Synthetic catalytic pores

Article abstract of DOI:10.1021/ja029845w

Catalytic activity of a synthetic multifunctional pore is studied in large unilamellar vesicles under conditions where substrate and synthetic catalytic pore (SCP) approach the membrane either from the same side (cis catalysis) or from opposite sides (trans catalysis). A synthetic supramolecular rigid-rod β-barrel with excellent ion channel characteristics is identified as SCP using 8-acetoxypyrene-1,3,6-trisulfonate (AcPTS) as model substrate. The key finding is that application of supportive membrane potentials increases the initial velocity of AcPTS esterolysis (v₀). This results in an increase of V_max beyond experimental error (+30%), whereas KM increases less significantly. Long-range electrostatic steering by the membrane potential, possibly guiding substrates into the transmembrane catalyst and, more importantly, accelerating product release (f_off = 1.3) is discussed as one possible explanation of this global reduction of catalyst saturation. Control experiments show, inter alia, that similarly strong changes do not occur with opposing membrane potentials. Copyright

Full text of DOI:10.1021/ja029845w

Published on Web 06/07/2003
Synthetic Catalytic Pores
Naomi Sakai,* Nathalie Sorde´, and Stefan Matile*
Department of Organic Chemistry, UniVersity of GeneVa, CH-1211 GeneVa, Switzerland
Received December 20, 2002; E-mail: naomi.sakai@chiorg.unige.ch; stefan.matile@chiorg.unige.ch
In this communication, we report experimental evidence for the
existence of synthetic catalytic pores (SCPs, Figure 1).1,2 Substrate
conversion during substrate translocation through a membrane-
spanning SCPs has captivated our attention since the discovery of
synthetic multifunctional pores (SMPs)1 with esterase and ion
channel activity.2a The scenario of substrate-loaded vesicles “spitting
out” products formed on the way through an SCP was particularly
attractive because it suggested that the rate of substrate binding
(kon) and substrate and/or product release (koff) may be manipulated
with, e.g., concentration gradients or membrane potentials.
To implement conditions appropriate to test this hypothesis, large
unilamellar vesicles composed of egg yolk phosphatidylcholine
(EYPC LUVs) were loaded with 8-acetoxypyrene-1,3,6-trisulfonate
(AcPTS) as model substrate of choice.2a-e With the obtained EYPC-
LUVs⊃AcPTS,3,4 trans esterolysis5 was initiated by extravesicular
addition of SCP1, a rigid-rod â-barrel with internal arginine-histidine
dyads that forms highly stable, multifunctional ion channels.6,7
Product formation after pore addition (Figure 2A, ii) was monitored
continuously by following the increase in emission of product with
time (Figure 2A, b).2a,8,9 Nonlinear dependence of the initial velocity
of product formation on substrate concentration (Figure 2B, O)
provided kinetic data for trans esterolysis in unpolarized spherical
membranes (Table 1, entry 1).
Figure 2. (A) Increase of product concentration with time after addition
of (i) valinomycin and (ii) SCP1 to EYPC-LUVs⊃AcPTS at (a) E ≈ -50
mV and (b) E ) 0 mV;8 (c) membrane polarization during experiment a;11
(iii-v): see text.9,13 (B) Initial velocity of product formation as a function
of substrate concentration at E ) 0 mV (O) and E ≈ -50 mV (b) with
curve fit.
Table 1. Kinetic Data for Esterolysis of AcPTS by SCP1
d
e
entry
cond.a
E (mV)b
KM (µM)c
V
max (pM/s)c
fon
foff
1
2
3
4
trans
trans
cis
0
-50
0
4.1 ( 0.7
4.7 ( 1.0
6.1 ( 0.9f
8.1 ( 1.9
170 ( 16g
222 ( 26g
202 ( 16f
213 ( 28
1.10
0.78
1.30h
1.05
cis
-50
a Conditions: see text.3-5,8 b Membrane potential. c Michaelis constant
Vesicle polarization was achieved by means of a potassium
diffusion potential as described previously, and verified by an
increase in emission of potential-sensitive Safranin O (Figure 2A,
c, v).10,11 Taking advantage of differences in the pH profiles of
and maximal velocity calculated by fitting the experimental data to V0 )
(Vmax x cAcPTS,total)/(KM + cAcPTS,total). fon ) foff KM0 mV/KM
. )
e foff
d
-50 mV
-50
0 mV
Vmax
mV/Vmax
.
f See ref 18. g Key data concerning the impact of
polarization on trans catalysis. h Key data concerning eventual electrostatic
steering (error ( 0.19).15
esterolysis (maximal at 5 < pH < 6), and pore formation (maximal
at pH < 5) by SCP1,2a,c,12 it was possible to perform trans esterolysis
in polarized membranes. The application of membrane potentials
E ≈ -50 mV13 resulted in an increase of Vmax beyond experimental
error (+30%) together with an eventual minor increase of KM
(+15%, Table 1, entry 2 versus entry 1).14
One possible explanation of these changes is that inside-negative
potentials guide internal, anionic substrates into the SCP1 and the
anionic products into the external media.15 Assuming a Briggs-
0mV
Haldane mechanism,16 factors fon ) kon-50mV/kon
and foff )
koff-50mV/koff0mV could be calculated (Table 1, entry 2).3 The obtained
values supported acceleration of substrate/product binding/release
(i.e., foff g fon g 1) by constructive electrostatic steering in trans
catalysis, with foff > 1 as key factor beyond experimental error.17
The key observation in control experiments with cis esterolysis3
was the absence of comparably strong changes upon polarization.
This difference supported validity of the changes seen with trans
esterolysis.5,18 Within experimental error, V0 decreased only slightly
because KM (+33%) increased more than Vmax (+5%, Table 1, entry
4 versus entry 3). Hindered substrate binding by interfering
electrostatic steering was one possible explanation for these less
relevant changes (fon ) 0.78, Table 1, entry 4).19,15 These
complementary trends with cis and trans esterolysis in polarized
membranes hinted at the attractive scope of a novel approach toward
Figure 1. Trans catalysis in polarized membranes. The influence of inside-
negative Nernst potentials (red) on the conversion of intravesicular anionic
substrates (AcPTS) by externally added SCP1 into anionic products (HPTS)
6
is assessed. SCP1 /SMP12a-e in lipid bilayers (EYPC) are depicted in blue
(hydrophilic), gold (hydrophobic), and black (rigid-rod and â-sheet scaffold),
R-amino acid residues pointing exterior of the barrel are black on white,
internal ones white on blue (single-letter abbreviations: L, Leu; H, His; R,
Arg; V, valinomycin).
9
7776
J. AM. CHEM. SOC. 2003, 125, 7776-7777
10.1021/ja029845w CCC: $25.00 © 2003 American Chemical Society

C O M M U N I C A T I O N S
nm, λex ) 522 nm) was followed as a function of time simultaneously to
the continuous detection of trans AcPTS esterolysis in polarized LUVs
(Figure 2A, a).8 Potentials were quantified from Safranin O emission
intensities using calibration curves.3, 10, 13
vectorial control of catalysis and, most importantly, corroborated
the existence of catalytic pores.
Acknowledgment. We thank R. Gurny for access to an
osmometer, two reviewers for helpful suggestions, and the Swiss
NSF for financial support (2000-064818.01 and National Research
Program “Supramolecular Functional Materials” 4047-057496).
(12) Data not shown.
(13) We were unable to find conditions where an “initial burst” of depolarization
from E ≈ -100 mM to E ≈ -50 mM can be avoided (Figure 2A, iv).
This observation was rationalized by initially poor selectivity between
AcPTS efflux and Na+ influx (i.e., depolarization by compensating,
valinomycin-mediated K+ antiport) until each pore is loaded with anionic
AcPTS blockers,6b “hopping” toward the exterior and inhibiting concurrent
cation exchange. (Note that, given the near absence of anion/cation
selectivity of SCP1 at pH 5.5,6b about 50% initial depolarization is
consistent with this explanation.) Decrease of Safranin O emission to the
original value for E ≈ 0 mV upon addition of excess melittin confirmed
existence of partial polarization (E ≈ -50 mV, Figure 2A, c) during
esterolysis (Figure 2Aa).
Supporting Information Available: Experimental details (PDF).
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(14) Initial velocities of product formation were nearly identical to that at E )
0 mV (Table 1, entry 1) in absence of either valinomycin or K+/Na+
gradient.3 The error level in Table 1 is affected by experimental
inaccessibility of data points at high substrate concentration; clear
differences in V0 with/without polarization were reproducibly observed
in trans catalysis with >3 µM substrate, measured always in parallel at
fixed substrate concentration (Figure 2B, filled versus empty circles).3
(15) Origins of the voltage dependence of esterolysis other than electrostatic
steering were considered. (a) Contributions from potential-induced asym-
metric changes of the pKa’s of internal histidines:2c Relevance of this
interesting potential expression of remote control in catalysis is not
supported by ohmic ion-channel behavior,6b relatively flat pH profiles
around pH 5.5, and increase (rather than decrease) of V0 in trans catalysis
upon polarization.2a,c,12 (b) Changes in SCP1 concentration/ conformation
upon polarization are not supported by potential-independence of membrane-
bound-SCP1 concentration found by fluorescence depth quenching experi-
ments3 and ohmic ion-channel behavior.6b (c) Relevant contributions from
substrate conversion in the external media are not supported by the found
increase in Vmax and KM upon polarization (rather than unchanged Vmax
and reduced KM expected for accelerated efflux, i.e., increased external
substrate concentration, upon polarization) and experimental evidence for
blockage of cation exchange6b during 1,3,6-pyrenetrisulfonate efflux.2a,c,e
under relevant conditions.2a,6b,12 (d) Exclusive substrate conversion in the
internal media is excluded by the found dependence of trans catalysis on
substrate concentration.
(1) Synthetic catalytic pores (SCPs): pores constructed from abiotic scaffolds
that catalyze substrate conversion during substrate translocation across
the same pore; “synthetic multifunctional pores” (SMPs): pores con-
structed from abiotic scaffolds with additional function(s); all SCPs are
SMPs, but not all SMPs are SCPs.
(2) Catalytic pores have not been reported previously. SMPs with ion channel
and catalytic activity: (a) Baumeister, B.; Sakai, N.; Matile, S. Org. Lett.
2001, 3, 4229-4232. (b) Som, A.; Matile, S. Eur. J. Org. Chem. 2002,
3874-3883. (c) Baumeister, B.; Som, A.; Das, G.; Sakai, N.; Vilbois, F.;
Gerard, D.; Shahi, S. P.; Matile, S. HelV. Chim. Acta 2002, 85, 2740-
2753. (d) Baumeister, B.; Matile, S. Macromolecules 2002, 35, 1549-
1555. (e) Som, A.; Sakai, N.; Matile, S. Bioorg. Med. Chem. 2003, 11,
1363-1369. Molecular recognition within pores: (f) Bayley, H.; Cremer,
P. S. Nature 2001, 413, 226-230. (g) Das, G.; Talukdar, P.; Matile, S.
Science 2002, 298, 1600-1602. Covalent capture within pores: (h)
Mindell, J. A.; Zhan, H.; Huynh, P. D.; Collier, R. J.; Finkelstein, A.
Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 5272-5276. (i) Shin, S.; Luchian,
T.; Cheley, S.; Braha, O.; Bayley, H. Angew. Chem., Int. Ed. 2002, 41,
3707-3709. Reactions within vesicles: (j) Walde, P.; Ichikawa, S. Biomol.
Eng. 2001, 18, 143-177. Synthetic ion channels/pores: (k) Gokel, G.
W.; Mukhopadhyay, A. Chem. Soc. ReV. 2001, 30, 274-286.
(3) See Supporting Information.
(4) Osmotic stress from intravesicular AcPTS was compensated by sucrose.
EYPC-LUVs⊃AcPTS, trans catalysis, E ) 0 mV: Inside; x mM AcPTS,
100-3.2 x mM sucrose, 100 mM KCl, 5 mM TES, pH 7.0, x ) 3-18
mM, outside; 100 mM sucrose, 100 mM KCl, 10 mM MES, pH 5.5.2a,c,5,8
EYPC-LUVs, cis catalysis, E ) 0 mV: Inside; 100 mM KCl, 5 mM TES,
pH 7.0, outside; 100 mM KCl, 10 mM MES, pH 5.5.2a,c,5,8 For E ) -50
mV: NaCl instead of KCl was used for outside buffer in cis and trans
esterolysis.
(16) Fersht, A. Enzyme Structure and Mechanism, 2nd ed.; W. H. Freeman
and company: New York, 1985. Here, kcat is assumed to be koff of product,
and effects of membrane polarization on koff of substrate and product to
be the same.3
(17) Electrostatic steering in enzymes, e.g.: Wade, R.; Gabdoulline, R. R.;
Lu¨demann, S. K.; Lounnas, V. Proc. Natl. Acad. Sci. U.S.A. 1998, 95,
5942-5949.
(18) The influence of internal HfR mutation on catalyst characteristics was
as follows (cis esterolysis, E ) 0 mV):a
(5) “Trans catalysis”: addition of substrate (here: intravesicular) and catalyst
(here: extravesicular) from opposite sides of the membrane; “cis
catalysis”: addition of substrate and catalyst from the same side of the
membrane (here: extravesicular).2c
(6) (a) Sorde´, N.; Matile, S. J. Supramol. Chem. 2003. In press. (b) Sakai,
N.; Sorde´, N.; Das, G.; Perrottet, P.; Gerard, D.; Matile, S. Org. Biomol.
Chem. 2003, 1, 1226-1231.
(7) As in previous reports on p-octiphenyl â-barrels, we reiterate that the
depicted suprastructures (as in Figure 1) can be viewed as, at worst, a
productive working hypothesis supported by all data on structure and
function available today; all concentrations indicated for SCP1/SMP1 refer
to tetramers.
b
d,e
kcat
[min-1
KM
[µM] (kcat/KM)/kMeIm
(kcat/KM)/kuncat
[M-1
c
d
pore
]
kcat/kuncat
]
SMP1 0.13f 0.7f 9.6 × 105 f 5.0 × 103
7.1 × 109
SCP1 0.24 6.1 2.0 × 105
9.2 × 103
1.5 × 109
(8) Continuous detection of AcPTS esterolysis (representative original data:
Figure 2A (a and b): Safranin O (60 nM) and AcPTS (0 µM for trans,
1-8 µM for cis esterolysis) were added to EYPC-LUVs⊃AcPTS (95 (
15 µM PC, cis esterolysis: 0 mM AcPTS.; trans: 3-18 mM AcPTS) in a
stirred and thermostated fluorescence cuvette (2 mL).4 Changes in
fluorescence intensity I of HPTS (λem ) 510 nm, λex ) 415.5 nm) were
then recorded as a function of time during addition of valinomycin (Figure
2A, i; for a: 600 nM, b: 0 nM) and SCP1 (Figure 2A, ii; 50 nM).
Concentration of HPTS was determined from HPTS emission intensities
using calibration curves.3
(9) The origin of initial “burst” in fluorescence emission of HPTS after the
addition of SCP1 is unknown (Figure 2A, iii). However, clear dependence
on substrate concentration suggested that these “bursts” do not originate
from accumulation of reactive intermediates.3
a Conditions, see refs 2a, 3, and 8. b ∆∆GES0 ) ∆GES0 (SCP1) (-29.6
kJ/mol) - ∆GES0 (SMP1) (-35.0 kJ/mol) ) +5.4 kJ/mol [assuming
KM ) KD(substrate)]. c Catalysis by 4(5)-methylimidazole: kMeIm
)
0.0032 M-1 s-1 2a
.
d Autohydrolysis: kuncat ) 4.34 × 10-7 s-1 (pH
5.5). e ∆∆GTS ) ∆GTS° (SCP1) (-52.2 kJ/mol) - ∆GTS (SMP1)
0
0
(-56.0 kJ/mol) ) +3.8 kJ/mol. f Data from 2a.
(19) It was intriguing to note that the effect of membrane polarization on fon
was more pronounced in cis esterolysis than in trans esterolysis. These
results might suggest that the rate-limiting step in cis esterolysis is binding
of the substrate, while that in trans esterolysis is dissociation of the product.
This difference may originate from an increased binding rate in trans
esterolysis due to the locally increased substrate concentration at the
entrance of SCP1 (about 3000×).
(10) (a) Sakai, N.; Matile, S. J. Am. Chem. Soc. 2002, 124, 1184-1185. (b)
Sakai, N.; Houdebert, D.; Matile, S. Chem.sEur. J. 2003, 9, 223-232.
(11) Continuous detection of membrane potential (representative original
data: Figure 2Ac): Fluorescence intensity I of Safranin O (c; λem ) 581
JA029845W
9
J. AM. CHEM. SOC. VOL. 125, NO. 26, 2003 7777