Rethinking retinoic acid self-regulation: A signaling robustness network approach.

All-trans retinoic acid (ATRA) signaling is a major pathway regulating numerous differentiation, proliferation, and patterning processes throughout life. ATRA biosynthesis depends on the nutritional availability of vitamin A and other retinoids and carotenoids, while it is sensitive to dietary and environmental toxicants. This nutritional and environmental influence requires a robustness response that constantly fine-tunes the ATRA metabolism to maintain a context-specific, physiological range of signaling levels. The ATRA metabolic and signaling network is characterized by the existence of multiple enzymes, transcription factors, and binding proteins capable of performing the same activity. The partial spatiotemporal expression overlap of these enzymes and proteins yields different network compositions in the cells and tissues where this pathway is active. Genetic polymorphisms affecting the activity of individual network components further impact the network composition variability and the self-regulatory feedback response to ATRA fluctuations. Experiments directly challenging the robustness response uncovered a Pareto optimality in the ATRA network, such that some genetic backgrounds efficiently deal with excess ATRA but are very limited in their robustness response to reduced ATRA and vice versa. We discuss a network-focused framework to describe the robustness response and the Pareto optimality of the ATRA metabolic and signaling network.