XPNPEP2 is associated with lymph node metastasis in prostate cancer patients

Abstract

As we reported in our previous studies, TMTP1, a tumor-homing peptide, selectively targets highly metastatic tumors and their metastatic foci. Aminopeptidase P2 (XPNPEP2) is a receptor for TMTP1 tumor-homing peptide. However, the biological and clinical significance of Aminopeptidase P2 in human cancers remains unknown. In this study, the high-density multiple organ tumor tissue array was employed for the analysis of XPNPEP2 expression profiles in human specimens. The results showed that XPNPEP2 was moderately expressed in the normal prostate tissues, but significantly decreased in the prostate cancer. Hence we used TCGA, IHC, and ELISA to further analyze the expression of XPNPEP2 in tissues and serum of prostate cancer patients. In general, XPNPEP2 expression was lower in prostate cancer tissue than in normal prostate tissue, but was higher in prostate cancer tissues with local invasion and LN metastasis than in tissues with localized Pca. Western blot clarified XPNPEP2 had a secreted form in the serum. Then the serums of 128 Pca patients, 70 healthy males and 40 prostate hyperplasia patients were obtained for detecting serum XPNPEP2 levels.The results indicated that the concentration of XPNPEP2 in serums of Pca patients with LN metastasis (142.7 ± 14.40 ng/mL) were significantly higher than levels in Pca patients without LN metastasis (61.63 ± 5.50 ng/mL) (p < 0.01). An ROC analysis revealed that the combination of PSA and XPNPEP2 was more efficient than PSA or XPNPEP2 alone for predicting LN metastasis, especially for Pca patients with low serum PSA levels. In summary, serum XPNPEP2 levels when combined with PSA levels may result in increased sensitivity for predicting LN metastasis in Pca patients, especially for patients with low serum PSA levels.

Figure 1

Immunostaining of XPNPEP2 expression in human prostate tissue. (A) and (C) A prostate cancer tissue microarray PR1921 and 30 BPH tissues were employed for staining with anti-XPNPEP2, the representative figures were shown (A), and XPNPEP2 immunostaining scores was presented (C). Scale bar, 50 um. **p < 0.01. (B) and (D) XPNPEP2 expression in Pca subdivided into localized, locally invasive and LN-metastatic Pca was also analyzed (B, D). Scale bar, 100 um. *p < 0.05, **p < 0.01.

Figure 2

The correlation between XPNPEP2 and metastasis. In the same FFPE section, the XPNPEP2 expression in the occult metastases and in large foci were analyzed (A), scale bar, 10 um. The cross-cancer XPNPEP2 mutation analysis was demonstrated using data extracted from the cBioPortal online analysis tool (the cBioPortal for Cancer Genomics) (B).

Figure 3

Serum XPNPEP2 levels in normal and prostate cancer patients. Western blot assay clarified secreted XPNPEP2 in serums of Pca patients (A). The ELISA assay detected serum XPNPEP2 levels in normal, BPH and Pca patients using a human XPNPEP2 ELISA pair set, the scatter plots were showed (B). (C) Plots illustrated the serum XPNPEP2 levels in Pca patients with local invasion versus without local invasion. (D) Plots showing the serum XPNPEP2 levels in Pca patients with LN metastasis versus without LN metastasis.

Figure 4

The correlation between PSA and XPNPEP2. The corresponding serum levels of PSA in Pca patients were analyzed (A, B), A correlation analysis between serum XPNPEP2 and PSA levels was performed (C). The XPNPEP2 and PSA were simultaneous stained by IHC in the serial sections, it is not in accord with the two gene expression in two representative Pca patients (D), scale bar, 20 um.

Figure 5

ROC curve tests of serum XPNPEP2 and PSA. ROC curves of XPNPEP2, PSA and both of them for patients with LN metastasis versus patients without (A). Outliers of PSA (>500 ng/ml) were excluded and ROC curves were generated for XPNPEP2, PSA and both of them for patients with LN metastasis versus patients without LN metastasis (B).