The Modified Messenger

Thirty Years in the Wilderness of mRNA

In October 1995, administrators at the University of Pennsylvania School of Medicine called Hungarian biochemist Katalin Karikó into an office and delivered an ultimatum. Over the preceding six years, every research grant application she had submitted to the National Institutes of Health, the American Heart Association, and corporate foundations had been rejected. Her department chief informed her that synthetic messenger RNA (mRNA) was an un-fundable dead end: if she wished to remain on the faculty track leading to professorship, she had to abandon mRNA immediately. When Karikó refused—during the exact week she was awaiting surgery for a suspected malignancy while her husband was stranded in Hungary by a six-month visa delay—the university demoted her off the tenure track, cut her salary below that of the laboratory technicians working beside her, and stripped her of her laboratory space.

Karikó had arrived in Philadelphia ten years earlier, in 1985, after her laboratory in Szeged ran out of funds. Because the Hungarian socialist government banned citizens from exporting more than a hundred dollars in foreign currency, she and her husband Béla Francia had sold their Russian Lada car on the black market for £900 in British notes, sewed the cash inside their two-year-old daughter Susan's teddy bear, and boarded a one-way flight to the United States.

Her goal was to turn messenger RNA—the transient single-stranded molecule that carries genetic instructions from nuclear DNA to cellular ribosomes—into a therapeutic platform. Rather than manufacturing complex proteins inside industrial bioreactors, doctors could inject synthetic mRNA directly into patients, instructing the patient's own ribosomes to synthesize therapeutic proteins or vaccine antigens on site for a few days before the mRNA naturally degraded without altering genomic DNA.

Throughout the 1990s, the molecular biology consensus dismissed therapeutic mRNA as impossible for two concrete reasons. First, unadorned single-stranded RNA is rapidly cleaved by extracellular ribonuclease (RNase) enzymes. Second—and far more lethal—whenever researchers injected in vitro-transcribed mRNA into mice or human cultured cells, the recipient's innate immune system reacted as if invaded by an RNA virus, unleashing an inflammatory storm of tumor necrosis factor-alpha ($\text{TNF-}\alpha$) and type I interferons ($\text{IFN-}\alpha/\beta$) that shut down protein synthesis and killed experimental animals from systemic shock.

Shunted from bench to bench at Penn, Karikó kept running transcriptions early in the morning and on weekends. In 1998, while standing over the departmental Xerox copy machine, she struck up a conversation with Drew Weissman, an immunologist newly recruited from the NIH to develop an HIV vaccine using human dendritic cells—the sentinel cells of the immune system.

When Weissman applied Karikó's synthetic mRNA to his cultured human dendritic cells, the cells exploded with inflammatory cytokines. Rather than abandoning the molecule, Karikó and Weissman designed a controlled comparison between 2000 and 2004 to find out why synthetic mRNA triggered endosomal Toll-like receptors—specifically TLR3, TLR7, and TLR8—when mammalian cells are already packed with their own native RNA. Exposing human dendritic cells to purified RNA from different biological sources, they found that bacterial RNA and lab-synthesized mRNA triggered massive cytokine storms, mammalian cytoplasmic mRNA triggered a small response, and mammalian transfer RNA (tRNA) triggered zero inflammatory cytokines at all.

Karikó recognized the structural difference. Lab-made mRNA transcribed by bacteriophage T7 RNA polymerase is assembled exclusively from four unmodified ribonucleoside triphosphates: adenosine ($A$), cytidine ($C$), guanosine ($G$), and uridine ($U$). Native mammalian tRNA, by contrast, has up to twenty-five percent of its nucleosides chemically modified after transcription.

Synthesizing mRNAs with dozens of modified nucleosides, Karikó and Weissman discovered that the immune alarm centered on unmodified uridine. When they replaced standard uridine with pseudouridine ($\Psi$)—a natural rotational isomer in which the uracil ring attaches to the ribose sugar via a carbon-carbon $\text{C5–C1}'$ bond rather than the standard nitrogen-carbon $\text{N1–C1}'$ bond—the modified mRNA slipped past TLR7, TLR8, and intracellular protein kinase R (PKR) completely undetected, eliminating inflammation while boosting protein translation tenfold.

When they submitted their discovery to Nature in 2005, editors rejected it within twenty-four hours as an incremental tweak before Immunity published it that August. Even in 2013, Penn refused to reinstate Karikó to the faculty track and sold off the patent rights as she left for BioNTech. Seven years later, in November 2020, the Pfizer-BioNTech and Moderna COVID-19 vaccines—built directly on Karikó and Weissman's pseudouridine modification to slip the SARS-CoV-2 spike protein blueprint past human Toll-like receptors—demonstrated 95 percent clinical efficacy, leading to billions of administered doses and the 2023 Nobel Prize in Physiology or Medicine.