-[Cnnccr I§i0lo_gy
`
`Thertipy Z5, 4i7l~1t7(). Sepictnbciv‘October 30(lfil; ©I.’,0ll3 1,/(1“¢icS Biusciencc
`
`Review
`
`Rhebbing up mTOR
`New Insights on TSCI and TSC2, and the Pathogenesis oi Tuberous Sclerosis
`
`David J. Kwiutkowski
`‘Correspondence to: David J. Kwiutlvtowslti, M.D., PIt.D.; Brigham K Women‘s
`
`Hospital; 22l Langwoud Ave, LMRC 302,‘ Boston, MOSSD(hUSBllS 021 I5 USA; Tel.
`617.278.0384; Fax: 6I7.734.2248; Email: ill<@rirs.liwlt.lmrvurd.eriu
`
`ABSTRACT
`
`Tuberous sclerosis is an autosomal dominant human genetic disorder in which distinctive
`tumors called hamartomas develop. Germline mutations in either TSCl or TSCQ cause
`this syndrome, and hamartomas typically display second hit events with loss oi
`the
`remaining normal allele. Studies initiated in Drosophila have identilied a role for the Tscl
`and Tsc2 genes in the regulation at cell and organ size, and genetic interaction studies
`have placed them in the Pl3K~Al<t—mTOR~SoK pathway. Biochemical studies have shown
`that activated Al<t phosphorylates TSC2 in the TSCi /TSC2 protein complex, inactivating
`it; while TSCI/TSC2 has GAP activity For the Rheb GTPase la member oi the ras lam-
`ily), and activated Rheb-GTP activates mTOR. Thus,
`in cells lacking TSCI or TSC2 there
`are increased levels of Rheb—GTP which leads to activation of mTOR, leading to cell size
`increase and growth. These developments provide enhanced understanding oi this
`signaling pathway and fundamental insights into the pathogenesis of tuberous sclerosis,
`; and o en the
`ossibili
`of
`treatment
`ior hamartomas In
`several
`harmacolo ic
`P
`p
`y
`P
`9
`.
`i approaches.
`3
`
`Received 09/0 I /03; Attepted 09/02/03
`
`Previously published unline as ti (B&T Epubliwtion at:
`http://www.Iundesliiosrienretum/journals/tlit/tot.pl1p?volume:2&issue:5
`
`tuberous sclerosis, TSC, TSCI, TSCZ, mTOR,
`Rheb. rapamycin, hamartoma
`
`ACKNOWLEDGEMENTS
`
`Supported by the NIH i\ilNDS (N331 535,
`the LAM lioundation,
`and the
`Rothherg Courage Iiund. The author ;1pologi'/res to 3
`those whose work was not cited here due to space 3
`f TUBEROUS SCI.EROSIS--THE CLINICAL SYNDROME
`
`‘
`
`
`
`Tuberous sclerosis (TSC) is an autosomal dominant genetic disorder charactcrr/.cd by
`development of unusual tumors in several organ systems.’ The tumors QTTSC are termed
`lramartomas to indicate that there is overgrowth at‘ relatively mature appearing cells, in dis-
`tinct contrast to common tmiligimticics. TSC liariizirtornas involve the skin, brain, and kid»
`neys in most patients. The vast majority ot‘TSC lianiartonias display limited growth
`potential and do not require intervention. A small traction display persistent growth,
`necessitating surgical control. However, progression to malignancy is very rare in TSC, and
`has been seen only in TSC renal hamartomas, termed angiotnyoliopoma (AML), at 21 lirc-'—
`quency ofabout 1.5-2‘)/0 at‘ all TSC patients.2
`Another retnarkzible lcati.ire ot7TSC hamzirromzis is that they first appear at dilftierent
`; ages during the parici'1t’s lite, and occasionally spontaneously rcsolvepl For example. cardiac
`‘
`rhabdomyomas are often present at birth, but then typically disappear during childhood.
`The major morbidity ol7TSC is due to the hallrnzirk cortical tubers, from which is derived
`the natne tuberous sclerosis (Fig. 1). Cortical tubers are focal regions at‘ cliSOI'g21t‘1l'/.(’(l cor-
`tical lntnination that contain both giant cells (80 to 150 microns in dizunetcr) and dys-
`plastic neurons. both oliwhich have disrupted radial oricritzttiorisi A TSC patient may have
`as many as 50 cortical tubers. Although they are thought to be relatively static lesions, they
`lead to seizure disorders in the vast majority OFTSC patients. and commonly contribute
`to developmental delay, mental retardation. and autism.
`Another unusual ieeiture ofTSC is the occurrence of‘ 21 unique praliterzitive lung disease
`termed Iymphangioleiomyomzitosis (LAM)
`in which there is both cystic change and
`; smooth muscle cell (SMC) proliferzitionl This lesion is seen nearly exclusively in adult
`female TSC patients, suggesting that estrogcns contribute to its development.
`
`TSC GENES, MUTATIONS AND THE TWO HIT MECHANISM
`
`TSC occurs due to inactivating mutations in either of‘ TSCI or T§(:2./*"'('lTl1cse genes
`are both relzitively large (25 and 41 exons, respectively), and encode the proteins lmmzirtin
`(l30kDa) and tuberin (200l<Da), respectively. Due to the frequent severe clinical effects of‘
`TSC, large TSC tlimilies are rare, and about 2/3 ot‘TSC patients are sporadic cases due to
`i new mutational events. Comprehensive mutational atulyses indicate that about 85% oil
`
`Roxane Labs., Inc.
`Exhibit 1018
`Page 001
`
`

`
`RHEISBING UP MTOR. ROLES O17 TSCI AND TSC2
`
`
`
`Figure l. Brain lesions in tuberous sclerosis. A. Brain MR) of an 8—yeamld
`boy using the FLAIR technique demonstrates several cortical tubers (white,
`cortical lesions, indicated by arrows) and subependymal nodules lining the
`ventricles (arrowheads). Courtesy of J. Egelhoff, Cincinnati, OH, B. High
`power view of a cortical tuber shows giant cells (white diamonds). Cortical
`surface is to the right. Courtesy oil. Chan, Boston, MA.
`
`rate.("S Clinical surveys also indicate that patients with TSCI
`mutations have symptoms and clinical features that are milder on
`average than patients with TSC2 mutations, although there is
`considerable overlap in the two sets of patients.8 Linkage studies
`provide evidence against a third TSC gene.6 Patients in whom
`mutations cannot be found are likely partially explained by
`mosaicism, which compromises efforts at mutation ClCtCCtiOn.8
`TSC harnartomas, particularly renal AMLs, often display loss of
`heterozygosity (LOH) for the wild type allele of either TSC1 or
`TSC2, consistent with a two hit mechanism for complete inacti—
`vation of either TSCI or T5C2.f"()"0 LAM lesions require micro—
`dissection of the SMCs to demonstrate LOH events and such methods
`have been used to show that patients with LAM but without other
`features ofTSC also have two hit inactivation ofTSC2.” Brain
`lesions have failed to showed evidence of LOH,9 even with micro-
`dissection,” and alternative mechanisms of gene inactivation or cell
`admixture likely explain this finding. LOH is also seen in the tumors
`(kidney cystadenomas, liver hemangiomas, and extremity angiosar—
`comas) that develop in T5c1+/— and Ttc2+/— mice.)-M5
`Consistent with a relative lack of malignancy in TSC patients,
`limited surveys of human cancer specimens have failed to show
`evidence of either TSC1 or TSCB mutation, with the possible
`exception of bladder carcinoma.l("l8
`
`TSCI /TSC2, Pl3K AND mTOR SIGNALLING
`
`Major progress has occurred in recent years in our understanding
`of the biochemical function of the TSCI and TSC2 gene products,
`tuberin and hamartin. These advances were initiated by seminal
`studies in Drosophila, which led to placement of TSCl and TSC2
`in the Pl3K—Al<t-mTOR—S6K signaling pathway. Here we review
`this work, emphasizing the role of the TSC1 and TSCQ genes in this
`pathway, and indicating the complementary nature of investigation
`carried out in Drosophila and in mammalian systems.
`The initial studies used FRT/FLP recombination to identify T551
`and T12 as targets for recessive mutations that affect fly eye size.l9’2Z
`Cells homozygous null for either Ft‘! or 7}(‘..) develop and differentiate
`
`lnR
`
`Pi3K92E
`
`%—— Pten
`
`Akt 7
`
`Tsc 1/ Tsc2
`
`Rheb
`
`Tor
`
`Figure 2. Genetic model for the control of
`cell and organ size in Drosophila. Genetic
`epistasis experiments provide support for
`the model
`that
`is
`shown (see text
`for
`details). Arrows indicate positive actions
`and bars
`represent negative actions.
`Although
`interaction
`between
`each
`sequential pair of genes in this pathway is
`clear,
`the model
`falls down when more
`extended comparisons are made (e.g.,
`results of Aktl and Sok overexpression are
`not additive.57). This
`likely reflects that
`each of the genes in this pathway have
`functions other than controlling the gene
`shown immediately below it, and there is
`feedback compensation throughout
`the
`pathway when one element is perturbed.
`
`mammalian cells indicating that Tscl
`and Tsc2 occur as a con1plex,33*24
`overexpression of both proteins but
`not either alone led to an opposite
`phenotype in which cell and organ
`size is reduced.
`
`These observations led to genetic
`interaction studies between 731/ and
`7}c2, and elements of the Pl3K—Akt-
`
`56/(
`
`mTOR—S6K pathway which had pre—
`viously been recognized to have an
`important role in the control of cell
`size in Drosophila.l9‘2l‘25‘l6 Ablation
`of F6] or T562 had effects that were
`dominant
`(epistatic)
`to
`those of
`homozygous loss of the insulin recep-
`tor, l’l3K, and Aktl. Loss of Tsirl or F52 was synergistic with loss of
`Pten. leading to further enlargement ofthe eye. On the other hand,
`loss of 56k30 or 7?)r23 was dominant in effect to loss of 7}(/ or TJI2.
`7}c1'/"7br"/“ embryos survived longer
`than T501’/‘ embryos, and
`organs with 7}c/’/‘Tor’/‘ cells were more normal
`than those with
`P614" cells.35 T561‘/’77)r*/‘56k*/’ embryos survived to adulthood and
`were semifertile.37 These studies positioned 7}r1/ Tt2 just above 7227‘
`in the Pl3K—Al<t—Tor-S6K signalling pathway (Fig. 2).
`These genetic studies were complemented by biochemical studies,
`which provided a partial mechanistic basis for the observations. Tsc2
`is phosphorylated by Aktl. and this event leads to the disruption of
`the Tscl-Tsc2 complex in vivo.3(’ In addition, expression ofa mutant
`Tsc2 which cannot be phosphorylated by Akti along with wild type
`Tscl completely suppressed the effects of Akt overexpression on
`cell/organ size whereas wild type Tsc2 did not have this effect.2(’
`Studies utilizing mammalian cells were performed in parallel,
`after the initial Drosophila findings. To understand these results we
`review the l’l3K—Akt—mTOR—S6K pathway in mammals (Fig.
`;3).38'3ll
`When recruited to the cell membrane by binding to activated, phos—
`phorylated growth factor receptors or through interaction with ras,
`Pl3K catalyzes the synthesis of 3' phosphoinositides. Those lipids
`then recruit Akt to the membrane where it is phosphorylated on two
`sites, one site by either autophosphorylation or another kinase
`(S473), and the second by the PDK1 kinase (T308). Akt then
`phosphorylates multiple substrates involved in regulation of apop—
`tosis, cell energy metabolism, and the cell cycle. including Bad,
`procaspase—9, IKK, CREB, FKHR/AFX/FOX transcription factors,
`
`Roxane Labs., Inc.
`Exhibit 1018
`Page 002
`
`

`
`RHEBBINC} UP :\/ITOR, ROLES Oi’ TSCI ANI) TSC2
`
`Pt(4)P
`PE(4,5)P2
`
`—+
`
`P|{3t4)P2
`Pt{3,4=5)P3
`
`
`
`——*
`
`Pl{4)P
`Pl(4,5}P2
`
`
`
`Rheb—GTP
`
`l, Pi
`—~——~——é~>-
`
`cc» ‘.7C7 U"U
`It.2‘
`
`ATP
`
`it
`
`PA
`
`vi
`t /‘AA
`
`one are
`
`@ 4
`
`other
`% \‘+P
`/ kinases
`A‘
`t‘*’
`
`l
`
`i
`protein translation, cell growth
`
`
`
`
`
`Figure 3. Signalling pathway model tor the function at TSCl and TSC2 in mammalian cells. A phosphorylateol growth toctor receptor is shown at upper
`lett, to which a P|3K molecule is binding. This leads to conversion at the indicated phosphoinositides to 3’—phosphoinositides, which leads to recruitment oi
`Akt to the membrane in a position where it can be phosphorylated and activated by PDKl and a second ltinase. PTEN tunctions to terminate this signaling
`pathway by acting as a 3' phosphatase on these phosphoinositides. Activated pAl<t phosphorylates TSC2 which inactivates its GAP activity. When active,
`TSCl/TSC2 complex serves as a GAP tor Rheb, reducing levels at Rheb~GTP. Rheb—GTP activates mTOR by an uncertain mechanism (thus 2 arrows). ATR
`phosphatidic acid (PA), and amino acids (AA) all intluence mTOR activity, although the sensing mechanisms are unknown and liltely indirect. Active rnTOR
`phosphorylates 4E-Bljl and SoKl. p4E-BPl
`releases tram elF4E, permitting formation oi the elF4F translation complex. pSoKl phosphorylotes so, and
`together they activate the translational machinery. For simplicity only the main pathway involving TSCl, TSC2, and mTOR is shown. Arrows indicate posi—
`tive actions and bars represent negative actions.
`
`lation site identified (S2448). Regulation of mTOR kinase activity
`is complex, and recent advances have highlighted its occurrence in a
`large complex including at least two binding partners: raptor and
`G[3L.31”33 mTOR’s kinase and autokinase activity are modulated in
`cells both by growth factor receptor kinase signalling and nutrient
`(amino acid) levels, ATP levels, and phosphatidic acid levels. mTOR
`two principal downstream targets: 36K]
`and /+E~BPi.
`Phosphorylation and activation of SGKI occurs in a multi~step
`process with several kinases (PDKL ERK) involved, with phosv
`phorvlation by mTOR
`a critical step at T389. Activated S6Kl
`phosphorylates the rihosomal subunit 56 leading to activation of‘
`protein translation, specifically mRNAs that contain a 5’
`tract of"
`oligopyrimidines, including many ribosomal proteins and translation
`Factors. Unphosphorylated 4E—Bi’1 binds to and sequesters the
`translation initiation factor elF4E. Phosphorylation of -4E—BPl
`
`translation ()fOrl1i{l1iI1€ dec;2u'b()Xyl21se, c—myc, hypoxia—inducible fac-
`tor lalpha, and a number of cell cycle proteins.7‘°
`in cultured mammalian cells lacking either Tscl or ”'l“sc2, there is
`constitutive high level pltosphotylation of S6Kl, 4E~BP1, and (in
`neuroepithelial cells) STAT3.34"39 (Similar
`results are seen in
`Drosophila S2 cells.33) Treatment of these cells with rapamycin, the
`highly specitic inhibitor of mTOR, rapidly reverses this phosphor)!-
`lation signature. Activation oi‘ S6Kl
`in these cells is also relatively
`resistant
`to withdrawal of amino acids
`tron)
`the culture
`media.
`2537"” Concordant with this activation of‘ mTOR in the
`absence oli either Tscl or Tsc2, there is an increased 3 pltcise per cent
`and growth rate in Tscl and 'l‘scZ null cells compared to controls
`under some CflndlIi()!1S.3S"V’3() Phosphorylated S6 is also prominent
`in these cultured cells.
`lmmunohistochemistry analyses have cone
`firmed that this signature oi” mTOR activation is present in several
`
`Roxane Labs., Inc.
`Exhibit 1018
`Page 003
`
`

`
`l{Hl;'l3l5Il\l(§ U1’ .\/l'l‘()R, l{(i)l.l.{.‘§ (>F'l'S(IE z\I\’l)'1'S(i2
`
`Mammalian Al<r phosphorylates Tsc2 at 2-4 critical sites, including
`S939 and Tl462,3l5'42’/*3 and in one study overexpression of’ a
`dominant active Akt led to accelerated degradation oFTscl and Tsc2
`by a ubiquitination/proteasome patliway/*2 l\/lutation of S939 and
`TM62 of Tscl
`to A. which prevents phosphorylation, enhances
`inhibition of 56K activity by Tsc2 in transient transfection assays,
`while mutation oiithese sites to D, which mimics phosphorylation,
`renders Tsc2 less effective at inhibiting S6K than wild type "l‘sc2.~i8/6
`mTOR kinase and autokinase activity parallels the degree ofiactiva»
`tion of SGKI
`in these assays.” in addition, overexpression oli
`Tsci/'Ilsc2 in similar assays leads to reduced phosphorylation of
`/iE~BPl, and blocks amino acid~induced phosphorylation of
`
`ENTER RHEB
`
`Thus, the above studies determined that Tscl/Tsc2 was a critical
`intermediate in the signalling pathway from l’l3K to mTOR and
`downstream elements, serving as a brake upon mTOR activity.
`l“lowever, the mechanism or this effect was unknown until recently
`when multiple investigators discovered that Tscl /Tsc2 functioned as
`a CTl’ase activating protein (CAP)
`for
`the heretofore relatively
`obscure but evolutionarily conserved member of‘ the rzis
`family,
`Rheb. Studies in yeast had shown that loss of Rheb mimicked nitro—
`gen starvation with G0/G] arrest,/"5 suggesting a potential role in this
`pathway. Rheb is unusual in comparison to other ras family mem-
`bers in that it has an arginine at the third residue of the G1 box,“
`and occurs in cells with relatively high amounts of bound GTP47
`Genetic screens in the fly led to identification of 19/26!) as a gene
`that caused eye (and other organ) enlargement when its transcrip-
`tion was increased and eye size reduction when it was homozygous-
`ly ablated./$50 Rheb overexpression has effects on cell size and cell
`cycle that are entirely similar to the effects olwliscl or Tsc2 lt)ss.43’50
`Genetic interaction studies show that
`loss of Tor is dominant to
`overexpression of“ Rheb, while loss of
`attenuates but does not
`fully block eye enlargement mediated by Rl1C‘.b.48’49 Further, overex~
`pression oi”Tsci/Tsc2 blocked the cell size effects of overexpression
`of Rheb, while loss of Rheb was dominant in effect to loss ofTscl or
`Tsc2/*9 In addition, reduction in Rheb expression partially or com—
`pletely rescued the embryonic lethality of loss of 73‘r].‘*9’5l in vitro
`analyses indicated that R,heb—GTP levels, Rheb stimulated S6l< acti-
`vation, and cell growth were independent of nutrient availabi ity,
`while loss oiiRheb completely blocked S6l< actixration.”/’8‘495l
`This W()Tl{ was again complemented and extended by concurrent
`studies on mammalian cells. Rheb—CTP levels were increasec in
`T56?” cells compared with controlssl
`ln transient
`transiec ion
`assays. coexpression of Tscl/Tsc2 markedly lowers the leve
`of‘
`Rheb~GTT’ and SGKI activation.5l”5” in l\’lH3T.°> cells,
`insulin
`stimulation leads to a l.8—lold increase in Rheb—GTP levels which is
`blocked by wortmannin (l’l3K inhibitor) but not
`rapamyci
`.52
`Overexpression ofiRheb leads to high level activation 0? S6Ki and
`phosphorylation of‘ 4E—BP1, while reduction in Rheb levels by
`siRNA reduces growth liactoninduced activation of S6K1.52‘5(’ This
`activity is completely independent of Pl3K and MAPK activation or
`function, as assessed by inhibitor treatmer1ts.53>54'5(’ In addition,
`rapamycin but not wortmannin completely inhibits the phospho—
`rylation of‘ SGKI
`induced by Rheb overexpression,535353 as does
`expression olia l<inase—dead mTOR,S3 and treatment with the Farrie-
`syl translerase inhibitor FTl—277.<’6 Nutrient deprivation had little
`
`hand, a dominant negative Rheb (D60K) blocked activation of
`SGKI by amino acids.”
`in vitro studies also demonstrated that purified Tscl/Tsc2 had
`specific and high level GAP activity for the Rheb GTPase that
`required both proteiris.53'(’4 Mutation of Rheb to block prenylation
`at its C~terminal reduced but did not eliminate its ability to stimu»
`late S6Kl activity in transient expression ;1ss;1ys.T’
`Thus,
`these studies in mammalian cells lead to the model oi’
`l’l3K—Al<t—TSC1/TSC24Rheb—mTOR signaling that
`is shown in
`(Fig. 3)
`ln normal quiescent cells,
`l’l3K and Akt are inactive, the
`TSCI/TSC2 complex is active as :1 CAP For Rheb, there are low lev-
`els of Rheb—GTl’, and mTOR is inactive. In response to growth fac-
`tor stimulation, PBK and Akt become activated, TSC2 is phospho-
`rylated by Altt and the TSCl/TSC2 complex becomes inactive as a
`CAP, so that Rheb—CTP levels rise, stimulating mTC)R. In cells lacle
`ing TSC1 or TSC2, there is no CAP for Rheb, and Rheb'GTl’ lev-
`els are high. leading to constitutive activation of mTOR and phos-
`phorylation oiiS6Ki and 4F,—BPl. This linear model captures most
`ol7 the reported results but simplifies many aspects of" this pathway,
`including the multiple targets oFr\l<t, and the complex hierarchical
`phosphorylation oiS(,iKl that regulates its activity, as just two exam«
`ples. In addition, not all the data from mammalian systems supports
`the above model.37 Moreover,
`there is compelling data from
`Drosophila that S6l< activation is under separate control from both
`I’l3K and Aktl, suggesting that
`there are parallel pathways oli
`Pl3K—Al<tl and mTOR—S6l< that converge and interconnect rather
`than existing in a linear sequence.49'57 However, this signaling circuit
`is certain to have complex Feedback and compensatory mechanisms
`built in, which can make interpretation of the effects of‘ loss of" one
`gene on other members of the pathway difficult. For example, both
`Tscl and Tsc2 null cells show a profound reduction in activation of‘
`Altt in response to serum and other growth l'E1ct’ors,37'3il This appears
`to be due to a major reduction in PDGFR expression in these cells.”
`These observations may explain why the TSCI and TSC2 genes are
`rarely involved in malignancy.
`
`OTHER FUNCTIONS OF TSCI/TSC2
`
`The TSCI/TSC2 complex has size 330l<Da and the GAP domain
`OFTSCZ comprises about i0 l<Da of this complex. This alone sug-
`gests that there are other functions of the complex. The only other
`consistently reported binding partner of the TSCi/TSC2 complex
`isl4~3—3, through p51 21 0.39'5859 The importance otthis binding is
`uncertain, but it may regulate the function oii'l‘SCl/TSC2.”
`TSCI has been reported to bind to e’/.rin and other ERM Family
`proteins, and appears to be involved in adhesion events and rho sig-
`naling to the actin cytosl<eleton.(’ll TSC2 has been reported to have
`a role in the membrane localization oil polycystin~l
`in renal epithe-
`lial cells.“ A role for the ’liS(i.‘l/’li‘SCZ complex in beta~catenin sig
`naling has also been noted,“ \Whether any of‘ these observations are
`independent oli or relate to the role OTTSCI/TSC2 in the Plf5K sig—
`naling pathway is unknown.
`
`THERAPEUTIC PROSPECTS
`
`Three years ago treatment for the various manifestations ofTSC
`appeared an unreachable goal. There are now several
`treatment
`approaches under active consideration: rapamycin and analogues,
`larnesyl
`transierase inhibitors (FTls), angiogenesis inhibitors, and
`irite*rleroi'i—y (IFNV). Rapamycin is uniquely attractive as a therapeutic,
`
`Roxane Labs., Inc.
`Exhibit 1018
`Page 004
`
`

`
`RHEBBING U1" MTOR. R(,)l..F.S OF TSC1 A.\Il') TSC2
`
`Analogues CCL779 and RADOOI with improved pharmacologic
`properties are also available. Longer term trials with CCl—779 and
`rapamycin are underway in several TSC mouse models, and clinical
`trials using rapamycin in TSC patients are at an early stage (Bissler
`and Franz, personal communication). Given the requirement for
`membrane locali"/.ation for Rheb hinction, FTls are also an attractive
`
`approach, particularly since Rheb blockade may explain the in vivo
`mechanism of action of" l3Tls which do not appear to liunction by
`inactivation of ras.i[l'5(’ Several TSC hamartornas are characterized
`by aberrant vascular channels, which appears to be due to expression
`of VEGF by cells lacking Tscl or T.sc2.(’4“70’7l Thus, anti—angiogenic
`therapies could also be valuable in TSC. Finally, the increased STAT
`expression and activation in Tsc null cell
`lines,36 as well as clinical
`and mouse observations suggest that lFNy may be therapeutic in
`TSC.65‘(’(’ Trials of IFN—y treatment in TSC mouse models are also
`
`A critical clinical issue in tuberous sclerosis is the molecular basis
`of epileptogenesis in TSC. Alterations in expression of several neu-
`rotransmitter receptors have been seen in tuber giant cells compared
`to normal neurons, and are likely to underlie this problem.67 mTOR
`has been shown to regulate protein synthesis at the synapse,68 and be
`involved in memory formation, so that it seems likely that aberrant
`mTOR activation in tuber giant cells accounts for their large size
`and abnormal expression patterns.
`\3(/ihether rapamycin and ana-
`logues might be effective in control of the CNS manifestations of
`TSC is an exciting though perhaps distant prospect. Nonetheless,
`rapamycin is effective in extending survival in Tsc mouse brain mod«
`els (Onda, Meil<,le, Kwiatl<owsl<i, unpublished observations).
`
`CONCLUSIONS AND FUTURE PROSPECTS
`
`in the field oliTSC
`Now is 9. time of tremendous excitement
`researcl , given major progress in the identification of the signaling
`pathway the TSC proteins participate in, and the discovery olisever»
`al potential therapeutic approaches. The intense scrutiny brought to
`beat on these genes is very likely to yield significant further advances
`in the near future. Questions that seem particularly important or
`fruitful at this time include the following. \X/hat is the mechanism
`by which Rheb—GTP influences mTOR activity? Since rapamycin
`rapidly dephosphorylates SGKI, S6, and /iE—BPl
`in
`cells,3“‘39 is phosphatase regulation in mammalian cells under the
`control of mTOR as suggested by some studies?(’9 Are all of the
`reported cellular findings in the absence oliTSCl or TSC2 due to
`the activation of Rheb and mTOR, or does the complex have addi-
`tional Functions? Will rap-amycin or its analogues, FTls, angiogene—
`inhibitors, or
`IFNy prove to be of clinical benefit
`to TSC
`
`References
`
`i:.I‘lgl3lI(lZ
`
`(Soinez M, Sanipsou V], \Vhitternore V. The tuberous sclerosis complex. Oxrrord,
`Oxlord University l’re.~s, I999.
`til.
`er
`Dreyer
`Al—Saleeru T.
`\Vessner
`l.l,, Sclieitliaoer B\\'-', Rutersoii K. Roach
`,’\/lztligriaiit tumors oliihe kidney. brain, and soli tissues in «children and young adults with
`the tuberous sclerosis Complex. (jziiitter l‘)‘,>8; 83:2?.ll8~‘;(>.
`llenske El’. New tlexelopriiems in the iieurobiology or the tuberous sclerosis
`ieurology H9‘); ‘~'.’»:l.%8«'i—‘)<}.
`ion and ch.ir:it‘«
`l(lc‘H{ill
`-C‘. The lltiro;>e;ir
`.:hroinosoi‘.ie lo tuberous sclerosis {,:oiisot'tiurn,
`
`ll ill3—l 3.
`teri.«:atis>ii oi the tuberous \(lU|'U3l.\ gene on Cl}l‘()fH0,sLiIYiL‘ in. Cell
`l‘)‘)%,
`.
`lLil\SSci\ B, Vcrlioeii 5, et al
`vzin Slegtcrihorst M, de lloogr R.
`l"lerni.ins C,
`.\'ulli\t
`.\l
`ltlentilicatiori of the tuberous \CiCR)Sl's gene T$'€,'i
`on cliromosome ‘)q.5-’i. iiticrice W97;
`
`(1. Lhezulle ll‘. Reeve Ml’, Sampson IR, l\'wi:itl<owsi<i D]. Molt\‘ul;u‘ genetic :ulv'.incc.s in tuber-
`ous sclerosis. Hum Genet 2000:
`i07:*}7/l l/i.
`
`9.
`
`'lioml»<in.» S_ (‘K 3!_
`lLlYl.1$/.C‘I}'l( K,
`I,
`til
`lones AC, Shvriinsuiular MM, Thomas .\l\‘<7, M
`
`2-and PllC|l()[VPik correlations in 130
`Comprehen.~;ivc imitation .1nalysis oliTS(Il and '
`families with tuberous sclerosis. Am l llum Genet 199‘): (i/i:l 5l)‘>vl.’il5.
`llabora Sl.., lozwiak S, Franz [Ji’\’, Roberts PS. Nicto A, Chung I, ct al. Mutational aiuilr»
`sis in a cohort ol224 tuberous sclerosis patients indicates increased severity oti'I'SC2, com~
`pared with TS(M.l, disease in multiple organs. Am l llum Genet 2001; (><‘§:('H—8(l.
`er al
`llcnskc Ill’. Scheithauer B\‘i'. Short Ml’, Wolliiianti R, Naliinias l, llornigold
`Allelic loss is tirequcni
`in tuberous sclerosis l<idnc-y lesions but rare in brain lesions. Am Vl
`llum (ient-t l‘)‘)<i: ‘3‘):4()()-»(>.
`. Au KS, Hebert AA, Roach i
`\
` ;\’ortlu'up ll. (loinpletc inactivation olitlie 'l'S(‘.2 gene leads
`to formation <ililianiai'torii
`...i\|!1Vl Hum Genet
`i‘)‘)‘l; (i'3:l7‘)0--‘3.
`(iarsillo T, Astrinidis A, Henslte El’. Mutations in the tuberous sclerosis complex gene
`T5("’ are :1 cause of sporadic pulmonary lymphangioleiomyomatosis. Proc Nzitl Acad Sci
`USA 2l'l()0: 9‘/:6()85~‘)(l.
`
`.
`
`15.
`
`H.
`
`Niida Y, Stemmer~Rachamimov A0, logrip M, Tapon ll‘). Perez R, K\Vlfl(l(()W~3l<i D], et al.
`Survey (,Yl‘i()fh1ill(‘ nuuaiions in tuberous sclerosis complex (TS(,Il liaimrtotnas siiggests dill
`lcrent genetic riieclmiiisiris for pzitliiigeiiesis of‘ TSC lesions. Am J Hum Genet 2001;
`($‘):493—5()3.
`
`\‘</iuren HP», Kwiatl<owsl«,i D]. Tsci1(+/-) mice develop
`Onda H, Lueclc A, Marks l’\‘i/,
`tumors in multiple sites that express gelsolin and are influenced by genetic bacl<gronntl.
`l
`Clin invest W9‘);
`l(l4:(>87-95.
`
`’olia_v;ishi T, Minowa O, Kuno J, Mitani H. llino 0, Nod.iT. Renal carcitiogeriesis, hepat-~
`ic hemangioinaiosis. aiiol embryonic lethality caused by .1 germvline 'li\'t2 mutation in (nice.
`Caiicer Res 199‘); 59:12(l6—\ l.
`
`obayashi T, lvlinowa O, Sugitani Y. Takai S, Mirani H, Kobayashi E, CI al. A gernvline
`Tsci mutation causes tumor development and embryonic lethality that are similar, but not
`identical
`to,
`those caused by Tscl mutation in mice. Proc Natl Acad Sci USA 2001:
`‘.l8:8762—7.
`
`l labuchi T. Knowles MA. Mutation olithe
`lornigold N, Devlin Al, Davies AM. Avcyard
`‘)q3/i gene TSCl in sporadic bladder cancer. Oncogene 1999: l8:Z(i57-61.
`’arry 1., Maynar JH, Patel A, Hodges AK, von De-imling A, Sampson {l R, et al. Molecular
`analysis ()li[liCTS(:l and TSC2 tumour suppressor genes in sporadic glial and glioneuronal
`tumours. Hum Genet 2000:
`l()7’:3‘S0-6.
`
`llrl, Patel A, (IlilTor«.l SC, Morrissey (I. Mallet l{R, et al. Analysis olithc
`’arrv L, Mavnari
`'Ii.'\(.l and'l'S(I2 genes in sporadic renal cell carcinomas. Br yl (lancer Z001: 8S:l22(s«3().
`. Ciao X, Pill} ll). "SCl
`
`
`
`.’llI(l T$(ilZ tumor sir I iressors .'nu:1roni7e insulin siwii;iliriu in celll l la e z;
`
`
`growth. (ienes Dev 300 l;
`l5: 1 jl<‘§,5~‘I2.
`
`lUHL.[l()l\\\\’l[l\'liSL..l. io:nu.1goiu‘/e ins‘ulinsig~
`loner (iii H1:.'iiigH,Xx.iT. l)I'0.\()pllil&1llixtil
`ualing in regulating cell growth. cell }1l'()lllT‘fil[il)I), and organ sire. (iell Z001;
`l()’v:,35.7-(S8.
`. Tapon N, ho N,
`)icl<son Bl, ili’€i$IH,\I\
`_lF.. l'l:ii'ili:n'.m IK. The tlrosoplulu rulierous sclero-
`.\'l\ colnplex gene luunologr
`restrict
`cell growth and cell pmlileiatiori.
`(iell
`2U(I E,
`l()§:.3"é5«3‘3.
`
`ho N, Rubin Lilv . gigas, a Drosuphila homolog of tuberous sclerosis gene product—Z. reg«
`ulates the cell cycle. Cell 199‘): ‘)(3:53.‘)--3‘).
`l‘l/anlc TL, Yeung IS, llenslic El’. Harnartin, the product otithe tuberous sclerosis l {TSCIJ
`gene, interacts with ruberin and appears to be localized to cytoplasmic vesicles. Cancer Res
`I998; 58:/i7(‘iC)~7ll.
`
`23.
`
`vi
`
`. van Slegtenhorst M. Nellist M, Nagellterlccn B, Chcadle J. Snell R. van den Ouwclantl A,
`or al.
`interaction between haniartin and tuberin.
`the TSCI and TSC3 acne products.
`Human Mol Genet 1998; 7:l(l334l(l38.
`
`. Ciao X, Zhang Y, Arrazola P, llino O, Kobayashi T, Yeung RS, er al. "lsc rumour suppres—
`sot proteins antagonize ainim3—acid¥l'QR signalling. Nat Cell Biol 3002; ’i:(3‘)‘Jv7<)—i.
`Potter (fl, Pedrara H}. Xu T. Ala regulate.» growth by directly pho.~:plior_vl.uing Tac2.
`Cell Biol 2003; /i:(s‘iR--(:5.
`
`l.eth;iiit_v oli l)iosopl'iila
`l lemniings~Mics7czal< M. llilioinas (1.
`.\/lontague
`',
`R.iClll1lCl’Sl(
`lacking 'li.‘>(I tumor suppressor function rescued by reducing tlS(2l\' signaling. (ienes Dev
`2002: l(x:2(i27—3Z.
`
`.’\Kn
`
`18.
`
`(iillllng li, I,eeVl'l', Navolanic PM. Steclrmn LS, Shelton JG, Blaloclr \Vl., Ct al. Involvement
`olil’l3K/Ala parliway in cell cycle pr0grt’s»‘ii)n, ;ipoptosis, and neoplastic :r.mslisrin.ttiori: A
`target For cancer clieinotherapy. l,eul<emia 30(?,%;
`l7:5‘)(lv(i05.
`jacinto F, Hall MN. Tor signalling in bugs, brain and hrziwu. Nat Rev Mol Cell Biol 2005;
`4:11‘/r26.
`
`.
`
`.
`
`Huang S, Bjorusti M. Hougluon Pl. Raparnyciris: Mechanisin olaction and cellular resist-
`
`ance. Cancer Biol Ther 20(),’l; 2:2’ ' “.32.
`.\lorul<i Y. Long X. Yoshino K. Oshiro N, Hitlayat S, et al. Raptor, a binding parts
`l’lLlY€l K,
`net or target ot'rap,unj,'cin (TOR). mctlizites TOR action. Cell 2003; 110:1‘/7A8‘).
`. Kim DH, Sarlitissov dos T), Ali $.\'l. King
`l.'lR’l{ RR. ErcliLnnentsbroinage ll. ct ill.
`
`rnTC>R iiueracts with raptor to lorin a
`utrientvs-erisiti\'<.' complex that signzils to the cell
`growth macliinery, (Lell ZUHZ;
`ll(l:l(;$—."/s,
`Kim li)l l. 5/
`issi,-V dos D. An .'\i,\i. L;liel< RR, liuntur l<\", hrdiument —l$roinagc ll, ct al.
`
`,
`regulator ol
`the rapaniycin~sensiti\*c }7.1Il1W.lV required for tht nutrt
`ant sciisitivc interaction between raptor and ntll)l{. Moi (fell ZUU3:
`l
`l:L\"‘)3 ‘)(l/i
`
`
`
`Kwi;itluiwsl~ti D}, Zhang l l,
`l3.ii'itlt\r;i ll.. llciber ' ‘r l{f\'l. (ilog
`ct
`.\l. el<l l.‘r\l‘.t’fl
`
`
`al. A rnousu model ot'TS(.l
`revc.1l~‘ .\\
`-c*pentient lctliality tlom lixci
`ii<'n\.tngiom
`
`vulstion or p7l)S(> lunmc 3.CflVlly
`in ilsci null cells. Hum Mol
`(,;'CnCT ZU02;
`
`
`33.
`
`M.
`
`Roxane Labs., Inc.
`Exhibit 1018
`Page 005
`
`

`
`RHliBBlN(} UP M'l"()R, R0l..li,‘$ 013 TSCI AND "I'S(I2
`
`()2.
`
`63,
`
`(H.
`
`(55,
`
`(57.
`
`(>8.
`
`(,i)_
`
`lVl.’il( BC, 'liil<c-nrzrrn K, Kc-ncrxun Hl,, Mimn RT, Ycirrrg RS. Th: rrih¢rin~h;trn;rrrin cum-
`plcx rrcgaiively rc-grilarcs Berti «crnenirr signaling .1ctivity.] Biol Chem 2005; 278:59~’i7_'31,
`Kcnerson Hl,, Aicher ll), ilirnc ll), Ycung RS. Acrivarctl ttrainriralitrri Lztrgcr nlirapartrycixr
`pathway in the patlirigwiresis ulirirlwr<iris sClc‘.rl)\‘is corriplcx |‘§’,H2ll tumors. (Iaircct Res 2002;
`(33:56-/r5~5().
` .
`ln<',lkl(Z€S V(h-
`F,l~H;ixltc'ntirc N, Vlritlkcr V, Xlrzrng H, l{\~'ittil<t>\vsl<i ll]. l.<,l1iS oliiliscl nr ls
`c,nl'.rr cntlnrlrelizil growth lircmr pmtlirciinn rhmirgh m.rrnm;rli.m target of’ rap.rntycin.
`Cancer Rcscrrrclx 2003; in })l't‘s\.
`llinu fl, K()l7;r)';r_\‘l‘,[ T, Milan; H, I’rg‘V§[r|ir)[] of l1(‘[cdi[;)ry c;1r(i[rr)gc|1c3is_ Pm,» J:}_p;[|’]
`2003; 78:50—}2,
`l:I‘£\I)I, D, ct til. A.ssuci:rti<.>rr lictxvccn .1
`66. Dabora SL, Roberts I‘, Nicto .~\, Perez, R, ltmwialt S.
`liiglr-expressing iritt‘rlErt>ti—g.inirn;r Allele and Al lowcr l’lL‘qlICIlC}' «)lil<l(ll"lL‘)' trrig:ornyolipomas
`in TSCIZ pziticrm, Am _I Hunt Genet 9.002: 712780-33,
`\Vlritc R. Hun Y, Stlrcitlrzrrrcr B, Lynch DR, llenskc lil’. (Lririu l’l3. Sclesrtivc .rltcr.'rtions in
`gltrriitimtc xmci CARA I'k‘C(‘P(l)|' strbrmir tt‘.l{l\I,-\ r.\ptcssion in <lysplasti<‘ rictmitis zmd giant
`cells olicmticzzl ttrlicrs. Ann Ncrirol 2001; ~’l‘):(F/"A78,
`l<l1£1HA, Pepiu AM, Soxsin \‘V$. Scmtonirr .