Activation of extracellular signal-regulated protein kinase 5 is essential for cystitis- and nerve growth factor-induced calcitonin gene-related peptide expression in sensory neurons

Abstract

Background: Cystitis causes considerable neuronal plasticity in the primary afferent pathways. The molecular mechanism and signal transduction underlying cross talk between the inflamed urinary bladder and sensory sensitization has not been investigated.

Results: In a rat cystitis model induced by cyclophosphamide (CYP) for 48 h, the mRNA and protein levels of the excitatory neurotransmitter calcitonin gene-related peptide (CGRP) are increased in the L6 dorsal root ganglia (DRG) in response to bladder inflammation. Cystitis-induced CGRP expression in L6 DRG is triggered by endogenous nerve growth factor (NGF) because neutralization of NGF with a specific NGF antibody reverses CGRP up-regulation during cystitis. CGRP expression in the L6 DRG neurons is also enhanced by retrograde NGF signaling when NGF is applied to the nerve terminals of the ganglion-nerve two-compartmented preparation. Characterization of the signaling pathways in cystitis- or NGF-induced CGRP expression reveals that the activation (phosphorylation) of extracellular signal-regulated protein kinase (ERK)5 but not Akt is involved. In L6 DRG during cystitis, CGRP is co-localized with phospho-ERK5 but not phospho-Akt. NGF-evoked CGRP up-regulation is also blocked by inhibition of the MEK/ERK pathway with specific MEK inhibitors U0126 and PD98059, but not by inhibition of the PI3K/Akt pathway with inhibitor LY294002. Further examination shows that cystitis-induced cAMP-responsive element binding protein (CREB) activity is expressed in CGRP bladder afferent neurons and is co-localized with phospho-ERK5 but not phospho-Akt. Blockade of NGF action in vivo reduces the number of DRG neurons co-expressing CGRP and phospho-CREB, and reverses cystitis-induced increases in micturition frequency.

Conclusions: A specific pathway involving NGF-ERK5-CREB axis plays an essential role in cystitis-induced sensory activation.

Figure 1

Cystitis increased CGRP mRNA and protein levels in L6 DRG. After CYP treatment for 48 h, the CGRP expression was examined in L6 DRG by immunohistochemistry (A-C) and real-time PCR (D). CGRP was largely expressed in small diameter sensory neurons (A, B). The average number of DRG neurons per unit area expressing CGRP was significantly increased post CYP treatment (C). The relative level of CGRP mRNA was also increased in L6 DRG during cystitis (D). Bar = 80 μm. *, p < 0.05 vs control. n = 5-6 animals for each group.

Figure 2

NGF immuno-neutralization attenuated cystitis-induced CGRP expression in L6 DRG. The number of CGRP immunoreactive neurons was significantly higher in L6 DRG from animals treated with CYP and control IgG (A, C) when compared to those from animals treated with NGF neutralizing antibody and CYP (B, C). Blockade of NGF activity in cystitis animals also reduced CGRP transcription in the L6 DRG (D). *, p < 0.05 vs CYP + IgG. n = 5 animals for each group. Bar = 60 μm.

Figure 3

Co-localization of CGRP with phospho-ERK5 but not phospho-Akt in L6 DRG during cystitis. Double immunostaining showed that a subpopulation of CGRP cells (A, green cells indicated by yellow arrows) also expressed phospho-ERK5 (B, red nuclear staining). In contrast, CGRP cells did not express phospho-Akt (D-F, blue arrows indicated CGRP cells, white arrows indicated Akt cells).

Figure 4

Retrograde NGF-induced CGRP expression was mediated by the MEK/ERK pathway but not by the PI3K/Akt pathway. In two-compartmented culture, the gangalia was pre-treated with vehicle (vehicle, A, B, E), or one of the inhibitors against the MEK/ERK pathway (C, F: U0126; G: PD98059), or the PI3K/Akt pathway (D, H: LY294002) followed by NGF treatment of the nerve terminals (B, C, D). After 12 h of NGF treatment, the level of CGRP was increased in the sensory neuronal cell bodies by 2 fold when compared to control (E). Both U0126 (F) and PD98059 (G: PD) pre-treatment blocked the retrograde NGF-facilitated CGRP expression in DRG. LY294002 (LY) pre-treatment did not affect the level of CGRP in the ganglia induced by retrograde NGF (H). Bar = 120 μm. *, p < 0.05 vs vehicle. Results were from 4 independent experiments for each treatment.

Figure 5

Co-localization of CGRP with phospho-CREB in bladder afferent neurons during cystitis. Double immunostaining showed that CGRP cells (A, green cells indicated by yellow arrows) often expressed phospho (p)-CREB (B, red nuclear staining, C was the merged photomicrograph of A and B indicating co-localization of CGRP and p-CREB: yellow arrows) during cystitis. In FB-labeled bladder afferent neurons (D, blue cells indicated by yellow arrows), CGRP (E, green cells) was also co-expressed with p-CREB (F: red nuclear staining; G was the merged photomicrograph of D, E and F indicating bladder afferent neurons expressing both CGRP and p-CREB indicated by yellow arrows). White arrows indicated CGRP neurons that did not have p-CREB. Bar = 60 μm.

Figure 6

Co-localization of phospho-CREB with phospho-ERK5 but not phospho-Akt in L6 DRG during cystitis. Double immunostaining showed that phospho-CREB in L6 DRG during cystitis (A, D: red nuclear staining) was co-localized with phospho-ERK5 (B, C: green nuclear staining) but not phospho-Akt (E, F: green cells). Bar = 60 μm.

Figure 7

NGF immuno-neutralization reduced the level of CREB phosphorylation in CGRP neurons during cystitis and reversed cystitis-induced bladder overactivity. The number of phospho (p)-CREB immunoreactive neurons was significantly higher in L6 DRG from animals treated with CYP and control IgG (A, C) when compared to those from animals treated with NGF neutralizing antibody and CYP (B, C). NGF neutralization also reduced cystitis-induced increases in the number of L6 DRG neurons co-expressing CGRP and p-CREB (D), and reversed cystitis-induced increases in micturition frequency (E). *, p < 0.05 vs CYP + IgG. #, p < 0.05 vs vehicle treatment. n = 4 animals for each group. Bar = 40 μm.

Figure 8

Schematic diagram illustrates the putative mechanism for NGF signal transduction that mediates the interaction of the inflamed urinary bladder and bladder sensory neurons in the DRG. Cystitis-induced NGF expression in the urinary bladder binds to TrkA and activates ERK5 and CREB in DRG via retrograde transport. The NGF-ERK5-CREB axis leads to CGRP up-regulation in bladder afferent neurons (A). The NGF-induced Akt activation (B) may contribute to TRPV1 sensitization [61] but not the CGRP up-regulation in the DRG. In addition to regulating CGRP expression, activation of CREB in bladder afferent neurons may also control gene expression of other neuropeptides (NPP) (C). These multi-branches of signaling pathways triggered by NGF may together mediate cystitis-induced bladder hyperactivity.