Synthesis of sulfonamide conjugates of Cu(ii), Ga(iii), In(iii), Re(v) and Zn(ii) complexes: Carbonic anhydrase inhibition studies and cellular imaging investigations

Article abstract of DOI:10.1039/c4dt03206c

Carbonic anhydrase IX (CA IX) is currently generating great interest as a marker of tumour hypoxia and a potential chemotherapeutic target. In order to test the principle that a CA IX inhibitor could be used for targeting PET or SPECT metallic radioisotopes to tumours we have prepared a number of conjugates involving aryl-sulfonamides or an acetazolamide derivative linked to a range of copper, indium, rhenium, 99m-technetium and zinc complexes. Radiolabelled ⁶⁴Cu and ^99mTc analogues of the 'cold' Cu and some of the Re complexes were prepared in good radiochemical incorporation. Inhibition of various human carbonic anhydrase isoforms (I, II, IX and XII) was tested with the 'cold', non-radiolabelled complexes, and compared with an acetazolamide standard (AZA). The molecular structure of a new, tri-sulfonated porphyrin-labeled sulfonamide was determined using synchrotron X-ray crystallography. This journal is

Full text of DOI:10.1039/c4dt03206c

Volume 44 Number 11 21 March 2015 Pages 4775–5212
Dalton
Transactions
An international journal of inorganic chemistry
www.rsc.org/dalton
Featuring the themed issue: Dalton Discussion 15: Metal ions in medical imaging
ISSN 1477-9226
PAPER
Jonathan R. Dilworth, Sofia I. Pascu, Philip A. Waghorn et al.
Synthesis of sulfonamide conjugates of Cu(II), Ga(III), In(III), Re(V) and Zn(II)
complexes: carbonic anhydrase inhibition studies and cellular imaging
investigations

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Synthesis of sulfonamide conjugates of Cu(II),
Ga(III), In(III), Re(V) and Zn(II) complexes: carbonic
anhydrase inhibition studies and cellular imaging
investigations†
Cite this: Dalton Trans., 2015, 44,
4859
Jonathan R. Dilworth,*a Sofia I. Pascu,*b Philip A. Waghorn,*a Daniela Vullo,c
Simon R. Bayly,a Martin Christlieb,d Xin Suna and Claudiu T. Supuranc
Carbonic anhydrase IX (CA IX) is currently generating great interest as a marker of tumour hypoxia and a
potential chemotherapeutic target. In order to test the principle that a CA IX inhibitor could be used for
targeting PET or SPECT metallic radioisotopes to tumours we have prepared a number of conjugates
involving aryl-sulfonamides or an acetazolamide derivative linked to a range of copper, indium, rhenium,
99m-technetium and zinc complexes. Radiolabelled 64Cu and 99mTc analogues of the ‘cold’ Cu and
some of the Re complexes were prepared in good radiochemical incorporation. Inhibition of various
human carbonic anhydrase isoforms (I, II, IX and XII) was tested with the ‘cold’, non-radiolabelled com-
plexes, and compared with an acetazolamide standard (AZA). The molecular structure of a new, tri-sulfo-
nated porphyrin-labeled sulfonamide was determined using synchrotron X-ray crystallography.
Received 17th October 2014,
Accepted 19th January 2015
DOI: 10.1039/c4dt03206c
www.rsc.org/dalton
factor HIF-1α and its targets. Under normoxic conditions
HIF-1α is hydroxylated and binds the ubiquitin ligase von
Introduction
Carbonic anhydrases (CA) are zinc metalloenzymes that cata- Hippel–Lindau (VHL) protein, targeting the HIF-1α for degra-
lyse the interconversion between CO2 and HCO3−, maintaining dation. Loss or mutation in VHL triggers the hypoxic pheno-
pH balance in blood plasma and transporting carbon dioxide type. The tumour associated upregulation of CA IX is the result
out of tissue. Sixteen isoforms of carbonic anhydrase exist to of a strong transcriptional activation of the CA9 gene by
date, located in the cytosol, mitochondria and cellular mem- HIF-1α.3 Studies on non-small cell lung carcinomas have
branes.1 Four transmembrane isozymes are implicated in the shown CA IX association with proteins involved in angiogen-
control of cell proliferation and cellular transformation. Par- esis, apoptosis inhibition and cell–cell adhesion disruption
ticular interest surrounds the transmembrane CA IX isozyme, suggesting a strong relationship of this enzyme to a poor clinical
which has been shown to be overexpressed in certain cancer outcome.4 Therefore, current research interest focuses in the
lines including renal, cervical squamous, ovarian, colorectal, development of CA inhibitors as an approach to the treatment of
bladder and non-small cell lung carcinoma. Strong association cancers where CA IX is over- expressed. Similar interest surrounds
between CA IX expression and intra-tumoural hypoxia has also CA XII, a second tumour associated CA isozyme which has been
been demonstrated.2 At a molecular level the response to identified in renal cell carcinoma.5 It is similarly regulated by
hypoxia is the stabilization and activation of the transcription tumour hypoxia but by different biochemical pathways to the CA
IX isozyme.3 Although CA XII is a less effective catalyst for the
CO2 hydration with respect to CA IX, its catalytic activity is associ-
ated with the acidification of the tumour environment under
hypoxia. The inhibition of these tumour associated CA isozymes
aSiemens Oxford Molecular Imaging Laboratory, Department of Chemistry,
University of Oxford, South Parks Road, Oxford, OX1 3QR, UK.
E-mail: jon.dilworth@chem.ox.ac.uk
bDepartment of Chemistry, University of Bath, Claverton Down, BA2 7AY, UK.
could represent a novel therapeutic approach for the manage-
ment of hypoxic tumours which over- express these proteins.6,7
E-mail: s.pascu@bath.ac.uk
cUniversity of Florence, NEUROFARBA Department, Pharmaceutical Sciences Section,
Selectivity for CA IX over the other CA enzymes has been
investigated through docking studies/molecular modeling and
X-ray crystal structure analysis8 and it was found that the gua-
nidine moiety of an arginine residue is specific to the CA IX
enzyme, which also has a larger active site than the CA I and
CA II isozymes.8 Supuran et al. have carried out some compre-
Via U. Schiff 6, 5019 Sesto Fiorentino, Florence, Italy
dCRUK/MRC Gray Institute for Radiation Oncology and Biology, University of Oxford,
Old Road Campus Research Building, Headington, Oxford, OX3 7DQ, UK
†Electronic supplementary information (ESI) available. CCDC 1028969. For ESI
and crystallographic data in CIF or other electronic format see DOI: 10.1039/
c4dt03206c
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 4859–4873 | 4859

View Article Online
Paper
Dalton Transactions
Fig. 1 Schematic representations of: acetazolamide (AZA), 4-(2-aminoethyl) benzenesulfonamide (ABS), (4-sulfamoylphenylethyl thioureido)
fluorescein and of three known reagents radiolabelled with [18-F], [11-C] or [64-Cu] for PET imaging in vivo.
hensive screening studies on libraries of sulfonamides to find catalytic cycle of CO2 processing.16–18 Recently, the Supuran
the most selective compounds for CA binding. Inhibition group and collaborators also reported on the organometallic
studies with a series of aromatic and heterocyclic sulfonamides [(Cp-R)M(CO)3] (M = Re or 99m-Tc) arylsulfonamide conjugates
showed that CA IX and CA XII are sulfonamide avid.9,10 4-(2- which were found to be selective for targeting of human Car-
Aminoethyl) benzenesulfonamide (ABS) was shown to be an bonic Anhydrase IX.8c This paper reports on the synthesis and
efficient and specific CA IX inhibitor with an approximate characterisation of new conjugates of arylsulfonamides
selectivity of more than 60 fold for CA IX over the CA I isozyme. (grouped under three different ligand classes: thiosemicarbazo-
Another sulfonamide derivative acetazolamide (AZA), a known nate-, amidothiolate- and porphyrin-based) and their corres-
inhibitor of CA IX, is a clinically approved drug (Diamox)11 with ponding metal complexes, i.e. complexes of technetium,
a strong binding affinity for CA IX and CA XII in comparison rhenium, gallium, indium copper and zinc in their non-radio-
with the CA I and CA II enzymes.3,9
active or radioactive form (for 64Cu and 99mTc). Cellular imaging
Derivatives of AZA and ABS have been investigated as tests were also performed on three (fluorescent) porphyrin
bifunctional chelators for imaging. However only very few ligands and one indium porphyrin complex, aiming to shed light
examples of these derivatives labelled with radioisotopes have into the cell inclusion of these conjugates and explore the effect
been reported thus far (Fig. 1).12 Preliminary radiolabelling of these sulfonamide derivatives on CA-mediated mechanisms.
studies with 2-[18F]-3,5,6-trifluor-3′-sulphamoylbenzanilide Their capacity to inhibit a series of carbonic anhydrase isozymes
have demonstrated good correlation between 18F retention and were performed via CA enzyme inhibition assays: the observed
the level of the CA IX expression in vitro and favourable binding constants were used to assess potential candidates for
tumour to muscle uptake ratios for CA IX expressing HT-29 use as SPECT or PET imaging agents for tumour hypoxia.
xenograft models in vivo.13 Bifunctional compounds of these
sulfonamides have been developed to incorporate a fluorescent
tag in the form of fluorescein14 to enable in vitro fluorescence
imaging. The compound (4-sulfamoylphenylethyl thioureido)
fluorescein has already been shown to stain selectively cells
which up-regulate the CA IX isozyme. In silico docking studies
Results and discussion
Arylsulfonamides conjugated with 64-copper(II) or zinc(II)
bis(thiosemicarbazonate) complexes
The complex diacetyl-bis(N4-methylthiosemicarbazonato)-
of the enzyme suggested that hydrogen bonding accounts for
the increased affinity for CA IX over CA II with a guanidine
residue specific to CA IX participating in hydrogen bonding
with the carbonyl oxygen of the fluorescein tail of the inhibi-
tor.8 Upon incorporation of a linker such as EDTA or DTPA for
the binding of copper (Fig. 1) interest could be focused on PET
radiolabelling studies.15 Metal-complexed sulfonamides are
particularly interesting as in some instances the binding
affinity of the conjugate has been shown to exceed that of the
parent sulfonamide due to the simultaneous binding of the
sulfonamide to the catalytic zinc centre within the CA active
site, and of the metal ion to a histidine residue critical to the
copper(II), CuATSM, Fig. 2, has been shown to be selective for
hypoxic tissue in imaging studies, where the PET radioisotopes
60Cu, 61Cu or 64Cu were used,19–21 and, as such it has been
used clinically for the detection of hypoxia in cancer.22,23 It
has been shown24 that Cu(II) bis(thiosemicarbazonates) are
stable in vitro in serum and can also be used to radiolabel
bombesin at room temperature without compromising the bio-
logical targeting capabilities of the protein. Here, analogous
bis(thiosemicarbazide) system was used to form new conju-
gates with arylsulfonamides (see Fig. 2 and Experimental
section). The possible impact on enzyme inhibition of the
4860 | Dalton Trans., 2015, 44, 4859–4873
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Fig. 2 Schematic representations of CuATSM and the new conjugates
emerging from the H2ATSM/en precursor, H2L1, ZnL1 and CuL1.
coordination of zinc or copper to the bis(thiosemicarbazone)
ligand conjugate was explored. Synthesis of the bis(thiosemi-
carbazone)-sulfonamide H2L1 was accomplished in an 82%
yield via the carbodiimide-mediated coupling of 4-sulfamoyl-
benzoic acid (ABS) with the previously described precursor
H2ATSM/en.20 The complexes ZnL1 and CuL1 were produced in
yields of 81% and 44% respectively by room temperature reac-
tion of H2L1 in methanol solution with zinc(II) acetate and
copper(II) acetate respectively. 64CuL1 was prepared in excellent
radiochemical incorporation (>95%) by reaction of a DMSO
solution of H2L1 with 64Cu(II) in aqueous acetate buffer.
Fig. 3 Schematic representations of new protected arylsulfonamide
conjugated ligands and complexes with Re and 99mTc.
Arylsulfonamides conjugated to 99m-technetium(V) or ‘cold’
rhenium(V) complexes
In the case of Tc and Re labelling we used the known MAG3,
DADS and MAGpy ligands (see Fig. 3) to study the effect of
overall charge and ligand structures on CA enzyme inhibition.
The MAG3 and DADS ligands have been used extensively with
99m-Tc and the complexes are rapidly and specifically excreted
from the body via the kidneys,25,26 a desirable attribute for
radiotracers. The technetium(V) or rhenium(V) complexes of
MAG3 and DADS share the same charge (−1) but have structural
differences, while the neutral complex of MAGPy is different in
both charge and structure.25,27 For the sulfonamide conjugates,
high efficiency of CA IX binding requires a favourable steric fit
into the inhibitor binding site of the CA IX enzyme28 and
secondary interactions such as H-bonding may be important.
In this study, three known ligands for Tc and Re (MAG3,
DADS and MAGPy), all bearing a pendant carboxylic acid
group were synthesised and are reported. ABS (Fig. 1) and
Fig. 4 Schematic representations of the new protected acetazolamide
conjugated ligands and complexes with Re(V) and 99mTc(V).
glycinated acetazolamide (Gly-AZA) containing
a reactive
amino group available for linking the sulfonamide were first
synthesised. The proligands H3L2Trt, H2L3Trt and H2L4Trt
(where Trt = trityl, Fig. 3 and 4) were obtained via standard
amide formation using a coupling agent (EDC or BOP, Experi-
mental section). These conjugates were then successfully
labelled with 99mTc as the Tc(V) mono-oxo dervatives. Because
technetium and rhenium are in the same group of the periodic
table and share similar chemical and biological properties,29
the bioassay results of the rhenium and technetium complexes
are directly comparable. Therefore, due to the need to use a
‘cold’ HPLC confirmation method for the 99mTc complexes
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 4859–4873 | 4861

View Article Online
Paper
Dalton Transactions
Fig. 5 Structural representations of the free base porphyrin derivatives investigated.
and to proceed to further studies on CAs inhibtion studies the coupling to the porphyrin scaffold using either the NHS acti-
isostructural rhenium complexes were also synthesised and vated ester or using BOP as a coupling agent with the por-
fully characterised, as described in Experimental section.
phyrin precursor of choice, see Experimental section and ESI.†
Fig. 5 gives the schematic representation of the new free base
porphyrin derivatives synthesised.
Arylsulfonamide conjugates with porphyrins and
metalloporphyrins
To explore if metalloporphyrins conjugated to sulfonamides Molecular structure of H2L9 by synchrotron X-ray diffraction
would also target CA IX a range of porphyrins (Fig. 5) conju-
gated to ABS or GlyAZA were synthesised together with their
indium(III) and gallium(III) complexes. These were of interest
as they provide the possibility of using the intrinsic fluo-
rescence of the porphyrins to study cellular uptake mecha-
nisms and tissue distributions in vivo. The peripheral
porphyrin substituents were selected deliberately aiming to
Crystals of trisulfonated porphyrin H2L9 were grown by the
slow infusion of acetone into an aqueous THF solution (1 : 1
v/v THF–water) of the compound. The crystals produced extre-
mely weak but promising X-ray diffraction patterns only when
a synchrotron X-ray source was used (SRS Daresbury, Station
9.8). The small crystal size (0.02 × 0.02 × 0.35 mm), the large
size of the unit cell and asymmetric unit, the weak scattering
vary the overall charge and water solubility. The amide and sul-
of the atoms involved and the extensive disorder of solvating
fonamide monofunctionalised tetraarylporphyrin or tripyridyl-
water molecules made solving and refining this data set extre-
mely challenging. The molecular structure is shown in Fig. 6
arylporphyrin ligands (H2L7 through to H2L13) were
synthesised from the carboxylate using standard coupling reac-
(CCDC 1028969). The sulfonamide bond lengths and angles
tions with the appropriate amino sulfonamide. Whilst ABS is
commercially available as the free amine, the structure of AZA
is not suitable for direct amide coupling. To achieve a reliable
nucleophilic amine source a Boc-glycine linker was introduced
according to Scheme 1. Deprotection of the Boc group with
TFA yielded a final functionalised AZA derivative suitable for
Scheme 1 Schematic representation for the synthesis of derivatised
acetazolamide for the aryl carboxylate-functionalised porphyrin coup-
i
ling: (a) conc. HCl, EtOH, reflux 3 h; (b) Boc-glycine, NEt3, Butyl chloro-
formate, MeCN, −10 °C to rt. 12 h; (c) TFA, rt. 3 h.
Fig. 6 Ball and stick representation of the molecular structure of H2L9.
4862 | Dalton Trans., 2015, 44, 4859–4873
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
are all in accordance with published X-ray crystal structures for which showed aqueous solubility, for physiologically dominant
aromatic sulfonamide structures.30 Though the data set was hCA I and II and cancer-associated hCA IX and XII by assaying
31
extremely weak, a solution was found and the structure deter- the CA catalysed hydration of CO2 and the data obtained is
mination allowed the confirmation of the connectivity and summarised in Table 1. The entry numbers 1–20 used in the
thus the identity of this compound, in close agreement with following text refer to those listed in Table 1.
1H and 13C NMR, and ESI-MS data, and thus the formation of
In the cases of the bis(thiosemicarbazonates) studied (i.e.
the desired functionalised porphyrin complex. The short with Compound entries with no’s 1–3 in Table 1), the free
amide C–N bond (C(45)–N(5) 1.26 Å) found in H2L9 is ligand and the metal complexes all show low Ki values for
indicative of significant delocalization over the amide group, CA IX and CA XII inhibition, that for the Cu complex being
and extensive disorder in the sulfonamide arm necessitated almost equivalent to ABS (given at the bottom of the table, and
modeling over several sites during refinement. Despite used as a control) itself. However, the discrimination between
this, the overall conectivity is clear and shed light into the CA IX and the other isoforms is poor. The anionic Re com-
molecular structure of the first water soluble porphyrin- plexes under entries 4–6 appear to be active, and promising
sulfonamide derivative. The crystal packing is described by a substrates to be used in this context however they show selecti-
series of H-bonding interactions (Fig. S1 in ESI†) which vity differences that are difficult to rationalise on the basis of
network between solvent water molecules, sulfonate groups structure or targeting group. The rhenium complex under
and sulfonamide amine groups, with H-bonding distances entry 5 with a pendant sulfonamide has the best overall per-
ranging between ca. 2.15 Å (between the sulfonamide amine formance in this assay. Intriguingly, the free base porphyrin
nitrogen and water molecule) for (N(6)–O(13)) and ca. 2.77 Å H2L7 and its corresponding indium complex, which possess
(between a sulfonate oxygen and water molecule) for (O(8)–O no sulfonamide functionality, both show reasonably strong
(14)). The intermolecular interactions observed include some binding affinities for the CA enzymes but in both cases show
weak aromatic interactions (with short range interactions of poor selectivity for CA IX and XII over both CA I and II. Intro-
ca. 2.75 Å) whereby π–π stacking takes place via the T-shape duction of the sulfonamide entity ABS (to give compound
stacking of phenyl rings on adjacent porphyrins (Fig. S2 in H2L8) increases the binding affinity for the CA IX and XII
ESI†).
enzymes and dramatically increases the selectivity for these
enzymes over the non-target CA I and II. In terms of CA IX
inhibition with the porphyrins and metal complexes it is reas-
suring that the porphyrin derivative (see entry 7 in Table 1)
with no sulfonamide shows no selectivity for CA IX whereas
CA IX inhibition assays
Enzyme inhibition data was determined, for all those com-
pounds which could be isolated in satisfactory yields and
Table 1 Enzyme inhibition studies (binding affinity data Ki (nM)) for sulfonamide series with calculated selectivity ratiosa,b
Ki (nM)
Selectivity ratios
CA I/CA IX
No
Compound
CA I
CA II
CA IX
CA XII
CA II/CA IX
CA XII/CA IX
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Control
Ref. 14b
Control
H2L1
335
11.4
33.1
432
1030
1020
78
470
9.3
9.0
62
28
9.0
105
89
1270
82
87
54
38
68
435
38
49
52
59
86
518
39
79
33
73
6.8
6.1
87
88
85
84
80
46
71
72
720
71
66
73
42
35
792
45
97
3.2
5.7
—
5.4
0.4
3.6
4.1
11.6
0.8
1.0
1.0
20.0
11.6
1.9
6.1
11.6
9.2
15.8
7.3
4.6
5.7
16.8
1.0
63.6
10
54.2
7.5
0.33
1.0
11.2
12.5
0.9
0.9
0.2
0.7
0.8
1.0
0.1
1.0
1.0
1.0
1.9
1.0
1.7
1.9
1.3
1.4
0.7
1.0
1.5
1.1
1.2
0.1
0.02
—
ZnL1
CuL1
[ReL2]−
[ReL3]−
[ReL5]−
H2L7
1180
1110
1190
46
111
818
841
46
5750
841
770
651
84
0.5
InClL7
H2L8
91
15.1
15.2
22.1
0.7
13.2
22.1
15.7
12.5
1.4
1.0
7.7
13.3
1.0
4.9
1080
442
126
2650
442
451
820
433
391
2970
655
77
GaClL8
InClL8
H2L9
InClL9
H2L10
H2L11
GaClL11
InClL11
H2L12
H2L13
InClL13
ABS
39
3986
520
81
160
12
2100
250
1300
AZA
FTES
25
24
0.5
1.9
45
a ABS = arenebenzylsulfonamide, AZA = actetazolamide, FTES = (4-sulfamoylphenylethyl thioureido) fluorescein. b Errors in the range of 5–10%
of the reported value, from three different determinations. CA I and CA II are Human (cloned) isozymes. CA IX is the catalytic domain of the
human, cloned isozyme. CA XII is the catalytic domain of the human recombinant isozyme, CO2 hydrase assay method.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 4859–4873 | 4863

View Article Online
Paper
Dalton Transactions
the free base compound shown under entry 9 (having one aryl group has for the CA enzymes prevents the sulfonamide
replaced by a pendant arylsulfonamide, H2L8) shows strong groups from binding preferentially to the CA IX and CA XII
discrimination between CA IX and other isoforms and is one enzyme isoforms. Several of the compounds do however show
of the most effective compounds of those screened in this particularly favourable binding affinities such as the free base
study. However the introduction of either indium or gallium to porphyrins H2L8 and the sulfonate variant H2L9.
give the metal complexes results in a loss in selectivity particu-
Docking studies, in line with those already reported by
larly with the indium complexes. Disappointingly, from the per- Supuran et al. will be carried out in future investigations to
spective of using these compounds for PET or SPECT imaging elucidate in full the structure activity relationships and the CA
incorporation of a metal–halide centre (e.g. In(III)–Cl or Ga(III)– inhibition mechanism.8b
Cl) this generally appears to reduce the selectivity for CA IX over
the other isoforms significantly. Free-base sulfonated porphyrin Confocal fluorescence imaging tests
H2L9 (entry 12) and its corresponding In–Cl complex (entry 13)
provide an exception to this with the In complex showing both To assess the binding potential of the sulfonamide groups to
stronger inhibition and higher selectivity. As with the metal the CA IX receptors, confocal fluorescence studies were per-
complexes mentioned above there is no obvious correlation of formed with compounds H2L8, H2L9 and H2L10 on the HCT 116
the observed binding/selectivity characteristics with structure. It colon carcinoma cell line. The HCT 116 line has been trans-
is likely that subtle factors involving secondary interactions fected to over-express the CA IX enzyme (without the need for
within the binding pocket are dictating the overall inhibition hypoxic culturing) and denoted CA IX positive. The empty-vector
patterns. Based on these results complexes under entries 5 and cell line, i.e. showing no expression of the CA IX enzyme, was
13 are under current investigations in our laboratories, and are denoted CA IX negative and was used as a control.
being taken forward for in vitro and in vivo studies using the
Cells lines were cultured according to known protocols40
99mTc and 111In labelled variants respectively. For the AZA-deriva- and as described below. Cells were seeded on to glass cover
tised metallo–porphyrin compounds (H2L11, H2L12 and H2L13) slips and left to adhere for 12 h overnight. Cells were incu-
we observed a lower binding affinity to the CA XI and XII bated with free-base porphyrin compounds H2L8, H2L9, H2L10
enzymes, whilst for the ABS metallo–porphyrin species we see an at 10 μM concentrations for 2 h at 37 °C. After 2 h, the cells
increased binding affinity for the CA XII enzyme. The shift from were washed three times with PBS and confocal images
H2TPP to the pyridinium-functionalised porphyrin species recorded using a Zeiss LSM 510 META microscope, irradiating
results in a reduction in the selectivity for CA IX and XII over at 405 nm with emission collected above 565 nm. In each case
both CA I and II but the binding affinity to CA IX and CA XII is a sample of 5 images were recorded. Images shown in Fig. 7
still favourable for both AZA- (H2L13) and ABS- (H2L10) functiona- and 8 and ESI† represent a fair sample of the recorded images.
lised species.
The two most notable observations are: (a) the different por-
Introduction of indium to the tetrapyrrole ring results in phyrin species investigated, whereby R = Ph (H2L8), 4-O3SC6H4
complete lack of selectivity towards CA IX and XII with a (H2L9) and MeNC5H4 (H2L10) show distinct cellular distri-
+
diminished binding affinity to both enzymes. This suggests butions; and (b) there is no distinguishable difference between
that either the free base porphyrin provides additional non- the uptake in the positive CA IX expressing cell line and the
covalent bonding interactions that enhance the overall binding empty vector cell line. Cellular imaging suggests that the uptake
affinity and selectivity towards the CA IX and CA XII enzymes and localisation is strongly driven by the porphyrin unit and not
or that the ubiquitous CA enzymes I and II provide additional by the sulfonamide groups. The H2TPP-type derivative H2L8
interactions with the metallo porphyrins which enhance the shows strong cytosol uptake with no nuclear staining.
binding interactions with these enzymes more so than with CA
The sulfonate species H2L9 shows only very weak internalis-
IX and CA XII. Shifting from H2TPP to the sulfonate species ation, as may be expected from the negative charge associated
(H2L9) still shows a strong affinity for CA IX and XII over both with the conjugate, whilst the pyridinium analogue H2L10
CA I and II but the binding affinity for all the enzymes is dra- shows strong nuclear uptake in both cell lines. For compound
matically reduced suggesting that the sulfonate groups act to H2L10, nuclear uptake was confirmed by colocalisation with
inhibit binding, conversely introduction of the metal ion dra- the nuclear stain DAPI (as shown in Fig. 8). The nuclear
matically increases the binding affinity and selectivity for uptake of the pyridinium species mirrors the nuclear localis-
CA IX and XII over CA I and II. In general, all of the porphyrin ation observed by confocal fluorescence in the case of a posi-
species show a uniform medium strength binding affinity to tively charged tetra(methylquinolinium) porphyrin identified
CA IX and CA XII enzymes. All sulfonamide species show an as a G-quadruplex interactive agent.42 In line with observations
enhanced selectivity for CA IX and CA XII over CA I and CA II from the binding affinities measurements, it appears that the
in comparison to the porphyrin species alone indicating that porphyrin unit dominate the cellular uptake of the conjugate.
the presence of the sulfonamide porphyrin conjugate has a The porphyrin sulfonamides still had some affinity for the CA
positive effect towards improving selectivity. The selectivity enzymes. However, the precise discrimination in terms of
ratios and binding affinities however are much poorer than the specific selectivity for the tumour expressing CA IX (and CA
control sulfonamides ABS and AZA on their own and the FTES XII) enzyme(s) with respect to the other isoforms is not sup-
ligand suggesting that the intrinsic affinity that the porphyrin ported by this study.
4864 | Dalton Trans., 2015, 44, 4859–4873
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Fig. 8 Confocal fluorescence imaging showing the cellular uptake at
37 °C and 2 h incubation time: (a) H2L10 (λex = 405 nm, λem > 565 nm);
(b) DAPI nuclear stain (λex = 405 nm, λem = 465–535 nm); (c) overlay
showing the simultaneous imaging with H2L10 and DAPI nuclear stain;
(d) overlay of H2L10, DAPI nuclear stain and brightfield image. Scalebar =
micrograph width = 150 µm.
selectively to the membrane bound CA IX enzymes. The selecti-
vity of the fluorescein derivative is only observed by confocal
fluorescence after incubation for 24 h with a clear delineation
of the cell membrane in CA IX expressing cells. As such uptake
studies with the sulfonate compounds were repeated with 24 h
incubation at 37 °C for H2L9 and InClL9. After 24 h it is
observed that internalisation is now strongly evident with
uniform uptake throughout the cytoplasm. No apparent differ-
ence between the HCT 116 CA IX positive and negative cell
lines is observed. However it was noted that the cell lines after
24 h incubation showed clear signs of cell death, and uptake
mechanisms independent of CA IX expression cannot be ruled
out. Further studies will be required to measure the toxicity of
Fig. 7 Confocal fluorescence imaging: cellular uptake images of H2L8
(a) and (b); H2L9 (c) and (d); H2L10 (e) and (f) in HCT 116 (CA IX positive,
denoted HCT 116+ and CA IX negative denoted HCT 116−) cell lines.
Typical bright-field images for the confocal fluorescence micrographs
are provided in ESI†. Conditions are described in the text.
Radiolabelling tests for CA IX binding by the porphyrin 111In
(Cl)L9
Whilst the porphyrin species do not show the same favourable these compounds. Incubation at concentrations orders of mag-
nitude lower to reduce toxicity results in much weaker fluo-
rescence and subsequently cellular uptake is not discernible
differentials between CA IX and XII over both CA I and II as
the small sulfonamide groups on their own, the medium
strength binding affinities and reasonable selectivity ratios are from background autofluorescence. Whilst the porphyrins are
fluorescent their quantum yields are still very weak (0.06 and
not discouraging for all of the derivatives. Of particular interest
is the sulfonate derivative which when present as the metal
0.03 for H2L9 and its corresponding indium derivative InClL9,
complex InClL9 shows between 5–10 fold increased selectivity respectively) when compared to fluorescein (0.95) and so low
concentrations (nanomolar range) of the compounds cannot
for CA IX over CA I and CA II. Whilst the H2TPP and pyridi-
be efficiently detected by fluorescence. To investigate further
nium species show by confocal fluorescence rapid cellular
uptake by a mechanism independent of CA IX expression, the the porphyrins at lower concentrations radiolabelling studies
have been performed.
sulfonate species displayed less cellular association. This is
probably a result of the negative charge on the porphyrin. This
Compound H2L9 was radiolabelled using an adapted
may make it more suitable for CA binding in the same manner method with respect to previously published conditions42 with
111InCl3 and in vitro uptake studies carried out in the same
as the fluorescein derivative. FTES shows little cellular uptake
two cell lines as used for confocal fluorescence tests. After 1, 3,
as a result of its negatively charged free acid group but binds
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 4859–4873 | 4865

View Article Online
Paper
Dalton Transactions
over 18 h of only 1.5%. This again suggests that there is no CA
mediated mechanism of cell inclusion of these porphyrin
conjugates.
A closer inspection of the CA binding data (Table 1)
suggested that the sulfonamide species InClL9 has an
increased binding affinity of 5–10 fold in a CA IX expressing
cell line based on binding affinities alone. This should lead to
increased membrane and potentially increased internalised
associated activity. On the basis of radiolabelling studies no
such increase in uptake or binding is observed, which is in
agreement with the confocal fluorescent studies and suggests
strongly that uptake and/or binding is not receptor-mediated
via initial carbonic anhydrase recognition. Furthermore, the
anionic sulfonamides show no apparent recognition for the
CA IX receptors and the weak uptake is most probably driven
by the intrinsic accumulation of the porphyrin entity. Given
the rapid and strong cellular uptake of the cationic and
neutral porphyrin species it was felt unlikely that they would
confer any better differential, with their cellular uptake domi-
nated purely by the porphyrin entity and by a pathway inde-
pendent of sulfonamide binding to CA IX enzymes.
Fig. 9 Uptake of [111-In]-labelled H2L9 species in HCT116 (CA IX posi-
tive) and HCT116 (CA IX negative) cell lines showing % activity associated
with the cell membrane and internalised for each cell line. Experiments
were carried out in triplicate and error bars indicate one standard devi-
ation of the mean.
Conclusions
A range of new sulfonamide conjugated metal complexes invol-
ving bis(thiosemicarbazonate), amidothiolate and porphyrin
6 and 18 h incubation (37 °C) of compound H2L9 with both ligands have been prepared with a view to developing PET and
the CA IX positive and CA IX negative cell lines the cells were SPECT agents based on metallic radionuclides to target CA IX
washed with PBS (to remove activity in the supernatant), 1 M expression in vivo. These compounds were all fully character-
HCl (to remove activity associated with the cell membrane) ised and successful, radiolabelling protocols resulting in a
and 1 M NaOH (to lyse the cells of internalised activity) high radiochemical incorporation were established for the
sequentially, and the supernatant, membrane associated 64-Copper and 99m-Technetium complexes. Those complexes
activity and internalised associated activity isolated respect- with solubility and stability in an aqueous environment suit-
ively and measured separately. Cell percentage uptake is given able for the CA enzyme inhibition assay were screened for
as the amount of activity internalised i.e. the activity in the activity, by testing their ability to inhibit CA IX and other iso-
1 M NaOH wash with respect to the total activity and the forms of the enzyme in vitro. A series of sulfonamide por-
amount of activity associated with the membrane i.e. the phyrin conjugates has been synthesised and fully
activity in the 1 M HCl wash with respect to the total activity. characterised. Water soluble analogues have been developed
The uptake data suggests low uptake in line with the fluo- for greater biocompatibility with no sign of acid hydrolysis of
rescence images. Identical weak membrane associated activity the amide bonds under sulfonating conditions and no sign of
is measured for both cell lines suggesting that no sulfonamide alkylation of the thiadiazole ring under methylating con-
binding occurs in the CA IX positive cell line (Fig. 9).
ditions. A library of sulfonamide derivatives has been studied
Cell uptake is observed which may be mediated by sulfona- for their binding affinities to specific CA receptors to establish
mide binding but is more likely to arise from the intrinsic their affinity for the poor prognostic CA IX and CA XII
uptake of the porphyrin conjugate. A higher percentage of enzymes. Our findings for the rhenium derivatives are compar-
radioactivity is associated with the CA IX positive cell line, able with those recently reported for organometallic [(Cp-R)-
which might suggest some positive sulfonamide response, but M(CO)3] (M = Re or 99m-Tc) Arylsulfonamide conjugates inves-
there is potential for variation in the amount of protein in the tigated for the selective targeting of human Carbonic Anhy-
different cell population, despite the same initial cell concen- drase IX.8c In general the porphyrin sulfonamide conjugates
tration. Detailed protein assays are beyond the scope of this displayed positive selectivity for the CA IX and CA XII enzymes
work and were not completed so the difference in internalised though metal insertion into the porphyrin resulted in loss of
activity as a result of difference in cell population cannot be selectivity and reduced binding affinity. Conversely the sulfo-
immediately ignored. Comparable changes in cell uptake in nate porphyrins showed the reverse effect showing improved
both cell lines over the time course are observed for the total binding and increased selectivity for CA IX and CA XII on
cell associated activity with a difference in associated activity indium complexation. Fluorescent studies in a CA IX positive
4866 | Dalton Trans., 2015, 44, 4859–4873
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
and CA IX negative cell line indicate (after 2 h and 24 h incu-
Diacetyl(N-4-methylthiosemicarbazonato)(N-4-aminoethyl-
bation at 37 °C) that there is no obvious difference in the cellu- amidophenyl-4-sulphamoylthiosemicarbazonato)zinc(II), ZnL1.
lar binding or uptake between the two cell lines for the H2L1 (40 mg, 85 µmol) and Zn(OAc)2·2H2O (22 mg, 100 µmol)
sulfonyl porphyrin species. A strong structural correlation with were stirred together in methanol (5 ml) at room temperature
cellular uptake is observed with the neutral tetraphenyl por- for 2 h forming a bright yellow solution. Distilled water (20 ml)
phyrins showing strong cytoplasm uptake, the cationic por- was added and the solution evaporated to remove the metha-
phyrins showing nuclear accumulation whilst the sulfonyl nol. The solution was extracted with ethyl acetate (3 × 10 ml)
porphyrins display only weak cell uptake. Radiolabelling and the combined organic fractions dried (MgSO4), evaporated
studies with an indium(III)-labelled sulfonated ABS porphyrin and dried under vacuum to give ZnL1 as a bright yellow solid
identified identical weak membrane binding for the CA IX (37 mg, 69 µmol, 81%).
positive and CA IX negative cell lines and comparable cell
1H NMR (300 MHz, DMSO-d6) δ ppm 8.72 (t, J = 4.5 Hz, 1H,
internalisation suggesting that the conjugates cellular distri- CH2NHCvO), 7.97 (d, J = 8.5 Hz, 2H, ArH), 7.88 (d, J = 8.5 Hz,
bution and uptake was dominated by the porphyrin core. 2H, ArH), 7.49 (s, 2H, SO2NH2), 3.51–3.44 (m, 2H, CH2), other
Although these results showed no direct correlations between CH2 obscured by water peak, 3.16 (d, J = 5.0 Hz, 2H,
the structures and physical properties of the complexes and CH3NHCvS), 2.20 (s, 3H, CH3CvN), 2.18 (s, 3H, CH3CvN).
the observed activities, two compounds based on the por- HRMS ES+: found 535.0320, calc. for [M
+
H]+
phyrin motif were identified as being good candidates to take (C16H23N8O3S3Zn) 535.0347. HPLC Rt = 10.12 min.
forward to cellular and in vivo investigations.
Diacetyl(N-4-methylthiosemicarbazonato)(N-4-aminoethyl-
amidophenyl-4-sulphamoylthiosemicarbazonato)copper(II),
CuL1. H2L1 (40 mg, 85 µmol) and Cu(OAc)2·2H2O (22 mg,
100 µmol) were stirred together in methanol (5 ml) at room
temperature for 2 h forming a brown solution. Distilled water
(20 ml) was added and the solution evaporated to 1/3 volume
Experimental procedures
General
All reagents were purchased from Sigma-Aldrich, Merck to produce CuL as a brown precipitate, which was collected by
Chemicals or Alfa-Aesar and were used as received unless filtration and dried under vacuum (20 mg, 38 µmol, 44%).
otherwise stated. Na[99mTcO4] was obtained from a 99Mo/99mTc HRMS ESI+: found 534.0364, calc. for [M
generator (GE Healthcare, UK). H2ATSM and H2ATSM/en were (C16H23N8O3S3Cu) 534.0351. HPLC Rt = 11.98 min.
prepared according to published methods.32,33 ESI contains
+
H]+
further details on HPLC and radioHPLC methods A–C used Syntheses of technetium and rhenium conjugates
and intermediates’ syntheses.
Diacetyl(N-4-methylthiosemicarbazone)(N-4-aminoethyl-
N-(2-Oxo-2-((2-oxo-2-((4-sulfamoylphenylethyl)amino)ethyl)-
amidophenyl-4-sulphamoylthiosemicarbazone), H2L1. H2ATSM/ amino)ethyl)-2-(tritylthio)acetamide TrtMAG3-ABS, TrtH3L2.
en (150 mg, 0.51 mmol), 4-sulfamoylbenzoic acid (104 mg, Succinimido-2-((triphenylmethyl)thio)acetate (1) was prepared
0.51 mmol), N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide following literature procedures.3,34 To a solution of (1) (3.44 g,
hydrochloride (119 mg, 0.62 mmol) and hydroxybenzotriazole 7.98 mmol) in 1,2-dimethoxyethane (35 mL) and DMF (18 mL)
(84 mg, 0.62 mmol) were placed under N2 in a schlenk flask was added a solution of triglycine (1.52, 8.04 mmol) and
and DMF (5 ml) added. The yellow mixture was stirred for 18 h NaHCO3 (1.38 g) in distilled water (18 mL). The resulting solu-
at room temperature and water (50 ml) added generating a tion was stirred at room temperature for 3 h and then concen-
yellow precipitate. This was collected by filtration, washed with trated in vacuo to remove 1,2-dimethoxyethane. Dilution with
water and diethylether, and dried in vacuo to provide H2L distilled water (35 mL) and treatment with 50% aqueous citric
(197 mg, 0.42 mmol, 82%).
acid (6 mL) gave a white solid of Trt-MAG3 (2), which was
1H NMR (300 MHz, DMSO-d6) δ ppm 10.34 (s, 1H, recrystallized in ethanol (2.90 g, 71.9%).
NHNvC), 10.24 (s, 1H, NHNvC), 8.88 (t, J = 5.3 Hz, 1H,
To a solution of (2) (393 mg, 0.777 mmol) in DMF (5 mL)
CH2NHCvS), 8.60 (t, J = 5.3 Hz, 1H, CH2NHCvO), 8.39 (q, J = was added ABS (155 mg, 0.774 mmol), EDC (149 mg,
4.40 Hz, 1H, CH3NHCvS), 8.00 (d, J = 8.50 Hz, 2H, ArH), 7.89 0.777 mmol), HOBT (104 mg, 0.770 mmol) and TEA (24 µL).
(d, J = 8.50 Hz, 2H, ArH), 7.49 (s, 2H, SO2NH2), 3.78 (td, J = 5.7, The reaction was left in ice bath for 1 h and then at RT for
5.3 Hz, 2H, CH2NHCvS), 3.54 (td, J = 5.7, 5.3 Hz, 2H, 1 day. The solvent was evaporated thereafter. The residue was
CH2NHCvO), 3.02 (d, J = 4.51 Hz, 2H, CH3NHCvS), 2.23 (s, re-dissolved in ethyl acetate (10 mL) and washed with 10%
3H, CH3CvN), 2.21 (s, 3H, CH3CvN). 13C{1H} NMR citric acid, 3% NaHCO3 and saturated brine, 20 mL each. The
(75.5 MHz, DMSO-d6) δ ppm 178.5 (CvS), 178.1 (CvS), solids, which precipitated when washed with NaHCO3 and
165.9 (CvO), 148.4 (CvN), 148.0 (CvN), 146.3 (Ar), 137.2 saturated brine, were collected. The organic solution was col-
(Ar), 128.0 (Ar), 125.6 (Ar), 43.9 (CH2), 38.9 (CH2), lected and dried under vacuum. The solutions of the raw
31.3 (CH3NH), 11.8 (CH3CvN), 11.7 (CH3CvN). MS ES+: product were combined and recrystallized from ethanol–water
found 495.1, calc. for [M + Na]+ (C16H24N8NaO3S3) 495.1; ES−: to give TrtH3L2 (440 mg, 82.6%).
found 471.1, calc. for [M − H]− (C16H23N8O3S3) 471.1. HPLC Rt
= 10.86 min.
1H NMR (d6-DMSO, δ, 300 MHz): 2.75 (t, J = 7.2 Hz,
2H), 2.84 (s, 2H), 3.25 (m, 2H), 3.61 (m, 4H), 3.68 (d, J = 4.9 Hz,
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 4859–4873 | 4867

View Article Online
Paper
Dalton Transactions
2H), 7.21–7.37 (m, 17H), 7.71 (d, J = 8.3 Hz, 2H), 7.92
N-(2-Oxo-2-((2-oxo-2-((2-oxo-2-((2-oxo-2-((5-sulfamoyl-1,3,4-
(t, J = 5.3 Hz, 1H), 8.12 (t, J = 5.7 Hz, 1H), 8.22–8.25 (m, thiadiazol-2yl)amino)ethyl)amino)ethyl)amino)ethyl)amino)-
2H). 13C NMR (d6-DMSO, δ, 75 MHz): 35.2, 36.4, 42.5, 42.6, ethyl)-2-(tritylthio)acetamide, TrtMAG3-glyAZA, H3L5Trt. Trt-
42.8, 42.8, 66.4, 126.1, 127.3, 128.5, 129.5, 129.5, 142.5, 144.0, MAG3 (2) (50.5 mg, 0.100 mmol) was added to DMF (5 mL)
144.4, 168.2, 169.1, 169.5, 169.6. ES/MS(+): m/z = 688.2915 and mixed with glyAZA (35.1 mg, 0.100 mmol), EDC (20.0 mg,
(Calc. m/z = 688.2264) [M + H]+. HPLC (Method B): Rt
11.88 min.
=
0.105 mmol), HOBT (13.5 mg, 0.100 mmol) and TEA (20 µL).
The reaction was left in ice bath for 1 h and then at room
N,N′-(3-Oxo-3-((4-sulfamoylphenylethyl)amino)propane-1,2- temperature for 1 day. The solvent was evaporated and the
diyl)bis(2-(tritylthio)acetamide, TrtDADS-ABS, H3L3Trt. Ethyl residue was re-dissolved in ethyl acetate (10 mL). The solution
2,3-bis(triphenylmethylthioacetylamino)propanoate (3) was was washed with 1 M HCl, 3% NaHCO3 and saturated brine,
prepared following the literature procedure.35 To 3 (1.05 g, 20 mL each. The organic solution was collected and dried
1.33 mmol) in a 1 : 1 mixture of THF and water (80 mL), NaOH under vacuum. The raw product was recrystallized from
(100 mg, 2.5 mmol) was added. The mixture was heated to ethanol–water to give H3L5Trt (41.0 mg, 56.6%).
reflux for 3 h under N2. The mixture was adjusted to pH = 2 by
1H NMR (d6-DMSO, δ, 300 MHz): 2.84 (s, 2H), 3.63 (d, J =
addition of 6 M HCl and was concentrated to 30 mL under 5.6 Hz, 2H), 3.71 (d, J = 5.0 Hz, 2H), 3.77 (d, J = 5.9 Hz, 2H),
reduced pressure. The precipitate that formed was collected by 4.08 (d, J = 5.5 Hz, 2H), 7.24–7.34 (m, 15H), 8.11–8.19 (m, 4H),
filtration and dried under vacuum to give Trt-DADS (4) 8.35 (s, 2H), 13.11 (s, 1H). 13C NMR (d6-DMSO, δ, 75 MHz):
(825 mg, 84.2%). To a solution of 4 (147 mg, 0.199 mmol) in 36.3, 42.2, 42.4, 42.7, 66.4, 127.3, 128.5, 129.5, 144.5, 161.5,
DMF (5 mL) was added ABS (40.0 mg, 0.200 mmol), EDC 164.9, 168.2, 169.4, 169.6, 170.0, 171.5. ES/MS(−): m/z =
(37.8 mg, 0.200 mmol), HOBT (26.7 mg, 0.198 mmol) and TEA 723.1169 (Calc. m/z = 723.1478) [M − H]−. HPLC (Method A):
(6 µL). The reaction was left in ice bath for 1 h and then at RT Rt = 9.88 min.
overnight. The solvent was evaporated and the residue re-dis-
N5-(2-Oxo-2-((5-sulfamoyl-1,3,4-thiadiazol-2-yl)amino)ethyl)-
solved in ethyl acetate (20 mL). The solution was washed with N1-(pyridine-2-ylmethyl)-2-(2-(tritylthio)acetamido)pentanedi-
10% citric acid, 3% NaHCO3 and saturated brine, 20 mL each. amide, TrtMAGpy-glyAZA or H2L6trt. Trt-MAGPy (5) (79.7 mg,
The organic layer was collected and dried under vacuum. The 0.144 mmol) was mixed with BOP (76.4 mg, 0.173 mmol) and
raw product was combined and recrystallized from ethanol– DIPEA (90 µL, 0.52 mmol) in DMF (5 mL) in ice bath for
water to give TrtH3L3 (108 mg, 59.0%).
20 min., then glyAZA (synthesised as above, (c) (60.7 mg,
1H NMR (d6-DMSO, δ, 300 MHz): 2.70 (m, 2H), 2.74 (s, 2H), 0.173 mmol) was added. The mixture was stirred at RT over-
2.80 (s, 2H), 3.11 (m, 2H), 3.21 (m, 2H), 4.11 (m, 1H), 7.19–7.34 night. The solvent was evaporated, and 3 M HCl (20 mL) and
(m, 32H), 7.71 (d, J = 8.4 Hz, 2H), 7.86 (t, J = 5.7 Hz, 1H), ethyl acetate (20 mL) were used consecutively to wash off the
8.00–8.07 (m, 2H). 13C NMR (d6-DMSO, δ, 75 MHz): 35.2, 36.4, impurities. The solid was further purified by silica gel chrom-
36.5, 40.4, 41.1, 53.3, 66.3, 126.1, 127.2, 128.5, 129.5, 142.5, atography using DCM–MeOH (9/1 by volume) to give H2L6Trt
144.0, 144.4, 144.5, 167.8, 168.1, 169.6. ES/MS(+): m/z = (32.3 mg, 29.0%).
941.541 (Calc. m/z = 941.284) [M + Na]+. HPLC (Method B in
ESI†): Rt = 15.37 min.
1H NMR (d4-CD3OD, δ, 500 MHz): 1.87–2.15 (m, 2H),
2.29–2.42 (m, 2H), 3.08 (d, J = 1.26 Hz, 2H), 4.15–4.25 (m, 2H),
N′-(Pyridine-2-ylmethyl)-N5-(4-sulfamoylphenylethyl)-2-(2-tri- 4.27 (dd, JA = 7.88, JB = 5.99 Hz, 1H), 4.53 (s, 2H), 4.61 (br, 2H),
tylthio)acetamido)pentanediamide, TrtMAGpy-ABS, H2L4Trt. 7.09–7.52 (m, 17H), 7.80 (td, JA = 7.72, JB = 1.58 Hz, 1H), 8.48
Trt-MAGpy (5) was synthesized following the literature pro- (d, J = 4.41 Hz, 1H). 13C NMR (d4-CD3OD, δ, 125 MHz): 28.7,
cedure.36 Compound 5 (111 mg, 0.200 mmol) was added to 32.6, 37.2, 43.6, 45.4, 54.7, 68.5, 123.0, 123.9, 128.2, 129.2,
DMF (5 mL) and mixed with ABS (40.0 mg, 0.200 mmol), EDC 130.8, 139.0, 145.6, 149.7, 158.9, 163.3, 166.6, 170.7, 171.4,
(37.8 mg, 0.200 mmol), HOBT (26.7 mg, 0.198 mmol) and TEA 173.6, 175.7. ES/MS(+): m/z = 795.1804 (Calc. m/z = 795.1812)
(6 µL). The reaction was left at RT for 1 day. The solvent was [M + Na]+. HPLC (Method A): Rt = 12.04 min.
evaporated and the residue was washed with 0.1 M HCl (3 ×
Syntheses of rhenium complexes
10 mL) and ethyl acetate (2 × 10 mL). The solid was dried
under vacuum to give H2L4Trt (37.8 mg, 25.7%).
Sodium
1H NMR (d6-DMSO, δ, 300 MHz): 1.77 (m, 2H), 1.96 (t, J = amido)ethyl)amido)ethyl)-2-thioacetamido-oxorhenium(V), Na-
7.7 Hz, 2H), 2.65 (t, J = 6.8 Hz, 2H), 2.76 (s, 2H), 3.14 (td, JA = [ReOMAG3-ABS], (5) (240 mg,
Na[ReOL2]. Trt-MAG3
N-(2-oxo-2-((2-oxo-2-((4-sulfamoylphenylethyl)-
6.6 Hz, JB = 19.3 Hz, 2H), 4.09 (td, JA = 5.4 Hz, JB = 7.8 Hz, 1H), 0.474 mmol) and sodium acetate (258 mg, 3.14 mmol) were
4.24 (d, J = 5.7 Hz, 2H), 7.08–7.33 (m, 18H), 7.63 (d, J = 8.0 Hz, mixed in dry methanol (60 mL), followed by addition of precur-
2H) Ar-H; 7.86 (t, J = 5.5 Hz, 1H), 8.13 (d, J = 7.6 Hz, 1H), 8.36 sor [ReOCl3(PPh3)2] (395 mg, 0.474 mmol). After being refluxed
(d, J = 3.9 Hz, 1H), 8.41 (t, J = 5.7 Hz, 1H). 13C NMR (d6-DMSO, for 4 h, the solution turned orange. After cooling, the solution
δ, 75 MHz): 26.9, 31.5, 34.8, 35.7, 40.0, 41.8, 53.6, 66.9, 123.9, was filtered and evaporated to dryness. The residue was chro-
124.3, 125.9, 126.7, 127.7, 129.1, 129.3, 141.6, 143.2, 143.8, matographed on silica gel with DCM–methanol. The orange
143.9, 144.1, 155.3, 170.2, 172.8, 173.2. ES/MS(+): m/z = eluent was collected and dried to give Na[ReO-MAG3] (6) (185 mg,
736.2386 (Calc. m/z = 736.2628) [M + H]+. HPLC (Method A): 80.6%). After 6 (48.4 mg, 0.100 mmol) was dissolved in phosphate
Rt = 12.04 min.
buffer (10 mL, 0.1 M, pH = 6.0), ABS (20.0 mg, 0.100 mmol) in
4868 | Dalton Trans., 2015, 44, 4859–4873
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
acetonitrile (1 mL) was added to the solution, followed by EDC
5-[4-(N-Propyl benzamide)]-10,15,20-tri(4-pyridyl)-porphyrin
(19.1 mg, 0.100 mmol) and TEA (14 μL). The mixture was stirred was synthesised employing general amide coupling method B
at room temperature overnight. After the solvent was evaporated, (ESI†). 5-(4-Carboxyphenyl)-10,15,20-tris(4-pyridyl)porphyrin 9
the residue was chromatographed on silica gel with DCM–metha- (see ESI†) (75 mg, 0.113 mmol), DIPEA (38.7 mg, 52.3 μL
nol (3/1 by volume). The orange eluent was collected and dried to 0.300 mmol), propylamine (7.97 mg, 8.04 μL, 0.135 mmol),
give Na[ReOMAG3-ABS] (37.1 mg, 55.6%).
BOP (67.5 mg, 0.146 mmol), CHCl3 (10 mL). 5-[4-(N-propyl
1H NMR (d4-CD3OD, δ, 500 MHz): 2.79–2.91 (m, 2H), 3.41 (t, benzamide)]-10,15,20-tri(4-pyridyl)-porphyrin was purified via
J = 7.3 Hz, 2H), 3.78 (d, J = 17.2 Hz, 1H), 4.02 (d, J = 17.3 Hz, silica gel chromatography (0% to 5% MeOH in CHCl3). Yield:
1H), 4.26 (d, J = 18.0 Hz, 1H), 4.28 (d, J = 18.1 Hz, 1H), 4.40 (d, 65.2 mg, 0.093 mmol, 82%.
J = 18.0 Hz, 1H), 4.65 (d, J = 16.6 Hz, 1H), 4.74 (d, J = 18.2 Hz,
1H NMR (300 MHz, d6-DMSO, 25 °C): δ 9.40 (d, 6H, J = 6.1,
1H), 4.99 (d, J = 16.6 Hz, 1H), 7.39 (d, J = 8.3, 2H), 7.82 (d, J = CHpy), 9.14 (m, 8H, CHpyrr), 8.91 (m, 6H, CHpy), 8.65 (t, 1H, J =
8.2, 2H), 13C NMR (d4-CD3OD, δ, 125 MHz): 34.3, 41.1, 41.5, 43.5, 5.5, CONH), 8.27 (d, 2H, J = 7.9, ArH), 8.15 (d, 2H, J = 8.1, ArH),
49.9, 54.5, 57.7, 127.8, 130.5, 142.7, 144.0, 191.4, 194.7, 195.8. 3.52
(m,
2H,
CONHCH2CH2CH3),
1.81
(m,
2H,
ES/MS(−): m/z = 644.0281 (Calc. m/z = 643.9832) [M]−. IR: CONHCH2CH2CH3), 1.07 (t, 2H, J = 6.9, CONHCH2CH2CH3),
978 cm−1 (RevO), 1616 cm−1 (CvO). UV-Vis: λabs(MeOH)/nm −2.95 (br s, 2H, ring NH). ESI-MS calcd for C44H35N8O [M +
(ε/M−1 cm−1) = 224 (2.52 × 104), 274 (sh, 5.73 × 103), 339 (sh, 3.23 × H]+ 703.2928, found 703.2935.
102), 402 (1.27 × 102), 485 (22). HPLC (Method C): Rt = 3.03 min.
5-[4-(N-Propyl benzamide)]-10,15,20-tri(4-pyridyl)-porphyrin-
N,N′-(3-Oxo-3-((4-sulfamoylphenylethyl)amido)propane-1,2- chloroindium(III), InClL7. This was prepared using general
diyl)bis(2-thioacetamido)oxorhenium(V) Na[ReODADS-ABS], metal complexation method 1. H2L7 (50.0 mg, 0.072 mmol,
or Na[ReOL3]. Na[ReODADS] (7) was synthesized from (4) NaOAc (35.0 mg, 0.429 mmol), InCl3 (32.0 mg, 0.143 mmol),
according to the method described for Na[ReOMAG3] (6); HOAc (10 ml). Yield: 51.2 mg, 0.061 mmol, 84%.
(50.1 mg, 70.5%). The title compound was synthesized by
1H NMR: (300 MHz, d6-DMSO, 25 °C): 8.99 (s, 8H, CHpyr),
coupling 7 with ABS using the same method as described 8.50 (t, 1H, J = 5.4, CONH), 8.31 (2 × d, 2 × 2H, ArH), 8.22 (m,
above for the MAG3 system (35.8 mg, 51.5%). 1H NMR (d4- 6H, o-Ph), 7.85 (m, 9H, o,m,p-Ph), 3.43 (m, 2H,
CD3OD, δ, 500 MHz): 3.03 (m, 2H), 3.20 (t, J = 7.4, 2H), 3.78 (d, CONHCH2CH2CH3), 1.75 (m, 2H, CONHCH2CH2CH3), 1.02 (t,
J = 17.4 Hz, 1H), 3.84 (d, J = 17.1 Hz, 1H), 4.14 (d, J = 17.4 Hz, 3H, CONHCH2CH2CH3). ESI-MS Calculated for C48H36InN5O
1H), 4.43 (d, J = 16.9 Hz, 1H), 4.75 (d, J = 8.4 Hz, 1H), 5.00 (d, [M − Cl]+, 812.1875, found 812.1855; HPLC: (method A in
J = 12.9 Hz, 1H), 7.50 (d, J = 8.6, 2H), 7.91 (d, J = 8.5, 2H), (N.B.: ESI†): Rt = 10.64 min.
the 1H resonance of the other NCH2CHN proton was over-
5-[4-(N-(4-Sulphamoylphenethyl)benzamide)]-10,15,20-tri-
lapped with other peaks). 13C NMR (d4-CD3OD, δ, 125 MHz): phenylporphyrin, [H2TPP-ABS], H2L8. The title compound
32.8, 40.1, 42.9, 43.5, 59.2, 69.6, 126.3, 129.1, 141.3, 142.4, was synthesised employing general peptide coupling method
194.2, 195.8, 196.4. ES/MS(−): m/z = 632.9900 (Calc. m/z = 1. 8 (see ESI†) (250 mg, 0.331 mmol), 4-(2-aminoethyl)benzene-
632.9941) [M]−. IR: 962 cm−1 (RevO), 1581 cm−1 (CvO). sulfonamide (100 mg, 0.500 mmol), DIPEA (64.6 mg, 87.1 μL,
UV-Vis: λabs(MeOH)/nm (ε/M−1 cm−1) = 226 (2.93 × 104), 249 0.500 mmol), DMF (10 mL). Purification was achieved by silica
(sh, 1.10 × 104), 300 (sh, 3.15 × 103), 332 (sh, 1.58 × 103), 402 gel chromatography (5% MeOH in CHCl3). Yield: 236 mg,
(1.83 × 102), 489 (11). HPLC (Method 3): Rt = 4.27 min.
N-(2-Oxo-2-((2-oxo-2-((2-oxo-2-((2-oxo-2-((5-sulfamoyl-1,3,4-
0.281 mmol, 85%.
1H NMR (300 MHz, d6-DMSO, 25 °C): δ 8.99 (t, 1H, J = 5.5,
thiadiazol-2yl)amido)ethyl)amido)ethyl)amido)ethyl)amido)- CONH), 8.81 (m, 8H, CHpyrr), 8.29 (d, 2H, J = 8.1, ArH), 8.23
ethyl)-2-(thio)acetamido Na[ReOMAG3-glyAZA] or Na[ReOL5]. (d, 2H, J = 8.4, ArH), 8.17 (m, 6H, o-ArH), 7.84 (d, 2H, J = 8.4,
The title compound was synthesized by coupling 6 with glyAZA ArH), 7.77 (m, 9H, m,p-ArH), 7.55 (d, 2H, J = 8.4 ArH), 7.37 (s,
using an analogous method to that described for the synthesis 2H, SO2NH2), 3.69 (m, 2H, CONHCH2CH2), 3.06 (t, 1H, J = 7.2,
1
of (45.1 mg, 64.1%). H NMR (d4-CD3OD, δ, 500 MHz): 2.91 (d, CONHCH2CH2). 13C NMR (125 MHz, d6-DMSO, 25 °C): δ 166.3,
J = 6.9, 2H), 3.82 (d, J = 17.2 Hz, 1H), 4.05 (d, J = 17.2 Hz, 1H), 144.0, 143.9, 142.2, 141.2, 134.2, 134.1, 131.0 (bs), 129.3, 128.1,
4.29 (d, J = 18.6 Hz, 1H), 4.32 (d, J = 18.8 Hz, 1H), 4.42 (d, J = 127.0, 125.9, 120.3, 120.2, 119.1, 34.9, 25.3. Mass Spectrum
18.2 Hz, 1H), 4.73 (d, J = 17.1 Hz, 1H), 4.79 (d, J = 18.1 Hz, 1H), ESI-MS calcd for C53H40N6O3S [M + H]+ 841.2961, found
5.05 (d, J = 17.1 Hz, 1H). 13C NMR (d4-CD3OD, δ, 125 MHz): 841.2768. HPLC: (Method A) Rt = 11.81 min. Elem. Anal.:
δ(ppm) = 41.1, 43.6, 54.5, 56.1, 57.7, 164.61, 191.5, 194.7, 195.8. Found: C; 75.2%, H; 4.9%, N; 10.1%. Calc.: C; 75.7%, H; 4.8%,
ES/MS(−): m/z = 680.9280 (Calc. m/z = 680.9650) [M]−. IR: N; 10.0%.
977 cm−1 (RevO), 1629 cm−1 (CvO). UV-Vis: λabs(MeOH)/nm (ε/
5-[4-(N-(2-Oxo-2-(5-sulphamoyl-1,3,4-thiadiazol-2-ylamino)-
M−1 cm−1) = 221 (1.43 × 104), 270 (sh, 4.61 × 103), 337 (sh, 3.18 × ethyl)benzamide)]-10,15,20-triphenyl-porphyrin, [H2TPP-AZA],
102), 397 (1.03 × 102), 491 (18). HPLC (Method A): Rt = 6.62 min. H2L11. The title compound was synthesised employing general
5-[4-(N-Propyl benzamide)]-10,15,20-tri(4-pyridyl)-porphyrin, amide coupling method 1. 8 (330 mg, 0.436 mmol), glyAZA
[TPyP-propyl,
H2L7]. 4-(10,15,20-Triphenylporphyrin-5-yl)- (180 mg, 0.512 mmol), DIPEA (66.2 mg, 89.2 μL, 0.512 mmol),
benzoic acid37 and 5-(4-carboxyphenyl)-10,15,20-tris(4-pyridyl)- DMF (10 mL). Purification was achieved via silica gel chrom-
porphyrin,38 were synthesised according to literature methods atography (2 to 8% MeOH in CHCl3). Yield: 256 mg,
with analytical data in accord with the literature values (ESI†).
0.292 mmol, 67%.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 4859–4873 | 4869

View Article Online
Paper
Dalton Transactions
1H NMR (500 MHz, d6-DMSO, 25 °C): δ 13.39 (s, 1H, CONH- 148.06, 147.8, 144.1, 141.4, 134.5, 134.4, 134.2, 132.6, 132.5,
thiadiazole), 9.44 (t, 1H, J = 5.5, CONH), 8.84 (m, 8H, CHpyrr), 132.5, 128.2, 127.0, 125.9, 121.4, 121.3, 120.3, 45.5. Mass Spec-
8.40 (s, 2H, SO2NH2), 8.37 (m, 4H, ArH), 8.23 (m, 6H, o-ArH), trum ESI-MS calcd for C49H33InN9O4S2 [M − Cl]+ 990.1130,
7.84 (m, 9H, m,p-ArH), 4.45 (d, 2H, J = 5.6, CONHCH2CONH), found 990.1090. HPLC: (Method A) Rt = 9.99 min.
−2.90 (s, 2H, ring NH). 13C NMR (125 MHz, d6-DMSO, 25 °C):
5-[4-(N-(2-Oxo-2-(5-sulphamoyl-1,3,4-thiadiazol-2-ylamino)ethyl)-
δ 169.6, 166.8, 164.5, 161.3, 144.5, 141.1, 134.2, 134.2, benzamide)]-10,15,20-triphenyl-porphyrin(chlorogallium(III)), [Ga-
133.0, 131.5, 128.1, 127.0, 126.1, 120.3, 120.1, 118.9, 43.0. (Cl)TPP-AZA] or [GaClL11]. The title compound was synthesised
ESI-MS calcd for C49H36N9O4S2 [M + H]+ 878.2326, found employing general metal complexation method 1. H2L11
878.2275. HPLC: (Method A) Rt = 10.82 min. Elem. Anal.: (50.0 mg, 0.057 mmol), NaOAc (28.0 mg, 0.341 mmol), GaCl3
Found: C; 66.8%, H; 3.8%, N; 14.4%. Calc.: C; 67.0%, H; 4.0%, (21.0 mg, 0.114 mmol), AcOH (10 mL). A maroon coloured
N; 14.4%.
solid was obtained. Yield: 45.3 mg, 0.046 mmol, 81%.
1H NMR (500 MHz, d6-DMSO, 25 °C): δ 9.13–8.91 (m, 8H,
or CHpyrr), 8.76 (br s, 1H, CONH), 8.40 (d, 2H, J = 7.3, ArH), 8.31
5-[4-(N-(4-Sulphamoylphenethyl)benzamide)]-10,15,20-tri-
phenylporphyrin(chloroindium(III)),
[In(Cl)TPP-ABS]
[InClL8]. The title compound was synthesised employing (d, 2H, J = 7.5, ArH), 8.22 (m, 6H, o-ArH), 7.86 (m, 9H, m-p-
general metal complexation method 1. H2L8 (74.5 mg, ArH), 4.20 (d, 2H, J = 4.9, CONHCH2). 13C NMR (75.5 MHz,
0.087 mmol), NaOAc (43.6 mg, 0.532 mmol), InCl3 (39.2 mg, d6-DMSO, 25 °C):
δ 165.9, 147.3, 147.0, 143.6, 140.8,
0.177 mmol), AcOH (10 mL). A maroon coloured solid was 134.2, 132.3, 132.2, 128.4, 127.2, 126.1, 120.4, 119.4, 45.1.
obtained. Yield: 69.7 mg, 0.070 mmol, 81%.
ESI-MS calcd for C49H33GaN9O4S2 [M − Cl]+ 944.1347, found
1H NMR (300 MHz, d6-DMSO, 25 °C): δ 9.11 (t, 1H, J = 5.6, 944.1334. HPLC: (Method A) Rt = 10.11 min. Elem. Anal.:
CONH), 9.13–9.01 (m, 8H, CHpyrr), 8.26 (d, 2H, J = 7.8, ArH), Found: C; 60.5%, H; 3.6%, N; 12.2%. Calc.: C; 60.0%, H; 3.4%,
8.19 (m, 8H, o-Ph, ArH), 7.95 (d, 2H, J = 8.2, ArH), 7.89 (m, 9H, N; 12.8%.
m,p-Ph), 7.51 (d, 2H, J = 8.0, ArH), 3.71 (m, 2H, CONHCH2CH2),
5-[4-(N-(4-Sulphamoylphenethyl)benzamide)]-10,15,20-tri-
3.04 (t, 2H, J = 7.6, CONHCH2CH2). 13C NMR (75.5 MHz, (4-sulfonylphenyl)porphyrin, [H2TSPP-ABS], H2L9. The title
d6-DMSO, 25 °C):
δ 166.1, 148.0, 147.8, 147.7, 147.4, compound was synthesised employing the general sulfonation
144.5144.1, 141.9, 141.3, 134.4, 134.2, 133.9, 132.54, 132.51, method (ESI†). H2L7 (70.0 mg, 0.083 mmol), conc. H2SO4
132.4, 130.5, 128.1, 127.1, 125.9, 125.8, 121.5, 121.3, 120.1, 38.2. (2 mL). Yield: 66.6 mg, 0.058 mmol, 70%.
27.4. ESI-MS calcd for C53H38InN6O3S [M − Cl]+ 953.1765, found
1H NMR (500 MHz, d6-DMSO, 25 °C): δ 9.05 (t, 1H, J = 5.7,
953.1789. HPLC: (Method A) Rt = 10.12 min. Elem. Anal.: Found: CONH), 8.82 (m, 8H, CHpyrr), 8.29 (d, 2H, J = 8.0, ArH), 8.21
C; 63.9%, H; 4.0%, N; 8.7%. Calc.: C; 64.4%, H; 3.9%, N; 8.5%. (m, 6H, ArHSO Na), 8.17 (d, 2H, J = 8.1 Hz, ArH), 8.05 (m, 6H,
3
5-[4-(N-(4-Sulphamoylphenethyl)benzamide)]-10,15,20-tri- ArHSO Na), 7.87 (d, 2H, J = 8.1, ArH), 7.61 (d, 2H, J = 7.8, ArH),
3
phenylporphyrin(chlorogallium(III)), [Ga(Cl)TPP-ABS] or 7.37 (s, 2H, SO2NH2), 3.71 (m, 2H, CONHCH2CH2), 3.14 (t, 1H,
[GaClL8]. The title compound was synthesised employing J = 7.5, CONHCH2CH2). 13C NMR (75.5 MHz, d6-DMSO, 25 °C):
general metal complexation method 1. H2L8 (68.5 mg, δ 166.1, 147.8, 144.1, 143.8, 142.5, 141.1, 134.24, 134.16, 133.7,
0.082 mmol), NaOAc (40.1 mg, 0.489 mmol), GaCl3 (28.7 mg, 131.4 (broad), 131.2, 129.1, 125.9, 124.6, 120.1, 119.8, 118.9, 35.6,
0.163 mmol), AcOH (10 mL). A maroon coloured solid was 26.4. ESI-MS calcd for C53H37N6O12S4 [M − 3Na]3−/3 359.0451,
obtained. Yield: 57.3 mg, 0.061 mmol, 74%.
found 359.0467. HPLC: (Method B, ESI†) Rt = 15.20 min.
5-[4-(N-(2-Oxo-2-(5-sulphamoyl-1,3,4-thiadiazol-2-ylamino)-
1H NMR (500 MHz, d6-DMSO, 25 °C): δ 8.92 (m, 8H, CHpyrr),
8.76 (t, 1H, J = 5.5, CONH), 8.32 (d, 2H, J = 7.6, ArH), 8.21 (d, ethyl)benzamide)]-10,15,20-tri(4-sulfonylphenyl)-porphyrin,
2H, J = 7.7, ArH), 8.26 (m, 6H, o-Ph), 7.88 (d, 2H, J = 8.1, ArH), [H2TSPP-AZA] or H2L12. The title compound was synthesised
7.85 (m, 9H, m,p-Ph), 7.49 (d, 2H, J = 7.9, ArH), 3.75 (m, 2H, employing the general sulfonation method (ESI†). H2L10
CONHCH2CH2), 3.15 (t, 2H, J = 7.3, CONHCH2CH2). ESI-MS (50.0 mg, 0.057 mmol), conc. H2SO4 (2 mL). Yield: 48.6 mg,
calcd for C53H38GaN6O3S [M − Cl]+ 907.1982, found 907.1965. 0.041 mmol, 72%.
HPLC: (Method A) Rt = 10.05 min. Elem. Anal.: Found: C;
67.8%, H; 4.3%, N; 8.7%. Calc.: C; 67.5%, H; 4.1%, N; 8.9%.
1H NMR (500 MHz, d6-DMSO, 25 °C): δ 8.90 (t, 1H, J = 5.7,
CONH), 8.88 (m, 8H, CHpyrr), 8.37 (m, 4H, ArH), 8.21 (d, 6H, J =
5-[4-(N-(2-Oxo-2-(5-sulphamoyl-1,3,4-thiadiazol-2-ylamino)ethyl)- 8.1, ArHSO Na), 8.07 (d, 6H, J = 8.0, ArHSO Na), 4.20 (d, 2H, J =
3
3
benzamide)]-10,15,20-triphenyl-porphyrin(chloroindium(III)), [In- 5.6, CONHCH2CONH), −2.91 (s, 2H, ring NH). 13C NMR
(Cl)TPP-AZA] or [InClL11]. The title compound was synthesised (125 MHz, d6-DMSO, 25 °C): 174.9, 171.9, 165.9, 165.3, 147.8,
employing general metal complexation method 1 (ESI†). H2L11 143.9, 141.2, 134.3, 134.2, 133.7, 131.6 (broad), 125.9, 124.2,
(40.0 mg, 0.046 mmol), NaOAc (22.0 mg, 0.273 mmol), InCl3 119.8, 119.7, 119.3, 45.5. ESI-MS calcd for C49H32N9O13S5 [M −
(19.7 mg, 0.091 mmol), AcOH (10 mL). A purple coloured solid 3Na]3−/3 371.3580, found 371.3576. HPLC: (Method B, ESI†)
was obtained. Yield: 34.0 mg, 0.033 mmol, 72%.
Rt = 16.11 min.
1H NMR (500 MHz, d6-DMSO, 25 °C): δ 9.06–9.03 (m, 2H,
5-[4-(N-(4-Sulphamoylphenethyl)benzamide)]-10,15,20-tri-
CHpyrr), 8.99 (m, 7H, CHpyrr, CONH), 8.43 (d, 2H, J = 8.1, ArH), (4-N-methylpyridinium
phenyl)-porphyrin
tri-iodide,
8.36 (d, 2H, J = 8.3, ArH), 8.25 (m, 6H, o-ArH), 7.90 (m, 9H, m-, [H2TMePyP-ABS], H2L10. H2L10 was synthesised employing
p-ArH), 4.20 (d, 2H, J = 5.9, CONHCH2). 13C NMR (125 MHz, d6- the general alkylation method (ESI†). The tripyridylporphrin
DMSO, 25 °C): δ 175.8, 175.0, 172.1, 165.9, 148.12, 148.09, ABS conjugate [H2TPyP-ABS] synthesis described in ESI† (65.0 mg,
4870 | Dalton Trans., 2015, 44, 4859–4873
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
0.077 mmol), MeI (109 mg, 47.9 μL, 0.77 mmol), DMF. Yield: CHpyrr), 9.02–9.11 (br s, 2H, CHpyrr), 8.87–9.00 (br s, 6H, CHpy),
114 mg, 0.090 mmol, 92%. 8.36 (d, 2H, J = 7.5, ArH), 8.30 (d, 2H, J = 7.5, ArH), 7.81 (d, 2H,
1H NMR (500 MHz, d6-DMSO, 25 °C): 9.46 (m, 6H, CHpy), J = 8.0, ArH), 7.55 (d, 2H, J = 8.0, ArH), 7.36 (s, 2H, SO2NH2),
9.11 (m, 8H, CHpyrr), 9.02 (t, 1H, J = 5.4, CONH), 8.99 (m, 6H, 4.66–4.87 (br s, 9H, 3 × CH3), 3.69 (m, 2H, CONHCH2CH2),
CHpy), 8.32 (d, 2H, J = 8.7, ArH), 8.30 (d, 2H, J = 8.8, ArH), 7.81 3.09–3.16 (t, 2H, J = 5.7, CONHCH2CH2). 13C NMR (75.5 MHz,
(d, 2H, J = 8.3, ArH), 7.56 (d, 2H, J = 8.4, ArH), 7.34 (s, 2H, d6-DMSO, 25 °C): δ 166.7, 156.6, 156.5, 147.8, 146.5, 146.4,
SO2NH2), 4.70 (m, 3 × 3H, 3 × N-CH3), 3.70 (m, 2H, 144.8, 144.3, 144.2, 142.9, 133.8, 133.3, 132.4, 132.3, 129.8,
CONHCH2CH2), 3.07 (t, 2H, J = 6.9, CONHCH2CH2), −3.06 (m, 126.2, 126.1, 117.3, 116.8, 48.1, 47.9, 35.1, 26.2. Mass Spec-
1H) 13C NMR (125 MHz, d6-DMSO, 25 °C): δ 166.2, 156.6, trum ESI-MS calcd for C53H42ClInN9O3S [M
−
3Cl]3+/3
156.5, 144.3, 144.2, 143.9, 143.1, 142.1, 134.4, 134.3, 132.1, 344.7281, found 344.7298. HPLC: (Method C in ESI†) Rt
=
129.3, 126.0, 125.8, 124.1, 121.9, 115.4, 114.7, 114.5, 111.8, 12.05 min.
47.9, 34.5, 25.3. ESI-MS calcd for C53H44N9O3S [M − 3I]3+/3
General method for copper-64 radiolabelling
64Cu-ATSM was prepared from H2-ATSM according to a pub-
269.1143, found 269.1143, ESI-MS calcd for C53H44N9O3S [M −
3I]3+/2 444.1717, found 444.1708. HPLC: (Method C) Rt
=
12.22 min. Elem. Anal.: Found: C; 50.3%, H; 3.3%, N; 9.8%. lished method.2,39 64CuL was prepared by adding 64Cu-
Calc. [H2L10 + MeOH]: C; 49.9%, H; 3.7%, N; 9.7%. (OAc)2(aq) (70 μL of a 1 MBq μL−1 solution) to a solution of
5-[4-(N-(2-Oxo-2-(5-sulphamoyl-1,3,4-thiadiazol-2-ylamino)- H2L in DMSO (100 μL, 1 mg mL−1). After 20 min stirring at
ethyl)benzamide)]-10,15,20-tri(4-N-methylpyridiumyl)-porphyrin room temperature HPLC analysis was carried out using a Plati-
tri-iodide, [H2TMePyP-AZA]I3 or H2L13. Compound H2L13 was num C18 column (5 μm particle size, dimensions 250 ×
synthesised employing the general alkylation method 3.0 mm). A 20 min gradient method with acetonitrile–water as
described in ESI.† [H2TPyP-AZA] (50.0 mg, 0.038 mmol), MeI the mobile phase was used: 0 min 5% MeCN; 10 min 95%
(54.3 mg, 23.8 μL, 0.38 mmol), DMF (see ESI†). Yield: 52.9 mg, MeCN; 15 min 95% MeCN; 18 min 5% MeCN, 20 min 5%
0.040 mmol, 88%.
MeCN. Flow rate = 1.0 mL min−1. The radiochemical purity of
1H NMR (300 MHz, d6-DMSO, 25 °C): δ 9.46 (d, 6H, J = 6.6, 64CuL (Rt = mins) was 99%. 0.9% saline solution (to <5% v/v
CHpy), 9.36 (t, 1H, J = 5.4, CONH), 9.26–9.05 (m, 8H, CHpyrr), ethanol) for use in PET imaging experiments. The specific
8.93–9.03 (m, 6H, CHpy), 8.40 (d, 2H, J = 8.0, ArH), 8.36 (d, 2H, activity of the administered tracer was approximately 20 MBq
J = 7.8, ArH), 4.71 (2 s, 3 × 3H, 3 × N-CH3), 4.43 (d, 2H, J = 5.8, mL−1 (0.2 MBq μg−1 of H2L).
CONHCH2CONH), −3.05 (br s, 1H, ring NH). Mass Spectrum
ESI-MS calcd for C49H41N12O4S2 [M − 3I]3+/3 308.0640, found
308.0621, ESI-MS calcd for C49H41N12O4S2 [M
General method for indium-111 radiolabelling
−
3I]3+/2
The ligand was prepared as a 1.0 mg mL−1 solution in DMSO
or distilled water. For indium labeling of water soluble por-
phyrins 10 μL of the stock porphyrin solution was diluted with
90 μL of pH 4.5 sodium acetate buffered solution and heated
at 115 °C for 30 min in an Eppendorf tube in the presence of
20 μL 111InCl3 (<10 MBq per experiment). For indium
labelling of tetraphenyl porphyrin derivative 10 μL of the stock
solution was diluted with 90 μL of AcOH and heated at 120 °C
for 1 h in the presence 10 μL of a 0.1 M NaOAc (aq.) solu-
tion and 20 μL 111InCl3 (<10 MBq per experiment). AcOH was
then driven off under nitrogen and the residue re-suspended
in 50 μL of CHCl3. The CHCl3 layer was washed repeatedly
with 100 μL aliquots of distilled water until no further activity
was associated with the aqueous washings. For all porphyrin-
based compounds the 20 min HPLC gradient methods
(analytical HPLC) of was performed by one of three methods
below:
462.0960, found 462.0993. HPLC: (Method C) Rt = 13.08 min.
5-[4-(N-(4-Sulphamoylphenethyl)benzamide)]-10,15,20-tri-
(4-sulfonylphenyl)-porphyrin(chloroindium(III)),
[In(Cl)-
TSPP-ABS], [InClL9]. The title compound was synthesised
employing general metal complexation method 3. H2L9
(40.0 mg, 0.035 mmol), InCl3 (15.5 mg, 0.070 mmol) in pH 4.5
NaOAc buffer (5 mL). Yield: 29.9 mg, 0.023 mmol, 66%.
1H NMR (300 MHz, d6-DMSO, 25 °C): δ 9.07 (t, 1H, J = 5.4,
CONH), 8.94 (s, 8H, CHpyrr), 8.29 (s, 4H, ArH), 8.18 (d, 6H, J =
8.0, ArHSO Na), 8.08 (d, 6H, J = 8.2, ArHSO Na), 7.60 (d, 2H, J =
3
3
8.0, ArH), 7.32 (d, 2H,
J
=
8.0, ArH), 3.65 (m, 2H,
CONHCH2CH2), 3.00 (t, 2H, J = 7.4, CONHCH2CH2) 13C NMR
(75.5 MHz, d6-DMSO, 25 °C): δ 166.3, 147.72, 147.69, 147.5,
147.4, 145.9, 144.1, 141.7, 140.3, 134.2, 134.0, 132.6 (broad),
128.1, 125.9, 125.7, 124.2, 120.8, 120.4, 35.0, 25.0. Mass Spec-
trum ESI-MS calcd for C53H35ClInN6O12S4 [M − 3Na]3−/3
408.3313, found 408.3345. HPLC: (Method B) Rt = 13.23 min.
5-[4-(N-(4-Sulphamoylphenethyl)benzamide)]-10,15,20-tri-
(4-N-methylpyridiniumylphenyl)-porphyrin (chloroindium-(III))
General method for technetium-99m radiolabelling
trichloride [In(Cl)TMePyP-ABS]Cl3, [InClL10]. The title com- The radiolabelling was carried out according to the procedure
pound was synthesised employing general metal complexation described previously,35 using [99mTcO4]− with SnCl2 as the
method 3. H2L10 (50.0 mg, 0.039 mmol), InCl3 (17.4 mg, reducing agent in sodium carbonate buffer at a temperature of
0.079 mmol) in pH 4.5 NaOAc buffer (5 mL). Yield: 30.3 mg, 95 °C. Briefly, the ligand (L2 to L6, 0.1 mg) was dissolved in
0.027 mmol, 68%.
TFA (190 μL) and treated with triethylsilane (10 μL) to remove
1H NMR (500 MHz, d6-DMSO, 25 °C): δ 9.41–9.67 (m, 6H, the Trt-group. After removal of the solvent under a stream of
CHpy), 9.23–9.35 (br s, 4H CHpyrr), 9.13–9.23 (br s, 3H, CONH, N2, sodium carbonate solution (80 μL, 0.1 mol L−1, pH = 9.5)
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 4859–4873 | 4871

View Article Online
Paper
Dalton Transactions
was added to the residue. Next, SnCl2 solution (3 μL, 1 mg (FP7 Metoxia project). SIP is grateful for funding from the EU
mL−1) in citrate buffer (0.1 mol L−1, pH = 5.0) and [99mTcO4]− Commission via O2SENSE ERC Consolidator grant, also from
(200 μL, 20 MBq) solution were added, and the reaction the Royal Society, STFC and MRC and FP7 MC ITN PRO-
mixture was vigorously stirred and allowed to react at 95 °C for SENSE. EPSRC Mass Spectrometry Service at Swansea and
1 h. The radiochemical purity and labelling efficiency were National Crystallography Service are thanked for assistance
determined using radio-HPLC (Method C in ESI†) or radio- with mass spectrometry and synchrotron crystallography
TLC. For radio-TLC, acetonitrile–water (1/1 by volume) was respectively. STFC funded our access to SRS Daresbury
used for the determination of 99mTc-colloid in the silica gel Station 9.8 and Dr John Warren is acknowledged for support
TLC (Rf = 0) and 2-butanone–ethyl acetate (2/3 by volume) was and training. We also thank Sophia Sarpaki for helpful
used for the determination of 99mTc-pertechnetate (Rf = 1.0).40 discussions.
In the cases when the radiochemical purity of the radiotracer
did not exceed 95%, the raw product was purified using a C18
Sep-Pak™ cartridge. The cartridge was pre-treated using
ethanol (5 mL), followed by water (5 mL), before the raw
References
product was loaded. The loaded cartridge was rinsed twice
with water (2 × 5 mL) to release the 99mTc-pertechnetate and
any other hydrophilic impurities, and the technetium-labelled
product was eluted with ethanol (1 mL). The radiochemical
purity of the purified radiotracer was then determined using
the method described above. All the 99mTc complexes had
RCP’s of >98% after purification.
1 G. L. Semenza, Cancer Metastasis Rev., 2007, 26, 223–224.
2 N. J. P. Beasley, C. C. Wykoff, P. H. Watson, R. Leek,
H. Turley, K. Gatter, J. Pastorek, G. J. Cox, P. Ratcliffe and
A. L. Harris, Cancer Res., 2001, 61, 5262–5267.
3 C. C. Wykoff, N. J. P. Beasley, P. H. Watson, K. J. Turner,
J. Pastorek, A. Sibtain, G. D. Wilson, H. Turley, K. L. Talks,
P. H. Maxwell, C. W. Pugh, P. J. Ratcliffe and A. L. Harris,
Cancer Res., 2000, 60, 7075–7083.
Confocal fluorescence microscopy
4 D. Vullo, M. Franchi, E. Gallori, J. Pastorek, A. Scozzafava,
S. Pastorekova and C. T. Supuran, J. Enzyme Inhib. Med.
Chem., 2003, 18, 403–406.
5 O. Tureci, U. Sahin, E. Vollmar, S. Siemer, E. Gottert,
G. Seitz, A.-K. Parkkila, G. N. Shah, J. H. Grubb,
M. Pfreundschuh and W. S. Sly, Proc. Natl. Acad.
Sci. U. S. A., 1998, 95, 7608–7613.
6 J.-Y. Winum, D. Vullo, A. Casini, J.-L. Montero,
A. Scozzafava and C. T. Supuran, J. Med. Chem., 2003, 46,
2197–2204.
HCT116 cells were seeded as monolayers in T75 tissue culture
flasks, and cultured in Dulbecco’s Modified Eagle’s Medium
(DMEM) supplemented with 10% foetal bovine serum, L-gluta-
mine, penicillin and streptomycin. Cells were maintained at
37 °C in a 5% CO2 humidified atmosphere and grown to
approximately 85% confluence before being split using
2.5% trypsin. For microscopy, cells were seeded onto cham-
bered cover-glass slides and incubated41 for 12 h to ensure
adhesion.
7 J.-Y. Winum, D. Vullo, A. Casini, J.-L. Montero,
A. Scozzafava and C. T. Supuran, J. Med. Chem., 2003, 46,
5471–5477.
Porphyrin complexes were prepared as 10 mM solution in
DMSO or distilled water, and diluted to 10 μM with DMEM,
and incubated with the cells at 37 °C. Prior to imaging, the
solution was replaced with 1 mL fresh DMEM. Background
autofluorescence was measured by imaging the cells in
1 mL of DMEM medium only. The fluorescent uptake of
porphyrin complexes was imaged by laser-scanning confocal
microscopy using a Zeiss LSM 510 META microscope irra-
diating at 405 nm with emission filtered between 565 and
615 nm.
8 (a) V. Alterio, R. M. Vitale, S. M. Monti, C. Pedone,
A. Scozzafava, A. Cecchi, G. De Simone and C. T. Supuran,
J. Am. Chem. Soc., 2006, 128, 8329–8335; (b) M. M. Ghorab,
M. S. Alsaid, M. Ceruso, Y. M. Nissan and C. T. Supuran,
Bioorg. Med. Chem., 2014, 22, 3684–3695; (c) D. Can,
B. Spingler, P. Schmutz, F. Mendes, P. Raposinho,
C. Fernandes, F. Carta, A. Innocenti, I. Santos,
C. T. Supuran and R. Alberto, Angew. Chem., Int. Ed., 2012,
51, 3354–3357.
9 D. Vullo, A. Innocenti, I. Nishimori, J. Pastorek,
A. Scozzafava, S. Pastorekova and C. T. Supuran, Bioorg.
Med. Chem. Lett., 2005, 15, 963–969.
Acknowledgements
We are grateful to Dr. Paul Burke and Prof. Franklin Aigbirhio
for the production of 64Cu at the Wolfson Brain Imaging, 10 D. Vullo, M. Franchi, E. Gallori, J. Pastorek, A. Scozzafava,
Addenbrooke’s Hospital, University of Cambridge and to Prof.
Adrian Harris, Weatherall Institute for Molecular Medicine,
S. Pastorekova and C. T. Supuran, Bioorg. Med. Chem. Lett.,
2003, 13, 1005–1009.
University of Oxford, for the provision of HCT116 cells and 11 C. T. Supuran, A. Scozzafava and A. Casini, Med. Res. Rev.,
helpful discussions. We also gratefully acknowledge funding 2003, 23, 146–189.
from CRUK (C5255/A8591), Siemens Molecular Imaging Ltd., 12 D. Le Bars, S. K. Luthra, V. W. Pike and E. R. Swenson,
Technology Strategy Board and the EU. This work was funded Appl. Radiat. Isot., 1988, 39, 671–675.
by Cancer Research UK, Siemens Molecular Imaging, and the 13 E. E. Graves and A. J. Giaccia, Oncology, 2007, 21, 368–
Technology Strategy Board. CTS received funding from the EU
376.
4872 | Dalton Trans., 2015, 44, 4859–4873
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
14 (a) L. Dubois, K. Douma, C. T. Supuran, R. K. Chiu, 27 A. Davison, A. G. Jones, C. Orvig and M. Sohn, Inorg.
M. A. M. J. van Zandvoort, S. Pastorekova, A. Scozzafava, Chem., 1981, 20, 1629–1632.
B. G. Wouters and P. Lambin, Radiother. Oncol., 2007, 83, 28 V. Garaj, L. Puccetti, G. Fasolis, J.-Y. Winum, J.-L. Montero,
367–373; (b) L. Dubois, S. Peeters, N. G. Lieuwes,
A. Scozzafava, D. Vullo, A. Innocenti and C. T. Supuran,
N. Geusens, A. Thiry, S. Wigfield, F. Carta, A. Mcintre,
Bioorg. Med. Chem. Lett., 2005, 15, 3102–3108.
A. Scozzafava, J.-M. Dogne, C. T. Supuran, A. L. Harris, 29 G. Liu and J. Hnatowich Donald, Anti-Cancer Agents Med.
B. Masereel and P. Lambin, Radiother. Oncol., 2011, 99(3),
424–431.
15 (a) M. Rami, J.-Y. Winum, A. Innocenti, J.-L. Montero,
Chem., 2007, 7, 367–377.
30 W. A. Siddiqui, S. Ahmad, H. L. Siddiqui, T. Hussain and
M. J. Parvez, J. Chem. Crystallogr., 2010, 40, 116–121.
A. Scozzafava and C. T. Supuran, Bioorg. Med. Chem. Lett., 31 R. G. Khalifah, J. Biol. Chem., 1971, 246, 2561–2573.
2008, 18, 836–841; (b) J. Lau, J. Pan, Z. Zhang, N. Hundal- 32 P. D. Bonnitcha, A. L. Vavere, J. S. Lewis and J. R. Dilworth,
Jabal, Z. Liu, F. Benard and K.-S. Lin, Bioorg. Med. Chem.
Lett., 2014, 24, 3064–3068.
16 C. T. Supuran, Met.-Based Drugs, 1995, 2, 331–336.
17 C. T. Supuran, G. Manole and M. Andruh, J. Inorg.
Biochem., 1993, 49, 97–103.
J. Med. Chem., 2008, 51, 2985–2991.
33 J. L. J. Dearling, J. S. Lewis, G. E. D. Mullen, M. J. Welch
and P. J. Blower, J. Biol. Inorg. Chem., 2002, 7, 249–
259.
34 D. Brenner, A. Davison, J. Lister-James and A. G. Jones,
Inorg. Chem., 1984, 23, 3793–3797.
18 M. Jude Kevin, L. Banerjee Abir, K. Haldar Manas,
S. Manokaran, B. Roy, S. Mallik, D. K. Srivastava and 35 A. Capretta, R. B. Maharajh and R. A. Bell, Carbohydr. Res.,
W. Christianson David, J. Am. Chem. Soc., 2006, 128,
3011–3018.
19 Y. Fujibayashi, H. Taniuchi, Y. Yonekura, H. Ohtani,
J. Konishi and A. Yokoyama, J. Nucl. Med., 1997, 38,
1155–1160.
20 J. S. Lewis, D. W. McCarthy, T. J. McCarthy, Y. Fujibayashi
and M. J. Welch, J. Nucl. Med., 1999, 40, 177–183.
21 A. L. Vavere and J. S. Lewis, Cu-ATSM, Dalton Trans., 2007,
4893–4902.
1995, 267, 49–63.
36 R. Rajagopalan, G. D. Grummon, J. Bugaj, L. S. Hallemann,
E. G. Webb, M. E. Marmion, J.-L. Vanderheyden and
A. Srinivasan, Bioconjugate Chem., 1997, 8, 407–415.
37 M. Balaz, A. E. Holmes, M. Benedetti, G. Proni and
N. Berova, Bioorg. Med. Chem., 2005, 13, 2413–2421.
38 J. P. C. Tome, M. G. P. M. S. Neves, A. C. Tome,
J. A. S. Cavaleiro, M. Soncin, M. Magaraggia, S. Ferro and
G. Jori, J. Med. Chem., 2004, 47, 6649–6652.
22 F. Dehdashti, M. A. Mintun, J. S. Lewis, J. Bradley, 39 (a) S. R. Bayly, R. C. King, D. J. Honess, P. J. Barnard,
R. Govindan, R. Laforest, M. J. Welch and B. A. Siegel,
Eur. J. Nucl. Med. Mol. Imaging, 2003, 30, 844–850.
23 F. Dehdashti, P. W. Grigsby, J. S. Lewis, R. Laforest,
B. A. Siegel and M. Welch, J. Nucl. Med., 2008, 49, 201–205.
H. M. Betts, J. P. Holland, R. Hueting, P. D. Bonnitcha,
J. R. Dilworth, F. I. Aigbirhio and M. Christlieb, J. Nucl. Med.,
2008, 49, 1862–1868; (b) K. Ogawa, T. Mukai, Y. Inoue,
M. Ono and H. Saji, J. Nucl. Med., 2006, 47, 2042–2047.
24 R. Hueting, M. Christlieb, J. R. Dilworth, E. G. Garayoa, 40 K. Ogawa, T. Mukai, Y. Inoue, M. Ono and H. Saji, J. Nucl.
V. Gouverneur, M. W. Jones, V. Maes, R. Schibli, X. Sun and
D. A. Tourwe, Dalton Trans., 2010, 39, 3620–3632.
25 A. R. Fritzberg, C. C. Kuni, W. C. Klingensmith III,
J. Stevens and W. P. Whitney, J. Nucl. Med., 1982, 23,
592–598.
Med., 2006, 47, 2042–2047.
41 R. L. Arrowsmith, P. A. Waghorn, M. W. Jones, A. Bauman,
S. K. Brayshaw, Z. Hu, G. Kociok-Köhn, T. L. Mindt,
R. M. Tyrrell, S. W. Botchway, J. R. Dilworth and S. I. Pascu,
Dalton Trans., 2011, 40, 6238–6252.
26 W. C. Klingensmith 3rd, A. R. Fritzberg, V. M. Spitzer, 42 E. Izbicka, R. T. Wheelhouse, E. Raymond, K. K. Davidson,
D. L. Johnson, C. C. Kuni, M. R. Williamson, G. Washer
R. A. Lawrence, D. Sun, B. E. Windle, L. H. Hurley and
and R. Weil 3rd, J. Nucl. Med., 1984, 25, 42–48.
D. D. Von Hoff, Cancer Res., 1999, 59, 639–644.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 4859–4873 | 4873