Monsanto's Corn Trap Repeats in MIT's Sewage Bacteria

Scientists at MIT engineered bacteria that consume pollutants in wastewater and then self-destruct when the job finishes — a kill-switch meant to prevent uncontrolled proliferation in sewage systems.

The bacteria would theoretically degrade contaminants and vanish, leaving no trace in the municipal infrastructure that treats what cities flush away. It sounds like precision engineering applied to a real problem. Is precisely the sales pitch that got us into trouble before.

In 1996, Monsanto released Bt corn across North America with equally confident promises. The toxin would kill crop pests and only crop pests. The modification would stay contained to the field because corn doesn't migrate. Except the pollen did. Within two years, researchers found Monarch butterfly larvae feeding on milkweed near Bt cornfields were ingesting enough toxin to harm them at developmentally significant rates.

The problem wasn't that the organism escaped — it was that the trait did, and no fence stops wind. The structural trap repeats. You identify a confined problem, engineer an organism to solve it, assure the public that safeguards will contain both the organism and its engineered trait. Then discover that "contained" meant something different to evolution than it meant to your press release. Sewage systems are not fields. They are mobile, networked infrastructure designed to move material around.

A kill-switch is a control that works until it doesn't, and it doesn't the moment selective pressure favors bacteria that ignore it.

A kill-switch is a control that works until it doesn't. It doesn't the moment selective pressure favors bacteria that ignore it or drift to treatment plants where the deletion never expressed. We know this ending now. We watched it unfold with Bt crops and documented exactly why containment failed in open systems. The question isn't whether engineered bacteria in sewers would behave differently from engineered pollen in cornfields, it's why we would expect them to, when every system you design now faces the same pressure that broke the last one. You can't build a better mousetrap if you're still using a sieve to hold the mice.