Bacterial Enzyme Successfully Transcribes Eight-Letter Synthetic DNA Alphabet

When researchers first synthesized Hachimoji DNA in 2019, they proved that an eight-letter genetic code could form a stable double helix, but the molecule remained a static chemical architecture. The breakthrough published on September 2, 2026, in Nature Communications changes that status from static to active. A team led by Dong Wang at the University of California San Diego Skaggs School of Pharmacy and Pharmaceutical Sciences demonstrated that Escherichia coli RNA polymerase—a natural bacterial enzyme—can accurately read and transcribe this expanded eight-letter alphabet into RNA. By capturing the process at a resolution of 2.42 to 2.75 angstroms using cryo-electron microscopy, the researchers showed that the enzyme recognizes synthetic bases using the exact same structural checkpoints it applies to natural DNA.[2][4]

The implication of this transcription event is profound: the fundamental machinery of life is not strictly bound to the four letters—adenine (A), thymine (T), cytosine (C), and guanine (G)—that evolution settled on. If a natural enzyme can process synthetic genetic information without requiring wholesale re-engineering, the barrier to creating custom biological systems is significantly lower than previously assumed. The strongest counter-argument to this optimism remains fidelity and translation. While the enzyme successfully transcribed the synthetic bases P, Z, B, and S, the system still faces mispairing challenges, and no laboratory has yet translated this expanded RNA into functional proteins. The leap from transcription in a biochemical assay to a fully functioning eight-letter organism remains vast.[1][6]

The UC San Diego team focused on RNA polymerase because it performs the critical first step in gene expression. When a cell needs to utilize genetic instructions, this enzyme moves along the DNA strand to produce an RNA copy. To observe how the E. coli enzyme handled the unnatural bases, the researchers utilized high-resolution structural imaging that resolves details smaller than the width of a single atom. They discovered that the enzyme does not treat the synthetic letters as foreign anomalies; instead, it identifies them through the same biochemical signals used for ordinary base pairs, seamlessly incorporating them into the transcription bubble.[2][4]

A parallel study published by Wang’s team in the Proceedings of the National Academy of Sciences on August 12, 2026, pushed the boundaries of genetic recognition even further. The researchers investigated a different synthetic base pair that lacks the hydrogen bonds traditionally considered essential for DNA stability. "Our findings reveal that hydrogen bonding between base pairs is not required to trigger active-site conformational changes in cellular RNA polymerase during catalysis," wrote Dong Wang and his co-authors in the PNAS study. This dismantles the long-held assumption that canonical hydrogen bonding is an absolute requirement for cellular transcription.[5]

The structural snapshots captured by the team provide a molecular-level account of this flexibility. By observing the enzyme in the act of opening the transcription bubble, the team confirmed that the synthetic letters did not cause the RNA polymerase to stall. The enzyme continued extending the RNA strand after encountering the artificial base pairs, establishing a mandatory prerequisite for eventually producing messenger RNA that contains expanded genetic codes.[2][4]

Despite the structural success, the system is not flawless. The researchers observed that the enzyme occasionally mispaired the synthetic bases during the transcription process. To address this fidelity gap, they introduced a modified synthetic base called Z*, which sharply reduced one major type of mispairing. This adjustment offers a viable route toward more reliable expanded genetic codes, though it underscores the precise engineering hurdles that remain before synthetic biology can safely deploy these alphabets in living cells.[4]

The ability to transcribe artificial genetic information dramatically increases the design space for synthetic biology. With four letters, the number of possible DNA sequences is enormous, but adding four more chemical building blocks expands that capacity exponentially. Researchers aim to leverage this expanded alphabet to develop molecules capable of recognizing specific biological targets. Previous studies have already utilized expanded genetic alphabets to create synthetic DNA molecules capable of recognizing liver cancer cells, pointing toward highly targeted future diagnostics.[1][3]

The next verifiable checkpoint for this technology will be translation—proving that the cellular machinery can take the transcribed eight-letter RNA and use it to synthesize entirely new proteins. Until that occurs, Hachimoji DNA remains a powerful transcription model rather than a complete biological operating system. Yet, by proving that a central component of the cellular machinery can handle genetic information that does not exist in natural biology, the UC San Diego team has redefined the limits of genetic engineering. The four-letter code is no longer the only biological language scientists can manipulate.[1][6]