Quaternary ammonium ions can externally block voltage-gated K+ channels. Establishing a theoretical and experimental model that predicts KDS and the selectivity of K+ over Na+ ions

Article abstract of DOI:10.1016/S0022-2860(00)00774-2

The physicochemical basis for the high ion selectivity of potassium channels is poorly understood. In the present studies, external blockade of cloned voltage-gated potassium channels with alkyl quaternary ammonium ions are analyzed from a model derived from theory and experimental data. Atomic mass units, electrostatic potential residing on the nitrogen atom, the COSMO van der Waals solvent accessible surface, the Onsager solvation model, and the isodensity PCM solvation model are computed at the semi-empirical and the ab initio levels of theory. A structure-activity relationship (SAR) exists between the calculated values and the experimentally obtained K_D (mM). The SAR model gives us K_D predictions and when K⁺ and Na⁺ are incorporated into the model, it dramatically predicts the selectivity of K⁺ over Na⁺ ions.

Full text of DOI:10.1016/S0022-2860(00)00774-2

Journal of Molecular Structure 562 (2001) 63±78
www.elsevier.nl/locate/molstruc
Quaternary ammonium ions can externally block voltage-gated K1
channels. Establishing a theoretical and experimental model that
predicts KDs and the selectivity of K1 over Na1 ions
*
M.F. Wempe
Department of Pharmacology and Physiology, University of Rochester Medical Center,
601 Elmwood Avenue, Box 711, Rochester, NY 14642, USA
Received 24 July 2000; accepted 10 October 2000
Abstract
The physicochemical basis for the high ion selectivity of potassium channels is poorly understood. In the present studies,
external blockade of cloned voltage-gated potassium channels with alkyl quaternary ammonium ions are analyzed from a model
derived from theory and experimental data. Atomic mass units, electrostatic potential residing on the nitrogen atom, the
COSMO van der Waals solvent accessible surface, the Onsager solvation model, and the isodensity PCM solvation model
are computed at the semi-empirical and the ab initio levels of theory. A structure±activity relationship (SAR) exists between the
calculated values and the experimentally obtained KD (mM). The SAR model gives us KD predictions and when K1 and Na1 are
incorporated into the model, it dramatically predicts the selectivity of K1 over Na1 ions. q 2001 Elsevier Science B.V. All
rights reserved.
Keywords: Structure±activity relationship; Voltage-gated potassium channel; Bio-cationic focusing mechanism
* Tel.: 11-716-275-2659; fax: 11-716-273-2652.
Abbreviations: BDEMA, butyldiethylmethyl ammonium ion; BDMIPA, butyldimethylisopropyl ammonium ion; BEDIPA, butylethyldiiso-
propyl ammonium ion; BisTEAEtOEt, bis triethyl ethyl ether ammonium ion; BTEA, butyltriethyl ammonium ion; BTPA, butyltripropyl
ammonium ion; DecTEA, decyltriethyl ammonium ion; DEDABCO, 1,4-diethyl-1,4-diazabicyclo[2.2.2]octane ammonium ion; DEDMA,
diethyldimethyl ammonium ion; DEDIPA, diethyldiisopropyl ammonium ion; DEDPA, diethyldipropyl ammonium ion; DEMPA, diethyl-
methylpropyl ammonium ion; DEMTolA, a-diethylmethyl toluene ammonium ion; DIEtOEt, bis iodoethyl ether; DMDABCO, 1,4-dimethyl-
1,4-diazabicyclo[2.2.2]octane ammonium ion; DMDIPA, dimethyldiisopropyl ammonium ion; DMDPA, dimethyldipropyl ammonium ion;
DMPIPA, dimethylpropylisopropyl ammonium ion; EDIPMA, ethyldiisopropylmethyl ammonium ion; EDIPPA, ethyldiisopropylpropyl
ammonium ion; EDMIPA, ethyldimethylisopropyl ammonium ion; ETMA, ethyltrimethyl ammonium ion; ETPA, ethyltripropyl ammonium
ion; HexTEA, hexyltriethyl ammonium ion; ITol, a-iodotoluene; MTPA, methyltripropyl ammonium ion; OctTEA, octyltriethyl ammonium
ion; TBA, tetrabutyl ammonium ion; TBEA, tributylethyl ammonium ion; TBMA, tributylmethyl ammonium ion; TBPA, tributylpropyl
ammonium ion; TEMA, triethylmethyl ammonium ion; TEA, tetraethyl ammonium ion; TEPA, triethylpropyl ammonium ion; TETolA, a-
triethyltoluene ammonium ion; THexA, tetrahexyl ammonium ion; TMA, tetramethyl ammonium ion; TMIPA, trimethylisopropyl ammonium
ion; TPA, tetrapropyl ammonium ion; TPATolA, a-tripropyl toluene ammonium ion; TPentA, tetrapentyl ammonium ion; 3DEMAPrPh, 3-
phenyl-(N-diethylmethyl) propyl ammonium ion; 3IPrPh, 3-iodo-1-phenylpropane; 3PhenODEMAPr, 3-phenoxy-(N-diethylmethyl)propane
ammonium ion; 3PhenOIPr, 3-phenoxy-iodopropane; 3PhenOTEAPr, 3-phenoxy-(N-triethyl)propane ammonium ion; 3TEAPrPh, 3-phenyl-
(N-triethyl) ammonium ion.
0022-2860/01/$ - see front matter q 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-2860(00)00774-2

64
M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
1. Introduction
eight of the known mammalian homologs from the
Shaker subfamily.
Models illustrating biological activity as a function
of chemical structure are called structure±activity
relationship (SAR) models [1]. SAR models can
provide us with qualitative and/or quantitative predic-
tions. Physical-organic chemists have used the
Hammett equation to investigate the relationship of
structure and function, by way of electronic factors
that can help decipher chemical mechanism and
chemical trend [2]. Pharmaceutical companies have
used SAR models in their efforts toward rational
drug design and SAR models are useful to the
environmental and toxicological community [3].
Perhaps SAR models can help decipher biological
functions that evolved early along the phylogenetic
tree of life. For example, potassium (K1) channels
are believed to have occurred as early as Protists,
which evolved from prokaryotes nearly 1400 million
years ago, yet the physicochemical basis for the high
ion selectivity of potassium channels is poorly to
moderately understood. Hence, analyzing the inter-
actions of alkyl quaternary ammonium (QA) ions
with voltage-gated potassium channels became a
goal. Using experimentally published KD (mM) data,
QA ions are theoretically compared to properties
computed at semi-empirical (AM1) [4] and ab initio
(HF/3-21G) molecular orbital (MO) levels of theory
[5].
Potassium channels are comprised of four a-
subunits that concurrently allow ion selectivity.
Potassium channels are subdivided into three structural
classes and differ by the number of transmembrane
domains (TM) found in the a-subunits [6]. The
voltage-gated family consists of 6TM domains (S1±
S6) with one P region between S5 and S6. Consisting
of a positively charged arginine or lysine residue at
every third position, the fourth transmembrane
segment (S4) has a highly conserved and repeating
amino acid motif. The charged subset makes the
essential contributions to the mechanism of voltage-
dependent activation. Voltage-gated potassium
channels are activated by membrane depolarizations
but may have a range of voltage sensitivities and
gating kinetics. There are four gene subfamilies in
the voltage-gated family: Shaker(Kv1.#), Shab(Kv2.#),
Shaw(Kv3.#), and Shal(Kv4.#). Each subfamily may
have several members; for example, Kv1.1±Kv1.8 are
Water-solvated proteins, potassium channels
are pores that allow certain ions to pass [7]. These
channels catalyze ion transport across the lipid bilayer
membrane. Molecular cloning and mutagenesis
studies demonstrate that all potassium channels have
nearly the same pore construction. The biophysical
evidence indicates the pore region to contain multiple
binding sites. These cationic binding sites have a high
af®nity for K1 and function as a selectivity ®lter [8].
The four a-subunits form a cone, which cradle this
selectivity ®lter. Compared to Na1 ions, K1 channels
are much more selective for K1 ions. Indeed, K1
channels can be thousands of times more permeable
to K1 than to Na1; throughput rates may be as high as
108 ions per second. Therefore, K1 channels are long
narrow pore channels that produce an ion ¯ux. Potas-
sium channels also contain a critical amino acid
sequence, commonly called the `potassium channel
signature sequence'. Mutating the amino acids of
the signature sequence can disrupt the preference of
K
1 over Na1.
The amino acids that in¯uence ion selectivity and
the conductance of the pore are essential. The initial
work that probed the pore region utilized charybdo-
toxin (CTX), a small peptide found in scorpion
venom. When CTX is externally applied, it inhibits
voltage-gated potassium channels. CTX blocks the
potassium channel pore, interacting electrostatically
with the channel. Mackinnon and Miller [9] mutated
negatively charged amino acids on the external loops
of the Shaker potassium channel (S1±S2, S3±S4, and
S5±S6 linkers) and found that external loops S1±S2
and S3±S4 did not affect CTX blockade. However,
mutations in the S5±S6 linker produced CTX-
insensitive channels and include aspartate 431, threo-
nine 449, and proline 451.
CTX is too bulky to enter the narrow pore region of
a potassium channel, so smaller cations are used to
probe various structural questions. For example, tetra-
ethyl ammonium ion (TEA) is a K1-channel blocker.
TEA is impermeable to cloned K1 channels that were
expressed in Xenopus oocytes, but internally and
externally block outwardly rectifying currents. Muta-
genesis studies show threonine 449 to have a profound
effect on the external TEA sensitivity. When threo-
nine 449 was mutated to tyrosine or phenylalanine,

M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
65
the external sensitivity of TEA increased and was a
function of the number of a-subunits containing this
modi®cation. The proposed hypothesis was that TEA
blocked the channel pore by interacting with the four
tyrosines at the extracellular entryway.
Jarolimek et al. [10,11] examined the selectivity of
QA ions on the external binding site of transfected
Xenopus oocytes (Kv3.1, Kv2.1, and Kv1.1). The external
potencies of Kv3.1 were (TEA . TPA . TBA) and
Kv2.1 were (TBA . TEA . TPA). QA ions blocked
the potassium current in a fast, but reversible, concen-
tration-dependent manner. Jarolimek et al. [10,11]
observed that, with Kv3.1, the free energies of binding
[DDGBinding ˆ 2RT ln(1 M/KD)] correlated with
changes in the computed free energies of hydration
[DDGHydration].
Reaction Field (SCRF) solvation models were
computed [12,13]. The SCRF methods use a solvent,
in this case water (e) of 78.39, as a continuum of
uniform dielectric constant (e). Two solvation models
were computed. The Onsager reaction ®eld model
where the solute occupies a ®xed spherical cavity
radius (a0), and the IPCM model where the solute
resides in an intuitive, but ¯exible, shape.
2.2. Chemicals
Iodomethane, iodoethane, iodopropane, iodo-
butane, triethylamine, tripropylamine, tributylamine,
diethylmethylamine, diisopropylethylamine, ethyl-
diisopropylamine, tetramethylammonium bromide,
tetrabutylammonium bromide, tetrapropylammonium
bromide, a-chlorotoluene, 3-bromo-1-phenylpropane,
2-chloroethyl ether, 2-butanone, ethyl acetate,
diethyl ether, methanol, 1,4-diazabicyclo[2.2.2]
octane (DABCO), phenol, sodium metal, sodium
iodide, N,N,N0,N0-tetraethylenediamine, 1-bromo-3-
chloropropane, and gallamine were purchased from
Sigma±Aldrich Chemical Co.
2. Experimental
2.1. Computational methods
Chemical structures were drawn with CS Chem-
Draw Ultraw (version 5.0, Cambridge Soft Corpora-
tion) and copied into CS Chem3D Prow (version 5.0,
Cambridge Soft Corporation). A molecular mechanics
(MM) minimization was performed with a root-mean-
square (RMS) gradient of 0.001. The cs mopac proe
application was used to generate molecular geome-
tries in the Gaussian Z-matrix format. These input
®les were transferred to a UNIX-based Silicon
Graphics O2e Workstation running Gaussianw 98
(Gaussian, Inc.), which performed the semi-empirical
Austin-Model 1 (AM1), and ab initio (HF/3-21G)
molecular orbital (MO) calcualations [5]. At each
geometry optimization, the molecular outputs in
their Cartesian Coordinates (CC) were transferred
into CS Chem 3D Prow and heats of formation, dipole
moments, Mulliken charges on the nitrogen, electro-
static potential (ESP) on the nitrogens, and
2.3. Instrumental analyses
2.3.1. GC/MS analysis
GC/MS analyses were performed with a Hewlett±
Packard 5890 gas chromatograph ®tted with a HP-1
crosslinked methyl siloxane column (25 m £ 0.2 mm
i.d.; and 0.5 mm ®lm thickness) and coupled to a
Hewlett±Packard 5972A mass selective detector.
The carrier gas was He. 1.0 ml samples were injected
in the splitless mode. The temperatures of the injector
port and transfer line were 2308C. The column
temperature was held at 708C for 5 min, increased to
2208C at a rate of 108C/min, and held at 2208C for
358C/min. Samples were analyzed in the electron
impact (EI) mode at 70 eV, and the mass selective
detector was calibrated with per¯uorotributylamine.
Conductor-like
Screening
solvation
Model
(COSMO) calculated with cs mopac proe. The
COSMO model generates a conducting polygonal
surface around the molecule at the van der Waal's
distance giving a solvent-accessible surface (SAS).
Water, with a dielectric constant of 78.4 at 258C,
was used as the solvent. In addition to the gas-phase
ab initio computations, a series of Self-Consistent
2.3.2. LC/MS (ESP)
A Hewlett±Packard series 1100 LC/MSD with an
electrospray (ES) inlet operating in the positive ion
mode was used to verify the atomic mass (amu) for the
quaternary ammonium ions. Water was used as the
solvent.

66
M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
Table 1
Chemical results (ES ˆ Electrospray analysis in the positive (1) ion mode)
QA ion
1H NMR
ES
BDEMA
(D2O) 0.76±0.81 (t, 3H), 1.11±1.25 (t & hex, 8H), 1.50±1.58 (p, 2H),
2.78 (s, 3H), 3.02±2.08 (t, 2H), 3.13±3.20 (q, 4H)
(D2O) 0.75±0.80 (t, 3H), 1.17±1.24 (d & p, 8H), 1.52±1.66 (p, 2H), 2.80
(s, 6H), 3.07±3.13 (t, 2H), 3.50±3.61 (hep, 1H)
(D2O) 0.77±0.79 (t, 3H), 1.15±1.17 (d, 12H), 1.22±1.28 (t, 3H), 2.98±
3.13 (m, 4H), 3.27±3.34 (q, 2H), 3.52±3.61 (m, 2H), 3.69±2.81 (hex,
2H)
(144.2 amu)
BDMIPA
BEDIPA
(144.2 amu)
(186.2 amu)
BisTEAEt
BisTEAEtOEt
(D2O) 1.15±1.20 (t, 18H), 3.25±3.32 (q, 12H), 3.53 (s, 4H)
(D2O) 1.08±1.12 (t, 18H), 3.17±3.25 (q, 12H), 3.34±3.37 (t, 4H), 3.75±
3.78 (t, 4H)
(M 1 I, 357.2 amu)
(M 1 I, 357.2 amu)
BTEA
BTPA
(D2O) 0.78±0.83 (t, 3H), 1.08±1.17 (t, 9H), 1.20±1.27 (hex, 2H), 1.48±
1.54 (p, 2H), 2.98±3.04 (t, 2H), 3.09±3.16 (q, 6H)
(D2O) 0.75±0.80 (t, 12H), 1.18±1.26 (hex, 2H), 1.46±1.57 (m, 8H),
2.97±3.06 (m, 8H)
(158.2 amu)
(200.2 amu)
DEDABCO
DEDIPA
(D2O) 1.32±1.36 (t, 6H), 3.57±3.65 (q, 6H), 3.91 (s, 12H)
(D2O) 1.32±1.37 (t, 6H), 1.44±1.46 (d, 12H), 3.36±3.43 (q, 4H), 3.89±
3.93 (hep, 2H)
(M 1 I, 297.1 amu)
(158.2 amu)
DEDMA
DEMPA
(D2O) 1.19±1.24 (t, 6H), 2.90 (s, 6H), 3.22±3.29 (q, 4H)
(D2O) 0.76±0.81 (t, 3H), 1.09±1.14 (t, 6H), 1.51±1.59 (hex, 2H), 2.77
(s, 3H), 2.97±3.03 (t, 2H), 3.11±3.18 (q, 4H)
(102.1 amu)
(130.2 amu)
DEMTolA
DIEtOEt
(D2O) 1.18±1.23 (t, 6H), 2.70 (s, 3H), 3.06±3.21 (m, 4H), 4.25 (s, 2H),
7.36±7.37 (bs, 5H)
(178.2 amu)
(DMSO-d6) 3.28±3.33 (t, 4H), and 3.67±3.71 (t, 4H)
GC/MS m/z 326, 199,
155(100%)
DMDABCO
DMDIPA
DMPIPA
(D2O) 3.32 (s, 6H), 3.99 (s, 12H)
(M 1 I, 269.0 amu)
(130.2 amu)
(D2O) 1.19±1.21 (d, 12H), 2.66 (s, 6H), 3.69±3.73 (hep, 2H)
(D2O) 0.77±0.82 (t, 3H), 1.19±1.21 (d, 6H), 1.58±1.68 (hex, 2H), 2.81
(s, 6H), 3.04±3.10 (t, 2H), and 3.51±3.60 (het, 1H)
(D2O) 1.13±1.20 (d & t, 15H), 2.77 (s, 3H), 3.19±3.21 (q, 2H), 3.48±
3.61 (hep, 2H)
(130.2 amu)
EDIPMA
EDIPPA
ETPA
(144.2 amu)
(172.2 amu)
(172.2 amu)
±
(D2O) 0.90±0.95 (t, 3H), 1.31±1.33 (d, 12H), 3.18±3.23 (m, 5H), 3.36±
3.45 (q, 2H), 3.67±3.73 (hex, 2H), 3.82±3.96 (hep, 2H)
(D2O) 0.75±0.80 (t, 9H), 1.08±1.10 (t, 3H), 1.48±1.56 (hex, 6H), 2.94±
3.00 (t, 6H), 3.09±3.16 (q, 2H)
Gallamine
(D2O) 1.06±1.15 (t, 27H), 3.20±3.35 (m, 18H), 3.54±3.60 (t, 6H), 4.38±
4.42 (t, 6H), 6.66±6.69 (d, 2H), 7.02±7.07 (t, 1H)
(DMSO-d6) 4.60 (s, 2H), 7.21±7.39 (m, 3H), and 7.39±7.41 (d, 2H)
(D2O) 0.78±0.83 (t, 9H), 1.55±1.63 (hex, 6H), 2.84 (s, 3H), 3.03±3.09
(t, 6H)
I-Tol
GC/MS m/z 218, 127, 91(100%)
MTPA
(158.2 amu)
TBEA
TBMA
TBPA
(D2O) 0.93±0.99 (t, 9H), 1.24±1.28 (t, 3H), 1.31±1.43 (hex, 6H), 1.60±
1.71 (p, 6H), 3.16±3.21 (t, 6H), 3.27±3.34 (q, 2H)
(D2O) 0.73±0.78 (t, 9H), 1.14±1.21 (hex, 6H), 1.45±1.56 (p, 6H), 2.79
(s, 3H), 3.02±3.07 (t, 6H)
(214.2 amu)
(200.2 amu)
(228.3 amu)
(D2O) 0.74±0.79 (t, 12H), 1.14±1.22 (hex, 6H), 1.42±1.57 (m, 8H),
2.95±3.04 (m, 8H)
(D2O) 1.06±1.12 (t, 12H), 3.06±3.13 (q, 8H)
TEA
(130.2 amu)
(116.1 amu)
(144.2 amu)
TEMA
TEPA
(D2O) 1.12±1.17 (t, 9H), 2.78 (s, 3H), 3.13±3.20 (q, 6H)
(D2O) 0.77±0.82 (t, 3H), 1.07±1.12 (t, 9H), 1.49±1.56 (hex, 2H), 2.93±
2.98 (t, 2H), 3.07±3.14 (q, 6H)
TETolA
TMA
(D2O) 1.12±1.20 (t, 9H), 2.99±3.06 (q, 6H), 4.20 (s, 2H), and 7.33±7.38
(bs, 5H)
(192.2 amu)
(74.1 amu)
(D2O) 3.02 (s, 12H)

M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
67
Table 1 (continued)
QA ion
1H NMR
ES
TMIPA
TPA
TPATolA
(D2O) 1.20±1.23 (d, 6H), 2.87 (s, 9H), 3.10±3.57 (hep, 1H)
(D2O) 0.79±0.84 (t, 12H), 1.50±1.60 (hex, 8H), 3.00±3.06 (t, 8H)
(DMSO-d6) 0.86±0.91 (t, 9H), 1.71±1.79 (hex, 6H), 3.01±3.07 (t, 6H),
4.53 (s, 2H), 7.47±7.54 (m, 5H)
(102.1 amu)
(186.2 amu)
(234.2 amu)
3DEMAPrPh
3IprPh
(DMSO-d6) 1.13±1.18 (t, 6H), 1.90±1.95 (p, 2H), 2.58±2.63 (t, 2H),
2.92 (s, 3H), 3.19±3.32 (m, 6H), and 7.20±7.33 (m, 5H)
(DMSO-d6) 1.97±2.07 (p, 2H), 2.61±2.66 (t, 2H), 3.16±3.21 (t, 2H),
7.15±7.29 (m, 5H)
(206.2 amu)
GC/MS m/z 246, 119, 91(100%);
3PhenODEMAPr
(DMSO-d6) 1.15±1.24 (t, 6H),
2.07±2.13 (p, 2H), 2.97 (s, 3H),
3.31±3.41 (m, 6H), 4.02±4.06 (t,
2H), 6.92±6.95 (m, 3H), and
7.26±7.31 (t,2H)
(222.2 amu)
3PhenOIPr
(DMSO-d6) 2.13±2.21 (p, 2H), 3.33±3.38 (t, 2H), 3.96±4.07 (t, 2H),
6.91±6.95 (m, 3H), 7.25±7.30 (m, 2H)
3PhenOTEAPr
GC/MS m/z 262, 169, 135, 107,
94(100%)
(DMSO-d6) 1.13±1.20 (t, 9H),
2.04±2.09 (p, 2H), 3.24±3.33
(m, 8H), 4.03±4.07 (t, 2H),
6.91±6.94 (m, 3H),7.26±7.31 (t,
2H)
(236.2 amu)
3TEAPrPh
(DMSO-d6) 1.05±1.16 (t, 9H), 1.86±1.91 (p, 2H), 2.60±2.65 (t, 2H),
3.11±3.32 (m, 8H), 7.19±7.32 (m, 5H)
(220.2 amu)
2.3.3. NMR analysis
2.4.1. General procedure for alkyl QA salts
1
The H NMR spectra were recorded at 300 MHz
using a General Electric QE-300 spectrometer. The
chemical shifts are reported in ppm and the splitting
patterns are: s, singlet; d, doublet; t, triplet; q, quartet;
p, pentet; hex, hextet; hep, heptet; m, multiplet; br,
broad; and bs, broad singlet (see Table 1).
The tertiary amine (0.05 mol) was dissolved in 2-
butanone (25 ml) and alkyl halide (0.05 mol) added
dropwise. Caution: when iodomethane was used, it
was dissolved in 2-butanone (5±10 ml) prior to the
(exothemic) dropwise addition. In most cases, a
condenser was attached and the reaction mixture heated
to re¯ux with a sandbath (110±1208C). The reaction
mixture was chilled in an ice bath and the solid was
collected by ®ltration. The product was transferred into
a round-bottom ¯ask and dried under vacuum
(0.6 mmHg). The isolated QA salt was recrystallized
twice from ethyl acetate, ®ltered, and dried in vacuo.
2.4. Synthesis
Some quaternary ammonium salts are available;
however, many are not and were synthesized.
Aromatic mono- and di-cationic QA ions were synthe-
sized and gallamine, a tri-cationic QA ion, is available
commercially (Schemes 1±3). Converting a tertiary
amine into a quaternary salt has been named the
Menshutkin reaction [14]. Implementing the Finkel-
stein reaction, the aromatic alkyl iodides were
prepared from their corresponding alkyl bromides
[15]. The compounds were characterized by 1H
NMR and atomic mass veri®ed by HPLC/MS with
electrospray (ES) operating in the positive ion mode.
2.4.2. Aromatic QA synthesis
(1) a-Chlorotoluene (17.25 ml, 0.15 mol) was
dissolved in acetone (150 ml). Sodium iodide (30 g,
0.2 mol) was added and the mixture stirred (24 h) at
room temperature. The reaction mixture was cooled in
an ice bath, BuÈchner ®ltered, and the ®ltrate concen-
trated under reduced pressure to afford a yellow oil.
The yellow oil was vacuum distilled (0.4 mmHg) with

68
M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
Scheme 1.
a vacuum-jacketed Vigreux column equipped with a
fraction collector (55±588C); 64% yield. a-Iodoto-
luene (4.4 g, 0.02 mol) was dissolved in 2-butanone
(2 ml) and tertiary amine (2.1 equivalents) added
dropwise. For triethylamine and diethylmethyl
amine, the reactions were exothermic but the tripropyl
amine proceeded at room temperature. The isolated
ammonium salts were recrystallized twice with ethyl
acetate, ®ltered, and dried in vacuo.
(2) 1-Bromo-3-phenylpropane (20 g) in acetone
I
Cl
I
NR
3
NaI ∆
NR3 ∆
NEt3 = TETolA
NMeEt2 = DEMTolA
NPr3 = TPATolA
Acetone
2-butanone
1)
2)
Br
I
NR
I
3
NEt3 = 3TEAPrPh
NMeEt2 = 3DEMAPrPh
NR3
2-butanone
∆
NaI ∆
Acetone
I
NR
3
Cl
I
OH
3)
O
O
EtOH
1) Na
O
NEt3 = 3PhenOTEAPr
NMeEt2 = 3PhenODEMAPr
NaI ∆
Acetone
2) 1-bromo-3-
chloropropane
NR3 ∆
2-butanone
Scheme 2.

M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
69
Scheme 3.
(20 ml) was added to acetone (100 ml) containing NaI
(20 g). The mixture was stirred for 20 h at room
temperature. The ¯ask was cooled in an ice bath,
®ltered, and ®ltrate concentrated under reduced pres-
sure. The oil was vacuum distilled (0.6 mmHg) with a
vacuum-jacketed Vigreux column equipped with a
fraction collector (90±918C) to give 3IPrPh, 63%
yield. 3IPrPh (2.46 g, 0.01 mol) was dissolved in 2-
butanone (10 ml) and triethylamine (2.1 ml) or
diethylmethylamine (1.9 ml) added drop wise. The
reaction ¯asks were sealed, crimp-top, and heated
(24 h) using a sandbath (100±1108C). The reaction
mixtures were slowly cooled to room temperature
and the solvent was removed under vacuum. The
crude products were recrystallized from ethyl acetate,
®ltered, and dried in vacuo.
(3) Phenol (18.82 g) was dissolved in absolute
ethanol (250 ml) and sodium metal (4.6 g) added in
small pieces with stirring [16]. 1-Bromo-3-chloro-
propane (29.7 ml, 47.23 g, 0.3 mol) was added and
the reaction was heated (2 h) with a sandbath tempera-
ture (110±1208C). The reaction vessel was cooled to
room temperature and the unreacted phenol removed
by BuÈchner ®ltration. The ®ltrate was concentrated
under reduced pressure and vacuum distilled
(0.87 mmHg; 82±838C) to give 55% of 3-phenoxy-
1-chloropropane; GC/MS: m/z M 1 2 172, 170, 94
(100%). The chloro-ether (0.329 mol) was diluted in
acetone (500 ml), sodium iodide (61.6 g, 0.411 mol)
added, and stirred at room temperature (24 h). The
reaction vessel was cooled in an ice bath and the
solid ®ltered. The ®ltrate was concentrated under
reduced pressure and vacuum distilled (0.4±
0.45 mmHg, 95±1008C) to give 3-phenoxy-1-iodo-
propane (57%). 3PhenODEMAPr and 3PhenoTEAPr
were synthesized analogous to the procedure
described above in two.
2.5. Di-cation synthesis
(1) DABCO (5.6 g, 0.05 mol) in MeOH (50 ml)
was stirred and 2.1 equivalents of alkyl halide
(14.9 g MeI, or 16.4 g EtI) was added dropwise
[17]. The mixture was re¯uxed (3.5 h) with a sandbath
(110±1208C). The reaction vessels were cooled in an
ice bath. The products were collected by ®ltration,
recrystallized twice from ethyl acetate, ®ltered, and
dried in vacuo; DMDABCO (85% yield) and
DEDABCO (72% yield).
(2) 2,20-Dichloroethyl ether (25 ml, 30 g) was
diluted in acetone (500 ml) containing NaI (75 g,
2.4 equivalents) and heated to re¯ux (36 h). The

70
M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
Table 2
QA and experimental KD (ND ˆ No Data NE ˆ No effect data
(3) Tetraethylethylene diamine (4.3 g, 25 mmol)
was dissolved in MeOH (25 ml) and EtI (2.1 eq.,
8.25 g) added drop wise with stirring. The mixture
was re¯uxed (7.5 h) with a sandbath (110±1208C).
The reaction vessel was cooled in an ice bath and
product collected by BuÈchner ®ltration to give
900 mg.
published by Jarolimek et al. [10,11])
QA ion
amu
KDa
KDb
Kv1.1
Kv2.1
Kv3.1
1 TMA
74.097
234
ND
78
NE
ND
ND
ND
ND
ND
ND
ND
4.62
ND
ND
ND
ND
ND
ND
ND
10
180 (65)
65
2 ETMA
88.113
102.128
102.128
116.144
116.144
130.160
130.160
130.160
130.160
130.160
144.175
144.175
144.175
144.175
158.191
158.191
158.191
158.191
172.207
172.207
186.222
186.222
186.222
200.238
200.238
214.253
214.253
228.269
242.285
242.285
298.347
354.410
3 DEDMA
4 TMIPA
5 TEMA
15
ND
11
ND
6.90 (6.8)
ND
3. Results and discussion
6 EDMIPA
7 DMDPA
8 DEMPA
9 TEA
ND
53
The external binding data, reported as KD (mM) for
the cloned Kv1.1, Kv2.1, and Kv3.1 channels, are
presented in increasing amu, Table 2 [10,11]. The
ab initio data, computed at the Hartree±Fock (HF)/
3-21G basis set, are summarized in Table 3. The
isodensity PCM (IPCM) calculations were performed
with Gaussian 98 and are summarized in Table 4.
The Onsager method1 uses a ®xed spherical cavity
to represent the QA ion. The IPCM method has an
iterative process where the cavity shape may change
during the SCF cycles. Because the molecular shape
can change, the isodensity surface has an intrinsic
shape for the cavity.
13
9.2
1.80
0.085
ND
0.48
ND
ND
0.64
ND
ND
ND
0.60
1.90
ND
ND
ND
ND
ND
36
10 DMPIPA
11 DMDIPA
12 TEPA
ND
0.082
ND
13 BDMIPA
14 EDIPMA
15 BDEMA
16 DEDPA
17 BTEA
18 DEDIPA
19 MTPA
20 ETPA
ND
ND
0.093
0.38
ND
ND
ND
ND
ND
7.9
ND
ND
21 EDIPPA
22 HexTEA
23 TPA
ND
1.74
1.34 (1.2)
ND
Plotting the Jarolimek et al. Kv3.1, Kv2.1, Kv1.1 KD data
against the computed COSMO VDW revealed a
39
24 BEDIPA
25 TBMA
26 BTPA
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.12
ND
ND
0.04
0.9
Ê 2
correlation (Fig. 1). As COSMO VDW (A )
Ê 2
ND
approached 80 A , the experimental KD sharply
ND
27 OctTEA
28 TBEA
29 TBPA
0.63
ND
decreased to give high af®nity. Even though the
experimental data does not allow a complete compar-
ison, the same initial trend is observed for all three of
these voltage-gated K1 channels.
ND
30 DecTEA
31 TBA
0.19
10
Because potassium channels have four a-subunits,
and because voltage-gated subfamilies and subfamily
homologs have sequence modi®cations at or near the
pore region, affords us an analogy to quadrupole mass
spectrometry [18]. Ion focusing occurs by means of a
mass separation and a function of vibrations within a
high-frequency electric quadrupole ®eld. In quadru-
pole mass spectrometry, the ion beam is directed in a
longitudinal direction residing between four parallel
metal rods. There is a voltage, a direct current (dc), at
32 TPentA
33 THexA
0.01
ND
0.04
0.0015
reaction vessel was cooled to room temperature and
the solid ®ltered. The ®ltrate was reduced to one-third
volume under reduced pressure, the solid was ®ltered,
and the red liquid ®ltrate vacuum distilled (0.8±
0.9 mmHg; 78±848C) to give 46.4 g, 70% DIEtOEt.
DIEtOEt (4.1 g, 12.7 mmol) was diluted in 2-buta-
none (20 ml) and triethylamine (4.0 ml) added drop
wise. A re¯ux condenser was attached and heated to
re¯ux (3.5 h). The reaction was cooled in an ice bath
and the product ®ltered. The product recrystallized
from ethyl acetate, ®ltered, and dried in vacuo to
give BisTEAEtOEt in 80%.
1
When Onsager model calculations are carried out on systems
having dipole moments of zero, these systems will not exhibit
solvent effects; thereby giving the same result as that for the gas
phase calculation (See Ref. 38, pp. 237±238). In Table 3, so that the
QA ions could emphatically be evaluated, the alkyl quaternary
ammonium ions were computed using the C1 point group.

M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
71
Table 3
Ab initio data at the HF/3-21G basis set2
QA ion
Mopac
amu
Mopac
COSMO
Gaussian
dipole
Onsager
dipole
Mopac
dipole
Mopac
ESP(N)
Gaussian
ESP(N)
1 TMA
74.097
88.113
56.60
62.63
68.51
67.10
73.71
72.14
83.75
81.03
77.96
79.59
76.80
85.81
87.17
81.60
88.59
93.25
93.49
86.31
94.66
100.62
93.17
108.35
107.77
100.39
116.94
115.60
123.66
122.70
130.18
138.94
134.80
168.77
199.22
0.002
1.734
1.405
2.415
1.550
2.710
1.339
3.008
1.844
3.771
1.446
2.274
5.669
1.274
5.014
1.055
2.927
1.186
4.562
1.472
1.626
9.462
2.132
2.645
7.020
1.565
14.336
6.211
4.730
19.512
4.170
2.607
3.327
0.000
1.326
1.033
1.887
1.070
2.016
1.210
2.233
1.269
2.787
1.107
1.697
4.415
0.949
3.998
0.848
2.478
0.810
3.118
0.959
1.141
7.889
1.305
2.071
5.066
1.236
12.523
4.530
3.441
17.543
2.777
1.708
2.355
0.000
1.484
1.163
2.229
1.214
2.354
1.380
2.580
1.457
3.224
1.281
2.046
5.105
1.084
4.774
0.999
2.904
0.934
3.588
1.147
1.323
9.362
1.536
2.570
5.814
1.441
14.96
5.302
4.127
20.62
3.227
1.977
2.737
0.3866
0.3535
0.3575
0.3055
0.3271
0.2167
0.2324
0.2208
0.3212
0.2032
0.2036
0.2714
0.3485
0.1611
0.3065
0.1174
0.3261
0.1538
0.2392
0.1144
0.2197
0.2550
0.0972
0.1244
0.7266
0.0043
0.3050
0.8023
0.6449
0.3031
0.4320
0.4836
0.1912
217.9153
217.9245
217.9332
217.9319
217.9414
217.9398
217.9394
217.9440
217.9486
217.9426
217.9462
217.9514
217.9441
217.9535
217.9454
217.9541
217.9529
217.9589
217.9491
217.9563
217.9604
217.9538
217.9589
217.9620
217.9531
217.9605
217.9542
217.9599
217.9626
217.9544
217.9645
217.9671
217.9687
2 ETMA
3 DEDMA
4 TMIPA
5 TEMA
102.128
102.128
116.144
116.144
130.160
130.160
130.160
130.160
130.160
144.175
144.175
144.175
144.175
158.191
158.191
158.191
158.191
172.207
172.207
186.222
186.222
186.222
200.238
200.238
214.253
214.253
228.269
242.285
242.285
298.347
354.410
6 EDMIPA
7 DMDPA
8 DEMPA
9 TEA
10 DMPIPA
11 DMDIPA
12 TEPA
13 BDMIPA
14 EDIPMA
15 BDEMA
16 DEDPA
17 BTEA
18 DEDIPA
19 MTPA
20 ETPA
21 EDIPPA
22 HexTEA
23 TPA
24 BEDIPA
25 TBMA
26 BTPA
27 OctTEA
28 TBEA
29 TBPA
30 DecTEA
31 TBA
32 TPentA
33 THexA
34 DetMTolA
178.160
192.175
206.191
220.207
222.186
234.222
236.201
142.147
97.32
102.20
114.44
119.87
120.99
123.98
125.43
75.16
6.117
5.687
4.429
4.006
5.314
5.019
0.4600
0.4481
0.1899
0.1099
0.2080
0.2651
0.2413
0.2387
0.2538
0.2872
0.3311
0.5819
0.8174
0.4509
0.3970
0.2592
0.3350
0.3338
217.9509
217.9572
217.9468
217.9535
217.9415
217.9642
217.9488
217.7709
217.7709
217.7887
217.7887
217.3371
217.8341
217.8757
217.8757
217.8401
217.8268
217.8313
35 TETTolA
36 3DEMAPrPh
37 3TEAPrPh
12.521
12.189
15.965
3.596
10.439
9.982
13.096
12.320
17.703
2.849
38 3PhenODEMAPr
39 TPATolA
13.716
2.303
40 3PhenOTEAPr
41 DMDABCO
15.381
0.002
13.035
0.000
15.858
0.001
42 DEDABCO
43 BisTEAEt
170.178
230.272
230.272
510.463
87.03
118.06
141.48
229.33
1.089
1.850
1.750
7.317
0.684
1.209
1.303
5.887
0.801
1.441
1.530
7.267
44 BisTEAEtOEt
45 Gallamine

72
M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
Table 4
Ab initio IPCM calculations
Cation
RHF/3-21G IPCM energya
Solvation energy (kcal/mol)b
Gas phase (hartree)
Solvated (hartree)
Lithium
Sodium
27.18709449495
2160.673746247
2596.006592796
2211.526989298
2250.347416928
2289.167010349
2289.165821293
2327.981632645
2327.981112056
2366.807726051
2366.801847776
2366.792193539
2366.801679414
2366.801322713
2405.620632008
2405.621501226
2405.610158258
2405.621653102
2444.440914695
2444.432334435
2444.427071094
2444.435579598
2483.255304252
2483.238879181
2522.071302026
2522.073767020
2522.055961237
2560.894731654
2560.901295362
2599.709759275
2599.712593174
2638.532779680
2677.348080752
2677.345081698
2832.638482803
2987.915751167
2517.430408978
2556.247474925
2595.068298086
2633.886224901
2669.506179935
2672.708206409
2708.319413298
2419.442372784
2497.090415478
2654.681986427
2806.803740473
21549.46036518
27.47083168467
2160.893555923
2596.161129575
2211.613440338
2250.430209574
2289.246372742
2289.246017223
2328.059337940
2328.058705627
2366.882587043
2366.877628047
2366.867256009
2366.876402761
2366.876588394
2405.693345797
2405.694505871
2405.683219118
2405.696608644
2444.512130613
2444.505228582
2444.498228031
2444.506148320
2483.323760779
2483.307997250
2522.143588864
2522.143491205
2522.123235535
2560.966316477
2560.969264476
2599.782153699
2599.781741318
2638.597059246
2677.421567633
2677.408620325
2832.699066235
2987.983955721
2517.507465135
2556.321452017
2595.153797892
2633.989242981
2669.589906580
2672.780656022
2708.416948819
2419.736548332
2497.372254697
2654.930593484
2807.018644548
21549.88245380
2178.05
2137.93
296.970
254.249
251.953
249.801
250.324
248.761
248.691
246.976
247.553
247.102
246.890
247.230
245.629
245.811
245.846
247.035
244.689
245.742
244.652
244.283
242.957
243.372
245.361
243.753
242.215
244.920
242.651
245.428
243.391
240.336
246.114
239.871
238.017
242.799
248.353
246.421
253.652
264.645
252.539
245.459
261.205
2184.598
2176.857
2156.003
2134.854
2264.865
Potassium
TMA
ETMA
DEDMA
TMIPA
TEMA
EDMIPA
DMDPA
DEMPA
TEA
DMPIPA
DMDIPA
TEPA
BDMIPA
EDIPMA
BDEMA
DEDPA
BTEA
DEDIPA
MTPA
ETPA
EDIPPA
HexTEA
TPA
BEDIPA
TBMA
BTPA
OctTEA
TBEA
TBPA
DecTEA
TBA
TPentA
THexA
DEtMTolA
TETTolA
3DEMAPrPh
3TEAPrPh
3PhenO-DEMAPr
TPATolA
3PhenO-TEAPr
DMDABCO
DEDABCO
BisTEAEt
BisTEAEtOEt
Gallamine
a
Energy in hartree units; 1 hartree ˆ 4.3597482 £ 10218J; 1 hartree ˆ 627.5095 kcal/mol ˆ 27.2116 eV.
DE ˆ (solvated energy 2 gas phase energy) £ 627.5095; DE are reported in kcal/mol.
b

M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
73
opposite rods and their phases are shifted 1808 from
one another and are superimposed by a high-
frequency ®eld. In the case of a prede®ned current,
which can be changed mechanically, only particles of
a particular mass/charge (m/z) ratio will reach the
outlet slot. Hence, in the case of a K1 channel with
four a-subunits and a voltage difference between the
extra- and intracellular compartments, modifying
amino acids near the channel pore will make subtle
changes to pore size and to the electrostatic focusing
power governed by a m/z relationship.
Structurally examining these three voltage-gated
pore structures, position 449 preserves the alcohol
functionality but the size and pKa are different
(Scheme 4). A phenol has a typical pKa of 10 and
sp3-hybridized alcohols usually have pKas between
15 and 19. Kv3.1 Y449 and Kv2.1 Y449 generate an
electronically rich, resonance stabilized, anion that
increases the electrostatic attractive force for a cation,
Coulomb's law [19]. And even though Kv3.1 and Kv2.1
have Y449, they are in different chemical environ-
ments. The incipient conjugative bases, having a
formal charge of 21, can be stabilized through
different
intramolecular
hydrogen
bonding.
Compared to threonine at position 449, mutagenesis
studies show phenylalanine to have a strong af®nity
enhancement. Phenylalanine demonstrates that an
aromatic ring at the outer proximity of the channel
pore provides a major attractive affect. The amino
acids 451±458 differ but play minor roles. Kv3.1
451±458 contain polar amino acids and have non-
bonding electrons that may stabilize a cation. By
comparison, Kv1.1 contains more non-polar amino
acids and fewer lone pair of electrons. In addition,
Kv1.1 (456) and Kv2.1 (456) are lysines (K) whose 11
charge is expected to cause a repulsive effect with a
cation. Kv1.1 K456 has an adjacent tryptophan (W) 454
whose aromatic p system will stabilize the 11 charge
through ®eld-effects, whereas Kv2.1 is aliphatic leucine
454 and will not effectively stabilize the 11 charge
[20].
The computed mopac dipole and electrostatic
potentials (ESP) have been plotted against the Jaro-
limek et al. Kv3.1, Kv2.1, and Kv1.1 KDs data (Fig. 2).
Once again, insuf®cient experimental data prevents
an absolute comparison but some trends have
emerged. It appears as though Kv1.1 was more sensitive
to QA size change; most likely, an anomaly re¯ecting
Fig. 1. Jarolimek et al. Kv3.1 and Kv1.1 KDs versus alkyl QA ion amu
and COSMO VDW.

74
M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
Scheme 4.
the molecular (i.e. symmetry, dipole moment, proxi-
mity effects, and the relative degrees of freedom) and
solvated differences from elongating the QA ion chain
unidirectionally rather than symmetrically [21±37].
Kv3.1, Kv2.1, and Kv1.1 show evidence of ESP (N) sensi-
tivity around 0.38. Whether all of these channels have
a similar ESP (N) pattern between 0.4 and 0.8 remains
to be determined.
The isodensity PCM (IPCM) solvation method
computes the potential energies for both the gas and
solvated phases [12,13]. The energy differences for
the QA ions have been converted from units of
Hartree into kcal/mol [38]. The DEs of solvation
were plotted against COSMO SAS and adjusted to
include the K1 and Na1 COSMO SAS values (Fig.
3). By extrapolating COSMO values from Fig. 1, we

M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
75
Fig. 2. Kv3.1, Kv2.1, Kv1.1 versus Mopac dipole and ESP on the alkyl QA ions.

76
M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
Fig. 3. Alkyl QA ion COSMO versus calculated DGsolvation
.
can estimate KDs for K1 and Na1 ions. In all cases,
Kv3.1, Kv2.1, and Kv1.1, a high KD for Na1 and a small,
high af®nity, KD for K1 ion is predicted.
generates a moderate but quite crucial ®eld effect
that affords a KDs between 20 and 30 mM.
On the basis of these results, we can predict the KD
values for Shaker T449Y. The three benzyl derived
QA ions (DEtMTolA, TETolA, and TPATolA) will
have a combination of strong and weak ®eld effects.
These different ®eld effects will affect the m/z ratio
(Table 5). These three aromatic QA ions will suffer
strong ®eld effects, the 11 charge is stabilized by the
phenyl p-system. The stabilization decreases the
overall charge, the `effective charge' (Zeff), to
give a larger (m/z) ratio than what one may
expect. A larger m/z ratio will lead to a higher
KD (65±85 mM). The expected weak ®eld effect
will result from an increase in mass, giving us a
4. Conclusion
This study intertwines physical bio-organic chem-
istry and its relationship to molecules in motion (i.e.
Stokes±Einstein equation, Nernst±Einstein equation,
etc.) [39,40]. Even though there may be higher extra-
cellular concentrations of Na1 than K1, the presented
theoretical and experimental model predicts potassi-
um ion selectivity through a `bio-cationic focusing
mechanism', much like that associated with quad-
rupole mass spectrometry. Nature has created the
potassium channel where it is focused for a m/z of a
K1 ion, not a Na1 ion. In addition, the computed
DEsolvation (Na1 2 K1) was 240.96 kcal/mol. An
absolute DE of 41 kcal/mol encompasses the
Nernst±Einstein relationship and how it predicts that
a hydrated potassium ion, at the same physiological
kinetic energy, will diffuse at a rate signi®cantly
faster than a hydrated sodium ion. Modifying the
amino acids near the pore region will result in
subtle changes that alter size (proximity effects)
and the electronic environment that ultimately
control the channels focusing power. Field effects
predicted
TETolA . DEtMTolA. However, the strong ®eld
KD
order
of
TPATolA .
effect will diminish upon moving the phenyl ring
Ê
by 4 or so A. Adding distance between the ammo-
nium ion and phenyl ring, accomplished by adding
±(CH2)2±, will change the electronic distribution
throughout the p and s-framework. From the
decreased shielding thus formed, 3TEAPrPh and
3PhenOTEAPr will give KDs of about 3. Changing
one of the ethyl groups to a methyl group,
3DEMAPrPh and 3PhenODEMAPr, respectively,

M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
77
Table 5
Shaker T449Y predicted KD values
QA ion
z
amu
COSMO
m/z
Shaker T449Y
Est.KD
Strong ®eld effect
DetMTolA
TETTolA
TPATolA
1
1
1
178.160
192.175
234.222
97.32
102.20
123.98
97.32
102.20
123.98
65
70
85
3DEMAPrPh
3TEAPrPh
1
1
206.191
220.207
114.44
119.87
114.44
119.87
20±30
,3
Weak ®eld effect
3PhenODEMAPr
3PhenOTEAPr
1
1
222.186
236.201
120.99
125.43
120.99
125.43
20±30
,3
Moderate to strong ®eld effect
DMDABCO
DEDABCO
BisTEAEt
2
2
2
2
3
142.15
170.78
230.27
246.27
510.27
75.16
87.03
71.07
85.09
100
80
118.06
141.48
229.33
115.13
123.13
170.16
20±40
5±10
,1
BisTEAEtoEt
Gallamine
[2] L.P. Hammett, Physical Organic Chemistry, McGraw-Hill,
New York, 1940.
will in¯uence the mass/charge ratio (m/z; amu/z).
For example, even though Gallamine (an aromatic
and tri-cation QA ion) and DEtMTolA (an
aromatic and mono-cationic QA ion) have similar
computed amu/z ratios, their KDs will be very
different. DEtMTolA suffers from a strong ®eld
effect, causing the `effective charge' Zeff to be
! 1 1 and affords a high KD, whereas Gallamine
will have a mild ®eld effect with a Zeff < 13
resulting in high af®nity, low KD (,1).
[3] J.D. McKinney, A. Richard, C. Waller, M.C. Newman, F.
Gerberick, Toxicol. Sci. 56 (2000) 8±17.
[4] B.H. Besler, K.M. Merz Jr., P.A. Kollman, J. Comput. Chem.
11 (1990) 431±439.
[5] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A.
Robb, J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgomery,
R.E. Stratmann, J.C. Burant, S. Dapprich, J.M. Millam, A.D.
Daniels, K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi,
V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C.
Adamo, S. Clifford, J. Ochterski, G.A. Petersson, P.Y. Ayala, Q.
Cui, K. Morokuma, D.K. Malick, A.D. Rabuck, K. Raghava-
chari, J.B. Foresman, J. Cioslowski, J.V. Ortiz, B.B. Stefanov,
G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts,
R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A.
Nanayakkara, C. Gonzalez, M. Challacombe, P.M.W. Gill, B.G.
Johnson, W. Chen, M.W. Wong, J.L. Andres, M. Head-Gordon,
E.S. Replogle, J.A. Pople, Gaussian 98 (Revision A.7),
Gaussian, Inc., Pittsburgh PA, 1998.
Acknowledgements
M.F.W thanks Professor M.W. Anders at the
University of Rochester Medical Center for the
research opportunities and to the Department of
Chemistry at the University of Rochester for the
access to their facilities.
[6] Y. Kurachi, L. Yeh Jan, M. Lazdunski (Eds.), Current Topics
in Membranes: Potassium Ion Channels/Molecular Structure,
Function, and Diseases, vol. 46, Academic Press, New York,
1999.
References
[7] J.O. Dolly, D.N. Parcej, J. Bioengng Biomem. 28 (1996) 231±
253.
[1] C. Hansch, T. Fujita, J. Am. Chem. Soc. 86 (1964) 1616.
[8] D.A. Doyle, J.M. Cabral, R.A. Pfuetzner, A. Kuo, J.M. Gulbis,

78
M.F. Wempe / Journal of Molecular Structure 562 (2001) 63±78
S.L. Cohen, B.T. Chait, R. Mackinnon, Science 280 (1998)
69±76.
[24] S. Das, D.K. Hazra, Indian J. Chem. Sec. A 27 (1988) 1073±
1079.
[9] R. MacKinnon, C. Miller, Science 245 (1989) 1382±1385.
[10] W. Jarolimek, K.V. Soman, M. Alam, A.M. Brown, P¯uÈgers
Arch-Eur. J. Physiol. 430 (1995) 672±681.
[25] G.H. Haggis, J.B. Hasted, T.J. Buchanan, J. Chem. Phys. 20
(1952) 1452±1465.
[26] R.D. Singh, R. Gopal, Bull. Chem. Soc. Jpn 45 (1972) 2088±
2090.
[11] W. Jarolimek, K.V. Soman, M. Alam, A.M. Brown, Mol.
Pharm. 49 (1996) 165±171.
[27] C.H. Hale, T. De Vries, J. Am. Chem. Soc. 70 (1948) 2473±
24375.
[12] J. Tomasi, M. Persico, Chem. Rev. 94 (1994) 2027±2094.
[13] D. Sitkoff, K.A. Sharp, B. Honig, J. Phys. Chem. 98 (1994)
1978±1988.
[28] T.H. Mead, O.L. Hughes, H. Hartley, J. Chem. Soc. (1933)
1207±1214.
[14] J. March, Advanced Organic Chemistry/Reactions, Mechan-
isms, Structure, 4th ed., Wiley, New York, 1992 (pp. 411±
413).
[29] D. Bax, C.L. de Ligny, A.G. Remijnse, Recl. Trav. Chim.-
Pays-Bas. 91 (1972) 965±988.
[30] S.R.C. Hughes, D.H. Price, J. Chem. Soc. A (1967) 1093±
1096.
[15] J. March, Advanced Organic Chemistry/Reactions,
Mechanisms, Structure, 4th ed., Wiley, New York, 1992
(pp. 430±431).
[31] J. Barthel, L. Iberl, J. Rossmaier, H.J. Gores, B. Kaukal, J.
Solut. Chem. 19 (1990) 321±337.
[16] L. Peyron, J. Peyron, Bull. Soc. Chim. Fr. (1950) 1062±1064.
[17] S. Oae, B. Hovarth, C. Zalut, R. Harris, J. Org. Chem. 24
(1959) 1348±1349.
[32] P.K. Muhuri, D.K. Hazra, J. Chem. Soc. Faraday Trans. 87
(1991) 3511±3513.
[33] K. Bauge, J.W. Smith, J. Chem. Soc. (1964) 4244±4247.
[34] J.S. Jha, S. Singh, R. Gopal, Bull. Chem. Soc. Jpn 48 (1975)
2782±2784.
[18] R.M. Silverstein, G.C. Bassler, T.C. Morrill, Spectrometric
Identi®cation of Organic Compounds, 5th ed., Wiley, New
York, 1991 (chap. 2).
[35] R.L. Kay, D.F. Evans, J. Phys. Chem. 70 (1966) 2325±2335.
[36] E.A. Richardson, K.H. Stern, J. Am. Chem. Soc. 82 (1960)
1296±1302.
[19] T.L. Brown, H.E. LeMay Jr., B.E. Bursten, Chemistry Ð The
Central Science, 7th ed., Prentice Hall, Upper Saddle River,
NJ, 1997 (p. 43).
[37] O. Exner, Dipole Moments in Organic Chemistry, Georg
Thiene, Stuttgart, 1975.
[20] J. March, Advanced Organic Chemistry/Reactions, Mechan-
isms, Structure, 4th ed., Wiley, New York, 1992 (pp. 12±19).
[21] R. Gopal, R. Kumar, D.K. Agarwal, J. Indian Chem. Sec. A 14
(1976) 803±804.
[38] J.B. Foesman, áleen Frisch (Eds.), Exploring Chemistry with
Electronic Structure Methods 2nd. ed., Gaussian Inc, Pitts-
burgh, PA, 1995.
[22] R. Gopal, R. Patak, J. Indian Chem. Sec. A 14 (1976) 804±
805.
[39] P.W. Atkins, Physical Chemistry, 4th ed., Freeman, New
York, 1990 (chap. 25).
[23] D.K. Agarwal, K. Singh, J. Indian Chem. Soc. 53 (1976) 482±
485.
[40] B. Hille, Ionic Channels of Excitable Membranes, Sinauer
Associates, Sunderland, MA, 1984 (chap. 7).