H I G H L I G H T S

NATURE REVIEWS | CANCER VOLUME 4 | JUNE 2004 | 419

Rapamycin, the immunosuppressantinhibitor of the kinase TOR, is in clin-ical trials for the treatment of severaldifferent cancers, but the exact mech-anism by which this drug preventstumour growth is unclear. In the Juneissue of Nature Medicine, WilliamSellers and colleagues show that arapamycin analogue inhibits prostatetumour growth by two independentpathways in mice — induction ofmitochondrial-mediated apoptosisand inhibition of proliferationthrough the downregulation of thetranscription factor Hif-1.

TOR kinase activity can be regu-lated by the oncogenic serine/threo-nine kinase AKT.When targeted to themouse ventral prostate, transgenic Aktdrives the development of prostateintra-epithelial neoplasia. Sellers andcolleagues show that in these micecomplete inhibition of Tor activity bythe oral rapamycin derivative RAD001induces a return to normal cell morphology within the intraluminaltissue of the prostate epithelium. Theauthors noted not only an increase inapoptotic cells within the regressingtumours, but also a concomitant inhibition of proliferation.

So how does Tor inhibition bothinduce apoptosis and inhibit prolifera-tion? The authors addressed this bycrossing the transgenic-Akt mice withmice specifically expressing humanBCL2 in the ventral prostate. BCL2prevented the apoptotic death of cells followingtreatment with therapamycin

analogue, indicating that the mito-chondrial apoptosis pathway is impor-tant for part of the rapamycin-medi-ated efficacy in this tumour model.However, BCL2 expression did notreverse the rapamycin-mediated inhi-bition of proliferation.

Which genes drive Akt-mediatedproliferation? The authors used anintricate series of microarray- andcomputer-based technologies to showthat the hypoxia-responsive proteinHif-1α is upregulated in both the Akt-and Akt/BCL2-expressing tumoursand that its activity is suppressed inRAD001-treated mice. Although others have published links betweenHIF-1α, TOR and AKT previously, themechanism by which TOR regulatesHIF-1α activity remains unclear.

Data from this mouse model indi-cates that human prostate cancer cellsthat overexpress BCL2 or that have TOR-independent activation of HIF-1αcouldbe resistant to TOR inhibitors. Moreimportantly, the current evaluation ofRAD001 in patients with advancedprostate cancer might not prove infor-mative without a corresponding analysisof tumour genotype.

Nicola McCarthyReferences and links

ORIGINAL RESEARCH PAPER Majumder, P. K.et al. mTOR inhibition reverses AKT-dependentprostate intraepithelial neoplasia throughregulation of apoptotic and Hif1 dependentpathways. Nature Med. 10, 594–601 (2004)FURTHER READING Denmeade, S. R. & Isaacs,J. T. A history of prostate cancer treatment. NatureRev. Cancer 2, 389–396 (2002)

RapTOR strikes again

T U M O R I G E N E S I S IN BRIEF

Hypoxia induces adhesion molecules on cancer cells:a missing link between Warburg effect and induction ofselectin-ligand carbohydrates.Koike, T. et al. Proc. Natl Acad. Sci. USA 101, 8132–8137 (2004)

Metastatic spread through blood vessels requires tumour cells toexpress carbohydrate selectin ligands that mediate their adhesionto endothelial cells, allowing them to enter the vasculature. Koikeet al. show that hypoxia induces increased expression of selectinligands and the enzymes involved in their synthesis in coloncancer cells, and that this requires hypoxia-inducibletranscription factors.

Prognostic significance of a short sequence insertion inthe MCL-1 promoter in chronic lymphocytic leukemia.Moshynska, O. et al. J. Natl Cancer Inst. 96, 673–682 (2004)

Moshynska et al. identified MCL1 promoter insertions inperipheral-blood lymphocytes from 18 of 57 patients withchronic lymphocytic leukaemia (CLL). These led to increasedexpression of the anti-apoptotic MCL1 protein, a member ofthe BCL2 family. MCL1 promoter insertions were associatedwith poor response to chemotherapy and decreased survival,indicating a potential use in identifying high-risk CLLpatients.

Alkylating DNA damage stimulates a regulated form ofnecrotic cell death.Zong, W.-X. et al. Genes Dev. 14 May 2004 (doi:10.1101/gad.1199904)

Alkylating agents induce tumour-cell death, even in cells thatlack components of the apoptotic machinery. Zhong et al.show that these drugs induce necrosis in apoptosis-resistantcells due to the activation of the DNA-repair protein PARP,which causes depletion of ATP within the cell. This outcomeonly occurs in cells — including most cancer cells — thatmaintain ATP production via aerobic glycolysis, indicatingthat the metabolic status of cells specifies their fate in responseto alkylating agents.

Suramin inhibits death receptor-induced apoptosis in vitro and fulminant apoptotic liver damage in mice.Eichhorst, S. T. et al. Nature Med. 16 May 2004 (doi:10.1038/nm1049)

Suramin, a derivative of urea, is a chemotherapeutic agent, but itseffect on apoptosis is unknown. Eichhorst et al. show thatsuramin inhibits CD95-mediated cell death and under certaincircumstances will inhibit chemotherapy-induced apoptosis. Thisindicates that the use of suramin as a chemotherapeutic agent —based on its ability to inhibit cytokine signalling — needs to bere-evaluated.

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

C H E M O S E N S I T I V I T Y

P R O G N O S T I C M A R K E R S

M E TA S TA S I S