NATURE REVIEWS | CANCER VOLUME 1 | OCTOBER 2001 | 5

Murderous hairpins

T H E R A P E U T I C S T R AT E G I E S

One significant challenge in the fieldof cancer therapeutics is to find a wayof overcoming cellular p53 inactivity.Reporting in the 30 August issue ofNature, Kenneth Raj and colleaguesnow show that a virus — the adeno-associated virus (AAV) — mightprovide an answer to this problem byselectively inducing apoptosis in cellsthat lack active p53.

The normal function of thetumour suppressor p53 is to pre-vent cell multiplication and pro-mote apoptosis. However, p53 isinactivated in most types of humancancer, which allows cancerous cellsto survive.

Raj and colleagues infected celllines — that were either null or wildtype for p53 — with AAV, and usedflow cytometry to assess DNA con-tent and cell fate. The p53 wild-typecells arrested briefly at G2, then re-entered the cell cycle after severaldays, whereas those that lacked p53activity underwent apoptosis. Thiseffect could also be seen in tumoursin nude mice that were induced byinjection with p53–/– or p53+/+ cells:AAV reduced the incidence of p53–/–

tumours by 17% compared with con-trols, and caused established tumoursto regress.

So, how does AAV induce apopto-sis? The authors could mimic theeffects of AAV with etoposide, a DNA-damaging drug, indicating that AAVinfection might elicit a DNA-damage

response. Moreover, treatment withcaffeine — an inhibitor of the DNA-damage signalling protein ATM (atax-ia–telangiectasia mutated) — allowedAAV-infected cells to continue prolif-erating. ATM-null cells were similarlyunaffected by AAV. This indicates thatAAV might act by inducing an ATM-dependent DNA-damage response.

But how is this response pro-duced? AAV-like particles that con-tain recombinant baculovirus DNAdid not affect cell growth, indicatingthat AAV DNA is responsible for trig-gering the DNA-damage response.AAV DNA is single-stranded, withhairpin loops at both ends, andmimics abnormal DNA. Might thisinformation signal DNA damage tocells? Indeed, injecting cells with anoligonucleotide that corresponds tothe AAV hairpin — but has no cod-ing sequence — preferentially killedp53-null cells but not wild-type cells.

What is unclear is why thisDNA-damage response kills p53-null cells, but leaves p53-wild-typecells unscathed. Further work willreveal the exact pathways involvedin this process. But for now, thispotential new strategy for cancertreatment is exciting news.

Sandra Clark

References and linksORIGINAL RESEARCH PAPER Araj, K., Ogston, P.and Beard, P. Nature 412, 914–916 (2001)FURTHER READING Vogelstein, B. and Kinzler, K. W.Nature 412, 865–866 (2001)

Blocked drainsOesophageal cancer is a particularly insidious disease, and itsincidence is rising in the developed world. It has usually begun tometastasize by the time it is diagnosed, and cure rates are low. Ifthe metastatic escape routes of cancer cells could be predicted, wemight be able to block their migration. The many lymphaticvessels that drain a tumour are the logical pathways for cancercells to follow, but which ones will tumour cells use to spread todistant sites? A team led by Michael Griffin at the University ofNewcastle, UK, has developed a simple method that tracks themost likely metastatic highways for oesophageal cancer. Byinjecting a blue dye and a radiolabelled colloid into the tumourmargins, they can follow lymphatic drainage during surgery andremove those lymph nodes to which the radioactivity — andhence the tumour cells — are draining. In the 40 patientsmonitored so far, immunohistochemistry has confirmed that thetechnique is 95% accurate.WEB SITEPress release:http://www.ncl.ac.uk/press.office/press.release/content.phtml?ref=999787371

Proteasome blocker for myelomaScreening for breast cancer is currently performed using standard,film mammography, but digital mammography — a techniquethat uses computers and detectors to produce an image of thebreast — could replace it in the future. A large, multicentre trial— the digital mammographic imaging screening trial — has beenlaunched to compare the efficacies of these screening methods,and the aim is to recruit 49,500 women in the United States andCanada over the next 18 months.

Potential benefits of digital mammography include increasedeffectiveness of detecting cancer in younger women — who havedenser breasts — as it has better contrast resolution. This hastraditionally been a problem with film mammography.

The trial also aims to evaluate cost effectiveness. Although the costof equipment will initially be large, digital mammography mightresult in fewer false-positives and, consequently, fewer callbacks. Inthe long run, this could save money, as well as patient anguish.

The trial will run for several years, but the results, whenannounced, could change the future of breast-cancer screening.WEB SITEThe DMIST trial: http://cancernet.nci.nih.gov/cgi-bin/srchcgi.exe

TRIAL WATCH

Courtesy of the American College of Radiology.

© 2001 Macmillan Magazines Ltd