Balancing tumour proliferation and sustained cell cycle arrest through proteostasis remodelling drives immune niche compartmentalisation in breast cancer

Balancing tumour proliferation and sustained cell cycle arrest through proteostasis remodelling drives immune niche compartmentalisation in breast cancer | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Balancing tumour proliferation and sustained cell cycle arrest through proteostasis remodelling drives immune niche compartmentalisation in breast cancer View ORCID Profile Cenk Celik , View ORCID Profile Eloise Withnell , Shi Pan , Tooki Chu , View ORCID Profile John Labbadia , View ORCID Profile Maria Secrier doi: https://doi.org/10.1101/2025.01.08.632014 Cenk Celik 1 UCL Genetics Institute, Department of Genetics, Evolution and Environment, University College London , Gower Street, London, WC1E 6BT, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Cenk Celik Eloise Withnell 1 UCL Genetics Institute, Department of Genetics, Evolution and Environment, University College London , Gower Street, London, WC1E 6BT, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Eloise Withnell Shi Pan 1 UCL Genetics Institute, Department of Genetics, Evolution and Environment, University College London , Gower Street, London, WC1E 6BT, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site Tooki Chu 1 UCL Genetics Institute, Department of Genetics, Evolution and Environment, University College London , Gower Street, London, WC1E 6BT, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site John Labbadia 2 Institute of Healthy Ageing, Department of Genetics, Evolution and Environment, University College London , Gower Street, London, WC1E 6BT, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for John Labbadia Maria Secrier 1 UCL Genetics Institute, Department of Genetics, Evolution and Environment, University College London , Gower Street, London, WC1E 6BT, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Maria Secrier For correspondence: m.secrier{at}ucl.ac.uk Abstract Full Text Info/History Metrics Supplementary material Data/Code Preview PDF Abstract Uncontrolled proliferation is a hallmark of cancer, yet tumour cells can enter G0 arrest by halting the cell cycle reversibly (quiescence) or irreversibly (senescence) to survive under stress and in hostile tumour microenvironments (TME). G0 arrested cells contribute to drug tolerance, metastasis and recurrence, but their identification remains challenging due to their rarity and elusive regulatory pathways. Here, we quantify G0 arrest and proliferation decisions in single-cell and spatially profiled breast primary tumours to unveil adaptive responses driving immune compartmentalisation. We identify a genomically-constrained and prolonged G0 arrest state resembling that of dormant precursors of cancer progression. This state featured adaptive transcriptional reprogramming, including unfolded protein response stress, reduced mitochondrial translation and epithelial-mesenchymal plasticity enabled by semaphorin signalling. Spatial transcriptomics revealed G0 arrest pockets encapsulated by APOE + lipid-associated macrophages, myofibroblastic CAFs and immature perivascular cells, suggestive of an immunosuppressive niche contrasting with cytotoxic environments in proliferative hotspots and displaying distinct drug sensitivities. To facilitate future research, we provide a foundation model capturing G0 arrest and proliferation with 94% accuracy in single cell data, available at https://github.com/secrierlab/G0-LM . Our findings provide new insights into the spatial organisation of cell cycle arrest in breast cancer, highlighting the role of G0 states in tumour heterogeneity and adaptation. Introduction The unchecked proliferation of tumour cells is a hallmark of cancer 1 , driving tumour progression and treatment resistance 2 , 3 . However, proliferation can halt when cells exit the cell cycle in G0, either reversibly (quiescence or dormancy) 4 – 6 or irreversibly (senescence) 7 , often in response to intrinsic or environmental stresses. These pauses allow cancer cells to withstand hostile conditions within the tumour microenvironment (TME), thereby promoting drug tolerance and therapeutic persistence 8 , 9 . Quiescence, in particular, may either be an inherent trait of drug-tolerant persister cells or an acquired adaptive state 10 , 11 . During tumour evolution, quiescent cells facilitate immune evasion 12 , 13 and niche adaptation 14 , 15 , fostering minimal residual disease and increasing the risk of relapse 11 , 16 , 17 . Maintaining a balance between proliferation and cell cycle arrest is essential for tumour adaptation. Breast cancer is the quintessential example where cell cycle dynamics underpin clinical outcomes, with receptor-positive subtypes exhibiting slower proliferative capacities, but late recurrence through clinical dormancy, while triple-negative breast cancer (TNBC) is marked by aggressive, rapidly proliferating cells and consequently bears the worst prognosis 18 , 19 . Epigenetic mechanisms have been implicated in controlling cell cycle switches in ER+ tumours: epithelial-mesenchymal plasticity helps maintain slow proliferation of disseminated tumour cells and prevents reawakening from dormancy 20 , 21 , while chromatin remodelling aids dormancy-driven resistance to endocrine therapies 22 . Cell cycle activity in breast tumours is also influenced by interactions with the TME, which further impacts therapy outcomes. For instance, Baldominos et al. 23 used spatial transcriptomics to describe a quiescent, fibroblast-enriched niche capable of evading immune detection in TNBC. The clinical relevance of G0 arrest extends beyond breast cancer, with reports of slow cycling, revival stem cell phenotypes linked with poor prognosis and chemoresistance in colorectal cancer 24 – 26 , or quiescent PROM1 + cells implicated in recurrence and resistance in paediatric high-grade gliomas 27 . However, studying G0 arrested cells remains challenging due to their diverse phenotypes 28 , 29 , scarcity within tumours and plastic switching between proliferative and quiescent states. The often short-lived nature of this state complicates its identification, while the signalling pathways controlling G0 arrest are still poorly understood. Additionally, the interplay between intrinsic factors 30 , 31 and the tumour ecosystem in shaping G0 arrest and proliferation decisions is crucial in determining invasiveness 32 and therapy resistance 33 – 35 , but remains inadequately explored. We previously developed a robust method to quantify G0 arrest in tumours using bulk or single-cell transcriptomic data, identifying genetic modulators like CEP89 that influence G0 arrest capacity 36 . Here, we apply this method to examine G0 arrest and proliferation phenotypes across breast cancer cohorts at single-cell and spatial transcriptomics resolution. We identify a stress-induced, G0 arrested, hybrid epithelial-mesenchymal (E/M) state in primary breast tumours, which appears longer-lived, presents unfolded protein response (UPR) and is linked to immune evasion via the semaphorin/plexin axis. We further uncover spatially segregated G0 arrest and proliferative niches throughout tumours with distinct TME dynamics and drug efficacy. Finally, we develop a foundation model to facilitate the identification of G0, slow- and fast-cycling states in single cell RNA-seq breast cancer data, available at https://github.com/secrierlab/G0-LM . Our findings emphasise the pervasive role of G0 arrest in breast cancer, providing a foundation for future research into its implications to tumour adaptation, progression and the cellular diversity that shapes breast cancer phenotypes. Results The landscape of G0 arrest and proliferation decisions in primary breast tumours We collated 138,727 cells across various breast cancer subtypes, leveraging 10X Genomics single-cell RNA sequencing (scRNA-seq) datasets from three studies 37 – 39 deposited at the Curated Cancer Cell Atlas of the Weizmann Institute of Science 40 . Our atlas covered both non-invasive (including ductal carcinoma in situ [DCIS] and neoplastic) and invasive breast cancers— hormone-positive (ER+ or PR+), HER2+ and triple-negative breast cancer (TNBC)—from 43 patients, and four healthy individuals ( Fig. 1a-c ; Extended Data Fig. 1a-c). A heterogeneous landscape of cell populations was observed across these tumours and healthy tissues, including epithelial, fibroblast, myeloid, lymphoid and endothelial cells, collectively representing most cell types within the breast cancer TME ( Fig. 1e-f ; Extended Data Fig. 1b; Supplementary Table 1). Download figure Open in new tab Fig. 1 An overview of the single cell breast cancer compendium and analysis steps. a Schematic representation of the methodology employed in this study to analyse G0 arrest and proliferation states in single cell datasets. b Breast cancer subtypes are illustrated with distinct colours, where thicker bars (top) indicate the number of patients per subtype and thinner bars (bottom) represent the number of cells per tumour type. c Summary of the number of patients and corresponding cell counts in each study cohort. d Cell labels denoting malignant and normal epithelial cells inferred using inferCNV analysis. e UMAP visualisation displays the identified cell classes along with detailed cell type annotations. f Heat map of the top highly expressed genes per cell type, with annotations colour-coded to correspond with those in panel e . We used inferCNV 41 to identify malignant cells ( Fig. 1d ), harbouring a total of 432 copy number alterations, including amplifications in chromosomes 1q, 8q and 16p, as well as deletions in 3p,16q and 17p (Extended Data Fig. 1d). Genes affected by these alterations included known oncogenes 18 such as ERBB2 and MYC , and tumour suppressors BRCA1 and BRCA2 . We used our previously developed G0 arrest scoring method 36 to assess the proliferative capacity and cell cycle arrest frequency in malignant single cells (Extended Data Fig. 2a). To capture both extreme and intermediate phenotypes of proliferation and cell cycle arrest, we used stringent cut-offs defining three populations: fast-cycling (12.5%), slow-cycling (87.5%) and G0 arrested (12.5%) cells (see Methods, Fig. 2a ; Supplementary Table 2). The differential abundance test demonstrated that the G0 arrested and fast-cycling labels of the tumour cells were indeed distinct cellular states (Extended Data Fig. 2b-c), confirming the two extreme proliferation capacities of these cells. Download figure Open in new tab Fig. 2 G0 arrest and fast cycling phenotypes in primary breast cancer. a Workflow for calculating G0 arrest scores, with UMAP embedding showing malignant cells annotated by discrete cell cycle states. b Expression comparison of proliferation markers MKI67 and TOP2A across different cell cycle states. Wilcoxon rank-sum test p-values are indicated on the plots. c Gene Ontology term comparison across G0 arrested, fast-cycling, and slow-cycling states, with dots representing outliers, boxes extending to the 25th and 75th percentiles, and the median normalised enrichment scores (NES) displayed. P-value calculated using Kruskal-Wallis test. d Hallmark pathway enrichment analysis contrasting G0 arrested and fast-cycling cells. e-f Percentage of fast-cycling, slow-cycling and G0 arrested cells across breast cancer subtypes (e) and individual patients (f) . Fast-cycling cells upregulated conventional proliferation markers ( MKI67 , TOP2A ), cell cycle checkpoints, MYC/E2F targets and other pathways linked with cell proliferation in ( Fig. 2b-d ). In contrast, G0 arrested cells demonstrated p53-sustained DNA damage response ( Fig. 2c ). Thus, the distinct cell cycle states captured within these primary breast tumours bear the expected hallmarks of proliferation and quiescence, providing confidence that we had captured these switches. We also ensured that our G0 scoring method did not mistakenly capture low-quality cells 42 , as indicated by MALAT1 expression (Extended Data Fig. 2d), nor a senescent state, as all malignant cells lacked senescence markers such as β-galactosidase (Extended Data Fig. 2e). One would expect the prevalence of these cell cycle states to align with the classical breast cancer molecular subtypes. Indeed, the ER+ tumours presented the highest fraction of G0 arrested cells, whilst TNBC samples were abundant in fast-cycling cells ( Fig. 2e-f ), consistent with the clinical features of these subtypes 18 , 19 . Importantly, although G0 arrested cells were sometimes a minority, they were nevertheless present across all molecularly defined subtypes ( Fig. 2e ), highlighting this state is widespread in primary tumours and likely mediates dynamic cell adaptation. Patients with fast proliferating ER+ tumours had shorter relapse-free survival, whereas those with higher G0 arrest levels manifested longer latency before relapse (Extended Data Fig. 2f), resembling the indolent phenotypes driving metastatic latency described by Montagner et al. 43 . No significant survival differences were observed in TNBC (Extended Data Fig. 2g), indicating that our cell cycle scoring is not simply recapitulating the molecular subtypes of breast cancer but rather an independent clinically relevant feature, particularly in ER+ tumours. Genomic drivers of proliferation/G0 arrest capacity in single cells and large cancer cohorts Unsurprisingly, G0 arrest was prevalent in primary tumour cells, especially within ER+ breast cancers. Cancer cells often exit the cell cycle as an adaptive response to stress during tumour development. While many G0 arrested cells in our cohort likely represent a transient, short-lived state reflecting rapid adaptation, others may be trapped in a deeper arrest, akin to dormant precursors associated with late recurrence. To interrogate which of these states were predominantly captured in our data, we reasoned that the genetic makeup of the proliferating and G0 arrested cells, reflecting their evolutionary distance, could also reflect the timing of G0 arrest: if we are simply capturing a short-lived cell cycle arrest state, the genomic profiles of these cells should resemble those of their proliferating counterparts; conversely, locking cells in a longer-lived G0 state would pre-empt the accumulation of further genomic alterations which may be observed in the fast-cycling cells. To determine the nature of these states, we examined the genomic profiles of proliferating and G0 arrested cells, as determined by inferCNV. Surprisingly, G0 arrested cells displayed fewer global CNA alterations, especially in ER+ patients ( Fig. 3a-b , Extended Data Fig. 3), suggesting these cells may be “ stuck ” at an earlier tumour evolutionary stage, before the acquisition of several additional changes which can be observed in proliferating cells. It is therefore possible these cells have been in cell cycle arrest for longer than would be expected during a transient stress response. This aligns with our evidence that the G0 arrest score reflects the depth of cell cycle arrest 36 . Furthermore, the G0 cells also presented elevated markers of dormancy programmes previously reported in breast cancer (Extended Data Fig. 2h). Hence, by selecting the extreme G0 arrest and proliferation phenotypes, we have naturally enriched for a population exhibiting longer-lived cell cycle arrest. Download figure Open in new tab Fig. 3 Genomic differences between G0 arrested and fast-proliferating ER+ breast tumours. a Chromosomal arm-level alterations in G0 arrested ER+ subclones, highlighting shared and unique genomic changes between subclones, as inferred by SCEVAN. Differentially expressed genes between G0 arrested and fast-cycling cells at each chromosomal location are highlighted on the side. b Chromosomal arm-level alterations in fast-proliferating ER+ subclones, illustrating key genomic differences compared to G0 arrested states. c Clone tree of G0 arrested ER+ tumour cells, with subclones colour-coded according to Reactome pathway enrichment of the correspondingly altered genes. d Clone tree of fast-proliferating ER+ tumour cells, with subclones colour-coded by Reactome pathway enrichment. Copy number analysis in G0 arrested ER+ cells identified eight subclones with shared features, including losses on chromosome 3p (p26.2-p25.3) and amplification of 11q (q12.2-q13.1), linked with poor prognosis 44 ( Fig. 3a ). Subclone 2 (19% of cells) was enriched for mitochondrial respiration changes ( Fig. 3c ), consistent with reduced energy demands of G0 arrest 45 . We also noticed copy-number driven response to starvation and metabolic stress (subclone 4), as well as subclonally altered translation and angiogenesis-regulating SLIT / ROBO pathways (subclone 8), capturing other features commonly linked with quiescence. Subclone 1 showed enrichment in interferon alpha and beta signalling, linked with chemotherapy resistance in inflammatory breast cancer 46 , along with amplifications in CRABP2 and MIA genes implicated in epithelial-mesenchymal transition (EMT) and metastasis 47 , 48 . Other subclones (3, 5 and 7) were linked with cell motility and extracellular matrix remodelling, suggesting the acquisition of pre-migratory features. Fast cycling ER+ cells also exhibited interferon signalling (subclones 5 and 6), and metabolic changes (subclone 4) but differed in dominant features including HSF1-mediated stress response (subclone 2, 24% of cells) and inflammation-driven CNAs (subclones 1 and 7, 37% of cells), suggestive of a genome-mediated adaptation in response to an immunogenic environment 49 , 50 ( Fig. 3d ). TNBC showed similar genomic adaptations (Extended Data Fig. 3), with G0 cells exhibiting additional alterations in glucose metabolism and keratinisation (subclone 5), and fast-cycling cells showing interleukin signalling changes (subclone 1). Gene-level CNA analysis highlighted known enablers of proliferation in fast-cycling cells, including TP53 deletions, MYC , PTEN and EGFR amplifications 51 , 52 (Extended Data Fig. 4a). Notably, the histone acetyltransferase CREBBP , a tumour suppressor in TNBC that drives aggressive growth when lost 53 , was deleted in fast-cycling cells but amplified in G0 cells, reflecting a potential genomic switch contributing to divergent proliferation/arrest phenotypes. The chromatin modulator ASXL2, promoting proliferation in ER+ breast cancer through ERα activation 54 , was amplified in fast-cycling cells and deleted in slower proliferating tumours from TCGA (Extended Data Fig. 4b). The most prevalent alteration in G0 cells was ZFHX3 loss, associated with reduced proliferation in ER+ breast cancer 55 and prostate cancer 56 and EMT induction in mammary epithelial cell lines 57 . Intriguingly, ZFHX3 deletion also ranked among top alterations in fast-growing metastatic tumours in the MET500 cohort but not in primary tumours from TCGA (Extended Data Fig. 4b-c). Thus, ZFHX3 may mark the precursors of successful metastatic colonisation. In conclusion, key features distinguishing proliferating from G0 cells appear selectively maintained through CNAs during cancer evolution. This suggests genetics may partly drive phenotypic plasticity at the extremes of cell cycle behaviour, mirroring the “ heritable plasticity ” described by Schiffman et al. 58 . Master regulators of proliferation-arrest switches in breast cancer cells Beyond genomic alterations, gene regulatory networks (GRNs) can offer insights into tumour cell cycle regulation. We hypothesised that GRNs in G0 arrested cells could reveal conserved and subtype-specific transcription factors (TFs) driving cell cycle arrest. Using pySCENIC 59 , we identified six TF modules governing G0 arrest in invasive breast tumours ( Fig. 4 ; Extended Data Fig. 5). These modules regulate cell cycle arrest (M1-M2), anti-tumour immunity via JAK-STAT signalling and interferon stimulation 60 (M3), integrated stress responses (ISR) through ATF3, ATF4 and DDIT3 (M4) 61 , and metastasis-related processes such as EMT (M5) ( Fig. 4a-b ). Using our previously developed methodology to infer EMT macro-states in single cell data 62 , we were able to determine that these cells occupied a hybrid E/M state ( Fig 4c ). Additionally, a SOX9-regulated stemness module (M6) was also observed, which may manifest in a minor subpopulation of G0 arrested TNBC cells ( Fig 4a-b ). Download figure Open in new tab Fig. 4 Gene regulatory networks modulating G0 arrest. a The left heat map shows AUCell enrichment scores for each regulon across breast cancer subtypes, as inferred by SCENIC. Middle heat map displaying correlation scores of the top 100 transcription factors (TFs) in G0 arrested cells. Vertical (right) and horizontal (top) bars highlight modules (M1–M6) with corresponding TFs, along with motifs for selected TFs. Inferred binding motifs enriched in selected TFs are shown on the right. b UMAP plot of G0 arrested cells (left) with regulon activities of selected TFs, categorised into biological functions. c Normalised enrichment scores (NES) for DNA replication, unfolded protein response, and hallmark EMT pathways, assessed by cell cycle status, and EMT score distribution by cell cycle status; dots indicate outliers, with boxes extending from the 25th to the 75th percentiles and the median indicated. Statistical significance ( p value) was calculated using the Kruskal-Wallis test. d Heat map showing the enrichment of proteostasis pathways based on cell cycle state. e-f Ribosomal (e) and mitochondrial (f) translation enrichment, illustrating pathway activity depending on the cell cycle state (left), along with a breakdown of gene-level activity across the pathway (right). The gene-level activity is calculated as the Spearman correlation between the expression of the gene and the pathway score. g Upregulated and downregulated proteostasis pathways across cell cycle states.The bar charts on the top reflect the number of pathways up- and downregulated for each category, in red and blue, respectively. Fast-cycling cells showed five GRN modules linked to proliferation, biosynthesis (M1-M3), IFN type II signalling (M4), and stemness/dormancy regulated by SPDEF and NR2F1 (M5) 63 (Extended Data Fig. 5a). MYBL2 and FOXM1 regulation were enriched across molecular subtypes (Extended Data Fig. 5b). Subtype-specific regulons highlighted distinct cell cycle control. In ER+ tumours, G0 arrest was most prominently regulated by classical luminal markers ( GATA3 , FOXA1 ). TNBC tumours displayed higher regulon specificity scores of MYC and RUNX3 , well-known proliferation regulators (Extended Data Fig. 5d). Notably, PR+ cells exhibited stem-like characteristics in both fast cycling ( SOX6 , SOX8 ) and arrested cells ( POU5F1 / OCT4 ), whereas cell fate commitment factors ( SOX2 , SPDEF ) predominantly affected fast cycling cells in HER2+ cancers. Slow-cycling cells activated GRNs shared by proliferating and arrested cells, regulating proliferation, transcription, cell fate commitment and ISR. These cells displayed significant activity of the BCL11A , ATF3 and JUN regulons (Extended Data Fig. 5e-f). Their hybrid characteristics suggest a transitional state, enabling rapid shifts between proliferation and arrest to adapt to the tumour niche. Proteostasis pathways govern stress adaptation and G0 arrest persistence in breast cancer Proteostasis pathways are critical for maintaining protein folding, degradation and overall proteome integrity, enabling cells to adapt to stress and to balance G0 arrest and proliferation 64 . Our GRN analysis highlighted the importance of proteostasis mechanisms, such as the UPR ( Fig. 4c ), across cell cycle states. We observed a gradient of reduced expression of translation, endoplasmic reticulum (ER) proteostasis and mitochondrial proteostasis genes correlating with cell cycle rates, with the lowest activity at G0 ( Fig. 4d ). While ribosomal translation remained active across all malignant cells, mitochondrial translation showed a declining trend ( Fig. 4e-f ), with most related genes negatively correlated with cell cycle arrest ( Fig. 4f ), reflecting the lower energy demands of G0 cells. Further breakdown of proteostasis pathways unveiled an upregulation of HSP90 molecules, PI3K complexes, ER signal peptidases and peroxiredoxins in fast proliferating cancer cells ( Fig. 4g ). In contrast, G0 arrested cells downregulated retrograde transport and signal recognition particles, whereas slow-cycling cells showed a downregulation of extracellular and nuclear proteostasis. UPR activity increased in both G0 arrested and slow-cycling cells, though ISR genes were largely downregulated ( Fig. 4g ). This aligns with reports that in addition to promoting gene expression, ATF4 and ATF3 can act as transcriptional repressors under specific stresses to regulate glucose metabolism and lipid homeostasis 65 – 67 . Overall, this analysis demonstrates the remarkable capacity of G0 arrested cells to remodel their proteostasis network to achieve a “ slow flow ” state of reduced but viable activity within the tumour. Cell cycle state-dependent rewiring of tumour cell-TME interactions in single cell data Given that the TME influences the induction and maintenance of growth arrest 68 , we next interrogated whether tumour-TME interactions are rewired when cells switch between proliferation and cell cycle arrest. First, we employed NicheNet 2.0 69 to explore the influence of various ligands expressed by immune and stromal cells on cancer cell cycle decisions. TGFB1 emerged as the top ligand in G0 arrested cells, driving expression of cell cycle checkpoint regulators ( EGR1 , FOS , KLF6 ) and suppressing cyclin-dependent kinase activity ( Fig. 5a , Extended Data Fig. 6a-b), reinforcing the known role of integrin-activated TGF-β1 in inducing cell cycle arrest 70 , 71 . Notably, TGFB1 -induced expression of MGP in G0 arrested cells may enhance chemoresistance via extracellular matrix interactions 72 . Additionally, Oncostatin M (OSM)-induced TIMP1 expression could disrupt cell cycle dynamics by modulating CDK1 and CDKN1A 73 , 74 . Stromal ligands, including FGF1 , TGFB3 , IGF1 and TNF , strongly correlated with genes regulating all cell cycle phases in fast-cycling cells (Extended Data Fig. 6c-d; Supplementary Table 3). Notably, T cell-specific interferon gamma ( IFNG ) elevated HLA-A expression in these cells (Extended Data Fig. 6c-d). Download figure Open in new tab Fig. 5 Cell cycle state-specific interactions with the TME. a The top 30 ligands expressed by cells in the TME and their corresponding targets in G0 arrested cancer cells, as inferred by NicheNet. Rows highlight ligands on immune and stromal cells, alongside their activity depicted by a white-red colour gradient. The ligands are grouped manually into biologically relevant categories, annotated to the right. Columns depict downstream targets in G0 arrested cancer cells. Violin plots (top) display the expression levels of these target genes. b Ligand-receptor interactions between G0 arrested and fast-cycling cancer cells, as inferred by CellPhoneDB. Sender cells are shown on the left, receptor cells are shown on the right. The purple ligand-receptor pairs indicate interactions in which G0 arrested cells function solely as receivers, regardless of the origin of the sender. The green ligand-receptor pairs denote interactions where fast-cycling cells exclusively serve as receivers, irrespective of the origin of the sender. The grey ligand-receptor pairs mark interactions that are not unique in either category of cell types. c Ligand-receptor interactions established uniquely between cells in the TME and G0 arrested cells (left panel) or fast-cycling cells (right panel), illustrating distinct communication networks depending on the cell cycle state. The rows depict the sender cells in the TME, the columns highlight distinct ligand-receptor pairs with the receiver tumour cell, as inferred by CellPhoneDB. Using CellPhoneDB 75 , we assessed ligand-receptor interactions across different cells in the TME. G0 arrested cells showed reduced communication with other cells, including proliferating or arrested tumour cells (Extended Data Fig. 6e, Figure 5b ). However, semaphorin-plexin signalling, known to promote EMT and metastasis 76 , 77 , was uniquely observed between G0 cells and various immune cells, including CD8 T cells and tumour-associated macrophages (TAMs) ( Fig. 5c ). TAMs and DCs/macrophages also influenced hormone signalling (STS-ESR1) and ERBB4-driven growth pathways in G0 cells, respectively ( Fig. 5c ; Supplementary Table 3). Conversely, in fast-cycling cells we observed CAF-induced activation of the Wnt pathway alongside myeloid stimulation of EGFR and ERBB2 ( HER2 ), all known to drive proliferation 78 , endothelial-linked Ephrin signalling, implicated in cell adhesion and migration 79 , and increased communication with CD4/CD8 T cells ( Fig. 5c ). The LGALS9 - P4HB pairing between fast-cycling cells and TAMs has been linked with immunosuppression 80 . Retinoic acid signalling appeared in both G0 ( ALDH1A3 - RARA ) and fast-cycling ( ALDH1A1 - RXRB , ALDH1A3 - RARG ) cells, though its role may vary, supporting stemness maintenance 24 in G0 and proliferation via the non-canonical Wnt 81 in fast-cycling cells ( Fig. 5c ). Shared interactions involving stromal collagens, TGFB3 and tumour cell-specific integrins (α2β1, αVβ6, αVβ8), mostly mediating interactions with CAFs, may regulate ECM organisation in both states (Extended Data Fig. 6e-f). To assess the potential for CD8 T cell exhaustion, we analysed naive, effector memory, pre-exhausted, exhausted and short-lived exhausted cell (SLEC) subsets (Extended Data Fig. 7a-d). In line with our previous observations of increased immune reactivity, fast-cycling cells presented more interactions with exhausted T cells than G0 cells (Extended Data Fig. 7f), mediated by cytokines like ANGPTL4 , shown to reprogram CD8 T cell metabolism 82 (Extended Data Fig. 7e), and TIGIT - NECTIN2 signalling (Extended Data Fig. 7f). However, T cell exhaustion appeared more strongly linked with DCs and TAMs (Extended Data Fig. 7f). Spatial organisation of proliferation and G0 arrest phenotypes unveils extensive niche remodelling While single-cell data offer insights into the intrinsic regulation of proliferation and cell cycle arrest in cancer cells, the spatial emergence of these states within tumour tissue remains unclear, despite being essential for understanding how these dynamics drive tumour growth. To investigate this, we examined G0 arrest and proliferation phenotypes using spatial transcriptomics data from 12 breast cancer 10x Visium slides (see Methods). Using our SpottedPy method 83 , we mapped G0 arrest and fast-cycling tumour hotspots and spatial interactions within the TME (first panel in Fig. 6a , Extended Data Fig. 8a). Figure 6a illustrates the spatial organisation of these hotspots in two patient tissues (additional patients in Extended Data Fig. 8a). While most of the tissue was composed of tumour cells with intermediate cycling capacity, the extreme proliferation or cell cycle arrest areas were dispersed throughout the tissue rather than collocated in a single region, suggestive of the tumour’s ability to fine-tune proliferation rates locally to adapt to varying stressors. Download figure Open in new tab Fig. 6 Spatial organisation of G0 arrest and high proliferation hotspots in breast cancer. a Spatial transcriptomics images showing hotspots identified by SpottedPy in two patient samples. First panel: Hotspots are colour-coded by cell cycle state: green (fast-cycling) and purple (G0 arrested), while grey marks the remaining tumour spots. Second panel: Overlap between G0 arrested tumour hotspots, hybrid EMT hotspots and semaphorin signalling hotspots. Third panel: Overlap between G0 arrest hotspots, proteostasis signalling and immature perivascular-like cell (PVL) hotspots. Fourth panel: Overlap between G0 arrest hotspots and APOE + lipid-associated TAMs. b-d Bubble plots representing relative median Euclidean distances to G0 arrest cancer hotspots (left) or fast cycling hotspots (right) for (b) selected cancer hallmarks and pathways, (c) proteostasis pathways and (d) immune and stromal cell hotspots. Circles with a black contour depict significant p-values (i.e. hotspots that are significantly closer to either G0 or fast cycling pockets). e Ligand-receptor interaction analysis between APOE + lipid-associated TAMs, myCAFs and immature PVL hotspots as senders and tumour cells as receivers, in relation to G0 arrested and fast-cycling hotspots. Purple indicates ligand-receptor interactions unique to G0 arrested tumour cells, green indicates those unique to fast cycling cells. f Drug-niche hotspots identified in the same two patients from a, showcasing areas within the tumour where the drug is predicted to be most effective and their overlap with G0 arrest cancer hotspots (left, shown for pegcetacoplan) or fast cycling hotspots (right, shown for daunorubicin citrate). g Distances between drug-effective regions and either G0 arrest or fast-cycling hotspots. The drugs shown were found to be differentially associated with G0 arrest or fast cycling areas within the tumour by drug2cell. h Box plots depicting tumour-drug area distances for the two drugs most strongly correlated with G0 arrest and fast-cycling niches. We next analysed the spatial proximity of G0 arrest/fast-cycling hotspots and overlap with key signatures, including EMT, hypoxia, angiogenesis, proteostasis, apoptosis and various cell types within the TME. We found that tumour spots in a hybrid E/M state 62 (and to a lesser extent a fully mesenchymal state) were significantly closer to G0 arrest hotspots ( Fig. 6a second panel, 6b), although some intra- and inter-patient variability could be observed (Extended Data Fig. 8b). These observations suggest an interplay between G0 arrest and E/M transformation akin to that observed in the single cell data. In addition, G0 arrest hotspots tended to be proximal to pockets of hypoxia and angiogenesis, but lacked apoptosis hallmarks, indicating they were not sites of programmed cell death ( Fig. 6b , Extended Data Fig. 8b). Furthermore, we revisited the proteostasis pathways observed in single-cell data and found collagen processing and folding in the ER, as well as extracellular proteostasis activity in the vicinity of G0 arrest hotspots ( Fig 6a-c ). Given that mediating collagen production is a primary function of the UPR, our data suggest that G0 cells remodel their proteostasis network to influence not only intrinsic proteome homeostasis, but also the structure/composition of their surrounding environment 81 . Interestingly, G0 arrest hotspots appeared encapsulated within APOE + lipid-associated TAM pockets and were near immature perivascular cells (PVL) and myofibroblastic cancer-associated fibroblasts (myCAFs), while remaining distant from lymphoid cells (T, NK and B cells) and monocytes ( Fig. 6a,d ). This pattern suggests a tumour protective niche that may specifically enable the immune evasion of G0 arrested cancer cells, while the fast-proliferating areas are more prone to cytotoxic T cell recognition. We noted enriched SEMA4D - PLXNB2 interactions in the neighbourhood of several G0 arrest hotspots, corroborating the modulatory role of G0 states by semaphorin signalling observed in single-cell data ( Fig. 6a-b , e). Further spatial ligand-receptor analysis between APOE + lipid-associated TAMs and G0 arrested cells ( Fig. 6e ) confirmed enriched SEMA4D - PLNXB2 pairing ( Fig. 6e , Extended Data Fig. 9a). Plexin B2 interactions with semaphorin IV have been implicated in tumour invasion through cytoskeletal reorganisation 84 , 85 and have also been reported by Borrelli et al. 86 to play a key role in metastatic colonisation by triggering an epithelialization switch in the metastatic precursor cells as a requirement for adaptation in the new environment. As our analysis suggests an EMT rather than mesenchymal-epithelial transition (MET) activation in these areas, this could imply a pleiotropic role for semaphorin IV-plexin B2 signalling capable of toggling between EMT and MET depending on the context (primary tumour/metastasis). Therefore, the G0 arrest hotspots display some characteristics akin to those of seeder cells of metastases, suggestive of areas that are actively adapting to microenvironmental pressures. We also found the JAG1 - CD46 ligand-receptor pair enriched, indicating a potential activation of the complement response in the G0 arrested cells ( Fig. 6e ). Indeed, significant activity of the complement pathway was detected in the proximity of G0 arrest hotspots ( Fig. 6b ). In addition, collagen/integrin signalling, particularly between the collagen 3/6 families and integrin ɑ11β1, was prevalent in mediating interactions between G0 arrested cells, APOE + lipid-associated TAMs, myCAFs and immature perivascular cells ( Fig 6b , Extended Data Fig. 9). Integrin ɑ11β1 has been shown to regulate CAF stiffness and promote metastatic potential in non-small cell lung cancer 87 . In conjunction with the enriched extracellular proteostasis and collagen processing signals observed proximal to the G0 cancer niche ( Fig. 6c ), these analyses suggest that the niche may be extensively remodelled, both through a reorganisation of the extracellular matrix and of the vasculature surrounding the G0 cancer cells. This may be enabled through the co-option of APOE + lipid-associated TAMs and myCAFs which may be secreting factors to enable the remodelling. Another tantalising hypothesis is that the G0 cancer cells could be secreting some of these factors, as shown in our previous work 87 reporting multiple collagen and integrin molecules to be potentially secreted by quiescent cells. The extensive intrinsic and extrinsic remodelling of G0 cells and their immediate environment may promote intravasation into the neighbouring blood vessels, which are likely being actively expanded as suggested by the presence of immature perivascular cells. Segregation of G0 arrest and proliferative areas suggests differential responses to drugs The observed spatial segregation between G0 arrest and proliferative areas might inform therapy response. Using drug2cell 88 , which predicts targets of drugs from the ChEMBL database based on their single cell expression, we identified molecular hotspots with distinct drug sensitivities. This analysis revealed that proliferative niches prominently overlapped with daunorubicin citrate-responsive regions ( Fig. 6f ). Daunorubicin, a DNA replication inhibitor commonly used in leukaemia, targets vascular mimicry structures–an adaptive mechanism supporting tumour growth and survival. Its ability to suppress mediators such as TGFB1 and MMP2 89 has potential utility in breast cancer, particularly when combined with tamoxifen to eliminate both cancer cells and stem cells 90 . In contrast, G0 arrest hotspots were associated with potential efficacy for pegcetacoplan, a complement pathway C5 inhibitor. This adds to our earlier findings that G0 niches display reduced complement pathway activity ( Fig. 6e ). Pegcetacoplan’s role in modulating inflammation 91 suggests it could counter immune evasion mechanisms within quiescent tumour niches. These findings underscore the spatial heterogeneity of drug-niche interactions and propose potential combination therapies targeting different tumour compartments. In summary, our analysis delineates distinct quiescent, potentially pre-migratory niches within primary tumours, rich in tumour-promoting signals associated with local stress adaptation. Surprisingly, these niches appear to be dispersed across the tumour, rather than clustering at the leading edge, hinting at proliferative and G0-hybrid E/M patterns that fluctuate throughout the tissue as a response to intrinsic and extrinsic stress factors. A foundation model for G0 arrest and proliferation in single cells To generalise the identification of G0 arrest, slow-cycling, and fast-cycling states in other single-cell datasets, we employed a foundation model framework similar to ones we have developed previously 92 to train a classifier of these states (Extended Data Fig. 10a). The proposed single-cell Large Language Model (scLLM), G0-LM, demonstrated robust performance in classifying G0 arrested, slow-cycling and fast-cycling states, achieving an average AUROC score of 0.94 across these categories (Extended Data Fig. 10b), underscoring the model’s efficacy in distinguishing biologically distinct cell cycle states. Additionally, the embeddings generated by G0-LM displayed notable patterns consistent with biological expectations. Despite being trained solely for classification, the embeddings revealed a continuum, showing a gradual transition of cells from fast-cycling to slow-cycling and ultimately into the G0 arrest state (Extended Data Fig. 10c-d). This highlights the model’s capacity to capture progressive state changes. The ability of G0-LM, powered by scLLM, to delineate such trends suggests a high sensitivity to the complex gene expression landscapes associated with the G0 arrest phenotype. To enable future analyses of G0, slow and fast proliferation phenotypes in single cell breast cancer data, we provide the G0-LM model at https://github.com/secrierlab/G0-LM . Discussion Our study reveals the interplay of cell cycle states within primary breast tumours, identifying proliferative and quiescent populations with distinct genetic and transcriptional profiles shaped by the TME. We observed a subtype-dependent prevalence of these states: ER+ tumours harboured a greater proportion of G0 arrested cells, while TNBC samples were rich in fast-cycling cells ( Fig. 2d ), reflecting known differences in clinical outcomes. G0 cells in ER+ tumours had fewer CNAs and expressed dormancy markers, suggesting they may be stably arrested at an earlier evolutionary stage rather than simply responding to transient stress. In contrast, fast proliferating cells upregulated the dormancy marker NR2F1 , indicating the potential to develop a more agile, plastic and epigenetically-driven dormant state, not observed here and likely only activated by dormancy-promoting immune cells 61 or upon treatment. Both cell states upregulated stemness modules, which may be decoupled from the strict cell cycle programme, although enacted through different transcription factors. The identification of a slow-cycling population bridging proliferative and quiescent characteristics suggests these cells could facilitate dynamic state shifts in response to TME pressures, enhancing adaptability. G0 arrested cells were often in a hybrid E/M state, aligning with the “ go-or-grow ” model of cancer 27 , 93 – 95 . They also exhibited UPR and reduced mitochondrial translation, allowing them to remain in a viable but metabolically reduced state over prolonged periods. It has been shown that mitochondrial stress couples activation of the ISR with reduced mitochondrial translation 96 , 97 . This is highly consistent with our observations in G0 arrested cells, suggesting that mitochondrial remodelling and metabolic adaptation may be what underlies activation of ISR-associated TFs (i.e. ATF4, ATF3 and DDIT3/CHOP). Since cytosolic ribosomal gene levels remained stable, translation may be reduced through the activation of the ISR, allowing G0 arrested cells to be “ primed ” for an increase in protein synthesis in case exit from G0 is required 98 . Our study also highlights the role of extrinsic factors in sustaining cell cycle arrest. Spatial profiling of breast tumours revealed patient-specific activity hotspots, identifying a G0 arrest cancer niche enriched in myCAFs and APOE + lipid-associated TAMs, primed by semaphorin-plexin signalling and perivascular remodelling. Extracellular proteostasis signalling within these niches suggests active communication and remodelling of the ECM and TME through multiple collagen-integrin interactions. Due to imperfect resolution of the Visium data, the source of secreted signals remains unclear, but two scenarios are plausible: (1) the APOE + lipid-associated TAMs and myCAFs may send signals, promoting an arrested, partially EMT-transformed state in tumour cells, or (2) tumour cells themselves may secrete signals to remodel the ECM and vasculature, which is developing with the help of the immature pericytes, and potentially co-opting TAMs and myCAFs to evade immune recognition. This spatial pattern suggests G0 arrested cells in primary breast tumours may already acquire traits enabling them to occupy pre-metastatic niches, with APOE + lipid-associated TAMs and myCAFs priming these cells for migration and colonisation. These interactions sustain quiescence and remodel the niche, potentially facilitating intravasation into the nascent vasculature 86 , 99 , 100 . These findings highlight the potential of therapies targeting cancer-TME interactions to destabilise quiescent niches, particularly in ER+ tumours with high proportions of G0 arrested cells. While our findings align partly with Baldominos et al. 23 , who noted a hypoxic quiescent niche containing immune-suppressive fibroblasts and exhausted T cells in TNBC tumours, we did not observe T cells near G0 hotspots. This is likely because Baldominos et al. 23 . specifically selected T cell-resistant tumour cells, which would be more likely to be rarer quiescent cells situated within the cytotoxic and exhausted proliferative hotspots observed in our analyses. Thus, our analysis highlights a different G0 arrest phenotype within the developing breast tumour that is likely not simply epigenetically driven as a reaction to T cell attack. In our single cell and spatial analyses, we find some hints that G0 arrested cells may exhibit markers of a long-lived, dormant-like phenotype with hybrid E/M and tumour invasion/migration features. This indicates that our analysis captures not only rapid adaptation to changing microenvironmental stress but may also pinpoint tumour regions likely to harbour precursors of metastasis and recurrence. While not all G0 arrested cells become successful persister cells or metastatic seeders, our method offers a systematic “ first scan ” for potential pre-migratory areas in spatial data. However, fast proliferating tumours show worse outcomes in breast cancer, and Lucas et al. 101 recently suggested that high-proliferating clones in single-cell DNA-sequencing datasets have increased metastatic potential in lung cancer. While Lucas et al. 101 did not distinguish G0 cells as a separate category and metastatic potential may vary in breast cancer, we cannot rule out that dormancy-priming by NR2F1 in fast-cycling cells may enhance their ability to switch into this programme under favourable microenvironmental conditions or therapy. This plasticity may outcompete the more constrained plasticity of metabolically inert G0 cells. The positioning of fast-cycling cells within cytotoxic and exhausted niches suggests that these areas may respond to immunotherapy, though this also emphasises the difficulty in eradicating the entire tumour, as immunotherapy is unlikely to be effective in G0 cancer pockets. Together, these analyses lay the groundwork for future studies, once several limitations are addressed. Our definition of G0 arrest relies on a conservative threshold on gene expression scores, which could be adjusted to capture gradual state changes, especially in spatial data. However, the dynamic emergence and extinction of cancer niches cannot be inferred from a single time point. Future studies should use longitudinally profiled samples, ideally both from primary and matched metastatic tumours, to gain insights into how these niches evolve over time, their stability and potential to generate migratory precursors to metastasis. Secondly, while we used CNAs to estimate the relative timing of G0 arrest in relation to proliferative cells, we acknowledge that transcriptomic copy number estimation can be prone to error and susceptible to the reference employed. While identified genomic changes previously implicated in cell proliferation and breast cancer provide confidence in these findings, future studies should combine single-cell DNA/RNA sequencing with lineage tracing to draw more robust conclusions. Thirdly, this study has not explored epigenetic or metabolic adaptations required to stabilise G0 arrest. Although seminal studies such as by Rosano et al. 22 noted that long-term quiescence is primarily driven by non-genetic adaptations to endocrine therapies, our analyses suggest that, at least in a pre-treatment setting, genomic alterations may predispose tumour cells to favour either fast proliferation or prolonged arrest. Further studies employing methylation, chromatin conformation and metabolomics are needed to explore the non-genetic mechanisms contributing to the evolutionary role of G0 arrest in cancer. We also acknowledge that the limited number of spatial transcriptomics samples may not fully capture the spatial variability in breast cancer, and the lack of single-cell resolution in the Visium platform complicates the inference of cell-specific signals, despite our deconvolution efforts. Lastly, while we identified intermediate (slow-cycling) cells as potential mediators of state transitions, further research is required to elucidate the molecular mechanisms behind these transitions and the TME factors involved. In conclusion, our study advances the understanding of cell cycle regulation in breast tumours by revealing cell-intrinsic and extrinsic factors that shape proliferative dynamics. We identify a long-lived population of G0 arrested cancer cells with genetically-constrained transcriptional plasticity, alongside spatially organised niches characterised by immunosuppressive and cytotoxic microenvironments. These insights are particularly relevant for ER+ cancers, where targeting quiescent populations and disrupting dormancy-associated signalling networks, such as semaphorin-mediated pathways, may reduce recurrence risk. Additionally, our spatial mapping of proliferation and G0 arrest hotspots offers a foundation for tailoring therapies to the unique microenvironmental contexts of each niche. Methods Data collection, integration and cell type annotation Single-cell RNA sequencing matrices for breast cancer, covering both non-invasive and invasive breast carcinoma samples, were sourced from the Curated Cancer Cell Atlas ( https://www.weizmann.ac.il/sites/3CA/ ). These datasets underwent standardisation based on droplet-based studies conducted by Gao et al. 37 , Qian et al. 39 , and Pal et al. 38 . To preprocess and integrate the samples, the standard Seurat pipeline with SCTransform() v2 normalisation 102 was employed, wherein low-quality cells (with mitochondrial transcript percentage > 20) were removed and cell cycle phases were regressed out. Subsequently, the Leiden community detection algorithm was applied with a resolution of 1.2 on the 29 principal components of the integrated data, followed by uniform manifold approximation projection (UMAP) for dimension reduction. Cell type annotation was performed at two resolutions. Initially, main cell types were identified using cell type signatures from PanglaoDB 103 and ScType R package 104 . Then, for further refinement, the CELL type iDentification tool on the Deeply Integrated Human Single-Cell Omics platform 105 was employed, by utilising averaged expressions of Leiden communities to accurately annotate cell types. Additionally, the top two differentially expressed genes of the annotated Leiden communities were calculated on the integrated data with a log2 fold-change of 0.25 and minimum percentage of 0.25 using the likelihood-ratio test for single-cell feature expression 106 . To identify malignant cells within the epithelial cluster, the inferCNV R tool 41 was utilised to infer copy number alterations (CNAs) from gene expression data. InferCNV was configured with the following parameters: a cutoff of 0.1, deemed suitable for 3’ sequencing droplet-based assays, denoising = TRUE , HMM_type = “i6” for Hidden Markov Model (HMM)-based CNA prediction, and analysis_mode = “subclusters” to identify clonal CNAs. A heat map depicting the relative expression intensities across each chromosome was generated, revealing regions of the tumour genome that were over- or under-abundant compared to normal cells. Differential gene expression analysis We conducted differential gene expression analysis on the integrated dataset using Seurat’s FindAllMarkers() function on the RNA layer of the Seurat object with parameters min.pct set to 0.5 and logfc.threshold of 0.25 for only positively upregulated genes for both cell type annotations and cell cycle states. Quantification of G0 arrest in malignant single cells and spatial spots First, based on inferred CNAs, the malignant cells were subsetted to implement a combined z-score, as previously defined by us in Wiecek et al. 36 . Z-scores of the 27 upregulated and 112 downregulated genes in the G0 arrest signature were then calculated using the GSVA R package. Next, a final G0 arrest score was obtained by subtracting the two scores. Finally, the G0 arrest scores were used for the discretisation of malignant cells into three categories: (1) G0 arrested, (2) fast-cycling and (3) slow-cycling. The cut off values of −3.83 and 3.03 were estimated by employing the maxstat R package on the G0 arrest scores, which utilises maximally selected rank statistics with several p-value approximation 107 . The single-cell differential abundance testing was then performed on the cells with two extreme proliferating characteristics, and the neighbourhoods on a nearest neighbour graph were identified by employing the miloR R package 108 . In spatial transcriptomics, z-scores for G0 arrest signatures within tumour spots were calculated similarly. Given the heterogeneity within spatial spots, we defined cut-off values using the 25 th and 75 th percentiles of G0 arrest scores. Tumour spots with z-scores below the 25 th percentile were classified as fast-cycling, while those above the 75 th percentile were classified as quiescent. This quantile-based approach accounted for the spatial variability inherent in tumour microenvironments and maintained consistency with the single-cell classification criteria. Genomic alterations in primary tumour cells We used SCEVAN R package 109 to investigate tumour clonality, Reactome Pathway enrichment of each subclone and oncoprint-like plots highlighting clone-specific alterations in fast-cycling and G0 arrested tumours. Genomic associations with G0 arrest/proliferation in single cells and large cancer cohorts Gene-level copy number changes called by inferCNV in the single cell data were compared between G0 arrested and fast-cycling cell populations, and changes significantly enriched in either category were identified using a Fisher’s exact test, with Benjamini-Hochberg adjustment of p-values for multiple testing. RNA sequencing data and gene-level copy number data from primary tumours from the TCGA Pan-Cancer Atlas were downloaded using TCGAbiolinks R package 110 . FPKM-normalised RNA-sequencing data and segmented copy number data from metastatic tumours were downloaded from the MAGI MET500 server ( https://met500.med.umich.edu/downloadMet500DataSets ), along with processed data on recurrent molecular aberrations from the MET500 study 111 . To map copy number segments to genes, the GenomicRanges R package was used to convert segment data into genomic ranges, which were annotated with gene-level data using the annotatr R package. A list of 723 known cancer driver genes was sourced from the COSMIC database. Each sample was assigned a G0 score as described in Wiecek et al. 36 using our previously developed QuieScore R package ( https://github.com/dkornai/QuieScore ). Tumours with a G0 score above 1.5 were classified as slow-cycling, while those with scores below −1.5 were considered fast-cycling. Logistic regression was employed to predict high levels of proliferation versus G0 arrest in TCGA and MET500 cohorts, with model selection performed using the stepAIC() function from the MASS R package. For model building, samples were split into training (75%) and testing (25%) datasets using the createDataPartition() function. A 10-fold cross-validation procedure was implemented using the trainControl() function from the caret R package. The logistic regression model trained on the MET500 dataset achieved an accuracy of 62.5% with an AUC of 0.66, whilst the model trained on TCGA dataset has an accuracy of 72.7%, with an AUC of 0.78. Gene enrichment analysis Hallmark gene set enrichment analysis on single cells was performed using the escape R package. Visualisation of the heatmap was done with the dittoSeq R package. Gene Ontology (GO) terms related to DNA replication and the unfolded protein response, as well as the HALLMARK_P53_PATHWAY and HALLMARK_EPITHELIAL_MESENCHYMAL_TRANSITION gene sets from MSigDB 112 were assessed by scDECAF R package 113 . The top GO terms associated with the gene regulatory network modules were identified using the enrichR R package with the GO_Biological_Process_2023 gene sets. The EMT score was calculated as described in Malagoli-Tagliazucchi et al. 62 . Endoplasmic reticulum stress-related pathways scores were computed using the irGSEA R package 114 , with curated gene signature lists as input. Dormancy signatures were taken from Ren et al. 114 , Ruth et al. 115 , Cheng et al. 116 , Kim et al. 117 and Montagner et al. 43 . Cell-cell interactions TME-receiver cell interactions were assessed by two methods (1) Estimation of putative ligands and downstream genes in the receiver population The estimation of ligands in the TME potentially driving gene expression in G0 arrest or fast-cycling cells involved utilising NicheNet 2.0 R package 69 . Differentially expressed genes with a log2 fold-change greater than 0.5 and an adjusted p-value lesser than 0.05 were considered for ligand prioritisation in the receiver population. Background genes were defined as genes expressed in the receiver population (at least 5% of the cells). Ligands were not constrained to be expressed solely by putative sender cells within the dataset, allowing for the assessment of TME ligands produced by non-immune cells such as tumour cells or the stromal compartment. The prioritisation of ligands was based on the area under the precision-recall (AUPR) scores, indicating whether a gene significantly correlated with their respective ligand-target regulatory potential. The ligand-target regulatory potential reflects the strength of association between a ligand and a target gene based on prior knowledge, presented in NicheNet’s integrated weighted networks. Finally, the predicted interaction network map was plotted using the DiagrammeR R package. (2) Estimation of ligand-receptor pairs To infer ligand-receptor pairs between the TME and receiver cells regardless of their effects on downstream gene expression, the differentially expressed interaction module of CellPhoneDB 75 was employed, with a threshold of 0.1 percent gene expression in each cell. This method utilises a list of differentially expressed genes sorted by adjusted p-values calculated by Seurat’s FindAllMarkers() function for each cell type. In spatial transcriptomics, we applied the LIANA+ package’s CellPhoneDB module 118 to identify spatially relevant ligand-receptor pairs. Gene regulatory networks Single-cell regulatory network inference and clustering (pySCENIC) 59 was employed to infer gene regulatory networks from single cells across G0 arrested, fast-cycling and slow-cycling tumours. The inference of gene regulatory networks was based on hg38 v10nr_clust databases , focusing on regions around 10 kb, 100 bp downstream, and 500 bp upstream of transcription start sites (TSS), using score and ranking matrices provided by the SCENIC pipeline. The top 100 GRNs were grouped by cancer subtype for AUCell scoring, and Spearman’s correlation was applied to identify conserved GRN modules. Binding motifs for selected transcription factors were inferred from the motifs-v10nr_clust-nr.hgnc-m0.001-o0.0.tbl database. Deconvolution of T cell states A comprehensive human reference T cell atlas 119 was used to query the T cell states. For projecting CD8 T cell subtypes onto single cells, ProjecTILs R package 120 was employed. Spatial analysis Breast cancer Visium slides were sourced from Barkley et al. 121 (slides 0-2), 10X Genomics 122 (slides 3-5) and Wu et al. 35 (slides 6-12). These data sets were merged into a unified AnnData python format for analysis. Pre-processing, normalisation and cell type deconvolution were carried out as conducted in previous studies 83 . A total of 32,845 spatially profiled spots were analysed. Spots were retained if they showed at least 100 genes with a minimum count of 1 per cell, had over 250 counts per spot, and less than 20% mitochondrial counts per cell. Cellular deconvolution was performed using cell2location 123 with a scRNA-seq breast cancer dataset from Wu et al. 35 , comprising 100,064 cells from 26 patients across 21 cell types. These included cancer epithelial cells (basal, cycling, Her2, LumA, LumB), B-cells (naïve and memory), CAF subtypes (myCAF-like and iCAF-like), perivascular-like cells (immature, cycling, differentiated), T-cells (cycling, CD 4, CD 8 T cells), cycling myeloid cells, dendritic cells, endothelial cells (ACKR1, CXCL12, RGS5), LYVE1-expressing lymphatic endothelial cells, luminal progenitors, mature luminal cells, macrophages, monocytes, myoepithelial cells, NK cells, NKT cells, and plasmablasts. Tumour cells in the spatial transcriptomics dataset were identified using the STARCH python package, which infers copy number alterations 124 . Hotspot analysis was conducted using SpottedPy python package 83 , where the EMT, hypoxia and UPR and angiogenic hallmarks 35 , as well as hybrid E/M and semaphorin signalling signatures were scored using scanpy’s score_genes() function. G0 arrest and fast-cycling tumour spots were calculated using the gene sets described in Wiecek et al. 36 . Drug-niche interactions To interrogate drug-niche interactions, we utilised drug2cell 87 to score candidate drugs and their targets with differential correlations to drug sensitivity by leveraging gene expressions in each cell state, then employed our SpottedPy 83 method to evaluate the spatial distances between drug niches and cell cycle states. The large language model The G0-LM model is an advanced architecture comprising three specialised networks, adapted from the scBERT framework 125 to address the specific challenges of G0 arrest in single-cell analysis. Its two-stage training pipeline bridges the gap between large pre-trained datasets and the specialised requirements of G0 arrest data 92 . In the first stage, the model’s embedding space is refined using a Low-Rank Adaptation (LoRA) approach 126 , a Parameter Efficient Fine-Tuning (PEFT) strategy that enables adaptation to scRNA-seq data while reducing overfitting risks. The G0-LM architecture follows that of EMT-LM 92 and incorporates a multiplication layer that leverages fine-tuned language models to amplify signals from highly expressed genes. Attention vectors generated by these models act as weighted enhancements of raw gene expression data, selectively highlighting relevant cellular features. To optimise model focus, genes with low expression or peripheral relevance to the EMT process are systematically filtered, concentrating learning on critical pathways. For classification, a Multilayer Perceptron (MLP) serves as the final module to yield predictions. Ultimately, this fusion network provides a computational model of cell states under G0 arrest, distilling complex gene expression profiles into accurate cell state outputs. Statistical analysis Groups were compared using a two-sided Student’s t -test, Wilcoxon rank-sum test, ANOVA or Kruskal-Wallis test, as appropriate. Adjustments for multiple testing were made using the Benjamini-Hochberg method when necessary. False discovery rates (FDR) were calculated by taking the −log10 of the adjusted p -values. Graphs were generated using either the ggplot2 and ggpubr R packages or the matplotlib and seaborn python libraries. Data availability The single-cell RNA sequencing datasets used in this study are publicly available through the Curated Cancer Cell Atlas of the Weizmann Institute of Science ( https://www.weizmann.ac.il/sites/3CA/breast ), with raw files accessible via the Gene Expression Omnibus (GEO) under accession numbers GSE148673 38 and GSE161529 39 , and through the ArrayExpress database of EMBL-EBI under accession number E-MTAB-8107 37 . Pre-processed and annotated single-cell Seurat objects are available on Zenodo (DOI: 10.5281/zenodo.14001194). Gene sets for calculating G0 arrest cell cycle score are available at https://github.com/secrierlab/CancerG0Arrest . The human CD8 T cell reference dataset is provided in R dataset format and can be downloaded from Figshare (DOI: 10.6084/m9.figshare.41414556). Processed spatial transcriptomics data is also publicly available on Zenodo (DOI: 10.5281/zenodo.10371890). Code availability All scripts used in this study are publicly available at https://github.com/secrierlab/G0-breast-cancer-atlas . The G0-LM model is available at https://github.com/secrierlab/G0-LM . Authors’ contributions M.S. designed and supervised the study. C.C. and M.S. designed the experiments and interpreted the data. C.C. performed all analyses, except for the following: E.W. developed SpottedPy, processed the spatial data and conducted the initial SpottedPy analyses; T.C. performed logistic regression analyses of genomic alterations in fast/slow proliferating tumours in bulk datasets from TCGA and MET500, as well as enrichment analyses of genomic changes in G0 arrest and fast-cycling single cell data; S.P. developed the G0-LLM model and analysed the corresponding data. J.L. provided a curated signature list for UPR and led the interpretation of the proteostasis-linked analyses. C.C., M.S., S.P. and J.L. wrote the manuscript. All authors read and approved the manuscript. Ethics declarations All datasets employed in this study are publicly available and comply with ethical regulations, with approval and informed consent for collection and sharing already obtained by the relevant organizations. Competing interests The authors declare no competing interests. Acknowledgements M.S., C.C. and S.P. were supported by a UKRI Future Leaders Fellowship (MR/T042184/1). E.W. was supported by a studentship award from the Health Data Research UK-The Alan Turing Institute Wellcome PhD Programme in Health Data Science (218529/Z/19/Z). Work in M.S.’s lab was supported by a BBSRC equipment grant (BB/R01356X/1) and a Wellcome Institutional Strategic Support Fund (204841/Z/16/Z). J.L. was supported by BBSRC grants BB/T013273/1 and BB/W014890/1. We would like to acknowledge Dr Andrew Holding for his input on visualising hotspot overlaps in the spatial transcriptomics slides, and Dr Kevin Litchfield and Dr Andrea Castro for their help investigating the therapeutic relevance of the G0 arrest signature. Footnotes https://github.com/secrierlab/G0-breast-cancer-atlas https://github.com/secrierlab/G0-LM References 1. ↵ Hanahan , D . Hallmarks of Cancer: New Dimensions . Cancer Discov 12 , 31 – 46 ( 2022 ). OpenUrl Abstract / FREE Full Text 2. ↵ Ebinger , S. et al. Characterization of Rare, Dormant, and Therapy-Resistant Cells in Acute Lymphoblastic Leukemia . Cancer Cell 30 , 849 – 862 ( 2016 ). OpenUrl CrossRef PubMed 3. ↵ Francescangeli , F. et al. Dormancy, stemness, and therapy resistance: interconnected players in cancer evolution . Cancer Metastasis Rev 42 , 197 – 215 ( 2023 ). OpenUrl PubMed 4. ↵ Agudo , J. et al. Targeting cancer cell dormancy . Nat Rev Cancer 24 , 97 – 104 ( 2024 ). OpenUrl CrossRef PubMed 5. Phan , T.G. & Croucher , P.I . The dormant cancer cell life cycle . Nat Rev Cancer 20 , 398 – 411 ( 2020 ). OpenUrl CrossRef PubMed 6. ↵ Weston , W.A. & Barr , A.R . A cell cycle centric view of tumour dormancy . Br J Cancer 129 , 1535 – 1545 ( 2023 ). OpenUrl CrossRef PubMed 7. ↵ Truskowski , K. , Amend , S.R. & Pienta , K.J . Dormant cancer cells: programmed quiescence, senescence, or both? Cancer Metastasis Rev 42 , 37 – 47 ( 2023 ). OpenUrl CrossRef PubMed 8. ↵ Pu , Y. et al. Drug-tolerant persister cells in cancer: the cutting edges and future directions . Nat Rev Clin Oncol 20 , 799 – 813 ( 2023 ). OpenUrl CrossRef PubMed 9. ↵ Recasens , A. & Munoz , L . Targeting Cancer Cell Dormancy . Trends Pharmacol Sci 40 , 128 – 141 ( 2019 ). OpenUrl CrossRef 10. ↵ Aissa , A.F. et al. Single-cell transcriptional changes associated with drug tolerance and response to combination therapies in cancer . Nat Commun 12 , 1628 ( 2021 ). OpenUrl CrossRef PubMed 11. ↵ Xue , J.Y. et al. Rapid non-uniform adaptation to conformation-specific KRAS(G12C) inhibition . Nature 577 , 421 – 425 ( 2020 ). OpenUrl CrossRef PubMed 12. ↵ Malladi , S. et al. Metastatic Latency and Immune Evasion through Autocrine Inhibition of WNT . Cell 165 , 45 – 60 ( 2016 ). OpenUrl CrossRef PubMed 13. ↵ Ribas , A . Adaptive Immune Resistance: How Cancer Protects from Immune Attack . Cancer Discov 5 , 915 – 9 ( 2015 ). OpenUrl Abstract / FREE Full Text 14. ↵ Barkan , D. et al. Metastatic growth from dormant cells induced by a col-I-enriched fibrotic environment . Cancer Res 70 , 5706 – 16 ( 2010 ). OpenUrl Abstract / FREE Full Text 15. ↵ Sosa , M.S. , Bragado , P. & Aguirre-Ghiso , J.A . Mechanisms of disseminated cancer cell dormancy: an awakening field . Nat Rev Cancer 14 , 611 – 22 ( 2014 ). OpenUrl CrossRef PubMed 16. ↵ Ashraf , H.M. , Moser , J. & Spencer , S.L . Senescence Evasion in Chemotherapy: A Sweet Spot for p21 . Cell 178 , 267 – 269 ( 2019 ). OpenUrl CrossRef PubMed 17. ↵ Basile , K.J. & Aplin , A.E . Resistance to chemotherapy: short-term drug tolerance and stem cell-like subpopulations . Adv Pharmacol 65 , 315 – 34 ( 2012 ). OpenUrl CrossRef PubMed 18. ↵ Harbeck , N. et al. Breast cancer . Nat Rev Dis Primers 5 , 66 ( 2019 ). OpenUrl CrossRef PubMed 19. ↵ Nolan , E. , Lindeman , G.J. & Visvader , J.E . Deciphering breast cancer: from biology to the clinic . Cell 186 , 1708 – 1728 ( 2023 ). OpenUrl CrossRef PubMed 20. ↵ Aouad , P. , Quinn , H.M. , Berger , A. & Brisken , C . Tumor dormancy: EMT beyond invasion and metastasis . Genesis 62 , e23552 ( 2024 ). OpenUrl CrossRef PubMed 21. ↵ Aouad , P. et al. Epithelial-mesenchymal plasticity determines estrogen receptor positive breast cancer dormancy and epithelial reconversion drives recurrence . Nat Commun 13 , 4975 ( 2022 ). OpenUrl CrossRef PubMed 22. ↵ Rosano , D. et al. Long-term Multimodal Recording Reveals Epigenetic Adaptation Routes in Dormant Breast Cancer Cells . Cancer Discov 14 , 866 – 889 ( 2024 ). OpenUrl CrossRef PubMed 23. ↵ Baldominos , P. et al. Quiescent cancer cells resist T cell attack by forming an immunosuppressive niche . Cell 185 , 1694 – 1708 e19 ( 2022 ). OpenUrl CrossRef PubMed 24. ↵ Malla , S.B. et al. Pathway level subtyping identifies a slow-cycling biological phenotype associated with poor clinical outcomes in colorectal cancer . Nat Genet 56 , 458 – 472 ( 2024 ). OpenUrl CrossRef PubMed 25. Qin , X. et al. An oncogenic phenoscape of colonic stem cell polarization . Cell 186 , 5554 – 5568 e18 ( 2023 ). OpenUrl CrossRef PubMed 26. ↵ Tape , C.J . Plastic persisters: revival stem cells in colorectal cancer . Trends Cancer 10 , 185 – 195 ( 2024 ). OpenUrl CrossRef PubMed 27. ↵ Antonica , F. et al. A slow-cycling/quiescent cells subpopulation is involved in glioma invasiveness . Nat Commun 13 , 4767 ( 2022 ). OpenUrl CrossRef PubMed 28. ↵ Coller , H.A. , Sang , L. & Roberts , J.M . A new description of cellular quiescence . PLoS Biol 4 , e83 ( 2006 ). OpenUrl CrossRef PubMed 29. ↵ Marescal , O. & Cheeseman , I.M. Cellular Mechanisms and Regulation of Quiescence . Dev Cell 55 , 259 – 271 ( 2020 ). OpenUrl CrossRef PubMed 30. ↵ Barr , A.R. et al. DNA damage during S-phase mediates the proliferation-quiescence decision in the subsequent G1 via p21 expression . Nat Commun 8 , 14728 ( 2017 ). OpenUrl CrossRef PubMed 31. ↵ Heldt , F.S. , Barr , A.R. , Cooper , S. , Bakal , C. & Novak , B . A comprehensive model for the proliferation-quiescence decision in response to endogenous DNA damage in human cells . Proc Natl Acad Sci U S A 115 , 2532 – 2537 ( 2018 ). OpenUrl Abstract / FREE Full Text 32. ↵ Attieh , Y. & Vignjevic , D.M . The hallmarks of CAFs in cancer invasion . Eur J Cell Biol 95 , 493 – 502 ( 2016 ). OpenUrl CrossRef PubMed 33. ↵ Rakina , M. , Larionova , I. & Kzhyshkowska , J . Macrophage diversity in human cancers: New insight provided by single-cell resolution and spatial context . Heliyon 10 , e28332 ( 2024 ). OpenUrl CrossRef 34. Sammut , S.J. et al. Multi-omic machine learning predictor of breast cancer therapy response . Nature 601 , 623 – 629 ( 2022 ). OpenUrl CrossRef PubMed 35. ↵ Wu , S.Z. et al. A single-cell and spatially resolved atlas of human breast cancers . Nat Genet 53 , 1334 – 1347 ( 2021 ). OpenUrl CrossRef PubMed 36. ↵ Wiecek , A.J. et al. Genomic hallmarks and therapeutic implications of G0 cell cycle arrest in cancer . Genome Biol 24 , 128 ( 2023 ). OpenUrl CrossRef PubMed 37. ↵ Gao , R. et al. Delineating copy number and clonal substructure in human tumors from single-cell transcriptomes . Nat Biotechnol 39 , 599 – 608 ( 2021 ). OpenUrl CrossRef PubMed 38. ↵ Pal , B. et al. A single-cell RNA expression atlas of normal, preneoplastic and tumorigenic states in the human breast . EMBO J 40 , e107333 ( 2021 ). OpenUrl CrossRef PubMed 39. ↵ Qian , J. et al. A pan-cancer blueprint of the heterogeneous tumor microenvironment revealed by single-cell profiling . Cell Res 30 , 745 – 762 ( 2020 ). OpenUrl CrossRef PubMed 40. ↵ Gavish , A. et al. Hallmarks of transcriptional intratumour heterogeneity across a thousand tumours . Nature 618 , 598 – 606 ( 2023 ). OpenUrl CrossRef PubMed 41. ↵ Tickle , T. , Tirosh , I. , Georgescu , C. , Brown , M. & Haas , B . inferCNV of the Trinity CTAT Project . ( Klarman Cell Observatory , Broad Institute of MIT and Harvard , 2019 ). 42. ↵ Clarke , Z.A. & Bader , G.D. MALAT1 expression indicates cell quality in single-cell RNA sequencing data . bioRxiv , 2024.07.14.603469 ( 2024 ). 43. ↵ Montagner , M. et al. Crosstalk with lung epithelial cells regulates Sfrp2-mediated latency in breast cancer dissemination . Nat Cell Biol 22 , 289 – 296 ( 2020 ). OpenUrl CrossRef PubMed 44. ↵ Ormandy , C.J. , Musgrove , E.A. , Hui , R. , Daly , R.J. & Sutherland , R.L . Cyclin D1, EMS1 and 11q13 amplification in breast cancer . Breast Cancer Res Treat 78 , 323 – 35 ( 2003 ). OpenUrl CrossRef PubMed Web of Science 45. ↵ Du , Y. , Gupta , P. , Qin , S. & Sieber , M . The role of metabolism in cellular quiescence . J Cell Sci 136 ( 2023 ). 46. ↵ Provance , O.K. & Lewis-Wambi , J . Deciphering the role of interferon alpha signaling and microenvironment crosstalk in inflammatory breast cancer . Breast Cancer Res 21 , 59 ( 2019 ). OpenUrl CrossRef PubMed 47. ↵ Wu , J.I. , Lin , Y.P. , Tseng , C.W. , Chen , H.J. & Wang , L.H . Crabp2 Promotes Metastasis of Lung Cancer Cells via HuR and Integrin beta1/FAK/ERK Signaling . Sci Rep 9 , 845 ( 2019 ). OpenUrl CrossRef PubMed 48. ↵ Xie , T. , Tan , M. , Gao , Y. & Yang , H . CRABP2 accelerates epithelial mesenchymal transition in serous ovarian cancer cells by promoting TRIM16 methylation via upregulating EZH2 expression . Environ Toxicol 37 , 1957 – 1967 ( 2022 ). OpenUrl CrossRef PubMed 49. ↵ Habiba , U.e. , et al. The multifaceted role of IL-12 in cancer . Advances in Cancer Biology - Metastasis 5 ( 2022 ). 50. ↵ Shibabaw , T. , Teferi , B. & Ayelign , B . The role of Th-17 cells and IL-17 in the metastatic spread of breast cancer: As a means of prognosis and therapeutic target . Front Immunol 14 , 1094823 ( 2023 ). OpenUrl CrossRef PubMed 51. ↵ Santoro , A. et al. p53 Loss in Breast Cancer Leads to Myc Activation, Increased Cell Plasticity, and Expression of a Mitotic Signature with Prognostic Value . Cell Rep 26 , 624 – 638 e8 ( 2019 ). OpenUrl CrossRef PubMed 52. ↵ Uribe , M.L. , Marrocco , I. & Yarden , Y . EGFR in Cancer: Signaling Mechanisms , Drugs, and Acquired Resistance. Cancers (Basel ) 13 ( 2021 ). 53. ↵ Peck , B. et al. 3D Functional Genomics Screens Identify CREBBP as a Targetable Driver in Aggressive Triple-Negative Breast Cancer . Cancer Res 81 , 847 – 859 ( 2021 ). OpenUrl Abstract / FREE Full Text 54. ↵ Park , U.H. et al. ASXL2 promotes proliferation of breast cancer cells by linking ERalpha to histone methylation . Oncogene 35 , 3742 – 52 ( 2016 ). OpenUrl CrossRef PubMed 55. ↵ Dong , G. et al. ZFHX3 Promotes the Proliferation and Tumor Growth of ER-Positive Breast Cancer Cells Likely by Enhancing Stem-Like Features and MYC and TBX3 Transcription . Cancers (Basel ) 12 ( 2020 ). 56. ↵ Hu , Q. et al. ZFHX3 acts as a tumor suppressor in prostate cancer by targeting FTO-mediated m(6)A demethylation . Cell Death Discov 10 , 284 ( 2024 ). OpenUrl CrossRef PubMed 57. ↵ Hu , Q. et al. ZFHX3 is indispensable for ERbeta to inhibit cell proliferation via MYC downregulation in prostate cancer cells . Oncogenesis 8 , 28 ( 2019 ). OpenUrl CrossRef PubMed 58. ↵ Schiffman , J.S. et al. Defining heritability, plasticity, and transition dynamics of cellular phenotypes in somatic evolution . Nat Genet 56 , 2174 – 2184 ( 2024 ). OpenUrl CrossRef PubMed 59. ↵ Aibar , S. et al. SCENIC: single-cell regulatory network inference and clustering . Nat Methods 14 , 1083 – 1086 ( 2017 ). OpenUrl CrossRef PubMed 60. ↵ Lan , Q. et al. Type I interferon/IRF7 axis instigates chemotherapy-induced immunological dormancy in breast cancer . Oncogene 38 , 2814 – 2829 ( 2019 ). OpenUrl CrossRef PubMed 61. ↵ Fan , P. & Jordan , V.C. PERK , Beyond an Unfolded Protein Response Sensor in Estrogen-Induced Apoptosis in Endocrine-Resistant Breast Cancer . Mol Cancer Res 20 , 193 – 201 ( 2022 ). OpenUrl Abstract / FREE Full Text 62. ↵ Malagoli Tagliazucchi , G. , Wiecek , A.J. , Withnell , E. & Secrier , M. Genomic and microenvironmental heterogeneity shaping epithelial-to-mesenchymal trajectories in cancer . Nat Commun 14 , 789 ( 2023 ). OpenUrl CrossRef PubMed 63. ↵ Sosa , M.S. et al. NR2F1 controls tumour cell dormancy via SOX9- and RARbeta-driven quiescence programmes . Nat Commun 6 , 6170 ( 2015 ). OpenUrl CrossRef PubMed 64. ↵ Hipp , M.S. , Kasturi , P. & Hartl , F.U . The proteostasis network and its decline in ageing . Nat Rev Mol Cell Biol 20 , 421 – 435 ( 2019 ). OpenUrl CrossRef PubMed 65. ↵ Ku , H.C. & Cheng , C.F . Master Regulator Activating Transcription Factor 3 (ATF3) in Metabolic Homeostasis and Cancer . Front Endocrinol (Lausanne ) 11 , 556 ( 2020 ). OpenUrl CrossRef PubMed 66. Neill , G. & Masson , G.R . A stay of execution: ATF4 regulation and potential outcomes for the integrated stress response . Front Mol Neurosci 16 , 1112253 ( 2023 ). OpenUrl CrossRef PubMed 67. ↵ Ryoo , H.D . The integrated stress response in metabolic adaptation . J Biol Chem 300 , 107151 ( 2024 ). OpenUrl CrossRef PubMed 68. ↵ Tomasin , R. & Bruni-Cardoso , A . The role of cellular quiescence in cancer - beyond a quiet passenger . J Cell Sci 135 ( 2022 ). 69. ↵ Browaeys , R. , Saelens , W. & Saeys , Y . NicheNet: modeling intercellular communication by linking ligands to target genes . Nat Methods 17 , 159 – 162 ( 2020 ). OpenUrl CrossRef PubMed 70. ↵ Derynck , R. , Turley , S.J. & Akhurst , R.J . TGFbeta biology in cancer progression and immunotherapy . Nat Rev Clin Oncol 18 , 9 – 34 ( 2021 ). OpenUrl CrossRef PubMed 71. ↵ Richardson , L. , Wilcockson , S.G. , Guglielmi , L. & Hill , C.S . Context-dependent TGFbeta family signalling in cell fate regulation . Nat Rev Mol Cell Biol 24 , 876 – 894 ( 2023 ). OpenUrl CrossRef PubMed 72. ↵ Sterzynska , K. et al. The Role of Matrix Gla Protein (MGP) Expression in Paclitaxel and Topotecan Resistant Ovarian Cancer Cell Lines . Int J Mol Sci 19 ( 2018 ). 73. ↵ Cheng , G. et al. Higher levels of TIMP-1 expression are associated with a poor prognosis in triple-negative breast cancer . Mol Cancer 15 , 30 ( 2016 ). OpenUrl CrossRef PubMed 74. ↵ Taube , M.E. , Liu , X.W. , Fridman , R. & Kim , H.R . TIMP-1 regulation of cell cycle in human breast epithelial cells via stabilization of p27(KIP1) protein . Oncogene 25 , 3041 – 8 ( 2006 ). OpenUrl CrossRef PubMed Web of Science 75. ↵ Troulé , K. et al. CellPhoneDB v5: inferring cell-cell communication from single-cell multiomics data . arXiv:2311.04567 ( 2023 ). 76. ↵ Evans , E.E. et al. Antibody Blockade of Semaphorin 4D Promotes Immune Infiltration into Tumor and Enhances Response to Other Immunomodulatory Therapies . Cancer Immunol Res 3 , 689 – 701 ( 2015 ). OpenUrl Abstract / FREE Full Text 77. ↵ Malik , M.F. , Ye , L. & Jiang , W.G . Reduced expression of semaphorin 4D and plexin-B in breast cancer is associated with poorer prognosis and the potential linkage with oestrogen receptor . Oncol Rep 34 , 1049 – 57 ( 2015 ). OpenUrl CrossRef PubMed 78. ↵ Qin , K. et al. Canonical and noncanonical Wnt signaling: Multilayered mediators, signaling mechanisms and major signaling crosstalk . Genes Dis 11 , 103 – 134 ( 2024 ). OpenUrl CrossRef PubMed 79. ↵ Guo , X. , Yang , Y. , Tang , J. & Xiang , J . Ephs in cancer progression: complexity and context-dependent nature in signaling, angiogenesis and immunity . Cell Commun Signal 22 , 299 ( 2024 ). OpenUrl CrossRef PubMed 80. ↵ Avsar , G. & Pir , P . An integrated study to decipher immunosuppressive cellular communication in the PDAC environment . NPJ Syst Biol Appl 9 , 56 ( 2023 ). OpenUrl CrossRef PubMed 81. ↵ Osei-Sarfo , K. & Gudas , L.J . Retinoic acid suppresses the canonical Wnt signaling pathway in embryonic stem cells and activates the noncanonical Wnt signaling pathway . Stem Cells 32 , 2061 – 71 ( 2014 ). OpenUrl CrossRef PubMed 82. ↵ Ding , S. et al. Deficiency of angiopoietin-like 4 enhances CD8(+) T cell bioactivity via metabolic reprogramming for impairing tumour progression . Immunology 170 , 28 – 46 ( 2023 ). OpenUrl CrossRef 83. ↵ Withnell , E. & Secrier , M . SpottedPy quantifies relationships between spatial transcriptomic hotspots and uncovers environmental cues of epithelial-mesenchymal plasticity in breast cancer . Genome Biol 25 , 289 ( 2024 ). OpenUrl CrossRef PubMed 84. ↵ Gurrapu , S. et al. Reverse signaling by semaphorin 4C elicits SMAD1/5- and ID1/3-dependent invasive reprogramming in cancer cells . Sci Signal 12 ( 2019 ). 85. ↵ Smeester , B.A. et al. SEMA4C is a novel target to limit osteosarcoma growth, progression, and metastasis . Oncogene 39 , 1049 – 1062 ( 2020 ). OpenUrl CrossRef PubMed 86. ↵ Borrelli , C. et al. In vivo interaction screening reveals liver-derived constraints to metastasis . Nature 632 , 411 – 418 ( 2024 ). OpenUrl CrossRef PubMed 87. ↵ Navab , R. et al. Integrin alpha11beta1 regulates cancer stromal stiffness and promotes tumorigenicity and metastasis in non-small cell lung cancer . Oncogene 35 , 1899 – 908 ( 2016 ). OpenUrl CrossRef PubMed 88. ↵ Kanemaru , K. et al. Spatially resolved multiomics of human cardiac niches . Nature 619 , 801 – 810 ( 2023 ). OpenUrl CrossRef PubMed 89. ↵ Fu , M. et al. Combination of targeted daunorubicin liposomes and targeted emodin liposomes for treatment of invasive breast cancer . J Drug Target 28 , 245 – 258 ( 2020 ). OpenUrl CrossRef PubMed 90. ↵ Guo , J. et al. Effects of stealth liposomal daunorubicin plus tamoxifen on the breast cancer and cancer stem cells . J Pharm Pharm Sci 13 , 136 – 51 ( 2010 ). OpenUrl CrossRef PubMed 91. ↵ Weitz , I.C . Pegcetacoplan: A New Opportunity for Complement Inhibition in PNH . J Blood Med 14 , 239 – 245 ( 2023 ). OpenUrl CrossRef PubMed 92. ↵ Pan , S. , Withnell , E. & Secrier , M . Classifying epithelial-mesenchymal transition states in single cell cancer data using large language models . bioRxiv , 2024.08.16.608311 ( 2024 ). 93. ↵ Akhmetkaliyev , A. , Alibrahim , N. , Shafiee , D. & Tulchinsky , E . EMT/MET plasticity in cancer and Go-or-Grow decisions in quiescence: the two sides of the same coin? Mol Cancer 22 , 90 ( 2023 ). OpenUrl CrossRef PubMed 94. Basu , S. , Dong , Y. , Kumar , R. , Jeter , C. & Tang , D.G . Slow-cycling (dormant) cancer cells in therapy resistance, cancer relapse and metastasis . Semin Cancer Biol 78 , 90 – 103 ( 2022 ). OpenUrl CrossRef PubMed 95. ↵ Landau , D.A. et al. Evolution and impact of subclonal mutations in chronic lymphocytic leukemia . Cell 152 , 714 – 26 ( 2013 ). OpenUrl CrossRef PubMed Web of Science 96. ↵ Mick , E. et al. Distinct mitochondrial defects trigger the integrated stress response depending on the metabolic state of the cell . Elife 9 ( 2020 ). 97. ↵ Quiros , P.M. et al. Multi-omics analysis identifies ATF4 as a key regulator of the mitochondrial stress response in mammals . J Cell Biol 216 , 2027 – 2045 ( 2017 ). OpenUrl Abstract / FREE Full Text 98. ↵ Smith , P.R. , Pandit , S.C. , Loerch , S. & Campbell , Z.T . The space between notes: emerging roles for translationally silent ribosomes . Trends Biochem Sci 47 , 477 – 491 ( 2022 ). OpenUrl CrossRef PubMed 99. ↵ Kato , S. et al. Semaphorin 4D, a lymphocyte semaphorin, enhances tumor cell motility through binding its receptor, plexinB1, in pancreatic cancer . Cancer Sci 102 , 2029 – 37 ( 2011 ). OpenUrl CrossRef PubMed 100. ↵ Zhang , C. et al. CD100-plexin-B1 induces epithelial-mesenchymal transition of head and neck squamous cell carcinoma and promotes metastasis . Cancer Lett 455 , 1 – 13 ( 2019 ). OpenUrl CrossRef PubMed 101. ↵ Lucas , O. et al. Characterizing the evolutionary dynamics of cancer proliferation in single-cell clones with SPRINTER . Nat Genet ( 2024 ). 102. ↵ Hao , Y. et al. Integrated analysis of multimodal single-cell data . Cell 184 , 3573 – 3587 e29 ( 2021 ). OpenUrl CrossRef PubMed 103. ↵ Franzen , O. , Gan , L.M. & Bjorkegren , J.L.M . PanglaoDB: a web server for exploration of mouse and human single-cell RNA sequencing data . Database (Oxford ) 2019 ( 2019 ). 104. ↵ Ianevski , A. , Giri , A.K. & Aittokallio , T . Fully-automated and ultra-fast cell-type identification using specific marker combinations from single-cell transcriptomic data . Nat Commun 13 , 1246 ( 2022 ). OpenUrl CrossRef PubMed 105. ↵ Li , M. et al. DISCO: a database of Deeply Integrated human Single-Cell Omics data . Nucleic Acids Res 50 , D596 – D602 ( 2022 ). OpenUrl CrossRef PubMed 106. ↵ McDavid , A. et al. Data exploration, quality control and testing in single-cell qPCR-based gene expression experiments . Bioinformatics 29 , 461 – 7 ( 2013 ). OpenUrl CrossRef PubMed Web of Science 107. ↵ Hothorn , T. & Lausen , B . On the exact distribution of maximally selected rank statistics . Computational Statistics & Data Analysis 43 , 121 – 137 ( 2003 ). OpenUrl CrossRef 108. ↵ Morgan , M. & Dann , E . miloR: Differential neighbourhood abundance testing on a graph . R package version 1.10.0 edn ( 2023 ). 109. ↵ De Falco , A. , Caruso , F. , Su , X.D. , Iavarone , A. & Ceccarelli , M . A variational algorithm to detect the clonal copy number substructure of tumors from scRNA-seq data . Nat Commun 14 , 1074 ( 2023 ). OpenUrl CrossRef PubMed 110. ↵ Colaprico , A. et al. TCGAbiolinks: an R/Bioconductor package for integrative analysis of TCGA data . Nucleic Acids Res 44 , e71 ( 2016 ). OpenUrl CrossRef PubMed 111. ↵ Robinson , D.R. et al. Integrative clinical genomics of metastatic cancer . Nature 548 , 297 – 303 ( 2017 ). OpenUrl CrossRef PubMed 112. ↵ Liberzon , A. et al. The Molecular Signatures Database (MSigDB) hallmark gene set collection . Cell Syst 1 , 417 – 425 ( 2015 ). OpenUrl CrossRef PubMed 113. ↵ Hediyeh-zadeh , S. et al. Identification of cell types, states and programs by learning gene set representations . bioRxiv , 2023.09.08.556842 ( 2023 ). 114. ↵ Ren , Q. et al. Gene expression predicts dormant metastatic breast cancer cell phenotype . Breast Cancer Res 24 , 10 ( 2022 ). OpenUrl CrossRef PubMed 115. ↵ Ruth , J.R. et al. Cellular dormancy in minimal residual disease following targeted therapy . Breast Cancer Res 23 , 63 ( 2021 ). OpenUrl CrossRef PubMed 116. ↵ Cheng , Q. et al. A signature of epithelial-mesenchymal plasticity and stromal activation in primary tumor modulates late recurrence in breast cancer independent of disease subtype . Breast Cancer Res 16 , 407 ( 2014 ). OpenUrl CrossRef PubMed 117. ↵ Kim , R.S. et al. Dormancy signatures and metastasis in estrogen receptor positive and negative breast cancer . PLoS One 7 , e35569 ( 2012 ). OpenUrl CrossRef PubMed 118. ↵ Dimitrov , D. et al. LIANA+ provides an all-in-one framework for cell-cell communication inference . Nat Cell Biol 26 , 1613 – 1622 ( 2024 ). OpenUrl CrossRef PubMed 119. ↵ Sade-Feldman , M. et al. Defining T Cell States Associated with Response to Checkpoint Immunotherapy in Melanoma . Cell 175 , 998 – 1013 e20 ( 2018 ). OpenUrl CrossRef PubMed 120. ↵ Andreatta , M. et al. Interpretation of T cell states from single-cell transcriptomics data using reference atlases . Nat Commun 12 , 2965 ( 2021 ). OpenUrl CrossRef PubMed 121. ↵ Barkley , D. et al. Cancer cell states recur across tumor types and form specific interactions with the tumor microenvironment . Nat Genet 54 , 1192 – 1201 ( 2022 ). OpenUrl CrossRef PubMed 122. ↵ 10X Genomics, I . ( 10X Genomics, Inc. ). 123. ↵ Kleshchevnikov , V. et al. Cell2location maps fine-grained cell types in spatial transcriptomics . Nat Biotechnol 40 , 661 – 671 ( 2022 ). OpenUrl CrossRef PubMed 124. ↵ Elyanow , R. , Zeira , R. , Land , M. & Raphael , B.J . STARCH: copy number and clone inference from spatial transcriptomics data . Phys Biol 18 , 035001 ( 2021 ). OpenUrl CrossRef PubMed 125. ↵ Yang , F. et al. scBERT as a large-scale pretrained deep language model for cell type annotation of single-cell RNA-seq data . Nature Machine Intelligence 4 , 852 – 866 ( 2022 ). OpenUrl CrossRef 126. ↵ Hu , E.J. , et al. LoRA: Low-Rank Adaptation of Large Language Models . arxiv preprint ( 2021 ). View the discussion thread. Back to top Previous Next Posted January 13, 2025. Download PDF Supplementary Material Data/Code Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Balancing tumour proliferation and sustained cell cycle arrest through proteostasis remodelling drives immune niche compartmentalisation in breast cancer Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Balancing tumour proliferation and sustained cell cycle arrest through proteostasis remodelling drives immune niche compartmentalisation in breast cancer Cenk Celik , Eloise Withnell , Shi Pan , Tooki Chu , John Labbadia , Maria Secrier bioRxiv 2025.01.08.632014; doi: https://doi.org/10.1101/2025.01.08.632014 Share This Article: Copy Citation Tools Balancing tumour proliferation and sustained cell cycle arrest through proteostasis remodelling drives immune niche compartmentalisation in breast cancer Cenk Celik , Eloise Withnell , Shi Pan , Tooki Chu , John Labbadia , Maria Secrier bioRxiv 2025.01.08.632014; doi: https://doi.org/10.1101/2025.01.08.632014 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Cancer Biology Subject Areas All Articles Animal Behavior and Cognition (7624) Biochemistry (17651) Bioengineering (13871) Bioinformatics (41882) Biophysics (21424) Cancer Biology (18566) Cell Biology (25461) Clinical Trials (138) Developmental Biology (13365) Ecology (19867) Epidemiology (2067) Evolutionary Biology (24290) Genetics (15590) Genomics (22476) Immunology (17714) Microbiology (40331) Molecular Biology (17148) Neuroscience (88483) Paleontology (666) Pathology (2828) Pharmacology and Toxicology (4817) Physiology (7635) Plant Biology (15114) Scientific Communication and Education (2044) Synthetic Biology (4286) Systems Biology (9815) Zoology (2268)