One of the main challenges in modern biology is understanding how and when genes are turned on. As our knowledge of transcription regulation has matured, bioinformatic analyses have allowed increasingly quantitative predictions of gene expression. The ultimate goal of such analyses is to predict gene expression from concentrations of proteins in the cell, based on protein- DNA networks. Yet this is a highly ambitious task in Eukaryotes, since their gene expression is determined by chromatin conformation, epigenetic factors, promoter and enhancer states, rates of transcription initiation, elongation, transcript processing and termination, and mRNA export and stability. This thesis is focused on the co-occurrence of transcriptional elongation and pre-mRNA splicing, the process in which introns are removed from the pre-mRNA transcript. Splicing is an important regulatory step because aberrant splicing leads to either reduced or non-functional protein expression, and alternative splicing expands the repertoire of functional proteins encoded by the genome. The co-transcriptional nature of splicing implies that the kinetics of elongation in relation to splicing are important for the outcome of splicing decisions. Co-transcriptional splicing (CTS) has been extensively studied, but quantitative models of transcription networks that predict gene expression timecourses have yet to incorporate CTS considerations. Here I constructed kinetic models of CTS. Initially, I built a model of constitutive CTS and developed methods to fit nascent RNA-seq data to the model. Fitting this model to published datasets indicated that only a subset of genes can be expected to process all of their introns co- transcriptionally. Detailed data-mining of high-throughput datasets and genomes revealed patterns of compensatory signatures in sequence, chromatin and polymerase data, suggesting an evolutionary selection towards splicing co- transcriptionally. Next I expanded the model to include alternative splicing reactions. Despite the exponential combinatorial complexity, all possible isoforms resulting from up to nine introns can be simulated. A further expansion of the model considers separate reactions at the 5' and 3' ends of introns, which allows for simulation of phenomena such as exon definition and polymerase-mediated recruitment. Together, these novel tools can be used to test quantitative predictions of genome-wide splicing outcomes, or be incorporated into larger gene expression models

The spliceosome catalyzes the removal of introns from pre-messenger RNA (mRNA) and subsequent pairing of exons with remarkable fidelity. Some exons are known to be skipped or included in the mature mRNA in a cell type- or context-dependent manner (cassette exons), thereby contributing to the diversification of the human proteome. Interestingly, splicing is initiated (and sometimes completed) co-transcriptionally. Here, we develop a kinetic mathematical modeling framework to investigate alternative co-transcriptional splicing (CTS) and, specifically, the control of cassette exons' inclusion. We show that when splicing is co-transcriptional, default splice patterns of exon inclusion are more likely than when splicing is post-transcriptional, and that certain exons are more likely to be regulatable (i.e. cassette exons) than others, based on the exon-intron structure context. For such regulatable exons, transcriptional elongation rates may affect splicing outcomes. Within the CTS paradigm, we examine previously described hypotheses of co-operativity between splice sites of short introns (i.e. 'intron definition') or across short exons (i.e. 'exon definition'), and find that models encoding these faithfully recapitulate observations in the fly and human genomes, respectively.

Understanding the functions of multi-cellular organs in terms of the molecular networks within each cell is an important step in the quest to predict phenotype from genotype. B-lymphocyte population dynamics, which are predictive of immune response and vaccine effectiveness, are determined by individual cells undergoing division or death seemingly stochastically. Based on tracking single-cell time-lapse trajectories of hundreds of B cells, single-cell transcriptome, and immunofluorescence analyses, we constructed an agent-based multi-modular computational model to simulate lymphocyte population dynamics in terms of the molecular networks that control NF-κB signaling, the cell cycle, and apoptosis. Combining modeling and experimentation, we found that NF-κB cRel enforces the execution of a cellular decision between mutually exclusive fates by promoting survival in growing cells. But as cRel deficiency causes growing B cells to die at similar rates to non-growing cells, our analysis reveals that the phenomenological decision model of wild-type cells is rooted in a biased race of cell fates. We show that a multi-scale modeling approach allows for the prediction of dynamic organ-level physiology in terms of intra-cellular molecular networks.

BAFF, an activator of the noncanonical NFκB pathway, provides critical survival signals during B cell maturation and contributes to B cell proliferation. We found that the NFκB family member RelB is required ex vivo for B cell maturation, but cRel is required for proliferation. Combined molecular network modeling and experimentation revealed Nfkb2 p100 as a pathway switch; at moderate p100 synthesis rates in maturing B cells, BAFF fully utilizes p100 to generate the RelB:p52 dimer, whereas at high synthesis rates, p100 assembles into multimeric IκBsome complexes, which BAFF neutralizes in order to potentiate cRel activity and B cell expansion. Indeed, moderation of p100 expression or disruption of IκBsome assembly circumvented the BAFF requirement for full B cell expansion. Our studies emphasize the importance of p100 in determining distinct NFκB network states during B cell biology, which causes BAFF to have context-dependent functional consequences.

The NF-κB protein RelB controls dendritic cell (DC) maturation and may be targeted therapeutically to manipulate T cell responses in disease. Here we report that RelB promoted DC activation not as the expected RelB-p52 effector of the noncanonical NF-κB pathway, but as a RelB-p50 dimer regulated by canonical IκBs, IκBα and IκBɛ. IκB control of RelB minimized spontaneous maturation but enabled rapid pathogen-responsive maturation. Computational modeling of the NF-κB signaling module identified control points of this unexpected cell type-specific regulation. Fibroblasts that we engineered accordingly showed DC-like RelB control. Canonical pathway control of RelB regulated pathogen-responsive gene expression programs. This work illustrates the potential utility of systems analyses in guiding the development of combination therapeutics for modulating DC-dependent T cell responses.

Targeted deletion of BAFF causes severe deficiency of splenic B cells. BAFF-R is commonly thought to signal to nuclear factor κ-light-chain-enhancer of activated B cells (NF-κB)-inducing kinase dependent noncanonical NF-κB RelB. However, RelB-deficient mice have normal B-cell numbers. Recent studies showed that BAFF also signals to the canonical NF-κB pathway, and we found that both RelB and cRel are persistently activated, suggesting BAFF signaling coordinates both pathways to ensure robust B-cell development. Indeed, we report now that combined loss of these 2 NF-κB family members leads to impaired BAFF-mediated survival and development in vitro. Although single deletion of RelB and cRel was dispensable for normal B-cell development, double knockout mice displayed an early B-cell developmental blockade and decreased mature B cells. Despite disorganized splenic architecture in Relb(-/-)cRel(-/-) mice, generation of mixed-mouse chimeras established the developmental phenotype to be B-cell intrinsic. Together, our results indicate that BAFF signals coordinate both RelB and cRel activities to ensure survival during peripheral B-cell maturation.

The transcription factor NF-κB is a regulator of inflammatory and adaptive immune responses, yet only IκBα was shown to limit NF-κB activation and inflammatory responses. We investigated another negative feedback regulator, IκBε, in the regulation of B cell proliferation and survival. Loss of IκBε resulted in increased B cell proliferation and survival in response to both antigenic and innate stimulation. NF-κB activity was elevated during late-phase activation, but the dimer composition was stimulus specific. In response to IgM, cRel dimers were elevated in IκBε-deficient cells, yet in response to LPS, RelA dimers also were elevated. The corresponding dimer-specific sequences were found in the promoters of hyperactivated genes. Using a mathematical model of the NF-κB-signaling system in B cells, we demonstrated that kinetic considerations of IκB kinase-signaling input and IκBε's interactions with RelA- and cRel-specific dimers could account for this stimulus specificity. cRel is known to be the key regulator of B cell expansion. We found that the RelA-specific phenotype in LPS-stimulated cells was physiologically relevant: unbiased transcriptome profiling revealed that the inflammatory cytokine IL-6 was hyperactivated in IκBε(-/-) B cells. When IL-6R was blocked, LPS-responsive IκBε(-/-) B cell proliferation was reduced to near wild-type levels. Our results provide novel evidence for a critical role for immune-response functions of IκBε in B cells; it regulates proliferative capacity via at least two mechanisms involving cRel- and RelA-containing NF-κB dimers. This study illustrates the importance of kinetic considerations in understanding the functional specificity of negative-feedback regulators.

Understanding human-specific patterns of brain gene expression and regulation can provide key insights into human brain evolution and speciation. Here, we use next-generation sequencing, and Illumina and Affymetrix microarray platforms, to compare the transcriptome of human, chimpanzee, and macaque telencephalon. Our analysis reveals a predominance of genes differentially expressed within human frontal lobe and a striking increase in transcriptional complexity specific to the human lineage in the frontal lobe. In contrast, caudate nucleus gene expression is highly conserved. We also identify gene coexpression signatures related to either neuronal processes or neuropsychiatric diseases, including a human-specific module with CLOCK as its hub gene and another module enriched for neuronal morphological processes and genes coexpressed with FOXP2, a gene important for language evolution. These data demonstrate that transcriptional networks have undergone evolutionary remodeling even within a given brain region, providing a window through which to view the foundation of uniquely human cognitive capacities.

Infantile-onset Pompe Disease (IOPD), caused by mutations in lysosomal acid alpha-glucosidase (Gaa), manifests rapidly progressive fatal cardiac and skeletal myopathy incompletely attenuated by synthetic GAA intravenous infusions. The currently available murine model does not fully simulate human IOPD, displaying skeletal myopathy with late-onset hypertrophic cardiomyopathy. Bearing a Cre-LoxP induced exonic disruption of the murine Gaa gene, this model is also not amenable to genome-editing based therapeutic approaches. We report the early onset of severe hypertrophic cardiomyopathy in a novel murine IOPD model generated utilizing CRISPR-Cas9 homology-directed recombination to harbor the orthologous Gaa mutation c.1826dupA (p.Y609*), which causes human IOPD. We demonstrate the dual sgRNA approach with a single-stranded oligonucleotide donor is highly specific for the Gaa^c.1826 locus without genomic off-target effects or rearrangements. Cardiac and skeletal muscle were deficient in Gaa mRNA and enzymatic activity and accumulated high levels of glycogen. The mice demonstrated skeletal muscle weakness but did not experience early mortality. Altogether, these results demonstrate that the CRISPR-Cas9 generated Gaa^c.1826dupA murine model recapitulates hypertrophic cardiomyopathy and skeletal muscle weakness of human IOPD, indicating its utility for evaluation of novel therapeutics.

Understanding human-specific patterns of brain gene expression and regulation can provide key insights into human brain evolution and speciation. Here, we use next-generation sequencing, and Illumina and Affymetrix microarray platforms, to compare the transcriptome of human, chimpanzee, and macaque telencephalon. Our analysis reveals a predominance of genes differentially expressed within human frontal lobe and a striking increase in transcriptional complexity specific to the human lineage in the frontal lobe. In contrast, caudate nucleus gene expression is highly conserved. We also identify gene coexpression signatures related to either neuronal processes or neuropsychiatric diseases, including a human-specific module with CLOCK as its hub gene and another module enriched for neuronal morphological processes and genes coexpressed with FOXP2, a gene important for language evolution. These data demonstrate that transcriptional networks have undergone evolutionary remodeling even within a given brain region, providing a window through which to view the foundation of uniquely human cognitive capacities.