Bidirectional regulation of motor circuits using magnetogenetic gene therapy

Abstract

Here, we report a magnetogenetic system, based on a single anti-ferritin nanobody-TRPV1 receptor fusion protein, which regulated neuronal activity when exposed to magnetic fields. Adeno-associated virus (AAV)-mediated delivery of a floxed nanobody-TRPV1 into the striatum of adenosine-2a receptor-Cre drivers resulted in motor freezing when placed in a magnetic resonance imaging machine or adjacent to a transcranial magnetic stimulation device. Functional imaging and fiber photometry confirmed activation in response to magnetic fields. Expression of the same construct in the striatum of wild-type mice along with a second injection of an AAVretro expressing Cre into the globus pallidus led to similar circuit specificity and motor responses. Last, a mutation was generated to gate chloride and inhibit neuronal activity. Expression of this variant in the subthalamic nucleus in PitX2-Cre parkinsonian mice resulted in reduced c-fos expression and motor rotational behavior. These data demonstrate that magnetogenetic constructs can bidirectionally regulate activity of specific neuronal circuits noninvasively in vivo using clinically available devices.

Fig. 1.. Generation and validation of ferritin-binding nanobodies for magnet stimulation.

Serial dilutions of nanobodies or BSA (100 μl) were incubated on plates coated with human spleen ferritin (1 μg/ml) and binding quantified by ELISA after incubation with anti-HA-HRP antibody. (A) Quantification of anti-ferritin nanobodies from CDR3 groups 6 and 26 binding to human spleen ferritin. (B) Oscillating magnetic field treatment (465 kHz, 30mT) significantly increases calcium-dependent SEAP release from 293T cells transfected with Nb-GFP-TRPV1^Ca2+, Nb-Ft-2-TRPV1^Ca2+, and Nb-Ft-14-TRPV1^Ca2+ and GFP-mFerritin. Data analyzed by two-tailed, unpaired t test with Welch’s correction. Nb-GFP-TRPV1^Ca2+, basal versus RF *P = 0.04, n = 8 and 8; Nb-Ft-2-TRPV1^Ca2+, basal versus RF *P = 0.02, n = 7 and 8; Nb-Ft-14-TRPV1^Ca2+, basal versus RF *P = 0.04, n = 7 and 8. (C) Normalized Fluo-4 fluorescence (ΔF/F₀) in Neuro2A cells expressing Nb-Ft-2-TRPV1^Ca2+ with (786 cells) or without (1659 cells) magnet treatment. Data were analyzed by two-way ANOVA with Tukey’s multiple comparison test, ****P < 0.0001. (D) Cumulative change in ΔF/F₀ of Neuro2A cells expressing Nb-Ft-2-TRPV1^Ca2+ with (786 cells) or without (1659 cells) magnet treatment. Data were analyzed by Mann Whitney U test, ****P < 0.0001. (E) Changes in RCaMP fluorescence normalized to baseline fluorescence (ΔF/F₀) with magnet treatment of HEK-293T cells expressing RCaMP alone (54 cells), TRPV1^Ca2+ (107 cells), or Nb-Ft-2-TRPV1^Ca2+ (101 cells). Error bars represent means ± SEM. (F) Cumulative change in ΔF/F₀ with magnet treatment of HEK-293T cells expressing RCaMP alone (54 cells), TRPV1^Ca2+ (107 cells), or Nb-Ft-2-TRPV1^Ca2+ (101 cells). AUC was calculated for the period of magnet exposure between 48 and 180 s (after focus adjustment) and analyzed by ordinary one-way ANOVA with Tukey’s multiple comparison test, ****P < 0.0001. Data are shown as means ± SD.

Fig. 2.. Selective viral-mediated expression of Nb-Ft-TRVP1^Ca2+ in striatal iSPNs elicits parkinsonian motor behavior.

(A) Schema of the double-floxed Cre-dependent AAV vector expressing the Nb-Ft-TRVP1^Ca2+ or mCherry under the control of the JET promoter; immunostaining for HA in A2a-Cre mice demonstrates selective expression of the AAV- Nb-Ft-TRVP1^Ca2+ in the dorsal striatum. (B and C) Selective transduction of D2-type iSPN with RNA in situ hybridization. Colocalization of D2-type (green) iSPN with mCherry (red). (D) Protocol and schema for DMF neural activation to assess motor behavior. (E) Example of altered motor activity during bilateral striatal pre-DMF (green), DMF (red), and post-DMF (blue) stimulation; individual lines represent the path of a single mouse. Effect of DMF stimulation on (F) ambulation bout duration and (G) freezing of gait in mCherry (blue bars/dots, n = 10) and Nb-Ft-TRVP1^Ca2+ (red bars/dots, n = 10) mice. Error bars show SEM. **P < 0.01, ****P < 0.0001 with two-tailed, unpaired t test with Welch’s correction.

Fig. 3.. Nb-Ft-TRPV1^Ca2+ in D2 iSPNs stimulated with high DMF and low DMF treatment produce freezing of gait.

(A) Schema for low DMF titration. (B) Magnetic field strength and individual activity heatmap example of motor activity during bilateral striatal low DMF titration in Nb-Ft-TRPV1^Ca2+ expressing mice. Freezing of gait in different magnetic field gradients in the low DMF titration ranges (blue bars, n = 5) in Nb-Ft-TRPV1^Ca2+ expressing mice. (C) Protocol and schema for high DMF and low DMF stimulation in A2a-Cre mice injected with the double-floxed Cre-dependent AAV vector expressing the Nb-Ft-TRPV1^Ca2+ or mCherry under the control of the JET promoter. Representative tracking data of change of location in (D) Nb-Ft-TRPV1^Ca2+ and (E) mCherry A2a-Cre mice during high DMF (red line), low DMF (blue line), and no DMF (green line) treatment. Effect of high DMF and low DMF stimulation on (F) ambulation and (G) freezing of gait in Nb-Ft-TRPV1^Ca2+ (red dot/bars, n = 8) and mCherry (blue dots/bars, n = 5). Error bars show SEM. *P < 0.05, **P < 0.01, ****P < 0.0001 with two-tailed, unpaired t test with Welch’s correction.

Fig. 4.. Nb-Ft-TRPV1^Ca2+ increases c-fos expression and cerebral metabolism upon DMF treatment.

(A) Immunostaining for c-fos (cyan) and HA or RFP (red) of the dorsal striatum of A2a-Cre mice euthanized 1 hour following 15 min of DMF exposure. (B) Percentage (%) of c-fos–positive neurons over total number of iSPNs expressing HA or RFP (n = 5 mice per group/3 sections per mice). (C) Correlation between c-fos expression and freezing of gait during DMF exposure (n = 5 mice per group). (D) Protocol and schema for ¹⁸F-FDG micro-PET/CT scan procedure and analysis with representative micro-PET/CT imaging of A2a-Cre mouse brains expressing the Nb-Ft-TRPV1^Ca2+ or the mCherry control virus following DMF exposure. ¹⁸F-FDG uptake in the dorsal striatum calculated as the percentage of injected dose/weight (%ID/g). (E) DMF over baseline ratio. (F) DMF over baseline normalized to the cerebellum, in mCherry (blue bars/dots, n = 8) and Nb-Ft-TRPV1^Ca2+ (red bars/dots, n = 8) mice. (G) Correlation between DMF over baseline ¹⁸F-FDG uptake ratio and freezing of gait during DMF exposure (n = 10). Error bars show SEM. *P < 0.05, **P < 0.01, ***P < 0.001 with two-tailed, unpaired t test with Welch’s correction. Anti-HA staining to detect Nb-Ft-TRVP1^Ca2+ expression, anti-RFP staining to detect mCherry expression.

Fig. 5.. Nb-Ft-TRPV1^Ca2+ increases calcium transients in the dorsal striatum.

(A and B) Schema and AAV expression of the double-floxed Cre-dependent AAV vector encoding the Nb-Ft-TRPV1^Ca2+ or mCherry under the control of the JET promoter (red) combined with double-floxed Cre-dependent AAV vector expressing GCaMP6s under the control of the Syn promoter (green). (C) Behavioral methodology to assess calcium transients during pre-DMF (green), DMF (red), and post-DMF (blue) treatment. (F) Activity trace example of altered motor activity during bilateral striatal pre-DMF (green), DMF (red), and post-DMF (blue) stimulation in an Nb-Ft-TRPV1^Ca2+ mouse with heatmap of GCaMP fluorescence aligned to pre-DMF (left), DMF (middle), and post-DMF (right) of the same animal shown in activity trace example expressing the Nb-Ft-TRPV1^Ca2+ vector. (G) Mean fluorescence aligned to pre-DMF, DMF, and post-DMF in Nb-Ft-TRPV1^Ca2+ group (n = 3). (D) Activity trace example of altered motor activity during bilateral striatal pre-DMF (green), DMF (red), and post-DMF (blue) stimulation in an mCherry mouse with heatmap of GCaMP fluorescence aligned to pre-DMF (left), DMF (middle), and post-DMF (right) for the same animal shown in activity trace example expressing the mCherry vector. (E) Mean fluorescence aligned to pre-DMF, DMF, and post-DMF in the mCherry group (n = 4). (H) Freezing of gait (FOG). (I) Average ∆F/F (%) during pre-DMF, DMF, and post-DMF stimulation. (J) Correlation between ∆F/F (%) calcium transients and FOG during DMF treatment. Error bars show SEM. *P < 0.05 with two-tailed, unpaired t test with Welch’s correction.

Fig. 6.. TMS treatment increases Nb-Ft-TRPV1^Ca2+–mediated motor freezing and calcium transients in the dorsal striatum.

(A) Schema of the TMS setup and A2a-Cre mice with bilateral injection of the double-floxed Cre-dependent AAV vector expressing the Nb-Ft-TRPV1^Ca2+ or mCherry under the control of the JET promoter (red) combined with double-floxed Cre-dependent AAV vector expressing GCaMP6s under the control of the Syn promoter (green) in the dorsal striatum. (B) Activity trace example of altered motor activity during baseline (left) and TMS (right) treatment in an mCherry mouse (blue) and Nb-Ft-TRPV1^Ca2+ mouse (red). Distance ambulated during baseline (left) and TMS (right) treatment in the (C) mCherry group (blue) and (D) Nb-Ft-TRPV1^Ca2+ group (red) (n = 4 to 5; data presented as the means ± SEM in gray). (E) Average distance during baseline and TMS treatment in mCherry (blue) and Nb-Ft-TRPV1^Ca2+ groups. (F) Mean fluorescence during baseline in Nb-Ft-TRPV1^Ca2+ (red) and mCherry (blue) mice (n = 3; data presented as the means ± SEM). (G) Mean fluorescence during 20 s of TMS triggers every 2 s in Nb-Ft-TRPV1^Ca2+ (red) and mCherry (blue) mice (n = 3 to 4; data presented as the means ± SEM). (H) Grouped bar graph of Nb-Ft-TRPV1^Ca2+ (red) and mCherry (blue) mice during baseline and 20 s of TMS triggers every 2 s (n = 3 to 4). Error bars show SEM. *P < 0.05, **P < 0.01 with two-tailed, unpaired (or paired when baseline versus TMS comparison were done) t test with Welch’s correction.

Fig. 7.. Selective viral-mediated expression of Nb-Ft-TRPV1^Ca2+ in the striatopallidal pathway elicits parkinsonian motor behavior in WT mice.

(A) Schema of the double-floxed Cre-dependent AAV vector expressing the Nb-Ft-TRPV1^Ca2+ or mCherry under the control of the JET promoter in striatal iSPN projecting to the Gp achieved by Gp injection of a retrograde AAV2 vector carrying Cre-GFP; immunostaining for GFP (green) and HA (red) in WT mice demonstrates selective expression of the AAV- Nb-Ft-TRPV1^Ca2+ in the dorsal striatum iSPN projecting to the GPe; white arrows show colocalization of GFP with HA. (B) Example of altered motor activity during bilateral striatopallidal pre-DMF (green), DMF (red), and post-DMF (blue) stimulation; lines represent the mouse’s path. Effect of DMF stimulation on (C) ambulation bout duration and (D) freezing of gait in mCherry (blue bars/dots, n = 8) and Nb-Ft-TRPV1^Ca2+ (red bars/dots, n = 8) mice. (E) Immunostaining for c-fos (cyan) and HA (Nb-Ft-TRVP1^Ca2+; red) or RFP (mCherry; red) of the dorsal striatum of WT mice euthanized 1 hour following 15 min of DMF exposure. White arrows show colocalization. (F) Percentage (%) of c-fos–positive neurons over the total number of iSPNs expressing HA or RFP (n = 3 mice per group/3 sections per mice). (G) Schema of the double-floxed Cre-dependent AAV vector expressing the Nb-Ft-TRPV1^Ca2+ or TRPV1^Ca2+ under the control of the JET promoter in striatal iSPNs projecting to the Gp achieved by Gp injection of a retrograde AAV2 vector carrying Cre-GFP. Effect of DMF stimulation on (H) ambulation bout duration and (I) freezing of gait in TRPV1^Ca2+ (blue bars/dots, n = 5) and Nb-Ft-TRPV1^Ca2+ (red bars/dots, n = 7) mice. Error bars show SEM. **P < 0.01, ***P < 0.001 with two-tailed, unpaired t test with Welch’s correction.

Fig. 8.. Mutant Nb-Ft-TRPV1^Cl− inhibits neuronal activity of STN projection neurons and rescues motor impairment in parkinsonian mice.

(A) Schema of inhibition system with mutant Nb-Ft-TRPV1^Cl− membrane channel. (B) Normalized change in MQAE fluorescence intensity (ΔF/F₀) of Neuro2A cells expressing mCherry, TRPV1^Ca2+, or Nb-Ft-TRPV1^Cl− with low AMF application in vitro. (C) AUC analysis of chloride imaging for the Nb-Ft-TRPV1^Cl− channel. (D) Protocol and schema for generation of parkinsonian PitX2-Cre mice with unilateral injection of 6-OHDA in the MFB, followed by randomization-based contralateral rotations induced by intraperitoneal administration of apomorphine (0.25 mg/kg) and ipsilateral intracranial injection of the double-floxed Cre-dependent AAV vector expressing the mutant Nb-Ft-TRPV1^Cl− under the control of the JET promoter in the STN. Representative immunostaining of the selective expression of Nb-Ft-TRPV1^Cl− (red) in STN neurons expressing Cre recombinase (green). (E) Percentage of change from pre-DMF treatment in contralateral rotations induced with apomorphine (0.25 mg/kg, ip) in parkinsonian PitX2-Cre mice expressing the Nb-Ft-TRPV1^Cl− or the mCherry control viruses in the STN during DMF and post-DMF. (F) Immunostaining for c-fos (cyan) and HA or mCherry (red) of STN of PiX-2-Cre mice euthanized 1 hour following 15 min of DMF exposure with apomorphine-induced rotations (0.25 mg/kg). (G) Number of transduced neurons per area in STN between mCherry and Nb-Ft-TRPV1^Ca2+ groups. (H) Percentage (%) of c-fos–positive neurons over the total number of STN neurons expressing HA or RFP (n = 4 mice per group/3 sections per mice). Error bars show SEM. *P < 0.05, **P < 0.01 with two-tailed, unpaired t test with Welch’s correction. Data were analyzed by Mann Whitney U test, ****P < 0.001. Anti-HA staining to detect Nb-Ft-TRVP1^Ca2+ expression, anti-RFP staining to detect mCherry expression.