Researchers at the National Cancer Institute have developed a functional assay to determine which mutations in the BRCA2 gene may increase the risk of tumorigenesis.
The assay expands on the ability of segregation analysis to analyze mutations in the gene by examining how they affect embryonic stem cells.
Mouse Brca2 is essential for the viability of mouse ESCs, and the assay is based on the ability of the human BRCA2 gene to complement the loss of the Brca2 gene in the mouse cells.
When introduced into a Brca2-deficient mouse ESC, functionally normal human BRCA2 variants can compensate for the mouse Brca2 deficiency. However, harmful or deleterious variants of human BRCA2 do not have this ability, leading to their identification in this test.
The research appeared July 6 in the online Nature Medicine and will be published in this month’s print edition of the journal.
Shyam Sharan, an investigator at the National Cancer Institute’s Center for Cancer Research and the senior author on the paper, spoke with CBA News this week about the assay, its potential application in drug discovery, and how he and his colleagues plan to further develop it.
Could you give me a little background on this work?
For more than 10 years, I have been working on BRCA1 and BRCA2 genes, and my main interest is in understanding how mutations in these genes cause cancer, to understand the process that leads to tumorigenesis.
We use mice as a model system. We are creating mutations in these genes that have been identified in humans, and then we want to see what happens.
We wanted to pick from the human population [mutations] that have been identified in people who actually get the disease. These mutations are throughout the length of the protein.
For me, it has been a big puzzle how these mutations cause cancer. We hope that if we understand the process of how it causes cancer, we may find therapeutics or ways of delaying or stopping the tumorigenic process.
As we started to make these mutations, we realized that a lot of these variants that have been identified in humans are not really deleterious, or tumorigenic. There are some that behaved the way they were expected to — by changing one amino acid, they were changing the entire reading frame, or making the whole protein unstable.
We realize that people are now going for genetic screening to find out if they have a mutation, and if they have a mutation what it means. If you have a deleterious mutation in these genes, the risk of developing breast cancer increases from 13 percent in the general population in the US to about 35 percent to 85 percent. So a deleterious mutation represents a major increase in risk.
Then we began to realize, it is important to determine how the mutation causes cancer, but even before that, it is important to distinguish neutral variants and deleterious variants.
Although our mouse system was great at making mutations in the human BRCA2 gene and looking at them in mice, it would take us two years to find an answer. It also took up a lot of space, money, and time.
About 10 years ago, I realized that in mouse embryonic stem cells, if we removed both copies of the Brca2 gene, the cells do not survive. If we put a mutated human BRCA2 gene into mouse ESCs, if the mutation is deleterious, the cells will not survive. If the cells survive, we can ask questions about the genes’ function and the effect of the mutation.
That led to the development of the mouse ESC-based system to study the human variants of the BRCA2 gene. We can study three to five variants in about three months, so that is a big difference from what were doing before.
How is this assay relevant to drug discovery?
Right now, the way the assay is set up, it can be used to screen the human variants. There are already 1,200 variants identified in the BRCA2 gene. I do not have the numbers for 2008, but by 2004, 70,000 people had gotten their gene sequenced.
Not all mutations in BRCA2 genes may be sensitive to the same drugs. Now we can start to treat these mutations that we have in tissue culture with DNA-damaging agents or different drugs, and see if there is a difference in their sensitivity.
Depending on the mutation, it may be sensitive to gamma radiation, so it may be treatable via chemotherapy. Or it may not be sensitive to gamma radiation, but it may be sensitive to something else. One can extrapolate this information and say that the compound may be something that can be used for treating tumors.
This provides a simple system to screen for sensitivity to different drugs, although we have not yet used it for that application.
How does the assay work?
There are two parts to this. One is the molecular biology part, where we make the mutation in a human BRCA2 gene, so we have about 200 kilobases of the human gene cloned in vectors.
Then we make the desired one nucleotide change and go into the tissue culture lab, where we take human genomic DNA and put it into mouse ESCs using electroporation. We select for the cells with this human DNA.
Then we start to screen them against compounds in 96-well plates, and look at the drugs’ effect on cell proliferation. Right now this is done manually, but it can be done via robotics for greater throughput.
Right now we have a limited number of known compounds that we are testing — things such as cisplatin and mitomycin C, to which BRCA2 mutated genes are known to be sensitive. We have not yet used this assay to screen a small compound library.
Are you planning to develop this assay to be done in higher throughput, such as 384- or 1,536-well plates?
At present we are able to do, as I said, three to five mutations in three months, because [we] are doing every step more systematically to see how things are working. Now we are trying to do things in a less systematic way to see if we can reduce the time necessary, and we have reduced the time to two months.
The only reason we have not done it in 384-well plates is that we do not currently have plate readers, although I see no reason that it should not work in 384-well plates.
What is the next step in this work?
Reducing the time necessary to screen the variants against compounds and simplifying the procedure. We would also like to show that this system is not limited to studying BRCA2 variants, but variants of other genes as well, as long as it has an effect on the cell function, such as cellular survival, proliferation, or DNA damage.
We are currently developing a similar system for BRCA1 as a proof of principle that the system can be used for other genes in addition to BRCA2.
It becomes important as we find all of these variants to make sense of them.