The Relationship between the Functional Complexity and the Molecular Organization of the Aivtennapedza Locus of Drosophzla Melanogaster

The Antp locus is involved in the development of the thorax of the larval and adult Drosophila. The absence of Antp' function during embryogenesis results in the larval mesothorax exhibiting characteristics of the prothorax and an ensuing lethality; the loss of Antp+ function in the development of the adult thorax causes specific portions of the leg, wing and humeral imaginal discs to develop abnormally. Every Antp mutation, however, does not cause all of these developmental defects. Certain mutant alleles disrupt humeral and wing disc development without affecting leg development, and they are not deficient for the wild-type function required during embryogenesis. Other Antp mutations result in abnormal legs, but do not alter dorsal thoracic development. Mutations of each type can complement to produce a normal adult fly, which suggests that there are at least two discrete functional units within the locus. This hypothesis is supported by the fact that each of the developmental defects arises from the alteration of a different physical region within the Antp DNA. These observations indicate that the complete developmental role of the Antp locus is defined by the spatial and temporal regulation of the expression of several individual functional units. ACH body segment of the Drosophila melanogaster larva and adult expresses E a characteristic set of cuticular structures (such as antennae and wings) and the specific set that is expressed depends in part on the expression of genes within the Bithorax Complex (BX-C) and/or the Antennapedia Complex (ANT-C). The identities of those segments that compose the posterior portion of the fly, from the metathorax through the eighth abdominal segment, are determined by genes of the BX-C (LEWIS 1978; SANCHEZ-HERRERO et al. 1985), whereas the identities of the first two thoracic segments (prothorax and mesothorax), as well as the development of the gnathal segments of the head (mandibular, maxillary and labial), result from the expression of genes within the ANT-C (KAUFMAN 1983 and unpublished observations). The ANT-C and the BX-C appear to be individually responsible for the identities of the head and abdomen, respectively. However, the proper identities of the thoracic segments depend on the expression of specific loci within both gene complexes Genetics 114: 919-942 November, 1986. 920 M. K. ABBOTT AND T. C. KAUFMAN (STRUHL 1982; KAUFMAN and ABBOTT 1984). Therefore, the analysis of the function of these particular genes and how they interact to promote normal thoracic development, be it directly or indirectly, should provide insight into the processes by which the developmental roles of the BX-C and ANT-C are defined. The Antennapedia (Antp) locus is one member of the Ant-C that is involved in the normal development of the larval and adult thorax. Embryos which are deficient for Antp+ function fail to make a normal mesothoracic segment, such that larvae developing from these embryos possess a second thoracic segment expressing characteristics of the prothorax (WAKIMOTO and KAUFMAN 198 1). Likewise, the elimination of Antp+ function, either during embryogenesis or the larval stage, results in the abnormal development of specific regions of the adult mesothorax, as well as the prothorax and metathorax (STRUHL 1981; KAUFMAN and ABBOTT 1984). The regions of the adult thorax which require Antp' function include portions of all three pairs of legs and specific regions of the dorsal mesothorax. Clones of mesothoracic leg tissue which are deficient for Antp+ function since early in embryogenesis are transformed to antennal tissue, suggesting that Antp+ function promotes a mesothoracic identity by the suppression of head development (STRUHL 1981). The scope of this Ant$+ function is not, however, confined solely to the mesothorax. The transformation of prothoracic and metathoracic leg tissue to antenna in somatic clones which are deficient for the wild-type function of either the Sex combs reduced (Scr) or the Ultrabithorax (Ubx) loci, as well as Antp' function, has resulted in the hypothesis that Antp+ function is actually required in all thoracic segments to suppress head development (STRUHL 1982). Furthermore, based on morphological criteria, DUNCAN (1982) and STRUHL (1983) have suggested that Antp is expressed in the first seven abdominal segments, but is apparently nonfunctional in the presence of BX-C gene expression. The purpose of the aforementioned investigations was to define the primary developmental role of the Antp locus; consequently, the mutant alleles which were utilized in these studies were characterized as being completely deficient for its wild-type function. Animals which possess Antp null alleles terminate development at the end of embryogenesis, or shortly thereafter, and exhibit the mesothorax to prothoracic transformation described above. Not all Antp alleles, however, produce these early developmental defects (DENELL 1973; LEWIS et al. 1980b). Nevertheless, the effects of these particular Antp alleles could be accounted for if it were assumed that they reduce the level of, but do not eliminate, Antp' function; i.e., they are hypomorphic alleles according to the critera of MULLER (1932). The preliminary characterization of several new X-ray-induced Antp alleles showed that there were significant qualitative differences in the genetic behavior of various Antp alleles (ABBOTT and KAUFMAN 1983). As a result of these observations, a detailed developmental-genetic investigation of a number of Antp alleles was undertaken. The results of this study, which are presented in this report, suggest that the differential genetic behavior exhibited by various ANTEh'NAPEDIA FUNCTIONAL COMPLEXITY 92 1 Antp alleles is due to the mutation of independent functional units which reside within the locus. Based on the nonoverlapping developmental defects caused by different sets of these alleles, we suggest that the Antp locus is composed of at least two discrete functional units. Some of these alleles are associated with chromosomal rearrangements; therefore, it has been possible to correlate each of the finctional units with a specific physical region within the Antp DNA. MATERIALS AND METHODS culture conditions: All fly stocks were maintained at 25" on media consisting of cornmeal, molasses and agar. This media was supplemented with live baker's yeast. Mutant chromosomes used: For clarity, the following nomenclature has been adopted in order to distinguish between the dominant and recessive Antp mutations: a recessive allele is designated with a superscript b e p i n g with a lower case "r" and followed by a specific mutant designation (e.g., Ant$' I o or rWZO), and a dominant allele possesses a superscript without the lowercase "r" (e.g., Antp" or CB). The Antp mutations utilized in this investigation are listed in Table 1. AI1 mutations not referenced within this report may be found in LINMLEY and GRELL (1968). The Blunt-short bristles (Bsb) mutation is described by BAKER (1980). X-ray mutagenesis: T o recover mutations in the Antp locus, a typical Fn mutagenesis screen was employed. Males homozygous for the third chromosome mutations red (353.6) and ebony (3-70.7) were X-irradiated with a dose of approximately 4000 r. These males were then mated with virgin females heterozygous for the third chromosome balancers TM3 and CxD. Between 40-50 females and 25 irradiated males were allowed to mate in bottles. After 2-4 days, the females were transferred to bottles containing fresh media and the males were discarded. From the F1 progeny, either * red e/TM3 (* red e designates the mutagenized chromosome) or * red e/CxD virgin females were collected and mass mated with Df(3R)Scr p* es/TM3 males. Several days later, single females were transferred to shell vials. The resulting F:! progeny from each female was scored for the presence of the * red e/Df(3R)Scr individuals. If no * red e/Df(3R)Scr progeny were recovered, then a stock of that * red e chromosome was established using the * red e/TM3 progeny. Polytene chromosome cytology: The polytene chromosomes from all of the X-rayinduced mutant stocks were examined to determine if they possessed visible rearrangements. These chromosome preparations were made from the salivary glands of * red elred e third-instar larvae. The mutant heterozygous larvae could be distinguished by the red pigmentation of their Malpighian tubules caused by the red mutation. These animals were raised at 18" in uncrowded conditions to maximize larval growth. The chromosome preparations were made according to the method of LEFEVRE (1976) and were examined with the phase contrast optics of a Zeiss Photomicroscope 111. Complementation studies: Complementation crosses were performed in order to establish allelic relationships between the X-ray-induced mutations. Typically, 101 5 virgin females were mated with 5-10 males in half-pint bottles. Unless specified, progeny were raised at 25". Scoring of the P1 progeny began on about the ninth day and continued for the next 6-7 days. A minimum of 100 progeny were scored for each cross. Somatic recombination experiments: Somatic clones homozygous for Antp' mutations were generated by X-ray-induced mitotic recombination. Males carrying an Antfir allele were mass mated with virgin females heterozygous for TM3 and a chromosome carrying the mutations Kinked (Ki), pink-peach ( p p ) , Minute(3)w Prickly (Pr) and Bluntshort bristles (Bsb). Progeny from this cross were collected for either 12or 24-hr intervals on 60 X 15 mm Petri dishes containing standard media. At various times after the midpoint of the collection period, the embryos or larvae were irradiated with 500 922 M. K. ABBOTT AND T. C. KAUFMAN