InteractiveFly: GeneBrief

Kruppel homolog 1: Biological Overview | References

Juvenile hormone (JH) given at pupariation inhibits bristle formation and causes pupal cuticle formation in the abdomen of Drosophila due to its prolongation of expression of the transcription factor Broad (BR). In a microarray analysis of JH-induced gene expression in abdominal integument, it was found that Krüppel homolog 1 (Kr-h1) was up-regulated during most of adult development. Quantitative real-time PCR analyses showed that Kr-h1 up-regulation begins at 10 h after puparium formation (APF), and Kr-h1 up-regulation occurs in imaginal epidermal cells, persisting larval muscles, and larval oenocytes. Ectopic expression of Kr-h1 in abdominal epidermis using T155-Gal4 to drive UAS-Kr-h1 results in missing or short bristles in the dorsal midline. This phenotype was similar to that seen after a low dose of JH or after misexpression of br between 21 and 30 h APF. Ectopic expression of Kr-h1 prolonges the expression of BR protein in the pleura and the dorsal tergite. No Kr-h1 was seen after misexpression of br. Thus, Kr-h1 mediates some of the JH signaling in the adult abdominal epidermis and is upstream of br in this pathway. It was also show that the JH-mediated maintenance of br expression in this epidermis is patterned and that JH delays the fusion of the imaginal cells and the disappearance of Dpp in the dorsal midline (Minakuchi, 2008).

In the Coleoptera and Lepidoptera, the epidermal cells make larval, pupal, and adult cuticles sequentially. By contrast, in higher Diptera such as Drosophila melanogaster, the adult epidermis on the head and the thorax is derived from imaginal discs, and the adult epidermis on the abdomen is formed by imaginal cells derived from the histoblast nests that make larval cuticle, but do not divide during larval life. After puparium formation, the histoblasts begin to proliferate rapidly, displacing the larval epidermal cells that subsequently die after making the pupal cuticle. This process is complete by 40 h after puparium formation (APF). Juvenile hormone (JH) application to D. melanogaster during the final larval instar or during the prepupal stage has little effect on the adult differentiation of the head and the thoracic epidermis, but it prevents the normal adult differentiation of the abdominal epidermis that is derived from the histoblasts. After JH treatment, the histoblasts continue to divide to form the imaginal epidermis, but the normal outgrowth of abdominal bristles is prevented, and a second pupal, rather than adult, cuticle is formed (Minakuchi, 2008).

In insects ecdysteroids trigger molting, while JH determines the nature of the molt. When JH is present, the ecdysteroid-induced molt is to another like stage; whereas in its absence, metamorphosis ensues. Little is known about how the JH signal is mediated in preventing insect metamorphosis. The broad (br) gene, an ecdysone-induced transcription factor in the Broad-Tramtrack-Bric-a-brac (BTB) family, is the key regulator of the onset of metamorphosis, since amorphic D. melanogaster mutants of br (npr) can develop normally until the final larval instar but fail to begin metamorphosis. In the silkworm Bombyx mori, RNAi knock-down of br in imaginal discs and primordia resulted in their failure to undergo metamorphosis properly. It has been shown that the treatment of either D. melanogaster or M. sexta with a JH mimic (JHM) at the onset of adult development induces the re-expression of the br gene in the abdominal epidermis and that misexpression of br during the adult development of D. melanogaster results in the truncation of bristles and the formation of pupal cuticle by the imaginal cells, both of the abdomen and of the disc-derived head and thorax. In hemimetabolous insects such as the milkweed bug Oncopeltus fasciatus, br is expressed during embryonic development and each nymphal molt, then disappears at the molt to the adult. In this animal, JH is necessary to maintain br expression during the nymphal stages. Clearly, br can be regulated by JH. Yet little is known about the genes that are either upstream or downstream of br in the JH signaling pathway (Minakuchi, 2008).

Therefore a genome-wide analysis of JH-regulated genes was performed in the abdominal integument of D. melanogaster to which pyriproxyfen, a JHM, had been applied at the time of pupariation to suppress the adult differentiation of abdominal histoblasts. One of the up-regulated genes was Krüppel-homolog 1 (Kr-h1, CG9167). This study shows that the misexpression of Kr-h1 in the epidermal cells results in missing or short bristles in the dorsal midline of the adult fly, a phenotype similar to that seen after treating wild-type animals with a low dose of JH and to that seen after br was misexpressed early in adult development. This action of KR-H1 was found to be accompanied by the prolongation of BR expression in the abdominal epidermis, indicating that Kr-h1 is a key regulator functioning upstream of br in the JH signaling pathway. It was also found that JHM treatment delayed the development of the abdominal epidermis, thus altering the timing of Dpp expression in the dorsal midline (Minakuchi, 2008).

This study has identified Kr-h1 as one of the genes up-regulated by JHM treatment of Drosophila at pupariation that then persists during the entire pupal-adult developmental period. Moreover, the presence of KR-H1 during early adult development can induce the abnormal re-expression of br in the abdomen that results in the formation of a second pupal cuticle (Zhou, 2002; Minakuchi, 2008).

KR-H1 is a zinc-finger type transcription factor with three putative isoforms with different N-terminal sequences (Pecasse, 2000). There are two main isoforms with the β isoform being expressed mainly during nervous system development in the embryo (Beck, 2004). Normally Kr-h1α is expressed at low levels in midembryogenesis, at high levels during larval life, then declines rapidly after pupariation (Pecasse, 2000) and is not expressed again until just before adult eclosion (Beck and Richards, personal communication to Minakuchi, 2008). KR-H1α appears to be necessary for metamorphosis since most of the mutants lacking Kr-h1α function die at the time of head eversion to the pupa (Pecasse, 2000) or shortly thereafter (Minakuchi, 2008).

In insects ecdysteroids cause the molt and JH is present during larval life to ensure that the molt is to another larval stage by preventing the developmental program-switching action of ecdysteroids necessary for metamorphosis. In most holometabolous insects where the epidermal cells are polymorphic so that they produce sequentially larval, then pupal, then adult cuticles, this switching occurs in the final larval instar when the JH titer declines and ecdysone appears in the absence of JH. By contrast, in the highly derived Drosophila, the onset of metamorphosis triggered by ecdysone in the absence of JH results in the death of most of the larval tissues and the development of the pupa and subsequent adult from the imaginal discs. One exception is the larval abdominal epidermis which switches from production of larval cuticle to that of pupal cuticle. The subsequent adult cuticle is then made by imaginal cells derived from the abdominal histoblasts that begin proliferation shortly after pupariation. Importantly, in Drosophila JH cannot prevent the metamorphosis of the imaginal discs or the proliferation of the histoblasts but can delay the onset of metamorphosis; it also causes the formation of a pupal rather than an adult cuticle by the new imaginal cells of the abdomen (Minakuchi, 2008).

Kr-h1α is regulated at least in part by 20-hydroxyecdysone (20E) and in turn regulates the ecdysone-regulated processes (Pecasse, 2000; Beckstead, 2005). It shows a dynamic pattern of binding to certain ecdysteroid-regulated chromosomal sites during the 20E-induced cell death of the salivary glands at metamorphosis (Beck, 2005). In mutants that lack Kr-h1α function, the normal ecdysteroid cascade of transcription factors at pupariation is disrupted with some appearing precociously and others being delayed or reduced in amount (Pecasse, 2000). The result of this misregulation is retention of the salivary glands and death around the time of head eversion that signals completion of pupal development. Interestingly, although overexpression of Kr-h1 suppressed the initial morphogenesis of mushroom body neurons, its loss caused no detectable defects in neuronal morphogenesis, but rather affected the patterning of EcR-B1 expression in the central nervous system at the onset of metamorphosis (Shi, 2007). Thus, KR-H1 is clearly necessary for the proper coordination of the ecdysone response (Minakuchi, 2008).

Interestingly, the few escapers among the Kr-h1α mutants formed cryptocephalic pupae that developed into adults with pigmented eyes and wings but no adult abdominal differentiation beyond the proximal segments (Pecasse, 2000). Thus, they resemble pharate adults formed after treatment with JH at the time of pupariation. In these JH-treated animals, it was found in this study that Kr-h1 is up-regulated in imaginal abdominal epidermal cells, derivatives of imaginal discs (wing, leg, and eye), persisting larval muscles necessary for eclosion and wing-spreading behavior, and in larval oenocytes during the ecdysteroid rises for pupal head eversion and adult development. Yet the adult head and thorax appeared grossly normal after JH treatment. Likewise, there was no significant difference in the number of persisting larval muscles in JH-treated pupae, but the normal differentiation and attachment of adult muscles and the outgrowth of abdominal bristles were either inhibited or delayed at 45 h APF has been seen for the thoracic muscles and the abdominal bristles. The larval oenocytes are involved in lipid metabolism during growth and larval development, then persist through much of adult development where they appear critical for normal utilization of the stored lipid. In other insects such as Tenebrio, the oenocytes have been shown to produce ecdysteroids. Whether the presence of KR-H1 in these oenocytes during this latter period is responsible for any of the defects seen in JH-treated animals is unknown (Minakuchi, 2008).

The role of JH in the normal developmental expression of Kr-h1α in the Drosophila larva is not known. In the red flour beetle, Tribolium castaneum, the transcript level of Kr-h1 is high during larval life, decreases at the end of the final larval stage and disappears just before pupation, then remains very low during the pupal stage and the ensuing adult development just as in Drosophila (Pecasse, 2000). Importantly, RNAi-mediated knockdown of Kr-h1 in young (pre-final instar) Tribolium larvae resulted in precocious metamorphosis, indicating that Kr-h1 is necessary for mediating JH signals in normal larvae. The finding that Kr-h1α reappears abnormally in the abdomen of pupae that were treated with JH at pupariation suggests that its appearance is the normal larval response to ecdysone in the presence of JH. Clearly further work is necessary to work out the details of the normal hormonal regulation of Kr-h1 in the Drosophila larva (Minakuchi, 2008).

In the brain of the honeybee, Apis mellifera, a homolog of Kr-h1 was identified as one of the genes down-regulated by queen mandibular pheromone (Grozinger, 2003) and up-regulated during the transition to foraging behavior of the adult, which is initiated by JH (Grozinger, 2006). Whether JH directly or indirectly controls the transcription of this gene has not been determined (Minakuchi, 2008).

In Drosophila at the onset of metamorphosis, 20E induces a cascade of transcription factors including the different isoforms of br, E74, and E75 that serve to regulate tissue-specific genes involved in metamorphosis. For most tissues, this is the first appearance of Broad (BR), a BTB-domain containing transcription factor, which is necessary for metamorphosis. At this time, br is apparently regulated by KR-H1 since this protein has been localized to the 2B5 br gene site on the salivary gland chromosomes (Beck, 2005). In the abdominal epidermis, br specifies pupal cuticle formation, whether the cells are initially larval or whether the cells are the imaginal cells derived from the histoblasts. This study found that misexpression of Kr-h1α during adult development caused the re-expression of br in the imaginal epidermis of both the pleura and the dorsal abdominal tergites, but that misexpression of br during normal adult development did not lead to Kr-h1 misexpression. These data strongly suggest that the JH-induced KR-H1α is acting somehow to prevent the permanent cessation of br expression in at least some of the imaginal abdominal epidermal cells during the onset of adult development. The nature of this action is not understood including whether or not it acts directly or indirectly on br transcription (Minakuchi, 2008).

Importantly, this interaction of Kr-h1α and br is not essential for the normal expression of BR during the late third larval instar since in Kr-h1α mutants, BR is present at the time of wandering as in wild-type larva. Whether the normal time course of br activation and expression occurs in these mutants has yet to be studied (Minakuchi, 2008).

The ectopic expression of Kr-h1α in the abdominal epidermis resulted in missing or short bristles in the dorsal midline. A similar phenotype was observed after low JHM was applied at pupariation, or after misexpression of br during early adult development between 21 and 30 h APF. The abdominal epidermis around the dorsal midline is the most sensitive to JHM treatment. As the dose of JHM is increased, more bristles on the tergite are affected. In wild-type Canton S, 100 ng of pyriproxyfen prevents the outgrowth of abdominal bristles, resulting in a bald abdomen with few or no bristles. When JHM was applied to JH-resistant lines such as the methoprene-tolerant mutant, the outgrowth of the majority of abdominal bristles was not completely blocked except for the bristles in the dorsal midline (Minakuchi, 2008).

By misexpressing Kr-h1 with the Gal4 driver T155 that expresses in the abdominal histoblasts and the derivative imaginal epidermal cells, it was not possible to mimic the complete loss of bristles as seen with the higher dose of JHM. Whether this is due to the strength of the T155 driver or whether this indicates the involvement of other pathways in the action of JH is unknown. The fact that BR persists longer in the dorsal midline cells of the tergite under control of KR-H1 driven by T155 than it does in the dorsolateral cells suggests that patterning elements may also be affected by JH. Zhou (2002) has showed that misexpression of the various isoforms of br between 30 and 39 h caused a truncated bristle phenotype, at early times on the head and thorax and at later times on the abdomen. Misexpression of BR-Z1 between 44 and 60 h has little effect on bristle formation but causes the formation of pupal cuticle. The present study has shown that the misexpression of br at even earlier times in developing adults (between 21 and 30 h APF) caused the loss of bristles and hairs in the dorsal midline of the tergite. This effect is mimicked by the ectopic expression of Kr-h1α in the larval and imaginal epidermis that in turn causes the prolongation of BR protein in the dorsal midline until at least 70 h (at 29°C). Presumably this BR misexpression prevented hair formation, bristle outgrowth and normal adult cuticle formation. Whether this effect of Kr-h1α on br expression is constrained to the dorsal midline by Dpp signaling or is modulated by other patterning genes is unclear (Minakuchi, 2008).

Interestingly, the pattern of BR expression in Kr-h1α-containing cells is different from the JH-induced BR pattern. In the Kr-h1α-directed BR expression, there appear to be more BR-expressing cells near the mid-dorsal line and fewer such cells laterally. Also, in the a5 region, more of the Kr-h1α-directed cells express BR than in the JH-treated animals, whereas fewer of those in a3 and a4 express BR. Thus, the KR-H1α-directed BR pattern looks like a triangle with most of the cells at the dorsal midline and very few cells in the lateral area expressing BR. Possibly this patchy BR staining in the anterior tergite directed by Kr-h1α is dependent on other local patterning elements that are not disturbed by JH. Further study of this difference is warranted (Minakuchi, 2008).

The histoblast nests start rapid mitosis after puparium formation, and the cell division continues until the imaginal cells from several nests migrate and fuse. After the imaginal cells replace the larval epidermal cells, they start the formation of abdominal bristles followed by the secretion of adult cuticle. Using the enhancer trap line (dpp-lacZ), it has been shown that Dpp expression is confined to the pleura and the dorsal midline in the posterior edge of the anterior compartment at 45 h APF. This study also observed a similar pattern of Dpp expression. Importantly, the spatial patterns of Dpp expression were very similar between control and JHM-treated animals, but the expression of Dpp and the differentiation of the abdominal epidermis were delayed in JHM-treated animals. Zhou (2002) has reported that the proliferation of imaginal epidermal cells on the abdomen occurs normally after JHM treatment at puparium formation such that there os no apparent difference in the size of the imaginal nest between JHM-treated and control animals at 18 and 30 h APF. This study investigated the migration of imaginal cells between 29 and 45 h APF in detail; it was found that JHM treatment at puparium formation delayed the migration of imaginal epidermal cells and/or the death of the larval epidermal cells during this period. Since death of the larval cells is tightly linked to the migration of the adult cells, one may not be able to separate the effects on the two. Whether JH delays the earlier rapid proliferative events needs to be restudied. These results indicate that JHM treatment disrupts the coordination of events in development of the adult abdominal epidermis, including the timing of Dpp expression, which is likely a consequence of the delayed development. Whether the JH-induced re-expression of br in the abdominal epidermis mediates this effect on dpp expression or whether there is a direct effect of JH on dpp expression is unknown (Minakuchi, 2008).

These studies have shown for the first time that the JH-mediated maintenance of br expression is patterned in the developing adult dorsal abdominal epidermis (the ventral epidermal cells were not investigated). The cells in the a1 region, the anterior-most of the segment, do not express BR in response to JHM. These cells, under normal conditions, show high levels of hedgehog (hh), and do not receive wingless (wg) and optomotor-blind (omb) signals. The lack of BR expression in these cells may be due to the lack of Wg signal. Wg signaling has been shown to be required for BR expression in the follicle cells of the dorsal appendage primordia of egg chambers. However, the loss of br expression in the cells in the p2 region is not likely to be caused by lack of Wg signal, because Wg is present at higher levels in the p2 region than in the p1 region where BR is expressed. Thus, various patterning elements apparently are also involved in the regulation of the JHM-induced br re-expression during adult development (Minakuchi, 2008).

Roles of Drosophila Kruppel-homolog 1 in neuronal morphogenesis

The molecular mechanisms underlying remodeling of neural networks remain largely unknown. In Drosophila, widespread neural remodeling occurs during metamorphosis, and is regulated by ecdysone. Kruppel-homolog 1 (Kr-h1) is a zinc finger transcription factor known to play a role in orchestrating ecdysone-regulated transcriptional pathways and, furthermore, implicated in governing axon morphogenesis. Interestingly, in honey bee workers, neural expression of the Apis mellifera homolog of Kr-h1 is enhanced during their transition to foraging behavior when there is increased neurite outgrowth, branching, and synapse formation. This study assessed the role(s) of KR-H1 in Drosophila neuronal remodeling and morphology by characterizing the effect of Kr-h1 expression on neuronal morphology through Drosophila larval, pupal, and adult stages. Increased expression of Kr-h1 led to reduced branching in individual neurons and gross morphological changes in the mushroom bodies (MBs), while knocking down Kr-h1 did not produce any obvious changes in neural morphology. Drosophila Kr-h1 is normally expressed when MB neurons do not undergo active morphogenesis, suggesting that it may play a role in inhibiting morphogenesis. Further, loss of endogenous KR-H1 enhanced the neuronal morphogenesis that is otherwise delayed due to defective TGF-beta signaling. However, loss of KR-H1 alone did not affect neuronal morphogenesis. In addition, Kr-h1 expression remains strongly linked to ecdysone-regulated pathways: Kr-h1 expression is regulated by Usp, which dimerizes to the Ecdysone receptor, and Kr-h1 expression is essential for proper patterning of the ecdysone receptor isoforms in the late larval central nervous system. Thus, although KR-H1 has a potential for modulating neuronal morphogenesis, it appears physiologically involved in coordinating general ecdysone signaling (Shi, 2007)

Krüppel-homolog is essential for the coordination of regulatory gene hierarchies in early Drosophila development

Drosophila development is marked by two major morphogenetic processes: embryogenesis and metamorphosis. While insect metamorphosis is known to be controlled by the steroid hormone ecdysone, relatively little is known concerning the hormonal control of embryogenesis. This study shows that many ecdysone-regulated transcripts of metamorphosis are also expressed in a wavelike manner during embryogenesis, suggesting that these genes also participate in an embryonic ecdysone response. At metamorphosis, the Krüppel-homolog (Kr-h) gene, coding for a zinc finger protein, is required during the prepupal ecdysone response. Kr-h mutants die at the prepupal-pupal transition. In these mutants, the expression of several ecdysone-regulated genes is disrupted, and it is concluded that Kr-h is a key modulator of the hormonal response. While Kr-h is expressed in many tissues at metamorphosis, in embryos expression is restricted to neurons. This study investigated its role during early Drosophila development using new alleles with an earlier lethality than those previously described. Although only minor morphological defects are in these mutants, that Kr-h expression is necessary for the early development of Drosophila, and, during metamorphosis, Kr-h acts as a modulator of the expression of many of these ecdysone-regulated genes (Beck, 2004).

We used DNA probes specific to Kr-hα and β were used for in situ hybridization to visualize tissues in which Kr-h transcripts are expressed. Kr-hβ, the major embryonic transcript, was detected from stage 12 of embryogenesis, around 8 h after egg laying, confirming the RT-PCR observations. Kr-hβ is located in the brain, the ventral nerve chord, and in some cells located laterally in a pattern suggesting a connection with the PNS. Kr-hβ RNA is found in these same areas until stage 17, at the end of the embryonic development. A similar pattern of expression of Kr-h transcripts was reported by Brody 2002) in late embryos. In contrast, no Kr-hα transcripts were detected in situ, perhaps because of the low expression level of this isoform (Beck, 2004).

To follow the KR-H protein, a polyclonal antibody directed against the KR-Hα and β common C terminal region was generated. The observations confirm the in situ hybridization results. KR-H is observed as early as stage 12 in the same tissues as those where Kr-hβ transcripts were observed. To determine the cell type expressing KR-H, a monoclonal antibody recognizing Elav, a protein specifically expressed in neurons, was used. KR-H is detected in all Elav-expressing cells: in the brain, the ventral nerve chord, and in the PNS, suggesting that KR-H is also specifically expressed in neurons. To confirm that Kr-h is associated with neuronal identity, its distribution was examined in Sca-gal4, UAS-glide/gcmM24A; UAS-glide/gcmM21G embryos, which overexpress glide/gcm in neuroblasts, thus promoting the glial destiny of these cells and reducing the number of neurons. In these embryos, KR-H was weakly expressed in the few differentiated neurons. Therefore, KR-H is expressed only in those cells having a neuronal identity. In addition, KR-H was detected in a few cells not expressing Elav, notably between the main lobes of the brain. However, Kr-h is not essential for neuronal identity as Elav staining was normal in Df(2L)Kr-h7.1 homozygotes. Attempts to raise polyclonal (rabbit, rat) or monoclonal (mouse) antibodies recognizing the 54 amino acid β-specific N-terminal fragment were unsuccessful (Beck, 2004).

Despite the dramatic changes in the transcripts of the ecdysone hierarchies, no obvious embryonic mutant phenotype was observed. Several putative pathways or structures were detected by immunostaining mutant embryos with antibodies against Dachsund, Neurotactin, Fasciclin II, Fasciclin III, and the BP102 serum, each of which stain different parts of the nervous system. α Spectrin and a serum recognizing trachea were also tested, β3 Tubulin , which is expressed in the visceral and somatic mesoderm, Hindsight expressed in CNS, glial cells, midgut, PNS, and trachea. However, none of these antibodies revealed clear abnormalities. Only when using the 22C10 antibody, directed against the Futsch protein found in axon and neuron cellular bodies, in certain homozygous Df(2L)Kr-h7.1 and heterozygous or homozygous Kr-h^7id embryos, was the presence of one extra neuron observed. These occasional ectopic neurons are not always found in the same segment and do not possess an axon, as only the cellular body is visible (Beck, 2004).

As seen in previous studies (Pecasse, 2000; Schuh, 1986), Kr-hβ is the preponderant embryonic isoform as detected both by transcript and in situ analyses. Unlike Kr-h expression at metamorphosis, where Kr-hα expression is found at similar relative levels in many tissues (Pecasse, 2000), Kr-hβ expression is specific to neuronal cells. It is not however necessary for the determination of the neuronal lineages as seen by Elav staining of homozygous Df(2L)Kr-h7.1 embryos, which present a normal neuronal pattern. Although these studies have not revealed major structural changes during embryogenesis, the occasional ectopic neurons in both mutants are consistent with the suggestion of a role for Kr-h in the elaboration of the nervous system derived from observations of disruptions in misexpression studies. Since KR-H spatial expression is very similar to Elav, it may prove to be a useful marker of the neuronal system (Beck, 2004).

Based on the analyses of its role in metamorphosis (Pecasse, 2000), Kr-h is defined as a modulator of the hormonal response, and this description is also appropriate in embryos. The consequences of early Kr-h mutations are as dramatic as at metamorphosis and the molecular consequences of mutant Kr-h alleles are similarly complex as they lead to errors in the timing and level of expression of genes of the ecdysone regulatory hierarchies and, perhaps more noticeably in these experiments, to failures in isoform switching. Kr-h is necessary for the correct expression of E74A, E74B, and E75C (Beck, 2004).

Between 8 and 14 h, E74A expression is aberrant in the absence of Kr-h transcripts, and this appears to depend primarily on Kr-hα, since expression is not restored in the presence of the Kr-h^7id allele where there are normal Kr-hβ levels in this period. E74B expression is similarly aberrant in the absence of Kr-h being broadly expressed between 8 and 24 h. The 12–16-h peak seen in control heterozygotes is restored in a dose-dependant manner by the Kr-h^7id allele, suggesting that Kr-hβ represses E74B expression from 16 h onwards. A similar effect is seen at metamorphosis where E74B is expressed prematurely in Kr-h¹ prepupae (Pecasse, 2000). A further striking interaction concerns E75C expression. In the absence of Kr-h, E75C transcripts are abundant between 8 and 16 h. Their repression is restored by a single dose of Kr-h^7id, suggesting that Kr-hβ is sufficient for this repression. Note however the correlation between Kr-hα transcripts and E75C transcripts in these animals from 16 h onwards (Beck, 2004).

While each change is in itself modest, if one considers the levels of transcripts of different isoforms present at a given time point, the ratios of the expression of these key regulatory genes are significantly perturbed (compare for example E74A, E75A, E75B, and E75C transcript levels in 10-12-h embryos of the Df(2L)Kr-h7.1/+ and Df(2L)Kr-h7.1/Df(2L)Kr-h7.1 genotypes. Note that since Kr-h is limited in its expression to neuronal cells, it is possible that the total embryo RT-PCR approach averages down the impact of the mutation in those tissues, although the Kr-h mutations may have indirect effects on gene expression in other tissues (Beck, 2004).

The idea that Kr-h is necessary to assure a fine balance in the regulatory hierarchies is supported by observations that the expression of key genes is highly dose dependant on the different Kr-h alleles. An unexpected result was the survival until the prepupal stage of combinations of Kr-h⁶, Kr-h⁷, and Kr-h^7id with Df(2L)Kr-h7.1. This was perhaps most striking for Kr-h^7id as, unlike Kr-h^7id embryos, the Df(2L)Kr-h7.1/Kr-h^7id embryos hatch and the larvae reach pupariation. Kr-hβ transcripts are reduced in these embryos compared to Kr-h^7id and are closer to control levels, suggesting that excess Kr-hβ transcript levels cause the block in late embryogenesis. Levels of other ecdysteroid-regulated transcripts are also closer to controls than Df(2L)Kr-h7.1 embryos. The idea that a reduction in Kr-hβ transcripts is necessary for early development is consistent with one interpretation of observations of Kr-h expression during metamorphosis (Pecasse, 2000) where the programmed disappearance of Kr-hα during the early to mid-prepupal period may be crucial for the transition from pupariation to pupation. This problem of balance may be further complicated by the observation that the KR-H protein binds near the Kr-h locus in salivary gland polytene chromosomes suggesting that there may be an element of autoregulation at the locus. These same studies show that KR-H localization is dynamic, and current efforts to understand its mode of action are focussed on its role during the metamorphic response to hormone (Beck, 2004).

Krüppel-homolog, a stage-specific modulator of the prepupal ecdysone response, is essential for Drosophila metamorphosis

A P-element-induced prepupal mutant of Drosophila melanogaster was charaterized that after an apparently normal embryonic and larval development fails to complete head eversion, an essential step in metamorphosis. The P-element insertion disrupts an ecdysone-regulated transcript which, although expressed during embryonic and larval stages, appears critical for preparing the late prepupal response to ecdysone. By a combination of molecular and genetic studies, in which new alleles were uncovered, it was shown that the locus is complex, containing at least two distinct promoters. These studies on the Kr¸ppel-homolog (Kr-h), add to a growing body of evidence that specific isoforms of a number of key genes are implicated in both embryogenesis and metamorphosis (Pecasse, 2000).

Metamorphosis involves the extensive remodelling of the Drosophila body plan. While certain larval tissues are destined for histolysis, others are reprogrammed. In parallel the imaginal discs differentiate into adult tissues. These events are not simultaneous in all tissues and a precisely timed series of events is initiated by the rise in ecdysone titre at the end of the third larval instar. Its progression requires the subsequent fluctuations of hormone titre in mid- and late prepupae. In the prepupal Kr-h lethal alleles the early phase appears to progress normally, at least until gas bubble formation at 3 to 4 h after pupariation, and it is only in midprepupae that the underlying genetic programmes show the first signs of disturbances which become morphologically apparent in late prepupae. Gas bubble displacement was followed in individual prepupae over several hours. By successive movements in both the anterior and the posterior parts, the gas bubble plays a role in separating cuticles and the animal finally separates completely from the puparium at head eversion, which marks pupation. In Kr-h lethal alleles, movements of the bubble appear random as though the neuromuscular system is uncoordinated. Equally, an aberration in the ecdysone response is seen by the persistence of the salivary gland which normally histolyses some 14 h after pupariation. Similar observations have been reported for EcR, βFTZ-F1, and DHR3 mutants dying in the prepupal or early pupal period (Pecasse, 2000).

The differentiation of the adult head, thorax, and abdomen appear to a certain extent independent. While all but a few animals fail to achieve head eversion, the leg and wing discs in some individuals fuse to form a thorax which most closely resembles the adult structure. In certain cases one can observe head characteristics, notably eye pigmentation, without head eversion. Perhaps significantly a similar aberrant differentiation producing a cryptocephalic phenotype has also been described for compound mutants of two key genes of the ecdysone response, E74 and BR-C. This suggests that Kr-h also acts in this regulatory pathway. The cuticle of the abdomen (derived from nests of histoblasts) forms, but with the exception of bristle formation in the anterior segments it looks more like larval or pupal cuticle, as though metamorphosis had not taken place in this part of the body. Such a differential response of body compartments, including a gradient of response in the abdomen, has been observed in classical approaches, notably the response to topical applications of juvenile hormone in late larvae or prepupae (Pecasse, 2000).

Like many genes involved in the ecdysone response, Kr-h has alternate promoters that give rise to transcripts which may encode at least two protein isoforms which contain different domains characteristic of transcription factors. It cannot be excluded that Kr-hγ will also produce a novel isoform, but to date it has not been possible to characterise this minor transcript. Similarly it is not possible to exclude the possibility that transcripts that retain the short introns are in fact translated into truncated proteins. An antibody against a region common to the two major isoforms detects Kr-h at a restricted number of chromosomal sites in larval and prepupal salivary glands. Further antibodies will be necessary to distinguish the proteins that are synthesised and to determine their stage and tissue distribution. Such diversity when integrated with combinatorial possibilities of the other genes, notably by direct protein-protein interactions, may be fundamental to the stage- and tissue specific response to ecdysone (Pecasse, 2000).

The absence of Kr-hα transcripts in the Kr-h¹ mutant has modest effects on the level of expression of key genes of the ecdysone regulatory hierarchies but, perhaps more importantly, causes a shift in their time of expression, both in vivo and in cultured glands. This disturbs the fine balance in regulatory factors necessary for the successful progression of the hormonal response as a midprepupal rift appears at the time when the salivary gland is preparing for the late prepupal response to ecdysone. Since this coincides with the time of disappearance of Kr-h transcripts in wild-type glands, it suggests that while the high level of Kr-h transcripts found in larvae and early prepupae is not vital, as larval development proceeds normally in Kr-h1, maintaining activity for the first 5 to 6 h of the prepupal period is essential. As there is premature induction of EcR transcripts in the mutant, this may indicate that Kr-h has a role in delaying the appearance of the ecdysone receptor. In this respect the fact that the profile of Kr-h transcripts in prepupal glands resembles that of DHR3, shown to be necessary for the progression to the prepupal ecdysone response, and the similarities in their response to hormone treatment of 2-h prepupal glands, may reflect a shared regulatory mechanism. However, while the molecular consequences for the ecdysone regulatory hierarchy in Kr-h mutants are more subtle than those seen in DHR3 mutants, the lethal phenotype appears more temporally restricted. Up to 90% of Kr-h mutants die as prepupae, whereas only 25% of DHR3 mutants die at this stage, the others dying later in pupal development (Pecasse, 2000).

Initially, because of the failure of the salivary gland to histolyse, it is suspected that the late prepupal response to hormone did not occur. However, it is now clear that this response is initiated, since E74A, E75A, E75B, and E93 transcripts are induced although the induction of the latter is clearly suboptimal. Note that the RT-PCR assay detects basal levels of E93 prior to the late prepupal induction from 10 h onwards. As it has been suggested that E93 is a key regulator of programmed cell death, it is possible that the gland survives up to 36 h after pupariation because apoptosis of the salivary gland is blocked in Kr-h. However, although the regulation of a certain number of genes known to be involved in apoptosis is modified, the fact that their transcripts are abundant in mutant glands that escape histolysis suggests that other factors are involved (Pecasse, 2000).

Search PubMed for articles about Drosophila Kr-h1

Beck, Y., Pecasse, F. and Richards, G. (2004). Kr¸ppel-homolog is essential for the coordination of regulatory gene hierarchies in early Drosophila development. Dev. Biol. 268(1): 64-75. PubMed citation: 15031105

Beck, Y., Dauer, C. and Richards, G. (2005). Dynamic localisation of KR-H during an ecdysone response in Drosophila. Gene Expr. Patterns 5(3): 403-9. PubMed Citation: 15661647

Beckstead, R. B., Lam, G. and Thummel, C. S. (2005) The genomic response to 20-hydroxyecdysone at the onset of Drosophila metamorphosis. Genome Biol. 6 (R99): 1-13. PubMed citation: 16356271

Brody, B., Stivers, C., Nagle, J. and Odenwald, W. F. (2002). Identification of novel Drosophila neural precursor genes using a differential embryonic head cDNA screen. Mech. Dev. 113: 41-59. PubMed citation: 11900973

Grozinger, C. M. and Robinson, G. E. (2006). Endocrine modulation of a pheromone-responsive gene in the honey bee brain. J. Comp. Physiol. A 193: 461-470. PubMed citation: 17192826

Grozinger, C. M., et al. (2003). Pheromone-mediated gene expression in the honey bee brain. Proc. Natl. Acad. Sci. 100: 14519-14525. PubMed citation: 14573707

Minakuchi, C., Zhou, X. and Riddiford, L. M. (2008). Krüppel homolog 1 (Kr-h1) mediates juvenile hormone action during metamorphosis of Drosophila melanogaster. Mech. Dev. 125(1-2): 91-105. PubMed citation: 18036785

Pecasse, F., Beck, Y., Ruiz, C. and Richards, G. (2000). Krüppel-homolog, a stage-specific modulator of the prepupal ecdysone response, is essential for Drosophila metamorphosis. Dev. Biol. 221(1): 53-67. PubMed citation: 10772791

Shi, L., Lin, S., Grinberg, Y., Beck, Y., Grozinger, C. M., Robinson, G. E. and Lee, T. (2007). Roles of Drosophila Kruppel-homolog 1 in neuronal morphogenesis. Dev. Neurobiol. 67(12): 1614-26. PubMed citation: 17562531

Schuh, R., et al (1986). A conserved family of nuclear proteins containing structural elements of the finger protein encoded by Kr¸ppel, a Drosophila segmentation gene. Cell 47: 1025-1032. PubMed citation: 3096579

Zhou, X., Zhou, B., Truman, J. W. and Riddiford, L. M. (2004). Overexpression of broad: a new insight into its role in the Drosophila prothoracic gland cells. J. Exp. Biol. 207:1151-1161. PubMed citation: 14978057

date revised: 29 June 2008

Home page: The Interactive Fly © 2008 Thomas Brody, Ph.D.