Establishing normal metabolism and differentiation in hepatocellular carcinoma cells by culturing in adult human serum

Genome-scale metabolic models in translational medicine: the current status and potential of machine learning in improving the effectiveness of the models

The genome-scale metabolic model (GEM) has emerged as one of the leading modeling approaches for systems-level metabolic studies and has been widely explored for a broad range of organisms and applications. Owing to the development of genome sequencing technologies and available biochemical data, it is possible to reconstruct GEMs for model and non-model microorganisms as well as for multicellular organisms such as humans and animal models. GEMs will evolve in parallel with the availability of biological data, new mathematical modeling techniques and the development of automated GEM reconstruction tools. The use of high-quality, context-specific GEMs, a subset of the original GEM in which inactive reactions are removed while maintaining metabolic functions in the extracted model, for model organisms along with machine learning (ML) techniques could increase their applications and effectiveness in translational research in the near future. Here, we briefly review the current state of GEMs, discuss the potential contributions of ML approaches for more efficient and frequent application of these models in translational research, and explore the extension of GEMs to integrative cellular models.

Differential Anti-Fibrotic and Remodeling Responses of Human Dermal Fibroblasts to Artemisia sp., Artemisinin, and Its Derivatives

Fibrosis is a ubiquitous pathology, and prior studies have indicated that various artemisinin (ART) derivatives (including artesunate (AS), artemether (AM), and dihydroartemisinin (DHA)) can reduce fibrosis in vitro and in vivo. The medicinal plant Artemisia annua L. is the natural source of ART and is widely used, especially in underdeveloped countries, to treat a variety of diseases including malaria. A. afra contains no ART but is also antimalarial. Using human dermal fibroblasts (CRL-2097), we compared the effects of A. annua and A. afra tea infusions, ART, AS, AM, DHA, and a liver metabolite of ART, deoxyART (dART), on fibroblast viability and expression of key fibrotic marker genes after 1 and 4 days of treatment. AS, DHA, and Artemisia teas reduced fibroblast viability 4 d post-treatment in up to 80% of their respective controls. After 4 d of treatment, AS DHA and Artemisia teas downregulated ACTA2 up to 10 fold while ART had no significant effect, and AM increased viability by 10%. MMP1 and MMP3 were upregulated by AS, 17.5 and 32.6 fold, respectively, and by DHA, 8 and 51.8 fold, respectively. ART had no effect, but A. annua and A. afra teas increased MMP3 5 and 16-fold, respectively. Although A. afra tea increased COL3A1 5 fold, MMP1 decreased >7 fold with no change in either transcript by A. annua tea. Although A. annua contains ART, it had a significantly greater anti-fibrotic effect than ART alone but was less effective than A. afra. Immunofluorescent staining for smooth-muscle α-actin (α-SMA) correlated well with the transcriptional responses of drug-treated fibroblasts. Together, proliferation, qPCR, and immunofluorescence results show that treatment with ART, AS, DHA, and the two Artemisia teas yield differing responses, including those related to fibrosis, in human dermal fibroblasts, with evidence also of remodeling of fibrotic ECM.

Transcriptome-based biomarker prediction for Parkinson's disease using genome-scale metabolic modeling

Parkinson's disease (PD) is the second most common neurodegenerative disease in the world. Identification of PD biomarkers is crucial for early diagnosis and to develop target-based therapeutic agents. Integrative analysis of genome-scale metabolic models (GEMs) and omics data provides a computational approach for the prediction of metabolite biomarkers. Here, we applied the TIMBR (Transcriptionally Inferred Metabolic Biomarker Response) algorithm and two modified versions of TIMBR to investigate potential metabolite biomarkers for PD. To this end, we mapped thirteen post-mortem PD transcriptome datasets from the substantia nigra region onto Human-GEM. We considered a metabolite as a candidate biomarker if its production was predicted to be more efficient by a TIMBR-family algorithm in control or PD case for the majority of the datasets. Different metrics based on well-known PD-related metabolite alterations, PD-associated pathways, and a list of 25 high-confidence PD metabolite biomarkers compiled from the literature were used to compare the prediction performance of the three algorithms tested. The modified algorithm with the highest prediction power based on the metrics was called TAMBOOR, TrAnscriptome-based Metabolite Biomarkers by On-Off Reactions, which was introduced for the first time in this study. TAMBOOR performed better in terms of capturing well-known pathway alterations and metabolite secretion changes in PD. Therefore, our tool has a strong potential to be used for the prediction of novel diagnostic biomarkers for human diseases.

Plasma metabolome of healthy and Rhodococcus equi-infected foals over time

BACKGROUND

Foals that develop pulmonary ultrasonographic lesions on Rhodococcus equi (R. equi) endemic farms are treated with antibiotics because those at risk of developing clinical pneumonia (~20%) cannot be recognised early. Candidate biomarkers identified using metabolomics may aid targeted treatment strategies against R. equi.

OBJECTIVES

(1) To describe how foal ageing affects their plasma metabolome (birth to 8 weeks) and (2) to establish the effects that experimental infection with Rhodococcus equi (R. equi) has on foal metabolome.

STUDY DESIGN

Experimental study.

METHODS

Nine healthy newborn foals were experimentally infected with R. equi as described in a previous study. Foals were treated with oral antibiotics if they developed clinical pneumonia (n = 4, clinical group) or remained untreated if they showed no signs of disease (n = 5, subclinical group). A group of unchallenged foals (n = 4) was also included in the study. By the end of the study period (8 weeks), all foals were free of disease. This status was confirmed with transtracheal wash fluid evaluation and culture as well as thoracic ultrasonography. Plasma metabolomics was determined by GC-MS weekly for the study duration (8 weeks).

RESULTS

Foals' plasma metabolome was altered by ageing (birth to 8 weeks) and experimental infection with R. equi as demonstrated using multivariate statistical analysis. The intensities of 25 and 28 metabolites were altered by ageing and infection (p < 0.05) respectively. Furthermore, 20 metabolites changed by more than 2-fold between clinical and subclinical groups.

MAIN LIMITATIONS

The number of foals is limited. Foals were experimentally infected with R. equi.

CONCLUSIONS

Ageing and R. equi infection induced changes in the plasma metabolome of foals. These results provide an initial description of foal's plasma metabolome and serve as background for future identification of R. equi pneumonia biomarkers.

Improving physiological relevance of cell culture: the possibilities, considerations, and future directions of the ex vivo coculture model

In vitro models provide an important platform for the investigation of cellular growth and atrophy to inform, or extend mechanistic insights from, logistically challenging in vivo trials. Although these models allow for the identification of candidate mechanistic pathways, many models involve supraphysiological dosages, nonphysiological conditions, or experimental changes relating to individual proteins or receptors, all of which limit translation to human trials. To overcome these drawbacks, the use of ex vivo human plasma and serum has been used in cellular models to investigate changes in myotube hypertrophy, cellular protein synthesis, anabolic and catabolic markers in response to differing age, disease states, and nutrient status. However, there are currently no concurrent guidelines outlining the optimal methodology for this model. This review discusses the key methodological considerations surrounding the use of ex vivo plasma and serum with a focus in application to skeletal muscle cell lines (i.e., C2C12, L6, and LHCN-M2) and human primary skeletal muscle cells (HSMCs) as a means to investigate molecular signaling in models of atrophy and hypertrophy, alongside future directions.

Disturbance of the Warburg effect by dichloroacetate and niclosamide suppresses the growth of different sub-types of malignant pleural mesothelioma in vitro and in vivo

Background: Inhalation of asbestos fibers is the most common cause of malignant pleural mesothelioma (MPM). In 2004, the United States Food and Drug Administration approved a combination of cisplatin with pemetrexed to treat unresectable MPM. Nonetheless novel treatment is urgently needed. The objective of this study is to report the combination effect of dichloroacetate (DCA) or niclosamide (Nic) Nic in MPM.

Materials and methods: The effect of a combination of DCA and Nic was studied using a panel of MPM cell lines (H28, MSTO-211H, H226, H2052, and H2452). Cell viability was monitored by MTT assay. Glycolysis, oxidative phosphorylation, glucose, glycogen, pyruvate, lactate, citrate, succinate and ATP levels were determined by corresponding ELISA. Apoptosis, mitochondrial transmembrane potential, cell cycle analysis, hydrogen peroxide and superoxide were investigated by flow cytometry. Cell migration and colony formation were investigated by transwell migration and colony formation assays respectively. The in vivo effect was confirmed using 211H and H226 nude mice xenograft models.

Results and conclusion: Cell viability was reduced. Disturbance of glycolysis and/or oxidative phosphorylation resulted in downregulation of glycogen, citrate and succinate. DCA and/or Nic increased apoptosis, mitochondrial transmembrane depolarization, G2/M arrest and reactive oxygen species. Moreover, DCA and/or Nic suppressed cell migration and colony formation. Furthermore, a better initial tumor suppressive effect was induced by the DCA/Nic combination compared with either drug alone in both 211H and H226 xenograft models. In H226 xenografts, DCA/Nic increased median survival of mice compared with single treatment. Single drug and/or a combination disturbed the Warburg effect and activated apoptosis, and inhibition of migration and proliferation in vivo. In conclusion, dichloroacetate and/or niclosamide showed a tumor suppressive effect in MPM in vitro and in vivo, partially mediated by disturbance of glycolysis/oxidative phosphorylation, apoptosis, ROS production, G2/M arrest, and suppression of migration and proliferation.

Constraint-Based Reconstruction and Analyses of Metabolic Models: Open-Source Python Tools and Applications to Cancer

The influence of metabolism on signaling, epigenetic markers, and transcription is highly complex yet important for understanding cancer physiology. Despite the development of high-resolution multi-omics technologies, it is difficult to infer metabolic activity from these indirect measurements. Fortunately, genome-scale metabolic models and constraint-based modeling provide a systems biology framework to investigate the metabolic states and define the genotype-phenotype associations by integrations of multi-omics data. Constraint-Based Reconstruction and Analysis (COBRA) methods are used to build and simulate metabolic networks using mathematical representations of biochemical reactions, gene-protein reaction associations, and physiological and biochemical constraints. These methods have led to advancements in metabolic reconstruction, network analysis, perturbation studies as well as prediction of metabolic state. Most computational tools for performing these analyses are written for MATLAB, a proprietary software. In order to increase accessibility and handle more complex datasets and models, community efforts have started to develop similar open-source tools in Python. To date there is a comprehensive set of tools in Python to perform various flux analyses and visualizations; however, there are still missing algorithms in some key areas. This review summarizes the availability of Python software for several components of COBRA methods and their applications in cancer metabolism. These tools are evolving rapidly and should offer a readily accessible, versatile way to model the intricacies of cancer metabolism for identifying cancer-specific metabolic features that constitute potential drug targets.

Optimisation of the HepaRG cell line model for drug toxicity studies using two different cultivation conditions: advantages and limitations

The HepaRG cell line represents a successful model for hepatotoxicity studies. These cells are of human origin and are differentiated in vitro into mature and functional hepatocyte-like cells. The objective of this research was to compare two different culture protocols, Sison-Young et al. 2017 (hereinafter referred as Sison) and Gripon et al. 2002 (hereinafter referred as Biopredic) for HepaRG cells in order to optimise this model for drug metabolism and toxicity testing studies. HepaRG cells obtained from the same batch were cultured according to the described protocols. Using both protocols, differentiated HepaRG cells retained their drug metabolic capacity (major phase I/II enzymes) and transporters, as well as their morphological characteristics. Morphologically, HepaRG cells cultured after the Biopredic protocol formed more apical membranes and small ductular-like structures, than those cultivated using the Sison protocol. Also, the efflux activity of multidrug resistance protein 1 (MDR1) and multidrug resistance-associated protein 1 (MRP1) as well as the activity of uridine-glucuronosyltransferase (UGT) and glutathione S-transferase (GST) were significantly reduced in HepaRG cultured using the Sison protocol. Applying well-established drug cocktails to measure cytochrome P450 (CYPs) activity, we found that production of the corresponding metabolites was hampered in Sison-cultured HepaRG cells, indicating that the activity of CYP1A2, CYP2C9, CYP3A4, CYP2B6 and CYP2C19 was significantly reduced. Moreover, HepaRG sensitivity to well-known drugs, namely diclofenac, amiodarone, imipramine and paracetamol, revealed some differences between the two culture protocols. Furthermore, the HepaRG cells can be maintained with higher viability and sufficient CYPs activity and expression (i.e. CYP3A4, CYP1A2 and CYP2B6) as well as liver-specific functions, using Biopredic compared with the Sison culture protocol. These maintained liver-specific functions might be dependent on the prolongation of the culture conditions in the case of the Biopredic protocol. In conclusion, based on the metabolic activity of HepaRG cells using the standard protocol from Biopredic, we believe that this protocol is optimal for investigating drug metabolism and pharmacokinetic screening studies.

Different Serum, Different Protein Corona! The Impact of the Serum Source on Cellular Targeting of Folic Acid-Modified Chitosan-Based Nanoparticles

The nanoparticle (NP) protein corona represents an interface between biological components and NPs, dictating their cellular interaction and biological fate. To assess the success of cellular targeting, NPs modified with targeting ligands are incubated with target cells in serum-free culture medium or in the presence of fetal bovine serum (FBS). In the former, the role of the corona is overlooked, and in the latter, the effects of a corona that does not represent the one forming in humans nor the respective disease state are considered. Via proteomic analysis, we demonstrate how the difference in the composition of FBS, sera from healthy human volunteers, and breast cancer patients (BrCr Pt) results in the formation of completely different protein coronas around the same NP. Successful in vitro targeting of breast cancer cells was only observed when NPs were incubated with target cells in the presence of BrCr Pt sera only. In such cases, the success of targeting was not attributed to the targeting ligand itself, but to the adsorption of specific serum proteins that facilitate NP uptake by cancer cells in the presence of BrCr Pt sera. This work therefore demonstrates how the serum source affects the reliability of in vitro experiments assessing NP-cell interactions and the consequent success or failure of active targeting and may in fact indicate an additional reason for the limited clinical success of drug targeting by NPs in cancer.

First Small-Molecule Inhibitors Targeting the RNA-Binding Protein IGF2BP2/IMP2 for Cancer Therapy

The RNA-binding protein IGF2BP2/IMP2/VICKZ2/p62 is overexpressed in several tumor entities, promotes tumorigenesis and tumor progression, and has been suggested to worsen the disease outcome. The aim of this study is to (I) validate IMP2 as a potential target for colorectal cancer, (II) set up a screening assay for small-molecule inhibitors of IMP2, and (III) test the biological activity of the obtained hit compounds. Analyses of colorectal and liver cancer gene expression data showed reduced survival in patients with a high IMP2 expression and in patients with a higher IMP2 expression in advanced tumors. In vitro target validation in 2D and 3D cell cultures demonstrated a reduction in cell viability, migration, and proliferation in IMP2 knockout cells. Also, xenotransplant tumor cell growth in vivo was significantly reduced in IMP2 knockouts. Different compound libraries were screened for IMP2 inhibitors using a fluorescence polarization assay, and the results were confirmed by the thermal shift assay and saturation-transfer difference NMR. Ten compounds, which belong to two classes, that is, benzamidobenzoic acid class and ureidothiophene class, were validated in vitro and showed a biological target specificity. The three most active compounds were also tested in vivo and exhibited reduced tumor xenograft growth in zebrafish embryos. In conclusion, our findings support that IMP2 represents a druggable target to reduce tumor cell proliferation.

Influence of vitamin D treatment on functional expression of drug disposition pathways in human kidney proximal tubule cells during simulated uremia

Effects of cholecalciferol (VitD₃) and calcitriol (1,25-VitD₃), on the expression and function of major vitamin D metabolizing enzymes (cytochrome P450 [CYP]2R1, CYP24A1) and select drug transport pathways (ABCB1/P-gp, SLCO4C1/OATP4C1) were evaluated in human kidney proximal tubule epithelial cells (hPTECs) under normal and uraemic serum conditions.hPTECs were incubated with 10% normal or uraemic serum for 24 h followed by treatment with 2% ethanol vehicle, or 100 and 240 nM doses of VitD₃, or 1,25-VitD₃ for 6 days. The effects of treatment on mRNA and protein expression and functional activity of select CYP enzymes and transporters were assessedUnder uraemic serum, treatment with 1,25-VitD₃ resulted in increased mRNA but decreased protein expression of CYP2R1. Activity of CYP2R1 was not influenced by serum or VitD analogues. CYP24A1 expression was increased with 1,25-VitD₃ under normal as well as uraemic serum, although to a lesser extent. ABCB1/P-gp mRNA expression increased under normal and uraemic serum, with exposure to 1,25-VitD₃. SLCO4C1/OATP4C1 exhibited increased mRNA but decreased protein expression, under uraemic serum + 1,25-VitD₃. Functional assessments of transport showed no changes regardless of exposure to serum or 1,25-VitD₃.Key findings indicate that uraemic serum and VitD treatment led to differential effects on the functional expression of CYPs and transporters in hPTECs.

In Vitro Infection with Hepatitis B Virus Using Differentiated Human Serum Culture of Huh7.5-NTCP Cells without Requiring Dimethyl Sulfoxide

An estimated two billion people worldwide have been infected with hepatitis B virus (HBV). Despite the high infectivity of HBV in vivo, a lack of easily infectable in vitro culture systems hinders studies of HBV. Overexpression of the sodium taurocholate co-transporting polypeptide (NTCP) bile acid transporter in hepatoma cells improved infection efficiency. We report here a hepatoma cell culture system that does not require dimethyl sulfoxide (DMSO) for HBV infection. We overexpressed NTCP in Huh7.5 cells and allowed these cells to differentiate in a medium supplemented with human serum (HS) instead of fetal bovine serum (FBS). We show that human serum culture enhanced HBV infection in Huh7.5-NTCP cells, e.g., in HS cultures, HBV pgRNA levels were increased by as much as 200-fold in comparison with FBS cultures and 19-fold in comparison with FBS+DMSO cultures. Human serum culture increased levels of hepatocyte differentiation markers, such as albumin secretion, in Huh7.5-NTCP cells to similar levels found in primary human hepatocytes. N-glycosylation of NTCP induced by culture in human serum may contribute to viral entry. Our study demonstrates an in vitro HBV infection of Huh7.5-NTCP cells without the use of potentially toxic DMSO.

Rescue Potential of Supportive Embryo Culture Conditions on Bovine Embryos Derived from Metabolically Compromised Oocytes

Elevated non-esterified fatty acid (NEFA), predominantly palmitic acid (PA), concentrations in blood and follicular fluid are a common feature in maternal metabolic disorders such as obesity. This has a direct negative impact on oocyte developmental competence and the resulting blastocyst quality. We use NEFA-exposure during bovine oocyte in vitro maturation (IVM) as a model to mimic oocyte maturation under maternal metabolic stress conditions. However, the impact of supportive embryo culture conditions on these metabolically compromised zygotes are not known yet. We investigated if the addition of anti-apoptotic, antioxidative and mitogenic factors (namely, Insulin-Transferrin-Selenium (ITS) or serum) to embryo culture media would rescue development and important embryo quality parameters (cell proliferation, apoptosis, cellular metabolism and gene expression patterns) of bovine embryos derived from high PA- or high NEFA-exposed oocytes when compared to controls (exposed to basal NEFA concentrations). ITS supplementation during in vitro culture of PA-exposed oocytes supported the development of lower quality embryos during earlier development. However, surviving blastocysts were of inferior quality. In contrast, addition of serum to the culture medium did not improve developmental competence of PA-exposed oocytes. Furthermore, surviving embryos displayed higher apoptotic cell indices and an aberrant cellular metabolism. We conclude that some supportive embryo culture supplements like ITS and serum may increase IVF success rates of metabolically compromised oocytes but this may increase the risk of reduced embryo quality and may thus have other long-term consequences.

Modifying nutritional substrates induces macrovesicular lipid droplet accumulation and metabolic alterations in a cellular model of hepatic steatosis

BACKGROUND AND AIMS

Nonalcoholic fatty liver disease (NAFLD) begins with steatosis, where a mixed macrovesicular pattern of large and small lipid droplets (LDs) develops. Since in vitro models recapitulating this are limited, the aims of this study were to develop mixed macrovesicular steatosis in immortalized hepatocytes and investigate effects on intracellular metabolism by altering nutritional substrates.

METHODS

Huh7 cells were cultured in 11 mM glucose and 2% human serum (HS) for 7 days before additional sugars and fatty acids (FAs), either with 200 µM FAs (low fat low sugar; LFLS), 5.5 mM fructose + 200 µM FAs (low fat high sugar; LFHS), or 5.5 mM fructose + 800 µM FAs (high fat high sugar; HFHS), were added for 7 days. FA metabolism, lipid droplet characteristics, and transcriptomic signatures were investigated.

RESULTS

Between the LFLS and LFHS conditions, there were few notable differences. In the HFHS condition, intracellular triacylglycerol (TAG) was increased and the LD pattern and distribution was similar to that found in primary steatotic hepatocytes. HFHS-treated cells had lower levels of de novo-derived FAs and secreted larger, TAG-rich lipoprotein particles. RNA sequencing and gene set enrichment analysis showed changes in several pathways including those involved in metabolism and cell cycle.

CONCLUSIONS

Repeated doses of HFHS treatment resulted in a cellular model of NAFLD with a mixed macrovesicular LD pattern and metabolic dysfunction. Since these nutrients have been implicated in the development of NAFLD in humans, the model provides a good physiological basis for studying NAFLD development or regression in vitro.

Generation of new hepatocyte-like in vitro models better resembling human lipid metabolism

In contrast to human hepatocytes in vivo, which solely express acyl-coenzyme A:cholesterol acyltransferase (ACAT) 2, both ACAT1 and ACAT2 (encoded by SOAT1 and SOAT2) are expressed in primary human hepatocytes and in human hepatoma cell lines. Here, we aimed to create hepatocyte-like cells expressing the ACAT2, but not the ACAT1, protein to generate a model that - at least in this regard - resembles the human condition in vivo and to assess the effects on lipid metabolism. Using the Clustered Regularly Interspaced Short Palindromic Repeats technology, we knocked out SOAT1 in HepG2 and Huh7.5 cells. The wild type and SOAT2-only-cells were cultured with fetal bovine or human serum and the effects on lipoprotein and lipid metabolism were studied. In SOAT2-only-HepG2 cells, increased levels of cholesterol, triglycerides, apolipoprotein B and lipoprotein(a) in the cell media were detected; this was likely dependent of the increased expression of key genes involved in lipid metabolism (e.g. MTP, APOB, HMGCR, LDLR, ACACA, and DGAT2). Opposite effects were observed in SOAT2-only-Huh7.5 cells. Our study shows that the expression of SOAT1 in hepatocyte-like cells contributes to the distorted phenotype observed in HepG2 and Huh7.5 cells. As not only parameters of lipoprotein and lipid metabolism but also some markers of differentiation/maturation increase in the SOAT2-only-HepG2 cells cultured with HS, this cellular model represent an improved model for studies of lipid metabolism.

In Vitro Hepatitis C Virus Infection and Hepatic Choline Metabolism

Choline is an essential nutrient required for normal neuronal and muscular development, as well as homeostatic regulation of hepatic metabolism. In the liver, choline is incorporated into the main eukaryotic phospholipid, phosphatidylcholine (PC), and can enter one-carbon metabolism via mitochondrial oxidation. Hepatitis C virus (HCV) is a hepatotropic positive-strand RNA virus that similar to other positive-strand RNA viruses and can impact phospholipid metabolism. In the current study we sought to interrogate if HCV modulates markers of choline metabolism following in vitro infection, while subsequently assessing if the inhibition of choline uptake and metabolism upon concurrent HCV infection alters viral replication and infectivity. Additionally, we assessed whether these parameters were consistent between cells cultured in fetal bovine serum (FBS) or human serum (HS), conditions known to differentially affect in vitro HCV infection. We observed that choline transport in FBS- and HS-cultured Huh7.5 cells is facilitated by the intermediate affinity transporter, choline transporter-like family (CTL). HCV infection in FBS, but not HS-cultured cells diminished CTL1 transcript and protein expression at 24 h post-infection, which was associated with lower choline uptake and lower incorporation of choline into PC. No changes in other transporters were observed and at 96 h post-infection, all differences were normalized. Reciprocally, limiting the availability of choline for PC synthesis by use of a choline uptake inhibitor resulted in increased HCV replication at this early stage (24 h post-infection) in both FBS- and HS-cultured cells. Finally, in chronic infection (96 h post-infection), inhibiting choline uptake and metabolism significantly impaired the production of infectious virions. These results suggest that in addition to a known role of choline kinase, the transport of choline, potentially via CTL1, might also represent an important and regulated process during HCV infection.

Current status and applications of genome-scale metabolic models

Genome-scale metabolic models (GEMs) computationally describe gene-protein-reaction associations for entire metabolic genes in an organism, and can be simulated to predict metabolic fluxes for various systems-level metabolic studies. Since the first GEM for Haemophilus influenzae was reported in 1999, advances have been made to develop and simulate GEMs for an increasing number of organisms across bacteria, archaea, and eukarya. Here, we review current reconstructed GEMs and discuss their applications, including strain development for chemicals and materials production, drug targeting in pathogens, prediction of enzyme functions, pan-reactome analysis, modeling interactions among multiple cells or organisms, and understanding human diseases.