Derivation and expansion using only small molecules of human neural progenitors for neurodegenerative disease modeling

Abstract

Phenotypic drug discovery requires billions of cells for high-throughput screening (HTS) campaigns. Because up to several million different small molecules will be tested in a single HTS campaign, even small variability within the cell populations for screening could easily invalidate an entire campaign. Neurodegenerative assays are particularly challenging because neurons are post-mitotic and cannot be expanded for implementation in HTS. Therefore, HTS for neuroprotective compounds requires a cell type that is robustly expandable and able to differentiate into all of the neuronal subtypes involved in disease pathogenesis. Here, we report the derivation and propagation using only small molecules of human neural progenitor cells (small molecule neural precursor cells; smNPCs). smNPCs are robust, exhibit immortal expansion, and do not require cumbersome manual culture and selection steps. We demonstrate that smNPCs have the potential to clonally and efficiently differentiate into neural tube lineages, including motor neurons (MNs) and midbrain dopaminergic neurons (mDANs) as well as neural crest lineages, including peripheral neurons and mesenchymal cells. These properties are so far only matched by pluripotent stem cells. Finally, to demonstrate the usefulness of smNPCs we show that mDANs differentiated from smNPCs with LRRK2 G2019S are more susceptible to apoptosis in the presence of oxidative stress compared to wild-type. Therefore, smNPCs are a powerful biological tool with properties that are optimal for large-scale disease modeling, phenotypic screening, and studies of early human development.

Figure 1. Derivation of neural epithelial cells.

(A) Plated EBs differentiating in the presence of both PMA and CHIR for 6 days. (B) Phase-contrast images of neural epithelial cells on the indicated days after splitting. (C) Immunostaining of hESC-derived neural epithelial cells with antibodies raised against the indicated neural progenitor markers. Nuclei are counterstained with Hoechst. (D) Immunostaining of spontaneously differentiated neural epithelial cells for TUBBIII and NEUN, for GFAP and S100-beta after astrocyte differentiation, as well as O4 and OLIG2 after spontaneous differentiation, indicating oligodendrocyte formation. Scale bars are 100 µm. See also Figure S1.

Figure 2. Neural epithelial cells can be respecified along both the dorsoventral and rostrocaudal axes.

(A) qRT-PCR analysis for the indicated marker on neural epithelial cells cultured under the indicated conditions for 6 days. Error bars represent the standard deviation from 3 independent cultures. (B) Immunostaining of neural epithelial cells cultured with the indicated concentration of PMA or CHIR for 6 days for PAX3, NKX2.2, NKX6.1, FOXA2, and SOX1. (C) qRT-PCR analysis of neural epithelial cells cultured with 1 µM RA for 1 or 8 days for the indicated rostrocaudal marker. Error bars represent the variation from 2 independent cultures. Scale bars are 100 µm. See also Figure S3.

Figure 3. Differentiation of PNS neurons and mesenchymal cells from neural epithelial cells.

(A) Summary of isolation and differentiation protocol used in this study. (B) Immunostaining of differentiated neural epithelial cells for HNK-1. (C) Neural epithelial cells were treated with CHIR for 2 days and then switched to BMP4 or serum-containing medium for 2 additional days. In both cases, the cultures show cells positive for the neural plate border/neural crest markers PAX7 and SLUG, whereas only some cells are still positive for PAX6. (D) Immunostaining of smNPCs differentiating in the presence of BMP4 for PERIPHERIN and BRN3A. (E) qRT-PCR demonstrating the upregulation of PERIPHERIN and BRN3A in neural epithelial cells differentiated for 8 days in the presence of BMP4, but not PMA, following two weeks of maturation. (F) More than 40% of cells are double positive for PERIPHERIN and TUBBIII after patterning with BMP4 and maturation. Error bars represent variation from 2 independent cultures. Scale bars are 100 µm. CHIR = CHIR99021, DM = dorsomorphin, FCS = fetal calf serum, PMA = purmorphamine, and SB = SB43152.

Figure 4. Directed differentiation of neural epithelial cells into mDANs.

(A) Summary of isolation and differentiation protocol used in this study. (B) hESC-derived neural epithelial cells were differentiated into mDANs and immunostained for TH and FOXA2 and counterstained for nuclei with Hoechst. (C) Immunostaining of neural epithelial cell–derived mDANs for TH and TUBIII and counterstained for nuclei with Hoechst. (D) qRT-PCR analysis of neural epithelial cell–derived cultures for the indicated markers of mDAN specification on day 21. Error bars show standard deviation from 3 different experiments. (E) Efficiency of mDAN formation for 3 independent neural epithelial cell lines. Error bars represent the variation between 2 independent cultures. Scale bars are 100 µm. CHIR = CHIR99021, DM = dorsomorphin, PMA = purmorphamine, and SB = SB43152.

Figure 5. Directed differentiation of neural epithelial cells into MNs.

(A) Summary of isolation and differentiation protocol used in this study. (B) hESC-derived neural epithelial cells were differentiated into MNs and immunostained for ISLET1, CHAT, and MAP2 and counterstained for nuclei with Hoechst. (C) Immunostaining showing colocalization of ISLET1, SMI32 and CHAT. (D) Immunostaining of neural epithelial cell–derived MNs showing colocalization of HB9 and TUBIII. (E) qRT-PCR analysis of neural epithelial cell–derived cultures for the indicated markers of MN specification on day 21 of differentiation. Error bars show standard deviation of 2 independent experiments. (F) MN differentiation efficiency from neural epithelial cells was approximately 50% as determined by TUBBIII and HB9 colocalization. Error bars represent variation from 3 independent cultures. Scale bars are 100µm. CHIR = CHIR99021, DM = dorsomorphin, PMA = purmorphamine, RA = all-trans retinoic acid, and SB = SB43152.

Figure 6. smNPCs-derived neurons become electrophysiologically mature and mature in vivo.

(A) On average, the recorded membrane potential from smNPC-derived neurons was −35±2 mV (n = 12) and the cell membrane capacitance was 31.88±4.36 pF (n = 12). These values are consistent with previously published results of neurons differentiated from human stem cells , , . The net of transmembrane currents, elicited by the voltage steps from holding potential −70 mV to +20 mV with 10 mV increments (the above panel shows the stimulation paradigm). (B) Current-voltage relationship of inward and outward currents, measured on the peak and normalized to cell capacitance (n = 8). (C) Cells demonstrate spontaneous firing of action potentials (APs) like neurons. Right panel shows more detailed view on the unitary action APs. See also Figure S8 for additional data and recording of miniature potentials. (D)–(H) smNPCs integrate and mature in vivo. Transplanted human cells were identified using human Nuclei (hNuclei)- and human NCAM (hNCAM)- specific antibodies. (D) Two weeks after transplantation into the midbrain of immunodeficient mice, smNPCs differentiated into DCX-positive and TUBBIII (E) neurons. (F) Two weeks post transplantation, smNPC are negative for SOX2. (G) smNPCs form synapses already two weeks post transplantation, as shown by SYNAPTOPHYSIN and hNCAM staining. (H) Only after treatment with PMA and FGF8 for 8 days before transplantation did TH+ neurons form. For further analysis and long-term survival after eight weeks, see Figure S9.

Figure 7. smNPCs are suitable for modeling of neurological diseases such as LRRK2 G2019S induced Parkinson’s Disease.

Patient-specific human iPSCs from two patients with Parkinson’s disease harboring LRRK2 G2019S were differentiated in parallel with two lines derived from healthy age- and sex-matched control donors. After two weeks of maturation, the cultures were replated as single cells. Medium was switched to N2 medium or N2 medium supplemented with 5 µM 6-OHDA, or 10 µM 6-OHDA, or 100 nM Rotenone to induce additional cytotoxic stress. Apoptosis was assessed by double-staining for TH and cleaved CASPASE3 (CASP) in duplicate wells for each line and concentration. (A) Example picture showing TH+/CASP3- (empty arrowhead) and TH+/CASP3+ neurons (arrowhead). (B) When normalized to the average number of apoptotic cells detected in the wild-type cultures, 6-OHDA and rotenone lead to a higher cell death, with an even higher increase in cells carrying LRRK2 G2019S. Error bars represent the variation from duplicate wells. (C) When normalizing each concentration to the average apoptosis in TH+ neurons from healthy controls, an increase of 46% can be observed in LRRK2 G2019S over wild type cultures in all stressor concentrations used. Error bars represent S.D. *** indicates p<0.001, according to Student’s t-test. See also Figure S11 for primary, unnormalized data.

Figure 8. Summary of smNPCs.

Diagram illustrating the conditions used to derive, propagate, and differentiate smNPCs. CHIR = 99021, DM = dorsomorphin, FCS = fetal calf serum, PMA = purmorphamine, RA = all-trans retinoic acid, and SB = SB43152.