To accommodate growth, insects must periodically replace their exoskeletons. After shedding the old cuticle, the new soft cuticle must sclerotize. Sclerotization has long been known to be controlled by the neuropeptide hormone bursicon [Fraenkel, 1962; Cottrell, 1962), but its large size of 30 kDa has frustrated attempts to determine its sequence and structure. Using partial sequences obtained from purified cockroach bursicon (Honegger, 2002), the Drosophila gene CG13419 was identified as a candidate bursicon gene. CG13419 encodes a peptide with a predicted final molecular weight of 15 kDa, which likely functions as a dimer. This predicted Bursicon protein belongs to the cystine knot family, which includes vertebrate transforming growth factor-ß (TGF-ß) and glycoprotein hormones. Point mutations in the bursicon gene cause defects in cuticle sclerotization and wing expansion behavior. Bioassays show that these mutants have decreased Bursicon bioactivity. In situ hybridization and immunocytochemistry reveal that Bursicon is co-expressed with crustacean cardioactive peptide (CCAP). Transgenic flies that lack CCAP neurons also lack Bursicon bioactivity. These results indicate that CG13419 encodes Bursicon, the last of the classic set of insect developmental hormones. It is the first member of the cystine knot family to be assigned a defined function in invertebrates. Mutants show that the spectrum of Bursicon actions is broader than formerly demonstrated (Dewey, 2004).

bursicon transcripts are expressed in a set of the neurons that contain CCAP. Transgenic Drosophila bearing targeted ablations of CCAP neurons (CCAP KO flies) are defective in wing expansion and tanning and have low survival to adulthood (Park, 2003). In situ hybridization of the CNS of CCAP KO larvae using a bursicon antisense probe shows a significant reduction in the number of labeled neurons, as would be expected if CCAP neurons express bursicon. In some preparations, a few surviving neurons weakly express the CG13419 message in the posterior VNS, consistent with previous findings (Park, 2003). Similarly, it was predicted that homogenates of CNSs from the CCAP KO flies would not contain bursicon bioactivity if tested in the neck-ligated Sarcophaga bioassay. Indeed, homogenates from CCAP KO flies contain little or no bursicon activity in the Sarcophaga bioassay, whereas homogenates of the relevant control flies (LacZ) produced a score similar to that of wild-type flies (Dewey, 2004).

The phenotype of burs mutants is similar to that of rickets (rk) mutants, which also fail to sclerotize, pigment, and expand their wings (Baker, 2002). The rk gene encodes a G protein-coupled receptor with a large, leucine-rich extracellular domain and was proposed to be a Bursicon receptor. It belongs to a subfamily of receptors whose ligands include the vertebrate glycoprotein hormones (Baker, 2002), which are also cystine-knot proteins. The similarity in phenotype between burs and rk mutants suggests that they may represent the ligand and receptor for this hormonal signaling pathway. Nevertheless, flies that were transheterozygous for mutant alleles of both genes (rk⁴/+; burs^Z1091/+ and rk⁴/+; burs^Z5569/+) showed no defects in wing expansion or tanning. Additional detailed experiments will need to be carried out to determine whether these two molecules interact (Dewey, 2004).

The homology to peptide sequences from cockroach bursicon and other insect species, the phenotypes and reduction in Bursicon activity in mutants of this gene, and the localization of its transcripts to neurons known to produce bursicon in other insect species, make a convincing case that CG13419 encodes the tanning hormone Bursicon. Hence, 42 years after its discovery, the structure of the last of the classic insect developmental hormones has finally been elucidated. Bursicon is the first member of the cystine knot family of signaling molecules to be assigned a defined hormonal function in invertebrates. Knowledge of the molecular identity of Bursicon now provides tools for the analysis of its function, including the mechanism of its tightly timed action in triggering the biochemical events (Andersen, 1996) that lead to cuticle hardening. The colocalization of Bursicon with CCAP, a peptide that activates the ecdysis motor program, poses the fascinating problem of how these two hormones are differentially regulated to produce two temporally distinct behaviors. The understanding of the neuroendocrine control of cuticle sclerotization may also have potential for the development of novel insect-specific pest control agents and measures (Dewey, 2004).

GENE STRUCTURE

Exons - 3

PROTEIN STRUCTURE

Amino Acids - 173

Structural Domains

CG13419 was identified using a modified protein BLAST search of the Drosophila genome using the P. americana partial bursicon peptide sequences obtained by microsequencing (Honegger, 2002). Sequence analysis of the genomic clone and its corresponding cDNA has revealed that the coding sequence is 522 nucleotides long. The gene contains three short exons (130, 125, and 267 nucleotides, respectively) and two introns (64 and 58 nucleotides, respectively). The CG13419 gene product is predicted to be a 173 amino acid preprotein (19 kDa). Removal of the predicted N-terminal signal sequence of 33 amino acids would result in a mature protein of 140 amino acids, approximately 15 kDa. CG13419 is a member of the 10-membered cystine knot protein family (Vitt, 2001), which typically forms dimers. Members of this family contain six cysteine residues that form the knot and an optional additional cysteine that may be important for dimerization. They include the glycoprotein hormones, TGF-β, platelet-derived growth factor, and the mucins. The sequence of CG13419 is most similar to that of the mucin subfamily, which also includes signaling molecules such as the Bone Morphogenic Protein antagonists. Previous data indicate that bursicon functions as a 30 kDa dimer in the cockroach (Kostron, 1999). The available sequence information for the homologous partial peptide sequences from the Anopheles gambiae and Apis mellifera genomes show 83% identity to CG13419. The cysteines within the cystine knot domain are conserved in these species (Dewey, 2004).

EVOLUTIONARY HOMOLOGS

In an effort to characterize the insect molting hormone bursicon from the cockroach, Periplaneta americana, amino acid sequences with high identity of Cu,Zn-superoxide dismutase (SOD) of Drosophila virilis were identified. Antisera against a conserved region of SOD, and a sequence unique to Periplaneta SOD were produced and used to test whether bursicon might be a form of SOD. Western blots of one- and two-dimensional gels revealed that the dimeric form of SOD and bursicon have a similar molecular mass (30 kDa). The two proteins can be separated, however, according to their different isoelectric points. Bursicon is identified in two-dimensional gels by elution from four unique spots not labeled by the anti-SOD antisera. In sections of Periplaneta nerve cords, the antisera labeled glial material surrounding neuronal somata close to the neural sheath. Bursicon, however, is contained in unique cell pairs in the ganglia of the ventral nerve cord. These neurons were labeled with new antisera produced against novel sequences of one of the four above-mentioned bursicon active spots. The results show unequivocally that SOD and bursicon are distinctly different proteins. Furthermore, the anti-SOD antisera provided a tool to isolate and sequence bursicon (Kostron, 1999).

Bursicon is the final neurohormone released at the end of the molting cycle. It triggers the sclerotization (tanning) of the insect cuticle. Until now, its existence has been verified only by bioassays. In an attempt to identify this important neurohormone, bursicon was purified from homogenates of 2,850 nerve cords of the cockroach Periplaneta americana by using high performance liquid chromatography technology and two-dimensional gel electrophoresis. Bursicon bioactivity was found in four distinct protein spots at approximately 30 kDa between pH 5.3 and 5.9. The protein of one of these spots at pH 5.7 was subsequently microsequenced, and five partial amino acid sequences were retrieved. Evidence is presented that two of these sequences are derived from bursicon. Antibodies raised against the two sequences labeled bursicon-containing neurons in the central nervous systems of P. americana. One of these antisera labeled bursicon-containing neurons in the crickets Teleogryllus commodus and Gryllus bimaculatus, and the moth Manduca sexta. A cluster of four bilaterally paired neurons in the brain of Drososphila melanogaster was also labeled. In addition, this antiserum detected three spots corresponding to bursicon in Western blots of two-dimensional gels. The 12-amino acid sequence detected by this antiserum, thus, seems to be conserved even among species that are distantly related (Honegger, 2002).

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.