Researchers throughout the world have been studying telomerase closely since its discovery by Carol Greider and Elizabeth Blackburn at the University of California, Berkeley, in 1985.

The enzyme is responsible for protecting and replicating telomeres, the DNA structures at the end of chromosomes that limit the number of times a cell can divide.

However, it is also implicated in cancer, where it counteracts normal telomere shortening and allows cancer cells to divide without limit.

While generally inactive in normal cells, telomerase is notably active in cancer cells, with estimations ranging between 85 and 90 per cent of all tumours having detectable levels of telomerase. This specific association between telomerase and cancer has led the global hunt to understand the enzyme's composition and thus its potential as a drug target.

While human telomerase was discovered in 1989, the full composition of the enzyme complex has remained undetermined. One of the reasons is that it is extraordinarily rare, with perhaps as few as 20 molecules contained in each cell.

The enzyme complex is known to contain two core components - telomerase RNA (hTR ), which acts as the template for synthesis of the telomere DNA, and telomerase reverse transcriptase (hTERT), which is the catalytic protein component.

Now, a team from Sydney's Children's Medical Research Institute (CMRI) has done two things: it has identified the final component of telomerase and it has developed a method to highly purify the enzyme. The team, led by Dr Scott Cohen, published its results in Science on March 30. (DOI: 10.1126/science.1138596)

"Telomerase was known to contain two core components - an RNA and a core protein," Cohen says. "It was also known to contain some mixture of other proteins but the identity of these other proteins has remained a mystery for almost 20 years. A total of 32 different proteins had been proposed as being part of telomerase. We've found that telomerase isn't 32 proteins - telomerase is just two."

In addition to the telomerase RNA and the hTERT protein, Cohen's team has identified just one other component, dyskerin, an RNA binding protein that is known to bind to many different RNAs but which itself was only discovered in 2000. To get to this stage, however, reasonable quantities of the enzyme had to be found and then purified, and this is where a little bit of good luck stepped in.

"The primary challenge of human telomerase is that although it is present in a wide range of cancers, within each cancer cell the amount of telomerase is vanishingly small," Cohen says. "Telomerase is likely the rarest enzyme in the human cell, so to get enough telomerase to purify it and study it we required a large amount of cancer cells."

To find help in generating enough telomerase in order to have a closer look at it, Cohen searched the internet and found someone with the expertise necessary to produce industrial quantities of cancer cells.

"The first hit I came up with was George Lovrecz at CSIRO, and within two weeks he was sending me 293 cells," Cohen says. "He runs this bioreactor and just whips them out."

Fate and the modern enzyme

The CMRI, and Dr Roger Reddel's Cancer Research Unit in particular, have been interested in telomerase for a number of years. It was here that Reddel and one of his PhD students, Tracy Bryan - who just happens now to be married to Cohen - first discovered another mechanism besides telomerase for maintaining telomeres. This mechanism, Alternative Lengthening of Telomeres (ALT), is thought to be used in between 10 and 15 per cent of cancers.

Reddel's lab cloned the hTERT gene in 1998, and fortunately one of the first things it did was to produce polyclonal antibodies from vaccinated sheep. These antibodies have proved invaluable to Cohen's work, he says.

"There's a lot of fate involved in this - like me ending up [at the CMRI] in the first place," Cohen says. "This antibody was sitting in the freezer since 1998, and then I got hold of it. You have to purify your antibody before use, that's very important, and this thing works like a charm. It's a great antibody and has the perfect characteristics. I could immuno-precipitate the enzyme, which was step one. That's a good start."

Cohen next exploited the ability of telomerase to bind to the end of chromosomes to purify it further. "There was a paper published in the journal Biochemistry that didn't get much attention - most people in the telomerase field are biologists, so they don't read it. As a chemist and having done enzymology, those kinds of things are right up my alley.

"[The paper] looked at how telomerase binds with a telomeric DNA substrate - depending on what the last 3'-letter of the telomeric DNA is, it binds really tight. I then used that to develop another affinity purification."

The human telomere repeat is TTAGGG and it needs three repeats for tight binding, Cohen says. "If it ends in GGG, that is the tightest binding - it just sticks like a rock. If it ends in TTA, that's the weakest, and the enzyme substrate dissociates in minutes."

Cohen synthesised the DNA substrate (TTAGGG)3 bearing a 5'-biotin, stuck it on an Avidin bead, and then collected the beads with the enzyme on the DNA. "If you then add dTTP (deoxythymidine triphosphate) and dATP (deoxyadenosine triphosphate) and no G, just T and A, then the enzyme will add TTA. That's the weakest binding, so it falls off.

"Only an active enzyme can add TTA - if it's not active or not folded right it's going to stay as GGG and will stay bound. That's what really set this apart - I could get only the active complex. In immuno-affinity purification the antibody recognises the sequence - it can't tell if it's active or inactive or even if it's properly folded, it just recognises the sequence.

"A lot of people do an IP and they show their protein on a Western blot but that just means they have protein, it doesn't say anything about activity. We can say we have only active enzymes, and it's very rare in enzymology to be able to make that distinction."

Reddel, the senior author of the Science paper, says the purification process Cohen has invented "is one of the cleverest I've seen. In about 12 hours he can purify telomerase more than 100-million fold."

Protein identification

Identifying the molecules in the purified enzyme was still a challenge, because the amount of telomerase he extracted from 100 grams of cancer cells was only 100 nanograms.

The identification was achieved by Drs Mark Graham, Nicolai Bache and Phil Robinson in CMRI's mass spectrometry unit, which is funded by CMRI's popular Jeans for Genes campaign.

The answer was rather simple. Apart from the two components telomerase is already known to contain, there is only one more protein present in the active enzyme, dyskerin. Cohen was not surprised at its relative simplicity, however.

"A lot of the papers that proposed these other protein associations used indirect methods. After all, no one had purified the enzyme complex. If you want to know the function of the protein you can use over-expression methods ... but things tend to go haywire. If your conclusions are based on over-expression you have to take them with a grain of salt.

"And to detect telomerase activity, most people use a PCR-based assay, which is very sensitive but not very accurate. There are many other artefacts in the assay that can give you a false negative or a false positive.

"So between over-expressing your protein, followed by a PCR-based assay, you're going to get all kinds of crap. I would say 25 or so of the papers I dismissed outright just because of the methods. The paper describing dyskerin was one of the few that I found convincing so I wasn't terribly surprised when it was there and I wasn't too surprised when everything else wasn't there."

In the Science paper, the team proposes that telomerase is a dimer, made up of two identical halves. The molecular weight of hTERT is127 kDa, dyskerin (57 kDa) and telomerase RNA (150 kDa) - add them up and you get 334 kDa, half the weight of the active telomerase enzyme complex (~670 kDa).

The next logical step is x-crystallography to determine the 3D structure of the enzyme, although this will require the creation of "synthetic" telomerase, the endogenous version being so minute. "It's not going to be easy, but whoever determines the detailed 3D structure of telomerase will have a landmark achievement on their hands."

Background

Scott Cohen did his initial training in chemistry at San Diego State University in his native US. He completed his PhD at Caltech and then moved to the University of Colorado, where he worked with Tom Cech, who won the 1989 Nobel Prize for Chemistry for his discovery of the catalytic properties of RNA. While Cohen was at the lab working on ribozymes, Cech was starting to become interested in telomerase.

In 1997, an Australian postdoc also with an interest in telomerase joined the Cech lab. This was Tracy Bryan, who during her PhD studies with Roger Reddel at the CMRI discovered the Alternative Lengthening of Telomeres (ALT) mechanism. Cohen and Bryan later married and decided to move to Australia.

"In 2001 our positions were ending and she always had a standing offer to come back to (the CMRI) in one form or another," Cohen says. "She has her own lab here and her program is in close co-ordination with Roger's program. When we first moved here I was at the University of Sydney doing chemistry, but that didn't work out."

Reddel offered him a place at the CMRI in December 2003, which Cohen accepted with pleasure. "I thought it was about time I learned some biology", he says.

Reddel mentioned how interesting telomerase was and challenged Cohen to look more closely. "As I started looking around and reading the literature, it occurred to me that no one knew what this stuff was, which sounded like a much more interesting question to me," Cohen says. "I thought, 'I'll see if I can purify it', so I did."