Influence of cAMP and protein kinase A on neurite length from spiral ganglion neurons

Abstract

Regrowth of peripheral spiral ganglion neuron (SGN) fibers is a primary objective in efforts to improve cochlear implant outcomes and to potentially reinnervate regenerated hair cells. Cyclic adenosine monophosphate (cAMP) regulates neurite growth and guidance via activation of protein kinase A (PKA) and Exchange Protein directly Activated by Cylic AMP (Epac). Here we explored the effects of cAMP signaling on SGN neurite length in vitro. We find that the cAMP analog, cpt-cAMP, exerts a biphasic effect on neurite length; increasing length at lower concentrations and reducing length at higher concentrations. This biphasic response occurs in cultures plated on laminin, fibronectin, or tenascin C suggesting that it is not substrate dependent. cpt-cAMP also reduces SGN neurite branching. The Epac-specific agonist, 8-pCPT-2'-O-Me-cAMP, does not alter SGN neurite length. Constitutively active PKA isoforms strongly inhibit SGN neurite length similar to higher levels of cAMP. Chronic membrane depolarization activates PKA in SGNs and also inhibits SGN neurite length. However, inhibition of PKA fails to rescue neurite length in depolarized cultures implying that activation of PKA is not necessary for the inhibition of SGN neurite length by chronic depolarization. Expression of constitutively active phosphatidylinositol 3-kinase, but not c-Jun N-terminal kinase, isoforms partially rescues SGN neurite length in the presence of activated PKA. Taken together, these results suggest that activation of cAMP/PKA represents a potential strategy to enhance SGN fiber elongation following deafness; however such therapies will likely require careful titration so as to promote rather than inhibit nerve fiber regeneration.

Figure 1

cAMP exerts a biphasic effect on SGN neurite length. A. Representative images of spiral ganglion cultures maintained in NT-3 or NT-3 with cpt-cAMP (1 mM, 10mM) as indicated for 48 h and labeled with anti-NF200. Scale bars=100 μm and 25 μm (bottom right panel). B. Mean neurite length in SG cultures maintained in NT-3-negative control medium, NT-3 (50 ng/ml, NT-3-positive control), NT-3 with cpt-cAMP (0.1–10 mM), or cpt-cAMP (1 mM). Each condition was performed in duplicate (2 wells/repetition) and at least 60 neurites were measured for each condition. The conditions were repeated in at least three separate culture preparations (≥ 6 wells total). The total number of neurites scored in each condition is indicated in each bar. *p<0.05, one way ANOVA followed by a Holm-Sidak post-hoc comparison. C. Normalized average SGN survival in cultures maintained in NT-3-negative control medium, NT-3 (50 ng/ml, NT-3-positive control), NT-3 with cpt-cAMP (0.1–10 mM), or cpt-cAMP (1 mM). 100% represents survival in NT-3. *p<0.05, one way ANOVA followed by a Holm-Sidak post-hoc comparison compared to NT-3-positive control. D. Mean neurite length in SG cultures maintained NT-3 with or without forskolin (FSK, 0.1–10 μM). *p<0.05, one way ANOVA followed by a Holm-Sidak post-hoc comparison compared to NT-3-positive control. E. Normalized average SGN survival in SG cultures maintained NT-3 with or without forskolin (FSK, 0.1–10 μM). *p<0.05, one way ANOVA followed by a Holm-Sidak post-hoc comparison compared to NT-3-positive control.

Figure 2

Influence of cAMP on neurite branching. A. Mean primary branching index in cultures maintained in control medium, NT-3 with or without cpt-cAMP (0.1–10 mM), or cpt-cAMP (1 mM). B. Mean secondary branching index in cultures maintained in control medium, NT-3 with or without cpt-cAMP (0.1–10 mM), or cpt-cAMP (1 mM). *p<0.05, one way ANOVA followed by a Holm-Sidak post-hoc comparison compared with NT3-positive control cultures. *p<0.05, one way ANOVA followed by a Holm-Sidak post-hoc comparison compared with NT3-negative control cultures. **p<0.05, one way ANOVA followed by a Holm-Sidak post-hoc comparison compared with NT3-negative control cultures.

Figure 3

The biphasic effect of cpt-cAMP on SGN neurite length is not dependent on culture substrate. Mean neurite length in SG cultures plated on fibronectin or tenascin C and maintained in NT-3 with or without cpt-cAMP (1, 10 μM) or forskolin (FSK, 1 μM). The total number of neurites scored in each condition is indicated in each bar. *p<0.05, one way ANOVA followed by a Holm-Sidak post-hoc comparison.

Figure 4

8-pCPT-2’-O-Me-cAMP, an Epac selective cAMP analog, does not alter SGN neurite length or survival. A. Mean neurite length of SGNs treated with 8-pCPT-2’-O-Me-cAMP (200 μM) or Rp-cAMPS (1 mM) in the presence of NT-3 was not significantly different from control SGNs maintained in NT-3 (p>0.05, one way ANOVA). The total number of neurites scored in each condition is indicated in each bar. B. Normalized average SGN survival in cultures treated with 8-pCPT-2’-O-Me-cAMP (200 μM) or Rp-cAMPS (1 mM) in the presence of NT-3 was not significantly different from control SGNs maintained in NT-3 (p>0.05, one way ANOVA).

Figure 5

Expression of GPKA, a constitutively active PKA isoform, reduces SGN neurite length. A&B. SG cultures transfected with GFP (A, green) or GPKA (B, green) and immunostained with anti-NF200 (red). Scale bar=100 μm. C. Higher magnification image of a SGN transfected with GPKA and immunostained with anti-NF200. Scale bar=25 μm. There are a few transfected non-neuronal cells (GFP-positive, NF200-negative) in each panel of A, B, and C. D. Mean neurite length of SGNs expressing GFP or GPKA in the presence of NT-3 (*p<0.001, Mann-Whitney rank sum test). The total number of neurites scored in each condition is indicated in each bar. E. Normalized average survival of SGNs expressing GFP or GPKA in the presence of NT-3 (*p<0.05, Student’s t-test). F&G. Correlation of neurite length with expression level based on GFP fluorescence intensity in arbitrary units in a subset of SGNs expressing GFP (F) or GPKA (G). Pearson’s correlation coefficients were r=−0.19 for GFP and r=−0.50 (p<0.001) for GPKA.

Figure 6

GPKA reduces SGN neurite length regardless of nuclear targeting. A. Mean neurite length of SGNs expressing GFP, GPKAnes, or GPKAnls in the presence of NT-3 (*p<0.05, one way ANOVA followed by a Holm-Sidak post-hoc comparison). The total number of neurites scored in each condition is indicated in each bar. B. Representative images of SGNs transfected with GPKAnls and immunostained with anti-NF200 (red). Scale bar=25 μm. There is also a transfected non-neuronal cell (GFP-positive, NF200-negative). C. Normalized average survival of SGNs expressing GFP, GPKAnes, or GPKAnls in the presence of NT-3 (*p<0.05, one way ANOVA followed by a Holm-Sidak post-hoc comparison).

Figure 7

Expression of constitutively active PI3-K, but not constitutively active JNK, isoforms rescues neurite length in SGNs expressing GPKA. A. SGN co-transfected with P110 and GPKA and immunostained with anti-NF200 (red). Scale bar=25 μm. A portion of the neurite travels in and out of the focal plane. There are a few transfected non-neuronal cells (GFP-positive, NF200-negative). B. Mean neurite length of SGNs co-transfected with GPKA and either empty vector, P110, or MKK-JNK1 (*p<0.05 by one way ANOVA followed by a Holm-Sidak post-hoc comparison). The total number of neurites scored in each condition is indicated in each bar. C. Normalized average survival of SGNs transfected with GPKA, GPKA+P110, or GPKA+MKK-JNK1 (*p<0.05, one way ANOVA followed by a Holm-Sidak post-hoc comparison).

Figure 8

PKA activity is not required for the reduction of neurite length by depolarization. A. Spiral ganglion cultures were maintained in NT-3, NT-3+30K, or NT-3+80K with or without Rp-cAMPS (1 μM) for 48 h as indicated. Mean neurite length in cultures maintained NT-3+30K+Rp-cAMPS and NT-3+80K+Rp-CAMPS was not significantly different (p>0.05, one way ANOVA) from cultures maintained in NT-3+30K and NT-3+80K, respectively. B. Spiral ganglion cultures were transfected with expression plasmids encoding GFP-tagged PKA inhibitory peptide (GPKI) or GFP. Cultures were maintained in NT-3, NT-3+30K, or NT-3+80K for 48 h after transgene expression as indicated. Mean neurite length in cultures maintained in NT-3+30K+GPKI and NT-3+80K+GPKI were not significantly different (p>0.05) from mean neurite length in cultures maintained in NT-3+30K+GFP and NT-3+80K+GFP, respectively, but both were significantly different than NT-3+GFP (p<0.05, one way ANOVA followed by a Holm-Sidak post-hoc comparison). The total number of neurites scored in each condition is indicated in each bar.