Indocyanine green was purchased from Acros Organics (Geel, Belgium). Cell culture reagents were purchased from Biological Industries (Beit Haemek, Israel). The RNeasy mini-isolation kit was from Qiagen (Valencia, CA, USA). TaqMan reverse transcription reagents and fluorescent MGB probes were from Applied Biosystems (Foster City, CA, USA). Primary antibodies for β-actin and C219 were from Biotest (Kfar Saba, Israel) and the primary antibody for OATP 1A2 was from Assay BioTech (Sunnyvale, CA, USA). Secondary horseradish peroxidase-conjugated goat anti-rabbit and goat anti-mouse antibodies were from Jackson ImmunoResearch (West Grove, PA, USA). The bicinchoninic acid (BCA) protein assay kit was from Pierce (Rockford, IL, USA). Skim milk was obtained from Difco (Franklin Lakes, NJ, USA). Pentobarbital sodium (Pental) was from CTS Kiryat Malachi, Israel. Heparin sodium was from Rotexmedica (Trittau, Germany). All the other reagents were purchased from Sigma–Aldrich (Rehovot, Israel).

The Madin–Darby canine kidney (MDCK) II cells transfected with cDNA coding for MDR1 (MDCK-MDR1) and wild-type (MDCK-CT) cells were kindly provided by Dr. Alfred Schinkel (The Netherlands Cancer Institute). The human colorectal adenocarcinoma cell line HT-29 was a generous donation of Dr. Ioav Cabantchik (The Hebrew University). MDR1 was induced in HT-29 cells (HT-29-MDR1) by incubation of control HT-29 cells (HT-29-CT) with 300 ng/mL colchicine in growth medium for a minimum of 9 weeks. All cell types were grown in Dulbecco’s modified Eagle’s phenol-free low-glucose medium (DMEM) supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 100 units/mL penicillin, and 100 μg/mL streptomycin at 37°C in a 5% CO₂ incubator, and HT-29-MDR1 cells were in addition continuously incubated with colchicine.

The experimental studies and protocols were approved by the Animal Care and Use Committee of the Hebrew University (Protocol # MD-12-13573) and the procedures followed were in accordance with institutional guidelines. Male CD1-nude-white mice (6–7 weeks old) were purchased from Harlan Laboratories (Rehovot, Israel) and housed in the specific pathogen-free facility (SPF) unit at the Ein Kerem campus of the Hebrew University. The mice, weighing 31 ± 2 g, had free access to food (a standard diet) and water and were maintained on a 12:12-h automatically timed light/dark cycle.

Probes chosen for these studies included two compounds previously reported to be MDR1 substrates, ICG (Portnoy et al., 2012) and rhodamine 800 (On et al., 2011), as well as several additional cyanine dyes: IR-775, IR-783, XenoLight DiR, and Genhance 750. Accumulation studies were conducted using MDCK-CT [which express endogenous, canine MDR1 (Feng et al., 2008)], MDCK-MDR1, HT-29-CT, and HT-29-MDR1 that were seeded separately in 96 well plates at 8 × 10⁴ cells/well and reached confluence, forming monolayers. The tested probe was dissolved in dimethyl sulfoxide (DMSO) and diluted in DMEM (maximal final DMSO concentration in the incubation medium 0.001%). One hundred microliter of the probe at one of nine concentrations (range 4 × 10^-7 M–1.0 × 10^-4 M) was added to each well of both MDCK-CT and MDCK-MDR1 plates. Following 1 h incubation with the probe, cells were washed three times with PBS and emission signal was quantified using Cytation 3 Cell Imaging Multi Mode Reader (BioTek, Winooski, VT, USA). In a separate set of experiments, control MDCK cells were incubated with IR-783 (2 × 10^-5 M) in 24 well plates and accumulation was measured as described above at the presence or the absence of verapamil (200 μM). Plates were scanned by Typhoon FLA 9500 biomolecular imager (GE Healthcare Life Sciences, Piscataway Township, NJ, USA).

Transport of IR-783 across cell monolayers was evaluated as described before (Portnoy et al., 2012). Briefly, MDCK-CT cells or MDCK-MDR1 cells were seeded at 2 × 10⁶ cells/well on microporous polycarbonate membrane filters. Cells were grown until transepithelial resistance reached 200 Ω or more (measured by Millicell-ERS, Millipore Corporation, Billerica, MA, USA), with daily replacement of medium. The experiments began with replacing the medium with fresh DMEM containing the probe (1.25 × 10^-5 M). Transport of the probes was tested for both the apical to basolateral (A to B), and the basolateral to apical (B to A) directions. Aliquots of 100 μL were taken from the receiver compartments every 30 min with replacement with fresh DMEM. Probe emission intensity was measured by the Cytation 3 Reader. The studies were performed in triplicate in a humidified incubator on two different days. Papp and the ER were calculated as described before (Portnoy et al., 2012).

HT-29-CT and HT-29-MDR1 xenografts were initiated by subcutaneous injection of 2.8 × 10⁶ cells into the flanks of the mice. HT-29-CT and HT-29-MDR1 cells were injected to the right and left flanks, respectively. Three weeks after grafting, mice were subjected to optical imaging. On each study day, 10 μL/gr ICG or IR-783 (8 mg/kg or 4 mg/kg, respectively, in 10 parts DDW:two parts filtered sucrose phosphate buffer 9.3%) was injected into the tail vein. Mice were repetitively scanned over a time period of 90 min, while body temperature was kept on a 37°C platform. This period was selected based on preliminary studies demonstrating that longer imaging periods do not increase the sensitivity to detect differences between control and MDR1 tumors and lead to greater mortality.

At the completion of the scans, mice were sacrificed under pentobarbital sodium anesthesia (200 mg/mL, 350 mg/kg) and cardiac blood samples (40 μL) were collected into heparinized 96-well plates. Subsequently, tumors were collected. Along the collection procedure, tissue and blood samples were protected from light and kept on ice. Immediately after that, they were scanned by the Typhoon FLA 9500 imager, to verify the results obtained in vivo (because the in vivo signal might potentially reflect emission from tissues underneath the tumor). The tissues were then frozen in liquid nitrogen and kept in -80°C until further analysis.

For analysis of probe uptake kinetics in vitro, we calculated the area under the emission-intensity-concentration curve (AUC) using the trapezoidal rule. Larger AUC implies lower uptake or greater efflux of the probe. Due to non-linearity of the emission intensity with regard to substrate concentrations, E_max and EC₅₀ values are not reported.

For analysis of the in vivo dye kinetics, regions of interest (ROIs) were drawn over the tumors (0.25 ± 0.1 cm²) using Living Image 4.3.1 (PerkinElmer, Waltham, MA, USA). A constant size ROI (0.1 cm²) was drawn near every tumor and used as a reference background region as recommended by the manufacturer of the in vivo imaging system (Caliper LifeSciences, 2012). The emission intensity was expressed in radiant efficiency units ([photons/second/steradian]/microwatt; [p/s/sr]/μW). Areas under the concentration-time curve (AUC) were calculated using Phoenix WinNonlin 6.3. The ex vivo emission intensity was analyzed using ImageJ 1.47V (National Institute of Health, Bethesda, MD, USA).

Analysis of mRNA levels was conducted as described before (Rubinchik-Stern et al., 2015). Briefly, total RNA was isolated using RNeasy mini-isolation kit. RNA integrity and purity were verified by ND-1000 spectrophotometer (NanoDrop Technologies, Inc). cDNA was synthesized from 2 μg of total RNA that had A260/A280 and A230/A260 ratios of 1.8–2.0 and 2.0–2.2, respectively. Reverse transcription was performed using TaqMan reverse transcription reagents. The reactions were run as follows: 25°C for 10 min, followed by 37°C for 120 min, then 85°C for 5 min. The real-time PCR assay was carried out with the use of gene specific FAM-labeled fluorescent MGB probes in StepOnePlus real-time PCR system (Applied Biosystems) on a fast mode. Samples were run in triplicate. The reaction final volume was 10 μL for each sample. The relative mRNA levels in each sample were normalized to the housekeeping gene β-actin. Changes in mRNA expression of target genes from the control cells were expressed relative to that of the vehicle control group.

Several tumors were large enough to allow measurement of both the mRNA and the protein levels of MDR1. From these tumors, whole cell lysates were prepared as previously described (Rubinchik-Stern et al., 2015), by tissue homogenization in cold radioimmunoprecipitation assay (RIPA) buffer. Lysates were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Stacking and separating gels were made of 5 and 10% acrylamide, respectively. Each lane was loaded with 20 μg protein samples of whole cell lysates. Following separation, the proteins from the gels were transferred to nitrocellulose membranes using a Mini Trans-Blot Cell (Bio-Rad Laboratories, Inc.). Membranes were blocked in Tris-buffered saline containing 0.1% Tween 20 (TBS-T) and 5% skim milk powder and probed overnight at 4°C with primary antibodies against β-actin (1:1,000), P-gp (recognizes both the human and the murine MDR1; 1:600), and OATP1A2 (SLCO1A2; 1:500). Antibodies were diluted in 5% w/v BSA in 0.1% Tris-Buffered Saline with Tween-20 and 0.02% sodium azide. The blots were then incubated for 1 h with horseradish peroxidase-conjugated secondary antibodies (goat anti-rabbit) at 1:10,000 dilutions and developed by enhanced chemiluminescence. β-Actin was used as the internal control.

Results are reported as mean ± SD, unless otherwise indicated. Values of MDR1 and control tumors were compared using the Wilcoxon signed rank test (InStat; GraphPad, La Jolla, CA, USA). A p value ≤ 0.05 was considered significant.

MDR1 overexpression in HT-29-MDR1 cells was confirmed by western blot analysis. This analysis further demonstrated that the expression of a representative uptake transporter, OATP1A2 is not significantly altered in these cells (Figure 1). Among the tested compounds, XenoLight and GenHance accumulated in MDCK-CT cells only scarcely, whereas rhodamine 800 and IR-775 demonstrated a quenching phenomenon at high concentrations (Mann et al., 2015). ICG and IR-783 accumulation increased with the dye concentration in all cell types. Both ICG and IR-783 appeared to accumulate to a lesser extent in MDR1-overexpressing MDCK and HT-29 cells than in their respective controls [Table 1; Figures 2A–D; see also (Portnoy et al., 2012)]. At 1.25 × 10^-5 M, the emission intensity of both ICG and IR-783 in HT-29-CT cells was 1.5-fold (P < 0.01) greater compared to HT-29-MDR1 cells (Figure 2E). The ERs of IR-783, calculated from the permeability assays, were 1.2 and 2.0 in MDCK-CT and MDCK-MDR1 cells, respectively, resulting in a net ER of 1.7 (Figure 2F). At the presence of verapamil, the emission intensity of control MDCK cells incubated with IR-783 was 2.9-fold higher than in cells incubated with the vehicle (Figures 2G,H). The ER of ICG and the effect of verapamil on its accumulation have been published before (Portnoy et al., 2012). Because ICG and IR-783 have shown the best imaging properties among the probes studied in vitro, they were selected for farther evaluation.

In vivo, MDR1 tumors did not significantly differ in size from control tumors, although variability in size across animals was observed for both tumor types (Table 2). Optical imaging of mice treated with ICG demonstrated tumor-specific signal accumulation (Figure 3). Elevated emission intensity in tumors located on the right flanks of the mice indicated higher ICG concentrations in HT-29-CT tumors, compared to their opposing HT-29-MDR1 tumors (Figures 3A,B). The AUC of emission intensity calculated individually for each mouse was greater in CT compared to MDR1 tumors, with individual CT/MDR1 AUC ratios of 2.2, 2.0, 1.8, 1.5, and 1.3 (mean 1.7; Figure 3C). The difference in emission intensity between the two tumor types were not significant in vivo or ex vivo (p > 0.05; Figures 3D,E).

IR-783 accumulated mostly in the largest control tumors (0.57 cm², 0.65 cm²). In mice which bared these tumors the probe accumulation appeared to be greater in control versus MDR1 tumors (Figure 4). However, in mice baring smaller control tumors (0.08 cm², 0.22 cm², 0.16 cm²), IR-783 emission intensity was greater in MDR1 than in control tumors (Figure 4C). Hence, the mean IR-783 did not differ between the tumor types (Figures 4D,E). No relationship was observed between MDR1 tumor size and emission intensity (data not shown).

RT-PCR analysis of xenograft extracts demonstrated significantly elevated MDR1 mRNA levels in HT-29-MDR1 tumors, indicating that these cells maintained MDR1 overexpression in vivo (Figure 5A). Interestingly, the expression of the murine Mdr1a that might contribute to ICG and IR-783 accumulation differed between control and the MDR1 xenografts (Figure 5B). However, the magnitude of difference was well below that observed for MDR1. Western blotting of proteins from tumors that were large enough to provide material for both mRNA and protein expression analyses confirmed MDR1 overexpression at the protein level (∼30-fold greater than in control tumors; Figure 5C). No correlation was found between MDR1 expression (at either the mRNA or the protein level) and ICG emission intensity (data not shown).