Mammalian synthetic transcriptional regulation

Synthetic biology circuits in mammalian cells for health related applications

Mammalian synthetic biology tools, platforms and applications

Ron Weiss, MIT

What can we do with gp-write technology?

Maybe cancer-resistant cells. Can we create customized immune cells? Can we create infection-resistant cells? Those types of cells really are- they have multicellular interactions, but they are individual cells. Another level is this notion of coherent tissues in being able to benefit from this gp-write project. Some of the things we're interested in pursuing includes programmable organoids and beginning to think about renewable anti-aging tissues, or modifiable genomes on the fly to create new functions or capabilities potentially in vivo.

Programmable organoids

This is he notion that we can take iPSCs and reprogram them, edit them, put new genetic circuits in them, proliferate them, and then make 2d or 3d patterns by design and use that as a mechanism to create desired tissue, either mimicking normal developmental pathways or maybe trying to create a variance of those. So you need precise spatial, temporal and cell type control.

Good for disease modeling and organ transplantation.

Here's a quick example. We can take iPSC cells, engineer them, and then make Gata6 at precise levels in this population. Low gata6, mid, and high. The low gata6 becomes becoming ectoderm and high is mesoderm and medium cells become either ectoderm, mesoderm or other based on neighbors. So this initial population of cells ends up differentiating and doing symmetry breaking. You should be able to derive any cell type from endoderm or ectoderm or mesoderm cell layers. Wet let the system develop and we got a large amount of cell types.

We got a liver bud next to a brain, and the liver bud had liver functionality.

We took this 2d environment and moved it to a 3d situation where we grew these organoids in something called hanging draw. hiPS/gata6 derived liver organoids, grew for 25. months. They are highly deeply vascularized. A lot of nutrients can peentrate this entire organoid, that's on the bottom left there, because of these nutrients we can reach all the cells. The organoids became huge-- we're calling them meganoids. They are large enough to play hacky-sac with or something. They are 100x larger in terms of other organoids.

You can begin to think about disease, modeling and transplantation.

We'd like to move forward on programmable organoids to enhance those organoids and think about the notion of renewable anti-aging tissue and rewritable genomes for organoids.

We would like to have new genetic variants, we'd like to be able to create mechanisms to sense everything there genetically, and create different 3d shapes, and being able to dynamically rewrite the genome so do this in an efficient way but also to actual tissues, so that doesn't allow us to use existing technologies for reconfiguring what the human genome is within each of these cells. We want to add/remove functionality as these tissues are operating.

We need synthetic biology building blocks. We need dynamic rewrite of the genome.

I'm not going to go into the detail of these things. We have made some progress: Systematic transfer of prokaryotcix sensors and circuits to mammalian cells (ACS Synthetic Biology). We have been able to create sensors for many of the things in cells, like intracellular mammalian sensors, like protein sensors and miRNA sensors, a Universal RNA-based logic evaluator that operates in mammalian cells (Nature BIotechnology).

Also be able to engineer cell-cell interactions either through long range cell communication-- phloretin which diffuses quickly, or synNotch for short range. We also have cadherins for physical interaction. We are programming cells to interact with each other and then create these 3d shapes hopefully leading to tissues.

What we really need is this notion of in situ or in vivo genome dynamic write platform. I think this builds up on the "landing pad" concept from earlier today, oligos to change small places in the genome. We've been able to create multi landing pad cell lines that have 3 months of expression. We'd like to have 100s of landing pads in the genome to integrate very large fragments. We've been able to generate 60 kb into each of our landing pads. We can introduce multiple DNA at a time.

BioFPGA-- reconfigurable human genome. Like FPGAs. The genome already allows for reconfiguration; then we provide small signals like small oligos which we can provide to the tissue and then those signals will direct the rewrite in the genome itself, essentially rewiring the genomes on the fly to allow us to reprogram the function on the fly. We have some ideas about bioFPGAs but they are preliminary.


Q: I am interested in the idea of dynamically rewriting the geonme as you go along. Cells seem to do that with chromatin rather than editing DNA bases. Why go down the DNA bases route?

A: Chromatin is certainly one of them. One of the ways you can turn things on and off. The question I have though is, I'm not sure chromatin editing allows you to basically rewire those signals in new ways. I'd like to have genetic circuits where I can connect promoters to new genes that will just by providing these signals.. I'm not sure there's a way to do that with just silencing specific portions of the genome. I'm not sure that's possible. You can think about this exponential search space, to associate any promoter with any gene, yeah I'd like to be able to do that.

Q: Ron, so, like you, we have been very focused on landing pads and so on. But HACs are also a real probability. Have you looked at them? What has been your experience and compare contrast?

A: HACs are very difficult to work with. We tried to do that with early tech and they were not very stable. There's some exciting new tech like presented yesterday here. I'm very excited about what she might be able to do with her HACs. I'd love to use HACs.