Individual-based null models and resampling methods for the analysis of heterochronic differences in development

Date

1999-12

Journal Title

Journal ISSN

Volume Title

Publisher

Texas Tech University

Abstract

Application of parametric bootstrapping and randomization methods to the analysis of heterochrony (differential timing of developmental events at the cellular or organismal level) may allow scientists to quantitatively test hypotheses about the effects of environmental variability on heterochrony. Herein, I focused on quantitative analysis and modeling of heterochrony, for (1) hierarchical (cell lineages), and (2) serial cossification sequences) developmental systems. For (1), I used the published data on embryogenesis of Caenorhabditis elegans (Schnabel et al., 1997) to simulate and compare cell lineages ('ABarpp' sublineage) at different temperatures (20 vs. 25C). The simulations suggested that the relative frequencies of the shared sequences were significantly different between groups (X^=3569, p<10'^ 10,000 iterations/group), indicating an increase in heterochrony as a response to increased temperature. The presence or absence of bilateral asymmetry in the timing of cell divisions was tested based on a null model of random sequence. Results of Wilcoxon tests on paired differences between the left and right ranks indicated that a) terminal cells on the right side of the embryo were likely to divide earlier than those on the left; and b) higher temperatures increased the degree of asymmetry. The effects of physical constraints (e.g., handedness of the early embryonic body plan) that may lead to the emergence of such an asymmetric pattern were discussed. For (2), I analyzed the ossification patterns of the vertebrae and branchial arches in the zebrafish. Danio rerio, exposed to different levels of thyroid hormone (T3). i.e., 1 ng/ml and 5 ng/ml. Onset of ossification showed a dose-dependent response to T3, such that onh at a higher dose, significant differences were observed (paired Wilcoxon tests; control vs. 1 ng/ml T3. p = 0.083; control \ s. 5 ng/ml, p = 0.002; 1 ng/ml T3 vs. 5 ng/ml T3, p = 0.004). On the other hand, the rate of vertebral ossifications was positively correlated with age for individuals exposed to 1 ng/ml T3 whereas at 5 ng/ml T3, number of ossifications reached an asymptote over time. Randomized Mantel's tests suggested that branchial arches responded to 5ng/ml T3 more than they did to 1 ng/ml T3, whereas vertebral development was accelerated at both dosages. Randomized sign tests revealed that there is a tendenc>' for the vertebrae to ossify from an anterior to posterior direction (control: p = 0.02; 1 ng/ml: p = 0.007; 5 ng/ml: p = 0.008). However, vertebrae that belong to the Weberian apparatus ossified independently of the more caudal vertebrae. This allometry suggests that the genetic and epigenetic mechanisms governing the development of cranial, Weberian, and trunk regions are at least partially decoupled. The importance of such modularity in generating novel morphs is discussed in the light of the genetic {Hox genes) and epigenetic (endocrine) mechanisms.

Description

Keywords

Caenorhabditis elegans, Zebra danio, Heterochrony (Biology), Ossification, Developmental cytology, Temperature, Thyroid hormones

Citation