You are here

| Direct Reprogramming

Reprogramming of Pericyte-Derived Cells of the Adult Human Brain into Induced Neuronal Cells



Findings published on October 4th 2012 in Cell Stem Cell, have shown it is possible to reprogramme mouse and human pericytes, brain cells typically found surrounding the endothelial cells of the brain’s capillary network, into neurons. The results from Benedikt Berninger’s group at the Ludwig-Maximilians University Munich could have potential therapeutic implications for patients with traumatic and degenerative brain disorders.

Cell cultures were established from 30 brain samples from patients undergoing surgery intended to treat epilepsy.  These cultures were found to be enriched in NG2/PDGFRβ/SMA-positive pericytes and were devoid of beta-III-tubulin expressing neuroblasts or surviving neurons. Cultures were then transfected with a retrovirus expressing Sox2 and Mash1.  Within 4 weeks 48% of cells became neuronal-like, displaying characteristic cytomorphology and expressing beta-III-tubulin. The development of polarized morphology of these neuronal-like cells was also recorded using time-lapse video microscopy.

Using live imaging the pericyte population was noted not to undergo any further cell division following the retroviral transfection and expression of markers of neuronal development (including beta-III-tubulin). The authors infer this as evidence for the direct conversion of adult human nonneuronal pericyte cells into human pericyte derived induced neuronal cells (hPdiN). Of these Sox2/Mash1 double positive hPdiN, 36% underwent cell death compared to 3% of untransfected cells.  The total efficiency of reprogramming was 19%.

The authors used genetic fate mapping in mice to confirm that the neo-neurons were derived from original pericytes. Transgenic mice were engineered to express yellow fluorescent protein and β-galactosidase in microvessel-associated pericytes. Cerebral cortex cultures were prepared from these transgenic mice and transfected with the Sox2/Mash1 retrovirus, significantly increasing the proportion of beta-III-tubulin-positive cells to 92%.

The authors then characterized the neuronal function of hPdiNs. 71% (n=17) of hPdiN coexpressing Sox2/Mash1 responded to step-current injection with the generation of a single low peak action potential indicative of an immature neuronal response. However, integration of hPdiNs into the murine embryonic neocortex following their co-culture facilitated the production of repetitive firing patterns indicative of a more mature response. The cells also displayed a more complex cytomorphology and immunoreactivity for the inhibitory neurotransmitter β-aminobutyric acid (GABA).

Although there is a requirement for further work to improve the efficiency of reprogramming, these results indicate that an in vivo population of cerebral cells could potentially be used to reconstitute damaged neurons. These techniques could also obviate the need for invasive neuronal transplantation required in strategies generating neuronal cells from ex vivo iPSC or fibroblasts.



M. Karow et al., “Reprogramming of Pericyte-Derived Cells of the Adult Human Brain into Induced Neuronal Cells,” Cell Stem Cell, 11: 471-76, 2012.

Pang ZP, et al. “Induction of human neuronal cells by defined transcription factors”. Nature. 2011 May 26;476(7359):220-3. doi: 10.1038/nature10202.