295
`
`Inhibitors of mammalian target of rapamycin as novel antitumor agents:
`From bench to clinic
`Shile Huang & Peter J Houghton*
`
`Address
`Department of Molecular Pharmacology
`St Jude Children's Research Hospital
`332 N Lauderdale
`Memphis
`TN 38105-2794
`USA
`Email: peter.houghton@stjude.org
`
`*To whom correspondence should be addressed
`
`Current Opinion in lnvestigational Drugs 2002 3(2):295-304
`© PharmaPress Ltd ISSN 1472-4472
`
`Rapamycin and its derivatives, CCl-779 and RAD-001, inhibit the
`mammalian
`target of rapamycin
`(mTOR), downregulating
`translation of specific mRNAs required for cell cycle progression
`from GI to S phase. Preclinically, mTOR inhibitors potently
`suppress growth and proliferation of numerous tumor cell lines in
`culture or when grown in mice as xenografts. CCl-779 and RAD-
`001 are being developed as antitumor drugs and are undergoing
`clinical trials. Clinically, CCI-779 has shown evidence of
`antitumor activity but induced relatively mild side effects in
`patients. Here we discuss potential antitumor mechanisms and
`resistance mechanisms of mTOR inhibitors, and summarize the
`current status of these compounds as novel antitumor agents.
`
`Keywords Antitumor, cell cycle, mammalian target of
`rapamycin (mTOR), rapamycin
`
`Introduction
`Malignant disease is characterized by genetic mutations or
`compensatory changes in cells that result in unregulated
`population growth due to increased proliferation or decreased
`cell death. Since the early 1950s, extensive chemical synthesis
`and screening programs have resulted in clinical trials of
`many potential anticancer agents. In some cases, cytotoxic
`agents have significantly increased survival rates in adult
`diseases and notably for children with hematologic as well as
`solid tumors. However, relatively few cytotoxic agents have
`proven to be useful against a wide spectrum of cancers and
`· most cause significant toxicity. The reasons for the relatively
`poor activity of cytotoxic agents are numerous. Most do not
`target the transforming event, rather they induce forms of
`damage that lead to necrosis or activation of cell-suicide,
`apoptosis. As a consequence, agents are cytotoxic to both
`malignant tumor cells and normal healthy cells, often causing
`severe side effects. Understanding pathways known to be
`critical to the growth and survival of tumors is essential to
`developing potentially selective treatments. Validation of this
`concept has been met with compounds such as imatinib
`(Gleevec, STl-571; Novartis AG). This compound inhibits
`tyrosine kinases such as BCR/ ABL in chronic myeloid
`leukemia and c-KIT in gastrointestinal stromal tumors. Less
`success has been met by targeting activated Ras with
`inhibitors of farnesyltransferase.
`
`Rapamune; Wyeth-Ayerst
`(sirolimus,
`Rapamycin
`Laboratories; Figure 1), an immunosuppressant, has emerged
`as a potent inhibitor of a signaling pathway that may be
`deregulated in some forms of cancer,
`leading to both
`
`increased growth and malignant characteristics of cells. It is a
`lipophilic macrolide,
`that
`selectively
`inhibits
`a
`serine/theronine kinase, specifically, the mammalian .target Qf
`rapamycin
`(mTOR). mTOR
`lies
`downstream
`of
`phosphatidylinositol 3-kinase (P13K) in the Pl3K signaling
`pathway. Rapamycin was originally isolated as a fungicide
`from the soil bacteria Streptomyces hygroscopicus, collected
`from Easter Island (known as Rapa Nui to the natives) in the
`South Pacific in 1975 [1,2]. Structurally similar
`to
`the
`immunosuppresive reagent FK-506 (tacrolimus; Fujisawa
`Pharmaceutical Co Ltd), rapamycin was initially developed
`for transplant rejection [3,4] and was approved by the US
`Food and Drug Administration in September 1999 and the
`European Commission in March 2000. While rapamycin was
`being developed as an immunosuppressant, it was also found
`to exert potent antitumor activity in vitro and in vivo [5-7].
`However, perhaps because its mechanism of action was
`unknown at that time, rapamycin was not developed as a
`cancer therapeutic.
`
`The potential for rapamycin as a cancer therapeutic was
`refocused, in part, by studies of Dilling et al [8]. These were
`the first studies
`to demonstrate potent and selective
`inhibition of growth by rapamycin. Rapamycin potently
`inhibited
`the growth of rhabdomyosarcoma cells at
`concentrations of approximately 1 ng/ml, whereas human
`colon cancer cells were
`inhibited only at micromolar
`concentrations in culture (Table 1). Proliferation of many
`rhabdomyosarcoma cells is regulated by an autocrine loop
`involving secretion of type II insulin-like growth factor (IGF(cid:173)
`Il) and signaling through the type I IGF receptor [9,10].
`Indeed, the cell lines most sensitive to rapamycin were
`dependent on this autocrine pathway.
`
`Additional findings from various research groups around
`the world support rapamycin as a good candidate for a
`cancer therapeutic agent. In many malignant cells in culture,
`rapamycin can act as a cytostatic agent by arresting cells in
`Gl phase. Another potential mode of its antitumor action is
`via the induction of apoptosis. Rapamycin potently inhibits
`proliferation
`or
`growth
`of
`cells
`derived
`from
`rhabdomyosarcoma,
`neuroblastoma,
`glioblastoma,
`medulloblastoma,
`small
`cell
`lung
`cancer
`[8,11-16],
`osteoscarcoma [17], pancreatic carcinoma [18,19], breast and
`prostate carcinoma [20-22], murine melanoma and leukemia,
`and B-cell lymphoma [6,23-25] (Table 1).
`
`Despite the antiproliferative effects of rapamycin, it has poor
`water-solubility and stability in solution, precluding its
`formulation for parenteral use as an anticancer agent. Two
`rapamycin ester analogs, CCI-779 (Wyeth-Ayerst Research;
`Figure 1) and RAD-001 (everolimus; Novartis AG; Figure 1),
`with
`improved pharmaceutical properties, but similar
`cellular effects to rapamycin [16,20-22,26,27,28•,29•], are
`currently undergoing antitumor phase l1I and phase I
`clinical trials, respectively. This review will discuss potential
`antitumor and resistance mechanisms of rapamycin and its
`derivatives, and summarize the preliminary data about
`these compounds, from the bench to the clinic.
`
`ARGENTUM
`Exhibit 1017
`
`000001
`
`

`

`296 Current Opinion in lnvestigational Drugs 2002 Vol 3 No 2
`
`Table 1. Sensitivity of different tumor cell lines to rapamycin.
`Cell Line
`Rhabdomyosarcoma cells
`Rh1 (10% serum)
`Rh1 (serum free)
`Rh18
`Rh28
`Rh30
`Colon carcinoma cells
`GCjc1
`VRC/c1
`Ca Co
`HCTS
`HCT29
`HCT11 6
`Small cell lung cancer cells
`H69
`H345
`H510
`Neuroblastoma cells
`NB-SD
`NS-1643
`NB-EB
`NB-1691
`NB-1382.2
`Glioblastoma cells
`SJ-G2
`SJ-G3
`Medul/ablastoma cells
`DAOY
`Osteoblasts-like osteosarcoma cells
`ROS 17/2.8
`Pancreatic cancer cells
`Panc-1
`MiaPaCa-2
`Leukemia cells
`RBL·2H3
`B-cell lymphoma cells
`BKS-2
`L1 .2
`' NFS1 .1
`WEHl-279
`Thymoma cells
`EL4
`
`Rapamycin (IC..,)
`(ng/ml)
`4680
`3.6
`0.1
`8.0
`0.37
`(ng/ml)
`9800
`1280
`1570
`8400
`> 10,000
`> 10,000
`(nM)
`- 1
`- 1
`- 1
`(ng/ml)
`- 1
`- 1
`2.9
`18
`639
`(ng/ml)
`0.5
`> 10,000
`(ng/ml)
`- 1
`(nM)
`< 100
`(ng/ml)
`1
`3
`(nM)
`- 10
`(ng/ml)
`0.31
`0.27
`0.31
`0.60
`(ng/ml)
`0.17
`
`Reference
`[8]
`
`[8]
`
`[11 , 12]
`
`(13]
`
`(13]
`
`[16)
`
`(17]
`
`[18,19)
`
`[23]
`
`(25)
`
`[25)
`
`Antitumor mechanism of inhibitors of mTOR
`Rapamycin and its analogs, CCI-779 and RAD-001, are the
`most potent and selective inhibitors of mTOR reported so
`far. The three agents share a common mechanism of
`anti tumor action, ie, inhibiting mTOR, which, links mitogen
`stimulation to protein synthesis and cell cycle progression.
`
`mTOR
`the antitumor mechanism of
`To better understand
`rapamycin and its derivatives, we will briefly review the
`emerging cellular role of mTOR. mTOR is referred to by
`various other names, some of w hich are derived from its
`binding partner FK-506-bind ing protein, FKBP12 (discussed
`below). These names are FRAP (FKBP12 and rapamycin(cid:173)
`associated protein), RAFTl (rapamycin and FKBP12 target
`1), RAPTl (rapamycin target 1) and SEP (sirolimus effector
`protein). In the mid-1990s, mTOR was identified as a
`mammalian serine/ threonine kinase of approximately 289
`in humans, mice and ra ts [30-33]. TOR proteins
`kDa
`represent a class of evolutionarily conserved kinases in
`the yeasts, Saccharomyces cerevisiae and
`eukaryotes. In
`Schizosaccharomyces pombe, two TOR genes, TORl and TOR2,
`
`have been cloned, which share 67% identity and encode
`proteins of approximately 280 kDa [34-36]. In the fruit fly,
`Drosophila melanogaster, a single TOR ortholog,
`termed
`dTOR, has been cha racterized, sharing 38% identity with
`TOR2 from Saccharomyces cerevisiae [37,38]. mTOR shares
`approximately 45% identity with TORl and TOR2 from the
`yeast Saccharomyces cerevisiae, and 56% identity with dTOR in
`overall sequence [39,40]. Human, mouse and rat mTOR
`proteins share 95% identity at the amino acid level [40,41].
`mTOR contains a catalytic kinase domain and a FKBP12-
`rapa mycin binding (FRB) domain near the C-terminus, and
`up to 20 tandemly repeated HEAT (Huntingtin, EF3, A
`subunit of PP2A and TOR) motifs at the N-terminus, as well
`as FAT (FRAP-ATM-TRRAP) and FATC (FAT C-terminus)
`domains (Figure 2). Since the C- terminus of mTOR shares
`strong homology to the ca talytic domain of Pl3K, mTOR is
`considered a member of the PIK-related kinase fa mily,
`which also includes MECl, TELl, RAD3, MEl-41, DNA-PK,
`ATM, ATR and TRRAP [42••]. Both PI3K and potentially
`protein kinase B (PKB; Akt) lie upstream of mTOR, w hereas
`ribosomal p70S6 kinase (p70S6K) and eukaryotic initiation
`factor-4E (eIF-4E) binding protein isoforms (4E-BP1-3) are the
`
`000002
`
`

`

`Inhibitors of mammalian target of rapamycin as novel antitumor agents: From bench to clinic Huang & Houghton 297
`
`Figure 1. Molecular structures of rapamycin, CCl-779 and RAD-001 .
`
`(J ,, H OH
`
`' y"'y
`
`0
`
`OH
`
`0
`
`~0.--Y-OH
`y ' CH3
`
`0
`
`O~OH
`
`Qyo
`
`0
`
`0
`
`0
`
`CH3
`
`rapamycin
`(Wyeth·Ayerst)
`
`CCl-779
`(Wyeth·Ayerst)
`
`RAD·001
`(Novartis)
`
`best characterized downstream mTOR effector molecules.
`Increasing evidence has implicated mTOR as a central
`controller of cell growth and proliferation. mTOR may
`directly or indirectly regulate translation initiation, actin
`organization, membrane traffic and protein degradation,
`protein kinase C signaling, ribosome biogenesis and tRNA
`synthesis, as well as transcription [42•• ]. Recent results also
`suggest
`that mTOR may sense cellular ATP
`levels,
`suppressing protein synthesis when ATP levels decrease
`[43].
`
`Specificity of rapamycin action
`Rapamycin inhibits proliferation and growth of many
`tumor cells, which is clearly a consequence of binding
`mTOR. Whether this action is a consequence of inhibiting
`mTOR kinase activity per se is less clear. Rapamycin cannot
`directly bind to mTOR. It first has to bind to the 12 kDa
`cytosolic immunophilin, FKBP12, found in mammalian
`cells,
`to
`form
`the FKBP12-rapamycin complex. The
`complex then interacts with the FRB domain in mTOR
`inhibits
`function of mTOR. High
`(Figure 2), and
`concentrations of rapamycin together with FKBP12 are
`required to inhibit mTOR kinase activity in vitro and
`mTOR autophosphorylation. However, the specificity of
`rapamycin action can be demonstrated in vivo as certain
`mutations in the FRB domain of mTOR affect FKBP12-
`rapamycin binding, and significantly reduce the cellular
`sensitivity of rapamycin. The first rapamycin-resistant
`alleles, TOR1 -1 and TOR2 -1, identified in a Saccharomyces
`cerevisiae genetic screen were shown to confer dominant
`resistance. These mutant TOR proteins lost the ability for
`FKBP-rapamycin
`complex
`binding
`[44].
`Similarly,
`mammalian cells also became highly
`resistant
`to
`.
`l
`.
`(S
`II 2005)
`d .
`th
`rapamycm w 1en a mutat10n
`er
`-? e
`occurre m
`e
`2035
`FRB domain of mTOR, which resulted in decreased affinity
`for binding of FKBP12-rapamycin complex [14,45,46]. In
`the yeast, Saccharomyces cerevisiae, decreased RBPl, a
`
`homolog of mammalian FKBP12, or mutation at Tyr8' , led
`to decreased binding of rapamycin and conferred a
`recessive resistance phenotype [47].
`
`Figure 2. Schematic representation of mTOR domains.
`
`2549 aa
`
`11 HEAT 11 HEAT
`
`HEAT repeals
`
`FRB
`
`FATC
`
`HEAT Huntingtin, EF3, A subunit of PP2A and TOR, FAT FRAP(cid:173)
`ATM-TRRAP, FRB FKBP12-rapamycin binding, CD catalytic
`domain, FATC FAT C-terminus.
`
`Potential models fQr rapamycin inhibition of mTOR
`Small molecule kinase inhibitors act directly, regulating
`kinase activity generally by competition for ATP binding.
`However, whether
`the FKBP12-rapamycin complex or
`rapamycin alone directly inhibits the kinase activity of mTOR
`is still controversial. In vitro, rapamycin inhibited the modest
`increase in kinase activity of immunoprecipitated mTOR
`induced by insulin [48]. The FKBP12-rapamycin complex also
`inhibited the autokinase activity of mTOR, aliliough a much
`higher concentration of rapamycin was needed in vitro ilian in
`vivo
`to
`inhibit the activity of mTOR [49]. Conversely,
`treatment of cells with
`rapamycin did not alter
`the
`
`autophosphorylation level of Ser2'"1, and had little or no effect
`on the kinase activity of immunoprecipitated mTOR [37,49].
`More recently, an alternative model for mTOR function has
`been proposed. Specifically, mTOR may repress phosphatase
`activity associated with downstream targets. The inhibition of
`mTOR induced by bound FKBP12-rapamycin complex, may
`result
`in activation of
`this phosphatase, which
`then
`dephosphorylates downstream effector molecules such as
`p70S6K [50 .. ,51•]. Consistent with
`this model, we [52,
`Houghton & Huang, unpublished data] have suggested that
`mTOR regulates the catalytic subunit of PP2A associated with
`p44/42 mitogen-activated protein (MAP) kinases in some
`
`000003
`
`

`

`298 Current Opinion in lnvestigational Drugs 2002 Vol 3 No 2
`
`p44/ 42
`inhibits
`rapamycin
`cells,
`these
`In
`cells.
`phosphorylation on Thr202
`IGF-1 stimulation.
`following
`to confirm
`the
`However, more studies are necessary
`general ity of this phosphatase model. An alterna tive model is
`for mTOR to act as a scaffold and for the FKBP12-rapamycin
`complex to disrupt higher order mTOR-protein complexes.
`
`Rapamycin inhibition of mTOR-control/ed signaling
`pathways
`Although specific details of how rapamycin inhibits function of
`mTOR remain to be resolved, it has been widely accepted that
`inhibition of mTOR by rapamycin blocks growth factor
`stimulation of 40S ribosomal p70S6 kinase and phosphorylation
`of 4E-BP1 (also designated PHAS-l). This results in a 15 to 20%
`inhibition of overall protein translation and arrests cell cycle
`progression in Gl. Consistent with this observation, mTOR
`controls the synthesis of essential proteins involved in cell cycle
`progression (cyclin Dl and ornithinine decarboxylase) [53,54]
`
`and survival (c-Myc) [55] . A scheme of mTOR-controlled
`signaling pathways based on rapamycin effects is shown in
`Figure 3. 4E-BP1, the suppressor of e!F-4E, has been reported to
`be a direct substrate for mTOR in cells [56,57]. /11 vil.ro, mTOR
`selectively phosphorylates 4E-BP1 at least at two and possibly
`four Ser /Thr residues (Thr37
`, Thr'w', Thr"' and Ser"') in the N(cid:173)
`termina] region [58••,59]. Phosphorylation of 4E-BP1 appears to
`be an ordered process [58••,59,60]. Phosphorylation of Ser"'
`depends on phosphorylation
`of
`all
`three Ser /Thr
`phosphorylation sites [59,60], whereas mutations of Thr37
`and/or Thr46 to Ala(s) prevents phosphorylation of Ser65 and
`Thr"', suggesting that phosphorylation of Thr37 and Thr46 serves
`as a requisite 'priming' event [51•]. It appears that mTOR also
`plays a critical role in regulating the phosphorylation of Ser'"
`and Thr"'. In the presence of rapamycin, 4E-BP1 becomes
`hypophosphorylated and associates with e!F-4E. This prevents
`formation of the eIF-4F initiation complex and cap-dependent
`translation of mRNA.
`
`Figure 3. Rapamycin-inhibited signaling pathways controlled by mTOR.
`
`===Hi=====
`
`IGF-IR
`
`Pl3K
`
`FKB 12-rapamycin
`
`§J--l _P_KB_IA_kt---1~ l
`I Phosphatase I t--8
`/ "Yr~-
`l
`
`1 p70S6
`86 I
`
`4E-BP1
`(PHAS-1)
`
`elF-4E
`
`I
`
`Arrows represent acti~ation , .whereas bars represent inhibition. IGF-IR type I insulin-like growth factor, IRS1 insulin receptor substrate 1,
`P13K phosphat1dyllnos1tol 3-kmase, PTEN phosphatase and tensin homolog deleted on chromosome ten, PKB/Akt protein kinase B, FKB12-
`rapamycin FK-506-binding protein 12-rapamycin complex, mTOR mammalian target of rapamycin, elF-4E eukaryotic initiation factor-4E,
`4E-BP1 (PHAS-1) elF-4E-binding protein 1, SG 40S ribosomal protein, p70S6K p70S6 kinase.
`
`000004
`
`

`

`Inhibitors of mammalian target of rapamycin as novel antitumor agents: From bench to clinic Huang & Houghton 299
`
`Ribosomal p7056K represents the other well characterized
`downstream target of mTOR. Two p7056 kinases have been
`characterized, namely, p7056Kl and p7056K.2. The activation
`of both these kinases can be inhibited by rapamycin [61,62].
`mTOR may directly or indirectly phosphorylate p7056Kl at
`Thr2~J or Thr:ll«' [S0 .. ,63,64,6S .. ,66•,67]. Phosphorylation of
`these two residues is blocked by rapamycin. Furthermore,
`mutation of either of these residues can abrogate the ability of
`rapamycin to inhibit p7056K activation. p7056K functions to
`increase translation of S' terminal oligopyrimidine (STOP)
`for elements of
`the
`tract mRNAs, primarily coding
`translational machinery,
`such as
`ribosomal proteins,
`elongation factors, the poly(A) binding protein [61] and JGF-II
`[68]. Inhibition of mTOR by rapamycin thus selectively causes
`decreased translation of STOP-containing mRNAs.
`
`In addition to pathways controlling translation initiation,
`mTOR has been implicated in regulating the retinoblastoma
`protein (pRb), RNA polymerase (Pol) l/II/llI-transcription
`and translation of rRNA and tRNA, and phosphatases (PP2A,
`PP4, PP6) [69]. It seems that these pathways are cell type(cid:173)
`dependent. For example, in vascular smooth muscle cells,
`rapamycin may act upstream of pRb to slow or arrest cell
`cycle transit [70]. In this model, rapamycin inhibits activation
`of cyclin-dependent kinases
`(CDKs), which results
`in
`hypophosphorylation of pRb protein, and
`inhibits cells
`progressing from Gl to S-phase [70]. ln T-lymphocytes,
`rapamycin induces Gl arrest, in part through inhibition of
`
`activation of CDKl (p34''k2) and the formation of the cyclin E(cid:173)
`p33"'k2 complex [71,72] . Gl arrest by rapamycin may also be
`due to prevention of the degradation of CDK inhibitory
`protein p27K'P' that occurs when cells are stimulated by growth
`factors [73,74]. This is further supported by the observation
`that p27K'P' deficient fibroblasts are somewhat resistant to
`rapamycin as determined by assaying for DNA synthesis [7S] .
`In NIH3T3 cells, rapamycin inhibits the Gl to S transition in
`part through decrease of cyclin Dl mRNA level and protein
`stability [76], or delay of the expression of cyclin A [77].
`
`Antitumor activity of rapamycin
`As previously mentioned, rapamycin has been approved as an
`immunosuppressive drug for organ transplantation by the
`FDA So far, rapamycin has been used clinically in organ
`transplantation with great success, particularly in kidney
`transplantation [78,79]. This is because rapamycin can inhibit T(cid:173)
`cell activation and proliferation. Increasing evidence indicates
`that rapamycin is not only a potent immunosuppressant, but
`also a promising antitumor agent. As reviewed above (Table 1),
`rapamycin potently inhibits the growth of many tumor cell
`lines in vitro, and has demonstrated antitumor activity in both
`xenograft and syngeneic murine tumor models.
`
`However, rather than acting as a cytostatic, rapamycin induces
`cell death under some conditions. Early data show that
`rapamycin induces programmed cell death or apoptosis of B(cid:173)
`cells [24,2S]. Consistent with these findings, recent studies
`indicate that rapamycin alone can also induce apoptosis of
`certain rhabdomyosarcoma cells [14,lS], and monocyte- and
`CD34-derived dendritic cells [80]. When combined with other
`in vitro,
`rapamycin enhanced
`chemotherapeutic agents
`cisplatin-induced apoptosis in human small cell lung cancer cell
`lines
`[11], potentiated
`apoptosis of
`the murine T(cid:173)
`lymphoblastoid cell line S49, induced by dexamethasone [81]
`
`and augmented cisplatin- or camptothecin-induced cytotoxicity
`in DAOY human medulloblastoma cell lines [16]. Rapamycin
`produced additive cytotoxicity with S-fluorouracil and
`cyclophosphamide in a Colon 38 tumor model [7]. At present,
`little is known about the molecular mechanism by which
`rapamycin induces apoptosis of tumor cells. However, Huang
`et al [lS] have observed that the responses of malignant and
`normal cells to rapamycin are qualitatively different. When
`treated with rapamycin, cells with wild-type pS3 arrest in Gl
`phase and maintain viability. In contrast, when grown under
`autocrine conditions (ie, serum-free)
`in
`the presence of
`rapamycin, pS3 mutant cells accumulate in Gl phase, but
`progress to S-phase and undergo apoptosis. More than 90% of
`apoptotic Rh30 cells (mutant pS3 alleles, Argm ~Cys273) were
`BrdU-labeled, suggesting that the cells died after initiating
`replication. Thus, rapamycin-induced death appea1·s to be a
`consequence of continued cell cycle progression, suggesting
`that pS3 senses inhibition of mTOR and co-operates to reinforce
`a Gl arrest. This model has been further tested using Rh30
`infected with adenovirus expressing wild-type pS3 (Ad-pS3)
`and pS3 +/+or pS3-/- murine embryo fibroblasts (MEFs) [lS].
`Restoring the pS3-mediated Gl checkpoint by Ad-pS3 infection
`causes Rh30 rhabdomyosarcoma cells to arrest in Gl and
`prevents rapamycin-induced apoptosis (Figure 4). Similarly,
`when exposed to rapamycin, pS3 -/- MEF cells continue cell
`cycle progression, and become apoptotic, whereas pS3 +I+
`MEF cells arrest in Gl and remain viable. Exactly why cells that
`fail to arrest in Gl in the presence of rapamycin die is currently
`unknown.
`
`Mechanisms of resistance to rapamycin
`As shown by Dilling et al [8], under similar conditions of
`growth, various cell lines demonstrated several thousand(cid:173)
`fold differences in sensitivity to rapamycin. The mechanism
`for this intrinsic resistance is under investigation. Cells may
`also acquire resistance either with or without mutagenesis.
`
`Mutations in FKBP12, mTOR and p7056K
`Budding yeast Saccharomyces
`cerev1siae
`treated with
`rapamycin irreversibly arrested in the Gl phase. However,
`when yeast TORl and TOR2 were genetically mutated to
`TORl-1 and TOR2-1, these strains were completely resistant
`to the growth-inhibitory effect of rapamycin. These resistant
`alleles encode proteins that have reduced affinity for
`binding the FKBP12-rapamycin complex [44]. Also in yeast,
`a recessive resistance phenotype was associated with
`decreased RBPl, a homolog of mammalian FKBP12, or a
`mutation altering Tyr89
`, leading to decreased binding of
`rapamycin [47]. In mammalian cells, resistance to rapamycin
`selected after mutagenesis
`is
`related
`to a dominant
`phenotype consistent with mutation in mTOR [4S]. Similar
`to results in yeast, mTOR mutants are associated with
`decreased affinity for binding of the FKBP12-rapamycin
`complex. High-level resistance to rapamycin is obtained
`when a mutant mTOR (Ser2035 ~Ile2 035), having reduced
`affinity for binding the FKBP12-rapamycin complex, is
`expressed [14,46]. mTOR is essential for activation of
`ribosomal p70S6Kl
`through phosphorylation of
`the
`rapamycin-sensitive sites at Thr229 or Thr389
`[63,64].
`Substitution of either of these residues can also abroga te the
`ability of rapamycin to inhibit p70S6K activation. Whether
`this results in resistance to the growth inhibitory effect of
`rapamycin is less clear, and may be cell context-specific.
`
`000005
`
`

`

`300 Current Opinion in lnvestigational Drugs 2002 Vol 3 No 2
`
`Figure 4. Protective effect of the tumor suppressor p53 on rapamycin-induced apoptosis.
`
`QJ
`
`QJ
`
`() c
`() en
`~
`0
`::::I
`:;::
`QJ
`32
`"C
`.Q
`E
`::::I
`'6 ·o.
`e a..
`
`Control
`
`Rap 100
`
`Ad-~gal + Rap 100
`
`Ad-p63
`
`Ad-p63 + Rap 100
`
`101 102 103 104 100
`
`101
`
`102
`
`103
`
`104
`
`Annexin V-FITC fluorescence
`
`Rh30 rhabdomyosarcoma cells were infected with either adenovirus expressing wild··type p53 (Ad-p53) or Ad-~-galactosidase (~-gal) at a
`multiplicity of infection of 1. After 24 h, medium was replaced with serum-free N2E, and cells were grown for a further 6 days in the absence
`or presence of 100 ng/ml rapamycin (Rap 100). Cells were harvested and apoptosis determined by quantitative FACs analysis (ApoAlert).
`Left panels show dual staining for propidium iodide uptake and annexin V-FITC. Right panels show corresponding distribution of annexin V(cid:173)
`FITC staining in populations of cells. (Adapted from Huang et al [15].)
`
`Decrease of 4E-BP1 protein expression
`Mechanisms of acquired resistance (without use of mutagens)
`or intrinsic resistance have received less attention. Murine
`BC3Hl cells selected for acquired resistance demonstra ted
`reduced levels of p27K'"', and consistent w ith this, embryo
`fibroblasts with disrupted p27K;pi are relatively resistant to
`rapa mycin, as determined by inhibition of DNA precursors
`[75]. These data are consistent with Gl arrest, being in part
`due to rapamycin stabilizing this cyclin-dependent kinase
`inhibitor protein. Recently, rapa mycin-resistant cell
`lines,
`Rh30/ Rapa10K and C2, have also been obtained by growing
`
`Rh30 parental rhabdomyosa rcoma cells in the continuous
`presence of increasing concentra tions of rapamycin, without
`prior mutagenesis (82, Huang & Houghton, unpublished
`data]. When resistan t clones were grown w ithout rapamycin
`for 6 or 10 weeks, they reverted to rapa mycin sensiti vity (in
`terms of IC50 by growth in hibition assay). Thus, acquired
`rapamycin resistance was unstable.
`
`Analysis of several clones revea led increased levels of the c(cid:173)
`Myc protein. Of interest, these rapa mycin-resistan t clones
`exhibited increased anchorage-independent grow th in soft
`
`000006
`
`

`

`Inhibitors of mammalian target of rapamycin as novel antitumor agents: From bench to clinic Huang & Houghton 301
`
`aga r. Consisten t with increased c-Myc, the levels of the 4E(cid:173)
`BP suppressor proteins bound to eIF-4E were significantly
`lower (approximately 10-fold), as were total cellular levels of
`4E-BP proteins. Steady state levels of 4E-BP transcripts
`remained unaltered, however, the rate of synthesis appeared
`to be decreased in rapamycin-resistant clones [Huang &
`Houghton, unpublished data ]. ln clones that reverted to
`rapamycin sensitivity, total levels of 4E-BP1 became similar
`to those in parental cells. ln some cases, intrinsic resistance
`also appears to relate to low 4E-BP:eIF4E levels. ln colon
`carcinoma cells, very low levels of 4E-BP were detected,
`whereas eIF4E levels were similar to those in sensitive
`tumor cell lines. In contrast, no significant changes were
`determined for p70S6 kinase levels or activity between
`parental and resistant clones.
`
`Development of CCI-779 and RAD-001 as
`antitumor agents
`CC/-779
`rapamycin-42, 2,2-
`inhjbitor-779;
`(cell cycle
`CCI-779
`is
`an
`ester of
`bis(hydroxymethyl)-propanoic acid)
`rapamycin (Figure 1), which was developed by Wyeth(cid:173)
`Ayerst as an antitumor agent. Like rapamycin, CCl-779
`by
`inhibition
`of mTOR,
`preventing
`the
`acts
`phosphorylation of 4E-BPs and p70S6K
`[20-22,26].
`However, in contrast to rapamycin, CCI-779 is stable in
`aqueous solution, hence
`it can be
`formulated
`for
`intravenous administration. In preclinical tests, CCI-779
`possesses similar antitumor profiles to rapamycin [16,20-
`22,26,28•,29• ]. CCI-779 potently
`inhibits growth of
`numerous cultured human tumor cell Jines including
`human breast, prostate, pancreatic and small cell lung
`carcinomas, glioblastoma, medulloblastoma, melanoma,
`rhabdomyosarcoma and T-cell leukemia, with IC50 values
`in the nanomolar range [16,20-22] . These observations have
`been further supported by the significant inhibitory effect
`of CCI-779 on growth of human tumor xenografts in
`athymic nude mice [16,20,21,24]. When combined with
`cisplatin or camp tothecin, CCI-779 showed additive
`cytotoxicity in subcutaneous implants of human brain
`tumors [16]. Interestingly, in vivo CCI-779 also inhibited
`the growth of human U251 malignant glioma cells that
`were resistant
`to
`rapamycin
`in vitro, although
`the
`mechanism is unknown [16]. ln addition, more recent
`results have
`revealed
`that
`the
`tumor
`suppressor
`phosphatase and tensin homolog deleted on chromosome
`ten (PTEN)-mutated or -deficient cancer cells are more
`sensitive to CCI-779 [21,28•,29•]. PTEN acts as a major
`negative regulator of the PI3K/ Akt signaling pathway [83-
`85]. Loss of PTEN by deletion or mutation occurs in as
`many as 50% of all solid human tumors [86], resulting in
`activation of Akt. Conceptually, this could activate mTOR(cid:173)
`dependent pathways, hence forming the basis for CCl-779
`hypersensitivity of PTEN deficient cells. However, it is
`unclear if Akt activates mTOR i11 situ, as mutation of
`putative Akt phosphorylation sites in the C-terminus of
`mTOR does not abrogate insulin stimulation of p70S6K
`activation [87]. Of considerable interest, however, is the
`report by Podsypanina et al [28•] showing inhibition of
`in PTEN +/- mice. These
`neoplastic
`transformation
`anima ls develop spontaneous multifocal complex atypical
`hyperplasia
`in
`the uterine secretory epithelium
`that
`progresses
`to neoplastic
`transformation. Tumor cells
`
`demonstrate elevated levels of phosphorylated Akt and
`activated p70S6K. While CCI-779 had no effect on Akt
`activation, as anticipated, it normalized p70S6K activity.
`Whether, loss of PTEN function consistently sensitizes
`tumor cells
`to
`rapamycin analogs
`remains
`to be
`d emonstrated.
`
`These preclinical results have revealed that CCl-779 exhibits
`in some instances, cytotoxic
`impressive cytostatic, and
`properties, and may be valuable to delay tumor progression
`and to improve survival when used alone, or in combination
`with other chemotherapeutic agents.
`
`Early data from phase I trials have shown promise of CCI-
`779 in treatment of some cancers [88•]. In European phase I
`clinical trials [89], CCl-779 was administered as a weekly 30-
`min intravenous infusion in 18 patients with different types
`of advanced solid tumors. Doses ranged from 7.5 to 220
`mg/m2 /week. After ;::.: 8 weekly doses, significant tumor
`regressions were observed in two patients (receiving 15
`mg/m2 /week) with lung metastasis of renal cell carcinomas
`and in one patient (receiving 22.5 mg/m2 /week) with a
`two
`neuroendocrine tumor of the lung. Additionally,
`patients experienced
`tumor stabilization. It
`is unclear
`whether the very high dose levels used in this study are
`necessary to elicit antitumor activity. Considering the high
`plasma levels of CCI-779 measw-ed in patients (1000-fold
`greater than required for rapamycin activity in vitro), it is
`conceivable that CCI-779 is acting through a secondary
`mechanism independent of mTOR.
`
`In the US, phase I clinical trials were also conducted to
`determine the safety and tolerability of CCl-779 [90]. CCI-
`779 was administered as a daily intravenous 30-min infusion
`for 5 days every 2 weeks. Of 45 patients with various types
`of cancer, nine achieved some evidence of tumor response.
`Importantly, results from the above two trials indicate that
`CCl-779 is well tolerated in patients with only mild side
`effects, such as acneform rash, mild mucositis, some
`thrombocytopenia, and elevated triglyceride and cholesterol
`levels. Rapamycin and its analogs potently inhibit T-cell
`activation, thus the potential for immunosuppression in
`patients treated with CCI-779 was anticipated. Interestingly
`modest
`evidence
`of
`CCl-779
`induced
`only
`immunosuppression. Rapamycins also
`inhibit
`insulin
`signaling, but in the clinical trials reported there was no
`toxicity consistent w ith inhibition of insulin signaling. CCI-
`779 is cw-rently in phase Ill clinical trials. In athymic nude
`mouse xenografts, apart from breast and prostate cancers,
`human glioblastoma
`(U87MG)
`tumors and pancreatic
`carcinoma were also very sensitive to CCI-779. Thus, with
`an expanded focus, it is probable that clinical trials will
`identify other tumor types that are sensitive to CCI-779.
`
`RAD-001
`is another
`(40-0-(2-hydroxyethyl)-rapamycin)
`RAD-001
`rapamycin analog that is being developed