The Drosophila Broad-Complex encodes a family of related proteins containing zinc fingers.

The Broad-Complex (BR-C) is essential for metamorphosis in Drosophila melanogaster. This locus is coextensive with the 2B5 ecdysone-responsive early puff and is necessary for puffing and transcription of many subsequently activated late genes in the developing salivary gland. Mapping of 31 cDNA clones indicates that approximately 100 kb of the genome is devoted to the synthesis of many BR-C RNAs. Sequence analyses of these cDNA clones show that the BR-C encodes a family of related proteins characterized by a common core amino-terminal domain fused to alternate carboxy domains each containing a pair of zinc fingers. Most proteins also contain domains rich in distinctive amino acids located between the common core and zinc finger regions. BR-C mutant alleles resulting from chromosomal rearrangements at 2B5 are associated with deletions of 5'-untranslated sequences, separation of the core coding domain from the downstream zinc finger domains, or a P element insertional disruption of a zinc finger coding sequence. We infer that the BR-C directly regulates late gene expression by specifying the synthesis of a family of proteins with DNA binding potential.

Delineation of cis-acting sequences required for expression of Drosophila mojavensis Adh-1.

The control of expression of the Adh-1 gene of Drosophila mojavensis has been analyzed by transforming ADH null Drosophila melanogaster hosts with P element constructs which contain D. mojavensis Adh-1 having deletions of different extent in the 5' and 3' ends. Adh-1 expression in the D. melanogaster hosts is qualitatively similar to expression in D. mojavensis, although expression is quantitatively lower in transformants. Deletions of the 5' end indicate that information required for normal temporal and tissue expression in larvae is contained within 70 bp of the transcription start site. However, deletion constructs to -70 are deficient in ovarian nurse cell expression, whereas the additional upstream sequences present in constructs containing deletions to -257 do support expression in the ovary. Comparison of the nucleotide sequence in the -257 to -70 region of Adh-1 of four species: D. mojavensis and Drosophila arizona, which express Adh-1 in the ovary, and Drosophila mulleri and Drosophila navojoa, which do not, has led to the identification of regions of sequence similarity that correlate with ovary expression. One of these bears a striking similarity to a conserved sequence located upstream of the three heat shock genes that have constitutive ovarian expression and may be an ovarian control element. We have identified an aberrant aspect of Adh-1 expression. In transformants which carry an Adh-1 gene without a functional upstream Adh-2 gene Adh-1 expression continues into the adult stage instead of ceasing at the onset of metamorphosis. In transformants with a functional Adh-2 gene, Adh-1 expression ceases in the third larval instar stage and aberrant expression in the adult stage does not occur.

Relationships between protein isoforms and genetic functions demonstrate functional redundancy at the Broad-Complex during Drosophila metamorphosis.

Metamorphosis in holometabolous insects is an ecdysone-dependent process by which the larval form is replaced by a reproductive, adult form. At the onset of metamorphosis ecdysone induces a set of early genes which coordinate tissue-specific responses to hormone. The Broad-Complex (BR-C) early gene, which acts as a global regulator of tissue-specific responses to ecdysone, encodes a family of zinc-finger DNA binding proteins known as Z1, Z2, Z3, and Z4. Genetically the BR-C encodes three complementing functions, br, rbp, and 2Bc, and a class of npr1 alleles that fail to complement any of the other genetic functions. The effects of BR-C mutations on metamorphic development are highly pleiotropic, yet little is known about the roles of individual BR-C proteins in directing the required responses to ecdysone. Because the BR-C is a vital regulator of metamorphosis it is essential to establish the relationships between BR-C genetic functions and protein products. We present here the first general and definitive study of these relationships. Using heat-inducible transgenes we have rescued lethality associated with each of the complementing genetic functions and have restored transcriptional activity of tissue-specific BR-C(+)-dependent target genes. Our data lead us to conclude that br+ function is only provided by the Z2 isoform. We find that Z1 transgenes provide full rbp+ function, while Z4 provides partial function. Likewise, while Z3 provides full 2Bc+ function, Z2 also provides partial function. These results indicate possible functional redundancy or regulatory dependence (via autoregulation) associated with the rbp+ and 2Bc+ functions. The establishment of these relationships between BR-C genetic functions and protein isoforms is an important step toward understanding the roles of BR-C proteins in directing metamorphic responses to ecdysone.

A switch in broad-complex zinc-finger isoform expression is regulated posttranscriptionally during the metamorphosis of Drosophila imaginal discs.

The Broad-Complex (BR-C) is a key member of the 20-hydroxyecdysone regulatory hierarchy that coordinates changes in gene expression during Drosophila metamorphosis. The family of transcription factors encoded by the BR-C share a common amino-terminal domain which is fused by alternative splicing to one of four pairs of C2H2 zinc-finger domains (Z1, Z2, Z3, and Z4). In this study, we examine the temporal expression of transcripts encoding each BR-C zinc-finger isoform-including the newly discovered fourth zinc-finger domain-during the metamorphosis of imaginal discs which form the integumental structures of the adult head and thorax. We find that all BR-C zinc-finger RNA isoforms are induced as a primary response to 20-hydroxyecdysone. However, induced BR-C RNA isoforms exhibit two divergent expression profiles. The Z2, Z3, and Z4 RNA isoforms accumulate to high levels at the beginning of the ecdysone response and abruptly disappear after several hours. In contrast, the Z1 RNA isoform continues to accumulate while the others decline, resulting in a switch in relative isoform levels. Using probes specific to different regions of the BR-C, we show that the switch in BR-C RNA isoform expression appears to be posttranscriptionally regulated, presumably by ecdysone-responsive factors. We propose that this switch results from a change in splice acceptor site choice. Finally, we present a model describing how this temporal switch in isoform expression could mediate changes in BR-C function, from transcriptional activation to repression and vice versa, that are critical for coordinate downstream target gene expression.

From Beaumont to poison ivy: marine sponge cell aggregation and the secretory basis of inflammation.

We have studied Microciona prolifera cells as a model for inflammation and secretion. Dissociated in Ca-, Mg-free seawater with 2.5 mM EDTA, the cells aggregate when exposed to Ca (greater than 5 mM) and Ca ionophores. Extracellular Ca is not required over the course of aggregation; brief pulses of Ca suffice. Aggregation was induced by A23187 in excess EDTA after cells were prepared by pulse Ca. It appeared that Ca ionophore stimulated the secretion of Microciona aggregation factor (MAF) to a locus or in a form inaccessible to external EDTA. Pulse-induced aggregation depended on MAF because it was inhibited by MAF fragments, which are ligands for MAF-binding sites. Sponge cells were preloaded with three fluorescent dyes that monitor aspects of stimulus-secretion coupling: 1) 3,3'-dipropylthiadicarbocyanine iodide (dis-C3-(5)), a carbocyanine dye presumed to report changes in membrane potential; 2) 9-aminoacridine (9AA), which presumably reports secretion from acid vesicles; and 3) chlortetracycline (CTC), presumed to report mobilization of membrane-associated Ca. Exposure of cells either to constant Ca or to pulse Ca stimuli caused prompt decreases in the fluorescence of cells with diS-C3-(5) and increases in fluorescence of cells with 9AA. In contrast, although constant Ca provoked decreases in fluorescence of cells with CTC, a pulse Ca was without effect. Moreover, inhibitors of stimulus-response coupling (e.g., aspirin, sodium salicylate, 5 mM; diclofenac, 100 microM) inhibited sponge aggregation induced by either constant or pulse stimuli. In contrast, like the endogenous mediator of inflammation, leukotriene B4, trienoic alkyl catechols (urushiol) from poison ivy provoked aggregation. These studies suggest the utility of this marine model for analysis of stimulus-response coupling in cells of higher species that also respond to secretagogues in the absence of external Ca.

Broad-complex transcription factors regulate thoracic muscle attachment in Drosophila.

The Broad-Complex, a 20-hydroxyecdysone-regulated gene, is essential for the development of many tissues during metamorphosis. In Broad-Complex mutants of the rbp complementation group, dorsoventral indirect flight muscles (DVM) are largely absent, and the dorsal longitudinal indirect flight muscles, tergotrochanteral muscles, and remaining DVM often select incorrect attachment sites. The Broad-Complex encodes a family of zinc-finger-containing transcription factors, and it is hypothesized that Broad Complex proteins containing the Z1 zinc-finger pair (BRC-Z1) mediate rbp+ function. We provide additional strong support for this hypothesis by showing that heat-shock-induced BRC-Z1 expression rescues the thoracic muscle defects of rbp mutants completely. BRC-Z4 induction can also rescue the thoracic musculature, but BRC-Z2 and -Z3 can not. Thus, the effect is specific to BRC-Z1 and its closest relative, BRC-Z4. Formation of muscle primordia from imaginal myoblasts appears normal in rbp mutants. However, the myotendinous junctions linking the DVM to the dorsal epidermis are weak, and the muscles detach during pupal life and subsequently degenerate. The data indicate that rbp mutations disrupt the cell-cell interactions between developing muscles and epidermal tendon cells as they recognize and attach to one another. Using a BRC-Z1-specific monoclonal antibody, we show that both the developing muscles and epidermal tendon cells express BRC-Z1 at the time of pupation, before mutant muscles begin to detach. We conclude that 20-hydroxyecdysone acts through the Broad-Complex to control the development of thoracic myotendinous junctions.

The Drosophila Broad-Complex early gene directly regulates late gene transcription during the ecdysone-induced puffing cascade.

The ensemble of tissue-specific changes that drives Drosophila metamorphosis is initiated by the steroid hormone ecdysone and proceeds through a transcriptional cascade comprised of primary response transcriptional regulators and secondary response structural genes. The Broad-Complex (BR-C) primary response early gene is composed of several distinct genetic functions and encodes a family of related transcription factor isoforms. Our objective in this study was to determine whether individual BR-C isoforms directly regulate secondary response target genes. A cluster of 10 salivary gland-specific secondary response L71 late genes are dependent on the BR-C rbp+ genetic function. Transgenic animals expressing individual BR-C isoforms were tested for their ability to provide the BR-C rbp+ genetic function by monitoring the transcriptional activation of the L71 genes. We found that the BR-C Z1 isoforms could complement the transcriptional defects seen in rbp mutants but the Z2, Z3, and Z4 isoforms could not. We conclude that the BR-C rbp+ function is provided by the BR-C Z1 isoform in prepupal salivary glands. L71 gene rescue was restricted to the prepupal salivary gland, suggesting the involvement of additional factors in L71 gene regulation. Interestingly, we found that the overexpression of Z3 or Z4 isoforms in BR-C+ salivary glands repressed L71 expression, indicating that BR-C proteins might also function as transcriptional repressors. Molecular mapping and characterization of the regulatory sequences that control L71-6 expression revealed several Z1 isoform binding sites. Mutagenesis of these Z1 binding sites resulted in the failure to activate late gene expression in vivo when measured by transgenic reporter genes. We conclude that the BR-C early gene directly activates late gene transcription by interacting with late gene cis-acting regulatory elements and that this interaction is responsible for the temporal linkage of early and late ecdysone-induced gene expression.

Adolescent coitus and infant birthweight.

Seventy-nine percent of a group of adolescent mothers reported coitus during pregnancy. There was a significant difference between the mean birthweight and gestational age of infants born to mothers who reported coitus within 90 days of delivery compared to the mean birthweight and gestational age of infants of mothers who reported no gestational age of infants of mothers who reported no coitus within 90 days of delivery. Thirty-nine percent of the variance in birthweight was explained by maternal factors alone, but only 1.0% of the variance in birthweight was accounted for by the interval between last coitus and delivery. Prepregnancy weight (r = 0.380) and weight gain during pregnancy (r = 0.327) contributed 19.1% and 7.4% of the variance in birthweight, respectively.

PIP: The subjects for this study are 99 (consecutive) adolescents who received prenatal care in the Rochester Adolescent Maternity Project (RAMP) between August 1980 and December 1981 and their infants. Data was collected from semi-structured interviews about the adolsecents' coital activitiy during pregnancy and the interval between last coitus and delivery. Demographic, obstetric and medical data were collected from each subject's chart. Neonatal data included birthweight and gestational age. T-tests, chi-square and multivariate analyses were performed. Results indicate that 79% of the study population reported coitus during pregnancy. 71% of these respondents stated that coitus occurred within 90 days of delivery. There was a significant difference between the mean birthweight and the mean gestational age of infants of mothers who reported coitus within 90 days of delivery compared to the infants of mothers who reported no coitus within 90 days of delivery. On multivariate analysis, prepregnancy weight (r=.380) and weight gain during pregnancy (r=.327) were both significantly correlated with birthweight. 39% of the variance in birthweight was explained by maternal factors alone, but only 1.0% of the variance in birthweight was accounted for by the interval between last coitus and delivery. As found by others, prepregnancy weight, weight gain during pregnancy, smoking frequency, and race were the major maternal factors accounting for the variance in birthweight. It is concluded that coitus alone during pregnancy in adolescents does not account for a significant amount of the variance in infant birthweight.

Dynamic expression of broad-complex isoforms mediates temporal control of an ecdysteroid target gene at the onset of Drosophila metamorphosis.

Metamorphosis in Drosophila melanogaster is orchestrated by the steroid hormone ecdysone, which triggers a cascade of primary-response transcriptional regulators and secondary effector genes during the third larval instar and prepupal periods of development. The early ecdysone-response Broad-Complex (BR-C) gene, a key regulator of this cascade, is defined by three complementing functions (rbp, br, and 2Bc) and encodes several distinct zinc-finger-containing isoforms (Z1 to Z4). Using isoform-specific polyclonal antibodies we observe in the fat body a switch in BR-C isoform expression from the Z2 to the other three isoforms during the third instar. We show that the 2Bc(+) function that corresponds presumably to the Z3 isoform is required for the larval fat body-specific expression of a transgenic construct (AE) in which the lacZ gene is under the control of the ecdysone-regulated enhancer and minimal promoter of the fat body protein 1 (Fbp1) gene. Using hs(BR-C) transgenes, we demonstrate that overexpression of Z1, Z3, or Z4, but not Z2, is able to rescue AE activity with faithful tissue specificity in a BR-C null (npr1) genetic context, demonstrating a partial functional redundancy between Z1, Z3, and Z4 isoforms. We also show that continuous overexpression of Z2 during the third instar represses AE, while conversely, expression of Z3 earlier than its normal onset induces precocious expression of the construct. This finding establishes a tight correlation between the dynamic pattern of expression of the BR-C isoforms and their individual repressive or inductive roles in AE regulation. Altogether our results demonstrate that the balance between BR-C protein isoforms in the fat body mediates, in part, the precise timing of the ecdysone activation of the AE construct but does not modulate its tissue specificity.