Biology subjects

Cetin-Atalay, R.

Publications and source records attributed to Cetin-Atalay, R..

12 recordsLinked to original sources

bioRxiv2026

Transfer Learning Enables Drug-Target Interaction Prediction in Data-Scarce One-Carbon Metabolism

Predicting drug-target interactions (DTIs) with deep learning offers opportunities to accelerate drug discovery, yet performance is constrained by the scarcity of target-specific training data. This is a particular challenge for mitochondrial one-carbon (1C) pathway enzymes, which are attractive therapeutic targets but remain pharmacologically understudied. Mitochondrial 1C metabolism supplies glycine, reducing equivalents, and 1C units critical for nucleotide synthesis, and has emerged as a key pathway in cancer and fibrosis. SHMT2 and MTHFD2, two key 1C enzymes, support collagen production in fibroblasts, blocking either prevents TGF-{beta}-induced glycine and collagen accumulation. Here, we developed transfer learning-based deep learning models to predict interactions between approved drugs and SHMT2 or MTHFD2 despite minimal target-specific training data, pre-training on large datasets from related enzymes before fine-tuning to these targets. Virtual screening of the DrugBank library identified six candidates, three of which, Carbimazole, Crizotinib, and GSK2018682 reduced TGF-{beta}-induced collagen production and glycine accumulation in human lung fibroblasts, demonstrating transfer learning as a strategy for repurposable drug identification in data-scarce metabolic targets.

bioinformatics↗

bioRxiv2025

Molecular Contrastive Learning with Graph Attention Network (MoCL-GAT) for Enhanced Molecular Representation

Learning the representation of molecules is crucial for drug discovery but is often hindered by the scarcity of labeled experimental data, which limits the performance of supervised machine learning models. While self-supervised learning (SSL) offers a solution by leveraging vast unlabeled chemical databases, many existing methods focus on learning from either local structural information or global molecular properties, but not both simultaneously. We introduce MoCL-GAT, a novel contrastive and transfer learning-based SSL framework that addresses this gap by simultaneously learning from two complementary objectives. It combines a local contrastive task on molecular subgraphs to capture fine-grained chemical environments with a global predictive task to learn holistic molecular descriptors. This dual-objective approach, powered by a Graph Attention Network, is designed to create more robust, versatile, and transferable molecular representations. Pre-trained on 1.9 million compounds, MoCL-GAT was fine-tuned on diverse benchmarks. It achieved state-of-the-art performance on molecular property prediction tasks, with an AUROC of 0.928 on BBBP and 0.749 on SIDER, and top-ranking RMSEs of 0.570 for ESOL and 1.818 for FreeSolv. Critically, fine-tuned models consistently and significantly outperformed models trained from scratch, confirming the value of pre-training. These results validate that MoCL-GATs dual-objective approach learns highly effective and transferable representations, enabling more accurate and data-efficient predictions for key cheminformatics challenges.

bioinformatics↗

bioRxiv2024

Role of Arginine and its Metabolism in TGF-β-Induced Activation of Lung Fibroblasts

Arginine is a conditionally essential amino acid with known roles in protein production, nitric oxide synthesis, biosynthesis of proline and polyamines, and regulation of intracellular signaling pathways. Arginine biosynthesis and catabolism have been linked to TGF-{beta}-induced activation of fibroblasts in the context of pulmonary fibrosis; however, a thorough study on the metabolic and signaling roles of arginine in the process of fibroblast activation has not been conducted. Here, we used metabolic dropouts and labeling strategies to determine how activated fibroblasts utilize arginine. We found that arginine limitation leads to activation of GCN2 while inhibiting TGF-{beta}-induced mTORC1 activation and collagen protein production. Extracellular citrulline could rescue the effect of arginine deprivation in an ASS1-dependent manner. Using metabolic tracers of arginine and its precursors, we found little evidence of arginine synthesis or catabolism in lung fibroblasts treated with TGF-{beta}. Extracellular ornithine or glutamine were the primary sources of ornithine and polyamines, not arginine. Our findings suggest that the major role for arginine in lung fibroblasts is for charging of arginyl-tRNAs and for promotion of mTOR signaling. Highlights- Arginine depletion inhibits TGF-{beta}-induced transcription in human lung fibroblasts (HLFs). - Arginine is not significantly catabolized in HLFs either through NOS or ARG dependent pathways. - Extracellular glutamine and ornithine are the primary sources of polyamines in lung fibroblasts. - The primary role of arginine in lung fibroblasts is for signaling through mTOR and GNC2.

cell biology↗

bioRxiv2024

HIF-1α regulates mitochondrial function in bone marrow-derived macrophages, but not in tissue-resident alveolar macrophages

HIF-1 plays a critical role in shaping macrophage phenotype and effector function. We have previously shown that tissue-resident alveolar macrophages (TR-AMs) have extremely low glycolytic capacity at steady-state, but can shift toward glycolysis under hypoxic conditions. Here, using inducible HIF-1 knockout (Hif1a-/-) TR-AMs and bone marrow-derived macrophages (BMDMs) and show that TR-AM HIF-1 is required for the glycolytic shift under prolyl hydroxylase inhibition, but is dispensable at steady-state for inflammatory effector function. In contrast, HIF-1 deletion in BMDMs led to diminished glycolytic capacity at steady-state and reduced inflammatory capacity, but higher mitochondrial function. Gene set enrichment analysis revealed enhanced c-Myc transcriptional activity in Hif1a-/- BMDMs, and upregulation of gene pathways related to ribosomal biogenesis and cellular proliferation. The findings highlight the heterogeneity of HIF-1 function in distinct macrophage populations and provide new insight into how HIF-1 regulates gene expression, inflammation, and metabolism in macrophages.

cell biology↗

bioRxiv2024

ATF4 and mTOR regulate metabolic reprogramming in TGF-β-treated lung fibroblasts

Idiopathic pulmonary fibrosis is a fatal disease characterized by the TGF-{beta}-dependent activation of lung fibroblasts, leading to excessive deposition of collagen proteins and progressive replacement of healthy lung with scar tissue. We and others have shown that fibroblast activation is supported by metabolic reprogramming, including the upregulation of the de novo synthesis of glycine, the most abundant amino acid found in collagen protein. How fibroblast metabolic reprogramming is regulated downstream of TGF-{beta} is incompletely understood. We and others have shown that TGF-{beta}-mediated activation of the Mechanistic Target of Rapamycin Complex 1 (mTORC1) and downstream upregulation of Activating Transcription Factor 4 (ATF4) promote increased expression of the enzymes required for de novo glycine synthesis; however, whether mTOR and ATF4 regulate other metabolic pathways in lung fibroblasts has not been explored. Here, we used RNA sequencing to determine how both ATF4 and mTOR regulate gene expression in human lung fibroblasts following TGF-{beta}. We found that ATF4 primarily regulates enzymes and transporters involved in amino acid homeostasis as well as aminoacyl-tRNA synthetases. mTOR inhibition resulted not only in the loss of ATF4 target gene expression, but also in the reduced expression of glycolytic enzymes and mitochondrial electron transport chain subunits. Analysis of TGF-{beta}-induced changes in cellular metabolite levels confirmed that ATF4 regulates amino acid homeostasis in lung fibroblasts while mTOR also regulates glycolytic and TCA cycle metabolites. We further analyzed publicly available single cell RNAseq data sets and found increased expression of ATF4 and mTOR metabolic targets in pathologic fibroblast populations from the lungs of IPF patients. Our results provide insight into the mechanisms of metabolic reprogramming in lung fibroblasts and highlight novel ATF4 and mTOR-dependent pathways that may be targeted to inhibit fibrotic processes.

cell biology↗

bioRxiv2024

Critical role of Gα12 and Gα13 proteins in TGF-β-induced myofibroblast differentiation.

Myofibroblast differentiation, characterized by accumulation of cytoskeletal and extracellular matrix proteins by fibroblasts, is a key process in wound healing and pathogenesis of tissue fibrosis. Transforming growth factor-{beta} (TGF-{beta}) is the most powerful known driver of myofibroblast differentiation. TGF-{beta} signals through transmembrane receptor serine/threonine kinases that phosphorylate Smad transcription factors (Smad2/3) leading to activation of transcription of target genes. Heterotrimeric G proteins mediate a distinct signaling from seven-transmembrane G protein coupled receptors, not commonly linked to Smad activation. We asked if G protein signaling plays any role in TGF-{beta}-induced myofibroblast differentiation, using primary cultured human lung fibroblasts. Activation of Gs by cholera toxin blocked TGF-{beta}-induced myofibroblast differentiation without affecting Smad2/3 phosphorylation. Inhibition of Gi by pertussis toxin, or siRNA-mediated combined knockdown of Gq and G11 had no significant effect on TGF-{beta}-induced myofibroblast differentiation. A combined knockdown of G12 and G13 resulted in a drastic inhibition of TGF-{beta}-stimulated expression of myofibroblast marker proteins (collagen-1, fibronectin, smooth-muscle -actin), with siG12 being significantly more potent than siG13. Mechanistically, a combined knockdown of G12 and G13 resulted in a substantially reduced phosphorylation of Smad2 and Smad3 in response to TGF-{beta}, which was accompanied by a significant decrease in the expression of TGF{beta} receptors (TGFBR1, TGFBR2) and of Smad3 under siG12/13 conditions. In conclusion, our study uncovers a novel role of G12/13 proteins in the control of TGF-{beta} signaling and myofibroblast differentiation.

cell biology↗

bioRxiv2023

Mitochondrial One-Carbon Metabolism is Required for TGF-β-Induced Glycine Synthesis and Collagen Protein Production

A hallmark of Idiopathic Pulmonary Fibrosis is the TGF-{beta}-dependent activation of lung fibroblasts, leading to excessive deposition of collagen proteins and progressive scarring. We have previously shown that synthesis of collagen by lung fibroblasts requires de novo synthesis of glycine, the most abundant amino acid in collagen protein. TGF-{beta} upregulates the expression of the enzymes of the de novo serine/glycine synthesis pathway in lung fibroblasts through mTORC1 and ATF4- dependent transcriptional programs. SHMT2, the final enzyme of the de novo serine/glycine synthesis pathway, transfers a one-carbon unit from serine to tetrahydrofolate (THF), producing glycine and 5,10-methylene-THF (meTHF). meTHF is converted back to THF in the mitochondrial one-carbon (1C) pathway through the sequential actions of MTHFD2 (which converts meTHF to 10-formyl-THF), and either MTHFD1L, which produces formate, or ALDH1L2, which produces CO2. It is unknown how the mitochondrial 1C pathway contributes to glycine biosynthesis or collagen protein production in fibroblasts, or fibrosis in vivo. Here, we demonstrate that TGF-{beta} induces the expression of MTHFD2, MTHFD1L, and ALDH1L2 in human lung fibroblasts. MTHFD2 expression was required for TGF-{beta}-induced cellular glycine accumulation and collagen protein production. Combined knockdown of both MTHFD1L and ALDH1L2 also inhibited glycine accumulation and collagen protein production downstream of TGF-{beta}; however knockdown of either protein alone had no inhibitory effect, suggesting that lung fibroblasts can utilize either enzyme to regenerate THF. Pharmacologic inhibition of MTHFD2 recapitulated the effects of MTHFD2 knockdown in lung fibroblasts and ameliorated fibrotic responses after intratracheal bleomycin instillation in vivo. Our results provide insight into the metabolic requirements of lung fibroblasts and provide support for continued development of MTHFD2 inhibitors for the treatment of IPF and other fibrotic diseases.

cell biology↗

bioRxiv2022

HIF-1α induces glycolytic reprogramming in tissue-resident alveolar macrophagesto promote survival during acute lung injury

Cellular metabolism is a critical regulator of macrophage effector function. Tissue-resident alveolar macrophages (TR-AMs) inhabit a unique niche marked by high oxygen and low glucose. We have recently shown that in contrast to bone marrow-derived macrophages (BMDMs), TR-AMs cannot utilize glycolysis and predominantly rely on mitochondrial function for their effector response. It is not known how changes in local oxygen concentration that occur during conditions such as acute respiratory distress syndrome (ARDS) might affect TR-AM metabolism and function; however, ARDS is associated with progressive loss of TR-AMs, which correlates with the severity of disease and mortality. Here, we demonstrate that hypoxia robustly stabilizes HIF-1 in TR-AMs to promote a glycolytic phenotype. Hypoxia altered TR-AM metabolite signatures, cytokine production, and decreased their sensitivity to the inhibition of mitochondrial function. By contrast, hypoxia had minimal effects on BMDM metabolism. The effects of hypoxia on TR-AMs were mimicked by FG-4592, a HIF-1 stabilizer. Treatment with FG-4592 decreased TR-AM death and attenuated acute lung injury in mice. These findings reveal the importance of microenvironment in determining macrophage metabolic phenotype, and highlight the therapeutic potential in targeting cellular metabolism to improve outcomes in diseases characterized by acute inflammation.

cell biology↗

bioRxiv2021

Protein Domain-Based Prediction of Compound-Target Interactions and Experimental Validation on LIM Kinases

Predictive approaches such as virtual screening have been used in drug discovery with the objective of reducing developmental time and costs. Current machine learning and network- based approaches have issues related to generalization, usability, or model interpretability, especially due to the complexity of target proteins structure/function, and bias in system training datasets. Here, we propose a new computational method "DRUIDom" to predict bio- interactions between drug candidate compounds and target proteins by utilizing the domain modularity of proteins, to overcome problems associated with current approaches. DRUIDom is composed of two methodological steps. First, ligands/compounds are statistically mapped to structural domains of their target proteins, with the aim of identifying physical or functional interactions. As such, other proteins containing the mapped domain or domain pair become new candidate targets for the corresponding compounds. Next, a million-scale dataset of small molecule compounds, including the ones mapped to domains in the previous step, are clustered based on their molecular similarities, and their domain associations are propagated to other compounds within the same clusters. Experimentally verified bioactivity data points, obtained from public databases, are meticulously filtered to construct datasets of active/interacting and inactive/non-interacting compound-target pairs ([~]2.9M data points), and used as training data for calculating parameters of compound- domain mappings, which led to 27,032 high-confidence associations between 250 domains and 8,165 compounds, and a finalized output of [~]5 million new compound-protein interactions. DRUIDom is experimentally validated by syntheses and bioactivity analyses of compounds predicted to target LIM-kinase proteins, which play critical roles in the regulation of cell motility, cell cycle progression, and differentiation through actin filament dynamics. We showed that LIMK-inhibitor-2 and its derivatives significantly block the cancer cell migration through inhibition of LIMK phosphorylation and the downstream protein cofilin. One of the derivative compounds (LIMKi-2d) was identified as a promising candidate due to its action on resistant Mahlavu liver cancer cells. The results demonstrated that DRUIDom can be exploited to identify drug candidate compounds for intended targets and to predict new target proteins based on the defined compound-domain relationships. The datasets, results, and the source code of DRUIDom are fully-available at: https://github.com/cansyl/DRUIDom. Author SummaryDrug development comprises several interlinked steps from designing drug candidate molecules to running clinical trials, with the aim to bring a new drug to market. A critical yet costly and labor-intensive stage is drug discovery, in which drug candidate molecules that specifically interact with the intended biomolecular target (mostly proteins) are identified. Lately, data-centric computational methods have been proposed to aid experimental procedures in drug discovery. These methods have the ability to rapidly assess large molecule libraries and reduce the time and cost of the process; however, most of them suffer from problems related to producing reliable biologically relevant results, preventing them from gaining real-world usage. Here, we have developed a new method called DRUIDom to predict unknown interactions between drugs/drug candidate compounds and biological targets by utilizing the modular structure of proteins. For this, we identify the domains, i.e., the evolutionary and functional building blocks of proteins, where these potential drug compounds can bind, and utilize this information along with protein domain annotations to predict new drug targets. We have tested the biological relevance of DRUIDom on selected proteins that play critical roles in the progression of numerous types of cancer. Cell-based experimental results indicated that predicted inhibitors are effective even on drug-resistant cancer cells. Our results suggest that DRUIDom produces novel and biologically relevant results that can be directly used in the early steps of the drug discovery process.

bioinformatics↗

bioRxiv2020

CROssBAR: Comprehensive Resource of Biomedical Relations with Deep Learning Applications and Knowledge Graph Representations

Systemic analysis of available large-scale biological and biomedical data is critical for developing novel and effective treatment approaches against both complex and infectious diseases. Owing to the fact that different sections of the biomedical data is produced by different organizations/institutions using various types of technologies, the data are scattered across individual computational resources, without any explicit relations/connections to each other, which greatly hinders the comprehensive multi-omics-based analysis of data. We aimed to address this issue by constructing a new biological and biomedical data resource, CROssBAR, a comprehensive system that integrates large-scale biomedical data from various resources and store them in a new NoSQL database, enrich these data with deep-learning-based prediction of relations between numerous biomedical entities, rigorously analyse the enriched data to obtain biologically meaningful modules and display them to users via easy-to-interpret, interactive and heterogenous knowledge graph (KG) representations within an open access, user-friendly and online web-service at https://crossbar.kansil.org. As a use-case study, we constructed CROssBAR COVID-19 KGs (available at: https://crossbar.kansil.org/covid_main.php) that incorporate relevant virus and host genes/proteins, interactions, pathways, phenotypes and other diseases, as well as known and completely new predicted drugs/compounds. Our COVID-19 graphs can be utilized for a systems-level evaluation of relevant virus-host protein interactions, mechanisms, phenotypic implications and potential interventions.

bioinformatics↗

bioRxiv2018

DEEPScreen: High Performance Drug-Target Interaction Prediction with Convolutional Neural Networks Using 2-D Structural Compound Representations

The identification of physical interactions between drug candidate chemical substances and target biomolecules is an important step in the process of drug discovery, where the standard procedure is the systematic screening of chemical compounds against pre-selected target proteins. However, experimental screening procedures are expensive and time consuming, therefore, it is not possible to carry out comprehensive tests. Within the last decade, computational approaches have been developed with the objective of aiding experimental studies by predicting novel drug-target interactions (DTI), via the construction and application of statistical models. In this study, we propose a large-scale DTI interaction prediction system, DEEPScreen, for early stage drug discovery, using convolutional deep neural networks. One of the main advantages of DEEPScreen is employing readily available simple 2-D images of compounds at the input level instead of engineered complex feature vectors that displayed limited performance in DTI prediction tasks previously. DEEPScreen learns complex features inherently from the 2-D molecular representations, thus producing highly accurate predictions. DEEPScreen system was trained for 704 target proteins (using ChEMBL curated bioactivity data) and finalized with rigorous hyper-parameter optimization tests. We compared the performance of DEEPScreen against shallow classifiers such as the random forest, logistic regression and support vector machines, to indicate the effectiveness of the proposed deep learning approach. Additionally, we compared DEEPScreen with other deep learning based state-of-the-art DTI predictors on widely used benchmark datasets and showed that DEEPScreen produces better or comparable results to the top performers. The method proposed here can be employed to computationally scan a large portion of the recorded drug candidate compound and protein spaces to aid the experimentalists working in the field of drug discovery and repurposing by providing a preselection of interesting novel DTIs.

bioinformatics↗

bioRxiv2018

Differential alteration of IL-8 in liver cancer stem cell enrichment in response to PI3K/Akt/mTOR inhibitors and sorafenib

Liver cancer stem cells (LCSCs) are derived from damaged and transformed Hepatic progenitor cells (HPCs) during precancerous cirrhosis stage. Ras/Raf/MAPK and PI3K/AKT/mTOR signaling pathways are significantly deregulated in liver cancer. The activation of PI3K/AKT/mTOR pathway in LCSC population is one of the reasons for acquired resistance to Sorafenib in advanced Hepatocellular carcinoma (HCC) patients. Therefore, identifying novel inhibitors targeting this pathway acting on LCSCs is highly essential. We therefore elucidated the bioactivities of small molecule kinase inhibitors on LCSCs acting through PI3K/Akt/mTOR pathway in comparison with DAPT (CSC inhibitor), DNA intercalators and Sorafenib. For this purpose, CD133+/EpCAM+ cells originated from HCC cells were analyzed by flow cytometry and effective inhibitors on LCSCs were further tested for their potential combinatorial effects. Treatment of cells with Sorafenib, and DNA intercalators resulted in enrichment of CD133+/EpCAM+ cells. Yet, mTOR inhibitor Rapamycin, and Notch pathway inhibitor DAPT significantly reduced CD133/EpCAM positivity. Combination studies revealed that sequential treatment strategy, which involves treatment of cells with Rapamycin prior to Sorafenib treatment, decreased the ratio of LCSCs as opposed to Sorafenib treatment alone or Sorafenib treatment prior to Rapamycin. The effect of the inhibitors were also demonstrated with LCSC sphere formation. Additionaly, a large panel of genes involved in cancer pathways were analyzed using Nanostring(R) nCounter(R) Technology to identify the differentially expressed genes in Rapamycin, Sorafenib or DAPT treated cells. Pathways involved in stemness (Wnt and Notch pathways) were differentially regulated between Rapamycin or DAPT treated cells and Sorafenib treated cells. Interleukin 8 (IL-8), FLNC, FLNA expressions were down-regulated upon treatment with DAPT or Rapamycin, yet up-regulated upon Sorafenib treatment. Following IL-8 inhibition CD133/EpCAM positivity of cells decreased significantly, indicating that IL-8 signaling is crucial for the conservation of stemness features of cancer cells.\n\nConclusionPI3K/Akt/mTOR pathway inhibitors alter hepatic CSC composition and gene expression in favor or to the detriment of cancer stem cell survival. Blockade of IL-8 signaling provides a promising therapeutic approach for prevention of LCSC enrichment.

cancer biology↗