Every living thing known to science writes its genetic instructions using the same four-letter alphabet: the DNA bases A, T, C and G. A study published this month shows that one of biology's most fundamental enzymes can also faithfully read an artificial genetic alphabet doubled to eight letters — a result that helps explain how cells might one day be engineered to run on expanded genetic code.
The study, published Sept. 2 in Nature Communications, comes from the lab of Dong Wang, a professor at the University of California San Diego's Skaggs School of Pharmacy and Pharmaceutical Sciences. Wang's team used high-resolution cryo-electron microscopy — powerful enough to resolve details smaller than the width of a single atom — to capture RNA polymerase, the enzyme responsible for transcribing DNA into RNA, in the act of copying synthetic genetic letters that do not exist anywhere in nature. A separate rundown of the same UC San Diego structural work walks through how the imaging was carried out.
Doubling the genetic alphabet
The synthetic letters belong to "hachimoji" DNA — from the Japanese for "eight" and "letter" — a system first built in 2019 by a team led by chemist Steven Benner, which added four new laboratory-made nucleotides to the natural four, forming two additional base pairs that behave chemically like their natural counterparts. Wang's group used bacterial RNA polymerase from E. coli and determined four cryo-EM structures, at resolutions between 2.42 and 2.75 angstroms, showing the enzyme locking onto these synthetic base pairs using essentially the same molecular contacts it uses for ordinary DNA. A related study from the same lab, published in August in the Proceedings of the National Academy of Sciences, found that RNA polymerase can even read a synthetic base pair that lacks the hydrogen bonds normally thought necessary to hold genetic letters together.
The findings matter because reading synthetic DNA accurately, without a cell's machinery stumbling or introducing errors, is a prerequisite for using expanded genetic alphabets in any real biological system. Researchers have already shown that hachimoji-based DNA strands can be engineered to selectively bind to liver cancer cells in laboratory settings, an early proof of concept for using non-natural genetic letters in diagnostics. Wang's structural work provides a mechanistic explanation for why that kind of transcription works at all inside a cell's normal enzymatic machinery, rather than requiring an engineered enzyme built from scratch.
It is worth being precise about what this study does and doesn't show. The experiments were done with purified bacterial RNA polymerase and synthetic DNA templates in a laboratory setting — not inside living cells or organisms — so questions remain about how faithfully this transcription would hold up amid the full complexity of a living genome. The hachimoji system itself remains a wholly synthetic construct that does not occur in nature, and researchers are still years away from engineered organisms that stably use expanded genetic codes for practical purposes. The university's own summary of the work frames the applications — new diagnostics, engineered biological systems, non-natural therapeutic compounds — explicitly as future possibilities, not demonstrated products.
Wang's group says its next steps involve testing how other cellular machinery, including DNA-copying enzymes and ribosomes, handles these expanded letters — building toward a picture of whether synthetic genetic alphabets could someday work as a stable, heritable addition to a living organism's toolkit, rather than a curiosity confined to test tubes.
