Reduction in voltage-gated K+ channel activity in primary sensory neurons in painful diabetic neuropathy: role of brain-derived neurotrophic factor

Abstract

Abnormal hyperexcitability of primary sensory neurons plays an important role in neuropathic pain. Voltage-gated potassium (Kv) channels regulate neuronal excitability by affecting the resting membrane potential and influencing the repolarization and frequency of the action potential. In this study, we determined changes in Kv channels in dorsal root ganglion (DRG) neurons in a rat model of diabetic neuropathic pain. The densities of total Kv, A-type (IA) and sustained delayed (IK) currents were markedly reduced in medium- and large-, but not in small-, diameter DRG neurons in diabetic rats. Quantitative RT-PCR analysis revealed that the mRNA levels of IA subunits, including Kv1.4, Kv3.4, Kv4.2, and Kv4.3, in the DRG were reduced approximately 50% in diabetic rats compared with those in control rats. However, there were no significant differences in the mRNA levels of IK subunits (Kv1.1, Kv1.2, Kv2.1, and Kv2.2) in the DRG between the two groups. Incubation with brain-derived neurotrophic factor (BDNF) caused a large reduction in Kv currents, especially IA currents, in medium and large DRG neurons from control rats. Furthermore, the reductions in Kv currents and mRNA levels of IA subunits in diabetic rats were normalized by pre-treatment with anti-BDNF antibody or K252a, a TrkB tyrosine kinase inhibitor. In addition, the number of medium and large DRG neurons with BDNF immunoreactivity was greater in diabetic than control rats. Collectively, our findings suggest that diabetes primarily reduces Kv channel activity in medium and large DRG neurons. Increased BDNF activity in these neurons likely contributes to the reduction in Kv channel function through TrkB receptor stimulation in painful diabetic neuropathy.

Figure 1. Reduction in the current densities of total Kv, DAP-sensitive I_A, and TEA-sensitive I_K in medium DRG neurons from diabetic rats

A and B, representative traces showing different types of Kv currents in medium DRG neurons from a control and diabetic rat. The neurons were held at −90 mV and depolarized from −70 to 60 mV in 10-mV increments (inset). C, I-V curves show the current densities of total Kv, I_A, and I_K in medium DRG neurons from rats in the control (n = 22 cells) and diabetic (n = 24 cells) groups. D, voltage-dependent activation kinetics (G-V curves) of total Kv, I_A, and I_K in medium DRG neurons from control and diabetic rats. The V_0.5 values of total Kv currents (control rats: 2.77 ± 2.11 mV, n = 22; diabetic rats: 10.57 ± 1.71 mV, n = 25, P < 0.05) and I_K (control rats: 4.85 ± 2.43 mV, n = 17; diabetic rats: 13.92 ± 3.20 mV, n = 20, P < 0.05), but not I_A (control rats: 10.18 ± 5.59 mV, n = 19; diabetic rats: 1.35 ± 4.51 mV, n = 22, P > 0.05), were significantly different between the control and diabetic rats (t-test). There was no significant difference in the k value of total Kv currents (control rats: 17.75 ± 0.75, n = 22; diabetic rats: 19.39 ± 0.70, n = 25, P > 0.05), I_A (control rats: 18.10 ± 2.21, n = 19; diabetic rats: 16.09 ± 0.82, n = 22, P > 0.05), and I_K (control rats: 15.05 ± 0.59, n = 17; diabetic rats: 16.45 ± 0.61, n = 20, P > 0.05) between the control and diabetic rats (t-test). *P < 0.05 compared with the corresponding value in the control group (two-way ANOVA).

Figure 2. Reduction in the current densities of total Kv, DAP-sensitive I_A, and TEA-sensitive I_K in large DRG neurons from diabetic rats

A and B, original traces show different types of Kv currents in large DRG neurons from a control and diabetic rat. Neurons were held at −90 mV and depolarized from −70 to 60 mV in 10-mV increments (inset). C, I-V curves show differences in the current densities of total Kv, I_A, and I_K in large DRG neurons between the control (n = 19 cells) and diabetic (n = 17 cells) groups. D, voltage-dependent activation kinetics of total Kv, I_A, and I_K in large DRG neurons from control and diabetic rats. The V_0.5 values of total Kv currents (control: 1.38 ± 2.21 mV, n = 19; diabetic: 11.98 ± 2.66 mV, n = 17, P < 0.05) and I_K (control: 4.67 ± 2.57 mV, n = 10; diabetic: 17.16 ± 3.71 mV, n = 11, P < 0.05), but not I_A (control: −4.71 ± 3.02 mV, n = 19; diabetic: 3.33 ± 3.97 mV, n = 17, P > 0.05), were significantly different between control and diabetic rats (t-test). There was no significant difference in the k value of total Kv currents (control: 19.66 ± 0.95, n = 19; diabetic: 20.47 ± 0.75, n = 17, P > 0.05), I_A (control: 16.51 ± 1.56, n = 19; diabetic: 13.25 ± 0.84, n = 17, P > 0.05), and I_K (control: 16.93 ± 1.06, n = 10; diabetic: 20.19 ± 2.15, n = 11, P > 0.05) between the two groups (t-test). *P < 0.05 compared with the corresponding value in the control group (two-way ANOVA).

Figure 3. Lack of changes in the total Kv, DAP-sensitive I_A, and TEA-sensitive I_K in small DRG neurons from diabetic rats

A and B, current traces showing different types of Kv currents in small DRG neurons from a control and diabetic rat. Neurons are held at −90 mV and depolarized from −70 to 60 mV in 10-mV increments (inset). C, comparison of the current densities of total Kv currents, I_A, and I_K in small DRG neurons between the control (n = 41 cells) and diabetic (n = 39 cells) group. D, steady-state activation (G-V) curves of total Kv, I_A, and I_K in small DRG neurons from control and diabetic rats. There was no significant difference in the V_0.5 value of total Kv currents (control: 3.80 ± 1.98 mV, n = 36; diabetic: 1.01 ± 2.02 mV, n = 39, P > 0.05), I_A (control: 10.40 ± 4.99 mV, n = 36; diabetic: 8.58 ± 3.81 mV, n = 37, P > 0.05), and I_K (control: 17.29 ± 3.06 mV, n = 23; diabetic: 13.12 ± 2.43, n = 15, P > 0.05) between the control and diabetic rats (t-test). Also, the k values showed no difference in total Kv currents (control: 16.22 ± 0.69, n = 36; diabetic: 17.01 ± 0.71, n = 39, P > 0.05), I_A (control: 15.29 ± 1.55, n = 36; diabetic: 16.85 ± 1.47, n = 37, P > 0.05), and I_K (control: 17.53 ± 2.31 mV, n = 23; diabetic: 17.49 ± 1.42, n = 15, P > 0.05) between the two groups (t-test).

Figure 4. Lack of differences in the total Kv, DAP-sensitive I_A, and TEA-sensitive I_K in IB₄-positive and IB₄-negative small DRG neurons between the control and diabetic group

A and B, I-V curves show the similar amplitudes of the Kv current density in IB₄-positive and IB₄-negative neurons at different potentials in the control (IB₄-positive, n = 13 cells; IB₄-negative, n = 5 cells) and diabetic (IB₄-positive, n = 31 cells; IB₄-negative, n = 5 cells) groups.

Figure 5. Changes in the mRNA levels of individual Kv subunits in the DRG from diabetic rats

A, differences in the mRNA levels of I_A subunits (Kv1.4, Kv3.4, Kv4.2, and Kv4.3) in the DRG between control and diabetic rats. B, lack of differences in the mRNA levels of I_K subunits (Kv1.1, Kv1.2, Kv2.1, and Kv2.2) in the DRG between control and diabetic rats (n = 8 rats, in each group). *P < 0.05 compared with the corresponding value in the control group (t-test).

Figure 6. Effects of BDNF treatment on the total Kv, DAP-sensitive I_A, and TEA-sensitive I_K currents in DRG neurons from control rats

A, I-V curves show that BDNF treatment had no effect on the current densities of total Kv, I_A, and I_K in small DRG neurons (n = 12). B, I-V curves show that BDNF treatment reduced the current densities of total Kv, I_A, and I_K in medium DRG neurons (n = 13). C, I-V curves show that reduced the current densities of total Kv, I_A, and I_K in large DRG neurons (n = 12). D, K252a, but not K252b, abolished the BDNF effects on total Kv, I_A, and I_K in medium DRG neurons (n = 7 in each group). E, K252a, but not K252b, blocked the BDNF effects on total Kv, I_A, and I_K in large DRG neurons (n = 8 in each group). *P < 0.05 compared with the corresponding value in the control or vehicle (K252b) group (two-way ANOVA).

Figure 7. Effects of the anti-BDNF antibody on the current densities of total Kv, DAP-sensitive I_A, and TEA-sensitive I_K currents in DRG neurons from diabetic rats

A, I-V curves show that treatment with the anti-BDNF antibody (50 ng/ml) slightly increased the current densities of total Kv, I_A, and I_K in small DRG neurons (n = 8). B, I-V curves show that treatment with the anti-BDNF antibody profoundly increased the current densities of total Kv, I_A, and I_K in medium DRG neurons (n = 13). C, I-V curves show that treatment with the anti-BDNF antibody substantially increased the current densities of total Kv, I_A, and I_K in large DRG neurons (n = 9). Note that the boiled anti-BDNF antibody had no effects on the Kv current density in small, medium, or large DRG neurons. *P < 0.05 compared with the corresponding value in the control group (two-way ANOVA).

Figure 8. Effects of K252a on the current densities of total Kv, DAP-sensitive I_A, and TEA-sensitive I_K currents in DRG neurons from diabetic rats

A, I-V curves show that treatment with K252a (300 nM) slightly increased the current densities of total Kv, I_A, and I_K in small DRG neurons (n = 11). B, I-V curves show that treatment with K252a substantially increased the current densities of total Kv, I_A, and I_K in medium DRG neurons (n = 10). C, I-V curves show that treatment with K252a profoundly increased the current densities of total Kv, I_A, and I_K in large DRG neurons (n = 9). *P < 0.05 compared with the corresponding value in the diabetic control group (two-way ANOVA).

Figure 9. Effects of K252a and anti-BDNF antibody on the current densities of total Kv, DAP-sensitive I_A, and TEA-sensitive I_k currents in DRG neurons from control rats

A, I-V curves show that treatment with K252a (300 nM, n = 11) and anti-BDNF antibody (50 ng/ml, n = 9) only had a small effect on the current density of I_A in small DRG neurons. B, I-V curves show the effect of treatment with K252a (n = 9) and anti-BDNF antibody (n = 10) on the current densities of total Kv, I_A, and I_k in medium DRG neurons. C, I-V curves show the lack of effect of treatment with K252a (n = 8) and anti-BDNF antibody (n = 8) on the current densities of total Kv, I_A, and I_k in large DRG neurons. *P < 0.05 compared with the corresponding value in the control group (two-way ANOVA).

Figure 10. Reduction in the mRNA levels of I_A subunits by BDNF in DRG neurons

A, effects of BDNF treatment on the mRNA levels of I_A subunits (Kv1.4, Kv3.4, Kv4.2, and Kv4.3) in DRG neurons from control rats (n = 4 samples in each group; t-test). B, effects of treatment with the anti-BDNF antibody (50 ng/ml) or K252a (300 nM) on the mRNA levels of Kv1.4, Kv3.4, Kv4.2, and Kv4.3 subunits in DRG neurons from diabetic rats (n = 5 samples in each group; repeated measures ANOVA). *P < 0.05 compared with the corresponding value in the control group.

Figure 11. Differences in the distribution pattern of BDNF immunoreactive neurons in the DRG between control and diabetic rats

A, representative confocal images show a greater number of BDNF immunoreactive neurons in medium and large neurons from a diabetic rat. Colocalization of BDNF and Nissl (a neuronal marker) is indicated in yellow when the two images are digitally merged. Images are single confocal optical sections. B, the histogram shows the distinct differences in the distribution of BDNF immunoreactive neurons in different sized DRG neurons between control and diabetic rats. *P < 0.05 compared with the corresponding value for the nontreated diabetic group (χ² test).