Optimized surface markers for the prospective isolation of high-quality hiPSCs using flow cytometry selection

Abstract

hiPSC derivation and selection remains inefficient; with selection of high quality clones dependent on extensive characterization which is not amenable to high-throughput (HTP) approaches. We recently described the use of a cocktail of small molecules to enhance hiPSC survival and stability in single cell culture and the use of flow cytometry cell sorting in the HTP-derivation of hiPSCs. Here we report an enhanced protocol for the isolation of bona fide hiPSCs in FACS-based selection using an optimized combination of cell surface markers including CD30. Depletion of CD30(+) cells from reprogramming cultures almost completely abolished the NANOG and OCT4 positive sub-population, suggesting it is a pivotal marker of pluripotent cells. Combining CD30 to SSEA4 and TRA-1-81 in FACS greatly enhanced specificity and efficiency of hiPSC selection and derivation. The current method allows for the efficient and automated, prospective isolation of high-quality hiPSC from the reprogramming cell milieu.

Figure 1. Conventional pluripotency markers SSEA4 and TRA-1-81 do not discriminate between partially and fully reprogrammed hiPSCs.

(a) Flow cytometry analysis of hiPSC lines generated by sorting human FTc1 fibroblasts directly into 96-well plate in SMC4 media 6 weeks post infection with polycistronic reprogramming lentivirus (expressing OCT4, SOX2, and KLF4). (b) SSEA4⁺ TRA-1-81⁺ cell lines were analyzed by qRT-PCR for the endogenous expression of pluripotency markers (NANOG, OCT4 and SOX2) and transgene. Error bars represent standard deviation of duplicates. Differences in gene expression of pluripotency markers in all FTc1 cell lines are significant (p-values < 0.05) relative to the H1 ESCs sample. (c) SSEA4⁺ TRA-1-81⁺ cell lines FTc1-C9 and FTc1-C12 were stained for expression of TRA-1-81 (green) and NANOG (red) and examined under fluorescence microscopy. Nuclei were stained with DAPI (blue). Scale bar is 400 μm.

Figure 2. Surface marker expression in hPSCs and somatic cells.

(a) Expression of indicated surface markers was compared to that of NANOG and OCT4 in an array of cell types including hPSCs (hESCs; n = 2, hiPSCs; n = 16) cultured in conventional media and on MEFs, hiPSCs cultured in SMC4 media on Matrigel (n = 11), partially reprogrammed cells cultured in SMC4 media on Matrigel (n = 6), primary adult cells (fibroblasts, n = 10; adipose stem cells, n = 1; myoblasts, n = 1), and hPSC-derived differentiated cells (embryoid bodies, n = 4; definitive endoderm, n = 1; trophectoderm, n = 1; monolayer differentiation, n = 3). Expression of indicated genes was examined by qRT-PCR, and values normalized within each set to that of hESC. (b) Expression of indicated surface markers was examined by qRT-PCR in normal tissue samples (n = 48) and results are depicted as a heat map. Distances between samples and assays are calculated for hierarchical clustering based on the ΔCT values using the Pearson's Correlation. The ΔCT range values are between -3 (High) and 15 (Low).

Figure 3. Screening CD molecules for ability to identify iPSCs from non-iPSC contaminants reveals CD30 as a specific pluripotency marker.

A partially differentiated hiPSC line (upper left) was stained with a combination of three surface markers: SSEA4 and TRA-1-81, and either TRA-1-60, CD200, CD90, CD9, CD50, or CD30. Two sub-populations were gated as shown (upper right): SSEA4⁻ TRA−1-81⁻ (non-iPSCs; grey area) and SSEA4⁺ TRA-1-81⁺ (iPSCs; black line). Histograms show expression of the indicated markers within each gate.

Figure 4. CD30 surface marker correlates with NANOG expression.

(a) Three partially reprogrammed cell lines FTc1-C8, FTc1-C17 and FTc1-C18, and one fully reprogrammed hiPSC line FTc1-c19 were stained for the surface markers TRA-1-60 (green), CD30 (red) and NANOG (blue), and examined under fluorescence microscopy. Scale bar is 200 μm. (b) Four cell lines generated by sorting SSEA4⁺ TRA-1-81⁺ cells were analyzed by qRT-PCR for expression of NANOG. (c) The two NANOG⁺ cell lines FTc8-C1 and FTc7-C23 and the two NANOG⁻ cell lines FTc10-C3 and FTc10-c9 were analyzed by flow cytometry for expression of the surface markers: SSEA4, TRA-1-81, CD30, and CD9. The histograms (right) depict the gated population of cells in the dot plots (left).

Figure 5. Selection of hiPSCs from a reprogramming pool using FACS is highly enhanced by combining CD30 with the conventional pluripotency markers SSEA4 and TRA-1-81.

(a) 37 days post initiation of reprogramming of FTc63 human fibroblasts, cells were stained for the surface markers: SSEA4, TRA-1-81, and CD30. Four different sub-populations of the reprogramming pool were separated by FACS as indicated: SSEA4⁻ TRA-1-81⁻ (S- T-), SSEA4⁺ TRA-1-81⁺ (S+ T+), SSEA4⁺ TRA-1-81⁺ CD30^- (S+ T+ C-), SSEA4⁺ TRA-1-81⁺ CD30⁺ (S+ T+ C+). Sorted cells were collected and seeded on Matrigel-coated plate in SMC4 medium. (b) The 4 different sorted sub-populations were analyzed for the expression of OCT4 (red) and NANOG (green) by fluorescence microscopy. Nuclei were stained with DAPI (blue). Scale bar in merged color image is 1000 μm. (c) The 4 different sorted sub-populations were analyzed by qRT-PCR for the endogenous expression of pluripotency markers, reprogramming factors, and indicated surface markers. Established hiPSC line FTi112 cultured in SMC4 on Matrigel, hESC lines H1 and HuES9 cultured in conventional media and on feeder cells were used as references. Podocalyxin (PODXL) is the carrier for TRA-1-60 and TRA-1-81. Error bars represent standard deviation of duplicates. The asterisks denote corresponding p-values (* <0.05, ** < 0.001).

Figure 6. hiPSC selection by sorting SSEA4⁺ TRA-1-81⁺ CD30⁺ cells yields pluripotent feeder-free hiPSC lines with stable genome and tri-lineage differentiation potential.

Four weeks post initiation of reprogramming of FTc91 human fibroblasts, hiPSCs were separated by FACS and using the SSEA4, TRA-1-81, and CD30 markers as described above. Three iPSC lines FTi115, FTi116, and FTi117 were established and characterized. (a) Flow cytometry analyses of indicated hiPSCs. (b) Indicated iPSC lines were immunostained for expression of NANOG (green) and OCT4 (red). Nuclei were stained with Hoechst dye (blue). Scale bar is 200 μm. (c) Nanog+ and Oct4+ FTi117 iPSCs were quanitified by intracellular flow cytometry. (d), (e) qRT-PCR analysis of gene expression of pluripotency markers (d) and transgene (e) in indicated hiPSC lines. The lentivirus used for reprogramming expressed the transgenes as a single polycistronic cassette. Transgene levels were measured by a TaqMan primer-probe set within the viral WPRE element. Error bars represent standard deviation of duplicates. The asterisks denote corresponding p-values (* <0.05, ** < 0.001). (f) FTi115 (p11), FTi116 (p10), and FTi117 (p8) maintained a normal karyotype (46, XY) over extended period of culture in SMC4 and on Matrigel coated plates. (g) Differentiation potential of FTi117 was tested by seeding 5-day old embryoid bodies on Matrigel-coated plate and staining cells 2 weeks later for markers of ectoderm (Nestin and Tuj1), mesoderm (αSMA), and endoderm (Foxa2 and AFP). Scale bar is 100 μm. (h) Histological sections of teratoma derived from FTi117 iPSCs. Panels show neuroepithelia (left), adipocytes (middle), gut epithelia (right).