Substrate-selective supramolecular tandem assays: Monitoring enzyme inhibition of arginase and diamine oxidase by fluorescent dye displacement from calixarene and cucurbituril macrocycles

Article abstract of DOI:10.1021/ja904165c

A combination of moderately selective host-guest binding with the impressive specificity of enzymatic transformations allows the real-time monitoring of enzymatic reactions in a homogeneous solution. The resulting enzyme assays ("supramolecular tandem assays") exploit the dynamic binding of a fluorescent dye with a macrocyclic host in competition with the binding of the substrate and product. Two examples of enzymatic reactions were investigated: the hydrolysis of arginine to ornithine catalyzed by arginase and the oxidation of cadaverine to 5-aminopentanal by diamine oxidase, in which the substrates have a higher affinity to the macrocycle than the products ("substrate-selective assays"). The depletion of the substrate allows the fluorescent dye to enter the macrocycle in the course of the enzymatic reaction, which leads to the desired fluorescence response. For arginase, p-sulfonatocalix[4]arene was used as the macrocycle, which displayed binding constants of 6400 M^-1 with arginine, 550 M^-1 with ornithine, and 60 000 M^-1 with the selected fluorescent dye (1-aminomethyl-2,3-diazabicyclo[2.2.2]oct-2-ene); the dye shows a weaker fluorescence in its complexed state, which leads to a switch-off fluorescence response in the course of the enzymatic reaction. For diamine oxidase, cucurbit[7]uril (CB7) was used as the macrocycle, which showed binding constants of 4.5 × 10⁶ M^-1 with cadaverine, 1.1 × 10⁵ M^-1 with 1-aminopentane (as a model for the thermally unstable 1-aminopentanal), and 2.9 × 10⁵ M^-1 with the selected fluorescent dye (acridine orange, AO); AO shows a stronger fluorescence in its complexed state, which leads to a switch-on fluorescence response upon enzymatic oxidation. It is demonstrated that tandem assays can be successfully used to probe the inhibition of enzymes. Inhibition constants were estimated for the addition of known inhibitors, i.e., S-(2-boronoethyl)-L- cysteine and 2(S)-amino-6-boronohexanoic acid for arginase and potassium cyanide for diamine oxidase. Through the sequential coupling of a "product- selective" with a "substrate-selective" assay it was furthermore possible to monitor a multistep biochemical pathway, namely the decarboxylation of lysine to cadaverine by lysine decarboxylase followed by the oxidation of cadaverine by diamine oxidase. This "domino tandem assay" was performed in the same solution with a single reporter pair (CB7/AO).

Full text of DOI:10.1021/ja904165c

Published on Web 07/23/2009
Substrate-Selective Supramolecular Tandem Assays:
Monitoring Enzyme Inhibition of Arginase and Diamine
Oxidase by Fluorescent Dye Displacement from Calixarene
and Cucurbituril Macrocycles
Werner M. Nau,* Garima Ghale, Andreas Hennig, Hu¨seyin Bakirci, and
David M. Bailey
School of Engineering and Science, Jacobs UniVersity Bremen, Campus Ring 1,
D-28759 Bremen, Germany
Received May 22, 2009; E-mail: w.nau@jacobs-university.de
Abstract: A combination of moderately selective host-guest binding with the impressive specificity of
enzymatic transformations allows the real-time monitoring of enzymatic reactions in a homogeneous solution.
The resulting enzyme assays (“supramolecular tandem assays”) exploit the dynamic binding of a fluorescent
dye with a macrocyclic host in competition with the binding of the substrate and product. Two examples of
enzymatic reactions were investigated: the hydrolysis of arginine to ornithine catalyzed by arginase and
the oxidation of cadaverine to 5-aminopentanal by diamine oxidase, in which the substrates have a higher
affinity to the macrocycle than the products (“substrate-selective assays”). The depletion of the substrate
allows the fluorescent dye to enter the macrocycle in the course of the enzymatic reaction, which leads to
the desired fluorescence response. For arginase, p-sulfonatocalix[4]arene was used as the macrocycle,
which displayed binding constants of 6400 M-1 with arginine, 550 M-1 with ornithine, and 60 000 M-1 with
the selected fluorescent dye (1-aminomethyl-2,3-diazabicyclo[2.2.2]oct-2-ene); the dye shows a weaker
fluorescence in its complexed state, which leads to a switch-off fluorescence response in the course of the
enzymatic reaction. For diamine oxidase, cucurbit[7]uril (CB7) was used as the macrocycle, which showed
binding constants of 4.5 × 106 M-1 with cadaverine, 1.1 × 105 M-1 with 1-aminopentane (as a model for
the thermally unstable 1-aminopentanal), and 2.9 × 105 M-1 with the selected fluorescent dye (acridine
orange, AO); AO shows a stronger fluorescence in its complexed state, which leads to a switch-on
fluorescence response upon enzymatic oxidation. It is demonstrated that tandem assays can be successfully
used to probe the inhibition of enzymes. Inhibition constants were estimated for the addition of known
inhibitors, i.e., S-(2-boronoethyl)-L-cysteine and 2(S)-amino-6-boronohexanoic acid for arginase and
potassium cyanide for diamine oxidase. Through the sequential coupling of a “product-selective” with a
“substrate-selective” assay it was furthermore possible to monitor a multistep biochemical pathway, namely
the decarboxylation of lysine to cadaverine by lysine decarboxylase followed by the oxidation of cadaverine
by diamine oxidase. This “domino tandem assay” was performed in the same solution with a single reporter
pair (CB7/AO).
Introduction
on the competitive encapsulation of a fluorescent dye and an
enzymatic product by macrocyclic hosts to monitor enzymatic
The monitoring of enzymatic processes is of fundamental
importance for the understanding of biological phenomena.1
Inspired by indicator displacement and synthetic pore strate-
gies,2-13 we have recently introduced a label-free method based
reactions (Scheme 1).14-16 In our previous examples, the
investigated enzymes transformed a weak competitor (substrate)
into a strong competitor (product) which displaced the fluores-
(9) Das, G.; Talukdar, P.; Matile, S. Science 2002, 298, 1600–1602.
(10) Sorde, N.; Das, G.; Matile, S. Proc. Natl. Acad. Sci. U.S.A. 2003,
100, 11964–11969.
(11) Litvinchuk, S.; Tanaka, H.; Miyatake, T.; Pasini, D.; Tanaka, T.;
Bollot, G.; Mareda, J.; Matile, S. Nat. Mater. 2007, 6, 576–580.
(12) Mora, F.; Tran, D.-H.; Oudry, N.; Hopfgartner, G.; Jeannerat, D.;
Sakai, N.; Matile, S. Chem.sEur. J. 2008, 14, 1947–1953.
(13) Sakai, N.; Mareda, J.; Matile, S. Acc. Chem. Res. 2008, 41, 1354–
1365.
(1) Reymond, J.-L.; Fluxa`, V. S.; Maillard, N. Chem. Commun. 2009,
34–46.
(2) Inouye, M.; Hashimoto, K.; Isagawa, K. J. Am. Chem. Soc. 1994,
116, 5517–5518.
(3) Koh, K. N.; Araki, K.; Ikeda, A.; Otsuka, H.; Shinkai, S. J. Am.
Chem. Soc. 1996, 118, 755–758.
(4) Niikura, K.; Anslyn, E. V. J. Chem. Soc., Perkin Trans. 2 1999, 2769–
2775.
(5) Wiskur, S. L.; Ait-Haddou, H.; Lavigne, J. J.; Anslyn, E. V. Acc.
Chem. Res. 2001, 34, 963–972.
(14) Hennig, A.; Bakirci, H.; Nau, W. M. Nat. Methods 2007, 4, 629–
632.
(6) Nguyen, B. T.; Anslyn, E. V. Coord. Chem. ReV. 2006, 250, 3118–
3127.
(15) Bailey, D. M.; Hennig, A.; Uzunova, V. D.; Nau, W. M. Chem.sEur.
J. 2008, 14, 6069–6077.
(7) Zhang, T.; Anslyn, E. V. Org. Lett. 2007, 9, 1627–1629.
(8) Zhu, L.; Anslyn, E. V. J. Am. Chem. Soc. 2004, 126, 3676–3677.
(16) Praetorius, A.; Bailey, D. M.; Schwarzlose, T.; Nau, W. M. Org.
Lett. 2008, 10, 4089–4092.
9
11558 J. AM. CHEM. SOC. 2009, 131, 11558–11570
10.1021/ja904165c CCC: $40.75  2009 American Chemical Society

Substrate-Selective Supramolecular Tandem Assays
A R T I C L E S
Scheme 1. Product- versus Substrate-Selective Tandem Assays for Monitoring Enzymatic Activity
cent dye from the macrocycle and effected a concomitant change
in its fluorescent properties. This method can be described as a
“product-selective supramolecular tandem assay”, because the
enzymatic product acts as the analyte responsible for the stronger
competitive binding and associated fluorescence response
(Scheme 1, top). We have further suggested that the macrocyclic
hosts in these types of continuous enzyme assays serve as cheap
substitutes for antibodies, which have become indispensable in
enzyme assays.17-20 We also demonstrated that the lower
selectivity of macrocycles14,15 (as opposed to highly specific
antibodies) toward the binding of different, structurally related
products21,22 can be used as an advantage in the development
of enzyme assays, because it allows access to several enzymes
affecting closely related functional group interconversions.
ably strong binding and slow release kinetics,23 which would
prevent a real-time response in a substrate-specific enzyme
assay. In contrast, macrocyclic receptors bind analytes more
weakly and reversibly through supramolecular interactions,
which is essential for substrate-selective tandem assays to be
performed.
Besides the conceptual advancement, our present study is
original in that we expand the range of tandem assays to two
new enzymes (arginase and diamine oxidase versus the previ-
ously studied amino acid decarboxylases) and two additional
enzyme classes (a hydrolase (EC3) and an oxidoreductase (EC1)
versus the previously documented examples of several lyases
(EC4)). Moreover, we apply an additional reporter pair, namely
cucurbit[7]uril (CB7)/acridine orange (AO)24 and demonstrate
for the first time the potential of tandem assays for inhibitor
screening. Finally, we show that the working principle of tandem
assays is not limited to enzymatic reactions in which the charge
status in substrate versus product is altered, and where several
orders of magnitude difference in binding constants apply.
Instead, it can be extended to more subtle structural variations,
such as the size or shape of substrate and product, which are
also associated with much smaller variations in affinity to the
macrocycle (factor of 10).
We now find that macrocycles in combination with simple
fluorescent dyes not only can substitute antibodies in product-
selective enzyme assays but also allow for a different line of
enzyme assays in which the substrate binds more tightly to the
receptor and competes with the fluorescent dye. This establishes
the concept of a “substrate-selective tandem assay” (Scheme 1,
bottom) in which the enzymatic reaction occurs with the
uncomplexed substrate. The latter is in a rapid dynamic
supramolecular equilibrium with the host-substrate complex,
allowing for a continuous fluorescence signaling of the ongoing
enzymatic reaction. The possibility of using substrate-selective
macrocycles substantially broadens the applicability of tandem
assays and opens new opportunities, because comparable
enzyme assays with continuous monitoring based on substrate-
specific antibodies are nonfeasible. Biological receptors bind
analytes through antibody-antigen interactions which, while
highly specific and exceedingly strong, suffer from an unfavor-
Results
Conceptual Approach. Once an enzymatic reaction of interest
has been identified for which a supramolecular tandem assay
should be set up according to Scheme 1, a suitable reporter pair
needs to be identified. The macrocyclic component of the
reporter pair must necessarily display a sufficiently large binding
differentiation between the substrate and the product of the
enzymatic reaction. As will be seen herein, even a rather small
variation in binding constants by 1 order of magnitude can be
sufficient to conduct the assay. Furthermore, it should display
a sizable binding with either the substrate (for a substrate-
selective assay) or with the product (for a product-selective
assay); i.e., the respective binding constant should be sufficiently
(17) Tawfik, D. S.; Green, B. S.; Chap, R.; Sela, M.; Eshhar, Z. Proc.
Natl. Acad. Sci. U.S.A. 1993, 90, 373–377.
(18) MacBeath, G.; Hilvert, D. J. Am. Chem. Soc. 1994, 116, 6101–6106.
(19) Geymayer, P.; Bahr, N.; Reymond, J.-L. Chem.sEur. J. 1999, 5,
1006–1012.
(20) Venn, R. F. In Principles and Practice of Bioanalysis, 2nd ed.; Venn,
R. F., Ed.; CRC Press: Boca Raton, FL, 2008.
(23) Houk, K. N.; Leach, A. G.; Kim, S. P.; Zhang, X. Angew. Chem.
2003, 115, 5020–5046; Angew. Chem., Int. Ed. 2003, 42, 4872-
4897.
(21) Lagona, J.; Mukhopadhyay, P.; Chakrabarti, S.; Isaacs, L. Angew.
Chem. 2005, 117, 4922–4949; Angew. Chem., Int. Ed. 2005, 44,
4844-4870.
(24) Shaikh, M.; Mohanty, J.; Singh, P. K.; Nau, W. M.; Pal, H.
Photochem. Photobiol. Sci. 2008, 7, 408–414.
(22) Rekharsky, M. V.; Inoue, Y. Chem. ReV. 1998, 98, 1875–1917.
9
J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009 11559

A R T I C L E S
Nau et al.
large to effect a sizable complexation within the substrate
concentration range appropriate for the particular enzymatic
conversion. Additionally, the fluorescent dye must be selected
such that it shows a large change in fluorescence properties upon
complexation under the conditions, particularly the pH, required
for the enzymatic reaction. Also important, its concentration
must be adjusted such that the dye competes with the substrate
or product in binding to the macrocycle. For example, when
the fluorescent dye displays very strong binding, it can be used
in a more dilute form, presuming that its fluorescence is still
instrumentally detectable. In order for the reporter pair to
produce a sufficiently large fluorescence response, a sizable
fraction of dye must become complexed or uncomplexed in the
course of the enzymatic reaction. This degree of dye uptake or
displacement can be conveniently varied through the concentra-
tion of added macrocycle, and it can be predicted from direct
fluorescence titrations of the dye with the respective macrocycle.
The concentration of macrocycle should ideally lie within the
titration range where the fluorescence intensity of the dye is
most sensitive to the addition of macrocycle, i.e., not in the
plateau region of nearly quantitative dye complexation. Once a
reporter pair and the concentrations of the individual components
have been set up, the robustness of the system for a potential
enzyme assay can be tested in the absence of the actual enzyme
by carrying out competitive fluorescence titrations with the
substrate and product of the enzymatic reaction. As an additional
control experiment, to assess interactions of the fluorescent dye
with the enzyme, the fluorescence of the dye or of the reporter
pair can be monitored upon addition of enzyme, in the absence
of substrate. While the procedure just outlined may sound quite
complex and involved, it is actually quite intuitive, as will be
shown in the following.
p-sulfonatocalix[4]arene (CX4) and cucurbit[7]uril (CB7) were
known to bind with poor to moderate selectivity to different
amino acids at pH 6, but only the former showed the dif-
ferentiation in binding between arginine [K ) (2800 ( 100)
M-1] and ornithine [K ) (210 ( 10) M-1] required for a tandem
assay for arginase. The preferential binding of arginine, the
substrate, to CX4 is also the prerequisite for performing the
tandem assay in the substrate-selective mode. Moreover, a
suitable fluorescent dye (1-aminomethyl-2,3-diazabicyclo[2.2.2]
oct-2-ene, DBO, K ) 60 000 ( 16 000 M-1; see Supporting
Information) was already known for this macrocycle,28,29 thereby
setting up the reporter pair for the desired tandem assay.30
In our present study, the CX4/DBO reporter pair was first
employed to remeasure, by competitive fluorescence displace-
ment titrations, the binding constants of the amino acids at the
alkaline pH most suitable for the enzyme arginase.31-33 The
resulting values at pH 9.5 (K ) (6400 ( 250) M-1) for arginine
and K ) (550 ( 130) M-1 for ornithine) revealed essentially
the same differentiation (a factor of 12 difference) as that at
pH 6 (factor of 13)14 with a slight increase in absolute binding
constants by a factor of ca. 2.5. This can be accounted for by
the different electrolyte concentrations employed previously for
pH 6 (10 mM NH4OAc)14 and presently for pH 9.5 (no buffer),
since inorganic cations are well-known to show a competitive
binding to the calixarene, thereby lowering the observed binding
constants with analytes.34 It should be noted that the difference
in binding constants for the arginine/ornithine pair is much
smaller than that previously found for the amino acid/biogenic
amine pairs because the binding with the calixarene macrocycle
is based on a more subtle amino acid residue recognition
(preferential binding of the larger arginine)35-37 as opposed to
the previously applicable recognition of charge status (largely
favored binding of the bis-cationic diamines).14 In general, CX4
is well-known to recognize positive charges but displays a poor,
but for our purposes sufficient, selectivity toward other structural
variations.28,38,39
The observed preferential binding for the substrate arginine
should allow the setup of a substrate-selective tandem assay
according to Scheme 2. Accordingly, arginase converts a
(28) Bakirci, H.; Nau, W. M. AdV. Funct. Mater. 2006, 16, 237–242.
(29) Bakirci, H.; Koner, A. L.; Dickman, M. H.; Kortz, U.; Nau, W. M.
Angew. Chem., Int. Ed. 2006, 45, 7400–7404.
(30) The DBO chromophore has the additional advantage, owing to its
exceedingly long fluorescence lifetime, of being amenable to detection
by nanosecond time-resolved fluorescence (Nano-TRF), which can
lead to an improved background suppression and increased robustness
of enzyme assays; cf. Hennig, A.; Roth, D.; Enderle, T.; Nau, W. M.
ChemBioChem 2006, 7, 733–737. Sahoo, H.; Hennig, A.; Florea,
M.; Roth, D.; Enderle, T.; Nau, W. M. J. Am. Chem. Soc. 2007,
129, 15927–15934. Hennig, A.; Florea, M.; Roth, D.; Enderle, T.;
Nau, W. M. Anal. Biochem. 2007, 360, 255–265.
Binding Studies and Assay Working Principles. The first
substrate-selective supramolecular tandem assay was developed
for the enzyme arginase, which hydrolyzes the guanidinium
group of arginine to yield the amino acid ornithine and urea as
products. Arginase is involved in asthma,25 immune response,26
and sexual arousal,27 such that arginase assays are presently of
considerable current interest for use both in academic laboratory
settings and in an industrial high-throughput screening format.
Recall that the tandem assay principle requires a differential
binding of the substrate and the product of the enzymatic
reaction with the supramolecular receptor, in our case
the macrocyclic host. From our previous study,14 both
(31) Baggio, R.; Cox, J. D.; Harper, S. L.; Speicher, D. W.; Christianson,
D. W. Anal. Biochem. 1999, 276, 251–253.
(32) Kuchel, P. W.; Nichol, L. W.; Jeffrey, P. D. J. Biol. Chem. 1975,
250, 8222–8227.
(33) Greenberg, D. M. Methods Enzymol. 1955, 2, 368–374.
(34) Bakirci, H.; Koner, A. L.; Nau, W. M. Chem. Commun. 2005, 5411–
5413.
(35) Douteau-Gue´vel, N.; Coleman, A. W.; Morel, J.-P.; Morel-Desrosiers,
N. J. Chem. Soc., Perkin Trans. 2 1999, 629–634.
(36) Kalchenko, O. I.; Perret, F.; Morel-Desrosiers, N.; Coleman, A. W.
J. Chem. Soc., Perkin Trans. 2 2001, 258–263.
(37) Perret, F.; Morel, J.-P.; Morel-Desrosiers, N. Supramol. Chem. 2003,
15, 199–206.
(25) Ricciardolo, F. L. M.; Zaagsma, J.; Meurs, H. Expert Opin. InVest.
Drugs 2005, 14, 1221–1231.
(38) Ballester, P.; Shivanyuk, A.; Far, A. R.; Rebek, J. J. Am. Chem. Soc.
2002, 124, 14014–14016.
(26) Bronte, V.; Zanovello, P. Nat. ReV. Immunol. 2005, 5, 641–654.
(27) Christianson, D. W. Acc. Chem. Res. 2005, 38, 191–201.
(39) Bakirci, H.; Koner, A. L.; Nau, W. M. J. Org. Chem. 2005, 70, 9960–
9966.
9
11560 J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009

Substrate-Selective Supramolecular Tandem Assays
A R T I C L E S
Scheme 2. Binding Equilibria in a Substrate-Selective Switch-Off Supramolecular Tandem Assay for Arginasea
a It should be noted that the dye, substrate, and product are in a rapid dynamic competitive equilibrium for encapsulation within the CX4 macrocycle.
relatively stronger competitor (arginine) to relatively weaker
competitors (ornithine and urea), thereby allowing the dye to
compete more efficiently in forming a complex with the
macrocycle as the enzymatic reaction proceeds and depletes the
substrate. The concentration of uncomplexed substrate is rapidly
replenished through its dynamic chemical equilibrium with its
corresponding macrocycle-substrate complex.40 Overall, be-
cause the fluorescence of the dye is lower in the inclusion
complex, the fluorescence of the system should diminish as a
stronger competitor is enzymatically converted into a weaker
one (“switch-off” response, see Scheme 2 and Figure 1).
below), and the observed variation in binding proclivity was
comparable to the differential binding of the same set of guests
with the smaller CB6 (factor of 70).47,48 Note that amylamine
serves as a model for 5-aminopentanal, which is the primary
enzymatic product of the diamine oxidase reaction but which
is thermally unstable under the reaction conditions.49-51
Our previous product-selective tandem assay studies utilizing
CB7 employed the fluorescent dye Dapoxyl to monitor the
reaction.14 The fluorescence response of this dye upon CB7
encapsulation stems primarily from a complexation-induced pKa
shift;52 it is consequently strongly pH dependent, which
precludes its use at neutral pH and above. Since the optimum
pH for diamine oxidase lies at pH 7.2, an alternative dye had
to be utilized. Recently,24 the fluorescent dye AO has been
shown to undergo significant changes to its fluorescence
properties upon encapsulation by CB7. A direct fluorescence
titration of this dye with CB7 under the recommended conditions
(pH 7.2, 10 mM (NH4)2PO4 buffer, see Supporting Information)
led to an approximately 6.5-fold increase in its fluorescence
intensity and afforded a binding constant of (2.9 ( 0.1) × 105
M-1 (lit. 2.0 × 105 M-1 at pH 7).24 Application of the CB7/
AO reporter pair in a substrate-selective tandem assay for
monitoring the enzymatic oxidation of cadaverine (Scheme 3)
should consequently lead to a “switch-on” response in fluores-
cence intensity over time, as a strong competitor (cadaverine)
is transformed into a weaker one (initially 5-aminopentanal).
Enzymatic Assays. Enzymatic activity of arginase (available
as partially purified enzyme) was investigated by adapting the
conditions of Greenberg.33 When the enzyme (140 nM) was
added to an arginine (0.1-10 mM) solution containing the CX4/
DBO reporter pair (200 µM/100 µM), the enzymatic activity
was immediately signaled by a continuous decrease in fluores-
cence intensity, the absolute magnitude of which depended on
the absolute concentrations of substrate (Figure 2a). According
We next turned our attention to the enzyme diamine oxidase,
which converts its substrate cadaverine to 5-aminopentanal.
Diamine oxidase plays important roles in cancer, tumor growth,
and apoptosis;41-43 it also regulates cellular polyamine levels
and, as such, is implicated in cell growth and proliferation
processes.44-46 We have previously studied the binding of
cadaverine to the CB7 macrocycle in its capacity as a product
of the enzymatic decarboxylation of lysine.14-16 The primary
driving force in the encapsulation of alkylamines is the attractive
interaction between their cationic ammonium sites and the
electronegative oxygen atoms of the cucurbituril portal carbo-
nyls, as evidenced by the higher binding affinity of the bis-
cationic cadaverine to CB7 [K ) (4.5 ( 1.3) × 106 M-1; lit.:
K ) 1.4 × 107 M-1 in 10 mM NH4OAc buffer, pH 6] than the
monocationic 1-aminopentane [amylamine, K ) (1.1 ( 0.1) ×
105 M-1; see Figure 1b]. Both binding constants were deter-
mined by competitive displacement titrations in 10 mM
(NH4)2PO4 buffer, pH 7.2 (using the CB7/AO reporter pair, see
(40) For the investigated macrocycles, a rapid complexation kinetics has
been experimentally confirmed through NMR measurements. These
established a fast guest exchange (milliseconds or faster) between
arginine and CX4, as well as between cadaverine and CB7, ensuring
a fast response to the much slower (minutes to hours, Figure 2)
enzymatic conversion.
(41) Buffoni, F.; Ignesti, G. Mol. Genet. Metab. 2000, 71, 559–564.
(42) Pietrangeli, P.; Mondov`ı, B. NeuroToxicology 2004, 25, 317–324.
(43) Toninello, A.; Pietrangeli, P.; De Marchi, U.; Salvi, M.; Mondov`ı,
B. Biochim. Biophys. Acta, ReV. Cancer 2006, 1765, 1–13.
(44) Tabor, C. W.; Tabor, H. Annu. ReV. Biochem. 1984, 53, 749–790.
(45) Walters, D. R. Phytochemistry 2003, 64, 97–107.
(47) Rekharsky, M. V.; Ko, Y. H.; Selvapalam, N.; Kim, K.; Inoue, Y.
Supramol. Chem. 2007, 19, 39–46.
(48) Mock, W. L.; Shih, N. Y. J. Org. Chem. 1983, 48, 3618–3619.
(49) Mann, P. J. G.; Smithies, W. R. Biochem. J. 1955, 61, 101–105.
(50) Hill, J. M. Biochem. J. 1967, 104, 1048–1055.
(51) Medda, R.; Padiglia, A.; Floris, G. Phytochemistry 1995, 39, 1–9.
(52) Koner, A. L.; Nau, W. M. Supramol. Chem. 2007, 19, 55–66.
(46) Gerner, E. W.; Meyskens, F. L. Nat. ReV. Cancer 2004, 4, 781–792.
9
J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009 11561

A R T I C L E S
Nau et al.
Figure 1. (a) Competitive fluorescence titration plots (λex ) 365 nm, λem ) 450 nm) of L-arginine (b), L-ornithine (O), and urea (0) in H2O, pH 9.5, with
the CX4/DBO reporter pair (200 µM/100 µM). (b) Competitive fluorescence titration plots (λex ) 485 nm, λem ) 510 nm) of cadaverine (b) and amylamine
(O) in 10 mM (NH4)2PO4 buffer, pH 7.2, with the CB7/AO reporter pair (8 µM/0.5 µM). The solid lines correspond to the fitted curves analyzed according
to a competitive binding function; cf. Experimental Section. Plots (c) and (d) were obtained by dividing the fitted competitive binding curve of the substrate
through that of the product, thus representing the expected fluorescence differentiation (I∞/I0) between substrate and product in a tandem assay. Values of
I∞/I0 smaller and larger than 1 indicate a switch-off and switch-on fluorescence response, respectively, in the course of an enzyme assay, which is also
qualitatively illustrated by the arrows in (a) and (b).
Scheme 3. Binding Equilibria in a Substrate-Selective Switch-On Supramolecular Tandem Assay for Diamine Oxidasea
a It should be noted that the dye, substrate, and product are in a rapid dynamic competitive equilibrium for encapsulation within the CB7 macrocycle.
to Scheme 2, arginase transforms the “strong” competitor
arginine into the weaker competitor ornithine, which allows a
greater fraction of the fluorescent dye (DBO) to compete with
the enzymatic product and therefore to bind to the macrocycle
(CX4). Thus, a significant decrease in fluorescence was observed
(switch-off fluorescence response), which approached a plateau
region as the enzymatic reaction neared completion.
The enzymatic activity of diamine oxidase (available as crude
enzyme extract) was followed with the CB7/AO reporter pair
(8 µM/0.5 µM) under conditions (10 mM (NH4)2PO4 buffer,
pH 7.2,) similar to the ones used by Aarsen and Kemp.53
Monitoring the enzymatic oxidation of cadaverine led to the
(53) Aarsen, P. N.; Kemp, A. Nature 1964, 204, 1195–1195.
9
11562 J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009

Substrate-Selective Supramolecular Tandem Assays
A R T I C L E S
Figure 2. (a) Evolution of fluorescence intensity (λex ) 365 nm, λem ) 450 nm) of the CX4/DBO (200 µM/100 µM) system during enzymatic hydrolysis
of arginine by arginase at different substrate concentrations at 25 °C. The reaction was initiated by addition of arginase (140 nM) at t ) 0 min. (b) Evolution
of fluorescence intensity (λex ) 485 nm, λem ) 510 nm) of the CB7/AO (8 µM/0.5 µM) system during enzymatic oxidation of cadaverine to 5-aminopentanal
at different substrate concentrations at 37 °C. The reaction was initiated by addition of crude diamine oxidase extract (1 unit/ml) at t ) 0 min. Note that the
different initial fluorescence intensities are due to varying degrees of dye displacement at different substrate concentrations. Background fluorescence (in the
absence of dye) accounts for not more than 10% of the total fluorescence intensity.
expected switch-on response in fluorescence intensity over time,
as the strong competitor cadaverine is transformed into a weaker
competitor, initially 5-aminopentanal and its cyclization product
∆1-piperideine (2,3,4,5-tetrahydropyridine), and ultimately larger
trimers.49-51 Again, the appearance of a plateau region in the
time-resolved fluorescence traces signaled the completion of the
enzymatic reaction (notable at high substrate concentrations,
Figure 2b).
previously investigated by other assay methods31-33,53,55-58 can
now be sensitively detected and continuously monitored by
fluorescence of an added dye in the presence of a suitable
macrocycle. While the qualitatiVe signaling of enzymatic activity
is readily observable, its quantification from the recorded
fluorescence traces is less straightforward, due to the complexity
of the multiple and inter-related reaction equilibria and kinetics
involved (see Schemes 2 and 3). In particular, the fluores-
cence traces (Figure 2) do not provide a direct measure of the
absolute reaction rate. For example, the initial slopes of the
fluorescence traces are steepest in an intermediary concentration
range (around 0.5-2 mM for arginase and 20-30 µM for
diamine oxidase) and are reached far below the respective KM
values of the enzymes (5.14 mM for arginase59 and 1.28 mM
for diamine oxidase58). This is a manifestation of the tandem
assay principle, because at higher concentrations the excess
substrate is still capable of displacing the fluorescent dye from
the macrocycle, although the enzymatic reaction has already
considerably progressed. However, the development of new
enzyme assays is currently not driven by the challenge to extract
absolute enzyme kinetics (which are already known for these
two enzymes27,58) but to supply convenient tools for rapid
screening of relatiVe enzymatic activity in the presence of either
a series of potentially pharmaceutically relevant inhibitors or a
biotechnologically engineered library of enzyme mutants.60
Therefore, we have investigated in further detail the utility of
supramolecular tandem assays by studying the effects of
inhibitors on the enzymatic transformations.
Through comparison of the competitive titration plots of the
substrates (in this case arginine and cadaverine) and products
(ornithine and amylamine), it is possible to approximately
predict the change in fluorescence response during the enzymatic
transformations at various substrate concentrations (Figure 1)
and to identify the substrate concentration range in which the
largest fluorescence changes should occur. For this purpose, the
ratios of fluorescence intensities observed at comparable
substrate and product concentrations were plotted (Figure 1c
and d). As can be seen, the titration plots predict a maximum
fluorescence decrease by ca. 35% at 2 to 6 mM substrate
concentration for the arginase assay (Figure 1c), which is even
somewhat more pronounced in the actual enzyme assay
(decrease by nearly 50%, Figure 2a), where it occurs in a similar
concentration range (1-5 mM). For the diamine oxidase assay,
the titration plots predict a maximum fluorescence enhancement
of 3 near a 20 µM substrate concentration (Figure 1d), which
in this case is not reached in the actual assay (maximum factor
of 1.5 increase). Control experiments revealed that diamine
oxidase, which was used as a crude hog kidney extract and
consequently contained large amounts of unknown impurities,54
caused a sizable reduction of the fluorescence enhancement of
the CB7/AO reporter pair (see below), which is held responsible
for this quantitative variation. Nevertheless, the concentration
range in which the largest fluorescence differentiation is
expected according to the competitive fluorescence titrations
(15-30 µM, Figure 1d) agrees very well with the experimental
results of the diamine oxidase assay runs (20-30 µM, Figure
2b).
Monitoring of Inhibitory Activity. We have investigated
known inhibitors of both enzymes in the corresponding substrate-
selective tandem assays. For example, S-(2-boronoethyl)-L-
cysteine (BEC) and 2-(S)-amino-6-boronohexanoic acid (ABH)
are known inhibitors of arginase.27 We studied arginase inhibi-
tion at substrate concentrations of 0.25-0.5 mM which pre-
sented a favorable balance between a rapid and sufficiently
(55) Mondovi, B.; Rotilio, G.; Finazzi, A.; Scioscia-Santoro, A. Biochem.
J. 1964, 91, 408–415.
The results for arginase and diamine oxidase confirm the
viability of substrate-selective supramolecular tandem assays
according to Schemes 2 and 3. The activity of both enzymes
(56) Tabor, H. J. Biol. Chem. 1951, 188, 125–136.
(57) Costa, M. T.; Rotilio, G.; Agro, A. F.; Vallogini, M. P.; Mondovi,
B. Arch. Biochem. Biophys. 1971, 147, 8–13.
(58) He, Z.; Nadkarni, D. V.; Sayre, L. M.; Greenaway, F. T. Biochim.
Biophys. Acta, Protein Struct. Mol. Enzymol. 1995, 1253, 117–127.
(59) Xie, X. Y.; Wang, C. X.; Wang, Z. Y. J. Therm. Anal. Calorim.
2004, 76, 275–284.
(54) Note that purification of crude hog kidney diamine oxidase extracts
can lead to more than a 1000-fold increase in its activity; cf. refs 55
and 56.
(60) Copeland, R. A. Anal. Biochem. 2003, 320, 1–12.
9
J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009 11563

A R T I C L E S
Nau et al.
Figure 3. Determination of arginase inhibition by (a) BEC and (b) ABH at 25 °C and corresponding dose-response curves. The inhibition was determined
in the presence of 140 nM arginase in H2O (pH 9.5). 100 µM DBO and 200 µM CX4 were used as a reporter pair. Arginine concentrations were 0.5 mM
for BEC and 0.25 mM for ABH.
Figure 4. Determination of diamine oxidase inhibition by KCN (0-1320 µM) at 37 °C upon addition of 1 unit/mL diamine oxidase extract at t ) 0 min.
Monitored with the CB7/AO reporter pair (8 µM/0.5 µM) in 10 mM ammonium phosphate buffer at pH 7.2 with the cadaverine concentration held at 30 µM.
(a) Selected continuous fluorescence traces (λex ) 485 nm, λem ) 510 nm) upon addition of enzyme (raw data) for the determination of the initial rates. (b)
Dose-response curve.
sensitive fluorescence response (see Figure 2a). In fact, addition
of either inhibitor in high concentration (100 µM) completely
suppressed the fluorescence response, signaling efficient inhibi-
tion. We additionally assessed the extent of inhibition by
recording fluorescence decays at intermediary inhibitor con-
centrations. The initial decrease in fluorescence intensity versus
time was taken as a relatiVe reaction rate to allow the analysis
of the resulting dose-response curves (Figure 3a and b) with
the Hill equation.60 The IC50 values were 3.7 ( 0.7 µM for
BEC and 0.22 ( 0.04 µM for ABH, which can be readily
converted into the inhibition constants KI by considering the
was taken as a relatiVe reaction rate for the dose-response curve
over a cyanide concentration range of 0-1320 µM (Figure 4b).60
The resulting IC50 value for cyanide was 210 ( 110 µM, which
compares very well with literature reports of inhibition constants,
which are in the range 80-380 µM.61-63 Incidentally, as can
be seen from Figure 4a, the fluorescence intensities reach a
plateau at a lower level in the presence of inhibitor, which is
fully in agreement with the fact that cyanide acts as a mixed
noncompetitive inhibitor of diamine oxidase,61 thereby irrevers-
ibly deactivating the enzyme and resulting in only partial
conversion. More potent inhibitors of diamine oxidase were also
investigated and found to have a similar inhibition potential as
reported in the literature; e.g., the IC50 value of semicarbazide
was estimated to lie below 10 µM.62,63 However, due to the
crude nature of the enzyme extract, no detailed quantification
was performed.
60
1
enzyme concentration (IC50 ) KIapp + /2[E].) The resulting
values for KI (3.6 ( 0.7 µM for BEC and 0.15 ( 0.04 µM for
ABH) are in good agreement with previously reported values
of 2.2 and 0.11 µM, respectively,27 and nicely reflect the
approximately 20 times more potent inhibitory activity of ABH.
It is worthwhile in this context to consider the potential
undesirable effects which an inhibitor could have in a screening
based on a substrate-selective supramolecular tandem assay. The
most relevant aspect is that an inhibitor might itself bind
significantly to the macrocyclic host, which would inevitably
reduce its ability to inhibit the enzyme. Fortunately, a significant
Inhibition of diamine oxidase by cyanide was studied
similarly with the substrate-selective tandem assay involving
the CB7/AO reporter pair. A substrate concentration of 30 µM
was selected, which produced a strong fluorescence response
in the absence of inhibitor (see Figure 2b). In this case, the
initial increase in fluorescence intensity versus time (Figure 4a)
(61) He, Z. W.; Zou, Y.; Greenaway, F. T. Arch. Biochem. Biophys. 1995,
319, 185–195.
(62) Mondov´ı, B.; Costa, M. T.; Agro`, A. F.; Rotilio, G. Arch. Biochem.
Biophys. 1967, 119, 373–381.
(63) Mondovi, B.; Rotilio, G.; Costa, M. T. In Methods in Enzymology;
Taylor, H., Tabor, C. W., Eds.; Academic Press: New York, 1971;
Vol. 17B, pp 735-740.
9
11564 J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009

Substrate-Selective Supramolecular Tandem Assays
A R T I C L E S
binding of an inhibitor to the host would lead to immediate
dye displacement, which would be readily identified by a sudden
change in fluorescence intensities upon addition of the com-
pound. In the case of the arginase assay, the fluorescence read-
out would be artificially increased, while, for the diamine
oxidase assay, a decreased intensity would be observed. Such
changes, which are in fact opposite to the fluorescence response
expected for the enzymatic reactions themselves, were not
observed for the presently investigated inhibitors,64 such that
this complication did not need to be further considered. Despite
the potential interference of the macrocycle present in the assay
mixture, a substrate-selective supramolecular tandem assay
offers an intrinsic possibility to identify potential false negatives
by considering all fluorescence intensity outliers, which show,
upon addition of inhibitor, an immediate fluorescence change
opposite to the expected direction.67
Effect of Macrocycle on the Enzymatic Reaction. The non-
negligible binding of the enzymatic substrate with the macro-
cycle, which constitutes the working principle of any substrate-
selective tandem assay, leads to a compulsory modulation of
the enzyme kinetics because the free substrate concentration is
effectively lowered due to partial complexation.68 In fact,
macrocycles,69-71 including the presently employed CB7,72 can
even be used as (apparent) inhibitors of enzymatic reactions,
e.g., of proteases, whenever they bind the substrate and their
concentration is sufficiently high.71-75 However, the required
concentrations of macrocycles to successfully employ them in
tandem assays (µM) are much lower than the concentrations
previously employed to achieve sizable inhibition effects (mM),
such that no large effect on the enzyme kinetics was a priori
effected. To verify this conjecture, we have monitored the
enzymatic transformations by alternative methods.
same time scale as in the supramolecular tandem assay in the
presence of CX4 (by fluorescence). For diamine oxidase, an
alternative, established enzyme-coupled assay was used which
exploits the enzymatic (with peroxidase) oxidation of the dye
o-dianisidine by hydrogen peroxide, the enzymatic byproduct
(Scheme 3).53 The depletion of the dye (which shows no
detectable binding with CB7) was followed UV-spectrophoto-
metrically at 440 nm to monitor the progress of the enzymatic
reaction in the absence and presence of CB7 (see Supporting
Information). While a strong retardation of the enzymatic
reaction was observed at concentrations above 0.1 mM of CB7,
lower concentrations such as those selected for the enzymatic
assays (8 µM) displayed quite small effects, particularly on the
initial rates relevant for inhibitor studies (see above). This can
be rationalized by considering the percentage of bound substrate
in each of these cases; for instance, when more than 30 µM,
i.e., a large excess, of substrate is used with 8 µM CB7, a
maximum of 26% of the substrate is complexed. However, when
0.1 mM of CB7 is utilized, the substrate is quantitatively (>99%)
complexed which necessarily leads to a reduced conversion rate.
Thus, although the presence of the macrocycle does affect the
enzyme kinetics in a substrate-selective tandem assay, the
conditions can be selected such that this interference becomes
either insignificant or at least practically acceptable. Most
important, applications in inhibitor screening are based on the
effects of additives on relatiVe enzyme kinetics, such that small,
but constant, influences on the absolute rates are generally
tolerable.
Monitoring Multistep Enzymatic Transformations by Domino
Tandem Assays. We also tested the possibility whether the
tandem assay principle would be suitable to monitor a multistep
enzymatic reaction, namely the sequential transformation of
lysine to cadaverine by lysine decarboxylase followed by the
oxidation of cadaverine to 5-aminopentanal (Vide supra) by
diamine oxidase.
For arginase, owing to the high substrate concentrations
employed in the assay, the reaction could be independently
1
monitored by H NMR (data not shown). Integration of the
signals from the δ-CH2 protons verified that the transformation
in the absence of CX4 (by H NMR) did indeed occur on the
As a common reporter pair for such a “domino tandem assay”
we selected CB7/AO, and as common working conditions for
both enzymes we chose pH 7.2 and 37 °C. The initial
decarboxylation of lysine to cadaverine results in a dramatic
increase (ca. ×10 000) in affinity for the analyte toward the
macrocycle (CB7).14 This results in the dye (AO) being ejected
from its macrocyclic host (CB7) as is shown in Scheme 4,
accompanied by a drastic decrease of the overall fluorescence
intensity of the system (Figure 5). The plateau region (∼140
min) demonstrates the exhaustion of substrate as the conversion
to cadaverine nears completion. After this point, addition of
the second enzyme (second arrow), diamine oxidase, initiates
the conversion of cadaverine to 5-aminopentanal and its
cyclization products (Vide supra). These products, in turn, have
a lower binding affinity to CB7 than cadaverine resulting in an
increase in the overall fluorescence intensity as the equilibrium
shifts (Figure 5), allowing a greater amount of the dye (AO) to
once again become encapsulated by CB7 (Scheme 4).
Noteworthy is that the fluorescence of the system does not
fully recover to its initial intensity, which is ostensibly due to
two factors: First, a non-negligible residual binding of the
product 5-aminopentanal (Figure 1) or its cyclization products49-51
applies, e.g., amylamine (K ) 1.1 × 105 M-1, our model for
5-aminopentanal) competes with cadaverine (K ) 4.5 × 106
M-1) more effectively for CB7 encapsulation than does lysine
(870 M-1),14 thereby allowing a lower fraction of dye to re-
enter the macrocyclic cavity. Second and more important,
diamine oxidase is being added as a crude enzyme extract,
1
(64) KCN causes interferences at high millimolar concentrations due to
competitive metal ion (K+) binding to the carbonyl portals of CB7;
cf. refs 65 and 66.
(65) Marquez, C.; Hudgins, R. R.; Nau, W. M. J. Am. Chem. Soc. 2004,
126, 5806–5816.
(66) Marquez, C.; Huang, F.; Nau, W. M. IEEE Trans. Nanobiosci. 2004,
3, 39–45.
(67) To further identify false negatives, a secondary screening is routinely
performed in high-throughput applications. In the case of supramo-
lecular tandem assays, this could be done in the simplest case by
adding the reporter pair after incubation with the enzyme. Such an
additional single-point determination would conveniently eliminate
the influence of the inhibitor-macrocycle binding on the enzyme
kinetics. Alternatively, a secondary screen with a different reporter
pair could be performed.
(68) This establishes an important contrast to product-selective tandem
assays, where a complexation of the enzymatic product would
generally have no effect on the enzyme kinetics, or potentially a
beneficial effect in cases where product inhibition applies.
(69) Uekama, K.; Hirayama, F.; Irie, T. Chem. ReV. 1998, 98, 2045–2076.
(70) McGarraghy, M.; Darcy, R. J. Inclusion Phenom. Macrocyclic Chem.
2000, 37, 259–264.
(71) Koralewska, A.; Augustyniak, W.; Temeriusz, A.; Kanska, M.
J. Inclusion Phenom. Macrocyclic Chem. 2004, 49, 193–197.
(72) Hennig, A.; Ghale, G.; Nau, W. M. Chem. Commun. 2007, 1614–
1616.
(73) Irwin, W. J.; Dwivedi, A. K.; Holbrook, P. A.; Dey, M. J. Pharm.
Res. 1994, 11, 1698–1703.
(74) Matsubara, K.; Ando, Y.; Irie, T.; Uekama, K. Pharm. Res. 1997,
14, 1401–1405.
(75) Monteiro, J. B.; Chiaradia, L. D.; Brandao, T. A. S.; Dal Magro, J.;
Yunes, R. A. Int. J. Pharm. 2003, 267, 93–100.
9
J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009 11565

A R T I C L E S
Nau et al.
Scheme 4. Binding Equilibria in a “Domino” Tandem Assay, Combining a “Switch-Off” and “Switch-On” Fluorescence Response As Well As
Both Product- And Substrate-Selective Binding
requiring the addition of a 140 times larger amount by weight
than the partially purified lysine decarboxylase. This results in
the addition of a large excess of unknown impurities54 which
lead to a reduction in fluorescence intensity, most likely by
partial displacement of dye from the complex. In fact, we were
independently able to estimate the extent of this second effect
(dashed line in Figure 5) by adding the same amount of crude
enzyme extract to a solution containing only the CB7/AO
reporter pair (8 µM/0.5 µM) and lysine (30 µM) without addition
of lysine decarboxylase, thereby mimicking the conditions after
complete conversion (and assuming negligible binding of the
product). Accordingly, impurities in the crude enzyme extract
are largely responsible for the lower fluorescence intensity at
the end of the domino tandem assay. Notwithstanding, both
enzymatic reactions can be reliably followed by this uncon-
ventional approach.
Discussion
Enzymatic assays frequently require the use of fluorescently
or radioactively labeled substrates or cofactors, or they depend
on a recognition of the reaction products through subsequent
binding to antibodies in competition with added fluorescently
labeled antigens.1,19 Also popular are assays in which the
reaction products are converted into chromophoric or fluorescent
secondary products, either catalyzed by another added enzyme
(enzyme-coupled assays) or by chemical follow-up reactions
with added (functional-group selective) reagents.1 The various
assay types require, in particular those involving radioactive
labels, antibodies, and chemical follow-up reactions, multiple
incubation steps or heterogeneous workup, which preclude, with
few exceptions,19 a direct and continuous monitoring of the
reaction progress. For example, the established arginase assays
require either multiple incubation steps, heating to 100 °C, a
colorimetric detection of the enzymatic byproduct urea,32,33 or
the use of chromophoric derivatives like 1-nitro-3-guanidi-
nobenzene.31 The colorimetric enzyme-coupled assay53 (see
Results) and oxygen-consumption based assays55,56 common for
the determination of diamine oxidase concentration and activity
are inherently less sensitive than fluorimetric assays, and only
the latter allow for implementation into fluorescence microplate
readers and up-scaling to high-throughput screening formats,76,77
as well as the use of state-of-the-art detection techniques
including time-resolved fluorescence.77,78
As documented by the kinetic traces in Figure 2 and the
inhibition studies in Figures 3 and 4, the assays developed herein
for arginase and diamine oxidase allow a real-time, direct, and
sensitive monitoring of the progress and inhibition of the
Figure 5. Evolution of fluorescence intensity (λex ) 485 nm, λem ) 510
nm) of the CB7/AO (8 µM/0.5 µM) system during enzymatic decarboxy-
lation of lysine (30 µM) initiated by lysine decarboxylase (40 µg/mL, t )
0 min), followed by oxidation of the intermediary cadaverine to 5-amino-
pentanal by diamine oxidase (5.6 mg/mL, t ≈ 140 min), at pH 7.2 and 37
°C. The dashed line represents the fluorescence intensity of the system when
the crude diamine oxidase extract is added without addition of lysine
decarboxylase.
(76) Goddard, J.-P.; Reymond, J.-L. Trends Biotechnol. 2004, 22, 363–
370.
(77) Hennig, A.; Florea, M.; Roth, D.; Enderle, T.; Nau, W. M. Anal.
Biochem. 2007, 360, 255–265.
(78) Hennig, A.; Roth, D.; Enderle, T.; Nau, W. M. ChemBioChem 2006,
7, 733–737.
9
11566 J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009

Substrate-Selective Supramolecular Tandem Assays
A R T I C L E S
enzymatic reaction in homogeneous solution by fluorescence.
They entirely bypass the use of antibodies, radioactive labels,
covalently attached fluorescent probes, chromogenic or fluoro-
genic substrates or cofactors,1,31 chemical follow-up reactions,
multiple incubation steps or heating, and heterogeneous workup.
Instead, they operate by simple addition of two additives (a
macrocycle and a dye), which are either commercially available
(CX4, CB7, AO) or readily synthesized (DBO).
the enzymatic conversion.23 To become specific, antibodies
typically display binding constants in the range 107-109 M-1
19,23
and “on rates” in the range 103-106 M-1 s-1
,
which
corresponds to “off rates” in the range 10-1-10-6 s-1. The
release of a complexed substrate, which would be relevant in a
substrate-specific antibody assay, would consequently take
seconds to days (1/koff), far too slow for a continuous monitoring
of enzymatic reactions. Note also that the complexation of the
substrate would itself require an additional incubation step. In
fact, while exceptions are known, it is good practice to
equilibrate (incubate) antibodies for typically 5-20 min in
homogeneous assays.19 This limits enzyme assays involving
substrate-specific antibodies to indirect examples, in which the
function of the antibody is essentially to assess conversion
through a single-point measurement83,84 and not to replenish
the free substrate through a dynamic equilibrium.
We have referred to assays which exploit indicator displace-
ment from macrocycles according to the working principles
illustrated in Schemes 1-3 as “tandem assays”. Tandem assays
exploit a differential, reversible, and competitive intermolecular
binding of three potential guests (substrate, product, and dye)
with a synthetic receptor and therefore present a genuinely
supramolecular approach to the design of enzyme assays.
Principles of supramolecular chemistry have been previously
utilized in enzyme assays, including the vesicles with synthetic
pores pioneered by Matile and co-workers,9-13 and case studies
of tailor-made fluorescent chemosensors, which chelate the
substrate or product of an enzymatic reaction.79-82 In contrast
to the known supramolecular approaches, tandem assays allow
a continuous monitoring of the enzymatic reaction and bypass
the need for the construction of specific fluorescent chemosen-
sors, respectively. Most importantly, they can be simply devised
by screening a library of reporter pairs composed of different
macrocycles and common fluorescent dyes and testing them for
differential binding and a fluorescence response (these are the
two prerequisites for the development of any tandem assay),
under the enzymatic reaction conditions.
The development of tandem assays for arginase and diamine
oxidase presents a very good example of how powerful this
approach can be. Although our own “library” is presently still
vanishingly small with only four reporter pairs employed until
now (CX4/DBO,14,28,29 CB7/Dapoxyl,14,52 CB7/AO,24 and CB6/
3-amino-9-ethylcarbazole16), it was nevertheless sufficiently
large to find at least one suitable reporter pair. For example,
CB7 does not show the required differential binding toward
arginine and ornithine, but CX4 does, such that the CX4/DBO
reporter pair was selected for the arginase assay. Conversely,
the fluorescent dye Dapoxyl does not show a sufficient
fluorescence response at pH 7, the preferred condition for the
diamine oxidase assay, but AO does, such that the CB7/AO
reporter pair was preferred in this case.
For comparison, common macrocyclic receptors like cyclo-
dextrins and calixarenes typically show binding constants in the
range 102-106 M-1and “on rates” in the range 106-109 M-1
85-91
s-1
,
which corresponds to “off rates” in the range 1-107
s-1. The release rates of a substrate bound to a macrocycle is
consequently faster (seconds to microseconds) in relation to the
typical times of enzymatic reactions in enzyme assays (minutes
to hours).40 The reversibility of the guest-macrocycle com-
plexation, ensured by the mM to µM binding constants in
combination with a rapid exchange dynamics,23,88 are conse-
quently critical parameters of supramolecular tandem assays,
which ensure the reporter pair to respond sufficiently rapidly
and precisely to the depletion of substrate as affected by the
enzymatic conversion.
Any sequestration of the substrate by a receptor will inevitably
lower its apparent concentration and consequently result in a
lowering of the absolute enzymatic reaction rate. This peculiarity
of substrate-selective tandem assays68 is not a primary concern
in inhibition studies, where hits are based on clear-cut relative
effects in a large series of investigated compounds. Additionally,
we could show that adverse effects on the reaction rate can be
minimized by working under conditions in which only a small
fraction of substrate is complexed, while still allowing the
reporter pair to “observe” the enzymatic reaction through its
response to the chemical equilibrium changes. This can be
ensured, in particular, by working at low macrocycle concentra-
tions with an excess of substrate. For example, the diamine
oxidase assay functions quite well by using a macrocycle
concentration of 8 µM and substrate concentrations above 30
µM, conditions under which less than 26% of the substrate are
complexed. This results in comparably small and tolerable
The tandem assay approach was originally inspired by
antibody-based indicator displacement assays and introduced
for enzymatic reactions, which afford products that bind strongly
to the macrocycle (product-selective assays, Scheme 1, top).
The presently designed tandem assays are substrate-selective,
i.e., based on competitive complexation of the substrate as the
stronger competitor. This conceptual step from “product-
selective” to “substrate-selective” is nontrivial, as can be seen
from a comparison between the potential of synthetic receptors
(macrocycles) and their biological counterparts (antibodies).19
The latter would bind the substrate too tightly (in some cases
irreversibly on the pertinent time scale) and moreover release
it too slowly to result in a real-time fluorescence response to
(83) Wong, R. C.; Burd, J. F.; Carrico, R. J.; Buckler, R. T.; Thoma, J.;
Boguslaski, R. C. Clin. Chem. 1979, 25, 686–691.
(84) Karvinen, J.; Laitala, V.; Ma¨kinen, M.-L.; Mulari, O.; Tamminen,
J.; Hermonen, J.; Hurskainen, P.; Hemmila¨, I. Anal. Chem. 2004,
76, 1429–1436.
(85) Monti, S.; Flamigni, L.; Martelli, A.; Bortolus, P. J. Phys. Chem.
1988, 92, 4447–4451.
(86) Petrucci, S.; Eyring, E. M.; Konya, G. In ComprehensiVe Supramo-
lecular Chemistry; Atwood, J. L., Ed.; Pergamon: New York, 1996;
Vol. 8, pp 483-497.
(87) Franchi, P.; Lucarini, M.; Pedulli, G. F.; Sciotto, D. Angew. Chem.,
Int. Ed. 2000, 39, 263–266.
(79) Vance, D. H.; Czarnik, A. W. J. Am. Chem. Soc. 1994, 116, 9397–
9398.
(88) Zhang, X.; Gramlich, G.; Wang, X.; Nau, W. M. J. Am. Chem. Soc.
2002, 124, 254–263.
(80) Mizukami, S.; Nagano, T.; Urano, Y.; Odani, A.; Kikuchi, K. J. Am.
Chem. Soc. 2002, 124, 3920–3925.
(89) Bakirci, H.; Nau, W. M. J. Photochem. Photobiol. A: Chem. 2005,
173, 340–348.
(81) Wongkongkatep, J.; Miyahara, Y.; Ojida, A.; Hamachi, I. Angew.
Chem., Int. Ed. 2006, 45, 665–668.
(90) Bohne, C. Langmuir 2006, 22, 9100–9111.
(91) Pace, T. C. S.; Bohne, C. AdV. Phys. Org. Chem. 2008, 42, 167–
223.
(82) Chen, X.; Jou, M. J.; Yoon, J. Org. Lett. 2009, 11, 2181–2184.
9
J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009 11567

A R T I C L E S
Nau et al.
effects on the absolute reaction rate, as was demonstrated
through control experiments with an alternative assay.
It was our idea to apply the supramolecular tandem assay
principle to such a cascade of enzymatic reactions and monitor
a multistep enzymatic reaction sequence by coupling product-
with substrate-selective sensing. In essence, the first reaction
could be signaled through the binding of a product with the
macrocycle, which could then serve as the substrate for a second
reaction. In an extreme situation, an analyte could go through
a series of enzymatic transformations (or an entire metabolic
pathway) that all affect its binding affinity to the macrocycle,
and by combining the tandem assay principle with a sequential
addition of the different enzymes, it should be possible to
continually monitor these transformations in real time and in a
“one-pot” fashion, potentially by using a single reporter pair.
This “domino” tandem assay technique should be especially
powerful when a metabolite undergoes transformations that
alternate its affinity to the macrocycle, e.g., from weak to strong
and back to weak.
The enzymatic reaction sequence from arginine (strong
competitor) to ornithine (weak competitor) to putrescine (very
strong competitor) to 1-aminobutanal (weaker competitor) fulfills
this requirement, as well as the reaction sequence from lysine
(weak competitor) to cadaverine (very strong competitor) to
1-aminopentanal (weaker competitor). Unfortunately, we lacked
access to the enzyme ornithine decarboxylase and could
therefore not examine the associated three-step enzymatic
process, but the two enzymes involved in the reaction sequence
involving cadaverine were available: lysine decarboxylase (in
partially purified form) and diamine oxidase (as crude extract).
The working principle of the corresponding domino tandem
assay is illustrated in Scheme 4. As can be seen from Figure 5,
the two assays can indeed be sequentially combined to afford
first a fluorescence decrease due to decarboxylation and then,
upon addition of diamine oxidase, a fluorescence increase due
to oxidative deamination (on-off-on response). The fact that
the fluorescence does not fully recover after the enzymatic
conversion by diamine oxidase is attributed to the use of this
enzyme as a crude enzyme extract, which results in partial dye
displacement due to large amounts of biological “impurities”
(dashed line in Figure 5). But regardless of this interference
caused by the lack of selectivity, tandem assays are sufficiently
robust to function also in more complex biological mixtures,
as was previously demonstrated through comparative experi-
ments with purified enzymes, crude extracts, and whole cells
expressing a particular enzyme. This may be quite surprising,
but the concentration of biological impurities contained in a
certain enzyme preparation is expected to remain constant as
the enzymatic reaction proceeds. The interference is conse-
quently “static” in nature and does not influence the time-
resolved change in fluorescence intensity, which presents the
optical read-out of the assay and which is diagnostic and specific
for the enzymatic reaction.
The substrate/product differentiation in the previously reported
tandem assays for decarboxylase assays14-16 was due to an
increase in the net positive charge of the competitor during
enzymatic transformation, which resulted in a 100-10000-fold
difference in binding constants of substrate and product with
cation-receptor macrocycles (CX4, CB6, and CB7). The working
principle for the diamine oxidase tandem assay is similar
(charge-selective binding), except that in this case the substrate
binds only ca. 40 times more strongly than the product. As
demonstrated for arginase, however, the application of supramo-
lecular tandem assays is not restricted to enzymatic reactions,
in which the charge status of the substrate is changed. Note
that arginine and ornithine are expected to have a very similar
charge status between pH 6 and 9.5. Enzymatic reactions in
which the size (or shape) of the competitor changes can be
similarly explored, owing to the fact that binding to any
macrocycle is also significantly size selective (hole-size selectiv-
ity) as well as functional-group selective.28,38,39 Although in
this case the differentiation with respect to the binding constants
of substrate and product can be frequently quite small (factor
of only 12 for the arginine/ornithine couple), the fluorescence
response in the arginase tandem assay is sufficiently robust to
reliably monitor the effects of added inhibitors and varying
substrate concentration.
The results for arginase thus demonstrate that the assay
principle can also be expanded to enzymatic reactions, which
affect more subtle structural variations than a change in charge
status, and even a factor of 10 difference in binding constants
between substrate and product can be sufficient to ensure a
sizable fluorescence response through the tandem assay working
principle. In other words, even poorly selective macrocycles28,38,39
can be successfully employed, which presents another illustrative
example of the shift toward differential as opposed to highly
selective receptors in the research area of molecular recogni-
tion.92,93 As shown for the arginase assay, even a high affinity
of the macrocycle is no principal requirement. The binding
constant of CX4 with arginine, for example, is only on the order
of 103 M-1, that is, a weak or at best moderate binding. The
binding constant does, however, determine the concentration
range in which substrate conversion can be reliably monitored,
and the latter also depends on the investigated enzyme.
In the course of our investigations on the enzymatic reactions
involving amino acids, we became aware of the interrelationship
of the various enzymes as well as the intermediary polyamines
in fundamental biochemical reaction pathways during cell
growth and differentiation.44-46 Elevated levels of arginase and
diamine oxidase activity, for example, are found in cancerous
tissues. Enzymatic conversion of arginine by arginase yields
ornithine which in turn undergoes decarboxylation to putrescine,
which is finally oxidized by diamine oxidase.58,94 Cadaverine
adapts a similarly central position in a metabolic pathway
involving decarboxylation of lysine as precursor and deamina-
tion of 1-aminopentanal as product.95,96
The domino tandem assay shows that it is possible to monitor,
in real time, quite complex, but biologically highly relevant,
metabolic pathways by a very simple method. Additionally and
very interesting to note, similar enzymatic reaction cascades
with alternating on-off-on optical signaling have been pro-
posed to be promising for biocatalyst-stimulated logic gate
operations in the context of biocomputing.97,98 In any case, our
results present a proof-of-principle; they demonstrate nicely the
(92) Lavigne, J. J.; Anslyn, E. V. Angew. Chem., Int. Ed. 2001, 40, 3118–
3130.
(93) Das, G.; Matile, S. Chem.sEur. J. 2006, 12, 2936–2944.
(94) Nikolic, J.; Stojanovic, I.; Pavlovic, R.; Sokolovic, D.; Bjelakovic,
G.; Beninati, S. Amino Acids 2007, 32, 127–131.
(97) Niazov, T.; Baron, R.; Katz, E.; Lioubashevski, O.; Willner, I. Proc.
Natl. Acad. Sci. U.S.A. 2006, 103, 17160–17163.
(95) Watson, A. B.; Brown, A. M.; Colquhoun, I. J.; Walton, N. J.; Robins,
D. J. J. Chem. Soc., Perkin Trans. 1 1990, 2607–2610.
(96) Shoji, T.; Hashimoto, T. Plant Cell Physiol. 2008, 49, 1209–1216.
(98) Freeman, R.; Sharon, E.; Tel-Vered, R.; Willner, I. J. Am. Chem.
Soc. 2009, 131, 5028–5029.
9
11568 J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009

Substrate-Selective Supramolecular Tandem Assays
A R T I C L E S
Fluka) as well as ornithine, dianisidine, urea, and potassium
cyanide (all Sigma-Aldrich) were obtained in the highest purity
available and used as received. Cucurbit[7]uril was synthesized
in >95% purity, following established synthetic protocols.66,115,116
The arginase inhibitors ABH and BEC were received as
ammonium salts from Alexis Biochemicals (Lausen, Switzer-
land). Diamine oxidase (crude extract from hog kidney, 0.18
units/mg solid), peroxidase (type II from horseradish, 181
purpurogallin units/mg), and lysine decarboxylase (partially
purified, 1.6 U/mg) were from Sigma-Aldrich. Arginase (partially
purified, from cow liver, 100-200 units/mg solid) was from
Fluka.
Absorption measurements were performed with a Varian Cary
4000 spectrophotometer. For bovine liver arginase an extinction
coefficient of 1.1 × 105 M-1 cm-1 at 278 nm was used.117 For
fluorescence measurements, a Varian Eclipse fluorimeter (argi-
nase assays: λex ) 365 nm, λem ) 450 nm, diamine oxidase
assays: λex ) 485 nm, λem ) 510 nm) was used. Continuous
assays for arginase (16 µg/mL, 140 nM) were performed
unbuffered at pH 9.5 in the presence of 100 µM DBO and 200
µM CX4 in a variable-temperature cell holder at 25.0 ( 0.1 °C.
Continuous assays for diamine oxidase (1 unit/ml) were measured
in 10 mM (NH4)2PO4 buffer, pH 7.2, with a mixture of 0.5 µM
acridine orange, 8 µM CB7, and 30 µM of cadaverine in a
variable-temperature cell holder at 37.0 ( 0.1 °C. Dose-response
curves of the inhibitors BEC and ABH were obtained in a similar
manner using 0.5 mM and 0.25 mM of arginine. Inhibition
kinetic traces for KCN were obtained with 30 µM cadaverine
for a series of KCN concentrations (0-1320 µM).
potential of the tandem assay principle and reveal the inherent
advantage of employing differentially selective as opposed to
highly specific receptors.
Conclusions
We have expanded the range of supramolecular tandem
assays from product- to substrate-selective variants. We
demonstrated that even small differences in binding constants
of substrate and product are sufficient not only to monitor
the activity of two important enzymes in real time in a
homogeneous solution but also to study inhibition effects.
Tandem assays are exceptionally adaptable and flexible,
which is attractive in view of the ever-expanding range of
substrates and enzymes. Instead of designing a specific
chemosensor or raising a specific antibody, generally for the
product of an enzymatic reaction, tandem assays can be
simply devised by screening a library of reporter pairs. Even
by combining all commercially available macrocycles and
fluorescent dyes, or focusing on the large diversity of host-
dye complexes already investigated,2-4,14,16,24,28,29,52,99-113
a very large library could readily be built up. This approach
has enormous potential for enzyme assay development, and
the effort, time, and resources related to testing such a library
should compete well with alternative strategies in assay
development. Owing to the moderate but known selectivity-
of different macrocycles, the search can also be rationally
limited (e.g., to cation-receptor macrocycles for reactions
involving positively charged substrates and products) and a
once identified reporter pair can be with high probability
transferred to enzymes affecting similar functional group
interconversions; e.g., the CB7/Dapoxyl reporter pair was
previously employed for not less than six amino acid
decarboxylases.14,15
For analytical analysis of the titrations, we define [D]0, [C]0,
and [H]0 as the total concentrations of dye, competitor (substrate
or product), and host (macrocycle). [D], [C], and [H] are the
concentrations of uncomplexed dye, uncomplexed competitor,
and uncomplexed host. [H•D] and [H•C] are the concentrations
of the host-dye and host-competitor complex, and KC and KD
are the association constants of the competitor and dye with the
host.
The fluorescence intensity (I) in the course of the titration can
be expressed as a linear combination of the fluorescence intensity
of the uncomplexed dye (ID) and that of the host-dye complex
(IH•D), weighted by their molar fractions according to eq 1. IH•D
was extrapolated from host-guest titrations (in the absence of
competitor) fitted according to a 1:1 binding model.28,52,118
Experimental Section
DBO was synthesized according to a literature procedure114
and purified by precipitation as the sulfate salt from diethyl ether.
Acridine orange, cadaverine, putrescine, ornithine, and CX4 (all
(99) Miskolczy, Z.; Biczok, L.; Megyesi, M.; Jablonkai, I. J. Phys. Chem.
B 2009, 113, 1645–1651.
[D]
[D]0
[H•D]
[D]0
I )
ID +
IH•D
(1)
(100) Bhasikuttan, A. C.; Mohanty, J.; Nau, W. M.; Pal, H. Angew. Chem.,
Int. Ed. 2007, 46, 4120–4122.
(101) Liu, Y.; Han, B. H.; Chen, Y. T. J. Phys. Chem. B 2002, 106, 4678–
4687.
Upon appropriate substitution one obtains eq 2, with the
concentration of uncomplexed host as variable; the latter is defined
by a cubic equation (eq 3).119
(102) Wagner, B. D.; Stojanovic, N.; Day, A. I.; Blanch, R. J. J. Phys.
Chem. B 2003, 107, 10741–10746.
(103) Nau, W. M.; Mohanty, J. Int. J. Photoenergy 2005, 7, 133–141.
(104) Baglole, K. N.; Boland, P. G.; Wagner, B. D. J. Photochem.
Photobiol. A: Chem. 2005, 173, 230–237.
KD[H]
I ) ID + (IH•D - ID)
(2)
(105) Arunkumar, E.; Forbes, C. C.; Smith, B. D. Eur. J. Org. Chem. 2005,
2005, 4051–4059.
1 + KD[H]
(106) Wang, R.; Yuan, L.; Macartney, D. H. Chem. Commun. 2005, 5867–
5869.
0 ) a[H]3 + b[H]2 + c[H] - d, where
a ) KCKD, b ) KC + KD + KCKD [D] + [C] - [H] ,
(107) Mohanty, J.; Bhasikuttan, A. C.; Nau, W. M.; Pal, H. J. Phys. Chem.
B 2006, 110, 5132–5138.
(
)
0
0
0
c ) K [C] - [H] + K [D] - [H] + 1, and d ) -[H]
(
C
)
0
(
)
0
0
D
0
0
(108) Liu, Y.; Li, C. J.; Guo, D. S.; Pan, Z. H.; Li, Z. Supramol. Chem.
2007, 19, 517–523.
(3)
(109) Wang, R.; Yuan, L.; Ihmels, H.; Macartney, D. H. Chem.sEur. J.
2007, 13, 6468–6473.
The fitting was implemented in OriginPro 7.5 (OriginLab
Corporation, Northampton, MA), by using a subroutine to solve
(110) Megyesi, M.; Biczok, L.; Jablonkai, I. J. Phys. Chem. C 2008, 112,
3410–3416.
(111) Halterman, R. L.; Moore, J. L.; Mannel, L. M. J. Org. Chem. 2008,
73, 3266–3269.
(115) Kim, J.; Jung, I.-S.; Kim, S.-Y.; Lee, E.; Kang, J.-K.; Sakamoto, S.;
Yamaguchi, K.; Kim, K. J. Am. Chem. Soc. 2000, 122, 540–541.
(116) Day, A.; Arnold, A. P.; Blanch, R. J.; Snushall, B. J. Org. Chem.
2001, 66, 8094–8100.
(117) Harell, D.; Sokolovsky, M. Eur. J. Biochem. 1972, 25, 102–108.
(118) Nau, W. M.; Zhang, X. J. Am. Chem. Soc. 1999, 121, 8022–8032.
(119) Wang, Z.-X. FEBS Lett. 1995, 360, 111–114.
(112) Montes-Navajas, P.; Corma, A.; Garcia, H. ChemPhysChem 2008,
9, 713–720.
(113) Neelakandan, P. P.; Hariharan, M.; Ramaiah, D. J. Am. Chem. Soc.
2006, 128, 11334–11335.
(114) Hudgins, R. R.; Huang, F.; Gramlich, G.; Nau, W. M. J. Am. Chem.
Soc. 2002, 124, 556–564.
9
J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009 11569

A R T I C L E S
Nau et al.
the cubic eq 3 with the Newton-Raphson method. The module is
Supporting Information Available: Host-dye titrations and
spectrophotometric assays of diamine oxidase to study the effect
of added macrocycle. This material is available free of charge
via the Internet at http://pubs.acs.org.
available from the authors upon request.
Acknowledgment. This work was supported by the Deutsche
Forschungsgemeinschaft (DFG Grant NA-686/5-1 “Supramolecular
Tandem Assays for Monitoring Enzymatic Activity”) and by the
Fonds der Chemischen Industrie.
JA904165C
9
11570 J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009