Limits...
Presynaptic localization of Smn and hnRNP R in axon terminals of embryonic and postnatal mouse motoneurons.

Dombert B, Sivadasan R, Simon CM, Jablonka S, Sendtner M - PLoS ONE (2014)

Bottom Line: Smn and hnRNP R are localized in close proximity to each other in axons and axon terminals both in vitro and in vivo.We also provide new evidence for a direct interaction of Smn and hnRNP R in vitro and in vivo, particularly in the cytosol of motoneurons.These data point to functions of SMN beyond snRNP assembly which could be crucial for recruitment and transport of RNA particles into axons and axon terminals, a mechanism which may contribute to SMA pathogenesis.

View Article: PubMed Central - PubMed

Affiliation: Institute for Clinical Neurobiology, University Hospital Wuerzburg, Wuerzburg, Germany.

ABSTRACT
Spinal muscular atrophy (SMA) is caused by deficiency of the ubiquitously expressed survival motoneuron (SMN) protein. SMN is crucial component of a complex for the assembly of spliceosomal small nuclear ribonucleoprotein (snRNP) particles. Other cellular functions of SMN are less characterized so far. SMA predominantly affects lower motoneurons, but the cellular basis for this relative specificity is still unknown. In contrast to nonneuronal cells where the protein is mainly localized in perinuclear regions and the nucleus, Smn is also present in dendrites, axons and axonal growth cones of isolated motoneurons in vitro. However, this distribution has not been shown in vivo and it is not clear whether Smn and hnRNP R are also present in presynaptic axon terminals of motoneurons in postnatal mice. Smn also associates with components not included in the classical SMN complex like RNA-binding proteins FUS, TDP43, HuD and hnRNP R which are involved in RNA processing, subcellular localization and translation. We show here that Smn and hnRNP R are present in presynaptic compartments at neuromuscular endplates of embryonic and postnatal mice. Smn and hnRNP R are localized in close proximity to each other in axons and axon terminals both in vitro and in vivo. We also provide new evidence for a direct interaction of Smn and hnRNP R in vitro and in vivo, particularly in the cytosol of motoneurons. These data point to functions of SMN beyond snRNP assembly which could be crucial for recruitment and transport of RNA particles into axons and axon terminals, a mechanism which may contribute to SMA pathogenesis.

Show MeSH

Related in: MedlinePlus

Smn deficiency in SMA type I axon terminals invivo.(A, B) Representative motor endplates from E18 Smn+/+; SMN2tg and Smn−/−; SMN2tg Diaphragm stained against Smn and SynPhys. Acetylcholine receptors (AChR) and postsynaptic nuclei were visualized by ω-BTX and DAPI, respectively (scale bar: 5 µm). In (A) Smn deficiency is visible by highly reduced immunoreactive signals, as highlighted in the white box, whereas in (B) the number of Smn particles per NMJ is decreased in SMA type I motor endplates, as indicated by white arrowheads. (A, B) In SMA type I axon terminals (n = 3, N = 32) mean Smn signal intensity was significantly reduced (0.43±0.09, P = 0.0220, t = 6.629, DF = 2) in comparison to control motor endplates (set as ‘1’, n = 3, N = 43), whereas SynPhys signals (Smn−/−; SMN2tg 1.15±0.19, P = 0.5221, t = 0.7694, DF = 2) and the size of the presynaptic compartment (Control 49.48±13.94 µm2; Smn−/−; SMN2tg 36.56±7.464; P = 0.4596, t = 0.8174, DF = 4) were comparable. (C) Representative images from E18 Smn+/+; SMN2tg and Smn−/−; SMN2tg spinal cord cross sections immunolabeled with Smn and ChAT. Quantitative analysis revealed a significant decrease in cytosolic Smn immunoreactivity in SMA type I motoneurons in comparison to Smn+/+; SMN2tg cells (Smn+/+; SMN2tg set as ‘1’, n = 6, N = 107; Smn−/−; SMN2tg 0.46±0.05, n = 6, N = 85; P<0.0001, t = 11.23, DF = 5). ChAT signal intensity was not statistically affected (Smn−/−; SMN2tg 0.83±0.21; P = 0.4638, t = 0.7928, DF = 5). (D) Representative Western Blot with cytosolic and nuclear fractions from E18 control and Smn−/−; SMN2tg spinal cord extracts. Histone H3 and α tubulin were used as markers for nuclear and cytosolic fractions, respectively, and as standardization proteins for quantitative analysis. In SMA type I spinal cord extracts cytosolic and nuclear Smn were significantly reduced by 64% (0.36±0.08, N = 10, P<0.0001, t = 8.480, DF = 9) and 86% (0.14±0.03, N = 10, P<0.0001, t = 26.39, DF = 9), respectively, in comparison to Smn+/+; SMN2tg extracts (set as ‘1’, N = 10).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4206449&req=5

pone-0110846-g007: Smn deficiency in SMA type I axon terminals invivo.(A, B) Representative motor endplates from E18 Smn+/+; SMN2tg and Smn−/−; SMN2tg Diaphragm stained against Smn and SynPhys. Acetylcholine receptors (AChR) and postsynaptic nuclei were visualized by ω-BTX and DAPI, respectively (scale bar: 5 µm). In (A) Smn deficiency is visible by highly reduced immunoreactive signals, as highlighted in the white box, whereas in (B) the number of Smn particles per NMJ is decreased in SMA type I motor endplates, as indicated by white arrowheads. (A, B) In SMA type I axon terminals (n = 3, N = 32) mean Smn signal intensity was significantly reduced (0.43±0.09, P = 0.0220, t = 6.629, DF = 2) in comparison to control motor endplates (set as ‘1’, n = 3, N = 43), whereas SynPhys signals (Smn−/−; SMN2tg 1.15±0.19, P = 0.5221, t = 0.7694, DF = 2) and the size of the presynaptic compartment (Control 49.48±13.94 µm2; Smn−/−; SMN2tg 36.56±7.464; P = 0.4596, t = 0.8174, DF = 4) were comparable. (C) Representative images from E18 Smn+/+; SMN2tg and Smn−/−; SMN2tg spinal cord cross sections immunolabeled with Smn and ChAT. Quantitative analysis revealed a significant decrease in cytosolic Smn immunoreactivity in SMA type I motoneurons in comparison to Smn+/+; SMN2tg cells (Smn+/+; SMN2tg set as ‘1’, n = 6, N = 107; Smn−/−; SMN2tg 0.46±0.05, n = 6, N = 85; P<0.0001, t = 11.23, DF = 5). ChAT signal intensity was not statistically affected (Smn−/−; SMN2tg 0.83±0.21; P = 0.4638, t = 0.7928, DF = 5). (D) Representative Western Blot with cytosolic and nuclear fractions from E18 control and Smn−/−; SMN2tg spinal cord extracts. Histone H3 and α tubulin were used as markers for nuclear and cytosolic fractions, respectively, and as standardization proteins for quantitative analysis. In SMA type I spinal cord extracts cytosolic and nuclear Smn were significantly reduced by 64% (0.36±0.08, N = 10, P<0.0001, t = 8.480, DF = 9) and 86% (0.14±0.03, N = 10, P<0.0001, t = 26.39, DF = 9), respectively, in comparison to Smn+/+; SMN2tg extracts (set as ‘1’, N = 10).

Mentions: To validate the specificity of the observed presynaptic Smn staining invivo, whole mount preparations from three E18 Smn−/−; SMN2tg mouse Diaphragms were analyzed and compared with controls (Fig. 7), revealing a significant reduction of the mean Smn signal intensity of 57% in SMA type I NMJs (0.43±0.09, P = 0.0220, n = 3, N = 32) in comparison to control samples (n = 3, N = 43), whereas neither the size of the presynaptic compartment nor SynPhys signal intensities were significantly altered at this developmental stage (Fig. 7A, B). We also investigated cytosolic Smn immunoreactivity in the corresponding E18 Smn−/−; SMN2tg (n = 6, N = 85) motoneuron cell bodies in spinal cord cross sections, detecting a significant decrease of 54% (0.46±0.05, P<0.0001) in comparison to Smn+/+; SMN2tg cells (n = 6, N = 107) (Fig. 7C). These two results were at variance with previous studies reporting profound loss of Smn protein in the range of 80% in brain extracts from these mice [59]. Therefore, we analyzed cytosolic and nuclear fractions from four E18 SMA type I spinal cords and corresponding control tissue in order to obtain more robust biochemical data and to validate the aforementioned immunohistochemical quantitative analysis (Fig. 7D). Smn protein levels were significantly reduced by 86% (0.14±0.03, n = 10, P<0.0001) in nuclear and by 64% (0.36±0.08, n = 10, P<0.0001) in cytosolic fractions of Smn−/−; SMN2tg spinal cord, respectively. With respect to the underlying biological variances derived from independent embryos and litters invivo we concluded from these data that the differences determined by immunohistochemistry were in line with the reduction of cytosolic Smn protein quantified by biochemical analysis, thus confirming the specificity of the applied Smn antibody also invivo.


Presynaptic localization of Smn and hnRNP R in axon terminals of embryonic and postnatal mouse motoneurons.

Dombert B, Sivadasan R, Simon CM, Jablonka S, Sendtner M - PLoS ONE (2014)

Smn deficiency in SMA type I axon terminals invivo.(A, B) Representative motor endplates from E18 Smn+/+; SMN2tg and Smn−/−; SMN2tg Diaphragm stained against Smn and SynPhys. Acetylcholine receptors (AChR) and postsynaptic nuclei were visualized by ω-BTX and DAPI, respectively (scale bar: 5 µm). In (A) Smn deficiency is visible by highly reduced immunoreactive signals, as highlighted in the white box, whereas in (B) the number of Smn particles per NMJ is decreased in SMA type I motor endplates, as indicated by white arrowheads. (A, B) In SMA type I axon terminals (n = 3, N = 32) mean Smn signal intensity was significantly reduced (0.43±0.09, P = 0.0220, t = 6.629, DF = 2) in comparison to control motor endplates (set as ‘1’, n = 3, N = 43), whereas SynPhys signals (Smn−/−; SMN2tg 1.15±0.19, P = 0.5221, t = 0.7694, DF = 2) and the size of the presynaptic compartment (Control 49.48±13.94 µm2; Smn−/−; SMN2tg 36.56±7.464; P = 0.4596, t = 0.8174, DF = 4) were comparable. (C) Representative images from E18 Smn+/+; SMN2tg and Smn−/−; SMN2tg spinal cord cross sections immunolabeled with Smn and ChAT. Quantitative analysis revealed a significant decrease in cytosolic Smn immunoreactivity in SMA type I motoneurons in comparison to Smn+/+; SMN2tg cells (Smn+/+; SMN2tg set as ‘1’, n = 6, N = 107; Smn−/−; SMN2tg 0.46±0.05, n = 6, N = 85; P<0.0001, t = 11.23, DF = 5). ChAT signal intensity was not statistically affected (Smn−/−; SMN2tg 0.83±0.21; P = 0.4638, t = 0.7928, DF = 5). (D) Representative Western Blot with cytosolic and nuclear fractions from E18 control and Smn−/−; SMN2tg spinal cord extracts. Histone H3 and α tubulin were used as markers for nuclear and cytosolic fractions, respectively, and as standardization proteins for quantitative analysis. In SMA type I spinal cord extracts cytosolic and nuclear Smn were significantly reduced by 64% (0.36±0.08, N = 10, P<0.0001, t = 8.480, DF = 9) and 86% (0.14±0.03, N = 10, P<0.0001, t = 26.39, DF = 9), respectively, in comparison to Smn+/+; SMN2tg extracts (set as ‘1’, N = 10).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4206449&req=5

pone-0110846-g007: Smn deficiency in SMA type I axon terminals invivo.(A, B) Representative motor endplates from E18 Smn+/+; SMN2tg and Smn−/−; SMN2tg Diaphragm stained against Smn and SynPhys. Acetylcholine receptors (AChR) and postsynaptic nuclei were visualized by ω-BTX and DAPI, respectively (scale bar: 5 µm). In (A) Smn deficiency is visible by highly reduced immunoreactive signals, as highlighted in the white box, whereas in (B) the number of Smn particles per NMJ is decreased in SMA type I motor endplates, as indicated by white arrowheads. (A, B) In SMA type I axon terminals (n = 3, N = 32) mean Smn signal intensity was significantly reduced (0.43±0.09, P = 0.0220, t = 6.629, DF = 2) in comparison to control motor endplates (set as ‘1’, n = 3, N = 43), whereas SynPhys signals (Smn−/−; SMN2tg 1.15±0.19, P = 0.5221, t = 0.7694, DF = 2) and the size of the presynaptic compartment (Control 49.48±13.94 µm2; Smn−/−; SMN2tg 36.56±7.464; P = 0.4596, t = 0.8174, DF = 4) were comparable. (C) Representative images from E18 Smn+/+; SMN2tg and Smn−/−; SMN2tg spinal cord cross sections immunolabeled with Smn and ChAT. Quantitative analysis revealed a significant decrease in cytosolic Smn immunoreactivity in SMA type I motoneurons in comparison to Smn+/+; SMN2tg cells (Smn+/+; SMN2tg set as ‘1’, n = 6, N = 107; Smn−/−; SMN2tg 0.46±0.05, n = 6, N = 85; P<0.0001, t = 11.23, DF = 5). ChAT signal intensity was not statistically affected (Smn−/−; SMN2tg 0.83±0.21; P = 0.4638, t = 0.7928, DF = 5). (D) Representative Western Blot with cytosolic and nuclear fractions from E18 control and Smn−/−; SMN2tg spinal cord extracts. Histone H3 and α tubulin were used as markers for nuclear and cytosolic fractions, respectively, and as standardization proteins for quantitative analysis. In SMA type I spinal cord extracts cytosolic and nuclear Smn were significantly reduced by 64% (0.36±0.08, N = 10, P<0.0001, t = 8.480, DF = 9) and 86% (0.14±0.03, N = 10, P<0.0001, t = 26.39, DF = 9), respectively, in comparison to Smn+/+; SMN2tg extracts (set as ‘1’, N = 10).
Mentions: To validate the specificity of the observed presynaptic Smn staining invivo, whole mount preparations from three E18 Smn−/−; SMN2tg mouse Diaphragms were analyzed and compared with controls (Fig. 7), revealing a significant reduction of the mean Smn signal intensity of 57% in SMA type I NMJs (0.43±0.09, P = 0.0220, n = 3, N = 32) in comparison to control samples (n = 3, N = 43), whereas neither the size of the presynaptic compartment nor SynPhys signal intensities were significantly altered at this developmental stage (Fig. 7A, B). We also investigated cytosolic Smn immunoreactivity in the corresponding E18 Smn−/−; SMN2tg (n = 6, N = 85) motoneuron cell bodies in spinal cord cross sections, detecting a significant decrease of 54% (0.46±0.05, P<0.0001) in comparison to Smn+/+; SMN2tg cells (n = 6, N = 107) (Fig. 7C). These two results were at variance with previous studies reporting profound loss of Smn protein in the range of 80% in brain extracts from these mice [59]. Therefore, we analyzed cytosolic and nuclear fractions from four E18 SMA type I spinal cords and corresponding control tissue in order to obtain more robust biochemical data and to validate the aforementioned immunohistochemical quantitative analysis (Fig. 7D). Smn protein levels were significantly reduced by 86% (0.14±0.03, n = 10, P<0.0001) in nuclear and by 64% (0.36±0.08, n = 10, P<0.0001) in cytosolic fractions of Smn−/−; SMN2tg spinal cord, respectively. With respect to the underlying biological variances derived from independent embryos and litters invivo we concluded from these data that the differences determined by immunohistochemistry were in line with the reduction of cytosolic Smn protein quantified by biochemical analysis, thus confirming the specificity of the applied Smn antibody also invivo.

Bottom Line: Smn and hnRNP R are localized in close proximity to each other in axons and axon terminals both in vitro and in vivo.We also provide new evidence for a direct interaction of Smn and hnRNP R in vitro and in vivo, particularly in the cytosol of motoneurons.These data point to functions of SMN beyond snRNP assembly which could be crucial for recruitment and transport of RNA particles into axons and axon terminals, a mechanism which may contribute to SMA pathogenesis.

View Article: PubMed Central - PubMed

Affiliation: Institute for Clinical Neurobiology, University Hospital Wuerzburg, Wuerzburg, Germany.

ABSTRACT
Spinal muscular atrophy (SMA) is caused by deficiency of the ubiquitously expressed survival motoneuron (SMN) protein. SMN is crucial component of a complex for the assembly of spliceosomal small nuclear ribonucleoprotein (snRNP) particles. Other cellular functions of SMN are less characterized so far. SMA predominantly affects lower motoneurons, but the cellular basis for this relative specificity is still unknown. In contrast to nonneuronal cells where the protein is mainly localized in perinuclear regions and the nucleus, Smn is also present in dendrites, axons and axonal growth cones of isolated motoneurons in vitro. However, this distribution has not been shown in vivo and it is not clear whether Smn and hnRNP R are also present in presynaptic axon terminals of motoneurons in postnatal mice. Smn also associates with components not included in the classical SMN complex like RNA-binding proteins FUS, TDP43, HuD and hnRNP R which are involved in RNA processing, subcellular localization and translation. We show here that Smn and hnRNP R are present in presynaptic compartments at neuromuscular endplates of embryonic and postnatal mice. Smn and hnRNP R are localized in close proximity to each other in axons and axon terminals both in vitro and in vivo. We also provide new evidence for a direct interaction of Smn and hnRNP R in vitro and in vivo, particularly in the cytosol of motoneurons. These data point to functions of SMN beyond snRNP assembly which could be crucial for recruitment and transport of RNA particles into axons and axon terminals, a mechanism which may contribute to SMA pathogenesis.

Show MeSH
Related in: MedlinePlus