CoChat
Sign Up
Cited by 223Cancer Genomics and DiagnosticsCancer-related Molecular PathwaysPancreatic and Hepatic Oncology Research

The DOI matches its Crossref record; 74 of the 74 references it cites were found in a registry or an index.

Ordered and deterministic cancer genome evolution after p53 loss

Open access
Baslan, Timour, Morris, John P., Zhao, Zhen, Reyes, Jose, Ho, Yu-Jui, Tsanov, Kaloyan M., Bermeo, Jonathan, Tian, Sha, Zhang, Sean, Askan, Gokce, Yavas, Aslihan, Lecomte, Nicolas, Erakky, Amanda, Varghese, Anna M., Zhang, Amy, Kendall, Jude, Ghiban, Elena, Chorbadjiev, Lubomir, Wu, Jie, Dimitrova, Nevenka, Chadalavada, Kalyani, Nanjangud, Gouri J., Bandlamudi, Chaitanya, Gong, Yixiao, Donoghue, Mark T. A., Socci, Nicholas D., Krasnitz, Alex, Notta, Faiyaz, Leach, Steve D., Iacobuzio-Donahue, Christine A., Lowe, Scott W.·2022·Nature
DOI 10.1038/s41586-022-05082-5PMC PMC9402436
Open this paper in CoChat — free accountOpen PDF

Abstract

Although p53 inactivation promotes genomic instability 1 and presents a route to malignancy for more than half of all human cancers 2,3 , the patterns through which heterogenous TP53 (encoding human p53) mutant genomes emerge and influence tumorigenesis remain poorly understood. Here, in a mouse model of pancreatic ductal adenocarcinoma that reports sporadic p53 loss of heterozygosity before cancer onset, we find that malignant properties enabled by p53 inactivation are acquired through a predictable pattern of genome evolution. Single-cell sequencing and in situ genotyping of cells from the point of p53 inactivation through progression to frank cancer reveal that this deterministic behaviour involves four sequential phases— Trp53 (encoding mouse p53) loss of heterozygosity, accumulation of deletions, genome doubling, and the emergence of gains and amplifications—each associated with specific histological stages across the premalignant and malignant spectrum. Despite rampant heterogeneity, the deletion events that follow p53 inactivation target functionally relevant pathways that can shape genomic evolution and remain fixed as homogenous events in diverse malignant populations. Thus, loss of p53—the ‘guardian of the genome’—is not merely a gateway to genetic chaos but, rather, can enable deterministic patterns of genome evolution that may point to new strategies for the treatment of TP53- mutant tumours.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41586-022-05082-5

Cites

74 of 74 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Lineage tracing of incipient cancer cells after sporadic p53 inactivation in mouse PDAC. a, Schematic of KPCLOH: fluorescent tracking of p53 LOH in Kras-driven pancreatic tumorigenesis. b, Representative haematoxylin and eosin (H&E) staining (left) and mKate/GFP immunofluorescence (IF, right) of SP (mKate+) versus DP (mKate+GFP+) cells in a PDAC-bearing (red outline) mouse. c, Representative H&E staining (top) and Kate/GFP/Ki-67 immunofluorescence (bottom) in the adjacent (Adj.) premalignant tissue (left) versus focal PDAC (right). The solid outline indicates ADM and AFL. The dashed outline shows PanIN. The arrowheads indicate Ki-67+ cells. d, DP cell frequency in ADM, AFL, PanIN and PDAC. n = 6. e, Representative H&E (left) and mKate/GFP immunofluorescence (right) of SP (red dots) versus DP cells in a mouse without PDAC. Inset: H&E (top) and immunofluorescence (bottom) analysis of SP cells within a DP structure (indicate by an asterisk (*)). f, Representative H&E (top) and Kate/GFP/Ki-67 immunofluorescence (bottom) analysis of ADM SP lesions (solid lines) observed in a mouse without PDAC. The arrowheads indicate Ki-67+ SP cells. g, Characterization of SP lesions in KPCLOH mice without PDAC. n = 43 lesions, n = 7 mice. HG, high grade; LG, low grade; w/n, within. h, The percentage of Ki-67+ DP and SP cells in adjacent premalignant and PDAC tissue. n = 8. i, The percentage of Ki-67+ SP and DP cells in lesions of the indicated size in KPCLOH mice without frank PDAC. n = 9. j, The relative growth of 500 DP or SP cells sorted before (pre-tumour, n = 6) and after (PDAC, n = 4) frank PDAC development. k, The survival of mice transplanted with 100–1,000 SP cells sorted from KPCLOH mice with (solid line, 12 injections, 6 each from 2 mice) or without (dashed line, n = 10) frank PDAC. For b and e, the experiments were repeated at least three times with similar results. For d, data are mean ± s.d. For the box plots in h and i, the centre line shows the median, the box limits show the 25th and 75th percentiles, and the whiskers show the range; outliers are shown. For h and i, significance was assessed using two-tailed Wilcoxon's rank-sum tests. Scale bars, 1 mm (b and e) and 50 μm (c and f). Source data
Fig. 1 Lineage tracing of incipient cancer cells after sporadic p53 inactivation in mouse PDAC. a, Schematic of KPCLOH: fluorescent tracking of p53 LOH in Kras-driven pancreatic tumorigenesis. b, Representative haematoxylin and eosin (H&E) staining (left) and mKate/GFP immunofluorescence (IF, right) of SP (mKate+) versus DP (mKate+GFP+) cells in a PDAC-bearing (red outline) mouse. c, Representative H&E staining (top) and Kate/GFP/Ki-67 immunofluorescence (bottom) in the adjacent (Adj.) premalignant tissue (left) versus focal PDAC (right). The solid outline indicates ADM and AFL. The dashed outline shows PanIN. The arrowheads indicate Ki-67+ cells. d, DP cell frequency in ADM, AFL, PanIN and PDAC. n = 6. e, Representative H&E (left) and mKate/GFP immunofluorescence (right) of SP (red dots) versus DP cells in a mouse without PDAC. Inset: H&E (top) and immunofluorescence (bottom) analysis of SP cells within a DP structure (indicate by an asterisk (*)). f, Representative H&E (top) and Kate/GFP/Ki-67 immunofluorescence (bottom) analysis of ADM SP lesions (solid lines) observed in a mouse without PDAC. The arrowheads indicate Ki-67+ SP cells. g, Characterization of SP lesions in KPCLOH mice without PDAC. n = 43 lesions, n = 7 mice. HG, high grade; LG, low grade; w/n, within. h, The percentage of Ki-67+ DP and SP cells in adjacent premalignant and PDAC tissue. n = 8. i, The percentage of Ki-67+ SP and DP cells in lesions of the indicated size in KPCLOH mice without frank PDAC. n = 9. j, The relative growth of 500 DP or SP cells sorted before (pre-tumour, n = 6) and after (PDAC, n = 4) frank PDAC development. k, The survival of mice transplanted with 100–1,000 SP cells sorted from KPCLOH mice with (solid line, 12 injections, 6 each from 2 mice) or without (dashed line, n = 10) frank PDAC. For b and e, the experiments were repeated at least three times with similar results. For d, data are mean ± s.d. For the box plots in h and i, the centre line shows the median, the box limits show the 25th and 75th percentiles, and the whiskers show the range; outliers are shown. For h and i, significance was assessed using two-tailed Wilcoxon's rank-sum tests. Scale bars, 1 mm (b and e) and 50 μm (c and f). Source dataCC BY · PubMed Central
Fig. 2 Recurrent and conserved CNAs targeting PDAC drivers shape the evolution of malignant genomes after p53 inactivation. a, Matching genome-wide copy-number profiles of SP and DP cells isolated from a polyploid KPCLOH PDAC. The red arrows indicate distinguishing alterations. b, Frequency plot of recurrent CNAs from sequencing-sorted DP (n = 14) and SP (n = 24) cells after PDAC development. The chromosomes highlighted in grey denote regions recurrently altered in SP samples and analysed for synteny with human PDAC data. The filled red trace denotes chromosome 6 gains found in a subset of DP samples. The vertical dashed lines denote the location of PDAC driver genes. c, Human–mouse synteny Circos rendering of selected alterations on mouse chromosomes 5 and 9. The red and blue colouring denotes gains and deletions with matching species synteny, respectively. The grey colouring denotes no matching genomic intervals in directionality (for example, gains or loss in both species). Selected PDAC-relevant genes are shown. d, Chromosome 9 deletion frequency plot in KPCLOH (n = 22), KPCmut (n = 16) and KPCmut/shSmad4 mouse PDACs (n = 7). Chr, chromosome.
Fig. 2 Recurrent and conserved CNAs targeting PDAC drivers shape the evolution of malignant genomes after p53 inactivation. a, Matching genome-wide copy-number profiles of SP and DP cells isolated from a polyploid KPCLOH PDAC. The red arrows indicate distinguishing alterations. b, Frequency plot of recurrent CNAs from sequencing-sorted DP (n = 14) and SP (n = 24) cells after PDAC development. The chromosomes highlighted in grey denote regions recurrently altered in SP samples and analysed for synteny with human PDAC data. The filled red trace denotes chromosome 6 gains found in a subset of DP samples. The vertical dashed lines denote the location of PDAC driver genes. c, Human–mouse synteny Circos rendering of selected alterations on mouse chromosomes 5 and 9. The red and blue colouring denotes gains and deletions with matching species synteny, respectively. The grey colouring denotes no matching genomic intervals in directionality (for example, gains or loss in both species). Selected PDAC-relevant genes are shown. d, Chromosome 9 deletion frequency plot in KPCLOH (n = 22), KPCmut (n = 16) and KPCmut/shSmad4 mouse PDACs (n = 7). Chr, chromosome.CC BY · PubMed Central
Fig. 3 Distinct and ordered phases of genome evolution accompany the benign-to-malignant switch. a, Breakpoint-based phylogenetic tree of single SP (n = 130) and DP (n = 55) cells sequenced from PDAC sample T2 (left). The red arrow indicates a split in the neighbour-joining tree and clonal sweep of SP cells. Distance is based on statistical considerations of breakpoint similarity/dissimilarity (Methods). Sweeping SP cells share a clonal relationship with a false-discovery rate (FDR) not exceeding a threshold value of t = 0.01. Right, breakpoint-based phylogenetic tree of single SP cells (n = 171) sequenced from pre-tumour sample P3. The clone track denotes a lineage that underwent genome doubling (navy). The clonal relationship between diploid and polyploid cells is computed with an FDR not exceeding a threshold value of t = 0.01. Colour codes for ploidy, lineage and copy number are provided. b, Matched H&E and immunofluorescence of lesions that underwent LMD (yellow outlines) (top). Bottom, matched copy-number profiles of lesions collected by LMD. Scale bar, 50 μm.
Fig. 3 Distinct and ordered phases of genome evolution accompany the benign-to-malignant switch. a, Breakpoint-based phylogenetic tree of single SP (n = 130) and DP (n = 55) cells sequenced from PDAC sample T2 (left). The red arrow indicates a split in the neighbour-joining tree and clonal sweep of SP cells. Distance is based on statistical considerations of breakpoint similarity/dissimilarity (Methods). Sweeping SP cells share a clonal relationship with a false-discovery rate (FDR) not exceeding a threshold value of t = 0.01. Right, breakpoint-based phylogenetic tree of single SP cells (n = 171) sequenced from pre-tumour sample P3. The clone track denotes a lineage that underwent genome doubling (navy). The clonal relationship between diploid and polyploid cells is computed with an FDR not exceeding a threshold value of t = 0.01. Colour codes for ploidy, lineage and copy number are provided. b, Matched H&E and immunofluorescence of lesions that underwent LMD (yellow outlines) (top). Bottom, matched copy-number profiles of lesions collected by LMD. Scale bar, 50 μm.CC BY · PubMed Central
Fig. 4 Deterministic principles govern the selection of genomic rearrangements after p53 LOH. a, Breakpoints in LOH cells from sample P2 associated with chromosome 11 deletion reflecting the lineage heterogeneity of cells undergoing LOH events. b, Quantification of distinct p53 LOH/chromosome 11 deletion breakpoints in 7 KPCLOH pre-tumour mice. c, Quantification of acquired CNAs in SP cells from pre-tumour mice (n = 7) compared with DP premalignant cells (n = 6). Statistical analysis was performed using a two-tailed Mann–Whitney U-test; P = 0.00338. d, Quantification of CNAs identified in pre-SP cells from seven mice according to CNA class. Statistical analysis was performed using a two-tailed Mann–Whitney U-test; P = 0.0041. e, Recurrent chromosome 9 deletions identified in pre-SP cells. Distinct deletion events are uniquely coloured. The vertical grey line marks the location of Tgfbr2. f, Genome-wide copy-number profiles of a polyploid single cell and its inferred diploid precursor illustrating the genomic relationship and genome doubling. The diagonal red lines denote CNA-associated breakpoints used to infer lineage (Extended Data Fig. 5). g, Heat-map analysis of all of the identified polyploid pre-SP cells (n = 132) in pre-tumour mice (n = 7). P1 and P5 illustrate instances in which the emerging polyploid lineage is diversifying genomically. h, Quantification of CNA events per class (that is, deletion versus gain) in SP cells sequenced from tumour (n = 6) and pre-tumour mice (n = 7). Statistical analysis was performed using a two-sided t-test for enrichment of gains in polyploid cells; P = 0.005. i, Illustration of the heterogeneity/homogeneity of selected recurrent gains and deletions in KPCLOH PDACs. The segments (blue lines) at multiple- or single-copy-number states indicate heterogeneity and homogeneity, respectively. j, Quantification of CNA segment homogeneity (Methods) based on single-cell copy-number data of SP cells from PDAC mice. n = 4. For c, d, and h, recurrent CNAs were computed using the algorithm CORE (Methods). Box plots are as defined in Fig. 1. Source data
Fig. 4 Deterministic principles govern the selection of genomic rearrangements after p53 LOH. a, Breakpoints in LOH cells from sample P2 associated with chromosome 11 deletion reflecting the lineage heterogeneity of cells undergoing LOH events. b, Quantification of distinct p53 LOH/chromosome 11 deletion breakpoints in 7 KPCLOH pre-tumour mice. c, Quantification of acquired CNAs in SP cells from pre-tumour mice (n = 7) compared with DP premalignant cells (n = 6). Statistical analysis was performed using a two-tailed Mann–Whitney U-test; P = 0.00338. d, Quantification of CNAs identified in pre-SP cells from seven mice according to CNA class. Statistical analysis was performed using a two-tailed Mann–Whitney U-test; P = 0.0041. e, Recurrent chromosome 9 deletions identified in pre-SP cells. Distinct deletion events are uniquely coloured. The vertical grey line marks the location of Tgfbr2. f, Genome-wide copy-number profiles of a polyploid single cell and its inferred diploid precursor illustrating the genomic relationship and genome doubling. The diagonal red lines denote CNA-associated breakpoints used to infer lineage (Extended Data Fig. 5). g, Heat-map analysis of all of the identified polyploid pre-SP cells (n = 132) in pre-tumour mice (n = 7). P1 and P5 illustrate instances in which the emerging polyploid lineage is diversifying genomically. h, Quantification of CNA events per class (that is, deletion versus gain) in SP cells sequenced from tumour (n = 6) and pre-tumour mice (n = 7). Statistical analysis was performed using a two-sided t-test for enrichment of gains in polyploid cells; P = 0.005. i, Illustration of the heterogeneity/homogeneity of selected recurrent gains and deletions in KPCLOH PDACs. The segments (blue lines) at multiple- or single-copy-number states indicate heterogeneity and homogeneity, respectively. j, Quantification of CNA segment homogeneity (Methods) based on single-cell copy-number data of SP cells from PDAC mice. n = 4. For c, d, and h, recurrent CNAs were computed using the algorithm CORE (Methods). Box plots are as defined in Fig. 1. Source dataCC BY · PubMed Central
Fig. 5 Whole genomes, targeted capture and single-cell sequencing corroborate evolutionary principles in human disease. a, The copy-number landscape of diploid TP53 biallelic PDAC compared with diploid TP53-mono/WT PDAC from the COMPASS dataset. b, PDAC ploidy according to TP53 allelic state from COMPASS dataset. TP53 biallelic mutant PDAC are significantly more likely to exhibit polyploidy. Statistical analysis was performed using the Fisher exact test; P = 10−6. c, Quantification of CNA events, as computed using the algorithm CORE (Methods) per class (that is, deletion versus gain) in all polyploid (n = 137) and diploid (n = 156) human PDACs from the COMPASS trial. Statistical analysis was performed using a two-sided t-test for gain/amplification enrichment in polyploid cells; P = 2.2 × 10−16. d, Kernel-density estimation of normalized homogeneity (Methods) of CNAs genome wide from targeted capture (MSK-IMPACT) of PDAC (n = 1,076 total) cases according to ploidy and TP53 mutation status. Chromosomal gains/amplifications are significantly more likely to be heterogenous. Statistical analysis was performed using a two-sample Kolmogorov–Smirnov test; P < 0.005. Empirical cumulative distribution function measurements are shown in Extended Data Fig. 10. e, Disease type in polyploid and diploid PDAC with biallelic TP53 inactivation from the MSK-IMPACT dataset. Statistical analysis was performed using a Fisher exact test; P = 0.003. Box plots are as defined in Fig. 1. Source data
Fig. 5 Whole genomes, targeted capture and single-cell sequencing corroborate evolutionary principles in human disease. a, The copy-number landscape of diploid TP53 biallelic PDAC compared with diploid TP53-mono/WT PDAC from the COMPASS dataset. b, PDAC ploidy according to TP53 allelic state from COMPASS dataset. TP53 biallelic mutant PDAC are significantly more likely to exhibit polyploidy. Statistical analysis was performed using the Fisher exact test; P = 10−6. c, Quantification of CNA events, as computed using the algorithm CORE (Methods) per class (that is, deletion versus gain) in all polyploid (n = 137) and diploid (n = 156) human PDACs from the COMPASS trial. Statistical analysis was performed using a two-sided t-test for gain/amplification enrichment in polyploid cells; P = 2.2 × 10−16. d, Kernel-density estimation of normalized homogeneity (Methods) of CNAs genome wide from targeted capture (MSK-IMPACT) of PDAC (n = 1,076 total) cases according to ploidy and TP53 mutation status. Chromosomal gains/amplifications are significantly more likely to be heterogenous. Statistical analysis was performed using a two-sample Kolmogorov–Smirnov test; P < 0.005. Empirical cumulative distribution function measurements are shown in Extended Data Fig. 10. e, Disease type in polyploid and diploid PDAC with biallelic TP53 inactivation from the MSK-IMPACT dataset. Statistical analysis was performed using a Fisher exact test; P = 0.003. Box plots are as defined in Fig. 1. Source dataCC BY · PubMed Central
Extended Data Fig. 1 Fluorescent linkage reports sporadic p53 loss of heterozygosity in the KPCLOH PDAC model. a, Schematic of breeding, embryonic stem cell engineering, allelic configuration, and staging of KPCTRE-shRNA ESC PDAC. Dashed line defines tumour mass detected by ultrasound, K, kidney. b, Representative whole mount bright field and fluorescent gross pathology of PDAC arising in context of adjacent premalignant (Pre-M) tissue in KPCshRenilla and KPCshp53 mice. c, Survival curve of KPCTRE-shRenilla (n = 33) and KPCTRE-shp53 (n = 21) mice. Significance of difference in survival curves assessed by log rank (Mantel-Cox) test. d, Kate and GFP flow cytometry of primary cultures of dissociated shRenilla and shp53 KPCTRE pancreas following PDAC development at indicated passages. e, PCR detection of recombined versus wild type alleles for Kras and p53 in primary cultures from indicated samples at passage 6. f, Representative flow cytometry plot distinguishing single Kate positive (SP) from double Kate/GFP positive (DP) cells after PDAC development in a KPCshRenilla mouse. g, PCR detection of recombined versus wild type alleles for Kras, p53, as well as shRNA and RIK transgenes in DP and SP cells sorted from KPCshRenilla mice following PDAC development and cell lines generated from PDAC arising in KPCshp53 mice. h, (Top) Digital PCR detecting relative levels of recombined conditional p53 allele, WT p53, and GFP targeted CHC cassette and (bottom) KrasG12D and WT Kras alleles in DP (n = 16) and SP (n = 19) cells sorted from KPCTRE-shRenilla PDAC. i, Relative copy number of chromosome 11 inferred from sparse whole genome sequencing from PDAC arising in KPCshRenilla and KPCshp53 mice (n = 4 each). Normalized segment values are centred around a mean value of 1 with segment values below 1 indictive of deletion events. j, Representative, matched immunofluorescence of GFP and mKate and immunohistochemistry of p53 in sequential sections in a KPCshRenilla PDAC. h, mean ± S.D. Scale bars b, 1 cm, j 50 μm. e and g were repeated at least twice with similar results, and j was repeated 3 times with similar results. Gel source data for e and g, see Supplementary Fig. 1. Gating strategy of d and f, see Supplementary Fig. 2. Source data
Extended Data Fig. 1 Fluorescent linkage reports sporadic p53 loss of heterozygosity in the KPCLOH PDAC model. a, Schematic of breeding, embryonic stem cell engineering, allelic configuration, and staging of KPCTRE-shRNA ESC PDAC. Dashed line defines tumour mass detected by ultrasound, K, kidney. b, Representative whole mount bright field and fluorescent gross pathology of PDAC arising in context of adjacent premalignant (Pre-M) tissue in KPCshRenilla and KPCshp53 mice. c, Survival curve of KPCTRE-shRenilla (n = 33) and KPCTRE-shp53 (n = 21) mice. Significance of difference in survival curves assessed by log rank (Mantel-Cox) test. d, Kate and GFP flow cytometry of primary cultures of dissociated shRenilla and shp53 KPCTRE pancreas following PDAC development at indicated passages. e, PCR detection of recombined versus wild type alleles for Kras and p53 in primary cultures from indicated samples at passage 6. f, Representative flow cytometry plot distinguishing single Kate positive (SP) from double Kate/GFP positive (DP) cells after PDAC development in a KPCshRenilla mouse. g, PCR detection of recombined versus wild type alleles for Kras, p53, as well as shRNA and RIK transgenes in DP and SP cells sorted from KPCshRenilla mice following PDAC development and cell lines generated from PDAC arising in KPCshp53 mice. h, (Top) Digital PCR detecting relative levels of recombined conditional p53 allele, WT p53, and GFP targeted CHC cassette and (bottom) KrasG12D and WT Kras alleles in DP (n = 16) and SP (n = 19) cells sorted from KPCTRE-shRenilla PDAC. i, Relative copy number of chromosome 11 inferred from sparse whole genome sequencing from PDAC arising in KPCshRenilla and KPCshp53 mice (n = 4 each). Normalized segment values are centred around a mean value of 1 with segment values below 1 indictive of deletion events. j, Representative, matched immunofluorescence of GFP and mKate and immunohistochemistry of p53 in sequential sections in a KPCshRenilla PDAC. h, mean ± S.D. Scale bars b, 1 cm, j 50 μm. e and g were repeated at least twice with similar results, and j was repeated 3 times with similar results. Gel source data for e and g, see Supplementary Fig. 1. Gating strategy of d and f, see Supplementary Fig. 2. Source dataCC BY · PubMed Central
Extended Data Fig. 2 Linkage in cis with mutant p53 retains inducible shRNA expression following p53 LOH in the KPCCis-shRNA PDAC model and functional and histological characterization of the premalignant to malignant transition captured by stage dependent analysis of KPCLOH mice. a, Schematic of breeding, embryonic stem cell engineering, allelic configuration, and cohort generation of the KPCR172H-CIS-TRE-shRNA ESC PDAC GEMM. Note that founder LSL-p53R172H; CHC double homozygotes were utilized to ensure segregation of the conditional mutant p53 allele in cis with the collagen homing cassette. b, Representative whole mount bright field and fluorescent gross pathology of PDAC and adjacent premalignant (Pre-M) tissue in KPCR172H-Cis-TRE-shRenilla mice. c, PCR detecting recombination of the conditional p53R172H allele versus WT p53 allele (left), RIK allele (centre), and targeted CHC versus WT Col1a1 allele (right) in primary cancer cell lines derived from PDAC developing in KPCR172H-Cis-TRE-shRenilla mice (n = 3). Note the absence of WT p53 and WT Col1a1, but the maintenance of the targeted CHC allele. d, Schematic depiction of maintenance of GFP linked shRNA in cis with the conditional mutant p53R172H allele during p53 LOH and PDAC progression in KPCR172H-Cis-TRE-shRNA mice. e, Tabular results of tumours developing after injection of indicated number of DP and SP cells sorted from PDAC bearing KPCLOH mice into immunocompromised, nude mice. f, Survival curve of injected recipients detailed in e. g, Flow cytometry of GFP and Kate in a representative tumour resulting from injection of PDAC associated SP cells (left) and flow cytometry of tumours resulting from injection of 25000 DP cells sorted from 3 of 4 mice as indicated in e. Tumours were composed of either exclusively SP cells (Donor 1), predominantly SP cells (Donor 2), or exclusively DP cells (Donor 3). h, Absolute copy number of chromosome 11 inferred from sparse whole genome sequencing from GFP positive tumours (as shown in C) arising from cells with focal p53 deletion. Red and black segments denote diploid and polyploid tumours respectively. i, Histological characterization of SP lesions (yellow outlines) observed before frank PDAC development in “Pre-tumour” KPCLOH mice. These mice are defined by the lack of clear tumour development by ultrasound. j, Age at collection of Pre-tumour (n = 7) and PDAC (n = 6) KPCLOH mice subjected to single cell genomic analysis. j, mean ± S.D. Scale bars. b, 1cm, i, 50 μm. c was repeated at least twice with similar results. Gel source data for c, see Supplementary Fig. 1. Gating strategy for g, see Supplementary Fig. 2. Source data
Extended Data Fig. 2 Linkage in cis with mutant p53 retains inducible shRNA expression following p53 LOH in the KPCCis-shRNA PDAC model and functional and histological characterization of the premalignant to malignant transition captured by stage dependent analysis of KPCLOH mice. a, Schematic of breeding, embryonic stem cell engineering, allelic configuration, and cohort generation of the KPCR172H-CIS-TRE-shRNA ESC PDAC GEMM. Note that founder LSL-p53R172H; CHC double homozygotes were utilized to ensure segregation of the conditional mutant p53 allele in cis with the collagen homing cassette. b, Representative whole mount bright field and fluorescent gross pathology of PDAC and adjacent premalignant (Pre-M) tissue in KPCR172H-Cis-TRE-shRenilla mice. c, PCR detecting recombination of the conditional p53R172H allele versus WT p53 allele (left), RIK allele (centre), and targeted CHC versus WT Col1a1 allele (right) in primary cancer cell lines derived from PDAC developing in KPCR172H-Cis-TRE-shRenilla mice (n = 3). Note the absence of WT p53 and WT Col1a1, but the maintenance of the targeted CHC allele. d, Schematic depiction of maintenance of GFP linked shRNA in cis with the conditional mutant p53R172H allele during p53 LOH and PDAC progression in KPCR172H-Cis-TRE-shRNA mice. e, Tabular results of tumours developing after injection of indicated number of DP and SP cells sorted from PDAC bearing KPCLOH mice into immunocompromised, nude mice. f, Survival curve of injected recipients detailed in e. g, Flow cytometry of GFP and Kate in a representative tumour resulting from injection of PDAC associated SP cells (left) and flow cytometry of tumours resulting from injection of 25000 DP cells sorted from 3 of 4 mice as indicated in e. Tumours were composed of either exclusively SP cells (Donor 1), predominantly SP cells (Donor 2), or exclusively DP cells (Donor 3). h, Absolute copy number of chromosome 11 inferred from sparse whole genome sequencing from GFP positive tumours (as shown in C) arising from cells with focal p53 deletion. Red and black segments denote diploid and polyploid tumours respectively. i, Histological characterization of SP lesions (yellow outlines) observed before frank PDAC development in “Pre-tumour” KPCLOH mice. These mice are defined by the lack of clear tumour development by ultrasound. j, Age at collection of Pre-tumour (n = 7) and PDAC (n = 6) KPCLOH mice subjected to single cell genomic analysis. j, mean ± S.D. Scale bars. b, 1cm, i, 50 μm. c was repeated at least twice with similar results. Gel source data for c, see Supplementary Fig. 1. Gating strategy for g, see Supplementary Fig. 2. Source dataCC BY · PubMed Central
Extended Data Fig. 3 Genome evolution following p53 inactivation is characterized by the emergence of recurrent and conserved copy number alterations that target known PDAC drivers. a, Representative genome-wide copy number profile of SP and DP cellular populations isolated from a diploid KPCLOH PDAC. b, Representative Zoom-in chromosomal views of copy number alterations acquired in KPCshp53 PDAC mice. Normalized segment values are centred around a mean value of 1 with segment values below and above 1 indictive of deletion and gain events, respectively. Vertical grey bar denotes location of PDAC driver gene. c, Frequency plot of acquired copy number events in 16 cell lines derived from PDACs arising in mice harbouring the hotspot p53 allele; R172H. Grey bar denotes chromosome 9 which encodes regulators of the TGF- β pathway. d, Frequency plot of copy number landscape of human PDAC (TCGA). Dotted black lines denote location of driver genes; TP53, KRAS, CDKN2A. Grey bar denotes location of recurrent events that are syntenic to chromosomes found altered in KPCLOH PDAC genomes. e, short hairpin mediated suppression of Smad4 in KPCcis/shRNA accelerates disease onset and is associated with worse survival. OS = overall survival, TFS = tumour free survival. Log rank p values for TFS is 0.0051 and OS is 0.0016. f, Frequency plot of copy number landscape of cancer cell lines from KPCcis-shSmad4 mice. Grey bar denotes chromosome 9 loss which is alleviated via shSmad4 perturbation. g, Genome-wide copy number profile of a tumour arising in KPCmut/shRenilla. Arrows denote distinguishing genomic alterations including deletion of chromosome 9. h, Zoom-in-view of chromosome 9 relative copy number in tumours arising in KPCmut/shSmad4 and KPCmut/shRenilla (n = 4 each).
Extended Data Fig. 3 Genome evolution following p53 inactivation is characterized by the emergence of recurrent and conserved copy number alterations that target known PDAC drivers. a, Representative genome-wide copy number profile of SP and DP cellular populations isolated from a diploid KPCLOH PDAC. b, Representative Zoom-in chromosomal views of copy number alterations acquired in KPCshp53 PDAC mice. Normalized segment values are centred around a mean value of 1 with segment values below and above 1 indictive of deletion and gain events, respectively. Vertical grey bar denotes location of PDAC driver gene. c, Frequency plot of acquired copy number events in 16 cell lines derived from PDACs arising in mice harbouring the hotspot p53 allele; R172H. Grey bar denotes chromosome 9 which encodes regulators of the TGF- β pathway. d, Frequency plot of copy number landscape of human PDAC (TCGA). Dotted black lines denote location of driver genes; TP53, KRAS, CDKN2A. Grey bar denotes location of recurrent events that are syntenic to chromosomes found altered in KPCLOH PDAC genomes. e, short hairpin mediated suppression of Smad4 in KPCcis/shRNA accelerates disease onset and is associated with worse survival. OS = overall survival, TFS = tumour free survival. Log rank p values for TFS is 0.0051 and OS is 0.0016. f, Frequency plot of copy number landscape of cancer cell lines from KPCcis-shSmad4 mice. Grey bar denotes chromosome 9 loss which is alleviated via shSmad4 perturbation. g, Genome-wide copy number profile of a tumour arising in KPCmut/shRenilla. Arrows denote distinguishing genomic alterations including deletion of chromosome 9. h, Zoom-in-view of chromosome 9 relative copy number in tumours arising in KPCmut/shSmad4 and KPCmut/shRenilla (n = 4 each).CC BY · PubMed Central
Extended Data Fig. 4 Single-cell genome analysis after PDAC development reveals two discrete genomic states distinguishing DP and SP populations. a, Hierarchal, copy number clustering heatmaps of SP and DP single cells sequenced from 4 polyploid PDAC. Colour code for lineage (L), ploidy (P), and chromosome copy number are provided. Sample annotation and number of single-cells sequenced are provided. b, Hierarchal, copy number clustering heatmaps of SP and DP single cells sequenced from 2 diploid PDAC. Colour code for lineage (L), ploidy (P), and chromosome copy number are provided. Sample annotation and number of single-cells sequenced are provided. Red arrows point to alterations acquired in DP cells that are not observed in matching SP cells. c, Genome-wide copy number profiles of representative SP single cells sequenced from PDAC samples T1 and T3 illustrating p53 null rearranged genomes. Red arrows indicate selected recurrent alterations. d, Breakpoint based phylogenetic tree of single SP (n = 130) and DP (n = 55) cells sequenced from PDAC T1. Colour codes for ploidy, lineage, and copy number information are provided. Red arrow points to split in the neighbour-joining tree demarcating the clonal sweep of SP cells. Phylogenetic distance is based on statistical considerations of breakpoint similarity/dissimilarity (Methods). SP cells constituting the sweeping clone share a clonal relationship with a False Discovery Rate (FDR) not exceeding a threshold value of t = 0.01. e, Matched genome wide copy number profiles of diploid and polyploid SP cells sequenced from T5. Red arrows indicate shared common alterations designating clonal relationships between cells.
Extended Data Fig. 4 Single-cell genome analysis after PDAC development reveals two discrete genomic states distinguishing DP and SP populations. a, Hierarchal, copy number clustering heatmaps of SP and DP single cells sequenced from 4 polyploid PDAC. Colour code for lineage (L), ploidy (P), and chromosome copy number are provided. Sample annotation and number of single-cells sequenced are provided. b, Hierarchal, copy number clustering heatmaps of SP and DP single cells sequenced from 2 diploid PDAC. Colour code for lineage (L), ploidy (P), and chromosome copy number are provided. Sample annotation and number of single-cells sequenced are provided. Red arrows point to alterations acquired in DP cells that are not observed in matching SP cells. c, Genome-wide copy number profiles of representative SP single cells sequenced from PDAC samples T1 and T3 illustrating p53 null rearranged genomes. Red arrows indicate selected recurrent alterations. d, Breakpoint based phylogenetic tree of single SP (n = 130) and DP (n = 55) cells sequenced from PDAC T1. Colour codes for ploidy, lineage, and copy number information are provided. Red arrow points to split in the neighbour-joining tree demarcating the clonal sweep of SP cells. Phylogenetic distance is based on statistical considerations of breakpoint similarity/dissimilarity (Methods). SP cells constituting the sweeping clone share a clonal relationship with a False Discovery Rate (FDR) not exceeding a threshold value of t = 0.01. e, Matched genome wide copy number profiles of diploid and polyploid SP cells sequenced from T5. Red arrows indicate shared common alterations designating clonal relationships between cells.CC BY · PubMed Central
Extended Data Fig. 5 Single-cell genome sequencing of SP cells from Pre-tumour mice reveals an intermediate evolutionary genomic state. a, Hierarchal, copy number clustering heatmaps of SP single cells sequenced from 7 Pre-tumour mice. Colour code for lineage (L), ploidy (P), and chromosome copy number are provided. Sample annotation and number of single-cells sequenced are provided. b, Bar-plot quantification of percent SP cells sequenced that were polyploid across Pre-tumour mice analysed. c, Breakpoint based phylogenetic tree of single SP cells (n = 171) sequenced from Pre-mouse P1. Colour codes for ploidy, lineage, and copy number information are provided. Phylogenetic distance is based on statistical considerations of breakpoint similarity/dissimilarity (Methods). Clone track denotes lineage that underwent genome doubling (purple). Clonal relationship between diploid and polyploid single cells is computed with a FDR not exceeding a threshold value of t = 0.01. d, Genome-wide copy number profiles of representative single cells of highly rearranged diploid Pre-SP cells (top panel) and its genetically traced polyploid counterpart (top panel) from Pre-Tumour Sample P3. Black arrows denote distinguishing copy number alterations and their breakpoint positions used in inferring phylogenetic relationships.
Extended Data Fig. 5 Single-cell genome sequencing of SP cells from Pre-tumour mice reveals an intermediate evolutionary genomic state. a, Hierarchal, copy number clustering heatmaps of SP single cells sequenced from 7 Pre-tumour mice. Colour code for lineage (L), ploidy (P), and chromosome copy number are provided. Sample annotation and number of single-cells sequenced are provided. b, Bar-plot quantification of percent SP cells sequenced that were polyploid across Pre-tumour mice analysed. c, Breakpoint based phylogenetic tree of single SP cells (n = 171) sequenced from Pre-mouse P1. Colour codes for ploidy, lineage, and copy number information are provided. Phylogenetic distance is based on statistical considerations of breakpoint similarity/dissimilarity (Methods). Clone track denotes lineage that underwent genome doubling (purple). Clonal relationship between diploid and polyploid single cells is computed with a FDR not exceeding a threshold value of t = 0.01. d, Genome-wide copy number profiles of representative single cells of highly rearranged diploid Pre-SP cells (top panel) and its genetically traced polyploid counterpart (top panel) from Pre-Tumour Sample P3. Black arrows denote distinguishing copy number alterations and their breakpoint positions used in inferring phylogenetic relationships.CC BY · PubMed Central
Extended Data Fig. 6 Phylogenetic tree inference of genome doubling timing based on copy number, minimal event distance metric. Phylogenetic reconstruction of diploid and polyploid single cells sequenced from Pre-Tumour 3 (top) and Pre-Tumour 1 (bottom) based on minimum event distance (MED) metric as described by Kauffman et al. Phylogenetic tree and associated heatmaps are depicted. Each branch corresponds to a single sequenced cell. Purple circle indicates node in the tree where diploid and polyploid cells share a branching relationship. Statistic adjacent to circle denotes branching support values calculated via 200 bootstrap resampling iterations. Algorithm provided support tree panels with bootstrap confidence statistics on branch/node relationships are provided in Supplementary Fig. 4 and 5 for Pre-Tumour 1 and 3, respectively.
Extended Data Fig. 6 Phylogenetic tree inference of genome doubling timing based on copy number, minimal event distance metric. Phylogenetic reconstruction of diploid and polyploid single cells sequenced from Pre-Tumour 3 (top) and Pre-Tumour 1 (bottom) based on minimum event distance (MED) metric as described by Kauffman et al. Phylogenetic tree and associated heatmaps are depicted. Each branch corresponds to a single sequenced cell. Purple circle indicates node in the tree where diploid and polyploid cells share a branching relationship. Statistic adjacent to circle denotes branching support values calculated via 200 bootstrap resampling iterations. Algorithm provided support tree panels with bootstrap confidence statistics on branch/node relationships are provided in Supplementary Fig. 4 and 5 for Pre-Tumour 1 and 3, respectively.CC BY · PubMed Central
Extended Data Fig. 7 In-situ genomic analysis directly links level of genome rearrangement with histopathological phenotypes during PDAC progression. a, Matched H&E and immunofluorescence for mKate/GFP in sequential sections after PDAC development in a KPCLOH mouse. White circles denote positions of SP and DP lesions with premalignant morphology and SP cells with PDAC morphology subjected to laser microdissection (LMD). b, Top panels - images of microdissected lesions noted in (a) - yellow lines denote boundaries of LMD. H&E as well as IF images are displayed. Bottom panels - corresponding genome wide copy number profiles of microdissected premalignant and malignant lesions. Red arrows denote distinguishing copy number alterations. c, Frequency plot of aggregate lesions collected by LMD and sequenced for each category. DP, n = 10; Pre SP, n = 9; SP-PDAC, n = 7. d, Matched GFP and Kate immunofluorescence and DNA FISH of chromosome 2, 9, and 10 in DP and SP cells within a focus of PDAC. Asterisks indicate cells with FISH signals consistent with polyploidy and loss of chromosome 9. e, Quantification of DNA-FISH foci in SP cells identified in KPCLOH mice before frank PDAC development (n = 5). For details of quantification, see Methods. f, DAPI based flow cytometric nuclear profiling of sorted Pre-SP cells capable of colony formation when plated at low density in-vitro (from Fig. 1j) (top-panel) and corresponding copy number profile with the sequencing imputed ploidy for the sample (bottom panel). MEF; Mouse Embryonic Fibroblasts. Scale bars a 1 mm, b 50 μm, d 10 μm. Gating strategy for f, see Supplementary Fig. 2. Source data
Extended Data Fig. 7 In-situ genomic analysis directly links level of genome rearrangement with histopathological phenotypes during PDAC progression. a, Matched H&E and immunofluorescence for mKate/GFP in sequential sections after PDAC development in a KPCLOH mouse. White circles denote positions of SP and DP lesions with premalignant morphology and SP cells with PDAC morphology subjected to laser microdissection (LMD). b, Top panels - images of microdissected lesions noted in (a) - yellow lines denote boundaries of LMD. H&E as well as IF images are displayed. Bottom panels - corresponding genome wide copy number profiles of microdissected premalignant and malignant lesions. Red arrows denote distinguishing copy number alterations. c, Frequency plot of aggregate lesions collected by LMD and sequenced for each category. DP, n = 10; Pre SP, n = 9; SP-PDAC, n = 7. d, Matched GFP and Kate immunofluorescence and DNA FISH of chromosome 2, 9, and 10 in DP and SP cells within a focus of PDAC. Asterisks indicate cells with FISH signals consistent with polyploidy and loss of chromosome 9. e, Quantification of DNA-FISH foci in SP cells identified in KPCLOH mice before frank PDAC development (n = 5). For details of quantification, see Methods. f, DAPI based flow cytometric nuclear profiling of sorted Pre-SP cells capable of colony formation when plated at low density in-vitro (from Fig. 1j) (top-panel) and corresponding copy number profile with the sequencing imputed ploidy for the sample (bottom panel). MEF; Mouse Embryonic Fibroblasts. Scale bars a 1 mm, b 50 μm, d 10 μm. Gating strategy for f, see Supplementary Fig. 2. Source dataCC BY · PubMed Central
Extended Data Fig. 8 Selective and ordered patterns of genome evolution during cancer initiation. a, Unique breakpoint patterns associated with chromosome 11 deletions identify independent p53 LOH lineages. Grey dots illustrate normalized raw read count values. Black lines illustrate segmented data. Vertical grey line denotes location of Trp53. Number of single cells sequenced representing each unique breakpoint event is provided. b, Boxplot quantification of unique lineages based on chromosome 11 deletion breakpoints in Pre-SP diploid cells (n = 7 samples) compared to PDAC-SP polyploid cells (n = 4 samples). Mann-Whitney U two-sided test of significance for number of LOH events in Pre-tumour and PDAC-SP samples, p-value = 0.03. c, Boxplot quantification of normalized read count mappability data from a census single cell genotyping approach (Methods) from PDAC-DP (n = 6), PDAC-SP (n = 6), and Pre-tumour SP (n = 7) single-cell sequencing at eGFP, Trp53, mKate, and Clp2 (control) sequences. d, Zoom in chromosomal views illustrating intra- (from within one animal) and inter- (between different animals) alteration heterogeneity of Kras gains (chromosome 6). e, Boxplot quantification of acquired copy number alterations detected in single-cell sequencing data from DP (n = 6), Pre-SP diploids (n = 7), and PDAC-SP samples (n = 6). p-value < 0.05 for all pairwise two-sided Mann-Whitney U test of significance with exact values provided in figure. f, Genome-wide illustration of independent genome doubling events observed in polyploid cells from non-tumour bearing sample P1 (left panels) and P2 (right panels). Red Arrows denote distinguishing p53 LOH events. g, Frequency plot depiction of diploid SP PDACs from the KPCLOH model (n = 7). Box plots are as defined in Fig. 1. Source data
Extended Data Fig. 8 Selective and ordered patterns of genome evolution during cancer initiation. a, Unique breakpoint patterns associated with chromosome 11 deletions identify independent p53 LOH lineages. Grey dots illustrate normalized raw read count values. Black lines illustrate segmented data. Vertical grey line denotes location of Trp53. Number of single cells sequenced representing each unique breakpoint event is provided. b, Boxplot quantification of unique lineages based on chromosome 11 deletion breakpoints in Pre-SP diploid cells (n = 7 samples) compared to PDAC-SP polyploid cells (n = 4 samples). Mann-Whitney U two-sided test of significance for number of LOH events in Pre-tumour and PDAC-SP samples, p-value = 0.03. c, Boxplot quantification of normalized read count mappability data from a census single cell genotyping approach (Methods) from PDAC-DP (n = 6), PDAC-SP (n = 6), and Pre-tumour SP (n = 7) single-cell sequencing at eGFP, Trp53, mKate, and Clp2 (control) sequences. d, Zoom in chromosomal views illustrating intra- (from within one animal) and inter- (between different animals) alteration heterogeneity of Kras gains (chromosome 6). e, Boxplot quantification of acquired copy number alterations detected in single-cell sequencing data from DP (n = 6), Pre-SP diploids (n = 7), and PDAC-SP samples (n = 6). p-value < 0.05 for all pairwise two-sided Mann-Whitney U test of significance with exact values provided in figure. f, Genome-wide illustration of independent genome doubling events observed in polyploid cells from non-tumour bearing sample P1 (left panels) and P2 (right panels). Red Arrows denote distinguishing p53 LOH events. g, Frequency plot depiction of diploid SP PDACs from the KPCLOH model (n = 7). Box plots are as defined in Fig. 1. Source dataCC BY · PubMed Central
Extended Data Fig. 9 Genomic heterogeneity of acquired gains and amplifications in KPCLOH PDAC. a, Hierarchal clustering tree of copy number profiles from tumour 2 (T2). Subpopulation of cells enriched for subclonal chromosome 18 gain are annotated on bar underneath the clustering dendrogram. Red vertical arrow indicates alteration. b, Zoom in chromosomal view of sub-clonal gains found in PDAC sequenced from KPCLOH model at single-cell resolution. Dashed red lines communicate reference copy number states for diploid and polyploid genomes, two and four, respectively. c, Histogram illustration of copy number values of YAP amplifications identified in two PDAC samples sequenced at single-cell resolution. Histograms illustrate heterogeneity of YAP amplifications. d, Overlay of three representative single-cell, genome-wide copy number profiles derived from two sequenced PDACs. Diagonal red arrow indicates amplifications identified in only a single sequenced cell. e, Genome-wide aggregate plot of all single-cells sequenced from KPCLOH PDAC T1. Thickness of blue line is proportional to percentage of cells carrying a given alteration at a copy number state. Red arrows point to alterations on chromosome 5 (gain) and 9 (deletion) found heterogeneously and homogenously, respectively.
Extended Data Fig. 9 Genomic heterogeneity of acquired gains and amplifications in KPCLOH PDAC. a, Hierarchal clustering tree of copy number profiles from tumour 2 (T2). Subpopulation of cells enriched for subclonal chromosome 18 gain are annotated on bar underneath the clustering dendrogram. Red vertical arrow indicates alteration. b, Zoom in chromosomal view of sub-clonal gains found in PDAC sequenced from KPCLOH model at single-cell resolution. Dashed red lines communicate reference copy number states for diploid and polyploid genomes, two and four, respectively. c, Histogram illustration of copy number values of YAP amplifications identified in two PDAC samples sequenced at single-cell resolution. Histograms illustrate heterogeneity of YAP amplifications. d, Overlay of three representative single-cell, genome-wide copy number profiles derived from two sequenced PDACs. Diagonal red arrow indicates amplifications identified in only a single sequenced cell. e, Genome-wide aggregate plot of all single-cells sequenced from KPCLOH PDAC T1. Thickness of blue line is proportional to percentage of cells carrying a given alteration at a copy number state. Red arrows point to alterations on chromosome 5 (gain) and 9 (deletion) found heterogeneously and homogenously, respectively.CC BY · PubMed Central
Extended Data Fig. 10 Whole genomes and targeted capture sequencing corroborates evolutionary principles in human PDAC. a, Bar-plot rendering of recurrent deletion event frequency as computed using the algorithm CORE (Methods) comparing p53 bi-allelically mutant vs. p53 mono-allelic and wild type (WT) diploid PDAC in COMPASS dataset. Cytoband of recurrent events is depicted on X-axis. Events were selected based on a threshold p-value of < 0.005. b, Frequency plot illustration of acquired copy number alterations in polyploid, p53 bi-allelically mutant human PDAC in COMPASS dataset. c, Bar-plot quantification of the frequency of recurrent CNAs (blue for deletion and red for gain) in polyploid PDAC genomes compared to diploid genomes in COMPASS dataset. Alterations are rank ordered according to CORE score. A p-value of < 0.005 was used as cut-off threshold for event inclusion. d, Empirical Cumulative Distribution Function (ECDF) for CNA event in p53 bi-allelically mutant diploid (left) and polyploid (right) datasets from MSK-IMPACT PDAC datasets. e, Bar-plot depiction of normalized homogeneity score for deletion and gain event type in the COMPASS dataset based on cell fraction and purity estimates from FACETs algorithm. f, Cox-regression survival of clinically annotated MSKCC PDACs samples. Metastatic cases, which are enriched for p53-biallelic polyploid samples have worst survival, p-val < 0.001.
Extended Data Fig. 10 Whole genomes and targeted capture sequencing corroborates evolutionary principles in human PDAC. a, Bar-plot rendering of recurrent deletion event frequency as computed using the algorithm CORE (Methods) comparing p53 bi-allelically mutant vs. p53 mono-allelic and wild type (WT) diploid PDAC in COMPASS dataset. Cytoband of recurrent events is depicted on X-axis. Events were selected based on a threshold p-value of < 0.005. b, Frequency plot illustration of acquired copy number alterations in polyploid, p53 bi-allelically mutant human PDAC in COMPASS dataset. c, Bar-plot quantification of the frequency of recurrent CNAs (blue for deletion and red for gain) in polyploid PDAC genomes compared to diploid genomes in COMPASS dataset. Alterations are rank ordered according to CORE score. A p-value of < 0.005 was used as cut-off threshold for event inclusion. d, Empirical Cumulative Distribution Function (ECDF) for CNA event in p53 bi-allelically mutant diploid (left) and polyploid (right) datasets from MSK-IMPACT PDAC datasets. e, Bar-plot depiction of normalized homogeneity score for deletion and gain event type in the COMPASS dataset based on cell fraction and purity estimates from FACETs algorithm. f, Cox-regression survival of clinically annotated MSKCC PDACs samples. Metastatic cases, which are enriched for p53-biallelic polyploid samples have worst survival, p-val < 0.001.CC BY · PubMed Central
Extended Data Fig. 11 Single cell sequencing corroborates evolutionary principles in human PDAC. a, Copy number based hierarchal clustering heatmap of 9 human PDAC (4 diploid and 5 polyploid) sequenced at single-cell genome resolution. Ploidy, Sample ID, and copy number colour schema are provided. b, Chromosomal zoom-in-view of aggregate single-cell segments from polyploid tumours illustrating heterogeneity of amplifications at MYC and KRAS, contrasting with the homogeneity of deletion events at TP53 and SMAD4. c, Histogram of copy number values across sequenced single cells for selected amplicons found in two different polyploid PDAC cases. Vertical dashed line denotes reference copy number state of 4. d, Normalized homogeneity score (Single cell homogeneity score – Methods) of copy number alterations across all polyploid single cells sequenced according to copy number state. Red colouring denotes gains and amplifications. Blue colouring denotes deletions. Grey colouring denotes references state (copy number = 4). e, Flow cytometric measurements of nuclear DNA content (e.g. ploidy) in human PDAC samples analysed at single-cell resolution. Asterisk denotes the polyploid population from which single-cells were gated and sorted. Median ploidy values of gated polyploid populations compared to sequencing inferred ploidy values are tabulated. FCS; flow cytomeric. SCS; single-cell sequencing.
Extended Data Fig. 11 Single cell sequencing corroborates evolutionary principles in human PDAC. a, Copy number based hierarchal clustering heatmap of 9 human PDAC (4 diploid and 5 polyploid) sequenced at single-cell genome resolution. Ploidy, Sample ID, and copy number colour schema are provided. b, Chromosomal zoom-in-view of aggregate single-cell segments from polyploid tumours illustrating heterogeneity of amplifications at MYC and KRAS, contrasting with the homogeneity of deletion events at TP53 and SMAD4. c, Histogram of copy number values across sequenced single cells for selected amplicons found in two different polyploid PDAC cases. Vertical dashed line denotes reference copy number state of 4. d, Normalized homogeneity score (Single cell homogeneity score – Methods) of copy number alterations across all polyploid single cells sequenced according to copy number state. Red colouring denotes gains and amplifications. Blue colouring denotes deletions. Grey colouring denotes references state (copy number = 4). e, Flow cytometric measurements of nuclear DNA content (e.g. ploidy) in human PDAC samples analysed at single-cell resolution. Asterisk denotes the polyploid population from which single-cells were gated and sorted. Median ploidy values of gated polyploid populations compared to sequencing inferred ploidy values are tabulated. FCS; flow cytomeric. SCS; single-cell sequencing.CC BY · PubMed Central
Extended Data Fig. 12 Schematic illustration of the deterministic principles governing the evolution of p53 mutant pancreatic cancer genomes. DP; double positive, p53wt/flox, SP; single positive, p53wt/−. Vertical dashed line illustrates point at which independent LOH clones emerge (first line) and selective sweeps that result in cancer progression (second line). Black arrow heads point to independent genome doubling events that can occur during transformation. Colour codes for deletion and gain CNAs are provided. Colour gradation is proportional to number of events acquired.
Extended Data Fig. 12 Schematic illustration of the deterministic principles governing the evolution of p53 mutant pancreatic cancer genomes. DP; double positive, p53wt/flox, SP; single positive, p53wt/−. Vertical dashed line illustrates point at which independent LOH clones emerge (first line) and selective sweeps that result in cancer progression (second line). Black arrow heads point to independent genome doubling events that can occur during transformation. Colour codes for deletion and gain CNAs are provided. Colour gradation is proportional to number of events acquired.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 73 references

Cited by

Sorted by citations

Items 101–200 of 223

Related

See all

Metadata

Authors
Baslan, T., Morris, J. P., Zhao, Z., Reyes, J., Ho, Y.-J., Tsanov, K. M., Bermeo, J., Tian, S., Zhang, S., Askan, G., Yavas, A., Lecomte, N., Erakky, A., Varghese, A. M., Zhang, A., Kendall, J., Ghiban, E., Chorbadjiev, L., Wu, J., Dimitrova, N., Chadalavada, K., Nanjangud, G. J., Bandlamudi, C., Gong, Y., Donoghue, M. T. A., Socci, N. D., Krasnitz, A., Notta, F., Leach, S. D., Iacobuzio-Donahue, C. A., & Lowe, S. W.
Affiliations
  • Memorial Sloan Kettering Cancer Center (US)
  • University of North Carolina at Chapel Hill (US)
  • Ontario Institute for Cancer Research (CA)
  • Cold Spring Harbor Laboratory (US)
  • Technical University of Sofia (BG)
  • Philips (United States) (US)
Corresponding author
Lowe, Scott W.
ORCID
Venue
Nature
Publisher
Springer Science and Business Media LLC
Published
2022-08-17
Volume
608
Issue
7924
Pages
795-802
ISSN
  • 0028-0836 (ISSN-L)
  • 1476-4687
DOI
10.1038/s41586-022-05082-5
PMC
PMC9402436
OpenAlex
W4292122046
Cited by
221
References
73
Citations by year
2026: 29 · 2025: 46 · 2024: 71 · 2023: 63 · 2022: 12
FWCI
20.37
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Cold Spring Harbor Laboratory Institutional Repository (Cold Spring Harbor Laboratory) (acceptedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
Funders
  • Howard Hughes Medical Institute (P30 CA008748)
  • Simons Foundation (519054)
  • Damon Runyon Cancer Research Foundation (P30 CA008748)
  • Pancreatic Cancer Action Network
  • Memorial Sloan-Kettering Cancer Center (CA008748, P30 CA008748)
  • Wellcome Trust
  • Cold Spring Harbor Laboratory
  • National Institutes of Health (CA008748, 5p30ca045508-35)
  • Philips Research Americas
  • National Cancer Institute (P01CA087497, CA008748, P30-CA045508, P01CA013106, P30CA008748)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cancer Research › Cancer Genomics and Diagnostics
  • Health Sciences › Medicine › Oncology › Cancer-related Molecular Pathways
  • Health Sciences › Medicine › Oncology › Pancreatic and Hepatic Oncology Research
Keywords
TP53 mutation, cancer genome evolution, pancreatic ductal adenocarcinoma, Trp53, genomic deletions, genome doubling, gene amplification, carcinogenesis, precancerous lesions, histological stage, single-cell genotyping, tumor heterogeneity
MeSH terms
Adenocarcinoma, Animals, Models, Genetic, Pancreatic Neoplasms, Genes, p53, Tumor Suppressor Protein p53, Genome, Gene Deletion, Disease Progression, Evolution, Molecular, Loss of Heterozygosity, Carcinoma, Pancreatic Ductal, Mice, Carcinogenesis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review