A europium complex that selectively stains nucleoli of cells

Article abstract of DOI:10.1021/ja056303g

A europium complex selectively staining the nucleolus of NIH 3T3, HeLa, and HDF cells is reported. This complex possesses not only the advantage of the long lifetime of europium emission (0.3 ms), but also a chromophore that allows excitation at a relativ

Full text of DOI:10.1021/ja056303g

Published on Web 01/31/2006
A Europium Complex That Selectively Stains Nucleoli of Cells
Junhua Yu,† David Parker,*,† Robert Pal,† Robert A. Poole,† and Martin J. Cann‡
Contribution from the Department of Chemistry and School of Biological and Biomedical
Sciences, Durham UniVersity, South Road, Durham DH1 3LE, United Kingdom
Received September 13, 2005; E-mail: david.parker@durham.ac.uk
Abstract: A europium complex selectively staining the nucleolus of NIH 3T3, HeLa, and HDF cells is
reported. This complex possesses not only the advantage of the long lifetime of europium emission (0.3
ms), but also a chromophore that allows excitation at a relatively long wavelength (λmax ) 384 nm) and
gives rise to an acceptable quantum yield (9%). The complex can be used both in live cell and fixed cell
imaging, giving an average intracellular concentration on the order of 0.5 µM. Strong binding to serum
albumin has been demonstrated by examination of the analogous gadolinium complex, studying relaxivity
changes with increasing protein concentration. The intracellular speciation of the complex has been examined
by circularly polarized emission spectroscopy and is consistent with the presence of more than one europium
species, possibly protein bound.
Introduction
has been loaded into cells to indicate the localization of
nucleoli.7 An advantage of this method is the specificity of the
The nucleolus of the cell was discovered over a century ago,
yet its biological roles remain to be fully elucidated.1 It is widely
accepted that nucleoli are nuclear sites of ribosomal RNA
transcription, processing, and ribosome assembly.2 The nucleolus
is a dynamic nuclear compartment formed during interphase
through fusion of initially separate nucleolar materials. The
mechanisms inducing the conglutination of various nucleolar
components are still under investigation.3 Alternative functions
to a role in ribosome biogenesis have been proposed; for
example, signal recognition, particle RNA maturation, a potential
role for growth factors and growth-factor-related proteins, and
potential links between nucleoli and tumorigenesis and aging
have also been postulated.4 The nucleolus may also be the target
of certain viruses.5
Studies of nucleoli and nucleolus-related processes often
require the visualization of nucleoli. Silver nitrate staining of a
group of argyrophilic proteins highlights the nucleolar organizer
regions (NORs) and partially exposes the localization of the
nucleolus.6 However, the limited staining exhibited as dark spots
under light microscopy has inhibited widespread application,
especially multicolor fluorescent labeling experiments. The most
extensively used methods to stain the nucleolus require the
employment of antibodies against a certain component of the
nucleolus. For example, the anti-fibrillarin monoclonal antibody
antibody against its antigen, allowing researchers to examine
the salient targeted biomolecules or processes. However, this
requires not only a specific antibody that is usually expensive
and often not commercially supplied, but also another dye that
needs to be conjugated to this antibody, or even an “anti-
antibody” antibody to detect this antibody.8 Though there are
reports employing green fluorescent protein (GFP),9 the devel-
opment of stable mammalian cell lines expressing GFP fusion
proteins remains demanding work.10 Low MW molecular probes
for nucleoli are very rare. The only commercial nucleolar stain
is “SYTO RNA-Select”, a green fluorescent cell stain.11 This
dye gives rise to an enhanced fluorescence on binding to RNA.
Compared to common organic fluorescent dyes and semi-
conductor quantum dots (QDs), lanthanide complexes have the
advantage of a long luminescence lifetime, large Stokes shift,
and relatively high quantum yield. The latter feature is achieved
when appropriate sensitizers are incorporated into the structure,
(7) (a) Dean, K. A.; von Ahsen, O.; Go¨rlich, D.; Fried, H. M. J. Cell Sci.
2001, 114, 3479. (b) Korgaonkar, C.; Hagen, J.; Tompkins, V.; Frazier, A.
A.; Allamargot, C.; Quelle, F. W.; Quelle, D. E. Mol. Cell. Biol. 2005, 25,
1258.
(8) (a) Lin, C.-Y.; Li, C.-C.; Huang, P.-H.; Lee, F.-J. S. J. Cell Sci. 2002, 115,
4433. (b) Leary, D. J.; Terns, M. P.; Huang, S. Mol. Cell. Biol. 2004, 15,
281. (c) Leopardi, R.; Roizman, B. Proc. Natl. Acad. Sci. U.S.A. 1996, 93,
4572.
(9) Snaar, S.; Wiesmeijer, K.; Jochemsen, A. G.; Tanke, H. J.; Dirks, R. W. J.
Cell Biol. 2000, 151, 653.
(10) Spector, D. L.; Goldman, R. D. In LiVe Cell Imaging: A Laboratory
Manual; Goldman, R. D., Spector, D. L., Eds.; Cold Spring Harbor: New
York, 2005; pp 25-31.
† Department of Chemistry.
‡ School of Biological and Biomedical Sciences.
(1) Leung, A. K. L.; Andersen, J. S.; Mann, M.; Lamond, A. I. Biochem. J.
2003, 376, 553.
(11) (a) Haugland, R. P. A Guide to Fluorescent Probes and Labeling
Technologies, 10th ed.; Molecular Probes: Eugene, OR, 2005. Parker, D.
Coord. Chem. ReV. 2000, 205, 109; Pandya, S.; Yu, J.; Parker, D. Dalton
Trans., in press. (b) Phimphivong, S.; Saavedra, S. S. Bioconjugate Chem.
1998, 9, 350. (c) Gao, X.; Yang, L.; Petros, J. A.; Marshall, F. F.; Simons,
J. W.; Nie, S. Curr. Opin. Biotechnol. 2005, 16, 63. Bruchez, M., Jr.;
Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, A. P. Science 1998, 281, 2013.
Jaiswal, J. K.; Simon, S. M. Trends Cell Biol. 2004, 14, 497. (d) Cha, A.;
Snyder, G. E.; Selvin, P. R.; Bezanilla, F. Nature 1999, 402, 809. Root, D.
D. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 5685. Maurel, D.; Kriazeff, J.;
Mathis, G.; Trinquet, E.; Ansanay, H. Anal. Biochem. 2004, 329, 253.
(2) Yusupov, M. M.; Yusupova, G. Zh.; Baucom, A.; Lieberman, K.; Earnest,
T. N.; Cate, J. H. D.; Noller, H. F. Science 2001, 292, 883.
(3) Scheer, U.; Hock, R. Curr. Opin. Cell Biol. 1999, 11, 365.
(4) (a) Jacobson, M. R.; Pederson. T. Proc. Natl. Acad. Sci. U.S.A. 1998, 95,
7981. (b) Pederson, T. J. Cell Biol. 1998, 143, 279. (c) Grummt, I. Genes
DeV. 2003, 17, 1691. (d) Johnson, F. B.; Marciniakt, R. A.; Guarente, L.
Curr. Opin. Cell Biol. 1998, 10, 332.
(5) Hiscox, J. A. Arch. Virol. 2002, 147, 1077.
(6) Derenzini M. Micron 2000, 31, 117.
9
2294
J. AM. CHEM. SOC. 2006, 128, 2294-2299
10.1021/ja056303g CCC: $33.50 © 2006 American Chemical Society

A Eu Complex That Selectively Stains Cell Nucleoli
A R T I C L E S
Scheme 1. Synthesis of the Europium Complex [EuL]3+
Figure 1. Absorption (dashed line) and total emission spectra of [EuL]3+
(5 µM) in water (dotted line) and in the presence of L-Asn (solid line, pH
7.3, 300 µM).
allowing intramolecular energy transfer to, for example, the
europium(III) ion.11a The development of time-gated microscopy
renders lanthanide complexes even more attractive as probes,
by virtue of their micro- to millisecond lifetime.11b The large
Stokes shift (excitation in the near UV and emission in the green
or red) and narrow half-height peak widths (<10 nm) are par-
ticularly useful for multicolor imaging. Moreover, while QDs
are considered more stable to photobleaching, their large vol-
umes and poor cell permeability may limit their utility.11c Lumi-
nescent lanthanide complexes have been applied as fluorescence
resonance energy transfer (FRET) donors in the study of a
variety of extracellular processes.11d However, the investigation
of their cell imaging properties is virtually unexplored.
Figure 2. Luminescence images of NIH 3T3 cells loaded with [EuL]3+
:
(a) cells incubated with the complex (500 µM, 1 h) in DMEM, (b) the
same as (a) but the cells were prefixed by methanol. Images were taken on
a Zeiss Axiovert 200M epifluorescence microscope, examining thioxanthone
fluorescence with a GFP filter setting (objective, Plan-Apochromat, 63×/
1.40 oil DIC; excitation filter, band-pass 450 ( 30 nm; emission filter,
band-pass 510 ( 25 nm). Scale bar: 10 µm.
Here, we report the surprising behavior of a europium(III)
complex ([EuL]3+) which may constitute a good alternative to
the commercial nucleolar stain. Not only does it possess a red
the desired ligand, L, in modest yield after purification by
alumina chromatography. The Eu(III) complex was isolated
following reaction with anhydrous europium(III) trifluoromethane-
sulfonate in MeCN.
As reported in our previous work,12 sensitized emission of
the aqua complex of [EuL]3+ (Figure 1) has a quantum yield
of 9% in water, about one-fifth of the fluorescence quantum
yield for the thioxanthone chromophore, following excitation
at 384 nm. The luminescence lifetime of the europium ion is
0.30 ms, much longer than the usual fluorescence lifetimes of
organic dyes (nanosecond scale).
emission contrasting with the green fluorescence of SYTO
RNA-Select (for multicolor imaging), but the complex also
possesses a long emission lifetime, avoiding autofluorescence
from biomolecules when using time-gating microscopy. The
complex was originally studied for its ability to act as a
chemosensor for anions, binding the citrate anion with very good
chemoselectivity in competitive media.12
Cell Localization Behavior. The complex [EuL]3+ permeates
into NIH 3T3, HeLa, and HDF cells when the complex is loaded
in DMEM (Dulbecco’s modified Eagle’s medium) with 10%
NBS (natal bovine serum) at 37 °C in 5% CO2/air. Some bright
spots are then evident in the nuclei of cells excited by near-UV
light (epifluorescence microscopy; excitation filter, band-pass
450 ( 30 nm; emission filter, 510 ( 25 nm). Moreover, pale
luminescence can also be observed in the nucleus and cytoplasm
(Figure 2a). Loading concentrations down to 100 µM were
examined and found to achieve bright luminescence images of
cells (see the Supporting Information, Figures 1S and 2S). The
use of time-gating microscopy or epifluorescence microscopy
equipped with better filters for europium ion emission (excitation
filter, 380 ( 10 nm; emission filter, LP 550 nm) will lower the
concentration required and increase the signal/noise ratio.
A key consideration is whether the cells studied remain viable
and healthy over the period of examination. From the bright
field images, more than 90% of the cells were considered to be
healthy, over a 24 h period, when the loading concentration of
[EuL]3+ was 100 µM. The observation of the nucleolus over
Results and Discussion
Synthesis. The synthesis of the macrocyclic ligand, L,
followed standard methods. Reaction of thiophenol with 2-chloro-
6-methylnicotinic acid gave the thioether 1 (Scheme 1), which
underwent electrophilic cyclization in the presence of poly-
(phosphoric acid) to yield the azathioxanthone. Selective bro-
mination of the benzylic site (NBS/AIBN/CCl4) gave the
mono(bromomethyl) derivative 2. Controlled alkylation of the
trans-disubstituted cyclen diamide 3 (K2CO3, MeCN) afforded
(12) (a) Parker, D.; Yu, J. Chem. Commun. 2005, 3141. The binding affinity of
[EuL] for citrate reported is 3 × 106 M-1, and in a simulated extracellular
ionic background, Kd is 0.28 mM. (b) Osaki, F.; Kanamori, T.; Sando, S.;
Sera, T.; Aoyama, Y. J. Am. Chem. Soc. 2004, 126, 6520.
9
J. AM. CHEM. SOC. VOL. 128, NO. 7, 2006 2295

A R T I C L E S
Yu et al.
Figure 4. The NIH 3T3 cells were incubated in the presence of the complex
(500 µM) in DMEM for 4 h. Images were taken here showing thioxanthone
fluorescence on a Zeiss Axiovert 200M epifluorescence microscope
(objective, Plan-Apochromat, 63×/1.40 oil DIC; excitation filter, BP 450
( 30 nm; emission filter, BP 510 ( 25 nm). Acquisition time: left panel,
230 ms; right panel, 100 ms. Scale bar: 14 µm.
Figure 3. Colocalization of [EuL]3+ and SYTO RNA-Select dye: upper
row, images by live cell loading (complex (100 µM, 4 h) and SYTO dye
(500 nM, 30 min)); lower row, images by fixed cell loading (similar to the
upper row but the cells were fixed before complex loading); left column,
Eu3+ luminescence; middle column, SYTO dye fluorescence; right column,
merged image. Images were taken using a Zeiss LSM 500 META confocal
microscope (for europium luminescence, excited at 405 nm, detected after
filtration with an LP 505 filter; for the SYTO dye, excited at 488 nm, Ar
laser, detected after filtration with a BP 505-550 band-pass filter). Scale
bar: 20 µm.
Parallel observations were made with HeLa cells (human
endothelial carcinoma cells; see the Supporting Information,
Figure 11S) and HDF cells (human dermal fibroblasts, a non-
transformed primary cell line with a lower nucleolar num-
ber, Supporting Information, Figure 12S), with localization also
clearly evident in the nucleoli. The consistent and stable
luminescence of the complex in live cells renders the complex
a new candidate for live cell imaging.
Though live cell imaging is rapidly developing,13 fixed cells
are still frequently used.14 The complex can also be applied to
visualize fixed cells as shown in Figure 2b. The cells were fixed
using methanol, resulting in luminescence characteristics similar
to those found in live cell loading. In the profiles shown in
Figure 2, the light intensity of the nucleoli, from both fixed
cell loading and live cell loading, is 3 times higher than that in
the cytoplasm and 2 times higher than that of the nuclear
membrane. This intensity profile permits the attainment of an
exclusive image of the nucleolus by adjusting the excitation
intensity or the gain value of the microscope (Figure 4).
The cells were fixed in methanol, resulting in luminescence
characteristics similar to those found in live cell loading. In the
profiles shown in Figure 2, the light intensity of the nucleoli,
from both fixed cell loading and live cell loading, is 3 times
higher than that in the cytoplasm and 2 times higher than that
of the nuclear membrane. This intensity profile permits the
attainment of an exclusive image of the nucleolus by adjusting
the excitation intensity or the gain value of the microscope
(Figure 4).
The number of the nucleoli in the cell depends on its mitotic
stage and its health status. The nucleolus is visible before mitosis
(interphase), dissolves at prophase, and then becomes discernible
again at telophase.15 As is evident from the images, almost all
cells examined have more than two nucleoli. This is consistent
with the fact that NIH 3T3 cells,16 like some malignant cells,
may have a higher nucleolus number, related to cell proliferative
activity.17 For the nontransformed HDF cells, on the other hand,
a single nucleolus was typically observed.
such extended time periods provides strong evidence that the
observed cells are healthy. Cell viability experiments were
undertaken, coloading with calcein-AM, a dye used to assess
cell viability in most eukaryotic cells. In live cells, the
nonfluorescent calcein-AM is converted to green fluorescent
calcein, following acetoxymethyl ester hydrolysis catalyzed by
intracellular esterases. Cells were examined by microscopy using
appropriate excitation and emission filters so that the red
europium and green calcein emission could be examined
separately. Cells loaded with complex or calcein-AM only were
used as a control. For the coloaded cells (500 µM complex, 3
h of loading), green and red emission was observed in >97%
of the cells examined, consistent with good cell viability.
When the complex was loaded at 4 °C, europium lumines-
cence was still clearly observed, indicating that the complex is
unlikely to enter the cells by endocytosis. The relatively even
distribution of the complex in the cytoplasm also supports this
premise. Molecules and conjugates entering cells via endocytosis
are believed to be initially trapped in endocytotic vesicles and
then transferred into endosomes. Only molecules and conjugates
escaping from the endocytosis pathway can survive to act as
probes.12b
Colocalization experiments were undertaken to confirm that
the bright spots observed in the nucleus were related to the
luminescence of the complex localizing at the nucleolus (Figure
3). Live cells were loaded with the complex or the commercially
available nucleolar stain SYTO RNA-Select. The bright spots
of the image (upper row, red), i.e., the luminescence of the
europium complex (excited here by a laser at 405 nm and
detected after filtration with an LP 505 nm filter; N.B. only the
emission of the europium ion can be recorded at this setting),
colocalize with those (upper row, green) obtained at the setting
for SYTO RNA-Select dye (excited at 488 nm, Ar laser, detected
after filtration with a BP 505-550 filter). The colocalization is
evident as the bright yellow spots at the nucleoli, after merging
(upper row, right column, Figure 3). This strongly suggests that
the targeted organelles are the nucleoli and that the observed
luminescence derives from the europium complex.
The colocalization of fixed cells loaded with the complex
and the SYTO dye demonstrates the excellent potential for this
(13) (a) Shav-Tal, Y.; Singer, R. H.; Darzacq, X. Nat. ReV. Mol. Cell Biol. 2004,
5, 855. (b) Watson, P.; Jones, A. T.; Stephens, D. J. AdV. Drug DeliVery
ReV. 2005, 57, 43.
(14) Molecular Biology of the Cell, 4th ed.; Alberts, B., Johnson, A., Lewis, J.,
Raff, M., Roberts, K., Walter, P., Eds.; Garland Science: New York, 2002;
Chapter 9.
(15) Stevens, B. J. J. Cell Biol. 1965, 24, 349.
(16) Jainchill, J. L.; Aaronson, S. A.; Todaro, G. J. J. Virol. 1969, 4, 549.
(17) Horky, M.; Kotala, V.; Anton, M.; Wesierska-Gadek, J. Ann. N. Y. Acad.
Sci. 2002, 973, 258.
9
2296 J. AM. CHEM. SOC. VOL. 128, NO. 7, 2006

A Eu Complex That Selectively Stains Cell Nucleoli
A R T I C L E S
Figure 5. Luminescence spectrum of NIH 3T3 cells loaded with the
complex. The cells at 80% confluency were loaded with the complex (200
µM) for 5 h, washed with PBS eight times, and then harvested with trypsin.
The mixture was centrifuged, and the pellet was resuspended in PBS (1.5
mL) for spectral measurement. The cells loaded with the complex were
excited at 384 nm. The final spectrum was “baseline corrected”.
Figure 6. CPL spectrum of [EuL]3+ in aqueous solution (black), in cells
(red), and in aqueous L-Asn solution (green). The concentrations of the
complex and L-Asn are 5 and 300 µM, respectively; the cells were the same
sample as described in Figure 5. The excitation wavelength was 384 nm.
complex as a nucleolus stain. The fluorescence image of fixed
cells loaded with the SYTO dye is slightly different from that
obtained by live cell loading. The latter shows more intense
localization of the dye in the whole nucleus compared to its
cytoplasm distribution, and the bright spots (nucleoli) are not
quite so clearly distinguished. Overall, the complex behaves
similarly in each method (Figure 3).
the corresponding Gd complex was prepared and its relaxivity
determined in the absence and presence of serum albumin, a
protein present both in the cell growth medium and within the
cell. The relaxivity of [GdL]3+ was measured to be 7.1 mM-1
s-1 (37 °C, 60 MHz) as the triflate salt and 5.2 mM-1 s-1 as
the citrate adduct. Incremental addition of protein to a 0.1 mM
solution containing the complex caused the relaxivity to rise
steeply. In the presence of 0.1 mM serum albumin the relaxivity
increased to 54 mM-1 s-1. Further addition of protein caused
no significant relaxivity change. When citrate (0.1 mM final
concentration) was added to this solution, the relaxivity fell to
16.2 mM-1 s-1, and the inverse addition experiment gave a
solution with a relaxivity of 13 mM-1 s-1. Taken together, these
behaviors are consistent with reversible binding of the complex
to the protein as suggested in the luminescence experiments.
Citrate binds preferentially to the Eu complex in the presence
of protein. The citrate adduct itself (q ) 0) also binds to protein,
giving rise to a smaller relaxivity enhancement in the ternary
adduct.
Complex Concentration and Speciation. The average
concentration of the complex in cells was estimated by mea-
suring the concentration of europium in a given cell cohort. By
harvesting cells loaded with the complex (200 µM for
5 h) in a 100 mm Petri dish following treatment with trypsin,
the number of labeled cells suspended in a 0.4 mL volume of
PBS (phosphate-buffered saline) was counted using laser sorting
flow cytometry. The total concentration of europium in this
sorted sample (71400 cells) was measured by ICP optical
emission spectroscopy. Given that the studied NIH 3T3 cell may
be estimated to have a mean volume of 4000 µm3, each cell
was calculated to contain 2 × 10-15 mol of Eu (1.2 × 109 Eu
complexes), equivalent to an intracellular concentration of
0.5 µM.
Circularly polarized luminescence (CPL) spectroscopy can
also be a good fingerprint of europium complex speciation.18
CPL spectra for [EuL]3+ were recorded in the presence of citrate,
L-Asp, L-Asn, malate, and oxalate and for the complex localized
in the NIH 3T3 cells (Figure 6). The latter spectrum did not
match well with any single-component adduct (see also Figures
9S and 10S in the Supporting Information), the observed
spectrum partially exhibiting characteristics of the adduct with
L-Asn (negative band at 592 nm, positive band in the ∆J ) 2
manifold). Taken together, the spectral studies are consistent
with intracellular formation of one or more europium complexes
in which the bound waters have been displaced, forming one
or more ternary species with a doubly or triply charged anion
(probably not simply citrate itself, nor any simple phospho-
anion), each of which is likely to be structurally related to
R-hydroxy/R-amino-1,4-dicarboxylic acid derivatives.
As reported earlier, the emission of the complex is particularly
sensitive to citrate, with the ratio of the luminescence intensity
of the ∆J ) 2 (616 nm) to ∆J ) 0 (579 nm) bands increasing
from 3.5 to reach a limiting value of 22.12 In addition, L-Asp,
L-Asn, and malate, each with a 2-hydroxy/amino-1,4-butane-
dicarboxylate structure, have a similar influence on the emission
spectrum of [EuL]3+ (Figure 1 and unpublished data). At the
same time, other common biological anions, such as lactate,
phosphate, bicarbonate, ATP, ADP, AMP, and chloride, change
this ratio only slightly, the value remaining around 3.5. The
addition of t-RNA, CT-DNA, and serum albumin to the aqueous
solution of the complex decreased the luminescence inten-
sity of the complex (see the Supporting Information, Figures
3S-8S). On the other hand, the presence of citrate in the mixture
of DNA/RNA or protein and the complex gave rise to the
characteristic emission of the citrate adduct, no matter which
species (nucleic acid/protein or citrate) was added first.
Conclusions
The luminescence spectrum of the complex localized in cells
exhibits a ratio of the bands centered at 618 and 580 nm (∆J )
2/∆J ) 0 ratio) of 15 ( 1 (Figure 5), consistent with a ternary
adduct involving binding to a species that is structurally similar
to citrate. This ternary “complex” may also undergo reversible
binding to an endogenous protein. To address this aspect further,
A cationic europium(III) complex has been defined with
surprisingly good and selective staining of the nucleolus of NIH
(18) (a) Bruce, J. I.; Dickins, R. S.; Govenlock, L. J.; Gunnlaugsson, T.; Lopinski,
S.; Lowe, M. P.; Parker, D.; Peacock, R. D.; Perry, J. J. B.; Aime, S.;
Botta M. J. Am. Chem. Soc. 2000, 122, 9674. (b) Atkinson, P.; Bretonniere
Y.; Parker, D.; Muller, G. HelV. Chim. Acta 2005, 88, 391.
9
J. AM. CHEM. SOC. VOL. 128, NO. 7, 2006 2297

A R T I C L E S
Yu et al.
poured onto cold concentrated aqueous sodium hydroxide solution (300
cm3) with vigorous stirring. The light yellow precipitate that formed
was collected via filtration. The product was recrystallized from warm
EtOH. The crystals that formed upon standing were filtered and dried
thoroughly to yield the title compound as a pale yellow crystalline solid.
Yield: 4.61 g (90%). Mp: 145-7 °C. Anal. Calcd for C13H9NOS: C,
68.72; H, 3.89; N, 6.10; S, 14.00. Found: C, 68.43; H, 3.95; N, 6.17;
3T3, HeLa, and HDF cells. This complex can be used in both
live cell imaging and fixed cell imaging, with the advantages
of the red emission and the long luminescence lifetime of the
europium ion. This behavior serves to highlight the opportunities
available for the definition of new families of emissive lan-
thanide complexes, suitable for use in selective cell staining
experiments and amenable to examination by time-resolved
microscopy.
1
S, 13.99. H NMR (CDCl3, 500 MHz): δ 8.73 (1H, d, J ) 8.2 Hz,
H4), 8.60 (1H, d, J ) 8 Hz, H6), 7.65 (2H, m, H8,9), 7.48 (1H, m, H7),
7.31 (1H, d, J ) 8.2 Hz, H3), 2.70 (3H, s, CH3). 13C NMR (100 MHz,
CDCl3): δ 25.2 (CH3), 121.7 (C3), 124.8 (C7), 127.3 (C9), 127.5 (C6),
129.6 (C9′), 130.4 (C4), 133.5 (C8), 137.5 (C6′), 138.5 (C4′), 158.0 (C1′),
162.1 (C2), 181.0 (C5). ESMS+: m/z 228 (MH), 250 (MNa), 477
[2MNa].
Experimental Section
General Remarks. All the solvents were distilled before use. All
the other reagents (Aldrich-Sigma) were used as received. RNA-Select
green fluorescent cell stain was obtained from Molecular Probes.
UV/vis spectra and luminescence spectra were recorded with a Perkin-
Elmer spectrometer (UV/vis/NIR Lambda 900) and Fluorolog-3
spectrometer (Horiba- Jobin Yvon) in a 1 cm optical path quartz cell
at room temperature, respectively. Silica gel (TLC standard grade,
Aldrich) was used for column chromatography. ESI-MS spectra were
recorded using a VG Platform II electrospray mass spectrometer.
Accurate masses were determined on a Thermo-Finnigan LTQ FT mass
spectrometer. 1H and 13C NMR was recorded on a Bruker Avance-400
(400 MHz), Varian Mercury-200 (200 MHz), or Varian Unity-500 (500
MHz) instrument. Epifluorescence images were taken on a Zeiss
Axiovert 200M epifluorescence microscope with a digital camera;
confocal images were taken on a Zeiss LSM 500 META confocal
microscope with 405 nm diode laser excitation and an LP 505 emission
filter for europium complex luminescence and with 488 nm argon laser
excitation and a BP 505-550 filter for SYTO dye fluorescence. Flow
cytometric analysis and sorting was conducted using a DakoCytomation
Inc. MoFlo multilaser flow cytometer (Fort Collins, CO). Samples were
interrogated with a 100 mW 488 nm solid-state laser (FSC, SSC) and
a 104 mW 351/363 nm argon ion laser. Fluorescence signals were
detected through an interference filter centered at 450 nm with a 65
nm band-pass. Fluorescence signals were collected in the logarithmic
mode. The data were analyzed using Summit v4.0 (DakoCytomation)
software. CPL spectra were measured at the University of Glasgow,
as reported previously,18 with the assistance of Dr. R. D. Peacock.
Inductively coupled plasma optical emission spectroscopic measure-
ments to determine Eu concentration were made using a Jobin-Yvon
Ultima 2 spectrometer. Relaxivity measurements were made at 37 °C
and 60 MHz on a Bruker Minispec mq60 instrument. The mean value
of three separate measurements was recorded.
2-Bromomethyl-1-azathioxanthone, 3. 2-Methyl-1-azathioxanthone
(1.00 g, 4.41 mmol) was dissolved in CCl4 (30 mL), and the reaction
was heated to 80 °C under argon. N-Bromosuccinimide (392 mg, 0.5
equiv) was added along with benzoyl peroxide (10 mg) with stirring
and the reaction monitored using TLC (SiO2, toluene/DCM/MeOH, 49:
1
49:2) and H NMR, adding further NBS and benzoyl peroxide. After
15 h and the addition of 2.5 equiv of NBS, the crude reaction mixture
was allowed to cool to room temperature and filtered. The solvent was
removed under reduced pressure and the residue purified by column
chromatography on silica gel (toluene/DCM, 90:10), to yield the product
as a bright yellow crystalline solid. Yield: 0.57 g (42%). Mp: 190-2
°C. Anal. Calcd for C13H8NOSBr: C, 50.99; H, 2.62; N, 4.58; S, 10.46.
Found: C, 51.13; H, 2.92; N, 4.81; S, 9.98. 1H NMR (500 MHz,
CDCl3): δ 8.84 (1H, d, J ) 8.1 Hz, H4), 8.58 (1H, dd, J ) 8.0 Hz,
H6), 7.65 (2H, m, H8,9), 7.57 (1H, d, J ) 8.1 Hz, H3), 7.52 (1H, m,
H7), 4.61 (2H, s, CH2Br). 13C NMR (100 MHz, CDCl3): δ 32.6 (CH2),
121.8 (C3), 125.8 (C4), 126.7 (C9), 127.2 (C7), 129.1 (C9′), 130.2 (C6),
133.3 (C8), 137.5 (C6′), 139.2 (C4′), 158.6 (C1′), 161.1 (C2), 180.6 (C5).
ESMS+: m/z 306 (MH), 328 (MNa).
trans-Bis(Phe ester) Cyclen 4. Under argon, 1,7-bis(benzyloxycar-
bonyl)-1,4,7,10-tetraazacyclododecane19 (2.4 g, 4.5 mmol), N-(2-
bromoacetyl)phenylalanine ethyl ester (3.2 g, 10.0 mmol), and potas-
sium carbonate (8.5 g) in DMF (50 mL) were stirred at room temper-
ature overnight. Then DCM (200 mL) was added, and the solution was
washed with water (100 mL × 10) and dried over anhydrous potassium
carbonate. The solvents were removed under reduced pressure, resulting
in the trans-diCbz diamide compound as an oily product that was used
directly in the next step without further purification. Yield: 4.3 g (98%).
1H NMR (CDCl3): δ 7.19-7.34 (20H, m, ArH); 5.08 (4H, s, ArCH2O);
4.81 (2H, m, R-H (Phe)); 4.18 (4H, t, J ) 7.2 Hz, OCH2CH3); 3.32
(8H, br s, cyclen); 3.11 (8H, br s, NCH2 and ArCH2C); 2.76 (8H, br s,
cyclen); 1.19 (6H, t, J ) 7.2 Hz, OCH2CH3). 13C NMR (100 MHz,
CDCl3): δ 171.7, 170.8, 156.7, 136.6, 129.35, 129.25, 128.6, 128.55128.4,
128.2, 128.15, 126.8, 67.3, 61.3, 54.6, 53.1, 37.9, 14.1. HRMS: m/z
calcd for C50H63N6O10(MH) 907.4536, found 907.4531.
Ligand and Complex Synthesis. 6-Methyl-2-thiophenoxynicotinic
acid, 1. 2-Chloro-6-methylnicotinic acid (5.00 g, 29.2 mmol) and
thiophenol (3.80 g, 34.5 mmol) were dissolved in DMF (30 cm3) with
stirring, followed by copper(I) bromide (0.25 g, 17.5 mmol) and
K2CO3 (6.00 g, 43.5 mmol). The mixture was heated for 15 min at
130 °C followed by 18 h at 150 °C, generating a light yellow solution.
The mixture was cooled and treated with water (170 cm3) to give a
yellow suspension, which was washed with ether (3 × 80 cm3). The
aqueous solution was acidified with acetic acid, yielding a very light
yellow precipitate upon cooling, which was filtered, washed with water,
and dried thoroughly to yield the title compound as a pale yellow,
crystalline solid. Yield: 5.90 g (83%). Mp: 170-2 °C. Anal. Calcd
for C13H11NO2S: C, 63.67; H, 4.49; N, 5.73; S, 13.06. Found: C, 63.68;
H, 4.44; N, 5.60; S, 12.98. 1H NMR (400 MHz, CDCl3): δ 13.40 (1H,
br s, -OH), 8.13 (1H, d, J ) 8 Hz, H4), 7.42-7.52 (5H, m, H2′-6′),
7.09 (1H, d, J ) 8 Hz, H5), 2.23 (3H, s, CH3); 13C NMR (100 MHz,
CDCl3): δ 24.8 (CH3), 119.8 (C5), 121.4 (C4), 129.6 (C3), 130.1 (C3′,5′),
131.8 (C1′), 136.1 (C2′,6′), 139.9 (C4′), 160.4 (C2), 160.8 (C6), 167.4
(COOH). ESMS+: m/z 246 (MH), 268 (MNa).
The diester (4.1 g, 4.5 mmol) was dissolved in ethanol (50 mL) and
hydrogenated over 0.1 g of Pd/C (20%, w/w) and 40 psi of hydrogen
at room temperature for 4 days. The solvents were removed under
reduced pressure, giving a colorless oil. The residue was recrystallized
from ethyl ether and DCM, yielding a light yellow solid. Yield: 2.7 g
1
(93%). H NMR (400 MHz, CDCl3): δ 7.30-7.15 (10H, m, ArH);
4.89 (2H, m, R-H (Phe)); 4.20 (4H, q, J ) 7.2 Hz, OCH2CH3); 3.25
(8H, br s, NCH2 CO, and ArCH2C); 2.75 (16H, br s, cyclen); 1.30
(6H, t, J ) 7.2 Hz, OCH2CH3). 13C NMR (100 MHz, CDCl3): δ 172.0,
170.4; 136.5, 129.5, 129.3, 128.4, 127.0; 61.7, 60.9; 53.1, 52.9; 47.4,
37.3, 14.1. HRMS: m/z calcd for C34H51N6O6(MH) 639.3800, found
639.3797.
L. Under argon, the diamide 4 (1.5 g, 2.2 mmol), 2-bromomethyl-
1-azaxanthone (0.34 g, 1.1 mmol), and potassium carbonate (5 g) were
stirred in acetonitrile (40 mL) at room temperature for 48 h. DCM
2-Methyl-1-azathioxanthone, 2. Polyphosphoric acid (60 cm3) was
added to 6-methyl-2-thiophenoxynicotinic acid (5.50 g 22.4 mmol) and
the mixture heated at 120 °C for 4 h under argon with stirring. The
resulting brown liquid was cooled to room temperature and then slowly
(19) Kovacs, Z.; Sherry, A. D. J. Chem. Soc., Chem. Commun. 1995, 185.
9
2298 J. AM. CHEM. SOC. VOL. 128, NO. 7, 2006

A Eu Complex That Selectively Stains Cell Nucleoli
A R T I C L E S
(200 mL) was added, and the solution was washed with water (80 mL
× 3) and dried over sodium sulfate. The product was purified by column
chromatography on alumina with DCM/EtOAc/CH3OH (250:15:4) as
eluant, yielding a light yellow solid product. Mp: 66-8 °C. Yield: 0.32
with 1 mL of trypsin (0.25%, w/v). The mixture was diluted to 10 mL
with PBS and centrifuged. The precipitates were collected and
resuspended to 1.5 mL of PBS for spectral measurements. The cells
were fixed by precooled methanol (-20 °C) for 15 min, washed with
PBS for 5 min twice, and then loaded with the complex in DMEM as
described above. For colocalization experiments, the complex (100 µM)
was incubated with NIH 3T3 cells seeded on coverslips in DMEM for
4 h, and then the medium was replaced with fresh medium in the
presence of SYTO dye (500 nM) for 30 min. The coverslips were then
washed with PBS at least eight times and mounted onto slides for
microscopy. The “fixed cells” were fixed by methanol prior to the above
complex-loading procedure.
1
g (34%). H NMR (400 MHz, CDCl3): δ 8.77 (1H, d, J ) 8.0 Hz,
H1), 8.60 (1H, d, J ) 8.0 Hz, H8), 7.71-7.65 (2H, m, H5 + H7), 7.59-
7.51 (2H, m, H2 + H6), 7.31-7.12 (10H, m, ArH); 4.81 (2H, dd, R-H
(Phe)); 4.15 (4H, q, J ) 7.2 Hz, OCH2CH3); 3.82 (2H, s, ArCH2N);
3.26-3.16 (8H, m, cyclen); 3.04 (4H, s, NCH2CO); 2.82 (4H, br s,
cyclen); 2.77 (4H, br s, cyclen); 2.05 (4H, s, ArCH2C); 1.23 (6H, t, J
) 7.2 Hz, OCH2CH3). HRMS (ESMS+): m/z calcd for C47H58O7N7S
(M+) 864.4197, found 864.4191.
[EuL]3+. The ligand (80 mg, 0.09 mmol) and Eu(CF3SO3)3 (134
mg, 0.26 mmol) were boiled under reflux in acetonitrile (3 mL) for
36 h. The solution was added slowly to diethyl ether (500 mL) and
filtered and the crude solid dissolved in a mixture of DCM (20
mL) and toluene (6 mL). The solution was allowed to slowly evaporate
at room temperature; when most of the DCM had evaporated, a
precipitate was collected, which was dried under vacuum, yielding a
gray solid. Yield: 60 mg (44%). HRMS (ESMS+): m/z calcd for
C49H57O13N7EuF6S3 1314.2305, found 1314.2311. UV/vis (H2O): λmax
) 384 nm; λem ) 446 nm (ligand fluorescence (φem ) 44%), plus
bands associated with Eu emission (φem ) 9%). τEu(H2O) ) 0.32 ms,
τEu(D2O) ) 0.49 ms, q ) 1.1.
Cell viability studies were carried out by loading either the NIH
3T3 or HDF cells with complex (500 µM, 4 h) or with calcein-AM (5
µM, 30 min) and separately with both complex and dye. The cells were
handled in a 12-well plate, washed with PBS, and mounted onto slides
for examination as described above. Cells were examined by microscopy
using the appropriate excitation and emission filters, so that the red
europium emission and green calcein fluorescence could be distin-
guished separately. For the coloaded cells, emission was observed for
>97% of the cells examined, corresponding to both the complex and
the dye, and was judged to be indicative of good cell viability.
Acknowledgment. We thank EPSRC, the Royal Society,
ONE-NE (Durham County Partnership/STBE), and the EC-
EMIL and DiMI networks of excellence for support.
The Gd analogue was made by an analogous method. HRMS
(ESMS+): m/z calcd for C49H57O13N7GdF6S3 1319.2321, found
1319.2379. r1p(H2O, 310 K, 60 MHz) ) 7.3 mM-1 s-1
.
Cell Culture and Complex Loading. NIH 3T3, HeLa, or HDF cells
were incubated under 5% carbon dioxide/air at 37 °C, in DMEM with
4.5 g/L glucose, L-glutamine, and pyruvate, supplemented with 10%
natal bovine serum and a 1% penicillin/streptomycin mixture. For
microscopy, cells seeded on coverslips were incubated with the complex
dissolved in fresh medium at 37 or 4 °C in a 12-well plate for the
indicated time. The cells were washed with PBS at least five times
and mounted onto slides for measurements. For luminescence spectral
measurement, the cells were incubated in a 100 mm Petri dish with
the complex, washed with PBS at least five times, and then harvested
Supporting Information Available: Time course and con-
centration dependence of complex loading with NIH 3T3 cells,
selected images of HeLa and HDF cells costained with RNA-
Select and the europium complex, and absorption, luminescence,
and CPL spectra of the complex in the presence of various
anionic species. This material is available free of charge via
the Internet at http://pubs.acs.org.
JA056303G
9
J. AM. CHEM. SOC. VOL. 128, NO. 7, 2006 2299