Making the proteins that living cells cannot make

I'm stoked to publicly share something I have been working on since joining Speculative Technologies: A roadmap for its next coordinated research program called Macromolecular Additive Manufacturing. This technology could lead to new multi-component proteins made from up to one hundred distinct subunits, which would be over a ten-fold increase versus what can be made with living cells.  

But why should anyone care about this? The technology could unlock artificial ‘antibodies’ that engage several targets simultaneously, vaccines made from multiple parts of infectious agents for robust immunity to many pathogens, or rapid prototyping of cell-free factories to synthesize chemicals with spatially organized enzymes.

A single persistent challenge in manufacturing and designing synthetic proteins inspired this program: People making things from protein are beholden to using living cells to make them. There are intrinsic limitations to the sort of proteins that cells can produce, and in turn, this limits the diversity of products that people can create. 

In particular, this program targets multi-component proteins made from multiple distinct protein subunits (i.e. parts) that bind to one another into larger assemblies. Biology is able to create powerful machines made from dozens of distinct parts — such as ribosomes, flagella, and ATP synthases — and people cannot make anything rivaling this complexity.

How are cells bottlenecking our ability to make new and intricate multi-component proteins? New protein assemblies composed from many distinct pieces are metabolically intensive and hard for cells to make in large amounts. Subtle nuances in design of the gene sequences coding for the protein can impact folding and yield such that optimizing production of new products can be unpredictable. There are only a limited range of modifications that cells can add to proteins and downstream chemical processes for adding these non-protein parts are challenging because the proteins are free floating in liquid. Finally, the specialized cells that we use to make proteins like therapeutic antibodies are sensitive to small variations in environmental conditions such as temperature and carbon dioxide. 

Naysayers might say that computational algorithms for designing proteins are the path to new multi-component proteins. However, such tools are bad at modeling proteins that have no precedent in nature because machine learning models are trained on data for proteins that were taken from living cells. Moreover, all the improvements in the world to these computational approaches will be unable to get around the aforementioned limits intrinsic to cellular synthesis.

The consequences of these bottlenecks are profound. For instance, the pharmaceutical industry’s development of protein therapies only focuses on the proteins that can be made with cells. There is the unsaid premise that pursuit of designs beyond what cells can make is futile. This rubs against trends in modern biologics where several protein pieces are mixed and matched into single therapies, such as with the emergence of multispecific antibodies. At the end of the day, biopharma must be able to scalably synthesize the therapy of interest, and this demands limiting their search to the proteins that can be made by cells in large amounts using existing cell-based manufacturing infrastructure. 

This program intends to lay the groundwork for macromolecular additive manufacturing to make the next generation of multi-component proteins. This process entails attaching the first piece to a support that is situated outside of cells, with additional pieces added one at a time to build up the final product. Before that, each of the smaller parts might be made in cells or using any other synthetic technique appropriate for the component at hand. 

Taken together, this program pragmatically merges cellular manufacturing with alternative techniques from chemistry — such as using supports as inspired from solid-phase synthesis — to create otherwise unrealizable multi-component proteins. The roadmap details the approach, why it has not happened yet, why we think it is possible, and what the consequences could be in much more detail. 

We are seeking feedback from the community about macromolecular additive manufacturing and how to support it moving forward. Please reach out to Speculative Technologies if you have ideas about how to make this happen!