Search

Fig. S7. The effect of HSP2 on oxygenation of subcutaneous NSCLC A549-luc tumor xenografts after 3 weeks of treatment (same tumors as shown in Fig. S1). (A) Representative tumor images of spatial distribution of oxygen saturation [sO2 (O2)] after the oxygen challenge and the amplitude of signal change (Î"sO2) for each treatment group from breathing air to breathing 100% oxygen. The change in sO2 (Î"sO2) is calculated by subtracting the average tumor sO2 while breathing air from the average tumor sO2 while breathing oxygen. The sizes of the shown tumors were: Control: 111 mm3; HSP2: 51.9 mm3. (B & C) Quantification of sO2 (O2) and Î"sO2 for each treatment group. For statistical analysis, the levels in treated tumors were compared to the levels in control tumors with a Welch 2-sample t-test. ***, P < 0.001. Box is between 25th and 75th percentile, line at median.

Fig. S6 Analysis of correlation between tumor sizes and MSOT metrics. (A, B, & C) sO2 (Air), sO2 (O2), and Î"sO2 as derived from MSOT data. No significant correlation was detected between any MSOT metric with tumor sizes measured ex vivo. (D) Summary of p-values and r correlations of MSOT and OE-MSOT metrics with tumor sizes. For statistical analysis, the levels in treated tumors were compared to the levels in control tumors with a Welch 2-sample t-test or Pearson r correlation.

Fig. S1. HSP2 effectively suppresses the growth, oxygen consumption rates, and ATP generation in subcutaneous A549-luc NSCLC tumor xenografts. (A) Sizes of H1299 tumors (mm3) from mice treated with HSP2 for 1.5 weeks (mid-point) or treated with HSP2/CycT for 3 weeks (end-point). (B) Photos showing excised tumors from NOD SCID mice treated with saline (control) or HSP2 (25 mg/kg I.V every 3 days) for 3 weeks (n=4/group). (B) Average weights for control and treated tumors shown in A. (C) Average body weights of control and treated mice bearing tumors shown in A. (D) Oxygen consumption rates (OCRs) measured in cells from control and treated tumors. (E) ATP levels measured in cells from control and treated tumors. Data are plotted as mean {plus minus} SEM. For statistical analysis, the levels in treated tumors were compared to the levels in control tumors with a Welch 2-sample t-test. *, P < 0.05.

Fig. S8. HSP2 and CycT do not affect iron metabolism and hypoxia induction of HIF1�. (A) Iron stain did not detect any difference between control and treated tumor tissues. Shown are montages, 10X, and 40X images of control, HSP2 and CycT-treated tumor tissue sections stained with Iron Stain. The blue rectangles in montages, 10X images the regions shown in 10X and 40X images, respectively. Scale bar: montage, 1 mm for controls and 500 µm for HSP2 and CycT treated tissues; 10X, 200 µm, 40X 50 µm. (B) Representative IHC images of H1299 NSCLC tumor tissue sections and graphs showing the levels of transferrin receptor TFRC in control, HSP2, and CycT-treated tumors. (C) Representative IHC images of H1299 tumor tissue sections and graphs showing the levels of ferroportin 1, SLC40A1, in control, HSP2, and CycT-treated tumors. Shown are montages and 10X images of control, HSP2 and CycT-treated tumor tissue sections stained with DAPI or antibodies against the indicated protein. The white rectangles in DAPI images denote the regions shown in 10X images. Scale bar: montage, 1 mm (for control tumor sections) and 500 µm (for HSP2 and CycT treated tumor sections); 10X, 20 µm. Protein levels were quantified, and data are plotted as mean {plus minus} SEM. The values shown in the graphs are averages of signals quantified from three independent IHC experiments. For statistical analysis, the levels in treated tumors were compared to the levels in control tumors with a Welch 2-sample t-test. P *

Fig. S3. Treatment with HSP2 or CycT does not affect mitochondrial structural proteins in tumor tissues. (A) Representative IHC images of H1299 NSCLC tumor tissue sections and graphs showing the levels of the mitochondria marker TOMM20 in control, HSP2, and CycT-treated tumors. (B) Representative IHC images of H1299 tumor tissue sections and graphs showing the levels of the mitochondria marker VDAC1 in control, HSP2, and CycT-treated tumors. Data are plotted as mean {plus minus} SD. For statistical analysis, the levels in treated mice were compared to the levels in untreated mice with a Welch 2-sample t-test. No statistical significance was detected in any group.

Fig. S2. HSP2 and CycT did not significantly affect blood and liver functions. (A) Average red blood cell count in mice bearing subcutaneous lung tumor xenografts treated with or without HSP2 or CycT. (B) Average hemoglobin levels in mice treated with or without HSP2 or CycT. (C) Average serum ALT (alanine transaminase) levels treated with or without HSP2 or CycT. (D) ATP levels measured in cells from livers excised from mice treated with HSP2, CycT, or control saline. Data are plotted as mean {plus minus} SD. For statistical analysis, the levels in treated mice were compared to the levels in untreated mice with a Welch 2-sample t-test.

Fig. S4. MSOT images from control and HSP2- or CycT-treated tumors of similar size. The tumors were from the same experiment shown in Fig. 2. The quantified values for all parameters shown in Fig. 2 were not changed when tumor sizes were matched. (A) The spatial distribution of baseline deoxyhemoglobin (Hb; blue), oxyhemoglobin (HbO2; red), and total hemoglobin (THb). Maps were constructed using back-projection and spectral-unmixing overlaid on the grayscale image of the corresponding animals in each treatment group (control, HSP2 and CycT). (B) Representative tumor images of spatial distribution of baseline oxygen saturation in air. (C) Representative tumor images of spatial distribution of oxygen saturation [sO2 (O2)] after the oxygen challenge and the amplitude of signal change (Î"sO2) for each treatment group from breathing air to breathing 100% oxygen. Tumor sizes are listed in Materials and Methods. Specifically, the sizes of the three shown tumors were: Control: 176 mm3, HSP2: 158 mm3, CycT: 138 mm3.

Fig. S5. (A & B) The effect of HSP2 on subcutaneous NSCLC H1299-luc tumor xenografts after 1.5 weeks of treatment (mid-point for treatment shown in Fig. 2). (A) Photos of excised tumors from mice treated with saline (control) or HSP2. (B) Average masses for control and treated tumors shown in A. (C-E) Comparison of sO2 and Î"sO2 levels in mid-point and end-point control and HSP2-treated tumors. (C) Representative tumor images of spatial distribution of oxygen saturation [sO2 (O2)] after the oxygen challenge in mid-point and end-point control and HSP2-treated tumors. (D) Quantification of sO2 levels in in mid-point and end-point control and HSP2-treated tumors. (E) Representative tumor images of the amplitude of signal change (Î"sO2) from breathing air to breathing 100% oxygen in mid-point and end-point control and HSP2-treated tumors. The change in sO2 (Î"sO2) is calculated by subtracting the average tumor sO2 while breathing air from the average tumor sO2 while breathing oxygen. (F) Quantification of Î"sO2 levels in mid-point and end-point control and HSP2-treated tumors. Tumor sizes are listed in Materials and Methods. Specifically the sizes of the four shown tumors were: Control: mid-point: 54.4 mm3, end-point: 369 mm3; HSP2: mid-point: 35.7 mm3, end-point 88.4 mm3. ***, P < 0.001; *, P < 0.05. The P values for the comparison of end-point HSP2-treated tumors and mid-point control tumors are listed under the plots in D and F.

Multispectral optoacoustic tomography (MSOT) is an emerging noninvasive imaging modality that can detect real-time dynamic information about the tumor microenvironment in humans and animals. Oxygen enhanced (OE)-MSOT can monitor tumor vasculature and oxygenation during disease development or therapy. Here, we used MSOT and OE-MSOT to examine in mice the response of human non–small cell lung cancer (NSCLC) xenografts to a new class of antitumor drugs, heme-targeting agents heme-sequestering peptide 2 (HSP2) and cyclopamine tartrate (CycT). HSP2 inhibits heme uptake, while CycT inhibits heme synthesis in NSCLC cells, where heme is essential for ATP generation via oxidative phosphorylation. HSP2 and CycT can inhibit ATP generation and thereby suppress NSCLC cell tumorigenic functions. MSOT showed that treatment of NSCLC tumors with HSP2 or CycT reduced total hemoglobin, increased oxygen saturation, and enhanced the amplitude of response to oxygen gas breathing challenge. HSP2 and CycT normalized tumor vasculature and improved tumor oxygenation, where levels of several hypoxia markers in NSCLC tumors were reduced by treatment with HSP2 or CycT. Furthermore, treatment with HSP2 or CycT reduced levels of angiogenic factor VEGFA, its receptor VEGFR1, and vascular marker CD34. Together, our data show that heme-targeting drugs HSP2 and CycT elicit multiple tumor-suppressing functions, such as inhibiting angiogenic function, normalizing tumor vasculature, alleviating tumor hypoxia, and inhibiting oxygen consumption and ATP generation.Significance:Heme-targeting agents HSP2 and CycT effectively normalize tumor vasculature and alleviate tumor hypoxia, raising the possibility of their combination with chemo-, radio-, and immunotherapies to improve antitumor efficacy.See related commentary by Tomaszewski, p. 3461