August 3, 2020

The Niche

Knoepfler lab stem cell blog

Horwitz Interview Zooms in on Allen Cell Institute: the GoogleMaps of Cells

Allen Institute for Cell Science
Allen Institute for Cell Science

One of the most exciting developments in the biomedical sciences in 2014 was the announcement of the new Allen Institute for Cell Science. This new institute, which will be focused on cell biology, received $100 million in funding from high-tech pioneer and philanthropist, Paul Allen. I am very curious about how the institute will tackle cell biology from a new angle so I invited the inaugural Director, Dr. Alan Rick Horwitz, to do an interview and he kindly agreed.

Paul: It sounds as though a major area of focus for the Institute will be on stem cells including IPSCs. Can you talk a bit about why?

Rick Horwitz
Rick Horwitz of Allen Institute for Cell Science

Horwitz: There are a lot of possible good choices. One could work with yeast or with mice. Some colleagues say that when you take cells out of the organism there can be changes. However, human genetics is quite advanced, when compared to the mouse, and we know a lot about human genetics, especially as it pertains to disease. While we are not a translational institute, but rather a basic science institute, with IPSC there’s an opportunity to inform to disease.

The gene editing that we plan to do with the IPSC lines gives us the potential to differentiate them into a variety of cells. There is also potential they will be useful for others in their research. The cells will be available for the community. These were major factors in the choice.

Paul: What method to make IPSCs? Any work on ES cells?

Horwitz: We won’t make the cells ourselves. We will use probably 2-5 of the hIPSC lines openly available. There will probably be no ES cell work.

Paul: There’s a ton of data out there on IPSCs so what will be your different angle on this?

Horwitz: Few are really looking at iPS cells from a cell biologic point of view and none as an integrative system. If we perturb a particular molecule, we’ll take an integrative and systems view. That will be unique. That’s our space. It’s technically very large, however. There are a large number and spectrum of well characterized and robust markers of biological activities and molecular machines. For example, as the cells divide, how many mitochondria are there and when and where are they active? Same thing of ribosomes. Where are the ribosomes versus the mitochondria and the actin filaments? We’ll ask ‘How does this system change with perturbations?’ A huge component is computational. The workflow will be as follows. Gene edit the stem cell lines, making hundreds of them and study them visually as they differentiate into cardiac myocytes, epithelia, myeloid cells, etc. and they change environments, as they respond to perturbations and natural variations, These data will then be used for computational models

The cell biological problem is localized transient events. That information is what we have to capture. That’s our goal. And then to determine how does the whole system respond?

Paul: Imaging and microscopy have been mentioned frequently in the pieces on the Institute. Will imaging be the main workhorse technology for the Institute? Do you plan for substantial roles for other methods at the Institute as well such as molecular biology or ‘omics technologies such as looking at the epigenome?

Horwitz: All imaging. We won’t do ‘omics. We will look at gene expression to relate that to the behavior, however. We hope to synergize with others in doing that kind of work. Proteomic networks and analyses will greatly inform our imaging.

Paul: You’ve compared what the Institute will do to a “Google Maps” of cells. That’s a cool analogy. When cell biologists like me go to the data generated by the institute what will we see? Live cell imaging? Markers? Cytoplasmic? Cell Surface? Nuclear/chromatin?

Horwitz: If you look at the genome, it’s like the white pages of the phone book. What the genes do is the yellow pages. If you want to understand how San Francisco works, it’d be hard to figure out by looking at the phone book. Compare that to Google maps and integrate the map information with census data, and make it dynamic; suddenly you see how SF works, that is what it does and how it does it.

Zoom in on the cell and see a ribosome and zoom in more and see molecules, etc. The goal is to create an enormous multiscale visual database.

Paul: Specific diseases or tissues as focus?

Horwitz: Cardiomyocytes to start with, its derivation from hiPSCs is reproducible and well-characterized. That’s where we’ll start. And also, one kind of epithelial cell.

Cardiomyopathies, cancer perturbances are on the radar already, but not for initial studies.

Paul: What will be the role of gene editing via CRISPR-Cas9? Can you give a specific example?

Horwitz: Fluorescent reporters at first. We want to mark a significant group of molecular machines, both in the IPS cells and in differentiated derivatives. Key mutations for disease models as well. A challenge is to prioritize which molecules to focus on. We’ve started talking to experts in the community – what are the best markers and molecules, that are expressed well, the GFP tag won’t perturb function, etc. Community involvement is critical here.

Paul: Looking to the future, where do you see the Institute in say 5 years? Do you see local interactions and collaborations in the Seattle area?

Horwitz: At the end of 5 years I’d like to have a big library of gene edited cells. We’d like to have some predictive models that work well. We want to be creating a significant system wide library as cells do a few core things/activities and molecular machines. Finally, we would like a significant database of images showing the locations of cellular components and activities, and the start of a meaningful animated output to display them.

There are highly synergistic activities at UW and elsewhere in town.

Paul: What excites you most about cells and Cell Biology?

Horwitz: Systems biology and predictive models of cellular behavior.