Fig 2: Validation of expression changes of ANGPT2 and CCL3 in transcription and protein level. (A): Validation of microarray data by quantitative Real-Time PCR (qRT-PCR). The expression patterns of ANGPT2 and CCL3 increased 2.4-fold (P < 0.01) and 3.6-fold (P = 0.02) in aortic tissues of AAD compared with healthy control. GAPDH was used as housekeeping gene. **: P < 0.01, *: P < 0.05. (B,C): Western blot for ANGPT2 and CCL3 protein from aortic tissues of healthy control and AAD patients. ß-actin serves as an internal control. ANGPT2 increased up to 4.2 folds in AAD patients, whereas CCL3 did not change.

Fig 3: ROC analysis of ANGPT2.

Fig 4: Clinicopathological characteristics of CCL4 and Angpt2 expression in OSCC tissue. (A,B) The expression profiles of CCL4 and ANGPT2 in tumor tissue and normal tissue specimens were analyzed in records downloaded from the TCGA database. (C,D) OSCC and adjacent normal tissue specimens from China Medical University Hospital were subjected to IHC evaluations with CCL4 or Angpt2 antibodies, and the positive staining percentages of the areas were scored as: 1 (<20%), 2 (20–40%), 3 (40–60%), 4 (60–80%), or 5 (80–100%). (E) A positive correlation was observed between CCL4 and Angpt2 expression (Spearman’s rank correlation coefficient, r2 = 0.4919).

Fig 5: ANGPT2 upregulation during HG exposure. A Representative immunoblot of ANGPT2, STX4 and DNM2. B Densitometric analysis of ANGPT2 immunoblot signals (n = 3). *P < 0.05 vs. NG (normal glucose). C Densitometric analysis of STX4 immunoblot signals (n = 3). No statistical significance vs. NG. D Densitometric analysis of DNM2 immunoblot signals (n = 3). No statistical significance vs. NG. E Representative immunohistochemical (IHC) staining of ANGPT2 in diabetic kidney. F Statistical data of IHC staining of ANGPT2 (n = 5). **P < 0.01 vs. Control