Abstract
The majority of the heredity of melanoma remains unexplained, however inherited copy number changes have not yet been systematically studied. The genetic environment is highly relevant to treatment stratification, and new gene discovery is therefore desirable. Using an unbiased whole genome screening approach for copy number we identify here a novel melanoma predisposing factor, familial duplications of gene PPP2R3B, encoding a regulatory unit of critical phosphatase PP2A. Significant correlation between expression of PPP2R3B in tumour tissue and survival in a large melanoma cohort was confirmed, and associated with a non-immunological expression profile. Mechanistically, construction and extensive characterization of a stable, inducible cellular model for PPP2R3B overexpression revealed induction of pigment cell switching towards proliferation and away from migration. Importantly, this was independent of the known microphthalmia-associated transcription factor (MITF)-controlled pigment cell phenotype switch, and was instead driven by uncharacterised gene C21orf91. Bioinformatic studies point to C21orf91as a novel target of MITF, and therefore a potential hub in the control of phenotype switching in melanoma. This study identifies novel germline copy number variants in PPP2R3B predisposing to melanocytic neoplasia, and uncovers a new potential therapeutic target C21orf91 in the control of pigment cell proliferation.
Introduction
Melanoma remains a major cause of morbidity and mortality. The majority of the heredity of melanoma remains unexplained, suggesting that multiple low prevalence variants are involved. Germline loss of function variants in CDKN2A in 2% of melanoma cases are the commonest known genetic predisposer1–3, with more recent discoveries of predisposing variants in genes BAP14 and POT15 at a similar frequency. With CDKN2A, affected individuals and their families are screened for melanoma under current guidelines6. Identification of new familial predisposing genes to melanocytic neoplasia is therefore desirable to identify families at higher risk of melanoma, however it also offers the potential to inform the understanding of the condition at molecular level. Whole genome germline copy number analysis has not yet been undertaken in melanoma cohorts. We sought to identify novel predisposing copy number changes for melanocytic neoplasia, using a rare disease approach.
Congenital melanocytic naevi (CMN; MIM 137550) is a rare congenital mosaic disorder of large and multiple moles, which predisposes affected individuals to melanoma. It is a valuable UV-independent model for the development of melanocytic naevi and melanoma, as the naevi develop in utero, and the melanoma most commonly arises post-natally but within the central nervous system7. It is also a good genetic model, as the causative somatic mutations are in classic melanoma hotspots in NRAS in 70%8,9, and BRAF in 7%9,10, with the remainder as yet unknown. Importantly for this study, despite the sporadic somatic nature of the disease, one third of cases have a first or second degree family history of CMN in the largest published cohort11,12, strongly suggesting predisposing germline genotypes to NRAS/BRAF somatic mutation in affected families. Previous work has delineated an enrichment of melanocortin-1-receptor (MC1R) variants in individuals with CMN12, again mirroring the genetic background of sporadic melanoma and underlining that this genetic model is UVR independent. We hypothesised that new predisposing copy number variants found via this rare disorder could predispose to melanoma in the normal population. Germline copy number in the CMN cohort was analysed using an unbiased whole genome approach, and findings validated in an independent adult sporadic melanoma cohort.
Results
Germline duplications involving PPP2R3B are found at increased frequency in individuals with melanocytic neoplasia
Using whole genome array comparative genomic hybridisation (CGH) of leukocyte DNA, duplications of Xpter were identified in 3/24 (12.5%) patients with CMN (Figure 1a), where only gene PPP2R3B was common to all three. No other novel copy number variant was seen in more than one patient. Control data from paediatric patients with other phenotypes from the same diagnostic testing facility identified duplications of this gene (with or without involvement of the two telomeric genes but not extending centromeric) in 13/4800, or 0.271%. Population data from normal individuals from the Database of Genomic Variants13 identified similar duplications in only 1/36,000, or 0.003%14, and 0.5% in nearly 7000 individuals in the MSSNG autism study15,16, with no difference between cases and controls (personal communication). Custom-designed TaqMan® copy number assays (Thermo Fisher Scientific, USA) were insufficiently robust for validation due to highly repetitive DNA sequence in this area. Custom-designed multiple ligation-dependent probe amplification (MLPA, MRC Holland) and targeted next generation sequencing (NGS) of the four most telomeric genes on Xp confirmed the array findings, detecting an overall prevalence of duplications involving PPP2R3B but not downstream SHOX in 5% of a cohort of 125 individuals with CMN (Figure 1c,d). NGS of leukocyte DNA from an adult melanoma cohort identified the same germline duplications in 4/178 (2.2%) (Figure 1b), confirming this copy number variant as enriched in populations with melanocytic neoplasia. Regional similarity search across Xp22.33 revealed three segmental duplications 5’ of PPP2R3B, and a high density of SINE and LINE repeats, but no segmental duplications between PPP2R3B and SHOX (Figure 1e). Sanger sequencing of leukocyte DNA from 48 CMN patients and 48 controls did not detect any undescribed variants or haplotype differences (data not shown).
Germline duplications involving PPP2R3B are found at increased frequency in individuals with melanocytic neoplasia
(a) Schematic of Xp22.33 region demonstrating the location of three novel duplications (yellow) found in 24 congenital melanocytic naevus (CMN) patients using whole genome array CGH of leukocyte DNA, with one identical parental duplication demonstrating inheritance. Described copy number variants from that region are shown below, duplications in blue, deletions in red, with each bar representing a single publication. The publication representing a duplication involving PPP2R3B described a single variant in a cohort of approximately 36,000 (asterisk, and see text for details), confirming that the CMN duplications are rare in the normal population. (b) PPP2R3B duplications in a UK non-syndromic melanoma cohort (n=179), and CMN cohort (n=8) shown by targeted next generation sequencing of PPP2R3B, in addition to the two telomeric genes (GTPBP6 and PLCXD1) and the next centromeric gene (SHOX). Data represent the ratio of corrected read depth (see text for details) across the whole of PPP2R3B with respect to the ratio across the whole of SHOX. Each bar represents an individual patient. PPP2R3B duplications called are shown in light blue: validation of the array CGH findings in the three CMN patients are clustered to the right of the figure, and new duplications in the melanoma cohort in the rest of the figure (n=4, 2.2%).
Validation of PPP2R3B duplications detected by array CGH
(c-d) Custom-designed MLPA® ratio plots validating copy number measurement of PPP2R3B (4 probes) and the two telomeric genes GTPBP6 and PLCXD1 (one probe each), to the left of each figure and less than 150bp in length; control probes of greater than 160kb targeting genes of known normal copy number across the genome are shown to the right. Example of normal copy number for all genes (c), and of a duplication of PPP2R3B and GTPBP6 and PLCXD1 in a CMN patient (d).
Low-Copy Repeats at Xp22.33
(e) The upper panel depicts a regional similarity search across Xp22.33 with YASS software (http://bioinfo.cristal.univ-lille.fr/yass/index.php) both forwards (green) and backwards (red) revealing three segmental duplications (LCRA, LCRB and LCRC) 5’ of PPP2R3B and a high-density of SINE and LINE repeats. No segmental duplications are detected 3’ to PPP2R3B before SHOX. The assembly gaps (red), local genes (purple) and the homology region (orange) with the Y chromosome are indicated.
Germline duplications of PPP2R3B lead to increased expression of PR70 in congenital melanocytic naevi
Owing to other potential genetic confounders within malignant tissue such as loss of Xp, effects of germline duplication on tissue expression in vivo was visualised in CMN tissue, known to have little somatic copy number variation17. Germline duplications were clearly associated with increased expression of PR70 in CMN tissue on immunohistochemistry compared to controls (Figure 2a,b).
(a-b) Germline duplications of PPP2R3B lead to increased protein expression within congenital melanocytic naevus tissue, compared to that of normal copy number controls. Immunohistochemical staining of formalin fixed paraffin embedded CMN tissue demonstrates negative or very low intensity PR70 staining in a patient with confirmed normal copy number of PPP2R3B (a), and widespread moderate intensity staining throughout the cytoplasm of naevus cells in a patient with a confirmed germline PPP2R3B duplication (b), both shown at 20X. Red arrow indicates nest of naevus cells with increased DAB staining.
(c-d) Increased PPP2R3B expression in melanoma tissue is correlated with improved melanoma-specific survival. (c) Kaplan-Meier curve generated from transcriptomic data from 703 FFPE melanoma tumours from the Leeds Melanoma Cohort, hazard ratio (HR) =0.69, (95% confidence interval 0.50-0.88), p = 0.004. The effect remains significant after adjusting for age, sex, AJCC (American joint committee for cancer) stage, vascular invasion, site, BRAF/NRAS mutation status and tumour invading lymphocytes (TILs). (d) Log intensity distribution of PPP2R3B DASL probe (ILMN_1689720) is close to a normal distribution.
(e) Improved melanoma specific survival observed with increased expression of PPP2R3B appears not to be immune mediated. Tumour expression of PPP2R3B correlates with expression of a large number of other genes in the genome: 596 positively correlated at FDR<0.05 with regression coefficient>0.20; 731 negatively correlated at FDR<0.05 with a regression coefficient<-0.2 (e). The genes positively correlated with PPP2R3B are predominantly enriched in non-immune pathways, consistent with the lack of association between PPP2R3B expression and TILs or any specific immune cell score (Figure S1). The genes negatively correlated with PPP2R3B expression are predominantly enriched in immune pathways.
Expression of PR70 is significantly associated with prolonged melanoma-specific survival (MSS)
Increased tissue expression of PPP2R3B in a larger melanoma cohort, although potentially influenced by various somatic changes in malignancy, was significantly associated with prolonged melanoma-specific survival (MSS) (Figure 2c,d). This effect remained significant after correction for known associations with survival, namely age, sex, AJCC (American joint committee for cancer) stage, vascular invasion, site, BRAF/NRAS mutation status and tumour invading lymphocytes (TILs). Unlike many known expression-survival associations in melanoma, transcriptomic pathway analysis demonstrated that this protective effect was not immune pathway-mediated (Figure 2e, Tables S1,S2), suggesting an alternative mechanism of action.
Creation of a stable inducible overexpression system to study PPP2R3B overexpression
Detailed characterisation of the effects and mechanisms of PPP2R3B overexpression were modelled by creation of a stable inducible over-expression system in two melanoma cell lines SKMEL2 and SKMEL30 (Figure 3a-c). Induction robustly and reproducibly led to PPP2R3B mRNA overexpression, and protein product PR70 overexpression.
(a-c) Generation of stable inducible overexpression model for PPP2R3B in melanoma cell lines SKMEL2 and SKMEL30. Diagrammatic summary of stable inducible PPP2R3B overexpression system (SKMEL2 pTRIPZ PPP2R3B and SKMEL30 pTRIPZ PPP2R3B), and generation of samples for reverse phase protein arrays (RPPA) and RNA sequencing (a). Validation of PPP2R3B overexpression in samples for Reverse Phase Protein Array and RNAseq, (b) qRT-PCR demonstrating increased PPP2R3B mRNA in induced cell lines at 4-6h and 16-18h hours (relative fold change in PPP2R3B expression, standardised to GAPDH, mean + SD of samples in quadruplicate), and, (c) Western blot confirming PR70 overexpression in induced cell lines at 4-6h and 16-18h hours with Vinculin loading control.
(d-j) Overexpression of PPP2R3B increases proliferation in melanoma cell lines. Increased proliferation following PPP2R3B overexpression using WST1 proliferation assay at 6 hours (d), and BrdU assay at 24 hours (e) (mean absorbance of colourmetric assay of eight replicates shown with error bar), and, by IncuCyte® cell count proliferation assay over 100 hours, measuring confluence (%) versus time (hours) in SKMEL2 pTRIPZ PPP2R3B (f) and SKMEL30 pTRIPZ PPP2R3B (g) (mean confluence at each time point of eight replicates shown with error bars). The mean confluence in each cell line is shown at timepoints of 50h (h) and 100h (i).
(j-l) Overexpression of PPP2R3B decreases cellular migration in melanoma cell lines. Scratchwound assay in SKMEL2 pTRIPZ PPP2R3B (j) and SKMEL30 pTRIPZ PPP2R3B (k) leads to decreased relative wound confluence (%) versus time (hours) compared to non-induced controls. Mean relative wound confluence in each cell line shown at 48h (l) (mean relative wound confluence at each time point of eight replicates shown with error bars). Statistically significant values are indicated by a single asterisk (p<0.05), a double asterisk p<0.01, a triple asterisk (p<0.001) or a quadruple asterisk (p<0.0001). All p-values for Students t-test, Prism v7.0 (Graphpad).
PPP2R3B overexpression leads to increased cellular proliferation and decreased migration in a 2D melanoma cell model
PPP2R3B overexpression led to clear pigment cell phenotype switching, most marked from 48 hours post induction onwards. This was measurable via significantly increased cellular proliferation by several alternative established methods (Figure 3d-i), and decreased migration in scratch assays couple with IncuCyte® monitoring (Figure 3j-l).
PPP2R3B overexpression does not significantly alter known melanoma signaling pathway activation
RNA sequencing pathway enrichment analysis identified suppression of the unfolded protein response and endoplasmic reticulum (ER) protein folding after induction of PPP2R3B (Table S3). Signaling pathway characterisation of 302 proteins using reverse phase protein arrays (RPPA, MD Anderson Core) pre- and post-induction of PPP2R3B demonstrated enrichment for mammalian target of rapamycin (mTOR) and hypoxia-induced factor 1 (HIF-1) signaling pathways, with prominent biological signatures of response to heat and stress (Figure 4a-d, Table S4). Functional protein association network analysis18 of differentially expressed proteins from RNA sequencing and RPPA (non-adjusted) created a strong network centred around these same pathways (Figure 4l). Immunoblotting provided validation of significantly decreased phosphorylation of AKT at 6-8 hours, however overall no dramatic effects on known melanoma signaling pathways were demonstrated (Figure 4e-k). Activation of CDC6, a known direct target of PR7019, was inconsistent across cell lines (Figure 4k). Increasing molar concentrations of PR70 up to a 1:1 ratio with that of the core PP2A enzyme increased PP2A activity towards its specific substrate pCDC6 in another cellular model, however further increases in concentration reduced phosphatase activity (Figure S3). PR70 was shown to be highly efficient at binding to the PP2A core enzyme when competing with another regulatory subunit B’γ1, and overexpression of V5-tagged PR70 in a mammalian cell line (C6) did not lead to free PR70 (Figure S4), suggesting either competitive binding of other PP2A holoenzymes or direct interactions with other effectors.
PPP2R3B overexpression affects mTOR/p70S6K1 and HIF-1 signalling pathways
(a-b) Heat map of protein expression observed by RPPA following overexpression of PPP2R3B in SKMEL2 pTRIPZ PPP2R3B (a) and SKMEL30 pTRIPZ PPP2R3B (b), demonstrating low background activity as expected from a controlled cellular model. (c-d) Volcano plot of log fold change in protein expression versus p-value for differentially expressed proteins common to both cell lines following PPP2R3B overexpression. Unadjusted P-values < 0.05 are shown in red at 6h (c) and 16h (d).
Overexpression of PPP2R3B dephosphorylates AKT, but does not consistently affect CDC6 activation. Western blot of lysate from uninduced and induced SKMEL2 pTRIPZ PPP2R3B and SKMEL30 pTRIPZ cells, for ERK, phospho-ERK (e), S6K1 and quantification (f, g), AKT and phospho-AKT (S308 and T473) (h), quantification of ratio of total to phospho-AKT (i), demonstrates PPP2R3B overexpression decreases phosphyorylation of AKT at 6-8 hours, but does not lead to strong activation of known melanoma signalling pathways. Western blot of CDC6 and phospo-CDC6 (j) and quantification of ratio of total to phospho-CDC6 (k), show PPP2R3B overexpression does not consistently alter CDC6 activation across the cell lines in this system, as confirmed by direct phosphatase activity measurements in another cellular model demonstrating that the interaction between PR70 and CDC6 may be dose-dependent (Figure S3).
Strong network of known protein interactions implicated following PPP2R3B overexpression identifies C21orf91 as a key new gene in these pathways.
Functional protein association network using STRING15 software demonstrating strong network of known signalling pathway interactions between all significant pre-adjustment RPPA hits and all significant RNAseq hits present in both cell lines following PPP2R3B overexpression at 16 hours (l). C21orf91 was the most differentially expressed gene in each cell line but is not currently known to be connected to these pathways.
PPP2R3B overexpression leads to a significant and sustained rise in expression of the previously uncharacterized gene C21orf91, independent of MITF
Given the lack of clear signaling pathway activation in the presence of a clear proliferative phenotype, an alternative candidate mediator was sought by unbiased methods. In this regard, relatively unknown gene C21orf91 (Refseq Gene ID:54149) was identified by RNA sequencing as the most significantly differentially-expressed gene in both cell lines, and validated at mRNA and protein levels (Figure 5, S1, Table S5). Crucially, knockdown of C21orf91 by siRNA reversed the increased proliferation associated with induction of PPP2R3B expression (Figure 6a-d), firmly tying C21orf91 to the pro-proliferative phenotype. Importantly, this increase was independent of microphthalmia-associated transcription factor (MITF), master regulator of melanocyte transcription and phenotype switching (Figure S2), as witnessed by the lack of MITF overexpression at mRNA and protein levels upon induction of PPP2R3B.
PPP2R3B overexpression leads to significant and sustained rise in expression of gene C21orf91.
C21orf91 was the most differentially expressed gene on PPP2R3B induction common to both cell lines and at both time points by RNA sequencing (Table S1), other than PPP2R3B itself. (a) Heat map from pathway signature analysis of RNAseq data at 6-8h and 16-20h, focusing on pro-proliferative anti-invasive melanoma signature genes38, clearly demonstrating increased expression of C21orf91 following PPP2R3B overexpression observed in both cell lines, at 6 and 16 hours. Validation of significantly increased C21orf91 expression following PPP2R3B overexpression at 6h and 16h hours, shown by (b) qRT-PCR relative fold change standardised to GAPDH, mean + SD of samples in quadruplicate and (c-d) Western blots with loading control vinculin, and (d) quantification of fold change observed in C21orf91 following overexpression of PR70.
Knockdown of C21orf91 rescues increased proliferation associated with PPP2R3 overexpression
IncuCyte® proliferation assay confluence (%) versus time (hours) for PPP2R3B-induced SKMEL2 pTRIPZ PPP2R3B, with and without knock-down of C21orf91, by siRNA1 (a), and siRNA2 (b), with means taken at 50h (c), and 100h (d) (mean confluence at each time point of eight replicates shown with error bar, standardised to uninduced control). Single asterisk p<0.05, double asterisk p<0.01 (Students t-test, Prism v7.0 (Graphpad)).
C21orf91 and MITF expression are significantly positively correlated in melanoma, independent of PPP2R3B
Positive correlation demonstrated in independent transcriptomic data from a cohort of 24 human derived melanoma cell lines, Spearman Rho 0.18 (p = 2 × 10−6) (e) and in human melanoma tissue (Spearman Rho 0.8 (95% CI 0.6-0.9, p<0.0001) (g). No correlation between expression of C21orf91 and PPP2R3B in either cohort (f) and (h), points to increased expression of C21orf91 as a new hub in melanocytic proliferation, not only a target of PPP2R3B.
Increased expression of C21orf91 is associated with genetic dependency on MITF in melanoma cell lines
CRISPR-Cas9 genome-scale knockout of MITF in melanoma cell lines (n=30) reveals increased expression of C21orf91 in cells with greater MITF dependency (i), suggesting that C21orf91 is downstream of MITF. Dependency score as described in https://depmap.org.
PR70 and C21orf91 are expressed throughout the cytoplasm with increased expression of C21orf91 in dividing cells
(i) Immunocytochemistry of SKMEL30 pTRIPZ PPP2R3B in uninduced cells at 10X (A) and induced cells at 10X (B) and 20X (C) confirms increased PR70 and C21orf91 expression throughout the cytoplasm following induction of PPP2R3B. PR70 is stained with Alex Fluor® 488 (green) secondary antibody, and nuclei stained with Hoescht (blue) and in (C) Phalloidin (actin) is visualised with a conjugated Alex Fluor® 647 (far red) antibody. C21orf91 expression is increased in dividing cells, 10X (D), 20X (E), 40X (F), stained with Alex Fluor® 488 (green) secondary antibody, and Hoescht nuclear stain (blue).
C21orf91 expression is positively correlated with MITF expression in melanoma
Given that MITF is the known master regulator of pro-proliferative phenotype switching in melanoma, and given that C21orf91 was not operating via MITF, we considered whether MITF could instead be operating via C21orf91. C21orf91 and MITF expression were indeed found to be significantly positively correlated in independent transcriptomic datasets from both melanoma cell lines and the melanoma patient cohort (Figure e-h), implying at least a key role for C21orf91 in the pro-proliferative state of melanoma, and potentially that MITF can operate via C21orf91. MITF dependency score and C21orf91 expression in melanoma cell lines were also found to be significantly associated in melanoma cell lines (i), data extrapolated from the Cancer dependency Map (Broad Institute) 20–23. PR70 subcellular localization by confocal microscopy was pan-cytoplasmic, including but not restricted to endoplasmic reticulum as previously suggested24, and increased but unaltered in distribution by overexpression (Figure 6i). C21orf91 expression was also demonstrated throughout the cytoplasm, and both PR70 and C21orf91 were noticeably increased in dividing cells (Figure 6i).
Discussion
Copy number in the genome has been less systematically explored than sequence variation due to technical constraints25, with copy number variation enriched in areas of low genome mappability26, however copy number variants (CNVs) are prevalent in genes for cell communication and RAS-pathway signaling, including serine threonine kinases and phosphatases27. Using a rare disease cohort, we identify here new germline duplications in the pseudoautosomal region 1 of the X chromosome which predispose to melanocytic neoplasia. The minimal common area included only gene PPP2R3B, which encodes PR70, ubiquitously expressed in the cytoplasm, and one of the ß” family of regulatory units of the critical phosphatase and regulator of the cell cycle PP2A28,29. PP2A is a heterotrimeric holoenzyme consisting of a structural A subunit, a catalytic C subunit, and a regulatory B subunit28,30,31. The numerous non-homologous regulatory subunits are classified into B, B’, B’’ and B’’’ subfamilies, implicated in control of enzyme activity and substrate specificity28,32. PP2A operates via key effector pathways RAS/MAPK, Wnt and AKT/mTOR28,29. As such, PP2A activity is intimately involved in malignancy and response to treatments33, and is a major focus of potential therapeutics33–36.
Our data demonstrate that PPP2R3B overexpression promotes proliferation of melanocytes, which could explain the predisposition to the development of a clinically-apparent melanocytic naevus or melanoma in the context of a somatic mutation in a melanocyte. Alternatively, the pro-proliferative germline environment could itself predispose to somatic mutation in the skin, via increased cell division or alteration of cell cycle regulation and the associated effects on DNA repair. Interestingly, in parallel, our data demonstrate clearly that increased PPP2R3B expression simultaneously improves survival in melanoma, with a similar significant protective effect of PPP2R3B expression seen in urothelial cancer and pancreatic cancer datasets from the TCGA database. In the melanoma cohort studied here, this protection is mediated via a non-immunological mechanism, and could potentially operate via phenotype switching towards proliferation and away from migratory potential. In support of this data, a comprehensive study of the role of PPP2R3B in melanoma at tumour as opposed to germline level found the region to be copy-number sensitive, with loss of the inactivated X in females and decreased expression in males linked to decreased distant metastasis-free survival. The authors proposed that the copy number sensitivity of this locus could explain the gender differences in melanoma incidence and survival37.
Pigment cell phenotype switching is classically controlled by a reciprocal relationship between MITF and POU3F238–40, however here was MITF-independent, and appears instead to be driven by the relatively unknown gene C21orf91. Although this mechanism could involve the AKT/mTOR/pS6K pathway, signalling pathway alterations were largely unimpressive, and indeed protein modelling demonstrated a decrease in phosphatase activity with an increasing ratio of PR70 to the core enzyme. Furthermore, significant association between MITF and C21orf91 expression in a melanoma cohort and pooled melanoma cell lines offers the intriguing possibility that MITF may itself operate via C21orf91, identifying this as a potential new hub in control of melanoma cell cycle regulation. Supportive evidence for a role of C21orf91 in this field includes its previous identification within a pro-proliferative anti-invasive transcriptomic signature in melanoma41, a central role in cell phenotype determination in neurological development42, and recognition as one of 180 key molecules in cross-species protein networks around the Ras-MAPK/PI3K pathways43.
Due to the highly repetitive nature of this region of the genome near the telomeric end of Xp, we found targeted NGS to be the most reliable way to detect and confirm duplications of PPP2R3B. Future screening of larger melanoma cohorts and families will likely require development of a diagnostic-grade test from the point of view of cost, which will allow assessment of the frequency across different cohorts, and of the penetrance of the melanoma phenotype associated with this variant.
Conclusions
We identify here germline duplications in the gene PPP2R3B predisposing to naevogenesis and melanoma in an important proportion of cases. Duplications increase melanocytic tissue expression of the protein product PR70, which confers a survival advantage via promotion of a pro-proliferative and anti-migratory pigment cell phenotype, itself driven by a novel MITF-independent mechanism mediated by C21orf91. This work offers novel insights into both the origins and behaviour of melanocytic neoplasia, and identifies C21orf91 as an important new and potentially targetable fulcrum in the control of proliferation.
Methods
Patient recruitment and sample collection
Congenital melanocytic naevi recruited from the department of paediatric dermatology, Great Ormond Street Hospital, UK and melanoma cohort from Professor Julia Newton-Bishop, University of Leeds. DNA was extracted from blood samples using standard methods in all cohorts.
Ethical Approval
All participants gave written informed consent as part of ethically approved studies
The study of CMN genetics was approved by the London Bloomsbury Research Ethics Committee (REC) of Great Ormond Street Hospital (GOSH)/UCL Institute of Child Health (ICH). Samples from the Leeds Melanoma cohort was approved by the North East – York Research ethics committee (Jarrow, Tyne and Wear, UK).
Array CGH
Whole genome array comparative genomic hybridization (CGH) was performed as per the manufacturer’s instructions on 29 germline DNA samples from the CMN cohort, using Roche Nimblegen 135K oligonucleotide arrays and sex-matched commercial pooled controls. 1-3 μg of patient DNA and control DNA (Megapool reference DNA, Male – EA-100M, Female – EA-100F, Kreatech, The Nederlands) was labelled with fluorescent dyes Cy3 (pink) and Cy5 (blue) respectively using NimbleGen Dual-Color DNA Labeling Kit according to the manufacturer’s protocol. Labelled samples were hybridized to the oligonucleotide array using the NimbleGen Hybridization System. Two-color array scanning was performed using a Molecular Devices GenePix 4400A (Molecular Devices inc, Sunneyville, CA, USA) at a resolution of 2.5 microns, with comparison of fluorescence from both dyes. Deva software from NimbleGen was used for data extraction, and data analysed using InfoQuant CGHFusion (version 5.7.0 on initial samples and later 6.1.0) or later samples Chromosome Analysis Suite 4.0 (ChAS 4.0, Thermofisher Scientific). Array CGH data has been submitted to http://www.ncbi.nlm.nih.gov/clinvar/.
MLPA®, DNA sample preparation and internal controls
MLPA® was performed as per MRC-Holland MLPA® DNA protocol version MDP-v003, optimized using 25ng of DNA per well, and a 1 in 2 dilution of the customised probemix, using four probes for PPP2R3B, and one each for GTPBP6 and PLCXD1 designed to dovetail with the probes in the reference SALSA® MLPA® probemix P200-A1 (MRC-Holland). Capillary electrophoresis of diluted PCR products with ROX-500 size standard was performed on an ABI 3730xl, and data analysed using GeneMarker® v2.4 (SoftGenetics, USA). Each plate contained the same controls of normal copy number and duplication for standardisation.
Targeted Next Generation Sequencing Panel
A SureSelect targeted panel (Agilent Technologies, UK) was designed to capture the whole genomic region encompassing the last 4 genes on the pseudoautosomal region of X and Y chromosomes, chrX: 198061-607558 chrY: 148061-557558 (hg19, GRCh37). Library preparation was carried out using the SureSelectXT kit under manufacturers instructions (Agilent Technologies, UK). Samples were run on a NextSeq instrument 500/550, read length of 2×150 bp (Illumina, USA). In total 183 samples were processed including a melanoma cohort (n= 168), CMN cohort (n=5) and control samples from the ALSPAC study (n=3). BAM files were inputted to DeepTools MultiBamSummary using the PPP2R3B_Moderately_Stringent_1_covered.bed probe coordinates file. Coverage across probe regions within the gene hg19 coordinates were extracted and averaged. 3 Control samples were used to create ‘normal expected’ coverage ratios of the genes compared to SHOX. All samples were normalised compared to these ratios and R studio was used to visualise and calculated gene coverage data across all samples.
Immunohistochemistry
The expression of protein product PR70 was assessed in formalin fixed paraffin embedded naevus tissue from patients with known PPP2R3B copy number status, both duplications and normal copy number, using an antibody against PR70 (see Table 1 for source, dilution) on the automated LEICA-Bond-Max immunostainer.
Antibodies
Cell lines and culture
SKMEL2 and SKMEL30 melanoma cell lines were obtained from Cell Services, Francis Crick Institute. SKMEL2 were cultured in DMEM 10% FBS Penicillin streptomycin and glutamine cells incubated at 37°C 10% CO2. SKMEL30 were cultured in RPMI 10% FBS Penicillin streptomycin and glutamine cells incubated at 37 50% CO2. Normal PPP2R3B copy number of both cell lines was verified as described above.
Generation of stable inducible PPP2R3B cells lines
Human myc-FLAG tagged PPP2R3B ORF clone from Origene (RC222908) was linearised and the insert DNA amplified using modified primers generating an N-terminal Myc tag. The In-Fusion® HD Cloning system (Takara, 638909) was used to allow directional cloning of the PPP2R3B insert into the AgeI-MluI site of the lentiviral vector pTRIPZ (GE Healthcare) resulting in the final PPP2R3B (tet-ON) construct, without the TurboRFP or shRNAmir-related elements of the parental pTRIPZ plasmid. Transduction of HEK 293T cells with pTRIPZ-PPP2R3B, psPAX2 and pMD2.G plasmids using Lipofectamine 2000 generated lentiviral particles used to infect SKMEL2 or SKMEL30 target cells using polybrene. Stable cell lines were selected using puromycin.
Sample preparation for Reverse Phase Protein Array and RNAseq
SKMEL30 pTRIPZ-PPP2R3B and SKMEL2 pTRIPZ-PPP2R3B cells were seeded and treated with doxycycline 1ug/ml for either 6 hours or 16 hours alongside untreated controls. Samples were collected in triplicate for both RNA extraction and protein lysis. RNA extraction and protein lysis was performed as described below. Confirmation of PPP2R3B induction in all replicates was performed by qPCR of reverse transcribed cDNA from extracted RNA, and Western Blotting for PPP2R3B.
Reverse Phase Protein Array – MD Anderson
Protein samples were diluted to 1.5ug/ul with complete lysis buffer. Samples were submitted to MD Anderson, Core Facility. Reported intensity values were log transformed to approximate normality and comparisons were performed using an un-paired t-test.
RNAseq
RNA integrity calculated using a Bioanalyser (Agilent). Library preparation using KAPA mRNA HyperPrep Kit (Roche) was automated using the Hamilton robot. Libraries were sequenced using a NextSeq 500 (Illumina, San Diego, US) with a 43-bp paired-end run. Data were trimmed for 3’ adapter sequences using Cutadapt 1.9.1, after which they were aligned to the ensemble GRCh38 release 86 human transcriptome using STAR 2.5.2a. Individual lane level replicates were merged using Samtools 1.8, after which raw gene counts were estimated using RSEM 1.3.0 and normalisation and differential expression calling was then performed using DESeq2. A corrected p-value of let than or equal to 0.05 was deemed to be significant. Pathway analyses based on genes reported in the various analyses were performed using Metacore (Clarivate Analytics)
Quantitative Real Time PCR
RNA was extracted using RNeasy mini kit (Qiagen, 74104), as per manufacturer’s instructions. 1ug of total RNA was reverse transcribed to cDNA using Quantitect reverse transcription kit (Qiagen, 205311), and quantitative real time PCR was performed, using Taqman probes and the StepOnePlus Real-Time PCR System (ABI, Abilene, TX, USA) (see Table 2 for details of Taqman probes). Relative fold change to the housekeeping gene GAPDH was calculated by ΔΔCT method.
Taqman® probes
Western Blotting
Protein lysate was quantified using Bradford protein assay, and 20ug of each sample was run on a 10% Bis-Tris protein gel. PVDF membrane was activated in methanol prior to transfer using a wet transfer reservoir. The membrane was washed three times in Tris Buffered Saline-Tween (TBST), then blocked using 3% albumin at room temperature for one hour. Membranes were then incubated in primary antibody in 3% albumin overnight at 4°C (see Table 1 for source, dilution). Membranes were washed three times in TBST then incubated in corresponding HRP-tagged secondary antibody in 5% milk for 1 hour at room temperature. Membranes were washed a further three times in TBST before adding ECL for two minutes followed by measurement of chemiluminescence using Amersham Imager 600. Intensity of bands was measured and quantified using Image J software.
Immunocytochemistry
Cells were seeded at a density of 4 × 104 cells per well in laminin coated 4 well chamber slides (Millicell® EZ slide, Merck Millipore, PEZGS0816). Cells were washed in cold PBS and fixed using 4% paraformaldehyde (PFA), followed by permeabilisation using 0.1% Triton X100. Slides were washed three times with PBS and incubated with blocking solution (1% Bovine serum albumin, 10% Donkey serum, 0.3M Glycine, 0.1% PBS Tween) for 1 hour at room temperature followed by primary antibody in blocking solution overnight at 4 degrees (see Table 1 for source, dilution). Slides were washed once with 1:10,000 Hoechst for nuclear staining, then washed three times with 0.1% PBS Tween before incubation with a fluorescent tagged appropriate secondary antibody at room temperature for one hour in the dark. Slides were again washed three times with 0.1% PBS Tween and were mounted prior to analysis using a Leica upright fluorescent microscope.
Proliferation Assays
WST1 proliferation Assay
SKMEL2 pTRIPZ PPP2R3B and SKMEL30 pTRIPZ PPP2R3B cells were seeded into a 96 well plate at a density of 1 × 104 cells per well. PPP2R3B overexpression was induced using doxycycline (1ug/ml) alongside un-induced controls. Plates were incubated for 48 hours at 37°C, prior to the addition of WST-1 reagent. Plates incubated in the dark at 37 °C for a further two hours. Absorbance of plates was read in a spectrophotometer at 450nm and 620nm. Results were adjusted for absorbance of blank media, and absorbance of WST1 dye (620nm) and averaged between the twelve wells for each condition. Absorbance was averaged between 12 replicates for each condition (mean + SD) and statistical significance was calculated by Students t-test.
BrDU Proliferation Assay
BrDU assay was carried out using BrdU Cell Proliferation ELISA Kit, (Abcam, ab126556), as per the manufacturer’s instructions. SKMEL2 pTRIPZ PPP2R3B and SKMEL30 pTRIPZ PPP2R3B cells were seeded into a 96 well plate at a density of 2 × 105 cells per well. PPP2R3B overexpression was induced using doxycycline (1ug/ml) alongside un-induced controls, and suggested assay controls. BrdU reagent diluted in media was added to all wells and plates were incubated for 24 hours at 37°C, at which point media was removed and Fixing Solution added to each well. Plates were incubated at room temperature for 30 minutes, then blotted dry before washing three times with Wash Buffer. Anti-BrdU monoclonal Detector Antibody was each well and plates incubated for 1 hour at room temperature. Plates were washed a further 3 times with Wash Buffer before adding peroxidase Goat Anti-Mouse IgG Conjugate and incubating for 30 minutes at room temperature. Plates were washed a final three times with Wash Buffer and once with distilled water before the addition of TMB Peroxidase Substrate Pipette. Plates were incubated for 30 minutes at room temperature in the dark. Stop Solution was added to each well and plates were absorbance at 450 nm was measured using a spectrophotometer. Absorbance was averaged between 12 replicates for each condition (mean + SD) and statistical significance was calculated by Students t-test.
IncuCyte® Cell Count Proliferation Assay
SKMEL2 pTRIPZ PPP2R3B and SKMEL30 pTRIPZ PPP2R3B cells were seeded into a 96 well ImageLock plate at a density of 1 × 104 cells per well. PPP2R3B overexpression was induced using doxycycline (1ug/ml) alongside un-induced controls, a total of 12 replicates per condition. The plate was then analysed using the IncuCyte® live-cell analysis system and set up to acquire 10X phase contrast images at a scanning interval of 60 minutes for 5 days, measuring percentage confluence. Confluence was averaged between 12 replicates for each condition (mean + SD) and statistical significance was calculated by Students t-test.
C21orf91 Knockdown
SKMEL2 pTRIPZ PPP2R3B were seeded into a 96 well ImageLock plate at a density of 1 × 104 cells per well. PPP2R3B overexpression was induced using doxycycline (1ug/ml) alongside un-induced controls, a total of 12 replicates per condition. Selected cells were transfected with Lipofectamine™ RNAiMAX using two different C21orf91 siRNAs at a concentration on 10nm or a Scrambled siRNA (Origene, SR310041). The plate was then analysed using the IncuCyte® live-cell analysis system and set up to acquire 10X phase contrast images at a scanning interval of 60 minutes for 5 days, measuring percentage confluence. Confluence was averaged between 12 replicates for each condition (mean + SD) and statistical significance was calculated by Students t-test.
Migration Studies
Scratch Wound Assay
SKMEL2 pTRIPZ PPP2R3B and SKMEL30 pTRIPZ PPP2R3B cells were seeded into a 96 well ImageLock plate at a density of 1 × 105 cells per well. PPP2R3B overexpression was indut.ced using doxycycline (1ug/ml) alongside un-induced controls, 12 replicates per condition. Plates were incubated at 37 degrees until all wells were confluent. WoundMaker™ was used to create a scratch in each well as per manufacturer’s instructions. Plates were washed with PBS to remove any debris and fresh media added to each well. The plate was then analysed using the IncuCyte® live-cell analysis system and set up to acquire 10X phase contrast images at a scanning interval of 60 minutes, scan type ‘Scratch Wound’, measuring relative wound confluence. Relative wound confluence was averaged between 12 replicates for each condition (mean + SD) and statistical significance was calculated by Students t-test.
Melanoma Cohort – Transcriptomic data
Transcriptomic data were generated in 703 FFPE tumours from the Leeds Melanoma Cohort, using the Illumina DASL® array.
Author contributions
SP designed and carried out cellular modelling, RPPA, RNAseq, qRT-PCR and Western blot validations, proliferation and migration assays and immunocytochemistry, analysis of all associated data, produced the figures, and wrote the online methods. LA-O and ACT carried out MLPA copy number validation, the pilot NGS panel, RT-PCR tissue arrays, and analysis of data. DL designed and advised on the cellular modelling, RPPA, RNAseq and validations. MH, JN and JNB contributed patients and samples, with phenotypic, genotypic, and expression data and the associated analysis, and reviewed the manuscript. LH and GE analysed the NGS panel copy number data. SH and AP analysed the RNAseq, RPPA and regional similarity search data and produced the relevant figures. LH and GE analysed the NGS panel copy number data. NW, HC and YX designed and carried out the PR70 phosphatase activity and pull-down experiments, and reviewed the manuscript. PA, JM, VMdS, CC, GT, JAPB, and SP contributed patients, samples and phenotypic information. SB, RW and SMB conducted some sequencing experiments. DB and DM advised and helped analyse immunocytochemistry. PB was involved in early discussions of the genetic data. CH contributed patients and samples. Michael H advised on proliferation and migration assays. LL contributed cell line expression data and analysis with the relevant figure. SL and DH contributed to analysis of the array CGH data and validation. JM, SS, MZ contributed control cohort data on copy number. MM, DZ and WLD helped optimise cellular modelling and validation. KP and VB contributed patients and samples and reviewed the manuscript. NS analysed immunohistochemistry. Paulina S performed knock-down experiments and analysis. PS was involved in early discussion of the genetic data and reviewed the manuscript. EH and GM contributed to early research design and to the manuscript. JD helped conceive and direct the cellular modelling and signalling pathway characterisation, and contributed to the manuscript. VAK conceived, designed and directed the research, designed and carried out the initial human genetics experiments, contributed patients, analysed data, and wrote the manuscript.
Competing interests
The authors declare no conflict of interest.
Supplementary
Melanoma Transcriptomic Data – Upregulated Pathways
The genes positively correlated with PPP2R3B are predominantly enriched in non-immune pathways, consistent with the lack of association between PPP2R3B expression and TILs (tumour infiltrating lymphocytes) or any specific immune cell score. There are only 24 pathways significantly enriched in these genes at FDR<0.05 (Table S1).
Using undirected interactions, the nodal genes in a network putting together all these pathways are: MAPK11 (betweenness=2237) and PLCB3 (betweenness=1120).
With directed interactions, the most important genes are MAPK11 (betweenness=103) and NFATC1 (betweenness=59).
Melanoma Transcriptomic Data – Upregulated Pathways
Melanoma Transcriptomic Data – Downregulated Pathways
The genes negatively correlated with PPP2R3B expression are predominantly enriched in immune pathways. There are 80 pathways, top 25 of which are shown in Table S2.
Using undirected interactions, the top 5 nodal genes in a network putting together all these pathways are: TGFB1 (betweenness=8395) and AURKA (betweenness=5133) and CD8A (betweenness=4893), RAC2 (4880) and LCK (betweenness=4855).
With directed interactions, the most important genes are LCK (betweenness=174), HLA-G (betweenness=110), MYD88 (betweenness=104), TGFB1 (betweenness=86) and CD8A (betweenness=82).
Melanoma Transcriptomic Data – Downregulated Pathways
Acknowledgements
We gratefully acknowledge the participation of all patients and families in this study, and research coordination by Mrs Jane White. VAK, AT and the work presented in this study were funded by the Wellcome Trust (Grant WT104076MA). SP was funded by Caring Matters Now Charity and by the Newlife Foundation. The work was supported by the GOSHCC Livingstone Skin Research Centre, and by the UK National Institute for Health Research through the Biomedical Research Centre at Great Ormond St Hospital for Children NHS Foundation Trust, and the UCL GOS Institute of Child Health.
References
