Induced pluripotent stem cells (iPSCs) are made by taking ordinary adult cells, like skin or blood cells, and introducing a set of four genes that rewind those cells back to a flexible, embryonic-like state. The entire process, from collecting a donor’s cells to confirming the final product, typically takes several weeks to a few months. What sounds simple in concept involves careful choices at every stage: which cells to start with, how to deliver the reprogramming genes, and how to verify that the result is safe and truly pluripotent.
The Four Genes That Reset a Cell
In 2006, Shinya Yamanaka’s lab discovered that just four transcription factors could reprogram a mature mouse cell into one that behaves like an embryonic stem cell. Those factors, Oct4, Sox2, Klf4, and c-Myc (often abbreviated OSKM), work together to activate the genes responsible for pluripotency, the ability to become virtually any cell type in the body. The same combination works in human cells. When all four are active inside a somatic cell, they gradually silence the cell’s specialized identity and switch on a developmental program that returns it to a blank-slate state.
Each factor plays a distinct role. Oct4 and Sox2 are master regulators of the pluripotent state, directly turning on the gene networks that define stem cells. Klf4 helps suppress genes tied to the cell’s original identity. c-Myc accelerates cell division and opens up tightly packed DNA so the other factors can access it. Some protocols swap or add factors (Lin28 is a common addition), but OSKM remains the standard combination.
Where the Starting Cells Come From
Fibroblasts, the cells found in skin and connective tissue, were the first cell type reprogrammed and remain the most commonly used source. A small skin biopsy provides enough fibroblasts to begin. But skin biopsies are mildly invasive, so researchers have developed protocols using cells that are easier to collect.
Blood is now one of the most practical starting materials. A standard draw of 2 to 4 milliliters of venous blood yields peripheral blood mononuclear cells that can be reprogrammed. Umbilical cord blood and even a finger prick’s worth of capillary blood also work. Beyond blood, kidney epithelial cells naturally shed into urine every day and can be collected noninvasively, and keratinocytes can be harvested from plucked hair follicles. More unusual sources include pancreatic beta cells, cells from the lining of joints, and mesenchymal cells from extracted wisdom teeth.
The choice of starting cell matters. Different cell types reprogram at different rates and efficiencies, and some carry epigenetic baggage from their original tissue that can linger in the final iPSC line.
How the Reprogramming Factors Get Inside
Getting four genes into a cell and having them work long enough to trigger reprogramming is one of the trickiest parts of iPSC production. Several delivery methods exist, each with trade-offs between efficiency and safety.
Viral Vectors
Yamanaka’s original method used retroviruses to carry the four genes into fibroblasts. Retroviruses are efficient because they insert their genetic cargo directly into the cell’s DNA, ensuring the reprogramming factors are expressed. The downside is permanent: those viral sequences stay in the genome, which raises the risk of disrupting important genes or reactivating later to drive tumor formation. Lentiviral vectors work similarly and share the same integration risk.
Sendai virus (SeV) offers a middle ground. It delivers genes into cells effectively but replicates in the cytoplasm without ever touching the cell’s DNA. In a large comparative study, Sendai virus achieved a reprogramming success rate of 92%, compared to 38.5% for a common non-viral alternative. The viral RNA gradually dilutes out as cells divide, leaving no trace in the genome after several passages.
Non-Viral, Integration-Free Methods
For clinical applications, the goal is a “footprint-free” iPSC line with no foreign DNA or leftover reprogramming molecules in the genome. Several strategies achieve this.
Episomal vectors, derived from Epstein-Barr virus but stripped of viral packaging, deliver the reprogramming genes as small DNA circles that sit outside the chromosomes. They work long enough to trigger reprogramming and are then lost during cell division. PCR analysis has confirmed no genomic integration in late-passage iPSC lines made this way, though efficiency is comparatively low. In fibroblasts specifically, episomal reprogramming succeeds about 27% of the time, well below viral methods.
Synthetic messenger RNA (mRNA) transfection is perhaps the cleanest approach. Researchers introduce lab-made mRNA encoding the four factors (sometimes with Lin28 added). The cell’s own machinery reads the mRNA and produces the reprogramming proteins, but mRNA degrades within about a day, leaving absolutely nothing behind in the genome. Early versions caused significant cell death because cells treated foreign RNA as a threat, but substituting modified versions of two RNA building blocks largely solved that problem. Reprogramming efficiency with mRNA reaches above 2% of treated cells, which is lower in raw percentage but produces high-quality, footprint-free lines. Karyotyping and genome-wide scans of mRNA-reprogrammed iPSCs have confirmed normal chromosomes with no detectable gains or losses compared to the donor cells.
Electroporation is commonly used to get non-viral cargo into cells. A brief electrical pulse opens temporary pores in the cell membrane, allowing episomal vectors or other molecules to slip inside.
The Production Timeline
A typical protocol using blood cells and electroporation follows a well-defined schedule. On day one, blood is drawn and the mononuclear cells are isolated over a few hours. These cells are then electroporated with the reprogramming factors and seeded onto coated culture plates in specialized reprogramming medium. This moment is designated Day 0.
Over the first ten days, cells are periodically moved to fresh plates, and the culture medium is gradually transitioned to one that supports pluripotent stem cell growth. By day 11, the medium switches entirely to stem cell maintenance formula, with daily changes from that point on. iPSC colonies, recognizable as tightly packed clusters with distinct borders, typically appear 10 to 20 days after electroporation.
Once colonies emerge, individual clones are picked and expanded. This selection step is critical: not every colony that forms is fully reprogrammed, so researchers look for colonies with smooth edges and characteristic round morphology. The selected clones are then grown across multiple plates until there are enough cells for thorough testing, which requires at least six full culture plates’ worth of cells.
Confirming the Cells Are Truly Pluripotent
Making cells that look like stem cells isn’t enough. A battery of tests confirms that reprogramming was complete and the resulting cells are genuinely pluripotent.
Alkaline phosphatase staining is one of the first checks. This enzyme is highly active in undifferentiated pluripotent cells and serves as a quick visual indicator. It’s used as a standard identity assay in cell banks maintained under manufacturing-grade conditions. Immunofluorescence staining then confirms that key pluripotency proteins, particularly Oct4, Nanog, and Sox2, are being produced by the cells’ own genes rather than leftover reprogramming factors.
The gold standard for proving pluripotency is the teratoma assay. iPSCs are injected under the skin of immunodeficient mice, and over 8 to 12 weeks, the cells form teratomas: benign tumors containing a jumble of tissues from all three embryonic germ layers (ectoderm, mesoderm, and endoderm). Finding nerve tissue, cartilage, and gut-like lining in the same tumor proves the cells can become virtually any tissue type. Newer computational alternatives like TeratoScore and PluriTest can assess pluripotency from gene expression data without requiring animal experiments.
Genomic Safety Checks
Reprogramming puts cells under stress, and that stress can introduce genetic errors. iPSC lines gain an average of 0.8 to 1.2 single-nucleotide mutations per cell per passage, roughly one-tenth the mutation rate seen in normal somatic cells during division. That’s reassuringly low, but certain types of damage still require screening.
Karyotyping checks for large-scale chromosomal abnormalities like extra or missing chromosomes. Genome-wide SNP arrays detect smaller copy-number changes, deletions or duplications, that karyotyping might miss. For clinical-grade lines, both tests are performed and compared against the original donor cells to confirm nothing was gained or lost during reprogramming.
One underappreciated risk involves mitochondrial DNA. Mutations that accumulated in the donor’s somatic cells over their lifetime get carried into the iPSC line and can actually be amplified during reprogramming. These mutations may affect cellular energy production and, more concerning for transplantation, could encode abnormal proteins that the immune system recognizes as foreign.
Scaling Up for Clinical Use
Making iPSCs in a research lab is one thing. Producing the billions of cells needed for a single patient’s therapy is another challenge entirely. Traditional flat culture plates can’t generate enough cells consistently, and manual handling introduces variability between batches.
Suspension bioreactors offer a path forward. Vertical-Wheel bioreactors, for example, create a low-shear, three-dimensional environment where stem cells grow as floating clusters rather than flat sheets. In testing, scaling from a 0.1-liter to a 0.5-liter reactor produced a 12-fold increase in total cell mass, significantly more than the 5-fold increase in vessel size would predict. A single 0.5-liter reactor generated roughly 183,000 islet equivalents from iPSC-derived cells in one run.
Still, the numbers needed for some therapies are staggering. Treating a single diabetes patient, for instance, requires approximately 7,000 to 12,000 islet equivalents per kilogram of body weight, translating to roughly a billion specialized cells for full insulin independence. That target remains beyond what any single platform can reliably deliver today, though closed-circuit bioreactor systems designed for clinical manufacturing conditions are narrowing the gap.