68
3
Fig S1 Synthetic scheme of 68 Ga-NOTA-G
3
-NGR2.

Fig S2 Spectrum of ESI-MS analysis of NOTA-G
3
-NGR2, m/z C
73
H
111
N
29
O
28
S
5
[M+2H] 2+ calcd,
1002.58; found, 1002.75.
The RCP of 68 Ga-NOTA-G
3
-RGD2 and 68 Ga-NOTA-G
3
-NGR2 was determined using a Miniscan TLC scanner (Bioscan, Washington, DC, USA) with Polygram SIL G (Macherey-Nagel GmbH, Germany) plates. The developing solvent was sodium citrate (0.1 M, pH = 5); radiolabeled peptide remained in the origin, whereas free 68 Ga moved with the solvent front (R f
= 0.8
 1.0).

Fig S32 Radiochemical purity was measured by Polygram SIL G plates. The developing solvent was sodium citrate (0.1M, pH=5), radiolabeled peptide remain at the origin, whereas free 68 Ga move with the solvent front (R f
= 0.8-1.0).
Cell extracts were obtained by lysing cells in cell lysis buffer (Beyotime, China) supplemented with protease inhibitors (Roche, Canada).Then the extracts was centrifugated at 12,000 rpm for 10 min at 4
°C, protein concentration of supernatant was determined with the Bradford Protein Assay Kit

(Beyotime, China). Cell extracts (40 μg of protein) were loaded on SDS-PAGE gels and transferred to a PVDF membrane (Life Technologies, Grand Island, NY, USA). Membranes were blocked with 5% nonfat dry milk and incubated with anti-CD13 monoclonal antibody (1:100, ABCAM) / anti-integrin α v monoclonal antibody (1:300, ABCAM) overnight and peroxidase-conjugated secondary antibody
(1:400, Life Technologies, Grand Island, NY, USA) for 2 h according to the manufacturer's instructions, developed using the ECL detection system (Thermo, USA) with ChemiDOC XRS+
(Bio-Rad, Hercules, CA, USA). Beta-actin was detected with anti-β-actin as an internal loading control.
Fig S43 Representative western blot analyses for CD13 receptor/integrin α v
expression of HT-1080 and
HT-29 cells. Beta-actin was used as a loading control. The data demonstrated that CD13 receptors and integrin α v
are overexpressed in HT-1080 cells, but not in HT-29 cells.
The cellular uptake and efflux studies were performed as described previously with some modifications(Chen et al. 2013). Briefly, HT-1080 cells were seeded into a 48-well plate at a density of
2.5 × 10 5 cells per well, and grown overnight. The tumor cells were then incubated with 68 Ga-labeled

tracers (~ 18 kBq/well) at 37 °C for 15, 30, 60, and 120 min. At the end of the incubation period, tumor cells were washed three times with chilled PBS, and harvested by trypsinization with 0.25% trypsin/0.02% EDTA (Beyotime, China). The cell suspensions were collected and measured in a gamma counter (Beijing PET CO., Ltd., China). The cellular uptake of labeled tracers was expressed as the percentage of total input dose after decay correction. Experiments were performed twice with triplicate wells.
For efflux studies, tumor cells were incubated with 68 Ga-labeled tracers (~ 18 kBq/well) for 1 h at
37 °C to allow internalization. The cells were then washed twice with PBS, and incubated in culture medium for 15, 30, 60, and 120 min. After washing three times with PBS, cells were harvested by trypsinization with 0.25% trypsin/0.02% EDTA (Beyotime, China). Cell suspensions were collected and measured in a gamma-counter. Experiments were performed twice in triplicate. Efflux data were presented as the percentage of the added dose after decay correction.
All 68 Ga-labeled tracers bound to HT-1080 cells. The uptake of 68 Ga-labeled tracers was maximal after
2 h incubation. For 68 Ga-NOTA-G
3
-RGD2 1.19 ± 0.21% of probe uptake occurred during the first hour, and 1.66 ± 0.14% at the end of the 2 h incubation. The cell uptake of 68 Ga-NOTA-G
3
-NGR2 was 1.25
± 0.13% during the first hour, and 1.58 ± 0.22% after 2 h incubation (Fig S4). HT-1080 cells had good retention of both tracers. For 68 Ga- NOTA-G
3
-RGD2, ~ 0.64% (from 1.32 to 0.68% of total input radioactivity) of the efflux was measured, compared with ~ 0.63% (from 1.28 to 0.65% of total input radioactivity) for 68 Ga-NOTA-G
3
-NGR2 at the end of the incubation (Fig S4). There was no significant difference in the cellular uptake and efflux of 68 Ga-NOTA-G
3
-NGR2 and 68 Ga-NOTA-G
3
-RGD2 in
HT-1080 cells at all the time points measured ( P > 0.05, n = 6).

Fig S54 Cellular uptake and efflux assays. a. Time-dependent cellular uptake of 68 Ga-NOTA-G
3
-RGD2 and 68 Ga-NOTA-G
3
-NGR2 ( n = 6, mean ± SD) into HT-1080 cells. b. Time-dependent cellular efflux of 68 Ga-NOTA-G
3
-RGD2 and 68 Ga-NOTA-G
3
-NGR2 ( n = 6, mean ± SD) in HT-1080 cells.
v
After imaging, mice were killed and tumors were cut and fixed in 10% formalin. Routine hematoxylin and eosin (H&E) and indirect immunohistochemistry of CD13 receptor/integrin α v
of HT-1080 tumor were carried out as described previously. The working dilution of anti-CD13 antibody and anti-integrin
α v
monoclonal antibody was all 1:100. Afterwards, slides were observed under an Olympus microscope and the images were captured. Brown-colored staining in the cell membrane was considered as a positive result of CD13/integrin α v
expression.

Fig S65 Hematoxylin and eosin (H&E) (a) and indirect immunohistochemistry of HT-1080 xenograft tumors expressed with CD13 receptor (b) and integrin α v
(c), brown color indicated CD13/integrin α v expression. Pictures were taken at 200 × magnification.
To follow tumor growth, 18 F-FDG microPET images were acquired before 68 Ga-NOTA-G
3
-RGD2 and 68 Ga-NOTA-G
3
-NGR2 peptide scanning. The mice were fasted for 6 h and then injected with 3.7
MBq 18 F-FDG; 10 min static scans and helical CT images were acquired at 1 h pi.
Fig S76 Decay-corrected axial, coronal and sagittal image slices of mice bearing HT-1080 tumor after intravenous administration of 18 F-FDG 1 h pi. Tumors are indicated using circles.