BF2-Azadipyrromethene Fluorophores for Intraoperative Vital Structure Identification

A series of mono- and bis-polyethylene glycol (PEG)-substituted BF2-azadipyrromethene fluorophores have been synthesized with emissions in the near-infrared region (700–800 nm) for the purpose of fluorescence guided intraoperative imaging; chiefly ureter imaging. The Bis-PEGylation of fluorophores resulted in higher aqueous fluorescence quantum yields, with PEG chain lengths of 2.9 to 4.6 kDa being optimal. Fluorescence ureter identification was possible in a rodent model with the preference for renal excretion notable through comparative fluorescence intensities from the ureters, kidneys and liver. Ureteral identification was also successfully performed in a larger animal porcine model under abdominal surgical conditions. Three tested doses of 0.5, 0.25 and 0.1 mg/kg all successfully identified fluorescent ureters within 20 min of administration which was sustained up to 120 min. 3-D emission heat map imaging allowed the spatial and temporal changes in intensity due to the distinctive peristaltic waves of urine being transferred from the kidneys to the bladder to be identified. As the emission of these fluorophores could be spectrally distinguished from the clinically-used perfusion dye indocyanine green, it is envisaged that their combined use could be a step towards intraoperative colour coding of different tissues.


Introduction
Dynamic fluorescence imaging has the potential to provide surgical guidance by identifying specific tissues or anatomical structures in real time as an operation proceeds [1]. At present, clinically approved camera hardware exists in a variety of formats, including open, laparoscopic and microscopic, which are capable of capturing high-quality fluorescence images during surgery [2]. Surgical resections can be guided in real-time with these devices due to their inherent ease of use, enabling continuous imaging during surgical procedures [3,4]. As a result, there has been a growing interest in developing fluorophores for in vivo and clinical applications using near-infrared (NIR) light with wavelengths between 700 nm and 1400 nm [5]. Due to low tissue autofluorescence and light attenuation, this wavelength range has been described as the clinical imaging window. This offers the potential of increased personalised surgeries with improved patient outcomes. The benefits can be viewed in two broad categories of ensuring the complete removal of diseased tissue thereby preventing reoccurrences or the prevention of accidental damage to vital structures which cause additional unintended patient complications. One such avoidable complication is accidental ureter damage most commonly encountered in complex lower abdominal surgeries [6,7]. Intraoperative ureteral injuries which are not immediately repaired in situ cause patient morbidity and mortality with a significant increase in time spent in hospital and cost of treatments. Unfortunately, the majority of accidental ureter damage goes unnoticed during the operation causing occurrences of renal failure and sepsis and necessitating further corrective surgery [8,9]. The identification of ureters at the outset of a surgery can be non-trivial due to their anatomical location behind the abdominal cavity often obscured by adipose tissue [6,9]. Cancerous and inflammation tissue or previous surgeries may also hinder their detection. Over half of ureter injuries occur during gynaecological operations with a quarter reported from general surgical procedures. Elevated risks have been reported for robotic or minimally invasive surgeries due to the inability to use the traditional tactile approach [10].
Real-time tissue observations through near-infrared fluorescence is having a growing clinical impact as they do not overly impede on the existing surgical workflow. If applied to ureter preservation, key desirable features would be that the fluorescent agent reach the ureters within minutes after administration, the NIR-wavelengths used for detection are distinguishable from clinically used indocyanine green (ICG) allowing both to be used in parallel (for different tissues), and has no toxicity such that a repeat dose could be used to confirm ureter integrity at the end of the operation. As a cyanine-derived dye, ICG is currently the only NIR fluorophore, with emission spanning 800-900 nm, approved by both the FDA and the European Medicines Agency [11]. As a vascularization assessment tool, it is used most commonly during bowel anastomosis surgeries, where it has been shown to reduce complications associated with the surgery and improve patient outcomes [12,13]. Due to its biliary excretion pathway from hepatocytes into the bile, it is unsuitable for renal excretion imaging [14,15]. Recently, other fluorophore classes with similar spectral wavelengths to ICG have been described as alternatives [16][17][18][19].
In this work, we have investigated the synthesis, physical and photophysical characteristics, and ureter imaging capabilities of NIR-AZA fluorophores 1 as their pegylated derivatives (Figure 1) [20,21]. Therapeutic pegylation is a well-established means to prolong the circulation half-life of a drug molecule but its application to in vivo imaging is relatively unexplored [22][23][24][25][26][27]. It was envisaged that an optimal PEG substitution of a fluorophore could be advantageous for several reasons. It is relatively inexpensive when compared to peptide or antibody conjugates, can increase the fluorescence brightness in aqueous biological media, can restrict the non-specific uptake into tissues and may promote efficient excretion via the renal pathway [22].
To this end, a series of pegylated derivatives of the NIR-AZA fluorophores were envisaged each with different degrees of pegylation ( Figure 1). The core fluorophore used was structurally based on the bis-arylmethoxy-substituted NIR-AZA 1a which has absorbance and emission maxima of 682 and 721 nm, respectively, with a fluorescence quantum yield of 0.35 [28]. Two design approaches were envisaged for PEG attachment to the fluorophore through either a mono or dual covalent connection of the PEG groups which would allow different degrees of pegylation to be explored ( Figure 1A,B). fluorophore through either a mono or dual covalent connection of the PEG groups which would allow different degrees of pegylation to be explored ( Figure 1A,B).

Synthesis
The strategy adopted for covalent PEG attachment was to functionalize through the para-phenolic oxygen substituents of the aryl rings at the α-pyrrole positions (Figure 1).
This required the synthesis of the bis-phenol 1b and mono-phenol-monomethoxy 2 as advanced intermediates (Scheme 1). The synthesis of 1b has been previously reported in three steps from commercial starting materials 4-hydroxyacetophenone and nitromethane [29]. Two routes were developed to 2, the first of which required the reaction of 1b with iodomethane in DMSO with cesium fluoride as base, which gave the mono-alkylated product in 83% yield (Scheme 1). An alternative five-step route which avoided the use of methyl iodide started from 4-methoxyacetophenone and 4-hydroxyacetophenone providing 2 in an overall 8% yield for the route (SI Figure S1) [30]. The conversion of 1b and 2 to their bis-or mono-N-hydroxysuccinimide esters was achievable in three steps through first phenolic O-alkylation with tert-butyl bromoacetate (Scheme 2). These nucleophilic substitutions were carried out under basic conditions in anhydrous DMSO at 40 °C or THF under reflux with monitoring by thin layer chromatography. Upon completion, work up involved silica gel chromatography purifications with 3 and 5 isolated in 84 and 86% yield, respectively. The NMR and mass spectrometry analysis confirmed the structures of both products. Next, tert-butyl ester cleavage of both were achieved with trifluoroacetic acid at room temperature in CH2Cl2 over 3 h. The carboxylic acid products 3a and 5a precipitated during the reaction allowing their isolation by filtration, following which they were washed with CH2Cl2 and dried under vacuum (SI Figure   Figure 1. General structure of NIR-AZA fluorophores showing position of PEG attachments via blue coloured aryl ring and non-pegylated reference fluorophore 1a used in this study. Design of (A) mono and (B) dual substituted pegylated (blue line) NIR-AZA fluorophores (red rectangle).

Synthesis
The strategy adopted for covalent PEG attachment was to functionalize through the para-phenolic oxygen substituents of the aryl rings at the α-pyrrole positions (Figure 1). This required the synthesis of the bis-phenol 1b and mono-phenol-monomethoxy 2 as advanced intermediates (Scheme 1). The synthesis of 1b has been previously reported in three steps from commercial starting materials 4-hydroxyacetophenone and nitromethane [29]. Two routes were developed to 2, the first of which required the reaction of 1b with iodomethane in DMSO with cesium fluoride as base, which gave the mono-alkylated product in 83% yield (Scheme 1). An alternative five-step route which avoided the use of methyl iodide started from 4-methoxyacetophenone and 4-hydroxyacetophenone providing 2 in an overall 8% yield for the route (SI Figure S1) [30].
Molecules 2023, 28, x FOR PEER REVIEW 3 of 22 fluorophore through either a mono or dual covalent connection of the PEG groups which would allow different degrees of pegylation to be explored ( Figure 1A,B).

Synthesis
The strategy adopted for covalent PEG attachment was to functionalize through the para-phenolic oxygen substituents of the aryl rings at the α-pyrrole positions ( Figure 1).
This required the synthesis of the bis-phenol 1b and mono-phenol-monomethoxy 2 as advanced intermediates (Scheme 1). The synthesis of 1b has been previously reported in three steps from commercial starting materials 4-hydroxyacetophenone and nitromethane [29]. Two routes were developed to 2, the first of which required the reaction of 1b with iodomethane in DMSO with cesium fluoride as base, which gave the mono-alkylated product in 83% yield (Scheme 1). An alternative five-step route which avoided the use of methyl iodide started from 4-methoxyacetophenone and 4-hydroxyacetophenone providing 2 in an overall 8% yield for the route (SI Figure S1) [30].
The conversion of 1b and 2 to their bis-or mono-N-hydroxysuccinimide esters was achievable in three steps through first phenolic O-alkylation with tert-butyl bromoacetate (Scheme 2). These nucleophilic substitutions were carried out under basic conditions in anhydrous DMSO at 40 °C or THF under reflux with monitoring by thin layer chromatography. Upon completion, work up involved silica gel chromatography purifications with 3 and 5 isolated in 84 and 86% yield, respectively. The NMR and mass spectrometry analysis confirmed the structures of both products. Next, tert-butyl ester cleavage of both were achieved with trifluoroacetic acid at room temperature in CH2Cl2 over 3 h. The carboxylic acid products 3a and 5a precipitated during the reaction allowing their isolation by filtration, following which they were washed with CH2Cl2 and dried under vacuum (SI Figure The conversion of 1b and 2 to their bis-or mono-N-hydroxysuccinimide esters was achievable in three steps through first phenolic O-alkylation with tert-butyl bromoacetate (Scheme 2). These nucleophilic substitutions were carried out under basic conditions in anhydrous DMSO at 40 • C or THF under reflux with monitoring by thin layer chromatography. Upon completion, work up involved silica gel chromatography purifications with 3 and 5 isolated in 84 and 86% yield, respectively. The NMR and mass spectrometry analysis confirmed the structures of both products. Next, tert-butyl ester cleavage of both were achieved with trifluoroacetic acid at room temperature in CH 2 Cl 2 over 3 h. The carboxylic acid products 3a and 5a precipitated during the reaction allowing their isolation by filtration, following which they were washed with CH 2 Cl 2 and dried under vacuum (SI Figure S2). Rather than attempt a direct acid/amine-PEG coupling it was preferential to first make and isolate the mono-and bis-substituted-activated esters 4 and 6. Room temperature reaction of the mono-and bis-carboxylic acids with N-hydroxysuccinimide in anhydrous DMSO in the presence of a carbodiimide coupling reagent EDCI showed complete conversion to esters 4 and 6 after 2 h as judged by analytical HPLC (Scheme 2). Both products could be isolated following the addition of 0.5 M HCl to the reaction mixture and extraction into CH 2 Cl 2 . The organic layer was washed with H 2 O and brine, dried over Na 2 SO 4 , filtered and reduced to dryness providing the coupling-enabled fluorochromes to be isolated as solids.
Molecules 2023, 28, x FOR PEER REVIEW 4 of 22 S2). Rather than attempt a direct acid/amine-PEG coupling it was preferential to first make and isolate the mono-and bis-substituted-activated esters 4 and 6. Room temperature reaction of the mono-and bis-carboxylic acids with N-hydroxysuccinimide in anhydrous DMSO in the presence of a carbodiimide coupling reagent EDCI showed complete conversion to esters 4 and 6 after 2 h as judged by analytical HPLC (Scheme 2). Both products could be isolated following the addition of 0.5 M HCl to the reaction mixture and extraction into CH2Cl2. The organic layer was washed with H2O and brine, dried over Na2SO4, filtered and reduced to dryness providing the coupling-enabled fluorochromes to be isolated as solids. The fluorochromes were then reacted with amino-PEG polymers with molecular weights of 0.55, 4.6 or 9 KDa for 4 and 0.55, 2.9 or 4.6 KDa for 6 (Scheme 3) [31]. Reactions were carried out in DMSO at rt with conversions monitored using analytical HPLC with all reactions reaching completion within 60 min. Reactions for 4 and 6 were carried out under identical conditions, though the equivalence of H2N-PEG-OH used was 1.05 for the former and 2.1 for the latter. Purifications entailed the removal of the solvent, redissolving in water and chromatography using a Sephadex G-25 column. For photophysical measurements and in vivo studies, the products were further purified by reverse phase semiprep chromatography to yield PEGylated fluorophores 7-12 in greater than 95% purity (Scheme 3). The fluorochromes were then reacted with amino-PEG polymers with molecular weights of 0.55, 4.6 or 9 KDa for 4 and 0.55, 2.9 or 4.6 KDa for 6 (Scheme 3) [31]. Reactions were carried out in DMSO at rt with conversions monitored using analytical HPLC with all reactions reaching completion within 60 min. Reactions for 4 and 6 were carried out under identical conditions, though the equivalence of H 2 N-PEG-OH used was 1.05 for the former and 2.1 for the latter. Purifications entailed the removal of the solvent, redissolving in water and chromatography using a Sephadex G-25 column. For photophysical measurements and in vivo studies, the products were further purified by reverse phase semi-prep chromatography to yield PEGylated fluorophores 7-12 in greater than 95% purity (Scheme 3).
The conversion of tert-butyl ester 3 to mono-pegylated NIR-AZA fluorophores 7-9 can be easily followed by 1 H NMR spectroscopy through the change in chemical shift of the methylene protons which is unobstructed by the large PEG proton signals further up-field ( Figure 2). For example, a chemical shift of 4.59 ppm is observed for tert-butyl ester 3 going to 5.04 ppm for N-succinimidyl ester 4 and returning to a similar chemical shift of 4.58 ppm for PEG-amide 7. Contrastingly, the methoxy methyl 1 H NMR signal remained relatively constant at 3.85-3.9 ppm for 3, 4 and 7. The covalent attachment of the PEG groups was further confirmed by MALDI Q-TOF mass spectrometry with the expected mass ranges obtained. The conversion of tert-butyl ester 3 to mono-pegylated NIR-AZA fluorophores 7-9 can be easily followed by 1 H NMR spectroscopy through the change in chemical shift of the methylene protons which is unobstructed by the large PEG proton signals further upfield ( Figure 2). For example, a chemical shift of 4.59 ppm is observed for tert-butyl ester 3 going to 5.04 ppm for N-succinimidyl ester 4 and returning to a similar chemical shift of 4.58 ppm for PEG-amide 7. Contrastingly, the methoxy methyl 1 H NMR signal remained relatively constant at 3.85-3.9 ppm for 3, 4 and 7. The covalent attachment of the PEG groups was further confirmed by MALDI Q-TOF mass spectrometry with the expected mass ranges obtained.

Photophysical properties of 7-12
Next, the photophysical characteristics of each pegylated fluorophore were examined in acetonitrile and water. The properties of 7-12 in acetonitrile were very similar with λmax of absorbances at 675 (±5) nm, emissions at 714 (±3) nm and quantum yields between 0.26 The conversion of tert-butyl ester 3 to mono-pegylated NIR-AZA fluorophores 7-9 can be easily followed by 1 H NMR spectroscopy through the change in chemical shift of the methylene protons which is unobstructed by the large PEG proton signals further upfield ( Figure 2). For example, a chemical shift of 4.59 ppm is observed for tert-butyl ester 3 going to 5.04 ppm for N-succinimidyl ester 4 and returning to a similar chemical shift of 4.58 ppm for PEG-amide 7. Contrastingly, the methoxy methyl 1 H NMR signal remained relatively constant at 3.85-3.9 ppm for 3, 4 and 7. The covalent attachment of the PEG groups was further confirmed by MALDI Q-TOF mass spectrometry with the expected mass ranges obtained.

Photophysical properties of 7-12
Next, the photophysical characteristics of each pegylated fluorophore were examined in acetonitrile and water. The properties of 7-12 in acetonitrile were very similar with λmax of absorbances at 675 (±5) nm, emissions at 714 (±3) nm and quantum yields between 0.26

Photophysical Properties of 7-12
Next, the photophysical characteristics of each pegylated fluorophore were examined in acetonitrile and water. The properties of 7-12 in acetonitrile were very similar with λ max of absorbances at 675 (±5) nm, emissions at 714 (±3) nm and quantum yields between 0.26 and 0.32 (Table 1). As aqueous solutions, both absorption and emission maxima were slightly red shifted at 685 (±3) and 723 (±2) nm, respectively. However, when quantum yields were compared as aqueous solutions, interesting differences were revealed between the mono 7-9 and dual 10-12 PEG-substituted derivatives of similar molecular weights. For example, both 9 and 12 have comparable overall PEG mass (i.e., 1 × 9 kDa for 9 and 2 × 4.6 kDa for 12), yet the quantum yield of 12 is four-fold higher than that of 9 (Table 1, entries 3 and 6). A similar trend was observed for mono-PEG 8 and dual-PEG 11 with quantum yields of 0.11 and 0.4, respectively (entries 2 and 5). The higher quantum yields for the dual substituted derivatives could be attributable to better shielding of the fluorophore from the bulk water by having two points of PEG attachment rather than one. A comparison of the absorption spectra of 8 with 11 and 9 with 12 showed that the mono-substituted compounds had slightly broadened spectra with full width at half maximum (fwhm) values of 109 versus 94 nm and 114 versus 93 nm, respectively ( Figure 3A,B). While only subtle differences were noted in the absorbance, this did have a larger impact in the fluorescence intensity which was markedly higher for the dual substituted derivatives 11 and 12. Clearly, PEG molecular weight alone is not the sole determinant of aqueous quantum yields with the double substitution providing a more favourable solvated micro-environment for the fluorophores, resulting in higher quantum yields. The impact of water was confirmed by comparing absorption and emission spectra of 12 in water with that of acetonitrile ( Figure 3C). This showed that absorbance was sharper in acetonitrile (fwhm 75 nm) with the quantum yield 2.5 times higher than water indicating that pegylation is a good strategy to impart aqueous solubility for fluorophores but the number and position of the PEG groups may be required to optimize quantum yield. between the mono 7-9 and dual 10-12 PEG-substituted derivatives of similar molecular weights. For example, both 9 and 12 have comparable overall PEG mass (i.e., 1 × 9 kDa for 9 and 2 × 4.6 kDa for 12), yet the quantum yield of 12 is four-fold higher than that of 9 (Table 1, entries 3 and 6). A similar trend was observed for mono-PEG 8 and dual-PEG 11 with quantum yields of 0.11 and 0.4, respectively (entries 2 and 5). The higher quantum yields for the dual substituted derivatives could be attributable to better shielding of the fluorophore from the bulk water by having two points of PEG attachment rather than one. A comparison of the absorption spectra of 8 with 11 and 9 with 12 showed that the monosubstituted compounds had slightly broadened spectra with full width at half maximum (fwhm) values of 109 versus 94 nm and 114 versus 93 nm, respectively ( Figure 3A,B). While only subtle differences were noted in the absorbance, this did have a larger impact in the fluorescence intensity which was markedly higher for the dual substituted derivatives 11 and 12. Clearly, PEG molecular weight alone is not the sole determinant of aqueous quantum yields with the double substitution providing a more favourable solvated micro-environment for the fluorophores, resulting in higher quantum yields. The impact of water was confirmed by comparing absorption and emission spectra of 12 in water with that of acetonitrile ( Figure 3C). This showed that absorbance was sharper in acetonitrile (fwhm 75 nm) with the quantum yield 2.5 times higher than water indicating that pegylation is a good strategy to impart aqueous solubility for fluorophores but the number and position of the PEG groups may be required to optimize quantum yield.  Assessment of the most promising fluorophores 11 and 12 using clinical imaging equipment was next carried out. Polytetrafluoroethylene (PTFE) tubing of 2 mm external diameter was filled with PBS solutions of 11 or 12 at concentrations of 5.0, 1.0 or 0.5 µM. Typically, a human ureter has a diameter of 4-6 mm and may be masked by adipose tissue, so prior to imaging, tubes were horizontally placed in a vessel and covered with ei- Assessment of the most promising fluorophores 11 and 12 using clinical imaging equipment was next carried out. Polytetrafluoroethylene (PTFE) tubing of 2 mm external diameter was filled with PBS solutions of 11 or 12 at concentrations of 5.0, 1.0 or 0.5 µM. Typically, a human ureter has a diameter of 4-6 mm and may be masked by adipose tissue, so prior to imaging, tubes were horizontally placed in a vessel and covered with either water or 1% intralipid solution to a depth of 1 cm to simulate the effect of light scattering through tissue [32]. Images were acquired using a Quest Spectrum clinical instrument fitted with an open camera positioned 30 cm from the tubes and excitation wavelength set at 680 nm and collection filters from 700 to 830 nm [3,33]. Encouragingly, in all cases the linear fluorescence from within the tubes could be detected with intensity decreasing with lower concentrations. As would be expected, fluorescence scattering was noted for tubes immersed in intralipid solution though this did not inhibit identifying their positions even at the lowest concentration of 0.5 µM (Figure 4). Assessment of the most promising fluorophores 11 and 12 using clinical imaging equipment was next carried out. Polytetrafluoroethylene (PTFE) tubing of 2 mm external diameter was filled with PBS solutions of 11 or 12 at concentrations of 5.0, 1.0 or 0.5 µM. Typically, a human ureter has a diameter of 4-6 mm and may be masked by adipose tissue, so prior to imaging, tubes were horizontally placed in a vessel and covered with either water or 1% intralipid solution to a depth of 1 cm to simulate the effect of light scattering through tissue [32]. Images were acquired using a Quest Spectrum clinical instrument fitted with an open camera positioned 30 cm from the tubes and excitation wavelength set at 680 nm and collection filters from 700 to 830 nm [3,33]. Encouragingly, in all cases the linear fluorescence from within the tubes could be detected with intensity decreasing with lower concentrations. As would be expected, fluorescence scattering was noted for tubes immersed in intralipid solution though this did not inhibit identifying their positions even at the lowest concentration of 0.5 µM (Figure 4).

Ureter Identification in Rodent Model
As a preliminary screening of in vivo brightness, equal 1 mg/kg concentrations of fluorophores 8, 9, 11, and 12 were administered by tail vein injections and accumulation of fluorophores in the bladder was estimated by comparison of fluorescence intensity at 60 min through the intact skin and tissue. It was found that while all fluorophores accumulated in the bladder 11 and 12 performed better than 8 and 9 based upon averaged fluorescence intensity measurements. The relative bladder emission intensity, taken 1 h post administration, for the four fluorophores 8:9:11:12 was 1.3:1:1.6:1.9 (SI Figure S4). While there was only a minor difference between 11 and 12, the marginally better 12 was chosen for more advanced in vivo studies.
To gain a first estimate range of concentrations of 12 useful for imaging, a dose deescalation study was undertaken using 0.2, 0.1 and 0.05 mg/kg of 12. It was confirmed that each dose gave rise to a sufficient accumulation of 12 in the bladder at 60 min such that fluorescence was clearly detectable through the intact tissue and skin (SI Figure S5). Urine samples were taken for each dosage and the presence of intact 12 was confirmed by fluorescence spectroscopy and HPLC ( Figure 5). While fluorescence intensities fluctuated due to variable urine volumes, in each case the expected emission maximum (721 nm) and HPLC retention time was comparable to a standard analysis of 12. Using a standard concentration versus fluorescence intensity of 12 for comparison, it could be estimated that 30% of 12 had been excreted via the renal pathway by this time (SI Table S1). samples were taken for each dosage and the presence of intact 12 was confirmed by fluo rescence spectroscopy and HPLC ( Figure 5). While fluorescence intensities fluctuated du to variable urine volumes, in each case the expected emission maximum (721 nm) an HPLC retention time was comparable to a standard analysis of 12. Using a standard con centration versus fluorescence intensity of 12 for comparison, it could be estimated tha 30% of 12 had been excreted via the renal pathway by this time (SI Table S1).  Next, surgical exposure of the lower abdomen was carried out prior to fluorophore administration such that direct observation of the ureters and other abdominal organs was possible. Fluorophore 12 was selected for this study being administered at a dose of 0.5 mg/kg with imaging continued for 90 min. Immediately following the tail vein injection, the blood vessel fluoresced with the kidneys also becoming fluorescent within seconds which was sustained throughout the 90 min (SI Figure S6). In each animal the ureters were identifiable within 20 min post administration and this maintained throughout the 90 min imaging period. A representative image (displayed both as fluorescence pseudo coloured green overlaid on the white light image and as fluorescence alone in black and white) is shown in Figure 6 panel A which was recorded after 35 min. In this image, both ureters (red arrows) are visible as is the highly emissive bladder (asterisk) (see SI movie S1). A clear predisposition of fluorescence from the kidneys relative to the liver and other organs existed throughout, confirming the preference for renal clearance of 12. A representative example of this is shown in Figure 6 panel B which was recorded 50 min post administration. In this case the animal was positioned to allow a clearer view of one ureter (red arrow), a kidney (circle), spleen (triangle) and liver (square) such that a relative fluorescence intensity quantification could be carried out for each organ ( Figure 6, panel B, left). An averaged pixel intensity was determined for four regions of interest of the ureter, kidney, spleen and liver as indicated by the yellow boxes (panel B, right). The emission ranged from highest to lowest for the ureter >> kidney >> liver > spleen with ratio intensity values of 4.8:2.8:1.1:1, respectively ( Figure 6B, SI Figure S7 for pixel analysis data). row), a kidney (circle), spleen (triangle) and liver (square) such that a relative fluorescence intensity quantification could be carried out for each organ ( Figure 6, panel B, left). An averaged pixel intensity was determined for four regions of interest of the ureter, kidney, spleen and liver as indicated by the yellow boxes (panel B, right). The emission ranged from highest to lowest for the ureter >> kidney >> liver > spleen with ratio intensity values of 4.8:2.8:1.1:1, respectively ( Figure 6B, SI Figure S7 for pixel analysis data).

Toxicity study of 12
A toxicology study of 12 was conducted using six weeks old male Sprague-Dawley rats weighing between 130 and 240 g. The objective of this study was to evaluate 12 using two different dosages administered by an intravenous (bolus) tail vein injection. Due to the fact that this was a preclinical acute toxicity study, high doses were used since the results would provide valuable information about this compound. The doses tested were (i) seven doses of 2 mg/kg administered on seven consecutive days amounting to a total of 14 mg/kg and (ii) a single injection of 10 mg/kg. PBS solutions of 12 were prepared at concentrations of 0.4 or 2 mg/mL, with PBS used as a control. Blood samples were acquired on day 8 and following euthanasia, organs were collected and weighed. No unscheduled deaths occurred during the study and no test item-related clinical signs were observed during the study. In comparison to control animals, there was no effect on body

Toxicity Study of 12
A toxicology study of 12 was conducted using six weeks old male Sprague-Dawley rats weighing between 130 and 240 g. The objective of this study was to evaluate 12 using two different dosages administered by an intravenous (bolus) tail vein injection. Due to the fact that this was a preclinical acute toxicity study, high doses were used since the results would provide valuable information about this compound. The doses tested were (i) seven doses of 2 mg/kg administered on seven consecutive days amounting to a total of 14 mg/kg and (ii) a single injection of 10 mg/kg. PBS solutions of 12 were prepared at concentrations of 0.4 or 2 mg/mL, with PBS used as a control. Blood samples were acquired on day 8 and following euthanasia, organs were collected and weighed. No unscheduled deaths occurred during the study and no test item-related clinical signs were observed during the study. In comparison to control animals, there was no effect on body weight during the course of the study (Figure 7). No organ weight changes occurred, at any of the doses tested, that were considered to be related to the administration of 12 (Table S2). weight during the course of the study (Figure 7). No organ weight changes occurred, a any of the doses tested, that were considered to be related to the administration of 12 (Table S2). The haematology parameters measured included erythrocyte count, hemoglobin thrombocyte count, leucocyte count, differential white cell count and reticulocyte count Blood chemistry for sodium, potassium, calcium, chloride, inorganic phosphorus, glu cose, urea, creatinine, total bilirubin, total protein, albumin, albumin/globulin ratio, tota cholesterol, triglycerides, alkaline phosphatase, alanine aminotransferase, aspartate ami The haematology parameters measured included erythrocyte count, hemoglobin, thrombocyte count, leucocyte count, differential white cell count and reticulocyte count. Blood chemistry for sodium, potassium, calcium, chloride, inorganic phosphorus, glucose, urea, creatinine, total bilirubin, total protein, albumin, albumin/globulin ratio, total cholesterol, triglycerides, alkaline phosphatase, alanine aminotransferase, aspartate aminotransferase and bile acids were determined. No statistically significant differences from controls for blood biochemistry parameters were observed during the study with for example, alanine transaminase (ALT), aspartate transaminase (AST), creatinine and urea levels comparable for both treated and control animal groups (Table S3). No statistically significant haematology differences were observed for white blood cells, neutrophils, lymphocytes, or monocytes relative to controls ( Table 2). A complete post-mortem examination was performed on all animals, including examination of the external surfaces, organs and tissues. No adverse macroscopic or microscopic findings were observed, and no histologic changes were detected in the liver tissue. At the injection sites, microscopic findings were observed with similar incidences and severities as those in controls and were considered to be related to the injection procedure. The study outcome recorded a No Observable Adverse Effect Level (NOAEL) at 10 mg/kg/day for 12.

Porcine Ureter Imaging
As a next step, ureteral identification with 12 was performed in porcine models under abdominal surgical conditions. Three experimental quantities were selected as 0.5, 0.25 and 0.1 mg/kg in order to identify a range of potential clinical doses (n = 2 for each dose). It is worth noting that as 12 incorporates a dual 4.6 KDa PEG substituent, only 6.3% of the total molecular weight arises from the fluorophore itself. With respect to just the fluorophore component of 12, these doses would be in the range of 0.028 to 0.0056 mg/kg. Images with 12 were acquired using a pre-clinical open-camera system (Perkin Elmer Solaris) by excitation at 660 nm and fluorescence collection between 700-800 nm while simultaneously capturing white light video [34]. This instrument also had the ability to image ICG using a 790 nm excitation and collection between 800-900 nm which allowed confirmation that both 12 and ICG could be visualised independently of each other. The goals were to identify the earliest time point of emissive ureters, establish ureter fluorescence longevity and confirm that 12 and ICG could be imaged independently of each other. Experimentally, images were recorded post administration at 10, 20, 40, 80 and 120 min (Figure 8, for other representative images see SI Figure S8).
Encouragingly, for all concentrations tested, ureters could be identified from the fluorescence image by 20 min post-administration. For the lowest dose of 0.1 mg/kg, ureteral fluorescence was still observable at 80 min and for the two higher doses the fluorescence remained strong until the end of the experiment at 120 min. An ICG reference phantom card, which was included in the imaging field of view (FOV), was non-emissive during this imaging, confirming wavelength fidelity. At this stage, administration of ICG (0.5 mg/kg) was carried out and ureters re-examined to check the emission from 12 (excitation 660 nm) was recordable without interference from ICG, while ICG emission (excitation 790 nm) was notable from the phantom card and other regions of the bowel (Figure 9). This confirmed the clinical potential for both 12 and ICG to be utilised simultaneously to colour code different tissues. with 12 were acquired using a pre-clinical open-camera system (Perkin Elmer Solaris) by excitation at 660 nm and fluorescence collection between 700-800 nm while simultaneously capturing white light video [34]. This instrument also had the ability to image ICG using a 790 nm excitation and collection between 800-900 nm which allowed confirmation that both 12 and ICG could be visualised independently of each other. The goals were to identify the earliest time point of emissive ureters, establish ureter fluorescence longevity and confirm that 12 and ICG could be imaged independently of each other. Experimentally, images were recorded post administration at 10, 20, 40, 80 and 120 min (Figure 8, for other representative images see SI Figure S8). Encouragingly, for all concentrations tested, ureters could be identified from the fluorescence image by 20 min post-administration. For the lowest dose of 0.1 mg/kg, ureteral fluorescence was still observable at 80 min and for the two higher doses the fluorescence remained strong until the end of the experiment at 120 min. An ICG reference phantom card, which was included in the imaging field of view (FOV), was non-emissive during this imaging, confirming wavelength fidelity. At this stage, administration of ICG (0.5 mg/kg) was carried out and ureters re-examined to check the emission from 12 (excitation 660 nm) was recordable without interference from ICG, while ICG emission (excitation 790 nm) was notable from the phantom card and other regions of the bowel (Figure 9). This confirmed the clinical potential for both 12 and ICG to be utilised simultaneously to colour code different tissues.

3D fluorescence intensity heat map
The passage of urine from the kidney to the bladder is controlled by a ureteral peristalsis mechanism. Ureteral peristalsis is the wave of muscle contraction that pumps urine, in pulses, through the ureter from the kidney into the bladder. This means that as long as

3D Fluorescence Intensity Heat Map
The passage of urine from the kidney to the bladder is controlled by a ureteral peristalsis mechanism. Ureteral peristalsis is the wave of muscle contraction that pumps urine, in pulses, through the ureter from the kidney into the bladder. This means that as long as a fluorophore remains within the urine and does not enter the cells comprising the ureter, then the emission intensity emanating from the ureter will rise and fall in response to the ureteral peristaltic phases [35]. This offers an opportunity to distinguish the ureter from any background fluorescence as it would not have this repetitive pulsed modulation of intensity. We have previously developed and reported on a 3-D emission intensity heat map display which could be used to record these dynamic changes in intensity signal [36,37]. To this end, a split display was adopted with normal video and fluorescence overlay pseudo coloured as white on the lower level and the fluorescence intensity map above. Fluorescence is displayed as a heat map from blue (no emission) to red (high emission) and with a normalized intensity scale from 0 to 1. This allows relative brightness to be compared for an entire field of view making spatial and temporal changes in intensity obvious. A representative example recorded over 5 s is shown in Figure 10, in which the intensity rise and fall can be tracked, due to a single peristaltic wave, in the heat map time view (SI Movie S2). It is hoped that such recording of these intermittent visualizations coupled to tissue recognition software would allow the ureter location to be identifiable for the entire surgery.

General
All reagents were used as received without further purification. Air-sensitive reagents were reacted in oven-dried glassware under nitrogen using syringe-septum cap techniques. All solvents were purified and degassed prior to use. Reactions were monitored using (i) thin-layer chromatography techniques (0.25 mm silica gel-coated aluminium plates (60 Merck F254) using 254 nm UV light for visualisation), (ii) reverse phase chromatography on a HPLC (Shimadzu) equipped with an analytical (YMC-Triart Phenyl, 4.6 × 150 mm I.D. S-5 µm) column, eluting with acetonitrile/water, and (iii) 1 H NMR monitoring. If required, products were further purified using (i) flash-column chromatography techniques with Merck silica gel 60 under pressure or (ii) reverse phase semi-pre-

General
All reagents were used as received without further purification. Air-sensitive reagents were reacted in oven-dried glassware under nitrogen using syringe-septum cap techniques. All solvents were purified and degassed prior to use. Reactions were monitored using (i) thin-layer chromatography techniques (0.25 mm silica gel-coated aluminium plates (60 Merck F254) using 254 nm UV light for visualisation), (ii) reverse phase chromatography on a HPLC (Shimadzu) equipped with an analytical (YMC-Triart Phenyl, 4.6 × 150 mm I.D. S-5 µm) column, eluting with acetonitrile/water, and (iii) 1 H NMR monitoring. If required, products were further purified using (i) flash-column chromatography techniques with Merck silica gel 60 under pressure or (ii) reverse phase semi-preparative HPLC using a semi-preparative (YMC-Triart Phenyl, 10 × 150 mm I.D. S-5 µm, 12nm) column. Products were analysed by 1 H NMR, 13 C NMR and 19 F NMR spectra recorded at 400 MHz, 101 MHz and 376 MHz, respectively, at rt and calibrated using residual non-deuterated solvent as an internal reference. Chemical shifts are reported in parts-per-million (ppm). Further analysis by reverse phase chromatography was carried out for final products using an HPLC (Shimadzu) as described above for reaction monitoring. ESI mass spectra were acquired using Waters LCT Classic in positive and negative modes as required. MALDI-TOF spectra were acquired using the Bruker 'autoflex maX' system with the MALDI Imaging Platform. All absorbance spectra were recorded with a Varian Cary 50 scan UV-visible spectrophotometer and fluorescence spectra were recorded with a Varian Cary eclipse fluorescence spectrophotometer. Organic solvents for absorbance and fluorescence experiments were of HPLC quality and Millipore filter HPLC grade water was used. The clinical Quest Spectrum Platform Imaging system (open camera) was used for in vivo rodent studies and the pre-clinical Perkin Elmer Solaris open-camera system was used for in vivo porcine ureter imaging. Compound 1b was synthesized as per literature procedures [29].
3.6. Dose De-Escalation Study for Fluorophore 12 N = 6 Sprague-Dawley rats aged 6-8 weeks old were anaesthetized with 4% isoflurane and 0.8 L/min O 2 and maintained at 2% isoflurane and 0.8 L/min O 2 for the duration of the experiment. Animals were injected with a dose of 0.2 mg/kg in the lateral tail vein, and were imaged as before for 90 min. After the imaging session was complete, the animals recovered in a warmed recovery cage until they regained full consciousness. One week later, the same group of animals underwent the same imaging parameters but the dose of fluorophore was reduced. This de-escalation of dose will occurred 2 further times with the doses dropping to 0.1 mg/kg and 0.05 mg/kg with a week's recovery between each imaging session. After the last imaging session, rats were euthanized by anaesthetic overdose followed by cervical dislocation.

Ureter Identification in Rodent Model Using 12
N = 3 female Sprague-Dawley rats aged 6-8 weeks old were anaesthetized as above. Animals were placed on a pre-calibrated heating pad in a sterile surgical environment. The hair from the abdomen was removed using depilatory cream. The surgical site was swabbed with alcohol and povidone iodine mixture. A midline incision of 3 cm and a left side incision was performed on the abdomen to expose the bladder, the left/right ureter, kidney, liver and spleen. The animals were injected via the lateral tail vein with 0.5 mg/kg of 12, and then were imaged as before. Rats were imaged for 90 min. After the imaging session, animals were euthanized by anaesthetic overdose cervical dislocation.

Toxicity Study of 12 in Rodents
This study was independently carried out in the test facility of Citoxlab France (subsidiary of Charles River), Evreux, France. The study was performed in a test facility certified by the French National Authorities for Good Laboratory Practice compliance and followed established practices and standard operating procedures of Citoxlab France. The study was conducted in compliance with animal health regulations, in particular: Council Directive No. 2010/63/EU and French decree No. 2013-118 on the protection of animals used for scientific purposes. The Citoxlab France Ethics Committee reviewed and approved the study plan in order to assess compliance as defined in Directive 2010/63/EU and in French decree No. 2013-118.
The objective of this study was to evaluate the potential toxicity of 12, following (i) 7 days of treatment of 2 mg/kg daily dose and (ii) a single dose of 10 mg/kg by intravenous route (bolus). Nine male rats (Sprague-Dawley) sourced from Charles River Laboratories Italia, Calco, Italy were used for the study. At the beginning of the treatment period, the animals were 6 weeks old and weighed between 130 and 240 g. Upon arrival at Citoxlab France, the animals were given a clinical examination to ensure good condition and were acclimated to the study conditions for 5 days before the beginning of the treatment period. Animals were housed in a secure rodent unit with conditions set as follows: temperature: 22 ± 2 • C, relative humidity: 50 ± 20%, light/dark cycle: 12 h/12 h, ventilation: 8 to 15 cycles/hour of filtered, non-recycled air. The animals were housed in groups of 3, in polycarbonate cages with stainless steel lids containing autoclaved sawdust. Each cage contained at least two objects for environmental enrichment. All animals had free access to SSNIFF rat/mouse pelleted maintenance diet and to tap water (filtered with a 0.22 µm filter) contained in bottles.
A PBS solution of 12 (2 mg/kg) was administered by intravenous route (bolus), with a constant dosage-volume of 5 mL/kg used for the treated group (n = 3). Each animal received this dosage daily for seven consecutive days. In the same manner, a single 10 mg/kg dosage was used to treat another group (n = 3). The quantity of dose administered to each animal was adjusted according to the most recently recorded body weight. The control animal group were administered PBS by intravenous route (bolus) for seven consecutive days (n = 3). Prior to blood sampling, the animals were deprived of food for an overnight period of 14 h. Blood samples were taken from the orbital sinus of the animals under light isoflurane anesthesia, into appropriate tubes.

Porcine Imaging Study
The present study is part of the ELIOS protocol (endoscopic luminescent imaging for oncology surgery), fully approved by the local Ethical Committee on Animal Experimentation (ICOMETH No. 38.2016.01.085), and by the French Ministry of Superior Education and Research (MESR) (APAFIS#87212017013010316298-v2). Six adult female pigs (Sus scrofa domesticus, mean weight: 9.6 ± 1.3 kg) were used. Animals were managed according to the directives of the European Community Council (2010/63/EU) and ARRIVE guidelines [38]. Following an acclimatation period in our animal keeping facility, the animals received premedication by means of an intramuscular injection of Zolazepam and Tiletamine 10 mg/kg (Zoletil ND, Virbac, Carros, France). Anesthesia induction was achieved through an intravenous injection of Propofol 3 mg/kg (Propofol Lipuro ND, B Braun, Sarlat, France) together with rocuronium 0.8 mg/kg (Esmeron ND, MSD, France). After intubation, the pigs were mechanically ventilated throughout the experiment and were sedated via an inhalation of isoflurane 2-3% (Isoflurin ND, Axience, Pantin, France). Three experimental doses of 12 as 0.5, 0.25 and 0.1 mg/kg (n = 2 for each dose) were administered by intravenous route (bolus) with imaging for 120 min. During the experiment, analgesia was ensured with intramuscular buprenorphine (Buprecare ND, Axience, Pantin, France) 0.01 mg/kg. At the end of the experimental procedure, pigs were sacrificed under deep anesthesia (Isoflurane 5%) with a lethal intravenous application of Pentobarbital 40 mg/kg (Exagon ND, Axience, Pantin, France).

Conclusions
Mono-and bis-PEGylated fluorophores 7-12 were synthesized in seven steps with greater than 95% purity, with each having similar emission spectra spanning the 700-800 nm wavelength region. Fluorophore bis-PEGylation resulted in higher aqueous fluorescence quantum yields with PEG chain lengths of 2.9 to 4.6 kDa of 11 and 12 being optimal. The advantage of substituting with two PEG groups was also seen in an in vivo rodent model screen of two mono-and two bis-substituted derivatives from which fluorophore 12 was selected for more in-depth studies. An in rodent dose de-escalation study with 12, showed that its emission could be detected from urine in the bladder at doses as low as 0.05 mg/kg. Rodent toxicity studies indicated 12 as a safe surgical tool with NOAEL established as 10 mg/kg/day. It was successfully employed for fluorescence ureter identification with the preference for renal excretion notable through comparative fluorescence intensities from the ureters, kidneys and liver. Ureteral identification with 12 was successfully performed in a larger animal porcine model under abdominal surgical conditions. Three tested doses of 0.5, 0.25 and 0.1 mg/kg all successfully identified fluorescent ureters within 20 min of administration and were sustained for 120 min. The use of 3-D emission heat map imaging allowed the spatial and temporal changes in intensity due to the peristaltic waves of urine being transferred from kidneys to the bladder to be clearly distinguished from surrounding tissues. As the emission of 12 was spectrally separable from the clinically used ICG it is envisaged that their combined use could be a step towards intraoperative colour coding of different tissues.