Research ArticleDRUG RESISTANCE

Regulation of the error-prone DNA polymerase Polκ by oncogenic signaling and its contribution to drug resistance

See allHide authors and affiliations

Science Signaling  28 Apr 2020:
Vol. 13, Issue 629, eaau1453
DOI: 10.1126/scisignal.aau1453
  • Fig. 1 Treatment of melanoma cells with BRAF or MAP kinase inhibitors modulates Polκ expression and localization.

    (A) qRT-PCR to detect the mRNA expression of Polκ relative to the DMSO control was performed on A375 cells treated with DMSO or 5 μM PLX4032 for 2, 8, 24, 48, or 72 hours. Mean ± SEM, n = 3 experiments; *P < 0.05, paired two-tailed t test. (B) Western blot analysis for Polκ was performed on A375 cells treated with DMSO (−) or 5 μM PLX4032 (+) for 48 hours. (C) Quantification of the Western blot data relative to the DMSO control. β-Actin or lamin-B1 served as the loading control. Mean ± SEM from five to eight independent experiments; ns = nonsignificant, **P < 0.01, paired two-tailed t test. (D) Immunofluorescence staining of A375 cells treated with DMSO or 5 μM PLX4032 for 24 hours. Scale bars, 10 μm. (E) Quantification of the percentage of cells with nuclear enrichment of Polκ. Mean ± SEM; at least 2500 cells were counted for each sample across 22 fields of view. ***P < 0.001, paired two-tailed t test. (F) Immunofluorescence staining of A375 cells treated with DMSO or MEK inhibitors (10 μM CI-1040 and 10 μM U0126) for 24 hours. Scale bars, 10 μm. (G) Quantification of the percentage of cells with nuclear enrichment of Polκ. Mean ± SEM; at least 1000 cells were counted for each sample across 18 fields of view. ***P < 0.001, paired two-tailed t test. (H) Immunofluorescence staining of A375 cells treated with DMSO or ERK inhibitors (1 μM ulixertinib and 1 μM SCH772984) for 24 hours. Scale bars, 10 μm. (I) Quantification of the percentage of cells with nuclear enrichment of Polκ. Mean ± SEM; at least 250 cells were counted for each sample across six fields of view. ***P < 0.001, paired two-tailed t test.

  • Fig. 2 mTOR signaling regulates Polκ subcellular localization.

    (A) Immunofluorescence staining of A375 cells treated with DMSO or 5 μM PLX4032 for 24 hours. Scale bars, 10 μm. (B) Quantification of the percentage of cells with phosphorylated S6 (p-S6; S240/244). Mean ± SEM; at least 400 cells were counted for each sample across eight fields of view. ***P < 0.001, paired two-tailed t test. (C) Immunofluorescence staining of A375 cells treated with 5 μM PLX4032 for 3, 6, 12, or 24 hours. The bar graph shows the percentage of cells with nuclear enrichment of Polκ (left axis, solid line) and the percentage of cells with p-S6 (right axis, dashed line). Mean ± SEM; at least 150 cells were counted for each sample across four to eight fields of view. The Pearson correlation coefficient (r) was calculated for the percentage of cells with nuclear enrichment of Polκ versus the percentage of cells with p-S6. (D) Immunofluorescence staining of A375 cells treated with DMSO, 0.5 μM rapamycin, 5 μM PP242, or 30 μM LY294002 for 6 hours. Scale bars, 10 μm. (E) Quantification of the percentage of cells with nuclear enrichment of Polκ after 6 hours of treatment. Mean ± SEM; at least 200 cells were counted for each sample across 8 to 12 fields of view. **P < 0.01, ***P < 0.001, paired two-tailed t test.

  • Fig. 3 The effects on Polκ and p-S6 are driver gene specific across tumor types.

    (A) Immunofluorescence staining of PC-9 cells treated with DMSO, 5 μM PLX4032, or erlotinib (0.1 or 1 μM) for 24 hours. Scale bars, 10 μm. (B) Quantification of the percentage of cells with nuclear enrichment of Polκ. Mean ± SEM; at least 100 cells were counted for each sample across four fields of view. ns = nonsignificant, ***P < 0.001, paired two-tailed t test. (C) Quantification of the percentage of cells with p-S6 (S240/244). Mean ± SEM; at least 100 cells were counted for each sample across four fields of view. ns = nonsignificant, ***P < 0.001, paired two-tailed t test. (D) Immunofluorescence staining of SK-BR3 cells treated with DMSO, 5 μM PLX4032, or lapatinib (0.1 or 1 μM) for 24 hours. Scale bars, 10 μm. (E) Quantification of the percentage of cells with nuclear enrichment of Polκ. Mean ± SEM; at least 200 cells were counted for each sample across seven fields of view. *P < 0.05, ***P < 0.001, paired two-tailed t test. (F) Quantification of the percentage of cells with p-S6 (S240/244). Mean ± SEM; at least 200 cells were counted for each sample across seven fields of view. *P < 0.05, ***P < 0.001, paired two-tailed t test.

  • Fig. 4 Exportin-1 plays a role in the subcellular localization of Polκ.

    (A) Immunofluorescence staining of A375 cells treated with DMSO or 5 μM PLX4032 ± 20 μM leptomycin B (LMB) for 6 hours. Scale bars, 10 μm. (B) Quantification of the percentage of cells with nuclear enrichment of Polκ after 3 or 6 hours. Mean ± SEM; at least 100 cells were counted for each sample across four to eight fields of view. **P < 0.01, ***P < 0.001, paired two-tailed t test. (C) Schema detailing the experiment design for the LMB recovery assay subsequently shown in (D) and (E). (D) Immunofluorescence staining of A375 cells treated with DMSO or 5 μM PLX4032 for 24 hours and then DMSO or 10 μM LMB for 24 hours. Scale bars, 10 μm. (E) Quantification of the percentage of cells with nuclear enrichment of Polκ. Mean ± SEM; at least 100 cells were counted for each sample across seven fields of view. *P < 0.05, **P < 0.01, ***P < 0.001, paired two-tailed t test.

  • Fig. 5 Polκ overexpression can lead to increased drug resistance.

    (A) Schema detailing how single-cell clones of A375 cells containing the doxycycline (dox)–inducible Polκ overexpression (OE) construct were created and how the dox− (− control) and dox+ (Polκ OE) populations were generated and then used to measure drug resistance. (B) Drug resistance of the dox− and dox+ populations of two different clones of dox-inducible Polκ overexpression cells to PLX4032 was determined by the CyQUANT Direct assay. For each population, the relative cell viability at each dose, compared with the 0 μM dose control (DMSO), was calculated. Mean ± SEM, n = 4 to 5 experiments; ns = nonsignificant, *P < 0.05, **P < 0.01, ***P < 0.001, Fisher’s least significant difference (LSD) test. A comparison between the two populations in their response to PLX4032 by two-way ANOVA gave a P value of 0.0130 for clone #1 and 0.0294 for clone #2. (C) Drug resistance of the dox− and dox+ populations of two different clones of dox-inducible Polκ overexpression cells to PD0332991 was determined by the CyQUANT Direct assay. For each population, the relative cell viability at each dose, compared with the 0 μM dose control (DMSO), was calculated. Mean ± SEM, n = 3 to 4 experiments; ns = nonsignificant, Fisher’s LSD test. A comparison between the two populations in their response to PD0332991 by two-way ANOVA gave a P value of 0.3785 for clone #1 and 0.2278 for clone #2.

  • Fig. 6 Knockout of MMR genes but not overexpression of Polκ causes increased mutagenesis.

    (A) Schema detailing how single-cell clones of A375 cells containing the SV40p-GFP and dox-inducible Polκ overexpression (OE) constructs were created and how the dox− (− control) and dox+ (Polκ OE) populations can be generated. Cas9 and sgRNAs against MMR genes (MLH1, MSH2, MSH6, or PMS2) or a negative control were added to the cells, and each knockout (KO) was validated using the Surveyor Mutation Detection Kit. Cells from each validated KO were treated for 14 days with media with or without dox, and then FACS was used to determine the percentage of GFP-negative versus GFP-positive cells in each population. (B and C) GFP intensity was measured by flow cytometry after SV40p-GFP, dox-inducible Polκ OE cells with or without MMR KO were treated with media with or without dox for 14 days. The bar graph shows the percentage of GFP-negative (gfp−) and GFP-positive (gfp+) cells for each sample from a representative experiment, performed using two different sets of sgRNAs against the MMR genes: sgRNAs #1 (B) and sgRNAs #2 (C).

  • Fig. 7 CRISPR of Polκ demonstrates pathways involved in drug resistance and immune response.

    (A) Schema detailing how A375 cells were treated with control or Polκ sgRNA and Cas9, single-cell clones were created, and the resulting cells were used to measure resistance to PLX4032. The graph shows the resistance of control and Polκ KO cells to PLX4032 as determined by the CyQUANT Direct assay. For each population, the relative cell viability at each dose, compared with the 0 μM dose control (DMSO), was calculated. Mean ± SEM, n = 3 experiments; **P < 0.01, all other comparisons were nonsignificant, Fisher’s LSD test. (B) RNA-seq was performed on the control and Polκ KO cells. Dual waterfall plot of top/bottom 50 gene sets from GSAA comparing POLK sgRNA versus control sgRNA ranked by normalized association score (NAS). (C) Heatmap of top leading edge genes from selected gene sets (up to 10 genes per gene set).

Supplementary Materials

  • stke.sciencemag.org/cgi/content/full/13/629/eaau1453/DC1

    Fig. S1. Validation of the Polκ antibody.

    Fig. S2. MAPK inhibition in melanoma cell lines increases POLK mRNA levels and changes the subcellular localization of Polκ.

    Fig. S3. DNA damage is not associated with the subcellular shift of Polκ.

    Fig. S4. Cell cycle inhibition is not responsible for the shift in the subcellular localization of Polκ.

    Fig. S5. MAPK inhibition decreases the abundance of p-S6.

    Fig. S6. BRAF inhibition induces ER stress, and other inducers of ER stress also change the subcellular localization of Polκ.

    Fig. S7. The effects on POLK mRNA levels are driver gene specific across tumor types.

    Fig. S8. Glucose starvation also modulates the subcellular localization of Polκ.

    Fig. S9. Inhibiting importin-β does not prevent the nuclear shift of Polκ.

    Fig. S10. Inhibition of the proteosome does not have a differential effect on the protein level of Polκ after treatment with DMSO or PLX4032.

    Table S1. qRT-PCR primers used in the study.

    Table S2. sgRNAs and validation primers used for MMR and Polκ KO.

    Table S3. RNA-seq and GSEA analysis on Polκ-CRISPR A375 cells.

    Data file S1. Differentially expressed genes between A375 cells with POLK sgRNA versus control sgRNA.

  • The PDF file includes:

    • Fig. S1. Validation of the Polκ antibody.
    • Fig. S2. MAPK inhibition in melanoma cell lines increases POLK mRNA levels and changes the subcellular localization of Polκ.
    • Fig. S3. DNA damage is not associated with the subcellular shift of Polκ.
    • Fig. S4. Cell cycle inhibition is not responsible for the shift in the subcellular localization of Polκ.
    • Fig. S5. MAPK inhibition decreases the abundance of p-S6.
    • Fig. S6. BRAF inhibition induces ER stress, and other inducers of ER stress also change the subcellular localization of Polκ.
    • Fig. S7. The effects on POLK mRNA levels are driver gene specific across tumor types.
    • Fig. S8. Glucose starvation also modulates the subcellular localization of Polκ.
    • Fig. S9. Inhibiting importin-β does not prevent the nuclear shift of Polκ.
    • Fig. S10. Inhibition of the proteosome does not have a differential effect on the protein level of Polκ after treatment with DMSO or PLX4032.
    • Table S1. qRT-PCR primers used in the study.
    • Table S2. sgRNAs and validation primers used for MMR and Polκ KO.
    • Table S3. RNA-seq and GSEA analysis on Polκ-CRISPR A375 cells.

    [Download PDF]

    Other Supplementary Material for this manuscript includes the following:

    • Data file S1 (Microsoft Excel format). Differentially expressed genes between A375 cells with POLK sgRNA versus control sgRNA.

Stay Connected to Science Signaling

Navigate This Article