The Structure, Function and Evolution of Biological Networks

Every cell must coordinate and regulate thousands of genes in a robust manner to ensure cellular function. Complex gene regulation is needed for most biological processes, such as development, physiological homeostasis, and response to environmental stresses. Networks of genes have evolved to coordinate and integrate biological processes within a cell and subsequently regulate the fate of the organism.

The Walhout lab wishes to understand biological networks and how these networks adapt to different conditions. We use systems biology approaches to dissect these complex networks. These approaches combine high-quality and large-scale genetic and biochemical data sets and uses computational modeling to integrate the data such that the organizing principles and emergent properties of biological systems are unveiled.

To examine biological networks we mainly use the model organism round worm Caenorhabditis elegans. Worms are highly adaptable, easy to manipulate, and have many analogs in human genetics. Furthermore, there are many genetic tools and worm-specific techniques that are not available for studying higher eukaryotes. Overall, our research involves two broad areas of biology:

1: Gene Regulatory Networks

2: Regulation of Metabolic Networks

1. Gene Regulatory Networks

Transcription factor activity mapping

From MacNeil et al., 2015. In vivo gene regulatory network based on transcription factor activity.

Gene regulatory networks capture physical and/or regulatory interactions between transcription factors (TFs) and their targets. Physical interactions include the direct binding of TFs to DNA targets, while regulatory interactions describe the effect of a TF on a downstream target gene, such as activation or repression. Not all physical interactions confer a regulatory output. Furthermore, regulatory interactions are not necessarily the result of direct binding. For instance, an activator can activate an activator of a gene, and removal of either results in a decrease in the mRNA levels of that gene.

In the Walhout lab, we systematically map both physical and regulatory interactions between TFs and their target genes. We utilize gene-centered techniques such as enhanced yeast one-hybrid assays to map protein-DNA interactions (PDIs) (Reece-Hoyes et al., 2011). We primarily focus on the nematode C. elegans (Fuxman Bass et al., 2016; MacNeil et al., 2015), but have also studied human regulatory genomic sequences (Fuxman Bass et al., 2015; Sahni et al., 2015). We have  mapped PDIs of ~15% of C. elegans protein-coding gene promoters, the largest PDI network to date (Fuxman Bass et al., 2016).

In addition to examining physical interactions, we also use gene-centered methods to study regulatory networks. Using RNAi to individual TFs in C. elegans transgenic strains containing intestine-specific reporter genes, we comprehensively mapped TF activity at the regulatory level and delineated the first tissue-specific in vivo gene regulatory network (MacNeil et al., 2015). This regulatory network displayed modest over lap with physical interactions, and allowed us to computationally model information flow from master regulatory genes to terminal nodes.

Finally, most genes function in more than one type of network, be it protein-DNA, protein-protein, gene regulatory, spatiotemporal, or a number of other parameters. We are one of the few laboratories that studies biological networks on a multiparameter scale. Our first studies in multiparameter networks examined dimerization partners, spatiotemporal expression patterns, and DNA-binding specificities for the C. elegans bHLH family of TFs (Grove et al., 2009). More recently we examined the evolutionary dynamics between C. elegans TF networks and networks of transcriptional cofators (Reece-Hoyes et al., 2013).


From Reece-Hoyes et al., 2013. A C. elegans TF-TF protein-protein interaction network, delineated by enhanced yeast two-hybrid assays.

Together with Chad Myers at the University of Minnesota, we have developed a set of tools that are publicly available to examine the association indices, or similarity between genes or TFs (see GAIN, SpotON, and MyBrid). In the future we will continue to map TF-DNA interactions and integrate this data with tissue expression and phenotypic data. We would also like to examine how gene regulatory networks are organized at the spatiotemporal level. Finally, we would like to examine how these gene regulatory networks change over the lifetime of an organism.

Recent publications:
Fuxman Bass, JI, Pons, C, Kozlowski, L, Reece‐Hoyes,JS, Shrestha, S, Holdorf, AD, Mori, A, Myers, CL, Walhout, AJM. (2016). A gene‐centered C. elegans protein–DNA interaction network provides a framework for functional predictions. Mol. Sys. Biol. 12: 884. doi: 10.15252/msb.20167131.

MacNeil LT, Pons C, Arda HE, Giese GE, Myers CL, Walhout AJM. (2015). Transcription Factor Activity Mapping of a Tissue-Specific In Vivo Gene Regulatory Network. Cell Systems 1, 152-162. doi: 10.1016/j.cels.2015.08.003.

Fuxman Bass JI, Sahni N, Shrestha S, Garcia-Gonzalez A, Mori A, Bhat N, Yi S, Hill DE, Vidal M, Walhout AJM. (2015). Human Gene-Centered Transcription Factor Networks for Enhancers and Disease Variants. Cell 161, 661–73. doi: 10.1016/j.cell.2015.03.003.

Sahni N*, Yi S*, Taipale M*, Fuxman Bass JI*, Coulombe-Huntington J*, Yang F, Peng J, Weile J, Karras GI, Wang Y, Kovács IA, Kamburov A, Krykbaeva I, Lam MH, Tucker G, Khurana V, Sharma A, Liu Y, Yachie N, Zhong Q, Shen Y, Palagi A, San-Miguel A, Fan C, Balcha D, Dricot A, Jordan DM, Walsh JM, Shah AA, Yang X, Stoyanova AK, Leighton A, Calderwood MA, Jacob Y, Cusick ME, Salehi-Ashtiani K, Whitesell LJ, Sunyaev S, Berger B, Barabási A, Charloteaux B, Hill DE, Hao T, Roth FP, Yu X, Walhout AJM, Lindquist S, Vidal M. (2015). Widespread Macromolecular Interaction Perturbations in Human Genetic Disorders. Cell 161, 647–60. doi: 10.1016/j.cell.2015.04.013.

Reece-Hoyes JS, Pons C, Diallo A, Mori A, Shrestha S, Kadreppa S, Nelson J, Diprima S, Dricot A, Lajoie BR, Ribeiro PS, Weirauch MT, Hill DE, Hughes TR, Myers CL, Walhout AJ. (2013). Extensive rewiring and complex evolutionary dynamics in a C. elegans multiparameter transcription factor network. Mol. Cell 51:116-27. doi: 10.1016/j.molcel.2013.05.018.

2: Regulation of Metabolic Networks

“You are what you eat” is a well-known phrase implying that you will be healthy if you eat nutritious food. While this statement is not literally true, it is well appreciated that our diet plays a fundamental role in our development, physiology and performance. Furthermore, diet and metabolism have been implicated in a variety of inherited and acquired diseases such as diabetes, cancer, and heart disease.

We use C. elegans to study how diet affects major life history traits such as development, reproduction, and aging. We used systems approaches to unravel the networks that govern different dietary responses. We found that feeding C. elegans worms two different species of bacteria can result in very different worm life history traits (MacNeil et al., 2013; Watson et al., 2013). Theses finds have implications for human diseases like Type 2 Diabetes and inborn diseases of metabolism, as well as investigations into the human intestinal microbiome. This work was featured on the cover of Cell, in the Boston Globe, and on the below video.


To further understand how diet affects life history traits, we used a combination of worm and bacterial genetics, as well as metabolite screens. With this interspecies systems biology approach we discovered that a diet of Comamonas aquatica provides the worms with vitamin B12, which in turn accelerates development and reduces fertility as compared to a diet of Escherichia coli (Watson et al., 2014). We are one of the few labs in the world using interspecies systems biology to understand the complex interactions between diet and physiology. This is an extremely exciting area of research with many applications to further our understanding of human diseases.


From Watson et al., 2016. A vitamin B-12 independent pathway of propionate breakdown.

From Watson et al., 2016. A vitamin B12 independent pathway of propionate breakdown.

Using a combination of systems biology, bacterial and worm genetics, computational biology, and biochemistry, we delineated a novel mechanism of propionate breakdown in the worm. Propionate buildup is a product of amino acid catabolism, and too much propionate is toxic to worms and mammals. The canonical mechanism of propionate breakdown is dependent on vitamin B12, and patients with mutations in the enzymes in this pathway are born with devastating, often fatal, acidemia diseases. We found that C. elegans uses a vitamin B12-independent mechanism of propionate breakdown, suggesting that in the wild C. elegans has the metabolic plasticity to rewire metabolism depending on available nutrients (Waston et al., 2016). We are furthering these studies to understand the transcriptional regulation of propionate metabolism, and also examining the transcriptional and regulatory regulation of other metabolites.

Nuclear hormone receptors (NHRs) are TFs which are regulated by the binding of a specific ligand to the TF. Ligand binding to a nuclear receptor results in a conformational change and activation of the receptor. The NHR then binds to its specific DNA element which results in a change of gene expression. Some important NHRs in mammalian physiology include the estrogen receptor, the glucocorticoid receptor, and the retinoic acid receptor.

Interestingly, while humans have only 48 of such NHRs, worms have more than 250! The ligands for and function of worm NHRs is largely unknown. We have started to delineate regulatory networks of metabolic genes and found this network to be enriched for nuclear NHRs (Arda et al., 2010; Watson et al., 2013). Many questions remain about NHRs in metabolic networks: What are these NHRs doing? How do they work? What are their ligands? We are also investigating how NHRs contribute to the overall function of the animal and how they are wired in different types of regulatory networks.


Yilmaz & Walhout, 2016. A Caenorhabditis elegans Genome-Scale Metabolic Network Model

In addition our genetic studies on delineating metabolic networks, we are also interested in computational models predicting C. elegans metabolic flux. Our bioengineer Safak Yilmaz developed a genome-scale metabolic network model of C. elegans along with a web tool called Wormflux. This model contains 1273 genes, 623 enzymes, and 1985 metabolic reactions, and can be integrated with gene expression data to examine biological processes such as dauer formation. Future studies will involve adding genes, enzymes, and reactions to the model, and modeling metabolic reactions in specific compartments based on protein localization. Additionally, we will integrate the E. coli metabolic network model with the C. elegans network model to delineate specific interspecies interactions.

Recent publications:
Watson E, Olin-Sandoval V, Hoy MJ, Li C, Louisse T, Yao V, Mori A, Holdorf AD, Troyanskaya OG, Ralser M, Walhout AJM. (2016). Metabolic network rewiring of propionate flux compensates vitamin B12 deficiency in C. elegans. eLife, doi: 10.7554/eLife.17670.

Yilmaz LS, Walhout AJM. (2016). A Caenorhabditis elegans Genome-Scale Metabolic Network Model. Cell Systems 2, 297–311. doi: 10.1016/j.cels.2016.04.012.

Watson E, MacNeil LT, Ritter AD, Yilmaz LS, Rosebrock AP, Caudy AA, Walhout AJ. (2014). Interspecies systems biology uncovers metabolites affecting C. elegans gene expression and life history traits. Cell. 156:759-70. doi: 10.1016/j.cell.2014.01.047.

MacNeil LT, Watson E, Arda HE, Zhu LJ, Walhout AJ. (2013). Diet-Induced developmental acceleration independent of TOR and insulin in C. elegans. Cell, 153:240-252. doi: 10.1016/j.cell.2013.02.049.

Watson E*, MacNeil LT*, Arda HE, Zhu LJ, Walhout AJ. (2013). Integration of metabolic and gene regulatory networks modulates the C. elegans dietary response. Cell, 153:253-266. doi: 10.1016/j.cell.2013.02.050.