Author contributions: T.C.: collection and assembly of data and data analysis and interpretation; W.Z. and S.B.: collection and assembly of data and data analysis; W.T.: conception and design and collection of data; H.H.: conception and design; R.L.: conception and design and data analysis and interpretation; J.Y.: conception and design, data analysis and interpretation, and manuscript writing.
Disclosure of potential conflicts of interest is found at the end of this article.
First published online in STEM CELLSEXPRESS September 28, 2011.
Spinal muscular atrophy (SMA) is one of the most common autosomal recessive disorders in humans and is a common genetic cause of infant mortality. The disease is caused by loss of the survival of motoneuron (SMN) protein, resulting in the degeneration of alpha motoneurons in spinal cord and muscular atrophy in the limbs and trunk. One function of SMN involves RNA splicing. It is unclear why a deficiency in a housekeeping function such as RNA splicing causes profound effects only on motoneurons but not on other cell types. One difficulty in studying SMA is the scarcity of patient's samples. The discovery that somatic cells can be reprogrammed to become induced pluripotent stem cell (iPSCs) raises the intriguing possibility of modeling human diseases in vitro. We reported the establishment of five iPSC lines from the fibroblasts of a type 1 SMA patient. Neuronal cultures derived from these SMA iPSC lines exhibited a reduced capacity to form motoneurons and an abnormality in neurite outgrowth. Ectopic SMN expression in these iPSC lines restored normal motoneuron differentiation and rescued the phenotype of delayed neurite outgrowth. These results suggest that the observed abnormalities are indeed caused by SMN deficiency and not by iPSC clonal variability. Further characterization of the cellular and functional deficits in motoneurons derived from these iPSCs may accelerate the exploration of the underlying mechanisms of SMA pathogenesis. STEM CELLS 2011;29:2090–2093.
Spinal muscular atrophy (SMA) is one of the most common autosomal recessive disorders and the most common genetic cause of infant mortality [1–3]. SMA is caused by loss of the survival of motoneuron (SMN) protein, resulting in the degeneration of alpha motoneurons in spinal cord. There are two SMN genes in humans, SMN1 and SMN2, and SMA is caused primarily by homozygous loss of the SMN1 gene [4, 5]. While the SMN2 gene encodes a functional protein, a defect in SMN2 RNA splicing results in the production of only 10% of the correctly spliced transcript, a level insufficient to support long-term motoneuron survival [6, 7]. Type 1 SMA is the most severe form with disease onset before 6 months of age and death within the first 2 years of life. Currently, there is no cure for SMA.
SMN exists diffusely in cytoplasm and in small, punctate structures termed “gem body” in nucleus . One function of SMN involves the biogenesis of small nuclear RNA ribonucleoprotein particles (snRNPs), which participate in pre-mRNA splicing [9, 10]. A reduction in SMN could generate defect in RNA splicing. However, SMN deficiency causes profound effects only on motoneurons but not on other cell types. SMN also exists in the axonal compartment of motoneurons where it is associated with heterogeneous nuclear RNP R and interacts with the 3′ untranslated region of β-actin mRNA [11–14]. Reduced SMN results in lower levels of β-actin mRNA and protein in axons and growth cones. Thus, SMN may be involved in transportation of mRNA to the growth cone during neuronal differentiation. Its activity in axon argues for a direct role of SMN in selective motoneuron degeneration in SMA.
It remains difficult to study SMA because of the scarcity of patient's motoneurons. Progress in developing patient-specific induced pluripotent stem cells (iPSCs) may overcome this hurdle [15–19]. Ebert et al.  established iPSC lines from a type 1 SMA patient and showed that these iPSCs exhibited reduced capacity to form mature motoneurons in vitro. As high intrinsic variability in differentiation exists among different iPSC lines [21–23], the reduced capacity to form motoneurons by the SMA iPSCs may be attributed to clonal variation rather than the underlying genetic defect. Establishment of iPSC lines from other SMA patients with similar phenotypes would address this concern. Here, we reported the establishment of five iPSC lines from a second type 1 SMA patient. These iPSC lines showed reduced motoneuron production and slower neurite development. Wild-type SMN expression in these iPSC lines restored normal motoneuron differentiation. Our study provides strong support for using the iPSC strategy to study the mechanisms of SMA pathogenesis.
MATERIALS AND METHODS
GM09677 fibroblasts were transduced with the retroviral vectors containing the four cell reprogramming genes, and independent colonies were isolated 4 weeks later. To restore SMN expression, SMA iPSCs were transduced with either HIV7/Neo or HIV7/SMN1, and G418-resistant colonies were pooled. For motoneuron differentiation, ESCs or iPSCs were first differentiated into neurospheres followed by seeding onto polyornithine-aminin-coated culture dishes in neural differentiation medium. Emergence of motoneurons occurred around day 35 after the initiation of differentiation (see Supporting Information for details).
We characterized the SMN genes in GM09677, primary fibroblasts from a type I SMA patient, before cell reprogramming. The result suggested that, similar to other type 1 SMA patients, GM09677 had a homozygous deletion in the SMN1 gene (Supporting Information Fig. S1) [4, 5]. Western blot confirmed that GM09677 cells expressed SMN at a lower level than IMR-90 normal fibroblasts (Fig. 1A). Nuclear gem bodies were readily detectable in HT1080 and IMR-90 cells but only rarely in GM09677 cells (Fig. 1B, 1C). Ectopic SMN1 expression in GM09677 cells via the transduction with HIV7/SMN1, a lentiviral vector containing the wild-type SMN1 cDNA, restored gem body formation (Fig. 1B, 1C).
To reprogram cells, GM09677 fibroblasts were transduced with the retroviral vectors carrying the oct4, sox2, klf4, and c-myc genes . Five stable iPSC lines were established and characterized (Supporting Information Figs. S1, S2, and S3). The pluripotency of these cell lines was revealed by their ability to differentiate in culture into cells expressing lineage-specific markers, including α-fetoprotein (endoderm), α-smooth muscle actin (mesoderm), nestin, and microtubule-associated protein 2 (ectoderm), and formation of teratomas in mice harboring differentiated cells from all three germ layers (Fig. 2).
To provide an isogenic control for disease phenotype analysis, SMA iPSCs were transduced with either HIV7/SMN1 or HIV7/Neo control containing only the neomycin-resistance gene, and G418-resistant iPSC colonies were pooled. Ectopic SMN expression on HIV7/SMN1 transduction restored gem body formation whereas HIV7/Neo transduction failed to show gem bodies (Fig. 3A). The transduced cells were subjected to differentiation using a well-established protocol . Like H9 cells, both SMA-23/Neo and SMA-23/SMN cells formed neuron-like cells in culture. Neurite outgrowth was initiated within a few hours after seeding of neurospheres derived from both cell pools. Neurites from SMA-23/SMN cells continued to extend at a similar speed as that from H9 cells, and the average neurite length at day 5 after seeding was also similar (Fig. 3B). In contrast, neurite outgrowth from SMA-23/Neo cells was severely delayed. After 5 days of seeding, the average neurite length from SMA-23/Neo cells was less than half of that from SMA-23/SMN or H9 cells (Fig. 3B). To ensure that this phenotype was not clone-specific, we established another two transduced pools derived from SMA-25 iPSCs. The two pools exhibited a similar phenotype as those derived from SMA-23 cells (Fig. 3B). Thus, SMN deficiency did not seem to interfere with iPSC differentiation but severely restricted neurite outgrowth. This phenotype is similar to the deficit exhibited by cultured SMA motoneurons isolated from mice or patients [14, 25]. Restoration of neurite outgrowth by SMN expression suggested that this deficit was indeed caused by SMN deficiency and not by iPSC clonal variation.
In these neuronal cultures, we detected cells positive for Olig2 and Islet1, two transcription factors important for motoneuron development (Fig. 4A, 4B). Many of those cells also expressed neuron-specific βIII-tubulin. Cells expressing HB9, a postmitotic motoneuron-specific transcription factor, were also present (Fig. 4A, 4B). However, the number of HB9+ cells in the SMA-23/Neo-derived culture was significantly reduced relative to that in the SMA-23/SMN- or H9-derived culture (Fig. 4C).
Thus, SMA iPSCs manifest some disease-associated phenotypes in differentiated neurons in vitro, and ectopic SMN expression rescues these phenotypes. Further characterization of the cellular and functional deficits in motoneurons derived from these iPSCs and a boost in motoneuron production may accelerate the exploration of the underlying mechanisms of SMA pathogenesis (Supporting Information Discussion).
We thank Jun Wu at the Animal Tumor Model Core for assisting teratoma formation, Shengbing Zang for interpreting teratoma tissue slides, and Mariko Lee at the Light Microscopy Imaging Core for microscopy.
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
The authors indicate no potential conflicts of interest.