HNCGAT: a method for predicting plant metabolite-protein interaction using heterogeneous neighbor contrastive graph attention network.

The prediction of metabolite-protein interactions (MPIs) plays an important role in plant basic life functions. Compared with the traditional experimental methods and the high-throughput genomics methods using statistical correlation, applying heterogeneous graph neural networks to the prediction of MPIs in plants can reduce the cost of manpower, resources, and time. However, to the best of our knowledge, applying heterogeneous graph neural networks to the prediction of MPIs in plants still remains under-explored. In this work, we propose a novel model named heterogeneous neighbor contrastive graph attention network (HNCGAT), for the prediction of MPIs in Arabidopsis. The HNCGAT employs the type-specific attention-based neighborhood aggregation mechanism to learn node embeddings of proteins, metabolites, and functional-annotations, and designs a novel heterogeneous neighbor contrastive learning framework to preserve heterogeneous network topological structures. Extensive experimental results and ablation study demonstrate the effectiveness of the HNCGAT model for MPI prediction. In addition, a case study on our MPI prediction results supports that the HNCGAT model can effectively predict the potential MPIs in plant.

Prediction and integration of metabolite-protein interactions with genome-scale metabolic models.

Metabolites, as small molecules, can act not only as substrates to enzymes, but also as effectors of activity of proteins with different functions, thereby affecting various cellular processes. While several experimental techniques have started to catalogue the metabolite-protein interactions (MPIs) present in different cellular contexts, characterizing the functional relevance of MPIs remains a challenging problem. Computational approaches from the constrained-based modeling framework allow for predicting MPIs and integrating their effects in the in silico analysis of metabolic and physiological phenotypes, like cell growth. Here, we provide a classification of all existing constraint-based approaches that predict and integrate MPIs using genome-scale metabolic networks as input. In addition, we benchmark the performance of the approaches to predict MPIs in a comparative study using different features extracted from the model structure and predicted metabolic phenotypes with the state-of-the-art metabolic networks of Escherichia coli and Saccharomyces cerevisiae. Lastly, we provide an outlook for future, feasible directions to expand the consideration of MPIs in constraint-based modeling approaches with wide biotechnological applications.

Identification of Plant Protein-Metabolite Interactions by Limited Proteolysis-Coupled Mass Spectrometry (LiP-MS).

The interactions between metabolites and proteins constitute crucial events in cell signaling and metabolism. In recent years, large-scale proteomics techniques have emerged to identify and characterize protein-metabolite interactions. However, their implementation in plants is generally lagging behind, preventing a complete understanding of the regulatory mechanisms governing plant physiology. Recently, a novel approach to identify metabolite-binding proteins, namely, limited proteolysis-coupled mass spectrometry (LiP-MS), was developed originally for microbial proteomes. Here, we present an adapted and accessible version of the LiP-MS protocol for use in plants. Plant proteomes are extracted and incubated with the metabolite of interest or control treatment, followed by a limited digestion by a nonspecific/promiscuous protease. Subsequently, a conventional shotgun proteomics sample preparation is performed including a complete digestion with the sequence-specific protease trypsin. Finally, label-free proteomics analysis is applied to identify structure-dependent proteolytic patterns corresponding to protein targets of the specific metabolite and their binding sites. Given its amenability to relatively high throughput, the LiP-MS approach may open a potent avenue for the discovery of novel regulatory mechanisms in plant species.

Studying Metabolism by NMR-Based Metabolomics.

During the past few decades, the direct analysis of metabolic intermediates in biological samples has greatly improved the understanding of metabolic processes. The most used technologies for these advances have been mass spectrometry (MS) and nuclear magnetic resonance (NMR) spectroscopy. NMR is traditionally used to elucidate molecular structures and has now been extended to the analysis of complex mixtures, as biological samples: NMR-based metabolomics. There are however other areas of small molecule biochemistry for which NMR is equally powerful. These include the quantification of metabolites (qNMR); the use of stable isotope tracers to determine the metabolic fate of drugs or nutrients, unravelling of new metabolic pathways, and flux through pathways; and metabolite-protein interactions for understanding metabolic regulation and pharmacological effects. Computational tools and resources for automating analysis of spectra and extracting meaningful biochemical information has developed in tandem and contributes to a more detailed understanding of systems biochemistry. In this review, we highlight the contribution of NMR in small molecule biochemistry, specifically in metabolic studies by reviewing the state-of-the-art methodologies of NMR spectroscopy and future directions.

Identification and Characterization of Robust Hepatocellular Carcinoma Prognostic Subtypes Based on an Integrative Metabolite-Protein Interaction Network.

Metabolite-protein interactions (MPIs) play key roles in cancer metabolism. However, our current knowledge about MPIs in cancers remains limited due to the complexity of cancer cells. Herein, the authors construct an integrative MPI network and propose a MPI network based hepatocellular carcinoma (HCC) subtyping and mechanism exploration workflow. Based on the expressions of hub proteins on the MPI network, two prognosis-distinctive HCC subtypes are identified. Meanwhile, multiple interdependent features of the poor prognostic subtype are observed, including hypoxia, DNA hypermethylation of metabolic pathways, fatty acid accumulation, immune pathway up-regulation, and exhausted T-cell infiltration. Notably, the immune pathway up-regulation is probably induced by accumulated unsaturated fatty acids which are predicted to interact with multiple immune regulators like SRC and TGFB1. Moreover, based on tumor microenvironment compositions, the poor prognostic subtype is further divided into two sub-populations showing remarkable differences in metabolism. The subtyping shows a strong consistency across multiple HCC cohorts including early-stage HCC. Overall, the authors redefine robust HCC prognosis subtypes and identify potential MPIs linking metabolism to immune regulations, thus promoting understanding and clinical applications about HCC metabolism heterogeneity.

Combination of network and molecule structure accurately predicts competitive inhibitory interactions.

Mining of metabolite-protein interaction networks facilitates the identification of design principles underlying the regulation of different cellular processes. However, identification and characterization of the regulatory role that metabolites play in interactions with proteins on a genome-scale level remains a pressing task. Based on availability of high-quality metabolite-protein interaction networks and genome-scale metabolic networks, here we propose a supervised machine learning approach, called CIRI that determines whether or not a metabolite is involved in a competitive inhibitory regulatory interaction with an enzyme. First, we show that CIRI outperforms the naive approach based on a structural similarity threshold for a putative competitive inhibitor and the substrates of a metabolic reaction. We also validate the performance of CIRI on several unseen data sets and databases of metabolite-protein interactions not used in the training, and demonstrate that the classifier can be effectively used to predict competitive inhibitory interactions. Finally, we show that CIRI can be employed to refine predictions about metabolite-protein interactions from a recently proposed PROMIS approach that employs metabolomics and proteomics profiles from size exclusion chromatography in E. coli to predict metabolite-protein interactions. Altogether, CIRI fills a gap in cataloguing metabolite-protein interactions and can be used in directing future machine learning efforts to categorize the regulatory type of these interactions.

Why and How to Dig into Plant Metabolite-Protein Interactions.

Interaction between metabolites and proteins drives cellular regulatory processes within and between organisms. Recent reports highlight that numerous plant metabolites embrace multiple biological activities, beyond a sole role as substrates, products, or cofactors of enzymes, or as defense or growth-regulatory compounds. Though several technologies have been developed to identify and characterize metabolite-protein interactions, the systematic implementation of such methods in the plant field remains limited. Here, we discuss the plant metabolic space, with a specific focus on specialized metabolites and their roles, and review the technologies to study their interaction with proteins. We approach it both from a plant's perspective, to increase our understanding of plant metabolite-dependent regulatory networks, and from a human perspective, to empower agrochemical and drug discoveries.

A Map of Protein-Metabolite Interactions Reveals Principles of Chemical Communication.

Metabolite-protein interactions control a variety of cellular processes, thereby playing a major role in maintaining cellular homeostasis. Metabolites comprise the largest fraction of molecules in cells, but our knowledge of the metabolite-protein interactome lags behind our understanding of protein-protein or protein-DNA interactomes. Here, we present a chemoproteomic workflow for the systematic identification of metabolite-protein interactions directly in their native environment. The approach identified a network of known and novel interactions and binding sites in Escherichia coli, and we demonstrated the functional relevance of a number of newly identified interactions. Our data enabled identification of new enzyme-substrate relationships and cases of metabolite-induced remodeling of protein complexes. Our metabolite-protein interactome consists of 1,678 interactions and 7,345 putative binding sites. Our data reveal functional and structural principles of chemical communication, shed light on the prevalence and mechanisms of enzyme promiscuity, and enable extraction of quantitative parameters of metabolite binding on a proteome-wide scale.