Biology subjects

Lehti, K.

Publications and source records attributed to Lehti, K..

14 recordsLinked to original sources

bioRxiv2026

TRAFIKK: systematic prediction and mechanistic interpretation of anticancer drug synergies

Effective drug combination therapies can improve cancer treatment, yet the mechanistic basis of drug synergy remains poorly understood. Most computational approaches prioritize predictive accuracy over molecular mechanistic interpretability, providing hence limited insights into how synergistic effects emerge across signalling contexts. We developed Trafikk, a molecular-signalling network-based framework that simulates drug perturbations in cell line-specific computational models to mirror functional outcomes in experimental combination screens. Across two independent large-scale datasets, Trafikk identified synergistic combinations with >77% recall. Functional response predictions revealed both conserved and context-dependent mechanisms. While AKT-MEK co-inhibition consistently disrupted coordinated survival and apoptotic signalling in 742 cell lines, PI3K-BCL2 synergy arose through distinct death programs shaped by cell-context-specific network constraints. Trafikk combines predictive performance with mechanistic interpretability, capturing how and why drug synergy emerges across cellular contexts. Source code, installation instructions and usage tutorial are freely available at https://github.com/druglogics/trafikk. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=147 SRC="FIGDIR/small/723755v1_ufig1.gif" ALT="Figure 1"> View larger version (33K): org.highwire.dtl.DTLVardef@9f09f0org.highwire.dtl.DTLVardef@1456ef5org.highwire.dtl.DTLVardef@177b11forg.highwire.dtl.DTLVardef@10d7435_HPS_FORMAT_FIGEXP M_FIG C_FIG

systems biology↗

bioRxiv2025

Single-Cell Integration of Chromatin Accessibility and Transcriptomics Reveals Regulatory Networks in Ovarian Tumor-Infiltrating Adaptive NK Cells

Natural killer (NK) cells are traditionally recognized for their rapid, non-specific responses against virus-infected or malignantly transformed cells, functioning as key effectors of innate immunity. However, a distinct subset known as adaptive NK (aNK) cells has been shown to acquire memory-like properties following viral infections, indicating their capacity for antigen-specific immune recall. Intriguingly, aNK cells have also been identified within the tumor microenvironment, where they can mediate tumor-specific recall responses. Yet, the regulatory mechanisms governing their function in tumor-infiltrating aNK cells remain largely undefined. In this study, we integrated publicly available multiomics datasets from ovarian cancer, including single-cell chromatin accessibility (scATAC-seq) and single-cell RNA sequencing (scRNA-seq), to identify chromatin-accessible regions and construct transcription factors (TF)-gene regulatory networks. To validate and extend these findings, we performed Smart-seq3 on NK cells isolated from ovarian tumors and applied SCENIC analysis to identify TF-driven gene regulation. By integrating results from both analyses, we identified PRDM1 and STAT2 as key TFs, along with their downstream targets CRCP and MTFP1, specifically enriched in tumor-infiltrating aNK cells. The expression levels of CRCP and MTFP1 positively correlated with NK cell infiltration in ovarian cancer tissues, suggesting their potential functions in supporting tumor-specific NK cell memory responses. In addition, external validation using data from the PROMIX clinical trial demonstrated that PRDM1 and STAT2 expression levels are positively associated with both overall survival and aNK cell-associated transcriptional features. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=111 SRC="FIGDIR/small/676040v1_ufig1.gif" ALT="Figure 1"> View larger version (37K): org.highwire.dtl.DTLVardef@6df026org.highwire.dtl.DTLVardef@1b16dc8org.highwire.dtl.DTLVardef@19e916borg.highwire.dtl.DTLVardef@15d431a_HPS_FORMAT_FIGEXP M_FIG C_FIG

immunology↗

bioRxiv2025

BooLEVARD: Boolean Logical Evaluation of Activation and Repression in Directed pathways

Boolean models are widely used for studying dynamic processes of biological systems. However, their inherent discrete nature limits their ability to capture continuous aspects of signal transduction, such as signal strength or protein activation levels. Although existing tools provide some path exploration capabilities that can be used to explore signal transduction circuits, the computational workload often requires simplifying assumptions that compromise the accuracy of the analysis. Here, we introduce BooLEVARD, a Python package designed to efficiently quantify the number of paths leading either to node activation or repression in Boolean models, which offers a more detailed and quantitative perspective on how molecular signals propagate through signaling networks. By focusing on the collection of non-redundant paths directly influencing Boolean outcomes, BooLEVARD enhances the precision of signal strength representation. We showcase the application of BooLEVARD in studying cell-fate decisions using a Boolean model of cancer metastasis, demonstrating its ability to identify critical signaling events. Furthermore, BooLEVARD provides a refined understanding of signaling dynamics, which can increase our understanding of disease development and drug responses. BooLEVARD is freely available at https://github.com/farinasm/boolevard.

systems biology↗

bioRxiv2024

The mechanical anisotropy of adipose tissues regulates ovarian cancer invasion

High-grade serous ovarian cancer, the most common and aggressive form of ovarian cancer, generally metastasises to visceral adipose tissues. In these tissues, the extracellular matrix through which ovarian cancer cells adhere and migrate is confined by the presence and preponderance of adipocytes. How cells migrate in this unique environment is not known, yet critical to understanding metastatic progression. To study these processes, we develop biomimetic organo-hydrogels that recreate structural and mechanical properties of human visceral adipose tissues. We show that ovarian cancer cells present invasive tropism towards organo-hydrogels, replicating the behaviour observed in native adipose tissues. This migration is facilitated by the mechanical anisotropy and microstructure of organo-hydrogels and adipose tissues, allowing the formation of cell force-induced migratory tracks, a process regulated by TGF{beta} in an MMP degradation-independent manner. These results highlight the contribution of adipocytes to tissue biophysical features as a key regulatory factor of ovarian cancer cell migration.

bioengineering↗

bioRxiv2024

Advancing Adoptive Cell Therapy: Optimized Expansion of Adaptive NK Cells for Solid Tumors

BackgroundImmune therapies are emerging as a critical component of cancer treatment, capable of delivering durable and potentially curative responses. While CAR-T cell therapy has proven effective for hematological malignancies, it faces challenges in treating solid tumors due to tumor antigen heterogeneity, an immunosuppressive tumor microenvironment, and physical barriers hindering CAR-T cell infiltration. NK cells, particularly adaptive NK (aNK) cells, offer a promising alternative due to their ability to recognize and kill tumor cells without prior sensitization and their resistance to immunosuppressive environments. PurposeThe study investigates the role of cytokines, specifically IL-21 and IL-15, in enhancing aNK cell expansion and activation using peripheral blood mononuclear cells (PBMCs) from healthy donors and tumor-infiltrating lymphocytes (TILs) from glioblastoma (GBM) patients. MethodsBuffy coats and GBM TILs were collected from Karolinska Hospital. NK cells were isolated and expanded in vitro with IL-15 and IL-21 cytokines and feeder cells (K562 and K562E). Furthermore, tumor lysate was added in the cultures to boost memory responses in aNK cells. NK cell functionality, cytotoxicity, and phenotyping was assessed using flow cytometry and statistical analysis (t-test and two-way ANOVA) used to validate the results. Further animal model was used to validate the cytotoxicity capacities of these cells against GBM tumors using a zebrafish model. ResultsIL-21 drives the expansion of aNK better than IL-15 controls, data shown in PBMCs and TILs derived from GBM patients and IPLA OVCA patients. Additionally, the use of tumor lysate as a booster for restimulation further amplifies the cytotoxic capacity of aNK cells against autologous tumors. The zebrafish model validates this method, by decreasing the tumor size in zebrafish animals after 3 days of injection. ConclusionThe results demonstrate that IL-21 is essential for the specific expansion of aNK cells, enhancing their aggressiveness towards tumor cells. Additionally, tumor lysate significantly increases the cytotoxic efficacy of aNK cells upon restimulation with the same tumor cells. These findings suggest that IL-21 plays a crucial role in the specific expansion and activation of aNK cells, enhancing their aggressiveness towards tumor cells. By optimizing the expansion protocol, this method aims to advance the clinical application of aNK cells in immunotherapies for solid tumors, offering a potential solution to the limitations faced by current CAR-T therapies.

immunology↗

bioRxiv2024

Enhanced NK cell proficiency in solid tumors through antigen-specific memory via NKG2C/A-HLA-E/ABC recruited to tumors by CXCR2

Natural killer (NK) cells have emerged as promising effectors in cancer immunotherapy due to their ability to recognize and eliminate tumor cells. To investigate the immunological memory and tumor reactivity of adaptive (a)NK cells in the context of desmoplastic tumors, we used human ovarian cancer as a model. Through in vitro culture systems resembling dendritic cell (DC)-mediated T cell activation, we demonstrated that aNK cells exhibit antigen-specific cytotoxic responses and memory generation towards ovarian tumor antigens. Furthermore, mature DCs presenting tumor-associated antigens induced the expansion of aNK cells, suggesting antigen-specific proliferation. Single-cell transcriptomics revealed a distinct genetic signature of aNK cells in tumor samples, characterized by a cytotoxic phenotype and interactions with myeloid cells, particularly DCs. The spatial analysis confirmed the intratumoral presence of aNK cells, with higher abundance in the tumor nest compared to conventional (c)NK cells. Functional assays demonstrated the cytotoxicity of expanded aNK cells against autologous ovarian tumors, accompanied by an activated receptor profile. Importantly, aNK cells displayed antigen-specific memory responses towards primary tumors, maintaining specificity over time. Blockade of NKG2C and HLA-E influenced aNK cell recall responses, indicating their roles in the adaptive NK cell immune memory. Additionally, CXCR2 was essential for efficient aNK cell migration toward tumors. These findings shed light on the therapeutic potential of aNK cells in ovarian cancer immunotherapy, highlighting their ability to develop immunological memory and effectively eradicate tumor cells.

immunology↗

bioRxiv2022

Efficacy and safety of glycosphingolipid SSEA-4 targeting CAR-T cells in solid tumors

Chimeric antigen receptor (CAR) T-cell immunotherapies for solid tumors face critical challenges such as heterogeneous antigen expression. We characterized SSEA-4 cell-surface glycolipid as a target for CAR-T cell therapy. SSEA-4 is mainly expressed during embryogenesis but is also found in several cancer types making it an attractive tumor-associated antigen. Anti-SSEA-4 CAR-T cells were generated and assessed pre-clinically in vitro and in vivo for anti-tumor response and safety. SSEA-4 CAR-T cells effectively eliminated SSEA-4 positive cells in all tested cancer cell lines whereas SSEA-4 negative cells lines were not targeted. In vivo efficacy and safety studies using NSG mice and the high-grade serous ovarian cancer cell line OVCAR4 demonstrated a remarkable and specific anti-tumor response at all CAR-T cell doses used. At high T cell doses, CAR-T cell-treated mice showed signs of health deterioration after a follow-up period. However, severity of toxicity was reduced with delayed onset when lower CAR-T cell doses were used. Our data demonstrate the efficacy of anti-SSEA-4 CAR-T therapy; however, safety strategies, such as dose-limiting and/or equipping CAR-T cells with combinatorial antigen recognition should be implemented for its potential clinical translation.

cancer biology↗

bioRxiv2022

YAP and collagen remodelling support cell proliferation and tumour aggressiveness in uterine leiomyosarcoma

Fibrillar collagen deposition, stiffness, and downstream signalling support the development of leiomyomas (LM), common benign mesenchymal tumours of the uterus, and are associated with aggressiveness in multiple carcinomas. Compared to epithelial carcinomas, however, the impact of fibrillar collagens on malignant mesenchymal tumours, including uterine leiomyosarcoma (LMS), remains elusive. In this study, we analyse the network morphology and density of fibrillar collagens combined with the gene expression of LMS, LM and normal myometrium (MM). We find that, in contrast to LM, LMS tumours present low collagen density and increased expression of collagen-remodelling genes, features associated with tumour aggressiveness. Using collagen-based 3D matrices, we show that the activity of MMP14, a central protein with collagen-remodelling functions particularly overexpressed in LMS, is necessary for LMS cell proliferation. In addition, we find that, unlike MM and LM cells, LMS proliferation and migration are not affected by collagen substrate stiffness. We demonstrate that LMS cell growth in low matrix adhesion microenvironments is supported by an enhanced basal YAP activity. Altogether, our results indicate that LMS cells acquire high collagen remodelling capabilities and are adapted to grow and migrate in low collagen and soft microenvironments. These results further suggest that matrix remodelling and YAP are potential therapeutic targets for this deadly disease.

cancer biology↗

bioRxiv2022

Mechanical confinement and DDR1 signalling synergise to regulate collagen-induced apoptosis in rhabdomyosarcoma cells

Fibrillar collagens promote cell proliferation, migration, and survival in various epithelial cancers and are generally associated with tumour aggressiveness. However, the impact of fibrillar collagens on soft tissue sarcoma behaviour remains poorly understood. Unexpectedly, we find here that fibrillar collagen-related gene expression is associated with favourable patient prognosis in rhabdomyosarcoma. By developing and using collagen matrices with distinct stiffness and in vivo-like microarchitectures, we uncover that the activation of DDR1 has pro-apoptotic and integrin {beta}1 pro-survival function, specifically in 3D rhabdomyosarcoma cell cultures. We demonstrate that rhabdomyosarcoma cell-intrinsic or extrinsic matrix remodelling promotes cell survival. Mechanistically, we find that the 3D-specific collagen-induced apoptosis results from a dual DDR1-independent and a synergistic DDR1-dependent TRPV4-mediated response to mechanical confinement. Altogether, our results indicate that dense microfibrillar collagen-rich microenvironments are detrimental to rhabdomyosarcoma cells through an apoptotic response orchestrated by the induction of DDR1 signalling and mechanical confinement. This mechanism helps to explain the preference of rhabdomyosarcoma cells to grow in and metastasise to low fibrillar collagen microenvironments such as the lung.

cancer biology↗

bioRxiv2021

Regulatory FOXP3+ T cells in uterine sarcomas are associated with favorable prognosis, low extracellular matrix expression and reduced YAP activation

PurposeUterine sarcomas are rare but deadly malignancies without effective treatment. The goal of this study was to characterize and identify potential mechanisms underlying observed variations in the immune microenvironment of different sarcoma subtypes, using integrated clinicopathological and molecular methods. Experimental designFifty-eight cases of uterine sarcoma with full clinicopathological annotation were analyzed for their immune landscape in the tumor microenvironment, gene, and protein expression. Cases included leiomyosarcoma (LMS; n=13), low-grade endometrial stromal sarcoma (ESS; n=16), undifferentiated uterine sarcoma (UUS; n=26), and YWHAE-FAM22 translocation-bearing ESS (YFAM; n=3). Image analysis was used to quantify immune cells and immune regulatory proteins. Gene ontology and network enrichment analysis of matching transcriptomic data was used to relate over- and under expressed genes to pathways and further to the immune phenotype and clinicopathological findings. ResultsImmune cell characterization revealed overall prevalence of regulatory T cells and the pro-tumor M2-like macrophages. Cytotoxic T cells were only found in ESS and UUS tumors. Expression of immune regulatory proteins was heterogeneous, with PD-L1 being undetectable. Hierarchical clustering of patients showed four immune signatures independent of tumor type, where infiltration of non-exhausted FOXP3+ cells and M1-like macrophages were associated with greater overall survival. High CD8+/FOXP3+ ratio in UUS and ESS was associated with poor survival and upregulation of extracellular matrix (ECM)-related genes and proteins and YAP nuclear localization. ConclusionsUterine sarcomas present distinct immune signatures with prognostic value, independent of tumor type. This study suggests that the ECM is a potential regulator of the immune microenvironment in uterine sarcomas.

cancer biology↗

bioRxiv2020

PROX1 transcription factor is a master regulator of myogenic and oncogenic features of rhabdomyosarcoma

Rhabdomyosarcoma (RMS) is an aggressive pediatric soft tissue cancer in need for novel therapies. Here we show that the PROX1 transcription factor, which is essential for normal myoblast differentiation, is highly expressed in RMS tumors. We demonstrate that PROX1 is needed for RMS cell stemness and growth in vitro, and for RMS tumor formation in mouse xenograft models. In addition, we unveil that PROX1 is an essential for myogenic properties in RMS. PROX1 depletion reprogrammed the RMS transcriptome to resemble benign mesenchymal stem cells and repressed many of the previously identified RMS effector transcripts and myogenic genes. By using proximity labeling and mass spectrometry, we found that PROX1 interacts with the NuRD and CoREST complexes containing class I HDACs. Our studies reveal a major role of PROX1-HDAC interaction in RMS and give insights that inhibiting this interaction could be a promising therapeutic approach.

cancer biology↗

bioRxiv2020

Multi-modal meta-analysis of cancer cell line omics profiles identifies ECHDC1 as a novel breast tumor suppressor

A caveat of cancer cell line models is that their molecular and functional profiling is subject to laboratory-specific experimental practices and data analysis protocols. The current challenge is how to make an integrated use of omics profiles of cancer cell lines for reliable discoveries. Here, we carried out a systematic analysis of nine types of data modalities using meta-analysis of 53 omics profiling studies across 12 research laboratories for 2018 cell lines. To account for relatively low consistency observed for certain data modalities, we developed a robust data integration approach that identifies reproducible signals shared among multiple data modalities and studies. We demonstrated the power of the integrative analyses by identifying a novel driver gene, ECHDC1, with tumor suppressive role validated both in breast cancer cells and patient tumors. Extension of the approach identified synthetic lethal partners of cancer drivers, including a co-dependency of PTEN deficient cells on RNA helicases. HighlightsO_LIA comprehensive meta-analysis of 53 multi-modal omics profiles of >2000 cancer cell lines from 12 research laboratories C_LIO_LIAn unexpected lack of consistency between TMT-labelled and non-labelled global proteomic profiles C_LIO_LIA non-parametric approach to integrate omics profiles from multiple laboratories and to identify robust molecular patterns in individual cell lines C_LIO_LIThe multi-modal data integration reveals novel drivers and potential therapeutic targets, including ECHDC1 in breast cancers and DDX27 in PTEN mutant cancers. C_LI

cancer biology↗

bioRxiv2018

FGFR4 phosphorylates MST1 to confer breast cancer cells resistance to MST1/2-dependent apoptosis

Cancer cells balance with the equilibrium of cell death and growth to expand and metastasize. The activity of mammalian sterile20-like kinases MST1/2 has been linked to apoptosis and tumor suppression via YAP/Hippo pathway dependent and independent mechanisms. With a kinase substrate screen we identified here MST1 and MST2 among the top substrates for fibroblast growth factor receptor 4 (FGFR4). In COS-1 cells, MST1 was phosphorylated at Y433 residue in an FGFR4 kinase activity-dependent manner, as assessed by mass spectrometry. Blockade of this phosphorylation by Y433F mutation induced MST1 activation, as reflected by increased autophosphorylation at T183 in FGFR4 overexpressing MDA-MB-231 cells. Importantly, the specific short-term inhibition or knockdown of FGFR4 also led to MST1/2 activation in conjunction with induction of MST1/2-dependent apoptosis in an endogenous model of HER2+ breast cancer cells. Moreover, FGFR4 knockdown increased the level of active nuclear MST1 coincidentally with cell polarization and membrane-association of YAP in three-dimensional breast cancer cell spheres. Consistently, FGFR4 overexpression correlated with reduced Hippo pathway-mediated, nuclear translocation-inhibiting YAP phosphorylation, and abysmal HER2+ breast carcinoma patient outcome in TCGA cohort. Our results reveal a novel mechanism for FGFR4 oncogenic activity via suppression of the stress-associated MST1/2-dependent apoptosis machinery in the tumor cells with prominent HER/ERBB signaling driven proliferation.

cancer biology↗

bioRxiv2018

Vulnerability of invasive glioma cells to lysosomal membrane instabilization

Diffusive by nature, glioma challenges clinical care by the impossibility of complete surgical resection of tumor, leaving the radio- and chemoresistant cells responsible for recurrence intact. We identified mammary-derived growth inhibitor (MDGI/FABP3) as invasive glioma biomarker. Here, we show that high MDGI expression associated with poor patient survival and promoted invasive glioma cell growth both in vitro and in vivo, while MDGI silencing drastically compromised patient-derived tumoroid viability via induction of lysosomal membrane permeabilization (LMP). This alternative cell death pathway provokes release of lysosomal hydrolases into the cytosol leading inevitably to the cell death. Our results show a novel functional role for MDGI in glioma cell invasion, survival, and maintenance of the lysosomal membrane integrity as well as an unsuspected sensitivity of glioma cells to an LMP-inducing drug, anti-histamine clemastine. In a preclinical study, clemastine-treatment significantly prolonged the survival of intracranial glioblastoma-bearing animals due to eradication of invasive glioma cells. This glioma cell vulnerability to LMP-inducing drugs opens new horizons for development of novel treatments and suggest re-positioning of an established drug for new indication.

cancer biology↗