Mechanisms of developmental robustness

Abstract

We present a review of noise buffering mechanisms responsible for developmental robustness. We focus on functions of chaperone Hsp90, miRNA, and cross-regulation of gap genes in Drosophila. The noise buffering mechanisms associated with these functions represent specific examples of the developmental canalization, reducing the phenotypical variability in presence of either genetic or environmental perturbations. We demonstrate that robustness often appears as a function of a network of interacting elements and that the system level approach is needed to understand the mechanisms of noise filtering.

Introduction

Biological organisms exhibit robustness in the face of environmental perturbations and potential lesions such as mutations. Throughout evolution, there has been a constant selection for inherited mechanisms that are able to keep a trait constant and reproduce stereotyped phenotypic outcomes. Genetic buffering is a term coined to describe these mechanisms. In the literature robustness and canalization are effectively used as synonyms of genetic buffering, however evolutionary nomenclature distinguishes these terms and understands under canalization genetic buffering that has evolved under natural selection in order to stabilize the phenotype and decreases its variability. Canalization allows the accumulation of ‘cryptic genetic variation’ caused by mutations that do not affect the phenotype. Canalized traits are phenotypically expressed only in particular environments or genetic backgrounds and become available for natural selection, a mechanism that can lead to the assimilation of novel traits.

The idea of canalization is first introduced in the 1940s, by the developmental biologist and geneticist C.H. Waddington. He noted that as opposed to mutants, wild type organisms are able to suppress phenotypic variation caused by genetic and environmental changes. In order to explain this fact C.H. Waddington hypothesized that in organism under natural selection the developmental processes are canalized: since cells make discrete fate decisions there must only be a finite number of distinct developmental trajectories possible and each such trajectory, called a chreod, must be stable against small perturbations (Waddington, 1962, Waddington, 1940, Waddington, 1957).

A metaphorical representation of canalization, the C.H. Waddington epigenetic landscape, represents a developing system as a mountainous landscape down which a ball rolls. The ball represents a cell and the selection of developmental pathway happens at each valley branchpoint by the action of embryonic inducing factors or genes. Each landscape valley corresponds to a collection of similar developmental trajectories. If to give the ball a little push toward the hill slope, it will return to the same valley as before, but will end up in another point than before the push. This analogy allows us to understand what canalization means: up to a certain threshold, any genetic variation or environmental noise will be ‘buffered’ and not affect development, but above this threshold, the cell would flip over into another developmental pathway. Contrary to homeostasis, which denotes the stability of the developmental system state, Waddington coined homeoresis as a term to describe the robustness of developmental pathways (Waddington, 1957).

Canalization is thought to be mediated by the buffering activity of specific genes; however, with the one notable exception of hsp90, there have been no facts successfully explaining the molecular mechanisms of canalization as a result of one gene action. On the contrary, new data point that canalization emerges at the level of gene regulatory networks (Wagner, 1996, Bergman and Siegal, 2003, Siegal and Bergman, 2002). For example, a biological system can be robust if the underlying genetic network has redundant interactions – where one interaction can compensate for the loss of another – and/or if the system has distributed robustness – when mutations in one pathway can be compensated by other processes in the system (Wagner, 2005). It is evident that if robustness is indeed an emergent property of a genetic network, we would expect it to break down in mutants affecting factors other than Hsp90. Each mutation in a system—and in particular those affecting regulatory factors—can potentially reduce redundancy and distributed robustness, and render the system more vulnerable to further perturbations. Indeed, Levy and Siegal (2008) identified over 300 mutants in yeast, which exhibit reduced robustness to environmental fluctuations, and Bergman and Siegal (2003) observed widespread loss of canalization when they simulated large numbers of single (one-step) mutants of evolved robust gene regulatory networks.

Thus, the mechanism of canalization and biological robustness remains controversial. However, this problem is at the heart of biology research over the past decade, and the number of publications on this topic is growing rapidly. Here, we summarize the current state of knowledge about several genetic buffering mechanisms.