AAV-Mediated Combination Gene Therapy for Neuropathic Pain: GAD65, GDNF, and IL-10

Abstract

Neuropathic pain is a chronic pain state characterized by nerve damage, inflammation, and nociceptive neuron hyperactivity. As the underlying pathophysiology is complex, a more effective therapy for neuropathic pain would be one that targets multiple elements. Here, we generated recombinant adeno-associated viruses (AAVs) encoding three therapeutic genes, namely, glutamate decarboxylase 65, glial cell-derived neurotrophic factor, and interleukin-10, with various combinations. The efficacy for pain relief was evaluated in a rat spared nerve injury model of neuropathic pain. The maximal analgesic effect was achieved when the AAVs expressing all three genes were administered to rats with neuropathic pain. The combination of two virus constructs expressing the three genes was named KLS-2031 and evaluated as a potential novel therapeutic for neuropathic pain. Single transforaminal epidural injections of KLS-2031 into the intervertebral foramen to target the appropriate dorsal root ganglion produced notable long-term analgesic effects in female and male rats. Furthermore, KLS-2031 mitigated the neuroinflammation, neuronal cell death, and dorsal root ganglion hyperexcitability induced by the spared nerve injury. These results suggest that KLS-2031 represents a promising therapeutic option for refractory neuropathic pain.

Keywords: Adeno-associated virus; Gene therapy; Neuropathic pain; glial cell-derived neurotrophic factor; glutamate decarboxylase 65; interleukin-10.

Graphical abstract

Figure 1

Expression of Genes after TF Injection AAV5-FLAG-GAD65 and AAV5-GDNF-V5/IL-10-HA were administered to the left L4 DRGs of naïve 8-week-old rats via TF injection. Ipsilateral DRGs were harvested 2, 4, and 12 weeks later. (A) FLAG-, V5-, and HA-tagged proteins (green) were visualized via immunostaining; nuclear counterstaining is in blue. Representative images for ipsilateral DRG at each time point are shown. Images were acquired with a 20× lens objective. Scale bar, 200 μm. (B) The numbers of immunopositive cells were counted using NIS-Elements BR software; n = 18 slices from six animals for each time point. Data are presented as mean ± SEM. Statistical differences among groups were assessed using Kruskal-Wallis one-way ANOVA, followed by Dunnett’s post hoc test. See also Figures S4 and S5.

Figure 2

Analgesic Efficacy of Three Therapeutic Genes Male rats received TF injections with the respective AAVs or PBS, 2 weeks after SNI. (A) Mechanical allodynia for rats receiving AAVs containing single genes (AAV5-GAD65, AAV1-GDNF, or AAV5-IL-10; 9 × 10⁸ VG) or a combination of three AAVs (AAV5-GAD65 + AAV1-GDNF + AAV5-IL-10; 3 × 10⁸ VG for each) was measured with vF tests, 4 weeks after administration. AAV5-GFP was used as control; n = 5 animals per group. #versus GFP; ∗versus PBS-treated SNI group. (B) Mechanical allodynia for rats receiving a combination of two AAVs (4.5 × 10⁸ VG for each) or three AAVs (3 × 10⁸ VG for each) was measured with vF tests, 4 weeks after administration. AAV5-GFP was used as control; n = 6 animals per group. #versus AAV5-GFP-treated SNI; ∗versus AAV5-GAD65 + GDNF + IL-10-treated SNI. (C) Mechanical allodynia for rats receiving the various virus ratios (1:9, 1:3, 1:1, 3:1, 9:1; total dose, 1 × 10⁹ VG) of KLS-2031 was measured with the vF test, 4 weeks after administration; n = 6 animals per group. ∗versus PBS-injected SNI group. Data are presented as mean ± SEM. Statistical differences among groups were assessed using Kruskal-Wallis one-way ANOVA, followed by Dunnett’s post hoc test.

Figure 3

Effect of KLS-2031 on Pain Alleviation in Rats with SNI (A, top) Experiment schedule is shown. (A, bottom) 2 weeks after rats underwent SNI surgery, they received TF injections of KLS-2031 at different doses (10⁸, 10⁹, or 10¹⁰ VG) or PBS (5 μL), and vF tests were conducted at the indicated time points; n = 6 animals per group. Statistical differences among groups were assessed for each time point using Kruskal-Wallis one-way ANOVA, followed by Dunnett’s post hoc test. ∗versus PBS-injected SNI group. (B) 2 weeks after rats underwent SNI surgery, they received TF injections of KLS-2031 (10¹⁰ VG) or PBS (5 μL). vF tests were conducted 4 weeks later. The PBS-injected SNI rats were orally administered duloxetine (30 mg/kg) or pregabalin (30 mg/kg), dissolved in 1 mL saline at 2 h and 1 h, respectively, prior to the vF test. The same quantity of saline was orally administered to the PBS- and KLS-2031-injected SNI groups; n = 6 animals per group. ∗versus PBS-injected SNI group. Data are presented as mean ± SEM. Statistical differences among groups were assessed using Kruskal-Wallis one-way ANOVA, followed by Dunnett’s post hoc test.

Figure 4

Equivalent Analgesic Efficacy of KLS-2031 in Male and Female Rats with SNI (Top) Experiment schedule is shown. (Bottom) 2 weeks after rats underwent SNI surgery, they received TF injections of KLS-2031 (1 × 10¹⁰ VG) or saline (5 μL), and vF tests were conducted at the indicated time points; n = 6 animals per group. ∗versus saline-injected SNI groups; †versus naïve. Data are presented as mean ± SEM. Statistical differences among groups were assessed for each time point using Kruskal-Wallis one-way ANOVA, followed by Dunnett’s post hoc test.

Figure 5

Neuroprotective Effect of KLS-2031 in DRG Ipsilateral DRGs (L4) from rats with SNI were harvested 4 weeks after TF injection of KLS-2031 (1 × 10⁹ VG). (A) Immunohistochemistry shows that apoptotic cells (cleaved caspase-3 marker; green [top]) were detected in the cytosol and nuclei (blue), and microglia (Iba1 marker; green [bottom]) formed clusters around neurons. Insets show enlargements of cleaved caspase 3- and Iba1-positive cells, indicated by asterisks and arrowheads, respectively. Images were acquired with a 20× lens objective. Scale bars, 100 μm (10 μm for insets). (B and C) For quantitative analysis, the proportions of caspase 3-positive cells (B) and circularity of microglia cells (C) were calculated; n = 6 animals per group; ∗versus saline-injected SNI group. Data are presented as mean ± SEM. Statistical differences among groups were assessed using Kruskal-Wallis one-way ANOVA, followed by Dunnett’s post hoc test. See also Figure S7.

Figure 6

Effects of KLS-2031 on the Small-Sized DRG Excitability Electrophysiology was performed in L4 DRGs via the whole-cell patch-clamp method. (A) RMPs were recorded in a whole-cell configuration within 10 s after patch break-in; n = 8 cells for naïve, SNI + saline, and SNI + KLS-2031 groups. (B) The rheobase currents of APs were elicited by a series of depolarizing currents from 0 to 700 pA (10 ms) in 50 pA step increments under current-clamp mode; n = 8 cells for naïve, SNI + saline, and SNI + KLS-2031 groups. (C) Input resistances were measured by membrane potential changes at a given hyperpolarizing current input (−200 pA, 200 ms); n = 8 cells for naïve, SNI + saline, and SNI + KLS-2031 groups. (D) Representative traces for current-clamp recordings from small-sized DRG neurons from naïve, SNI + saline, and SNI + KLS-2031 groups in response to prolonged (1 s), depolarizing 50 pA current injections. Dashed lines represent the RMP in naïve, small-sized L4 DRG neurons. (E) AP frequencies in response to 1 s current injections of various amplitudes in naïve rats (n = 8 cells), saline-injected SNI group (n = 8 cells), and KLS-2031-injected SNI group (n = 8 cells). ∗versus saline-injected SNI group; †versus naïve. Data are presented as mean ± SEM. Statistical differences among groups were assessed with Mann-Whitney U test.