Fig 1.
Effect of the Top3๐ท mutations on the transcriptome of 0โ2 hrs. old embryos.
A) The left and central panels show the changes compared to the wild-type (WT) transcriptome using adjusted p-values (Adjp) < 0.05 and log2-fold change < โ1 (downregulated) and >1 (upregulated). The Venn diagrams also show the allocation of the data to the 3 different mutants. Using the more stringent criteria Adjp < 0.0002 and log2-fold change < โ1, the Venn diagram on the right shows the fractions of the genes with reduced transcript levels for the 3 alleles. n: is the number of different genes identified in the three mutants. B) Differentially expressed transcript isoforms in the Top3๐ฝ26 null mutant compared to the wild type (WT). Adjp < 0.05 and absolute log2-fold change โฅ 1. The Venn diagram shows which fraction is also differentially expressed in the point mutants when analyzed with the same stringency parameters (left panel). The fractions of the differentially expressed transcript isoforms expressed at lower levels in the mutants are shown in the right diagram. CโE) RNA features of the genes, which produce the transcript isoforms, that depend on Top3๐ฝ to reach normal expression levels. C) Transcript isoforms from genes that produce long UTRs are over-represented in the group of RNAs that depend on Top3๐ฝ and its Tyr332 (Y332) residue to reach normal expression levels. D) Transcript isoforms from genes that encode longer CDS are expressed at higher levels in the wild type than in the Top3๐ฝ26 and Top3๐ฝY332F mutants. E) Transcript isoforms from genes that produce longer pre-mRNAs are expressed at higher levels in the wild type than in the Top3๐ฝ26 and Top3๐ฝY332F. Expression levels were compared to the Top3๐ฝ+ (WT) control in 0โ2 hrs. old embryos. p-value < 0.0001=****, p-value < 0.001=***, p-value < 0.01=**, p-value < 0.05=*. n = 50 for the null mutant Top3๐ฝ26 and Top3๐ฝY332F and n = 5 for Top3๐ฝโRGG.
Fig 2.
Characteristics of embryonic RNA targets of Top3ฮฒ
::eGFP. A) Longer RNAs are more likely targets of the enzymatic activity of Top3ฮฒ because they are more enriched in the IP with the wild-type Top3๐ฝ compared to the Top3๐ฝY332F. r measures the strength of the correlation and p stands for p-value. BโE) Transcript features of the top 46 embryonic RNA targets of Top3ฮฒ::eGFP. The complete sets of RNAs detected in the wild-type RNA-IP served as the standard. p-value < 0.0001=****, p-value < 0.001=***, p-value < 0.01=**, p-value < 0.05=*.
Fig 3.
Expression and localization of gene products of Top3ฮฒ targets.
The posterior poles of the embryos are shown to the right. Comparing shot mRNA and Shot protein localization in syncytial (shot mRNA) and cellularizing (Shot protein) wild-type and Top3๐ฝ mutant embryos shows lower signal levels and less localized signals in the posterior part of the embryos in the null mutant and Top3๐ฝY332F. Both were less reduced in Top3๐ฝโRGG mutant embryos. Note the shot mRNA localization in the pole cells in the wild type and Top3๐ฝโRGG mutant. Dhc64C localization at the posterior side of the somatic part of the embryo (arrowhead) was not detectable in the null mutant and Top3๐ฝY332F and reduced in Top3๐ฝโRGG mutant embryos. Kst localization at the apical membrane (arrowhead) was reduced in null mutants and Top3๐ฝY332F compared to the wild type. Top3ฮฒ protein is localized similarly to its targets. It is, however, absent from the null mutant which reveals the background levels of the staining. Top3ฮฒ is also present in the wild-type nuclei and in Top3๐ฝY332F but absent from nuclei in the Top3๐ฝโRGG embryo. The same imaging settings were used for the same protein or mRNA staining in the different genotypes. Exceptions are the insets which were overexposed to reveal the embryo. The scale bar is 10ยตm for the shot mRNA pictures and 20ยตm for the other micrographs.
Fig 4.
Overlap between reduced embryonic RNAs in Top3๐ท26 and the fly homologs of differentially expressed RNAs in several cancers and neurological diseases associated with human Top3๐ท.
A, B) Venn diagrams of the genes shared between the list of reduced RNAs in Top3๐ฝ26 embryos and the fly homologs of differentially expressed RNAs of the listed diseases. Numbers and frequencies are listed. The probability of random overlap is shown in % at the bottom left of each Venn diagram. C) Drosophila genes depending on Top3๐ฝ for normal expression levels in embryos were placed into a Venn diagram such that their position shows in which neurological disease their homologs are differentially expressed (note that JME was not included). Rop appeared in all lists. D) Drosophila genes depending on Top3๐ฝ for normal expression levels in embryos were placed into a Venn diagram such that their position shows in which cancer their homologs are differentially expressed. E) List of the 50 genes that depend on Drosophila Top3๐ฝ for their normal expression levels in embryos and whose homologs are also affected in all listed cancers associated with human Top3๐ฝ alterations.
Fig 5.
Aging effects of Top3๐ท mutants.
A) Climbing assays of wild type, Top3๐ฝโRGG, Top3๐ฝY332F, and Top3๐ฝ26 flies. A two-way ANOVA analysis was performed. The graph bars show the mean climbing index (C.I.) for the different genotypes. Error bars represent the standard error of the mean (S.E.M). Each dot represents a calculated individual C.I. The adjusted p-values for each significant difference are indicated on the graph. (p-value < 0.0001=****, p-value < 0.001=***, p-value < 0.01=**, p-value < 0.05=*.) B) In adults, null mutant NMJs contain more branches than the wild type, indicative of a compensatory mechanism. Arrowheads point to the lost synaptic integrity of the dorsal longitudinal flight muscle (DLM). NMJ synapses were visualized by staining motor neurons with anti-HRP antibodies. The progressive denervation of the DLM is particularly evident in the null and the Y332F mutant in the adult thoracic NMJs. The scale bar represents 20ยตm. All panels show maximum intensity projections with a z-step size of 0.7ฮผm. C) Life span shortening by Top3๐ฝ mutations. Top3๐ฝ26null and Top3๐ฝY332F mutants reduce long-term survival. Top3๐ฝโRGG, which showed only a minor reduction compared to a wild-type strain in other experiments, served as a control because it was induced in the same background as Top3๐ฝY332F.
Fig 6.
Top3๐ท counteracts the degenerative effect of (G4C2)49 repeat RNAs in the eye.
Expression of the UAS-(G4C2)49 transgene in wild-type (+) and Top3๐ฝ mutant flies using the GMR-GAL4 driver. All Top3๐ฝ mutants enhanced the degenerative phenotype of the (G4C2)49 RNA. The Top3๐ฝโRGG allele caused mild disruptions, and Top3๐ฝ26 and Top3๐ฝY332F mutants caused strong degeneration. The presented micrographs were from one set of experiments carried out in parallel; at least three replicates were performed. UAS-(G4C2)49 and GMR-GAL4 lines were heterozygous. + indicates the control genotype Top3๐ฝ+; the mutant alleles are shown as superscripts.
Fig 7.
Comparing Drosophila Top3ฮฒ targets and effects to the corresponding FMR1/FMRP and mammalian Top3ฮฒ data.
A) A previous mammalian study produced a list of overlapping mouse FMRP targets in brain extracts and Top3ฮฒ targets in HeLa cells [1]. The Venn diagram compares the list of Drosophila homologs of these mammalian targets to our list of direct targets of Top3ฮฒ. The list of the Drosophila homologs that were also found in both mammalian interactor screens is shown with the characteristics of the Drosophila gene. The length measurements stem from the longest pre-mRNA isoforms (S11 Table). B) Comparing RNASeq data from 0โ2 hrs old embryos from the Top3๐ฝ26 null mutant (Adjp < 0.0002; log2-fold change < โ1 and >1, respectively, according to S1 Table) and the FMR1 mutant [35]. The analysis considers reads per gene. Lists of the genes identified in the overlapping datasets are presented in the supplements (S13 Table). C) Genes encoding transcript isoforms that depend on Top3๐ฝ for their normal accumulation in young embryos (S4 Table) and transcripts that depend on FMR1 for their translation in quiescent oocytes [36]. The Venn diagram shows the overlap between the two datasets and the genes in the overlap are listed. D) Comparing RNASeq data from 0โ2 hrs old Top3๐ฝ26 embryos (Adjp < 0.0002; log2-fold change < โ1; S1 Table) and the FMR1 RNAi effect on translational repression [36].