Pattern formation and developmental mechanisms Good mileage from comparative studies in cell biology, gene regulation, development and evolution Editorial overview William McGinnis and Cheryll Tickle Current Opinion in Genetics & Development 2005, 15:355–357 Available online 23rd June 2005 0959-437X/$ – see front matter # 2005 Elsevier Ltd. All rights reserved. DOI 10.1016/j.gde.2005.06.005
William McGinnis Cell & Developmental Biology, 0349 University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0349, USA e-mail: [email protected]
William McGinnis is a Professor in the Section of Cell & Developmental Biology at the University of California, San Diego. His research is directed at understanding how Hox proteins diversify morphology during development and evolution, and how epidermal barriers are repaired after wounding. Cheryll Tickle Division of Cell & Developmental Biology, School of Life Sciences, University of Dundee, Dundee, DD15EH, UK e-mail: [email protected]
Cheryll Tickle is a Foulerton Research Professor of the Royal Society. Her research in the School of Life Sciences at Dundee University is focussed on vertebrate limb development in an effort to gain insights into the broader picture of embryonic development and vertebrate evolution.
Introduction Over the past fifteen years there has been a massive surge in the discovery of developmentally important genes, and more recently the complete genomes of many developmental model organisms have been sequenced. One initial surprise — although perhaps no longer surprising — is the conservation of developmental patterning gene functions in a variety of diverse organisms. A theme that is revealed in this issue is the continuing power of comparative studies — from flies and worms to vertebrates and vice versa, and between vertebrate models — in illuminating developmental processes. Some of the examples discussed here include specification of features of the body plan, eyes and appendages, and, more unexpectedly, generation of impermeable epidermal layers. At a higher level, genetic networks elucidated in model organisms have become so well understood that entire networks can now be applied as comparators to trace evolutionary relationships for organisms, such as hemichordates, that have been on the fringe of developmental biology in the modern era. Although there is plenty of mileage left in the use of model organisms to discover novel developmental control genes, there are many new and exciting challenges that have emerged: the challenge of relating genetic information to what cells actually do; the challenge of identifying complete sets of genes responsible for directing particular processes or for making specific structures; the challenge of finding gene regulatory elements without random tests of upstream, intronic and downstream DNA fragments; the challenge of revealing how gene diversification drives evolution; and the challenge of discovering the biological regulatory roles of non-coding RNAs.
The cell biology of development Although it is often said that mammals operate with only 200 cell types, that contention is based on an extremely broad definition of cell type. As we acquire better tools for measuring their properties, the number of recognized cell ‘types’ has proliferated; but, no matter what their type, cells have a limited repertoire of visible behaviours. They can adhere to each other or not; move or stay put; send and receive signals or ignore them; reveal their identities by differentiating, or remain as stem cells; divide or not; and decide to live or to die. Although all these processes seem active and decisive, cells can achieve their fates by rather passive routes as well. Carthew tells us how in his review on the mechanisms used by epithelial cells to assume their distinctive shapes, and in the process he reveals what soap bubbles have to do with the development of discrete cellular morphologies. Current Opinion in Genetics & Development 2005, 15:355–357
356 Pattern formation and developmental mechanisms
All cells sense their environment to some extent, but use environmental information to explore. Davies explains how tracheal development in fruit flies has been used as a system in which auto-fluorescent reporter proteins have enabled branching morphogenesis to be observed in action. He also describes how similar strategies have been applied to study branching of the ureteric bud of mammalian kidneys and development of vasculature in zebrafish, with the unexpected finding that branching systems are extensively remodelled. High-resolution studies have also begun to reveal, for the first time, details of cellular activity, such as the visualisation of fine processes put out by cells to sample their environment. Previously, it had only been possible to observe these details in transparent embryos, such as sea urchins, in which very similar exploratory cell behaviour occurs during gastrulation. Of the vast repertoire of cell behaviours, several operations are performed during the formation and subsequent development of the somites in vertebrate embryos. Changes in cell–cell adhesiveness and cell affinities first create borders between consecutive blocks of cells and then generate an epithelial structure; later, cell migration and rearrangements are involved in sclerotome and myotome formation. Kalcheim and Ben-Yair review recent work on the local signals that control these complex changes in cell behaviour, and the molecules that affect them. The picture emerging is that each individual somite consists of a surprising number of discrete, local cell-populations in which cells behave differently and have distinct fates. This harks back to ideas about branching morphogenesis in which, again, it is local differences in cell behaviour that are important. Local differences in cell behaviour also determine plant form. Branching pattern is dictated by the activity of different meristems. In her commentary, Leyser places recent work on auxin transport in an historical perspective and also discusses how this suggests a new mechanism for preventing growth of lateral branches. Xi, Kirilly and Xie review the similarities and differences between somatic stem cells and germ line stem cells, those highly topical cell types that engender both hope and controversy. Some basic lessons are emerging, an important one being the crucial nature of signals from a local cellular environment (the niche) in the maintenance of stem cell identity. Whether there are common intrinsic factors that specify ‘stem cell identity’ — as opposed to them being generally required for cell cycle progression, or maintenance of gene expression states — in different contexts seems much less clear at this time. Much more is known about the intrinsic factors that contribute to stem cell identity in plants. Despite the differences between animal and plant stem cells, there exists a deep knowledge, as reviewed by Vernoux and Benfey, concerning the specification of plant stem cells, and how stem cells operate in an organism that is ever-growing. Current Opinion in Genetics & Development 2005, 15:355–357
Gene regulation Buried within whole genome sequences are the key sequences to which transcription factors bind in order to regulate gene expression. Vavuori and Elgar discuss approaches that have been taken to predict individual transcription factor binding sites in large genomes. Here, comparative studies are essential tools, comparing genome sequences not only between widely different organisms but also between closely related species. The authors point out that although scanning immediate upstream regions in vertebrates can reveal gene regulatory elements it turns out that some of these regions might be a very long way away from the gene that they regulate. This topic is the focus of the article by Zuniga, who reviews three examples of distant regulatory elements that drive specific patterns of gene expression in developing vertebrate limbs. Another characteristic of these ‘long-range’ elements is that they might be shared by genes with unrelated functions. The cautionary tale about how the gene responsible for mouse ‘limb deformity’ was finally tracked down shows how this can lead to confusion. Zuniga also points out, in line with the Vavouri and Elgar article, that an important feature of these regulatory elements is that they are very highly conserved in a range of different vertebrates. In addition, she highlights that, on the basis of analysis of Drosophila genes, many more examples of such ‘long-range’ regulators could yet be discovered in vertebrates. In the past few years, RNA non-coding sequences have attracted increasing interest as a newly discovered mechanism for regulating gene expression. Harfe reviews the rapid progress that has been made in identifying vertebrate microRNAs and in measuring their activity in embryos. There are already indications, based on the effects of genetically manipulating the enzyme required to process microRNAs in zebrafish and in mouse, that microRNAs are essential for proper development; however, as Harfe points out, unpicking the roles of individual microRNAs looks to be a much sterner challenge.
Comparative developmental biology We all begin the journey from egg to ego as a single cell, and Minakhina and Steward review the earliest events in the specification of embryonic axes, which takes place in oocytes. In both Xenopus and Drosophila, the polarity of the oocyte is controlled by motor proteins that move ribonucleoproteins to discrete locations in the cell, and the specificity of this localization is dependent on the sequence of the 30 UTR of messenger RNAs, which are turning out to be much more information-rich than heretofore appreciated. A long-enduring mystery has been the disparity between the functional specificity of Hox proteins in anterior– posterior (AP) patterning and their rather promiscuous DNA binding properties. A partial solution to the mystery www.sciencedirect.com
Editorial overview McGinnis and Tickle 357
is revealed by the Extradenticle/Pbx proteins, which are cofactors that increase the DNA binding specificity of Hox proteins for a subset of their target genes. New evidence reviewed by Mahaffey suggests that a variety of zinc-finger transcription patterns also play a role in increasing the specific output of Hox proteins. Interestingly, members of each of these zinc-finger protein families, the fly homologs of which are called Buttonhead, Disconnected and Teashirt, are expressed in different zones of the AP axis in the embryos of flies and mammals, and seem to constitute an AP-patterning system that operates in parallel to, and in combination with, Hox proteins. Although Hox genes were labeled as some of the original ‘master control genes’, new master genes appear at regular intervals, and one of the best known is the Pax6 gene as a master of eye identity. Kozmik rightly notes that a more accurate label for Pax6 is as an ‘eye selector’, and reviews recent findings on the role of Pax transcription factors in the origin of pigment cells, photosensitive cells and eye organs. One intriguing possibility is that ancient genes related to members of the extant Pax class had roles in the determination of either pigment cell or photoreceptor cell identity. However, the occurrence of gene rearrangements that fused DNA binding domains of different Pax progenitors plausibly contributed to the development of more-complex photosensitive cells, and might have led to further elaborations of eye morphology in the animal kingdom. Although it is commonly accepted that the eye organs of insects and mammals originated in a light-sensitive organ that was specified in part by a Pax gene, just the opposite is true for the appendages of insects and mammals, which are believed to have evolved independently. Could the conventional wisdom be wrong? Pueyo and Couso review the many amazing similarities in gene expression pattern and
function between limb-patterning genes in Drosophila and vertebrates. They conclude that although it is still a heretical notion, and our concept of an appendage might have to be stretched, it is possible that our last common ancestor with flies had appendages in which the axial coordinates were specified by transcription factors from the Distalless, Dachshund and Homothorax families. Still in the realm of heretical notions, could Drosophila and Caenorhabditis elegans possibly be model organisms for dermatological research? Jane, Ting and Cunningham review the recent, surprising evidence suggesting that the development and repair of the epidermal barrier in insects, nematodes and mammals is controlled by transcription factors from the Grainyhead family. In part, this appears to be owing to a conserved role of Grainyhead proteins in activating the transcription of cross-linking enzymes that help to seal the epidermal barrier from a harsh world. As evidenced above, the pendulum has swung far in assuming that the structural conservation of transcription factors indicates that biological function will also be conserved, but Irish and Litt point out that this assumption is made at one’s peril. Genes that encode MADS transcription factors have duplicated and diverged into large families during the evolution of plants; surprisingly, some structural homologs within this family perform different developmental roles, whereas paralogs perform equivalent roles in different plant lineages. Nearly all are fascinated by their genealogical origins, and Gerhart, Lowe and Kirschner review recent findings on the development of the hemichordates, which represent some of our most ancient deuterostome relatives. The study of these animals promises to reveal the origins of notochords, organized dorsal nervous systems, and many other features possessed by chordates.
Current Opinion in Genetics & Development 2005, 15:355–357