Technology Review Article & Attendant FAQ

An article covering my work was published today in the online edition of MIT Technology Review Magazine. It was written by Antonio Regalado, and he was very thorough in his research; nobody else to date has asked for scanned documentation to back up my claim of winning a Class 1 License from the EPA!

The article was well written and had a great narrative flow, but some readers may be left with questions on the “crunchy bits”: details that don't make a great story, but might satisfy a sufficiently piqued interest. So, here's my attempt at an FAQ.

“In a spare bedroom of his mother's house in County Cork”

Yep. To clarify, it's my family home, not just my mother's. I don't live at home anymore; I rent a house several miles away with my wife, Niamh. However, the EPA license covers a location as much as an individual, so I wanted to avoid having to re-license whenever we move. Also, there was a spare room in the family home!

“Today, you can [transform E.coli] with Epsom salt and a […] brand of laxatives."

Yep, verified by yours truly. I've got a how-to written up which I'll be posting on this site in a little while. It's a far easier, more accessible method than the traditional “Calcium chloride and heat-shock” method, and it doesn't even require a centrifuge. It's based upon the protocol originally presented by Chung, Niemela and Miller, but it foregoes the DMSO, as it only offered marginal gains in transformation efficiency and is far harder to find outside a lab.

Specifically, I used local pharmacy-brand epsom salts and “Miralax” brand PEG-3350. Other brands of PEG-3350 should be available almost anywhere.

“Isolating [..] bioluminescent bacteria from squid [..] from a Cork fishmonger”

I have an old, badly written guide to this experiment here, on my personal blog. I generally don't advise people to replicate the experiment for safety/liability reasons, as there is some evidence to suggest that the V.phosphoreum bacteria produce a toxin. Quantities of the toxin and actual hazards presented by it are unknown to me. Anyone who does replicate that work is advised to take safety precautions.

“Changing bacteria […] may present a risk to the general public”

Without pulling the gloves off entirely, I think this comment deserves some attention. In the research lab I used to work in, there were people working with viable human viruses to create anti-cancer viral therapies. That's cutting edge stuff, and the potential is huge. But, there were no notices, policies or even discussions about what might not be appropriate to engineer into a human virus. That was left to the common sense of the people in the lab.

That's not to say that I foresaw or foresee any risks in the work conducted there, but it's disingenuous to turn around and point the “scare quotes” at someone else who's working with Class 1 organisms only, and who routinely warns against working with human cell lines at home.

Virtually every lab that works with DNA, whether for research or development reasons, uses engineered E.coli cells to “keep” the DNA in between revisions. The lab strains of E.coli that are used, and which I use, are so genetically incompetent that simply using them counts as “containment”. The same goes for B.subtilis 168, the strain I am working with at the moment. There is no risk of “accidentally” creating a dangerous pathogen here, and even a dedicated, trained and malign person would be very hard pressed to do a better job than nature at making dangerous pathogens.

Moving on.

“Germs that pose “negligible risk” to the public or the environment”

Here's an interesting difference between the US and EU approach to biosafety: In the USA, there's a “select list” of organisms whose DNA should be off-limits to people outside special containment labs. So, if I order DNA encoding a protein normally found in, say Y.pestis, the causative agent behind the bubonic plague, the synthesis company would be well advised to refuse or face legal liability. Not to mention that they'd better know why I'm ordering Y.pestis DNA in the first place.

In Europe, however, or at least in Ireland, there's no off-limits DNA per se. Rather, a holistic risk assessment must be conducted for every new project, and it must consider the hazards posed by the “donor strain” from which the DNA is derived, the “recipient strain” into which the DNA will be, er, hacked, and the “resultant GMM”. All three are considered for environmental and medical safety implications.

This is important because most genes in Y.pestis are entirely innocuous. Y.pestis’ DNA mostly codes for routine stuff that all bacteria do: polymerases for copying DNA and transcribing mRNA; functional RNAs to translate that mRNA into protein; pathways for metabolism, cell growth and division. In fact, I'd wager that far less than a hundredth of the functional code of Y.pestis encodes anything of pathogenic relevance.

Of course, why bother with a potentially hazardous gene when I can get a far more well-known alternative from harmless bugs? Because sometimes the interesting trait you want is only found in species X. For example, Y.pestis might encode a restriction enzyme that cuts DNA at a certain place, and there might be no alternatives. Restriction enzymes don't play any role in pathogenicity to humans; they are a selfish genetic system, like a gene-parasite of the bacteria, though they may also help protect the bacteria from bacteriophages. In the USA, I could not get the DNA for this genetic system legally, but in the EU I could, because the risk assessment procedure covers the resulting GMM more than the donor strain. If I wanted a gene encoding a toxin, that'd be a no-no, but a polymerase or restriction enzyme is OK.

..Don't worry, I'm still not planning to work with Y.pestis DNA.

"E. coli […] isn't so easy to work with."

To put that in context, it's not just the smell. It's got more to do with the broth media E.coli likes to grow in (which require pre-digestion with enzymes, as covered in another of my previous guides), the storage conditions (-80C freezers! -20C can be done, with difficulty), transport considerations (cells can't be easily dried out, so fragile tubes of liquid or agar media must be sent through letter-post) and ease of genetic manipulation (even with simple miralax/epsom salts methods).

To put it simply, E.coli is a darling of labs with the money and equipment to feed, store and transport a slightly awkward bacteria. There are solid reasons to love E.coli, and I still do, but for practical reasons I think B.subtilis is a winner for home-labs.

B.subtilis grows on potato-broth, which is about as easy to make as bacterial media get. It forms long-lived spores that can be stored at room temperature or can be easily spotted onto paper for postage. And it's “naturally competent”, meaning that in certain phases of growth it should be able to easily absorb and adopt DNA. That's something I'll be getting into this week, now that I've formally commenced work and notified the EPA of same.

One argument in favour of E.coli right now is the huge library of DNA available from various labs that work out-of-the-box in E.coli, the fruits of decades of genetic research. Yes, that DNA exists, and it won't work as-is with B.subtilis, but that's due to change.

For starters, amateurs and even some academic labs can't access the E.coli DNA anyway due to transfer-agreement licensing, gene patenting and policies preventing exchange of genetic material with anyone, let alone individuals.

Secondly, the cost of DNA synthesis is pretty soon going to make it easier/cheaper to re-compile DNA than ask for some from another lab, and at that point it'd be easy to recompile the whole library for another species such as B.subtilis.

“Garvey calls [it] “Indie Biotech Backbone 1.0,” and he plans to share it”

Yes. And also, “kinda”. I actually plan to sell it, because this is my career; making DNA and protein products to fill a community need. However, the DNA will be provided under an “Open-Source” MTA, conforming to the definition of “Open Source” presented by

Why an MTA? Because unlike text, which automatically inherits stupidly overpowered (for those big enough to pay) legal protection under Copyright, which can be translated to a more modern Creative-Commons License with a little notice, DNA has no such inherent protection. To protect DNA so that you can enforce open-source practices, an MTA makes sense; it's an established legal platform, it costs nothing to tack on, and it can include provisions like those called for by a “share-alike” license.

What I'd envision is an MTA that says, in plain English:

  1. This DNA has been sold to you, and it belongs to you. Beyond what's agreed here, I hold no further right to shape how you use, enjoy or adapt this DNA.
  2. You may share, modify, adapt, use or sell this DNA, provided you require agreement to this license from your customers or recipients of your derivative or re-shared DNA.
  3. You must also include the annotated base-pair code of the DNA, whether modified or not, with any works shared or sold to others. You may substitute with a link to the code rather than a printout, but the code must be downloadable in a standard format that can be easily used and adapted; FASTA or Genbank are recommended.
  4. You may not patent or otherwise legally encumber works or products derived from this DNA.

So, I'm planning to “sell” the DNA. But I want to sell it in a format that ensures community empowerment for DIYbioers, and encourages a forward-sharing platform for DNA engineering. World problems don't get solved by patents or secrecy, but by iterative innovation and free sharing of ideas.

 Share, if you like. Fediverse sharing is preferred, though.