The importin beta superfamily member RanBP17 exhibits a role in cell proliferation and is associated with improved survival of patients with HPV+ HNSCC

Abstract

Background: More than twenty years after its discovery, the role of the importin beta superfamily member Ran GTP-binding protein (RanBP) 17 is still ill defined. Previously, we observed notable RanBP17 RNA expression levels in head and neck squamous cell carcinoma (HNSCC) cell lines with disruptive TP53 mutations.

Methods: We deployed HNSCC cell lines as well as cell lines from other tumor entities such as HCT116, MDA-MB-231 and H460, which were derived from colon, breast and lung cancers respectively. RNAi was used to evaluate the effect of RanBP17 on cell proliferation. FACS analysis was used for cell sorting according to their respective cell cycle phase and for BrdU assays. Immunocytochemistry was deployed for colocalization studies of RanBP17 with Nucleolin and SC35 (nuclear speckles) domains. TCGA analysis was performed for prognostic assessment and correlation analysis of RanBP17 in HNSCC patients.

Results: RNAi knockdown of RanBP17, significantly reduced cell proliferation in HNSCC cell lines. This effect was also seen in the HNSCC unrelated cell lines HCT116 and MDA-MB-231. Similarly, inhibiting cell proliferation with cisplatin reduced RanBP17 in keratinocytes but lead to induction in tumor cell lines. A similar observation was made in tumor cell lines after treatment with the EGFR kinase inhibitor AG1478. In addition to previous reports, showing colocalization of RanBP17 with SC35 domains, we observed colocalization of RanBP17 to nuclear bodies that are distinct from nucleoli and SC35 domains. Interestingly, for HPV positive but not HPV negative HNSCC, TCGA data base analysis revealed a strong positive correlation of RanBP17 RNA with patient survival and CDKN2A.

Conclusions: Our data point to a role of RanBP17 in proliferation of HNSCC and other epithelial cells. Furthermore, RanBP17 could potentially serve as a novel prognostic marker for HNSCC patients. However, we noted a major discrepancy between RanBP17 RNA and protein expression levels with the used antibodies. These observations could be explained by the presence of additional RanBP17 splice isoforms and more so of non-coding circular RanBP17 RNA species. These aspects need to be addressed in more detail by future studies.

Keywords: HNSCC; HPV; Proliferation; RanBP17; Survival; circRanBP17.

Fig. 1

RNAi knockdown of RanBP17 inhibits cell proliferation independent of CDDP treatment. A XTT proliferation assay depicting the relative cell viability of the HNSCC cell line UM-SCC-3 after RNAi knockdown of RanBP17 using pool 1 siRNA (consists of si1, si2, si4, si5 as depicted in D) following incubation with CDDP (6.25, 12.5, 25, 50 and 100 µmol/L). Note the significant reduction in cell viability at 0 µmol/L CDDP in UM-SCC-3 cells after RanBP17 knockdown. B Effect of RanBP17 RNAi knockdown alone (using pool 1 siRNA) was tested in the indicated cell lines. Efficient RNAi knockdown of RanBP17 RNA using pool 1 siRNA is shown for HCT116^{p53 wt/wt} cells. C The 3 most responsive cell lines from B were used to evaluate the effect of all 8 single RanBP17 specific siRNAs (si1-8). D Depicted are the siRNA (si1-8) binding sites as well as their location within the RanBP17 reference sequence NM_022897.4. Also included is the respective information for the updated RanBP17 reference sequence NM_022897.5, which differs from NM_022897.4 by lacking the first 136 nucleotides. Underlined nucleotides refer to the nucleotide sequence of the adjacent exon (shown in brackets). NT_non target RNA, siRanBP17=silencing RNA (pool 1) directed against RanBP17. Statistical tests; A, B, C: Unpaired Students t-test, two-sided (n=4 for all analyses) *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001

Fig. 2

CDDP treatment of normal human epidermal keratinocytes (NHEK) inhibits RanBP17 expression. A In one preliminary Western Blot experiment, a reduction of RanBP17 protein expression was observed in NHEK, treated with 5 and 50 µmol/L CDDP. B RT-qPCR analysis was subsequently performed to evaluate the response of RanBP17 RNA expression levels on CDDP treatment. Statistical tests in B One way ANOVA with Tukey´s correction for multiple comparisons (n=4) **p<0.01

Fig. 3

Course of RanBP17 expression in HCT116^p53wt/wt and UM-SCC-3 tumor cells after CDDP treatment. A RanBP17 RNA expression at different CDDP concentrations (0, 0.78, 1.56, 3.13, 6.25, 12.5, 25, 50, 100 µmol/L) is shown for HCT116^p53wt/wt cells using 11 different primer pairs (B). The same is shown for the UM-SCC-3 cell line (C). Here only 3 primer pairs (P3, P5, P11) amplified PCR products. Background information for primer pairs P1-P11 are depicted in Supplementary Fig. S4. Statistical tests: One way ANOVA with Tukey´s correction for multiple comparisons (n=1-6) *p<0.05; **p<0.01; ****p<0.0001

Fig. 4

Expression of RanBP17 during the cell cycle. A FACS sorting yields 4 cell populations, representing the G1/G0, S and G2/M phases of the cell cycle as well as a population labeled „All“, which contains all phases for comparison. Western blot analysis points to increased expression levels of RanBP17 (antibody orb226830) during the S-phase of the cell cycle, which is particularly prominent in UT-SCC-26A. Cell cycle associated proteins (PCNA, cyclin-B1, CDK1) show an expected increase of their expression during S- and into G2/M-phase, thereby helping to validate the accuracy of cell sorting to the respective phase of the cell cycle. B Total cellular protein was significantly elevated in S phase cells compared with cells in the G1/G0 phase of the cell cycle, which is an expected finding for cells in division. A similar observation was made for total RNA levels. C No significant differences in RanBP17 RNA levels could be noted between the different cell cycle phases for UM-SCC-3 (n=4) and UT-SCC-26A (n=3) cells. Statistical tests: B and C: One way ANOVA with Dunn´s correction for multiple comparisons. *p<0.05. (A: Protein bands shown in the upper two panels (RanBP17-orb226830 and GAPDH) and the lower six panels (RanBP17-GTX70420, PCNA, Cyclin-B1, GAPDH, CDK1, Beta-Tubulin) are derived from lysates of two independent experiments). TP53 mutation in UM-SCC-3: p.R248Q (nuclear p53); TP53 mutation in UT-SCC-26A: p.Y236* (cytoplasmic p53)

Fig. 5

Treatment with the EGFR kinase inhibitor AG1478 arrests cells in G1 and leads to accumulation of RanBP17. A Treatment of UM-SCC-3 and UT-SCC-26A HNSCC cell lines with the EGFR kinase inhibitor AG1478 reduced cell proliferation (PCNA) but displayed an unexpected rise of RanBP17 protein expression levels (arrow, antibody orb226830). B Cell cycle sorting of UM-SCC-3 cells reveals a significant rise of RanBP17 and drop of PCNA in G1/G0 cells after AG1478 treatment. C Cell cycle analysis shows a reduction of cells in the S and elevation of cells in the G1/G0 phase of the cell cycle after AG1478 exposure (significant in UM-SCC-3) consistent with a block in the G1 phase as previously reported for this kinase inhibitor [21]. D A FACS based BrdU proliferation assay was performed for more detailed S-phase analysis. The S-phase was separated into three subcompartments (S1-S3, gates were adjusted manually for each sample to account for shifts in the whole cell population). A significant drop of cells in the early S phase (S1) was observed in both tested HNSCC cell lines. Cell numbers in S2 and S3 were significantly reduced in UM-SCC-3 but not UT-SCC-26A cells. Statistical tests: C (column chart) and D (left column chart): Unpaired Students t-test, two-sided; D (right column chart): One way ANOVA with Tukey´s correction for multiple comparisons. (n=4 for all analyses) *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. TP53 mutation in UM-SCC-3: p.R248Q (nuclear p53); TP53 mutation in UT-SCC-26A: p.Y236* (cytoplasmic p53)

Fig. 6

RanBP17 localizes to nuclear bodies distinct from SC35 domains. UM-SCC-3 and UT-SCC-26A cells were co-stained with the RanBP17-specific antibody HPA029568 and a SC35-specific antibody (clone SC-35), exhibiting a distinct nuclear localization of RanBP17 (arrowheads) that is different from SC35 domains. DAPI was used to counterstain the nucleus. TP53 mutation in UM-SCC-3: p.R248Q (nuclear p53); TP53 mutation in UT-SCC-26A: p.Y236* (cytoplasmic p53)

Fig. 7

RanBP17 does not localize to nucleoli. UM-SCC-3 and UT-SCC-26A cells were co-stained with the RanBP17-specific antibody HPA029568 and an antibody directed against Nucleolin (clone ZN004), exhibiting a distinct nuclear RanBP17 localization (arrowheads) that is different from nucleoli. DAPI was used to counterstain the nucleus. TP53 mutation in UM-SCC-3: p.R248Q (nuclear p53); TP53 mutation in UT-SCC-26A: p.Y236* (cytoplasmic p53)

Fig. 8

Timer2.0 analysis (https://timer.cistrome.org/) showing A the Gene_DE comparison of the Exploration tab for the different TCGA entities for RANBP17 and B the comparison for HPV+ (upper part) and HPV- (lower part) samples within the TCGA-HNSC (HNSCC, head and neck squamous cell carcinoma) cohort regarding RANBP17 expression and TP53 mutational status with correlating age adjusted Kaplan-Meier (cumulative survival) plot based on a 50% RanBP17 expression split (accessible at Gene_Outcome and Gene Mutation of the Exploration tab). C comparison of RanBP17 and CDKN2A (p16^INK4A). *: p-value < 0.05; **: p-value <0.01; ***: p-value <0.001 (Wilcoxon test). HPV-: human papilloma virus negative; HPV+: human papilloma virus positive; Abbreviations according to The Cancer Genome Atlas (TCGA). TPM: transcripts per kilobase million