A bright and responsive europium probe for determination of pH change within the endoplasmic reticulum of living cells

Article abstract of DOI:10.1039/c3cc42308e

A ratiometric Eu^III complex has been developed that localises selectively within the endoplasmic reticulum of living cells. Careful calibration, using a time-gated spectral imaging microscope, allows the intensity ratio of emission bands and the variation of excited state lifetime to be used for pH determination, with a pK_a of 7.15.

Full text of DOI:10.1039/c3cc42308e

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A bright and responsive europium probe for
determination of pH change within the
endoplasmic reticulum of living cells†
Cite this: Chem. Commun., 2013,
49, 5363
Received 29th March 2013,
Accepted 30th April 2013
Brian K. McMahon, Robert Pal and David Parker*
DOI: 10.1039/c3cc42308e
www.rsc.org/chemcomm
A ratiometric EuIII complex has been developed that localises the polarisability of the axial donor, to allow ratiometric analyses of
selectively within the endoplasmic reticulum of living cells. Careful various analytes such as citrate, lactate and bicarbonate.5,6 The mode of
calibration, using a time-gated spectral imaging microscope, allows action for these systems involved displacement of a coordinated H2O
the intensity ratio of emission bands and the variation of excited molecule, either through direct interaction of the analyte at the metal
state lifetime to be used for pH determination, with a pKa of 7.15.
centre or, in the case of pH monitoring, by reversible pH-dependent
intramolecular sulphonamide ligation.
The pursuit of non-invasive, responsive optical probes which target
Our previous pH probes have been based on the 1,4,7,10-tetra-
specific cellular organelles and signal concentration changes of azacyclododecane macrocyclic framework incorporating either
bioactive species in real time remains a challenging area of research.1 an azathiaxanthone (e = 6500 MÀ1 cmÀ1) or azaxanthone (e =
In particular, probes that can report changes to the pH environment 5700 MÀ1 cmÀ1) sensitising moiety and exhibiting quantum yields
of certain cellular compartments have the potential to be used in of 6% or less. With this in mind, we have set out to develop an
the monitoring of several pathological and physiological cellular optical pH probe with a higher quantum yield that incorporates
processes. The critical design factors for these probes require careful multiple sensitising moieties with large molar absorptivity, creating
consideration. Important factors include rapid cellular uptake, specific a complex with overall higher brightness, B, where B(l) = e(l)f.
organelle localisation, low cytotoxicity, the ability to be calibrated and A family of EuIII complexes based on triazacyclononane with three
the retention of probe performance in cellulo.
p-substituted aryl-alkynyl groups to harvest incident light has
Determination of the pH environment of biochemical events has recently been reported. With quantum yields as high as 50% and
commonly relied on the use of chromogenic pH indicators which often molar extinction coefficients in the range of 60 000 MÀ1 cmÀ1, it was
only take advantage of the lower wavelengths of the electromagnetic decided to create new pH probes based on such systems.7 A methyl
spectrum. However, the need for more advanced luminescent systems, substituted sulphonamide moiety that can bind reversibly to the
which can be employed for more efficient in vivo detection in real time, lanthanide ion, thereby changing the metal coordination environ-
has sparked the development of probes with longer emission life- ment, was incorporated into the ligand design (Scheme 1).
times.2,3 Such delayed emission has the advantage of overcoming the
The pyridylalkynylaryl based chromophore is known to allow
poor signal-to-noise ratio caused by short lived background emission excitation of EuIII over the range 337–365 nm and was prepared using
(autofluorescence) and light scattering from the surrounding biological established methods.7 Subsequent dialkylation of mono-BOC-triaza-
environment. It is well known that emissive lanthanide complexes are cyclononane, followed by BOC deprotection (TFA-CH2Cl2) resulted in
particularly well suited for this purpose, with an extensive variety of the bis-alkylated precursor ligand. Reaction with 2-methanesulfonate-
examples being reported.4 Another significant advantage of using N-methanesulfonylethylamine in MeCN in the presence of K2CO3 at
lanthanide based probes is their ability to measure fluctuations in 70 1C afforded the desired ligand. Prior to complexation with
the concentration of intracellular species through well established Eu(OAc)3, base hydrolysis of the ethyl phosphinate esters with NaOH
ratiometric methodology. This is most relevant with complexes of in a CD3OD : D2O (3 : 1) solution mixture was carried out. Purification
europium(III), as the spectral form and emission intensity are highly by preparative-HPLC gave [EuL1]+ in 33% yield.
dependent upon the symmetry and nature of the coordination sphere.
Examination of the EuIII emission spectra of [EuL1]+ as a function
In recent work, we have exploited the sensitivity of the DJ = 2 band to of pH (0.1 M NaCl, 298 K) revealed substantial and reversible
changes in the fine splitting of the DJ = 1 transition and also in
the form of the DJ = 2 and DJ = 4 spectral bands. Upon basification,
two new bands at 625 nm and 688 nm were observed. By plotting the
Department of Chemistry, Durham University, South Road, Durham, DH1 3LE, UK.
E-mail: david.parker@dur.ac.uk
change in the emission intensity ratio of selected bands (e.g. DJ =
2/[DJ = 0 + DJ = 1]) over the pH range 4–9 (Fig. 1) an 80% increase in
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c3cc42308e
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 5363--5365 5363

View Article Online
Communication
ChemComm
Table 1 Rate constants kb (msÀ1 Æ10%) for depopulation of the EuIII 5D0 excited
state and hydration state qa, (Æ0.2) for [EuL1]+ in the presence of 0.1 M NaCl or
0.4 mM HSA at varying pH values (lexc = 332 nm, 298 K)
[EuL1]
pH
k(H2O)
k(D2O)
qEu
0.1 M NaCl
4.5
7.0
9.5
4.5
9.5
1.54
1.72
1.75
0.99
1.26
0.80
1.34
1.46
0.72
1.23
0.6
0.2
0
0
0
0.4 mM HSAc
Scheme 1 Representation of pH dependent sulphonamide ligation.
a
q values were estimated according to ref. 8 and assume fast NH/DH
b
exchange for the unbound sulphonamide. Values represent the mean
c
of three separate measurements. At pH 9, with excess bicarbonate in
the ratio was found, with a protonation constant of 6.52(Æ0.05)
being estimated. Lifetime measurements for the EuIII excited state of
[EuL1]+, both in D2O and H2O, were recorded at various pH values. At
pH 9, a q value of zero was determined with partial hydration being
observed under more acidic conditions (q = 0.6 at pH 4). The more
intense emission observed at higher pH is characterised by an
extremely high overall emission quantum yield of 38%. At pH 6.5,
the quantum yield reduced to 16%.
H2O and D2O, k values were 1.33 and 1.05 msÀ1 respectively, q = 0.04.
significant changes were observed in the profile or relative intensities of
the EuIII emission bands. In relation to protein interference, at pH 9 no
major structural changes in the EuIII emission were observed, so that
the concept of direct binding of the protein functional groups to the
EuIII centre and thus displacement of the sulphonamide moiety can be
eliminated. As expected, upon acidification (pH 4), a different spectral
form to that seen in the absence of HSA was observed, suggesting
possible protein binding to the metal centre at lower pH. Excited state
lifetime measurements also confirmed that in both acidic (pH 4) and
basic (pH 9) media, [EuL1]+ had a hydration state of zero (Table 1).
Recently, a structurally similar complex to [EuL1]+, with three
methoxy pyridylalkynylaryl chromophore units exhibited selective
staining of the mitochondria in a variety of cell lines (NIH-3T3, CHO
and PC-3).7 However, it was not known whether replacement of one
of these antenna moieties with the methyl sulphonamide pendant
arm would have a significant effect on the localisation profile.
Indeed, cellular localisation studies with [EuL1]+, using a concen-
tration of 30 mM and incubation times of between 30 min and 48 h
showed little staining of the mitochondrial region of CHO and NIH-
3T3 cells. Nevertheless, co-staining experiments with ER Tracker
Green confirmed significant uptake in the endoplasmic reticulum
with 90(Æ2)% correspondence being determined (Fig. 2).
In order to gain insight into the response of [EuL1]+ in cellulo, its
emissive behaviour was examined in a ‘simulated extracellular’ ionic
background consisting of 0.9 mM hydrogen phosphate, 0.13 mM
citrate, 2.3 mM lactate, 30 mM bicarbonate, 0.1 M NaCl and 0.4 mM
human serum albumin (HSA). A shift in the protonation constant
associated with reversible intramolecular sulphonamide ligation to
7.10(Æ0.03) was observed, owing to a competitive protein interaction
with the rod-like alkynyl sensitising moieties and anion chelation at
the EuIII metal centre. Furthermore, the EuIII emission intensity
showed an overall increase upon acidification. Confirmation of
complex interference within this competitive medium was gained
by comparing the spectral form of [EuL1]+ at pH 9 to that recorded in
the absence of any protein or anions. Significant changes in the fine
splitting of the DJ = 1, DJ = 2 and DJ = 4 transitions were observed,
consistent with a change in the EuIII coordination environment.
Significant competition to intramolecular sulphonamide ligation
comes from intermolecular HCO3À binding and this was hypothesised
as the principle cause of the spectral changes. This aspect was
investigated by recording an emission spectrum of [EuL1]+ in the
presence of 30 mM HCO3À alone. The resulting spectral form was
identical to that observed when the mixed ‘simulated extracellular’
media was used. For each of the other biologically relevant anions, no
Incubation times of up to 24 h were investigated, but little
change in the intensity of the staining was observed after 4 h,
suggesting relatively quick localisation within the ER(Æ5%).
Cytotoxicity studies using an MTT assay determined an IC50 value of
greater than 100 mM for [EuL1]+, over a 24 h period,9 and ICP-MS
measurements of 4 Â 106 NIH 3T3 cells loaded with 30 mM [EuL1]
for 24 h at pH 5.6 indicated that the total intracellular concentration
was 390 mM. Interestingly, at higher pH (7.7), a decrease of ca. 50%
in intracellular complex concentration was measured, suggesting
uptake and accumulation processes for [EuL1]+ may occur more
efficiently at lower pH.
Fig. 2 (a) LSCM microscopy images of NIH 3T3 cells (mouse skin fibroblasts)
loaded with [EuL1]+ (4 h incubation, 30 mM complex concentration in the growth
medium); lexc = 355 nm, lem = 605–720 nm (b) corresponding ER Tracker Green
emission (lexc = 496 nm, lem = 505–530 nm) (c) RGB merge image (P = 0.90).
Fig. 1 Variation of the EuIII emission of [EuL1]+ as a function of pH (H2O, 5 mM
complex, 298 K, I = 0.1 M NaCl, lexc = 332 nm). Inset: plot of DJ = 2/(DJ = 0 + DJ = 1)
versus pH, showing the fit to the observed data for an apparent pKa = 6.52(Æ0.03).
c
5364 Chem. Commun., 2013, 49, 5363--5365
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
Previous studies by Kim et al.10 demonstrated that due to the large
proton permeability of the ER membrane, significant communication
occurs with the homeostatic mechanism that controls cytosolic pH.
Thus, the value of pHER is similar to that found within the cytoplasm
(7.1). Therefore, any knowledge gained on pH changes of the ER
region using [EuL1]+, provides information on overall cytosolic pH. In
order to ensure the probe retains its ability to respond to pH changes
in cellulo, a preliminary experiment was carried out where the pHER
was altered from 7.7 to 6.0 (by varying the pH of the growth medium
and adding nigericin (0.2 mM) to allow K+/H+ exchange) and the
response of [EuL1]+ determined by microscopy. By changing the pH of
the cell culture media from 7.7 to 6.0, a two fold increase in brightness
within the ER organelles was observed. Using the Leica SP5 II LSCM
microscope equipped with a high sensitivity hybrid detector (HyD),
changes in the emission intensities of the various individual selected
EuIII transitions were monitored. This involved extracting an average
intensity value via a contrast transfer function (CTF) for each band,
from 15 small (250  250 pixel) voxels (pixel size 120 nm  120 nm Â
0.772 mm) within the image which represented the stained ER regions
of the cell. The largest change in emission band ratio with pH was
observed for DJ = 2/DJ = 4, where the ratio increased from 1.8 at pH 7.7
to 2.2 at pH 6.0.
Fig. 3 Changes in the Eu emission intensity ratio DJ = 2/(DJ = 0 + DJ = 1) and
excited state lifetimes (t) for [EuL1]+ as a f(pH).
through ratiometric comparison of two emission bands or by
measurement of its excited state lifetime. Each method is
independent of complex concentration. It is believed that the
ER pH mimics that found within the cytoplasm, so that this
probe also provides information on mean cytosolic pH. Such
behaviour augurs well for the development of related responsive
systems (e.g. pM, pX) that can faithfully report concentration
changes in particular cell compartments of living cells.
A more detailed in cellulo calibration experiment was carried
out using a custom-built, time-gated spectral imaging micro-
scope, where spectra and excited state lifetimes for [EuL1]+
within the ER regions of the cell (4 h incubation) were recorded
at 11 different pH points between pH 7.5 and 6.0. An experi-
mental technique was employed using growth media with
different pH values, along with the addition of nigericin
(0.2 mM) at each loading to ensure pHER equilibration (ESI†).
To ensure that any changes in pCO2 within the incubation
chamber did not alter the pH of the growth medium, it was
necessary before loading to allow equilibration of the partial
pressure of CO2 in the solutions and re-adjust the pH (Â3). As
observed for the in vitro based pH titration carried out in the
‘simulated extracellular’ ionic background, acidification led to
an overall increase in emission intensity along with changes in
spectral form. By plotting DJ = 2/(DJ = 0 + DJ = 1) against pH, it
was apparent that a near linear response was observed between
the pH range 6.7 and 7.4, with an overall 40% change in ratio
measured within this region. Changes in the excited state
lifetimes were also observed, with values of 672 ms and 413 ms
recorded at pH 6.7 and pH 7.4, respectively. This 60% increase
in the lifetime upon lowering the pH within the ER, suggests
that changes in the structural form of the complex when the
sulphonamide moiety is unbound may be affecting certain
protein–complex interactions, resulting in longer lifetime
values due to a reduction in the quenching process. Comparing
the spectral intensity ratio changes with the pH/lifetime varia-
tion (Fig. 3), a crossover point occurs at pH 7.15(Æ0.05), in
agreement with the pKa value (7.10(Æ0.03)), obtained for [EuL1]
in a competitive anion/protein environment.
We thank the ERC (FCC 266804) for support.
Notes and references
1 (a) V. P. Torchillin, Annu. Rev. Biochem., 2006, 8, 343; (b) X. Chen,
Y. Zhou, X. Peng and J. Yoon, Chem. Soc. Rev., 2010, 39, 2120;
(c) M. H. Kim and B. R. Cho, Acc. Chem. Res., 2009, 42, 863.
2 (a) D. G. Smith, B. K. McMahon, R. Pal and D. Parker, Chem.
Commun., 2012, 48, 8520; (b) R. Pal and D. Parker, Org. Biomol.
Chem., 2008, 6, 1020; (c) R. Pal and D. Parker, Chem. Commun., 2007,
474.
3 (a) L. K. Truman, S. Comby and T. Gunnlaugsson, Angew. Chem., Int.
Ed., 2012, 51, 9624; (b) M. Liu, Z. Ye, C. Xin and J. Yuan, Anal.
Chim. Acta, 2013, 761, 149; (c) S. Shinoda and H. Tsukube, Analyst,
2011, 136, 431; (d) L. N. Sun, H. Peng, M. I. J. Stich, D. Achotz and
O. S. Wolfbeis, Chem. Commun., 2009, 5000.
4 (a) A. Thibon and V. C. Pierre, Anal. Bioanal. Chem., 2009, 394, 107;
(b) S. V. Eliseeva and J.-C. G. Bu¨nzli, Chem. Soc. Rev., 2010, 39, 189;
(c) E. J. New, D. Parker, D. G. Smith and J. W. Walton, Curr. Opin.
Chem. Biol., 2010, 14, 238–246.
5 (a) D. G. Smith, G.-L. Law, B. S. Murray, R. Pal, D. Parker and
K.-L. Wong, Chem. Commun., 2011, 47, 7347; (b) D. G. Smith, R. Pal
and D. Parker, Chem.–Eur. J., 2012, 18, 11604; (c) R. Pal, A. Beeby and
D. Parker, J. Pharm. Biomed. Anal., 2011, 56, 352; (d) R. Pal, D. Parker
and L. C. Costello, Org. Biomol. Chem., 2009, 7, 1525.
6 S. J. Butler and D. Parker, Chem. Soc. Rev., 2013, 42, 1652.
7 J. W. Walton, A. Bourdolle, S. J. Butler, M. Soulie, M. Delbianco,
B. K. McMahon, R. Pal, H. Puschmann, J. M. Zwier, L. Lamarque,
O. Maury, C. Andraud and D. Parker, Chem. Commun., 2013, 49,
1600–1602.
8 (a) R. S. Dickins, D. Parker, A. S. de Sousa and J. A. G. Williams, Chem.
Commun., 1996, 697; (b) A. Beeby, I. M. Clarkson, R. S. Dickins,
S. Faulkner, L. Royle, D. Parker, A. S. de Sousa, J. A. G. Williams and
M. Woods, J. Chem. Soc., Perkin Trans. 2, 1999, 493.
9 J. Carmichael, W. G. DeGraff, A. F. Gazdar, J. D. Minna and
J. B. Mitchell, Cancer Res., 1987, 47, 936.
10 (a) J. H. Kim, L. Johannes, B. Goud, C. Anthony, C. A. Lingwood,
R. Daneman and S. Grinstein, Proc. Natl. Acad. Sci. U. S. A., 1998,
95, 2997; (b) J. LLopis, H.-W. Liu, A. Miyawaki, M. G. Farquhar and
R. Y. Tsien, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 6803.
In summary, using [EuL1]+, this calibration procedure
provides an efficient method for determining pH change within
the endoplasmic reticulum of living mammalian cells, either
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 5363--5365 5365