Scientists Find Proteins Can Help Dictate DNA Sequences

Scientists studying how bacteria defend themselves from viruses have uncovered something they did not expect to find: an enzyme capable of building a precise DNA sequence without copying DNA or RNA. The discovery has revealed an unusual mechanism in which the physical shape of a protein helps determine the genetic sequence being produced.

The finding does not rewrite the rules of genetics overnight, but it has forced scientists to reconsider an assumption that has stood for decades. According to Stanford biochemist Alex Gao, the mechanism represents “a fundamentally new way that life produces DNA.”

DNA Synthesis Has Always Needed a Template

The familiar process of DNA replication depends on an existing genetic template. When a cell divides, the two strands of DNA separate, allowing enzymes to read each strand and build a matching partner. The result is two copies of the original genetic material.

Reverse transcription also follows the same basic principle. Retroviruses can create DNA from RNA, but they still require a nucleic acid sequence to guide the process.

Until now, scientists had not identified a system capable of producing a precise DNA sequence without reading DNA or RNA.

That is what made the discovery inside Escherichia coli, commonly known as E. coli, so surprising. Researchers studying a bacterial defense system found an enzyme producing DNA with a repeating pattern, even though there was no conventional template for it to copy.

Two Tiny Gates Control the Sequence

The bacterial system is called DRT3, short for defense-associated reverse transcriptase 3. It contains two enzymes, Drt3a and Drt3b, along with a short piece of noncoding RNA. Drt3b is responsible for the part of the system that caught researchers off guard.

The enzyme produces a DNA strand containing alternating C and A bases. Research published in Science traced that precision to two amino acid residues inside the enzyme’s active site. One acts as a gate that favors A, while the other favors C, preventing the same letter from being added repeatedly.

The physical structure of the enzyme therefore restricts what the DNA sequence can become. Instead of reading an instruction written in DNA or RNA, Drt3b uses its own molecular geometry to produce the repeating ACACAC pattern.

Gao told Science, “The protein itself serves as the blueprint for the DNA sequence. That was quite a surprise. This is a fundamentally new way that life produces DNA.”

The Other Enzyme Uses a Familiar Method

The second enzyme in the DRT3 system behaves much more conventionally. Drt3a reads a short ACACAC sequence within the system’s noncoding RNA and uses it as a template to produce a complementary strand containing alternating G and T.

The two enzymes therefore create complementary DNA strands through different mechanisms. Drt3b produces an AC repeat through structural gating, while Drt3a reads RNA and produces the matching GT sequence through a template-directed process.

Those strands can then pair together into double-stranded DNA. The final product looks like ordinary DNA, even though half of it was produced through a route scientists had not previously recognized.

How the Two Strands Come Together

The unusual process can be broken down into three parts:

• Drt3b creates an AC strand: Its active site physically favors alternating C and A bases without reading a nucleic acid template.
• Drt3a creates a GT strand: It follows a more familiar process by reading a repeating sequence in noncoding RNA.
• The strands pair together: Because AC and GT are complementary, they combine to form double-stranded DNA.

The system demonstrates that the same DNA molecule can be assembled through two different molecular strategies. One relies on a conventional template, while the other depends on the physical properties of a protein.

Scientists Found an 18-Part Molecular Ring

Researchers needed an extremely detailed view of the system to understand how the molecular gates worked. Using cryo-electron microscopy, they resolved the complex to 2.6 angstroms, allowing them to examine the structures involved at a remarkable level of detail.

The resulting image revealed an orderly molecular ring made from six copies of Drt3a, six copies of Drt3b, and six copies of the noncoding RNA. Together, the 18 components formed a highly organized structure.

Inside that ring, the two residues responsible for controlling the AC sequence occupy a cavity with highly specific geometry. The structure allows one repeating pattern while excluding alternatives, suggesting the enzyme is specialized for this particular task rather than generating DNA sequences freely.

The Strange DNA Acts as a Viral Defense

The system exists because bacteria are locked in a constant battle with viruses called bacteriophages, or phages. Under normal conditions, an enzyme called RecBCD destroys the unusual DNA repeats produced by DRT3, preventing them from accumulating inside the bacterial cell.

That changes when certain phages attack. Phage lambda, for example, produces a protein called Gam that blocks RecBCD so the virus can replicate more easily inside the bacterium.

By shutting down RecBCD, the virus also removes the mechanism that normally keeps DRT3 under control. The DNA repeats begin accumulating, and the infected bacterial cell eventually stops growing, preventing the virus from completing its normal replication cycle.

The defense system therefore turns part of the virus’s own strategy against it. A protein produced by the invading virus triggers the conditions that allow the bacterial defense mechanism to become active.

Another Research Team Found the Same Gates

A separate team from Columbia University and the University of Tokyo independently studied the same system. Their work identified the key residues, E22 and R241, and further examined how the RecBCD trigger helps activate the bacterial defense response.

Samuel Sternberg, one of the researchers involved in that work, described the finding as an unusual connection between protein structure and DNA production. “So in a very literal sense, information in the protein’s amino-acid sequence is being translated into a DNA sequence,” he said.

Other scientists have described the finding as a significant conceptual shift. However, researchers have also stressed that the discovery should not be interpreted as proof that proteins can generally write genetic information back into DNA.

The Central Dogma Still Stands

Some early coverage suggested that the discovery had broken the central dogma of molecular biology. The reality is more complicated. Francis Crick’s central dogma concerns how encoded sequence information moves between biological molecules, particularly the idea that information does not flow directly from proteins back into nucleic acids.

Drt3b does not appear to violate that principle. The enzyme is not reading its amino acid sequence and converting each amino acid into a corresponding DNA letter. Instead, its three-dimensional structure physically restricts which DNA bases can enter the active site.

Researchers have described the relationship as structural rather than encoding. The protein’s shape influences the resulting DNA sequence, but it does not function like a traditional genetic template.

That distinction keeps the central dogma intact while expanding scientists’ understanding of how proteins can control DNA synthesis.

Scientists Think There Could Be More

The discovery also has clear limitations. Drt3b produces one specific repeating sequence of alternating C and A, and there is no published evidence that scientists can redesign the enzyme to create any DNA sequence they choose.

The mechanism has also been studied in a bacterial system, so it does not demonstrate that human cells use the same strategy. Its immediate practical applications remain uncertain, although unusual bacterial defense systems have produced major scientific breakthroughs before.

Sternberg suggested that researchers may only be beginning to uncover this type of biology. “My suspicion is that this is the tip of the iceberg,” he said, pointing to the possibility that similar mechanisms have remained hidden because scientists did not know to search for them.

The discovery began with researchers trying to understand how bacteria survive viral attacks. Instead, they found two tiny amino acids capable of controlling a DNA sequence through nothing more than the shape of the enzyme around them. If life has been using other molecular tricks that nobody has thought to investigate, scientists may have only just started finding them.