
(8) Parker, C. W. In Handbook of Experimental Immunology; Weir, D.
M., Ed.; Blackwell Scientific Publications: Oxford, 1978; Vol. I, 18.1-
18.25.
(9) Iwaki, T.; Torigoe, C.; Noji, M.; Nakanishi, M. Biochemistry 1993,
32, 7589-7592.
(10) Lopatin, D. E.; Voss, E. W., Jr. Biochemistry 1971, 10, 208-216.
(11) Watt, R. M.; Voss, E. W., Jr. Immunochemistry 1977, 14, 533-
554.
† University of Texas.
‡ Universite´ Louis Pasteur.
X Abstract published in AdVance ACS Abstracts, March 15, 1996.
(1) For reviews on antibody structure, see: (a) Amzel, L. M.; Poljak, R.
J. Annu. ReV. Biochem. 1979, 48, 961-997. (b) Davies, D. R.; Chacko, S.
Acc. Chem. Res. 1993, 26, 421-427.
(2) Ligand, as used throughout this paper, has two meanings: It signifies
either a small molecule that binds within an antibody binding pocket or a
chelate for a metal ion. In addition, “cross-reactive” is an immunological
term indicating that the antibody is capable of binding a ligand other than
what it was designed for (i.e. the hapten). Cross-reactive is not meant to
indicate that a chemical reaction takes place within the antibody binding
pocket.
(3) Schultz, P. G.; Jacobs, J. W. Catalytic Antibodies: Generation and
Characterization. In EnVironmental Influences and Recognition in Enzyme
Chemistry; Liebman, J. F., Greenburg, A., Eds.; VCH Publishers, Inc.: New
York, 1988; pp 305-312.
(4) The upper affinity limit of a “normal” immune response has been
suggested to be ca. 1010 M-1. See: Foote, J.; Eisen, H. N. Proc. Natl. Acad.
Sci. U.S.A. 1995, 92, 1254-1256.
(12) Voss, E. W., Jr. Fluorescein Hapten: An Immunological Probe;
CRC Press, Inc.: Boca Raton, FL, 1984.
(13) Watt, R. M.; Herron, J. N.; Voss, E. W., Jr. Mol. Immunol. 1980,
17, 1237-1243.
(14) Levison, S. A.; Hicks, A. N.; Portmann, A. J.; Dandliker, W. B.
Biochemistry 1975, 14, 3778-3786.
(15) Brandt, W. W.; Smith, G. S. Anal. Chem. 1949, 21, 1313-1319.
0002-7863/96/1518-3192$12.00/0 © 1996 American Chemical Society

Molecular Recognition of a Monoclonal Antibody
J. Am. Chem. Soc., Vol. 118, No. 13, 1996 3193
Figure 1. Structures and abbreviations for the compounds used in this
study.
states17 that are sensitive to both solvent18 and quenchers such
as O2.19 As demonstrated in this study, when used as model
antibody ligands, these compounds can provide important
information on the nature of antibody-antigen binding.
The excited state properties of these compounds have led to
important applications in the areas of both electron transfer
studies and catalysis.20,21 As a result, the photophysical and
molecular recognition properties of this general class of
compounds have been studied in a variety of biological and
abiological organized media, such as DNA,22 micelles,23 clays,24
polyelectrolytes,25 and zeolites.26 Consequently, these systems
Figure 2. Synthetic scheme for the synthesis of the hapten 1 and the
corresponding ruthenium(II) analogue 2: (a) 4,4′-bipyridine, CH3CN,
reflux (35%); (b) 4-(3-bromopropyl)-4′-methyl-2,2′-bipyridine, CH3-
CN, reflux (45%). (c) For 1: [Co(bpy)2]Cl3‚H2O, H2O then TFA,
dioxane (30% from 5). For 2: [Ru(bpy)2]Cl2‚2H2O, MeOH, reflux (6)
then 1:1 H2O:TFA (67% from 5).
provide well-studied background for comparison and contrast.
Despite the extensive use of these metal complexes, antibody-
based studies have been limited to our previous report of a
(16) (a) Seddon, E. A.; Seddon, K. R. The Chemistry of Ruthenium;
Elsevior Science Publishing Co. Inc.: New York, 1984; pp 414-475. (b)
Shro¨der, M.; Stephenson, T. A. In ComprehensiVe Coordination Chemistry;
Wilkinson, G., Ed.; New York, 1987; Vol. 4, pp 327-357.
(17) For general reviews, see: (a) Balzani, V.; Boletta, F.; Gandolfi, M.
T.; Maestri, M. Top. Curr. Chem. 1978, 75, 1-64. (b) DeArmond, M. K.;
Myrick, M. L. Acc. Chem. Res. 1989, 22, 364-370. (c) Meyer, T. J. Pure
Appl. Chem. 1986, 58, 1193-1200. (d) Juris, A.; Balzani, V.; Barigelletti,
F.; Campagna, S.; Belser, P.; von Zelewsky, A. Coord. Chem. ReV. 1988,
84, 85-277. (e) Kalyanasundaram, K. Photochemistry of Polypyridine and
Porphyrin Complexes; Academic Press: San Diego, 1992; pp 87-164 and
339-361. (f) Roundhill, D. M. Photochemistry and Photophysics of Metal
Complexes; Plenum Press: New York, 1994; pp 165-210.
2+ 27
polyclonal antibody system that binds Ru(bpy)3
.
For the present monoclonal antibody study, a Co(dmbpy)-
(bpy)23+-methyl viologen hapten (1) was synthesized (Figure
2). Several elements went into the design of this hapten: The
kinetically inert cobalt(III) metal center was chosen based on
its previously reported stability to the conditions of immuniza-
3+
tion.28 In addition, the dimensions of Co(bpy)3 roughly
approximate those of Ru(bpy)3 ,
2+ 29 and we anticipated that an
antibody capable of binding the hapten 1 would also be capable
of accommodating Ru(bpy)32+ within the same binding pocket.
Finally, methyl viologen was tethered within close proximity
(18) (a) Nakamaru, K. Bull. Chem. Soc. Jpn. 1982, 55, 1639-1640. (b)
Caspar, J. V.; Meyer, T. J. J. Am. Chem. Soc. 1983, 105, 5583-5590.
(19) (a) Demas, J. N.; Diemente, D.; Harris, E. W. J. Am. Chem. Soc.
1973, 95, 6864-6865. (b) Demas, J. N.; Harris, E. W.; McBride, R. P. J.
Am. Chem. Soc. 1977, 99, 3547-3551.
(25) (a) Meisel, D.; Rabani, J.; Meyerstein, D.; Matheson, M. S. J. Phys.
Chem. 1978, 82, 985-990. (b) Kelder, S.; Rabani, J. J. Phys. Chem. 1981,
85, 1637-1640.
(20) (a) Kalayanasundaram, K. Coord. Chem. ReV. 1982, 46, 159-244.
(b) Meyer, T. J. Acc. Chem. Res. 1989, 22, 163-171.
(26) (a) Maruszewski, K.; Strommen, D. P.; Kincaid, J. R. J. Am. Chem.
Soc. 1993, 115, 8345. (b) Borja, M.; Dutta, P. K. Nature 1993, 362, 43-
45. (c) Yonemoto, E. H.; Kim, Y. I.; Shmehl, R. H.; Wallin, J. O.; Shoulders,
B. A.; Richardson, B. R.; Haw, J. F.; Mallouk, T. E. J. Am. Chem. Soc.
1994, 116, 10557-10563.
(21) For an excellent review on biocatalytic approaches to solar ener-
gy storage, see: Willner, I.; Willner, B. In Topics in Current Chemistry;
Springer-Verlag: New York, 1991; Vol. 159, pp 153-218.
(22) (a) Kelly, J. M.; Tossi, A. B.; McConnell, D. J.; OhUigin, C. Nucleic
Acids Res. 1985, 13, 6017-6034. (b) Kumar, C. V.; Barton, J. K.; Turro,
N. J. J. Am. Chem. Soc. 1985, 107, 5518-5523.
(27) Shreder, K.; Harriman, A.; Iverson, B. L. J. Am. Chem. Soc. 1995,
117, 2673-2674.
(23) (a) Lachish, U.; Ottolenghi, M.; Rabani, J. J. Am. Chem. Soc. 1981,
103, 3761-3764. (b) Schmehl, R. H.; Whitesell, L. G.; Whitten, D. G. J.
Am. Chem. Soc. 1981, 103, 3761-3764. (c) Thomas, J. K. In Mechanistic
Aspects of Inorganic Reactions; Rorabacher, D. B., Endicott, John F., Eds.;
ACS Symposium Series 198; American Chemical Society: Washington,
DC, 1982; pp 335-346.
(28) Iverson, B. L.; Lerner, R. A. Science 1989, 243, 1184.
(29) For the crystal structure of D-[Co(bpy)3]‚3Fe(CN)6‚8H2O, see:
Ohashi, Y.; Yanagi, K.; Mitsuhashi, Y.; Nagata, K.; Kaizu, Y.; Sasada, Y.;
Kobayashi, H. J. Am. Chem. Soc. 1979, 101, 4739-4740. For the crystal
structure of [Ru(bpy)3]‚2PF6, see: (a) Rillema, D. P.; Jones, D. S.; Levy,
H. A. J. Chem. Soc., Chem. Commun. 1979, 849-851. (b) Rilema, D. P.;
Jones, D. S.; Woods, C.; Levy, H. A. Inorg. Chem. 1992, 31, 2935-2938.
(24) Thomas, J. K. Acc. Chem. Res. 1988, 21, 275-280.

3194 J. Am. Chem. Soc., Vol. 118, No. 13, 1996
Shreder et al.
of the Co(dmbpy)(bpy)23+ metal center in an attempt to elicit a
2+
binding site capable of complexing Ru(bpy)3 and methyl
viologen simultaneously. Following immunization with 1, a
murine monoclonal antibody (AC1106) cross-reactive for both
1 and 2 was isolated. In this study, AC1106 is shown to exhibit
a highly structure dependent binding mode for derivatives of
2+
2+
Ru(bpy)3 and Ru(phen)3
.
Results and Discussion
Steady-State Luminescence Studies. Steady-state lumines-
cence spectroscopy proved to be a useful tool in detecting the
specific binding of various luminescent ruthenium(II) metal
complexes to the monoclonal antibody AC1106. Shown in
Figure 3a is the integrated steady-state emission intensity
2+
recorded for a fixed amount of Ru(dmbpy)(bpy)2 in the
presence of increasing amounts of AC1106. It can be seen that
2+
the addition of AC1106 to a solution of Ru(dmbpy)(bpy)2
results in enhanced luminescence from the metal complex. As
important controls, it was noted that upon titration of Ru-
2+
(dmbpy)(bpy)2 with both bovine serum albumin (BSA) and
nonspecific IgG antibody there was no significant enhancement
in the emission intensity. These control studies, therefore,
confirm that the luminescence enhancement seen upon titration
with AC1106 is the result of a binding event within the antibody
pocket and does not arise from nonspecific binding to a protein
surface. Shown for comparison in Figure 4 are the emission
spectra at the beginning and end of the titration with AC1106:
There is a slight blue shift observed for the emission maximum
of the AC1106-Ru(dmbpy)(bpy)22+ complex relative to free Ru-
(dmbpy)(bpy)22+. In addition, the change in the total integrated
emission intensity upon binding to the antibody to free metal
complex is a factor of 2.3, a value well above the experimental
error.
When these same experiments were performed with Ru-
(dmbpy)32+, similar results were obtained. Shown in Figure
3b is the integrated steady-state emission intensity measured
2+
for a fixed amount of Ru(dmbpy)3 in the presence of
2+
increasing amounts of AC1106. As with Ru(dmbpy)(bpy)2
,
a significant enhancement in emission intensity was observed
upon titration with AC1106 but no such enhancement was
observed when the antibody was replaced with either BSA or
nonspecific antibody. Again, there is a slight blueshift in the
emission maximum upon binding (see supporting information).
Here, the change in the total integrated emission intensity of
the antibody bound to free metal complex is a factor of 2.6.
The amplification in emission intensity noted upon binding
of these metal complexes to AC1106 provides a convenient way
to derive the corresponding association constants.30 Thus, for
Ru(dmbpy)(bpy)22+, the Ka was determined to be 1.2 × 107
Figure 3. (a) Steady-state luminescence titration of 1 × 10-6 M Ru-
2+
(dmbpy)(bpy)2 with AC1106 (b), BSA (4), and nonspecific IgG
antibody (O). Shown is an interpolative fit for the Ka of AC1106 binding
to Ru(dmbpy)(bpy)22+ of 1.2 × 107 M-1. (b) Steady-state luminescence
2+
titration of 1 × 10-6 M Ru(dmbpy)3 with AC1106 (b), BSA (4),
2+
M-1 (Figure 3a) whereas for Ru(dmbpy)3 a lower limit for
and nonspecific IgG antibody (O). Shown is an interpolative fit for the
Ka of AC1106 binding to Ru(dmbpy)32+ of 3.4 × 107 M-1. Also shown
the Ka of 5 × 107 M-1 was set based on the linear nature of
2+
this titration (see Figure 3b). Interestingly, an association
is the titration of Ru(bpy)3 with AC1106 (0). (c) Steady-state
2+
2+
luminescence titration of 1 × 10-6 M Ru(dmphen)3 with AC1106
constant for AC1106 binding to Ru(bpy)3 could not be
(b), BSA (4), and nonspecific IgG antibody (O). Also shown is the
determined using this method since there was no amplification
of the luminescence in this case (Figure 3b). This result serves
2+
titration of Ru(phen)3 with AC1106 (0). Shown is an interpolative
fit for the Ka of AC1106 binding to Ru(dmphen)32+ of 8.4 × 105 M-1
.
2+
to indicate that the affinity of AC1106 for Ru(bpy)3 is
relatively low.
elicit the immune response that gave rise to AC1106, namely,
unsubstituted 2,2′-bipyridine groups and 5. Because 5 is derived
from 4,4′-dimethyl-2,2′-bipyridine, the apparent “methyl-
specific” mode of binding might be taken to indicate that the
immunological process that produced AC1106 gave particular
emphasis to the substituted ligand of the hapten. This may be
because the greater complexity of this latter ligand provides
for more specific molecular recognition.
The steady state luminescence data demonstrate that the
presence of methyl groups in the 4 and 4′ positions of the 2,2-
bipyridine ligands in Ru(dmbpy)(bpy)2 and Ru(dmbpy)3
plays a crucial role in recognition by AC1106. Interestingly,
two types of metal ligands are present in the hapten 1 used to
2+
2+
(30) The data could best be fit by using a 1:1 ratio of antibody pockets
to bound metal complex, so this mode of binding was assumed under the
conditions of the experiments presented in this paper.

Molecular Recognition of a Monoclonal Antibody
J. Am. Chem. Soc., Vol. 118, No. 13, 1996 3195
Figure 4. Emission spectra at the beginning (- - -) and end (s) of the
2+
titration of Ru(dmbpy)(bpy)2 with AC1106 shown in Figure 3a.
Alternatively, or in conjuction with this idea, is that increased
hydrophobicity of the more highly methyl-substituted metal
complexes leads to a higher affinity for AC1106. Indeed, strict
methyl-group recognition could not be used as a rationalization
2+
for the increased affinity of Ru(dmbpy)3 over Ru(dmbpy)-
(bpy)22+: Because the antibody binding pocket of AC1106 was
induced by 1, this pocket should be shape-selective for this
hapten. As a result, one would not expect the four “additional”
Figure 5. Semilogarithmic plot of the luminescence decay trace of
Ru(dmbpy)(bpy)22+ (1.6 × 10-5 M) in the presence of AC1106 (4.9 ×
10-5 M in Ab binding sites) in N2- and air-saturated PBS. AC1106-
2+
2+
bound Ru(dmbpy)3 yielded similar results (see supporting informa-
methyl groups present on Ru(dmbpy)3 to afford favorable
tion). As determined by the residuals shown above each plot, the
luminescence decay traces of these species gave a satisfactory fit
(- - -) to a single-exponential decay process assigned to the fully bound
metal complex. Excited-state lifetimes for these species are given in
Table 1.
protein contacts with AC1106 when only two methyl groups
are present in the hapten. However, if the binding pocket of
AC1106 possesses considerable hydrophobic character, it may
exert a preference for more hydrophobic metal complexes.31
To probe the effect that shape selectivity has on binding,
experiments were carried out with Ru(dmphen)32+. This metal
complex, by virtue of its bulkier 1,10-phenanthroline-derived
ligands, is a poor steric mimic of the Co(bpy)33+-based hapten
centration. When AC1106 was titrated into a solution of Ru-
2+
(dmbpy)(bpy)2 or Ru(dmbpy)32+, a longer lived component
appeared in the luminescence decay profiles, in addition to the
component characteristic of free metal complex. This new
component was assigned to the antibody-bound metal complex,
in line with the results of the steady-state luminescence studies.
At higher concentrations of AC1106, the luminescent decay
traces of both metal complexes gave a satisfactory fit to a single-
exponential decay process in both air- and N2-saturated solution
(Figure 5). The excited-state lifetimes recorded under such
conditions, which are significantly longer than those recorded
in the complete absence of antibody, are assigned to the
antibody-bound metal complex.
when compared to its 2,2′-bipyridine-derived congener, Ru-
2+
(dmbpy)3
. Shown in Figure 3c is the integrated steady-
state emission intensity recorded for a fixed amount of Ru-
(dmphen)32+ in the presence of increasing amounts of AC1106.
As with the 2,2′-bipyridine-derived ruthenium(II) metal com-
plexes, a significant emission enhancement can be seen upon
titration with AC1106 whereas BSA or nonspecific antibody
exhibit no such effect. Using this enhancement as a means to
calculate the association constant for interaction between
AC1106 and the metal complex yields a value of 8.4 × 105
M-1. The 100-fold lowered affinity for Ru(dmphen)32+ versus
Ru(dmbpy)32+ can be explained by the relatively poor steric fit
of the former metal complex within the binding pocket of
AC1106. Interestingly, as in the 2,2′-bipyridine series, no
significant enhancement of the emission for the methyl-free Ru-
Shown in Table 1 are the excited-state lifetimes of the free
and AC1106-bound luminescent ruthenium(II) metal complexes
in air- and N2-saturated buffer (PBS: 10 mM phosphate, 120
2+
mM NaCl, pH 7.4). Significantly, for both Ru(dmbpy)(bpy)2
and Ru(dmbpy)32+, there is excellent agreement in the magni-
tudes of the amplification of the steady-state emission yield and
in the enhancement in the excited-state lifetimes seen upon
binding to AC1106. This agreement indicates that, in each case,
the entire racemic sample of Ru(dmbpy)(bpy)2 and Ru-
(dmbpy)3 is bound by the antibody. Thus, in the case that
2+
(phen)3 was seen upon titration with AC1106 (Figure 3c).
This observation again emphasizes that methylated ligands are
crucial components for recognition of metal complexes by
AC1106.
2+
2+
Time-Resolved Luminescence Studies. Time-resolved lu-
minescence spectroscopy was also used to study the interaction
of AC1106 with these same derivatives of Ru(bpy)32+ and Ru-
(phen)32+. In the absence of antibody, the luminescence profiles
recorded for the metal complexes could be analyzed satisfac-
torily in terms of a single exponential, in each case the derived
lifetimes depending on the concentration of dissolved oxygen
but being independent of laser power and chromophore con-
only a single enantiomer was bound the observed amplification
in luminescence intensity would demand that the luminescence
decay profiles were biphasic with one component having the
lifetime of unbound metal complex and the other having a
lifetime twice that found here for bound metal complex. On
this basis, we can discount selective enantiomeric binding with
this antibody and, instead, it appears that AC1106 binds both
enantiomers in a nonstereospecific manner.32 This result is
surprising considering the high degree of stereospecificity that
is often associated with antibody binding.33
(31) For the importance of solvent reorganization on ligand binding,
see: Chervenak, M. C.; Toone, E. J. J. Am. Chem. Soc. 1994, 116, 10533-
10539 and references therein.

3196 J. Am. Chem. Soc., Vol. 118, No. 13, 1996
Shreder et al.
Table 1. Excited State Lifetimes of the Indicated Ruthenium(II) Metal Complex
e
metal complex
τa (ns)a
τb (ns)b
τc (ns)c
τd (ns)d
840 [1.8]
∆
s.s./∆t.r.
Ka (M-1
)
2+
Ru(dmbpy)(bpy)2
310
210
460
320
640 (D2O)
590
1800
690 [2.2]
620 [3.0]
2.1/2.2
2.6/3.0
1.2 × 107
2+
Ru(dmbpy)3
670 [2.1]
g5 × 107
900 (D2O)
2+
Ru(bpy)3
380
520
920 [2.4] (30%)f
2400 [4.6] (50%)f
1200 [2.0] (30%)f
4100 [2.3] (50%)f
n.d.g/2.4
n.d.g/4.6
7.7 × 103
8.4 × 105
2+
Ru(dmphen)3
a In air-saturated buffer (PBS). b In N2-saturated PBS. c Bound by AC1106 in air-saturated PBS [enhancement of excited-state lifetime upon
binding versus a]. d Bound by AC1106 in N2-saturated PBS [enhancement of excited-state lifetime upon binding versus b]. e Amplification of
integrated steady-state emission yield of the AC1106-bound to free metal complex measured in air-saturated PBS/enhancement of the excited-state
lifetime of the AC1106-bound to free metal complex measured in air-saturated PBS. f Because of the lower affinity of AC1106 for this metal
complex, these excited-state lifetime values were derived at the indicated partial saturation. g n.d. ) not determined.
Table 2. Association Constants (Listed in Descending Order)
Single-exponential decay profiles were also maintained in the
Derived Via Competition ELISA for AC1106 and the Indicated
Compounds37
presence of oxygen, a known quencher of the excited state of
Ru(bpy)32+ and its simple derivatives.19 In the event that each
antibody-bound enantiomer were covered by AC1106 to dif-
ferent degrees, it seems reasonable to expect that these would
show disparate accessibility toward O2. In turn, this situation
would give rise to non-exponential decay profiles.34 We
conclude, therefore, that AC1106 provides a similar degree of
protection for the enantiomers of each metal complex.
compd
Ka (M-1
)
[Ru(mv2+-bpy)(bpy)2]4+
2 × 108
8 × 106
3 × 106
2 × 105
8 × 104
6 × 104
1
2+
Ru(dmbpy)3
5
mv2+-bpy
Ru(dmbpy)(bpy)2
2+
2+
Because of the high affinity of AC1106 for Ru(dmbpy)3
2+
2+
and the symmetrical nature of this metal complex, Ru(dmbpy)3
Similarly, when Ru(dmphen)3 was titrated with a large
excess of AC1106, a long-lived component assigned to the
antibody-bound metal complex was observed. Interestingly, the
increase in the excited-state lifetime of the AC1106-bound Ru-
(dmphen)32+ versus free Ru(dmphen)32+ was significantly higher
than that observed with the 2,2′-bipyridine derived metal
complexes (see Table 1). When Ru(phen)32+ was titrated with
AC1106, no new component appeared. Unlike the methyl-free
was chosen to probe the extent that the antibody-bound metal
complex is protected from excited-state deactivation by solvent.
Using a method developed previously to measure solvent
accessibility of luminescent ruthenium(II) metal complexes in
organized media,35 the excited-state lifetimes of free and
2+
antibody-bound Ru(dmbpy)3 were measured in both N2-
saturated H2O-PBS and D2O-PBS (see Table 1). Using these
values, it was determined that some 72% of the antibody-bound
2+
Ru(bpy)3 which bound to AC1106 weakly, no evidence for
2+
Ru(dmbpy)3 is inaccessible to excited-state deactivation Via
2+
binding of Ru(phen)3 to AC1106 in a similar concentration
interaction with solvent. The protection from solvent extended
to this metal complex by AC1106 is, in part, likely the result
of the tight Van der Waals contact between the bound Ru-
(dmbpy)32+ and protein contacts present in the antibody binding
pocket. This characteristic property of antibody binding is
known to exclude solvent from the region of contact between
an antibody and its antigen.36
regime was detected. Consequently, an upper limit to the
2+
association constant for AC1106 binding to Ru(phen)3 was
set at 1 × 103 M-1
.
Competition ELISA Studies. While the luminescent ru-
thenium(II) metal complexes discussed above proved to be
valuable probes of the molecular recognition properties of
AC1106, steady-state and time-resolved luminescence method-
ology was not applicable to non-emissive compounds, such as
[Ru(mv2+-bpy)(bpy)2]4+ or the hapten 1. As a result, a
competition enzyme-linked immunoabsorbant assay (ELISA)
was used. This colorimetric assay monitors the quantity of
antibody bound to a 1-OVA conjugate on the surface of a
microplate. By competing with binding sites on the plate, both
luminescent and nonluminescent molecules can be used to
inhibit the surface binding of AC1106. The resulting inhibition
can then be quantified to yield the relative affinities that AC1106
has for a set of molecules. Hence, the assay signal is dependent
on the amount of microplate-bound antibody, not on the intrinsic
spectroscopic properties of a molecule of interest.
At the higher concentrations required for the time-resolved
luminescence studies, a longer-lived component appeared when
2+
Ru(bpy)3 was titrated with a large excess of AC1106. As
before, this component was assigned to the AC1106-bound metal
complex. Deconvolution of the resulting biphasic luminescence
decay trace yielded excited-state lifetimes for the antibody-
bound Ru(bpy)32+ in air- and N2-saturated solution of 920 and
1200 ns, respectively. Because the concentration of both free
2+
and antibody-bound Ru(bpy)3 can be determined using the
initial luminescent intensities of the two species, use of the time-
resolved data allowed for a determination of the Ka for AC1106
2+
binding to Ru(bpy)3 of (7.7 ( 2.4) × 103 M-1
.
(32) The observed lack of stereospecific binding does not necessarily
mean that the association constants for the two enantiomers are identical.
However, it does imply that the binding pocket is sufficiently lacking in
shape selectivity such that both enantiomers are accommodated.
(33) Before the advent of monoclonal antibody technology, the earliest
molecular recognition studies of antibody-based shape-selectivity and
stereospecificity were made with polyclonal antibodies. For a classic and
authoritative example, see: Landsteiner, K. The Specificity of Serological
Reactions; Harvard University Press: Cambridge, MA, 1945.
(34) Davila, J.; Harriman, A. J. Am. Chem. Soc. 1990, 112, 2686-2690.
(35) Hauenstein, B. L., Jr.; Dressick, W. J.; Buell, S. L.; Demas, J. N.;
DeGraff, B. A. J. Am. Chem. Soc. 1983, 105, 4251-4255.
Table 2 shows the association constants for AC1106 binding
to selected compounds as determined by competition ELISA.
By comparing the affinities of AC1106 for various compounds,
information on what features are important for recognition by
AC1106 can be obtained. For example, as determined in the
steady-state luminescence method, Ru(dmbpy)32+ has a higher
affinity than Ru(dmbpy)(bpy)22+ for AC1106, again indicating
the importance of methyl groups present on the metal ligands
for complexation.37 Furthermore, AC1106 has a higher affinity
for [Ru(mv2+-bpy)(bpy)2]4+ than the pentacationic hapten 1, a
clear indication that more than simple charge-charge interac-
tions are part of the binding process. In addition, AC1106
(36) (a) Padlan, E. A. In Structure of Antigens; Van Regenmortel, M.
H. V., Ed.; CRC Press, Inc.: Boca Raton, FL, 1992; Vol. I, pp 29-42. (b)
Shabat, D.; Itzhaky, H.; Reymond, J.-L.; Keinan, E. Nature 1995, 374, 143-
146.

Molecular Recognition of a Monoclonal Antibody
J. Am. Chem. Soc., Vol. 118, No. 13, 1996 3197
exhibits significantly higher affinities for both [Ru(mv2+-bpy)-
(bpy)2]4+ and the hapten 1 than the methyl viologen containing
ligands, mv2+-bpy and 5, an indication that the presence of a
metal center is a critical recognition element for AC1106.
Finally, there is a dramatic preference of AC1106 for [Ru(mv2+
-
bpy)(bpy)2]4+ over Ru(dmbpy)(bpy)22+. When analyzed as the
sum of its parts, [Ru(mv2+-bpy)(bpy)2]4+ is equivalent to a Ru-
(dmbpy)(bpy)22+ metal complex to which methyl viologen has
been synthetically “appended”. Thus, the significant increase
in the association constant between AC1106 and [Ru(mv2+
-
bpy)(bpy)2]4+ indicates that AC1106 does indeed reserve part
of its binding site for methyl viologen.
Quenching with Methyl Viologen and Oxygen. Because
of the apparent affinity of AC1106 for methyl viologen,
quenching experiments were designed to search for evidence
2+
of simultaneous binding of Ru(dmbpy)(bpy)2 and methyl
Figure 6. Stern-Volmer plot derived from the titration of the AC1106-
2+
Ru(dmbpy)(bpy)2 complex (1.4 × 10-5 M) with methyl viologen.
viologen inside the binding pocket of AC1106. These experi-
ments relied upon the fact that methyl viologen is an avid
quencher of luminescence from ruthenium(II) metal complexes,
due to light-induced electron transfer. Presumably, the close
proximity of reactants bound within the same pocket would
result in immediate (i.e. static) quenching of the triplet excited
state of the metal complex (in this case, immediate refers to
the time window within the 10-ns laser pulse).38 In contrast,
surface-bound methyl viologen could quench the triplet state
of the metal complex located within the antibody pocket by
way of a diffusional process. In the latter case, the triplet
lifetime would decrease systematically with increasing concen-
tration of viologen whereas in the former case the lifetime of
the quenched species would be unresolvable from the excitation
pulse.
Excited-state lifetimes of the antibody-bound (b) and free (O) Ru-
2+
(dmbpy)(bpy)2 were obtained from deconvolution of the biphasic
luminescent decay traces observed in the presence of methyl viologen.
Bimolecular quenching constants of 3.4 × 108 and 1.7 × 109 M-1 s-1
2+
were measured for antibody-bound and free Ru(dmbpy)(bpy)2
respectively.
,
When this experiment was performed, static quenching by
methyl viologen was not observed. Instead, the AC1106-bound
2+
Ru(dmbpy)(bpy)2
was found to be subject to dynamic
quenching (see Figure 6). However, in addition to the com-
ponent assigned to the antibody-bound Ru(dmbpy)(bpy)22+, a
new component appeared with the addition of increasing
amounts (10-3 M) of methyl viologen (see supporting informa-
tion). Throughout the titration with methyl viologen, this new
Figure 7. Plot of relative initial intensities of the AC1106-Ru(dmbpy)-
component was found to have a shorter excited-state lifetime
2+
2+
(bpy)2 complex (b) and free Ru(dmbpy)(bpy)2 (O) as a function
of the concentration of added methyl viologen. The concentration of
each species is proportional to its respective initial luminescence
intensity.
2+ 39
relative to the AC1106-bound Ru(dmbpy)(bpy)2
.
This new
component was also found to be subject to dynamic quenching
effects by methyl viologen. Furthermore, during the titration
the concentration of this new species was observed to increase
2+
relative to that of the antibody-bound Ru(dmbpy)(bpy)2
antibody-bound Ru(dmbpy)(bpy)22+, respectively. The kq for
the new species was found to be identical, within experimental
error, to the corresponding kq for Ru(dmbpy)(bpy)22+ measured
(Figure 7).
A Stern-Volmer analysis proved useful in assigning this new
species. Shown in Figure 6 is the Stern-Volmer plot of the
excited-state lifetimes of the two components as a function of
methyl viologen concentration. Bimolecular quenching con-
stants (kq’s) for quenching by methyl viologen of 1.7 × 109
and 3.4 × 108 M-1 s-1 were measured for the new species and
in the absence of antibody (1.2 × 109 M-1 s-1). Consequently,
2+
the new species is assigned as free Ru(dmbpy)(bpy)2
.
2+
Both the departure of the antibody-bound Ru(dmbpy)(bpy)2
in the presence of methyl viologen and the dynamic nature of
the Stern-Volmer plot for the components shown in Figure 6
argue against the formation of a static, antibody-bound complex
of Ru(dmbpy)(bpy)22+ and methyl viologen. Interestingly, this
displacement behavior is not unique to this antibody-based
2+
(37) No significant, quantifiable inhibition was detected using Ru(bpy)3
,
Ru(dmphen)32+, or methyl viologen despite solution evidence that they bind
to AC1106. Competition ELISA is known to underestimate solution binding
affinities. However, the relative order of affinities determined within the
same assay is reliable. For a discussion of the caveats in using solid phase
immunoassays to measure association constants, see: Butler, J. E. In The
Structure of Antigens; Regenmortel, M. H. V., Ed.; CRC Press: Boca Raton,
FL, 1992; Vol. I, pp 209-259.
system.40 Methyl viologen has also been shown capable of
2+ 41
displacing polysulfonate-bound Ru(bpy)3
.
2+
Quenching by O2 was also investigated using Ru(dmbpy)3
.
Bimolecular quenching constants for quenching by O2 were
measured as 4.0 × 109 and 4.0 × 108 M-1 s-1 for the free and
antibody-bound Ru(dmbpy)32+, respectively. This reduction in
(38) As a model for such an event, [Ru(mv2+-bpy)(bpy)2]4+ in the
presence or abscence of AC1106 was found to be nonluminescent on the
nanosecond time scale. This indicates that the triplet excited state lifetime
in either case is less than 10 ns. For more on the photophysics of [Ru-
(mv2+-bpy)(bpy)2]4+ see ref 60.
2+
the kq for the antibody-bound Ru(dmbpy)3 can be easily
accounted for by a decrease in the diffusion coefficient for the
antibody-bound complex. In addition, the ability of AC1106
to shield the bound metal complex from oxygen, as it does with
(39) Similar displacement behavior was seen with our previously reported
polyclonal system27 derived from 2 (unpublished results). Thus, it may be
that this behavior is a general trend associated with this hapten design.

3198 J. Am. Chem. Soc., Vol. 118, No. 13, 1996
Shreder et al.
2+
solvent, is also likely to play a role. A similar reduction in the
for the AC1106-Ru(bpy)3 complex, namely 920 and 1200
ns, are noticeably shorter. A comparison of these two systems
clearly indicates the excited-state lifetime of an antibody-bound
Ru(bpy)32+ complex is influenced by the nature of the antibody
binding pocket.
kq for quenching by O2 has also been observed for DNA-bound
2+ 43
versus free Ru(phen)3
.
Comparisons to Other Types of Organized Media. En-
hanced luminescence has been seen with derivatives of Ru-
(bpy)32+ and Ru(phen)32+ when bound to certain hydrophobic,
AC1106 falls under the general category of metalloantibodies.
In order to both modify and study the binding properties of
antibodies, there has been an effort to incorporate metal cofactors
into the binding sites of antibodies. Antibody binding pockets
have been engineered to bind a variety of metals using site-
directed mutagenesis.47 In addition, antibodies elicited Via
immunization with metal-based haptens that bind metallopor-
phyrins,48 metal chelates of ethylenediaminetetraacetic acid,49
metal chelates of triethylenetetraamine,28 and mercuric ions50
have been reported. In many cases, seemingly subtle changes
in the metal or metal chelate result in significant changes in
the recognition properties of these antibodies.
2+
anionic media. For example, Ru(phen)3 has been proposed
to bury one of its three 1,10-phenanthroline ligands into
hydrophobic regions on the negatively-charged, double-stranded
DNA scaffold.42 As a consequence of this interaction, the
2+
luminescent lifetime of Ru(phen)3 increases in the presence
of calf-thymus DNA from 470 to 1300 ns in air-saturated
aqueous, buffered solution.43 Similarly, the excited-state lifetime
of Ru(dmphen)32+ increases from 1700 ns in N2-saturated H2O
to 3100 ns with partial immersion in a sodium dodecyl sulfate
(SDS) micelle. This same trend in the presence of SDS is also
observed for a variety of other 2,2′-bipyridine and 1,10-
phenanthroline derived ruthenium(II) metal complexes.35 Based
on the complementary nature of antibody binding44 and the
hydrophobic, cationic nature of the hapten, one might predict
the binding pocket of AC1106 to be both hydrophobic and
anionic. As with DNA and with SDS, the change of the
microenvironment inside the antibody binding pocket of AC1106
relative to the aqueous solution likely plays an important role
in modulating the excited state lifetime of these metal com-
plexes.
AC1106 is not an exception to this observation. The results
in this study indicate that AC1106 exhibits a very structure-
dependent mode of binding. To some degree, this binding mode
may be a consequence of the screening process used to find
this monoclonal antibody. AC1106 was chosen by screening
hybridoma cells derived from the immunization of the hapten
1 against 2. Accordingly, the cross-reactivity of AC1106 is a
compromise between one possible optimized fit for 1 and an
accommodating fit for 2. Indeed, the lack of stereospecific
binding seen for Ru(dmbpy)3 and Ru(dmbpy)(bpy)2 may
be a testament to the conciliatory position imposed by the
screening process. It is possible that a highly specific im-
munological strategy to bind 1 does not conform to stereospe-
cific binding of a seemingly analogous ruthenium(II) metal
complex.
2+
2+
Other factors may also be responsible for the enhanced
luminescence. The H2O/D2O experiment conducted with Ru-
(dmbpy)32+ is clear evidence of the ability of AC1106 to protect
the excited state of the antibody-bound metal complexes from
deactivation Via interaction with solvent. Along this same line
of reasoning, AC1106 also seems to have the ability to protect
the antibody-bound metal complex from excited-state deactiva-
tion by O2. Finally, protein contacts with the bound metal
complexes could result in a reduction of internal vibrational
modes. In turn, the more rigidified metal complex would be
less susceptible to excited-state deactivation caused by, for
example, metal-to-ligand stretching modes.45 In general, similar
factors have been invoked to explain the enhanced emission
Considering the relatively large area that has been reported
to be available to antibodies for surface-surface interactions
between antibody and antigen (up to 880 Å2)51 and the potential
for a variety of complementary interactions between 1 and
AC1106, a large degree of the specificity of AC1106 is
dependent on the presence or abscence of a surprisingly small
hydrophobic portion (i.e. the methyl groups) of the luminescent
metal complexes used in this study. This result is particularly
relevant to the field of catalytic antibodies. Here, despite the
relatively small size of haptenic transition state analogues, it
has been observed that in some cases a small percentage of
hapten-specific monoclonal antibodies are catalytic.52 This
observation, taken with the highly focused specificity of
2+
seen when Ru(phen)3 binds to double-stranded DNA.22b,46
Interestingly, in our previously reported polyclonal antibody
system elicited Via immunization with 2, the ruthenium(II)
analogue of the cobalt(III)-based hapten used in this study, we
measured excited-state lifetimes of 1300 and 1500 ns for the
polyclonal antibody-bound Ru(bpy)32+ in air- and N2-saturated
solution, respectively. The corresponding excited-state lifetimes
(47) (a) Roberts, V. A.; Iverson, B. L.; Iverson, S. A.; Benkovic, S. J.;
Lerner, R. A.; Getzoff, E. D.; Tainer, J. A. Proc. Natl. Acad. Sci. U.S.A.
1990, 87, 6654-6658. (b) Iverson, B. L.; Iverson, S. A.; Roberts, V. A.;
Getzoff, E. D.; Tainer, J. A.; Benkovic, S. J.; Lerner, R. A. Science 1990,
249, 659-652. (c) Wade, W. S.; Koh, J. S.; Nianhe, H.; Hoekstra, D. M.;
Lerner, R. A. J. Am. Chem. Soc. 1993, 115, 4449-4456. (d) Gregory, D.
S.; Martin, A. C. R.; Cheetham, J. C.; Rees, A. R. Protein Eng. 1993, 6,
29-35. (e) Barbas, C. F., III; Rosenblum, J. S.; Lerner, R. A. Proc. Natl.
Acad. Sci. U.S.A. 1993, 90, 6385-6389.
(48) (a) Schwabacher, A. W.; Weinhouse, M. I.; Auditor, M.-T. M.;
Lerner, R. A. J. Am. Chem. Soc. 1989, 111, 2344-2346. (b) Keinan, E.;
Sinha, S. C.; Sinha-Bagchi, A.; Benory, E.; Ghozi, M. C.; Eshhar, Z.; Green,
B. S. Pure Appl. Chem. 1990, 62, 2013-2019.
(49) (a) Reardan, D. T.; Meares, C. F.; Goodwin, D. A.; McTigue, M.;
David, G. S.; Stone, M. R.; Leung, J. P.; Bartholomew, R. M.; Frincke, J.
M. Nature 1985, 316, 265-268. (b) Love, R. A.; Villafranca, J. E.; Aust,
R. M.; Nakamura, K. K.; Jue, R. A.; Major, J. G., Jr.; Radhakrishnan, R.;
Butler, W. F. Biochemistry 1993, 32, 10950-10959.
(50) Wylie, D. E.; Lu, D.; Carlson, L. D.; Carlson, R.; Babacan, K. F.;
Schuster, S. M.; Wagner, F. W. Proc. Natl. Acad. Sci. U.S.A. 1992, 89,
4104-4108.
(40) One possible explanation for the antibody-based displacement
behavior is that disfavored electrostatic interactions result from simulta-
neously binding two dications next to each other in an antibody pocket. A
second plausible explanation is a binding scheme in which the positively
charged pyridine ring of methyl viologen affords some affinity for an
antibody binding site that recognizes Ru(dmbpy)(bpy)22+. In either case, it
is possible that a dual hapten design that incorporates a neutral electron
donor or acceptor tethered to Ru(bpy)32+ may provide an antibody binding
pocket capable of simultaneously binding the corresponding redox-active
pair .
(41) Meisel, D.; Matheson, M. S. J. Am. Chem. Soc. 1977, 99, 6577-
6581.
(42) Turro, N. J.; Barton, J. K.; Tomalia, D. A. Acc. Chem. Res. 1991,
24, 332-340.
(43) Tossi, A. B.; Kelly, J. M. Photochem. Photobiol. 1989, 49, 545-
556.
(44) Pressman, D. In Molecular Structure and Biological Specificity;
Pauling, L., Itano, H. A., Eds.; Waverly Press, Inc.: Baltimore, 1957; No.
2, pp 1-17.
(45) (a) Lumpkin, R. S.; Kober, E. M.; Worl, L. A.; Murtaza, Z.; Meyer,
T. J. J. Phys. Chem. 1990, 94, 239-243. (b) Maruszewski, K.; Strommen,
D. P.; Kincaid, J. R. J. Am. Chem. Soc. 1993, 115, 8345-8350.
(46) Barton, J. K.; Danishefsky, A. T.; Goldberg, J. M. J. Am. Chem.
Soc. 1984, 106, 2172-2176.
(51) (a) Kabat, E. A. Structural Concepts in Immunology and Immu-
nochemistry; Holt, Rinehart, and Winston, Inc.: New York, 1968; pp 82-
100. (b) Davies, D. R.; Padlan, E. A. Annu. ReV. Biochem. 1990, 59, 439-
473.

Molecular Recognition of a Monoclonal Antibody
J. Am. Chem. Soc., Vol. 118, No. 13, 1996 3199
indicated otherwise, and were reproducible to within 5%. Biphasic
decay profiles were analyzed according to If(t) ) A1 exp(-t/τ1) + A2
exp(-t/τ2) where τ1 and τ2 refer to luminescence lifetimes of the two
components and A1 and A2 refer to the initial luminescence intensities
of the respective components.
AC1106, would seem to indicate that small regions in the
structure of a ligand (or substrate) can dominate the antibody-
antigen interactions. If these dominant interactions do not
distinguish starting material from transition state, catalytic
activity will be limited.53
(b) Competition ELISA. The hapten or inhibitor of interest was
serially diluted by a factor of 2 in PBS across the length of a 8 × 12
well polypropylene microplate (Fisher Scientific) leaving one well near
the center of each row free of hapten or inhibitor as a control. To
each well was placed an aliquot (50 µL) of a 5 × 10-8 M solution of
AC1106 in PBS. These solutions were allowed to incubate overnight
at room temperature. At the end of this incubation period, an ELISA
plate was blocked with 5 mg/mL of bovine serum albumin (BSA,
Sigma) in PBS for 2 h at 37 °C and washed three times with PBS and
shaken dry. The solutions were transferred row by row using a
multichannel pipetter (1 × 12) to this plate and the incubation was
continued for an additional 20 min. The plate was then washed three
times with PBS and shaken dry. A solution of goat anti-mouse IgG
(0.8 mg/mL) (H+L, Pierce Chemical Co.) was diluted 1:500 in a 5
mg/mL solution of BSA in PBS. Fifty microliters of this secondary
antibody solution was portioned to each well and incubated for 1 h at
37 °C. The plate was then washed vigorously 10 times with PBS and
shaken dry. Fifty microliters of 1-Step ABTS (2,2′-azinobis(3-
ethylbenzothiazolin-6-sulfonic acid) diammonium salt, Pierce Chemical
Co.) was portioned to each well and was allowed to develop for
approximately 10 min. The color change in each well was monitored
spectrophotometrically at 405 nm using a EL311s automated microplate
reader (BIO-TEK Instruments, Inc.).
Summary and Conclusion
The characterization of AC1106 elicited Via immunization
with a Co(dmbpy)(bpy)23+-methyl viologen hapten (1) has been
described. AC1106 was found cross-reactive for a variety of
luminescent ruthenium(II) metal complexes which, when bound
by this monoclonal antibody, exhibit a 2.2-4.6-fold enhance-
ment in their excited-state lifetimes in air-saturated solution.
AC1106 was found to be highly specific for methylated
derivatives of Ru(bpy)32+ and Ru(phen)32+ in the order of Ru-
2+
2+
2+
(dmbpy)3 > Ru(dmbpy)(bpy)2 > Ru(dmphen)3 > Ru-
2+
(bpy)3 . Ru(phen)32+. The specificity of this antibody for
these metal complexes is the result of a combination of factors
that includes hydrophobicity and shape-selectivity. Through a
combination of steady-state and time-resolved luminescence
techniques, the affinities of AC1106 for these metal complexes
were found to range over four orders of magnitude from g5 ×
107 to e1 × 103 M-1
.
As demonstrated in this study, luminescent ruthenium(II)
metal complexes, as model antibody ligands, are useful spec-
troscopic probes of antibody-antigen interactions. In particular,
this study has focused on the unique recognition properties of
a single monoclonal antibody. Other antibodies may exhibit
considerably different recognition properties that result in
stereospecific binding or even greater modulation of the excited-
state lifetimes of Ru(bpy)32+ and its derivatives. Based on the
examples to date, the high specificity and affinity exhibited by
this class of antibodies should be useful in the design of
(c) Association Constants. The equation to solve for the association
constants using the steady-state luminescence data was derived from
the simple equilibrium expressed in eq 1 where Ka is the association
constant, X is the concentration of the AC1106-bound metal complex,
Rf is the fixed concentration of metal complex (1 × 10-6 M), and C is
the concentration of AC1106 in antibody binding sites. Because the
change in emission intensity of the ruthenium(II) metal complex (E)
observed upon titration with AC1106 is the product of the fraction of
the total metal complex that is antibody-bound and the total difference
in the integrated steady-state emission intensity of the antibody-bound
and free metal complexes (Ec), eq 2 can be written. Solving for X in
eq 1 and substituting into eq 2 yields eq 3. Using iterative, curve-
fitting procedures (KaleidaGraph, version 3.0.2) and eq 3, association
constants were derived from a plot of E versus C. After entering initial
guesses for Ka and Ec, these variables were allowed to float freely over
the iteration.
antibody-based systems that take advantage of the excited-state
2+
properties of Ru(bpy)3
.
Experimental Section
Methods. (a) Luminescence Spectroscopy. All luminescence
measurements were conducted in phosphate buffered saline (PBS: 10
mM sodium phosphate, 120 mM NaCl, pH 7.4). Steady-state lumi-
nescence spectra were recorded with a fully-corrected Perkin Elmer
LS5 spectrofluorimeter. Solutions for steady-state luminescence studies
were adjusted to possess an absorbance of less than 0.05 at the excitation
wavelength of 450 nm and were air-equilibrated. Luminescence spectra
were recorded between 500 and 800 nm and luminescence yields were
determined by integration of the spectrum over the entire range. Where
X
Ka )
(1)
(2)
(Rf - X)(C - X)
X
E ) E
cRf
2+
appropriate, Ru(bpy)3 was used as a reference compound.
All flash photolysis studies were made with a frequency-doubled,
Q-switched Quantel YG481 Nd:YAG laser (pulse width 10 ns;
maximum pulse energy 40 mJ). The laser intensity was attenuated
with crossed-polarizers and all studies were carried out under conditions
where the luminescence intensity was a linear function of laser power.
Solutions were adjusted to possess an absorbance of ca. 0.05 at 532
nm and were purged continuously during the experiment with N2, O2,
or mixtures thereof according to the needs of the experiment. Kinetic
studies were carried out at a fixed wavelength of 610 nm with 20
individual laser shots being averaged, corrected for changes in the
baseline, and analyzed with a microcomputer using nonlinear, least-
squares iterative procedures. Reported kinetic parameters were theresult
of a satisfactory fit to a single-exponential decay process, unless
E )
2
[(K C + K R + 1) - (K C + K R + 1)2 - 4K CR ]E
x
a
a
f
a
a
f
a
f
c
(3)
2KRf
The equation to solve for the association constants using competition
ELISA was derived from the relationships expressed in eqs 4 and 5.
For eq 4 (analogous to eq 1), Y is the concentration of the AC1106-
inhibitor complex, L is the concentration of the inhibitor, and Cf is the
fixed concentration of AC1106 (1 × 10-7 M in antibody binding sites).
The A405 value observed in the presence of inhibitor (A), proportional
to the concentration of free antibody, is related to the fraction of bound
antibody through eq 5 where Ao is the A405 value observed in the absence
of inhibitor. Solving for Y in eq 4 and substituting into eq 5 yields eq
6. Using iterative, curve-fitting procedures and eq 6, association
constants were derived from a plot of A versus L. After entering initial
(52) For example, see: (a) Yu, J.; Hsieh, L. C.; Kochersperger, L.;
Yonkovich, S.; Stephans, J. C.; Gallop, M. A.; Schultz, P. G. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 339-341. (b) Thorn, S. N.; Daniels, R. G.; Auditor,
M.-T. M.; Hilvert, D. Nature 1995, 373, 228-230.
(53) Wallace, M. B.; Iverson, B. L. J. Am. Chem. Soc. 1996, 118, 251-
252.

3200 J. Am. Chem. Soc., Vol. 118, No. 13, 1996
Shreder et al.
guesses for Ka and Ao, these variables were allowed to float freely over
the iteration.
Dichloromethane and acetonitrile were distilled from CaH2 under
N2. All reagents were of the highest grade available and were purchased
from the Aldrich Chemical Co. unless indicated otherwise. [Ru(4,4′-
dimethyl-2,2′-bipyridine)3]2+ (Ru(dmbpy)32+) and [Ru(4,7′-dimethyl-
1,10′-phenanthroline)3]2+ (Ru(dmphen)32+) were prepared according to
Y
Ka )
(4)
(5)
the general procedure of Palmer and Piper for the synthesis of Ru-
(L - Y)(Cf - Y)
2+ 58
(bpy)3
.
[Ru(4,4′-dimethyl-2,2′-bipyridine)(2,2′-bipyridine)2]2+ (Ru-
(dmbpy)(bpy)22+),59 [Ru(mv2+-bpy)(bpy)2]4+ 60 and mv2+-bpy60 (see
,
Figure 1) were synthesized according to literature procedures. Ruthe-
Y
-
nium metal complexes were converted to their NO3 salts Via
A ) Ao - A
oCf
purification on silica gel with 50:45:5 CH3CN:H2O:saturated KNO3.61
1
All literature compounds gave satisfactory H NMR, 13C NMR, and
mass spectrometric analyses. The hapten 1 and its ruthenium(II)
analogue 2 were synthesized according to the scheme shown in Figure
2.
A ) Ao -
2
[(K L + K C + 1) - (K L + K C + 1)2 - 4K LC ]A
x
a
a
f
a
a
f
a
f
o
[Co(2,2′-bipyridine)2(1-(3-(4′-methyl-2,2′-bipyridin-4-yl)propyl)-
1′-(carboxylpentyl)-4,4′-bipyridinium)]‚(PF6-)5 (1). Co(2,2′-bipyri-
dine)2Cl2]Cl‚H2O62 (19 mg, 0.038 mmol) and the metal ligand 5 (25
mg, 0.036 mmol) were dissolved in 3.0 mL of H2O and allowed to
stand at room temperature for 1.5 h whereupon the initially purple
solution turned a light yellow-orange. Dioxane (0.7 mL) and 1.0 mL
of trifluoroacetic acid were added and the resulting solution was allowed
to stand for 40 min. The solvent was removed at room temperature
under high vacuum and the resulting yellow oil was chromatographed
using reverse-phase (C2/C18) chromatography. Two light yellow bands
eluted using 15% CH3CN/H2O at a flow rate of 1.25 mL/min. The
last yellow band to elute, according to 1H NMR analysis, was the tert-
butyl ester of 1. The first yellow band to elute proved to be the
hapten: 1H NMR (CD3OD) δ 1.48 (m, 2H, CH2), 1.70 (m, 2H, CH2),
2.12 (m, 2H, CH2), 2.34 (t, 2H, COCH2), 2.36 (m, 2H, bpy-CH2CH2),
2.70 (s, 3H, CH3), 3.16 (t, 2H, bpy-CH2), 4.78 (t, 2H, N+-CH2), 4.94
(t, 2H, N+-CH2), 7.31-9.40 (30H, m, aromatic H’s); 13C NMR (CD3-
OD) δ 21.5, 25.2, 26.5, 31.8, 32.1, 32.5, 34.4, 62.2, 63.0, 128.3, 128.4,
128.4, 128.6, 128.7, 129.4, 129.5, 132.6, 132.7, 133.3, 133.4, 145.0,
145.1, 147.1, 147.3, 151.2, 151.3, 151.5, 151.7, 152.5, 152.6, 156.5,
157.0, 157.3, 158.9, 159.0, 160.1, 160.2, 160.3, 177.0. For analytical
purposes only, the hapten was converted to its PF6- salt and lyophilized
to yield 17 mg (30%) of a light orange solid: Anal. Calcd for C50H50-
CoF30N8O2P5: C, 38.04; H, 3.19; N, 7.10. Found: C, 38.20; H, 3.56;
N, 7.04.
(6)
2KaCf
Materials. (a) Preparation of Protein Conjugates. Hapten 1 (5
mg), 10 mg of keyhole limpet hemocyanin (Pierce Chemical Co.), and
50 µL of 0.1 M N-hydroxysulfosuccinimide (Pierce Chemical Co.) were
placed in 0.9 mL total volume of doubly distilled water and adjusted
to pH 7 with 0.1 M NaOH using a syringe. [1-Ethyl-3-(dimethylami-
no)propyl]carbodiimide (100 µL of 1 M) in double-distilled water was
added and the solution was briefly vortexed. After 1 h at room
temperature, 9.0 mL of PBS was added for a final protein concentration
of 1 mg/mL and the resulting solution was dialyzed exhaustively against
PBS at 4 °C using Spectra/Por 7 dialysis tubing, 50 000 MWCO
(Spectrum). Ovalbumin (OVA) conjugates of 1 and 2 were prepared
in a similar fashion.
(b) Production and Isolation of Antibodies. A male Balb/cJ mouse
(Jackson Laboratories, Bar Harbor, ME) was injected intraperitonially
with 150 µL of a 1 mg/mL solution of the hapten 1-KLH conjugate
in PBS emulsified with MPL + TDM adjuvant (RIBI Immunochem
Research, Inc., Hamilton, MT). A total of four injections, 21 days
apart, were performed. Fourteen days after the final injection, a final
intravenous injection was performed with 200 µL of a 1 mg/mL solution
of the hapten 1-KLH conjugate in PBS 3 days prior to removal of the
spleen. Hybridoma cells were produced from these spleen cells
according to literature procedure54 and were screened Via ELISA55
against both the 1-OVA and 2-OVA conjugates. One hybridoma line
was found positive for both conjugates and was used in the production
of ascites fluid. AC1106 was purified from ascites using NH4SO4
precipitation followed by protein G chromatography according to a
previously published protocol.56 Antibody concentration was deter-
mined spectroscopically using an A280 value of 1.4 for a 1.0 mg/mL
rabbit antibody solution (ꢀ280 ) 220 000 M-1 cm-1).57
(c) Preparation of ELISA Plates. In each well of a 8 × 12,
polystyrene assay plate (Corning Glass Works, Corning, NY) 50 µL
of a 0.5 µg/mL solution of the hapten 1-OVA conjugate was allowed
to evaporate at 37 °C overnight. Each well was fixed with 100 µL of
MeOH for 30 min after which the MeOH was shaken out and the
residual solvent was allowed to evaporate.
Syntheses. Proton nuclear magnetic resonance (1H NMR) and
carbon nuclear magnetic resonance (13C NMR) spectra were recorded
on a General Electric QE-300 NMR spectrometer using the residual
peaks in the deuterated solvents as internal standards. Elemental
analyses were performed by Atlantic Microlabs. Low-resolution FAB
spectra were obtained using a Finnigan-MAT TSQ-70 instrument.
3-Nitrobenzyl alcohol was used as the matrix for FAB mass spectra.
High-resolution mass spectra were recorded on a VG ZAB-2E
instrument. Reverse-phase chromatography was performed with a
FPLC system (Pharmacia LKB) using a HR10/10 column filled with
PepRPC 15 µm (C2/C18).
[Ru(2,2′-bipyridine)2(1-(3-(4′-methyl-2,2′-bipyridin-4-yl)propyl)-
1′-(carboxylpentyl)-4,4′-bipyridinium)]‚(PF6-)4‚4H2O (2). The pro-
tected ruthenium(II) metal complex 6 (94 mg, 0.059 mmol) in 10 mL
of 1:1 trifluoroacetic acid/water was allowed to stand at room
temperature for 45 min. The solvent was removed using a rotary
evaporator and the resulting solid purified on neutral alumina using a
gradient of MeOH/H2O. The purified solid was suspended in ap-
proximately 10 mL of H2O, NH4PF6 was added, and the insoluble
fraction was collected using vacuum filtration. The resulting orange
solid was then taken up in dry CH3CN and filtered to remove any
insoluble inorganic salts. The solvent was removed using a rotary
evaporator to yield 80 mg (88%) of the final product: 1H NMR (CD3-
CN) δ 1.40 (2H, m, CH2), 1.62 (2H, m, CH2), 2.03 (2H, m, CH2), 2.28
(2H, t, COCH2), 2.43 (2H, m, bpy-CH2CH2), 2.51 (3H, s, bpy-CH3),
2.93 (2H, t, bpy-CH2), 4.64 (2H, t, N+-CH2), 4.75 (2H, t, N+-CH2),
7.22-9.05 (30H, m, aromatic H’s); 13C NMR (CD3CN) δ 21.0, 24.5,
25.6, 31.2, 31.7, 39.9, 62.0, 62.5, 125.0, 125.9, 128.0, 128.1, 128.3,
129.1, 138.4, 146.4, 146.5, 150.6, 150.9, 151.3, 151.5, 152.0, 152.3,
152.4, 152.7, 157.1, 157.7, 157.8, 160.2, 160.7, 175.2; MS FAB, M+:
m/z 1331; HRMS, M+: 1331.2004 (Calcd for C50H50F18N8O2P3102Ru:
1331.2026). Anal. Calcd for C50H50F24N8O2P4Ru‚4H2O: C, 38.80;
H, 3.78; N, 7.24. Found: C, 38.67; H, 3.60; N, 7.24.
tert-Butyl-6-bromohexanoate (3). 6-Bromohexanoyl chloride (4.5
g, 21 mmol) and pyridine (1.8 mL, 22 mmol) were dissolved in 50 mL
(54) Galfre, G.; Howe, S. C.; Milstein, C.; Butcher, G. W.; Howard, J.
C. Nature 1977, 266, 550-552.
(58) Palmer, R. A.; Piper, T. S. Inorg. Chem. 1966, 5, 864-865.
(59) Dose, E. V.; Wilson, L. J. Inorg. Chem. 1978, 17, 2660-2666.
(60) Yonemoto, E. H.; Saupe, G. B.; Schmehl, R. H.; Hubig, S. M.;
Riley, R. L.; Iverson, B. L.; Mallouk, T. E. J. Am. Chem. Soc. 1994, 116,
4786-4795.
(55) Harlow, E.; Lane, D. Antibodies: A Laboratory Manual; Cold Spring
Harbor Laboratory: Cold Spring Harbor, New York, 1988; pp 564-565.
(56) Stephen, D. B.; Iverson, B. L. Biochem. Biophys. Res. Commun.
1993, 192, 1439-1444.
(61) Sessler, J. L.; Capuano, V. L.; Burrell, A. K. Inorg. Chim. Acta
1993, 204, 93-101.
(57) Ey, P. L.; Prowse, S. J.; Jenkin, C. R. Immunochemistry 1978, 15,
429-436.
(62) Vlcek, A. A. Inorg. Chem. 1967, 6, 1425-1427.

Molecular Recognition of a Monoclonal Antibody
J. Am. Chem. Soc., Vol. 118, No. 13, 1996 3201
of CH2Cl2 and cooled to 4 °C. tert-Butyl alcohol (2.5 mL, 27 mmol)
was added to the solution and stirred for 16 h at 4 °C whereupon a
fine white precipitate of pyridine hydrochloride formed. The resulting
suspension was washed with H2O (3 × 100 mL), 10% K2CO3 (1 ×
100 mL), and again with H2O (1 × 100 mL). The organic layer was
dried over Na2SO4 and distilled under high vacuum using a short path
distillation apparatus to yield 4.1 g (78%) of a clear liquid: 1H NMR
δ (CDCl3) 1.42 (s, 9H, CH3), 1.44 (m, 2H, CH2), 1.59 (m, 2H, CH2),
1.83 (m, 2H, CH2), 2.21 (t, 2H, CH2CO), 3.39 (t, 2H, CH2Br); 13C
NMR δ (CDCl3) 24.1, 27.5, 28.0, 32.4, 33.5, 35.2, 80.1, 172.8; MS
FAB, [M + H]+: 251 m/z; HRMS, [M + H]+: 251.0646 (calcd for
C10H20O2Br: 251.0647).
35.9, 62.7, 63.0, 81.5, 122.7, 123.7, 125.5, 126.3, 128.3, 147.1,147.2,
150.0, 150.5, 150.6, 151.2, 151.3, 152.3, 156.8, 157.3, 174.6; MS FAB,
M+: m/z 538; HRMS, M+: 538.3295 (calcd for C34H42N4O2: 538.3308).
[Ru(2,2′-bipyridine)2(1-(3-(4′-methyl-2,2′-bipyridin-4-yl)propyl)-
1′-(tert-butoxycarbonylpentyl)-4,4′-bipyridinium)]‚(PF6-)4 (6). The
metal ligand 5 (54 mg, 0.078 mmol) and cis-dichlorobis(2,2′-bipyr-
idine)ruthenium(II) dihydrate (35 mg, 0.067 mmol) were heated at reflux
in MeOH for 24 h and the solvent was removed using a rotary
evaporator. The resulting reddish solid was taken up in 20 mL of H2O
and precipitated with NH4PF6. The precipitate was collected using
vacuum filtration and purified on neutral alumina using a gradient of
MeOH/CH2Cl2 to yield 81 mg (76%) of a bright orange solid: 1H NMR
δ (CD3OD/CD3CN) 1.38 (s, 9H, C(CH3)3), 1.41 (m, 2H, CH2), 1.60
(m, 2H, CH2), 2.03 (m, 2H, CH2), 2.20 (t, 2H, CH2CO), 2.45 (m, 2H,
CH2), 2.51 (s, 3H, bpy-CH3), 2.94 (m, 2H, CH2), 4.64 (t, 2H, N+-
CH2), 4.78 (t, 2H, N+-CH2), 7.22-9.16 (30H, aromatic-H’s); 13C NMR
δ (CD3CN) 21.2, 25.0, 26.0, 28.3, 31.8, 32.0, 32.3, 35.7, 61.6, 62.3,
80.6, 125.2, 125.3, 125.4, 126.1, 126.9, 128.1, 128.4, 128.5, 128.7,
129.2, 138.6, 146.9, 147.4, 150.2, 150.4, 151.4, 151.6, 151.9, 152.5,
152.6, 152.9, 153.4, 157.7, 158.0, 173.5; MS FAB, M+: 1242 m/z;
HRMS, M+: 1242.3021 (calcd for C54H58F12N8O2P2102Ru: 1242.3010).
[1-(tert-Butoxycarbonylpentyl)-4,4′-bipyridinium]‚(Br-) (4). 4,4′-
Bipyridine (1.1 g, 7.0 mmol) and tert-butyl 6-bromohexanoate (0.34
g, 1.4 mmol) were heated at reflux in 12 mL of dry CH3CN for 48 h
and the solvent was removed using a rotary evaporator. The resulting
solid was purified on neutral alumina using a gradient of MeOH/CHCl3
to give 0.2 mg (35%) of a tan solid: 1H NMR δ (CDCl3) 1.31 (s, 9H,
CH3), 1.37 (m, 2H, CH2), 1.57 (m, 2H, CH2), 2.06 (m, 2H, CH2), 2.14
(t, 2H, CH2CO), 4.97 (t, 2H, N+-CH2), 7.69 (d, 2H, H3′ and H5′), 8.41
(d, 2H, H2′ and H6′), 8.77 (d, 2H, H3 and H5), 9.72 (d, 2H, H2 and H6);
13C NMR δ (CDCl3) 24.0, 25.2, 27.9, 31.4, 34.8, 61.0, 80.1, 121.4,
125.8, 140.8, 145.9, 151.2, 172.6; MS FAB, [M + H]+: 327 m/z;
HRMS, [M + H]+: 327.2072 (calcd for C20H27N2O2: 327.2073).
[1-(3-(4′-Methyl-2,2′-bipyridin-4-yl)propyl)-1′-(tert-butoxy-
carbonylpentyl)-4,4′-bipyridinium]‚(Br-)2 (5). 4-(3-Bromopropyl)-
4′-methyl-2,2′-bipyridine60 (200 mg, 0.69 mmol) and [1-(tert-butoxy-
carbonylpentyl)-4,4′-bipyridinium]‚(Br-) (300 mg, 0.74 mmol) were
heated at reflux in 25 mL of dry CH3CN for 40 h whereupon a yellow
precipitate formed. The suspension was cooled to room temperature
and filtered using vacuum filtration. The precipitate was washed with
diethyl ether (3 × 25 mL) and dried under vacuum to give 225 mg
(45%) of a mustard yellow solid: 1H NMR δ (CD3OD) 1.42 (s, 9H,
C(CH3)3), 1.45 (m, 2H, CH2), 1.67 (m, 2H, CH2), 2.10 (m, 2H, CH2),
2.27 (t, 2H, CH2CO), 2.45 (s, 3H, bpy-CH3), 2.56 (m, 2H, CH2), 2.97
(t, 2H, bpy-CH2), 4.76 (t, 2H, N+-CH2), 4.88 (t, 2H, N+-CH2), 7.29
(1H, d, H5′), 7.40 (1H, d, H5), 8.09 (s, 1H, H3′), 8.13 (s, 1H, H3), 8.49
(d, 1H, H6′), 8.54 (d, 1H, H6), 8.63-8.67 (m, 4H, (viologen) H3, H3′,
H5, and H5′), 9.27-9.34 (m, 4H, (viologen) H2, H2′, H6′, and H6), 9.32
(d, 2H,); 13C NMR δ (CD3OD) 21.2, 25.4, 26.6, 28.4, 32.2, 32.5, 32.8,
Acknowledgment. We would like to acknowledge the Searle
Foundation (Chicago Community Trust), the Welch Foundation
(F1188), and an NSF-PYI award (CHE-9157440) to B.L.I. for
financial support; Prof. Jonathan Sessler for his helpful sug-
gestions and comments; Dr. Edward Yonemoto for his synthetic
advice; Ms. Suman Olivelle and Mr. Brian Thompson for their
assistance with the hybridoma and ascites work, respectively;
and Michael Zingalis and Todd Russell for their synthetic
assistance.
Supporting Information Available: Addition spectra and
data plots (3 pages). This material is contained in many libraries
on microfiche, immediately follows this article in the microfilm
version of the journal, can be ordered from the ACS, and can
be downloaded from the Internet; see any current masthead page
for ordering information and Internet access instructions.
JA952014O