Porphyrin derivatives as potent and selective blockers of neuronal Kv1 channels

Article abstract of DOI:10.1039/c4cc05639f

Selective inhibitors of voltage-activated K⁺ channels are needed for the treatment of multiple sclerosis. In this work it was discovered that porphyrins bearing 2-4 carbon alkyl ammonium side chains predominantly blocked the Kv1.1 current whilst Kv1.2 was susceptible to a porphyrin bearing polyamine side chains.

Full text of DOI:10.1039/c4cc05639f

View Article Online
View Journal
ChemComm
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: K. Nolan, A. Al-
Sabi, D. Daly, O. Dolly and G. Kinsella, Chem. Commun., 2014, DOI: 10.1039/C4CC05639F.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/chemcomm

Page 1 of 4
ChemComm
View Article Online
DOI: 10.1039/C4CC05639F
Chemical communications
RSCPublishing
COMMUNICATION
Porphyrin derivatives as potent and selective blockers
of neuronal Kv1 channels
Cite this: DOI: 10.1039/x0xx00000x
D. Dalya,d, A. AlꢀSabib,d, G. K. Kinsellac, K. Nolana*, and J. O. Dollyb*
Received 00th July 2014,
Accepted 00th XXXX 2014
DOI: 10.1039/x0xx00000x
www.rsc.org/
interest was to determine the effects on Kv1 selectivity of porphyrin
substituent chain length (10ꢀ13), amide linkage (14), primary amine
groups (17, 18) and multiple charged groups (15, 16). Four of the
compounds purified were shown electrophysiologically to inhibit
Kv1.1, 1.2 or both homoꢀtetrameric channels expressed in
mammalian cells.
Selective inhibitors of voltage-activated K+ channels are
needed for the treaetment of multiple sclerosis. In this work it
was discovered that porphyrins bearing 2-4 carbon alkyl
ammonium side chains predominantly blocked the Kv1.1
current whilest Kv1.2 was susceptible to a porphyrin bearing
polyamine side chains.
ROC
COR'
COR
R'OC
Voltageꢀgated K+ channels (Kv1) control neuron excitability and
synaptic transmission; their alteration (e.g. mutation or expression
levels) underlies human diseases.2 Kv1 channels are membraneꢀ
spanning oligomeric sialoglycoproteins 3,4 (Mr ~400 k) consisting of
4 poreꢀforming α subunits and 4 cytoplasmicallyꢀassociated auxiliary
β proteins.5,6 When expressed in vitro, each of the major α subunit
genes [Kv1.1ꢀ1.7] yields a homoꢀtetrameric channel with distinct
biophysical and pharmacological profiles.7 Kv1 members, exposed
on demyelinated axons in patients suffering from multiple sclerosis
(MS), contribute to abnormal propagation of nerve signals with
resultant debilitating muscle weakness.8 Although aminopyridines
can inhibit these culprit channels, such therapy results in serious offꢀ
N
N
N
N
EDCI/NHS
DMF
4M HCl
Dioxane
HN
HN
NH
NH
TCPP
ROC
COR'
COR
R'OC
H2Cl
N
Boc
N
NH3Cl
15. R'
=
HN
NH3Cl
NH3Cl
NHBoc
10. R' =
6. R = HN
NHBoc
HN
HN
1. R = HN
HN
3. R = HN
N
N
11. R'
12. R'
=
=
NH3Cl 16. R' = ClH3
N
NHBoc
2. R =
BocHN
HN
NHBoc
7. R
=
HN
HN
NH3Cl
17. R' =
HN
N(CH3)2HCl
NHBoc
NHBoc
N(CH3
)
2
8. R =
2
2
4
13. R' =
14. R' =
NH3Cl
4. R =
5. R =
HN
18. R' =
HN
N(CH3)2HCl
4
9. R
=
HN
N(CH3)
2
NH3Cl
NHBoc
O
O
target effects including seizures, due to blockade of other K+ channel Scheme 1. Synthesis of porphyrins 10 – 18 from precursor TCPP.
types.8 Recently, Kv1.1ꢀ and 1.2ꢀcontaining channels were found to
be abnormally expressed in optic nerve demyelinated axons from a
Porphyrins 10ꢀ18 were prepared by the introduction of mono NꢀBoc
cuprizoneꢀinduced mouse model that resembles MS. Thus, finding
alkyl diamines using carbodiimide coupling as outlined in Scheme 1.
smaller extracellular inhibitors for such K+ channels should aid
Porphyrins 1ꢀ4 were prepared by treating TCPP with EDCI and the
respective mono NꢀBocꢀprotected alkyldiamines in the presence Nꢀ
progress towards developing new drugs to alloriate MSꢀrelated
hydroxysuccinimide.14 Final yields of the Boc protected prophyrins
symptoms. It is notable that symmetricallyꢀsubstituted porphyrins
bearing alkyl amino groups (cationicꢀcharged at physiological pH)
after silica gel chromatography ranged from 14ꢀ70 %. Porphyrin 5
was prepared using a modified Steglich esterification by reacting Nꢀ
tightly bind a KcsAꢀKv1.3 channel.1,10 This involves docking into a
distinctive part of the pore (e.g. outer turret region) of the Kv1.3
Boc aminoethanol with TCPP in the presence of EDCI and 4ꢀ
channel, as determined by solid state NMR.10 Therefore, porphyrins
dimethylaminopyridine (DMAP) as catalyst in DMF for 24 hrs.; it
can provide the 4ꢀfold symmetry required for interaction with Kv1
was isolated in 80% yield after silica gel chromatography.
channels, but a tetraꢀphenylꢀporphyrin with alkyl ammonium side
Preparation of porphyrins 6 and 7 was complicated by the need to
selectively protect primary over secondary amines. In the case of 6,
the required diprotected triamine was prepared according to the
chains blocks several different Kv1 members.11 Hence, the aim of
the present study was to carryout a structure activity study with
procedure of Leonor et al.15 and then treated with TCPP using the
various substituted prophyrins to determine their selectivity for
subtypes of Kv1 channels. This task was complemented using a
EDCI procedure described above; this resulted in a yield of 35%
model of the rat homologous Kv1.1 channel, derived from the
after silica gel chromatography. For porphyrin 7, the required
crystallographic structure of Kv1.2,12 to perform molecular
diprotected triamine was prepared16 and coupled to TCPP using the
modelling of its interaction with the porphyrins.. The prophyrins
examined in this (10ꢀ18) study are shown in Scheme 1. Of particular
EDCI protocol to give
7
in 66% yield after silica gel
This journal is © The Royal Society of Chemistry 2014
Chem.comm., 2014, 00, 1-3 | 1

ChemComm
Page 2 of 4
View Article Online
COMMUNICATION
Chemical communications
DOI: 10.1039/C4CC05639F
chromatography. The same EDCI coupling was employed for
making porphyrins 8 and 9 from the corresponding amines (N,Nꢀ
ethylenediamine and N,Nꢀdimethylꢀ1,3ꢀdiaminopropane) in 62% and
59% yield, respectively. Conversion of all of the Boc protected
porphyrins 1ꢀ7 to their respective amine hydrochloride salts, 10ꢀ16,
was achieved using 4M HCl in dioxane (Scheme 1); this afforded
pure products in quantitative yields after washing. Likewise, the
amine hydrochloride salts of 8 and 9 were generated to give 17 and
18 in quantitative yields. Compounds 10ꢀ18 were characterised by
1H NMR and highꢀresolution MALDIꢀTOF mass spectrometry.†
The reactivities of 10ꢀ18 were tested on homomeric (Kv1.1, 1.2,
1.3, 1.4 and 1.6) channels stably expressed in HEKꢀ293 cells, using
wholeꢀcell patch clamp recording.† Alignments of the pore amino
acid sequences of 5 neuronal Kv1 channels (Fig. 1) revealed mainly
conserved residues except at the outer and inner turret regions.17,18
Hence, it was anticipated that these Kv1 channels might show
Fig. 1 Alignment of the pore sequence in neuronal Kv1 subunits.
Letters highlight the residue differences at the outer / inner turret
and pore helix between each channel.
Fig. 2 Differential inhibition by compounds 10-12 and 15 of currents
mediated by various Kv1 channels expressed in HEK-293 cells. (A)
Representative current traces from Kv1.1 and Kv1.2 homomeric
channels in the absence (black) and presence of 10 ꢀM compound
10 or 15, generated by a step voltage from the holding potential to
20 mV, as shown. (B) Summary of the dissimilar pharmacological
profiles of five Kv1 channels shown for six compounds.
distinct susceptibilities to the aboveꢀnoted porphyrins bearing
different alkyꢀammonium side chains. Representative K+ current
traces from expressed Kv1.1 channel were blocked by 10 µM
compound 10 but not by the same concentration of 15, while the
opposite effect was observed for Kv1.2 channel (Fig. 2A). Only
compounds 10ꢀ12 and 15 inhibited certain Kv1 channels at 10 µM,
with 13 and 14 (Fig. 2B) and 16ꢀ18 (not shown) proving ineffective.
Notably, 10 predominantly blocks Kv1.1 and to a lesser extent
Kv1.4 channel, while 11 preferentially inhibits Kv1.1 and 1.2 K+
currents. On the other hand, the inhibitory effect of 12 was much
more pronounced on Kv1.1 than Kv1.2, 1.3 or 1.4, with Kv1.6 being
virtually nonꢀsusceptible (Fig. 2B). It is notable that porphyrins
containing 2, 3 or 4 carbon (10ꢀ12) displayed inhibition whereas 13
with 6 carbons was devoid of activity (Fig. 2B). Also, the
importance of an amide linking the alkylamine substituents to the
porphyrin became evident from 10 being active but not 14, in which
the amide was replaced with an ester. Another significant finding is
the loss of activity when the primary amines were replaced with
tertiary amines, revealed on relating the inhibition by compound 10
and 11 (Fig. 2B) to the inactive 17 and 18 (not shown). This lack of
blockade might be due to the absence of hydrogen bond (HB) donors
or size constraints. Moreover, increasing the number of terminal
amine groups, from one as in the case of 10 to two in 16, eliminated
activity, although this could also arise from the lack of a HB donor at
the amide linkage.
Interestingly, inserting an amino group instead of a methylene
moiety into 13, as is the case for 15, restores inhibition. This subtle
modification of placing both a charged group and HB donor in the
centre of the chain proved to be of great significance, because 15
almost exclusively inhibits Kv1.2 (Fig. 2A, B).
To quantify the potencies of compounds 10, 12 and 15 for
inhibiting Kv1.1, 1.2 and 1.4 channels, their IC50 values were
calculated from doseꢀresponse curves, obtained from Qpatch
recordings (Fig. 3). For Kv1.1 and 1.4, 10 gave IC50s of 13 ± 0.4 ꢀM
and 21 ± 1 ꢀM [n = 7], respectively, with Kv1.2 exhibiting
negligible sensitivity. On the other hand, Kv1.1 was somewhat more
Fig. 3 Distinct inhibitory potencies of 10, 12 and 15 on Kv1.1, 1.2
and 1.4 channels. Dose-response curves, obtained from Qpatch
recordings, for Kv1.1 (○), Kv1.4 (□) and Kv1.2 (●). Some of the error
bars fall within the data points. The IC50 values are given in the text.
2 | Chem. comm., 2014, 00, 1-3
This journal is © The Royal Society of Chemistry 2014

Page 3 of 4
Journal Name
ChemComm
View Article Online
DOI: 1C0.O10M39M/CU4CNCI0C5A6T39IOF N
susceptible to 12 (IC50 = 9 ± 1 ꢀM [n = 7]) but Kv1.2 and 1.4
displayed similar albeit lower sensitivities (IC50 = 21 ± 2 ꢀM and 19
± 3 ꢀM [n = 5]) (Fig. 3B). However, compound 15 is selective for
Kv1.2 having IC50= 12 ± 1 ꢀM [n = 5] with greatly diminished
inhibition of Kv1.4 > 1.1 (Fig. 3C). Notice the steeper concentration
dependence for 10 and 15 than 12, reflected by the observed Hill
coefficients of 1.9, 2.5, and 1.5. These values indicate cooperative
interactions of both compounds with Kv1.1 and 1.2 channels similar
to the behaviour reported for 10 with Kv1.3.11 In the latter study
which used Xenopus expression system, compound 10 inhibited
Kv1.1 and 1.6 with nM IC50 values, and displayed lower potency for
Kv1.2, 1.3, 1.4 and 1.5. Our stably transfected mammalian cells
tolerated >1 ꢀM of 10 (a concentration that made Xenopus cells
leaky)
and
showed
distinct
pharmacological
profiles.
Notwithstanding the lower potencies of the compounds tested herein,
these exhibited greater selectivity for Kv1.1 (as in case of 10 and
12), Kv1.2 (15) or both (11) with little blockade of Kv1.3 or 1.6
channels. These differences in potency and selectivity are probably
attributable to the use of amphibian and humanꢀderived cells. Our
study employed the latter because of the focus on generating Kv1
channels relevant to those over expressed in demyelinated neurons,
specifically Kv1.1 and to a lesser extent 1.2, as potential targets for
MS.9
Fig. 4 Porphyrin derivatives docking into a Kv1.1 channel. (A) Side
(left) and top (right) views of a representative compound 12,
(yellow), interacting with the Kv1.1 channel structure (derived from
Kv1.2), with its four α subunits represented in red (chain A),
blue (chain B), green (chain C) and orange (chain D). (B) A closer
side-view of Kv1.1 subunits interacting via residues from the outer
turret region and those lining the deep pore, with the side chains of
Some of the compounds tested also acted as gating modifiers. An compound 12. For clarity, the interacting residues of chain A (red)
example is shown in Fig. 2A where inhibition of Kv1.1 current by
10 ꢀM 10 is associated with slowing of its activation time course.
The time constant (τ) calculated by fitting the current traces at +20
mV potential with a single exponential function† indicated a ~5ꢀfold
difference [τcontrol = 1.74 ± 0.4 ms vs. τcompound 10 = 8.4 ± 2.3 ms, p =
0.029, n = 4, each]. Alteration of Kv1 channel gating by 10 accords
with that reported recently using Kv1 channels expressed in Xenopus
oocytes.11
are not labelled. (C) A typical flexibly docked side-view shows
residues from the outer and inner turret of Kv1.1 subunits
interacting with the side chains of compound 10 (in yellow). (D) A
side-view of the flexible docking reveals residues from the outer
and inner turret regions of all Kv1.2 subunits interacting with the
side chains of compound 15 (in yellow).
Interestingly, 15 was found to exhibit selectivity for the Kv1.2
channel over the other subtypes. Following a similar docking
procedure, 15 was flexibly docked into the mammalian Kv1.2
complex (PDB: 2A79).12 Here, docked 15 lies across the top of the
Kv1.2 tetramer, making contacts with each chain, and blocking the
entrance to the inner turret. Key interactions are predicted to occur
with residues Asp355 and Ser356 of the outer turret; Asp363 of the
pore helix, and Gly378 and Asp379 of the inner turret region. These
interactions could explain the absolute specificity of compound 15
for the Kv1.2 channel.
A possible mode of interaction between the porphyrins and the
Kv1.1 channel was derived from homology modelling (HM),19 using
the crystal structure of Kv1.2 as a template12 (Fig. 4A). The
geometric quality of the backbone conformation, residue interactions
and contact plus energy profile of the structure fall well within the
restrictions established for reliable structures (e.g. 94.2 % in core
and 5.7 % in allowed regions) and comparable to the template
used.20, 21 Autodock422 was used to refine the side chains from the
outer and inner turrets in the docking models for 12 and 10.
Predicted interactions between 12 and the channel included HBs
with: chain A (Glu353Gly374, Tyr375), chain B (Gly374), chain C
(Glu351, Ala352, GLu353, Gly374and Tyr375 ) and chain D
(Glu353; Gly374).
Slowing the activation kinetics, in addition to abilities of 10ꢀ12
and 15 to inhibit selective Kv1 channels would be advantageous, for
example, in deꢀaccelerating the activation of unwanted Kv1 channels
overꢀexpressed in demyelinated axons.9 For therapeutic applications
in future, the more selective porphyrinꢀbased inhibitors could be
improved by retaining the 4ꢀfold scaffold bearing the appropriate
side chains that exhibit selectivity for members of Kv1 channels but
with a nonꢀphotoꢀreactive core.
These residues are largely in the inner turret region (Fig. 4B). A
flexible docked Kv1.1 channel conformation was then used to
predict interactions with 10. Predicted interactions between 10 and
Kv1.1 channel include: chain A (Glu353, Gly374, Tyr375), chain B
(Gly374, Tyr375) chain C (Phe356, Asp361, Gly374, Tyr375) and
chain D (Val373, Gly374, Gly376); these residues are again largely
in the inner turret region (Fig. 4C). Our predicted binding pose for
10, with one substituent of the porphyrin derivative inserted into the
inner turret, is similar to that described by Gradl et al.1; this showed
the bound porphyrin penetrating into the selectivity filter of a Kv1.3
model, where the protonated amine favourably interacts with the K+
binding site. This docking model indicates that compound 10 when
compared with 12 has broader interactive sites and a shallower
interaction in the inner turret region, consistent with its steep
concentration dependence and Hill coefficients (Fig. 3A).
In Summary, this work provides a proof of principle for the
feasibility of developing small inhibitors for particular subtypes
of neuronal Kv1 channels.
Notes and references
This work was funded by a Principle Investigator grant from
Science Foundation Ireland (to J.O.D.) and the Programme for
Research in Third Level Institutions (PRTLI) Cycle 4. The
PRTLI is coꢀfunded through the European Regional
This journal is © The Royal Society of Chemistry 2014
Chem. comm., 2014, 00, 1-3 | 3

ChemComm
Page 4 of 4
View Article Online
COMMUNICATION
Chemical communications
DOI: 10.1039/C4CC05639F
Development Fund (ERDF), part of the European Union 16 B. R. Deng B, Johnson RS, Girdhar NK, Danumah C, Yamazaki T,
Structural Funds Programme 2007ꢀ2013. The authors sincerely Fenniri H., Macromolecules, 2012, 45, 7157.
thank the software vendors for their continuing support for 17 A. Grupe, K. H. Schroter, J. P. Ruppersberg, M. Stocker, T. Drewes,
academic research efforts, in particular contributions from S. Beckh, and O. Pongs, Embo J, 1990, 9, 1749.
Accelrys Discovery Studio. Dr. Liam O’Hara is thanked for 18 W. Stuhmer, J. P. Ruppersberg, K. H. Schroter, B. Sakmann, M.
generating the stable cell lines and Ms. Sharon Whyte for
manuscript preparation. Dr Declan Daly was funded by the Irish
Research Council
Stocker, K. P. Giese, A. Perschke, A. Baumann, and O. Pongs, Embo
J, 1989, 8, 3235.
19 A. Sali and T. L. Blundell, J Mol Biol, 1993, 234, 779.
20 R. A. Laskowski, J. A. Rullmannn, M. W. MacArthur, R. Kaptein,
and J. M. Thornton, J Biomol NMR, 1996, 8, 477.
21 A. L. Morris, M. W. MacArthur, E. G. Hutchinson and J. M.
Thornton, Proteins, 1992, 12, 345.
a School of Chemical Sciences, Dublin City University,
Glasnevin, Dublin 9, Ireland
bInternational Centre for Neurotherapeutics, Dublin City
University, Glasnevin, Dublin 9, Ireland
22 G. M. Morris, R. Huey, W. Lindstrom, M. F. Sanner, R. K. Belew, D.
cBiology Department, National University of Ireland Maynooth,
Maynooth, Co. Kildare, Ireland
S. Goodsell and A. J. Olson J. Computational Chemistry 2009, 16,
2785.23V. Martos, S. C. Bell, E. Santos, E. Y. Isacoff, D. Trauner,
and J. de Mendoza, Proc Natl Acad Sci U S A, 2009, 106, 10482.
dBoth authors contributed equally to this work
Corresponded authors: oliver.dolly@dcu.ie, kieran.nolan@dcu.ie
†Electronic Supplementary Information (ESI) available: DNA constructs
and expression of Kv1 channels, Electrophysiological recordings and data
analysis, Computational methods, Homology modelling, Molecular
docking, Synthesis of compounds 10ꢀ18 and their precursors and NMR
and mass spectra for compounds 10ꢀ18. See DOI: 10.1039/c000000x/
1
2
3
4
5
6
S. N. Gradl, J. P. Felix, E. Y. Isacoff, M. L. Garcia, and D. Trauner, J
Am Chem Soc, 2003, 125, 12668.
M. C. D'Adamo, L. Catacuzzeno, G. Di Giovanni, F. Franciolini, and
M. Pessia, Front Cell Neurosci, 2013, 7, 134.
D. N. Parcej, V. E. Scott, and J. O. Dolly, Biochemistry, 1992, 31,
11084.
F. C. Wang, D. N. Parcej, and J. O. Dolly, Eur J Biochem, 1999, 263,
230.
V. E. Scott, Z. M. Muniz, S. Sewing, R. Lichtinghagen, D. N. Parcej,
O. Pongs, and J. O. Dolly, Biochemistry, 1994, 33, 1617.
V. E. Scott, J. Rettig, D. N. Parcej, J. N. Keen, J. B. Findlay, O.
Pongs, and J. O. Dolly, Proc Natl Acad Sci U S A, 1994, 91, 1637.
O. Pongs and J. R. Schwarz, Physiol Rev, 2010, 90, 755.
S. I. Judge and C. T. Bever, Jr., Pharmacol Ther, 2006, 111, 224.
B. Bagchi, A. AlꢀSabi, S. Kaza, D. Scholz, V. B. O'Leary, J. O.
Dolly, and S. V. Ovsepian, PLoS One, 2014, 9, e87736.
7
8
9
10 C. Ader, R. Schneider, S. Hornig, P. Velisetty, E. M. Wilson, A.
Lange, K. Giller, I. Ohmert, M. F. MartinꢀEauclaire, D. Trauner, S.
Becker, O. Pongs, and M. Baldus, Nat Struct Mol Biol, 2008, 15,
605.
11 S. Hornig, I. Ohmert, D. Trauner, C. Ader, M. Baldus, and O. Pongs,
Channels (Austin), 2013, 7, 473.
12 S. B. Long, E. B. Campbell, and R. Mackinnon, Science, 2005, 309,
897.
13 D. Muller, I. Zeltser, G. Bitan, and C. Gilon, J Org Chem, 1997, 62,
411.
14 T. Gianferrara, A. Bergamo, I. Bratsos, B. Milani, C. Spagnul, G.
Sava, and E. Alessio, J Med Chem, 2010, 53, 4678.
15 M. Leonor, Moura, C., Paulo, A., Santos, I. C. and Vitor, R. F.,
Dalton Transactions, 2006, 47, 5630.
4 | Chem. comm., 2014, 00, 1-3
This journal is © The Royal Society of Chemistry 2014