Evaluation of asymmetric orthogonal cyanine fluorophores

Article abstract of DOI:10.1016/j.dyepig.2020.108712

Pentamethine cyanine (Cy5) fluorophores have proven to be versatile imaging agents (i.e., tracers) for a range of micro- and macroscopic imaging applications, including image-guided surgery. In this study the relationship between the structure of asymmetric Cy5 fluorophores and their photophysical properties was studied. To this end, seven Cy5 analogues, bearing orthogonal N-indole substituents (H, SO₃^?, or benzene), were synthesised and evaluated. In-depth analysis revealed that introduction of sulfonates enhanced the fluorescence brightness and photostability, while reducing the lipophilicity, serum binding and stacking tendency. The addition of benzene moieties induced a bathochromic shift of 10–20 nm, increased the lipophilicity (LogP = -1.56–1.23) and serum binding (67.3–93.8percent bound), as well as negatively impacted the brightness (0.74–42.9 · 10³ M^?1 cm^?1), photostability (24.4–90.6percent remaining), and stacking tendency. Chemical stability was uninfluenced by the substitution pattern. Additionally, the generation of a c[RGDyK]-based hybrid tracer based on one of these fluorophores in combination with a diethylenetriaminepentaacetic acid (DTPA) chelate and an ¹¹¹In-isotope was reported. This compound was evaluated in vitro using α_vβ₃-overexpressing Geβ3 cells and in vivo using a 4T1 mouse tumour model. Overall, the presented results imply that alterations of the asymmetrical orthogonal Cy5 fluorophore structure have impact on the (photo)physical properties. Furthermore, the orthogonal Cy5 fluorophore framework can readily be applied in tracer development.

Full text of DOI:10.1016/j.dyepig.2020.108712

Journal Pre-proof
Evaluation Of Asymmetric Orthogonal Cyanine Fluorophores
Albertus W. Hensbergen, Mathijs A.C. de Kleer, Michael S. Boutkan, Danny M.
van Willigen, Felicia A. van der Wijk, Mick M. Welling, Hans-Jürgen Wester, Tessa
Buckle, Fijs W.B. van Leeuwen
PII:
S0143-7208(20)31409-1
DOI:
https://doi.org/10.1016/j.dyepig.2020.108712
DYPI 108712
Reference:
To appear in: Dyes and Pigments
Received Date: 23 March 2020
Revised Date: 10 July 2020
Accepted Date: 11 July 2020
Please cite this article as: Hensbergen AW, de Kleer MAC, Boutkan MS, van Willigen DM, van der
Wijk FA, Welling MM, Wester H-J, Buckle T, van Leeuwen FWB, Evaluation Of Asymmetric Orthogonal
Cyanine Fluorophores, Dyes and Pigments, https://doi.org/10.1016/j.dyepig.2020.108712.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
©
2020 Published by Elsevier Ltd.

Albertus W. Hensbergen: Investigation, Formal Analysis, Writing - Original Draft
Mathijs A. C. de Kleer: Investigation, Writing - Review & Editing Michael S.
Boutkan: Investigation, Writing - Review & Editing Danny M. van Willigen:
Investigation, Writing - Review & Editing Felicia A. van der Wijk: Investigation,
Writing - Review & Editing Mick M. Welling: Investigation, Writing - Review & Editing
Hans-Jürgen Wester: Conceptualisation, Writing - Review & Editing Tessa Buckle:
Investigation, Writing - Review & Editing Fijs W. B. van Leeuwen:
Conceptualisation, Supervision, Writing - Review & Editing, Funding acquisition

1
EVALUATION OF ASYMMETRIC ORTHOGONAL
CYANINE FLUOROPHORES
2
3
a
a
a
4
5
6
Albertus W. Hensbergen , Mathijs A. C. de Kleer , Michael S. Boutkan ,
a
a
a
Danny M. van Willigen , Felicia A. van der Wijk , Mick M. Welling , Hans-
b
a
a,*
Jürgen Wester , Tessa Buckle , Fijs W. B. van Leeuwen
a
7
8
Interventional Molecular Imaging Laboratory, Department of Radiology, Leiden
University Medical Center, Albinusdreef 2, 2333 ZA Leiden, The Netherlands
b
9
0
Pharmazeutische Radiochemie, Technische Universität München, Walther-Meißner-
1
Straße 3, 85748 Garching, Germany
11
12
13
14
*Corresponding author: F. W. B. van Leeuwen, Interventional Molecular Imaging
Laboratory, Department of Radiology, Leiden University Medical Center, Albinusdreef
2, 2333 ZA Leiden, The Netherlands. Phone: +31715266029. E‐mail:
f.w.b.van_leeuwen@lumc.nl
15
HIGHLIGHTS
16
17
18
19
20
21
•
•
•
Sulfonates on cyanine fluorophores enhance brightness, photostability, and
reduce stacking
Extra benzene rings negatively impact photophysical properties of cyanine
fluorophores
Asymmetric orthogonal cyanine fluorophores can be used for hybrid tracer
development
22
ABSTRACT
2
3
4
Pentamethine cyanine (Cy5) fluorophores have proven to be versatile imaging
2
agents (i.e., tracers) for a range of micro- and macroscopic imaging applications,

25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
including image-guided surgery. We studied the composition of asymmetric Cy5
fluorophores in relation to the photophysical properties. To this end, seven Cy5
-
3
analogues, bearing orthogonal N-indole substituents (H, SO , or benzene) were
synthesised and evaluated. In-depth analysis revealed that introduction of sulfonates
enhanced the fluorescence brightness and photostability, while reducing the
lipophilicity, serum binding and stacking tendency. The addition of benzene moieties
induced a bathochromic shift of 10–20 nm, increased the lipophilicity (LogP = -1.56–
1.23) and serum binding (67.3–93.8% bound), as well as negatively impacted the
3
-1
-1
brightness (0.74–42.9 · 10 M · cm ), photostability (24.4–90.6% remaining), and
stacking tendency. Chemical stability was uninfluenced by the substitution pattern.
Additionally, we report the generation of a c[RGDyK]-based hybrid tracer based on
one of these fluorophores in combination with a diethylenetriaminepentaacetic acid
111
(DTPA) chelate and an In-isotope. This compound was evaluated in vitro using
α β -overexpressing Geβ3 cells and in vivo using a 4T1 mouse tumour model.
v
3
Overall, our results imply that alterations of the asymmetrical orthogonal Cy5
fluorophore composition have impact on the (photo)physical properties. Furthermore,
the orthogonal Cy5 fluorophore framework can readily be applied in tracer
development.
43
4
4
5
Keywords: fluorescence, cyanine fluorophores, fluorescence brightness,
photobleaching, image-guided surgery
4
46
1. INTRODUCTION
47
48
49
50
With the rise of fluorescence-guided surgery in oncology, the demand for specific and
highly fluorescent tracers has grown substantially. Next to fluorophores such as
indocyanine green and fluorescein being routinely used in clinical care [1, 2], various
ongoing clinical trials make use of investigative new imaging agents (i.e., tracers)

51
52
53
54
55
56
57
58
59
60
61
targeting receptors highly expressed on the membranes of tumorous cells [3-5].
Fluorophores from the cyanine (Cy) family are the most commonly used fluorophores
in image-guided surgery [6]. The composition of the fluorophore also impacts the
photophysical properties; the fluorescence quantum yield (Φ ) of a fluorophore can
F
be influenced by its substituents on either indole unit [7], but a clear trend between
the substitution pattern on indoles and the Φ has not been found [8, 9]. Ultimately,
F
when conjugated to a targeting vector, the fluorophore composition can considerably
influence the performance—e.g., receptor affinity, pharmacodynamics and
pharmacokinetics—of the corresponding tracer [10-13]. These effects have been
studied using either near-infrared heptamethine (Cy7) [14, 15] or far-red
pentamethine (Cy5) analogues [11, 16].
62
63
64
65
66
67
68
69
70
71
72
Not only the fluorophore composition, but also the manner the fluorophore is
incorporated in a (hybrid) tracer has shown to be of influence regarding tracer
performance [11]. Most fluorescent (and hybrid) tracers consist of the “end-labelled
fluorophore” design, meaning that the targeting moiety is coupled to the fluorescent
moiety but the two are not integrally intertwined. Literature also indicates that
shielding the fluorophore could reduce the (possibly negative) influence(s) exerted by
a fluorescent label [17]. This finding engendered the concept of using fluorophores as
spacers during functionalization [18-20]. While literature mostly describes cyanine
fluorophores containing symmetrical N-indole substituents to be used in tracer
design, recent reports also mention asymmetrical orthogonal cyanine fluorophores
[11, 21, 22].
73
74
75
76
77
The impact of the substituents of a fluorophore on tracer affinity,
pharmacodynamics, and pharmacokinetics is well described [16, 22-24]. For
asymmetrical penta- and heptamethine cyanine fluorophores, the influence of small
variations in these substituents (e.g., a single sulfonate or benzene substituent at the
(C4 and) C5 position of either indoles) can exert on the lipophilicity, serum binding

78
79
80
81
82
83
84
85
86
87
88
89
quantum yield, molar extinction coefficient, stacking behaviour, and chemical- and
photostability has also been studied [9, 13], as well as the influence of the
polymethine bridge length on the signal penetration depth [25]. For asymmetrical
orthogonal Cy5 fluorophores (i.e., a cyanine fluorophore that bears a single sulfonate
or benzene substituent at the (C4 and) C5 position of either indoles and has different
N-indole substituents suited for conjugation) these features have, to the best of our
knowledge, not yet been examined. Hence, the characterisation of an asymmetrical
1
2
1
2
orthogonal Cy5 framework (Phth-(R )Cy5(R )-COOH, where either or both R and R
-
3
equal H, SO , or benzene (Scheme 1)) was studied. In addition to this, we created a
novel c[RGDyK]-based hybrid tracer to study if asymmetrical orthogonal fluorophores
have potential for angiogenesis imaging purposes by using c[RGDyK]; an excellent
model for evaluating tracer development strategies [22, 26, 27].
90

91
92
93
94
Scheme 1. Synthesis scheme for fluorophores 1–7. a) AcOH, 90 °C, 24 h, b) 1-(4-
bromobutyl)pyrrolidine-2,5-dione in sulfolane, 90 °C, 72 h, c) 6-bromohexanoic acid in 1,2-
dichlorobenzene or acetonitrile, 95–120 °C, 72 h, d) AcOH/Ac
(3:1), r.t., 12 h.
2 2
O (1:1), 60–120 °C, 12 h; pyridine/Ac O
95
2. MATERIALS & METHODS
9
6
7
2.1.
Chemistry
9
2.1.1. General
9
8
9
All chemicals and solvents were obtained from commercial sources and used without
further purification. Dimethylformamide (DMF), dichloromethane (DCM) and
dimethylsulfoxide (DMSO) were dried with 4 Å molecular sieves, unless stated
otherwise. Column chromatography was performed with 40–63 µm silica from
Screening Devices (Amersfoort, The Netherlands). Dry column vacuum
chromatography (DCVC) was performed as published by Pedersen et al. [28] with
15–40 µm silica and Hyflo Supercell Celite (Fisher Scientific, Hampton, US). High-
performance liquid chromatography (HPLC) was performed a Waters HPLC system
using either a 1525EF or 2545 pump and a 2489 UV/VIS detector. For preparative
HPLC either a Dr. Maisch GmbH Reprosil-Pur 120 C18-AQ 10 μm (250 × 20 mm)
column or a XBridge Prep C8 10 μm OBD 250 × 30 mm column was used with a
gradient of 0.1% trifluoro acetic acid (TFA) in H O/acetonitrile (CH CN) 95:5 to 0.1%
9
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
00
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
2
3
TFA in H O/CH CN 5:95 in 40 min (12 or 25 mL/min, respectively) was employed.
2
3
For semi-preparative HPLC a Dr. Maisch GmbH Reprosil-Pur C18-AQ 10 μm (250 ×
10 mm) column was used and a gradient of 0.1% TFA in H O/CH CN 95:5 to 0.1%
2
3
TFA in H O/CH CN 5:95 in 40 min (5 mL/min) was employed. For analytical HPLC a
2
3
Dr. Maisch GmbH Reprosil-Pur C18-AQ 5 μm (250 × 4.6 mm) column was used and
a gradient of 0.1% TFA in H O/CH CN 95:5 to 0.1% TFA in H O/CH CN 5:95 in 40
2
3
2
3
min (1 mL/min) was employed. High-resolution mass spectrometry (HRMS) was

117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
performed on either on-line C18 nanoHPLC MS/MS with a system consisting of an
Easy nLC 1200 gradient HPLC system (Thermo Fisher, Bremen, Germany), and a
LUMOS mass spectrometer (Thermo Fisher, Bremen, Germany) with a gradient from
of 0.1% formic acid (FA) in H O/CH CN 98:2 to 0.1% TFA in H O/CH CN 20:80 in 30
2
3
2
3
min, or a Waters Acquity H-class UPLC (Waters, Milford, USA) using a Acquity UPLC
BEH C18 1.7 µm (2.1 × 50 mm) column with a gradient of 0.1% FA in H O/CH CN
2
3
98:2 to 0.1% FA in H O/CH CN 60:40 in 1.8 min (0.6 mL/min) coupled to a high-
2
3
resolution XEVO G2S-XTOF Mass Spectrometer (Waters, Milford, USA). Low-
resolution mass spectrometry (LRMS) was performed using a Bruker Microflex
matrix-assisted laser desorption ionisation time-of-flight (MALDI-TOF) mass
1
13
spectrometer. H and C NMR were performed on a Bruker AV-300 (300 MHz) or a
Bruker Ascend 850 (850 MHz) equipped with a CryoProbe (all from Bruker, Billerica,
United States) in deuterated solvents. Absorbance spectra were recorded using an
Ultrospec 2100 pro (Amersham Biosciences, Little Chalfont, UK). Fluorescence
spectra were recorded using a PerkinElmer LS-55 (Perkin Elmer, Waltham, USA).
1
32
33
2.1.2. Synthesis
1
Merrifield resin synthesis
134
135
136
137
138
139
140
141
142
143
As performed by Lopalco et al. [29] Chloromethyl polystyrene resin (8.3 g, 15.0
mmol), N-tert-butoxycarbonyl-(N-Boc) aminophenol (9,4 g, 45.0 mmol), TBAI (1,7 g,
4.5 mmol) and Cs CO (14.7 g, 45.0 mmol) were dissolved in acetone, refluxed at 70
2
3
°C overnight under nitrogen atmosphere. The resin was then washed extensively
with 100 mL DMF, H O, DMF, DCM and diethyl ether (Et O) and dried in vacuo. A 4-
2
2
(4-nitrobenzyl)pyridine test was used to determine reaction completion. 10.3 g (90%
isolated yield) of Merrifield resin was formed. Removal of the N-Boc protection group
was carried out by adding a solution of 20% TFA in DCM. After washing with DCM,
the resin was neutralised using 20% N,N-diisopropylethylamine (DiPEA) in DCM and
after washing with DCM the resin was used in further steps.

144
145
146
The
2,3,3-trimethyl-3H-indole-5-sulfonate-
and
1,1,2-trimethyl-1H-
benzo[e]indole-based precursors for the Cy5 fluorophores were synthesised as
previously published [11, 16, 22] and used without further purification.
147
2.1.2.1. Synthesis of phthalimide fluorophores
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
The synthesis of 1–3, 5, and 6 has previously been described [11]. The following
compounds were synthesized in a similar fashion, which is in short as follows: the
hemicyanine was prepared by heating the indole-based building block with N-
((1E,3E)-3-(phenylimino)prop-1-en-1-yl)aniline in a mixture of acetic acid/acetic
anhydride (AcOH/Ac O; 1:1) to 90 °C for 18 h followed by stirring at 120 °C for 2 h.
2
After full conversion of the starting material (as determined by using UV/Vis
spectroscopy (λmax,product ≈ 450 nm, λmax, starting material ≈ 390 nm)), the mixture was
cooled down to r.t. and precipitated in diethyl ether. After repeated centrifugation,
decanting and washing steps with diethyl ether and ethyl acetate (EtOAc), the
obtained solid was dissolved in a 1:1 DCM/DMF mixture and directly added to the
deprotected resin. The suspension was agitated at r.t. for 60 min. Then, the resin
was washed with DCM/DMF mixtures and the cyanine fluorophore was prepared by
adding the corresponding second indole-based building block in a mixture of
pyridine/Ac O (3:1). The resulting mixture was agitated at r.t. for 18 h. Crude
2
compounds were purified by means of DCVC (EtOAc/MeOH) and subsequent
preparative HPLC.
1
64
65
1-(5-carboxypentyl)-2-((1E,3E)-5-((E)-1-(4-(1,3-dioxoisoindolin-2-yl)butyl)-3,3-
1
dimethylindolin-2-ylidene)penta-1,3-dien-1-yl)-3,3-dimethyl-3H-indol-1-ium (1)
1
3
166
167
168
169
As previously described. [11] C NMR (850 MHz, CD OD δ 177.16, 169.86, 155.55,
3
155.50, 155.39, 143.47, 143.46, 142.65, 142.53, 135.41, 133.20, 129.75, 129.69,
126.67, 126.63, 126.58, 126.31, 126.16, 124.14, 123.41, 123.37, 112.07, 111.93,
104.44, 104.43, 104.33, 104.32, 50.58, 50.46, 44.81, 44.27, 38.11, 34.59, 31.13,

+
1
70
71
28.19, 27.92, 27.90, 27.33, 26.62, 25.69, 25.38. HRMS calculated [M] for
+
4
1
C H N O 670.3639 found 670.3643
43
48
3
172
173
174
1-(5-carboxypentyl)-2-((1E,3E)-5-((E)-1-(4-(1,3-dioxoisoindolin-2-yl)butyl)-3,3-
dimethylindolin-2-ylidene)penta-1,3-dien-1-yl)-3,3-dimethyl-3H-indol-1-ium-5-
sulfonate (2)
1
3
175
176
177
178
179
180
As previously described. [11] C NMR (850 MHz, CD OD) δ 177.22, 176.06, 169.94,
3
156.50, 155.18, 145.15, 143.30, 142.93, 142.83, 142.29, 135.45, 133.26, 129.82,
128.00, 127.25, 126.81, 124.18, 123.51, 121.29, 112.50, 111.15, 105.62, 104.21,
50.95, 50.19, 49.38, 44.75, 44.61, 40.40, 38.06, 34.60, 28.03, 27.95, 27.77, 27.34,
+
+
26.59, 25.72, 25.49. HRMS calculated [M] for C H N O S 750.3207, found
43
47
3
7
750.3214
181
182
183
184
(E)-2-((2E,4E)-5-(1-(5-carboxypentyl)-3,3-dimethyl-3H-indol-1-ium-2-yl)penta-2,4-
dien-1-ylidene)-1-(4-(1,3-dioxoisoindolin-2-yl)butyl)-3,3-dimethylindoline-5-sulfonate
(3)
13
As previously described. [11] C NMR (850 MHz, CD OD) δ 177.22, 176.31,
3
185
186
187
188
189
169.93, 156.62, 155.00, 145.18, 143.33, 143.07, 142.71, 142.19, 135.45, 135.40,
133.29, 129.88, 127.98, 127.27, 126.95, 124.18, 124.09, 123.56, 121.28, 112.63,
111.03, 105.72, 104.15, 51.06, 50.08, 45.14, 44.23, 40.40, 38.09, 34.56, 34.45,
28.35, 28.30, 27.96, 27.75, 27.33, 27.26, 26.62, 25.69, 25.64, 25.19. HRMS
+
+
calculated [M] for C H N O S 750.3207, found 750.3212
43
47
3
7
190
191
192
1-(5-carboxypentyl)-2-((1E,3E)-5-((E)-1-(4-(1,3-dioxoisoindolin-2-yl)butyl)-3,3-
dimethyl-5-sulfonatoindolin-2-ylidene)penta-1,3-dien-1-yl)-3,3-dimethyl-3H-indol-1-
ium-5-sulfonate (4)
1
93
94
1-(5-carboxypentyl)-2-((1E,3E)-5-((E)-1-(4-(1,3-dioxoisoindolin-2-yl)butyl)-3,3-
dimethyl-5-sulfonatoindolin-2-ylidene)penta-1,3-dien-1-yl)-3,3-dimethyl-3H-indol-1-
1

195
196
197
198
199
ium-5-sulfonate was obtained from 1-(5-carboxypentyl)-2,3,3-trimethyl-3H-indol-1-
ium-5-sulfonate (1.7 g, 4.8 mmol), 1-(4-(1,3-dioxoisoindolin-2-yl)butyl)-2,3,3-trimethyl-
3H-indol-1-ium-5-sulfonate
(264.0
mg,
0.6
mmol),
and
N-((1E,3E)-3-
(phenylimino)prop-1-en-1-yl)aniline hydrochloride (1.4 g, 5.3 mmol) in a 16% yield
1
(86 mg) as a blue solid after lyophilisation. H NMR (300 MHz, CD OD) δ 8.29 (dd, J
3
200
201
202
203
= 12.2, 6.8 Hz, 2H), 7.93 – 7.75 (m, 8H), 7.37 (d, J = 8.4 Hz, 2H), 6.69 (t, J = 12.3
Hz, 1H), 6.36 (d, J = 13.6 Hz, 1H), 4.18 (d, J = 7.0 Hz, 4H), 3.76 (s, 2H), 2.36 (t, J =
13
7.2 Hz, 2H), 1.85 (s, 6H), 1.79 – 1.63 (m, 15H), 1.57 – 1.43 (m, 2H). C NMR (850
MHz, CD OD) δ 175.61, 175.40, 175.10, 169.90, 156.25, 156.10, 144.92, 144.89,
3
204
205
206
207
208
143.40, 143.25, 142.65, 142.54, 135.62, 135.46, 135.41, 133.22, 132.87, 128.04,
128.02, 127.75, 124.16, 124.12, 124.08, 121.58, 121.34, 121.32, 111.76, 111.63,
50.59, 50.48, 45.03, 44.55, 40.39, 40.38, 38.07, 34.45, 28.11, 27.83, 27.22, 26.58,
+
+
2
25.65, 25.31, 24.50. HRMS calculated [M+H] for C H N O S 830.2736, found
43
47
3
10
830.2781
209
210
211
3-(5-carboxypentyl)-2-((1E,3E)-5-((E)-1-(4-(1,3-dioxoisoindolin-2-yl)butyl)-3,3-
dimethylindolin-2-ylidene)penta-1,3-dien-1-yl)-1,1-dimethyl-1H-benzo[e]indol-3-ium
(5)
1
3
212
213
214
215
216
217
As previously described. [11] C NMR (850 MHz, CD OD) δ 176.56, 174.20, 169.96,
3
154.99, 154.97, 143.65, 142.57, 140.79, 135.41, 135.25, 133.47, 133.22, 131.76,
131.17, 129.77, 129.39, 128.75, 126.63, 126.18, 126.10, 124.12, 123.42, 123.36,
112.12, 111.91, 104.35, 104.08, 57.57, 57.47, 57.37, 52.56, 50.47, 44.71, 44.54,
38.10, 34.60, 31.12, 28.17, 27.99, 27.56, 27.39, 26.52, 25.74, 25.64, 17.20. HRMS
+
+
4
calculated [M] for C H N O 719.3723, found 719.3734
47
49
3

218
219
220
1-(5-carboxypentyl)-2-((1E,3E,5E)-5-(3-(4-(1,3-dioxoisoindolin-2-yl)butyl)-1,1-
dimethyl-1,3-dihydro-2H-benzo[e]indol-2-ylidene)penta-1,3-dien-1-yl)-3,3-dimethyl-
3H-indol-1-ium (6)
1
3
221
222
223
224
225
226
227
As previously described. [11] C NMR (850 MHz, CD OD) δ 177.25, 176.71, 173.93,
3
169.95, 155.06, 154.83, 143.64, 142.45, 140.87, 135.45, 133.57, 133.30, 131.81,
131.15, 129.71, 129.42, 128.80, 126.63, 126.27, 125.97, 124.19, 123.40, 123.38,
112.20, 111.80, 104.44, 104.01, 57.67, 57.57, 57.47, 57.37, 57.27, 52.64, 50.36,
45.08, 44.17, 38.13, 34.58, 31.12, 28.55, 28.01, 27.54, 27.34, 26.64, 25.73, 25.34,
+
+
17.47, 17.38, 17.29, 17.20, 17.11. HRMS calculated [M] for C H N O 719.3723,
47
49
3
4
found 719.3733
228
229
230
3-(5-carboxypentyl)-2-((1E,3E,5E)-5-(3-(4-(1,3-dioxoisoindolin-2-yl)butyl)-1,1-
dimethyl-1,3-dihydro-2H-benzo[e]indol-2-ylidene)penta-1,3-dien-1-yl)-1,1-dimethyl-
1H-benzo[e]indol-3-ium (7)
231
232
233
234
235
236
237
3-(5-carboxypentyl)-2-((1E,3E,5E)-5-(3-(4-(1,3-dioxoisoindolin-2-yl)butyl)-1,1-
dimethyl-1,3-dihydro-2H-benzo[e]indol-2-ylidene)penta-1,3-dien-1-yl)-1,1-dimethyl-
1H-benzo[e]indol-3-ium was obtained from 3-(5-carboxypentyl)-1,1,2-trimethyl-1H-
benzo[e]indol-3-ium (2.4 g, 4.8 mmol), 3-(4-(1,3-dioxoisoindolin-2-yl)butyl)-1,1,2-
trimethyl-1H-benzo[e]indol-3-ium (557.2 mg, 1.2 mmol), and N-((1E,3E)-3-
(phenylimino)prop-1-en-1-yl)aniline hydrochloride (1.4 g, 5.28 mmol) in a 16% yield
1
(150.0 mg) as a blue solid after lyophilisation. H NMR (850 MHz, CD OD) δ 8.31 (q,
3
238
239
240
241
242
243
J = 13.1 Hz, 2H), 8.23 (dd, J = 23.6, 8.4 Hz, 2H), 8.01 (dd, J = 24.3, 8.3 Hz, 2H), 7.94
(d, J = 8.4 Hz, 2H), 7.82 – 7.77 (m, 2H), 7.77 – 7.71 (m, 2H), 7.64 (p, 7.75 Hz, 2H),
7.59 (dd, J = 22.0, 8.6 Hz, 2H), 7.48 (dt, J = 19.8, 7.5 Hz, 2H), 6.64 (t, J = 12.3 Hz,
1H), 6.33 (d, J = 13.6 Hz, 1H), 4.26 (dt, J = 15.3, 7.0 Hz, 4H), 3.77 (t, J = 6.4 Hz, 2H),
2.34 (t, J = 7.2 Hz, 2H), 2.02 (s, 5H), 1.99 (s, 5H), 1.93 – 1.84 (m, 5H), 1.72 (p, J =
13
7.65 Hz, 2H), 1.58 – 1.53 (m, 2H), 1.22 (s, 4H). C NMR (850 MHz, CD OD) δ
3

244
245
246
247
248
249
177.24, 176.03, 175.86, 169.94, 154.43, 154.37, 154.31, 154.26, 140.96, 140.88,
135.39, 135.15, 134.97, 133.46, 133.35, 133.20, 131.76, 131.70, 131.14, 129.45,
129.42, 128.76, 128.71, 126.58, 126.54, 126.14, 126.04, 124.11, 123.35, 123.31,
112.11, 112.04, 104.06, 103.98, 52.48, 52.39, 44.97, 44.43, 38.13, 34.60, 31.12,
+
28.50, 27.63, 27.61, 27.35, 26.54, 25.74, 25.63. HRMS calculated [M] for
+
4
C H N O 770.3952, found 770.3989
51
51
3
250
Synthesis of the hybrid tracer
2
51
52
c[RGDyK]
2
Synthesised as previously described[16].
253
254
255
256
257
258
259
DTPA(OtBu) -NHS
4
DTPA(O-tert-butyl
ester)4
(100.0
mg,
161.9
µmol),
dipyrrolidino(N-
succinimidyloxy)carbenium hexafluorophosphate (73.0 mg, 178.1 µmol) and DiPEA
(141.0 µL, 809.3 µmol) were dissolved in DCM (5 mL, dried on 4Å molecular sieves)
and stirred for 15 h at r.t. The crude product was used without further purification in
+
+
following reaction steps. HRMS calculated [M] for C H N O
12
715.4130, found
3
4
58
4
715.4193.
260
261
262
263
264
265
266
267
268
269
Phth-(SO )Cy5-c[RGDyK] (8)
3
Compound 3 (20.0 mg, 26.7 µmol), dipyrrolidino(N-succinimidyloxy)carbenium
hexafluorophosphate (12.0 mg, 29.3 µmol) and DiPEA (23.0 µL, 133.3 µmol) were
dissolved in DMSO (400 μL). After stirring for 30 min at r.t., full conversion was
confirmed by thin layer chromatography (TLC) and MALDI-TOF. c[RGDyK] (18.0 mg,
29.3 µmol) and DiPEA (9.3 µL, 53.0 µmol) were added and the solution was stirred at
r.t. for 72 h whereafter TLC and MALDI-TOF indicated full consumption of c[RGDyK].
H O (0.1% v/v TFA) (2.4 mL) and CH CN (0.1% v/v TFA) (3 mL) were added and the
2
3
compound was purified by reversed phase (RP)-HPLC and lyophilised, yielding a
1
blue solid (10.0 mg, 28% isolated yield). H NMR (850 MHz, CD OD) δ 8.28 (t, J =
3

270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
13.0 Hz, 1H), 8.19 (t, J = 13.0 Hz, 1H), 7.86 – 7.83 (m, 3H), 7.82 (dd, J = 8.2, 1.6 Hz,
1H), 7.78 (dd, J = 5.4, 3.0 Hz, 2H), 7.54 (d, J = 7.4 Hz, 1H), 7.44 (t, J = 7.7 Hz, 1H),
7.37 (d, J = 7.9 Hz, 1H), 7.32 (t, J = 7.4 Hz, 1H), 7.28 (d, J = 8.3 Hz, 1H), 6.99 (d, J =
8.5 Hz, 2H), 6.67 (d, J = 8.4 Hz, 2H), 6.61 (t, J = 12.3 Hz, 1H), 6.39 (d, J = 13.7 Hz,
1H), 6.21 (d, J = 13.3 Hz, 1H), 5.21 (br. s, J = 59.6 Hz, 1H), 4.74 (t, J = 7.0 Hz, 1H),
4.40 (t, J = 7.9 Hz, 1H), 4.29 – 4.23 (m, 2H), 4.19 (t, J = 7.4 Hz, 2H), 4.11 (br. s, 2H),
3.90 (dd, J = 10.9, 3.8 Hz, 1H), 3.74 (br. s, J = 6.1 Hz, 2H), 3.57 (s, 1H), 3.39 – 3.37
(m, 1H), 3.34 – 3.32 (m, 1H), 3.28 (s, 1H), 3.23 – 3.21 (m, 1H), 3.21 – 3.17 (m, 1H),
3.15 – 3.10 (m, 1H), 3.06 (t, J = 7.2 Hz, 2H), 2.86 (d, J = 7.9 Hz, 2H), 2.80 (dd, J =
16.4, 7.6 Hz, 1H), 2.65 (s, 1H), 2.58 (dd, J = 16.4, 6.5 Hz, 1H), 2.21 (t, J = 7.3 Hz,
2H), 2.03 (s, 1H), 1.99 (s, 1H), 1.95 (d, J = 11.7 Hz, 1H), 1.90 – 1.78 (m, 7H), 1.74 (s,
6H), 1.72 – 1.60 (m, 10H), 1.56 – 1.51 (m, 1H), 1.51 – 1.44 (m, 3H), 1.44 – 1.38 (m,
1H), 1.38 – 1.29 (m, 2H), 1.28 (s, 1H), 1.21 (s, 1H), 1.12 (s, 1H), 1.11 – 1.07 (m, 1H),
+
+
1.00 – 0.92 (m, 1H). HRMS calculated [M] C H N O S 1350.6107, found
70
86 12 14
1350.6104
285
286
287
288
289
290
291
292
293
294
NH -(SO )Cy5-c[RGDyK] (9)
2
3
As described by Gromov et al. [30], i.e., compound 8 (10.0 mg, 7.4 µmol) was
dissolved in DMF (6 mL), and CH NH (33 wt% in ethanol) (20 mL) was added. After
3
2
stirring at r.t. for 16 h, ethanol and CH NH were removed in vacuo whereafter the
3
2
product was redissolved in ethanol and precipitated in diethyl ether (45 mL) and
centrifuged subsequently. After decanting, the precipitate was washed with ethyl
acetate (45 mL), centrifuged again and this was repeated twice with diethyl ether (45
mL). The blue solid was desiccated, redissolved in H O/CH CN (4 mL; 1:1; 0.1% v/v
2
3
+
TFA) and used directly in the following steps. HRMS calculated [M] for
+
C H N O S 1220.6052, found 1220.6062
62
84 12 12

295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
DTPA-(SO )Cy5-c[RGDyK] (10)
3
Compound 9 (4.0 mg, 3.3 µmol) and DTPA(OtBu) -NHS (11.7 mg, 16.4 µmol) were
4
dissolved in DMF (1.5 mL) followed by addition of DiPEA (5.7 µL, 32.7 µmol). The
mixture was stirred for 4 h at r.t. until H O (0.1% v/v TFA) (2.5 mL) was added. The
2
crude product was purified by RP-HPLC and lyophilised. To the resulting blue solid a
95:5 mixture of TFA/H O was added, resulting in a reddish solution which was
2
shaken at room temperature for 2 h. Hereafter, the solvent was removed in vacuo
and final purification was performed using RP-HPLC, yielding a blue solid (270.0 µg,
1
5% isolated yield) after lyophilisation. H NMR (850 MHz, DMSO) δ 9.33 – 8.92 (m,
1H), 8.58 – 8.12 (m, J = 65.6, 35.6, 10.7 Hz, 3H), 7.87 – 7.75 (m, 1H), 7.74 – 7.57
(m, 2H), 7.42 – 7.32 (m, 2H), 7.30 – 7.21 (m, 2H), 7.04 – 6.86 (m, 2H), 6.80 – 6.53
(m, 10H), 4.66 – 4.51 (m, 1H), 4.47 – 4.36 (m, 1H), 4.36 – 4.29 (m, 1H), 4.26 (br s,
1H), 4.18 (s, 2H), 4.16 – 4.10 (m, 1H), 4.07 – 4.01 (m, 1H), 3.89 (dd, J = 11.1, 6.5
Hz, 1H), 3.68 – 3.59 (m, 2H), 3.51 (s, 2H), 3.46 – 3.42 (m, 5H), 2.99 (d, J = 7.3 Hz,
3H), 2.32 – 2.10 (m, 2H), 2.11 – 2.04 (m, 4H), 2.03 – 1.99 (m, 1H), 1.89 (s, 6H), 1.75
(s, 6H), 1.69 (s, 5H), 1.61 (s, 6H), 1.54 – 1.44 (m, 5H), 1.39 (dd, J = 13.5, 6.9 Hz,
2H), 1.35 (s, 2H), 1.23 (s, 2H), 1.15 (s, 4H), 1.05 (d, J = 7.0 Hz, 1H), 0.98 (d, J = 7.5
+
+
Hz, 1H), 0.92 – 0.76 (m, 4H). LRMS calculated [M+H] for C H N O S 1597.749,
76
107 15 21
found 1597.766
314
2.2.
Stability towards glutathione
315
316
317
318
319
320
321
In an in vivo setting, interaction with the endogenous nucleophile glutathione can
lead to unwanted adducts caused by cleavage. This experiment was performed
based on previously published procedures [9], with the following deviations: 20 µL of
the mixture was injected onto a Waters 1525EF HPLC system using a SunFire C18
100 Å, 3.5 μm (150 × 4.6 mm) column and a gradient of 0.1% TFA in H O/0.1% TFA
2
in CH CN 95:5 to 0.1% TFA in H O/0.1% TFA in CH CN 5:95 in 24 min. The mixture
3
2
3
was then incubated at 37 °C for 6 h whereafter another 20 µL was injected for HPLC

3
22
23
analysis. The stability of the fluorophores was calculated relative to the integration of
the chromatogram at t = 0 h.
3
3
24
25
2.3.
(Photo)physical properties
3
2.3.1. Lipophilicity
326
327
328
329
330
331
332
333
334
335
336
337
338
500 µL of 1-octanol was added to a 4 µM solution of compound in phosphate-
buffered saline (PBS; 500 µL) in a 2 mL Eppendorf tube (n = 6), or vice versa. The
tubes were vortexed at maximum speed using an IKA Vibrofix VF1 (IKA®-Werke
GmbH & Co. KG, Staufen, Germany) for 3 min and consecutively centrifuged for 5
min at 6,000 x g in an Eppendorf 5415D centrifuge (Eppendorf, Hamburg, Germany).
Hereafter, 100 μL of the organic phase as well as the aqueous phase was
transferred to a new Eppendorf and this was also done for the aqueous phase. To
the organic phase, 100 μL of PBS was added followed by 200 μL of ethanol (EtOH).
To the aqueous phase, 100 μL of 1-octanol followed by 200 μL of EtOH were added.
Of each mixture (n = 12 per compound), 200 µL was transferred to a white Lumitrac
96-wells plate and fluorescence was measured using a Perkin Elmer LS55 (Perkin
Elmer, Waltham, USA) equipped with a plate reader, whereafter the lipophilicity
(LogP ) was calculated.
o/w
339
2.3.2. Serum protein interaction
340
341
342
343
Serum protein interaction was measured (n = 3) as described previously [9] with the
following deviations: The samples were incubated for 4 hours and a control sample
was included by adding compound to the reservoir chamber containing foetal calf
serum (FCS).
344
2.3.3. Molar extinction coefficient (ε)
3
45
46
Molar extinction coefficient was assessed as described previously [9] with the
following deviations: Stock solutions in DMSO-d were diluted in DMSO, H O or PBS
3
6
2

3
47
48
to obtain a concentration range of 7.5–0.25 µM. All optical densities were measured
in 1 mL disposable plastic cuvettes (l=1 cm; Brand, Germany).
3
3
49
50
2.3.4. Relative fluorescence quantum yield (Φ ), Stokes shift and
F
3
Brightness
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
To determine the Φ , solutions in PBS were made for each fluorophore with a
F
concentration where Amax < 0.100. Then, the absorbance at λ = 605 nm (1, 3 and 4),
(2), λ = 630 nm (6 and 5), or λ = 640 nm (7) was measured. Fluorophores were
excited at the abovementioned wavelengths to record fluorescence spectra. Dilutions
were made by exchanging 500 µL of solution with PBS consequently, until at least 3
different concentrations were measured. The absorbance at the corresponding
wavelength was then correlated to the sum of the fluorescence emission. The
regression coefficient of the resulting plot was then compared to the regression
coefficient of a suitable, known reference compound (either Methyl-(SO )Cy5-
3
COOH, EuK-Cy5(Ar)-mas , or sulfonate-(Ar)Cy5(Ar)-COOH; Φ = 13% [9], 12%
3
F
[11], and 10% [13], respectively). The Stokes shift was determined using the
λex,max/λem,max values of the recorded spectra. The brightness of the fluorophores was
calculated by multiplying the corresponding molar extinction coefficient (ε) with the
corresponding quantum yield (Φ ). The brightness of 10 was determined by using the
F
molar extinction coefficient of the corresponding unconjugated fluorophore.
366
2.3.5. Stacking behaviour
367
368
369
These values were determined as described previously [9], with the following
deviations: DMSO stock solutions of each fluorophore were diluted to concentrations
ranging from 0.78 µM to 50 µM.
370
2.4.
Photostability
3
71
72
The photostability assay was carried out based on previously reported
methods [13], with the following deviations: the cuvettes were placed at a 5 cm
3

373
374
375
376
377
378
379
380
381
382
383
distance in front of a prototype Karl Storz camera setup (Karl Storz Endoskope
GmbH & Co. KG, Tuttlingen, Germany). This camera setup included an IMAGE1 S
H3-Z FI Three-Chip FULL HD camera head equipped with a 0° laparoscope in
combination with an IMAGE 1ꢀS CONNECT module, an IMAGE 1ꢀSꢀH3-LINK link
module and a Cy5-modified D-light C light source (590–680ꢀnm emission). A
standard eyepiece adapter containing a bandpass filter (640–720ꢀnm; Cat no.
20100034; Karl Storz Endoskope GmbH & Co. KG) was placed between the camera
and the laparoscope to image the Cy5 fluorescence. The sum of the fluorescence
emission was plotted for each fluorophore at the different time points. It must be
noted that the time required for direct illumination during surgical resection is well
below the 30-minute direct, constant illumination timeframe used in the current setup.
384
385
386
As a control for solubility issues, the most hydrophobic compound 7 was
measured without irradiation by the Storz camera while all other variables in the
setup were kept at the same level.
387
2.5.
Signal penetration depth
388
389
390
391
392
The signal penetration depth was carried out as described by van Willigen et al. [25],
with the following deviations: All fluorophores were dissolved in H O and
2
supplemented with human serum albumin and the point at which the capillaries
became indistinguishable from diffused light was determined by three independent
observers.
393
2.6.
Flow cytometry
394
395
396
397
398
Receptor affinity (k ) of the hybrid tracer for the α β integrin was measured in
D v 3
saturation experiments using Geβ3 cells as previously published [22]. Normalized
geometric means were fitted with equations in the GraphPad Prism 7 software. The
kD values were calculated using the “Binding–Saturation, One site–Total” nonlinear
regression equation (eq 1), with the background value set at 0.

(
1)
ꢋꢌꢍꢎꢏ
log ꢀꢁꢂꢃ = log ꢄ10ꢅꢆꢇ ꢈ ꢊ1 +
ꢕꢖ
ꢉ
ꢐꢑ,ꢒꢓꢔ
399
2.7.
Radiolabelling, in vivo tumour model, and biodistribution
1
11
400
401
402
403
404
405
406
407
408
409
Radiolabelling of 10 and the reference tracer 11 with In, the development of the
orthotopic 4T1 breast tumour model in nude BALB/c mice, and biodistribution
experiments were carried out according to previously described procedures [22]. The
local ethics committee of the Leiden University Medical Center (LUMC) approved all
animal experiments prior to execution. Experiments were performed in accordance
with the Experiments on Animals Act (Wod, 2014) and the applicable legislation in
the Netherlands in accordance with the European guidelines (EU directive no.
2010/63/EU) regarding the protection of animals used for scientific purposes. Mice
were kept in accordance to the Dutch law in a licensed establishment for use of
experimental animals (animal facility of the LUMC).
410 3. RESULTS
411
3.1.
Synthesis
412
413
414
415
All fluorophores were synthesised following the procedures as previously described
[11]. The asymmetrical orthogonal 2 could readily be conjugated to DTPA and
c[RGDyK] after liberating the amine, yielding the tracer 10 (Scheme 2). Subsequent
111
111
radiolabelling of this tracer with InCl then yielded the hybrid tracer In-10.
3

416
417
418
419
Scheme 2. Synthetic strategy for 10, a hybrid tracer based on the design by using a fluorophore as
linker. a) i) HSPyU, DiPEA, DMSO, r.t., 30 min, ii) c[RGDyK], DiPEA r.t., 72 h; b) CH NH (33 wt% in
EtOH), DMF, 4 h, r.t.; c) (i) DTPA(OtBu) -NHS, DMF, DiPEA, r.t., 4 h; (ii) TFA/H O (95:5), r.t., 2 h.
3
2
4
2
420
421
3.2.
Chemical stability
4
22
23
No differences in chemical stability towards glutathione (≥ 88%) independent of
4
substitution pattern were seen between the different fluorophore analogues. These

424
425
426
427
findings are in line with previous reports on the stability of cyanine fluorophores (i.e.,
cyanine fluorophores without an aryl ether bond in the polymethine chain) towards
glutathione [9, 13]. 7 was excluded from this comparison as its low polarity gave rise
to solubility issues at the required concentration.
428
429
3.3.
Lipophilicity and serum binding
430
431
432
433
434
435
436
The partition coefficient in octanol/water (LogP(o/w)) of the cyanine fluorophores can
be statistically significantly lowered by addition of sulfonates and statistically
significantly increased by the addition of an extra benzene moiety (p < 0.005, Table
1) which is in line with previously published results [31]. Interestingly enough, the
addition of two benzene moieties (resulting in 7) did not necessarily influence the
LogP(o/w) of the cyanine fluorophore to a statistically significant degree (p > 0.05,
Table 1), indicating a limitation to this extension.
437
Table 1. Lipophilicity (LogP(o/w)) and serum binding
Compound
LogP (o/w) (n = Serum binding
6)
(%
bound,
n=3)
1
2
3
4
5
1.13 ± 0.07
0.01 ± 0.01
-0.04 ± 0.01
-1.56 ± 0.01
1.05 ± 0.01
93.8 ± 1.3
87.1 ± 0.3
84.5 ± 0.3
67.3 ± 2.7
92.8 ± 0.9
6
7
1.23 ± 0.04
1.17 ± 0.03
92.7 ± 0.8
88.0 ± 1.7
438
439
440
441
442
443
444
Serum binding was influenced by molecular alterations to some extent (Table 1).
The addition of one or more sulfonate(s) statistically significantly decreased the
serum binding (p ≤ 0.0006), whereas the addition of one benzene moiety induced no
changes (p > 0.05). However, statistically significant changes were found when
comparing 7 (88.0 ± 1.7%) to 1 (93.8 ± 1.3%; p = 0.0026). An indication was found
that some correlation between the lipophilicity and serum binding exists, i.e., the

4
45
46
serum binding of a fluorophore increases as its lipophilicity of increases (r = 0.8975,
2
4
p = 0.0012; r = 0.8818; Figure SI22).
447
3.4.
Photophysical properties
448
449
450
Addition of sulfonate or benzene moieties can result in alterations to the
(photo)physical properties of the Cy5 fluorophore. The differences found in
brightness, photostability and stacking behaviour are reported below.
451
3.4.1. Brightness
452
453
454
455
456
457
458
459
460
461
462
463
All studied fluorophores exhibited a higher ε in DMSO than in H O and PBS (Table
2
2). The ε of the sulfonated Cy5 fluorophores was shown to be clearly related to the
site and number of substitutions (4 > 3 > 1 > 1; Table 2). In concurrence with
previous reports [9], the addition of one sulfonate at the C5 position of the carboxylic-
acid-containing indole moiety diminished the ε, while the addition of a sulfonate
moiety at the C5 position of the phthalimide-containing indole moiety caused an
increase in the ε; a trend that seems inapplicable to the benzo[e]indole-containing
fluorophores (Table 2). Interestingly, the addition of a sulfonate on the phthalimide-
bearing indole increased the ε to a larger extent than a sulfonate on the carboxylic-
acid-bearing indole. This finding was unexpected, as the phthalimide does not
participate in the conjugated system of the cyanine as it is separated from this
system by four methylenes.
464
465
466
467
The Φ of sulfonated Cy5 fluorophores 3, 2, and 4 was higher than the Cy5
F
fluorophore without C5 indole substitutions 1 and the benzo[e]indole-containing Cy5
fluorophores 6, 5, and 7 (Table 2). These data resulted in the following trend in
brightness: 4 >> 3 > 2 > 1 > 6 > 7 > 5 (Table 2).
4
68
69
4
Table 2. Photophysical properties of Cy5 fluorophores and hybrid tracer analogues.

Compound
Molar
coefficient
extinction
λ
ex,max/λem,max
Φ
(%)
F
Brightne
ss
Photostabi
lity (%
remaining
fluorescen
ce, n=4)
(Stokes shift; nm)
5
-1
-1
3
-1
(
ε; · 10 M · cm )
(· 10
M
-
1
·
cm )
DMSO
2
H O
PBS
DMSO
2
H O
PBS
PBS
PBS
1
2
3
4
5
1.86
1.36
2.23
3.06
1.78
0.92
0.82
1.29
2.60
0.46
0.93[11]
0.95[11]
1.46[11]
2.86
650/670 (20)
655/676 (21)
655/674 (19)
660/680 (20)
670/691 (21)
644/656 (12)
646/663 (17)
643/662 (19)
649/666 (17)
644/660 (16)
646/664 (18)
643/662 (19)
4
6
6
3.72
5.70
8.76
42.9
0.74
70.1 ± 6.0
85.0 ± 2.1
86.2 ± 4.3
90.6 ± 3.7
57.7 ± 2.4
649/666 (17) 15
§
0.37[11]
669/677 (8)
662/680 (18)
662/679 (17)
n.d.*
2
4
4
§
6
1.89
1.81
0.43
0.27
0.60[11]
0.50
670/692 (22)
689/710 (21)
663/691 (28)
2.40
2.00
53.8 ± 4.6
24.4 ± 5.6
*
7
n.d.
§
4
70
71
measurements influenced by poor solubility
4
* could not be measured due to poor solubility
472
3.4.2. Photostability
473
474
475
476
477
The performed experiments indicated that the addition of sulfonates
increased the photostability, but this effect was not dependent on the amount of
sulfonates (1 < 3 ≈ 2 ≈ 4; p < 0.05; Table 2). The photostability of the benzo[e]indole-
containing cyanine fluorophores was lower and dependent on the degree of
functionalisation (1 > 6 ≈ 5 > 7; p < 0.05 Table 2).
478
3.4.3. Stacking behaviour
479
480
481
482
483
484
485
486
When the presented fluorophores were analysed for their stacking behaviour, it
was found that the incorporation of benzo[e]indole moieties drove both H- and J-
stacking (i.e., a sandwich-type or brickstone arrangement, respectively) [32] in both
H O and PBS (Figure 1). Here it seems that the planar and hydrophobic nature of
2
these fluorophores (see Table 1, LogP(o/w)), drives the van der Waal’s forces that
cause formation of H- or J-aggregates [13]. Conversely, the introduction of a single
(tetrahedral) sulfonate moiety on either indole decreased H- and J-stacking (Figure 1)
as a result of Coulombic repulsion [32].
487

488
489
490
491
492
Figure 1. Stacking behaviour of 1–7 in different solvents. Stacking of the individual fluorophores was
measured at different concentrations (0.78–50 µM) in DMSO, H O and PBS. Data show severe stacking
O and PBS. Arrows indicate change in absorbance
2
for all benzene- and unsubstituted fluorophores in H
2
intensity between spectra.

493
3.5.
Signal penetration depth in porcine tissue
494
495
496
497
498
499
The fluorescence penetration depth in porcine tissue was measured using a
clinical grade Storz laparoscopic camera and compared to the brightness of the
corresponding fluorophore. In concordance with previously published results [25], all
fluorophores were visible up to about 7 mm of layers of porcine tissue (except for 6,
approximately 4 mm; Figure SI23). No clear correlation between the brightness of a
fluorophore and its tissue penetration could be established.
5
00
01
3.6.
In vitro and in vivo evaluation of fluorophore performance
5
in a hybrid tracer
502
503
504
505
506
2 was selected as this fluorophore composition has proven to be most favourable
in a previous study on end-labelled Cy5 fluorophores [16]. Flow cytometry
experiments have shown an affinity for the α β integrin in Geβ3 cells of k = 34.1 ±
v
3
D
13.3 nM (Figure SI24) for 10, which is comparable to the reference compound 11; kD
= 30.3 ± 5.7 nM) [16].
507

5
08
09
Figure 4. Biodistribution data of the hybrid tracer 10 vs. a reference compound 11 [16].
5
510
511
512
513
514
515
516
The biodistribution data (Figure 4) show that the uptake in tumorous tissue as
well as non-targeted tissue for 10 is significantly higher after 2 h than the uptake for
11. Only blood uptake (0.065 ± 0.034 %ID/g for 10 vs. 0.034 ± 0.0064 %ID/g for 11; p
> 0.05) after 2 h showed statistically nonsignificant changes between the two tracers.
These data imply that the tracer is cleared not only through the renal clearance
pathway, but also through the hepatobiliary pathway and that the clearance rate for
10 is much slower compared to 11.
517 4. DISCUSSION
518
519
520
521
522
523
By creating seven structurally different asymmetrical Cy5 fluorophores (Scheme 1),
the influences that substituents at the (C4 and) C5 position on either (or both)
indole(s) have on several (photo)physical properties (i.e., lipophilicity, serum protein
binding, molar extinction coefficient (ε), fluorescence quantum yield, stacking
behaviour, chemical- and photostability, and penetration depth in porcine tissue) of
cyanine fluorophores could be mapped.
524
525
526
527
528
529
530
531
532
533
Prior to these findings, it was Fisher et al. [8] and Spa et al. [9] who indicated
that no clear trend between cyanine substitution pattern on the indoles and Φ could
F
be established, but the data presented here suggest otherwise; i.e., sulfonates were
shown to increase the Φ of a fluorophore whereas extra benzene moieties
F
diminished the Φ (Table 2). Stacking can affect the photophysical characteristics of
F
a fluorophore in aqueous media, e.g., induce a batho- or hypsochromic shift or
greatly diminish the brightness of a fluorophore. The cyanine core (without C5-indole
substitutions) is prone to fluorophore–fluorophore stacking [33, 34] as its molecular
structure is relatively planar. The fluorophore–fluorophore aggregation was
diminished by the addition of sulfonates and increased by the addition of extra

534
535
536
537
538
539
540
benzene rings, which correlates to the solubility of these compounds in aqueous
media. The latter, however, is also dependent on moieties conjugated on the N-
indole position. Interestingly, the variations in brightness only revealed a small
variation in the in-depth tissue penetration of the different fluorophore. Overall, 4 had
shown to possess the most favourable photophysical properties and that the addition
of benzene moieties on the cyanine backbone diminishes the ε, Φ , and thereby the
F
brightness.
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
Another trend between the substitution pattern and a (photo)physical
characteristic was found in photobleaching. Fluorophores typically undergo
photobleaching upon exposure to a bright light source. This process, where the
molecular structure of the fluorophores is disintegrated, causes a loss in fluorescence
signal intensity [35, 36]. The results imply that the addition of one (or more)
sulfonate(s) (compound 2–4) statistically significantly increases the photophysical
stability of the fluorophore when comparing to the unsubstituted fluorophore 1 (Table
2, p < 0.05), whereas benzo[e]indole-containing analogues (compound 5–7)
statistically significantly decrease the photostability of the fluorophore (Table 2, p <
0.05). This trend is in line with previously published results [9, 13] and suggests that
benzo[e]indole-containing cyanine fluorophores need sulfonate substituents to
compensate their reduced photostability. Uniquely, a synthesised fluorophore could
function as a bifunctional linker after liberation of the amine. This ability to
functionalise both the carboxylic acid and the primary amine in an orthogonal manner
allows for new possibilities in fluorescent and hybrid tracer design. Previous reports
on α β -targeted molecular imaging indicated that end-labelled fluorophores with one
v
3
sulfonate yielded the most favourable properties when conjugated to a targeting
vector [16]. In this study, in vitro and in vivo evaluation of the hybrid tracer 10 showed
that the fluorophore could function as a scaffold when conjugated to both c[RGDyK]
and DTPA in an orthogonal fashion. These experiments, however, also illustrate that

561
562
563
564
the favourable fluorophore–receptor interaction observed with our prostate-specific-
membrane-antigen-targeted hybrid tracers [11] did not translate directly to α β . This
v
3
finding thus implies that the properties of the targeting vector should dictate the use
of specific fluorophores, be it in an end-labelled or linker fashion.
565
566
567
As a result of these studies, we hope to aid chemists in their scrutiny for the
optimisation of fluorescent and/or hybrid targeted tracers and as such, achieve better
patient care.
568 5. CONCLUSION
569
570
571
572
573
574
575
The influence of structural alterations of an indole moiety at the (C4 and) C5
position on selected properties of Cy5 fluorophores has been evaluated. By doing so,
it could be concluded that sulfonates are generally favourable for the photophysical
properties of
a
fluorophore, whereas extra benzene rings bring about
disadvantageous characteristics. Furthermore, it has been shown that with the
availability of (a)symmetrical orthogonal Cy5 fluorophores, a door has been opened
for new and innovative tracer designs and synthetic strategies.
576 6. ACKNOWLEDGEMENTS
577
578
579
580
This research was funded by an NWO‐TTW‐VICI grant (TTW 16141). The authors
thank Dr. Jan Wouter Drijfhout for the use of, and assistance with instruments, Bjorn
van Doodewaerd for support regarding HRMS measurements and Dr. Karthick Sai
Sankar Gupta for support regarding NMR measurements.
581 7. REFERENCES
582
583
584
[1] Alander JT, Kaartinen I, Laakso A, Patila T, Spillmann T, Tuchin VV, et al. A
review of indocyanine green fluorescent imaging in surgery. Int J Biomed Imaging.
2012;2012:940585.

5
85
86
[2] Marmor MF, Ravin JG. Fluorescein angiography: insight and serendipity a half
century ago. Arch Ophthalmol. 2011;129(7):943-8.
5
587
588
589
[3] Burggraaf J, Kamerling IM, Gordon PB, Schrier L, de Kam ML, Kales AJ, et al.
Detection of colorectal polyps in humans using an intravenously administered
fluorescent peptide targeted against c-Met. Nat Med. 2015;21(8):955-61.
590
[4] VEGF-targeted Fluorescent Tracer Imaging in Breast Cancer.
5
91
92
[5] Interest of Fluorescence in Salvage Surgery for Recurrence of Head and Neck
Cancer in Irradiated Area. 2012.
5
5
93
94
[6] Lavis LD, Raines RT. Bright ideas for chemical biology. ACS Chem Biol.
2008;3(3):142-55.
5
595
596
597
[7] Widengren J, Schwille P. Characterization of Photoinduced Isomerization and
Back-Isomerization of the Cyanine Dye Cy5 by Fluorescence Correlation
Spectroscopy. The Journal of Physical Chemistry A. 2000;104(27):6416-28.
598
599
600
[8] Fisher NI, Hamer FM, Pope WJ. A comparison of the absorption spectra of some
typical symmetrical cyanine dyes. Proceedings of the Royal Society of London Series
A - Mathematical and Physical Sciences. 1997;154(883):703-23.
601
602
603
604
[9] Spa SJ, Hensbergen AW, van der Wal S, Kuil J, van Leeuwen FWB. The
influence of systematic structure alterations on the photophysical properties and
conjugation characteristics of asymmetric cyanine 5 dyes. Dyes and Pigments.
2018;152:19-28.
605
606
607
608
[10] Buckle T, van Willigen DM, Spa SJ, Hensbergen AW, van der Wal S, de Korne
CM, et al. Tracers for Fluorescence-Guided Surgery: How Elongation of the
Polymethine Chain in Cyanine Dyes Alters the Pharmacokinetics of a Dual-Modality
c[RGDyK] Tracer. J Nucl Med. 2018;59(6):986-92.

609
610
611
612
[11] Hensbergen AW, Buckle T, van Willigen DM, Schottelius M, Welling MM, van der
Wijk FA, et al. Hybrid Tracers Based on Cyanine Backbones Targeting Prostate-
Specific Membrane Antigen: Tuning Pharmacokinetic Properties and Exploring Dye-
Protein Interaction. J Nucl Med. 2020;61(2):234-41.
613
614
615
[12] Schwechheimer C, Ronicke F, Schepers U, Wagenknecht HA. A new structure-
activity relationship for cyanine dyes to improve photostability and fluorescence
properties for live cell imaging. Chem Sci. 2018;9(31):6557-63.
616
617
618
[13] van der Wal S, Kuil J, Valentijn ARPM, van Leeuwen FWB. Synthesis and
systematic evaluation of symmetric sulfonated centrally C C bonded cyanine near-
infrared dyes for protein labelling. Dyes and Pigments. 2016;132:7-19.
619
620
621
[14] Choi HS, Nasr K, Alyabyev S, Feith D, Lee JH, Kim SH, et al. Synthesis and in
vivo fate of zwitterionic near-infrared fluorophores. Angew Chem Int Ed Engl.
2011;50(28):6258-63.
6
22
23
[15] Lee H, Mason JC, Achilefu S. Heptamethine cyanine dyes with a robust C-C
bond at the central position of the chromophore. J Org Chem. 2006;71(20):7862-5.
6
624
625
626
[16] Bunschoten A, van Willigen DM, Buckle T, van den Berg NS, Welling MM, Spa
SJ, et al. Tailoring Fluorescent Dyes To Optimize a Hybrid RGD-Tracer. Bioconjug
Chem. 2016;27(5):1253-8.
627
628
629
[17] Kuil J, Buckle T, Oldenburg J, Yuan H, Borowsky AD, Josephson L, et al. Hybrid
peptide dendrimers for imaging of chemokine receptor 4 (CXCR4) expression. Mol
Pharm. 2011;8(6):2444-53.
630
631
632
[18] Tan X, Luo S, Wang D, Su Y, Cheng T, Shi C. A NIR heptamethine dye with
intrinsic cancer targeting, imaging and photosensitizing properties. Biomaterials.
2012;33(7):2230-9.