Bcl-2 enhances Ca(2+) signaling to support the intrinsic regenerative capacity of CNS axons

Abstract

At a certain point in development, axons in the mammalian CNS undergo a profound loss of intrinsic growth capacity, which leads to poor regeneration after injury. Overexpression of Bcl-2 prevents this loss, but the molecular basis of this effect remains unclear. Here, we report that Bcl-2 supports axonal growth by enhancing intracellular Ca(2+) signaling and activating cAMP response element binding protein (CREB) and extracellular-regulated kinase (Erk), which stimulate the regenerative response and neuritogenesis. Expression of Bcl-2 decreases endoplasmic reticulum (ER) Ca(2+) uptake and storage, and thereby leads to a larger intracellular Ca(2+) response induced by Ca(2+) influx or axotomy in Bcl-2-expressing neurons than in control neurons. Bcl-x(L), an antiapoptotic member of the Bcl-2 family that does not affect ER Ca(2+) uptake, supports neuronal survival but cannot activate CREB and Erk or promote axon regeneration. These results suggest a novel role for ER Ca(2+) in the regulation of neuronal response to injury and define a dedicated signaling event through which Bcl-2 supports CNS regeneration.

Figure 1

Bcl-x_L supports survival but not axon regeneration of postnatal RGCs. (A) Transverse retinal sections from wt, Bcl-x_Ltg, and Bcl-2tg mice were stained with TUNEL and GAP-43 antibody on day 1 after optic nerve injury. Asterisks indicate TUNEL-positive cells; arrows indicate the nerve fiber layer (NFL). TUNEL-positive cells are present and the NFL is absent in the retinal sections of wt mice but not Bcl-x_Ltg or Bcl-2tg mice. GCL, ganglion cell layer; IPL, inner plexiform layer. Scale bar, 100 μm. (B) Number of TUNEL-positive cells in retinal sections of wt, Bcl-x_Ltg and Bcl-2tg mice (n=4/group). (C) Longitudinal optic nerve sections from wt, Bcl-x_Ltg, and Bcl-2tg mice labeled with GAP-43 antibody 1–2 days after optic nerve injury. Arrows point to the crush site. Labeled axons in Bcl-x_Ltg mice remain anterior to the crush site, while those in Bcl-2tg mice grew robustly past the lesion site. Scale bar, 250 μm. (D) Retinal axon regrowth in retina–brain slice cocultures prepared from wt, Bcl-x_Ltg, and Bcl-2tg mice. Values are mean±s.d. ^*P<0.01 versus wt, two-tailed t-test.

Figure 2

Bcl-2 acts intrinsically in neurons to support neuritogenesis and axon regeneration. (A–C) RGCs were isolated from wt (A), Bcl-x_Ltg (B), and Bcl-2tg (C) mice, incubated for 5 days, and stained with calcein. Surviving RGCs from Bcl-2tg mice extended longer axons than those from wt or Bcl-x_Ltg mice. (D–F) Percentage of surviving RGCs (D), percentage of surviving RGCs with axons longer than 3 body lengths (E), and average length of the longest axon from each RGC (F) (n=5 cultures/group). RGCs in Bcl-x_Ltg and Bcl-2tg mice had similar survival rates, but Bcl-x_Ltg RGCs had shorter neurites. (G) Western blot analysis of Bcl-2 and Bcl-x_L expression in stably transfected PC12 cells. (H, I) Percentage of surviving cells (H) and percentage of cells bearing neurites (I) (n=5 cultures/group). Bcl-x_L- and Bcl-2-expressing PC12 cells had similar survival rates, but significantly fewer Bcl-x_L-expressing cells had neurites. ^*P<0.01 versus wt, two-tailed t-test.

Figure 3

The growth-promoting effect of Bcl-2 is ER-dependent. (A) Schematic of DNA structures of Bcl-2, Bcl-x_L, and Bcl-2 mutants. (B) Western blot analysis of PC12 cell lines stably transfected with Bcl-2ER, Bcl-2TM, and Bcl-2MOM. (C) Subcellular location of Bcl-2ER, Bcl-2MOM, and Bcl-x_L shown by confocal microscopy. EGFP-Bcl-2ER (green) colocalizes with the ER marker calnexin (blue); EGFP-Bcl-2MOM and EGFP-Bcl-x_L colocalize with the mitochondrial marker (Mito) cytochrome c (red). Scale bar, 4 μm. (D) Percentage of dying cells after treatment with staurosporine (left) and percentage that extended neurites (n=4 cultures/group). ER, Bcl-2ER; TM, Bcl-2TM; MOM, Bcl-2MOM. ^*P<0.05 versus wt, two-tailed t-test.

Figure 4

Bcl-2-, but not Bcl-x_L-, expressing neurons display reduced ER Ca²⁺ content. (A–C) Representative trace (A) and quantitative analysis of basal (B) and TG-induced [Ca²⁺]_i (C), measured by Fura-2, in PC12 cells expressing a control (Cont), Bcl-2, or Bcl-x_L plasmid. Measurement of TG-induced Ca²⁺ change was carried out in the absence of extracellular Ca²⁺. (D, E) TG-induced [Ca²⁺]_i (D) and NGF-induced neurite outgrowth (E) in control (Cont), Bcl-2-, or Bcl-2+SERCA2b-expressing PC12 cells (n=4 cultures/group). (F) NGF-induced neurite outgrowth in control and Bcl-2-expressing PC12 cells treated with 0–10 mM BHQ (n=4/group). ^*P<0.01 versus control, two-tailed t-test.

Figure 5

Bcl-2 expression enhances intracellular Ca²⁺ signaling by reducing ER Ca²⁺ uptake. (A–C) Representative trace (A) and quantitative analysis of extracellular Ca²⁺- (B) and KCl-induced changes in [Ca²⁺]_i (C), measured by Fura-2 (n=4/group). PC12 cells stably transfected with control (Cont), Bcl-2, or Bcl-x_L plasmids were incubated in Ca²⁺-free medium for over an hour; where indicated, Ca²⁺ (1.0 mM free extracellular Ca²⁺ final concentration) was added. Changes in [Ca²⁺]_i were measured from its baseline to when the elevation of [Ca²⁺]_i reached its peak. (D) Comparison of KCl-induced [Ca²⁺]_i in cells expressing control (Cont), Bcl-2, Bcl-x_L, Bcl-2ER (ER), Bcl-2TM (TM), and Bcl-2MOM (MOM) genes (n⩾3/group). (E) Measurement of KCl-induced changes in [Ca²⁺]_i in cells expressing a control, Bcl-2, or Bcl-2+SERCA2b plasmid. ^*P<0.01 versus control, two-tailed t-test.

Figure 6

Bcl-2 activates CREB and Erk to promote neuritogenic response. (A) Western blot analysis of time-dependent phosphorylation of CREB and Erk in PC12 cells expressing control, Bcl-2, or Bcl-x_L plasmids treated with KCl (30 mM). Antibodies recognizing the unphosphorylated forms of CREB and Erk served as controls. (B, C) CREB-dependent (B) and Erk-dependent (C) reporter gene activities in stably transfected PC12 cells in the presence or absence of KCl (30 mM) (n⩾4/group). (D–F) Neurite outgrowth (D, F) and neuronal survival (E) in control (Cont) and Bcl-2-expressing cells (Bcl-2) incubated with 1 ng/ml NGF (n=4/group). Cultures were incubated in the absence (control) or presence of either a viral vector carrying a LacZ reporter gene or dominant-negative CREB (mCREB) (D, E) or the MEK inhibitor U0126 (100 nM) (F). ^*P<0.01, two-tailed t-test.

Figure 7

Optic nerve injury in Bcl-2tg mice increases the intracellular Ca²⁺ response and activates CREB and Erk in RGCs. (A–C) Representative trace (A) and quantitative analysis of basal (B) and KCl-induced changes in [Ca²⁺]_i (C), measured by Fura-2, in RGCs of wt, Bcl-2tg, and Bcl-x_Ltg mice (n⩾3/group). Changes in [Ca²⁺]_i were measured from its baseline to when the elevation of [Ca²⁺]_i reached its peak. Values are means±s.d. ^*P<0.01 versus wt, two-tailed t test. (D) Immunofluorescence staining for pCREB and pErk in retinal sections from adult wt and Bcl-2tg mice at day 1 after optic nerve crush. Scale bar, 50 μm.

Figure 8

Model of intracellular Ca²⁺ responses and subsequent signaling events in RGCs of wt, Bcl-x_Ltg, and Bcl-2tg mice. (A) Neural injury or axotomy results in an increase of extracellular Ca²⁺, a surge of Ca²⁺ influx, and elevation of [Ca²⁺]_i. In neurons of wt mice, the excess intracellular Ca²⁺ is absorbed by the ER and translocated to the mitochondria, where it initiates apoptosis. (B) Bcl-x_L expression, which is targeted to the mitochondria, prevents Ca²⁺-induced activation of the apoptotic signal and supports neuronal survival but does not affect ER Ca²⁺ uptake or the intracellular Ca²⁺ response to injury. (C) Expression of Bcl-2, however, is targeted primarily to the ER, where it reduces ER Ca²⁺ uptake and Ca²⁺ translocation to the mitochondria after axotomy, thereby preventing neuronal apoptosis. In addition, Bcl-2 reduces ER Ca²⁺ uptake, leading to greater elevation of [Ca²⁺]_i after injury, thereby activating CREB and Erk and stimulating a neuritogenic response and axon regeneration.