PROJECT SUMMARY ABSTRACT
To achieve robust control of cellular decision-making, including differentiation, a quantitative, mechanistic
description of signaling pathway dynamics and regulation is required. Indeed, signaling pathways dictate dif-
ferential gene expression, which is often the first step in differentiation. Measuring the concentration, distribu-
tion, and interactions of signaling proteins in live cells lays the foundation for building predictive models of sig-
naling. Quantitative imaging of live cells is especially important in cells and tissues that have feedback loops,
because feedback is an inherently dynamic situation.
The overall objective of this proposal is to to synergistically use systems biology modeling and cutting edge
quantitative imaging to build a predictive model of the NF-kappaB/Dorsal gradient, which patterns the dorsal-
ventral axis of the 1-3 hr old Drosophila embryo. The central hypothesis is that negative feedback decreases
the response time of the Dl gradient, through Dl/Cact interactions within the nuclei, thereby allowing the gradi-
ent to rapidly achieve proper levels. To test this hypothesis, the following Specific Aims will be performed:
Specific Aim 1: Determine the functional role of Cact in the nucleus. It is hypothesized that nuclear-
localized Cact ensures Dl levels rapidly reach steady state, so that gene expression is properly specified along
the DV axis. To test this hypothesis, measurements will be performed to capture the spatiotemporal dynamics
of Dl, Cact, and Dl/Cact complex in the nuclei (vs cytoplasm) along the DV axis of blastoderm stage embryos.
Specific Aim 2: Determine the functional role of the cact negative feedback loop. It is hypothesized that
weakening negative feedback will prevent the Dl gradient from reaching steady state, which will lead to impre-
cise gene expression domains. To test this hypothesis, methods described in Aim 1 will be used to measure
Dl, Cact, and Dl/Cact complex in embryos that lack the negative feedback loop. Expression patterns of Dl tar-
get genes will also be measured. Specific Aim 3: Understanding a system at the mechanistic and quantitative
level naturally leads to the formulation of a mathematical model. Ultimately, the long-term goal is to use predic-
tive models to aid in process control of biological decisions. The goal in this Aim is to make the Dl gradient
model predictive. Therefore, the model will be constrained with measured data, used to make a series of spe-
cific predictions, and used to design a list of optimal experiments.
The work will lead to several positive impacts, including, a detailed, mechanistic description impacts the
understanding of NF-κB in vertebrate systems, an understanding of the diversity of dynamical behavior exhibit-
ed by negative feedback in the NF-κB pathway, and approaches generalized to other systems.
