September 2007
In the September issue of the journal Cancer Cell, a team of NIDCR-supported scientists and their colleagues report on a novel protein called LZAP. Discovered by the group in 2005, LZAP appears to be a new growth-inhibiting tumor suppressor gene. In the latest paper, the scientists show that LZAP has biological activity that relates to tumor suppression, they define a biological function in the nucleus that correlates with that activity, and they move into clinical tumor samples to show that LZAP is frequently inactivated in squamous cell carcinomas of the head and neck. The Inside Scoop recently spoke about LZAP with Dr. Wendell G. Yarbrough, senior author on the paper and a head and neck cancer surgeon and research scientist at the Vanderbilt-Ingram Cancer Center in Nashville, Tennessee.
What does LZAP stand for?
LZAP is the acronym for “leucine-zipper-containing ARF-binding protein.” That’s quite a mouthful, I know. I can break it down for you. “Leucine-zipper-containing” indicates that LZAP has a zipper-like loop in its structure. “ARF-binding” reflects that LZAP binds to an important protein in the cell nucleus called ARF.
And what made LZAP such an interesting find?
Well, a few things. Let’s go back to ARF. As a signal is relayed into the nucleus to prompt abnormal growth or indicating the presence of a toxic compound, ARF will bind a specific protein. This protein-protein interaction clears the way for another protein called P53 to become active and trigger the transcription, or copying, of myriad genes whose protein products are needed to help the cell respond to the given threat. What’s interesting, about half of all tumors inactivate P53 to allow their abnormal growth. But some tumors also seem to target ARF. So this ARF/P53 pathway is of great interest to cancer researchers, and LZAP clearly plugs into the network. So to answer your question, LZAP has the potential at the very least to fill in some of the blanks in our understanding of gene transcription. It also could tell us more about the growth dysregulation that leads to certain cancers. How much more only time and a great deal of research will tell.
In this paper you jump off the ARF pathway onto the nearby NF-kB pathway in the cell nucleus. In the serpentine pathways of cell biology, how did you arrive there?
Well, that’s a good story. In our earlier work, we saw LZAP had growth inhibitory properties that indicated it might be a tumor suppressor gene. So we followed up to see if LZAP in fact had other tumor suppressor-like qualities. We discovered that when we temporarily shut off LZAP - or as scientists often say, “knocked it down”- the affected cells proliferated and produced clones, or copies of themselves, that are typical of the growth of a developing tumor.
Did these cells also behave like tumor cells?
Yes. When we inactivated LZAP, we also saw increased tumor like invasion through Matrigel. You may already know, Matrigel is a protein mixture that is commonly used in laboratories to mimic a tissue’s basement membrane, which tumor cells must breech before they are considered invasive cancer. That’s when the light bulb went off in our heads that LZAP might activate a family of enzymes called matrix metalloproteinases, or MMPs. Tumors frequently rely on MMPs to degrade connective tissues, allowing them to invade, metastasize, and colonize other tissues. In follow up experiments, we again knocked down LZAP and determined it led to an increase in MMP9.
And the MMPs brought you to the NF-kB pathway?
Precisely. Many of the major cellular pathways have been much better defined in recent years. So we knew from the scientific literature that MMP9 was mostly regulated by NF-kB. And that brought us to a protein called RelA.
Why RelA?
Because RelA is a transcriptionally active component of NF-kB. Let me back up for a second and define a few terms. NF-kB stands for nuclear factor-kappa B. It’s a complex of proteins that responds in the nucleus to various incoming signals, such as stress, radiation, and growth. But the operative word here is complex in the sense of a network. The network of proteins processes the incoming signal and prompts the transcription of a subset of genes to enable the appropriate cellular response. Not all cell types respond identically to a given stimulus, and that creates a degree of inherent variation in the response of tissue or organ to a threat. But greater specificity is needed in generating a cellular response. The NF-kB network has the ability to process incoming signals and fine tune transcription of the appropriate subset of genes, usually to promote cell survival or to increase inflammation, invasion, and angiogenesis, which is the formation of new blood vessels. That’s where RelA enters the picture. It is the major component of NF-kB in most cell types.
And did LZAP bind to RelA?
It did. LZAP bound directly to RelA. It suggested to us that when bound to RelA, LZAP regulates some aspect of the NF machinery. That is, LZAP seems to be an NF-kB inhibitor. This was interesting to us because dysregulation of NF-kB has been linked to inflammatory and autoimmune diseases. It’s also been linked to the development of tumors. So we followed up on the cancer lead and took a closer look at primary head and neck tumors.
What did you find?
We found about 30 percent of head and neck tumors have markedly decreased LZAP protein. Some have undetectable levels by Western blot, which is a standard laboratory test to measure protein levels in a cell.
In other words, there are many different types of head and neck cancers. But, within a certain subset of these tumors, LZAP clearly is dysregulated?
That’s what it suggested to us. With the biological activity that we already had seen in our laboratory experiments, we thought LZAP may be one of the tumor suppressors that is lost in these tumors. So, we knocked down LZAP in a group of cells, implanted them into mice, and waited to see what happened. We found that the loss of LZAP did not increase the number of tumors, but it dramatically increased the growth of the tumors.
The title of your paper describes LZAP as a "putative tumor suppressor gene." Obviously, the term "tumor suppression" covers a lot of biological territory and has become a very general, amorphous term for growth inhibition. What are the required criteria to nail down LZAP as a tumor suppressor gene?
It’s really a little bit in the eye of the beholder. I think if you knock out a gene in mice and tumors develop, most people would say that the inactivated gene is a bona fide tumor suppressor. Even without that, I think some genes would still qualify as tumor suppressors. But I think the knock out is the best first test for a tumor suppressor.
Where do you take this next?
There’s a ton of work to be done. Number one, what is the real role of LZAP in tumor suppression, if any?
Why do you say "if any?"
Well, you never really know. All of the data support the thought that LZAP has tumor suppressor-like qualities and the fact that it’s lost in some tumors is pretty strong evidence. But I guess as scientists we always want to see the proof. I’m very excited about the possibilities for LZAP, but I want to see the proof. In biology, unlike math and physics where you can kind of figure out things based on known facts and formulas, you have to test it and see what comes out.
What are the tests that need to be done?
We’re obviously going to knock it out in mice. We’re also going to study further the LZAP-RelA interaction. It would be nice to know whether LZAP triggers residual changes, such as in the levels of kinases, which are enzymes that are important in regulating the cell cycle. If so, what effect does it have on RelA activity? Then, of course, we’d like to look at other tumor types and see if LZAP is lost. Other squamous cell cancers come to mind, such as those in lung and esophagus. But also other tumors in which NF-kB seems to play an important role.
What about clinically? I realize it's pretty speculative at this point, but how might LZAP one day tie in to care for head and neck cancer patients?
I think clinically, besides looking at other tumor types for LZAP expression, we’re going to look at LZAP expression and see if it correlates with a particular clinical parameter, such as metastasis and angiogenesis.
So see where the story leads?
Right. Once again, we just have to do it. I don’t think we can make any big predictions. We just have to give it a try and see if there’s any correlation.
Thanks for talking about LZAP.
Glad to do it.