The polytene chromosomes of Rhynchosciara americana and R. hollaenderi, a pair of sibling species in the americana-like group of Rhynchosciara, were compared using a number of techniques, including in situ hybridization. With classical cytological techniques, the only differences observed were in the morphology of centromeric and telomeric heterochromatin, in the size of a DNA and RNA puff, and in the presence of an inversion polymorphism in R. hollaenderi. However, after in situ hybridization with rDNA and poly-r(A) probes, differences between the two species appeared at a number of sites. Differences in poly-r(A) sites were especially informative in establishing phylogenetic relationships between these two species and a third species currently being examined from this group. Chromosomal evolution between these species appears to have occurred mainly through differential amplification and transposition of repetitive sequence DNA, of which dA:dT tracts are an important component. The R. hollaenderi karyotype is tentatively considered more ancestral than that of R. americana because it has features present in the third Rhynchosciara species. Explanations for the monomorphisms observed in Rhynchosciara species and mechanisms of speciation in the group are considered within the context of the species' complex behavior.

Antibodies to specific nucleic acid conformations are amongst the methods that have allowed the study of non-canonical (Watson-Crick) DNA structures in higher organisms. In this work, the structural limitations for the immunological detection of DNA.RNA hybrid duplexes were examined using specific RNA homopolymers as probes for homopolymer polydeoxyadenylic acid (poly(dA)).polydeoxythymidylic acid (poly(dT))-rich regions of Rhynchosciara americana (Diptera: Sciaridae) chromosomes. Anti-DNA.RNA duplexes did not react with the complex formed between chromosomal poly(dA) and exogenous polyuridylic acid (poly(rU)). Additionally, poly(rU) prevented the detection of polyadenylic acid.poly(dT) hybrid duplexes preformed in situ. These results raised the possibility that three-stranded structures rather than duplexes were formed in chromosomal sites. To test this hypothesis, the specificity of antibodies to triple-helical nucleic acids was reassessed employing distinct nucleic acid configurations. These antibodies were raised to the poly(dA).poly(rU).poly(rU) complex and have been used here for the first time in immunocytochemistry. Anti-triplex antibodies recognised the complex poly(dA).poly(rU).poly(rU) assembled with poly(rU) in poly(dA).poly(dT)-rich homopolymer regions of R. americana chromosomes. The antibodies could not detect short triplex stretches, suggesting the existence of constraints for triple-helix detection, probably related to triplex tract length. In addition, anti-poly(dA).poly(rU).poly(rU) antibodies reacted with the pericentric heterochromatin of RNase-treated polytene chromosomes of R. americana and Drosophila melanogaster. In apparent agreement with data obtained in cell types from other organisms, the results of this work suggest that significant triple-helix DNA extensions can be formed in pericentric regions of these species.

The characterisation of sequences at chromosome ends of Rhynchosciara americana was continued with the screening of a genomic library using as a probe a short repeat identified in a previous report (M-22, 22 bp) which was found to be specific for noncentromeric termini of this species. Simple repeats, complex tandem and apparently dispersed repeats were present in the genomic clones analysed. Repetitive sequences do not define individual chromosome tips as they were found in all noncentromeric ends. A novel and unusually short tandem repeat type for dipteran chromosome ends (named M-16) composed of 16 nucleotides and frequently associated with M-22 arrays was characterised in this work. Islands of M-16 and M-22 tandem repeats were found in all the genomic clones analysed. Individual probes representative of each repetitive element hybridised not only to all noncentromeric ends of R. americana chromosomes but also to inter-telomeric bridges. This contrasted with the other repeat types which displayed sub-telomeric localisation as seen by double detection of hybridised probe and telomeric reverse transcriptase. Some stretches composed of M-16 and M-22 tandem repeats localised in different regions of the analysed genomic clones were either identical or showed sequence similarity that was unexpectedly higher than the mean sequence similarity observed among repeats within each of their tandem arrays. The occurrence of segmental duplications, as deduced by sequence analyses involving the two repeats that appeared to reach chromosome ends, might indicate the involvement of this type of duplication process in the chromosome end maintenance in this species.

An antiserum (called AScE/D) against the semiconserved D-domain of a Chironomus tentans ecdysone receptor protein (cEcR) gave indirect immunofluorescence signals at DNA puff sites in Trichosia pubescens. The signals varied in maximum intensity at different DNA puff sites. Control experiments using the secondary rhodamine-labeled anti-rabbit IgG alone, preimmune serum, affinity purified AScE/D (called pABcE/D) and AScE/D preabsorbed with expressing bacterial extract or highly purified bacterially expressed cEcR indicated that the signals obtained at these chromosomal sites were likely to be due to specific interaction between an endogenous sciarid EcR and antibodies against cEcR. This conclusion was supported by observation of signals at certain Ec-inducible primary RNA puff sites. AScE/D signals began to appear at DNA puff sites during L3, the stage when amplification initiates, but at most sites their mean intensity was low and not statistically significant. Sites with AScE/D signals of significant mean intensity at this stage already showed evidence of transcription. The number and strength of transcription signals increased during L4. Comparison of the developmental course of signals for AScE/D, DNA synthesis, RNA presence/synthesis, and puff size for several DNA puffs during late larval- prepupal development showed a closer relationship of AScE/D signals with the initiation of RNA synthesis than with the initiation of DNA synthesis. Therefore, although we cannot absolutely eliminate a direct involvement of EcR in the amplification process at some sites, this investigation gives stronger support for its direct involvement in transcription. Since AScE/D signals are observed at DNA puff sites from the time the latter begin amplification/transcription through their regression, it appears that Ec and EcR are necessary as a sustained stimulus at these regions.

The puffing pattern in polytene chromosomes from salivary glands of fourth-instar larvae of Trichosia pubescens was studied. It was found that the puffing pattern is rather constant during most of larval life but changes continuously in a precise sequential order during the period preceding pupation. During this period, characterized by drastic changes in the puffing pattern, amplification of specific genes and expansion of the DNA-puffs occur. The pattern of protein synthesis in the salivary gland has also been studied by means of SDS-polyacrylamide gel electrophoresis and fluorography. It changes drastically and continuously during the phases preceding pupation. These changes in protein synthesis could be correlated with the changes in the puffing pattern, essentially with the activity of the DNA-puffs.

The chromosomal localization of ribosomal DNA (rDNA) was studied in polytene and diploid tissues of four sciarid species, Trichosia pubescens, Rhynchosciara americana, R. milleri and Schwenkfeldina sp. While hybridization to mitotic chromosomes showed the existence of a single rDNA locus, ribosomal probes hybridized to more than one polytene chromosome region in all the species analyzed as a result of micronucleolar attachment to specific chromosome sites. Micronucleoli are small, round bodies containing transcriptionally active, probably extrachromosomal rDNA. In T. pubescens the rDNA is predominantly localized in chromosome sections X-10 and X-8. In R. americana the rDNA is frequently found associated with centromeric heterochromatin of the chromosomes X, C, B and A, and also with sections X-1 and B-13. Ribosomal probes in R. milleri hybridized with high frequency to pericentric and telomeric regions of its polytene complement. Schwfenkfeldina sp. displays a remarkably unusual distribution of rDNA in polytene nuclei, characterized by the attachment of micronucleoli to many chromosome regions. The results showed that micronucleoli preferentially associate with intercalary or terminal heterochromatin of all sciarid flies analyzed and, depending on the species, are attached to a few (Trichosia), moderate (Rhynchosciara) or a large (Schwenkfeldina sp.) number of polytene chromosome sites.

We have studied the distribution of DNA X RNA hybrids on polytene chromosomes with the aid of a goat antibody against DNA X RNA hybrids using the immunofluorescence technique. Fixed polytene chromosomes of the sciarid Trichosia pubescens (Diptera) show distinct, stage-specific labelling patterns throughout larval development. Controls for the staining procedure - including preincubation with hybrid-specific endoribonuclease H - prove that DNA X RNA hybrids are present on fixed chromosomes. They are revealed only under mild fixation conditions which do not efficiently immobilize all chromosomal proteins, indicating that some proteins have to be removed to make the antigens accessible to antibody. Certain fixation conditions may also cause local denaturation of chromosomal DNA, and some hybrids may possibly form during specimen preparation. After incorporation of radioactive uridine, a combination of phase contrast, fluorescent, and autoradiographic images of one and the same chromosomal preparation demonstrates that hybrid fluorescence is confined to transcriptionally active regions. Two puff classes can be distinguished. The first binds antibody and includes most RNA puffs and all DNA puffs so far studied; the second, comprising some RNA puffs, does not show bright fluorescence in spite of the fact that RNA synthesis is high as revealed by 3H-uridine incorporation. DNA X RNA hybrids are not found at DNA puff sites during the DNA amplification period; these sites contain detectable hybrids only when transcription is taking place. - Combination of the fluorescent technique with its excellent resolution and autoradiography should be helpful in studying detailed topological aspects of transcriptionally active chromosomal regions.

No abstract