Determination of the activity of heterofunctionalized catalysts from mixtures

Article abstract of DOI:10.1039/b605730f

A conceptually new approach for the rapid determination of the activity of heterofunctionalized catalysts is described. A small library of catalysts is synthesized via a one-pot synthesis and screened for activity without separating the library members. The screening of libraries with varying catalyst distributions and subsequent deconvolution allows the determination of the activity of each catalyst separately. This approach is applied to a four-component library of biomimetic catalysts active in the cleavage of HPNP, a model compound for RNA. The rate constants for the four catalysts obtained via the deconvolution procedure are in very good agreement with the values obtained via the classical one catalyst-one screening approach. Simulations provide a hint of the scope and limitations of the approach. the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006.

Full text of DOI:10.1039/b605730f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Determination of the activity of heterofunctionalized catalysts from
mixtures
Giovanni Zaupa, Marco Martin, Leonard J. Prins* and Paolo Scrimin*
Received (in Montpellier, France) 18th April 2006, Accepted 21st June 2006
First published as an Advance Article on the web 22nd August 2006
DOI: 10.1039/b605730f
A conceptually new approach for the rapid determination of the activity of heterofunctionalized
catalysts is described. A small library of catalysts is synthesized via a one-pot synthesis and
screened for activity without separating the library members. The screening of libraries with
varying catalyst distributions and subsequent deconvolution allows the determination of the
activity of each catalyst separately. This approach is applied to a four-component library of
biomimetic catalysts active in the cleavage of HPNP, a model compound for RNA. The rate
constants for the four catalysts obtained via the deconvolution procedure are in very good
agreement with the values obtained via the classical one catalyst–one screening approach.
Simulations provide a hint of the scope and limitations of the approach.
Introduction
determined with a minimum synthetic effort. To give a proof
of principle, we have validated a minimal catalyst mixture
The design of artificial structures with multiple different
functionalities is of strong current interest in the areas of
1
molecular recognition and catalysis. The presence of multiple
(
library) composed of four members. The catalytic activity of
each member indirectly determined via the library method
described above was compared to the values obtained via the
classical one catalyst–one screening approach. This approach
allows the determination of the activity of catalysts from
mixtures, an area that is still largely unexplored, without using
different recognition/catalytic units in a heterofunctionalized
molecular receptor allows a better tuning of its interaction
with a substrate with respect to a homofunctionalized system.
It is not surprising that in enzymes, which are the catalysts par
excellence, heterofunctionalization in the recognition and
6,7
any sophisticated apparatus.
2
catalytic sites is a rule. The current strategy to address
Results and discussion
heterofunctionalized structures is based on the use of a
molecular scaffold with the connector units masked by ortho-
Our long-standing interest in the design of artificial metallo-
nucleases for the cleavage of the phosphodiester bonds of
DNA and RNA prompted us to test our concept on catalysts
able to cleave an RNA model substrate (HPNP, 2-hydroxy-
3
gonally compatible protective groups. The desired function-
alities are subsequently introduced in a stepwise manner via a
series of deprotection–coupling steps. The disadvantage of this
approach is the prerequisite of a multistep synthesis to first
synthesize the protected scaffold molecule and next to intro-
duce the different functional groups.
8
propyl-4-nitrophenylphosphate). It is known that natural
phosphodiesterases typically require for activity the coopera-
2+
tive action of two or more metal ions (often Zn ) possibly in
Here we propose a conceptually different approach to
determine the activity of heterofunctionalized catalysts. Our
approach is based on the one-step synthesis of small mixtures
of catalysts, which are directly screened for catalytic activity
without prior separation of the individual components. Con-
sequently, the observed catalytic activity is a sum of the
activity of each catalyst present in the mixture multiplied by
conjunction with other functionalities acting as a general base
9
or general acid.
Therefore, simple catalysts providing a collection of such
functionalities seemed very suitable to verify the proposed
synthesis and screening approach. The catalysts are composed
of a platform molecule 1 to which three functionalized arms
are connected (Scheme 1). Scaffold 1 is a hexasubstituted
benzene, where the adjacent substituents alternate up and
down with respect to the benzene plane. This preorganiza-
tional effect of 1 has been widely used to create receptors and
4
its concentration. Screening of mixtures with varying catalyst
distributions, simply obtained by mixing the reagents in
different ratios, and subsequent deconvolution allows the
determination of the catalytic activity of each catalyst indivi-
10
catalysts, most notably by Anslyn and co-workers. As func-
5
dually.
tional arms we have used N,N-dimethylethylenediamine, A
and N,N-bis(2-pyridylmethyl)ethylenediamine, B. The basic
dimethylamino group in arm A may serve in a general acid/
base catalyzed reaction, whereas arm B serves as the metal
In this report we show that with this protocol the catalytic
activity of heterofunctionalized catalysts can be very rapidly
2
+ 11
.
binding site for Zn
University of Padova, Department of Chemical Sciences and ITM-
CNR, Padova Section, via Marzolo, 1, 35131 Padova, Italy. E-mail:
leonard.prins@unipd.it. E-mail: paolo.scrimin@unipd.it; Fax: +39-
A one-pot reaction of scaffold 1 with a mixture of A and B
gives a mixture of 4 potential catalysts AAA, AAB, ABB, and
BBB with a distribution depending on the initial ratio of A and
049-8275239; Tel: +39-049-8275276; Tel: +39-049-8275256
This journal is ꢀc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1493–1497 | 1493

View Article Online
2
+
reactions were performed in the presence of 120 mM Zn in
order to ensure saturation of all metal binding sites deriving
from arm B. Under these conditions the catalytic contribution
2
+
14
of non-ligated Zn
has also to be considered. For each
mixture, the initial rate was determined via a linear regression
over the first 10 000 s.
Independently, the catalytic activities of the pure catalysts
2
+
AAA, AAB, ABB, BBB and Zn
were measured under
identical conditions (Table 1). The observed rates for the
catalyst mixtures are the sum of the rate contribution of each
2+
catalyst present in the mixture (including free Zn ) following:
n
init = kAAA[AAA] + kAAB[AAB] + kABB[ABB] + kBBB
À1
[BBB] + kZn[Zn] in which vinit (M s ) is the experimentally
À1
observed rate and k (s ) the individual pseudo-first order rate
constants for the different catalysts. Since the catalyst con-
centrations are known for each mixture, the individual rate
constants (see Table 1) can be obtained via a simple deconvo-
lution procedure. The procedure amounts to the determina-
tion of the coefficients (the rate constants, k) of n-component
linear equations (with n being the number of catalysts present
in the library). The minimum number of libraries of different
composition is, hence, n. To minimize the error in our case we
have analyzed libraries with nine different compositions in-
2
+
stead of five (having four catalysts plus Zn ). The rate
Scheme 1 Synthetic procedure for the one-pot preparation of
libraries of different composition.
constants were determined in an iterative manner using the
1
software Scientist.
5
Comparison of the calculated values with the values ob-
tained from direct measurement of pure catalysts AAA, AAB,
ABB and BBB, shows, first of all, an excellent correlation
which unequivocally validates our proposed screening meth-
od. Secondly, this comes at the cost of a relatively large error
margin which is inversely related to the catalytic activity.
Thus, the largest error is observed for compound AAA caused
B added to scaffold 1. RP-HPLC analysis of the mixtures
showed that exclusively the four catalysts are present in the
mixtures, the peaks being assigned to each of them via ESI-MS
spectrometry (representative HPLC spectra of some of the
mixtures are given in Fig. 1). Within this series of mixtures, the
distribution of the catalysts ranged from 75 : 24 : 1 : 0 to 0 : 11 :
1
2
6
1 : 28 for catalysts AAA, AAB, ABB and BBB respectively.
by a very low activity not exceeding 10 times the uncatalyzed
À7 À1 8b
In order to validate the proposed screening method, pure
samples of each of the four catalysts were isolated via pre-
parative RP-HPLC.
reaction (kuncat = 3.9 Â 10
s
)
and therefore hardly
affecting the reactivity profile. To address this point more
generally, simulations of a library system were performed with
varying catalytic activities of the library members. Artificial
reactivity profiles were calculated for a simple reaction (first
order both in substrate and catalyst) using the following rate
equation, assuming identical catalyst mixture compositions as
Next, each mixture was screened for catalytic activity in the
cleavage of HPNP. Reactions were performed in a buffered
aqueous solution at pH 7.2 and 40 1C with an overall catalyst
1
3
concentration of 20 mM and a 10-fold excess of substrate. All
obtained by us experimentally: d[P]/dt = k
AAB[AAB] + fABB[ABB] + fBBB[BBB]) with [P] and [R] the
product and reagent concentrations, respectively. Factors
AAA–fBBB were used to systematically modulate the catalytic
1
[R](fAAA[AAA] +
f
f
activities of the four catalysts. Typically, a reactivity profile
was generated for the 9 catalyst mixtures and, next, the
obtained rates were taken as input values for the deconvolu-
tion procedure in order to calculate fAAA–fBBB. The factors
thus obtained were compared to the input values. The results
of three different simulations are given in Table 2.
Simulation I serves as a reference, with only catalyst AAA
having catalytic activity. In this case, the reactivity profile is
directly proportional to the relative amount of AAA present in
the libraries. Deconvolution of the obtained reactivity profile
gives accurately the right value for fAAA with negligible values
for the other catalysts. In simulation II, a situation is simu-
lated where two good catalysts (AAA and AAB) are present
(fAAA = fAAB = 1000). Also in this case, the fitting of the
Fig. 1 HPLC elution profiles of four representative mixtures.
1
494 | New J. Chem., 2006, 30, 1493–1497
This journal is ꢀc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006

View Article Online
Table 1 Observed rate constants (k
and that of the mixtures
c
) for the four catalysts in the cleavage of HPNP obtained both by measuring the activity of the pure catalysts
5
10 k
À1b
5
10 k
À1c
5
10 k
À1d
c
/s
Catalyst
c
/s
c
/s
AAA
AAB
ABB
BBB
0.4 (Æ0.1)
0.9 (Æ0.2)
2.2 (Æ0.4)
5.1 (Æ0.2)
0.5 (Æ0.1)
0.3 (Æ0.2)
0.9 (Æ0.4)
2.1 (Æ0.4)
5.1 (Æ0.7)
0.5 (Æ0.2)
—
0.9 (Æ 0.3)
2.1 (Æ 0.2)
—
—
b
2+
Zn
a
Reaction conditions: [cat] = 20 mM, [HPNP] = 200 mM, [Zn(NO
3
)
2
] = 120 mM, [HEPES] = 10 mM, pH = 7.2, T = 40 1C. From
c
d
measurements of the pure components. From measurements of the catalyst mixtures. From measurements of the catalyst mixtures, fixing the
homomeric species.
reactivity profile accurately provides the input values with
minimal error. Finally, a situation is simulated (III) in which
three species are catalytically active, but to a different extent
are given in the third column of Table 1. Again, an excellent
correlation was observed, but, importantly, this time also with
error margins that are very acceptable for this kind of kinetic
experiments, even for catalyst AAB.
(
fAAA = 1000, fAAB = 100, fABB = 10) which probably
resembles most closely a real experimental situation. In this
case the simulation shows that the activities of AAA and AAB
can be retrieved accurately with relatively small errors,
whereas the contribution of the poorly performing catalyst
ABB is not detected. This is obviously caused by the fact that
the contribution of ABB is not significantly affecting the
reactivity profile. These simple simulations seem to indicate
that the ability to retrieve the rate constant for a certain
catalyst is determined by the extent to which the catalyst
contributes to the overall activity. In other words, the rate
constants of catalysts that hardly outcompete the background
reaction, in our experimental study for instance catalyst AAA,
or that have a negligible activity with respect to another
dominating catalyst present in the mixture, are difficult to
determine with acceptable precision.
Encouraged by this approach, we decided to investigate
whether the analysis could be brought to a more sophisticated
level. Repeating the same kinetics and deconvolution proce-
dure at different substrate concentrations gives for each cata-
lyst the observed rate constant kobs as a function of the
substrate concentration S. Since kobs and S are related via
1/kobs = (K
parameters kcat and K
M
/kcat)/S + 1/kcat, in this way the Michaelis–Menten
M
can be addressed for each catalyst.
The obtained values for the heteromeric catalysts AAB and
ABB are given in Table 3, together with the values obtained
directly from measurements of the pure catalysts (BBB and
2
+
Zn ). It should be noted that for compound AAA no
saturation curve was obtained. The contribution of AAA
therefore was ignored, with no apparent effect on the outcome
of the deconvolution procedure.
Not entirely satisfied with the large error margins obtained
in the experimental study, especially for catalysts AAA and
AAB, we decided to apply a more practical approach. In the
catalyst library the only heteromeric catalysts are AAB and
ABB whose synthesis would require the use of orthogonal
Importantly, the values obtained for the heteromeric cata-
lysts AAB and ABB via the deconvolution approach are in
good agreement with the values obtained from the pure
catalysts. The values are all within a 10% range, except for
the kcat value of ABB (20%). Considering the large intrinsic
2
+
8,9
protective groups. The other catalysts (AAA, BBB, and Zn
)
error which is common for this kind of kinetic measurement
can be obtained in pure form, and consequently, their activity
can be studied separately. Therefore, the deconvolution pro-
cedure was repeated leaving only the rate constants for
catalysts AAB and ABB as unknowns. To have less variables,
we used only four catalyst mixtures instead of the previous
nine. The obtained rate constants for the heteromeric catalysts
(which is reflected by the relatively large error margins), and
the fact that only 4 catalyst libraries were used for each
Table 3 Michaelis–Menten parameters for the heteromeric catalysts
AAB and ABB obtained via measurement of the pure catalyst and via a
deconvolution of the mixtures
4
À1b
3
10 K
b
/M
4
À1c
3
10 K
c
Cat.
10 kcat/s
M
10 kcat/s
M
/M
Table 2 Simulations of reactivity profiles
2+
Zn
0.3 (Æ0.1)
2.4 (Æ0.4)
6.4 (Æ0.9)
10.7 (Æ0.2)
14.8 (Æ3.0)
6.7 (Æ1.0)
6.0 (Æ0.9)
5.6 (Æ0.7)
—
—
f
AAA
f
AAB
f
ABB
f
BBB
AAB
BBB
BBB
a
2.5 (Æ0.4)
5.8 (Æ0.8)
—
6.6 (Æ0.5)
6.1 (Æ0.6)
—
a
b
Simul. In
Out
In
Out
In Out
In Out
I
1000 1000 (Æ9)
0
2 (Æ2) 0 14 (Æ4) 0 À3 (Æ9)
Reaction conditions: [cat] = 20 mM, [HPNP] from 0.1 to 2 mM,
II
III
a
1000 994 (Æ1) 1000 1000 (Æ3)
0
6 (Æ4) 0 À3 (Æ2)
[
Zn(NO ] = 120 mM, [HEPES] = 10 mM, pH = 7.2, T = 40 1C.
3 2
)
1000 1000 (Æ2) 100 110 (Æ5) 10 3 (Æ4) 0
In denotes the input value. Out denotes the value obtained.
3 (Æ4)
b
c
From measurements of the pure components. From measurements
of the catalyst mixtures.
b
This journal is ꢀc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1493–1497 | 1495

View Article Online
substrate concentration, these results are very satisfactory. It is
worth emphasizing that this approach allows the Michaelis–
Menten parameters for AAB and ABB to be obtained without
ever having the pure catalysts in hand.
Experimental data of the isolated four catalysts
Catalysts AAA, AAB, ABB and BBB were isolated using
preparative RP-HPLC using a C18 column (5 mm). Purity
was confirmed by analytical RP-HPLC.
2
,4,6-Triethylbenzene-1,3,5-tris(N,N-dimethyl-ethylenediamine)-
1
carboxyamide, AAA. H NMR (250 MHz, CDCl
3
) d(ppm):
Conclusions
1
2
.23 (t, J = 7.5 Hz, 9H, CH CH ); 2.23 (s, 18H, N(CH ) );
2
3
3 2
In conclusion, we have reported a procedure to rapidly
synthesize and screen heterofunctionalized catalysts. Catalyst
preparation occurs in a one-step synthesis eliminating the use
of any protective group. Screening for catalytic activity is
directly performed on the mixtures without the need for
separation. The rate constants for each individual catalyst
are simply obtained by deconvoluting the overall rate con-
stants of a series of mixtures with different catalyst distribu-
tions. The proof-of-principle study described here has
validated the approach, but has also indicated some limita-
tions in determining the activity of poorly performing cata-
lysts. On the contrary, however, and of much more importance
considering catalyst discovery, this method very accurately
identifies the best catalyst for a given reaction. It is also worth
pointing out that with this approach we have eliminated the
frequently observed problem of false positive hits in combi-
natorial screening in cases where the high activity observed
simply results from the combined activities of numerous
weakly performing entities. In our opinion, these two advan-
tages create a very high potential for this screening method.
Currently, we are applying this protocol for the screening of
much larger libraries.
.50 (t, J = 5.8 Hz, 6H, CH NH ); 2.66 (q, J = 7.5 Hz, 6H,
2
2
CH CH ); 3.53 (m, 6H, CH NHCO); 6.39 (m br, 3H,
2
3
2
NH amide). ESI-MS (CH
+
3
OH): m/z found 543.4, calc. 543.5
(
M + K ). HPLC (Lichrosphere RP-C18: H
2
O–THF–
TFA = 87.5 : 12 : 0.5): 4.1 min (100%).
2,4,6-Triethylbenzene-1,3-bis(N,N-dimethyl-ethylenediamine)-
carboxyamide-5-(N,N-bis(2-pyridylmethyl)-ethylenediamine)-car-
1
boxyamide, AAB. H NMR (300 MHz, CD
t, J = 7.5 Hz, 6H, CH CH ); 1.20 (t, J = 7.5 Hz, 3H,
CH CH ); 2.19 (s, 12H, N(CH ); 2.44 (t, J = 6.3 Hz, 4H,
CH –N of arm A); 2.53 (q, J = 7.5 Hz, 4H, CH CH ); 2.65
3
CN) d(ppm): 1.09
(
2
3
2
3
3 2
)
–
2
2
3
(
q, J = 7.5 Hz, 2H, CH CH ); 2.81 (t, J = 5.9 Hz, 2H,
2 3
–
CH –N of arm B); 3.43 (m, 6H, CH NHCO); 3.85 (s, 4H,
2
2
pyr–CH –N); 6.68 (br, 2H, NH amide A); 7.14 (m, 2H, 5-H
2
3
pyr); 7.34 (d, J = 7.8 Hz, 2H, 3-H pyr); 7.62 (m, 2H, 4-H pyr);
3
.10 (br, 1H, NH amide B); 8.17 (d, J = 4 Hz, 2H, 6-H pyr).
8
ESI-MS (CH OH): m/z found 681.3 (100%), 697.4 (30%); calc.
3
+ +
81.4 (M + Na ), 697.5 (M + K ). HPLC (Lichrosphere
6
2
RP-C18: H O–THF–TFA = 87.5 : 12 : 0.5): 4.9 min (100%).
2
,4,6-Triethylbenzene-1-(N,N-dimethyl-ethylenediamine)-car-
boxyamide-3,5-bis(N,N-bis(2-pyridylmethyl)-ethylenediamine)-
1
carboxyamide, ABB. H NMR (300 MHz, CD CN) d(ppm):
3
1
.05 (t, J = 7.5 Hz, 3H, CH CH ); 1.13 (t, J = 7.5 Hz, 6H,
2
3
Experimental
CH
2
CH ); 2.19 (s, 6H, N(CH
3
)
3
2
); 2.44 (t, J = 6.3 Hz, 2H,
–
CH
Hz, 4H, –CH
8H, pyr-CH –N); 6.68 (br, 1H, NH amide A); 7.11 (m, 4H, 5-
2
–N of arm A); 2.57 (m, 6H, CH
2
CH
3
); 2.82 (t, J = 5.8
General procedure for the synthesis of the catalyst mixtures
2
–N of arm B); 3.45 (m, 6H, CH
2
NHCO); 3.86 (s,
To a solution of compound N,N-bis-(2-pyridylmethyl)ethy-
1
2
6
3
lenediamine B (3x equivalents with x = 0.1–0.9) and diiso-
propylethylamine (6 equiv.) in anhydrous CH Cl (1 mL) was
H pyr); 7.34 (d, J = 7.8 Hz, 4H, 3-H pyr); 7.62 (m, 4H, 4-H
3
pyr); 8.12 (br, 2H, NH amide B); 8.19 (d, J = 4 Hz, 4H, 6-H
2
2
added a solution of platform 1 (25 mg, 0.072 mmol, quantita-
pyr). ESI-MS (CH
3
OH): m/z found 835.4 (100%), 851.5
+ +
(30%); calc. 835.5 (M + Na ), 851.6 (M + K ). HPLC
tively obtained by treating the corresponding triscarboxylic
1
7
acid with SOCl
) in anhydrous CH
2
Cl
2
(0.3 mL) and stirred
(Lichrosphere RP-C18: H
min (100%).
2
O–THF–TFA = 87.5 : 12 : 0.5): 5.5
2
for 90 min in a closed vial. Next, a solution of N,N-dimethy-
lethylenediamine A [3(1 À x) + 2 equiv. with x = 0.1–0.9] and
diisopropylethylamine (6 equiv.) was added and the reaction
was stirred for 90 min after which an additional 2 equiv. were
added. After stirring over night, volatiles were removed under
reduced pressure and the residue was taken in CH Cl , washed
2,4,6-Triethylbenzene-1,3,5-tris(N,N-bis(2-pyridylmethyl)-ethy-
1
lenediamine)-carboxyamide, BBB.
DMSO-d ) d(ppm): 0.98 (t, J = 7.2 Hz, 9H, CH
q, J = 7.2 Hz, 6H, CH CH ); 2.68 (t, J = 6.4 Hz, 6H,
CH –N of arms); 3.37 (m, 6H, CH NHCO); 3.80 (s, 12H,
pyr-CH -N); 7.19 (m, 6H, 5-H pyr); 7.47 (d, J = 7.7 Hz, 6H,
-H pyr); 7.69 (m, 6H, 4-H pyr); 8.11 (br, 3H, NH amide); 8.39
H
NMR (300 MHz,
6
2
CH ); 2.36
3
(
2
3
2
2
–
2
2
with 1 M NaOH and brine, dried over Na SO and evaporated
2
4
3
2
to dryness.
3
3
(
d, J = 4 Hz, 6H, 6-H pyr). ESI-MS (CH
3
OH): m/z found
89.8; calc. 989.5 (M + Na ). HPLC (Lichrosphere RP-C18:
O–THF–TFA = 87.5 : 12 : 0.5): 6.7 min (100%).
HPLC analysis of the catalyst mixtures
+
9
Mixtures were analyzed using a Lichrosphere RP-C18 HPLC-
column using H O–THF–TFA = 87.5 : 12 : 0.5 as the eluent.
H
2
2
General procedure for the screening of the catalytic activity
Peak detection occurred at 230 nm with peak integration
directly yielding the ratios between the different components.
The measured ratios obtained this way were verified indepen-
dently using HPLC calibration curves of the separated cata-
lysts which gave a maximal error margin of 5%.
2
Cleavage experiments of HPNP were performed in H O
buffered at pH 7.2 with HEPES (10 mM) at 40 1C. For
solubility reasons the catalysts were added from a stock
solution in 50% H O–CH CN. The catalyst concentrations
2
3
1
496 | New J. Chem., 2006, 30, 1493–1497
This journal is ꢀc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006

View Article Online
were calculated using an apparent molecular weight based on
the measured HPLC distribution. Kinetic experiments were
performed by measuring the absorbance of p-nitrophenolate
at 400 nm at 2 min intervals for at least 3 h. After each
measurement the pH was determined to exclude changes. Initial
rates were calculated by linear regression over the first 10 000
seconds. Each measurement was performed at least in duplicate.
3 (a) M. B. Steffenson and E. E. Simanek, Angew. Chem., Int. Ed.,
2
004, 43, 5178–5180; (b) V. Del Amo, L. Siracusa, T. Markidis, B.
Baragana, K. M. Bhattarai, M. Galobardes, G. Naredo, M. N.
Perez-Payan and A. P. Davis, Org. Biomol. Chem., 2004, 2,
320–3328; (c) A. Madder, L. Li, H. De Muynck, N. Farcy, D.
˜
´
´
3
Van Haver, F. Fant, G. Vanhoenacker, P. Sandra, A. P. Davis and
P. J. De Clercq, J. Comb. Chem., 2002, 4, 552–562; (d) T. Opatz
and R. M. J. Liskamp, Org. Lett., 2001, 3, 3499–3502; (e) X.-T.
Zhou, A. Rehman, C. Li and P. B. Savage, Org. Lett., 2000, 2,
3
015–3018.
Assuming an independent contribution from each catalyst.
General procedure for deconvoluting the rates obtained for the
catalyst mixtures
4
5 A similar approach has been previously applied to supramolecular
assemblies, but the dynamic nature of these systems has always
prevented a validation of the results obtained for the heteromeric
species. See: (a) M. T. Reetz, T. Sell, A. Meiswinkel and G. Mehler,
Angew. Chem., Int. Ed., 2003, 42, 790–793; (b) P. Molenveld, J. F.
J. Engbersen and D. N. Reinhoudt, Angew. Chem., Int. Ed., 1999,
The initial rate is determined by the contribution of each
catalyst present in the mixture following: ninit = k
+ kAAB[AAB] + kABB[ABB] + kBBB[BBB] + kZn [Zn ].
[AAA]
2+
AAA
2
+
For each individual catalyst mixture the catalyst concentra-
tions are known, leaving the rate constants as unknowns.
3
6 K.-J. Johansson, M. R. M. Andreae, A. Berkessel and A. P. Davis,
8, 3189–3191.
1
5
Tetrahedron Lett., 2005, 46, 3923–3926.
P. Chen, Angew. Chem., Int. Ed., 2003, 42, 2832–2847.
(a) F. Manea, F. Bodar Houillon, L. Pasquato and P. Scrimin,
Angew. Chem., Int. Ed., 2004, 43, 6165–6169; (b) A. Scarso, U.
Least square fitting was performed with Scientist on nine
mixtures contemporaneously.
7
8
Determination of the Michaelis–Menten parameters
Scheffer, M. Go
C. Toniolo and P. Scrimin, Proc. Natl. Acad. Sci. U. S. A., 2002,
9, 5144–5149; (c) C. Sissi, P. Rossi, F. Felluga, F. Formaggio, M.
¨
bel, Q. B. Broxterman, B. Kaptein, F. Formaggio,
Michaelis–Menten parameters were obtained for the pure
catalysts under similar experimental conditions as before with
substrate concentrations ranging from 0.1 to 3 mM (8 points).
Initial rates were determined based on the first 1000 s and
the plot of vinit vs. S was fitted using the Scientist package
software with the Michaelis–Menten equation vinit = Vmax  S/
9
Palumbo, P. Tecilla, C. Toniolo and P. Scrimin, J. Am. Chem. Soc.,
2001, 123, 3169–3170; (d) P. Rossi, F. Felluga, P. Tecilla, F.
Formaggio, M. Crisma, C. Toniolo and P. Scrimin, Biopolymers,
2
000, 55, 496–501.
(a) J. R. Morrow and O. Iranzo, Curr. Opin. Chem. Biol., 2004, 8,
92–200; (b) N. Williams, B. Takasaki, M. Wall and J. Chin, Acc.
Chem. Res., 1999, 32, 485–493; (c) D. Wilcox, Chem. Rev., 1996,
6, 2435–2458.
0 (a) G. Hennrich and E. V. Anslyn, Chem.–Eur. J., 2002, 8,
219–2224; (b) M. Komiyama, S. Kina, K. Matsumura, J. Sumao-
ka, S. Tobey, V. M. Lynch and E. V. Anslyn, J. Am. Chem. Soc.,
002, 124, 13731–13736.
1 (a) J. K. Romary, J. D. Barger and J. E. Bunds, Inorg. Chem., 1968,
, 1142–1145; (b) A. Ojida, M.-A. Inoue, Y. Mito-oka and I.
9
1
M
(K + S). Next, the Michaelis–Menten parameters for AAB
9
and ABB were calculated by determining the observed rate
constants kobs for each of them at multiple substrate concen-
trations using the procedure identical as described before.
1
1
2
2
Acknowledgements
7
Hamachi, J. Am. Chem. Soc., 2003, 125, 10184–10185; (c) S.
Takebayahi, M. Ikeda, M. Takeuchi and S. Shinkai, Chem. Com-
mun., 2004, 420–421.
Support by MIUR (contract 2003037580, PRIN 2003) and the
University of Padova (Project CPDA054893) is gratefully
acknowledged.
1
2 An equal molecular absorption coefficient for the four catalysts at
230 nm was assumed, since at this wavelength the chromophore
that differs the four species (pyridine) has a very low absorbance
À1
À1
(
e = 393 L mol cm ). As validation, the obtained concentrations
References
were compared with concentrations obtained from calibration
curves of the catalysts separately. The maximum error was 3%.
1
(a) N. Srinivasan and J. D. Kilburn, Curr. Opin. Chem. Biol., 2004,
, 305–310; (b) L. Baldini, A. J. Wilson, J. Hong and A. D.
8
13 The average molecular weight of each mixture can be calculated
from the mixture distribution.
14 The cleavage of HPNP was measured using UV-Vis spectroscopy
following the increase in absorbance at 400 nm due to the release of
the nitrophenolate ion.
15 MicroMath Scientist for Windows, version 2.01.
16 K. Hanaoka, K. Kikuchi, Y. Urano and T. Nagano, J. Chem. Soc.,
Perkin Trans. 2, 2001, 1840–1843.
17 S. V. Kolotuchin, P. A. Thiessen, E. E. Fenlon, S. R. Wilson, C. J.
Loweth and S. C. Zimmerman, Chem.–Eur. J., 1999, 5, 2537–2547.
Hamilton, J. Am. Chem. Soc., 2005, 126, 5656–5657; (c) C.
Chamorro and R. M. J. Liskamp, J. Comb. Chem., 2003, 5,
7
94–801; (d) S. C. McClesky, M. J. Griffin, S. E. Schneider, J. T.
McDevitt and E. V. Anslyn, J. Am. Chem. Soc., 2003, 125,
114–1115.
1
2
(a) A. Fersht, Structure and Mechanism in Protein Science: A Guide
to Enzyme Catalysis and Protein Folding, W. H. Freeman, New
York, 3rd edn, 1999; (b) A. J. Kirby, Angew. Chem., Int. Ed. Engl.,
1996, 35, 707–724.
This journal is ꢀc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1493–1497 | 1497