Pattern formation is the developmental process that turns a bunch of differentiated cells into an organism as opposed to a tumorous lump. For example a teratoma (cancer of the ovary) contains all types of differentiated cells such as teeth, hair, muscle, skin, bone, brain etc. cells. But despite having all the same cells as a baby, it is clearly not a baby. It lacks the developmental pattern that packages cells into a recognizable, reproducible organism. Pattern formation occurs in all organisms, e.g. a plant looks different from a callus, but it is best understood and most intensively studied in the fruit fly, Drosophila melanogaster. Why? Drosophila has been a model genetic organism since the rediscovery of Mendel's laws. This means that there are enormous genetic and molecular resources available for the study of developmental processes in Drosophila. In addition, the eggs are big enough to see and develop externally so all the techniques of classical embryology can also be performed with Drosophila. In the 1980's there was a massive screen to detect mutations in pattern formation genes. Decades of work later, we are coming to realize that the developmental processes, first understood by genetic analysis and then by molecular cloning and analysis of these genes in Drosophila, are common to all organisms from flies to humans to plants. Thus we will start this section with an overview of Drosophila embryogenesis and pattern formation and then move to a discussion of the similar, but less well studied, processes in plants and humans.
A. Oogenesis and embryogenesis in Drosophila
B. Pattern Formation in Drosophila
C. Techniques for studying pattern formation
D. Axis formation - a specific example
E. Pattern formation and flowering in plants
F. Pattern formation in humans
A. Oogenesis and embryogenesis in Drosophila
Oogenesis: As with most female germ cells, there are support cells. In Drosophila there are two types of somatic support cells, nurse cells and follicle cells. The support cells function to deposit nutrients into the egg to support embryo growth until it becomes self-supporting. In addition, both types of support cells are important in the early events of pattern formation. They are involved in depositing and localizing axis determinants that transmit epigenetic information on the orientation of the maternal axis to the embryo.
Embryology: After fertilization, the zygotic nucleus divides rapidly many times, but without accompanying cell division, to generate a synticial cleavage embryo. These nuclei then migrate from the center of the egg to the periphery to form a syncitial blasula. The nuclei then cellularize to form the cellular blastula. Gastrulation and neurulation follow. In addition to formation of the gut cavity and brain, one of the major events in early development is segmentation. Segmentation also occurs in other animals such as chickens, amphibians, mammals but is less obvious in the adult animal. In the first stage of segmentation, the embryo forms parasegments which are two segments wide and slightly offset from the final segments. These parasegments then subdivide into single segments.
B. Pattern Formation in Drosophila
The body pattern is formed by a cascade of pattern formation genes. The general strategy is conserved in all organisms-plants and animals - but best understood in Drosophila.
There are about 20 genes which act in the mother to deposit axis information into the embryo. These genes fall into three independent groups; those that specify the anterior, those that specify the posterior and those which specify the two termini. The morphogens encoded by these genes are deposited as either mRNA (anterior and posterior) or protein (termini). These cytoplasmic determinants are deposited in a spatially restricted manner in the oocyte and thus generate a morphogen gradient. The products of these maternal effect genes are transcription factors (TFs). As the maternal effect mRNAs and proteins are localized, the TF they encode are localized so these TFs only activate genes in certain portions of the embryo. Thus they activate genes in a spatially restricted manner. The genes that are activated by the maternal effect genes are a group of genes called the gap genes. These genes also encode TF. The gap gene TFs activate other genes, the segmentation genes. These genes encode TFs, which in turn activate the homeotic genes. The homeotic genes also encode TFs which activate a series of downstream genes (many of which are also TFs). These genes are unique to each segment (portion) of the embryo and thus give each part of the developing embryo its own identity.
C. Techniques for studying pattern formation
The five main techniques for studying pattern formation are as follows:
1. Loss of function phenotype from either mutant analysis or antibody
blocking
2. Pathway analysis from epistatic interactions
3. Enhancer traps and marker genes
4. In situ hybridization - in both wild-type and mutant backgrounds
5. Molecular transcriptional analysis - binding of one gene product to
promoter of another
We will examine how these techniques have aided in the analysis of pattern formation by examining a specific example of axis formation.
D. Axis formation-a specific example
Defining the main axises of the organism is the first patterning event that must happen. It is also among the simplest of the subsequent patterning events. We will discuss how Drosophila makes one of its axises - the anterior.
Making the anterior:
Defining the anterior end of the oocyte, which is initially completely symmetrical, depends on marking the anterior end. This is done in the simplest possible way, a marker product is pumped in at the anterior end of the oocyte from the nurse cells. (How do the nurse cells know which is to be the anterior end?) The "stuff" pumped in at the anterior end - the anterior morphogen - is bicoid mRNA. This mRNA would normally diffuse through the egg but it becomes trapped at the anterior. The substances that trap the biocoid mRNA are the products of the swallow and exuperantia genes. The bicoid mRNA is recognized by a unique 3' tail. The exuperantia gene product attaches the bicoid mRNA to microtubules and the swallow gene product maintains this attachment (somehow).
The end result of the symmetrical deposition and selective retention of biocoid mRNA is a gradient of bicoid mRNA in the embryo. When the bicoid mRNA is translated this results in a gradient of bicoid protein with the highest concentration at the anterior end and the lowest at the posterior. The bicoid protein is a transcription factor. A variety of gap genes have binding sites for the bicoid transcription factor in their promotors. Depending on the number of binding sites for the bicoid transcription factor, different gap genes will have different sensitivities to the bicoid morphogen and will be activated at different bicoid concentrations. Because the bicoid protein is expressed in a gradient along the egg, this means that different positions in the egg and embryo will have different combinations of gap genes expressed. These different combinations of gap genes then set up different regions of the embryo which are defined by differing subsets of expressed genes.
Based on the above information you should be able to answer the following questions using examples of the different types of developmental techniques mentioned above.
Questions:
What would be the phenotype of bicoid null mutants?
What would eggs from mothers with null mutations in swallow and
exuperantia look like?
What would a swallow + bicoid double mutant look like?
How could you determine if the biocoid product deposited in the egg as
mRNA or protein?
How could you test the hypothesis that bicoid mRNA is the true anterior
morphogen?
How could you test if swallow and exuperantia genes are involved?
How could you determine if it was the bicoid 3' tail that was responsible
for anterior localization?
How could you determine if the swallow and exuperantia gene products were
localized in the anterior end? Would you expect them to be?
How could you tell if bicoid acts as a transcription factor for other
genes such as tailess?
The above description of anterior formation is simplified but sufficient to demonstrate the basic principal of gene cascades, differentiation as a result of differential gene expression and the ways in which developmental processes can be studied. If time permits we may also discuss:
Terminal Genes
Posterior Formation
Dorsal/Ventral formation
Cell ligand/receptor interactions
Plant Pattern Formation
Pattern formation in plants is less well studied than in Drosophila, but the truly surprising discovery that has emerged is that, despite the very different body organization of a fruit fly vs. a plant, the embryonic body plan is organized on the same principals.
Pattern formation in plants has been largely studied in one type of plant; Arabodopsis thaliana, the wall cress. Arabodopsis is a member of the mustard family and a typical dicot angiosperm. It has many advantages for developmental and genetic studies. It is physically small so it can be grow in (large) test tubes. It is fast growing (for a plant) with a 2 month generation time and it has a small genome which has led to its inclusions in the genome projects.
Body plan formation in Arabidopsis follows the same principles as in Drosophila. Plants are basically radially symmetrical so their primary axis is the apical-basal axis defined by the SAM and RAM (basically equivalent to anterior-posterior axis in animals). Thus plant embryos face the same problem as animal embryos - they must determine which end should form the SAM and which should form the RAM. As in animals, the initial axis determination is dependent on maternally supplied morphogens, in this case, auxins and cytokinins. These are expressed in a concentration gradient across the developing embryo. How this gradient is established is not understood although the endosperm seems to play a role. These hormones appear to act as transcriptional regulators (presumably coupled with a protein) to activate and/or repress other genes. The target genes of auxin and cytokinin are presumably gap genes. At least some of these gap genes have been identified and encode transcription factors. Presumably these transcription factors activate the radial and homeotic genes (plants do not have segmentation genes or their equivalent) to generate the final body pattern. Although the similarity between body pattern formation in animals and plants provides many clues, early pattern formation is difficult to study in plants as the embryo and zygote are very small and surrounded by vaste amounts of maternal tissue (endosperm, integuments, seed, etc.). Most experiments on plant embryology avoid this problem by doing in vitro experiments on calluses, making the assumption that a callus is equivalent to an embryo.
Questions:
How could you determine if the endosperm sets up the auxin/cytokinin
concentration gradients?
How could you determine if auxin and/or cytokinin activate genes?
How could you determine which genes are activated by auxin and/or
cytokinin?
How could you identify the protein(s) which are coupled to auxin/cytokinin
to act as transcriptional regulators?
How could you determine if gap genes control the homeotic genes directly
or act through other transcription factors as in animals?
Axis Mutants
In 1991, a giant mutagenesis screen was initiated with the aim of mutating every gene involved in axis formation, at least once (i.e. getting at least one allele of each gene). 44,000 plants were mutated, grown and self crossed to generate homozygous seedlings. These seedlings were then screened for abnormal phenotypes. This screen identified; 25,000 embryonic lethals, 5,000 seedlings with warped cotyledons and/or roots and 250 gap mutants. Complementation between these 250 gap mutations revealed approximately 30 complementation groups, hence different gap genes, as compared with the roughly 100 gap genes in Drosophila.
Gap Genes
The mutant phenotypes of the gap genes fall into three groups; those that delete the apical portion of the embryo (the apical group), those that delete the basal end (the basal group), those that delete the central portion (the central group) and those that delete both the apical and basal ends (the terminal group). Researchers are now attempting to identify the number of pathways used to make the embryonic axis by making double mutants, localizing of the areas of gap gene expression by in situ hybridization and determining the molecular nature of the gene by cloning and sequencing each gene.
Radial Pattern Mutants
In addition to the gap mutations there are also radial pattern mutants such as knolle which has no protoderm (hence no epidermis) and Kreule which has bloated epidermal cells.
Homeotic Genes
As in Drosophila and mammals, plants also have homeotic genes. The best studied are those which transform flower parts because these mutations are not lethal to the plant and many have been isolated by amateur botanists and horticulturists for their unusual flowers. For example, the gene called superman transforms the pistil into additional stamens. Other homeotic genes transform petals into carpels/stamens, etc. Unlike animal homeotic genes, the plant homeotic genes don't have homeoboxes and are not clustered.
Flowering Genes
A new type of gene that has been more recently isolated controls the switch from a SAM into a FLAM. Little is known about this type of gene but as it regulates a profound developmental switch of and embryonic cell type, it is certain to be interesting.