One of the greatest barriers to treating neural degeneration and neural tissue injury is the fact that neurons do not proliferate, and there are only a few areas of the brain that exhibit neurogenesis. It is possible to convert one somatic cell type to another with the best-known example being the reprogramming of fibroblasts to induced pluripotent stem (iPS) cells by “Yamanaka factors” (1). More recently, direct transdifferentiation of cells to neurons has been proposed for treating a number of neurodegenerative disorders (2). Specifically targeting certain somatic cells circumvents immune recognition and bypasses the need to revert back to an embryonic state thus avoiding the loss of essential age-related and epigenetic factors. Glial cells represent a model target since they are plentiful, proliferate in response to injury and most importantly, are highly plastic. However, while the conversion of glial cells to functional neurons represents an attractive therapeutic approach for a number of neurodegenerative disorders, there has so far been limited success (3).
Two recent papers from Qian et al (4) and Zhou et al (5) demonstrate that reduced expression of the RNA binding protein PTB (encoded by Ptbp1-polypyrimidine tract-binding protein) allows the direct conversion of glial cells to neurons. PTB and it’s neuronal analogue nPTB are both down regulated during neurogenesis (6) and in vitro knockdown of Ptbp1 is sufficient for conversion of mouse fibroblasts to neurons (7). Through regulation of a microRNA circuit comprising both positive and negative feedback loops, PTB inhibits the transcriptional repressor REST that functions to suppress neuronal gene targets.
In the work by Qian et al, conversion of mouse astrocytes to neurons was demonstrated by lentiviral-mediated shRNA knockdown (KD) of Ptbp1 (shPTB) resulting in neuronal morphology and expression of neuronal markers TUJ1 and MAP2 concomitant with suppression of astrocyte-specific genes. The response of shPTB converted neurons to synaptic input was also confirmed by patch clamp recordings. To demonstrate in vivo reprogramming of astrocytes, AAV expressing shPTB (together with a loxP-Stop-LoxP(LSL)-RFP) were injected into the substantia nigra of wild type and transgeneic mice expressing Cre driven by the astrocyte-specific marker GFAB. Over time, the percentage of RFP positive cells expressing mature neuronal markers increased in AAV-shPTB injected mice thus demonstrating astrocyte-to-neuron conversion. Subsequent experiments demonstrated shPTB-targeted astrocytes are able to mature into dopaminergic (DA) neurons offering a potential resource to restore dopamine levels in Parkinson’s disease. Using an established model in which DA neurons are ablated by 6-hydroxydpamine (6-OHDA), depletion of striatal dopamine leading to loss of neurons in the substantia nigra was examined following injection with empty AAV or AAV-shPTB. In contrast to control treated mice, treatment of mice with AAV-shPTB was able to restore the number of neuronal cell bodies together with a robust restoration of striatal dopamine and activity-induced dopamine release. Although the 6-OHDA mouse model does not fully recapitulate the phenotype of Parkinson’s, mice do exhibit motor phenotypes such as contralateral forelimb dysfunction. In young mice at least, treatment with AAV-shPTB was able to correct three motor phenotypes. Finally, to address clinical intervention, several antisense oligonucleotides (ASOs) were screened for their ability to reduce PTB in mouse astrocytes and the top performing ASOs were then assessed for their ability to restore motor function in the 6-OHDA model. ASOs were able to convert astrocytes to functional neurons and importantly were able to rescue impaired motor function.
In the paper by Zhou et al, PTB depletion was achieved by targeting with CRISPR-Cas13d (CasRx) which specifically targets RNA (8). Following screening of several gRNAs to edit and reduce Ptbp1 mRNA levels, the ability of PTB to convert Müller glial (MG) cells to neurons (Retinal ganglion cells – RGCs) was examined. An AAV expressing dual gRNAs targeting Ptbp1 and CasRx driven by the astrocyte-specific promoter GFAP (AAV-GFAP-CasRx-Ptbp1) was injected into the eyes of Rosa-CAG-LSL-tdTomato mice together with AAV-GFAP-Cre-GFP to identify MG cells. Conversion of MG cells to different subtypes of RGCs was confirmed by co-staining of tdTomato positive cells with several RGC markers such as Brn3a and Rbpms. No positive staining was observed in retinas of mice injected with control virus (AAV-GFAP-CasRx-control gRNAs).
To determine whether PTB depleted MG cells can replenish RGCs following injury, RGCs were depleted by injection with N-methyl-D-aspartate (NMDA). When AAV-GFAP-CasRx-Ptbp1 was injected 2-3 weeks later, the number of MG-converted RGCs was significantly increased as was the response to light stimulation. RGC projections to the dorsal lateral geniculate nucleus (dLGN) and superior colliculus (SC) in the brain are responsible for relaying visual information. In NMDA/AAV-GFAP-CasRx-Ptbp1 treated mice, tdTomato positve axons were observed in the dLGN and SC confirming the presence of newly formed axons. These MG-to-RGC derived projections partially restored visual responses as determined by visually evoked potentials (VEPs).
As with Qian et al, Zhou et al next examined whether PTB depleted astrocytes are capable of replenishing DA neurons in the substantia nigra. AAV-GFAP-CasRx-Ptbp1 together with AAV-GFAP-mCherry were injected into the striatum of wild type mice. High levels of CasRx expression was observed with PTB reduction confirmed by immunostaining. Co-staining of mCherry positive cells with neuronal markers revealed conversion of astrocytes to neuronal cells. The 6-OHDA mouse model of Parkinson’s was employed with AAV-GFAP-CasRx-Ptbp1 injected 3 weeks after 6-OHDA infusion. In contrast to control treated mice, a high percentage of cells from AAV-GFAP-CasRx-Ptbp1 injected mice expressed mCherry and dopamine neuron markers such as tyrosine hydroxylase (TH) and Slc6a3 suggesting derivation from astrocytes. Converted DA neurons appeared to be able to release dopamine since mCherry positive cells stained positive for vesicular monoamine transporter 2 (VMAT2), an essential protein required for dopamine release. Lastly, to elucidate whether AAV-GFAP-CasRx-Ptbp1 transduced astrocytes could alleviate motor function symptoms in the 6-OHDA PD mouse model, contralateral rotation and rotarod tests were performed. In both cases mice treated with AAV-GFAP-CasRx-Ptbp1 performed better.
Since it was demonstrated that MyoD can convert fibroblasts to myoblasts (9), several other transcription factors have been shown to induce transdifferentiation. Additionally, powerful tools such as Mogrify (10) exist to predict transcription factor “cocktails” for reprogramming a plethora of human cells. While additional studies are needed to address longevity of converted neurons, age-related restrictions, specificity and adequate targeting efficiency, the above work highlights a real potential for direct in vivo glial-to-neuronal fate switching by suppression of a single RNA binding protein thus providing a new approach for therapeutic application for Parkinson’s disease and other neurological disorders.
VectorBuilder offers second-generation recombinant AAV packaging for multiple serotypes (1, 2, 3, 4, 5, 6, 6.2, 7, 8, 9, rh10, DJ, DJ/8, PHP.eB, PHP.S, AAV2-retro, AAV2-QuadYF and AAV2.7m8). We offer shRNA, single and dual gRNA vectors and can package both ssAAV (single-stranded AAV) and scAAV (self-complementary AAV). Contact us to find out more, or for help designing your custom vectors. Use our award-winning online platform to design and order custom vectors specific to your research needs.
- Takahashi, K. & Yamanaka, S. Cell. 2006. 126: 663–676.
- Mollinari, C et al. Cell Death Dis. 2018. 9: 830.
- Gascón, S et al. Cell Stem Cell. 2017. 21: 18–34.
- Qian, H et al. Nature. 2020. 582: 550–556.
- Zhou, H et al. Cell. 2020. 3: 590–603.
- Hu, J et al. Biophys Rep. 2018. 4: 204–214.
- Xue, Y et al. Cell. 2013. 152: 82–96.
- Konermann, S et al. Cell. 2018. 173: 665–676.
- Davis, R, el al. Cell. 1987. 51: 987–1000.
- Rackman, O et al. Nature Genetics. 2016. 48: 331–335.