RhoA/ROCK regulation of neuritogenesis via profilin IIa-mediated control of actin stability - PubMed (original) (raw)

RhoA/ROCK regulation of neuritogenesis via profilin IIa-mediated control of actin stability

Jorge Santos Da Silva et al. J Cell Biol. 2003.

Abstract

Neuritogenesis, the first step of neuronal differentiation, takes place as nascent neurites bud from the immediate postmitotic neuronal soma. Little is known about the mechanisms underlying the dramatic morphological changes that characterize this event. Here, we show that RhoA activity plays a decisive role during neuritogenesis of cultured hippocampal neurons by recruiting and activating its specific kinase ROCK, which, in turn, complexes with profilin IIa. We establish that this previously uncharacterized brain-specific actin-binding protein controls neurite sprouting by modifying actin stability, a function regulated by ROCK-mediated phosphorylation. Furthermore, we determine that this novel cascade is switched on or off by physiological stimuli. We propose that RhoA/ROCK/PIIa-mediated regulation of actin stability, shown to be essential for neuritogenesis, may constitute a central mechanism throughout neuronal differentiation.

PubMed Disclaimer

Figures

Figure 1.

RhoA negatively regulates neuritogenesis. (A) Compared with cells transfected with a null HA plasmid (Control), RhoA inactivation (C3, 24 h) induces increased neurite sprouting and reduction of F-actin content. (B) Conversely, transfection of HA-tagged, constitutively active RhoA (V14RhoA) decreases neurite sprouting and increases accumulation of F-actin. (C) Quantitative comparison indicates that V14 (n = 22) decreases neurite length and number, compared with control cells (Control, n = 35), whereas C3 transfection (n = 22) increases process sprouting and elongation. Data are mean + SD values. *, P < 0.001, in comparison with control. Bars, 10 μm.

Figure 2.

RhoA kinase ROCK activity modulates neuritogenesis. (A) ROCK inhibition (Y-27632, 18 h) induces multiple neurite sprouting and reduction of F-actin content, when compared with PBS buffer-treated controls. (B) In contrast, overexpression of myc-tagged ROCK (OE ROCK) severely inhibits neurite budding, and favors increased F-actin accumulation. (C) Quantitative analysis indicates that Y-27632–treated cells (n = 29) significantly sprout more and longer processes than control cells (n = 25), and that ROCK overexpression (OE ROCK, n = 27) severely reduces neurite sprouting and extension. Data are mean + SD values. *, P < 0.001, in comparison with control. Bars, 10 μm.

Figure 3.

PIIa is a negative regulator of neurite sprouting. (A) Western blot of heterozygous (+/−) and PIIa-null mice (−/−) brain lysates probed for actin, PIIa, and PI showing that only PIIa expression is affected in PII−/− mice. Compared with PII+/− neurons, at 24 h, PII−/− cells exhibit multiple sprouting neurites with increased lengths. (B) PIIa-morpholino antisense oligonucleotides (ASPIIa) reduce PIIa levels by 70% in rat hippocampal neurons in culture, whereas PIIa-missense oligonucleotides (misASPIIa) do not alter PIIa expression, compared with untreated control cells (C). The unrelated Golgi protein mannosidase-2 (Mann2) and the closely related PI are unaffected by antisense treatment. At 24 h, misASPIIa neurons present a typical morphology, whereas ASPIIa cells display increased number of sprouting neurites and extreme elongation of most processes. (C) Quantitative analysis shows that mouse PII+/− and rat misASPIIa neurons display similar neurite number and length averages. PII−/− cells have a clear increase in the number, branching (up to quaternary, 4ry), and length of emergent processes (*, P < 0.001, PII−/− in comparison with PII+/−; PII+/−, n = 23; PII−/−, n = 21). Similarly, but more strikingly, ASPIIa rat hippocampal neurons exhibit increased neurite number, branching (up to quintenary, 5ry), and length when compared with misASPIIa cells (**, P < 0.001, ASPIIa in comparison with misASPIIa; misASPIIa, n = 26; ASPIIa, n = 25). Data are mean + SD values. (D) Western blot shows detectable levels of PIIa-GFP in transfected (T) versus untransfected cells (NT). Overexpression of PIIa-GFP (OEPIIa) leads to differentiation arrest, at 24 h, when compared with GFP-only transfected neurons (Control). (E) Comparative analysis shows that transfection of PIIa-GFP induces a clear decrease in neurite budding and extension when compared with the mock GFP transfection. Data are mean + SD values. *, P < 0.001; GFP, n = 35; PIIa-GFP, n = 31. (F) Reconstruction of observed phenotypes illustrates the morphologies induced by PIIa loss and gain of function. Bars, 10 μm.

Figure 4.

The effects caused by PIIa loss- and gain-of-function approaches are reversible. (A) PIIa overexpressing cells treated with PIIa antisense oligonucleotides (OEPIIa + ASPIIa) have more and longer neurites than overexpressing neurons only. (B) Expression of PIIa-GFP in a PIIa-antisense background (ASPIIa + OEPIIa) prevents the increase in neurite number and extension produced by severe PIIa reduction. (C) Quantitative analysis shows that both the gain- (OEPIIa) and loss-of-function (ASPIIa) phenotypes can be reciprocally and significantly rescued, in nascent neurite number and length. Data are mean + SD values. *, P < 0.001 for each pair; OEPIIa, n = 19; OEPIIa + ASPIIa, n = 19; ASPIIa, n = 30; ASPIIa + OEPIIa, n = 22. Bars, 10 μm.

Figure 5.

PIIa modulates actin dynamics in early neuronal differentiation. (A) Compared with buffer-treated (Control) neurons, antisense-treated cells display a generalized reduction of PIIa labeling along with lower F-actin fluorescence and higher G-actin levels. Overexpression of untagged PIIa (arrowhead) leads to increased F-actin content and decreased G-actin levels. Measurement of F- and G-actin indicates that, compared with the steady state, PIIa antisense increases the G-actin content, whereas OEPIIa cells have significantly more F-actin. Data are mean + SD values. *, P < 0.001, in comparison with control F-actin; **, P < 0.001, in comparison with control G-actin. (B) Untransfected cells are cytochalasin-D (cytoD) sensitive, sprouting numerous neurites with long, curled appearance. PIIa overexpressing cells (GFP positive, arrowhead) have higher cytoD resistance. Still, cytoD partially rescues the PIIa overexpression phenotype (compare with A). Cell measurements show that cytoD induces process sprouting and extension (cytoD, n = 32; control, n = 35) and that the OEPIIa phenotype (n = 26) is significantly reverted by cytoD-induced depolymerization of F-actin (OEPIIa + CytoD, n = 22). Data are mean + SD values. *, P < 0.001, in comparison with control; **, P < 0.001, OEPIIa in comparison with OEPIIa + CytoD. Bars, 10 μm.

Figure 6.

ROCK physically and functionally interacts with PIIa. (A) Brain cell homogenates immunoprecipitated with anti-PIIa antibody followed by blotting with anti-ROCK antibody (top) or immunoprecipitated with anti-ROCK antibody followed by blotting with anti-PIIa antibody (bottom). Both ROCK and PIIa are reciprocally immunoprecipitated. No signal is detected when protein G beads are incubated without antibodies (B). (B) Western blotting with the PIIa antibody of untransfected cells treated with the ROCK inhibitor (Y-27632, 6 h) or buffer-treated (control) cells, reveals that the amount of immunoprecipitated PIIa does not depend on the phosphorylation activity of ROCK (middle, PIIa, IP). Western blotting with the PIIa antibody of transfected (myc-ROCK) and un-transfected (control) cells, reveals that the amount of immunoprecipitated PIIa is higher in cells with higher levels of ROCK (middle, PIIa, IP). ROCK expression is detected with a myc-specific antibody (top, myc, IP). No PIIa signal is detected when extracts are incubated with beads without ROCK antibody (middle, PIIa, IP). To control for the presence and levels of PIIa, preimmunoprecipitation lysates were blotted with PIIa antibody (bottom, PIIa, lysate). (C) Western blots of 2D-SDS-PAGE samples from neuronal extracts blotted with anti-PIIa antibody. PIIa is present in two clearly distinguishable spots (control). The more negative form (+ to − labels the isoelectrical focusing direction) corresponds to a more phosphorylated form (AP-sensitive, histogram). Y-27632 (6 h) increases the percentage of dephosphorylated PIIa versus control (43% increase, histogram), whereas the more phosphorylated form is lower than in controls (histogram: only Δ% for the more dephosphorylated form is shown; data are mean + SD values; n = 3; P < 0.001 for AP and Y-27632 in comparison with control; 100% is the sum of both forms of PIIa). (D) Inhibition of ROCK activity (18 h) increases neurite sprouting (Y-27632 reconstruction and unmarked cells). This phenotype is significantly prevented by PIIa overexpression (OEPIIa + Y-27632 reconstruction and GFP-positive cells, arrowheads) and accompanied by an increase of the F-actin content. (E) ROCK overexpression induces a dramatic arrest of differentiation (Fig. 2 B and OE ROCK reconstruction). PIIa antisense on a ROCK overexpression background results in partial reversion of the ROCK phenotype, with longer neurites and reduced F-actin (arrowhead and OE ROCK + ASPIIa reconstruction). (F) Quantitative analysis indicates that overexpression of PIIa (OEPIIa, n = 21) causes an arrest in neurite budding and elongation. Inhibition of ROCK, in this experimental background (OEPIIa + Y-27632, n = 22), leads to a very significant recovery with neurites reaching control-like number and length, but not to the levels observed for ROCK inhibition alone (compare with Fig. 2 C). Overexpression of ROCK (OE ROCK, n = 27) inhibits neurite sprouting and extension. PIIa antisense in this case (OE ROCK + ASPIIa, n = 26) results in a partial phenotype rescue, as processes elongate further, but cells do not sprout more neurites. Data are mean + SD values. *, P < 0.001, OEPIIa + Y-27632 in comparison with OEPIIa; **, P < 0.001, OEROCK + ASPIIa in comparison with OEROCK; °, P < 0.05, OEROCK + ASPIIa in comparison with OEROCK. Bars, 10 μm.

Figure 7.

PIIa is a downstream effector of RhoA. (A) ROCK immunoprecipitation of cell extracts overexpressing constitutively active RhoA (V14) or inactive Rho (C3, 24 h), followed by Western blotting with PIIa antibody. V14RhoA increases the amount of PIIa immunoprecipitated by ROCK (middle, PIIa, IP), whereas C3 reduces the amount of PIIa immunoprecipitated by ROCK (middle, PIIa, IP). V14RhoA and C3 expression is detected with an HA-specific antibody (top, HA, IP). No signal is detected (B) when extracts are precipitated without ROCK antibody (middle, PIIa, IP). To control for the presence and levels of PIIa, preimmunoprecipitation lysates were blotted with PIIa antibody (bottom, PIIa, lysate). (B) C3 expression (24 h) enhances neurite sprouting (Fig. 1 A and C3 reconstruction). Co-transfection with PIIa-GFP (C3 + OEPIIa, 24 h) prevents the increased neurite sprouting and elongation characteristic of C3 expression (C3 + OEPIIa reconstruction) and favors actin polymerization at the distal fractions of budding neurites (arrowhead). Conversely, the overall F-actin content is reduced in comparison with PIIa overexpression alone (compare with Fig. 5 A). (C) Constitutively active RhoA induces a differentiation-arrested phenotype (Fig. 1 A and V14RhoA reconstruction). Reduction of PIIa levels partially reverts this arrest (V14RhoA + ASPIIa reconstruction); these cells show a less polymerized actin cytoskeleton, namely along elongated processes (arrowhead) but do not sprout more neurites (compare with Fig. 1 B). (D) Statistical analysis shows that PIIa antisense (V14 + ASPIIa, n = 28) incompletely rescues the V14-induced (V14, n = 22) phenotype as neurons extend longer neurites, but there is no detectable change in neurite number. On the contrary, compared with single HA-C3 transfection (C3, n = 22), co-transfection of HA-C3 and PIIa-GFP (C3 + OEPIIa, n = 24) results in a decrease of neurite number and length. Data are mean + SD values. *, P < 0.001, V14 + ASPIIa in comparison with V14; **, P < 0.001, C3 + OEPIIa in comparison with C3; °, P < 0.05, V14 + ASPIIa in comparison with V14. Bars, 10 μm.

Figure 8.

Extracellular stimuli mimic the RhoA/ROCK modulation of PIIa. (A) Neurons treated with different neurotrophins (NT-3, BDNF or NGF) at 50 ng/ml sprout numerous long neurites (Control) compared with untreated cells (compare with Fig. 1 A). Cells treated and transfected with HA-V14RhoA, myc-ROCK, or PIIa-GFP are refractory to these growth-promoting effects. (B) Neurons overexpressing ROCK (arrowhead and myc inset in the myc-ROCK panel), constitutively active RhoA (arrowhead and HA inset in the HA-V14 panel), or PIIa (arrowhead and GFP inset in the PIIa-GFP panel) interfere with the effect of the growth-promoting substratum laminin (unmarked cells in the respective panels). Cells either overexpressing inactive Rho (arrowhead and inset in the HA-C3 [24 h] panel), or treated with Y-27632 (18 h) or with low PIIa levels (ASPIIa) are able to differentiate in the presence of a growth unfavorable collagen-rich substratum that normally induces reduced sprouting and elongation (exemplified by unmarked cells in the HA-C3 panel). Transfected cells in individual insets are either labeled with specific antibodies against myc (inset in myc-ROCK panel) and HA (insets in HA-V14 and HA-C3 panels) or detected by GFP fluorescence (inset in PIIa-GFP panel). (C) Quantitative analysis of experiments shown in A and B. Measurements of neurite length and number shows that the effects of different extracellular stimuli on neurite sprouting and initial elongation can be modulated by modifying the activity of RhoA, ROCK and PIIa. The same number of cells was used for each experimental condition tested for every external stimuli: NT-3, n = 31; BDNF, n = 29; NGF, n = 34; Laminin, n = 23; and Collagen, n = 18. Data are mean + SD values; *, P < 0.001, in comparison with respective controls presented with same extracellular stimuli (grouped by lower horizontal bar). Cells were labeled with phalloidin. Bars, 10 μm.

Figure 9.

Proposed involvement of RhoA/ROCK/PIIa in the regulation of mammalian neuritogenesis. (A) At steady state, before neuritogenesis ensues, cellular shape is kept as the submembranous actin cytoskeleton is evenly stabilized (external signals are not depicted for simplicity). This depends, at least in part, on the dominance of the RhoA–ROCK–PIIa pathway described here, where RhoA recruits and activates ROCK which, in turn, phosphorylates PIIa. Importantly, in recently dissociated neurons most of the PIIa is phosphorylated (Fig. 6 C). The neuritogenic–arrest complex might comprise other yet unidentified proteins that influence ROCK–PIIa complex formation, regulate PIP2 function, and/or bring to close vicinity other actin-related proteins (such as cofilin or myosin). (B) Upon contact of a segment of the membrane with a growth- positive signal, RhoA is inactivated and locally dissociates from the membrane. This reduces the interaction between ROCK and PIIa and decreases phosphorylated PIIa (and probably reduces local accumulation of PIP2). This translates into a very localized shift in the local F/G actin ratio, favoring the monomeric form leading to actin instability and sprout formation. Localization of the incoming signal, whether qualitative or quantitative (gradient), would explain how a ubiquitously distributed intracellular pathway can be modulated to allow neuritogenesis at a particular spot of the cell sphere (between dashed lines).

Cited by

Circular RNA in Non-small Cell Lung Carcinoma: Identification of Targets and New Treatment Modalities.
Weidle UH, Birzele F. Weidle UH, et al. Cancer Genomics Proteomics. 2023 Dec;20(6suppl):646-668. doi: 10.21873/cgp.20413. Cancer Genomics Proteomics. 2023. PMID: 38035705 Free PMC article. Review.
What Is the Role of the Rho-ROCK Pathway in Neurologic Disorders?
Benarroch E. Benarroch E. Neurology. 2023 Sep 19;101(12):536-543. doi: 10.1212/WNL.0000000000207779. Neurology. 2023. PMID: 37722862 Free PMC article. No abstract available.
An Alzheimer's disease risk variant in TTC3 modifies the actin cytoskeleton organization and the PI3K-Akt signaling pathway in iPSC-derived forebrain neurons.
Cukier HN, Duarte CL, Laverde-Paz MJ, Simon SA, Van Booven DJ, Miyares AT, Whitehead PL, Hamilton-Nelson KL, Adams LD, Carney RM, Cuccaro ML, Vance JM, Pericak-Vance MA, Griswold AJ, Dykxhoorn DM. Cukier HN, et al. Neurobiol Aging. 2023 Nov;131:182-195. doi: 10.1016/j.neurobiolaging.2023.07.007. Epub 2023 Jul 17. Neurobiol Aging. 2023. PMID: 37677864
Phosphorylation of the actin-binding protein profilin2a at S137 modulates bidirectional structural plasticity at dendritic spines.
Cornelius J, Haak S, Rothkegel M, Korte M, Michaelsen-Preusse K. Cornelius J, et al. Front Cell Dev Biol. 2023 Feb 15;11:1107380. doi: 10.3389/fcell.2023.1107380. eCollection 2023. Front Cell Dev Biol. 2023. PMID: 36875774 Free PMC article.
Molecular Mechanisms Involved in the Regulation of Neurodevelopment by miR-124.
Gu X, Xu X, Jia C, Wang J, Zhang J, Gao Q, Chen J. Gu X, et al. Mol Neurobiol. 2023 Jul;60(7):3569-3583. doi: 10.1007/s12035-023-03271-5. Epub 2023 Feb 25. Mol Neurobiol. 2023. PMID: 36840845 Review.

References

1. Arber, S., F.A. Barbayannis, H. Hanser, C. Schneider, C.A. Stanyon, O. Bernard, and P. Caroni. 1998. Regulation of actin dynamics through phosphorylation of cofilin by LIM-kinase. Nature. 393:805–809. - PubMed
1. Bamburg, J.R. 1999. Proteins of the ADF/cofilin family: essential regulators of actin dynamics. Annu. Rev. Cell Dev. Biol. 15:185–230. - PubMed
1. Baum, P.D., and G. Garriga. 1997. Neuronal migrations and axon fasciculation are disrupted in ina-1 integrin mutants. Neuron. 19:51–62. - PubMed
1. Billuart, P., C.G. Winter, A. Maresh, X. Zhao, and L. Luo. 2001. Regulating axon branch stability: the role of p190 RhoGAP in repressing a retraction signaling pathway. Cell. 107:195–207. - PubMed
1. Bito, H., T. Furuyashiki, H. Ishihara, Y. Shibasaki, K. Ohashi, K. Mizuno, M. Maekawa, T. Ishizaki, and S. Narumiya. 2000. A critical role for a Rho-associated kinase, p160ROCK, in determining axon outgrowth in mammalian CNS neurons. Neuron. 26:431–441. - PubMed

RhoA/ROCK regulation of neuritogenesis via profilin IIa-mediated control of actin stability - PubMed (original) (raw)