Origin of life researchers can build cell like compartments, but keeping them fed is the hard part. A Minnesota team's preprint 'SpudCell' is a candidate workaround, though it still needs human help and quits after about five divisions.
A protocell is a simplified, cell-like compartment built from non-living chemistry. Origin-of-life researchers have been assembling them for years. The trouble is not building them. It is keeping them running.
A membrane is what makes a protocell a protocell. It also makes the protocell starve. Anything the membrane lets in, it also keeps out. A protocell can copy a piece of DNA or run a reaction, but only as long as its starting supplies last. Then it runs out and stops.
A team at the University of Minnesota, led by Kate Adamala with Aaron Engelhart, reports a candidate workaround. They call it SpudCell. The name riffs on Sputnik and on Adamala's line that she is "mostly made of potatoes." The system itself is a bottom-up synthetic compartment that copies its genome and divides for a few generations before it falls apart.
The setup works like this. SpudCell encloses a roughly 90,000-base genome split across several circular DNA molecules, with counts varying in published materials. The DNA is copied using machinery borrowed from bacteriophage Phi29, a virus that naturally replicates its own genome inside host cells. Earlier work had already run this machinery inside liposomes, the tiny artificial bubbles used here as cell-like compartments (Nature Communications, 2018). Transcription uses T7 RNA polymerase, another virus-derived enzyme. Translation, the step that actually builds proteins from the genetic instructions, uses the PURE system, a reconstituted set of molecular machines developed at the University of Tokyo that is supplied from outside.
The interesting trick is feeding. SpudCell does not make its own ribosomes or enzymes. Instead, it fuses with separate "feeder liposomes" that carry lipids, ribosomes, enzymes, and small molecules. Fusion is controlled by a fusion pore protein that SpudCell itself encodes. When the protein is produced, it studs the membrane and lets the feeder liposome empty its contents into the cell.
Division is mechanical, not biological. There is no cytoskeleton, no internal scaffolding to pull the cell apart, the way real cells do. Instead, the team engineered enough crowding of fusion pore proteins on the membrane that the physical stress splits the compartment in two.
The team then ran a selection experiment. A SpudCell variant that produced more fusion protein grew faster and outcompeted the parent line over five generations. The advantage grew when nutrients were scarce. The University of Minnesota frames this as the first demonstration of selection and competition in a fully bottom-up synthetic cell. Earlier milestones, including the 2010 Venter team's synthetic Mycoplasma and a 2019 Cambridge construct, modified existing cells rather than building from non-living parts.
The result is described in a preprint manuscript hosted on the lab's Biotic page, and reported by Ars Technica and The Guardian. It is not yet peer-reviewed.
The limits are real and source-stated. After about five generations, SpudCell fails. Genome partitioning is random, so daughter cells progressively lose plasmids. The system is also heavily dependent on supplied PURE translation machinery and feeder liposomes. Adamala's own framing, reported by Ars Technica, is that SpudCell is not alive.
Independent reaction has been mixed. Tom Ellis, who works on synthetic biology at Imperial College London, told The Guardian the result was "probably the field's biggest breakthrough in recent times." Philosopher John Dupré of the University of Exeter, in the same piece, questioned whether such systems add much beyond modified bacteria and argued the work does not address what he called the relational aspect of life, the way living organisms continuously interact with and reshape their surroundings.
Adamala and others have used the announcement to launch Biotic, a public-benefit institution meant to coordinate open synthetic-cell engineering. Co-founders include Drew Endy of Stanford, a longtime advocate of open, standardized biological parts. The institutional bet is that bottom-up synthetic cells become a platform rather than a single result.
What to take away: origin-of-life and synthetic-biology claims tend to use language that obscures how much human help a system still needs. A useful rubric is now on the table. Does the compartment feed itself, by making the molecular machines it needs from its own instructions? Or does it still rely on fed-in supplies? And does it keep dividing without help, or does it quit after a handful of generations? SpudCell, by its own authors' account, fails both tests. It is a candidate workaround for the starvation problem, not a finished synthetic cell.