Fig 2: No increase of Htt co-localization with Rab22A.Representative confocal microscopic immunofluorescent images of Htt and Rab22A in STHdhQ7/Q111 cells and the quantification analysis of co-localization. Treatments with 1 µM RP-cAMP or 1 µM 8-pCPT-2'-O-Me-cAMP for 48 hr leads to no change of co-localization between Gpr52 and Rab22A. Red: Rab22A; Blue: DAPI; Green: Htt (using antibody 2051); Scale bars: 20 µm.DOI: http://dx.doi.org/10.7554/eLife.05449.012

Fig 3: Rabgap1l interferes with Gpr52-mediated Htt modulation.(A) Left: Representative western-blots of STHdhQ7/Q111 cells transfected with the Gpr52 siRNA (Gpr52_si2) vs the control siRNA, along with or without the Rabgap1l siRNA. Rabgap1l knock-down blunts the Gpr52's effect on the Htt level. Bar plot: HTRF quantification of the transfected cells as indicated, showing consistent results with the western-blots; statistical analyses by the two-tailed Mann–Whitney U-test. (B) mRNA levels of Gpr52 (left Y-axis) or Gapdh (right Y-axis) vs the ones of Rabgap1l (X-axis) upon transfection with HTT siRNAs or control (Neg). Both Gpr52 and Rabgap1l mRNA levels are lowered upon HTT knock-down, and a correlation between the lowering is observed. (C) Left panels: Representative confocal microscopic immunofluorescent images of Htt and Rab39B in STHdhQ7/Q111 cells. Treatments with 1 µM RP-cAMP or 1 µM 8-pCPT-2'-O-Me-cAMP for 48 hr leads to increased co-localization between Htt and Rab39B. Red: Rab39B; Blue: DAPI; Green: Htt (using antibody 2051); The lower panels are magnified images from the left for visualizing the co-localization. Yellow pixels indicate co-localization. Scale bars: 20 µm. Right panels: co-localization parameters including Pearson's coefficient and overlap coefficient (mean and S.E.M.). Numbers in indicate the number of cells analyzed for each treatment from five or more biological replicates. (D) A model figure explaining the modulation of Htt levels by Gpr52. Gpr52 increases cAMP when activated, which leads to activation of an unknown GEF (Guanine Exchange Factor) that activates the downstream Rab39B protein. Rab39B then co-localize with Htt and translocate it to the ER, where the proteasomal degradation is prohibited due to lack of proteasomes in the ER. The Gpr52 gene locates in an intron of the Rabgap1l gene, which expresses the GAP (GTPase Activating Protein) for Rab39B, and thus blocks the modulation. Thus, Gpr52 and Rabgap1l provide balanced regulation of Htt in striatal cells, and the shared genomic loci may facilitate their balance in modulating Htt levels via co-regulated expression in striatal cells.DOI: http://dx.doi.org/10.7554/eLife.05449.010

Fig 4: Constitutively active Rab39B blunts Gpr52's effect.Left: Representative western-blots of Htt (ab1), consitutively active Rab39B (anti-FLAG for Rab39B_CA), and the loading control (Tubulin). ‘-’ indicates the empty vector or scrambled siRNA controls. Right: quantification of the normalized Htt signals, n = 4.DOI: http://dx.doi.org/10.7554/eLife.05449.013

Fig 5: Rabgap1l genomic information and siRNA validation.(A) Genomic loci of Rabgap1l and Gpr52. (B) qPCR quantification of the Rabgap1l mRNA level of STHdh cells transfected with Rabgap1l siRNAs or the non-targeting control siRNA (Neg). 50–80% knock-down could be achieved by siRNA transfection in these cells. (C) Western-blot (left) and HTRF (right) experiments showing that Rabgap1l knock-down by siRNA increases the Htt level in the STHdh cells. Hdh5 and B01 are Htt siRNAs used as positive controls. For HTRF, the 2B7/2166 antibody pair was used. Data are plotted as mean and S.E.M, n = 16 for non-targeting siRNA control (Neg) samples, and n = 12 for Rabgap1l siRNA transfected samples. ‘***’: P < 0.001 by the two-tailed Mann–Whitney U-test.DOI: http://dx.doi.org/10.7554/eLife.05449.011