Autologous iPSC-derived dopamine neuron transplantation in a nonhuman primate Parkinson’s disease model

Abstract

Autologous dopamine (DA) neurons are a new cell source for replacement therapy of Parkinson’s disease (PD). In this study, we tested the safety and efficacy of autologous induced pluripotent stem cell (iPSC)-derived DA cells for treatment of a cynomolgus monkey PD model. Monkey bone marrow mesenchymal cells were isolated and induced to iPSCs, followed by differentiation into DA cells using a method with high efficiency. Autologous DA cells were introduced into the brain of a cynomolgus monkey PD model without immunosuppression; three PD monkeys that had received no grafts served as controls. The PD monkey that had received autologous grafts experienced behavioral improvement compared with that of controls. Histological analysis revealed no overgrowth of grafts and a significant number of surviving A9 region-specific graft-derived DA neurons. The study provided a proof-of-principle to employ iPSC-derived autologous DA cells for PD treatment using a nonhuman primate PD model.

Transplantation in the nonhuman primate MPTP model of Parkinson's disease: update and perspectives

“…For example, VM tissue isolated from early stage fetuses showed a higher survival rate than that of older stage fetuses (Sladek Jr. et al, 1993b;Collier et al, 2002), as was previously shown in rodents (Fricker et al, 1997;Torres et al, 2007;Torres et al, 2008). Similarly, early DA progenitors derived from primate PSCs (D21-D28 neurospheres) produced larger grafts (Kikuchi et al, 2011) and showed a higher survival rate (Takagi et al, 2005;Wang et al, 2015) than did late DA neurons (D42 neurospheres; Sanchez-Pernaute et al, 2005;Takagi et al, 2005;Kikuchi et al, 2011;Doi et al, 2012;Emborg et al, 2013a). Nevertheless, the survival rate never exceeded 5 %.…”

“…This suggests that a minimal restoration of normal DA innervation in the striatum is sufficient for functional recovery, as previously shown in the rodent model (Grealish et al, 2014). However, a certain threshold of TH+ cell dose is needed to improve PD motor symptoms and was estimated to be approximately 10 000 TH+ cells per brain, whether derived from ESCs (Doi et al, 2012) or from iPSCs (Hallett et al, 2015;Wang et al, 2015). A lower number of TH+ cells led to poor dopamine reinnervation and no functional recovery (Sanchez-Pernaute et al, 2005;Hallett et al, 2015).…”

“…In contrast, single-cell suspensions of NSCs, DA progenitors or neurons enable precise counting of the transplanted cells, and 1 to 10 million of cells were usually injected per monkey brain (Table 1a). Looking back on several decades of transplantation studies in MPTP NHPs, it appears that the survival rate of transplanted DA neurons is disappointing, with generally more than 90 % of the cells dying after transplantation (Table 1a; Sanchez-Pernaute et al, 2005;Redmond Jr. et al, 2007;Doi et al, 2012;Emborg et al, 2013a;Wang et al, 2015).…”

“…Human ESCs (Kriks et al, 2011;Daadi et al, 2012;Doi et al, 2012;Grealish et al, 2014;Gonzalez et al, 2015Gonzalez et al, , 2016Chen et al, 2016) and monkey ESCs (Kawasaki et al, 2002;Sanchez-Pernaute et al, 2005;Takagi et al, 2005;Xi et al, 2012) were first used, recently followed by human iPSCs (Kikuchi et al, 2011;Kriks et al, 2011;Sundberg et al, 2013;Doi et al, 2014) and monkey iPSCs (Morizane et al, 2013;Sundberg et al, 2013;Wang et al, 2015).…”

“…Complete and robust midbrain specification was recently obtained via a floor plate intermediate stage from human (Kriks et al, 2011;Xi et al, 2012;Sundberg et al, 2013) and NHP PSCs (Xi et al, 2012;Hallett et al, 2015;Wang et al, 2015), using a modified dual-SMAD inhibition protocol (BMP/TGFβ inhibition; Chambers et al, 2009) and activation of Wnt signalling, an essential pathway in DA neuron development in the mouse (Castelo-Branco et al, 2003Joksimovic et al, 2009) and in humans (La Manno et al, 2016). Combining suspension culture with dual-SMAD inhibition, Wnt and SHH activation also led to robust VM differentiation, with correct midbrain GIRK2+ A9 and CALBINDIN+ A10 phenotypes, similar to fVM content (Kirkeby et al, 2012;Morizane et al, 2013;Doi et al, 2014;Grealish et al, 2014;Chen et al, 2016).…”