Studying a pair of SOCS reveals mechanisms for repressing a cancer-associated protein

Science Spotlight

Studying a pair of SOCS reveals mechanisms for repressing a cancer-associated protein

From the Cooper Lab, Basic Sciences Division

Nov. 20, 2017

Cells and tissues within the body must work together to accomplish important tasks, such as developing from an embryo to an adult or mounting an immune response. As with all collaboration, success requires communication. Cells communicate by using signaling proteins, such as cytokines, hormones and ephrins. These molecules are secreted by some cells and detected by receptors on the surface of other cells. Once activated, receptors transmit signals into the cell, where they are relayed to their final destination through a series of phosphorylation events. Because failure to control the timing of cellular communication can cause problems like inflammation or cancer, understanding how cells set the strength and duration of signaling is of great medical interest.

Toward this goal, the Cooper Laboratory (Basic Sciences Division) is studying a family of proteins known as suppressors of cytokine signaling (SOCS), which are thought to help reset the system by promoting degradation of downstream signaling proteins. “Previous work in the Cooper lab showed that SOCS2 and SOCS6 act as tumor suppressors in normal breast epithelial cells”, says graduate student Carissa Pilling, the lead author on a paper recently published in Scientific Reports. To determine which signaling pathways are regulated by SOCS2 and SOCS6, Ms. Pilling and Dr. Jonathan Cooper set out to identify proteins that they interact with.

SOCS proteins preferentially bind substrates that are phosphorylated on tyrosine residues and thus are actively involved in signaling, meaning that SOCS-substrate interactions may only take place under certain conditions. In addition, SOCS proteins generally promote degradation of their binding partners by bringing them in close proximity to machinery that marks them for destruction. Thus, in order to capture SOCS-substrate interactions, the authors had to set up experimental conditions that would maximize tyrosine phosphorylation and minimize protein degradation. Under these conditions, SOCS-bound proteins were isolated using affinity purification techniques and then identified using mass spectrometry.

Overall, the study revealed 53 proteins that interact with SOCS2 and 73 that interact with SOCS6. In addition to known SOCS interactors such as the E3 ubiquitin ligase Cullin5 and the adaptor proteins ElgB/C, the researchers found interactions with many proteins that are known to be phosphorylated, such as Ezrin, Crk, IRS4, EphA2, and β-Catenin. Interestingly, the list of interactors included proteins located in the nucleus, mitochondria and cell membrane, indicating that SOCS proteins can traffic to a variety of cellular compartments. 

Of particular interest for further study was the interaction of SOCS2 with EphA2, a receptor tyrosine kinase involved in embryonic development that is over-expressed in many breast cancers. By constructing a series of mutant SOCS2 and EphA2 proteins, the authors determined that the interaction between them requires tyrosine autophosphorylation of EphA2. SOCS2 over-expression leads to decreased EphA2 protein levels, indicating that the functional outcome of the SOCS2-EphA2 interaction might be degradation of EphA2. This observation is consistent with the role of SOCS2 in antagonizing signaling pathways.

To confirm that SOCS2 and EphA2 interact under normal, rather than artificial, conditions, the researchers stimulated EphA2 with its natural ligand, the ephrin EfnA1. They found that EfnA1treatment does indeed promote the EphA2-SOCS2 interaction, but that binding is delayed relative to EphA2 autophosphorylation. This result was surprising because SOCS2 should be able to interact with EphA2 as soon as autophosphorylation occurs.

Since EphA2 resides on the cell surface until it is stimulated by EfnA1, one possible explanation for the delayed binding of SOCS2 to EphA2 is that the interaction can only take place inside the cell. To test this hypothesis, the authors performed immunofluorescence microscopy. They determined that SOCS2 and EphA2 interact not at the cell surface but within endosomes, a cellular compartment that shuttles surface proteins to or from the membrane.

fluorescence microscopy image showing co-localization of SOCS2 and EphA2

SOCS2 (Flag-tagged) and EphA2 co-localize on endosomes (upper left panel, 120min). Co-localization does not require the SOCS2-CRL5 interaction (upper right panel) but does require the SOCS2 SH2 domain, which mediates the interaction with EphA2 (lower left panel), and stimulation by EfnA1 (lower right panel).

Image taken from the article.

Because endosomes can fuse with degradative compartments known as lysosomes, the authors realized that the SOCS2-EphA2 interaction might cause EphA2 to be degraded by the lysosome. This pathway would be different than the usual route of SOCS substrate degradation, which occurs via the proteasome. Consistent with this idea, SOCS2-mediated degradation of EphA2 was independent of the proteasome, and EphA2 levels increased when the lysosome was specifically inhibited.

Finally, the authors found that expression of EfnA1 is induced by SOCS2. Since EfnA1 triggers EphA2 internalization, this result reveals that SOCS2 down-regulates EphA2 both by inducing its internalization and promoting its localization to lysosomes. Together, these observations indicate that over-expression of SOCS2 could represent a therapeutic strategy for suppressing expression of EphA2 in cancer.

EphA2 is only one of the SOCS-binding proteins identified by Pilling and Cooper. Interestingly, many of the detected proteins are not known to be tyrosine phosphorylated, indicating that novel phosphorylation sites and/or biological roles of SOCS proteins are waiting to be discovered.



Pilling C and Cooper JA. “SOCS2 Binds to and Regulates EphA2 through Multiple Mechanisms.” Scientific Reports. 2017 Sept 7;7(1):10838. doi: 10.1038/s41598-017-11040-3.

This research was supported by United States Public Health Service and the National Science Foundation.