A team of researchers from NUS Institute for Health Innovation & Technology (iHealthtech), led by Associate Professor Shao Huilin and Associate Professor Brian Lim, has developed a first-of-its-kind technology to map out diverse protein interactions in cells using DNA barcodes.
The technology, dubbed TETRIS, can explicitly identify and quantify multiple interacting partners in large protein assemblies. By capturing the complex hierarchy of protein interactions within tumor cells, the technology uncovers detailed molecular mechanisms driving disease progression. This enables more precise diagnostics, allowing for the accurate sub-typing of cancers and the identification of aggressive forms of the disease in just a few hours, which was not possible previously.
Further, TETRIS provides vital insights from which doctors can tailor therapeutic strategies to individual patients. For instance, identifying the specific proteins and their interactions that contribute to cancer growth can lead to targeted therapies that improve patient outcomes.
The team’s findings were published in Nature Biomedical Engineering on 19 June 2024. The first authors of the study are Dr. Liu Yu and Dr. Noah Sundah. Both are research fellows from NUS iHealthtech.
Unmasking insidious cancer cells
Proteins are responsible for nearly all basic processes of life. Understanding how these building blocks of life interact with one another is a critical facet of biology and medicine.
Indeed, proteins interact extensively with one another to drive important functions and activities in health and disease—deciphering these interactions can not only lead to better predictions of cell behavior, but also have wide-ranging clinical applications, from improved disease diagnostics to developing more effective therapeutic strategies.
Current methods for studying these interactions, however, have limitations such as false results and incomplete profiling of protein interactions, among others.
The gold-standard approach—yeast-two hybrid assays—requires genetic manipulation and is limited to pairwise binary interactions, rendering it unsuitable for clinical samples. Another common method—mass spectrometry-based proteomics—often misses weak interactions due to extensive sample processing and remains binary in its evaluation.
All in all, these methods fall short of capturing the full spectrum of protein interactions, particularly the higher-order ones where multiple proteins interact to form large functional assemblies; changes in higher-order protein interactions are often linked to more aggressive types of cancer.
The NUS researchers turned to DNA nanotechnology for a solution.
“DNA is a programmable material and can be used to encode rich information while having predictable interactions, which enables us to craft sophisticated architectures with fine spatial control at the nanometer scale,” said Assoc Prof Shao, who led the design of TETRIS. She is also from the Department of Biomedical Engineering under the College of Design and Engineering at NUS.
Harnessing the advantages of DNA nanotechnology, TETRIS leverages hybrid molecular structures as smart encoders to map protein interactions directly in patient samples. Each encoder carries a target-recognizing antibody and a template DNA barcode.
In action, the encoders not only bind to interacting proteins, but also have their barcodes fused bilaterally with that of their neighboring units. The resultant barcodes thus capture all information—molecular identity and spatial relationship—and can be used to decode extensive protein interactions.
Unlike current methods, TETRIS measures both pairwise and higher-order protein interactions, thereby providing a comprehensive picture of the complex protein interactome.
“Think of proteins as delegates at a scientific conference. Each delegate spots a name tag with a unique barcode. When they interact, or ‘shake hands,’ TETRIS captures these interactions by linking their barcodes together,” said Assoc Prof Lim, who led the development of algorithms used to process the data collected by TETRIS.
“This creates a chain of interactions that we can subsequently read and decode via algorithms. Just like seeing who is chatting to whom at the conference, TETRIS enables us to see how proteins interact within cells, providing us with a lens through which we can understand and diagnose diseases more effectively.”
Assoc Prof Lim is also from the Department of Computer Science under the NUS School of Computing.
A standout feature of TETRIS lies in its ability to perform on-site encoding and decoding of protein interactions directly in clinical samples. The technology has been tested on biopsies of human breast cancer tissues, from which it accurately diagnosed cancer subtypes and revealed higher-order protein interactions associated with cancer aggressiveness.
Transforming the future of health care
TETRIS provides a more detailed and accurate picture of the molecular underpinnings of diseases—a boon for cancer diagnostics and treatments. Changes in higher-order protein interactions, which are hallmarks of aggressive cancers, can be more easily detected, thus leading to more informed, personalized clinical decisions.
Additionally, TETRIS is designed with scalability and adaptability in mind. The technology can process a large number of samples and generate results quickly using existing laboratory infrastructure—allowing it to be integrated into routine clinical workflows with minimal disruption.
For instance, the technology can be used in a doctor’s office, where samples obtained via fine-needle aspiration—a safer and minimally-invasive biopsy—can be rapidly analyzed to inform treatment decisions.
The NUS researchers plan to expand the application of TETRIS to other types of cancers and neurological diseases, potentially paving the way for novel diagnostic tools and therapeutic interventions across a broad spectrum of illnesses.
The team has filed two patents for the technology and hopes to commercialize the innovation.
