Wavelength-orthogonal photolysis of neurotransmitters in vitro

Article abstract of DOI:10.1039/c1cc15135e

Irradiation of a mixture of 4-methoxyphenacyl-caged (S)-glutamate and 4,5-dimethoxy-2-nitrobenzyl-caged γ-amino butyric acid (GABA) on neurons, at ～260 nm, evokes selective photorelease of (S)-glutamate (Glu) whereas photolysis at 405 nm causes selective photorelease of GABA.

Full text of DOI:10.1039/c1cc15135e

View Article Online / Journal Homepage / Table of Contents for this issue
Chemical Communications
www.rsc.org/chemcomm
Volume 48 | Number 5 | 18 January 2012 | Pages 617–772
ISSN 1359-7345
COMMUNICATION
Stuart J. Conway and Nigel J. Emptage et al.
Wavelength-orthogonal photolysis of neurotransmitters in vitro

Dynamic ArticVleiewLiAnrktiscle Online
ChemComm
Cite this: Chem. Commun., 2012, 48, 657–659
www.rsc.org/chemcomm
COMMUNICATION
Wavelength-orthogonal photolysis of neurotransmitters in vitrow
ab
b
Megan N. Stanton-Humphreys,z Ruth D. T. Taylor,z Craig McDougall,c Mike L. Hart,b
C. Tom A. Brown,c Nigel J. Emptage*b and Stuart J. Conway*a
Received 18th August 2011, Accepted 21st September 2011
DOI: 10.1039/c1cc15135e
Irradiation of a mixture of 4-methoxyphenacyl-caged (S)-glutamate
and 4,5-dimethoxy-2-nitrobenzyl-caged c-amino butyric acid
(GABA) on neurons, at B260 nm, evokes selective photorelease
of (S)-glutamate (Glu) whereas photolysis at 405 nm causes
selective photorelease of GABA.
Pairs of caging groups have been defined as wavelength-
selective if it is possible to selectively and sequentially photo-
lyse them using two different irradiating wavelengths in one
order, but that in the other order appreciable quantities of
both groups are cleaved. The term wavelength-orthogonal
photolysis has been applied to systems in which the two
groups can be cleaved with a high degree of selectivity in
either order. Wavelength-orthogonal photolysis was first
achieved by Bochet, who demonstrated that a 3,5-dimethoxy-
benzoin group could be selectively cleaved in the presence of a
4,5-dimethoxy-2-nitrobenzyl (DMNB) group, when irradiated
at 254 nm. Irradiation at 420 nm evoked cleavage of the
DMNB group and the 3,5-dimethoxybenzoin group remained
intact.2–5
The lack of photophysically distinct (wavelength-orthogonal)
caged biomolecules has been a significant shortcoming in the
toolkit of chemical probes for biological investigations. Yet a
plethora of applications exist for wavelength-orthogonally
protected tools:1 Wavelength-orthogonal protection of receptor
agonists and antagonists could enable temporal and spatial
control over receptor activation and inhibition; wavelength-
orthogonal protection of two neurotransmitters would allow
unprecedented study of synaptic integration of excitatory and
inhibitory inputs (Fig. 1); and wavelength-orthogonal caging
of residues within proteins would allow temporal and spatial
control over post-translation protein modification.
Wavelength-selective protecting groups have been employed
in solid phase synthesis6,7 and wavelength-orthogonal protecting
groups have been applied to surface lithography.8–10 The first
biological application of wavelength-selective caging groups
was exemplified by Kotzur et al. who demonstrated the
selective photorelease of thiol groups in a peptide.11 Lawrence
et al. then employed wavelength-selective caging groups to
control the photoactivation of cGMP- and cAMP-mediated
Our aim was to demonstrate proof-of-concept in the appli-
cation of wavelength-orthogonal protecting groups to the
study of biological problems. We focused on (S)-glutamic acid
(Glu) and g-amino butyric acid (GABA) as they are the two
most significant central nervous system (CNS) neurotransmitters
and evoke opposing biological responses. Glu is an excitatory
amino acid that mediates its action via activation of a large
family of Glu receptors, which comprise both ligand-gated cation
channels and G-protein-coupled receptors (GPCRs). GABA is
an inhibitory transmitter that exerts its action via activation
of both ligand-gated chloride channels and GPCRs. Con-
sequently, these transmitters have fast biological responses that
are clearly distinguishable in neurons. Placing Glu and GABA
under wavelength-orthogonal photocontrol is of interest when
attempting to mimic the complex signals received by neurons.
a Department of Chemistry, Chemistry Research Laboratory,
University of Oxford, Mansfield Road, Oxford, OX1 3TA, UK.
E-mail: stuart.conway@chem.ox.ac.uk; Fax: +44 (0)1865 285002;
Tel: +44 (0)1865 285109
b Department of Pharmacology, University of Oxford,
Mansfield Road, Oxford, OX1 3QT, UK.
E-mail: nigel.emptage@pharm.ox.ac.uk; Fax: +44 (0)1865 271853;
Tel: +44 (0)1865 271588
c School of Physics and Astronomy, University of St Andrews,
North Haugh, St Andrews, Fife, KY16 9SS, UK.
w Electronic supplementary information (ESI) available: Full chemical
and biological experimental details, additional eletrophysiological
experiments. See DOI: 10.1039/c1cc15135e
z These authors contributed equally to this work.
Fig. 1 Wavelength-orthogonal photolysis of caged glutamate and
GABA. Irradiation at 405 nm activates GABA receptors whereas
irradiation at 250–260 nm activates glutamate receptors.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 657–659 657

View Article Online
When a mixture of MPA-Glu and DMNB-GABA was
irradiated using a 405 nm diode, photolysis of DMNB-GABA
1
was observed by H NMR analysis (Fig. S2A and B, ESIw),
but MPA-Glu was unaffected. When an identical mixture was
irradiated using a UV lamp equipped with a filter having
maximum transmission at 285 nm (Fig. S1, ESIw), only MPA-
Glu was photolysed (Fig. S2C and D, Tables S1–S2 and
Fig. S3–4, ESIw). These results demonstrate that MPA-Glu is a
suitable orthogonal partner for DMNB-GABA. The relatively
long times required for photolysis in these experiments simply
reflect the need to photolyse sufficient material for NMR analysis
and do not reflect the in vitro kinetics of photorelease. The
in vitro photolysis of these groups occurs in the millisecond
timeframe (Fig. 3).
Fig. 2 DMNB-caged GABA (1),27 glycine (2) and Glu (3), 3,5-
dimethoxybenzoin-caged GABA (4), Glu (5), and MPA-caged Glu (6).
With the potential for wavelength-orthogonal photolysis
established, MPA-Glu (6) and DMNB-GABA (1) were applied
separately to hippocampal pyramidal neurons. MPA-Glu had
no effect when applied to the bath solution, indicated by a lack
of discernable increase in neuronal excitability, action potential
firing or change in holding potential. Additionally, when bath-
applied MPA-Glu was irradiated at 405 nm, no electrophysio-
logical response was observed (Fig. 3C, bottom panel), showing
that no Glu was released and that 405 nm light does not affect
the neurons. However, irradiation using a UV flash lamp
equipped with a 250–260 nm bandpass filter evoked a cation
current consistent with activation of Glu receptors (Fig. 3C,
top panel; MPA-Glu [25–125 mM] was photolysed at 250–260 nm
in 17 different cells a total of 30 times). DMNB-GABA showed
no activity on cells when applied to the bath solution. Irradiation
at 250–260 nm did not produce a response (Fig. 3D, bottom
panel), showing that no GABA was released and that 250–260 nm
light does not affect the neurons for the timescale of the experi-
ment. However, irradiation at 405 nm evoked an anion current
consistent with activation of GABA receptors (Fig. 3D, top
panel; DMNB-GABA [25–100 mM] was photolysed at 405 nm
in 20 different cells a total of 30 times). To establish that the
currents observed were evoked by activation of either Glu or
GABA receptors, photolysis was conducted in the presence or
absence of either Glu or GABA receptor antagonists. DNQX
is an antagonist of AMPA receptors, through which fast Glu
currents are mediated. Picrotoxin is an antagonist at the iono-
tropic GABAA receptors, which mediate the fast GABA response.
It was shown that photolysis of MPA-Glu at 250–260 nm evoked
a current, which was abolished in the presence of DNQX
(Fig. 3A, n = 5). Similarly, DMNB-GABA was shown to
produce a current when irradiated at 405 nm. This current was
abrogated in the presence of picrotoxin (Fig. 3B, n = 6).
signalling pathways.12 Schafer et al. have demonstrated the
¨
wavelength-selective photorelease of two oligonucleotides.13
Recently, Goguen et al. have reported the synthesis of wave-
length-selectively-caged phosphopeptides and used them to
study Wip 1 phosphatase.14 Kantevari et al. have published
an example of 2-photon, 2-color photolysis of GABA and
Glu, in which they show optically independent photolysis.15
Herein, we demonstrate wavelength-orthogonal photolysis of
Glu and GABA across a heterologous population of receptors
in hippocampal neurons.
Initially, the 3,5-dimethoxybenzoin and DMNB groups
employed by Bochet were used in the synthesis of caged
pairs of Glu and GABA (1–5, Fig. 2 and ESIw). However,
despite successful demonstration of wavelength-orthogonal
photolysis during 1H NMR studies, the 3,5-dimethoxybenzoin
caging group was found to be toxic to neurons. Although
3,5-dimethoxybenzoin-caged ATP has been used in vitro
previously without toxicity being reported,16 we were unable
to continue using this caging group. Therefore it was necessary
to identify another biocompatible caging group that could
be photolysed in an orthogonal manner to the DMNB
group. Previously, during the development of caged TRPV1
agonists,17,18 we investigated the use of the 4-methoxyphenacyl
(MPA) caging group.19 This group and its derivatives have been
employed by Givens and co-workers to cage phosphates,20
neurotransmitters21,22 and protein kinase A.23 The MPA
group typically displays a lmax of shorter wavelength than
the DMNB group and hence was likely to be wavelength
orthogonal to the DMNB group. Compounds similar to the
photolysis by-products from the MPA group (4-hydroxyaceto-
phenone, 4-methoxyacetophenone and 4-hydroxyphenylacetic
acid) were applied to the neurons and were found to be non-toxic.
Therefore, MPA-caged Glu was synthesised (6, Fig. 2 and
ESIw) and was observed to have a lmax = 283 nm and an
extinction coefficient (e) of 14 000. The UV/vis spectra of 1 and
6 show that MPA-Glu does not absorb at 405 nm in contrast
to DMNB-GABA, which has significant absorption at this
wavelength (Fig. S1, ESIw). This observation indicates that
it should be possible to remove the DMNB group in the
presence of the MPA group by irradiation at 405 nm. Although
both compounds absorb in the 250–285 nm region, the higher
extinction coefficient, relatively fast release and good quantum
yield of the 4-methoxyphenacyl group might contribute to
successful wavelength-orthogonal photolysis.24–26
With the potential for wavelength-orthogonal photolysis
in vitro established, both MPA-Glu (6) and DMNB-GABA
(1) were applied simultaneously to the bath solution of hippo-
campal pyramidal neurons. The compounds showed no effect
until photolysed (Fig. 3E). Irradiation of this mixture, using a
UV flash lamp equipped with a 250–260 nm bandpass filter,
evoked a cation current consistent with activation of Glu
receptors (Fig. 3E). Subsequent exposure of the mixture to
light of 405 nm wavelength caused an anion current, consistent
with activation of GABA receptors (Fig. 3E, n = 6). To confirm
that wavelength orthogonality rather than wavelength selectivity
was being observed, the order of irradiation was reversed.
c
658 Chem. Commun., 2012, 48, 657–659
This journal is The Royal Society of Chemistry 2012

View Article Online
anion and cation currents in the orthogonal experiments. A
significant advantage of our compounds is that they can be
used across multiple neurons, with areas of different receptor
expression, without the need for manipulation of irradiating
power or multiple control experiments.
In conclusion, we have synthesised MPA-Glu (6) and
DMNB-GABA (1) and shown that these two compounds can
be photoreleased in a wavelength-orthogonal manner. These
compounds were employed in the wavelength-orthogonal
activation of Glu and GABA receptors in hippocampal neurons.
These data represent the first example of 1-photon wavelength-
orthogonal photolysis in vitro, giving exquisite control over the
activation of Glu and GABA receptors. Our results prove the
principle that wavelength-orthogonal photolysis can be applied
to a variety of neurotransmitters and other biological systems.
The authors thank the EPSRC, Leverhulme Trust, MRC,
Welcome Trust, HEFCE, Oxford University & ORS for funding.
Notes and references
1 H.-M. Lee, D. R. Larson and D. S. Lawrence, ACS Chem. Biol.,
2009, 4, 409–427.
2 C. Bochet, Tetrahedron Lett., 2000, 41, 6341–6346.
3 C. Bochet, Angew. Chem., Int. Ed., 2001, 40, 2071–2073.
4 A. Blanc and C. G. Bochet, J. Org. Chem., 2002, 67, 5567–5577.
5 C. Bochet, Synlett, 2004, 2268–2274.
6 M. Kessler, R. Glatthar, B. Giese and C. G. Bochet, Org. Lett.,
2003, 5, 1179–1181.
7 M. Ladlow, C. H. Legge, T. Neudeck, A. J. Pipe, T. Sheppard and
L. L. Yang, Chem. Commun., 2003, 2048–2049.
8 A. del Campo, D. Boos, H. W. Spiess and U. Jonas, Angew. Chem.,
Int. Ed., 2005, 44, 4707–4712.
9 P. Stegmaier, J. M. Alonso and A. del Campo, Langmuir, 2008, 24,
11872–11879.
10 V. San Miguel, C. G. Bochet and A. del Campo, J. Am. Chem.
Soc., 2011, 133, 5380–5388.
11 N. Kotzur, B. Briand, M. Beyermann and V. Hagen, J. Am. Chem.
Soc., 2009, 131, 16927–16931.
12 M. A. Priestman, L. Sun and D. S. Lawrence, ACS Chem. Biol.,
2011, 6, 377–384.
13 F. Schafer, K. B. Joshi, M. A. H. Fichte, T. Mack, J. Wachtveitl
¨
and A. Heckel, Org. Lett., 2011, 13, 1450–1453.
14 B. N. Goguen, A. Aemissegger and B. Imperiali, J. Am. Chem.
Soc., 2011, 133, 11038–11041.
15 S. Kantevari, M. Matsuzaki, Y. Kanemoto, H. Kasai and G. C.
R. Ellis-Davies, Nat. Methods, 2010, 7, 123–125.
16 H. Thirlwell, J. E. Corrie, G. P. Reid, D. R. Trentham and
M. A. Ferenczi, Biophys. J., 1994, 67, 2436–2447.
Fig. 3 Wavelength-orthogonal photolysis of MPA-Glu (6) and DMNB-
GABA (1) on neurons. (A) Representative voltage clamp experiment
showing that illumination of 50 mM MPA-Glu by light at 250–260 nm
(arrow) gives activation of a glutamatergic current (top) that is suppressed
by 50 mM DNQX. (B) In the presence of 50 mM DMNB-GABA,
illumination by light at 405 nm results in activation of a GABAergic
current that is suppressed by 100 mM picrotoxin (bottom). (C) A
glutamatergic current is elicited in response to light at 250–260 nm
(top), but not in response to light at 405 nm (bottom) indicating that
photolysis of 125 mM MPA-Glu is wavelength specific. (D) Light
irradiation at 405 nm (top panel), but not light at 250–260 nm (bottom
panel) elicits a GABAergic current in the presence of 50 mM DMNB-
GABA. (E) Demonstration of wavelength-orthogonal photolysis. In the
presence of 125 mM MPA-Glu and 50 mM DMNB-GABA, irradiation at
250–260 nm results in a glutmatergic current, while subsequent irradiation
at 405 nm evokes a GABAergic current. (F) Wavelength-orthogonal
photolysis occurs with either order of wavelength presentation.
17 J. L. Carr, K. N. Wease, M. P. van Ryssen, S. Paterson, B. Agate,
K. A. Gallagher, C. T. A. Brown, R. H. Scott and S. J. Conway,
Bioorg. Med. Chem. Lett., 2006, 16, 208–212.
18 M. P. van Ryssen, N. Avlonitis, R. Giniatullin, C. McDougall,
J. L. Carr, M. N. Stanton-Humphreys, E. L. A. Borgstrom, C. T.
¨
It was demonstrated that photolysis first at 405 nm led to
activation of GABA receptors, and subsequent photolysis at
250–260 nm led to activation of Glu receptors (Fig. 3F, n = 3).
Hence, application of both compounds to the bath solution
allows rapid and interchangeable control of both Glu and
GABA channels, modulated only by the wavelength of irra-
diating light. No toxicity associated with irradiation of neurons
at 250–260 nm was observed in the time course of our experi-
ments, as assessed by physiological and visual means. The
control photolyses of the individual caged neurotransmitters
in vitro were conducted at the same concentration, and were
irradiated with the same power and for the same length of
time, as the experiments with the mixture of caged compounds.
Consequently, we are confident that we are observing pure
A. Brown, D. Fayuk, A. Surin, M. Niittykoski, L. Khiroug and
S. J. Conway, Org. Biomol. Chem., 2009, 7, 4695–4707.
19 J. Sheehan and K. Umezawa, J. Org. Chem., 1973, 38, 3771–3774.
20 R. Givens and L. Kueper, Chem. Rev., 1993, 93, 55–66.
21 R. Givens, A. Jung, C. Park, J. Weber and W. Bartlett, J. Am.
Chem. Soc., 1997, 119, 8369–8370.
22 K. Stensrud, J. Noh, K. Kandler, J. Wirz, D. Heger and
R. S. Givens, J. Org. Chem., 2009, 74, 5219–5227.
23 K. Zou, S. Cheley, R. S. Givens and H. Bayley, J. Am. Chem. Soc.,
2002, 124, 8220–8229.
24 C. Park and R. Givens, J. Am. Chem. Soc., 1997, 119, 2453–2463.
25 C. Bochet, J. Chem. Soc., Perkin Trans. 1, 2002, 125–142.
26 H. Yu, J. Li, D. Wu, Z. Qiu and Y. Zhang, Chem. Soc. Rev., 2010,
39, 464–473.
27 M. Wilcox, R. Viola, K. Johnson, A. Billington, B. Carpenter,
J. McCray, A. Guzikowski and G. Hess, J. Org. Chem., 1990, 55,
1585–1589.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 657–659 659