Figure 6. Identification of regulatory elements activated by FLCN loss in RPTEC and BHD tumors.

(A) Identification of transcriptional regulatory elements associated with loss of FLCN expression. Regulons were identified by iRegulon (Janky et al., 2014), using an input a list of differential expressed genes (Figure 3). (B) Upstream regulons enriched in FLCN^NEG RPTEC based on significantly upregulated genes derived from our transcriptomic data set (n = 711, FDR < 0.05 and logFC > 2). Transcription factors with normalized enrichment scores (NES) higher than 3.5 are shown, together with detected number of targets, motifs, and elements. ISREs are highlighted in orange. Upper part shows motifs enriched 500 bp upstream from transcription start site (TSS), lower part shows motifs enriched 20 kb around TSS. (C) Upstream regulons enriched in FLCN^NEG RPTEC based on significantly upregulated proteins derived from our proteomic data set (n = 498, p<0.05 and FC > 2). Transcription factors with normalized enrichment score (NES) higher than 3.5 are shown, together with number of targets, motifs, and elements detected. ISREs are highlighted in orange and E-boxes in purple. STAT1 appears twice due to the fact that iRegulon ranks this transcription factor to be the most likely upstream regulator for two sets of targets genes, containing slightly different ISRE-motifs 20 kb upstream from the TSS. (D) Two major enriched motif elements detected in iRegulon analysis of genes upregulated in FLCN^NEG RPTEC. Regulons can be assigned to E-box (in purple) or ISRE (in orange) motif group. (E) Bar graphs of RNA expression levels of genes associated with an E-box motif, derived from RPTEC transcriptomic data set. FLCN^POS values are shown in pink and FLCN^NEG values are shown in green. Significant p-values are indicated as *≤0.05, **≤0.01, ***≤0.001, ****≤0.0001. (F) Bar graphs of protein expression levels of genes associated with an E-box motif, derived from RPTEC proteomic data set. FLCN^POS values are shown in pink and FLCN^NEG values are shown in green. FNIP2 peptides were not detected in our proteomic experiment and therefore absent in the bar graph. Significant p-values are indicated as *≤0.05, **≤0.01, ***≤0.001, ****≤0.0001. (G) Bar graphs of RNA expression levels of genes associated with an ISRE motif derived from RPTEC transcriptomic data set. FLCN^POS values are shown in pink and FLCN^NEG values are shown in green. Significant p-values are indicated as *≤0.05, **≤0.01, ***≤0.001, ****≤0.0001. (H) Bar graphs of protein expression levels of genes associated with an ISRE motif derived from RPTEC proteomic data set. FLCN^POS values are shown in pink and FLCN^NEG values are shown in green. IRF9 and IRF4 peptides were not detected in our proteomic experiment and therefore absent in the bar graph. Significant p-values are indicated as *≤0.05, **≤0.01, ***≤0.001, ****≤0.0001. (I) Dot plot of protein expression levels of genes associated with an E-box (left) or ISRE motif (right) derived from BHD kidney tumor proteomic data sets (see Figure 6—figure supplement 1D and E), as compared to normal kidney tissue. FLCN^POS values are shown in pink and FLCN^NEG values are shown in green. STAT2 levels were below detection levels in these protein extracts.

Figure 6—figure supplement 1. Extended GSEA and iRegulon analysis of FLCN loss in RPTEC and BHD tumors.

(A) For Gene Set Enrichment Analysis (GSEA) genes or proteins were ranked based on p‐values, with genes/proteins significantly higher expressed in FLCN^NEG RPTEC on top of the list (hallmark gene sets, classic ES). Enriched hallmark gene sets are ranked by normalized enrichment score (NES), with gene sets enriched in FLCN^NEG marked in green and gene sets enriched in FLCN^POS marked pink. The size of the dot reflects the significance of the enrichment (FDR=false discovery rate). (B) Identification of master transcriptional regulators associated with FLCN expression. Regulons were identified with iRegulon tool (Janky et al., 2014), using an input a list of FLCN differential expressed genes derived from our transcriptomic and proteomic analyses (Figures 2 and 3). Tables showing upstream regulons enriched in FLCN^POS RPTEC based on significantly upregulated genes (n = 62, FDR < 0.05 and logFC > 2) and proteins (n = 138, p<0.05 and FC > 2). Transcription factors with normalized enrichment score (NES) higher than 3.5 are shown, together with detected number of targets, motifs, and elements. Upper part of table shows motifs enriched 500 bp upstream from transcription start site (TSS), lower part shows motifs enriched 20 kb around TSS. (C) Upregulation of target genes with an ISRE or E-box motif in FLCN^NEG RPTEC cell lines was validated by qRT-PCR. Results shown are representative for three independent experiments with three technical replicates. To determine quantitative gene expression, data levels were normalized to the geometric mean of two housekeeping genes. See Figure 6—figure supplement 1—source data 1 for raw qRT-PCR values and fold change calculations. (D) Absence of FLCN protein expression in two BHD tumors (BHD T1 and BHD T2) was verified by western blotting. GAPDH was used as loading control and western blot was performed three times. (E) Coomassie staining of gel containing samples for proteomic analyses of BHD T1, BHD T2 and two normal kidney lysates (HK_1 and HK_2). For measurements, 5-band fractionation was applied to all samples and two lanes of HK_2 were combined.