A little-known mouse protein disrupts cancer-causing chemical changes to genes associated with human colorectal cancer cells and potentially could be used to treat solid tumors, according to a new study from researchers at the Johns Hopkins Kimmel Cancer Center and the Chinese Academy of Sciences.
In the study, published Jan. 8 in the journal Nature Communications, the mouse version of the protein, called STELLA, disrupted a key epigenetic factor and impaired tumor growth better than the protein's human version. By pinpointing the amino acids (building blocks of a protein) responsible for the difference in activity, the research team developed and tested a drug strategy using those amino acids to treat colorectal cancer in cell lines and in a mouse model of cancer. Epigenetics refers to chemical alterations to genes that promote cancer growth and spread without mutating the DNA.
"For solid tumors — the major killers in cancer — there is a tremendous unmet need to develop new approaches to therapeutically block DNA methylation abnormalities," says corresponding co-author Stephen Baylin, M.D., the Virginia and D.K. Ludwig Professor of Oncology and Medicine at Johns Hopkins and co-director of the Johns Hopkins Kimmel Cancer Center Genetics and Epigenetics Program. "This is a novel approach to take the emerging need for epigenetic therapy in cancer forward in a very palpable way," adds corresponding co-author Xiangqian Kong, a principal investigator at the Guangzhou Institute of Biomedicine and Health, Chinese Academy of Sciences.
The new drug strategy is the culmination of an intensive investigation into ways to target and block proteins that facilitate cancer-specific epigenetic changes in cells. Epigenetics ("on top of" genetics) refers to chemical modifications to the genome that don't change the DNA sequence. If DNA is a cell's hardware, the epigenome is the software. Epigenetic changes to DNA, which include attachment or removal of methyl groups, dictate when and where certain genes are turned on or off, and to what degree.
Just as abnormalities in DNA can drive cancer, so can abnormalities in the epigenome. During the past decade, scientists have developed numerous therapies that block abnormal DNA methylation to help prevent cancer progression and metastasis. Currently, epigenetic therapies are approved for blood cancers, such as leukemia, but not solid tumors.
A prominent epigenetic target is UHRF1, a protein that is highly expressed in many solid tumors. UHRF1 acts as a guide that recruits another protein to add methyl groups to the DNA of tumor suppressor genes. If researchers are able to intercept that guide, they might prevent or even undo cancer-causing changes to the genome, Baylin and Kong say.
Mounting evidence since 2014 suggests that STELLA, a protein involved in the development of mouse embryos, grabs UHRF1 and sequesters it away. With that knowledge, Baylin, Kong and colleagues set out to investigate how and why STELLA inhibits UHRF1.
They quickly identified a difference in the activity of the protein's mouse version and its human counterpart: mouse STELLA (mSTELLA), but not human STELLA (hSTELLA), binds tightly to UHRF1. Comparing the two proteins, they found that mSTELLA and hSTELLA are only 31% identical at the amino acid level.
Next, the team performed structural studies and identified a small peptide region that accounted for the difference in activity between mSTELLA and hSTELLA. But would the mouse peptide work as well in human cancer cells? Putting it to the test, the researchers found that the mSTELLA peptide was required to effectively block UHRF1 and activate tumor suppressor genes in human colorectal cancer cells.
Based on those results, the team moved immediately to develop a drug strategy using mSTELLA to treat cancer. They designed a lipid nanoparticle therapy — an ultratiny drug delivery vehicle made of fatty molecules — to deliver the mSTELLA peptide as mRNA to cells (similarly to how most COVID-19 vaccines work). The therapy performed well in mice, activating tumor suppressor genes and impairing tumor growth.
Because UHFR1 is implicated as an oncogene in numerous types of cancer, the results have implications for treating many cancers, say Baylin and Kong: "We are really excited about moving this forward to take to patients."
Other researchers involved in the study were Ying Cui, Ray-Whay Chiu Yen and Srinivasan Yegnasubramanian at Johns Hopkins. Additional authors were Wenjing Bai, Jinxin Xu, Wenbin Gu, Danyang Wang, Weidong Rong, Xiaoan Du, Xiaoxia Li, Cuicui Xia, Qingqing Gan, Guantao He, Huahui Guo, Jinfeng Deng, Yuqiong Wu, Cheng Luo, Linping Wu and Jinsong Liu from the Chinese Academy of Sciences in Beijing, and Scott Rothbart from the Department of Epigenetics at the Van Andel Institute, Grand Rapids, Michigan, where Baylin also has an appointment.
The study was supported by the National Natural Science Foundation of China (grant #22107101), an investigator-initiated grant from Bristol Myers-Squibb — Celgene, the National Institute of Environmental Health Sciences (#R01ES011858), the Samuel Waxman Cancer Research Foundation, The Hodson Trust, the National Key R&D Program of China (#2022YFA1303100, #2023YFF0724200), the Guangdong Basic and Applied Basic Research Foundation (#2021B1515420002, #2023B1212060050), the Pearl River Talents Plan (#2021QN02Y734), the Science and Technology Projects in Guangzhou (#2024A04J4358), the Open Project of State Key Laboratory of Respiratory Disease (#SKLRD-OP-202213), the Lingang Laboratory (#LG-QS-202205-07), the Basic Research Project of Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences (# GIBHBRP23-03, GIBHBRP24-03) and the National Cancer Institute, National Institutes of Health (#R01CA283463).
Baylin is an inventor of MSP, which is licensed to MDxHealth in agreement with The Johns Hopkins University. He and JHU are entitled to royalty sale shares. He also has been granted a provisional patent related to the work.
