Stem cells (above) can differentiate into nearly any cell type in the body, making them appealing, albeit challenging, research tools. Credit: Alpha Tauri 3D Graphics/Shutterstock
When Shinya Yamanaka discovered that he could reprogram somatic cells into pluripotent stem cells in 2006, he seemingly set the stage for a new era in disease modelling and drug discovery. Induced pluripotent stem cells (iPSCs) can differentiate into a number of cell types, which could be exceptionally valuable for those developing therapies.
Yamanaka received the Nobel Prize in Physiology or Medicine for his work in 2012, and cell biologists now regularly use iPSCs in the lab. But the impact of iPSCs on drug discovery has been frustratingly modest.
“Everyone thought that stem cells were going to give us the human cell models that we required to make better drugs,” says Mark Kotter, founder and CEO of bit.bio, a cell biology and cell therapy company. “But if you want to do drug screening, you need billions of cells that are extremely consistent every time—and that wasn’t possible.”
Researchers still struggle to consistently differentiate human iPSCs into terminally differentiated cell types, such as neurons and muscle cells. “It’s time consuming, laborious, expensive, and it has low reproducibility,” says Farah Patell-Socha, vice president of research products at bit.bio.
Current protocols take stem cells through every stage of early development to reach the final cell type. The process can take four to more than eight weeks, with oligodendrocyte differentiation stretching up to 170 days. Typically only a small fraction of the initial pluripotent cells develop into the desired cell type. Also, because batches can be highly variable, each one must be validated independently. That takes skilled scientists, costly reagents and time.
Additionally, researchers still lack relevant cell models that are cell-type specific and carry the disease-associated alleles to study many human diseases. “People study Alzheimer’s in mice, but mice never show the full spectrum of Alzheimer’s symptoms, so it’s a flawed model,” says Patell-Socha.
Though the inherent promise of iPSCs for disease research and drug discovery has not changed, the question remains: What will it take to reach it?
Robust and reliable cell models
Whether in manufacturing or drug discovery, scaling a process usually requires moving from bespoke to standardized workflows. Bit.bio has developed a scalable technology platform capable of rapidly producing consistent batches of human cells based on a paradigm in biology called cell reprogramming. Instead of following traditional multi-step protocols driven by chemical cues, reprogramming can differentiate cells by direct activation of transcription factors.
The partnership between bit.bio and Charles River Laboratories, a contract research organisation for the biomedical research and drug development communities, offers access to these cells through their use in target discovery, high-throughput screening (HTS) and drug discovery services.
“We figured out a way to trick stem cells into accepting a new program,” Kotter says.
When reprogramming stem cells, researchers often introduce transcription factors that direct cell differentiation using vectors that randomly integrate into the genome. However, this can interfere with the function of cells and the process of gene silencing, in which a cell suppresses the expression of a certain gene. That can disrupt the production of transcription factors, stalling the process.
Several years ago, Kotter and his colleagues discovered they could bypass the gene silencing mechanism, with a gene targeting strategy called opti-ox (optimised inducible overexpression). The ‘trick’ was to insert transcription factor genes into genomic safe-harbour sites in the stem cell genome; this protects cellular function and once activated, those DNA sequences are rarely, if ever, silenced.
When opti-ox technology is applied to cellular reprogramming, it enables the precise differentiation of entire cultures of stem cells into any cell type at scale. In his early experiments with the approach, Kotter says he was shocked by the high purity and reduced timelines. “We achieved virtually 100% purity within days, which was 10 times faster than anything seen before.”
The execution of the genetic code, engineered into every iPSC, drives the consistent, deterministic induction of the same somatic cells every time, making them useful as a standard in scientific studies and as a scalable tool for drug discovery assays.
“These cell lines are a major advantage for my work because I have hundreds or thousands of drugs that I want to apply to as many cells as I can,” says Mariangela Iovino, group leader of small molecule drug discovery at Charles River Laboratories. She tested bit.bio’s first cell-line product to gauge its performance in drug discovery assays.
“You know what you are getting all the time, and it’s faster,” she says. “Before, it took me five weeks to get mature cells. Now, it only takes one to two weeks.”
A tool for drug discovery
Bit.bio’s aim is to code cells to produce consistent batches of every cell type in the human body that, ultimately, could enable novel medicines for diseases. The company is now focused on building its CNS, immune and muscle ioCells portfolio. Researchers can obtain human iPSC-derived glutamatergic neurons and GABAergic neurons, which can be used to study neurological and psychiatric diseases, such as autism and schizophrenia, and human iPSC-derived skeletal myocytes, which are useful in the study of muscular dystrophy or metabolic disorders.
Scientists at bit.bio have also engineered cells for Huntington’s disease, which is characterised by the progressive degeneration of nerve cells in the brain. Developing cellular disease models for Huntington’s disease has been difficult. “You can isolate neurons from transgenic animal models, but there are limitations to the translatability of findings from a rodent brain to a human brain,” says David Fischer, Chief Technology Officer Early Discovery at Charles River Laboratories. In addition, classical protocols to differentiate Huntington’s disease iPSCs to neurons take very long and can be inconsistent between lines.
The progression and severity of Huntington’s disease is known to be related to an expansion of CAG repeats in the HTT gene. The introduction of a disease-relevant expansion of 50 CAG repeats into a wild type ioGlutamatergic Neurons cell line background creates a reproducible standard model for researchers with a highly characterised isogenic control. “The wild type and the disease model are genetically identical except for the introduction of that CAG repeat, so it’s a perfect control,” Patell-Socha says.
In principle, Fischer says, that same methodology could be applied to any disease where a relevant phenotype can be modelled in vitro, including ALS and Alzheimer’s. Disease-relevant cell models could serve as a platform for screening novel compounds in drug discovery.
“Better in vitro human cell culture models will help us understand the efficacy and safety of potential drugs as early as possible,” Fischer says. “This would reduce the number of animals that we have to use in research, and also improve the success rate of the molecules we test and shorten the timeline from drug concept to patient.”
While that process is still evolving, the new platform represents a step forward in the scalability of iPSC-based cell models—and that alone should bring the research community closer to realizing their full potential.