When the first COVID-19 vaccines were distributed in Europe and the United States in 2021, the BioNTech/Pfizer vaccine required temperatures of down to minus 70 degrees Celsius, impossible for much of Africa. The continent ended the year with one of the lowest full vaccination rates globally, while Europe and the United States were already distributing booster shots. A research team at the University of Toronto has now tested whether vaccines could instead be manufactured locally: the MANGO device produces proteins from freeze-dried powder without grid power and without refrigeration, and has been tested at ten locations in seven countries.
1961: Birth of cell-free protein synthesis
The principle dates back to an experiment in 1961. Marshall Nirenberg and J. Heinrich Matthaei removed the inner machinery of bacterial cells—ribosomes, polymerases, and energy carriers—and found that these components produced proteins even outside a living cell. With this cell-free system, the two researchers deciphered the first codon of the genetic code, the triplet sequence UUU for the amino acid phenylalanine. Nirenberg received the 1968 Nobel Prize in Physiology or Medicine.
Six decades later, this technique is emerging as a production technology. Cell-free protein synthesis requires no living organisms, no fermentation vessels, and no refrigeration for raw materials. Keith Pardee's research group at the Leslie L. Dan Faculty of Pharmacy at the University of Toronto built a portable device from this principle: MANGO, short for Manufacturing on the Go.
2020: COVID reveals the system fault
The COVID-19 pandemic showed the consequences of centralized vaccine production. A few manufacturers in the United States and Germany produced the world's most sought-after vaccines. Air transport in refrigerated containers, storage in freezers, and limited local capacity became bottlenecks. Rich countries purchased contingents in advance, poorer nations waited.
Solutions to this kind of problem had been attempted before. The meningitis A vaccine MenAfriVac, introduced in 2010, was the first vaccine globally permitted to officially travel outside continuous refrigeration starting in October 2012, up to four days at 40 degrees Celsius. Over 350 million people in Africa's meningitis belt were vaccinated with it, and meningitis A epidemics have nearly disappeared from the region since then. Yet MenAfriVac was still manufactured in a factory in India.
The MANGO concept addresses the next level: not better logistics for centralized production, but decentralized production itself.
2026: MANGO at ten locations around the world
The device contains freeze-dried extracts from bacterial cells, including ribosomes, polymerases, transfer RNA molecules, and energy carriers. Similar to instant coffee, the components are reactivated with water. A computer-controlled pump and valve system guides synthesis and purification of the desired protein. For simple steps, a hand-cranked centrifuge suffices.
In May 2026, the Pardee team reported in Science Advances on the device's deployment at ten locations in seven countries: Canada, the United States, Brazil, Colombia, Chile, India, and other laboratories. In Recife and Bogotá, RT-LAMP enzymes were manufactured for diagnostic tests that detect infectious diseases like Zika, dengue, and chikungunya. Such enzymes previously depended on supplies from well-equipped laboratories. According to the study, the quality of the MANGO product was comparable to commercial gold standards.
Comparison: What decentralized diagnostics have already achieved
The impact that decentralized molecular biology tools can have is demonstrated by Cepheid's GeneXpert device. Since the WHO recommendation in 2010 for the Xpert MTB/RIF test, reliable tuberculosis diagnosis outside reference laboratories has been possible. Before this test, a confirmed TB diagnosis in countries without laboratory infrastructure took weeks to months. According to the WHO, over 130 countries now use the device.
What MANGO attempts goes beyond diagnostics: not just detection tests but production tools for biological therapeutics. To date, no portable device has produced a vaccine for clinical use in a situation where no central factory was available. The MANGO team is initially targeting diagnostic enzymes and research proteins. Whether the next step, decentralized manufacturing of vaccine candidates, will succeed remains an unanswered question.
Three hurdles to widespread use
The pilot sites show that the principle works technically. Three hurdles remain before broad implementation.
Regulation: Every locally produced diagnostic agent and every vaccine candidate requires regulatory approval. National and international authorities have so far no standardized procedures for decentrally produced biological therapeutics. The question of how an enzyme from a MANGO device in Recife gets certified remains open.
Knowledge transfer: The device should be operable without specialized molecular biology training. Pardee emphasizes that simplified protocols and open-source software have been developed. But local implementation requires trained personnel and time.
Funding: The U.S. National Science Foundation supports cell-free systems in biomedicine through its CFIRE program with 40 million dollars, awarded in 2024. A global rollout would require many times that amount, coupled with the interest of international health organizations like GAVI and CEPI in recognizing decentralized production as a standard option.