David J. Klinke

Extent of beta cell destruction is important but insufficient to predict the onset of type 1 diabetes mellitus.

Background Type 1 diabetes mellitus is characterized by an inability to produce insulin endogenously. Based on a series of histopathology studies of patients with recent onset of the disease, it is commonly stated that the onset of clinical symptoms corresponds to an 80-95% reduction in beta cell mass. Motivated by the clinical importance of the degree of beta cell destruction at onset, a meta-analysis was used to determine the validity of this common wisdom. Methods and Findings The histopathology results identifying insulin containing islets in patients younger than 20 years of age were extracted from three different studies. The results for 105 patients were stratified by duration of diabetic symptoms and age at onset. Linear regression and a non-parametric bootstrap approach were used to determine the dependence of residual beta cell mass to age at onset. The percentage reduction in beta cell mass was highly correlated (p<0.001) with the age of onset with the greatest reduction in beta cell mass in the youngest patients. As this trend had not been previously observed, an alternative physiology-based model is proposed that captures this age-dependence. Conclusions The severity in beta cell reduction at onset decreased with age where, on average, a 40% reduction in beta cell mass was sufficient to precipitate clinical symptoms at 20 years of age. The observed trend was consistent with a physiology-based model where the threshold for onset is based upon a dynamic balance between insulin-production capacity, which is proportional to beta cell mass, and insulin demand, which is proportional to body weight.

An empirical Bayesian approach for model-based inference of cellular signaling networks

BackgroundA common challenge in systems biology is to infer mechanistic descriptions of biological process given limited observations of a biological system. Mathematical models are frequently used to represent a belief about the causal relationships among proteins within a signaling network. Bayesian methods provide an attractive framework for inferring the validity of those beliefs in the context of the available data. However, efficient sampling of high-dimensional parameter space and appropriate convergence criteria provide barriers for implementing an empirical Bayesian approach. The objective of this study was to apply an Adaptive Markov chain Monte Carlo technique to a typical study of cellular signaling pathways.ResultsAs an illustrative example, a kinetic model for the early signaling events associated with the epidermal growth factor (EGF) signaling network was calibrated against dynamic measurements observed in primary rat hepatocytes. A convergence criterion, based upon the Gelman-Rubin potential scale reduction factor, was applied to the model predictions. The posterior distributions of the parameters exhibited complicated structure, including significant covariance between specific parameters and a broad range of variance among the parameters. The model predictions, in contrast, were narrowly distributed and were used to identify areas of agreement among a collection of experimental studies.ConclusionIn summary, an empirical Bayesian approach was developed for inferring the confidence that one can place in a particular model that describes signal transduction mechanisms and for inferring inconsistencies in experimental measurements.

Exosomes: improved methods to characterize their morphology, RNA content, and surface protein biomarkers.

As a type of secreted membrane vesicle, exosomes are an emerging mode of cell-to-cell communication. Yet as exosome samples are commonly contaminated with other extracellular vesicles, the biological roles of exosomes in regulating immunity and promoting oncogenesis remain controversial. Wondering whether existing methods could distort our view of exosome biology, we compared two direct methods for imaging extracellular vesicles and quantified the impact of different production and storage conditions on the quality of exosome samples. Scanning electron microscopy (SEM) was compared to transmission electron microscopy (TEM) as alternatives to examine the morphology of exosomes. Using SEM, we were able to distinguish exosomes from other contaminating extracellular vesicles based on the size distribution. More importantly, freezing of samples prior to SEM imaging made it more difficult to distinguish exosomes from extracellular vesicles secreted during cell death. In addition to morphology, the quality of RNA contained within the exosomes was characterized under different storage conditions, where freezing of samples also degraded RNA. Finally, we developed a new flow cytometry approach to assay transmembrane proteins on exosomes. While high-copy-number proteins could be readily detected, detecting low-copy-number proteins was improved using a lipophilic tracer that clustered exosomes. To illustrate this, we observed that exosomes derived from SKBR3 cells, a cell model for human HER2+ breast cancer, contained both HER1 and HER2 but at different levels of abundance. Collectively, these new methods will help to ensure a consistent framework to identify specific roles that exosomes play in regulating cell-to-cell communication.

Integrating Epidemiological Data into a Mechanistic Model of Type 2 Diabetes: Validating the Prevalence of Virtual Patients

Mathematical models are playing an increasing role in understanding the complexity of multifactorial diseases like type 2 diabetes. The objective of this study was to validate a population of virtual patients against a real population of patients with type 2 diabetes. A population of virtual patients was created that incorporates different underlying pathogenic lesions consistent with a type 2 diabetic phenotype. These virtual patients were created within the Metabolism PhysioLab platform, a non-linear coupled differential algebraic model that incorporates the salient causal mechanisms underlying glucose homeostasis and substrate metabolism. The weights of each individual virtual patient were determined to reproduce the diversity in a real type 2 diabetic population obtained from the NHANES III study. As a validation test, this virtual population reproduced a series of clinical studies that identify less invasive biomarkers for insulin sensitivity. This approach demonstrates how computational bridges can be constructed between statistical approaches common in epidemiology and deterministic approaches common in biomedical engineering.

Induction of Wnt-inducible signaling protein-1 correlates with invasive breast cancer oncogenesis and reduced type 1 cell-mediated cytotoxic immunity: a retrospective study.

Innate and type 1 cell-mediated cytotoxic immunity function as important extracellular control mechanisms that maintain cellular homeostasis. Interleukin-12 (IL12) is an important cytokine that links innate immunity with type 1 cell-mediated cytotoxic immunity. We recently observed in vitro that tumor-derived Wnt-inducible signaling protein-1 (WISP1) exerts paracrine action to suppress IL12 signaling. The objective of this retrospective study was three fold: 1) to determine whether a gene signature associated with type 1 cell-mediated cytotoxic immunity was correlated with overall survival, 2) to determine whether WISP1 expression is increased in invasive breast cancer, and 3) to determine whether a gene signature consistent with inhibition of IL12 signaling correlates with WISP1 expression. Clinical information and mRNA expression for genes associated with anti-tumor immunity were obtained from the invasive breast cancer arm of the Cancer Genome Atlas study. Patient cohorts were identified using hierarchical clustering. The immune signatures associated with the patient cohorts were interpreted using model-based inference of immune polarization. Reverse phase protein array, tissue microarray, and quantitative flow cytometry in breast cancer cell lines were used to validate observed differences in gene expression. We found that type 1 cell-mediated cytotoxic immunity was correlated with increased survival in patients with invasive breast cancer, especially in patients with invasive triple negative breast cancer. Oncogenic transformation in invasive breast cancer was associated with an increase in WISP1. The gene expression signature in invasive breast cancer was consistent with WISP1 as a paracrine inhibitor of type 1 cell-mediated immunity through inhibiting IL12 signaling and promoting type 2 immunity. Moreover, model-based inference helped identify appropriate immune signatures that can be used as design constraints in genetically engineering better pre-clinical models of breast cancer.Innate and type 1 cell-mediated cytotoxic immunity function as important extracellular control mechanisms that maintain cellular homeostasis. Interleukin-12 (IL12) is an important cytokine that links innate immunity with type 1 cell-mediated cytotoxic immunity. We recently observed in vitro that tumor-derived Wnt-inducible signaling protein-1 (WISP1) exerts paracrine action to suppress IL12 signaling. The objective of this retrospective study was three fold: 1) to determine whether a gene signature associated with type 1 cell-mediated cytotoxic immunity was correlated with overall survival, 2) to determine whether WISP1 expression is increased in invasive breast cancer, and 3) to determine whether a gene signature consistent with inhibition of IL12 signaling correlates with WISP1 expression. Clinical information and mRNA expression for genes associated with anti-tumor immunity were obtained from the invasive breast cancer arm of the Cancer Genome Atlas study. Patient cohorts were identified using hierarchical clustering. The immune signatures associated with the patient cohorts were interpreted using model-based inference of immune polarization. Reverse phase protein array, tissue microarray, and quantitative flow cytometry in breast cancer cell lines were used to validate observed differences in gene expression. We found that type 1 cell-mediated cytotoxic immunity was correlated with increased survival in patients with invasive breast cancer, especially in patients with invasive triple negative breast cancer. Oncogenic transformation in invasive breast cancer was associated with an increase in WISP1. The gene expression signature in invasive breast cancer was consistent with WISP1 as a paracrine inhibitor of type 1 cell-mediated immunity through inhibiting IL12 signaling and promoting type 2 immunity. Moreover, model-based inference helped identify appropriate immune signatures that can be used as design constraints in genetically engineering better pre-clinical models of breast cancer.

Inferring relevant control mechanisms for interleukin-12 signaling in naïve CD4+ T cells.

Interleukin‐12 (IL‐12) is a key cytokine involved in shaping the cell‐mediated immunity to intracellular pathogens. IL‐12 initiates a cellular response through the IL‐12 signaling pathway, a member of the Janus kinase/signal transducer and activator of transcription (JAK/STAT) family of signaling networks. The JAK/STAT pathway includes several regulatory elements; however, the dynamics of these mechanisms are not fully understood. Therefore, the objective of this study was to infer the relative importance of regulatory mechanisms that modulate the activation of STAT4 in naïve CD4+ T cells. Dynamic changes in protein expression and activity were measured using flow cytometry and these data were used to calibrate a mathematical model of IL‐12 signaling. An empirical Bayesian approach was used to infer the relative strengths of the different regulatory mechanisms in the system. The model predicted that IL‐12 receptor expression is regulated by a dynamic, autonomous program that was independent of STAT4 activation. In summary, a mathematical model of the canonical IL‐12 signaling pathway used in conjunction with a Bayesian framework provided high‐confidence predictions of the system‐specific control mechanisms from the available experimental observations.

Inferring predominant pathways in cellular models of breast cancer using limited sample proteomic profiling

BackgroundMolecularly targeted drugs inhibit aberrant signaling within oncogenic pathways. Identifying the predominant pathways at work within a tumor is a key step towards tailoring therapies to the patient. Clinical samples pose significant challenges for proteomic profiling, an attractive approach for identifying predominant pathways. The objective of this study was to determine if information obtained from a limited sample (i.e., a single gel replicate) can provide insight into the predominant pathways in two well-characterized breast cancer models.MethodsA comparative proteomic analysis of total cell lysates was obtained from two cellular models of breast cancer, BT474 (HER2+/ER+) and SKBR3 (HER2+/ER-), using two-dimensional electrophoresis and MALDI-TOF mass spectrometry. Protein interaction networks and canonical pathways were extracted from the Ingenuity Pathway Knowledgebase (IPK) based on association with the observed pattern of differentially expressed proteins.ResultsOf the 304 spots that were picked, 167 protein spots were identified. A threshold of 1.5-fold was used to select 62 proteins used in the analysis. IPK analysis suggested that metabolic pathways were highly associated with protein expression in SKBR3 cells while cell motility pathways were highly associated with BT474 cells. Inferred protein networks were confirmed by observing an up-regulation of IGF-1R and profilin in BT474 and up-regulation of Ras and enolase in SKBR3 using western blot.ConclusionWhen interpreted in the context of prior information, our results suggest that the overall patterns of differential protein expression obtained from limited samples can still aid in clinical decision making by providing an estimate of the predominant pathways that underpin cellular phenotype.

Modulating Temporal Control of NF-κB Activation: Implications for Therapeutic and Assay Selection

The activation of transcription factor NF-kappaB (nuclear factor-kappaB) plays a central role in the induction of many inflammatory response genes. This process is characterized by either oscillations or stable induction of NF-kappaB nuclear binding. Changes in dynamics of binding result in the expression of distinct subsets of genes leading to different physiological outcomes. We examined NF-kappaB DNA binding activity in lipopolysaccharide (LPS)-stimulated IC-21 cells by electromobility shift assay and nonradioactive transcription factor assay and interpreted the results using a kinetic model of NF-kappaB activation. Both assays detected damped oscillatory behavior of NF-kappaB with differences in sensitivity and reproducibility. 3,4-Dichloropropionaniline (DCPA) was used to modulate the oscillatory behavior of NF-kappaB after LPS stimulation. DCPA is known to inhibit the production of two NF-kappaB-inducible cytokines, IL-6 and tumor necrosis factor alpha, by reducing but not completely abrogating NF-kappaB-induced transcription. DCPA treatment resulted in a potentiation of early LPS-induced NF-kappaB activation. The nonradioactive transcription factor assay, which has a higher signal/noise ratio than the electromobility shift assay, combined with in silico modeling, produced results that revealed changes in NF-kappaB dynamics which, to the best of our knowledge, have never been previously reported. These results highlight the importance of cell type and stimulus specificity in transcription factor activity assessment. In addition, assay selection has important implications for network inference and drug discovery.

Inferring alterations in cell‐to‐cell communication in HER2+ breast cancer using secretome profiling of three cell models

Challenges in demonstrating durable clinical responses to molecular‐targeted therapies have sparked a re‐emergence in viewing cancer as an evolutionary process. In somatic evolution, cellular variants are introduced through a random process of somatic mutation and are selected for improved fitness through a competition for survival. In contrast to Darwinian evolution, cellular variants that are retained may directly alter the fitness competition. If cell‐to‐cell communication is important for selection, the biochemical cues secreted by malignant cells that emerge should be altered to bias this fitness competition. To test this hypothesis, we compared the proteins secreted in vitro by two human HER2+ breast cancer cell lines (BT474 and SKBR3) relative to a normal human mammary epithelial cell line (184A1) using a proteomics workflow that leveraged two‐dimensional gel electrophoresis (2DE) and MALDI‐TOF mass spectrometry. Supported by the 2DE secretome maps and identified proteins, the two breast cancer cell lines exhibited secretome profiles that were similar to each other and, yet, were distinct from the 184A1 secretome. Using protein–protein interaction and pathway inference tools for functional annotation, the results suggest that all three cell lines secrete exosomes, as confirmed by scanning electron microscopy. Interestingly, the HER2+ breast cancer cell line exosomes are enriched in proteins involved in antigen‐processing and presentation and glycolytic metabolism. These pathways are associated with two of the emerging hallmarks of cancer: evasion of tumor immunosurveillance and deregulating cellular energetics. Biotechnol. Bioeng. 2014;111: 1853–1863.

Scalable analysis of flow cytometry data using R/Bioconductor

Flow cytometry is one of the fundamental research tools available to the life scientist. The ability to observe multidimensional changes in protein expression and activity at single‐cell resolution for a large number of cells provides a unique perspective on the behavior of cell populations. However, the analysis of complex multidimensional data is one of the obstacles for wider use of polychromatic flow cytometry. Recent enhancements to an open‐source platform—R/Bioconductor—enable the graphical and data analysis of flow cytometry data. Prior examples have focused on high‐throughput applications. To facilitate wider use of this platform for flow cytometry, the analysis of a dataset, obtained following isolation of CD4+CD62L+ T cells from Balb/c splenocytes using magnetic microbeads, is presented as a form of tutorial. A common workflow for analyzing flow cytometry data was presented using R/Bioconductor. In addition, density function estimation and principal component analysis are provided as examples of more complex analyses. The compendium presented here is intended to help illuminate a path for inquisitive readers to explore their own data using R/Bioconductor (available as Supporting Information).