Near IR emitting isothiocyanato-substituted fluorophores: Their synthesis and bioconjugation to monoclonal antibodies

Article abstract of DOI:10.1039/b504334d

Two near IR emitting fluorophores, based on the phthalo-cyanine and naphthalocyanine chromophores, which also bear a single isothiocyanato group suitable for conjugation to proteins are reported; their utility as luminescent probes is demonstrated by conjugation to monoclonal antibodies and the ability of these conjugates to selectively bind cells bearing the relevant antigen. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b504334d

View Article Online / Journal Homepage / Table of Contents for this issue
C O M M U N I C A T I O N
Near IR emitting isothiocyanato-substituted fluorophores: their
synthesis and bioconjugation to monoclonal antibodies
Wubiao Duan,a Karen Smith,b Huguette Savoie,a John Greenmanb and Ross W. Boyle*a
a University of Hull, Department of Chemistry and Clinical Biosciences Institute, Hull,
UK HU6 7RX. E-mail: r.w.boyle@hull.ac.uk; Fax: +44 (0)1482 466410
b University of Hull, Postgraduate Medical School and Clinical Biosciences Institute, Hull,
UK HU6 7RX. E-mail: j.greenman@hull.ac.uk; Fax: +44 (0)1482 466410
Received 25th March 2005, Accepted 13th May 2005
First published as an Advance Article on the web 31st May 2005
Two near IR emitting fluorophores, based on the phthalo-
cyanine and naphthalocyanine chromophores, which also
bear a single isothiocyanato group suitable for conjugation
to proteins are reported; their utility as luminescent probes
is demonstrated by conjugation to monoclonal antibodies
and the ability of these conjugates to selectively bind cells
bearing the relevant antigen.
react with amines via an addition, as opposed to substitution,
reaction, thus simplifying conjugate purification. Recently, we
have been involved in the development of porphyrins bearing a
single isothiocyanato group for conjugation to proteins, both
as targeted photosensitisers for photodynamic therapy4 and
also fluorescence imaging of cellular systems.5,6 The porphyrins
however, emit in the red region of the spectrum, typically around
640 nm,7 whereas human plasma has an intense autofluorescence
band ranging from 600–650 nm,8 leading to poor signal to
background ratios.
Phthalocyanines and naphthobenzoporphyrazines have ab-
sorption and emission bands which are both bathochromically
and hyperchromically shifted relative to the porphyrins, and
these spectral properties can be fine tuned by peripheral
substitution.9 We now wish to report that, by judicial choice of
peripheral groups, it is possible to synthesise phthalocyanines
and naphthobenzoporphyrazines which absorb and emit in
the 700–850 nm region, and also bear a single amine-reactive
isothiocyanto group suitable for conjugating to protein based
biological targeting agents.
Considerable interest is currently being shown in luminescent
molecules which emit in the near IR region of the electro-
magnetic spectrum for a variety of analytical applications.1
Molecules with these spectroscopic properties are especially
well suited for biological imaging applications due to the
optical transparency of these media at wavelengths greater than
700 nm.2 In order to be truly useful for such applications it is
necessary to have not only an imaging component which absorbs
and emits at wavelengths longer than 700 nm, but also a targeting
component which binds the luminescent probe selectively to
the target tissue.3 Currently only three probes are commercially
available which absorb and emit in this region, Alexa FluorTM
700, Alexa FluorTM 750 and Cy7TM, these fluorophores are
available as amine reactive succinimidyl esters, however, unlike
fluorescein based probes, isothiocyanates of these dyes are
not available. Bioconjugations with isothiocyanato fluorophores
offer the advantage of not generating any by-products, as they
The most convenient precursor to the isothiocyanato group
is an amine and, as we have previously used this strategy with
porphyrin based systems,5,6 the readily available phthalocyanine
precursor 4-amino-1,2-benzenedicarbonitrile 1 was selected
to allow introduction of the amino functionality to both
macrocycles. As use of this precursor would inevitably mean
Scheme 1 Reagents and conditions: (i) ZnCl2, DBU, n-butanol, reflux 24 h (ii) CSCl2, CH2Cl2, RT, 4 h.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 3 8 4 – 2 3 8 6 T h i s j o u r n a l i s T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
2 3 8 4
©

View Article Online
that one amino-isoindole unit would be incorporated into both
phthalocyanine and naphthobenzoporphyrazine macrocycles,
and contribute little to bathochromically shifting the kmax rel-
ative to phthalocyanines, the bis-alkoxy substituted precursors,
3,6-dibutoxy-1,2-benzenedicarbonitrile 2 and 1,4-dibutoxy-2,3-
naphthalenedicarbonitrile 3 were chosen to provide the remain-
ing sub units. Selection of these latter precursors ensured the
presence of electron donating alkoxy groups at positions on
the macrocycles which have been shown to result in significant
bathochromic shifts relative to the unsubstituted analogues.9,10,11
Thus 1 and 2 and 1 and 3 respectively, were mixed in 1 : 3 molar
ratios with ZnCl2 and DBU in n-butanol and heated at reflux
under argon for 24 h. After cooling and evaporation of solvent
the mixtures were washed with methanol and chromatographed
to give the required mono amino phthalocyanato zinc com-
plex 4 and mono amino naphthobenzoporphyrazinato zinc
complex 5 in 3.3% and 1.8% yields respectively (Scheme 1).
Conversion of 4 and 5 to the corresponding isothiocyanates
6 and 7 was achieved by reaction with excess thiophosgene
in dichloromethane at room temperature. Compounds 6 and
visible spectroscopy and was found to be 4.2 : 1, 3.9 : 1 and
13.4 : 1 for anti-EpCAM, anti-CD146 and anti-CD104 respec-
tively with compound 6, and 4.2 : 1, 7.6 : 1 and 6.3 : 1 for the
same antibodies with compound 7. All three antibodies bind
to tumour associated antigens that are upregulated in various
cancers. The colon adenocarcinoma (LoVo) and lung large cell
carcinoma cell line (COR-L23) were selected to assay binding
of the conjugates using flow cytometry. The COR-L23 line
expresses the antigens recognised by all three antibodies whereas
the LoVo cell line only expresses EpCAM and CD104; thus in
addition to positive binding, non-specific binding can also be
investigated. Equivalent amounts of antibody were added in
every case.
Analysis of flow cytometry data¶ for conjugates of com-
pound 6 shows excellent correlation between conjugated and
non-conjugated anti-EpCAM and anti-CD104 antibodies with
CORL23 and LoVo cells indicating that the presence of the
macrocycle attached to the antibody has not affected binding
to the respective antigens (Fig. 3). The anti-CD146 conjugate
again shows virtually perfect overlap with binding of its non-
conjugated analogue for the antigen positive CORL23 cells,
with only a very minor level of non-specific binding to the
LoVo cells being introduced by conjugation of the macrocycle.
Comparison of these results with those from similar assays
using the conjugates of naphthobenzoporphyrazine 7 show some
differences for the anti-EpCAM and anti-CD146 antibodies
on CORL23 cells, with both showing a slight loss of binding
efficiency relative to the non-conjugated antibodies. The same
cell line with the anti-CD104 conjugate however shows excellent
binding, comparable to the unconjugated antibody. As the
loading ratio for the anti-EpCAM conjugate (4.2 : 1) was lower
than that for the anti-CD104 conjugate (6.3 : 1), levels of flu-
orophore bound to the antibody surface cannot be responsible
for the slightly lower binding seen with the former conjugate.
It is also very unlikely that the linkage of compound 7 itself
is inhibiting antigen-binding per se as the conjugates all bind
identically, when compared with the unconjugated antibodies,
against LoVo cells. It would appear therefore that conjugation of
different macrocycles can cause subtle differences in the ability
of an antibody to recognise its target antigen; however such
changes are relatively minor, i.e. the anti-EpCAM and anti-
CD146 conjugates still effectively bind all cells.
1
7 were characterised by H NMR, MS, elemental analysis and
UV–visible spectroscopy.†Values of kmax for 6 and 7 were 729 and
796 nm, respectively, and both showed the expected fluorescence
bands when excited at these wavelengths with Stokes shifts of
10–20 nm (Figs. 1 and 2). Interestingly, the compounds described
here, unlike many conventional fluorophores, have multiple
absorptions in the visible and infrared regions, other than the
kmax. This allows much larger Stokes shifts to be obtained e.g.
excitation on the secondary bands at 660 nm for 6 and 700 nm
for 7 results in Stokes shifts of approximately 100 nm for each
molecule.
The fluorescent probes described here are hydrophobic, a fac-
tor which can lead to aggregation of antibody conjugates. How-
ever, this would be expected to result in significantly increased
binding to cells which do not express the corresponding antigen,
and would be indicated by flow cytometry curves displaced
significantly to the right of those for the corresponding non-
conjugated antibody. Although his effect can be seen to a minor
degree in the case of compound 6 with the CD146 antibody and
LoVo cell line, it cannot be seen at all for compound 7 with the
same antibody and cells. As 7 is the more hydrophobic of the
two fluorophores, this physicochemical property does not seem
to result in aggregation of antibody conjugates.
Fig. 1 Absorbance and fluorescence spectra for 6 in DMSO.
Finally, as both phthalocyanines12 and naphthobenzopor-
phyrazines13 have been shown to be capable of photocytotoxicity,
and this would clearly have implications for imaging of live cells,
the photocytotoxicity of all conjugates was screened against the
cell line to which they demonstrated positive binding. In all
cases no difference was observed between irradiated and non-
irradiated cells, with 100% cell survival in both cases, even at
significantly higher concentrations than those used for imaging.
In conclusion, we have shown that fluorescent macrocycles
based on the phthalocyanine and naphthobenzoporphyrazine
core can be synthesised which absorb and emit in the spectral
region most favourable for biological imaging. The macro-
cycles incorporate isothiocyanate groups which allows facile
bioconjugation to monoclonal antibodies. The ability of these
conjugates to selectively bind to and identify cells which express
the corresponding antigens has also been demonstrated.
Fig. 2 Absorbance and fluorescence spectra for 7 in DMSO.
In order to demonstrate the utility of these fluorophores for
use in biological imaging applications both were conjugated
to three different monoclonal antibodies, anti-EpCAM, anti-
CD146 and anti-CD104 (Serotec, Oxford, UK) in 0.5 M
bicarbonate buffer (pH 9.2) and the conjugates were purified by
gel exclusion chromatography.ठThe degree of labelling (moles
of compound per mole of antibody) was determined by UV–
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 3 8 4 – 2 3 8 6
2 3 8 5

View Article Online
Fig. 3 Flow cytometry of conjugated antibodies.
The degree of labelling (the number of moles of compound 6 or 7
conjugated per mole of antibody) was calculated using spectroscopic
methods.
Acknowledgements
The authors wish to thank the BBSRC (21/E12509), Wellcome
Trust (059572, 066948) and Leverhume Trust (F/00181H) for
financial support.
¶ Flow cytometry: Cells were removed from culture vessels with 5 mM
EDTA. After washing in PBS the cells were counted, re-suspended in
PBS–0.25% w/v bovine serum albumin (BSA) and 2 × 105 cells were
added to each tube. The cells were labelled with 50 ll (5 lg ml−1) of
anti-EpCAM, anti-CD104 or anti-CD146 (conjugated/unconjugated)
for 1 h at 4 ◦C. After washing with PBS–BSA, cells were labelled with
50 ll (10 lg ml−1) of rabbit anti-mouse IgG-FITC (Serotec) for 1 h at
Notes and references
† Selected experimental and spectral data : 6 dH(400 MHz DMSO-D6)
1.16 (18H, m, CH2CH3), 1.51 (12H, m, CH2CH2CH3), 2.10 (12H, m,
OCH2CH2CH2CH3), 4.65 (12H, m, OCH2CH2CH2CH3), 7.7 (9H, m,
ArH); m/z (MALDI) 1068 (M + H+); kmax (DMSO/nm) 336, 660,
729; 7 dH(400 MHz DMSO-D6) 1.17 (18H, m, CH2CH3), 1.75 (12H,
m, CH2CH2CH3), 2.21 (12H, m, OCH2CH2CH2CH3), 5.12 (12H, m,
OCH2CH2CH2CH3), 6.35 (3H, m, benzene-H), 7.72(6H, m, naphlalene-
H), 8.85 (6H, m, naphthalene-H); m/z (MALDI) 1217 (M+); kmax
(DMSO/nm) 348, 700, 796.
‡ Cell lines and antibodies: The LoVo human colon adenocarcinoma
and CORL23 human lung large cell carcinoma cell lines (ECACC) were
grown in DMEM and RPMI respectively. All media was supplemented
with 10% v/v fetal calf serum, 2 mM L-glutamine, 100 lg ml−1
streptomycin, 100 lg ml−1 penicillin and 0.25 lg ml−1 amphotericin B
(Invitrogen). All cells were maintained at 37 ◦C in a humidified incubator
with a 5% CO2 atmosphere. The murine antibodies, anti-CD104 and
anti-CD146, were purchased from Serotec. The anti-EpCAM antibody
was a generous gift from Prof. S. Warnaar (University of Leiden). Anti-
CD104 binds to a 205 kDa glycoprotein involved in cell–cell adhesion
and is upregulated in colorectal and bladder carcinomas. Anti-CD146
binds to a cell adhesion molecule upregulated in melanoma cells. Anti-
EpCAM recognises a 40 kDa cell–cell adhesion molecule upregulated
on a variety of carcinomas, particularly colorectal anti-EpCAM and
anti-CD104 shows positive binding to both the LoVo and CORL23 cell
lines whereas the anti-CD146 only shows binding to CORL23.
4
◦C. A further wash with PBS–BSA was performed before analysis of
cells in a FACScalibur cytometer (BD Biosciences). Data is presented as
histograms plotting cell number against fluorescence intensity.
1 A. Gomez-Hens and M. P. Aguilar-Caballos, Trends Anal. Chem.,
2004, 23, 127.
2 P. C. Wilson, Photosensitising Compounds: Their Chemistry, Biology
and Clinical Use, Wiley Interscience, Chichester, 1989, 73.
3 R. Hudson and R. W. Boyle, J. Porphyrins Phthalocyanines, 2004, 8,
954.
4 R. Hudson, M. Carcenac, K. Smith, L. Madden, O. J. Clarke, A.
Pelegrin, J. Greenman and R. W. Boyle, Br. J. Cancer, 2005, 92,
1442.
5 J. M. Sutton, O. J. Clarke, N. Fernandez and R. W. Boyle,
Bioconjugate Chem., 2002, 13, 249.
6 O. J. Clarke and R. W. Boyle, Chem. Commun., 1999, 2231.
7 J. M. Sutton, N. Fernandez and R. W. Boyle, J. Porphyrins Phthalo-
cyanines, 2000, 4, 655.
8 Y. Cao, Y.-E. Lee Koo and R. Kopelman, Analyst, 2004, 129,
745.
9 M. J. Cook, A. J. Dunn, S. D. Howe, A. J. Thomson and K. J.
Harrison, J. Chem. Soc., Perkin Trans. I, 1988, 2453.
10 B. D. Rihter, M. E. Kenney, W. E. Ford and M. A. J. Rodgers, J. Am.
Chem. Soc., 1990, 112, 8064.
§ Conjugation of antibodies: Conjugation is carried out in a 1 ml reaction
11 B. D. Rihter, M. E. Kenney, W. E. Ford and M. A. J. Rodgers, J. Am.
Chem. Soc., 1993, 115, 8146.
−1
volume containing 1 mg ml antibody in 0.5 M bicarbonate buffer
(pH 9.2) and a 20 × molar excess of compound 6 or 7. The reaction
vessels were agitated gently at room temperature for 1 h and protected
from light. The conjugates were purified using Sephadex G25 columns
(Amersham) and eluted with phosphate buffered saline pH 7.4 (PBS).
12 A. C. Tedesco, J. C. G. Rotta and C. N. Lunardi, Curr. Org. Chem.,
2003, 7(2), 187.
13 P. Margaron, R. Langlois, J. E. van Lier and S. Gaspard, J. Pho-
tochem. Photobiol. B., 1992, 14(3), 187.
2 3 8 6
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 3 8 4 – 2 3 8 6