Figure?2b shows a representative example of the interface between tumor and nontumor tissue, demonstrating that FAP expression was entirely limited to the tumor area. cancers. FAP was commonly expressed by cultured GNS cells but absent from normal neurons and astrocytes. Within glioblastoma tissues, the strongest expression of FAP was around blood vessels. In fact, almost every tumor vessel was highlighted by FAP expression, whereas normal tissue vessels and cultured endothelial cells (ECs) lacked expression. Single\cell analyses of dissociated tumors facilitated a detailed characterisation of the main cellular components of the glioblastoma microenvironment and revealed that vessel\localised FAP is because of expression on both ECs and pericytes. Conclusion Fibroblast ADH-1 trifluoroacetate activation protein is expressed by multiple cell types within glioblastoma, highlighting it as an ideal immunotherapy antigen to target destruction of both tumor cells and their supporting vascular network. gene expression in large patient cohorts, we mined published microarray and RNA sequencing datasets. Microarray data from The Cancer Genome Atlas (TCGA) revealed a significant overexpression of in glioblastoma compared to normal brain (Physique?1a). By setting a conservative threshold for expression based on the mean?+?3??SD of the normal tissue samples, 39.6% of glioblastoma tissues (216/548 specimens) expressed above the threshold, whereas none (0/9) of the normal brain tissues did. To support these microarray\based analyses, we also analysed RNA sequencing data from TCGA (Physique?1b). This revealed that both primary and recurrent glioblastoma expressed at significantly higher levels compared to less aggressive low\grade gliomas, with no significant difference in expression between primary and recurrent tumors. Open in a separate window Physique 1 expression in transcriptomic analyses of glioblastoma and normal tissues. (a, b) gene expression values from TCGA microarray (a) and RNAseq (b) datasets. The expression value for each tissue sample is usually shown. Red lines represent the median of each group, while ADH-1 trifluoroacetate dotted lines represent the threshold for expression, based on [mean?+?(3??SD)] of the respective normal brain dataset. The proportion of samples in each group with expression above the threshold is usually indicated at the top of the graphs. In a, groups were compared by the MannCWhitney (gene expression values, measured by RNAseq, were obtained from the GTEx portal for 51 normal tissue types and compared to cultured skin fibroblasts (black arrow; positive control). Box plots show median and 25th and 75th percentile; points are displayed as outliers if they are above or below 1.5 times the interquartile range. Number of samples analysed per tissue type ranged from 4 to 803, with a mean of 325. Blue dotted arrow highlights the 13 regions of brain tissue analysed. The above analyses revealed that some glioblastoma tissues show particularly elevated expression. To determine whether such marked overexpression was associated with poorer prognosis, we compared survival time for patients in the top 10% (expression range for the microarray dataset (Physique?1c). Indeed, the expression was particularly enriched in the mesenchymal tumors (Supplementary physique 1), in keeping with the ADH-1 trifluoroacetate poor prognosis of this subtype. 27 , 28 Interestingly, though, this previous analysis did not detect Itga8 the association between expression level and overall survival that we did, likely because samples were stratified into quartiles rather than comparing the top and bottom 10% of the expression range. Supplementary physique 1 also shows that high expression was associated with overexpression of gene signatures for (1) vascular function; (2) immune system; and (3) extracellular matrix remodelling and interactions. The link with vascular genes is particularly interesting in light of other findings to be discussed below. To avoid off\tumor toxicity, an ideal immunotherapy target antigen shows low to negligible expression in healthy tissues. Previous studies suggest that FAP meets this criterion, 17 , 18 , 20 , 21 , 22 but other studies have countervailing data. 23 , 29 To help clarify ADH-1 trifluoroacetate this issue, we examined expression in a broad range of normal tissues by analysing the GTEx dataset, which includes RNAseq data from 53 different tissue types collected from healthy individuals, as well as cultured skin fibroblasts (Physique?1d). In agreement with results in Figure?1b, and the findings of others, 20 , 22 transcripts were essentially undetectable in all 13 brain regions examined (median 0.31 transcripts per million (TPM) averaged across the different regions for all those individuals). In contrast, cultured skin fibroblasts, which would be expected to express high levels of expression, although certain tissues did show marginally elevated expression, particularly uterus (median 34.7), cervix (median 15.5 and 15.0 for endocervix and ectocervix, respectively) and arteries (median 23.5, 17.1 and 14.0 for aorta, coronary and tibial arteries, respectively). Together, these analyses reveal.