Reading DNA in Motion: A Breakthrough in Label-Free Biosensing

Summary:

• Conventional surface-enhanced Raman spectroscopy (SERS) can yield incomplete or biased chemical fingerprints when different parts of large molecules, like DNA, sit too far from the sensor's surface.
• NTU CCEB researchers developed an electrical reorientation method that draws more of the DNA strand into SERS hotspots, capturing richer molecular data to enable advanced biosensing for DNA, RNA, and proteins.

 

As the blueprint of life, DNA holds the critical genetic instructions governing development and growth. By analyzing DNA, researchers can uncover vital disease biomarkers, trace epigenetic mutations, and accelerate drug discovery through the observation of drug-DNA interactions. However, studying molecules that are 10,000 times smaller than the width of a human hair requires extraordinary sensing power. To achieve this, scientists rely on Surface-Enhanced Raman Spectroscopy (SERS), a highly sensitive and portable technique that acts like a "scientific magnifying glass”. The analytical tool works by using the light-enhancing capability of metallic nanostructures to amplify the natural chemical signals of molecules as small as a single-molecule.

Despite its amplification power, conventional SERS faces a major technical bottleneck when analyzing larger, more complex biomolecules. Because typical SERS reads molecules from a static, fixed position, critical details are often missed. When a long, folded strand of DNA is placed on a sensor, different parts of the molecule sit at varying distances from the active "hotspots," resulting in an incomplete, biased, or highly obscured chemical fingerprint. Obtaining a complete, undistorted molecular profile is essential for the future of rapid disease diagnostics, therapeutic monitoring, and personalized medicine.

To overcome this long-standing limitation, lead author Cam Tu Tran, working under the guidance of Professor Ling Xing Yi and their research team at NTU CCEB, have developed a novel method to actively manipulate DNA on a sensor. In their study published in the Journal of the American Chemical Society (JACS), the team demonstrated a "dynamic electrochemical reorientation" technique. By applying tunable electrical fields, they successfully steered and flipped short DNA strands directly within the sensor's highly sensitive hotspots, allowing researchers to read their exact composition, length, and sequence with unprecedented accuracy. 

Electrochemical Strategy to Reconfigure DNA strands

While previous SERS approaches succeeded in revealing general details like the base composition of short DNA strands, deeper structural insights—such as the exact length of a strand or how its individual bases are arranged—remained elusive. These critical signals were frequently drowned out by dominant chemical groups or masked by background noise.

To bypass this barrier, Cam Tu and her coworkers developed a way to actively manipulate which parts of the DNA strand enter the SERS hotspots. By changing the electrical potential of the gold nanoparticle sensor surface, the team could dynamically alter how the negatively charged DNA interacts with the metal. Tuning this surface charge essentially coaxes the DNA to adopt different structural postures, or conformations, on demand.

“The most challenging part for the team was figuring out how to reorientate the DNA strands without having to chemically alter the sensor surface,” said Cam Tu. “Electrochemistry allows us to reorientate the DNA rapidly, in less than one minute, and systematically on the same sensor surface, while preserving its SERS enhancement.”

By combining these individual snapshots into a comprehensive SERS "superprofile," the researchers were able to synthesize a far more complete chemical fingerprint. This multi-view approach allowed them to successfully decode the DNA strand’s base composition, length, and exact sequence without relying on a single, static perspective.

The team then tested whether these additional perspectives gave more specific information about different DNA strands. They created a library of 44 short DNA oligonucleotides — small, single-stranded pieces of DNA. The strands also varied in length and in how their bases were arranged.

Teaching a computer to read the chemical fingerprints

Researchers paired the SERS superprofiles with a step-by-step machine learning framework designed to determine the proportion of each DNA base, the length of the strand, and how those bases are arranged.

Researchers further tested the system using DNA strands that were not included in its original training library, simulating the challenge of analyzing an unknown sample. One of these DNA strands differed subtly from sequences the machine learning system had previously encountered. Analysis of its superprofile indicated that it did not fit the known groups, allowing Cam Tu and coworkers to flag it as a sequence outside the existing library.

This identification ability could eventually be useful as a screening step, where unusual SERS superprofiles are flagged for further analysis.

Going beyond short strands of DNA

Cam Tu stresses that this study is a proof of concept. The experiments used short, single-stranded DNA strands ranging from six to 18 bases in length. Future work could examine longer DNA strands and more complicated structures such as DNA hairpins and double-stranded DNA.

In addition to studying changes in DNA conformation, researchers hope to study chemical modifications to DNA, including methylation and glycosylation. Detecting such changes could be particularly relevant to research into epigenetics, where chemical modifications influence how genetic information is used without changing the underlying DNA sequence.

In the longer term, the team envisages building a broad library of different molecule-surface configurations to support molecular characterisation across a wider range of applications. By manipulating DNA’s physical orientation on a sensor using electrical fields, this label-free technology bypasses slow, expensive chemical tagging to revolutionize molecular analysis. Potential applications could range from rapid biosensing for disease and mutation detection to supporting drug discovery through the study of drug-DNA interactions. In future, the approach could also be extended beyond DNA to RNA, peptides, and proteins, with potential applications in disease research.

“Looking ahead, similar approaches could potentially be applied to other biomolecules such as polysaccharides and proteins, as well as synthetic polymers that require structural analysis,” said Cam Tu. “Rather than being used solely for detection, SERS could also decode the specific chemical structures of target molecules.”

 

🎓  Find out more about CCEB's postgraduate programmes here. 

💬  Stay connected! Follow CCEB on LinkedIn, Instagram, and Facebook to keep up with our latest research highlights, student stories, and campus happenings. 

Story by Bella Tan Shixuan, Princessa, and Sumita Thiagarajan, NTU CCEB