C5b-9-activated, K(v)1.3 channels mediate oligodendrocyte cell cycle activation and dedifferentiation

Abstract

Voltage-gated potassium (K(v)) channels play an important role in the regulation of growth factor-induced cell proliferation. We have previously shown that cell cycle activation is induced in oligodendrocytes (OLGs) by complement C5b-9, but the role of K(v) channels in these cells had not been investigated. Differentiated OLGs were found to express K(v)1.4 channels, but little K(v)1.3. Exposure of OLGs to C5b-9 modulated K(v)1.3 functional channels and increased protein expression, whereas C5b6 had no effect. Pretreatment with the recombinant scorpion toxin rOsK-1, a highly selective K(v)1.3 inhibitor, blocked the expression of K(v)1.3 induced by C5b-9. rOsK-1 inhibited Akt phosphorylation and activation by C5b-9 but had no effect on ERK1 activation. These data strongly suggest a role for K(v)1.3 in controlling the Akt activation induced by C5b-9. Since Akt plays a major role in C5b-9-induced cell cycle activation, we also investigated the effect of inhibiting K(v)1.3 channels on DNA synthesis. rOsK-1 significantly inhibited the DNA synthesis induced by C5b-9 in OLG, indicating that K(v)1.3 plays an important role in the C5b-9-induced cell cycle. In addition, C5b-9-mediated myelin basic protein and proteolipid protein mRNA decay was completely abrogated by inhibition of K(v)1.3 expression. In the brains of multiple sclerosis patients, C5b-9 co-localized with NG2(+) OLG progenitor cells that expressed K(v)1.3 channels. Taken together, these data suggest that K(v)1.3 channels play an important role in controlling C5b-9-induced cell cycle activation and OLG dedifferentiation, both in vitro and in vivo.

Figure 1. MBP, PLP, and K_v1.3 mRNA expression during OLG differentiation

OPCs were cultured in defined medium for the indicated times, and the expression of K_v1.3, K_v 1.4, MBP, and PLP mRNA was determined. A. Expression of MBP and PLP mRNA, as determined by real-time PCR, significantly increased during OLG differentiation. B. K_v1.3 mRNA was present in the OPC cells (0 h), and the levels decreased significantly at 48h of differentiation (p=0.0007). C. The effect of differentiation on K_v protein expression was examined by western blotting and expressed as a ratio to β-actin. High levels of K_v1.3 protein were found in OPCs, and these levels decreased significantly during differentiation (p=0.029). The levels of K_v1.4 did not change significantly during differentiation. The expression of the mRNA or protein at the beginning of the experiment (0 h) was considered to be 100%. Results of three separate experiments are expressed as mean ± SEM, relative to the value at 0 h.

Figure 2. Effect of C5b-9 on K_v1.3 expression in OLGs

A. C5b-9 induces K_v1.3 expression. OLGs were cultured in defined medium and exposed to C5b-9 or C5D, and K_v1.3 expression was determined by western blotting. K_v1.3 protein expression was found to be low in the differentiated OLGs (CTR) and was significantly increased after exposure to C5b-9 (p=0.042). C5D serum had no significant effect on K_v1.3 expression, indicating that the increase seen was dependent on the assembly of the terminal pathway. The results are expressed as means ± SEM. Data are derived from three separate experiments. B. rOsK -1 inhibits K_v1.3 expression induced by C5b-9. OLGs were pretreated with rOsk-1 (15nM) for 1 h and then exposed to C5b-9 for 30 min. Lysates were analyzed by western blotting for K_v1.3 and β-actin expression. Exposure to rOsk-1 was able to reverse the effect of C5b-9 on K_v1.3 expression (p=0.007), indicating that channel expression is required for the C5b-9 effect. Results of three separate experiments are expressed as mean ± SEM, relative to the control (CTR).

Figure 3. Effect of rOsk-1 on C5b-9 induced outward currents in OLGs

K_v currents were elicited by 500-ms pulses stepped from a holding potential of −60 to +60 mV, preceded by a conditioning pulse of −110 mV for 200 ms. C5b-9 was added directly to the experimental chamber. After 5–8 minutes of incubation, the measurements were repeated, then 5nM rOsK-1 5nM final concentration was added to the experimental chamber and measurements were repeated. A–C. Examples of whole-cell current families derived from an unstimulated cell (A), the same cell stimulated with C5b-9 (B) and treated with 5nM rOsK-1 (C). D. Relative membrane conductance as a function of membrane potential for control (unstimulated) cell (dots), after stimulation with C5b-9 (squares) and after treatment with 5 nM rOsK-1 (triangles). E. Example of the current elicited by C5b-9 and blocked by rOsK-1. Point-by-point subtraction of the current traces obtained after application of rOsK-1, from the current traces evoked by C5b-9 revealed a moderately inactivating, K_v1.3-like current.

Figure 4. The effect LY294002 on C5b-9-induced K_v1.3 and outward current

A. OLGs were pretreated with the PI3K inhibitor LY294002 (10 or 20 µM) for 1 h, then exposed to C5b-9 for 30 min. Lysates were analyzed by western blotting for K_v1.3 as described above. Exposure to LY294002 (20 µM) was able to reverse the effect of C5b-9 on K_v1.3 (p=0.016). B. Point-by-point subtraction of the current traces elicited from the C5b-9-stimulated cells obtained prior to and after application of LY294002 (20 µM) revealed the whole-cell currents blocked by LY294002. The electrophysiological measurements were performed in 10 cells. C, D. OLGs were pretreated with the PI3K inhibitor LY294002 (10 or 20 µM) for 1 h, then exposed to C5b-9 for 30 min. Lysates were analyzed by western blotting for Akt and FOXO1 as described above. Exposure to LY294002 was able to reverse the effect of C5b-9 on Akt (C) and FOXO1 phosphorylation (D). Results of three separate experiments are expressed as mean ± SEM, relative to the CTR.

Figure 5. K_v1.3 expression induced by C5b-9 is dependent on Gi protein and ERK activation

OLGs were pretreated with PTX (500 ng/ml) for 4 h or with PD98059 or SP600125 for 1 h, then exposed to C5b-9 for 30 min. Lysates were analyzed by western blotting for K_v1.3 and β-actin expression as described above. Exposure to PTX was able to reverse the effect of C5b-9 on K_v1.3 (p=0.032) (A). A similar effect was seen when the MEK1 inhibitor PD98059 (p=0.05) was used (B), but JNK inhibition by SP600125 (C) had no effect on K_v1.3 protein expression.

Figure 6. Effect of rOsK-1 on the activation of Akt and ERK1

OLGs were cultured in defined medium and pretreated with rOsK-1 prior to stimulation with C5b-9. The effect of rOSK-1 on Akt (A) and ERK1 (B) phosphorylation was analyzed by western blotting. rOsK-1 was able to inhibit Akt phosphorylation (p=0.018) but had no effect on ERK1 activation.

Figure 7. Effect of K_v1.3 inhibition on DNA synthesis

OLGs were pretreated with rOsk-1 or 4AP for 1 h and then stimulated with serum C5b-9 for 24 h in the presence of 1 µCi [³H]thymidine. C5b-9 significantly increased [³H]thymidine incorporation when compared to C5b6 (p=0.005). Both 4AP (p=0.001) and rOsK-1 (p=0.003) blocked the [³H]thymidine incorporation induced by C5b-9. Three separate experiments were performed, and data are presented as means ± SEM.

Figure 8. Effect of rOsK-1 on C5b-9-induced down-regulation of MBP and PLP expression

OLGs were cultured in defined medium for 56 h, and then stimulated with C5b-9 for 6 h with or without rOsk-1 pretreatment. The expression of MBP and PLP mRNA was determined by real-time PCR. The expression of MBP and PLP decreased significantly after exposure to C5b-9 (p=0.05 for MBP and p=0.004 for PLP). Pretreatment with rOsk-1 abolished the down-regulation of MBP and PLP induced by C5b-9. Results of three separate experiments are expressed as mean ± SEM, relative to CTR (considered to be 100%).

Figure 9. Expression of NG2-positive cells in MS brain and colocalization with K_v1.3

Cryostat sections were immunostained for NG2 expression or double-stained for NG2 and K_v1.3. Expression of NG2 was abundant in MS active lesions, NAWM, and NAGM. NG2⁺ cells in active MS plaques (A) and NAWM (B) displayed multibranched processes. A smaller number of NG2⁺ cells were seen in the NAGM. These cells also appeared to have a lower number of multibranched processes (C). NG2⁺ cells (red deposits) also expressed K_v1.3 (black deposits) (D). E. Control for the immunoperoxidase reaction. A–E: original magnification ×400. Insert: original magnification ×1000.

Figure 10. C5b-9 deposits in MS brain and co-localization with NG2 cells

Cryostat sections of MS brain were stained for C5b-9 or double-stained for C5b-9 and NG2. C5b-9 deposits were found in active MS plaques (A) and NAWM (B). In double-staining analysis, NG2⁺ cells (red deposits) also co-localized with C5b-9 deposits (black deposits) (arrowheads) (C). D. Control for the immunoperoxidase reaction. A–D: original magnification, ×400. Insert: original magnification, ×1000.