Alpha-synuclein regulates nucleolar DNA double-strand break repair in melanoma

Abstract

Although an increased risk of the skin cancer melanoma in people with Parkinson's disease (PD) has been shown in multiple studies, the mechanisms involved are poorly understood, but increased expression of the PD-associated protein alpha-synuclein (αSyn) in melanoma cells may be important. Our previous work suggests that αSyn can facilitate DNA double-strand break (DSB) repair, promoting genomic stability. We now show that αSyn is preferentially enriched within the nucleolus in melanoma, where it colocalizes with DNA damage markers and DSBs. Inducing DSBs specifically within nucleolar ribosomal DNA (rDNA) increases αSyn levels near sites of damage. αSyn knockout increases DNA damage within the nucleolus at baseline, after specific rDNA DSB induction, and prolongs the rate of recovery from this induced damage. αSyn is important downstream of ataxia-telangiectasia-mutated signaling to facilitate MDC1-mediated 53BP1 recruitment to DSBs, reducing micronuclei formation and promoting cellular proliferation, migration, and invasion.

Fig. 1.. αSyn localizes to the GC of the nucleolus in melanoma cells.

(A) Brown 3,3′-diaminobemzidine (DAB) pigment labels αSyn antibody (LB509) staining enriched in a subset of metastatic melanoma cells (arrowheads). Scale bar, 10 μm. Table showing demographic information and staining pattern detected. −, no detectable staining; +, light staining; ++, moderate staining. (B) Formalin-fixed paraffin-embedded (FFPE) wide-excision patient biopsy was stained for αSyn (LB509), nuclear marker (RPA32), and 4′,6-diamidino-2-phenylindole (DAPI). The sample was imaged on the Zeiss 980 confocal microscope with Airyscan. White circles indicate RPA32 foci, and arrowheads denote LB509 foci. (C and D) SK-Mel28 cells were stained for αSyn (Syn1), nucleolar markers (nucleophosmin, treacle, and nucleostemin), and DAPI. Cells were imaged on the Zeiss 980 confocal microscope with Airyscan, and colocalization was analyzed in Imaris software. Error bars represent SEM with quantification from three biological replicates (total n = 30 to 36 nuclei per condition). ****P < 0.0001 by t test. (E) SK-Mel28 cells were stained for αSyn (Syn1), fibrillar center (FC) marker (RPA194), dense fibrillar component marker (fibrillarin), granular component (GC) marker (nucleophosmin), and DAPI. Cells were imaged on the Zeiss 980 confocal microscope with Airyscan oversampling and Joint Deconvolution and Channel Alignment post-processing. Three-dimensional (3D) renderings were produced using Imaris software. (F) SK-Mel28 cell pellets were fixed for 2 hours at RT. Thin sections were cut on an ultramicrotome (EM UC7), stained for αSyn (MJFR1, 1:75) (12-nm colloidal gold particles), and observed by a JEOL 1400 transmission electron microscope. Blue labeling denotes nuclear membrane. Red labeling denotes the outline of the nucleolus. Arrows point to representative MJFR1 staining. Quantification of three biological replicates (total n = 28 to 35 nucleoli per genotype). ****P < 0.0001 by t test. Error bars denote SEM. The different αSyn antibodies used in [(A) to (E)] were chosen after tissue-type and methodology optimization to determine which produced the most distinct and specific nuclear labeling.

Fig. 2.. αSyn interacts with γH2AX in the nucleolus and knocking out αSyn leads to increased ATM/ATR-driven H2AX phosphorylation.

(A and C) SK-Mel28 cells were stained for αSyn (Syn1), DSB marker (γH2AX), nucleolar mask (nucleostemin), and DAPI. Cells were imaged on the Zeiss 980 confocal microscope with Airyscan, and data were analyzed using Fiji (intensity) or Imaris (colocalization). Quantification represents three biological replicates (total n = 71 to 172 nuclei or nucleoli per condition/genotype for intensity or n = 30 nuclei per condition for colocalization). *P < 0.05, ***P < 0.001, and ****P < 0.0001 by t test or ANOVA. Error bars denote SEM for all graphs. N, nucleus; Nu, nucleolus. Same γH2AX quantification of control cells between [(A) and (C)]. (B) SK-Mel28 cells were seeded on poly-d-lysine (PDL)–coated coverslips and then fixed in 4% paraformaldehyde. PLA (Duolink) was completed using antibodies against Syn1 and γH2AX. Cells were imaged on the Zeiss 980 confocal microscope, and the number of foci per nucleus was measured using CellProfiler while masking for the nucleus using DAPI. Each figure shows representative images and quantification from three biological replicates (total n = 283 to 306 nuclei per genotype). ****P < 0.0001 by t test. Error bars denote SEM. (D) SK-Mel28 cells were lysed, and a nuclear fractionation was performed. Nuclear protein was run out on SDS–polyacrylamide gel electrophoresis (PAGE) and probed for γH2AX and total protein. Western blots were imaged on LI-COR CLx imager. ***P < 0.001 by ANOVA. Error bars denote SEM. Quantification from four biological replicates. (E and F) SK-Mel28 cells were treated with dimethyl sulfoxide (DMSO), KU-60019 (10 μM), VE-822 (0.1 μM), or NU-7441 (1 μM) for 24 hours. Cells were fixed and stained for Syn1, γH2AX, nucleostemin, and DAPI. Mean intensity of γH2AX signal within DAPI and nucleostemin masks analyzed using Fiji. **P < 0.01, ***P < 0.001, and ****P < 0.0001 by ANOVA. Error bars denote SEM. Quantification from three biological replicates (total n = 517 to 615 nuclei or nucleoli per condition).

Fig. 3.. αSyn is important in bleomycin and IR-induced DDR pathways.

(A and D) SK-Mel28 cells were treated with DMSO or bleomycin (100 μg/ml) for 1 hour and stained for Syn1, γH2AX, and DAPI. Mean intensity of γH2AX signal within DAPI and/or nucleostemin masks were analyzed using Fiji. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 by ANOVA. Error bars denote SEM. N, nucleus; Nu, nucleolus. Quantification from three biological replicates (total n = 45 to 757 nuclei or nucleoli per condition). Same γH2AX quantification between graphs in (A) with control cells. (B) SK-Mel28 cells were treated with DMSO or bleomycin (100 μg/ml) and lysed, and a nuclear fractionation was performed. Nuclear protein was run out on SDS-PAGE and probed for γH2AX and total protein. *P < 0.05 and ****P < 0.0001. Error bars denote SEM. Quantification from four biological replicates. (C) SK-Mel28 cells were treated with DMSO or bleomycin (100 μg/ml). Cells were processed according to the In-Cell Western manufacturer instructions and stained for γH2AX (800) and CellTag (700). **P < 0.01, ***P < 0.001, and ****P < 0.0001 by ANOVA. Error bars denote SEM. Quantification from three biological replicates (with four technical replicates per biological replicate). (E) SK-Mel28 cells were treated with DMSO or bleomycin (100 μg/ml). DNA damage in situ ligation followed by proximity ligation assay (DI-PLA) was completed. Number of foci per nucleus was measured within nuclear masking. Quantification from three biological replicates (total n = 171 to 231 nuclei per condition). ****P < 0.0001 by ANOVA. Error bars denote SEM. (F) SK-Mel28 cells were treated with 0Gy or 5Gy of x-ray ionizing radiation (IR) followed by a 20-min recovery period and stained for Syn1, γH2AX, NST, and DAPI. Mean intensity of γH2AX signal within DAPI masks were analyzed using Fiji. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 by ANOVA. Error bars denote SEM. Quantification from three biological replicates (total n = 358 to 388 nuclei per condition).

Fig. 4.. Inducing rDNA DSBs increases αSyn levels and localization to sites of damage and αSyn KO significantly increases γH2AX.

(A) SK-Mel28 cells were transfected with WT and H98A I-PpoI mRNA and stained for Syn1, γH2AX, nucleostemin, and DAPI. Quantification from three biological replicates (total n = 1867 to 2823 nuclei per condition). *P < 0.05 and ****P < 0.0001 by ANOVA. Error bars denote SEM. (B) SK-Mel28 cells were transfected with WT and H98A I-PpoI mRNA, cells were lysed, and a nuclear fractionation was performed. Nuclear protein was run out on SDS-PAGE and probed for γH2AX and total protein. Quantification from four to six biological replicates. *P < 0.05, **P < 0.01, and ****P < 0.0001 by ANOVA. Error bars denote SEM. (C) SK-Mel28 cells were transfected with Cas9 and guide RNAs (gRNAs). Cells were stained for γH2AX, treacle, RNA polymerase (Pol) II, and DAPI. γH2AX intensity was measured within a radius of treacle-identified nucleolar cap. Quantification from three biological replicates (total n = 126 nucleoli per condition). **P < 0.01 and ****P < 0.0001 by ANOVA. Error bars denote SEM. (D) SK-Mel28 cells were transfected with WT and H98A I-PpoI mRNA, and the DI-PL was completed. A treacle co-stain was included to identify nucleolar cap formation. The number of foci per nucleus was measured with DAPI masking. Each figure shows representative images and quantification from three biological replicates (total n = 104 to 119 nuclei per condition). ****P < 0.0001 by ANOVA. Error bars denote SEM. (E) SK-Mel28 control cells expressing synuclein-GFP or empty vector (EV)–GFP. Yellow arrows show targeting of laser-induced damage (LID) pulse in the nucleolus. Baseline (t = −6 s) and after LID (t = 6 and 23 s) images show accumulation of synuclein-GFP at DNA damage site. Data from graph show the calculated enrichment ratio at LID site (compared to an adjacent site in the nucleolus). Quantification from >20 cells over two biological replicates. ****P < 0.0001 by t test. Error bars denote SEM. TPMT, transmitted photomultiplier tube; LID/UR, laser-induced damage/undamaged region.

Fig. 5.. αSyn KO significantly impairs recovery of rDNA damage downstream of ATM signaling.

(A) SK-Mel28 cells were seeded in a black-welled PDL-coated 96-well plate and treated with DMSO, KU-60019 (10 μM), VE-822 (0.1 μM), or NU-7441 (1 μM) for 24 hours. Cells were then treated with WT and H98A I-PpoI mRNA for 6 hours in the presence of the inhibitors. Cells were processed according to the In-Cell Western manufacturer instructions and stained for γH2AX (800) and CellTag (700). Plates were imaged on the LI-COR CLx. *P < 0.05, **P < 0.01, and ***P < 0.001 by ANOVA. Error bars denote SEM. Quantification from six biological replicates (with three technical replicates per biological replicate). Normalization to control cells transfected with I-PpoI H98A and treated with DMSO. (B) SK-Mel28 cells (control/KO/rescue) were seeded on poly-l-lysine (PLL)–coated coverslips and then transfected with WT and H98A I-PpoI mRNA. At the indicated time point post-transfection, cells were fixed and stained for Syn1, γH2AX, nucleostemin, and DAPI. Cells were imaged on the Zeiss 980 confocal microscope, and data were analyzed using Fiji. Quantification from three biological replicates (total n = 1909 to 3194 nuclei per condition). Statistical labeling denotes significance between control and KO cell lines, ****P < 0.0001 by ANOVA. (C) SK-Mel28 cells (control/KO/rescue) were transfected with WT and H98A I-PpoI mRNA. After 24 hours, cells were lysed, and a nuclear fractionation was performed. Nuclear protein was run out on SDS-PAGE and probed for γH2AX and total protein. Quantification from five biological replicates. *P < 0.05, **P < 0.01, and ***P < 0.001 by ANOVA. Error bars denote SEM.

Fig. 6.. αSyn localizes to DSB-induced nucleolar caps and is important for nucleolar cap recovery.

(A) SK-Mel28 cells were seeded on PLL-coated coverslips and then treated with WT I-PpoI mRNA for 6 hours or actinomycin D (Act D; 100 ng/ml) for 1 hour before fixation. Cells were stained for Syn1, nucleolar caps marker (UBF), nucleostemin, and DAPI. Cells were imaged on the Zeiss 980 confocal microscope with Airyscan oversampling and Joint Deconvolution and Channel Alignment post-processing. 3D renderings and distance from cap analysis were produced using Imaris software. Quantification from five biological replicates. Error bars denote SEM. Statistical significance was calculated via nonlinear regression (99% confidence interval). (B) SK-Mel28 cells (control/KO/rescue) were seeded on PDL-coated eight-well Ibidi plates. Cells were transfected with 800 ng of GFP-treacle using Lipofectamine 3000. Twenty-four hours post-transfection, cells were treated with WT I-PpoI mRNA. Four hours after treatment, live-cell imaging was performed using the Zeiss Celldiscoverer 7 and imaged for 15 hours. Quantification from five biological replicates (>200 cells were analyzed per genotype per experiment). *P < 0.05 and ****P < 0.0001 by two-way ANOVA or Mantel-Cox test. AU, arbitrary units.

Fig. 7.. αSyn KO leads to reduced recruitment of MDC1 and 53BP1 to nucleolar caps and is associated with impaired cellular growth and dysregulated expression of transcripts associated with DNA repair pathways.

(A to C) SK-Mel28 cells were transfected with WT I-PpoI mRNA. Zero, 6, or 24 hours post-transfection, cells were stained for Syn1, UBF, 53BP1, and DAPI; or Syn1, treacle, BRCA1, and DAPI; or UBF, MDC1, and DAPI. Data were analyzed using Fiji. Micronuclei were hand counted with the experimenter blinded to cell condition. Quantification from three to four biological replicates (total n = 84 to 956 nuclei or cells per condition). *P < 0.05, **P < 0.01, and ****P < 0.0001 by ANOVA. Error bars denote SEM. (D) SK-Mel28 cells were plated on Matrigel-coated ImageLock plates. Graphs represent temporal progression of total confluence (proliferation), or wound closure using relative wound density as the metric to measure migration or invasion. Data represent time course of means of each cell line among three biological replicates (with 16 technical replicates per experiment). *95% confidence interval (CI), ***99% CI by nonlinear regression (migration) or simple linear regression (proliferation and invasion) analysis. n.s., not significant. (E) Total RNA was extracted from SK-Mel28 control and KO cells and sent for RNA-seq (n = 3 biological replicates per condition). Differentially expressed gene transcripts in KO cells compared to control were identified. These were cross-referenced to over 500 nucleolar-specific genes. Sixty-four genes were identified and plotted on a volcano plot using P value and fold change. All transcripts, greater than 1 SD and less than −1 SD, are plotted. Red labeling indicates RT-PCR validated genes. (F) All significant up-regulated and down-regulated transcripts underwent gene ontology analysis. Gene sets found to be significant after Benjamini-Hochberg multiple testing correction (P adj. < 0.05) are marked with an asterisk. BMP, bone morphogenetic protein.

Fig. 8.. Potential molecular mechanism by which αSyn facilitates nucleolar DDR and the downstream cellular consequences.