Abstract Cancer cell states are governed by coordinated signaling and metabolic networks, yet the mechanisms coupling metabolic activity to oncogenic signaling remain poorly understood. Here, we identify dihydropyrimidines (DHPs), metabolites generated during pyrimidine catabolism, as “signaling metabolites” that sustain aggressive cancer cell states. Depletion of DHPs, either by knockout of the pyrimidine catabolic enzyme DPYD or by expression of the DHP-degrading enzyme DPYS, suppressed STAT3 signaling and attenuated mesenchymal and inflammatory transcriptional programs. Mechanistically, DHPs stabilized DPYSL2, a JAK1-interacting adaptor protein required for efficient STAT3 activation. Structural modeling and thermal stability analyses further support a direct interaction between DHPs and DPYSL2. Accordingly, high DPYD expression was associated with EMT, inflammatory signaling, and poor-prognosis breast cancer subtypes across patient cohorts. Ultimately, our findings establish pyrimidine catabolism as a direct regulator of signaling competence and identify DHPs as signaling-active metabolites that couple cellular metabolic state to oncogenic transcriptional programs and mesenchymal identity.

The host-microbiome axis has been implicated in promoting anti-inflammatory immune responses. Yet, the underlying molecular mechanisms of commensal-mediated IL-10 production by regulatory B cells (Bregs) are not fully elucidated. Here, we demonstrate that bacterial CpG motifs trigger the signaling downstream of TLR9 promoting IκBNS-mediated expression of Blimp-1, a transcription regulator of IL-10. Surprisingly, this effect was counteracted by the NF-κB transcription factor c-Rel. A functional screen for intestinal bacterial species identified the commensal Clostridium sporogenes, secreting high amounts of short-chain fatty acids (SCFAs) and branched-chain fatty acids (BCFAs), as an amplifier of IL-10 production by promoting sustained mTOR signaling in B cells. Consequently, enhanced Breg functionality was achieved by combining CpG with the SCFA butyrate or the BCFA isovalerate thereby synergizing TLR- and mTOR-mediated pathways. Collectively, Bregs required two bacterial signals (butyrate and CpG) to elicit their full suppressive capacity and ameliorate T cell-mediated intestinal inflammation. Our study has dissected the molecular pathways induced by bacterial factors, which might contribute not only to better understanding of host-microbiome interactions, but also to exploration of new strategies for improvement of anti-inflammatory cellular therapy.

Downregulation of the urea cycle enzyme argininosuccinate synthase (ASS1) in multiple tumors is associated with a poor prognosis partly because of the metabolic diversion of cytosolic aspartate for pyrimidine synthesis, supporting proliferation and mutagenesis owing to nucleotide imbalance. Here, we find that prolonged loss of ASS1 promotes DNA damage in colon cancer cells and fibroblasts from subjects with citrullinemia type I. Following acute induction of DNA damage with doxorubicin, ASS1 expression is elevated in the cytosol and the nucleus with at least a partial dependency on p53; ASS1 metabolically restrains cell cycle progression in the cytosol by restricting nucleotide synthesis. In the nucleus, ASS1 and ASL generate fumarate for the succination of SMARCC1, destabilizing the chromatin-remodeling complex SMARCC1–SNF5 to decrease gene transcription, specifically in a subset of the p53-regulated cell cycle genes. Thus, following DNA damage, ASS1 is part of the p53 network that pauses cell cycle progression, enabling genome maintenance and survival. Loss of ASS1 contributes to DNA damage and promotes cell cycle progression, likely contributing to cancer mutagenesis and, hence, adaptability potential.

The epithelial-mesenchymal transition (EMT) program is crucial for transforming carcinoma cells into a partially mesenchymal state, enhancing their chemoresistance, migration, and metastasis. This shift in cell state is tightly regulated by cellular mechanisms that are not yet fully characterized. One intriguing EMT aspect is the rewiring of the proteoglycan landscape, particularly the induction of heparan sulfate proteoglycan (HSPG) biosynthesis. This proteoglycan functions as a co-receptor that accelerates cancer-associated signaling pathways through its negatively-charged residues. However, the precise mechanisms through which EMT governs HSPG biosynthesis and its role in cancer cell plasticity remain elusive. Here, we identified exostosin glycosyltransferase 1 (EXT1), a central enzyme in HSPG biosynthesis, to be selectively upregulated in aggressive tumor subtypes and cancer cell lines, and to function as a key player in breast cancer aggressiveness. Notably, ectopic expression of EXT1 in epithelial cells is sufficient to induce HSPG levels and the expression of known mesenchymal markers, subsequently enhancing EMT features, including cell migration, invasion, and tumor formation. Additionally, EXT1 loss in MDA-MB-231 cells inhibits their aggressiveness-associated traits such as migration, chemoresistance, tumor formation, and metastasis. Our findings reveal that EXT1, through its role in HSPG biosynthesis, governs signal transducer and activator of transcription 3 (STAT3) signaling, a known regulator of cancer cell aggressiveness. Collectively, we present the EXT1/HSPG/STAT3 axis as a central regulator of cancer cell plasticity that directly links proteoglycan synthesis to oncogenic signaling pathways.

Supplemental Figure S2. Circulating tumor cells express higher levels of TRPM2 compared with primary tumor cells in pancreatic cancer patients. A. Transcriptome analysis of mesenchymal genes, epithelial genes and TRPM2 on primary 4T1 tumor cells and circulating 4T1 tumor cells. The analysis was performed on dataset (GSE37244) published by LeBlue et al. (28). B. Microarray study for gene expression (GSE18670) changes in 6 samples from pancreatic cancer primary tumor (Primary) and circulating tumor cells (CTC) was described previously by Sergeant at al. (29). Here the same dataset was reanalyzed for the log2 expression ratio of TRMP2 in these samples. Data show TRPM2 expression in primary tumor cells (blue) and CTC (Red). *p<0.05 (2-tailed paired t-test).

Carcinomas often exhibit aggressive characteristics, such as enhanced migration abilities, through the execution of the epithelial-mesenchymal transition (EMT) program. Heparan sulfate (HS) is a polysaccharide expressed on the surface of aggressive cancer cells, which acts as a co-receptor to stimulate EMT-associated signaling pathways. However, despite HS’ role in cancer aggressiveness, the mechanisms governing its EMT-dependent biosynthesis remains poorly understood. Here, we characterized the HS chain elongation enzyme, exostosin glycosyltransferase 1 (EXT1), as an essential component of the EMT program. We identified an EMT-dependent expression of EXT1 and its selective upregulation in aggressive tumor subtypes and cell lines. Overexpression of EXT1 in epithelial cells is sufficient to induce HS biosynthesis, cell migration, and invasion, form tumors in mice, and activate the STAT3 pathway. Moreover, its knockout in aggressive cells significantly inhibited their EMT-associated characteristics. These findings demonstrate a cellular mechanism by which metabolic processes regulate signaling pathways to govern cell state.

Supplemental Figure S2. TRPM2kd AT3 and LLC cells resistant to H2O2 and neutrophil cytotoxicity. A. Relative expression of TRPM2 in control TRPM2+ cells TRPM2kd AT3 and LLC cells. B. TRPM2kd AT3 and LLC cells are resistant to H2O2 cytotoxicity. C. TRPM2kd AT3 and LLC cells are resistant to neutrophil cytotoxicity. **p<0.01

Supplemental figure legends

The Cancer Genome Atlas (TCGA) and analogous projects have yielded invaluable tumor-associated genomic data. Despite several web-based platforms designed to enhance accessibility, certain analyses require prior bioinformatic expertise. To address this need, we developed Gene ENrichment Identifier (GENI, https://www.shaullab.com/geni), which is designed to promptly compute correlations for genes of interest against the entire transcriptome and rank them against well-established biological gene sets. Additionally, it generates comprehensive tables containing genes of interest and their corresponding correlation coefficients, presented in publication-quality graphs. Furthermore, GENI has the capability to analyze multiple genes simultaneously within a given gene set, elucidating their significance within a specific biological context. Overall, GENI's user-friendly interface simplifies the biological interpretation and analysis of cancer patient-associated data, advancing the understanding of cancer biology and accelerating scientific discoveries.

Supplemental Figure S4. EMT increases TRPM2 expression and renders cells susceptible to neutrophil cytotoxicity. A. Representative images showing control and Snail overexpressing 4T1 cells. B. Snail expression in control and Snail overexpressing 4T1 cells. C. qPCR analysis of TRPM2, Vimentin (Vim.) and E-Cadherin (E-Cad.) expression in control and Snail overexpressing 4T1 cells. D. Extent of neutrophil cytotoxicity in co-culture with 4T1 control cells (Cont.) or 4T1 cells induced to undergo EMT by Snail overexpression (Snail). E. Representative images showing HMLE cells before (HMLE) and after prolonged treatment with 10 ng/ml TGFβ to induce EMT. F. qPCR analysis of TRPM2, Vimentin (Vim.) and E-Cadherin (E-Cad.) in control and TGFβâ^'treated HMLE cells. G. Extent of neutrophil cytotoxicity in co-culture with control (Cont.) or TGFβ-pretreated HMLE cells that have undergone EMT. H. Representative images showing control and Twist overexpressing HMLE cells. I. Twist expression in control and Twist overexpressing HMLE cells. J. qPCR analysis of TRPM2, Vimentin (Vim.) and E-Cadherin (E-Cad.) expression in control and Twist overexpressing HMLE cells. K. Extent of neutrophil cytotoxicity in co-culture with HMLE control cells (Cont.) or HMLE cells induced to undergo EMT by Twist overexpression (Twist). *p<0.05, **p<0.01.

Supplemental Figure S1. Neutrophils in the premetastatic lung outnumber and interact with disseminated tumor cells. Ly6G immunohistochemistry of metastases-laden lungs from 28 day 4T1 tumor-bearing mouse. Insets are high power magnification.

Supplemental Figure S3. Inverse correlation between E-Cadherin expression and TRPM2 expression in primary and metastatic 4T1 tumors. Immunofluorescent staining of TRPM2 (red) and E-Cadherin (green) in 4T1 primary tumor (A) in a lung metastatic 4T1 tumor (B).

Supplemental Figure S5. Metastatic seeding of BMP7 treated cells is not enhanced by neutrophil depletion in tumor baring mice. Representative images of lung associated GFP+ BMP7 treated cells in control (BMP7) and neutrophil depleted (BMP7 depletion) mice, 48 hrs following tail-vein injection.

Supplemental Figure S1. BAPTA protects tumor cells from H2O2 induced cell death. A. Extent of H2O2 (25 μM) cytotoxicity time course in 4T1 B. Annexin-V/PI analysis of vehicle and H2O2 treated (25 μM) 4T1 cells following overnight incubation. C. Effect of BAPTA-AM (5μM) on viability of 4T07, MC38, LLC, FARN and AT3 cells. D. Effect of BAPTA-AM (5μM) on neutrophil cytotoxicity in 4T07, MC38, LLC, FARN and AT3 cells. **p<0.01.

Ufmylation is a posttranslational modification in which the modifier UFM1 is attached to target proteins. This conjugation requires the concerted work of three enzymes named UBA5, UFC1, and UFL1. Initially, UBA5 activates UFM1 in a process that ends with UFM1 attached to UBA5’s active site Cys. Then, in a trans-thiolation reaction, UFM1 is transferred from UBA5 to UFC1, forming a thioester bond with the latter. Finally, with the help of UFL1, UFM1 is transferred to the final destination—a lysine residue on a target protein. Therefore, not surprisingly, deletion of one of these enzymes abrogates the conjugation process. However, how overexpression of these enzymes affects this process is not yet clear. Here we found, unexpectedly, that overexpression of UBA5, but not UFC1, damages the ability of cells to migrate, in a similar way to cells lacking UBA5 or UFC1. At the mechanistic level, we found that overexpression of UBA5 reverses the trans-thiolation reaction, thereby leading to a back transfer of UFM1 from UFC1 to UBA5. This, as seen in cells lacking UBA5, reduces the level of charged UFC1 and therefore harms the conjugation process. In contrast, co-expression of UBA5 with UFM1 abolishes this effect, suggesting that the reverse transfer of UFM1 from UFC1 to UBA5 depends on the level of free UFM1. Overall, our results propose that the cellular expression level of the UFM1 conjugation enzymes has to be tightly regulated to ensure the proper directionality of UFM1 transfer.

The intricate neuronal wiring during development requires cytoskeletal reorganization orchestrated by signaling cues. Because cytoskeletal remodeling is a hallmark of cell migration, we investigated whether metastatic cancer cells exploit axon guidance proteins to migrate. Indeed, in breast cancer patients, we found a significant correlation between mesenchymal markers and the expression of dihydropyrimidinase-like 2 (DPYSL2), a regulator of cytoskeletal dynamics in growing axons. Strikingly, DPYSL2 knockout in mesenchymal-like breast cancer cells profoundly inhibited cell migration, invasion, stemness features, tumor growth rate, and metastasis. Next, we decoded the molecular mechanism underlying this phenomenon and revealed an interaction between DPYSL2 and Janus kinase 1 (JAK1). This binding is crucial for activating signal transducer and activator of transcription 3 (STAT3) and the subsequent expression of vimentin, the promigratory intermediate filament. These findings identify DPYSL2 as a molecular link between oncogenic signaling pathways and cytoskeletal reorganization in migrating breast cancer cells.

Fumarate hydratase (FH) is an evolutionary conserved TCA cycle enzyme that reversibly catalyzes the hydration of fumarate to L-malate and has a moonlight function in the DNA damage response (DDR). Interestingly, FH has a contradictory cellular function, as it is pro-survival through its role in the TCA cycle, yet its loss can drive tumorigenesis. Here, we found that in both non-cancerous (HEK-293T) and cancerous cell lines (HepG2), the cell response to FH loss is separated into two distinct time frames based on cell proliferation and DNA damage repair. During the early stages of FH loss, cell proliferation rate and DNA damage repair are inhibited. However, over time the cells overcome the FH loss and form knockout clones, indistinguishable from WT cells with respect to their proliferation rate. Due to the FH loss effect on DNA damage repair, we assumed that the recovered cells bear adaptive mutations. Therefore, we applied whole-exome sequencing to identify such mutated genes systematically. Indeed, we identified recurring mutations in genes belonging to central oncogenic signaling pathways, such as JAK/STAT3, which we validated in impaired FH-KO clones. Intriguingly, we demonstrate that these adaptive mutations are responsible for FH-KO cell proliferation under TCA cycle malfunction.

The intricate neuronal wiring during development requires cytoskeletal reorganization orchestrated by signaling cues. Considering that cytoskeletal remodeling is a hallmark of cell migration, we inquired whether metastatic cancer cells exploit the axon guidance proteins to migrate. Indeed, in breast cancer patients, we found a significant correlation between the mesenchymal markers and the expression of dihydropyrimidinase-like 2 (DPYSL2), a regulator of cytoskeletal dynamics in growing axons. Strikingly, DPYSL2 knockout in mesenchymal-like cells profoundly inhibited cell migration, invasion, stemness features, tumor growth rate, and metastasis. Next, we aimed to decode the molecular mechanism underlying this phenomenon and revealed an interaction between DPYSL2 and Janus kinase 1 (JAK1). This binding is crucial for triggering signal transducer and activator of transcription 3 (STAT3) and subsequently expressing vimentin, the pro-migratory intermediate filament. Collectively, we identified DPYSL2 as a molecular link between oncogenic signaling pathways and cytoskeletal reorganization in migrating breast cancer cells. Statement of significance This study shows that the axon guidance adaptor protein DPYSL2 is essential for promoting breast cancer migration. Specifically, this protein interacts with JAK1 to govern STAT3 signaling and subsequently vimentin expression.

Breast cancer is a frequent heterogeneous malignancy and the second leading cause of mortality in women, mainly due to distant organ metastasis. Several animal models have been generated, including the widely used orthotopic mouse models, where cancer cells are injected into the mammary fat pad. However, these models cannot help monitor tumor growth kinetics and metastatic colonization. Cutting-edge tools to monitor cancer cells in real time in mice will significantly advance the understanding of tumor biology. Here, breast cancer cell lines stably expressing luciferase and green fluorescent protein (GFP) were established. Specifically, this technique contains two sequential steps initiated by measuring the luciferase activity in vitro and followed by the implantation of the cancer cells into mammary fat pads of nonobese diabetic-severe combined immunodeficiency (NOD-SCID) mice. After the injection, both the tumor growth and metastatic colonization are monitored in real time by the noninvasive bioluminescence imaging system. Then, the quantification of GFP-expressing metastases in the lungs will be examined by fluorescence microscopy to validate the observed bioluminescence results. This sophisticated system combining luciferase and fluorescence-based detection tools evaluates cancer metastasis in vivo, which has great potential for use in breast cancer therapeutics and disease management.