Scientists use human stem cell models to target deadly brain cancer

Malignant brain cancer is a devastating disease and it’s estimated that more than 16,000 patients will die of it this year. One of the most aggressive forms of brain cancer is gliomas, which originate from the support cells in the brain or spine that keep nerve cells happy and functioning. Unfortunately, there is no cure for gliomas and common treatments involving surgery, radiation and chemotherapy are not effective in fully eradicating these tumors.

3e2c6-glioma

Brain CT scan of human glioma.

In hopes of finding a cure, scientists have turned to animal models and human cell models derived from tumor biopsies or fetal tissue, to gain understanding of how gliomas form and what makes these type of tumors so deadly and resistant to normal cancer treatments.  These models have their limitations, and scientists continue to develop more relevant models in hopes of identifying new potential treatments for brain cancer.

Speaking of which, a CIRM-funded research team from the Salk Institute recently reported a new human stem cell-based model for studying gliomas in Nature Communications. The team figured out how to transform human induced pluripotent stem cells (iPS cells) into glioma tumor-initiating cells (GTICs) that they used to model how gliomas develop and to screen for drugs that specifically target this deadly form of cancer.

Making the Model

One theory for how gliomas form is that neural progenitor cells (brain stem cells) can transform and take on new properties that turn them into glioma tumor-initiating cells or GTICs, which are a subpopulation of cancer stem cells that are really good at staying alive and reproducing themselves into nasty tumors.

The Salk team created a stem cell model for glioma by generating GTICs in a dish from human iPS cells. They genetically manipulated brain progenitor cells (which they called induced neural progenitor cells or iNPCs) derived from human iPS cells to look and behave like GTICs. Building off of previous studies reporting that a majority of human gliomas have genetic mutations in the p53 and Src-family kinase (SFK) genes, they developed different iNPC lines that either turned off expression of p53, a potent tumor suppressor, or that ramped up expression of SFKs, whose abnormal expression are associated with tumor expansion.

The team then compared the transformed iNPC lines to primary GTICs isolated from human glioma tissue. They found that the transformed iNPCs shared many similar characteristics to primary GTICs including the surface markers they expressed, the genes they expressed, and their metabolic profiles.

Their final test of their stem cell model determined whether transformed iNPCs could make gliomas in an animal model. They transplanted normal and transformed iNPC lines into the brains of mice and saw aggressive tumors develop only in mice that received transformed cells. When they dissected the gliomas, they found a mixture of GTICs, more mature brain cells produced from GTICs, and areas of dead cells. This cellular makeup was very similar to that of advanced grade IV primary glioblastomas.

Screening for drugs that target glioma initiating cells

Now comes the applied part of this study. After developing a new and relevant stem cell model for glioma, the team screened their transformed iNPC lines with a panel of 101 FDA-approved anti-cancer drugs to see if any of them were effective at stopping the growth and expansion of GTICs. They identified three compounds that were able to target and kill both transformed iNPCs and primary GTICs in a dish. They also tested these compounds on living brain slices that were injected with GTICs to form tumors and saw that the drugs worked well at reducing tumor size.

The authors concluded that their transformed iNPCs are appropriate for modeling certain features of how GTICs develop into adult gliomas. Their hope is that this model will be useful for developing new targeted therapies for aggressive forms of brain cancer.

“Our results highlight the potential of hiPSCs for studying human tumourigenesis. Similar to conventional disease modeling strategies based on the use of hiPSCs, the establishment of hiPSC cancer models might facilitate the future development of novel therapeutics.”


Related Links:

Four Challenges to Making the Best Stem Cell Models for Brain Diseases

Neurological diseases are complicated. A single genetic mutation causes some, while multiple genetic and environmental factors cause others. Also, within a single neurological disease, patients can experience varying symptoms and degrees of disease severity.

And you can’t just open up the brain and poke around to see what’s causing the problem in living patients. It’s also hard to predict when someone is going to get sick until it’s already too late.

To combat these obstacles, scientists are creating clinically relevant human stem cells in the lab to capture the development of brain diseases and the differences in their severity. However, how to generate the best and most useful stem cell “models” of disease is a pressing question facing the field.

Current state of stem cell models for brain diseases

Cold Spring Harbor Lab, Hillside Campus, Location: Cold Spring Harbor, New York, Architect: Centerbrook Architects

Cold Spring Harbor Lab, Hillside Campus, Location: Cold Spring Harbor, New York, Architect: Centerbrook Architects

A group of expert stem cell scientists met earlier this year at Cold Spring Harbor in New York to discuss the current state and challenges facing the development of stem cell-based models for neurological diseases. The meeting highlighted case studies of recent advances in using patient-specific human induced pluripotent stem cells (iPS cells) to model a breadth of neurological and psychiatric diseases causes and patient symptoms aren’t fully represented in existing human cell models and mouse models.

The point of the meeting was to identify what stem cell models have been developed thus far, how successful or lacking they are, and what needs to be improved to generate models that truly mimic human brain diseases. For a full summary of what was discussed, you can read a Meeting Report about the conference in Stem Cell Reports.

What needs to be done

After reading the report, it was clear that scientists need to address four major issues before the field of patient-specific stem cell modeling for brain disorders can advance to therapeutic and clinical applications.

1. Define the different states of brain cells: The authors of the report emphasized that there needs to be a consensus on defining different cell states in the brain. For instance, in this blog we frequently refer to pluripotent stem cells and neural (brain) stem cells as a single type of cell. But in reality, both pluripotent and brain stem cells have different states, which are reflected by their ability to turn into different types of cells and activate a different set of genes. The question the authors raised was what starting cell types should be used to model specific brain disorders and how do we make them from iPS cells in a reproducible and efficient fashion?

2. Make stem cell models more complex: The second point was that iPS cell-based models need to get with the times. Just like how most action-packed or animated movies come in 3D IMAX, stem cell models also need to go 3D. The brain is comprised of an integrated network of neurons and glial support cells, and this complex environment can’t be replicated on the flat surface of a petri dish.

Advances in generating organoids (which are mini organs made from iPS cells that develop similar structures and cell types to the actual organ) look promising for modeling brain disease, but the authors admit that it’s far from a perfect science. Currently, organoids are most useful for modeling brain development and diseases like microencephaly, which occurs in infants and is caused by abnormal brain development before or after birth. For more complex neurological diseases, organoid technology hasn’t progressed to the point of providing consistent or accurate modeling.

The authors concluded:

“A next step for human iPS cell-based models of brain disorders will be building neural complexity in vitro, incorporating cell types and 3D organization to achieve network- and circuit-level structures. As the level of cellular complexity increases, new dimensions of modeling will emerge, and modeling neurological diseases that have a more complex etiology will be accessible.”

3. Address current issues in stem cell modeling: The third issue mentioned was that of human mosaicism. If you think that all the cells in your body have the same genetic blue print, then you’re wrong. The authors pointed out that as many as 30% of your skin cells have differences in their DNA structure or DNA sequences. Remember that iPS cell lines are derived from a single patient skin or other cell, so the problem is that studies might need to develop multiple iPS cell lines to truly model the disease.

Additionally, some brain diseases are caused by epigenetic factors, which modify the structure of your DNA rather than the genetic sequence itself. These changes can turn genes on and off, and they are unfortunately hard to reproduce accurately when reprogramming iPS cells from patient adult cells.

4. Improve stem cell models for drug discovery: Lastly, the authors addressed the use of iPS cell-based modeling for drug discovery. Currently, different strategies are being employed by academia and industry, both with their pros and cons.

Industry is pursuing high throughput screening of large drug libraries against known disease targets using industry standard stem cell lines. In contrast, academics are pursuing candidate drug screening on a much smaller scale but using more relevant, patient specific stem cell models.

The authors point out that, “a major goal in the still nascent human stem cell field is to utilize improved cell-based assays in the service of small-molecule therapeutics discovery and virtual early-phase clinical trials.”

While in the past, the paths that academia and industry have taken to reach this goal were different, the authors predict a convergence between the paths:

“Now, research strategies are converging, and both types of researchers are moving toward human iPS cell-based screening platforms, drifting toward a hybrid model… New collaborations between academic and pharma researchers promise a future of parallel screening for both targets and phenotypes.”

Conclusions and Looking to the Future

This meeting successfully described the current landscape of iPS cell-based disease modeling for brain disorders and laid out a roadmap for advancing these stem cell models to a stage where they are more effective for understanding the mechanisms behind disease and for therapeutic screening.

I agree with the authors conclusion that:

“Moving forward, a critical application of human iPS cell-based studies will be in providing a platform for defining the cellular, molecular, and genetic mechanisms of disease risk, which will be an essential first step toward target discovery.”

My favorite points in the report were about the need for more collaboration between academia and industry and also the push for reproducibility of these iPS cell models. Ultimately, the goal is to understand what causes neurological disease, and what drugs or stem cell therapies can be used to cure them. While iPS cell models for brain diseases still have a way to go before being more clinically relevant, they will surely play a prominent role in attaining this goal.

Meeting Attendees

Meeting Attendees