Targeting Acidity in Pancreatic Adenocarcinoma: Multispectral Optoacoustic Tomography Detects pH-Low Insertion Peptide Probes In Vivo

Abstract

Background: pH-low insertion peptides (pHLIP) can serve as a targeting moiety that enables pH-sensitive probes to detect solid tumors. Using these probes in conjunction with multispectral optoacoustic tomography (MSOT) is a promising approach to improve imaging for pancreatic cancer.

Methods: A pH-sensitive pHLIP (V7) was conjugated to 750 NIR fluorescent dye and evaluated as a targeted probe for pancreatic adenocarcinoma. The pH-insensitive K7 pHLIP served as an untargeted control. Probe binding was assessed in vitro at pH 7.4, 6.8, and 6.6 using human pancreatic cell lines S2VP10 and S2013. Using MSOT, semiquantitative probe accumulation was then assessed in vivo with a murine orthotopic pancreatic adenocarcinoma model.

Results: In vitro, the V7-750 probe demonstrated significantly higher fluorescence at pH 6.6 compared with pH 7.4 (S2VP10, P = 0.0119; S2013, P = 0.0160), whereas no difference was observed with the K7-750 control (S2VP10, P = 0.8783; S2013, P = 0.921). In the in vivo S2VP10 model, V7-750 probe resulted in 782.5 MSOT a.u. signal compared with 5.3 MSOT a.u. in K7-750 control in tumor (P = 0.0001). Similarly, V7-750 probe signal was 578.3 MSOT a.u. in the S2013 model compared with K7-750 signal at 5.1 MSOT a.u. (P = 0.0005). There was minimal off-target accumulation of the V7-750 probe within the liver or kidney, and probe distribution was confirmed with ex vivo imaging.

Conclusions: Compared with pH-insensitive controls, V7-750 pH-sensitive probe specifically targets pancreatic adenocarcinoma and has minimal off-target accumulation. The noninvasive detection of pH-targeted probes by means of MSOT represents a promising modality to improve the detection and monitoring of pancreatic cancer.

Figure 1

Schematic of probe insertion into cell membrane.

Figure 2

In vitro testing of probe binding to pancreatic cancer cells at different levels of media acidity. (A) S2VP10 cells treated with either V7–750 (top row) or K7–750 (bottom row) at different levels of media pH. The experiment was repeated for S2013 cells (B). Dosimetry quantifies the signal intensity for each cell line at different pH values (C).

Figure 3

Axial tomographic slices of the S2VP10 mouse demonstrate localization of V7–750 probe signal to the region of the tumor bed, while similar signal is not observed with K7–750 probe. The gray scale image represents a single wavelength, 850 nm, and serves as a background image. (A) Single slices from 47 mm– 50 mm of mice demonstrate distribution of V7–750 probe signal throughout the tumor bed at 4 hours, while minimal accumulation is observed with K7–750. (B) Orthogonal images demonstrate 3D accumulation of V7–750 and K7–750 within the mouse in the xyz-plane.

Figure 4

Comparison of MSOT signal intensity of probe accumulation among organs. A region of interest method was utilized using an elliptical ROI of 3.5 mm² in a region of pancreas tumor, liver, and kidney was utilized to determine the mean signal intensity for each ROI in each slice. (A) The highest ROI mean signal intensity for each organ and each mouse was averaged among 3 mice per cell line injected with V7–750 probe. The V7–750 probe resulted in 782.5 MSOT a.u. in S2VP10 mice and 578.3 MSOT a.u. in S2O13 mice at 4 h, but was greatly reduced to 73.5 and 34.2 MSOT a.u. at 24 h. (B) The highest ROI mean signal intensity for each organ and each mouse was averaged among 3 mice per cell line injected with K7–750 probe. The K7–750 probe accumulated within the kidney and only resulted in very limited pancreas tumor accumulation in either S2VP10 (5.3 MSOT a.u.) or S2013 (5.1 MSOT a.u.) models.

Figure 5

Analysis of the regions of interest of the pancreas show V7–750 accumulated with a maximum peak at 49.2 mm in S2VP10 and 48.4 mm in S2O13 model. The area utilized for region of interest measurements remained constant for each slice. Image quantification for each organ is shown in Figure 4.

Figure 6

Traditional planar fluorescent imaging of probe signal immediately following injection, at 4 hours post-injection, and at 24 hours post-injection. (A) Although both probes are distributed systemically with similar signal intensity, the V7–750 pHLIP localizes to the pancreas at the 4 hour time point with diminishing signal at 24h. (B) Ex vivo identification of V7–750 and K7–750 in liver, kidney, and pancreas tumor. Images were quantified in Supplemental Figure S6.