Breaking Barriers To Engineering Recalcitrant Organisms

[SOFIA] Okay, quick thought experiment. You've got a bacterium living in a hot spring, or eating cellulose, or fixing carbon straight from sunlight. It does something you desperately want — and you cannot, for the life of you, get a single piece of DNA into it. That wall, that's the whole story today.

[DANIEL] It's the unglamorous bottleneck of the entire field. Everybody wants to engineer the exotic organism. Almost nobody can, because step one — get your DNA inside the cell so it actually replicates or integrates — fails silently.

[SOFIA] And I want to be precise, because we've got a PhD audience but they're not all microbiologists. When we say "getting DNA in," for bacteria and archaea that's transformation — you're pushing naked DNA across the envelope. Or conjugation, cell-to-cell transfer through a pilus. Or electroporation, where you zap the membrane with an electric field and punch temporary holes.

[DANIEL] Not transfection. That word is for eukaryotic cells. People blur it constantly and it drives me a little crazy.

[SOFIA] It does, I've watched it happen. So why is this hard? Why can I transform E. coli in my sleep and not the interesting bug?

[DANIEL] Two big barriers. One is physical — the cell wall and membrane, which vary wildly. A thick Gram-positive wall, a mycobacterial waxy layer, a fungal cell wall. Getting through that is a materials problem. The second is that even if the DNA gets in, the cell's immune system chews it up. Restriction-modification systems — enzymes that cut DNA lacking the host's own methylation pattern. Your incoming plasmid looks foreign, so it's shredded.

[SOFIA] Right, the cell has a bouncer checking IDs. And the reason we care — this isn't academic — these non-model organisms are sitting on chemistry we want. Carbon fixation, breaking down lignocellulose into fuels, making molecules E. coli just can't.

[DANIEL] The bioeconomy argument. And it's real. But you can't do metabolic engineering on an organism you can't even mutate.

[SOFIA] So let's trace how the field chipped at this wall. And the earliest turning point I love is 1994, filamentous fungi. Aspergillus niger.

[DANIEL] The classic move with fungi was protoplasting — you enzymatically strip the entire cell wall off, leaving a naked protoplast, then coax DNA in. Laborious, and hard on the cells.

[SOFIA] And this paper basically said, what if we don't? They took germinating conidia — the spores, just starting to sprout — intact, and electroporated them. Got around a hundred colonies per microgram of plasmid. No protoplasting.

[DANIEL] Modest number. A hundred per microgram is nothing next to E. coli. But the point isn't the yield, it's that intact cells at the right growth stage — germinating — become permeable enough to skip the worst step. And a light enzyme pretreatment doubled it.

[SOFIA] That timing thing matters, right? The cell wall is being remodeled during germination, so it's briefly vulnerable.

[DANIEL] A window. Which is a theme that keeps coming back — you catch the cell in a physiological state where the barrier is soft.

[SOFIA] Then jump to 2005, and this one's very practical. Pseudomonas aeruginosa. Choi and colleagues.

[DANIEL] This is the one that makes me happy, actually, because it's so simple. The prior methods for getting chromosome-level DNA into P. aeruginosa were transduction — phage-mediated — and conjugation. Both slow, both finicky.

[SOFIA] And they came up with a ten-minute electrocompetent prep. Microcentrifuge, a few wash steps, done. And it gave a ten-thousand-fold jump in efficiency.

[DANIEL] Ten thousand-fold is the kind of number that makes me suspicious, so I want to be fair about what it means — it's relative to the older workflows for that organism, not some universal record. But the effect size is enormous and the method is trivially reproducible. That combination is rare. It replaced transduction and conjugation for that species outright.

[SOFIA] And notice the pattern — 1994 solved fungi one way, 2005 solved a pseudomonad another way. There's no universal method emerging. Every organism, a new hack.

[DANIEL] Which is exactly what the 2009 review nailed down. Aune and Aachmann surveyed the artificial transformation methods across recalcitrant bacteria — electroporation, chemical, the works — and the honest conclusion was: no single approach wins. You have to match the method's mechanism against the specific organism's barrier.

[SOFIA] Which sounds like a downer but I think it's the mature take. It reframes the problem. Instead of hunting for the one magic protocol, you diagnose: is my problem the wall, or the restriction system, and pick your tool accordingly.

[DANIEL] It elevates restriction-modification bypass as its own design axis. Methylate your plasmid to match the host, or knock out the host's restriction system first. That's a mechanistic strategy, separate from the physical delivery.

[SOFIA] Now here's a left turn I did not expect in this arc. 2012, mitochondria.

[DANIEL] Yeah, this one's a different flavor. Isolated mitochondria — from plants and mammals — will actively import double-stranded DNA on their own. Through VDAC, the voltage-dependent anion channel in the outer membrane.

[SOFIA] So the organelle has a built-in DNA door. That's wild. You're not forcing anything, the mitochondrion pulls it in.

[DANIEL] And there's a nice divergence in the detail. Plant mitochondrial import additionally needs ANT — the adenine nucleotide translocator — and the DNA needs terminal inverted repeats. Mammalian import doesn't require those. Same broad phenomenon, different molecular requirements.

[SOFIA] Which rhymes with the whole subject — conserved machinery, divergent logic. And that's almost exactly the phrasing of the 2014 natural transformation review.

[DANIEL] Right. Natural transformation — where the bacterium takes up DNA from its environment with no zapping, no help from us, because it evolved to. Documented in around eighty species. They share conserved uptake proteins — a pilus-like apparatus that pulls DNA across — but the regulation, when and why they turn competence on, is all over the map.

[SOFIA] And the review comes down on the why: not nutrition, not DNA as food. The evidence favors genetic diversity — grabbing new alleles — and repairing your own chromosome using the incoming DNA as a template.

[DANIEL] That's the falsifiable-ish claim, and I appreciate that they weighed the alternatives rather than just asserting it. If it were purely nutritional you'd expect uptake to track starvation more tightly than it does.

[SOFIA] And here's why natural competence is the dream for engineers — if the organism already has the door, you just knock. No electroporator, no protoplasting.

[DANIEL] If you can find or induce it. Which brings us to where the field's heading.

[SOFIA] 2018. The non-model prokaryote toolbox review, out of NREL and CU Boulder. And this is the synthesis moment — they take on the phototrophs, the autotrophs, the cellulose degraders, the organisms with real bioeconomy value, and say plainly: these are still locked out of metabolic engineering, and the design rules are scattered.

[DANIEL] So they try to codify it. Transformation, CRISPR tools, and — this is the part that matters — the regulatory barriers. Not government regulation, gene regulation. Do your promoters even fire in this chassis? Does your origin of replication work?

[SOFIA] That's the shift I love across these twenty-four years. We start at "can we physically get the DNA in," and we end at "the DNA's in, now does the genetic program run." The bottleneck moved downstream.

[DANIEL] And the field finally treats the chassis as a design variable instead of a given. You choose the organism and build the toolbox to fit it, informed by mechanism — the barrier, the restriction system, whether it's naturally competent.

[SOFIA] Twenty-four years from a hundred colonies of Aspergillus to a design manual for organisms nobody could touch. The wall's not gone, but we've got a real set of doors now.

[DANIEL] And a better map of which door fits which lock. That's the progress.

[SOFIA] Perfect place to leave it. Stick with us — more from The Dish after this.