Groundbreaking Discovery Reveals How Protein Sequence Information is Translated Directly into a DNA Sequence
For decades, biology students have learned a familiar path for genetic information. DNA provides instructions for RNA, and RNA helps cells build proteins from specific sequences of amino acids. A new discovery reported in Science in April 2026 adds a surprising new wrinkle to that picture. Researchers studying a bacterial defense system found a process in which features of a protein directly determine the sequence of newly synthesized DNA.
The system is called ‘defense-associated reverse transcriptase 3’, or DRT3. Scientists found that it can produce double-stranded DNA using two different reverse transcriptase enzymes, but only one of those enzymes follows the familiar rules.
The second enzyme does something researchers had not seen before. Instead of reading a DNA or RNA template, it uses amino acids in its own protein structure to control which DNA building blocks get added.
However, that finding does not mean cells have suddenly started converting ordinary proteins back into their original genes. The process is much more specific, but it still expands what scientists thought enzymes could accomplish.
DRT3 Builds DNA Using Two Very Different Methods
Deep Mind / Pexels / Normally, DNA synthesis depends on an existing nucleic acid template. During DNA replication, one DNA strand provides the sequence information needed to create its complementary partner.
‘Reverse transcriptases’ work differently, but they still need a nucleic acid template. These enzymes typically read RNA and use its sequence to produce DNA, a process famously used by retroviruses such as HIV.
DRT3 breaks from that pattern through a carefully coordinated system involving two reverse transcriptases. The researchers named these proteins Drt3a and Drt3b, and the system also includes a non-coding RNA molecule. Together, these components produce repetitive double-stranded DNA. One strand contains repeating GT units, while its complementary partner contains repeating AC units.
Drt3a handles the first part using a mechanism that researchers already understand. It reads a conserved ACACAC sequence found within the associated non-coding RNA and uses that sequence as a template. The enzyme then synthesizes a DNA strand containing repeating GT units. This process is unusual in its biological role, but it still follows the familiar rule that one nucleic acid sequence guides the creation of another.
The real surprise appears when Drt3b starts building the complementary strand. Scientists expected it to find another DNA or RNA template that could explain how it repeatedly selected the correct nucleotides.
Drt3b Uses Its Own Protein Structure to Guide DNA Synthesis
Third Man / Pexels / Researchers used high-resolution cryo-electron microscopy to examine Drt3b at extremely fine detail. The resulting structures allowed them to see how the enzyme interacts with nucleotides while building DNA.
Their analysis highlighted two amino acids inside the enzyme’s active site. A glutamic acid at position 26, known as ‘Glu26,’ and an arginine at position 253, called ‘Arg253,’ play central roles in nucleotide selection.
These amino acids create chemical environments that favor particular DNA building blocks. Their positions inside Drt3b help control the alternating incorporation of adenine and cytosine during DNA synthesis.
The result is a repeating AC sequence that forms the complementary strand. In effect, information contained in the protein helps determine information written into the newly produced DNA.
The research team tested the idea by changing key amino acids within Drt3b. When researchers disrupted these important residues, the enzyme lost its ability to produce the expected DNA sequence correctly.
Those changes also damaged the bacterial antiviral defense associated with DRT3. The result connected the unusual DNA synthesis mechanism with the system’s biological purpose rather than leaving it as an interesting laboratory reaction.