Cell Culture

The murine pancreatic ductal adenocarcinoma (PDAC) tumor (KPC) cell clone type 2838c3 was derived from KPCY mice [21]. We kindly received these cells from Dr. Stanger at the University of Pennsylvania. PDAC (2838c3) cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Gibco) supplemented with 10% FBS and 1% penicillin-streptomycin (PenStrep) at 37°C with 5% CO2. The cells were regularly tested for mycoplasma contamination.

Mouse Model

Six- to eight-week-old female BALB/c nude mice (BALB/cOlaHsd-Foxn1nu, Envigo, The Netherlands) were subcutaneously (s.c.) bilaterally injected, with 30μl of 3x105 tumor cells in PBS (PDAC clone type 2838c3) under general anesthesia (2.5% isoflurane/O2). Tumor growth was measured with calipers, and the three orthogonal diameters of the tumors were measured. The tumor volume was estimated using the formula Vt=D1*D2*D3*PI/6.

Reagents Used

Bremachlorin (Radapharma International B.V., The Netherlands) was protected from light when handled, and the reagent was divided into aliquots, which were stored at 4°C. IRDye®-800CW Carboxylate (LI-COR Biosciences, Lincoln, NE) is a commercially available fluorescent near-infrared dye. Briefly, 2 nmol IRDye®-800CW Carboxylate aliquots were prepared in PBS.

In Vivo Bremachlorin Uptake

Mice were randomly assigned to one of the three treatment groups: 6 hours Bremachlorin; repeated-imaging 24 hours Bremachlorin; and control when tumors reached a volume of 60-80 mm3. This volume was reached 13 days after inoculation of the tumor cells (Supplementary Fig. 3a). The mean tumor treatment volumes were not significantly different between the groups (Supplementary Fig. 3c). Bremachlorin was administered at 20 mg/kg, 3.5 mg/ml, via the i.v. route via the tail vein to the Bremachlorin receiving groups (n=8) [3, 5,6,7,8]. 6 hours Bremachlorin group (n=4) were housed in subdued light conditions for 6 hours. Repeated-imaging 24 hours Bremachlorin group were kept in subdued lighting for 24 hours but were imaged at four different time points: 3, 4.5, 6 and 24 hours. Control mice (n=4) received nothing and were housed in subdued lighting conditions for 6 hours before imaging. The animals were imaged on an IVIS Spectrum imager (Revvity, Waltham, MA) under general anesthesia (2.5% isoflurane/O2). The animals were imaged dorsally and on both the right and left sides, using an FOV C, medium binning, excitation at 605 nm and emission 660 nm with 5 seconds of exposure. Signal quantification was performed by drawing regions of interest (ROIs) around the tumor. The general background signal was corrected by subtracting the signal from the same sized ROI placed on the animal’s back and used for each repeated time point using Living Image® 4.7.3 software (Revvity). For the tumor-to-background ratio (TBR), the background skin signal was measured using the same sized ROI (of the tumor) placed next to the animal’s tumor and used for each repeated time point. The values represent the average radiant efficiency [p/s/cm2/sr]/[μW/cm2].

After imaging, the animals were sacrificed by cervical dislocation. The tumors and specific organs, liver, spleen, stomach, small intestine, kidney, muscle, skin of the tumor and pancreas, were removed, weighed and imaged on an IVIS Spectrum imager (Revvity) with the same settings as explained above. The tumors were cut in half, one part for histology analysis and the other part for fluorescence bio-distribution measurements. The samples were then flash frozen with liquid nitrogen and stored at -80°C overnight before homogenization.

Ex Vivo Bremachlorin Bio-distribution

For fluorescence quantification, the specific organs and tumors were disrupted using a TissueLyser II system (Qiagen, Venlo, The Netherlands), using precooled Eppendorf holders, 5-mm stainless steel beads, and 1x RIPA buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% SDS). A dilution series of Bremachlorin in RIPA buffer was performed to determine the linear range of the photosensitizer. 150 μl of homogenate solution was loaded into a microplate, 96 well, μclear, Schwarz high binding (cat. 665097 Greiner plates) and measured on a plate reader (Spectramax, Molecular Devices, USA) with excitation 605 nm and emission 660 nm settings.

Ex Vivo Bremachlorin Histology Analysis

Tumors from the 6 hour Bremachlorin group (n=4) and the control group (n=4) were subjected to fluorescence microscopy and histological analysis. The 6 hour time point was chosen because previous studies used this time point for PDT [5, 8]. Freshly frozen tumor tissue was cut into sections of 20 μm in width on starfrost glass slides, using a cryostat and stored at -80°C. The samples were shielded from light as much as possible. For Bremachlorin fluorescence confocal imaging, the sections were thawed and imaged using a Leica SP5 confocal microscope with 405 nm excitation, a band pass filter of 650-700 nm detection and 10x and 20x objectives. Immediately after, the sections were subjected to immunofluorescence analysis. The sections were stained with an anti-CD31 antibody and Alexa-Fluor 488 (BioLegend cat# 102514, 1:200) overnight at 4°C in a humidified chamber. The next day, the slides were carefully rinsed in PBS, incubated for 10 minutes with Hoechst 3342 (1:1000 PBS) and covered with glycerol (diluted in PBS, 1:3) and a glass cover slip and sealed with transparent nail polish. The slides were then stored at 4°C and imaged within 4 days of mounting. The slides were imaged with the same confocal microscope (Leica SP5) with a bright-field and 405 nm excitation and with a band pass of 430-470 nm detection for Hoechst and 488 nm excitation and with a band pass of 520-550 nm for CD31 detection. Images were analyzed for CD31 and Bremachlorin co-localization using the Coloc 2 plug-in with ImageJ-Fiji. This generated a Spearman rank correlation coefficient from the images derived from two tissue samples from each group: the control and 6 hour Bremachlorin groups. Two or three images were taken per sample. The coverslips from the same slides were carefully removed, and the sections were stained with hematoxylin and eosin (H&E) using a standard protocol and imaged on a NanoZoomer (Hamamatsu, Japan).

In Vivo Photodynamic Therapy (PDT)

Mice were randomly assigned to one of two treatment groups: the Bremachlorin-PDT group or the control group, when the tumor volume reached 60-80mm3. This volume was reached on average between days 14-16 after inoculation of the tumor cells (Supplementary Fig. 3b). The mean tumor treatment volumes were not significantly different between the groups (Supplementary Fig. 3d). Bremachlorin was administered at 20 mg/kg, 3.5 mg/ml, via the i.v. route via the tail vein to the PDT receiving group (n=5), and the mice were housed in subdued light conditions for 6 hours. Preventative pain relief agent (buprenorphine 0.1 mg/kg, 6 μl/gram) was administered s.c. both pre- and post-PDT. The illumination was performed under general anesthesia (2.5% isoflurane/O2). Mice were placed on a temperature-controlled stage, and the tumor was positioned loosely to the side of the mouse to ensure that the illumination beam contained only the tumor and the skin, the rest of the mouse was shielded with black paper. The tumors were then trans-dermally illuminated with a 662 nm laser (Millennia Pro solid-state CW 532 pump laser, Spectra Physics, Utrecht, The Netherlands) combined with a Matisse dye laser with DCM Dye (Syrah, Grevenbroich, Germany) and a frontal light diffuser (Medlight SA, Ecublens, Switzerland) at 116 mW/cm2 to a fluence of 116 J/cm2 (described previously [5]). Mice in the control group (n=5) received no drug or pain relief and were housed in subdued light conditions for 6 hours.

In Vivo 800CW Carboxylate Uptake

IRDye®-800CW Carboxylate (100ul of 2 nmol in PBS, LI-COR Biosciences) was i.v. administered via the tail vein 24 hours after PDT or control treatment. Imaging was performed under general anesthesia (2.5% isoflurane/O2) 24 hours after receiving 800CW Carboxylate administration using an IVIS Spectrum imager (Revvity). The mice were imaged dorsally and on both the right and left sides, using an FOV C, medium binning, excitation at 710 nm and emission 820 nm with 10 seconds of exposure. Signal quantification was performed by drawing regions of interest (ROIs) around the tumor. The background signal was corrected by subtracting the signal from the same-sized ROI placed next to the tumor on the mouse using Living Image® 4.7.3 software (Revvity). The values represent the average radiant efficiency [p/s/cm2/sr]/[μW/cm2].

Schematic of PDT Treatment and 800CW Carboxylate Administration and Imaging Ex Vivo Histology Analysis of 800CW Carboxylate

After 800CW Carboxylate imaging, the animals were sacrificed by cervical dislocation. The tumors were removed, flash frozen with liquid nitrogen and stored at -80°C. The frozen tumor tissues were cut at the cryostat into 10 μm thick sections, mounted on glass slides and stored at -80°C until use. For 800CW Carboxylate fluorescent imaging, the slides were imaged on an Odyssey® M Imaging System at 5 μm resolution with the 800 nm fluorescent channel (ex 785 nm/em 820 nm) (LI-COR Biosciences). Consequent sections were stained with hematoxylin and eosin (H&E) using a standard protocol and imaged on an NanoZoomer (Hamamatsu). The H&E stained sections were assessed blindly by a pathologist to annotate the tumor tissue and necrotic tissue. The percentage of the necrotic area was calculated by measuring the ratio of the necrotic area or fluorescent area to the tumor area and multiplying by 100. The H&E images were analyzed using NDP view 2 software (Hamamatsu), and the fluorescence images were analyzed using Empiria Studio software (LI-COR Biosciences).

Animal Protocol

The animal protocol was approved by the Institutional Animal Care and Bioethics Committee of the Erasmus MC, Rotterdam, The Netherlands (approval number AVD1010020209868 and work package (WP) number SP2200243). The experiments were conducted in accordance with the national guidelines (National CCD license 209868) and regulations established by the Dutch Experiments on Animals Act (WoD) and by the European Directive (2010/63/EU) on the Protection of Animals used for scientific purposes.

Data Analysis

All the statistical analyses were performed using GraphPad Prism 9.0 software (GraphPad Software, San Diego, CA, USA). For both the Bremachlorin and 800CW Carboxylate uptake in vivo fluorescence and for the tumor-to-background ratio, one-way ANOVA with Tukey’s multiple comparisons test was used. For the ex vivo bio-distribution two-way ANOVA with Bonferroni’s multiple comparison test was used. Differences were considered significant when * p<0.05, ** p<0.01, *** p<0.005, and **** p<0.0001.