Perturbation Biology: Inferring Signaling Networks in Cellular Systems

We present a powerful experimental-computational technology for inferring network models that predict the response of cells to perturbations, and that may be useful in the design of combinatorial therapy against cancer. The experiments are systematic series of perturbations of cancer cell lines by targeted drugs, singly or in combination. The response to perturbation is quantified in terms of relative changes in the measured levels of proteins, phospho-proteins and cellular phenotypes such as viability. Computational network models are derived de novo, i.e., without prior knowledge of signaling pathways, and are based on simple non-linear differential equations. The prohibitively large solution space of all possible network models is explored efficiently using a probabilistic algorithm, Belief Propagation (BP), which is three orders of magnitude faster than standard Monte Carlo methods. Explicit executable models are derived for a set of perturbation experiments in SKMEL-133 melanoma cell lines, which are resistant to the therapeutically important inhibitor of RAF kinase. The resulting network models reproduce and extend known pathway biology. They empower potential discoveries of new molecular interactions and predict efficacious novel drug perturbations, such as the inhibition of PLK1, which is verified experimentally. This technology is suitable for application to larger systems in diverse areas of molecular biology.     

A Refined Model of the Prototypical Salmonella SPI-1 T3SS Basal Body Reveals the Molecular Basis for Its Assembly

The T3SS injectisome is a syringe-shaped macromolecular assembly found in pathogenic Gram-negative bacteria that allows for the direct delivery of virulence effectors into host cells. It is composed of a “basal body”, a lock-nut structure spanning both bacterial membranes, and a “needle” that protrudes away from the bacterial surface. A hollow channel spans throughout the apparatus, permitting the translocation of effector proteins from the bacterial cytosol to the host plasma membrane. The basal body is composed largely of three membrane-embedded proteins that form oligomerized concentric rings. Here, we report the crystal structures of three domains of the prototypical Salmonella SPI-1 basal body, and use a new approach incorporating symmetric flexible backbone docking and EM data to produce a model for their oligomeric assembly. The obtained models, validated by biochemical and in vivo assays, reveal the molecular details of the interactions driving basal body assembly, and notably demonstrate a conserved oligomerization mechanism.     

Whole Genome Sequencing Identifies a Deletion in Protein Phosphatase 2A That Affects Its Stability and Localization in Chlamydomonas reinhardtii

Whole genome sequencing is a powerful tool in the discovery of single nucleotide polymorphisms (SNPs) and small insertions/deletions (indels) among mutant strains, which simplifies forward genetics approaches. However, identification of the causative mutation among a large number of non-causative SNPs in a mutant strain remains a big challenge. In the unicellular biflagellate green alga Chlamydomonas reinhardtii, we generated a SNP/indel library that contains over 2 million polymorphisms from four wild-type strains, one highly polymorphic strain that is frequently used in meiotic mapping, ten mutant strains that have flagellar assembly or motility defects, and one mutant strain, imp3, which has a mating defect. A comparison of polymorphisms in the imp3 strain and the other 15 strains allowed us to identify a deletion of the last three amino acids, Y₃₁₃F₃₁₄L₃₁₅, in a protein phosphatase 2A catalytic subunit (PP2A3) in the imp3 strain. Introduction of a wild-type HA-tagged PP2A3 rescues the mutant phenotype, but mutant HA-PP2A3 at Y₃₁₃ or L₃₁₅ fail to rescue. Our immunoprecipitation results indicate that the Y₃₁₃, L₃₁₅, or YFLΔ mutations do not affect the binding of PP2A3 to the scaffold subunit, PP2A-2r. In contrast, the Y₃₁₃, L₃₁₅, or YFLΔ mutations affect both the stability and the localization of PP2A3. The PP2A3 protein is less abundant in these mutants and fails to accumulate in the basal body area as observed in transformants with either wild-type HA-PP2A3 or a HA-PP2A3 with a V₃₁₀T change. The accumulation of HA-PP2A3 in the basal body region disappears in mated dikaryons, which suggests that the localization of PP2A3 may be essential to the mating process. Overall, our results demonstrate that the terminal YFL tail of PP2A3 is important in the regulation on Chlamydomonas mating.     

Cauli: A Mouse Strain with an Ift140 Mutation That Results in a Skeletal Ciliopathy Modelling Jeune Syndrome

Cilia are architecturally complex organelles that protrude from the cell membrane and have signalling, sensory and motility functions that are central to normal tissue development and homeostasis. There are two broad categories of cilia; motile and non-motile, or primary, cilia. The central role of primary cilia in health and disease has become prominent in the past decade with the recognition of a number of human syndromes that result from defects in the formation or function of primary cilia. This rapidly growing class of conditions, now known as ciliopathies, impact the development of a diverse range of tissues including the neural axis, craniofacial structures, skeleton, kidneys, eyes and lungs. The broad impact of cilia dysfunction on development reflects the pivotal position of the primary cilia within a signalling nexus involving a growing number of growth factor systems including Hedgehog, Pdgf, Fgf, Hippo, Notch and both canonical Wnt and planar cell polarity. We have identified a novel ENU mutant allele of Ift140, which causes a mid-gestation embryonic lethal phenotype in homozygous mutant mice. Mutant embryos exhibit a range of phenotypes including exencephaly and spina bifida, craniofacial dysmorphism, digit anomalies, cardiac anomalies and somite patterning defects. A number of these phenotypes can be attributed to alterations in Hedgehog signalling, although additional signalling systems are also likely to be involved. We also report the identification of a homozygous recessive mutation in IFT140 in a Jeune syndrome patient. This ENU-induced Jeune syndrome model will be useful in delineating the origins of dysmorphology in human ciliopathies.     

A Bayesian Approach for Modeling Cattle Movements in the United States: Scaling up a Partially Observed Network

Networks are rarely completely observed and prediction of unobserved edges is an important problem, especially in disease spread modeling where networks are used to represent the pattern of contacts. We focus on a partially observed cattle movement network in the U.S. and present a method for scaling up to a full network based on Bayesian inference, with the aim of informing epidemic disease spread models in the United States. The observed network is a 10% state stratified sample of Interstate Certificates of Veterinary Inspection that are required for interstate movement; describing approximately 20,000 movements from 47 of the contiguous states, with origins and destinations aggregated at the county level. We address how to scale up the 10% sample and predict unobserved intrastate movements based on observed movement distances. Edge prediction based on a distance kernel is not straightforward because the probability of movement does not always decline monotonically with distance due to underlying industry infrastructure. Hence, we propose a spatially explicit model where the probability of movement depends on distance, number of premises per county and historical imports of animals. Our model performs well in recapturing overall metrics of the observed network at the node level (U.S. counties), including degree centrality and betweenness; and performs better compared to randomized networks. Kernel generated movement networks also recapture observed global network metrics, including network size, transitivity, reciprocity, and assortativity better than randomized networks. In addition, predicted movements are similar to observed when aggregated at the state level (a broader geographic level relevant for policy) and are concentrated around states where key infrastructures, such as feedlots, are common. We conclude that the method generally performs well in predicting both coarse geographical patterns and network structure and is a promising method to generate full networks that incorporate the uncertainty of sampled and unobserved contacts.     

Risk of Cerebral Palsy and Childhood Epilepsy Related to Infections before or during Pregnancy

Background and Aim

Maternal infections during pregnancy have been associated with several neurological disorders in the offspring. However, given the lack of specificity for both the exposures and the outcomes, other factors related to infection such as impaired maternal immune function may be involved in the causal pathway. If impaired maternal immune function plays a role, we would expect infection before pregnancy to be associated with these neurological outcomes.

Methods/Principal Findings

The study population included all first-born singletons in Denmark between January 1 1982 and December 31 2004. We identified women who had hospital-recorded infections within the 5 year period before pregnancy, and women who had hospital-recorded infections during pregnancy. We grouped infections into either infections of the genitourinary system, or any other infections. Cox models were used to estimate adjusted hazard ratios (aHRs) with 95% confidence interval (CI). Maternal infection of the genitourinary system during pregnancy was associated with an increased risk of cerebral palsy (aHR = 1.63, 95% CI: 1.34–1.98) and epilepsy (aHR = 1.27, 95% CI: 1.13–1.42) in the children, compared to children of women without infections during pregnancy. Among women without hospital-recorded infections during pregnancy, maternal infection before pregnancy was associated with an increased risk of epilepsy (aHR = 1.35, 95% CI: 1.21–1.50 for infections of the genitourinary system, and HR = 1.12, 95% CI: 1.03–1.22 for any other infections) and a slightly higher risk of cerebral palsy (aHR = 1.20, 95% CI: 0.96–1.49 for infections of the genitourinary system, and HR = 1.23, 95% CI: 1.06–1.43 for any other infections) in the children, compared to children of women without infections before (and during) pregnancy.

Conclusions

These findings indicate that the maternal immune system, maternal infections, or factors related to maternal immune function play a role in the observed associations between maternal infections before pregnancy and cerebral diseases in the offspring.     

Quantitative Bioluminescent Imaging of Pre-Erythrocytic Malaria Parasite Infection Using Luciferase-Expressing Plasmodium yoelii

The liver stages of Plasmodium parasites are important targets for the development of anti-malarial vaccine candidates and chemoprophylaxis approaches that aim to prevent clinical infection. Analyzing the impact of interventions on liver stages in the murine malaria model system Plasmodium yoelii has been cumbersome and requires terminal procedures. In vivo imaging of bioluminescent parasites has previously been shown to be an effective and non-invasive alternative to monitoring liver stage burden in the Plasmodium berghei model. Here we report the generation and characterization of a transgenic P. yoelii parasite expressing the reporter protein luciferase throughout the parasite life cycle. In vivo bioluminescent imaging of these parasites allows for quantitative analysis of P. yoelii liver stage burden and parasite development, which is comparable to quantitative RT-PCR analysis of liver infection. Using this system, we show that both BALB/cJ and C57BL/6 mice show comparable susceptibility to P. yoelii infection with sporozoites and that bioluminescent imaging can be used to monitor protective efficacy of attenuated parasite immunizations. Thus, this rapid, simple and noninvasive method for monitoring P. yoelii infection in the liver provides an efficient system to screen and evaluate the effects of anti-malarial interventions in vivo and in real-time.     

Expression of Fungal Cutinase and Swollenin in Tobacco Chloroplasts Reveals Novel Enzyme Functions and/or Substrates

In order to produce low-cost biomass hydrolyzing enzymes, transplastomic lines were generated that expressed cutinase or swollenin within chloroplasts. While swollenin expressing plants were homoplasmic, cutinase transplastomic lines remained heteroplasmic. Both transplastomic lines showed interesting modifications in their phenotype, chloroplast structure, and functions. Ultrastructural analysis of chloroplasts from cutinase- and swollenin-expressing plants did not show typical lens shape and granal stacks. But, their thylakoid membranes showed unique scroll like structures and chloroplast envelope displayed protrusions, stretching into the cytoplasm. Unusual honeycomb structures typically observed in etioplasts were observed in mature chloroplasts expressing swollenin. Treatment of cotton fiber with chloroplast-derived swollenin showed enlarged segments and the intertwined inner fibers were irreversibly unwound and fully opened up due to expansin activity of swollenin, causing disruption of hydrogen bonds in cellulose fibers. Cutinase transplastomic plants showed esterase and lipase activity, while swollenin transplastomic lines lacked such enzyme activities. Higher plants contain two major galactolipids, monogalactosyldiacylglycerol (MGDG) and digalactosyldiacylglycerol (DGDG), in their chloroplast thylakoid membranes that play distinct roles in their structural organization. Surprisingly, purified cutinase effectively hydrolyzed DGDG to MGDG, showing alpha galactosidase activity. Such hydrolysis resulted in unstacking of granal thylakoids in chloroplasts and other structural changes. These results demonstrate DGDG as novel substrate and function for cutinase. Both MGDG and DGDG were reduced up to 47.7% and 39.7% in cutinase and 68.5% and 67.5% in swollenin expressing plants. Novel properties and functions of both enzymes reported here for the first time should lead to better understanding and enhanced biomass hydrolysis.     

Paclitaxel-Induced Apoptosis Is BAK-Dependent,but BAX and BIM-Independent in Breast Tumor

Paclitaxel (Taxol)-induced cell death requires the intrinsic cell death pathway, but the specific participants and the precise mechanisms are poorly understood. Previous studies indicate that a BH3-only protein BIM (BCL-2 Interacting Mediator of cell death) plays a role in paclitaxel-induced apoptosis. We show here that BIM is dispensable in apoptosis with paclitaxel treatment using bim^−/− MEFs (mouse embryonic fibroblasts), the bim^−/− mouse breast tumor model, and shRNA-mediated down-regulation of BIM in human breast cancer cells. In contrast, both bak ^−/− MEFs and human breast cancer cells in which BAK was down-regulated by shRNA were more resistant to paclitaxel. However, paclitaxel sensitivity was not affected in bax^−/− MEFs or in human breast cancer cells in which BAX was down-regulated, suggesting that paclitaxel-induced apoptosis is BAK-dependent, but BAX-independent. In human breast cancer cells, paclitaxel treatment resulted in MCL-1 degradation which was prevented by a proteasome inhibitor, MG132. A Cdk inhibitor, roscovitine, blocked paclitaxel-induced MCL-1 degradation and apoptosis, suggesting that Cdk activation at mitotic arrest could induce subsequent MCL-1 degradation in a proteasome-dependent manner. BAK was associated with MCL-1 in untreated cells and became activated in concert with loss of MCL-1 expression and its release from the complex. Our data suggest that BAK is the mediator of paclitaxel-induced apoptosis and could be an alternative target for overcoming paclitaxel resistance.     

Fluctuations in Species-Level Protein Expression Occur during Element and Nutrient Cycling in the Subsurface

While microbial activities in environmental systems play a key role in the utilization and cycling of essential elements and compounds, microbial activity and growth frequently fluctuates in response to environmental stimuli and perturbations. To investigate these fluctuations within a saturated aquifer system, we monitored a carbon-stimulated in situ Geobacter population while iron reduction was occurring, using 16S rRNA abundances and high-resolution tandem mass spectrometry proteome measurements. Following carbon amendment, 16S rRNA analysis of temporally separated samples revealed the rapid enrichment of Geobacter-like environmental strains with strong similarity to G. bemidjiensis. Tandem mass spectrometry proteomics measurements suggest high carbon flux through Geobacter respiratory pathways, and the synthesis of anapleurotic four carbon compounds from acetyl-CoA via pyruvate ferredoxin oxidoreductase activity. Across a 40-day period where Fe(III) reduction was occurring, fluctuations in protein expression reflected changes in anabolic versus catabolic reactions, with increased levels of biosynthesis occurring soon after acetate arrival in the aquifer. In addition, localized shifts in nutrient limitation were inferred based on expression of nitrogenase enzymes and phosphate uptake proteins. These temporal data offer the first example of differing microbial protein expression associated with changing geochemical conditions in a subsurface environment.