Candace L. Lynch

Dynamic Changes in the Copy Number of Pluripotency and Cell Proliferation Genes in Human ESCs and iPSCs during Reprogramming and Time in Culture

Genomic stability is critical for the clinical use of human embryonic and induced pluripotent stem cells. We performed high-resolution SNP (single-nucleotide polymorphism) analysis on 186 pluripotent and 119 nonpluripotent samples. We report a higher frequency of subchromosomal copy number variations in pluripotent samples compared to nonpluripotent samples, with variations enriched in specific genomic regions. The distribution of these variations differed between hESCs and hiPSCs, characterized by large numbers of duplications found in a few hESC samples and moderate numbers of deletions distributed across many hiPSC samples. For hiPSCs, the reprogramming process was associated with deletions of tumor-suppressor genes, whereas time in culture was associated with duplications of oncogenic genes. We also observed duplications that arose during a differentiation protocol. Our results illustrate the dynamic nature of genomic abnormalities in pluripotent stem cells and the need for frequent genomic monitoring to assure phenotypic stability and clinical safety.

Recurrent Variations in DNA Methylation in Human Pluripotent Stem Cells and Their Differentiated Derivatives

Human pluripotent stem cells (hPSCs) are potential sources of cells for modeling disease and development, drug discovery, and regenerative medicine. However, it is important to identify factors that may impact the utility of hPSCs for these applications. In an unbiased analysis of 205 hPSC and 130 somatic samples, we identified hPSC-specific epigenetic and transcriptional aberrations in genes subject to X chromosome inactivation (XCI) and genomic imprinting, which were not corrected during directed differentiation. We also found that specific tissue types were distinguished by unique patterns of DNA hypomethylation, which were recapitulated by DNA demethylation during in vitro directed differentiation. Our results suggest that verification of baseline epigenetic status is critical for hPSC-based disease models in which the observed phenotype depends on proper XCI or imprinting and that tissue-specific DNA methylation patterns can be accurately modeled during directed differentiation of hPSCs, even in the presence of variations in XCI or imprinting.

Specific lectin biomarkers for isolation of human pluripotent stem cells identified through array-based glycomic analysis

Rapid and dependable methods for isolating human pluripotent stem cell (hPSC) populations are urgently needed for quality control in basic research and in cell-based therapy applications. Using lectin arrays, we analyzed glycoproteins extracted from 26 hPSC samples and 22 differentiated cell samples, and identified a small group of lectins with distinctive binding signatures that were sufficient to distinguish hPSCs from a variety of non-pluripotent cell types. These specific biomarkers were shared by all the 12 human embryonic stem cell and the 14 human induced pluripotent stem cell samples examined, regardless of the laboratory of origin, the culture conditions, the somatic cell type reprogrammed, or the reprogramming method used. We demonstrated a practical application of specific lectin binding by detecting hPSCs within a differentiated cell population with lectin-mediated staining followed by fluorescence microscopy and flow cytometry, and by enriching and purging viable hPSCs from mixed cell populations using lectin-mediated cell separation. Global gene expression analysis showed pluripotency-associated differential expression of specific fucosyltransferases and sialyltransferases, which may underlie these differences in protein glycosylation and lectin binding. Taken together, our results show that protein glycosylation differs considerably between pluripotent and non-pluripotent cells, and demonstrate that lectins may be used as biomarkers to monitor pluripotency in stem cell populations and for removal of viable hPSCs from mixed cell populations.

Increased Risk of Genetic and Epigenetic Instability in Human Embryonic Stem Cells Associated with Specific Culture Conditions

The self-renewal and differentiation capacities of human pluripotent stem cells (hPSCs) make them a promising source of material for cell transplantation therapy, drug development, and studies of cellular differentiation and development. However, the large numbers of cells necessary for many of these applications require extensive expansion of hPSC cultures, a process that has been associated with genetic and epigenetic alterations. We have performed a combinatorial study on both hESCs and hiPSCs to compare the effects of enzymatic vs. mechanical passaging, and feeder-free vs. mouse embryonic fibroblast feeder substrate, on the genetic and epigenetic stability and the phenotypic characteristics of hPSCs. In extensive experiments involving over 100 continuous passages, we observed that both enzymatic passaging and feeder-free culture were associated with genetic instability, higher rates of cell proliferation, and persistence of OCT4/POU5F1-positive cells in teratomas, with enzymatic passaging having the stronger effect. In all combinations of culture conditions except for mechanical passaging on feeder layers, we noted recurrent deletions in the genomic region containing the tumor suppressor gene TP53, which was associated with decreased mRNA expression of TP53, as well as alterations in the expression of several downstream genes consistent with a decrease in the activity of the TP53 pathway. Among the hESC cultures, we also observed culture-associated variations in global gene expression and DNA methylation. The effects of enzymatic passaging and feeder-free conditions were also observed in hiPSC cultures. Our results highlight the need for careful assessment of the effects of culture conditions on cells intended for clinical therapies.

Normal Human Pluripotent Stem Cell Lines Exhibit Pervasive Mosaic Aneuploidy

Human pluripotent stem cell (hPSC) lines have been considered to be homogeneously euploid. Here we report that normal hPSC – including induced pluripotent - lines are karyotypic mosaics of euploid cells intermixed with many cells showing non-clonal aneuploidies as identified by chromosome counting, spectral karyotyping (SKY) and fluorescent in situ hybridization (FISH) of interphase/non-mitotic cells. This mosaic aneuploidy resembles that observed in progenitor cells of the developing brain and preimplantation embryos, suggesting that it is a normal, rather than pathological, feature of stem cell lines. The karyotypic heterogeneity generated by mosaic aneuploidy may contribute to the reported functional and phenotypic heterogeneity of hPSCs lines, as well as their therapeutic efficacy and safety following transplantation.

Restricted ethnic diversity in human embryonic stem cell lines

To the Editor: Human pluripotent stem cells (hPSCs) have the capacity to self-renew indefinitely and to differentiate into a wide array of cell types, which make them a potential source of essentially unlimited quantities of differentiated cells for basic and clinical research. The tremendous self-renewal of hPSCs might lead one to conclude that a small number of cell lines would be sufficient to meet all needs. However, it is becoming increasingly clear that the genetic background of human cell lines can have significant effects on experimental results. Although hundreds of associations between individual alleles and specific traits or diseases are discovered each year, a full understanding of the relationships between genetic variation and cellular or organismal phenotype is still remote1. The HapMap project2, along with many other studies, has made it clear that there are large differences in the allelic frequencies for many single-nucleotide polymorphisms (SNPs) among different ethnicities. Ethnicity can serve as a proxy for genetic variation to ensure diversity in genetic backgrounds in a study population. The availability of hPSCs from a variety of ethnic backgrounds would ensure the generalizability of results as well as increasing the likelihood that specific alleles or combinations of alleles of interest will be available for study. This is particularly important for the use of hPSCs and hPSC-derived cells for drug screening and toxicity studies. Several idiosyncratic drug effects have been attributed to specific genetic variations3, and the efficacy and toxicity of numerous drugs are presumed to be influenced by genetic factors. On the clinical side, hPSCs and their derivatives are potential sources for cell therapy, and recipients of cell transplants are more likely to find immunologic matches with donors who share their ethnic background. However, because most hPSC lines have been generated from de-identified material, the ethnic backgrounds of the donors are not known. We determined the ethnic origin of 47 human embryonic stem cell (hESC) lines (including 42 standard hESC lines and 5 parthenote hESC lines), 5 hiPSC lines and 58 non-pluripotent samples (19 blood and tissue samples and 39 primary cell lines). Although this study was not intended to be comprehensive, we included hESC lines derived in a variety of geographical locations in order to sample the spectrum of ethnic diversity present in hESC lines in general (Supplementary Table 1). Using genome-wide SNP genotyping and Bayesian analysis of population structure (BAPS; see Supplementary Methods), we determined that the ethnic origins of the hESC lines included in this study were quite restricted (Fig. 1, Supplementary Table 1). The large majority of hESC lines (43 of 47) were of European and East Asian ethnicity. The diversity of the hiPSC lines and nonpluripotent cells is also shown (Fig. 1, Supplementary Fig. 1). We note that the location of derivation of the hESC lines is not necessarily indicative of their ethnic origin. For instance, four of the five hESC lines described in one landmark publication4 were derived in the United States (Wisconsin) but reportedly5 came from blastocysts transported from Israel by collaborators. Our results on three of these lines indicate that they show a genetic profile similar to those of people from the region around Adygea (WA01), Tuscany (WA09) and Tuscany/Palestine (WA07), with indication of enzymatic activity3. We targeted residues Asp154, Glu155, Lys136 and Ser287 based on previous experiments showing increases in stability and/or enzymatic activity from mutations at these sites2,3. Mutations at Pro220 were tested based on a report of increased activity with alterations at this residue5; however, we were unable to find substitutions at Pro220 that increased activity or red-shift. The end result of this selection process was the variant RLuc/E155G/D162E/A164R/ L165I/M185V/Q235A/S287A, which we denote RLuc7-521. We used purified protein to assess the bioluminescence emission spectrum, enzymatic activity and quantum yield of the new variant (Supplementary Table 1). RLuc7-521’s signal was increased 1.6-fold on a photons per second per mole basis, and its emission spectra (Fig. 1a) showed a 40-nm red-shift compared to RLuc with both coelenterazine and the substrate analog coelenterazine-v. RLuc7-521’s quantum yield of 3.9 ± 0.1%, though higher than that of RLuc8.6535 (3.1 ± 0.2%), was lower than that of RLuc (5.3 ± 0.1%), indicating that further improvements may be achievable through additional mutagenesis. We tested RLuc7-521 as a mammalian reporter gene in cell culture and observed a nearly twofold improvement in light output compared to RLuc, and an almost identical intracellular stability (Fig. 1b). We also tested RLuc7-521 for small-animal imaging by injecting cells transiently transfected with RLuc or RLuc7-521 into mice. Western analysis of cells before injection showed equivalent levels of luciferase protein expression between conditions (Supplementary Fig. 2). Subsequent imaging of the mice showed a 3.3-fold increase in signal output for RLuc7-521 (Fig. 1c), a reflection of the improved light output of RLuc7-521 combined with reductions in signal attenuation by tissue for this red-shifted luciferase. In summary, RLuc7-521 is a seven-mutation variant of RLuc that shows a green-peaked (521-nm) emission spectrum, an identical intracellular stability, and at least a twofold increase in signal, with even greater signal gains in small-animal imaging due to decreased attenuation of its red-shifted photons. This new variant represents a direct replacement of Renilla luciferase for reporter-gene applications as it retains the temporal relationship between gene activation and luciferase activity while providing greater sensitivity.

Glycosyltransferase ST6GAL1 contributes to the regulation of pluripotency in human pluripotent stem cells

Many studies have suggested the significance of glycosyltransferase-mediated macromolecule glycosylation in the regulation of pluripotent states in human pluripotent stem cells (hPSCs). Here, we observed that the sialyltransferase ST6GAL1 was preferentially expressed in undifferentiated hPSCs compared to non-pluripotent cells. A lectin which preferentially recognizes α-2,6 sialylated galactosides showed strong binding reactivity with undifferentiated hPSCs and their glycoproteins, and did so to a much lesser extent with differentiated cells. In addition, downregulation of ST6GAL1 in undifferentiated hPSCs led to a decrease in POU5F1 (also known as OCT4) protein and significantly altered the expression of many genes that orchestrate cell morphogenesis during differentiation. The induction of cellular pluripotency in somatic cells was substantially impeded by the shRNA-mediated suppression of ST6GAL1, partially through interference with the expression of endogenous POU5F1 and SOX2. Targeting ST6GAL1 activity with a sialyltransferase inhibitor during cell reprogramming resulted in a dose-dependent reduction in the generation of human induced pluripotent stem cells (hiPSCs). Collectively, our data indicate that ST6GAL1 plays an important role in the regulation of pluripotency and differentiation in hPSCs, and the pluripotent state in human cells can be modulated using pharmacological tools to target sialyltransferase activity.

Molecular analyses of neurogenic defects in a human pluripotent stem cell model of fragile X syndrome

New research suggests that common pathways are altered in many neurodevelopmental disorders including autism spectrum disorder; however, little is known about early molecular events that contribute to the pathology of these diseases. The study of monogenic, neurodevelopmental disorders with a high incidence of autistic behaviours, such as fragile X syndrome, has the potential to identify genes and pathways that are dysregulated in autism spectrum disorder as well as fragile X syndrome. In vitro generation of human disease-relevant cell types provides the ability to investigate aspects of disease that are impossible to study in patients or animal models. Differentiation of human pluripotent stem cells recapitulates development of the neocortex, an area affected in both fragile X syndrome and autism spectrum disorder. We have generated induced human pluripotent stem cells from several individuals clinically diagnosed with fragile X syndrome and autism spectrum disorder. When differentiated to dorsal forebrain cell fates, our fragile X syndrome human pluripotent stem cell lines exhibited reproducible aberrant neurogenic phenotypes. Using global gene expression and DNA methylation profiling, we have analysed the early stages of neurogenesis in fragile X syndrome human pluripotent stem cells. We discovered aberrant DNA methylation patterns at specific genomic regions in fragile X syndrome cells, and identified dysregulated gene- and network-level correlates of fragile X syndrome that are associated with developmental signalling, cell migration, and neuronal maturation. Integration of our gene expression and epigenetic analysis identified altered epigenetic-mediated transcriptional regulation of a distinct set of genes in fragile X syndrome. These fragile X syndrome-aberrant networks are significantly enriched for genes associated with autism spectrum disorder, giving support to the idea that underlying similarities exist among these neurodevelopmental diseases.

Melanocytes Derived from Transgene-Free Human Induced Pluripotent Stem Cells

TO THE EDITOR Defects in melanocytes have been implicated in the etiology of a variety of human skin diseases and disorders (Lin and Fisher, 2007; Fistarol and Itin, 2010; Rees, 2011). There is long-standing interest in studying the development and dysfunction of human melanocytes, but there has not been a reliable and accessible system to study early events in human melanocyte differentiation. An in vitro system that reliably and efficiently produces normal human melanocytes from embryonic stage cells would allow us to better dissect the physiological and pathological development of melanocytes. Recent advances in stem cell biology have led to the establishment of human induced pluripotent stem cell (hiPSC) techniques that enable researchers to reprogram somatic cells to the pluripotent state (Takahashi et al., 2007). Differentiation of human and mouse pluripotent stem cells (PSCs) toward the melanocyte lineage has been reported (Yamane et al., 1999; Pla et al., 2005; Fang et al., 2006; Nissan et al., 2011; Ohta et al., 2011; Yang et al., 2011), but existing protocols have shortcomings that may limit their research and clinical applications. For example, the use of embryonic stem cells could lead to allogeneic immunoincompatibility of differentiated melanocytes and transplant recipients. In addition, the use of hiPSCs generated by integrative reprogramming strategies raises concerns about reactivation of retained transgenes, some of which are oncogenes. In addition, the current methods for melanocyte differentiation from hiPSCs require optimization in order to reproducibly generate high-purity melanocytes from multiple hiPSC lines. We have established a strategy to produce human melanocytes in vitro for use as a platform for pigment cell research and the development of cell-based therapies. We first derived transgene-free hiPSCs from two distinct types of skin cells: human primary melanocytes (HMs) and human dermal fibroblasts (HDF51) (Figure 1a and Supplementary Figure S1a online). We used a nonintegrative reprogramming approach mediated by Sendai virus–based vectors independently encoding POU5F1, SOX2, KLF4, and MYC (Fusaki et al., 2009; Macarthur et al., 2012). As shown in Figure 1b and Supplementary Figure S1b online, biomarkers of cellular pluripotency, including endogenous OCT4/POU5F1, NANOG, Tra-1-81, and UEA-I (Wang et al., 2011), were positive in HMi-506, HMi-503, and HDF51i-509 hiPSCs. Cells were also shown to be pluripotent using a gene expression diagnostic test (PluriTest; Muller et al., 2011), by differentiation into cells that express biomarkers relevant to all three germ layers in vitro (Figure 1c and Supplementary Figure S1c, S1d and S1e online) and by generation of teratomas (Supplementary Figure S1d online). Figure 1 Generation and differentiation of transgene-free human induced pluripotent stem cell (hiPSCs). (a) HMi-506 cells generated from human primary melanocyte (HM) cells using a Sendai virus–based reprogramming system were cocultured with mouse embryonic ... We newly developed two differentiation protocols based on previously reported methods. One protocol involves an aggregation-in-suspension step, whereas the other does not (Supplementary Figure S2 online). Both protocols generated cells displaying typical melanocyte morphology and pigmentation (Figure 1d) from hiPSCs after 30 days of directed differentiation, suggesting that the aggregation-in-suspension step is dispensable. The melanin granules that accumulated at the dendritic tips of differentiated cells were intensely stained by Fontana–Masson staining, indicating that the pigmentation of these cells was due to melanogenesis (Supplementary Figure S3 online). In addition, MITF (microphthalmia-associated transcription factor), a marker for melanocyte progenitors, was expressed in more than 90% of the differentiated derivatives after 30 days (Figure 1e and Supplementary Figure S4 online), which appears to be a higher differentiation efficiency than other reported protocols (Nissan et al., 2011; Ohta et al., 2011). As expected, MITF was not detected in the undifferentiated hiPSCs, and was present in the primary melanocytes (Figure 1e). Notably, our protocols resulted in similarly high levels of melanocyte differentiation for all four independent hiPSC lines examined, highlighting their reproducibility. Other melanocytic biomarkers including TYR (tyrosinase), MLANA (melan-A), TYRP1 (tyrosinase-related protein 1), PMEL (premelanosome protein), PAX3 (paired box 3), and SOX10 (SRY-box 10) were highly expressed in the differentiated derivatives (similar to primary melanocytes, Figure 2a and b). The melanin content and cell signaling involved in melanin production in the differentiated derivatives was increased by treatment with α-melanocyte-stimulating hormone (α-MSH) in a dose-dependent manner (Figure 2c and d and Supplementary Figure S5 online). These findings indicate that the differentiated derivatives possess molecular features of bona fide melanocytes and accurately mimic their ability to respond to α-MSH, which is the factor that activates melanogenesis and enhances skin pigmentation during the tanning response (Thody, 1999). Figure 2 Molecular and functional characterization of the melanocyte-like differentiated cells. (a) Heat map and dendrogram of melanocytic biomarkers showing that these transcripts were preferentially expressed in human primary melanocyte (HM) cells and HMi-506_Mel ... Genome-wide gene expression profiling and unsupervised hierarchical clustering revealed that the melanocytes (HMi-506_Mel Diff_1 and HMi-506_Mel Diff_2) differentiated from the HMi-506 cells were closely clustered with HMs and were distinct from all undifferentiated hiPSC samples (Figure 2e). As genetic abnormalities may occur in hiPSC genomes during the reprogramming and differentiation processes, we tested the genomic stability of the cells by comparing the differentiated derivatives with the parental primary melanocytes using high-resolution single-nucleotide polymorphism (SNP) genotyping and copy number variation analysis. As shown in Figure 2f, the HMi-506_Mel Diff derivatives and parental cells showed highly similar genotyping profiles, showing that the cellular genome remained stable during reprogramming and differentiation. Similar to human melanocytes in vivo, the differentiated derivatives in semiautologous skin reconstructs were located at the dermis–epidermis interface and interspersed with keratinocytes (Supplementary Figure S6a, S6b, S6c and S6d online), indicating their ability to integrate with the skin tissue of transplant recipients. Similar to the autologous dermal fibroblasts used for generating transgene-free hiPSCs, the differentiated derivatives stimulated limited proliferation of peripheral blood mononuclear cells that were isolated from the blood of the same individual in a mixed lymphocyte reaction assay (Supplementary Figure S6e online). These results attest to the clinical advantages of melanocytes differentiated from hiPSCs using the reprogramming and differentiation approaches described here. In this study, we have demonstrated that genetically stable melanocytes can be efficiently differentiated from transgene-free hiPSCs generated from two different types of cutaneous cells. This differentiation protocol takes less time than previously reported melanocytic differentiation protocols, and we showed that it is equally effective for multiple independent hiPSC lines. We performed a thorough investigation of the differentiated cells, including genome-wide gene expression analysis and SNP genotyping in addition to functional assays. Our approach can serve as an unlimited source of custom human melanocytes that can be used for novel approaches for modeling human skin disease (e.g., melanoma and vitiligo) and to provide material for transplantation.