Fig 1.
Identification of TTR as a novel interacting protein for the GABAA receptor δ subunit.
(A) Alignment of the amino-acid sequence of full length mouse transthyretin (NP_038725) with the sequence of the positive clone obtained from our yeast two-hybrid screening. (B-C) Co-immunoprecipitation of the HA-tagged TTR and the myc-tagged extracellular domain of the δ-subunit (extra-δ) after co-expression in HEK cells. (D) Confocal images showing the co-localization of the α6β3δ-receptors (green) and TTR (red) after co-expressed in HEK cells. The arrows show the coloclization sites. Scale bar, 10 μm. (E) The quantification of colocalization of the transfected TTR and α6β3δ receptors in HEK cell by intensity correlation analysis in Image J. The mean intensity correlation quotient(ICQ) is between 0 and +05, which means they are dependent staining.
Fig 2.
TTR interacts with the GABAA receptor δ subunit in vivo.
(A) Total staining of endogenous GABAA receptor δ subunit (red) and TTR (green) in cerebellar granule cells in culture. Boxed areas were enlarged for better view of the colocalization. The arrows point the coloclization sites. Scale bar: 5 μm. (B) Live cell staining of the endogenous surface level of TTR (green) and the GABAA receptor δ subunit (red) in cultured cerebellar granule cells. The arrows point the coloclization sites. Scale bar: 5 μm. (C) Live cell staining of cortical neurons overexpressed with hTTR (green) and rat δ subunit (red), revealing partial colocalization of the two signals. The arrows point the coloclization sites. Scale bar: 10 μm. (D) Co-immunoprecipitation of TTR and δ-receptors from mouse cerebellum extract. Brain lysate was immunoprecipitated with sheep anti-TTR antibody or normal sheep IgG as a control, and then immunoblotted with rabbit anti-δ antibody. (E) The quantification ofcolocalization of the TTR and δ-α6β3δ receptors in cerebellar neurons by intensity correlation analysis. The intensity correlation quotient(IQC) was shown for both total staining (TS) and live cell staining(LCS).
Fig 3.
TTR regulates the surface expression of δ-receptors in cultured cerebellar neurons.
(A) Knockdown TTR with mouse-specific shRNAs reduced the surface δ-staining (red), which was rescued by co-transection with human TTR (RFP-tagged, purple in the insert) that was resistant to mouse TTR shRNAs. Arrows point to the transfected cells. Scale bar, 10 μm. (B) Quantification of the surface δ staining signal shown in (A). * P < 0.05; ** P < 0.01. (C) Nucleotide sequence alignment showing the specificity of the TTR shRNA designed for mouse TTR, but not for human TTR. (D) Knockdown efficiency of the TTR shRNAs evaluated by cotransfecting mouse HA-TTR with control plasmid or with TTR specific shRNAs in HEK cells. TTR shRNAs significantly downregulated the expression level of mouse HA-TTR. (E) TTR shRNAs had no off-target effect directly on δ-GABAA-Rs when coexpressed in HEK cells. (F-G) Representative traces of THIP current recorded from cultured cerebellar granule cells either transfected with GFP as a control (F) or with TTR shRNA (G). (H-J) Summarized bar graphs showing the THIP current amplitude (H; Control: 107 ± 25 pA, n = 18; TTR shRNA: 55 ± 12 pA, n = 22; P < 0.05, Students’ t-test.), THIP current density (I; Control: 7.2 ± 1.7 pA/pF, n = 18; TTR shRNA: 2.9 ± 0.6 pA/pF, n = 22; P < 0.05), and membrane capacitance (J; Control: 17.4 ± 4.1 pF, n = 18; TTR shRNA: 19.8 ± 4.2 pF, n = 22; P > 0.2) changes after transfection of TTR shRNAs in cultured cerebellar granule cells. * P < 0.05.
Fig 4.
Overexpression of TTR in cultured cortical neurons increased the surface δ expression and tonic current.
(A) Overexpression of human TTR increased the surface δ signal in cortical neurons. Arrows point to the cells transfected with mCherry (control) or TTR. Scale bar, 10 μm. (B) Quantification of the surface δ signal intensity in control and TTR-overexpressing cells. * P < 0.04. (C-D) Representative traces of THIP currents recorded from cortical neurons transfected with mCherry (C, control) or human TTR (D). (E-G) Summary bar graphs for THIP current amplitude (E; Control: 22 ± 3 pA, n = 17; TTR: 33 ± 3 pA, n = 17; * P < 0.05, Student’s t-test), THIP current density (F; Control: 0.52 ± 0.08 pA/pF, n = 17; TTR: 0.79 ± 0.09 pA/pF, n = 17; * P < 0.05), and cell membrane capacitance (G; Control: 47 ± 5 pF, n = 17; TTR: 47 ± 4 pF, n = 17; P > 0.9).
Fig 5.
Effects of wide type human TTR peptides or engineered monomeric human TTR (M-hTTR) on δ-GABAA-Rs in cerebellar granule cells.
(A) Immunostaining of surface δ subunit (green) in cerebellar granule neurons after treatment with 14 ug/ml hTTR or M-hTTR for 1 day. Scale bar, 5 μm and 2.5 μm. (B-C) Quantification of the surface δ signal on soma (B) or dendrites (C). *** P < 0.001, ** P < 0.01.
Fig 6.
In vivo analysis of the δ-GABAA-Rs in the cerebellar granule layer in WT and TTR-/- mice.
(A) Immunostaining of the surface δ receptors in cerebellar slices from WT and TTR-/- mice (7–8 months old). Top row, low power image of the δ signal in cerebellar granule layer. Scale bar, 50 μm. Bottom row, high power image of the δ signal. Scale bar, 20 μm. (B) Quantification of the relative surface δ signal density in WT and TTR-/- mice (low power image). (C) Western blot analysis of the total and surface δ receptors in WT and TTR-/- cerebellar tissue. The total δ protein level showed no difference, while the biotinylated surface δ protein level decreased significantly in TTR-/- mice. Both total and the surface γ2 GABAA-Rs have no difference between WT and TTR-/- mice. (D-E) Quantification of the relative total and surface δ intensity from the Western blot analysis. (F-G) Surface δ staining and quantification in primary cerebellar neuronal cultures from WT or TTR-/- mice. Bar, 5μm.