Figure 1.
Imaging-based assay of BMP signaling for RNAi screening in Drosophila S2 cells.
(A) Imaging-based analysis of BMP signals in Drosophila S2 cells. Dpp, Scw and Flag-tagged Mad were co-transfected in S2 cells together with dsRNA in 384-well plates. Mad expressions were detected with anti-Flag/Alexa 488 anti-mouse IgG antibodies, and BMP signals were detected with anti-pMad/Alexa 647 anti-rabbit IgG antibodies. dsRNA against gfp served as a control to rule out any nonspecific effects caused by RNAi transfection, and the BMP type II receptor put served to validate that its knockdown inhibits pMad expression. (B) Imaging analysis of the BMP signals in S2 cells reflects the relative intensities that were regulated by various RNAi approaches (ds-gfp, ds-sax, ds-tkv, and ds-put). The relative intensities of BMP signals by RNAi from four independent experiments were measured with imaging-based analysis. The intensity of ds-gfp was set at 100%. ** indicates P<0.01 (ds-sax, ds-tkv, or ds-put was compared to ds-gfp). (C) The protocol for genome-wide RNAi screening of BMP signaling. The Arrayscan program enabled to calculate the number of DAPI (channel 1) as a total cell number; the number of Flag-Mad-positive cells (channel 2) within the DAPI-positive cells was counted as the Mad-positive cells, and then the number of pMad-positive cells (channel 3) within the Flag-positive cells was counted as the pMad-positive cells. The relative intensity of each sample was calculated as a percentage of pMad staining against Mad (Flag)-positive cells.
Table 1.
Candidate genes involved in BMP signaling in Drosophila S2 cells.
Figure 2.
Ter94 is involved in BMP signaling in Drosophila S2 cells.
(A) BMP signals are regulated by Ter94 in S2 cells. The loss of BMP signaling caused by Ter94 (3′-UTR) RNAi was restored by the ectopic expression of Ter94. pAW-Ter94 was used for the expression of Ter94 in S2 cells; whereas pBS served as a control vector. Tubulin was used for a loading control. (B) The relative intensities of pMad signals (compared to the pMad signal of ds-lacI) were measured with Western blotting. pMad signals were normalized by Tubulin expression; ds-lacI intensity was set at 100%. ** indicates P<0.01. (C) The relative protein level of Ter94 (compared to the protein level of ds-lacI-treated cells) was quantified. The Ter94 level was normalized by Tublin expression; ds-lacI intensity was set at 100%. ** indicates P<0.01. (D) Ter94 is involved in BMP signaling upstream of Smurf or independent of Smurf in S2 cells. Tubulin was used for a loading control. (E) The relative intensities of pMad signaling (compared to the pMad signal of ds-lacI) were measured with Western blotting. pMad signals were normalized by Tubulin expression; ds-lacI intensity was set at 100%. * indicates P<0.05 and ** indicates P<0.01. Note that the pMad signal of the double knockdown of Ter94 and Smurf is similar to that of Ter94 RNAi. (F) The loss of BMP signaling caused by Ter94 RNAi was restored by the ectopic expression of hVCP. The relative intensities of pMad signaling (compared to the pMad signal of ds-lacI) were measured with Western blotting. pAW-Ter94 or pAW-hVCP was used for the expression of Ter94 or hVCP in S2 cells, respectively. pMad signals were normalized by Flag-Mad expression; ds-lacI intensity was set at 100%. ** indicates P<0.01. dsRNA against lacI served as a control to rule out any specific effects caused by RNAi transfection, and put was used to validate that its knockdown inhibits pMad signals.
Figure 3.
VCP is involved in BMP signaling in mammalian cells.
(A) Alkaline phosphatase (ALP) assay in C2C12 cells. BMP6-induced ALP activity was abolished by the knockdown of Smad1 or VCP. ** indicates P<0.01. (B) BMP-induced activity of the luciferase reporter (BRE-Luc) in KS483 cells. BMP6-induced luciferase activity was inhibited by the overexpression of VCPR155H but was activated by the overexpression of VCPWT. * indicates P<0.05.
Figure 4.
Ter94 is involved in BMP signaling in vivo.
(A) Ter94 alleles and dpp show genetic interactions. The dpp null mutant is haploinsufficient. In a Smurf mutant background, the dpp heterozygote is partially viable. In Ter94 and Smurf double mutants, the viability of the dpp heterozygote is less than that of the Smurf mutant alone. (B–H) A dorsal view of race mRNA in early stage-6 embryos. dpp null (B), dpp heterozygote (C), wild-type phenotype (D), three copies of Dpp (+/P[dpp-P23]: E), transheterozygous for dpp null and Smurf mutant alleles (dppH46 Smurf+/dpp+ SmurfC15: F), transheteorzygous for dpp null, Smurf mutant and Ter94 mutant alleles (dppH46 Ter94+ Smurf+/dpp+ Ter94EY03468 SmurfC15: G, dppH46 Ter94+ Smurf+/dpp+ Ter94k15002 SmurfC15: H). (I) Number of race staining cells in early stage-6 embryos. ** indicates P<0.01 (compared with wild-type phenotype yw).
Table 2.
Genetic interactions between Ter94 and the BMP signals.