Tamoxifen is one of the most prescribed anti-breast-cancer drugs, but tumours becoming resistant hinder its efficacy in the clinic. There is therefore great interest in developing strategies to reduce resistance and sensitize breast cancer cells to tamoxifen. A groundbreaking study by Iorns et al. published in this issue of the Biochemical Journal suggests that a signal transduction pathway controlled by PDK1 (phosphoinositide-dependent kinase 1) plays a crucial role in regulating the sensitivity of breast cancer cells to tamoxifen. The implications of this study are that PDK1 or PI3K (phosphoinositide 3-kinase), Akt (also known as protein kinase B), S6K (S6 kinase) and mTOR (mammalian target of rapamycin) inhibitors, already being developed for cancer therapy, are likely to have additional utility in sensitizing breast tumours to tamoxifen. In this commentary we also discuss the possibility that inhibiting the PDK1 pathway may help overcome acquired resistance to other anti-cancer treatments.
- growth factor signalling pathway
- phosphoinositide-dependent kinase 1 (PDK1)
- phosphoinositide 3-kinase (PI3K)
- tamoxifen sensitization
Most breast cancers are ER (oestrogen receptor)-positive and depend on oestrogen to grow and proliferate. Tamoxifen is a pro-drug metabolized in the liver to 4-hydroxy-tamoxifen, which competes with oestrogen in binding to the ER, producing a non-functional nuclear complex that fails to stimulate transcription. Remarkably, for a drug developed over 30 years ago by ICI (now AstraZeneca), tamoxifen is still the world's most deployed anti-breast-cancer drug. It has undoubtedly contributed to the reduction in breast cancer mortality seen over the last decades. Despite these benefits, almost all breast cancer patients with metastatic disease, and many with non-metastatic disease, relapse and ultimately die due to tumours becoming resistant to tamoxifen.
Understanding the mechanisms underlying intrinsic and acquired resistance to tamoxifen is therefore of huge clinical importance and could result in novel strategies to overcome tamoxifen resistance, thus further improving survival of breast cancer patients. To address this important issue, in this issue of the Biochemical Journal, Iorns et al.  undertook an innovative dual-screening approach that combined a parallel RNAi (RNA interference) and chemical compound screen in oestrogen-dependent MCF7 breast cancer cells. The aim was to identify genes and compounds that could affect the sensitivity of cells to tamoxifen . The genetic screen involved an RNAi library targeting 779 kinases and related proteins, whereas the chemical screen used 1200 compounds made up of approved drugs and known kinase inhibitors. Remarkably, seven of the 20 most sensitizing hits identified from the genetic screen were key members of a well-studied intracellular signalling pathway controlled by PDK1 (phosphoinositide-dependent kinase 1), which is hyperactivated in the majority of breast and other cancers . Moreover, nine of the top ten tamoxifen-sensitizing hits in the chemical screen were also judged to inhibit PDK1 signalling .
THE PDK1 SIGNALLING PATHWAY
The PDK1 signalling pathway is activated by growth factors through PI3K (phosphoinositide 3-kinase) . PDK1 exerts its growth-promoting effects by phosphorylating and activating 23 downstream protein kinases including isoforms of Akt (also known as protein kinase B) and PKC (protein kinase C) . Many common cancer-driving mutations in genes such as PI3K, PTEN (phosphatase and tensin homologue deleted on chromosome 10) and Ras result in hyperactivation of the PDK1 signalling pathway . Moreover, PDK1 is reportedly overexpressed in a significant number of breast cancers . Activation of the PDK1 pathway stimulates cell growth, proliferation and survival, as well as promoting angiogenesis. Consistent with the key role this pathway plays, inhibition of PDK1 , Akt or mTOR (mammalian target of rapamycin) kinases suppresses tumour formation in genetically modified tumour-prone mice . For these reasons, pharmaceutical companies are devoting huge efforts to inhibit PDK1 pathway components in order to dampen down the over-stimulation of this signalling network for the treatment of breast cancer and other types of cancer . In the last year, clinical trials of several novel PI3K, mTOR and Akt inhibitors have commenced. There is great expectation that these drugs will play a considerable role in the fight against many types of tumour .
SUPPRESSING TAMOXIFEN RESISTANCE THROUGH PDK1 INHIBITION
Iorns et al.  convincingly demonstrate that independent RNAi probes, which reduce PDK1 expression, enhance sensitivity of MCF7 cells to 4-hydroxy-tamoxifen by nearly an order of magnitude. These treatments also enhance G1 cell-cycle arrest induced by tamoxifen. Reduction in the expression of several kinases activated by PDK1, such as Akt1 and different PKC isoforms, also sensitized cells to 4-hydroxy-tamoxifen. The RNAi results were complemented by the observation that drugs which inhibited Akt phosphorylation, such as triciribine, promoted dose-dependent sensitization to tamoxifen. A well-characterized selective inhibitor developed by Merck, Akti-1/2 , which prevents activation of Akt isoforms by PDK1 , also sensitized MCF7 cells to 4-hydroxy-tamoxifen to a similar extent as knock-down of PDK1 expression or treatment with triciribine . The authors also establish that reduction of PDK1 expression sensitizes MCF7 cells to ICI 182780 (fulvestrant), a drug that inhibits oestrogen signalling by a different mechanism to tamoxifen, by inducing degradation of ER-α. Taken together, these results indicate that suppression of the PDK1 signalling pathway sensitizes cells to inhibition of ER signalling rather than specifically affecting the mode of action of tamoxifen. The findings of Iorns et al.  are consistent with the conclusions of earlier studies which demonstrated that inhibition of PI3K, an upstream regulator of the PDK1 pathway, sensitized cells to endocrine therapies, whereas activation of Akt caused resistance to tamoxifen in breast cancer models [10,11].
Interestingly, some of the compounds identified as Akt activation inhibitors and tamoxifen sensitizers are clinically approved drugs for non-cancer applications. These compounds include the anti-psychotics thioridazine and methotrimeprazine (both from the phenothiazine group), the anti-depressant sertraline, the anti-arrhythmic drug amiodarone, and the anti-helmintic agent fenbendazole . It would be interesting to investigate whether breast cancer patients taking one of these drugs responded better to tamoxifen. It should be noted, however, that the mechanisms by which triciribine and the other drugs mentioned above inhibited Akt is not well-characterized. As triciribine is a simple nucleoside derivative, it is unlikely to be a highly selective kinase inhibitor. Moreover, for the majority of compounds, it was only shown that a single high concentration of 10 μM reduced phosphorylation of Akt at Ser473, one of the activating phosphorylation sites targeted by mTORC2 (mTOR complex-2). Whether these drugs inhibited phosphorylation of the key Akt-activating phosphorylation site at Thr308, the residue phosphorylated by PDK1, was not reported . More work is therefore required to determine how selective triciribine and other drugs were for different protein kinases and whether the ability of these compounds to reduce Akt Ser473 phosphorylation is mediated through inhibition of mTORC2, PDK1, PI3K or other mechanisms.
MECHANISM OF TAMOXIFEN ACTIVITY AND SENSITIZATION
An important question concerns the mechanism by which the PDK1 signalling pathway regulates the sensitivity of the cell to tamoxifen, as well as other agents that inhibit oestrogen signal transduction. A PubMed survey reveals over 500 papers reporting various forms of cross-talk between the PI3K/Akt and oestrogen signalling pathways, suggesting that the mechanisms by which PDK1 signalling impinges on the oestrogen pathway could be complex. As gene expression lies at the heart of oestrogen signalling, the ability of PDK1 to control the nuclear localization and function of the transcription factors, such as Foxo (forkhead box O) via Akt or SGK (serum- and glucocorticoid-induced protein kinase), may play a role in modulating gene expression responses controlled by ERs [12,13]. Reportedly, Akt also phosphorylates and controls the activity of the ER , which may represent another mechanism by which PDK1 could regulate oestrogen signalling in cells.
ROLE OF PDK1 IN CONTROLLING RESISTANCE TO CANCER THERAPIES BEYOND TAMOXIFEN
There are an increasing number of reports indicating that components of the PI3K/PDK1/Akt pathway control cancer cell resistance to cancer therapies other than tamoxifen. For example, inhibition of the PI3K pathway sensitizes breast cancer cells to various drugs that inhibit the ErbB (HER) receptor tyrosine kinases including lapatinib , gefitinib and erlotinib . A recent paper shows that activating mutations in the catalytic subunit of Class IA PI3K confers resistance to lapatinib in breast cancer cells that can be abrogated through the use of NVP-BEZ235, a dual inhibitor of PI3K/mTOR . Moreover, activation of the PI3K/PDK1 pathway has been shown to attenuate the efficacy of other commonly used anti-cancer chemotherapeutic agents such as methyl jasmonate , or cisplatin and docetaxel . These findings suggest that hyperactivation of the PDK1 signalling pathway might represent a common pathway that cancer cells use to escape from the growth-suppressing effects of cancer treatments.
The finding that inhibition of the PDK1 signalling pathway sensitizes breast cancer cells to tamoxifen and potentially other cancer therapies will be of great interest to pharmaceutical companies already involved in developing inhibitors against PDK1 signalling pathway components. It suggests that these drugs might not only counteract the tumour-promoting mutations, but will also have the additional benefit in sensitizing cancer cells to current chemotherapeutic regimes. Potent and selective inhibitors of PI3K (GDC-0941, NVP-BEZ235), Akt (GSK690693) and mTOR (rapamycin, NVP-BEZ235) have been developed and are being evaluated in clinical trials. The study by Iorns et al.  emphasizes that, in addition to testing the effectiveness of PDK1 pathway inhibitors at reducing tumour development and size, it would be important to evaluate whether these drugs sensitize tumours to existing therapies such as tamoxifen. Another major point to emerge from the study by Iorns et al.  is that an inhibitor targeting PDK1 itself would be particularly effective at sensitizing cells to tamoxifen. Thus far no sufficiently potent or selective PDK1 inhibitor suitable for clinical trials has been disclosed ; however, there seems to be an increasing realization within the pharmaceutical industry that PDK1 represents a promising anti-cancer drug target, reflected by the publication of 37 PDK1 patent applications in 2007 and 29 thus far in 2008 . The main reason why PDK1 has been somewhat neglected is that the focus of the majority of the research field has surrounded other trendier components of the pathway such as PI3K, Akt and mTOR. PDK1, being a simple kinase whose regulation and function is well-characterized, even fails to feature on many summary diagrams of the PI3K/Akt pathway! However, just because PDK1 is not on many researchers' radars does not mean it is not a key anti-cancer target. In our opinion, inhibiting PDK1 could have significant benefits over inhibitors of PI3K, mTOR or Akt. Unlike these compounds, a PDK1 inhibitor would reduce the activity of a large number of growth-promoting kinases including Akt, S6K, SGK, PKC and RSK (ribosomal S6 kinase) isoforms, which co-ordinate phosphorylation of diverse proteins that stimulate cell growth. Moreover, unlike other components of the PDK1 pathway, there is only a single isoform of PDK1. This may help in elaborating specific and selective inhibitors . Hopefully the finding that inhibiting PDK1 markedly sensitizes breast cancers to tamoxifen, and probably other chemotherapeutic agents, will generate further interest and impetus in prosecuting PDK1 as an anti-cancer drug target.
This work was supported by the Deutsche Forschungsgemeinschaft (to C. P.); and the UK Medical Research Council, as well as the pharmaceutical companies supporting the Division of Signal Transduction Therapy Unit (AstraZeneca, Boehringer-Ingelheim, GlaxoSmithKline, Merck & Co. Inc, Merck KgaA and Pfizer).
- © The Authors Journal compilation © 2009 Biochemical Society