Organotellurium scaffolds for mass cytometry reagent development

Article abstract of DOI:10.1039/c5ob00593k

Mass cytometry (MC) is a powerful tool for studying heterogeneous cell populations. In previous work, our laboratory has developed an MC probe for hypoxia bearing a methyl telluride mass tag. The methyl telluride was unoptimized, displaying stability and

Full text of DOI:10.1039/c5ob00593k

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Organotellurium scaffolds for mass cytometry
reagent development†
Cite this: DOI: 10.1039/c5ob00593k
Hanuel Park, Landon J. Edgar, Matthew A. Lumba, Lisa M. Willis and Mark Nitz*
Mass cytometry (MC) is a powerful tool for studying heterogeneous cell populations. In previous work,
our laboratory has developed an MC probe for hypoxia bearing a methyl telluride mass tag. The methyl
telluride was unoptimized, displaying stability and toxicity limitations. Here, we investigate three classes of
organotelluriums as MC mass tags: methyl tellurides, trifluoromethyl tellurides and 2-alkyl-tellurophenes.
NMR was used to compare the stability of these compounds in aqueous and organic solutions and the
compounds were analysed for toxicity in Jurkat cells. The methyl tellurides were moderately stable to
aerobic oxidation in organic solution under dry ambient conditions. The trifluoromethyl tellurides were
stable to aerobic oxidation in organic solution but decomposed in aqueous solution. The 2-alkyl-telluro-
phenes proved to be stable in both organic and aqueous solutions under ambient conditions and showed
limited toxicity (IC50 > 200 µM) in cell based assays. The synthetic feasibility, chemically stability, and limited
toxicity of tellurophenes suggests these groups will be good choices for MC reagent development.
Received 25th March 2015,
Accepted 22nd April 2015
DOI: 10.1039/c5ob00593k
www.rsc.org/obc
molecular weight metal isotopes, usually lanthanides. We
are interested in developing novel MC probes that can be used
Introduction
Characterization of single cells in tissue samples requires a to assay cellular biochemistry. Fig. 1 depicts the general
highly parameterized assay.1 Fluorescence-based flow cytome- requirements of an MC probe. Ideally, the mass tag should be
try (FC) has been the method of choice to study heterogeneous accessible in a high yielding synthesis amenable to isotope
cell populations as it allows for 5–10 parameters to be routi- incorporation, be stable under biologically-relevant conditions
nely analyzed.2 However, FC cannot be used for highly parame- and have low toxicity. Recently we reported an MC-probe
terized assays (>20 parameters) due to the spectral overlap of (Telox) for measuring cellular hypoxia.5 This probe used a
the fluorophores used for analyte detection.3 A solution to this 2-nitroimidazole as the activity group for hypoxic-specific lab-
problem is to substitute the optical detection and fluorescently elling and a methyl telluroether functionality as mass tag for
tagged antibodies in FC for mass detection with an induc- MC detection.5 Tellurium was chosen to be the element for
tively-coupled plasma mass spectrometer (ICP-MS) and detection as it is known to form stable bonds with carbon and
isotope-tagged antibodies. This technology, known as mass it has eight naturally occurring isotopes that can be accessed
cytometry (MC), is capable of detecting numerous bioorthogo- to generate a series of uniquely identifiable, biologically indis-
nal isotopes (theoretically > 100) with single mass unit resolu- tinguishable MC probes using the same chemistry.
tion over multiple orders of magnitude.1 MC allows
Although Telox demonstrated an important concept, the
experiments analogous to flow cytometry but with significantly telluroether functionality was unoptimized, having moderate
greater parameterization. MC has been used to detect and
quantify 34 cellular parameters simultaneously to reveal the
drug response across a human hematopoietic continuum.4
MC experiments can be done by using commercially
available reagents which are composed of antibodies labeled
with metal chelating polymers that bind a range of high
Department of Chemistry, University of Toronto, Toronto 80 St. George St., Canada.
E-mail: mnitz@chem.utoronto.ca; Tel: +1 416-946-0640
†Electronic supplementary information (ESI) available: Full experimental details
for all experiments and relevant compound spectra are included. See DOI:
10.1039/c5ob00593k
Fig. 1 The general requirements and design of a mass cytometry probe.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
stability and a metabolic LD50 value close to the required assay
concentration. These issues have motivated our laboratory to
investigate alternative organotellurium moieties as mass tags
for future activity-based probes.
Table 1 Organotellurium compounds investigated
1
6
The first organotellurium compound was synthesized by
Wöhler in 1840.6 Increasingly organotellurium compounds are
being investigated in living systems, although this area of
research remains underdeveloped.7,8 Tellurium has no known
biological role in prokaryotic or eukaryotic cells. In biological
systems, tellurium metabolism is poorly understood, however
it is presumed to follow the metabolic pathway of its analogue,
selenium. Microorganisms have been found to methylate
inorganic tellurium to volatile or ionic species for excretion.
Experimental evidence of this process is scarce due to the
instability of the metabolites.7 However the number reports of
cellular studies involving aryl, vinylic, alkynyl and alkyl telluro-
ethers in biological systems are increasing.9–14 The majority
of this research has been based upon the ability of aryl telluro-
ethers to mimic glutathione peroxidase activity providing, in
some cases, resistance to oxidative stress, and, in other cases,
disregulating redox homeostasis leading to apoptosis.15,16
Recent murine studies have shown diverse effects from the
expected toxicity of an amino acid-based aryl telluroether
to increased memory in mice treated with an alkyl
telluroether.17,18
2
3
7
8
4
5
9
10
11
Here we describe the synthesis, aqueous/aerobic stability,
and in vitro toxicity of a series of alkyl telluroethers and telluro-
phene functional groups. We hypothesized that the organo-
tellurium exhibiting the most favourable properties with
respect to these three criteria would be an improved mass tag
component for the next generation of MC probes.
Results and discussion
Methyl telluroethers: synthesis and stability
The organotellurium functionality that was initially investi-
gated was the methyl telluroether due to its small size and
ease of synthesis (Table 1, compounds 1–4, 8–9). Aryl telluro-
ethers were not investigated due to numerous reports of their
redox-activity in living systems and their reported instability
under ambient light.19 The methyl telluroethers were syn-
thesized from nucleophilic lithium methyl tellurolate, using a
modified procedure first established by N. Khun, followed by
reaction with the desired electrophile (Scheme 1).20 The syn-
thesis of compound 3 required quenching of the methyl tell-
uroate with water to generate the tellurol prior to the Michael-
style addition to methyl acrylate. The yields of these additions
ranged from 66 to 91%.
Scheme 1 The synthesis of compounds 1–5. The yields of the follow-
ing reactions were: 1 = 66%, 2 = 74%, 3 = 85%, 4 = 91%, and 5 = 50%.
The relative chemical stability of the methyl telluroethers
(1–4) were quantified using 1H NMR by integration of the CH3- lity of the compounds to be monitored without interference
Te signals with respect to a residual DMSO-d5 internal stan- from atmospheric water (Fig. 2).
dard. Samples were prepared as solutions of DMSO-d6 and
Compounds 1–4 all degraded significantly over the course
placed under a slow, continuous stream of dry ambient atmos- of the 24 h incubation. Compound 4 was the most stable alkyl
phere in a clear glass desiccator. This setup allowed the stabi- telluride investigated, degrading approximately 15% over
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 2 1H NMR stability experiment of compounds 1–11. (A) Compounds 1, 2 and 7. (B) Compounds 3, 4, 5 and 6. (C) Compounds 8, 9, 10 and 11.
The organotellurium compounds (∼150 µM) in DMSO-d6 with 1,3,5 – trioxane, the secondary internal standard. The compounds were kept in clear
glass 20 mL vials. The vials were kept in a moisture free environment for 24 hours with continuous supply of ambient atmosphere dried using a
series of bubblers containing phosphoric acid, potassium hydroxide and calcium sulfate. Aliquots were analyzed by 1H NMR and the ratio between
the DMSO-d5 internal standard and the protons β to the tellurium nucleus at time 0 were taken and normalized to generate a degradation plot.
Experimental error was calculated by generating triplicate integration data from individual NMR spectra. This error takes into consideration inte-
gration bias and instrument fluctuations.
24 hours (Fig. 2B). Compounds 2 (Fig. 2A) and 3 (Fig. 2B) the degradation of compound 3. The remaining organic com-
showed the greatest degradation, approximately 75% and 85%, ponents resulting from these degradations could not be identi-
respectively (Fig. 2A and 2B).
fied and may be lost due to their low molecular weight and
The alkyl tellurides are presumed to undergo oxidation volatility. To reduce the propensity for oxidative degradation of
under the experimental conditions. During incubation the the telluroethers the trifluoromethyl telluroethers, 5 and 10,
initially yellow solutions became colourless with the formation and the tellurophenes 6, 7 and 11 were investigated.
of white precipitate, at varying rates. This phenomena has
Trifluoromethyl telluroethers: synthesis and stability
been previously observed and is presumed to be the telluroxide
species forming polymeric structures or the formation of Compound 5 bearing a trifluoromethyl group is expected
TeO2.21,22 In addition, it has been observed previously that have reduced electron density at the tellurium center and thus
solutions of Telox exhibited 1H NMR absorptions consistent should oxidize more slowly. Compound 5 was synthesized
with chalcogen oxidation.5,23 Alkyl telluride compounds are by the generation of tetramethylammonium trifluoromethyl
also known to undergo hemolytic bond cleavage and this may tellurolate in situ by treating tellurium metal with trimethyl-
result in the observed small quantities of dimethyl ditelluride (trifluoromethyl)silane and tetramethylammonium fluoride.27
and dimethyl telluride.24–26 These species are volatile and, in Methyl-4-bromobutyrate was added to the solution to give the
compounds 1 and 2, are observed only in the early time points. product 5 in 50% yield (Scheme 1).
The same species are observed in the 8 hour sample from com-
The stability of the trifluoromethyl telluride 5 was evaluated
pound 3. In addition, methyl acrylamide was produced during under the same conditions as compounds 1–4 (Fig. 2B).
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Comparison of the structurally related compounds 4 and 5 lene using N-bromosuccinimide and silver nitrate gave the
supported our hypothesis that reducing electron density at the known coupling intermediate using the conditions of Wulff
Te centre would stabilize the compound, as no degradation et al.37 This compound was then coupled using CuCl in 30%
was observed over the 24 hour incubation.
BuNH2 with the desired acetylene component. Compound 6
required the addition of 4-pentynoic acid and compound 7
required the addition of 3-butyn-1-ol. The trialkylsilyl pro-
tected diacetylene compounds were deprotected using tetra-
butylammonium fluoride and cyclized into tellurophenes 6
and 7 using a basic solution of Rongalite and tellurium metal
in good yield (Scheme 2).
The stabilities of the two tellurophenes were studied using
the same protocol as the alkyl telluride species (Fig. 2A and B).
Compounds 6 and 7 had improved stability in comparison to
the methyl alkyl telluroethers. Both tellurophene compounds
degraded insignificantly over the 24 hour incubation having
similar stability to the trifluoromethyl telluroether 5.
Tellurophenes: synthesis and stability
Tellurophenes have not been evaluated in biological systems,
and only recently has the first water soluble tellurophene been
reported.28 Tellurophenes possess interesting photophysical
properties and have been investigated as light harvesting
agents for solar cell applications and in materials
chemistry.29–31 The chalcogen analogue, selenophenes, have
been investigated in biological systems with promise as antioxi-
dant molecules. Through computational analysis, ground
state aromaticity of tellurophenes suggest these compounds to
be more stabilized than selenophenes.32 We hypothesized that
the aromatic nature of the tellurophene would provide greater
chemical stability over the telluroether derivatives under the
desired biological conditions.
Benzylamine-conjugated organotellurium derivatives:
synthesis and stability
The tellurophenes were synthesized via the addition of Te2− Future MC probes will require the conjugation of the organo-
to a mono-functionalized diacetylene in a synthesis modified telluriums to biologically relevant functional groups. Here, two
from Stephens and Sweat’s initial report.33,34 In synthesizing conjugation reactions are considered that would provide a
these tellurophenes, the generation of Te2− was a key step. means to label primary amines; a carbamylation and an ami-
Commonly, an aqueous suspension of Te0 is treated with dation reaction. Furthermore, forming benzylamine derivatives
NaBH4. However, we found the use of a basic Rongalite of the organotellurium compounds 2, 4, 5 and 7, lead to com-
(NaHOCH2SO2) solution reproducibly generated Te2− and gave pounds (8–11) with more comparable partition coefficients for
higher yields of the desired tellurophene.35
cell toxicity studies (Schemes 3 and 4).
Compound was synthesized via
The required trialkylsilyl diacetylenes can be generated in
8
a
p-nitrophenyl
excellent yield using the Cadiot-Chodkiewicz cross-coupling carbonate ester generated by the treatment of compound 2 with
reaction (Scheme 2).36 Bromination of triisopropylsilyl acety- p-nitrophenyl chloroformate. The reactive carbonate could be
Scheme 2 The synthesis of compounds 6 & 7. The yields of the reactions were: 6 = 70% & 7 = 66%.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 3 Synthesis of compound 8. The yeild over two steps was 77%.
Scheme 4 Amidation of compounds 4–6 with benzylamine. The yields of the reactions were: 9 = 81%, 10 = 30%, & 11 = 81%.
purified and stored. Compounds 9 and 10 were synthesized Stability in PBS buffer
by hydrolysis of the methyl ester and, after isolation of the
carboxylic acid, coupling proceeded with DCC in the presence of Of the compounds evaluated, the trifluoromethyl telluroether-
NHS and benzylamine. Compound 11 was synthesized analo- amide (10) and the tellurophene-amide (11) exhibited the best
gously to 9 and 10 but required the use of column chromato- stabilities under aerobic conditions. To further validate these
graphy for purification.
compounds as potential MC mass tags, the stability was also
The stabilities of these benzylamine derivatives were studied in a buffered aqueous solution by dissolving the com-
assessed using the aforementioned 1H NMR assay (Fig. 2C). pounds in a 50/50 solution of DMSO-d6/PBS-d buffer. The com-
The rate of degradation of these compounds mirrored those of pounds were kept in an environment exposed to air
the underivatized compounds with the methyl telluride and ambient light at room temperature. 19F NMR was used
species, compounds 8 and 9, degrading 20–25% over the to study compound 10 using a trifluoroacetic acid internal
1
incubation and compounds 10 and 11 being stable over standard while H NMR was used to study the degradation of
the incubation. Interestingly the carbamate
8 was con- compound 11 using the DMSO-d5 internal standard. Disap-
siderably more stable than the parent alcohol 2 suggesting pointingly, as shown in Fig. 3, the trifluoromethyl tellur-
the alcohol may directly contribute to the degradation oether 10 showed 60% degradation after 24 hours. However,
mechanism.
under the same conditions the tellurophene 11 was stable.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
probe concentrations required in these experiments will typi-
cally be below ∼100 µM.5
Conclusion
MC is a powerful analytical tool. Tellurium has valuable charac-
teristics as a mass tag for MC; including eight stable isotopes,
small functional group size and low polarity. We have syn-
thesized and characterized various organotellurium compounds
functionalized for MC probe development. The alkyl telluride
species, although synthetically tractable, have limitations,
mainly due to poor compound stability. For future work, alkyl
telluride species lacking β-protons could be investigated to
improve stability. However, since the tellurophene moiety is
available in good yield, is chemically stable and is sufficiently
non-toxic it warrants further investigation as an MC mass tag.
The applications of this functionality may extend beyond MC;
for example, due to the similarity of tellurophenes and thio-
phenes, this moiety introduces a new building block for medic-
inal chemists.
Fig. 3 Stability studies of compounds 10 and 11 in 50% DMSO-d6,
PBS-d buffer solutions. The 19F NMR spectra of compound 10 and
1H NMR spectra of compound 11 were taken at shown time points.
Compound 10 used trifluoroacetic acid as the internal standard and
compound 11 used DMSO-d5.
Cellular toxicity
To finally investigate these organotellurium compounds as
potential probe moieties for MC, the metabolic toxicity of the
compounds was evaluated. Organotellurium compounds are
often described as toxic, with aryl telluroethers showing cellu-
lar toxicity below 100 µM across a range of cell lines under
different assay conditions.9–11,38,39 Here, we investigated the
toxicity of compounds 8–11 in Jurkat cells after a 24 hour incu-
bation using the metabolic probe WST-1 as per manufacturer’s
instructions. Compounds 8, 9 and 10 are expected to show
degradation over the time frame of the toxicity assay, based on
our NMR stability studies, and, as such these experiments
show the relative toxicities of the compounds and their result-
ing degradation products (Table 2). Compound 8 had an
apparent IC50 value of 610 µM, but with a large experimental
error due to the lack of solubility of compound 8 at higher con-
centrations. Compounds 9 and 10 were more toxic with an
IC50 < 200 µM, however, the tellurophene 11 was less toxic
with an IC50 of 280 µM. These data suggest that, in general,
the alkyl telluroethers and the tellurophenes are less toxic
than previously investigated aryl telluroethers.9–14 The IC50
values of the reported organotellurium compounds provide
promise as mass tags for activity based MC probes since the
Acknowledgements
NSERC (Natural Sciences and Engineering Research Council of
Canada) and Fluidigm Canada are gratefully acknowledged for
financial support.
Notes and references
1 O. Ornatsky, D. Bandura, V. Baranov, M. Nitz, M. A. Winnik
and S. Tanner, J. Immunol. Methods, 2010, 361, 1–20.
2 S. C. De Rosa, L. a. Herzenberg and M. Roederer, Nat. Med.,
2001, 7, 245–248.
3 S. C. Bendall, G. P. Nolan, M. Roederer and P. K. Chattopadhyay,
Trends Immunol., 2012, 33, 323–332.
4 S. C. Bendall, E. F. Simonds, P. Qiu, E. D. Amir,
P. O. Krutzik, R. Finck, R. V. Bruggner, R. Melamed,
A. Trejo, O. I. Ornatsky, R. S. Balderas, S. K. Plevritis,
K. Sachs, D. Pe’er, S. D. Tanner and G. P. Nolan, Science,
2011, 332, 687–696.
5 L. J. Edgar, R. N. Vellanki, A. Halupa, D. Hedley,
B. G. Wouters and M. Nitz, Angew. Chem., Int. Ed., 2014,
11473–11477.
6 F. Wöhler, Justus Liebigs Ann. Chem., 1840, 35, 111–112.
7 L. A. Ba, M. Döring, V. Jamier and C. Jacob, Org. Biomol.
Chem., 2010, 8, 4203–4216.
Table 2 IC50 values of the organotellurium compounds 8–11
8 R. Cunha, I. Gouvea and L. Juliano, An. Acad. Bras. Cienc.,
2009, 81, 393–407.
9 S. Shaaban, F. Sasse, T. Burkholz and C. Jacob, Bioorg. Med.
Chem., 2014, 22, 3610–3619.
10 L. Piovan, P. Milani, M. S. Silva, P. G. Moraes, M. Demasi
and L. H. Andrade, Eur. J. Med. Chem., 2014, 73, 280–285.
11 P. Du, N. E. B. Saidu, J. Intemann, C. Jacob and
M. Montenarh, Biochim. Biophys. Acta, 2014, 1840, 1808–1816.
Compound
IC50 (µM)
8
9
10
11
610 290a
180 60
130 20
280 30
a Compound 8 has a large experimental error due to the lack of
solubility. All cells could not be killed at the maximal concentration
acceptable for the experiment (2 mM).
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
12 W. Cao, Y. Gu, M. Meineck, T. Li and H. Xu, J. Am. Chem. 24 W. Bell, D. J. Cole-hamilton, P. N. Culshaw, A. E. D. Mcqueen
Soc., 2014, 136, 5132–5137. and J. C. Walton, J. Organomet. Chem., 1992, 430, 43–52.
13 S. G. N. Wollenhaupt, A. Thalita, W. G. Salgueiro, 25 R. U. Kirss and D. W. Brown, Organometallics, 1991, 10,
S. Noremberg, G. Reis, C. Viana, P. Gubert, F. A. Soares, 3597–3599.
R. F. Affeldt, D. S. Lüdtke, F. W. Santos, C. C. Denardin, 26 D. W. Brown, B. K. T. Higa and R. W. Gedridge, Organo-
M. Aschner and D. S. Avila, Food Chem. Toxicol., 2014, 64,
192–199.
14 P. Du, U. M. Viswanathan, K. Khairan, T. Buric,
metallics, 1991, 10, 3589–3596.
27 W. Tyrra, N. V. Kirij, D. Naumann and Y. L. Yagupolskii,
J. Fluorine Chem., 2004, 125, 1437–1440.
N. E. B. Saidu, Z. Xu, B. Hanf, I. Bazukyan, A. Trchounian, 28 T. M. Mccormick, E. I. Carrera, T. B. Schon and
F. Hannemann, I. Bernhardt, T. Burkholz, B. Diesel, D. S. Seferos, ChemComm, 2013, 49, 11182–11184.
A. K. Kiemer, K.-H. Schäfer, M. Montenarh, G. Kirsch and 29 A. a. Jahnke, B. Djukic, T. M. McCormick, E. Buchaca
C. Jacob, MedChemComm, 2014, 5, 25.
15 A. Müller, E. Cadenas, P. Graf and H. Sies, Biochem.
Pharmacol., 1984, 33, 3235–3239.
16 V. Jamier, L. a. Ba and C. Jacob, Chem. – Eur. J., 2010, 16,
10920–10928.
17 D. F. Meinerz, B. Comparsi, J. Allebrandt, D. Oscar,
Domingo, C. Hellmann, Y. Lee and D. S. Seferos, J. Am.
Chem. Soc., 2013, 135, 951–954.
30 G. L. Gibson, T. M. Mccormick and D. S. Seferos, J. Phys.
Chem., 2013, 117, 16606–16615.
31 M. Kaur, D. S. Yang, J. Shin, T. W. Lee, K. Choi, M. J. Cho
and D. H. Choi, Chem. Commun., 2013, 49, 5495–5497.
C. Mariano, D. B. Santos, A. Paula, P. Zemolin, M. Farina, 32 M. M. Campos-vallette and R. E. Clavijo C., Spectrosc. Lett.,
A. L. Dafre, J. B. T. Rocha, T. Posser and J. L. Franco,
Springerplus, 2013, 2, 10920–10928.
18 A. Cristina, G. Souza, C. I. Acker, B. M. Gai, J. Sebastião
and C. W. Nogueira, Neurochem. Int., 2012, 60, 409–414.
19 A. Ouchi, T. Hyugano and C. Liu, Org. Lett., 2009, 11, 4870–
4873.
1985, 18, 759–766.
33 D. P. Sweat and C. E. Stephens, J. Organomet. Chem., 2008,
693, 2463–2464.
34 T. J. Barton and R. W. Roth, J. Organomet. Chem., 1972, 39,
66–68.
35 S. Kotha and P. Khedkar, Chem. Rev., 2012, 112, 1650–1680.
20 N. Kuhn, P. Faupel and E. Zauder, J. Organomet. Chem., 36 J. P. Marino and H. N. Nguyen, J. Org. Chem., 2002, 67,
1986, 302, 6–8. 6841–6844.
21 G. Kirsch, M. M. Goodman and F. F. Knapp, Organo- 37 M. X.-W. Jiang, M. Rawat and W. D. Wulff, J. Am. Chem.
metallics, 1983, 2, 357–363. Soc., 2004, 126, 5970–5971.
22 K. Kobayashi, K. Tanaka, H. Izawa, Y. Arai and 38 L. Engman, N. Al-maharik, M. Mcnaughton and G. Powis,
N. Furukawa, Chem. – Eur. J., 2001, 7, 4272–4279. Biochim. Biophys. Acta, 2003, 11, 5091–5100.
23 W. Nakanishi, Y. Ikeda and H. Iwamura, Org. Magn. Reson., 39 M. McNaughton, L. Engman, A. Birmingham, G. Powis and
1982, 20, 117–122.
I. a. Cotgreave, J. Med. Chem., 2004, 47, 233–239.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem.