Drosophila melanogaster Strain, and Its Evolutionary Implications

Knowledge of multigenic family organization should provide insight into their mode of evolution. Accordingly, we characterized the 5s ribosomal gene family in the Drosophila melanogaster strain $06. The 5s genes in this strain display a striking HindIII restriction difference compared to the “standard” D. mehogaster 5s genes. The sequence of three q?06 5s genes was determined. We show that the HindIII restriction site heterogeneity within the fl 5s family most probably results from the same point mutation, suggesting that a single 5s variant was propagated into the 5s cluster of this strain. Furthermore, we demonstrate that the structural organization of the 5s genes in q?06 is a bipartite structure, i.e., that about 40% of the 5s genes constitute a HindIII+/HindIIImixed cluster, while those remaining constitute an homogeneous HindIIIcluster. The events which might lead to such an heterogeneous pattern are discussed from an evolutionary point of view. M ULTIGENIC families constituted by repeated sequences are present in all eukaryotic species. Although unit structure and family organization are highly variable among species, a high level of homogeneity is maintained within each, the different units of the family evolving, apparently, in unison. This process is usually referred to as concerted evolution (ZIMMER et al. 1980). Nonreciprocal exchanges, mainly unequal crossing-over and conversion events, are commonly thought of as phenomena affecting the evolution of repeated sequences (SMITH 1976; NACYLAKY and PETES 1982). These events are able to modify the frequency of a variant in a multigene family, leading thus to its spread or elimination. For example, an experimental determination of the rate of unequal mitotic crossing-over in Saccharomyces cereuisiae shows that it would be high enough to homogenize the 140 tandemly arranged 18s-28s ribosomal units (SzosTAK and Wu 1980). The authors calculate that there is a 0.5 probability that a ribosomal locus containing two equally abundant variants would be homogenized after 48,000 generations. However, in most cases, the rate of homogenization resulting from these events is too low to explain the observed levels of homogeneity (DOVER 1982). Thus, at least some of the phenomena affecting the structure of multigene famGeneral Hospital, Boston, Massachusetts 021 14. I Present address: Department of Molecular Biology, Massachusetts * To whom reprint requests should be addressed. The sequence data presented in this d c l e have been submitted to the EMBUGenBd Data Libraries under the accession number YOO614. ilies are faster, and/or are not strictly random, thus keeping only very similar copies as family members. Processes such as nonrandom recombination, transposition and RNA-mediated correction may be involved in concerted evolution (ARNHEIM 1983; STRACHAN, WEBB and DOVER 1985; MORZYCKAWROBLEWSKA et al. 1985). For instance, the dispersed structure of the Neurospora crassa 5s ribosomal gene family is best explained by transposition (SELKER et al. 1981). The occurrence of directional gene conversion has been demonstrated at the Ascobolus immersw b2 locus, where it leads to a preferential conservation of determined alleles, depending on the molecular nature of the mutation (HAMZA et al. 1986; HAMZA, NICOLAS and ROSSIGNOL 1987). Experimental approaches to the study of concerted evolution dynamics are difficult. Based on available structural criteria, multigenic families appear homogeneous and seem to have reached a state of equilibrium. Depending on the family, i.e., of nature of functional and structural constraints, different degrees of homogeneity are observed. Noncoding families, such as the 360 and 500 Drosophila satellites, display high levels of heterogeneity (STRACHAN, WEBB and DOVER 1985). This is not surprising since there are in this case few or no functional constraints. On the contrary, high levels of homogeneity are observed in the case of the coding families (ARNHEIM 1983). The Drosophila melanogaster 5s ribosomal genes constitute a multigenic family clearly displaying concerted evolution. In this species, the haploid genome contains approximately 160 clustered 5s genes (PROGenetics 118: 685-691 (April, 1988). 686 M.-L. Samson and M. Wegnez CUNIER and TARTOF 1975; PROCUNIER and DUNN 1978). The 5s genes are tandemly arranged (PROCUNIER and TARTOF, 1976) and are located in region 56F (PRENSKY, STEFFENSEN and HUGHES 1973). Each is 373 ? 7-bp long, divided into a 135 bp long transcribed region UACQ, JOURDAN and JORDAN 1977) and a 238 2 7-bp long nontranscribed spacer (TSCHUDI and PIRROTTA 1980). This size heterogeneity is due to the variable copy number (4, 5 or 6 copies) of the heptamer GCTGCCT downstream of the transcribed region. TSCHUDI and PIRROTTA (1980) sequenced four D. melanogaster 5s genes and reported three point differences between the sequenced copies. Three other base changes and one duplication of three bases were found by TSCHUDI, P RROTTA and JUNAKOVIC (1982) when sequencing another 5s gene. A two-nucleotide deletion in the coding region has also been reported by SHARP et al. (1984), who partially sequenced three 5s genes. All of these data demonstrate that the degree of homogeneity within the D. melanogaster 5s gene family is high, but that some heterogeneity of the family exists. In this paper, we analyze the heterogeneity of the Drosophila $06 strain, in which approximately 20% of 5 s units share a HindIII restriction site. This analysis provides some insight into the processes involved in the concerted evolution of the Drosophila 5s family. We will refer to the consensus 5s sequence determined by TSCHUDI and PIRROTTA (1980) as "standard." MATERIALS AND METHODS Drosophila strains: All strains were from the Gif collection. Several D. melanogaster isolates were used (Canton-S, 1~2, Charolles, Gruta, and Oregon-R), as well as the mutant strains min (PROCUNIER and TARTOF 1975), f l (COTI? et al. 1986) and the balancer stock CyO;TM31T(2;3)apy" (LINDSLEY and GRELL 1968). Molecular analyses: DNA from populations of adults was prepared by homogenizing about 100 flies in 4 ml of extraction buffer (100 mM EDTA, 200 pg/ml pronase, 50 mM Tris-HC1, pH 7.8) at 4". SDS was added to a concentration of 1% prior to incubation at 65" for 30 min. The pronase concentration was increased to 400 pg/ml, and the preparation was incubated for 3 hr at 37". After phenolchloroform extraction, DNA was precipitated by adding 3 ml of isopropanol at room temperature. The precipitate was rinsed with ethanol and redisolved in 500 pl of 1 mM EDTA, 10 mM Tris-HC1, pH 8. DNA was prepared from individual flies using an adaptation of the technique described by JUNAKOVIC, CANEVA and BALLARIO (1984). Each fly was homogenized with a glass stick in a 1.5-ml Eppendorf tube. Extraction buffer (500 pl of 0.2 M sucrose, 0.1 M Tris-HC1, pH 9.2, 50 mM EDTA, 0.5% SDS) was then added and the mixture heated for 10 min at 65". The mixture, after addition of potassium acetate (120 pl of a 5 M stock solution, pH 9), was kept 10 min on ice, then spun for 10 min. The supernatant, transferred to a fresh tube, was spun again for 10 min. DNA was then precipitated at -20" after addition of 400 p1 of isopropanol. The precipitate was ethanol rinsed and allowed to redissolve overnight in 50 ~1 10 mM Tris-HCI, pH 8, 1 mM EDTA prior to restriction digestion. Restriction digests were analyzed on 0.6% agarose gels. Southern analyses were performed under standard conditions (WAHL, STERN and STARK 1979; SMITH and SUMMERS 1980), except that nitrocellulose filters were baked for 45 hr at 80" in order to improve the retention of short restriction fragments (less than 600 bp). Scannin of 5s patterns obtained by Hind111 digestion of the go b: DNA was performed on several autoradiograms corresponding to different experiments and to various times of exposure. Genetical localization of 5s sequences in f l : y506 virgin females were crossed to CyO;TM3/T(2;3)apy" males and the F, males carrying Cy0 and T M 3 balancers were mated with min virgin females. Eight phenotypically distinguishable classes of F2 flies were recovered. They respectively bore all possible combinations of the first, second and third chromosomes of the $Ob strain. All were tested by Southern analysis for the presence or the absence of f l specific 5s sequences. In situ hybridizations were performed according to SPI. ERER et al. (1983). The 5s DNA probe (pBR7A including ten copies of D. melanogaster 5s genes, provided by V. PIRROTTA) was labeled with ['HIdGTP by nick translation. Cloning and sequencing of HindIII restriction fragments containing 5s s uences: Fifty micrograms of genomic DNA from the 7'' strain were cut with HindIII. The resulting restriction fragments were fractionated on a 0.6% agarose gel. Two size classes of DNA fragments, respectively around 375 and 750 nucleotide long, were electroeluted into dialysis bags, and then cloned into the HindIII site of pUC8. Screening for clones containing 5s sequences was performed using a D. melanogaster 5s probe, purified from pBR7A. Plasmid DNA was prepared according to the method described by BIRNBOIM and DOLY (1979). Drosophila sequences contained in the clones were purified on acrylamide gels (MAXAM and GILBERT 1980). A set of fragments with various 5' protruding ends were generated by cutting the inserts with appropriate restriction enzymes. Klenow labeling and additional restriction cuts provided fragments with a unique "P-labeled 3' end. These fragments were used in sequencing reactions as described by MAXAM and GILBERT (1980).