Genetic and Physical Studies of a Portion of T H E White Locus Participating in Transcriptional Interactions in Drosophila Adijlt Tissues Regulation and in Synapsis-dependent

We have identified and sequenced the portion of the white locus affected by an idiosyncratic set of white mutant alleles (the wJP alleles). The affected white locus portion (w‘p region) extends from ca. 590 base pairs (bp) to ca. 1270 bp 5’ to the apparent start site for the major white transcription unit. Based on the properties of these mutant alleles, we infer the existence of two distinct cis-acting regulatory elements in the w* region and a third element mapping 3’ to this region (3’ to position ca. -670). Our analysis allows us to define the apparent position of one of the two wSp region elements with substantial precision. Examination of the DNA sequences in this region suggests that it is functionally similar to the enhancers identified in vertebrates. This same element participates in synapsis-dependent genetic interactions, suggesting a largely unexpected relationship between enhancer-like, cis-acting genetic elements and the genetic elements responsible for the synapsisdependent genetic interactions in trans revealed by the existence of transvection effects. Our results further suggest that a presumptive regulatory locus (suppressor-of-whitespotted) regulates white transcription in adult tissues and is not involved in regulating white expression in larvae. We discuss the regulation of white expression in light of our studies. We also demonstrate unusual structures for an X-ray-induced deletion and a spontaneous deletion. HE wsf’ mutant alleles at the white locus of Drosophila have long been T recognized to have quite special properties as assessed by genetic and cytological studies (LINDSLEY and GRELL 1968 and references therein): These alleles produce a characteristic, drastic reduction in eye pigment deposition. Moreover, the wsp mutations disrupt the synapsis-dependent genetic interactions between synapsed white alleles revealed by the existence of transvection effects (GANS 1953; LEWIS 1954; GREEN 1959; KAUFMAN, TASAKA and SUZUKI 1973; JACK and JUDD 1979; BABU and BHAT 1980, 1981; BINCHAM 1980). Furthermore, in spite of their quite extreme effects on adult eye color phenotype, wSp mutations have no effect on pigmentation of larval malpighian tubules. This is in contrast to essentially all comparably extreme white mutant alleles which exert similar effects on pigmentation in both of these tissues Genetics 110: 479-494 July, 1985. 480 D. DAVISON E T AL. (LINDSLEY and GRELL 1968). This last observation strongly suggests that wsp mutations have tissue-specific effects on white locus expression. Molecular studies (ZACHAR and BINCHAM 1982) revealed several additional novel features of these mutations: First, the four wSp mutations affect the same portion of the locus. Second, the detailed structure of these mutations suggested that they affected a genetic element outside of the white locus transcription unit and, thus, were good candidates for regulatory mutations. Several recent studies ( O ' H A R E , LEVIS and RUBIN 1983; LEVIS, O'HARE and RUBIN 1984; PIRROTTA and BROCKL 1984; CHAPMAN and BINCHAM 1985) support the original proposal of ZACHAR and BINGHAM (1982) that the wsp mutations map outside of the white transcription unit. Furthermore, CHAPMAN, ZACHAR and BINGHAM (1985) demonstrated the existence of a novel type of suppressor locus (suppessor-of-white-spotted) that interacts specifically with the wsp mutant alleles. Mutant alleles at suppressor-of-white-spotted largely restore a wild-type eye color phenotype to wSp individuals. Thus, the wSp region represents a well-defined opportunity to study presumptive regulatory elements. Moreover, the effects of these mutations on synapsis-dependent genetic interactions at white suggest that these mutations may reveal unexpected properties of such elements. We report here the results of physical, genetic and transcriptional studies of mutations affecting the wsp region of the white locus. MATERIALS AND METHODS Fly strains and culture: Description of fly strains can be found in LINDSLEY and GRELL (1968), in ZACHAR and BINCHAM (1982) and in CHAPMAN and BINCHAM (1985). All files were reared at 2 1.5-22" throghout the developmental stages examined. Control of culture temperature is extremely important in visualizing the effects of su(wsp) on white transcript levels in mature adult tissues. At higher temperatures (greater than 24") the adult effects described are much less extreme (CHAPMAN and BINGHAM 1985; D. DAVISON, C. CHAPMAN and P. BINGHAM, unpublished observations). Adult eye color phenotypes described were assessed within 24-48 h after eclosion. DNA sequence determination: DNA sequences were determined essentially as described by SANCER, NICKLEN and COULSON (1977), SANGER and COULSON (1978) and BARNES, BEVAN and SON (1983). A description of the fragments sequenced is in Figure 9. Analysis of DNA sequence similarities: Analysis of nucleotide sequence similarities was performed in several ways. T h e dot matrix-plotting programs of CONRAD and MOUNT (1982) and PUSTELL and KAFATOS (1984) were used for initial examination of the sequence and for comparison to the BPV enhancer region (WEIHER and BOTCHAN 1984) and the SV40 enhancer core sequence (WEIHER, KONIG and GRUSS 1983). The GOAD and KANEHISA (1982) implementation of the NEEDLEMAN and WUNSCH ( 1 970) and SELLERS ( 1 974) (NWS) metric search procedures were then used to further refine the similarities noted in the dot matrix procedures. Analysis of the direct repeats in the su(zdp)p region (see RESULTS) was performed using the global alignment implementation of the NWS procedures (FITCH and SMITH 1983). Statistical analysis w d S performed by the method of GOAD and KANEHISA (1982); in particular, equations 4 though 7 were used. For all of these calculations the composition of the entire sequence in Figure 2 was used. T h e expectation (E) value is the average number of times one would expect to find a match of comparable quality in two randomly chosen sequences of the same length and composition (GOAD and KANEHISA 1982; KANEHISA 1984). For the range of E values calculated below, E values are good approximations of the frequency with which the matches in question should occur at random and we will refer to E values as expected frequencies in the text. T h e expectation values for the alignments in Figure 8 are 0.012 (SV40) and 0.048 (BPV). T h e relevant parameters for all WHITE LOCUS REGULATION 481 calculations are given below. The direct repeat in the su(w'p) interval (see DISCUSSION) has an E of 0.0002. Although the GOAD and KANEHISA (1982) method has been criticized (LIPMAN et al. 1984), we note that the E resulting from the method is an upper bound and, therefore, similarities are likely to be more statistically significant than this calculation indicates. Furthermore, the problem pointed out by LIPMAN et al. (1984) occurs at particular, unusual boundary conditions which are not applicable in the cases presented here. The data for the calculations of Figure 8 are: top, p = 0.228, I = 896 (in this specific case] = I ; W. GOAD, personal communication); bottom, p = 0.252, I = 896, J = 64. For the direct repeats near the su(wq)p region the data are: p = 0.255, 1 = 68, m = 46, r = 18, gl = 6, g2 = 2, I = I = 896. SI protection analysis of deletion breakpoints: Approximately 10 pg of Drosophila genomic DNA were mixed with approximately 50 ng of an MIS-cloned fragment (virion DNA) in 100 pl of 10 mM Tris, 1 mM EDTA (pH 7.4). The mixture was vortexed vigorously to shear genomic DNA slightly and then heated to 95" for 5 min. The mixture was immediately chilled on ice, NaCl was added to 0.2 M and the mixture was incubated at 65" for 1 h. After chilling, an equal volume of a solution containing 50 mM sodium acetate and 6 mM zinc sulfate was added to the mixture and followed by the appropriate amount of SI. Digestion was carried out at 37" for 15 min. SI protection products were analyzed by Southern transfer after fractionation on 1 % formaldehyde agarose gels (MANIATIS, FRITSCH and SAMBROOK 1982). Northern gel analysis: RNAs were analyzed after Northern transfer (MANIATIS, FRITSCH and SAMBROOK 1982) using the hybridization procedure of Hu and MFSSING (1982) modified by BINGHAM and ZACHAR (1 985). Purified head tissues and polyadenylated RNAs were prepared as previously described (BINGHAM and ZACHAR 1985). Larvae were washed from food, and polyadenylated RNAs were prepared as from adult tissues. white locus DNA sequence probes: white locus DNA sequences were originally cloned by BINGHAM, LEVIS and RUBIN (1981). All white cloned segments used herein were derived from these original segments or were retrieved from clone libraries using these sequences as probes (ZACHAR and BINGHAM 1982; LEVIS, BINGHAM and RUBIN 1982; D. DAVISON, C. CHAPMAN and P. BINGHAM, unpublished observations).