Ab7279

Human and mouse liver samples were fixed in 4% paraformaldehyde for 24 h, embedded in paraffin, and cut into 4-µm-thick sections. For immunohistochemistry, horseradish peroxidase (HRP)-conjugated secondary antibodies (PeproTech Inc., USA) were used for immunostaining with the following primary antibodies: rabbit polyclonal anti-Robo1 (1:50, ab7279, Abcam, Cambridge, MA, USA), rabbit polyclonal anti-collagen I (1:100, 14695-1-AP, ProteinTech, Chicago, IL, USA), and rabbit monoclonal anti-α-SMA (1:500, ab108424, Abcam). For immunofluorescence staining, tissue sections were incubated with primary antibodies against Robo1 (1:50, ab7279, Abcam), α-SMA (1:200, BM0002, Boster Biological Technology, Wuhan, China), and desmin (1:100, ab227651, Abcam), followed by incubation with Alexa Fluor 594- and Alexa Fluor 488-conjugated secondary antibodies (1:200, Invitrogen, Carlsbad, CA, USA). All sections were stained with 1 μg/ml 4′,6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich) to visualize cell nuclei. At least three liver sections were included in each group. The stained sections were viewed under a microscope (Nikon Eclipse Ci), and images were captured using a high-resolution digital camera (Nikon digital sight DS-FI2).

The rats (n = 4 per group) without undergoing MRI experiments were deeply anesthetized. The perilesional cortex was separated, and protein levels were determined by Western blotting, as previously described (Zhan et al., 2020 (link)). Proteins were transferred onto polyvinylidene difluoride membranes, followed by blocking them with 5% nonfat milk for 2 h and subsequently incubating membranes at 4°C overnight with primary antibodies: anti-GAP-43 (1:40000; Epitomics, #2259-1), SYN (1:320000; Epitomics, #1870-1), Netrin-1 (1:2,000; Abcam, ab126729), DCC (1:1,000; Abcam, ab125280), Slit-2 (1:10,000; Abcam, ab134166), Robo-1 (1:1,000; Abcam, ab7279), NogoA (1:20,000; Abcam, ab62024), NgR (1:40,000; Abcam, ab62024), RhoA (1:20,000; Cell signaling, 2117s), ROCK-2 (1:50,000; Abcam, ab125025), and GAPDH (1:1,60,000; GeneTex, GTX627408). After washing, membranes were incubated with secondary anti-rabbit (1:20,000; Applygen Technologies Inc., C1309) or anti-mouse (1:20,000; NeoBioscience, cat. ANM 02-1, Lot. 0912) IgG (H + L)-HRP for 1 h at room temperature. Immunoreactive protein bands were detected by using the SuperECL Plus kit (Applygen, China, cat. No. P1050) and chemiluminescent imager (VILBER, United States). The intensities of target proteins were quantified by ImageJ software.

Tissue samples from resected pancreatic ductal adenocarcinomas and clinical data were obtained from patients collected under the auspices of the APGI. Tissues were fixed in 4% formalin and embedded in paraffin. In total, 96 spot tumour microarrays were created using triplicate cores of tumour tissue from each case. Immunohistochemistry (IHC) was performed as described above, using the following antibodies: Robo1 (dil. 1/200, ab7279, Abcam), Robo2 (dil. 1/50, ab75014, Abcam). The intensity of staining is scored using a four-point scale (0, 1, 2 and 3) to represent none, weak, moderate and strong levels of expression. An estimate of the percentage of stained cells was made and multiplied by the intensity score to calculate the H-score: (1 × % weak) + (2 × % moderate) + (3 × % strong). All sections were scored by the investigator and then second-scored by an expert pathologist without knowledge of patient outcomes. Three cores per tumour were scored, and the median H-score was used. Tumours with median H-score less than 200 were regarded as ROBO1 low-expression group.