DNA-SWCNT Biosensors Allow Real-Time Monitoring of Therapeutic Responses in Pancreatic Ductal Adenocarcinoma

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is a highly desmoplastic cancer with limited treatment options. There is an urgent need for tools that monitor therapeutic responses in real time. Drugs such as gemcitabine and irinotecan elicit their therapeutic effect in cancer cells by producing hydrogen peroxide (H₂O₂). In this study, specific DNA-wrapped single-walled carbon nanotubes (SWCNT), which precisely monitor H₂O₂, were used to determine the therapeutic response of PDAC cells in vitro and tumors in vivo. Drug therapeutic efficacy was evaluated in vitro by monitoring H₂O₂ differences in situ using reversible alteration of Raman G-bands from the nanotubes. Implantation of the DNA-SWCNT probe inside the PDAC tumor resulted in approximately 50% reduction of Raman G-band intensity when treated with gemcitabine versus the pretreated tumor; the Raman G-band intensity reversed to its pretreatment level upon treatment withdrawal. In summary, using highly specific and sensitive DNA-SWCNT nanosensors, which can determine dynamic alteration of hydrogen peroxide in tumor, can evaluate the effectiveness of chemotherapeutics. SIGNIFICANCE: A novel biosensor is used to detect intratumoral hydrogen peroxide, allowing real-time monitoring of responses to chemotherapeutic drugs.

Figure 1.

Detection of H₂O₂ using DNA-SWCNT hybrid: a) Photoluminescence spectra obtained from SWCNT (10 μg mL⁻¹) after incubation with H₂O₂ concentrations from 0 – 200μM. b) Percentage of PL quenching has been plotted with concentration of H₂O₂. c) Raman spectra collected from SWCNT after incubation of H₂O₂ concentrations from 0 – 200μM.d) Raman intensity corresponding to G-band has been plotted with concentration of H₂O₂.

Figure 2.

Detection of H₂O₂ in gemcitabine-treated cancer cell and cytotoxicity study: a) Endogenous expression of H₂O₂ measured using PL signals corresponding to carboxy-H2DCFDA and SWCNT biosensor. b) Cell viability assay of PANC1 after exposure to 10μM GEM treatment at three different time points.

Figure 3.

Monitoring spatial distribution of H₂O₂ with varying concentrations of GEM: a) PANC1 treated with SWCNT. Raman spectra collected from each point in the dotted line. The color map in the dotted line is based on intensity at 1590 cm⁻¹. One such spectrum is shown in the inset and the corresponding point in the line scan is identified by the arrow sign. Color bar displays the range of Raman signals recorded in individual image. b) Spectra collected from each point has been plotted in a 3D graph, where x-axis represents the wavenumber (cm⁻¹), y-axis represents the coordinate of the line, and z-axis represents the intensity in arbitrary unit. c) A representative picture of the line scan on cells after treatment with 5μM GEM. Color bar displays the range of Raman signals recorded in each individual image. d) A representative picture of line scan on cells after treatment with 10μM GEM. Color bar displays the range of Raman signals recorded in each individual image. e) 3D representation of intensity profile from Figures a, c, and d, demonstrating effective decrease of Raman signals with increasing dose of GEM in vitro. The data set has been presented with offset in x axis to demonstrate their intensity profile without overlapping. f) Bar graph represents alteration of maximum intensity of Raman signals with increasing concentration of GEM. **** represents p value <0.0001. g) Cell viability assay of PANC1 after exposure of 5μM and 10μM GEM treatment for 72 hours. h) PL spectra obtained from SWCNT incubation with (red line) or without (black line) GEM. Both spectra overlap showing no effective influence of GEM on PL spectra of SWCNT.

Figure 4.

Detection of H₂O₂ and their spatial distribution in irinotecan-treated cancer cell: PANC1 cells incubated with 10 μg mL⁻¹ SWCNT overnight followed by irinotecan treatment for 72 hours. a) Cell without irinotecan treatment and b) cell with 120 μM irinotecan treatment. Color bar displays the range of Raman signals recorded in individual image. c) Spectra collected from each point have been plotted in a 3D graph where x-axis represents the wavenumber (cm⁻¹), y-axis represents the coordinate of the line, and z-axis represents the intensity in arbitrary unit. d) Cell viability assay of PANC1 after incubation with 120μM irinotecan for 72 hours. e) 3D representation of intensity profile from Figures a, and b, demonstrating decrease of Raman signals with irinotecan treatment in vitro. The data set has been presented with offset in x-axis to demonstrate their intensity profile without overlapping. f) Maximum Raman intensity presented in bar graph display the effect of irinotecan treatment on H₂O₂ production. **** represents p <0.0001.

Figure 5.

Depth profile of Raman G-band intensity: The tumor was scanned in longitudinal direction to find the depth corresponding to maximum Raman signals.

Figure 6.

Longitudinal live animal imaging: Tumor site of same mice imaged a) before initiation of treatment, b) after three treatments of GEM, and c) two weeks after withdrawal of treatment. The red color map was done based on the Raman signals from SWCNT at 1590 cm⁻¹ wavenumber. d) The maximum level of Raman signals from SWCNT corresponding to these three distinct treatment conditions is displayed for all three mice. * and ** indicate p<0.05 and <0.01, respectively.