The stargazin-related protein gamma 7 interacts with the mRNA-binding protein heterogeneous nuclear ribonucleoprotein A2 and regulates the stability of specific mRNAs, including CaV2.2 - PubMed (original) (raw)

Comparative Study

. 2008 Oct 15;28(42):10604-17.

doi: 10.1523/JNEUROSCI.2709-08.2008.

Anthony Davies, Karen M Page, David J Cox, Jerôme Leroy, Dominic Waithe, Adrian J Butcher, Priya Sellaturay, Steven Bolsover, Wendy S Pratt, Fraser J Moss, Annette C Dolphin

Affiliations

Comparative Study

Laurent Ferron et al. J Neurosci. 2008.

Abstract

The role(s) of the novel stargazin-like gamma-subunit proteins remain controversial. We have shown previously that the neuron-specific gamma7 suppresses the expression of certain calcium channels, particularly Ca(V)2.2, and is therefore unlikely to operate as a calcium channel subunit. We now show that the effect of gamma7 on Ca(V)2.2 expression is via an increase in the degradation rate of Ca(V)2.2 mRNA and hence a reduction of Ca(V)2.2 protein level. Furthermore, exogenous expression of gamma7 in PC12 cells also decreased the endogenous Ca(V)2.2 mRNA level. Conversely, knockdown of endogenous gamma7 with short-hairpin RNAs produced a reciprocal enhancement of Ca(V)2.2 mRNA stability and an increase in endogenous calcium currents in PC12 cells. Moreover, both endogenous and expressed gamma7 are present on intracellular membranes, rather than the plasma membrane. The cytoplasmic C terminus of gamma7 is essential for all its effects, and we show that gamma7 binds directly via its C terminus to a heterogeneous nuclear ribonucleoprotein (hnRNP A2), which also binds to a motif in Ca(V)2.2 mRNA, and is associated with native Ca(V)2.2 mRNA in PC12 cells. The expression of hnRNP A2 enhances Ca(V)2.2 I(Ba), and this enhancement is prevented by a concentration of gamma7 that alone has no effect on I(Ba). The effect of gamma7 is selective for certain mRNAs because it had no effect on alpha2delta-2 mRNA stability, but it decreased the mRNA stability for the potassium-chloride cotransporter, KCC1, which contains a similar hnRNP A2 binding motif to that in Ca(V)2.2 mRNA. Our results indicate that gamma7 plays a role in stabilizing Ca(V)2.2 mRNA.

PubMed Disclaimer

Figures

Figure 1.

Figure 1.

Effect of C-terminal truncation of γ7 on CaV2.2/β1b/α2δ-2 and CaV3.1 currents. A, Linearized diagram of γ7, indicating the approximate positions of the four transmembrane (TM) segments (black bars), the N-glycosylation site (V) at N45, and the position of the truncations at amino acids 217, 238, and 271. B, Left, Example traces elicited after cDNA injection into Xenopus oocytes by 100 ms step depolarizations to between −40 and 0 mV from a holding potential of −100 mV for CaV2.2/β1b/α2δ-2 (top), plus γ7 (middle) and plus γ7 (1–217) (bottom). The charge carrier was 10 m

m

Ba2+. The symbols beside the traces refer to the relevant data in the current–voltage relationship (right); CaV2.2/β1b/α2δ-2 (■; n = 19) plus γ7(1–217) (▵; n = 18) and plus γ7 (□; n = 8). Data are fit by a modified Boltzmann function as described in Materials and Methods, with V 50, act of −9.9, −9.8, and +1.2 mV, respectively, and G max of 19.6, 16.6, and 5.9 μS, respectively. C, Mean percentage of control peak I Ba for CaV2.2/β1b/α2δ-2 currents (black bar; n = 29) when coexpressed with γ7 (white bar; n = 31) or its truncated constructs, γ7(1–217) (hatched bar; n = 15), γ7(1–238) (gray bar; n = 12), and γ7(1–271) (striped bar; n = 16). The statistical significances compared with control are **p < 0.01. There was no effect of any of the constructs on the voltage for 50% steady-state inactivation, which from combined experiments was −60.2 ± 0.8 mV for controls (n = 16), −58.6 ± 1.0 mV in the presence of γ7 (n = 7), −61.7 ± 3.6 mV for γ7(1–217) (n = 5), and −59.8 ± 1.0 mV for γ7(1–271) (n = 5). D, Representative traces of peak Ba2+ currents in tsA 201 cells, cotransfected with CaV2.2, β1b and α2δ-2 with pMT2 as control (con) compared with γ7 (panel 1), γ7(1–217) (panel 2), and γ7(stop) (panel 3). Panel 4 shows CaV3.1 expression with pMT2 as control (con) compared with γ7. Currents were elicited by depolarization to +5 mV (CaV2.2, 1 m

m

Ba2+) or −10 mV (CaV3.1, 10 m

m

Ba2+), from a holding potential of −100 mV. Calibration bars refer to all traces for CaV2.2. E, Mean inhibition of Ba2+ currents (expressed as percentage of control ± SEM) induced by coexpression of CaV2.2 with γ7 (white bar; n = 21), γ7(1–217) (black bar; n = 15), γ7(stop) (hatched bar; n = 28), or CaV3.1 with γ7 (cross-hatched bar; n = 29). Statistical significance compared with the current size without γ7, **p < 0.01, Student's t test.

Figure 2.

Figure 2.

Effect of C-terminal truncation of γ7 on CaV2.2 protein. A, γ7 and truncated γ7 constructs were expressed at the expected sizes in COS-7 cells (top, γ7 I–II loop Ab; bottom γ7, C-terminal tail Ab). In each lane, the lower band corresponds to the expected protein molecular weight, and the upper band(s) correspond to either the mature glycosylated form or intermediate glycosylated species. As expected, γ7(1–217) and γ7(1–238) were not detected by the γ7(C-terminal tail) Ab. The specificity of the Abs is indicated by the lack of immunostaining in the absence of transfected constructs (−). Similar results were obtained in Xenopus oocytes (data not shown). The same amount of total protein was loaded in each lane (25 μg). B, Examples of Western blots showing the effect of γ7 (i) and γ7(1–271) (ii) and the lack of effect of γ7(1–217) (iii) and KV3.1b (iv) on the level of CaV2.2 protein expressed in COS-7 cells. con, Transfection with Kir–AAA cDNA, in place of γ7 (see Materials and Methods). The same amount of total protein was loaded in each pair of lanes (25 μg). C, Correlation between the effect of the γ7 and its various C-terminal truncated constructs on the level of CaV2.2 protein with their effect on CaV2.2 I Ba shown in Figure 1_C_. The effect of γ7, γ7(1–217), γ7(1–271), and γ7(1–238) on CaV2.2 protein levels represents the mean inhibition observed in 3–10 experiments. The linear fit has a correlation coefficient, r, of 0.936. All error bars are SEM.

Figure 3.

Figure 3.

Effect of γ7 and C-terminally truncated γ7(1–217) on CaV2.2 mRNA stability. A, Effect of γ7 on CaV2.2 mRNA degradation rate. Constructs were expressed in Xenopus oocytes either without (■) or with γ7 (▵) or with γ7(1–217) (□). In the absence of a γ7 construct, an equivalent amount of a similar sized transcript, a nonfunctional K+ channel Kir–AAA cDNA was used. After 24 h (T 0), actinomycin D (50 μg/ml) was added to the medium, and the CaV2.2 mRNA levels were measured at the times shown after this. The numbers of determinations from individual oocytes are between 6 and 12 for each data point; *p < 0.05 compared with CaV2.2 (Student's t test). The lines are apparent linear fits using errors as weight. B, Bar chart showing the percentage of CaV2.2, α2δ-2, and KCC1 mRNA present at time t after actinomycin D addition at time T 0. For CaV2.2 mRNA turnover, t = 9 h. Bars 1–3 are control (black bar; n = 11); + γ7 (white bar, n = 7) and + γ7(1–217) (hatched bar; n = 8). Bars 4–7 show the percentage of α2δ-2 mRNA and KCC1 mRNA present 9 and 24 h, respectively, after actinomycin D addition for the control condition (black bar; n = 8 and 9), + γ7 (white bar; n = 9 and 9). These times were chosen as the percentage of mRNA remaining in control conditions was similar for all mRNA species. *p < 0.05 compared with control, Student's t test. C, Degradation rate for endogenous CaV2.2 in the γ7–CFP PC12 cell line (○; n = 3) compared with control pcDNA3.1-transfected PC12 cell line (■; n = 5), after differentiation with NGF. The CaV2.2 mRNA level is expressed as percentage of time 0 (T 0), when actinomycin D was added. The half-life for endogenous CaV2.2 mRNA was 19.5 h in the pcDNA3.1-transfected PC12 cell line. In γ7-transfected PC12 cells, the half-life was 9.1 h (**p < 0.01, Student's t test). Inset, Reduction of endogenous CaV2.2 mRNA level in γ7–CFP PC12 cell line compared with control pcDNA3.1-transfected PC12 cell line. ***p < 0.001 versus control.

Figure 4.

Figure 4.

Short-hairpin RNA constructs knockdown γ7 levels in PC12 cells. A, Left, Endogenous γ7 (detected using γ7 I–II loop Ab; green) in a differentiated PC12 cell, together with F-actin (Alexa Fluor 594–phalloidin; red) and nuclear staining (DAPI; blue). Right, Control in the absence of primary γ7 Ab. Scale bar, 10 μm (applies to both images). The images are a _Z_-stack of five to eight confocal images. B, Fluorescence microscopy of PC12 cells stably transfected with a human γ7–CFP construct. These cells were transiently transfected with YFP and either negative control gnu shRNA (Ctrl; left) or a mixture of three human γ7 shRNAs (γ7; right) and examined after 5 d. Top row, Transfected cells are identified with YFP fluorescence. Middle row, CFP fluorescence is almost abolished in cells transfected with γ7 shRNA. Arrowhead in left indicates a cell transfected with the control shRNA that shows γ7–CFP fluorescence. Arrows in right indicate two cells transfected with γ7 shRNA that do not show γ7–CFP fluorescence. Bottom row, Bright field. Scale bars, 30 μm. C, Bar chart of the mean percentage of CFP fluorescence in PC12 γ7–CFP cells after transfection with γ7 shRNA (white bar) compared with transfection with control shRNA (black bar) or untransfected PC12 cells (UT, gray bar). ***p < 0.001 vs control. D, Effect of rat γ7 shRNA on γ7 mRNA level in PC12 cells. γ7 mRNA was quantified by q-PCR in PC12 cells transfected with GFP (white bar), negative control shRNA (Ctrl, black bar), or human γ7 shRNAs: γ7 96 (red bar), γ7 285 (green bar), γ7 500 (blue bar), or γ7 mix (cyan bar). ***p < 0.001 versus control. E, Effect of transfection with γ7 shRNA on endogenous γ7 protein levels in PC12 cells. Left, Cells transfected with control shRNA and GFP; right, cells transfected with rat γ7 shRNA and GFP. Top row, Bright-field image; middle row, GFP, with transfected cell outlined; bottom row, endogenous γ7 visualized by immunocytochemistry (red), showing reduced fluorescence in γ7 shRNA-transfected cell (outlined). Scale bar, 20 μm (applies to all images, which were taken on a conventional fluorescence microscope). F, Quantification of data including that given in E, showing reduction in mean fluorescence intensity (measured as fluorescence density in arbitrary units) in γ7 shRNA-transfected cells (white bar) compared with control shRNA (Ctrl, black bar) or untransfected cells (UT, gray bar); ***p < 0.001 compared with control.

Figure 5.

Figure 5.

Short-hairpin RNA constructs enhance endogenous CaV2.2 mRNA levels and calcium channel currents in PC12 cells. A, Effect of rat γ7 shRNA on endogenous Cav2.2 mRNA level in PC12 cells. Cav2.2 mRNA was quantified by q-PCR in PC12 cells transfected with GFP (white bar), negative control shRNA (Ctrl, black bar), or γ7 shRNAs: γ7 96 (gray bar), γ7 285 (hatched bar), γ7 500 (cross-hatched bar), or γ7 mix (striped bar). ***p < 0.001 versus control. B, Top, Representative traces of peak calcium channel currents recorded from differentiated PC12 cells transfected with control shRNA (left) and γ7 shRNA (right). Currents were elicited by a 100 ms depolarization step to +10 mV from a holding potential of −100 mV. Bottom, Current–voltage relationships for the two conditions (■, Ctrl shRNA; ○, γ7 shRNA). The charge carrier was 10 m

m

Ba2+. The knockdown of γ7 induces an increase of the peak current (−16.1 ± 0.9 pA/pF; n = 48) compared with control (−12.5 ± 0.9 pA/pF; n = 17; p < 0.05).

Figure 6.

Figure 6.

Identification of proteins interacting in a complex with γ7. A, Proteins coimmunoprecipitated with γ7–HA from stably transfected PC12 cells. Left, Coomassie blue staining. Right, Western blotting and immunodetection with anti-HA Ab of immunoprecipitated and control samples separated by SDS-PAGE. Solid arrow indicates γ7–HA. Small arrows indicate proteins coimmunoprecipitated with γ7–HA. Bands 1 and 2 were identified by peptide mass fingerprinting to contain hnRNP A2. Band 2 also contained hnRNP A3. The control lane is untransfected PC12 cells. Position of molecular weight markers is shown on the left. Representative of three experiments. B, Endogenous hnRNP A2 coimmunoprecipitates with transiently transfected γ7–HA but not with γ7(1–217)–HA in tsA 201 cells. Western blotting and immunodetection with anti-hnRNP A/B Ab (H-200, top row) and anti-HA Ab (bottom row) of input (left) and HA-immunoprecipitated samples (after 500 m

m

NaCl wash, right) separated by SDS-PAGE. Position of molecular weight markers is shown on the left. Blots are representative of three to five independent experiments, using anti-hnRNP A/B Abs from two different sources. C, CaVβ1b with a C-terminal HA tag was expressed and immunoprecipitated as described for γ7–HA. Its presence in the precipitate is confirmed in the top blot. The control lane is from cells not transfected with CaVβ1b–HA. Immunoblotting for endogenous hnRNP A2 shows it is present in the input lanes but absent from the immunoprecipitate. D, Coimmunoprecipitation of endogenous hnRNP A2 and γ7 proteins from PC12 cells. Western blotting and immunodetection with anti-hnRNP A2 Ab (EF-67, left) and γ7 C-terminus Ab (right) of input (left lane of both panels) and γ7 C-terminus Ab-immunoprecipitated samples (right lane of both panels) separated by SDS-PAGE. Control immunoprecipitations in which γ7 Ab was omitted are shown in the middle lane of each panel. Position of molecular weight markers is indicated between the panels. Blots are representative of two independent experiments. E, The interaction of the C terminus of hnRNP A2 with the cytoplasmic C-terminal tail of γ7 was independently confirmed by yeast cotransformation tests. Lane 1, Positive control (blue reaction product) showing CaV2.2 I–II loop (BI-II) pACT2 and β1b pAS2–1. Lane 2, Interaction between hnRNP A2 C terminus (hn-C) in pACT2 and γ7 C terminus (γ7-C) in pAS2–1. Lanes 3 and 4, Negative controls showing that the hnRNP A2 C terminus and the γ7 C terminus do not interact with β1b in the alternative vector. The filter was incubated with the X-gal (5-bromo-4-chloro-3-indolyl-b-

d

-galactopyranoside) substrate for 5 h. F, Regions of colocalization of hnRNP A2 with γ7–CFP (top row) or endogenous γ7 (bottom row) in SCG neurons. i, Immunolocalization of endogenous hnRNP A2 in SCG neuron cell bodies (top, red; bottom, green). ii, Localization of γ7–CFP (top, blue) or immunolocalization of endogenous γ7 (bottom, red). iii, Merger of images of i and ii, with the colocalized regions shown in white (top) or yellow (bottom). iv, Colocalized γ7 and hnRNP A2 pixels are also shown separately for clarity (white). Scale bar, 10 μm (applies to all images). No endogenous γ7 staining was observed in the absence of primary Ab (data not shown).

Figure 7.

Figure 7.

Binding of hnRNP A2 to CaV2.2 mRNA and effect of hnRNP A2 and γ7 on CaV2.2 currents. A, Evidence that one of the two potential A2RE sites in CaV2.2 mRNA binds to hnRNP A2. Endogenous hnRNP A2 (indicated by the arrow) from mouse brain lysate was pulled down by biotinylated oligonucleotides containing the consensus A2RE site (lane 2) and the potential site in CaV2.2 (lane 3) but not by beads alone (lane 1) or by a nonspecific sequence with the same composition as the consensus A2RE site (lane 4). The hnRNP A2 was detected by immunoblotting with anti-hnRNP A2 Ab (F16). Position of molecular weight markers is shown on the left. Representative of two experiments. B, Coimmunoprecipitation of endogenous CaV2.2 mRNA associated with hnRNP A2 in PC12 cells. Two days after transfection with HA–hnRNP A2 (HA-A2, lanes 1 and 3) or hnRNP A2 (A2, lanes 2 and 4), hnRNP A2 protein was immunoprecipitated (IP) from the whole-cell lysate (lanes 1 and 2) with HA antibody (lanes 3 and 4). Immunoblots show that HA Ab pulled down hnRNP A2 [top row, anti-hnRNP A2 (EF-67 Ab); middle row, anti-HA]. Coimmunoprecipitated RNA was extracted, reverse transcribed, and amplified by PCR. A specific PCR product corresponding to the endogenous CaV2.2 mRNA was amplified (35 cycles) only in the condition in which hnRNP A2 protein was pulled down (bottom row, lane 3). C, Left, Endogenous hnRNP A2 proteins were immunoprecipitated from PC12 cells with anti-hnRNP A2 Ab EF-67 (top row, lane 2). In this condition, a specific PCR product corresponding to the endogenous CaV2.2 mRNA was also amplified (35 cycles, bottom row). CaV2.2 mRNA was not detected in the control in which antibody was omitted (lane 3). Right, The immunoprecipitation was repeated, including an acetone precipitation step, to confirm the presence of endogenous hnRNP A2. D, Enhancement by hnRNP A2 but not hnRNP A1 of CaV2.2 currents recorded from Xenopus oocytes. Left, Example traces elicited by 50 ms step depolarizations to between −40 and 0 mV in 10 mV steps, from a holding potential of −100 mV for CaV2.2/β1b/α2δ-2 (top traces) or CaV2.2/β1b/α2δ-2 plus hnRNP A2 (bottom traces). The charge carrier was 10 m

m

Ba2+. Right, Bar chart shows the effect of hnRNP A2 coexpression with CaV2.2/β1b/α2δ-2 on peak CaV2.2 I Ba amplitude, expressed relative to the mean peak control current in each experiment. Control (black bar; n = 18), for both groups taken from experiments directly comparing hnRNP A1 and hnRNP A2: + hnRNP A2 (white bar; n = 21) and + hnRNP A1 (hatched bar; n = 19). These data are pooled from two separate experiments showing similar results. Statistical significance, *p < 0.05, **p < 0.01 (one-way ANOVA and Bonferroni's post hoc test). E, Inhibition of the hnRNP A2-mediated enhancement of I Ba in Xenopus oocytes by a low concentration of γ7. Bar chart compares the effect of two concentrations of γ7 cDNA (γ7: CaV2.2 ratio 1 and 0.25) and also shows the lack of enhancement by hnRNP A2 on peak CaV2.2 I Ba amplitude, when coexpressed with the lower concentration of γ7 (γ7: CaV2.2 ratio 0.25). Data are expressed as a percentage of the mean peak control CaV2.2 current in each experiment, and all data were recorded 2 d after cDNA injection. Control in the absence of γ7 or hnRNP A2 (white bars); + γ7 (ratio 1; hatched bar), + γ7 (ratio 0.25; cross-hatched bar), + hnRNP A2 (black bar, from data obtained in the same experiments), + hnRNP A2 and γ7 (ratio 0.25; gray bar). Number of determinations given above each bar. These data are pooled from five different batches of oocytes, in all of which similar results were observed. Statistical significance, **p < 0.01 (one-way ANOVA and Bonferroni's post hoc test).

Figure 8.

Figure 8.

Subcellular localization of γ7. A, Western blotting of a γ7 mutant in which the potential glycosylation site, N45 is mutated to A, and immunodetection with anti-γ7 I–II loop Ab. Lane 1, γ7; lane 2, γ7 N45A. The reduction in mass and sharpening of the band indicates that γ7 is normally glycosylated at this site. Positions of molecular weight markers are shown on the left. The data are representative of two independent experiments. B, Immunodetection of transiently transfected γ7 (green) in nonpermeabilized (left) and permeabilized (right) tsA 201 cells, using γ7 I–II loop Ab. The nuclear stain DAPI (blue) was used as a cell marker. Scale bar, 100 μm (applies to both images). No immunostaining was observed in any field in the absence of permeabilization. C, Immunostaining for γ7 using I–II loop Ab (Texas Red secondary Ab, left) colocalizes completely with γ7–CFP (middle), as shown by yellow regions in merged image (right), in permeabilized (top row) but not nonpermeabilized (bottom row) SCG neurons. Scale bar, 40 μm (applies to all images). D, Partial colocalization of γ7–CFP and an ER marker in SCG neurons. Left, γ7–CFP; middle, ER marker (ER-DsRed); right, overlay showing colocalization of DsRed with γ7–CFP (white). Scale bar, 10 μm (applies to all images).

Figure 9.

Figure 9.

Diagram of proposed function of γ7. A, The physiological distribution of hnRNP A2 and γ7. The hnRNP A2 (red circles) is localized primarily in the nucleus in which it binds to A2RE sequences on mRNAs and exits to the cytoplasm with these mRNAs, including that of CaV2.2 (green). This is thought to stabilize certain mRNAs, reducing degradation and therefore enhancing expression. Our results suggest this may be the case for CaV2.2. hnRNP A2 may also stabilize the mRNAs for transport. γ7 (dark blue) is present on the ER and is also associated with motile vesicles. B, After overexpression of γ7, it may sequester cytoplasmic hnRNP A2 and therefore increase degradation of CaV2.2 mRNA.

Similar articles

Cited by

References

    1. Ainger K, Avossa D, Diana AS, Barry C, Barbarese E, Carson JH. Transport and localization elements in myelin basic protein mRNA. J Cell Biol. 1997;138:1077–1087. - PMC - PubMed
    1. Black JL, 3rd, Lennon VA. Identification and cloning of putative human neuronal voltage-gated calcium channel gamma-2 and gamma-3 subunits: neurologic implications. Mayo Clin Proc. 1999;74:357–361. - PubMed
    1. Brice NL, Berrow NS, Campbell V, Page KM, Brickley K, Tedder I, Dolphin AC. Importance of the different β subunits in the membrane expression of the α1A and α2 calcium channel subunits: studies using a depolarisation-sensitive α1A antibody. Eur J Neurosci. 1997;9:749–759. - PubMed
    1. Brittis PA, Lu Q, Flanagan JG. Axonal protein synthesis provides a mechanism for localized regulation at an intermediate target. Cell. 2002;110:223–235. - PubMed
    1. Broadus J, Fuerstenberg S, Doe CQ. Staufen-dependent localization of prospero mRNA contributes to neuroblast daughter-cell fate. Nature. 1998;391:792–795. - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources