Par Pharm., Inc.
`Exhibit 1005
`Page 001
`
`

`
`This material may be protected by Copyright law (Title 17 U.S. Code)
`
`Rapamycins: Antifungal, Antitumor,
`Antiproliferative, and
`Immunosuppressive Macrolides
`Randall Ellis Morris
`
`What we know ira drop. What we don't know is an ocecm.
`lrattc Newlon
`
`P rogress in rapamycin (RPM) research has been
`
`rapid and is poised to accelerate even more
`dramatically. An Investigational New Drug applica(cid:173)
`tion (IND) for phase I ti·ials of RPM as a treatment
`for prospective graft recipients ""as approved less
`than 2 years after the first published reports 12 and
`
`•
`public disclosure3 of the apility of RPM to prolong
`graft survival in experimental animals. RPM is a
`macmlide fermentation product that has antifungal
`and antitumor activity. However, its effects on the
`immune system have generated the most interest
`because RPM is structurally similar to another new
`immunosuppressive macrolide, FK506. RPM is par(cid:173)
`ticularly intriguing because it inhibits the activation
`of immune cells by unique, relatively selective, and
`e.xtremely potent and highly effective mechanisms.
`For example, one half microgram of RPM adminis"
`tered daily to mouse recipients of completely mis(cid:173)
`matched heart allografts prolongs graft survival.
`When these mice are treated for only 2 weeks ·with
`higher doses of RPM, or when a single dose ofRPM is
`administered to rat heart allograft recipients, strain(cid:173)
`spe<;iflc unresponsiveness is induced, and grafts sur(cid:173)
`vive indefinitely in both species.
`The research on RPM is representative of a
`significant shift in emphasis in transplantation. from
`the macrocosmic world in which innovative surgical
`techniques predominated from the 1950s through
`the 1970s to our current focus on the microcosm of
`cellular and molecular immunopharmacology. A rev(cid:173)
`olution in the discovery, development, and clinical
`use of new strategies to control the immune response
`is clearly upon us: it took more than 35 years to
`
`From the Laharalti!V for Transplrmtation lmmwwlo.!IJ', Departmi!nl ~~
`Cardiollwraric SurgeiJ'· StmJfilrd Universi{J' School '![ i\!Jediciue, SttuJford,
`CA.
`llddrrss rrprint reqU£Sis to Randall Ellis Manir, MD, Lnbomtory• jar
`'JimiSpiantalimz Jmmurwlom·, Department qf Cardiollwracic Swgtl)", Stan(cid:173)
`Jim{ Universi!J' School'![ Medicine, Stm!fiml, Cl194305-5247.
`Copyright© /9!12 b)• WB. Smmdcts Compo'!)"
`O!J55-470XI!J2/0fJOUJ004$5.00!0
`
`accrue the four imperfect mainstays of immunosup(cid:173)
`pression for transplantation-steroids, azathioprine,
`anti-T-cell antibodies, and Gyclosporin A (GsA). In
`1992, six newxenobiotic immunosuppressants will be
`in clinical trials (Fig I).
`This new era in immunosuppression can be traced
`to the convergence of several lines of research: ( 1}
`the discovery and successful clinical use of GsA; (2)
`an increased understanding of the fundamental biol(cid:173)
`ogy of immune cells that enables the actions of
`different immunosuppressants to be better under(cid:173)
`stood and thus lay the foundation for more rational
`means to discover, develop, and use improved drugs;
`and (3) organized preclinical research programs
`designed to identify potentially valuable immunosup(cid:173)
`pressants and to generate the knowledge needed for
`these agents to be used intelligently in the clinic.
`Figure 2 shows the research program used for several
`years in the Laboratory for Transplantation Immunol(cid:173)
`ogy at Stanford University that enabled us to identify
`RPM3
`11 and the morpholinomethyl ester of mycophe(cid:173)
`"
`nolic acid (:.MPA) 12
`16 as immunosuppressants for
`-
`graft rejection. The mechanisms of action and immu(cid:173)
`nophamJacology of these two compounds, as well as
`19 deoxyspergualin (DSG)/o.21 and brequinar
`FK506, 17
`"
`sodium (BQR)22 have also been studied and com(cid:173)
`pared with one another in our laboratory.
`Our spectrum of experimental systems begins
`with in vivo mouse models that are so rapid, quantita(cid:173)
`tive, and inexpensive that we have been able to
`evaluate hundreds of molecules for suppression of
`alloimmunity. The vast majority of these drug candi(cid:173)
`dates fail during testing in rodents because they lack
`efficacy or safety, and they are discarded quickly so
`that our resources can be concentrated on com(cid:173)
`pounds with the greatest potential. Compounds that
`show promise are evaluated further in rodent models
`to identify those with the following ideal characteris(cid:173)
`tics: (I) unique mode of action; (2) high efficacy for
`the prevention or treatment of acute, accelerated, or
`chronic rejection; and (3} low toxicity. This Darwin(cid:173)
`ian selection process accomplishes two tasks: first, it
`insures that only the agents with the greatest paten-
`
`Tranrplantalian Rez,iewr, Val6, Na I (]anum)~, 1992: pp39-87
`
`39
`
`

`
`40
`
`Randall Ellis MwTis
`
`MIZORIBINE
`DEOXYSPERGUALIN
`FK506
`MYCOPHENOLIC ACID
`RAPAMYCIN
`BREQUINAR SODIUM
`
`CYCLOSPORINE
`OKT3 MAb & OTHER MAbs
`
`CORTISONE
`AZATHIOPRINE
`ANTI-T CELL Abs
`
`1950
`
`1960
`
`1970
`
`1980
`
`1990
`
`2000
`
`Figure 1. History of the use of drugs used to control graft rejection. All of the following xenobiotics recently cliscovt>recl to
`suppress graft rejection in preclinical models have advanced to clinical trials: the antimetabolites such as mizoribine
`(i'vlZR), l\ll'A in its prod rug form ofRS-6H4-3, and BQR; the cydosporine-like drug FK506, and drugs that define two ne11·
`classes of immunosuppressants, DSG and RPM.
`
`tial are advanced to the expensive nonhuman pri(cid:173)
`mate transplant model; and second, it prepares us to
`be able to use these compounds intelligently in
`nonhuman primates. The nonhuman primate model
`is important because it is highly predictive of the
`safety and efficacy of a test drug in humans. The sum
`of all knowledge produced from well-planned prcclin-
`
`ical studies is the essential roundation from which
`successful clinical trials arc designed and executed.
`New drug development is a highly complex, multidis(cid:173)
`ciplinary task, and our contribution to the develop(cid:173)
`ment and clinical usc of new immunosuppressants
`depends on very close collaboration with scientists
`and clinicians in the pharmaceutical industry.
`
`FUNDAMENTAL ~
`IMMUNOLOGY ""'~
`/+'x
`,4~=~=~~ n~\<__....~ __....
`@~
`
`I =o=, ~ ~//
`
`LIN VITRO __j
`'-------IN VIVO------'
`._ _____ DISCOVERY--------'
`
`' - - - - - DEVELOPMENT----'
`
`TRANSPLANTATION
`
`AUTOIMMUNE
`DISEASES
`L CLINICAL TRIALS _j
`Figure 2. Schematic represelllatiun of the program used at the Laboratory ofTransplantatiun Immunology al Stanford
`Unh·ersily to identify compounds 11ith immunosuppressive activities for transplantation and to develop these compounds
`lor clinical use [[Jr the prc\"t~ntinn and treatment of rejection. Fundamental knowledge oft he immune system coupled with
`an appreciation nf the characteristics of" the drug candida It' is ust'd to design experiments to prolile the activity uf the
`mmpound and define its mechanisms of action. Heterotopic transplantation of neonatal mouse heart allografts into the ear
`pinnae of mouse recipients and alloantigenic and mitogenic stimuli of" popliteal lymph nodP hyperplasia are used as rapid
`and quantitative bioassays before proceeding to the morf' laborious techniques of primarily vascularized heterotopic
`(abdominal) and secondarily v;u>cularized heterotopic (subrenal capsule) heart allogrart and xenograft transplantion in the
`rat. Assessment of" 1 he t>fTicacy and the safety or the compound in cynomolgus monkey recipients of" ht'lerutopic allografts
`precedes phase I clinical trials in transplant patients and patients with autoimmune diseases.
`
`

`
`Ca2+.DEPENDENT Ca2+.1NDEPENDENT
`
`41
`
`LIGANDS
`+
`
`LIGAND
`+
`
`,,.,.,,~, I
`
`~~ IL·2- ~~IL·2-
`~ CYTOI<INES
`
`IL·2
`
`CsA
`FK506
`
`~ ,,@~-~~FULLY
`~~~!~~D
`''i' - fi CELLS
`MZR ~ ~
`
`-
`
`.
`
`.
`
`.
`
`RPM
`(B CELL)
`
`MPA
`BQR
`(B CELL)
`
`DSG
`(B CELL)
`
`Figure 3. Schematic representation of the possible sites of action of the follmving iri1munosupprt'ssants on activated T
`rells: CsA and FK506 prevent the transcription of early phase cytokinc genes; RPM inhibits the signal transduction ofiL-2
`bound to its receptor and may have other antiproliferative effects unrelated to lymphokine signals; MZR, ~IPA, and BQR
`all inhibit purine (MZR, MPA) or pyrimidi11e (BQR) nucleotide synthesis; DSG seems to inhibit late stages ofT-cell
`maturation. RPM, .MZR, MPA, BQR, and DSG also act on activated B cells at the sites shown.
`
`Even more important than the relatively large
`number of new immunosuppressants that have been
`discovered is their variety. Each of these new mole(cid:173)
`cules ~uppresses the immune system by blocking
`distinctly different biochemical reactions that initiate
`the activation of immune cells that cause the many
`fonns of graft rejection (Fig 3). Brie~y, CsA a~d
`FK506 act soon after Ca2+-dependent T-cell activa(cid:173)
`tion to prevent the synthesis ofcytokines important
`for the perpetuation and amplification of the im(cid:173)
`25 RPM acts later to block multiple
`mune response.23
`•
`effects of cytokines on immune cel1s including the
`inhibition ofinterlcukin-2-(IL-2-)triggcred T-cell pro(cid:173)
`liferation,26:31 but its antiproliferative effects are not
`restricted solely toT and B cells. RPM also selectively
`inhibits the proliferation of growth factor-dependent
`and gro\vth factor-independent nonimmune cells.
`Nlizoribine (.MZR),32 MPA,33 and BQR34 arc antime(cid:173)
`tabolites that inhibit DNA synthesis primarily in
`lymphOcytes. These new antimetabolitcs arc more
`selective than azathioprine because these com(cid:173)
`pounds block the activity of enzymes restricted only
`to the de novo purine or pyrimidine bios;11thetic
`pathways. Lymphocytes are more dcpenden ton these
`pathways for nucleotide S)'Ilthesis than other cells.
`Recent revicws'J>.:rn discuss these and other immu(cid:173)
`nosuppressants. RPM has recently been the subject
`of four brief reviews,73941 a long rcview,+1 and has
`been included in reviews that have primarily focused
`on FK506. 1
`1 This review prmides a complete profile
`H
`of RPlVI from work published through the end of
`August 1991. Despite the progress made in under(cid:173)
`standing RPM since the first publication on this
`compound in 1975,1
`" the description of its ability to
`suppress graft rejection has stimulated rrnewed
`
`interest by a \\ide variety of investigators whose work
`has not yet been published. A~ a result, research on
`macrolide immunosuppressants has become fluid
`and extremely fast-paced. Because unpublished data
`generally are not available for evaluation, I have not
`referred to unpublished work or personal communi(cid:173)
`cations. However, I have relied on many studies of
`RPM from the Laboratory ofTransplantation Immu(cid:173)
`nology at Stanford Uni~'ersity that have yet to be
`published in full. In most of these cases, I have
`supplied the data f~om which conclusions in the text
`arc drawn.
`Because this review is being written relatively
`early in the research life of RPM, and because the
`majority of the work on this complex molecule has
`yet to be published, the material subsequently pre(cid:173)
`sented should be regarded more as a preview rather
`than as a review. At the very least, this article will
`provide a logical framework that ofher investigators
`can usc to organize and to evaluate new information
`on RPM as it is published, For many investigators
`with highly specialized interests, only selected sec(cid:173)
`tion~ \\ill be of usc. For others, it is essential to
`understand all that is knmm about a new and unique
`molecule such as RPM. Without an understanding of
`RPM that is both deep and broad, it \viii be difficult to
`meet the challenging tasks of usil1g RPM as a tool to
`learn more about the immune system, maximizing
`its therapeutic potential, and discovering new and
`improved members of this class of immunosuppres(cid:173)
`sant. If we strive to understand thoroughly the little
`that is now known about RPM, we \\ill make more
`efficient and rapid progress toward our goal of
`understanding all of the imRortant biological effects
`of this molecule.
`
`

`
`42
`
`Randall Ellis Jlcmi 1
`
`CYCLOSPORINE _..,
`
`FK506
`
`RAPAMYCIN
`
`~TE ----------------~E~V=EN~T--------------------
`1972· SPECIFIC SUPRESSION
`1975
`OF IMMUNE CELL
`ACTIVATION SHOWN
`IN VITRO AND IN VIVO
`
`FDA APPROVAL AND
`COMMON CLINICAL USE
`IN TRANSPLANTATION
`1\.._~
`SCREEN
`......_, FERMENTATION
`PRODUCTS FOR
`INHIBITORS OF
`THE MOUSE MIXED
`LYMPHOCYTE REACTION
`
`IMMUNOSUPPRESSIVE
`ACnVITY SHOWN IN
`VITRO AND IN VIVO
`
`UNIQUE, AAPAMYCIN·UKE
`STRUCTURE DEFINED
`
`1975
`
`1978
`
`1980
`
`19B3
`
`1982-
`1983
`
`19BS
`
`19B8
`
`ANTIFUNGAL ACTIVITY,
`LOW TOXICITY SHOWN
`
`IMMUNOSUPPRESSION
`OF AUTOIMMUNE
`DISEASE SHOWN
`
`UNIQUE MACROLIDE
`STRUCTURE DEFINED
`
`INVESTIGATION
`OF AAPAMYCIN
`AS AN IMMUNO·
`SUPPRESSANT FOR
`TRANSPLANTATION
`
`RAPAMYCIN
`SHOWN TO PROLONG
`ALLOGRAFT SURVIVAL
`
`Figure 4. Evolutionary path ofRlJJVI as an immunosup(cid:173)
`pressant for transplantation.
`
`In addition to reviewing the information on RPM,
`this article warns of the danger of inductive reason(cid:173)
`ing in which, in an adolescent field like immunology,
`arguing from highly specific cases to general laws
`
`often promotes the illusion of knowledge rather than
`its true acquisition. However, by interrelating infor(cid:173)
`mation concerning the structure, the molecular mech(cid:173)
`anisms, and the actions of RPM on ddi.ned cell t}l)es
`in vitro, its effects in \ivo, as well as its disposition in
`the body and its toxicity, new and important insights
`into the actions of RPM ran be gained. In general,
`the conceptual tools used in this review to analyze the
`data from experiments on RPM ran be applied to the
`study of many other immunosuppressants, especially
`other xrnobiotirs.
`Before dissecting and examining every aspect of
`RP.M in detail, it is worth reviewing the events that
`led to the attention RPM is now receiving. Figure +
`shows the relationship of the evolution of RPM as an
`immunosuppressant to the development of CsA and
`FK506 as immunosuppressants. Table I prm~des a
`more detailed outline of the sequence of the main
`events that have defined progress in RPM research in
`its first 15 yrars. 1 -r~i-2'1'1r"12 The ancestors of RPwf are
`CsA and FK506. As shown in Fig 4, CsA stimulated
`the organization of a rational screening program
`designed to discover other fermentation products
`with mechanisms of immunosuppressive action iden(cid:173)
`tical to CsA. The discovery ofFK506 was the product
`of this program,',, and when the structure of FK506
`was defined, its similarity to the structure of RPM
`was immediately recognized.;' Years bcrore, the
`structure of RPM had been determined as a conse-
`
`Table 1. History ofRPi'vf Drug Dewlopmenl: The First 15 Years
`
`Discove~!'
`
`Isolation !'rom Easler Island (Rapn Nui) soil
`sample and characterization of antimicro(cid:173)
`bial activity
`In vivo use:
`Toxicitv
`Pharm;cokinrlif's
`Bionvailabilit v
`Antifungal a~tivity
`Immunosuppression of autoimmune dis-
`case
`Elucidation of structure
`Antitumor activitv described
`Immunosuppression of allograft rejection
`RPI'vl alone
`
`lU'l\1 in combination with C:sA
`Di!Terentiation of effects ufRl'M and FJGDG
`on immune cells in vitro
`
`Diffrrentiation of eflects of RPivl and FK50G
`on immune system in vivo
`Demonstration of binding o!'RPivlto FK5Dfi
`binding protein
`
`:Year
`
`1975
`
`1978
`
`1977
`
`1980
`19HJ
`
`19H9
`
`1990
`19H9
`1990
`
`1990
`
`19H9
`
`Vezina, Kudelski, and Sehgal"~>
`Sehgal, Baker, and Vezina"
`
`Baker, Sidorowicz, Sehgal, et al 111
`
`;vrm·tel, Klicius, and Galet"1
`
`Findlay and Radics''"
`Douro~ and Suffness"
`
`Morris and .Meiser'
`C:alnc, Collier, Lim, et al'
`:\Ieiser, vVang, and Morris'
`Tocci, Matko\'ich, Collier, el al"
`.\fetcalfe and Richards'''
`Dumont, Siaruch, Knprak, et al"''
`Morris, vVu, and Shorthouse 1
`
`Harding, Galat, Ut'hling, et a(' 1
`
`

`
`Rajmn~rcim
`
`43
`
`0
`
`quence of the identification of RPM as an antifungal
`antibiotic (Table 1). Shortly after the antibiotic
`activities of RPM were described, it was found to
`have immunosuppressive activity. This was only a few
`years after the immunosuppressive activity of CsA
`was discovered, but ironically, RPM was not devel(cid:173)
`oped as an immunosuppressant at that time. In a
`review'11
`' of immunosuppressive agents published in
`1988, Devlin and Hargrave encouraged "a detailed
`comparison of the biological profile of these mac(cid:173)
`rolides [FK506 and RPM]." These investigators sug(cid:173)
`gestion was based on the structural similarity of both
`compounds and their known immunosuppressive
`activity.
`Sehgal was aware that investigators at the Labora(cid:173)
`tory for Transplantation Immunology at Stanford
`University had developed a quanta) bioassay for the
`evaluation of immunosuppressant potency and ef(cid:173)
`ficacy, had validated the assay with CsA,33 and had
`).!sed it to study FK506. 17 In I 988, he offered to
`provide us with enough RPM to enable us to deter(cid:173)
`mine whether its activity differed from FK506 in
`mouse as well as rat heart transplant recipients. As
`subsequently discussed, the activity of RPM is ex(cid:173)
`tremely dependent on the vehicle in which it is
`suspended and the route by which it is administered.
`Had our first experiment used suboptimal conditions
`for the administration of RPM, we would have found
`no difference in potency or efficacy between RPM
`and FK506 and might not have pursued our study of
`RPM. In retrospect, the mode of administration used
`at the outset was optimal and, under those condi(cid:173)
`tions, RPM was clearly more potent and effective
`than FIG06. This clear difference in pharmacological
`effect between these two structurally related mac(cid:173)
`rolides prompted our continued investigations of the
`activity of RPM. At the same time as these studies
`were being conducted, investigators at the University
`ofCambriqge, England, were testing the immunosup(cid:173)
`pres!jive activity of RPM in rodents, dogs, and pigs.2
`1 performed at Cambridge
`Simultaneous studies27
`2
`•
`'
`and by various groups of investigators at Merck
`Sharp and Dohme Research Laboratories, United
`States showed that RPM and FK506 affect immune
`cells quite differently in vitro.
`
`RAPAMYCIN At= OCH, R,= OH
`DEMETHOXYRAPAMYCIN Rt= H, R2= DH
`
`FKSOG
`
`PflODflUGS OF flAPAMYCIN
`
`(R1: OCH3. R2: see below):
`
`N,N·DIMETHYLGLYCINATE
`METHANE SULFONIC ACID SALT
`I
`o
`RZ"" i'o~N...._ CHlSOlH
`
`3·(N,N·DIETHYLAMINO!PROPIONATE HYDROCHLORIDE SALT
`0
`
`A,• l-.O~N......... HCI "
`
`4·(PYRFlOLIDINO)BUTYAATE HYDROCHLORIDE SALT
`('a
`o
`R,- l'o~N....) HCI
`
`Figure 5. Chemical structures of RPM, 29-demet hoxyra(cid:173)
`paQ1ycin, FK506, and the prodrugs of RPM.
`
`ries in the middle 1970s.1fi.H The aerial mycelium of
`
`this b~cterium is monopodally branched (Fig 6),
`contains sporophores terminated by short, coiled
`spore chains, and absorbs water. It was ultimately
`identified as belonging to the species Strcpton~yces
`hj'groscopicus, designated by Ayerst Research as strain
`AY B-994, and deposited in both the ARS culture
`collection of the United States Department ofAgri(cid:173)
`culture (assigned number NRRL 5491) and the
`American Type Culture Collection (A1~CC 29253). A
`structurally related compound,;n 29-demethoxyra(cid:173)
`pamycin (AY-24,668 [Fig 5]) is coproduced with
`RPM. Another culture isolated from the same soil
`sample and designated AY B-1206 produces higher
`levels of RPM than AY B-994 and little or no
`29-demethoxyTapamycin.:'7
`
`Fermentation, Purification, and
`In Vitro Antimicrobial Activity
`ofRPMs
`
`Origin and Characterization of the
`Bacterium Producing RPMs
`
`RPM (AY-22,989 [Fig 5]) is made by a filamentous
`bacterium from the streptomycete group that was
`isolated from an Easter Island soil sample by Vezina
`et al and Sehgal et al at Ayerst Research Laborato-
`
`Fermentation of RPM
`Soon after the availability of a pure strain of S
`hygroscopicw, fermentation conditions (type of media,
`media pH, and temperature) were varied to define
`its cultural characteristics. 16'rn Although this microbe
`grows and sporulates in a wide range of culture
`
`

`
`44
`
`Randall Ellis Manis
`
`Figure 6. (A) Photomicro(cid:173)
`graph of the filamentous bar(cid:173)
`terium, S hj•grascvpicus. that
`produces RPM (magnification
`X455). {B) Electron micro(cid:173)
`graph ofS ~vgmsmpicus ( magnifi(cid:173)
`cation X2,500). (Reprinted
`with permission.'")
`
`conditions, more narrowly defined conditions are
`necessaty for the optimum production ofRPNL RPM
`has been produced by aerobic submerged fermenta(cid:173)
`tion similar to that used for most antibiotics. Inocu(cid:173)
`lum is prepared in two stages in a medium contain(cid:173)
`ing soybean meal, glucose, (I\1Jl4) 2S04, and CaCOJ
`and used at 2%. For the fermentation in stirred
`vessels, the starting medium was soybean meal,
`glucose, (NH+) 2S04, and KHlO+. Glucose is fed
`continuously after the 2nd day and the pH was
`controlled at 6.0 with NH40H. 1vlaximum titers of
`RP1vi arc reached in 96 hours. Paper disc-agar diffu(cid:173)
`sion assays with Candida albicans are used to deter(cid:173)
`mine the antibiotic titer.
`The fermentation methods required to produce
`29-dcmethm:yraparnycin arc the same as those de(cid:173)
`scribed for RPM.51
`;
`
`Purification of RPMs
`Tbe purification scheme (Fig 7) adopted for the
`production of RPM was developed shortly after the
`identification of the antifungal activity of RPM and is
`subsequently summarizcd.+7 After fermentation, the
`pH of the beer is adjusted to 4.0. The mycelium,
`extracted with trichloroethane, is filtered off and the
`extract is dried with anhydrous sodium sulfate to
`
`FERMENT STREPTOMYCES HYGROSCOPICUS
`
`EXTRACT MYCELIUM WITH ORGANIC SOLVENTS
`
`APPLY CONCENTRATED EXTRACT TO SILICA GEL COLUMN
`
`ELUTE WITH ACETONE
`
`RAPAMYCINS
`
`Figure 7. Fermentation and isolation ofRP!\•ls.
`
`produce about 500 gm of oily residue from a 160-liter
`fermentation run. After extracting the residue with
`methanol, the extracts are evaporated to yield approx(cid:173)
`imately 50 gm of residue that is then dissolved in 15%
`acetone in hexane and mixed with silica gel. The
`dissolved RPM is adsorbed to the silica gel and
`remains bound to the gel after the mixture has been
`filtered and washed onto a column from which RPM
`is eluted with an acetone:hexane mixture. After
`evaporating the column eluate to dryness, the resi(cid:173)
`due is dissolved in ether from which pure ctystals of
`RPM are obtained. In this initial purification process,
`recoveries of RPM are on the order of 40%; I 0 L of
`broth produce 300 mg pure RPM. A more recent
`method of purification has been reported.57
`•
`Except for minor modifications, the methods
`described for the isolation of 29-demethm.-yraparny(cid:173)
`cin are the same as those used for RPivl ;x;
`
`In Vitro Antimicrobial Activity of RPMs and
`Mechanisms of Their Antimicrobial Actions
`The antimicrobial screening program at Ayerst Re(cid:173)
`search Laboratories identified RP:tvl for its antifungal
`activity. RPM inhibits the growth of yeasts and
`filamentous fungi including the derrnatophytes;Vficm(cid:173)
`sjJomm gypsewn and TriclwjJI!J'Ion granulosum. 11
`7 The
`'·'
`minihlum inhibitory concentrations (MIG) of RPM
`against ten strains of C albicans were in the range of
`less than 0.02 to 0.2 J-Lg/ ml, representing greater
`potency than that of amphotericin B, nystatin, or
`candicidin in this assay. RPM has no antibacterial
`activity. The spectrum of antimicrobial activity of
`29-demethoxyrapamyein is similar to RPlVI, but its
`potency is only about 25% that of RPM although
`nearly as potent as amphotericin B.ifi
`One study has investigated the mechanisms by
`which RPM mediates its antifungal effects,'>~~ and the
`results of this study are summarized in Table 2.
`Approximately 90 minutes after adding RPM to C
`albicans cultures, growth is inhibited and subse-
`
`

`
`R apamycinr
`
`45
`
`Table 2. Ylechanisms of Antifungal Actions of RPM
`
`EJ!i:ct qfRPM
`Treatment
`
`Not inhibited
`Not incrt"asrd
`
`Not inhibited
`
`Not inhibited
`
`Not inhibited
`
`Not inhibited
`
`Inhibited
`
`Inl\ibited
`
`Inhibited
`
`· Inhibited more for
`&~A than DNA
`
`Increased
`
`Sorbose retention bv C olbicom.
`Increased hemolvsi; of rat rt'd
`blood cells, efllux ofK+, Pi,
`UV-absorbing material from
`C albicanr.
`C albican.r anerobic glycoly'5is,
`aerqbic respiration.
`Protein synthesis by cell-free
`preparations of C albiams, E
`culi, rat liver, and mitochonc
`drial prt"parations ore albicallS.
`Amino add metabolism by glu(cid:173)
`tamic-oxaloacetic transami(cid:173)
`nase, glutamic-pyruvate
`transaminase in C albicans.
`Glucosamine and !\1-acetyl-glu(cid:173)
`cosamine incorporation into
`whole C albicam.
`Oxidative deamination of glu(cid:173)
`tamic and aspartic aci~ls inC
`albicanr.
`Incorporation nf glucosr i nlo
`man nan inC albicalls.
`Incorporation ofNa acetate and
`methionine into total lipid of
`C albicans.
`Incorporation of adenine and
`phosphate into ~~~A and Dl'\A
`of C albicam.
`Degradation ofnP-labelled intra(cid:173)
`: cellular macromolecules and
`lt"akage through C albicanr
`membrant'.
`
`qucntly yeast cells lose viability and begin to lyse. The
`candicidal actions of RPM differ from polyene antibi(cid:173)
`otics that increase yeast cell permeability by binding
`to sterols in the cytoplasmic membrane, thus causing
`leakage of cellular components. Not only do sterols
`not reverse the actions of RPM, but RPM does not
`increase the leakage of sorbose or the effiux of
`potassium, phosphate, or UV-absorbing materials
`from yeast cells.
`The effects of RPM on other metabolic systems of
`C albicans have also been investigated.:'» For 'example,
`RPM docs not inhibit anaerobic glycolysis or aerobic
`respiration, nor does it inhibit the incorporation of
`glucosaminc or N-acetyl-glucosamine. RPM does not
`inhibit protein synthesis in cell-free preparations ofC
`albicallS, rat liver, or mitochondria from C: albicans.
`Although RPM inhibits the incorporation of glu(cid:173)
`cose into mannan and acetate into lipids, the S}11thc(cid:173)
`sis of gluran is minimally affected, indicating that
`
`inhibition of cell wall S}11thesis is not the primary site
`of the antifungal action of RPM. iii
`TI1e most profound effects of RPM on C albicans
`may also provide clues to its actions on mammalian
`.I
`cells. For examplr, vrry low concentrations (.02 -
`J.Lg/ mL) of RPM inhibit the incorporation of adenine
`and phosphate into RNA and DNA. At the lVIIC lor
`RPM, phosphate-conhlining molecules leak out of
`the yeast cell membrane. The degradation of these
`molecules, presumably including nucleic acids, seems
`to be promoted in some way by RPM.;a
`
`Physi~9-Chemical Properties of
`RPMs
`Structure ofRPMs
`Although the initial analysis of the structure of RPM
`by infrared and nuclear ~agnetic resonance (NMR)
`spectroscopy did not provide the complete picture of
`its structurct these techniques indicated that RPM
`was a complete\y new type of macrolide antibiotic.
`Ultimately, x-ray crystallographic data clarified the
`1 RPM is a 31-membered macrqcy(cid:173)
`structure of RPM. 5
`'
`cle lactone containing an amide with a C 15 carbonyl
`and a lactone with a C21 carbonyl (Fig 5).Additional
`analyses of the ~'~C and 1H NJ.ViR spectra of RPM
`confirmed the x-ray crystal structure o(RPM.J<' X-ray
`studies showed that RPM in its solid crystal form is
`conformationally homogeneous; in solution however,
`RPM exists as a mixture of two conformational
`isomers caused by tram to cis amide isomerization via
`hindered rotation about the pipecolic acid N-CO
`bond. The ratio of trans to cis rotamers in chloroform
`solutions is 3, to 4: !.'"·''"
`Illustrations of the structure of RPM were initially
`59 a
`inconsistent: different cnantiomers were dra\\111
`novel numbering system was used/" and incorrect
`stereochemistry at C28 was represented.J<' Ulti(cid:173)
`mately, the correct structure was published,''' and the
`coordinates are deposited in the crystal data bank.
`Using advanced 2-dimensional NMR spectroscopic
`methods, new assignments of the proton and carbon
`spectra for the m~jor rotamer of RPM have been
`made and a new numbering system suggested.@
`The closest structural relati\'e to RPM is the
`antifungal and immunosuppressive macrolide FK506,
`which is also produced by a strcptomycetc.11 FK'i06 is
`a 23-mcmbcrcd macrocydr lactone that shares a
`unique hcmiketal masked a:,[3-cliketopipecolic acid
`amide substructure with RPiVI,;; but Jacks the C l-C6
`triene segment of RPM.
`The results of "C-labcllcd acetate and propionate
`
`

`
`46
`
`Randall Ellis Moni.1
`
`and HC-labelled methionine incorporation studies of
`the biosynthesis of RPM were consistent with the
`proposed polyketidc pathway in which the carbons of
`the lactone ring of RPM are derived from condensa(cid:173)
`tion of acetate and propionate units in a manner
`similar to that responsible for fatty acid synthesis.
`The methyl group of methionine is an efficient
`source for the three mcthoxy carbons of RPM.
`Because none of the labelled precursors was incorpo(cid:173)
`rated into either the cyclohexane or heterocyclic
`ring, these moieties probably orginate Ii·om the
`shikimate pathway and lysine, respectively.61
`When 1H and 11C NJviR, infrared, llV, mass
`spectroscopy, and optical rotaJy dispersion/circular
`dichroism (ORD/CD) analyses were used to com(cid:173)
`pare the structures ofRPIVI and 29-clemethox·yrapa(cid:173)
`mycin, these molecules were shown to be configuration(cid:173)
`ally identical at all chiral centers and to have identical
`structural features at all but C29. Like RPM, approx(cid:173)
`imately 20% of 29-demethoxyrapamycin in solution
`exists as the cis rotamcr form.GJ
`In addition to the naturally occurring 29-
`demethoxyrapamycin, amino acid ester analogues of
`RPJvi have been synthesized to produce three water
`soluble prodrugs of RPi\<f..t (Fig 5). The amine ftmc(cid:173)
`tions of the appended esters can be converted to
`water soluble salts that are cnzynmtically hydrolyzed
`in the plasma to produce RPM. Although RPM forms
`both monoester and diester ndducts depending on
`the reaction conditions, only monoester salts are
`
`Table 3. Physical and Chemical Properties ofRPl\I
`
`discussed because these are sufficiently water soluble
`to ob,~ate the need for the clisubstitutcd forms. The
`28-hydrox·yl group of RP:YI has been proposed as the
`site of esterification for each of these prodrugs, but
`this remains to be confirmed.
`
`Physical Properties of RPMs
`Table 3 lists the physical properties of RPM.17·;o,t..t
`Although 29-dcmethoxyrapamycin is also a white
`crystalline solid, it has a lower melting point (lOr to
`1 08°C) than RPM."~> Both RPM and its 29-demethoxy
`form are lipophilic and only minimally soluble in
`water. The water solubilities of both the mono-N,N(cid:173)
`dimethylglycinate methancsulfonic acid salt and the
`mono-N,N-diethylpropionatc hydrochloride salt pro(cid:173)
`drugs of RPM are more than 50 mg/mL. The water
`solubility of the mono-4-(pyrrolidino)butyrate hydro(cid:173)
`chloride salt prod rug is 15 mg/ mL.''1
`Because IvllCs for the antifungal acti,~ty of RPM
`in vitro vmy depending on the medium used and the
`length of the assay, it was suggested that RPM is
`unstable:H' Subsequent studies shmvecl that 5 J.,Lg/mL
`of RPM in uninoculatecl broth loses 80% of its
`antimicrobial activity after 7 clays of incubation at
`37°C. 17 Later analysis showed that 50% of the antimi(cid:173)
`crobial acti,~ty of l or 5 J.,Lg/mL concentrations of
`RP1vi are lost after only 24 hours of incubation in
`culture medium.;"
`High-pressure liquid chromatography (HPLC)
`has also been used to examine the stability of RPM in
`
`3l-.Membered macrm:ydic lactone Cj 1H,ttNOu FW = 91+.2
`3-'k I ratio of cis-trans rotamcrs abnut the pipecolic acid ?\-CO bond
`White, crystalline solid .MP 183-185 C
`
`Solubility:
`20 f,Lg/ mL in water
`sparingly soluble in rther
`soluble in methanol, ethanol, acetone, chloroform, methylene dichloride, trichloroethane, dimethyl forma(cid:173)
`m ide, dimethyl acetamide, dimethyl sulfoxide
`
`Stability (degradation by hydrolysis):
`
`Temperature
`
`25°C:
`
`37.5°C:
`
`Vehicle
`
`act>late buffer
`phosphate
`huller
`acetate buffer
`phosphate
`bufrer
`human
`plasma
`rat plasma
`
`pH
`
`3.3
`
`i. I
`3.3
`
`7.-J.
`
`TV:! (hrs) by HPLC
`
`3.1.8
`
`n.G
`9.CJ
`
`10.2
`
`3
`2.8:1
`
`

`
`different diluents subjected to different tempera(cid:173)
`tures'" (Table 3). RPM is particularly susceptible to
`degradation in plasma and in low and neutral pH
`buf