DmTRF4-1 and DmTRF4-2 are candidate genes for the homologue of S. cerevisiae Trf4/5. (A) Schematic representation of DmTRF4-1, DmTRF4-2, and ScTrf4. The location of the nucleotidyltransferase domain and the PAP-associated domain are indicated by dark gray and light gray boxes, respectively. Numbers to the right of the boxes indicate the total number of amino acids in the respective proteins. Homology/similarity (as a percentage) in the catalytic domains between DmTRF4-1 or DmTRF4-2 and ScTrf4 are indicated below each box, respectively. (B) Comparison of DmTRF4-1, DmTRF4-2, and ScTrf4. An amino acid alignment of the catalytic domain, consisting of the nucleotidyltransferase domain and the PAP-associated domain, of DmTRF4-1, DmTRF4-2, and ScTrf4. Identical and similar amino acid residues are boxed in black and gray, respectively. The two conserved catalytic aspartate residues are indicated by asterisks. The alignment was carried out using the CLUSTAL W program. (C) Expression of DmTRF4-1 and DmTRF4-2 during the Drosophila development. Northern blotting was successively performed on the same membrane. A portion (30 μg) of total RNA was applied to each lane. Rp49 mRNA served as a loading control.

DmTRF4-1 knockdown and DmTRF4-2 knockout mutants. (A) Physical map of the genomic location of DmTRF4-2 (CG17462). In CG17462^NP2505, P-element was inserted in the 5′-untranslated region of the CG17462 gene. Black and gray arrows indicate predicted genes. (B) RT-PCR was performed for total RNAs from WT (lane 1) and DmTRF4-2 mutant (CG17462^NP2505; lane 2) third-instar larvae. Expression of Act5C was used as an internal control. The cycle numbers used are indicated. (C) RT-PCR was performed for total RNAs from Act5c-GAL4/+ (control; lane 1) and Act5C-GAL4/UAS-Dmtrf4-1 IR (DmTRF4-1 KD; lane 2) third-instar larvae. Expression of Act5C was used as an internal control. The cycle numbers are indicated. The relative amount of the DmTRF4-1 transcript in Act5c-GAL4/UAS-DmTRF4-1 IR to that in Act5C-GAL4/+, normalized to Act5C mRNA, was also quantified graphically. The data represent the mean of three independent measurements. Error bars indicate ± the standard deviation. (D) Knockdown of the DmTRF4-1 level by ubiquitous expression of the DmTRF4-1 double-stranded RNAs in living flies (Act5C-GAL4/UAS-Dmtrf4-1 IR) caused third-instar-larval or early pupal lethality. Few transgenic flies survived until the late pupal stage, but when this happened the body of the pupa was significantly darkened. Note the darkened imaginal body in the DmTRF4-1 knockdown mutant, as indicated by the arrowhead (DmTRF4-1 KD; panel a). Early pupal lethality was not observed in control flies (Act5C-GAL4/+) (Control; panel b).

Effects of overexpression of the DmTRF4-1 in D. melanogaster. (A) RT-PCR was performed for total RNAs from Hsp70-GAL4/+ (control; lane 1), Hsp70-GAL4/UAS-Dmtrf4-1 IR (DmTRF4-1 KD; lane 2), Hsp70-GAL4/+; UAS-DmTRF4-1/+ (DmTRF4-1 OE; lane 3), and Hsp70-GAL4/UAS-Dmtrf4-1 IR; UAS-DmTRF4-1/+ (DmTRF4-1 KD and OE; lane 4) third-instar larvae incubated for 3 h at 37°C. Expression of Act5C was used as an internal control. The cycle numbers are indicated. The relative amount of the DmTRF4-1 transcript in each sample to that in Hsp70-Gal4/+, normalized to Act5C mRNA, was also quantified graphically. The data represent the mean of three independent measurements. Error bars indicate + the standard deviation. (B) The transgenic fly lines crossed the GMR-GAL4 driver, which induced GAL4 posterior to the MF of the eye disc. No phenotype were observed in control flies (GMR-GAL4/+; +; +) (control; panel a) and knockdown flies DmTRF4-1(GMR-GAL4/+; UAS-Dmtrf4-1 RI/+; +) (DmTRF4-1 KD; panel b). Overexpression of the DmTRF4-1 caused a rough eye phenotype (GMR-GAL4/+; +; UAS-DmTRF4-1/+), as indicated by the arrowhead (DmTRF4-1 OE; panel c). Knockdown of the DmTRF4-1, while overexpressing DmTRF4-1, suppressed a rough eye phenotype induced by overexpression of the DmTRF4-1 (GMR-GAL4/+; UAS-Dmtrf4-1 RI/+; UAS-DmTRF4-1/+) (DmTRF4-1 KD and OE; panel d). (C) Overexpression of DmTRF4-1 in the wing imaginal disc using the MS1096-GAL4 driver caused an underdeveloped wing (MS1096-GAL4/+; +; UAS-DmTRF4-1/+), as indicated by the arrowhead (DmTRF4-1 OE; panel a). No phenotype was observed in control flies (MS1096-GAL4/+) (control; panel b). (D) Overexpression of DmTRF4-1 in the whole eye imaginal disc using the ey-GAL4 driver resulted in a headless phenotype, as indicated by the arrowhead, and caused lethality at the pupal stage (ey-GAL4/+; UAS-DmTRF4-1/+) (DmTRF4-1 OE; panel a). No phenotype was observed in control flies (ey-GAL4/+) (control; panel b). These were the contents of the dissected pupa.

Overexpression of DmTRF4-1 causes extraordinary snRNA polyadenylation, which is DmRrp6 dependent. Total RNA (5 μg) (total; left panels), purified poly(A)⁺ RNA (1 μg) [poly(A)⁺; middle panels], deadenylated RNA (1 μg) [poly(A)⁻; right panels] were isolated from third-instar larvae of Hsp70-GAL4/+ (control; lanes 1, 7, and 13), Hsp70-GAL4/+; DmRrp6^f07001/+ (DmRrp6 KD; lanes 2, 8, and 14), Hsp70-GAL4/UAS-Dmtrf4-1 IR (DmTRF4-1 KD; lanes 3, 9, and 15), Hsp70-GAL4/UAS-Dmtrf4-1 IR; DmRrp6^f07001/+ (DmTRF4-1 KD and DmRrp6 KD; lanes 4, 10, and 16), Hsp70-GAL4/+; UAS-DmTRF4-1/+ (DmTRF4-1 OE; lanes 5, 11, and 17), Hsp70-GAL4/+; UAS-DmTRF4-1/DmRrp6^f07001 (DmTRF4-1 OE and DmRrp6 KD; lanes 6, 12, and 18) after induction of GAL4 expression by heat shock for 3 h. Specific radiolabeled probes (Table 1) for U1 (A), U2 (B), U4 (C), U5 (D), U6 (E), snRNA and Ribosomal protein S29 (RpS29), as a loading control (F), were hybridized and detected by autoradiography. The migration positions of the regular snRNA and polyadenylated snRNA are indicated on the right of the panels. The ratio of the polyadenylated snRNAs in Hsp70-GAL4/+; UAS-DmTRF4-1/DmRrp6^f07001 (DmTRF4-1 OE and DmRrp6 KD) to those in Hsp70-GAL4/+; UAS-DmTRF4-1/+ (DmTRF4-1 OE) are shown in lane 12. Each band intensity was quantified by using a BAS-3000 imaging analyzer and normalized to RpS29 mRNA.

DmTRF4-1, but not DmTRF4-2, exhibits PAP activity. (A) Purified GST fusion protein containing WT DmTRF4-1 (GST-DmTRF4-1 WT; lane 1) and mutant DmTRF4-1 (GST-DmTRF4-1 DD328,330AA; lane 2), in which aspartate residues at positions 328 and 330 have been changed to alanine, and WT DmTRF4-2 (GST-DmTRF4-2 WT; lane 3) were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis on a 7.5 or 12.5% polyacrylamide gel and stained with Coomassie brilliant blue (1.0 μg of each). The positions of the purified proteins are indicated by an asterisk. (B) DmTRF4-1 has PAP activity, but DmTRF4-2 has no activity in vitro. GST-DmTRF4-1 WT (lane 2 to 3) (250 ng), GST-DmTRF4-1 DD328,330AA (lane 4) (500 ng), and GST-DmTRF4-2 WT (lane 5 to 6) (350 ng) were incubated with 5′-end-labeled oligo(A)₁₅ in the presence of 0.5 mM Mn²⁺ at 30°C. A control reaction with no enzyme is shown in lane 1. (C) Mg²⁺ versus Mn²⁺ dependence. GST-DmTRF4-1 WT (250 ng) was incubated with 5′-end-labeled oligo(A)₁₅ in the presence of either Mg²⁺ (5 mM; lane 2), Mn²⁺ (0.5 mM; lane 3), or both Mg²⁺ and Mn²⁺ ions (5 and 0.5 mM, respectively; lane 4) at 30°C for 30 min. GST-DmTRF4-2 WT (350 ng) was incubated with 5′-end-labeled oligo(A)₁₅ in the presence of either Mg²⁺ (5 mM; lane 6), Mn²⁺ (0.5 mM; lane 7), or both Mg²⁺ and Mn²⁺ ions (5 and 0.5 mM, respectively; lane 8) at 30°C for 30 min. A control reaction with no cations is shown in lanes 1 and 5. (D) Nucleotide substrate dependence. GST-DmTRF4-1 WT (250 ng) was incubated with 5′-end-labeled oligo(A)₁₅ in the presence of either ATP (lane 1), GTP (lane 2), CTP (lane 3), or UTP (lane 4) at 30°C for 30 min. All PAP polymerase reaction products were resolved on a 20% polyacrylamide-8 M urea gels and visualized by using autoradiography.

Subcellular localization of DmTRF4-1 in living S2 cells. The S2 cells overexpressing V5-tagged DmTRF4-1 DD328,330AA was immunostained with anti-V5 antibody (red) (A and A′) and antifibrillarin (green) (B and B′), as a nucleolar marker, and stained with DAPI (blue) (C and C′). The merge is shown in (D and D′). The lower panels are a high-magnification view.

DmRrp6 mutant suppresses a rough eye phenotype induced by the overexpression of DmTRF4-1. (A) Physical map of the genomic location of DmRrp6. In DmRrp6^f07001, the P-element was inserted in the second intron of the DmRrp6 gene. Black and gray arrows indicate predicted genes. (B) RT-PCR was performed for total RNAs from WT (lane 1) and DmRrp6 mutant (DmRrp6^f07001; lane 2) third-instar larvae. A homozygote of the DmRrp6^f07001 is inviable. Expression of Act5C was used as an internal control. The cycle numbers used are indicated. The relative amount of the DmRrp6 transcript in DmRrp6^f07001 to that in the WT, normalized to Act5C mRNA, was also quantified graphically. The data represent the mean of three independent measurements. Error bars indicate ± the standard deviation. (C) DmRrp6 mutation suppresses the phenotype induced by overexpression of DmTRF4-1. No phenotype was observed in the WT fly (control; panel a) and DmRrp6 mutant (DmRrp6^f07001) (DmRrp6 KD; panel b). Overexpression of DmTRF4-1 posterior to the MF of the eye disc caused a rough eye phenotype (GMR-GAL4/+; +; UAS-DmTRF4-1/+) (DmTRF4-1 OE; panel c). DmRrp6 mutation slightly suppressed the rough eye phenotype induced by overexpression of DmTRF4-1 (GMR-GAL4/+; +; UAS-DmTRF4-1/DmRrp6^f07001) (DmTRF4-1 OE and DmRrp6 KD; panel d). Note that the rough eye phenotype is indicated by arrowheads.