At A Glance
Cynthia Kenyon, Herbert Boyer Professor of Biochemistry and Biophysics, University of California, San Francisco; Director, UCSF Hillblom Center for the Biology of Aging.
Education: 1981 — PhD, biology, Massachusetts Institute of Technology
1976 — BS, chemistry and biochemistry, University of Georgia.
Downstream from the discovery enabled by microarrays stands a scientist who has a focused set of genes on which to concentrate. Meet Cynthia Kenyon, a professor of biochemistry and biophysics at the University of California, San Francisco, whose lab is working through a list of some 100 C. elegans genes that affect the regulation of aging in the nematode.
Kenyon’s lab has already discovered mutations in the worm’s daf-2 gene, which encodes an insulin/IGF-1-like receptor, and can double the lifespan of C. elegans. Further, Kenyon’s team has found that this system is regulated by sensory neurons, and that signals from the reproductive system also apparently regulate aging.
BioArray News spoke with Kenyon this week to discuss her research and the role that microarrays play.
Would you describe yourself as a researcher on aging, or a C. elegans researcher?
We are studying aging, and we are using C. elegans to study it because it has a two-week lifespan, and you can also find and analyze genes really quickly and effectively. It’s really easy to do a lot of experiments on C. elegans because you don’t have to wait a long time to see if they are long-lived or not. But we are primarily interested in the problem of aging.
You’ve studied these animals since 1981, and I’ve read that this research has even affected what you choose to eat.
I eat a low-carb diet. We gave our worms sugar and it shortened their life span. So, I’ve been on this diet for three years. I have a great serum profile. And, they have done studies now that are upholding this. They took two obese groups of people and put one half of them on a low-carb diet, a very stringent version of the Atkins diet where you eat only 30 grams of carbs a day, and the other half on a low-fat diet. Neither group actually turned out to lose much weight at the end of a year. But, the low carb people had a much better responsiveness to insulin and [beneficial] triglyceride levels.
That insulin responsivity is a key insight in your research, no?
We have discovered that in order for these [nematode] mutants that had defective insulin responsiveness to live so long — they live twice as long — is because, basically, they are altered in the response pathway for insulin IGF-1.
In order for them to live so long, they have had to have a transcription factor. That pointed us in the direction of microarrays because it told us that gene expression was crucial for the changes in life span. So we needed to find out which genes were changed, and which of those were important. So we used microarrays to figure out which pattern of gene expression was changed along with mutants. Then we used RNAi to test the significance of the genes that changed. We published in 2003 [Kenyon, CJ, et al. “A C. elegans mutant that lives twice as long as wild type.” Nature. 2003 Jul 17;424(6946):277-83.]
What really helped us was Julie Ahringer [University of Cambridge] had made a whole-worm-genome RNA library. What worms do is that they eat whole bacteria. So the library is actually a bacteria library where you have 20,000 strains of bacteria, each expressing a different worm gene in the form of double-stranded RNA. You feed that to the worm, one by one. If you have a gene whose function you want to test, you just have the worms eat the bacteria expressing double-stranded RNA for that gene, and it knocks it down. So, we took our top 50 hits from the microarray and we fed worms the RNAi bacteria for each of those 50 genes to ask if they were important.
We found that a lot of the genes whose expression was turned up in the long-lived mutants were important for their life span. And the genes that were turned down were important, too, because if you turned them down in normal worms, the worms would live longer. That’s the first thing, so we could quickly move from an expression pattern to functional analysis. And there were a few things that I think were really significant for aging. The first was that each of the genes that we tested on their own didn’t have a very large effect. The ones that did have an effect, it was generally from 5 percent to about 30 percent change in life span. That suggested that the transcription factor that we knew about, the FoxO family member, was able to extend lifespan so much by bringing together the activities of lots of genes, each of which, on their own, had a smaller effect. In other words, the genes acted in a cumulative fashion to affect aging. The other thing is that a lot of these genes on their own, the effects are so small you probably wouldn’t pick them up in a conventional mutant hunt. The microarray analysis enabled us to see what the trends were, what the patterns were.
The other thing that was very interesting, and very relevant, is that the genes had a very broad array of functions, they didn’t just do one thing. Some were antioxidant genes, some were metabolic genes, some were anti-bacterial genes, some were chaperones. That was really interesting too, that the animal is pulling out all the stops, coming at the aging problem in different ways, using different kinds of genes. It was really very profound, a very important study, because it gave you really the big picture.
What techniques did you use to create your arrays?
We used PCR, and got primers, and we amplified genomic DNA sequences. The primers were designed by others to target exons.
We are now switching over to 70-mers oligos, which are better than the PCR products because they are more uniform from gene to gene.
When we do PCR reactions, not every reaction is the same. Some work better than others. So you don’t have completely uniform levels, and you have to try to equalize them. With the oligos, that uniformity is much better. And, we know a lot more now about the annotation of the worm genome, and in the last three or four years, we know a lot more about the genes. So the new set is more complete and accurate than the set we used.
What was your array strategy in your studies?
We did lots of arrays, more than 70. We did lots of mutants, we did time courses, we had long-lived mutants, short-lived mutants. And, we are very, very confident about the genes that we analyzed. We know that the expression is very consistently changed over a wide range of conditions. And it is also changed at pretty high levels — a lot of the genes have very large-fold changes. What we didn’t do in our study was say ‘Let’s see how every single gene changes.’ We’ve got the data, but we didn’t analyze it that way, we didn’t say: ‘Here are the genes that consistently change in a level that is 1.5-fold higher than the wild type.’ What we did is we looked for sets of genes whose expression just jumped out at you. Those were the genes that we analyzed. Genes whose levels were changed two-fold or more, in extremely consistent ways, made it into our top 50. We are now working our way down the next 50 in our list and finding that fewer and fewer are having effect on lifespan, suggesting that we were really right.
How did you begin using arrays?
In the C. elegans field, Stuart Kim’s lab [Stanford University] started doing microarrays with C. elegans. I had asked him if our lab could collaborate with him on our aging studies. I told him that we wanted to do arrays of all the different pathways, and do lots of them. He said, ‘Cynthia I don’t know, it sounds like a lot.’ So we set it up ourselves. We knew that we really needed microarrays.
I think the aging process is a system-wide phenomenon. In other words, the rate of aging is changing the whole animal. So we really need to know what was happening with the expression in the animal. A lot of labs either can’t afford to, or don’t really need it quite enough, to set up a whole microarray facility within their lab like we did.
How are you using arrays now?
We discovered that inhibiting mitochondrial respiration increased lifespan. We weren’t the only ones, there were several groups, but we did independently discover that at the same time. We think the mechanism is probably different from the insulin/IGF-1 pathway. We are using microarrays to analyze that, and also to analyze calorically restricted animals. And we discovered also that the reproductive system affects lifespan so we are also doing microarrays of that as well as sensory mutants that affect lifespan. We really have a lot of people in our labs that are doing the same kinds of things we did for the insulin IGF-1 pathway now for other pathways. And this should allow us to determine how much overlap there is between the pathways.
You have been doing integrated biology for quite a while. Do you need to add a physicist in the lab, or a software designer?
I don’t know. That might be helpful, but we haven’t done that yet. We are working on all these various pathways of aging. Some researchers are doing directed screens for genes that may control these different process. Others are doing microarrays to get a genomic fingerprint of the activities that are involved and associated with each processes. Now also, it could fit into the regulatory mechanisms. For example, you might find common regulatory sequences in front of the genes whose expression are changed under the various conditions, and that could point you in the direction of certain regulatory molecules, which could be the same that other people have discovered in the lab using more direct approach. They are really compatible, complementary approaches.
What would you like to see from these tools?
Inexpensive C. elegans microarrays. That would be nice.
Obviously, the more user-friendly they are, the better. We have people coming into the lab who really aren’t that interested in understanding [microarrays]. I have a very smart student who is interested in aging and the nervous system and she wants to find genes that are regulated by sensory perception, which we found to affect lifespan. And then she wants to study the genes. And so we thought, microarrays. That is the way to try to find the genes that are regulated. She doesn’t want to become an expert in everything, she just want to find some genes and study them to understand the molecular biology of their regulation. For someone like that, what you really want is for all the databases and software to be as user-friendly as possible, so that a person who just wants to use microarrays just to get some information, and then move on to something else, can use them very quickly.
The easier they are to use, the more attractive they will be for labs. The people in the labs, they do have to be able to satisfy themselves that the programs that they are using are valid. So there has to be enough information so that for them, and for me, too, to say: ‘Yes I got it, I saw what they are doing. Yes, I think that is good.’ I think it should be easy for them to implement them.
C. elegans is not just studied now by people like me who have been studying it for many years, but it’s being studied a lot by people whose main focus is mammalian genetics because a lot of the genes that they are interested in are also in worms. So you have a lot of people who would like to be able to exploit the valuable experimental features of C. elegans for their own uses. It would be really great if it were easy for any lab to start doing microarray analysis and just having a chip that they don’t have to make, and they could have them at a decent price. There’s a market, [and the microarray industry] should go there.