Fig 1: Characteristics of the local immune microenvironment in CAIS testicular tissue and immune correlation analysis of ASAP2. (A) Comparison of immune cell infiltration proportions between the complete androgen insensitivity syndrome (CAIS) group and the normal control group in testicular tissue. The box plots show that, after CIBERSORTx deconvolution, monocytes were significantly increased in the CAIS group (P < 0.01), whereas the infiltration proportions of M2 macrophages and activated natural killer cells were increased and that of naive B cells was decreased (P < 0.05); no statistically significant differences were observed for the remaining immune cell subsets. (B) Stacked bar plot showing the immune cell composition of each sample in the GSE125222 cohort. Different colors represent different immune cell subsets, and the height of each colored segment indicates its relative abundance within an individual sample. (C) Spearman correlation analysis between ASAP2 expression levels and the infiltration proportions of different immune cell subsets. The x-axis represents the correlation coefficient (R). Red dots indicate statistically significant correlations, whereas gray dots indicate correlations that did not reach statistical significance. ASAP2 expression was significantly positively correlated with naive B-cell infiltration (P < 0.05), whereas its correlations with the other immune cell subsets were not statistically significant. * P < 0.05, ** P < 0.01.
Fig 2: Single-gene gene set enrichment analysis based on ASAP2 reveals its associated pathway enrichment profile. (A) Running enrichment score plot of gene set enrichment analysis (GSEA) for ASAP2-associated gene sets. The figure shows representative significantly enriched pathways within the ranked gene list, including Epstein-Barr virus infection, the NOD-like receptor signaling pathway, and Parkinson disease. The curves at the top represent the running enrichment scores for each pathway, the vertical lines in the middle indicate the distribution positions of pathway-related genes within the ranked gene list, and the gray area at the bottom shows the variation trend of the ranked list metric. The table in the upper right corner lists the corresponding P values and adjusted P values for these pathways. (B) GSEA ridge plot of significantly enriched pathways associated with ASAP2. Different peak shapes indicate the enrichment distribution of core genes from each pathway within the ranked list, and different colors correspond to the adjusted P values. Significantly associated pathways included cytoskeleton in muscle cells, the tumor necrosis factor signaling pathway, antigen processing and presentation, the NOD-like receptor signaling pathway, proteasome, and several pathways related to neurodegenerative diseases.
Fig 3: Preliminary evaluation of ASAP2 expression in public transcriptomic cohorts and independent clinical samples. (A) Differential expression of ASAP2 in two independent public transcriptomic datasets. The left panel shows the GSE125222 testicular tissue cohort, and the right panel shows the GSE300211 patient-derived fibroblast cohort treated with dihydrotestosterone (DHT). The box plots show that ASAP2 was downregulated in the complete androgen insensitivity syndrome (CAIS) group compared with the control group in both datasets. The y-axis represents expression levels normalized by variance stabilizing transformation (VST). (B) Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) evaluation of ASAP2 mRNA expression in the independent PBMC pilot cohort. The dot plot on the left shows the relative expression levels of individual biological replicates and the overall trend, whereas the bar plot on the right presents the grouped quantitative results, showing that ASAP2 mRNA expression was lower in the CAIS group than in the control group. (C) Western blot evaluation of ASAP2 protein expression in the independent PBMC pilot cohort. The left panel shows representative immunoblot bands and a schematic of relative protein expression; the theoretical molecular weights of ASAP2 and β-actin were 112 kDa and 43 kDa, respectively. The bar plot on the right shows grayscale quantification normalized to β-actin, indicating reduced ASAP2 protein expression in the CAIS group. Quantitative data are presented as mean ± standard deviation (SD); ** P < 0.01.
Fig 4: Protein-protein interaction network construction and multi-algorithm feature selection for identification of candidate core molecules in complete androgen insensitivity syndrome (CAIS). (A) Protein-protein interaction (PPI) network constructed based on 328 overlapping candidate genes. Node color from dark to light indicates degree from high to low. (B) Top 10 hub genes in the PPI network ranked by degree, including BDNF, TLR4, APOE, IL7, HGF, ITGA8, VTN, ERBB3, GBP1, and SNCA. (C) Ten-fold cross-validation curve of the least absolute shrinkage and selection operator (LASSO) regression model. The y-axis represents binomial deviance, and the dashed lines indicate the position corresponding to the optimal penalty parameter. (D) Top 15 important genes identified by the random forest model, with feature importance ranked according to mean decrease Gini. (E) Top 15 feature genes identified by support vector machine-recursive feature elimination (SVM-RFE), with bar length representing the absolute weight of each feature. (F) Receiver operating characteristic (ROC) curves of the candidate core molecules ASAP2 and GRIN2D obtained after cross-screening by the three algorithms. Both genes showed apparent discriminative patterns within the small discovery cohort (area under the curve [AUC] = 1.000), which should be interpreted as an internal screening signal rather than evidence of clinical diagnostic utility.
Supplier Page from Abcam for Anti-ASAP2 antibody