Synthetic Proteins Co-Opt Oncogenic Signaling for Cancer Treatment

Synthetic Proteins Co-Opt Oncogenic Signaling for Cancer Treatment

A customizable approach spares healthy cells

Michelle Dotzert, PhD

One of the great challenges in cancer research is developing strategies to identify and target malignant cells, while sparing healthy cells. Cancer cells have several unique characteristics that differentiate them from healthy cells, which can be leveraged as therapeutic targets. Cell transmembrane receptors, for example, are often mutated or overexpressed in cancer cells, resulting in constitutively activated signaling pathways that provide constant growth signals. Synthetic proteins have been engineered to identify these overactive pathways and kill cancer cells without harming healthy cells, and demonstrate the potential for synthetic biology to create novel solutions to challenges in medicine.

Rewiring to our advantage

Synthetic biology is the engineering of biology, involving the synthesis of complex, biologically-based systems with functions that do not exist in nature. It has been applied to overcome challenges in manufacturing, agriculture, and even medicine.

In his lab at Stanford University’s School of Medicine, Dr. Michael Lin and his colleagues are developing ways to engineer proteins to carry out customized sensing or controlling functions. According to Lin, this research serves “to create tools for other biologists to use, and to develop new therapeutic methods for complex diseases”—cancer being a very complex disease.

Certain cell transmembrane receptors that are transiently activated in healthy cells, specifically ErbB-family receptor tyrosine kinases including ErbB1 and ErbB2, are subject to constitutive activation in cancer cells, which drives aberrant growth and proliferation. Various treatments have been designed to target these receptors, but exert an all-or-nothing effect by which they inhibit all the receptors or eliminate cells expressing them, creating challenges with toxicity.

Lin and his colleagues developed a synthetic system called RASER (Rewiring of Aberrant Signaling to Effector Release) to target cancer cells based on transmembrane receptor-mediated signaling cascades. The system is based on synthetic proteins designed to query the state of the signaling pathway associated with these cell transmembrane receptors and execute a therapeutic program if an oncogenic state is detected. In this way, the system rewires the oncogenic signal to deliver “cargo” to cancer cells. This cargo is an effector protein, “when released, [it] can be something that kills the cell or activates a transcriptional program,” explains Lin.

The pair of synthetic proteins serve different functions. “One protein contains a viral protease, the other a substrate for that protease and a programmable effector domain,” he says. “The two proteins assemble at an active oncoprotein to release effector in proportion to the total integrated signal.” Since the receptor protein is always “on,” sufficient cargo accumulates and exerts its effects on the cell.

This synthetic approach offers several advantages over existing treatment strategies. First, it is highly programmable, and may be designed to sense various oncoproteins and to activate different effectors. Further, it cannot be evaded via oncoprotein mutation. “We use binding domains from the natural effectors of the oncoprotein,” explains Lin. “This means the oncoprotein cannot mutate around recognition by RASER without losing its ability to activate cell survival or proliferation either.” Essentially, if the oncoprotein were to mutate and evade RASER, it would no longer function to promote cancerous cell growth.

Combining biochemistry and mathematical modeling

Depiction of hyperactive protein leading to cancer cell growth (top), and leveraging a hyperactive protein to promote cancer cell death (bottom).
Dr. Michael Lin, Stanford University

Identifying and understanding protein interactions was an important starting point in the development of RASER. Fortunately, the number of known protein-protein interactions continues to increase, and there are numerous publicly available databases containing quantitative data. “So, this knowledge is mostly provided to us, although we had to make our own measurements for some parameters like protein half-life,” explains Lin. Taken together, this information informed the selection of the components the team wanted to use.

Ideally, in the cells with hyperactive transmembrane receptors, the protease component of the synthetic proteins would be recruited to the membrane where it would cleave the substrate and release the cargo. First, the team used a mathematical model to determine the possible dynamic range of the system. “When the models produced usable dynamic ranges, then we knew we were on the right track,” says Lin.

“We then created a series of RASER proteins and tested them rigorously in cancer and non-cancer cell lines,” he explains. The cell lines were treated with a lentivirus expressing ErbB-RASER1C-Bid, or drugs used as first-line standards of care for metastatic breast and lung cancer (carboplatin and paclitaxel combination chemotherapy, and lapatinib). While the combination chemotherapy treatment killed both cancerous and healthy cells, and lapatinib cytotoxicity varied, RASER induced high rates of cell death in the cancerous cells with negligible effects on the healthy cells. Such high specificity is promising for overcoming toxicity challenges in the development of new cancer treatments.


According to Lin, the process of developing RASER “required a combination of extensive biochemical knowledge, some mathematical modeling, and very systemic testing.” While the process was intensive, and involved a lot of information management, Lin notes that future RASERs can build off this work, “so it will be easier going forward.”

“The biggest challenge is that our tools for rapid protein expression are not very physiological,” explains Lin. He and his colleagues found transient transfections were too variable, and led to too much overexpression. This is a common drawback with transient transfection, in which introduced nucleic acids only exist in the cell for a short period of time, but produce a high copy number of transfected genetic material and high protein expression. As a result, “we ended up making lentivirus constructs to perform most of our published results,” says Lin. Unlike transient transfection, lentiviral transduction integrates viral DNA into the host DNA, and is passed to subsequent generations during cell division.

The next challenge for RASER will be its introduction into a large number of tumor cells, and determining “how to enable it to propagate through the tumors,” says Lin. “We are working on making viral RASERs by starting simply, but we hope in the future to incorporate methods for making particular viruses more suitable for cancer cell delivery,” he explains.

Many options for synthetic biology in cancer treatment

RASER is not the only synthetic biology approach being used for cancer treatment. Chimeric antigen receptor T-cell (CAR T-cell) therapy is an adoptive T-cell immunotherapy in which the patient’s T cells are genetically engineered to produce chimeric antigen receptors (CARs), which enables them to recognize these antigens on tumor cells. There are several FDA-approved CAR T-cell therapies for specific forms of leukemia and lymphoma.

While promising, challenges remain for CAR T-cell therapies. For example, “CAR-T cells are still having trouble getting into solid tumors,” says Lin, adding, “viral RASER can be a uniquely targeted anti-cancer agent that can find its way to tumors anywhere in the body.”

Whether RASER will be effective in human tumors remains to be determined. However, its customizability and highly targeted activity make it a very attractive strategy for further development.