Leveraging machine learning to dissect role of combinations of amino acids in modulating the effect of zinc on mammalian cell growth.

Although the contributions of individual components of cell culture media are largely known, their combinatorial effects are far less understood. Experiments varying one component at a time cannot identify combinatorial effects, and analysis of the large number of experiments required to decipher such effects is challenging. Machine learning algorithms can help in the analysis of such datasets to identify multi-component interactions. Zinc toxicity in vitro is known to change depending on amino acid concentration in the extracellular medium. Multiple amino acids are known to be involved in this protection. Thirty-two amino acid compositions were formulated to evaluate their effect on the growth of CHO cells under high zinc conditions. A sequential machine learning analysis methodology was used, which led to the identification of a set of amino acids (threonine, proline, glutamate, aspartate, asparagine, and tryptophan) contributing to protection from zinc. Our results suggest that a decrease in availability of these set of amino acids due to consumption may affect cell growth in media formulated with high zinc concentrations, and in contrast, normal levels of these amino acids are associated with better tolerance to high zinc concentration. Our sequential analysis method may be similarly employed for high throughput medium design and optimization experiments to identify interactions among a large number of cell culture medium components.

A physiologically-based pharmacokinetic model for tuberculosis drug disposition at extrapulmonary sites.

Tuberculosis (TB) is a leading cause of mortality attributed to an infectious agent. TB primarily targets the lungs, but in about 16% cases can affect other organs as well, giving rise to extrapulmonary TB (EPTB). However, an optimal regimen for EPTB treatment is not defined. Although the recommended treatment for most forms of EPTB is the same as pulmonary TB, the pharmacokinetics of EPTB therapy are not as well studied. To address this gap, we formulate a whole-body physiologically-based pharmacokinetic (PBPK) model for EPTB that for the first time includes the ability to simulate drug concentrations in the pleura and lymph node, the most commonly affected sites of EPTB. Using this model, we estimate the time-dependent concentrations, at potential EPTB infection sites, of the following four first-line anti-TB drugs: rifampicin, ethambutol, isoniazid, and pyrazinamide. We use reported plasma concentration kinetics data to estimate model parameters for each drug and validate our model using reported concentration data not used for model formulation or parameter estimation. Model predictions match the validation data, and reported pharmacokinetic parameters (maximum plasma concentration, time to reach maximum concentration) for the drugs. The model also predicts ethambutol, isoniazid, and pyrazinamide concentrations in the pleura that match reported experimental values from an independent study. For each drug, the predicted drug concentrations at EPTB sites are compared with their critical concentration. Simulations suggest that although rifampicin and isoniazid concentrations are greater than critical concentration values at most EPTB sites, the concentrations of ethambutol and pyrazinamide are lower than their critical concentrations at most EPTB sites.

SUMOylation of Dorsal attenuates Toll/NF-κB signaling.

In Drosophila, Toll/NF-κB signaling plays key roles in both animal development and in host defense. The activation, intensity, and kinetics of Toll signaling are regulated by posttranslational modifications such as phosphorylation, SUMOylation, or ubiquitination that target multiple proteins in the Toll/NF-κB cascade. Here, we have generated a CRISPR-Cas9 edited Dorsal (DL) variant that is SUMO conjugation resistant. Intriguingly, embryos laid by dlSCR mothers overcome dl haploinsufficiency and complete the developmental program. This ability appears to be a result of higher transcriptional activation by DLSCR. In contrast, SUMOylation dampens DL transcriptional activation, ultimately conferring robustness to the dorso-ventral program. In the larval immune response, dlSCR animals show an increase in crystal cell numbers, stronger activation of humoral defense genes, and high cactus levels. A mathematical model that evaluates the contribution of the small fraction of SUMOylated DL (1-5%) suggests that it acts to block transcriptional activation, which is driven primarily by DL that is not SUMO conjugated. Our findings define SUMO conjugation as an important regulator of the Toll signaling cascade, in both development and host defense. Our results broadly suggest that SUMO attenuates DL at the level of transcriptional activation. Furthermore, we hypothesize that SUMO conjugation of DL may be part of a Ubc9-dependent mechanism that restrains Toll/NF-κB signaling.

Temporal analysis of melanogenesis identifies fatty acid metabolism as key skin pigment regulator.

Therapeutic methods to modulate skin pigmentation has important implications for skin cancer prevention and for treating cutaneous hyperpigmentary conditions. Towards defining new potential targets, we followed temporal dynamics of melanogenesis using a cell-autonomous pigmentation model. Our study elucidates 3 dominant phases of synchronized metabolic and transcriptional reprogramming. The melanogenic trigger is associated with high MITF levels along with rapid uptake of glucose. The transition to pigmented state is accompanied by increased glucose channelisation to anabolic pathways that support melanosome biogenesis. SREBF1-mediated up-regulation of fatty acid synthesis results in a transient accumulation of lipid droplets and enhancement of fatty acids oxidation through mitochondrial respiration. While this heightened bioenergetic activity is important to sustain melanogenesis, it impairs mitochondria lately, shifting the metabolism towards glycolysis. This recovery phase is accompanied by activation of the NRF2 detoxication pathway. Finally, we show that inhibitors of lipid metabolism can resolve hyperpigmentary conditions in a guinea pig UV-tanning model. Our study reveals rewiring of the metabolic circuit during melanogenesis, and fatty acid metabolism as a potential therapeutic target in a variety of cutaneous diseases manifesting hyperpigmentary phenotype.

Kinetic model of GPCR-G protein interactions reveals allokairic modulation of signaling.

Established models of ternary complex formation between hormone, G protein coupled receptor (GPCR), and G protein assume that all interactions occur under equilibrium conditions. However, recent studies have established that the lifetimes of these interactions are comparable to the duration of hormone activated GPCR signaling. To simulate interactions during such non-equilibrium conditions, we propose a kinetic model wherein the receptor undergoes rate-limiting transitions between two hormone-bound active states. Simulations, using experimentally measured parameters, demonstrate transient states in ternary complex formation, and delineate the phenomenon of GPCR priming, wherein non-cognate G proteins substantially enhance cognate G protein signaling. Our model reveals that kinetic barriers of slow receptor interconversion can be overcome through allokairic modulation, a regulatory mechanism of ternary complex formation and downstream signaling.

Mathematical model of the multi-amino acid multi-transporter system predicts uptake flux in CHO cells.

Supply and uptake of amino acids is of great importance to mammalian cell culture processes. Mammalian cells such as Chinese hamster ovary (CHO) cells express several amino acid (AA) transporters including uniporters and exchangers. Each transporter transports multiple AAs, making prediction of the effect of changed medium composition or transporter levels on individual AA transport rate challenging. A general kinetic model for such combinatorial amino acid transport, and a simplified analytical expression for the uptake rate as a function of amino acid concentrations and transporter levels is presented. From this general model, a CHO cell-specific AA transport model, to our knowledge the first such network model for any cell type, is constructed. The model is validated by its prediction of reported uptake flux and dependencies from experiments that were not used in model construction or parameter estimation. The model defines theoretical conditions for synergistic/repressive effect on the uptake rates of other AAs upon external addition of one AA. The ability of the CHO-specific model to predict amino acid interdependencies experimentally observed in other mammalian cell types suggests its robustness. This model will help formulate testable hypotheses of the effect of process changes on AA initial uptake, and serve as the AA transport component of kinetic models for cellular metabolism.

Boolean model for melanogenesis.

Melanogenesis is a highly regulated process through which the pigment melanin is produced in skin cells. Irregularities in the molecular events that govern the process of skin pigmentation can cause disorders like vitiligo. In order to understand the biology of disease progression, it is important to have an in depth understanding of intracellular events. Mathematical models provide an integrated view of intracellular signalling. There are very few models to date that incorporate intracellular processes relevant to melanogenesis and only one to our knowledge that simulates the dynamics of response to varying levels of input. Here, we report the formulation of the largest Boolean model (265 nodes) for melanogenesis to date. The model was built on the basis of a detailed interaction network graph published by Raghunath et al. Through additional manual curation of the reported interactions, we converted the graph into a set of Boolean rules, following the procedure of the first Boolean model (62 nodes) for melanogenesis published by Lee et al. Simulations show that the predicted response to varying UV levels for most of the nodes is similar to the predictions of the existing model. The greater complexity allows investigation of the sensitivity of melanin to additional nodes. We carried out perturbation analysis of the network through node deletion and constitutive activation to identify sensitivity of outcomes, and compared the nodes identified as sensitive to previous reports.

Corrigendum: Web-gLV: A Web Based Platform for Lotka-Volterra Based Modeling and Simulation of Microbial Populations.

[This corrects the article DOI: 10.3389/fmicb.2019.00288.].

Web-gLV: A Web Based Platform for Lotka-Volterra Based Modeling and Simulation of Microbial Populations.

The affordability of high throughput DNA sequencing has allowed us to explore the dynamics of microbial populations in various ecosystems. Mathematical modeling and simulation of such microbiome time series data can help in getting better understanding of bacterial communities. In this paper, we present Web-gLV-a GUI based interactive platform for generalized Lotka-Volterra (gLV) based modeling and simulation of microbial populations. The tool can be used to generate the mathematical models with automatic estimation of parameters and use them to predict future trajectories using numerical simulations. We also demonstrate the utility of our tool on few publicly available datasets. The case studies demonstrate the ease with which the current tool can be used by biologists to model bacterial populations and simulate their dynamics to get biological insights. We expect Web-gLV to be a valuable contribution in the field of ecological modeling and metagenomic systems biology.

Clathrin-Mediated Endocytosis Regulates a Balance between Opposing Signals to Maintain the Pluripotent State of Embryonic Stem Cells.

Endocytosis is implicated in the maintenance of embryonic stem cell (ESC) pluripotency, although its exact role and the identity of molecular players remain poorly understood. Here, we show that the clathrin heavy chain (CLTC), involved in clathrin-mediated endocytosis (CME), is vital for maintaining mouse ESC (mESC) pluripotency. Knockdown of Cltc resulted in a loss of pluripotency accompanied by reduced E-cadherin (E-CAD) levels and increased levels of transforming growth factor ß (TGF-ß) and extracellular signal-regulated kinase (ERK) signaling. We demonstrate that both E-CAD and TGF-ß receptor type 1 (TGF-ßR1) are internalized through CME in mESCs. While E-CAD is recycled, TGF-ßR1 is targeted for lysosomal degradation thus maintaining inverse levels of these molecules. Finally, we show that E-CAD interacts with ERK, and that the decreased pluripotency upon CME loss can be rescued by inhibiting TGF-ßR, MEK, and GSK3ß, or overexpressing E-CAD. Our results demonstrate that CME is critical for balancing signaling outputs to regulate ESC pluripotency, and possibly cell fate choices in early development.

Method for identification of sensitive nodes in Boolean models of biological networks.

Biological systems are often represented as Boolean networks and analysed to identify sensitive nodes which on perturbation disproportionately change a predefined output. There exist different kinds of perturbation methods: perturbation of function, perturbation of state and perturbation in update scheme. Nodes may have defects in interpretation of the inputs from other nodes and calculation of the node output. To simulate these defects and systematically assess their effect on the system output, two new function perturbations, referred to as 'not of function' and 'function of not', are introduced. In the former, the inputs are assumed to be correctly interpreted but the output of the update rule is perturbed; and in the latter, each input is perturbed but the correct update rule is applied. These and previously used perturbation methods were applied to two existing Boolean models, namely the human melanogenesis signalling network and the fly segment polarity network. Through mathematical simulations, it was found that these methods successfully identified nodes earlier found to be sensitive using other methods, and were also able to identify sensitive nodes which were previously unreported.

Mapping architectural and transcriptional alterations in non-lesional and lesional epidermis in vitiligo.

In vitiligo, chronic loss of melanocytes and consequent absence of melanin from the epidermis presents a challenge for long-term tissue maintenance. The stable vitiligo patches are known to attain an irreversible depigmented state. However, the molecular and cellular processes resulting in this remodeled tissue homeostasis is unclear. To investigate the complex interplay of inductive signals and cell intrinsic factors that support the new acquired state, we compared the matched lesional and non-lesional epidermis obtained from stable non-segmental vitiligo subjects. Hierarchical clustering of genome-wide expression of transcripts surprisingly segregated lesional and non-lesional samples in two distinct clades, despite the apparent heterogeneity in the lesions of different vitiligo subjects. Pathway enrichment showed the expected downregulation of melanogenic pathway and a significant downregulation of cornification and keratinocyte differentiation processes. These perturbations could indeed be recapitulated in the lesional epidermal tissue, including blunting of rete-ridges, thickening of stratum corneum and increase in the size of corneocytes. In addition, we identify marked increase in the putrescine levels due to the elevated expression of spermine/spermidine acetyl transferase. Our study provides insights into the intrinsic self-renewing ability of damaged lesional tissue to restore epidermal functionality in vitiligo.

Systematic comparison of the response properties of protein and RNA mediated gene regulatory motifs.

We present a framework enabling the dissection of the effects of motif structure (feedback or feedforward), the nature of the controller (RNA or protein), and the regulation mode (transcriptional, post-transcriptional or translational) on the response to a step change in the input. We have used a common model framework for gene expression where both motif structures have an activating input and repressing regulator, with the same set of parameters, to enable a comparison of the responses. We studied the global sensitivity of the system properties, such as steady-state gain, overshoot, peak time, and peak duration, to parameters. We find that, in all motifs, overshoot correlated negatively whereas peak duration varied concavely with peak time. Differences in the other system properties were found to be mainly dependent on the nature of the controller rather than the motif structure. Protein mediated motifs showed a higher degree of adaptation i.e. a tendency to return to baseline levels; in particular, feedforward motifs exhibited perfect adaptation. RNA mediated motifs had a mild regulatory effect; they also exhibited a lower peaking tendency and mean overshoot. Protein mediated feedforward motifs showed higher overshoot and lower peak time compared to the corresponding feedback motifs.

CHO Cells adapted to inorganic phosphate limitation show higher growth and higher pyruvate carboxylase flux in phosphate replete conditions.

Inorganic phosphate (Pi ) is an essential ion involved in diverse cellular processes including metabolism. Changes in cellular metabolism upon long term adaptation to Pi limitation have been reported in E. coli. Given the essential role of Pi , adaptation to Pi limitation may also result in metabolic changes in animal cells. In this study, we have adapted CHO cells producing recombinant IgG to limiting Pi conditions for 75 days. Not surprisingly, adapted cells showed better survival under Pi limitation. Here, we report the finding that such cells also showed better growth characteristics compared to control in batch culture replete with Pi (higher peak density and integral viable cell density), accompanied by a lower specific oxygen uptake rate and cytochrome oxidase activity towards the end of exponential phase. Surprisingly, the adapted cells grew to a lower peak density under glucose limitation. This suggests long term Pi limitation may lead to selection for an altered metabolism with higher dependence on glucose availability for biomass assimilation compared to control. Steady state U-13 C glucose labeling experiments suggest that adapted cells have a higher pyruvate carboxylase flux. Consistent with this observation, supplementation with aspartate abolished the peak density difference whereas supplementation with serine did not abolish the difference. This supports the hypothesis that cell growth in the adapted culture might be higher due to a higher pyruvate carboxylase flux. Decreased fitness under carbon limitation and mutations in the sucABCD operon has been previously reported in E. coli upon long term adaptation to Pi limitation, suggestive of a similarity in cellular response among such diverse species. © 2017 American Institute of Chemical Engineers Biotechnol. Prog., 33:749-758, 2017.

Identification of Unintuitive Features of Sumoylation through Mathematical Modeling.

Sumoylation is a multistep, multienzymatic post-translational modification in which a small ubiquitin-like modifier protein (SUMO) is attached to the target. We present the first mathematical model for sumoylation including enzyme mechanism details such as autosumoylation of E2 and multifunctional nature of SENP. Simulations and analysis reveal three nonobvious properties for the long term response, modeled as an open system: (i) the steady state sumoylation level is robust to variation in several enzyme properties; (ii) even when autosumoylation of E2 results in equal or higher activity, the target sumoylation levels are lower; and (iii) there is an optimal SENP concentration at which steady state target sumoylation level is maximum. These results are qualitatively different for a short term response modeled as a closed system, where e.g. sumoylation always decreases with increasing SENP levels. Simulations with multiple targets suggest that the available SUMO is limiting, indicating a possible explanation for the experimentally observed low fractional sumoylation. We predict qualitative differences in system responses at short post-translational and longer transcriptional time scales. We thus use this mechanism-based model to explain system properties and generate testable hypotheses for existence and mechanism of unexpected responses.

Ranking of Reactions Based on Sensitivity of Protein Noise Depends on the Choice of Noise Measure.

Gene expression is a stochastic process. Identification of the step maximally affecting noise in the protein level is an important aspect of investigation of gene product distribution. There are numerous experimental and theoretical studies that seek to identify this important step. However, these studies have used two different measures of noise, viz. coefficient of variation and Fano factor, and have compared different processes leading to contradictory observations regarding the important step. In this study, we performed systematic global and local sensitivity analysis on two models of gene expression to investigate relative contribution of reaction rate parameters to steady state noise in the protein level using both the measures of noise. We analytically and computationally showed that the ranking of parameters based on the sensitivity of the noise to variation in a given parameter is a strong function of the choice of the noise measure. If the Fano factor is used as the noise measure, translation is the important step whereas for coefficient of variation, transcription is the important step. We derived an analytical expression for local sensitivity and used it to explain the distinct contributions of each reaction parameter to the two measures of noise. We extended the analysis to a generic linear catalysis reaction system and observed that the reaction network topology was an important factor influencing the local sensitivity of the two measures of noise. Our study suggested that, for the analysis of contributions of reactions to the noise, consideration of both the measures of noise is important.

Mathematical modeling of combinatorial regulation suggests that apparent positive regulation of targets by miRNA could be an artifact resulting from competition for mRNA.

MicroRNAs bind to and regulate the abundance and activity of target messenger RNA through sequestration, enhanced degradation, and suppression of translation. Although miRNA have a predominantly negative effect on the target protein concentration, several reports have demonstrated a positive effect of miRNA, i.e., increase in target protein concentration on miRNA overexpression and decrease in target concentration on miRNA repression. miRNA-target pair-specific effects such as protection of mRNA degradation owing to miRNA binding can explain some of these effects. However, considering such pairs in isolation might be an oversimplification of the RNA biology, as it is known that one miRNA interacts with several targets, and conversely target mRNA are subject to regulation by several miRNAs. We formulate a mathematical model of this combinatorial regulation of targets by multiple miRNA. Through mathematical analysis and numerical simulations of this model, we show that miRNA that individually have a negative effect on their targets may exhibit an apparently positive net effect when the concentration of one miRNA is experimentally perturbed by repression/overexpression in such a multi-miRNA multitarget situation. We show that this apparent unexpected effect is due to competition and will not be observed when miRNA interact noncompetitively with the target mRNA. This result suggests that some of the observed unusual positive effects of miRNA may be due to the combinatorial complexity of the system rather than due to any inherently unusual positive effect of the miRNA on its target.

Non-coding RNA interact to regulate neuronal development and function.

The human brain is one of the most complex biological systems, and the cognitive abilities have greatly expanded compared to invertebrates without much expansion in the number of protein coding genes. This suggests that gene regulation plays a very important role in the development and function of nervous system, by acting at multiple levels such as transcription and translation. In this article we discuss the regulatory roles of three classes of non-protein coding RNAs (ncRNAs)-microRNAs (miRNAs), piwi-interacting RNA (piRNAs) and long-non-coding RNA (lncRNA), in the process of neurogenesis and nervous function including control of synaptic plasticity and potential roles in neurodegenerative diseases. miRNAs are involved in diverse processes including neurogenesis where they channelize the cellular physiology toward neuronal differentiation. miRNAs can also indirectly influence neurogenesis by regulating the proliferation and self renewal of neural stem cells and are dysregulated in several neurodegenerative diseases. miRNAs are also known to regulate synaptic plasticity and are usually found to be co-expressed with their targets. The dynamics of gene regulation is thus dependent on the local architecture of the gene regulatory network (GRN) around the miRNA and its targets. piRNAs had been classically known to regulate transposons in the germ cells. However, piRNAs have been, recently, found to be expressed in the brain and possibly function by imparting epigenetic changes by DNA methylation. piRNAs are known to be maternally inherited and we assume that they may play a role in early development. We also explore the possible function of piRNAs in regulating the expansion of transposons in the brain. Brain is known to express several lncRNA but functional roles in brain development are attributed to a few lncRNA while functions of most of the them remain unknown. We review the roles of some known lncRNA and explore the other possible functions of lncRNAs including their interaction with miRNAs.

IFN-γ signaling maintains skin pigmentation homeostasis through regulation of melanosome maturation.

Cellular homeostasis is an outcome of complex interacting processes with nonlinear feedbacks that can span distinct spatial and temporal dimensions. Skin tanning is one such dynamic response that maintains genome integrity of epidermal cells. Although pathways underlying hyperpigmentation cascade are recognized, negative feedback regulatory loops that can dampen the activated melanogenesis process are not completely understood. In this study, we delineate a regulatory role of IFN-γ in skin pigmentation biology. We show that IFN-γ signaling impedes maturation of the key organelle melanosome by concerted regulation of several pigmentation genes. Withdrawal of IFN-γ signal spontaneously restores normal cellular programming. This effect in melanocytes is mediated by IFN regulatory factor-1 and is not dependent on the central regulator microphthalmia-associated transcription factor. Chronic IFN-γ signaling shows a clear hypopigmentation phenotype in both mouse and human skin. Interestingly, IFN-γ KO mice display a delayed recovery response to restore basal state of epidermal pigmentation after UV-induced tanning. Together, our studies delineate a new spatiotemporal role of the IFN-γ signaling network in skin pigmentation homeostasis, which could have implications in various cutaneous depigmentary and malignant disorders.

Comparison of methods for identifying periodically varying genes.

Several methods have been reported for identifying periodically varying genes from gene expression datasets. We compare the performance of five existing methods and a combination of G-statistic and autocovariance (called GVAR) using simulated sine-function-based and cell-cycle-based datasets. Based on this analysis we recommend appropriate methods for different experimental situations (length of the time series, sampling interval and noise level). No single method performs the best under all tested conditions. None of the evaluated methods perform well at high noise levels for short time series data. At lower noise levels, GVAR performed the best.