Fig 2: XPA-dependent gene expression including the steroid regulators AKR1C1 and AKR1C2A. Venn diagrams for high expressed genes (left) and low expressed genes (right) in all four pairs of cell lines at a cut off level of FC1.5 or more. Overlaps are shown, with each XPA- and XPA+ cell pair represented in a different color. Only 11 “high” and 16 “low” genes (boxed numbers) overlap between all four comparisons (cf. Table 1). B. Orientation of part of the AKR1C gene cluster on human chromosome 10 p15.1, with nucleotide positions shown, from the GRCh38/hg38 assembly. The H3K27Acetylation track (a histone modification often found near active genes) is shown for the K562 (blue) and NHEK (pink) cell lines, to emphasize that gene activation is different in between cell lines. Direction of transcription is shown by gray arrows.C. Immunoblot for products of three of the most differentially expressed genes (AKR1C1, AKR1C2 and AKR1C3) in lysates from XPA-proficient and deficient cell lines. An XPF immunoblot serves as a loading reference. AKR1C1 and AKR1C2 protein levels were lower in XPA-deficient cells compared to proficient cells, while AKR1C3 protein level levels do not obviously correlate with expression. The * denotes a nonspecific band detected by the AKR1C3 antibody. The blots were scanned and the AKR1C protein level determined, relative to the respective XPF loading control. The numbers under the blots show this AKR1C/XPF ratio, normalized to 1 for the XPA+ cell line in each pair.

Fig 3: XPA is a nuclear protein resistant to leakage from the nucleus during DNA repairA. Immunofluorescence (IF) staining with 12F5 antibody (green) and DAPI (blue). Cells in the top two panels were stained without exposure to buffer containing detergent. This shows that XPA is localized in the nuclei of complemented XPA/XP12RO cells, regardless of exposure to UVC. The lower two panels show that brief exposure to a buffer containing 0.05% NP-40 causes leakage of XPA into the cytoplasm in untreated cells but not in cells exposed to 20 J/m2 UVC.B. In XPA/XP12RO cells, the leakage of XPA into the cytoplasm decreases with increasing doses of UVC (0 J/m2–60 J/m2). C. XPA/XP12RO cells were exposed to psoralen with or without UVA radiation (PUVA). The 2 h sample received 54 kJ/m2 (first dose 18 kJ/m2, second dose 36 kJ/m2) and the 2.5 h sample received 72 kJ/m2 (first dose 18 kJ/m2, second dose 54 kJ/m2). All cells were exposed briefly to a buffer containing 0.05% NP-40 before fixation for IF. XPA leaked to the cytoplasm in cells exposed to psoralen alone, but not in PUVA treated cells. D. Quantification showing the% of XPA/XP12RO cells with cytoplasmic XPA following PUVA treatment with increasing total doses of UVA. All cells were exposed briefly to a buffer containing 0.05% NP-40 before fixation for IF. The red bar shows XPA leakage in cells exposed only to psoralen, and the blue bar shows a reduction in leakage following DNA damage induction with increasing UVA doses.

Fig 4: XPA-dependent transcription related functions after retinoic acid transactivationA. Retinoic acid-dependence of RARB expression. qPCR results for XPA-proficient and deficient cell lines showing responses to RA treatment at 7 h. RARB gene expression was analyzed with GAPDH as control. The expression of RARB mRNA in RA treated cells was normalized to the DMSO-only controls. The black bars indicate the range of data in 2 experiments. B. Pathways for genes upregulated by RA treatment. IPA analysis assigned pathways to genes with higher expression (FDR = 0.05) in RA treated cells compared to DMSO controls. The scores are presented as negative log values of p-values. XPA+ cells (red) have higher significant differences than XPA- cells (blue), for some pathways shown on the x-axis. Transcription-related functions are highlighted in red. Mean values of comparisons between the four pairs of cell lines are shown here, and results for individual cell lines are given in Supplemental Fig. 1.

Fig 5: XPA expression, UVC sensitivity and NER complementationA. Immunofluorescence images using anti-XPA antibody (12F5) and DAPI showing the presence of XPA in all XPA-proficient cell lines and little or no XPA in the deficient cell lines. DAPI (blue) staining of nuclei included for reference. B. Immunoblot of the cell lines with anti-XPA antibodies (CJ1 and GTX103168), as well as alpha-tubulin or XPF antibodies as loading controls. C. Quantification of the immunoblots for the HeLa S3 cell lines shown in (B). The blue bars represent the alpha-tubulin control. The red bars represent the anti-XPA antibody CJ1. D. UVC clonogenic survival for XPA-disrupted cell lines KO38 and KO142 compared to wild type HeLa S3. Each point on the survival curves represents the mean of 3 independent experiments. KO38 is more sensitive to UVC compared to HeLa S3. KO142 has only slightly more sensitivity compared to HeLa S3. E. NER complementation assay. KO38 whole cell extracts complemented with purified XPA protein to produce incision products. 50 ng 1,3-intrastrand d(GpTpG)-cisplatin substrate was incubated with 45 µg whole cell extract protein. Various amounts of purified XPA protein were added to KO38 extracts. NER activity is detected with 44 and 88 µg of purified XPA combined with KO38 extracts. Incision products were observed as a characteristic ladder of bands where complementation was successful (lanes 7 and 8), like the incision products with NER proficient HeLa S3 extract (lane 1).