Imaginal Discs (Insects)

tmp43-10_thumb

The term imaginal disc is used to describe structures found in the larvae of the Holometabola. Holometabolous insects JL_ can be defined as those in which the final larval instar metamorphoses into a radically different adult during a quiescent pupal stage; they are thought to be a monophyletic group, distinct from the Hemimetabola. During the metamorphosis of Holometabola, the epidermis must form novel structures that were lacking in the larva, and in some cases, replace larval tissues that were lost. The cells that give rise to the new epidermal tissues of the adult imago are often referred to as histoblasts. When histoblasts are organized into morphologically distinct clusters, these structures are commonly referred to as imaginal discs or imaginal buds. This article discusses this definition, and briefly reviews some of the experimental studies examining the biology and, especially, the development of imaginal discs.

WHAT IS AN IMAGINAL DISC?

Usage of the terms imaginal disc and histoblast varies from author to author. This is because the number, arrangement, and development of the cells that give rise to adult structures during metamorphosis vary considerably between taxa, and even between different regions in a single larva. For instance, a few authors only use the terms imaginal disc and histoblast cell for those taxa where imaginal discs are formed in the embryo, several molts before metamorphosis. These early-developing discs may in some cases secrete a thin cuticle-like substance but they do not contribute appreciably to larval life, and thus can be considered as specialized, relatively undifferentiated structures set aside during embryogenesis for adult development. However, most authors have also used the terms for cells and structures that cannot be detected until the final larval instar and that are apparently derived from cuticle-secreting, differentiated larval cells. It is likely that late-developing imaginal discs are the more ancestral condition and that early-developing discs are the more derived condition within the Holometabola.
The morphology of imaginal discs also varies. The prototypical imaginal disc is a pocket or sack of cells that has invaginated from the larval epidermis, and is destined to form part or all of an adult appendage, compound eye, or genitalia. This sack evaginates (“everts”) during metamorphosis and contributes to the adult cuticle (Fig. 1). One
Eversion of D. melanogaster leg disc, shown in cross section. After Blair (1999).
FIGURE 1 Eversion of D. melanogaster leg disc, shown in cross section. After Blair (1999).
portion of the sack has thickened to form the disc epithelium, whereas the rest forms a thinner peripodial membrane; the space within the sack is termed the peripodial cavity (Fig. 1). However, the positions of the disc epithelium and peripodial regions vary, as does the disc’s degree of invagination from the larval epidermis, and some authors have defined several categories of discs or quasi-discs. For instance, some discs have invaginated a large distance from the larval epidermis, and remain connected to it by only a thin peripodial stalk. At the other extreme are those cases, as in the wings of some Coleoptera, where the regions of epidermis that form adult structures never invaginate from the larval epidermis, and can only be recognized as thickenings of the larval epidermis.
There is also variation in the development of different appendages. Becasue larvae do not have external wings, these must be formed during metamorphosis; the wing precursors usually appear prior to this and are often organized into disc-like structures in the larva or embryo. However, many holometabolous taxa have larval legs and antennae that were formed at embryonic stages. This likely represents the more ancestral state, as it resembles the non-metamorphic growth of appendages in hemimetabolous insects; legless larvae evolved several times in different homometabolous taxa. There is a great deal of variation in how larval appendages grow and change during metamorphosis to form the larger, more complex appendages of the adult. At one extreme, as in some Coleoptera, most or all the cells of the larval leg divide to give rise to the adult leg. At the other extreme, as in some Lepidopteran antennae, the dividing cells are organized into an invagi-nated imaginal disc. Between these extremes are cases, as in some Lepidopteran legs, in which each adult appendage is derived from several zones of dividing cells. As these zones are not disc-like in structure, more neutral terms like “adult primordia” are often used. The remaining non-dividing cells of the larval appendage can either die or persist to give rise to small regions of the adult appendage.
Thus, the organization of the cells fated to form adult structures into a morphologically identifiable imaginal disc is simply one of several ways these cells can be arranged. And although the term imaginal disc is used only for holometabolous insects, it is only the invaginated morphology of most discs that distinguishes them from structures like the external wing pads of hemimetabolous insects, which also are morphologically distinct and form adult wings during the final molt. Thus, the evolution of invaginated wing imaginal discs from the evaginated wing pad may have required only a morphological change, rather than the evolution of a radically different cell type or organ. In fact, a similar change has apparently evolved in some Hemimetabola, such as thrips, which have invaginated wing precursors instead of external wing pads.


EXPERIMENTAL STUDIES ON IMAGINAL DISCS

The biology of imaginal discs has been described in a number of species, but has only been experimentally analyzed in a few. As with other metamorphosing organs, there have been studies on the endocrine control of imaginal disc eversion, differentiation and cuticle secretion. Other types of experiments, such as ablation or transplantation of portions of imaginal discs, have been used to study aspects of developmental patterning, such as the development of pigmentation patterns in the wings of butterflies and moths. And in a sense, all studies of the evolution of adult Holometabola reflect on the histo-blast cells and discs from which portions of those adults are formed.
However, far and away the most extensively studied imaginal discs are those of the fruit fly, Drosophila melanogaster. This species has been the subject of intense genetic studies for the last century, and has become a model for the study of a variety of biological problems. Gaining an understanding of the imaginal discs in this species has been particularly critical because the entire adult epidermis is derived from imaginal discs or disc-like structures. Thus, any mutation that alters the external form of the adult D. melanogaster does so by altering the development of these imaginal structures. This has led to a large literature on the genetics, development, and cell biology of these structures and, via comparisons with other Drosophilids and Dipterans, to a number of evolutionary studies.
It should be noted that the imaginal discs of D. melanogaster are at one extreme in the spectrum of imaginal tissues, and it is not clear to what extent the mechanisms underlying this apparently derived state are shared by other Holometabola. As in other Cyclorrhaphan Diptera, the metamorphosis of D. melanogaster is unusual in several respects. First, the cuticle of last larval instar is not shed during the formation of the pupa (pupariation), as it is in most Holometabola. Rather, the larval cuticle is retained and converted into an outer covering, the pupal case. Second, in D. melanogaster the non-disc cells of the larval epidermis are polyploid (containing more than the dip-loid number of chromosomes), and these die during the early stages of metamorphosis. In many other Holometabola, large portions of the larval epidermis and appendages are retained in the adult. Finally, in D. melanogaster, the imaginal disc primordia are formed during embryonic development, rather than during the last larval instar as they are in some other Holometabola.
In D. melanogaster embryos, each imaginal disc primordium contains 10-40 cells, which divide during the three larval instars to form, by pupation, as many as 50,000 cells by late third instar. The arrangement of imaginal discs in the late third instar larva is shown in Fig. 2. The adult head is derived from a pair of fused eye-antennal discs, as well as pairs of proboscis (labial) and labral (cibarial or clypeo-labra) discs. The adult thorax is derived from the three pairs
Imaginal discs in late third instar D. melanogaster.
FIGURE 2 Imaginal discs in late third instar D. melanogaster.
of ventral leg discs and, dorsally, pairs of prothoracic (humeral), wing, and haltere discs. Each of these makes portions of the body wall as well as the appendages for which they are named. Each segment of the adult abdomen is formed from four pairs of small “histo-blast nests” and the genitals from the genital disc.
At the late third instar, each imaginal disc consists of a simple epithelial sack invaginated from the larval epithelium (Fig. 1). One surface of each sack forms the thickened and folded disc epithelium from which most of the adult structures are derived. The other surface of the sack is the thinner peripodial membrane, and each disc remains connected to the larval epithelium by a long, narrow stalk. During the first few hours of pupariation, each disc everts through the stalk, expands, and eventually sutures together with adjacent discs. The peripodial membrane is lost during this process, and the polyploid cells of the larval epidermis are histolyzed and replaced. The portions of each disc epithelium that are fated to form the appendages unfold and lengthen: the prospective legs and antennae form long tubes, whereas the prospective dorsal and ventral surfaces of the wing lengthen and flatten together. The disc cells then secrete the pupal cuticle, which is separated from the epithelial surface shortly thereafter. After a delay during which further morphogenesis and differentiation occurs, the adult cuticle is secreted.
Many aspects of the biology of imaginal discs of D. melanogaster have been examined over the years, and continue to be the subjects of intense research. These include the hormonal control of disc development and cuticle secretion, the molecular genetics of cell division and tissue growth, and various problems in cell biology. Perhaps the most striking advances, however, have been made in the field of developmental patterning, examining for the most part the large wing, leg, and eye-antennal discs. Space constraints prevent the discussion of many fascinating and important research areas, such
Signaling and anterior boundary cells in wing disc of D. melanogaster
FIGURE 3 Signaling and anterior boundary cells in wing disc of D. melanogaster.
as the patterning of the compound eye, the formation of sensory organs, and the planar polarity of the disc epithelium. Instead, the following discussion briefly reviews some early developmental events that help establish the identities and axes of the discs.

PATTERNING IMAGINAL DISCS IN D. MELANOGASTER

The identity of each disc primordium is specified during embryonic development. This specification relies on the disc-specific expression of a small number of genes, encoding largely transcription factors or their co-factors, and by stimulatory and inhibitory signals from neighboring cells. The localized expression of these genes and signals is inherited from earlier stages during which the embryo was subdivided into anterior-posterior and dorsal-ventral domains. The loss or mis-expression of these genes and signals can block or enlarge disc development, and in some cases cause dramatic “homeotic” transformations of disc identities.
How are different tissues formed within each disc? Even at late third instar most imaginal disc cells appear morphologically similar, as only a few (mostly sensory) elements have begun to differentiate terminally. However, the premetamorphic imaginal disc is in no sense a tabula rasa, as it has already been subdivided into a number of regions. In fact, most disc primordia are subdivided into anterior and posterior lineage “compartments” from the earliest stages of their development. This subdivision is controlled by the posterior-specific expression of the engrailed and invected transcription factors in each disc, which acts as a binary switch, controlling the choice between posterior and anterior identities and preventing cells from crossing between compartments. The wing and haltere discs are further subdivided into prospective dorsal and ventral lineage compartments, in this case by the dorsal-specific expression of the apterous transcription factor.
Why have lineage compartments? Cells in different lineage compartments have different signaling capabilities, and this has important developmental consequences. Posterior cells secrete the signaling molecule Hedgehog, but only anterior cells are capable of responding to that signal (Fig. 3). Because Hedgehog diffuses only a short distance into the anterior compartment it induces the formation of a specialized group of cells just anterior to the compartment boundary. The formation of this group of boundary cells is critical for appendage development, because boundary cells in turn secrete several important signals (Fig. 3) , including Decapentaplegic (a member of the Bone Morphogenetic Protein family of morphogens generated in the wing, dorsal leg, and dorsal antenna) and Wingless (a member of the Wnt family of morphogens generated in ventral leg and ventral antenna). Cells outside the boundary region judge their approximate position
Signaling and dorsal-ventral boundary cells in the wing disc of D. melanogaster.
FIGURE 4 Signaling and dorsal-ventral boundary cells in the wing disc of D. melanogaster.
in the disc by the levels of boundary signal they detect. Thus, these boundaries define important axes used to pattern the entire disc.
The subdivision of the wing into dorsal and ventral compartments by the dorsal expression of apterous plays a very similar role. This establishes reciprocal signaling between dorsal and ventral cells, this time via the Notch signaling pathway, that results in the specification of cells on either side of the dorsal-ventral boundary (Fig. 4). These cells secrete Wingless, which helps pattern the wing blade along the proximo-distal axis.
D. melanogaster imaginal discs are also specified along the proximo-distal axes by characteristic zones of gene expression. While these zones are not compartmental (the gene expression patterns shift during development and do not define cell lineage domains), these zones are nonetheless crucial for the subdivision of the discs into prospective proximal, intermediate, and distal regions.
Do the other Holometabola share these axis-defining patterning mechanisms? Many of the basic axis-defining genes are conserved in other insects, and the expression patterns of these genes in the few wing imaginal discs that have been examined are largely similar to D. melanogaster, even in taxa as distantly related as lepidopter-ans and winged ants. Most of the work on non-wing appendages in the Holometabola has been limited to the larval legs and antenna, rather than histoblast and adult tissues. While expression patterns of the proximo-distal patterning genes are largely conserved; there are some curious differences in the expression of compartmental signals in larval legs. While Engrailed is expressed in the posterior and Wingless just anterior to the compartment boundary, the boundary-specific expression of the signal Decapentaplegic observed in D. melanogaster is not conserved in other Holometabola. Therefore, compartmental BMP signaling, which plays a crucial role in patterning the highly derived imaginal discs of D. melanogaster, may not represent the ancestral state. With the advent of new methods for altering gene expression in taxa not amenable to standard genetic techniques, a growing number of laboratories are experimentally testing the roles played by these patterning genes in a variety of taxa.

Next post:

Previous post: