AAV transduction of dopamine neurons with constitutively active Rheb protects from neurodegeneration and mediates axon regrowth

Abstract

There are currently no therapies that provide either protection or restoration of neuronal function for adult-onset neurodegenerative diseases such as Parkinson's disease (PD). Many clinical efforts to provide such benefits by infusion of neurotrophic factors have failed, in spite of robust effects in preclinical assessments. One important reason for these failures is the difficulty, due to diffusion limits, of providing these protein molecules in sufficient amounts to the intended cellular targets in the central nervous system. This challenge suggests an alternative approach, that of viral vector transduction to directly activate the intracellular signaling pathways that mediate neurotrophic effects. To this end we have investigated the ability of a constitutively active form of the GTPase Rheb, an important activator of mammalian target of rapamycin (mTor) signaling, to mediate neurotrophic effects in dopamine neurons of the substantia nigra (SN), a population of neurons affected in PD. We find that constitutively active hRheb(S16H) induces many neurotrophic effects in mice, including abilities to both preserve and restore the nigrostriatal dopaminergic axonal projections in a highly destructive neurotoxin model. We conclude that direct viral vector transduction of vulnerable neuronal populations to activate intracellular neurotrophic signaling pathways offers promise for the treatment of neurodegenerative disease.

Figure 1

Transduction of SNpc neurons by adeno-associated virus (AAV)-Rheb vectors in normal adult male C57Bl/6 mice. The upper panels show immunoperoxidase staining for FLAG performed at 4 weeks after intranigral injection of each vector. For each vector, FLAG expression is observed as brown reaction product in the SNpc (arrows). No staining above background is observed in the SNpc on the control, noninjected side. FLAG immunostaining within the regions outlined by the squares is shown at higher magnification in the adjacent panels to the right. In the lower panels, immunofluorescence double-labeling for tyrosine hydroxylase (TH) (red) and FLAG (green) demonstrates that transgene expression is identifiable within dopamine neurons of the SNpc for each vector. Each vector was estimated to achieve efficiencies of transduction of dopamine neurons ranging from 80% in the caudal planes adjacent to the vector injection site to 60% in the rostral planes.

Figure 2

Phosphorylation of the mTor substrate 4E-BP1 by hRheb(S16H). (a) Western analysis of p-4E-BP1 expression in the ventral mesencephalon at 4 weeks after intranigral injection of adeno-associated virus (AAV) vectors. The upper panel shows a representative blot, and the lower panel shows a quantitative analysis based on the density of the p-4E-BP1 bands normalized for the β-actin band for each sample. hRheb(S16H) induced a 2.6-fold increase in the p-4E-BP1/β-actin ratio in comparison to the mean ratio for the four contralateral controls (P < 0.001, one-way ANOVA; P = 0.004 Tukey post-hoc analysis; n = 4 animals, each group). Neither hRheb(WT) nor hRheb(N153T) had significant effects. Successful transduction of the substantia nigra (SN) was confirmed in each case by western analysis of FLAG expression as shown (C, control side; E, experimental side). (b) Immunoperoxidase staining for p-4E-BP1 (with thionin counterstain) in the SN at 12 weeks following transduction with AAV hRheb(S16H). Brown reaction product is observed in neurons (arrows) on the injected side, but not on the contralateral control side or on either side following AAV green fluorescent protein (GFP) injection (Bar = 100 µm). An example of neuronal staining is shown in the right panel.

Figure 3

Effects of wild-type (WT) and constitutively active forms of hRheb on substantia nigra (SN) neurons in normal adult male C57Bl/6 mice. (a) Morphologic analysis of SN dopamine neurons at 4 weeks after intranigral injection of adeno-associated virus (AAV) hRheb(S16H). The upper panel shows a representative coronal section of the SN following tyrosine hydroxylase (TH) immunoperoxidase stain and thionin counterstain. The experimental (EXP) side injected with AAV hRheb(S16H) shows an increased density of staining. This increase is associated with an increase in the mean area of the TH-positive neurons, as shown in representative micrographs at higher power in the lower panels. This effect is shown quantitatively for mice receiving AAV Rheb(S16H) as a 35% increase in mean neuron size on the injected side in comparison to mean size on the contralateral control noninjected side (P < 0.001, one-way ANOVA and Tukey post-hoc analysis as shown; n = 80 neurons from four mice, each experimental group). AAV Rheb(N153T) also increased neuron size, by 19% (Tukey post-hoc analysis as shown; n = 80 neurons from four mice, each experimental group). (b) Morphologic analysis of SN neurons demonstrated by immunoperoxidase staining for the neuron marker NeuN. The EXP side shows an increased density of staining in the SNpc. As for the TH staining, this increase is associated with an increase in the mean cross-sectional area of the NeuN-positive neurons, as shown in representative micrographs at higher power in the lower panels. This effect on neuron size is general to other neuron phenotypes as shown quantitatively in the graph, which reveals a 42% increase in the size of NeuN-positive neurons in the SNpr (P < 0.001, one-way ANOVA and Tukey's post-hoc analysis as shown).

Figure 4

Relative effect of wild-type (WT) and constitutively active forms of hRheb on measures of striatal dopaminergic innervation in normal adult male C57Bl/6 mice. (a) Morphologic analysis of striatum at 4 weeks after intranigral injection of adeno-associated virus (AAV) hRheb(S16H). In the upper panel a representative coronal section of tyrosine hydroxylase (TH) immunoperoxidase staining reveals an increased density of staining on the experimental (EXP) side injected with AAV hRheb(S16H). As shown in the micrographs in lower panels, this increase is due in part to an increased number of TH-positive axons. The increase in optical density for TH peroxidase stain is shown quantitatively in panel (b). AAV hRheb(S16H) induced a 1.4-fold increase in optical density (OD) of TH peroxidase stain, expressed as percent of the contralateral, noninjected side, in comparison to AAV green fluorescent protein (GFP)-injected mice (P = 0.001, one-way ANOVA and Tukey post-hoc analysis as shown; n = 4 animals, each experimental group). AAV hRheb(S16H) also induced a 1.2-fold increase in optical density (OD) of peroxidase stain of the dopamine transporter (DAT) (P < 0.001, one-way ANOVA and Tukey post-hoc analysis as shown; n = 4 animals, each experimental group). (c) These morphologic effects of AAV hRheb(S16H) were accompanied by increases in biochemical measures of striatal dopaminergic innervation. In comparison to AAV GFP control injection, AAV hRheb(S16H) induced significant increases in striatal dopamine (DA) and its metabolite homovanillic acid (HVA), expressed as percent of the contralateral, noninjected control striatum (P = 0.003 and P < 0.001, respectively; t-test; n = 7 and 8, respectively). (d) To determine whether these morphologic and biochemical measures of increased striatal dopaminergic innervation were functionally significant, we administered amphetamine [2.5 mg/kg intraperitoneal (i.p.)] to mice at 4 weeks following intranigral injection of either AAV GFP or AAV hRheb(S16H). Amphetamine induces dopamine release from striatal terminals, and in the presence of a unilateral increase in functional dopaminergic innervation, it will induce a rotational behavior contralateral to the side with the increase (upper panel). Following AAV GFP injection, no sustained rotational behavior was observed (lower panel). However, after injection of AAV hRheb(S16H), a highly significant contralateral rotational behavior (depicted as negative on the ordinate) was observed following amphetamine injection (P = 0.009, t-test, n = 8 animals, each experimental group).

Figure 5

hRheb(S16H) protects substantia nigra (SN) dopamine neurons from 6-hydroxydopamine (6-OHDA)-induced neuron death. (a) Mice received intranigral injection of adeno-associated virus (AAV) green fluorescent protein (GFP) as control or AAV hRheb(S16H) and 3 weeks later they received an intrastriatal injection of 6-OHDA. The nigrostriatal dopaminergic projection was assessed at 4 weeks following 6-OHDA. (b) Representative coronal sections of the SN following tyrosine hydroxylase (TH) immunoperoxidase stain and thionin counterstain from mice that received either AAV GFP (upper) or AAV hRheb(S16H) (lower). In the mice that received AAV hRheb(S16H), the population of TH-positive dopamine neurons in the SNpc is relatively preserved. The neuroprotective effect of hRheb(S16H) is shown quantitatively; while only 43% of dopamine neurons survived following AAV GFP, 90% survived following AAV hRheb(S16H), due to a 2.2-fold increase in the absolute number of surviving neurons (P < 0.001, one-way ANOVA and Tukey post-hoc analysis as shown; n = 7 animals, each experimental group). The number of surviving neurons in the AAV hRheb(S16H) condition was not significantly different from the contralateral, noninjected control side (P = 0.65, NS). (c) Morphologic preservation of dopamine neurons in the SN was accompanied by relative preservation of SN dopamine and its metabolite homovanillic acid (HVA). For dopamine, one way ANOVA revealed a significant difference for the comparison between AAV GFP experimental (EXP) (6-OHDA-injected side) and the AAV hRheb(S16H) noninjected control (CON) (P = 0.03), but not for the AAV hRheb(S16H) experimental (EXP) comparison with the noninjected control (CON) (P = 0.9, NS). A similar result was obtained for HVA.

Figure 6

hRheb(S16H) preserves striatal dopaminergic innervation following intrastriatal 6-hydroxydopamine (6-OHDA). (a) Representative tyrosine hydroxylase (TH) immunoperoxidase stained coronal sections of the striatum from mice that received either adeno-associated virus (AAV) green fluorescent protein (GFP) (upper) or AAV hRheb(S16H) (lower) before receiving intrastriatal 6-OHDA. In the mice that received AAV hRheb(S16H), the TH-positive dopaminergic innervation of the striatum is substantially preserved. This neuroprotective effect of hRheb(S16H) is shown quantitatively as the optical density of immunoperoxidase stain measured over the entire striatum (outlined in blue). In the mice that received AAV GFP, optical density values were reduced to 32% of the control values, whereas in the mice treated with AAV hRheb(S16H), they were reduced only to 85% (P = 0.002, one-way ANOVA and Tukey post-hoc analysis as shown; n = 7 animals, each experimental group). (b) This morphologic preservation of dopamine innervation of the striatum was accompanied by relative preservation of striatal dopamine. In AAV hRheb(S16H)-treated mice (n = 8), mean striatal dopamine content was 20-fold greater than the mean value in AAV GFP-treated mice (n = 7) (P < 0.02, one-way ANOVA and Tukey post-hoc analysis as shown). Following a partial nigrostriatal lesion, there is a compensatory increase in the turnover of dopamine, indicated by an increase in the homovanillic acid (HVA)/dopamine ratio. In the AAV hRheb(S16H)-treated mice, there was a significant decrease in this ratio, signifying a partial restoration of dopaminergic innervation (P < 0.005, one-way ANOVA and Tukey post-hoc analysis as shown). (c) Following unilateral 6-OHDA lesion, administration of amphetamine induces ipsiversive rotation, due to the predominance of dopamine release on the intact side. Mice treated with AAV hRheb(S16H), unlike mice treated with AAV GFP, did not demonstrate ipsiversive rotational behavior, indicating relative preservation of nigrostriatal dopaminergic function following 6-OHDA lesion (P = 0.03, t-test,; n = 7 animals, each experimental group).

Figure 7

Restoration of the dopaminergic nigrostriatal projection by AAV hRheb(S16H) at 6 weeks following 6-hydroxydopamine (6-OHDA) lesion. (a) To assess the ability of hRheb(S16H) to restore the dopaminergic projection after injury, mice were first lesioned unilaterally by intrastriatal 6-OHDA injection, and then at 6 weeks postlesion, by which time degenerative processes are complete, the substantia nigra (SN) was transduced with AAV hRheb(S16H) or green fluorescent protein (GFP) control. Twelve weeks after adeno-associated virus (AAV) (18 weeks postlesion) mice were sacrificed for morphological assessment. (b) hRheb(S16H) transduction at 6 weeks postlesion does not affect the number of surviving SN TH-positive neurons, as shown by representative coronal sections of the SN (FR: fasciculus retroflexus) or by stereology counts as shown in the graph. In mice treated with AAV GFP control injection, the number of TH-positive neurons was reduced to 2237 ± 399 by 6-OHDA, a value 31% of the nonlesioned contralateral SN. In mice treated with AAV hRheb(S16H), TH-positive neurons were reduced to 1,925 ± 451, or 28%, of the nonlesioned contralateral control. (c) In mice treated with hRheb(S16H), there was a significant reinervation of the striatum, as shown by peroxidase staining for TH. In mice treated with AAV hRheb(S16H) (N = 4), striatal peroxidase staining on the lesioned side, measured as optical density over the entire striatum (outlined in blue), was 44.6 ± 0.4% of the contralateral control. In AAV GFP-injected mice (N = 4), it was 24.6 ± 3.0%, and in non-AAV-injected mice (N = 4), it was 26.6 ± 8.2%, both significantly different from AAV hRheb(S16H) [P = 0.04, ANOVA; Student–Newman–Keuls, as shown in (d)]. In addition, mice treated with AAV hRheb(S16H) showed an increased number of TH-positive axons in the medial forebrain bundle (MFB). These mice (N = 3) had a mean of 140 ± 5.6 axons on the lesioned side (71% of the contralateral control), whereas AAV GFP (N = 3) and non-AAV-injected mice (N = 4) had only 91.5 ± 12.9 (49%) and 84.9 ± 9.6 (43%) axons, respectively, a highly significant difference [P < 0.001, ANOVA; Tukey post-hoc comparisons as shown in (e)].

Figure 8

Schematic representation of Akt/Rheb/mTor signaling pathways. Following activation at the plasma membrane by phosphorylation by PDK1 and mTORC2, Akt phosphorylates and thereby inhibits the GTPase activity of the tuberous sclerosis complex (TSC). This inhibition allows accumulation of activated GTP-bound Rheb, which is a principal activator of the mTORC1 kinase. Two principal downstream substrates of mTORC1 are 4E-BP1 and p70S6K. Their phosphorylation mediates effects of mTORC1 activation (see reviews in refs. 41,42,44). Constitutively active hRheb(S16H) is resistant to GTPase activation by TSC, and it therefore maintains enhanced activation of mTORC1.