The chemical reaction that grows the food for roughly half the people alive was worked out in a German lab in 1909, industrialized by BASF in 1913, and has barely changed since. It takes nitrogen from the air and hydrogen from natural gas, jams them together at around 400 degrees Celsius and 200 atmospheres of pressure, and produces ammonia. This is the Haber-Bosch process. It fixes roughly 130 million tons of nitrogen a year, and by most estimates the food it makes possible feeds 40 to 50 percent of humanity. It also burns enough fossil fuel to account for close to 1.8 percent of global carbon dioxide emissions, by figures echoed across the fertilizer literature. One reaction, feeding half the world, quietly cooking the planet.

There is another way to pull nitrogen out of the air, and it does not require 200 atmospheres of anything. Bacteria have been doing it in soil for a billion years, using an enzyme called nitrogenase that works at ambient temperature and pressure. The catch is that the bacteria are stingy. Evolution tuned them to fix exactly as much nitrogen as they need for themselves and not one atom more, and to shut the whole apparatus off the instant they sense nitrogen already floating around in the soil, which in a fertilized cornfield is always. A microbe that lives on a corn root in Iowa is sitting on a buffet and refuses to share.

A Berkeley company called Pivot Bio has spent a decade engineering that stinginess out of the bacteria, and it now holds the deepest US patent stack in the field by a wide margin. Of the 28 US patents granted since 2018 on engineered nitrogen-fixing microbes for crops, Pivot Bio holds eight, more than any other single assignee. No one else has more than two. Its 2025 was the biggest patent year the company has ever had, five grants in twelve months, and across its full portfolio it now holds 17 US grants covering the microbes, the gene edits, the seed treatments, and the assays that prove the whole thing works in dirt. For a category most R&D directors have never heard of, that is a commanding position.

The off switch that got deleted

Read the actual claims and the mechanism stops being abstract. Pivot’s foundational 2018 patent describes exposing a corn plant to “engineered diazotrophic bacteria” carrying “at least one genetic variation introduced into a nitrogen fixation genetic regulatory network,” such that the microbes fix atmospheric nitrogen “in the presence of exogenous nitrogen.” That last phrase is the entire trick. Wild bacteria stop fixing when fertilizer is present. Pivot’s do not.

A later grant, issued in November 2024, names the specific target: a gene called glnD. The patent claims a bacterium in which the glnD gene is deleted, or has its ACT domain or UTase domain knocked out. In plain terms, glnD is the top of the sensory cascade that tells the cell how much nitrogen is around. Cut it, and the microbe loses the ability to sense that it should stop. Nitrogenase keeps running. The bacterium keeps fixing nitrogen it does not need and leaking the surplus to the plant root it lives on. Pivot’s own numbers, written into a 2025 composition patent, quantify the result: the engineered strains fix at least 1×10⁻¹⁷ millimoles of nitrogen per cell per hour and colonize corn roots at ten thousand cells per gram of root tissue. The same patent covers mixing the microbes directly into a bag of conventional fertilizer, because the point is not to replace fertilizer overnight. The point is to shave a slice off it every season.

There is a second piece of engineering in these patents that has nothing to do with biology and everything to do with getting a product to market. Pivot repeatedly describes its microbes as “non-intergeneric.” That word is regulatory, not scientific. US oversight of engineered microbes treats organisms carrying DNA from a foreign genus far more harshly than organisms edited using only their own genes. By deleting native genes rather than importing foreign ones, Pivot built its product to slot into the lighter regulatory lane. The gene edit and the regulatory strategy are the same design decision. That is the kind of detail that separates a company shipping on five million acres from one still writing grant applications.

From a refactored gene cluster to a corn field

The science traces to a single paper. In 2012, in the Proceedings of the National Academy of Sciences, a UC Berkeley and UCSF graduate student named Karsten Temme, working with David Zhao and MIT synthetic biologist Christopher Voigt, published “Refactoring the nitrogen fixation gene cluster from Klebsiella oxytoca.” They took the tangled genetic machinery for nitrogen fixation and rebuilt it from scratch out of synthetic parts, deliberately stripping away every scrap of native regulation so they could control it themselves. It was a demonstration that the cell’s own rulebook could be rewritten. Temme and a labmate, Alvin Tamsir, turned that demonstration into Pivot Bio, with Voigt as a co-founder.

The company that grew out of it is no longer a science project. Pivot’s corn product, PROVEN 40, is named for the up to 40 pounds of nitrogen per acre it is meant to replace, about a quarter of what a corn crop needs. US farmers used it on more than five million acres of corn, up from three million two years earlier, according to AgFunderNews, which also reported the company crossed $100 million in revenue and installed agtech investor Chris Abbott as CEO while Temme moved to chief innovation officer. Pivot raised a $430 million round that pushed its valuation toward $2 billion, Forbes reported. MIT News, profiling the company in early 2025, framed the pitch bluntly: microbial nitrogen delivered straight to the root, without the leaching and runoff of synthetic fertilizer.

Why the patent lead matters more than the acres

Here is the tension the patent data exposes. Pivot is not the only company chasing biological nitrogen, but it is the only one converting the chase into granted US patents. Ginkgo Bioworks and Bayer have been working this problem together since 2017, first through a joint venture called Joyn Bio and then, after Joyn folded into Ginkgo in 2022, through a renewed alliance announced in late 2025. On the specific patent theme of engineered nitrogen-fixing crop microbes, neither Bayer nor Ginkgo holds a single US grant. Kula Bio, which raised a fresh $25 million round in December 2025 according to its securities filing, took a different road entirely, skipping genetic engineering and instead feeding ordinary bacteria a carbon-rich diet so they have the metabolic energy to fix more nitrogen. Azotic, the British company behind a coating microbe called N-Fix, has two US grants.

The candor comes from inside the field. As one of the scientists who worked on the Ginkgo and Bayer effort told AgFunderNews, “Nobody’s yet hit a solution that can consistently replace 30% or 40% of synthetic fertilizer. This has always been the target.” Pivot’s PROVEN 40 is aimed at the low end of that very range, and independent agronomy on how much nitrogen these microbes reliably deliver in a bad-weather year remains a live argument. The patents do not settle whether the biology wins. What they settle is who owns the ground if it does.

That is the adjacent-possible bet worth watching. Green ammonia startups are trying to keep the 1913 reaction and swap its fossil hydrogen for clean hydrogen, a hard and expensive retrofit of a century-old furnace. Pivot’s patents describe skipping the furnace. If engineered soil microbes can eventually shoulder a third of a cornfield’s nitrogen, the map of a $190 billion fertilizer industry, and a meaningful chunk of global emissions, starts to redraw itself around a gene deletion rather than a chemical plant. The company that has quietly filed the most claims on that deletion is not a fertilizer giant. It is a startup founded on a graduate student’s decision to delete a bacterium’s ability to say no.

Method note. Patent counts come from a search of 9.3M US utility grants sourced from USPTO bulk grant XML, filtered to grants issued since January 2018 whose text matches engineered biological nitrogen fixation for crops, with assignee names consolidated across variant spellings. “Pivot Bio holds eight of 28” reflects grants matching that theme; the company’s full portfolio of 17 US grants includes filings indexed under broader “plant traits” language. Patent mechanism details are quoted from the granted claims and abstracts of the cited patents. Commercial figures, funding, and executive quotes are drawn from AgFunderNews, Forbes, MIT News, DCVC, and company filings, cited inline. Haber-Bosch energy and emissions figures come from the fertilizer-transition literature, including FAO and IOPscience analyses; estimates of the share of global population fed by synthetic nitrogen range from roughly 40 to 50 percent depending on method. A patent lead is a claim on future ground, not proof the underlying agronomy will deliver at scale.