Photo credit: www.sciencedaily.com
Researchers Unveil Interactive Map of U2OS Cells, Enhancing Understanding of Childhood Cancers
For over 400 years, scientists have been on a quest to map the complexities of human cells. Despite significant advancements, many elements of cellular structure and function remain mysterious.
“While we have cataloged the proteins within our cells, the intricate ways in which they interact to fulfill cellular functions are still largely undefined across different cell types,” remarked Leah Schaffer, Ph.D., a postdoctoral research scholar at UC San Diego School of Medicine.
In a groundbreaking study, Schaffer, alongside fellow researchers from UC San Diego and collaborating institutions including Stanford University, Harvard Medical School, and the University of British Columbia, has developed a comprehensive and interactive map of U2OS cells, which are linked to pediatric bone tumors. This pioneering work combines high-resolution microscopy with an analysis of protein interactions to shed light on subcellular architecture and protein complexes. The resulting map has unveiled previously unidentified protein functions and will enhance the understanding of how mutated proteins contribute to various diseases, including childhood cancers. This significant study is set to be published on April 9, 2025, in Nature.
“Textbooks may give the impression that we fully comprehend cellular structure, but in reality, we lack a detailed assembly guide for any human cell type,” explained co-senior author Trey Ideker, Ph.D., a professor in the Department of Medicine and an adjunct professor in the Jacobs School of Engineering at UC San Diego.
The research team employed a method known as affinity purification to isolate individual proteins and map their interactions. Analyzing over 20,000 images of cell interiors marked with fluorescent dye allowed them to illuminate the positions of proteins of interest, as cataloged in the Human Protein Atlas. By integrating these findings across more than 5,100 proteins, they discovered 275 unique protein assemblies within U2OS cells.
“Traditionally, there has been an inclination to believe that a single gene corresponds to one protein with one specific function,” noted co-senior author Emma Lundberg, Ph.D., an associate professor of bioengineering and pathology at Stanford University. “However, we are gradually realizing the prevalence of multifunctional proteins. This study showcases the significance of integrating diverse data to uncover these multifunctional roles.”
The research unearthed 975 previously unknown functions for various proteins within the cellular map. Notably, C18orf21, a recently identified protein, has been implicated in RNA processing, while the DPP9 protein, known for its ability to cleave proteins at particular sites, has been associated with interferon signaling, a crucial process for combating infections.
The interactive model was built using extensive data derived from scientific literature concerning protein functions. Co-first author Clara Hu, a biomedical sciences doctoral candidate in Ideker’s lab, explained that they utilized GPT-4—a sophisticated AI language model—to swiftly gather information regarding individual proteins and their collaborative roles within protein assemblies. This approach dramatically reduced the time needed compared to traditional research methodologies. The GPT-4 tool also identified common themes within protein assemblies and facilitated the naming of these groups, which were reflected in the interactive cell map.
“This unbiased approach allows us to explore how these components interconnect, particularly in the context of various diseases,” stated Schaffer.
The mapping exercise also enabled the researchers to pinpoint mutated proteins within the cellular framework. They identified 21 frequently mutated assemblies associated with childhood cancers, linking 102 mutated proteins to cancer development thanks to the findings. This insight may transform how molecular and cellular cancer research is governed.
“We must shift our focus away from isolated mutations, which are rare and sporadic, to examine the overarching cellular machinery compromised or manipulated by these mutations,” emphasized Ideker.
Schaffer further likened navigating the U2OS cell map to exploring an online geographic map, enhancing accessibility and discoverability for users interested in understanding the relationships among various proteins.
“As we refine the resolution, we can extract even greater detail,” Hu added. The research team is diligently working to enhance their map’s resolution, enabling users to participate in deep exploration with high-fidelity information.
Researchers believe that the U2OS cell atlas will not only deepen insights into childhood cancers but also provide foundational knowledge for scientists interested in mapping other cell types. This work can also serve as a basis for utilizing AI technologies to investigate the roles of less-understood proteins and their complexes, ultimately aiding in unveiling the mechanisms of diverse disease processes.
Additional contributors to the study include Gege Qian, Dorothy Tsai, Nicole M. Mattson, Katherine Licon, Robin Bachelder, Yue Qin, Xiaoyu Zhao, Christopher Churas, Joanna Lenkiewicz, and Jing Chen from UC San Diego; Kei Ono and Peter Zage from the same institution; Kyung-Mee Moon and Leonard J. Foster from the University of British Columbia; Abantika Pal, Neelesh Soni, Andrew P. Latham Aji Palar, Andrej Sali, and Ignacia Echeverria from the University of California, San Francisco; as well as Steven P. Gygi, Laura Pontano Vaites, Edward L. Huttlin, and J. Wade Harper from Harvard Medical School; and Anthony Cesnik, Ishan Gaur, Trang Le, William Leineweber, and Ernst Pulido from Stanford University.
This study received partial funding from the National Institutes of Health (NIH) through grants associated with the Bridge2AI Program and other initiatives as well as support from Schmidt Futures, the Wallenberg Foundation, and the Göran Gustafsson Foundation.
Source
www.sciencedaily.com