The Science Behind LanzaTech’s Success

As I explained in my first overview article in this series, LanzaTech is an industrial chemicals business. But to describe it as merely another commodity chemicals producer misses the point entirely.

At its heart, LanzaTech is a cutting-edge synthetic biology or “SynBio” venture. According to Wikipedia, SynBio is “…a multidisciplinary area of research that seeks to create new biological parts, devices, and systems, or to redesign systems that are already found in nature.”

As a generalist, your correspondent tends to shy away from SynBio companies. A few years ago, I dabbled with one – a company recently in the news due to its September 17 IPO via SPAC, Ginkgo Bioworks. I ended up losing interest in the company because the application of its technology seemed so trivial – at the time I looked at the firm, it was mostly producing perfume fragrances.

(My original assessment of Ginkgo turned out not to be completely accurate, according to this New York Times article. Ginkgo’s biological foundry apparently helped design genes that sped development of Moderna’s Covid-19 vaccine. Now that I understand more about the SynBio process, I may go back and take a closer look at Boston-based Ginkgo.)

Because the work that LanzaTech does – sequestering waste carbon from steel manufacturing and other high carbon footprint activities and creating industrial chemicals with it – is so important to the world in terms of maintaining quality of life in a post-fossil carbon world, I wanted to dig in to understand what was under the hood.

This article describes what is under LanzaTech’s hood in layman’s terms and more generally what kind of work goes on in a SynBio company.

LanzaTech SynBio Breakthroughs

Innovation turns up in some unusual places, but it’s hard to think of a place more unusual than rabbit droppings.

In 1994, scientists described an anaerobic (i.e., cannot survive exposure to oxygen) bacteria (Clostridium autoethanogenum), present in rabbit droppings, that takes in carbon monoxide or carbon dioxide and hydrogen as food and metabolizes them into ethanol in much the same way yeast metabolizes sugars into alcohol.

As molecules go, ethanol is an attractive one. Its backbone is made up of two carbon atoms; dangling off one of those carbons are three hydrogens, dangling off the other are two hydrogens and a hydroxide molecule (a hydrogen atom bound to an oxygen atom).

The careful reader will note that except for that oxygen atom, ethanol is made up of hydrogen and carbon, which are the elements in those enormously useful molecules that oil and gas companies spend so much time and energy digging up – hydrocarbons.

LanzaTech’s co-founder, Dr. Sean Simpson’s vision was two-fold. First, he thought that there should be a way to engineer a process by which huge colonies of C. autoethanogenum might be bred and their activity harnessed to produce commercial quantities of sustainable ethanol solely from above-ground carbon sources.

Second, he thought that if an ethanol-producing system could be designed, it should also be possible to genetically modify bacteria and breed colonies that could sustainably produce other hydrocarbons in commercial quantities as well.

The way in which that first vision has played out will be covered in my next article focusing on LanzaTech’s engineering.

The way in which that second vision has played out is a scientific miracle, involving alterations of the bacteria’s genetic instructions governing its metabolism.

The process of metabolism is carried out through the action of a globular proteins known as enzymes.  Enzymes speed (or “catalyze”) chemical reactions within a cell and otherwise regulate biochemical processes. For almost every reaction within a cell to be carried out, a specific enzyme must act as a catalyst.

The proteins that form enzymes are themselves made up of series of amino acids. Amino acids are created according to biochemical codes stored within an organism’s DNA, then translated and built using RNA.

The first step in LanzaTech’s scientific quest was to understand what enzymes were involved in the ethanol production process and what part of C. autoethanogenum’s DNA encoded them. (This process took around five years of concentrated scientific effort.) Once the team understood how the enzymes worked and in what part of the DNA their chemical composition was encoded, the next step was to figure out how the enzymes could be altered.

Altering the action of enzymes by making changes to an organism’s DNA is what is meant by the terrifying-sounding process of “genetic engineering.” Whereas your correspondent is probably not alone in associating the term genetic engineering with very noticeable morphological changes in organisms (like creating extinct wooly mammoths by modifying the genetic code of elephants), the work that LanzaTech is doing is much more subtle.

LanzaTech’s aim is to coax C. autoethanogenum to create metabolic waste of a slightly different chemical composition – a few more carbon atoms strung together, or carbon atoms strung together in a different arrangement – and to do that predictably at commercial scales.

The words “coding” and “engineering” may bring mechanistic images to your mind but remember that LanzaTech scientists are working with living organisms, not Lego blocks. As such, one big constraint is that the scientists had to figure out how to allow the bacteria to continue feeding and living while simultaneously producing a slightly different waste product.

A huge boost to the field of genetic engineering came about in 2012 with the development of a tool known as CRISPR-Cas9 gene editing. This technology – which rightfully earned its discoverers, Jennifer Doudna and Emmanuelle Charpentier a Nobel Prize in Chemistry – allows a scientist to easily encode strands of DNA and have them inserted into an organism’s genes.

Even with CRISPR-Cas9, though, the process of genetic engineering is hard. Bacteria, while simple organisms, have approximately a gazillion base pairs in each gene, and each part of the gene encodes instructions for some different enzyme to be created.

All these enzymes work together within a cell to produce one or more forms of metabolic waste. Rather than conceiving of this process as a linear assembly line, it is helpful to think of it as global supply chain. Some enzymatic reactions can take place in parallel, and their respective products are assembled later into another chemical; some enzymatic reactions can only take place after other reactions are complete.

If separate processes are not synched up in the correct way, the desired metabolite may not be created, or its creation might not be optimized. Strange things can and do happen when tweaking metabolic processes. For instance, increasing the output of one key enzyme might hamper, rather than improve, the cells’ metabolic effectiveness, so that reaction must be suppressed or slowed to allow another enzyme to maximize the efficiency of its output.

Long story short, in manipulating the genetic code, there are an enormous number of variables that must be taken into account. Because of this complexity and because it takes so long for human scientists to test every possible permutation of enzymatic action, LanzaTech realized it would have to incorporate advances in artificial intelligence into the SynBio process.

Using artificial intelligence to aid in the process of determining metabolic efficiency is given its own terrifying name: “Computational Biology”. Your correspondent is enormously indebted to LanzaTech’s Director of Computation Biology, Dr. Wayne Mitchell, for a private “Genetic Engineering and Computational Biology for Dummies” tutorial one Saturday afternoon.

LanzaTech scientists use AI to find the most likely few dozen genetic sequences to test, then have human scientists create those sequences, splice them into C. autoethanogenum, and breed small colonies to test how the genetic modifications manifest themselves in the bacteria’s metabolic process.

Dr. Mitchell explained this process in terms of the AI used in our phones’ mapping apps. There are numerous routes connecting A to B – the map’s AI routines discard the paths taking the traveler down time-consuming side streets, pointing them at the two or three likely best routes.

LanzaTech’s Bioinformatics group is essentially using AI to identify the few metabolic pathways that are likeliest to yield the best biochemical results in much the same way that your phone’s mapping app choses the street routing that are likely to yield the quickest travel time.

Once a small number of promising pathways are identified using AI, LanzaTech’s scientists begin work with CRISPR-Cas9 to modify DNA sequences to create the mutations that will take the organism’s biochemical processes down the pathways specified. (The next article will look at some of the tools LanzaTech has developed to do this.) Scientists make these modifications, breed small colonies of the modified bacteria, then send them through to another team of experimental scientists and engineers to test how the new strains perform.

The experimentalists are looking to answer a few key questions, such as:

  • Is the new organism producing the target molecule as waste?
  • How quickly can the new organism produce the target molecule?
  • How robust is the new organism to different gaseous inputs?

The testing process is an iterative one. Experimentalists and theorists go back and forth regarding the performance of a particular strain – experimentalists offering statistics related to observed performance, theorists using some AI tools and their own understanding and experience to consider how the observed statistics might be improved by further tweaks of the bacteria’s genetic code.

The end result of this amazing work – which in your correspondent’s mind is roughly as complex and difficult as placing an astronaut on Mars – is that LanzaTech can now modify their humble rabbit poop-dwelling bacteria to produce around 50 different chemical compounds using not much more than greenhouse gases for “food”!

However, no matter how impressive this feat is in terms of pure research, it is all just worth as much as a blue ribbon on a science fair project unless the process can be commercialized.

The good news is that LanzaTech worked simultaneously on the scientific research and on the engineering needed to operationalize it. With scientists and engineers working in parallel, the breakthroughs could easily be pulled from the lab and placed in the field. It is precisely this process of cutting-edge engineering that this series’ next article will cover.

Follow me on Forbes or LinkedIn to be notified when the engineering article is published!

Dr. Simpson and Dr. Mitchell were years ahead of me in seeing that civilization had to create new industrial paradigms if it were to survive in a post-climate change world. These brilliant people and their gifted and dedicated LanzaTech colleagues have created a new paradigm and are moving at full speed with impressive partners to make the new paradigm ubiquitous.

Intelligent investors take note.

Read more here: Source link