Volume 7
`Number 1
`February
`1986
`
`Endocrine
`Reviews
`
`TINGE-1S LIBRP1/47),
`Lii
`Univer54 of Wisr,or.sn
`53706
`1r305 Linden Dr . ;40.11SCM, Wis.
`.0 "t0 0
`
`MAR
`
`SUED QUARTERLY FOR
`he Endocrine Society
`N 0163-769X
`
`4
`
`MYLAN EXHIBIT - 1025
`Mylan Pharmaceuticals, Inc. v. Bausch Health Ireland, Ltd. - IPR2022-00722
`
`

`

`Editorial Board
`
`BOARD
`
`EDITOR Pentti Siiteri, Ph.D.,
`University of California San Francisco
`School of Medicine HSW-1656
`Department of Obstetrics/Gynecology-Reproductive Medicine
`San Francisco, CA 94143
`L. E. Braverman, M.D.
`University of Massachusetts Medical School, Worcester, MA
`J. W. Funder, M.D.
`Medical Research Center Prince Henry's Hospital, Melbourne, Victoria, Australia
`M. M. Grumbach, M.D.
`University of California San Francisco, San Francisco, CA
`E. Herbert, Ph.D.
`Oregon Health Sciences University, Portland, OR
`B. S. Katzenellenbogen, Ph.D.
`University of Illinois, Urbana, IL
`J. L. Kostyo, Ph.D.
`University of Michigan Medical School, Ann Arbor, MI
`H. H. Samuels, M.D.
`New York University Medical Center, New York, NY
`R. S. Sherwin, M. D.
`Yale University School of Medicine, New Haven, CT
`F. Sweet, Ph.D.
`Washington University Medical School, St. Louis, MO
`R. I. Weiner, Ph.D.
`University of California San Francisco, San Francisco, CA
`J. D. Wilson, M.D.
`University of Texas Southwestern Medical School, Dallas, TX
`S. S. C. Yen, M.D.
`University of California at San Diego, La Jolla, CA
`
`HEALTH
`
`SCIENCES L
`r
`University of
`1305 Linden
`VVisconsin IBRA .lt
`
`Madison, Wis. 53706
`:1 666
`MAR
`
`Contents
`
`Vol. 7, No. 1, February 1986
`
`Editorial Note
`
`Introduction
`
`The Molecular Basis of Gonadotropin-Releasing
`Hormone Action
`
`Hypogonadotropic Disorders in Men and Women:
`Diagnosis and Therapy with Pulsatile
`Gonadotropin-Releasing Hormone
`
`Pentti Siiteri
`
`P. Michael Conn
`
`P. Michael Conn
`
`Nanette Santoro, Marco Filicori,
`and William F. Crowley, Jr.
`
`

`

`Use of a Potent, Long Acting Agonist of
`Gonadotropin-Releasing Hormone in the Treatment
`of Precocious Puberty
`
`Induction of Ovulation in Primate Models
`
`Gonadotropin-Releasing Hormone Analog Design.
`Structure-Function Studies Toward the
`Development of Agonists and Antagonists: Rationale
`and Perspective
`
`Treatment of Prostate Cancer with Gonadotropin-
`Releasing Hormone Agonists
`
`Paul A. Boepple,
`M. Joan Mansfield,
`Margaret E. Wierman,
`Craig R. Rudlin,
`Hans H. Bode,
`John F. Crigler, Jr.,
`John D. Crawford, and
`William F. Crowley, Jr.
`
`Daniel Kenigsberg,
`Burt A. Littman, and
`Gary D. Hodgen
`
`Marvin J. Karten and
`Jean E. Rivier
`
`F. Labrie, A. Dupont,
`A. Belanger, R. St-Arnaud,
`M. Giguere, Y. Lacourciere,
`J. Emond, and G. Monfette
`
`Gonadal Regulation of Hypothalamic Gonadotropin-
`Releasing Hormone Release in Primates
`
`Tony M. Plant
`
`Treatment of Breast Cancer with Gonadotropin-
`Releasing Hormone
`
`Andrea Manni, Richard Santen,
`Harold Harvey, Allan Lipton, and
`Devorah Max
`
`Pharmacokinetics of Gonadotropin-Releasing
`Hormone and Its Analogs
`
`David J. Handelsman and
`Ronald S. Swerdloff
`
`24
`
`34
`
`44
`
`67
`
`75
`
`89
`
`95
`
`Mechanisms of Gonadotropin-Releasing Hormone
`Agonist Action in the Human Male
`
`Comparison of the Potential for Therapeutic
`Utilities with Gonadotropin-Releasing Hormone
`Agonists and Antagonists
`
`Special Issue Index
`
`Endocrine Society Guidelines for Membership
`Application
`
`Endocrine Society Nomination for Membership
`
`The Endocrine Society 68th Annual Meeting
`
`Registration Form
`
`S. Bhasin and R. S. Swerdloff
`
`106
`
`Brian H. Vickery
`
`115
`
`125
`
`126
`
`127
`
`128
`
`129
`
`

`

`AUTHOR AND SUBSCRIPTION INFORMATION
`
`Address manuscripts and correspondence to:
`Pentti K. Siiteri, Ph.D., Editor
`543 MeClay Road
`Novato, California 94947
`(415) 898-1581
`
`All business matters, including correspondence and remittance relating to nonmember subscriptions,
`back volumes and advertising, and orders and remittance for reprints, should be sent to ENDOCRINE
`REVIEWS, 428 E. Preston St., Baltimore, Md. 21202. Copies will not be replaced without charge
`unless we receive a request within 60 days of the mailing in the U.S. or within 90 days in all other
`countries.
`Bound volume number 6 will be available in early 1986 to subscribers only. The volume will be hand
`bound in black buckrum, and the journal name, volume number, and year of issue will be stamped in
`gold on the spine. The price is $45.00 (plus an additional $5.00 for orders outside the U.S.) with a
`$2.00 discount given for advance payment. Orders must be received by the publisher before December
`1, 1985.
`Change of Address: Notify both Williams & Wilkins, 428 E. Preston St., Baltimore, Md. 21202,
`and The Endocrine Society, 9650 Rockville Pike, Bethesda, Md. 20814, of all address changes.
`Publication of advertisements in the Journals of The Endocrine Society does not imply endorsement
`by the Society of the reagents or services offered.
`No part may be duplicated or reproduced without permission of the Publisher and the Editor.
`Printed in U.S.A.
`
`THE ENDOCRINE SOCIETY
`
`Bert W. O'Malley, President
`Sidney H. Ingbar, President-Elect
`
`Nicholas P. Christy, Secretary-Treasurer
`
`PUBLICATIONS COMMITTEE-HILTON A. SALHANICK, Chairman; BERT W. O'MALLEY,
`Ex-Officio; SIDNEY H. INGBAR, Ex-Officio; NICHOLAS P. CHRISTY, Ex-Officio; ALFRED E.
`WILHELMI, Ad Hoc; GLENN D. BRAUNSTEIN, MARY F. DALLMAN, HUNTER HEATH III, PHILIP
`TROEN.
`
`COUNCIL-GERALD D. AURBACH, LOUIS V. AVIOLI, LEWIS E. BRAVERMAN, DELBERT A.
`FISHER, LAWRENCE A. FROHMAN, RICHARD HORTON, SELNA L. KAPLAN, JACK L. KOSTYO,
`JACK H. OPPENHEIMER, JEAN D. WILSON.
`
`Norm C. KARPIN, Executive Director, 9650 Rockville Pike, Bethesda, Maryland 20814.
`
`Correspondence relating to membership applications and Society business including remission
`of members' dues and subscriptions should be sent to 'Nicholas P. Christy, Secretary-Treasurer,
`9650 Rockville Pike, Bethesda, Maryland 20814.
`
`ENDOCRINE REVIEWS (ISSN 0163-769X) is published quarterly by The Endocrine Society, 428 E. Preston St.. Baltimore,
`MD 21202. Member price: $20.00 ($27.00 foreign) for journal subscription. Subscription rates: individuals $45.00; foreign $55.0O;
`institutions $60.00, foreign $70.00; in-training $30.00 ($40.00 foreign); single copy $10.00 ($12.00 foreign). Subscription prices
`subject to change. Japanese Yen price is available from our sole agent U.S.-Asiatic Co. Ltd., 13-12, Shimbashi 1-Chome, Minato-
`Ku. Tokyo 105, Japan, telephone 03-502-6471. Second class postage paid at Baltimore, MD and at additional mailing offices.
`Postmaster, send address changes (Form 3579) to Williams & Wilkins, 428 E. Preston St., Baltimore, MD 21202. Indexed by
`Current Contents and Index Medicus. Copyright IL, 1986 by The Endocrine Society.
`
`

`

`0163-769X/86/0701-0044$02.00/0
`Endocrine Reviews
`Copyright© 1986 by The Endocrine Society
`
`Vol. 7, No. 1
`Printed in U.S.A.
`
`Gonadotropin-Releasing Hormone Analog Design.
`Structure-Function Studies Toward the Development of
`Agonists and Antagonists: Rationale and Perspective
`
`MARVIN J. KARTEN AND JEAN E. RIVIER
`Contraceptive Development Branch (M.J.K.), Center for Population Research, National Institute of Child
`Health and Human Development, National Institutes of Health, Bethesda, Maryland 20892; and The
`Clayton Foundation Laboratories for Peptide Biology (J.E.R.), The Salk Institute, La Jolla, California 92037
`
`Introduction
`
`ON JUNE 24, 1971, Andrew V. Schally announced
`
`the determination of the primary structure of por-
`cine GnRH at The Endocrine Society Meeting in San
`Francisco. This announcement was followed by publica-
`tions by Matsuo et al. (1, 2) and Baba et at. (3) on the
`proposed amino acid sequence for porcine GnRH and its
`synthesis and by Burgus et al. (4) who characterized
`ovine GnRH and found the sequence to be identical with
`that of porcine GnRH. The physiological and therapeutic
`importance attributed to the discovery of the new sub-
`stance was greatly increased by the prospect of the design
`of potent and long acting GnRH agonists and antago-
`nists. Since that time more than 2000 analogs of GnRH
`have been synthesized. The impact of research of GnRH
`and its analogs on clinical medicine recently led Ziporyn
`(5) to note, "There's almost no subspecialty of medicine
`that will be left untouched by the [research] advances
`associated with LHRH or its analogs." It is the intent of
`this article to provide a historical review of the major,
`and some minor, aspects of the chemical development of
`GnRH agonists and antagonists up to the present state
`of development (July 1, 1985). The synthetic chemical
`efforts have been devoted largely to increasing the affin-
`ity of the peptides to the GnRH receptor and their
`resistance to degradation or elimination in in vivo sys-
`tems, characteristics which, for the GnRH analogs, are
`generally interrelated.
`An annual compilation and review of structure-activity
`relationships of GnRH analogs is available in the Spe-
`cialist Periodical Reports which cover the literature pub-
`lished during 1971-1980 (volumes 4-13) (6-15). At irreg-
`ular intervals, Annual Reports in Medicinal Chemistry
`
`Address requests for reprints to: Dr. Marvin J. Karten, Contracep-
`tive Development Branch, Center for Population Research, National
`Institute of Child Health and Human Development, National Institutes
`of Health, Bethesda, Maryland 20892.
`
`furnish brief updates of the studies of structure-activity
`relationships of GnRH analogs (16-21). These reports,
`in conjunction with two recently published comprehen-
`sive monographs should provide the reader with a bal-
`anced account of the development of GnRH analogs (22-
`29).
`
`Synthesis, Purification, and Characterization of
`GnRH Analogs
`
`It is important to note that the rapid development of
`GnRH analogs was made possible through the extensive
`use of solid phase peptide synthesis (SPPS) introduced
`by Merrifield (30). As one of the codevelopers of the
`method (31), Stewart (32) has pointed out that the use
`of automated equipment for SPPS, benzhydrylamine-
`like resins for peptide amide synthesis (33, 34), and
`adequate methods for the purification of peptides, par-
`ticularly reverse phase HPLC (RP-HPLC) in recent
`years (35), have made the synthesis of mammalian
`<Glu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-
`GnRH,
`Gly-NH2, and its analogs, a relatively simple task. Al-
`though classical (solution) methods have been employed
`for the synthesis of GnRH (see Ref. 36 and references
`therein) and its analogs (37), it is quite clear that the use
`of SPPS and RP-HPLC were essential for the rapid
`exploration of structure-activity relationships as well as
`providing investigators with relatively large amounts of
`these substances for pharmacological, toxicological, and
`clinical studies.
`While the purity of the agonists synthesized either by
`SPPS or classical (solution) methods was always of con-
`cern in terms of the concomitant biological activity of
`potential racemization products, it was a critical factor
`in the biological evaluation of the antagonists. This was
`particularly true in the early stages of development when
`the GnRH inhibitory activities were very low and could
`be masked by a small amount of racemized material
`
`44
`
`

`

`February, 1986
`
`GnRH ANALOG DESIGN
`
`45
`
`present as a contaminant and acting as an agonist. For
`example, the high GnRH potency of [D-Alal GnRH
`(350-450% of GnRH) would barely be affected by a 10%
`racemization contaminant of [L-Alal GnRH (4% of
`GnRH) (38). However, a very weak antagonist, e.g. [D-
`Phe9GriRH, (39) could have its activity masked by the
`presence of small amounts of [L-Phe2]GnRH (40) with
`only 2-4% of the potency of GnRH but with full intrinsic
`activity in vitro. The separation of the possible diaster-
`corners, e.g. [D-Ala6]GnRH and [L-Alal GnRH, which
`could result from racemization, was eventually made
`feasible through the use of HPLC, thus eliminating one
`element of uncertainty from the interpretation of the
`biological results (35). Similarly, a preparation of [D-
`HislGnRH exhibiting 10% GnRH-like potency (39) was
`subsequently shown, when purified by HPLC, to be
`inactive either as an agonist or an antagonist (41).
`Racemization of histidine during peptide synthesis is
`well documented and difficult to prevent irrespective of
`whether classical methods of synthesis or SPPS are used.
`The widespread use of HPLC resulted from the recog-
`nition that classical methods of purification were inferior
`to HPLC as a tool for the separation of these diastereo-
`meric peptides. Most GnRH analogs reported in the
`literature have been characterized by amino acid analysis
`only, often without a quantitative determination of the
`unnatural amino acids. Investigators have relied on the
`presumed authenticity of the starting protected amino
`acid and on high coupling efficiency during the assem-
`bling of the peptides on the resins rather than pursuing
`rigorous methods of characterization of the peptides.
`Nuclear magnetic resonance (NMR) and mass spectrom-
`etry were employed in those cases where definite proof
`of structure was required. Optical rotations were gener-
`ally measured, and TLC and HPLC, in several systems,
`were used for proof of homogeniety.
`
`Development of GnRH Agonists
`
`The original incentive for the development of more
`potent GnRH agonists was the expectation that the
`knowledge of the LH-releasing and ovulation-inducing
`effects of GnRH observed in laboratory animals could be
`applied to the treatment of male and female infertility
`(42). However, the half-life of GnRH is very short (43,
`44) and more potent and longer acting analogs were
`thought to be necessary for practical clinical utility,
`regardless of any anticipated or unanticipated therapeu-
`tic applications. Potent agonists, referred to as supera-
`gonists, were rapidly produced and were subsequently
`discovered, along with GnRH, to have, ironically, anti-
`reproductive effects. They were available for reproduc-
`tive pharmacological evaluation (42, 45, 46) within 3 yr
`Of the structure elucidation of GnRH; this accounts for
`
`their rapid clinical exploration exemplified by the first
`demonstration (in 1978) of inhibition of reproductive
`function in women by a superagonist (47). Once super-
`agonists had been synthesized and their potential ther-
`apeutic value was recognized, further incentive, after
`1976, to seek structurally novel and longer acting ago-
`nists was provided by promising commercial considera-
`tions.
`The various biological assays and animal models that
`have been utilized for the testing of GnRH agonists have
`been recently reviewed (42, 48). The most widely used in
`vitro assays have been, initially, the dispersed pituitary
`cell for the measurement of LH and FSH secretions and,
`more recently, the receptor binding assay using purified
`pituitary membrane fractions for the estimation of the
`potencies of the analogs (49, 50). In vivo biological assays,
`which have been utilized to determine the potencies of
`GnRH agonists, include induction of ovulation, disrup-
`tion of the estrus cycle, stimulation of uterine growth,
`inhibition of pregnancy, stimulation of LH release in
`ovariectomized rats, and stimulation of LH/FSH release
`in immature rats using an infusion technique (42, 48).
`
`Prokethylamide (NEt) modifications
`
`The first important structural modification of GnRH
`leading to increased potency was discovered by Fujino
`and co-workers (37), who examined the effect of struc-
`tural modifications at the C terminus of GnRH. Although
`the des-amide of GnRH (GnRH free acid) exhibited very
`low GnRH potency
`in ovariectomized rats (51)
`and Pre-GnRH showed only 10% of the potency of
`GnRH in vitro (34), replacement of the glycine amide
`terminus with alkyl amines produced nonapeptide alkyl
`amides with significantly greater ovulation-inducing po-
`tency than GnRH itself (37). Thus, [Pro9-ethylamide
`(NEt)]GnRH, the most potent analog of the series, was
`5 times more potent than GnRH and more potent than
`either the Pro9-methylamide (NHMe) or Prokpropylam-
`ide (NHPr) modifications. The fluorinated ethylamide
`analog, [Pro9-NHCH2CF3]GnRH was subsequently re-
`ported by Coy et al. (52) to be twice as potent as [Pro9-
`NEt]GnRH in releasing LH when administered to im-
`mature male rats. The data of Fujino et al. (37), it was
`noted, suggested that the terminal glycine amide was not
`essential for high potency and that the total chain length
`of the peptide played an important role in the binding
`affinity of the analog for the pituitary receptor. It was
`also suggested that the introduction of this Pro-alkylam-
`ide moiety may also increase the duration of action of
`these analogs by virtue of their greater resistance to
`enzymatic degradation (53). These two desired proper-
`ties, greater binding affinity and enzymatic resistance to
`proteolysis, were to become the basis for the rational
`
`

`

`46
`
`KARTEN AND RIVIER
`
`Vol. 7, No. 1
`
`design and for the explanation of activity, or lack thereof,
`of all the GnRH analogs regardless of the site of struc-
`tural modification. The concept that protection from
`renal elimination would also lead to prolonged action
`was eventually incorporated into the design of GnRH
`analogs.
`
`D-Xace modifications
`The second important structural modification of
`GnRH, discovered by Monahan et al. (38), was the re-
`placement of the Gly6 residue with D-alanine yielding [D-
`AlaiGnRH with a potency of approximately 350-450%
`that of GnRH both in vitro and in ovariectomized rats.
`The corresponding [1, -Ala9 GnRH had only 4% of the
`potency of GnRH, and it was suggested that since the
`potencies determined in vivo were in agreement with the
`in vitro results, it was unlikely that the differences in the
`biological activities could be solely explained by differ-
`ences in clearance rates. Instead, the increased biological
`potency was attributed to the conformational stabilizing
`effect of the D-alanine which was favorable for binding
`(and activity) at the receptor. This study is also note-
`worthy for its suggestion that GnRH may conformation-
`ally contain a 0-II type bend (involving Ser-Tyr-Gly-
`GnRH and preferable
`Leu) which is stabilized in
`for binding at the receptor site. This point will be dis-
`cussed below when the contribution of conformational
`studies to the design of GnRH analogs, particularly
`antagonists, is reviewed. Also to be deferred for later
`discussion is our knowledge of the enzymatic degradation
`of peptides and its contribution to the design (if any) of
`GnRH analogs. However, it should be noted that, regard-
`less of whether the Gly6—Leu7 bond or the Tyr5-Gly6 bond
`is considered to be a major site of proteolytic cleavage,
`substitution of glycine by O-amino acids is likely to
`render either linkage more resistant to enzymatic deg-
`radation (26).
`
`Additive effects (or lack thereof) of Pro9-NEt and D-Xaa6
`modifications
`It is often assumed that the biological effect of com-
`bining several structural changes in one molecule will be
`additive (or, more correctly, multiplicative). Thus, ac-
`cording to this additivity rule (54), if one structural
`modification leads to a relative potency of a and a second
`modification to a relative potency of b then the combi-
`nation of both structural modifications in a single mol-
`ecule would be expected to yield an analog with a biolog-
`ical potency of a multiplied by b. The potential additivity
`of biological potency of the Pro9-NEt and D-Ale modi-
`fications was immediately tested and, in rapid succession,
`two reports [Coy et al. (55) and Fujino et al. (56)] on this
`important combination appeared. Infusion experiments
`
`with immature male rats showed that the three analogs,
`[D-Ale,Pro9-NEt]GnRH,
`GnRH, and [Pro9-
`NEt] GnRH, had LH/FSH releasing potency ratios of
`12-16:7-8:2.5, respectively, compared with GnRH (55).
`Similar ovulation-inducing potency ratios were observed
`(56) among the three analogs, but the potencies relative
`to GnRH were much higher. These in vivo results cor-
`related with the in vitro results of Vale et al. (57) who,
`using the stimulation of LH secretion from rat pituitary
`cells in culture by GnRH agonists as an index of potency,
`noted that the combination of the two structural modi-
`fications yielded an analog with a potency approximately
`equal to the product of the potencies of the individual
`modifications. The in vitro results of Fujino et al. (56)
`did not agree with these findings. Although it is self-
`evident that comparisons of biological data by different
`groups are valid only if the same bioassays are employed
`in precisely the same manner (48), nevertheless, many
`of the apparent disagreements in the data reported,
`regarding comparisons of potencies of superagonists, can
`be attributed to a disregard of this axiom. Thus, the
`validity of the extension of the additivity principle to the
`combination of a D-aromatic amino acid in position 6
`and the Pro9-NEt modification became a focal point of
`interest with the publication of apparently conflicting
`reports on this subject (57-59). The observation was
`made that agonists with D-aromatic amino acids such as
`[D-Trpl GriRH and [D-Trp6,Pro9-NEt]GnRH are much
`more potent (36 times and 144 times the in vitro potency
`of GnRH, respectively) than the corresponding substi-
`tutions with aliphatic amino acids, such as [D-Ale]
`LHRH and [D-Ala6,Pro9-NEt]GnRH, which are approx-
`imately 4 times and 14 times, respectively, the potency
`of GnRH (57). The subsequent binding studies of Perrin
`et al. (50) showed increased binding potencies for the
`Pro9-NEt modifications as compared with [D-Alas]
`GnRH or [D-TrpiGnRH; however, when compared with
`the biological potencies, in stimulating LH secretion in
`vitro, the increases were far less dramatic. The data for
`the D-Trp6 analogs are in accord with the subsequent
`binding studies of Barron et al. (60) but are not in accord
`with the earlier in vitro data reported by Coy et at. (58).
`In vivo measurements in immature rats comparing [D-
`LeulGnRH and [D-Leu6,Pro9-NEt]GnRH (Leuprolide),
`using integrated levels of LH over a 6-h period after
`injection, showed a similar additive effect (61), but the
`additive effect was reported not to hold true for the
`corresponding 0-aromatic amino acid modifications in
`this same assay system (59). Thus, in male rats, the Pro9-
`NEt modification was reported to decrease the potency
`of [n-Phe6]GnRH and [D-Trp6]GnRH by a factor of
`nearly 2 (59). However, it has been more recently re-
`ported that, using estrus suppression (62) as an index of
`agonist activity, [D-Trp6]GnRH and [D-Trp6,Pro9-NEt]
`
`

`

`February, 1986
`
`GnRH ANALOG DESIGN
`
`47
`
`GnRH were equipotent. Postcoital comparisons in rats
`(Naqvi, R. and M. Lindberg, unpublished observations)
`also indicated that the two analogs were equipotent. In
`women, it was reported that the two analogs were equi-
`potent with respect to the sc dose required for maximal
`LH release (63). Barron et al. (60) showed that MCRs in
`pregnant women were similar for both analogs. They
`concluded that since the NEt residue, which is reported
`to protect the peptide from postproline-cleaving enzyme
`activity, did not lead to a prolonged survival time in
`pregnant women, degradation by this enzyme in human
`tissues contributes minimally to GnRH clearance. Sup-
`port for this conclusion was found in reports that both
`analogs were equipotent in vivo (42, 64). Thus, the over-
`whelming evidence points to in vivo equipotency for the
`two D-Trp6 analogs irrespective of the in vitro results
`and binding studies supporting the additive effects on
`the biological potency of the D-Trp6 and Pro9-NEt mod-
`ifications.
`
`Hydrophobic modifications at position 6
`
`The trend toward seeking more potent agonists with
`increasing hydrophobic character resulted in the addition
`of two more superagonists to the growing list of analogs
`available for clinical exploration. [Nr-Bzl-D-His9,Pro9-
`NEt]GnRH (Histerelin) was designed by Rivier et al.
`(22) to have the characteristics of high water solubility
`at acidic pH and greater lipophilic character in vivo,
`while retaining high biological potency. A correlation
`was noted between the in vitro potencies of certain
`position 6 superagonists and their HPLC retention times
`at physiological pH. These correlations included [Nr-
`BzI-D-Hise,Pro9-NEt]GnRH and
`[D-Trp6,Pro9-NEt]
`GnRH, with the former being slightly more potent than
`the latter, in vitro. In vivo results showed a similar trend
`(48).
`It had previously been observed that the incorporation
`of 13-amino acids, with larger and more lipophilic side
`chains than in [D-LeulGnRH, such as [13-TrpiGnRH,
`resulted in more potent agonists (57, 59). Nadasdi and
`Medzihradszky (65) proposed a quantitative correlation
`to exist between the potency of position 6 substituted
`GnRH analogs and the calculated hydrophobicity of the
`amino acid side chain. It is accepted that increased
`lipophilicity of drugs is generally associated with greater
`retention of the drug in the body and, therefore, pro-
`longed duration of action (66). The retention may be the
`result of enhanced renal reabsorbtion or fat storage of
`nonionized fat-soluble compounds. Protection of the
`drug from renal excretion, through plasma protein bind-
`ing, will also affect its duration of action. Plasma protein
`binding generally increases, in a given series of analogs,
`With increasing hydrophobicity (66). With this in mind,
`
`Nestor et al. (62, 67) postulated that analogs with greater
`hydrophobicity could have an extended biological half-
`life resulting from a whole body depot effect. They would
`attribute this effect to a decreased rate of clearance of
`the analog from the circulation and the increased binding
`capacity of the analog for hydrophobic plasma carrier
`proteins (67). The results of the study on a wide range
`of hydrophobic analogs showed that the most potent
`ones were found in a hydrophobicity range, as measured
`by their retention time on RP-HPLC (68), greater than
`that of the D-Trp6 analogs. As an example, [D-Nal(2)6]
`GnRH (Nafarelin acetate), the most potent of this series,
`was reported to be 200 times more potent than GnRH in
`suppressing estrus in rats. It was pointed out, however,
`that analogs with greater hydrophobicity than Nafarelin
`acetate were less potent and this includes the analog
`incorporating the Pro9-NEt modification into Nafarelin
`acetate
`(62).
`Interestingly,
`[13-Nal(2)1GnRH and
`[D-Nal(1)6]GnRH were isolipophilic but the latter was
`4-fold less potent. Other examples (68, 69) also bear
`witness to the inadequacy of using hydrophobicity alone
`as a prediction of agonist potency (28).
`Nafarelin acetate, which was reported to be twice as
`potent as [D-TrplGnRH, [D-Trp6,Pro9-NEt]GnRH, or
`[N'-Bzl-D-His9,Pro9-NEt]GnRH in estrus suppression
`comparisons (26), became the last superagonist to be
`made available for clinical exploration. The improvement
`of the pharmacokinetics with the more hydrophobic ag-
`onist may appear to be supported by the comparisons of
`the reported half-lives of GnRH [tv, = 8 min, constant
`infusion, (44)], [D-Trp6]GnRH [t = 30 min, constant
`infusion, (44)], and [D-Nal(2)9GnRH Eta = 2.4 h, sc
`administration, (67)]; however, comparison of the three
`peptides under identical conditions is not available. The
`considerably longer plasma elimination half-lives re-
`ported for Nafarelin acetate in rats, monkeys, and hu-
`mans than reported for GnRH or [D-Leu6,Pro9-NEt]
`GnRH (Leuprolide) were attributed, at least in part, to
`the more extensive plasma binding of Nafarelin acetate
`(70).
`N", AP" -dialkyl-D-homoarginines were incorporated
`into position 6 of GnRH agonists (67, 71) as a result of
`successful GnRH antagonist investigations with this un-
`natural amino acid. The most potent, [Nw,AP"-diethyl-
`D-Hars,ProkNEt]GnRH, was only slightly less potent
`than [D-Nal(2)6]GnRH in the rat estrus suppression
`assay.
`
`Other C-terminally modified analogs
`
`Another structural modification which has generally
`led to increases in potency, in combination with 13-amino
`acids in position 6, is the a-aza-G1y10(-NHNHCO-) sub-
`stitution. A series of a-aza analogs of GnRH were syn-
`
`

`

`48
`
`KARTEN AND RIVIER
`
`Vol. 7, No. 1
`
`thesized by Dutta et al. (72, 73) with the expectation that
`the presence of an a-aza residue might be conformation-
`ally favorable, leading to higher binding affinity, and be
`more resistant to enzymatic degradation. Replacement
`of amino acids in position 6, 9, and 10 of GnRH by a-
`aza amino acids alone did not confer any potency advan-
`tage but when the a-aza-G1y10 modification was com-
`bined with, for example, the D-Ser(But)6 substitution,
`GnRH,
`the resulting analog, [D-Ser(But)6, a -
`currently undergoing clinical development, was consid-
`ered to be at least 5 times more potent than [D-
`Ser(But)6,Pro9-NEt]GnRH, Buserelin, (74), using induc-
`tion of ovulation as a measurement of potency (72, 73).
`(Buserelin has been, clinically, the most extensively stud-
`ied GnRH analog.) It was unclear, to Dutta et al., which
`individual factor was primarily responsible for this en-
`hancement of biological potency. [2-o-Nall, a-aza-Glyw]
`GnRH was reported to be slightly more potent than
`Nafarelin acetate and approximately 2.5 times more po-
`tent than the corresponding Pro9-NEt modification in
`the estrus suppression assay (26). However, Nestor (26)
`noted that if the a-aza-Gly1° substitution conferred high
`potency by virtue of its enhanced resistance to the post-
`proline-cleaving enzyme in rat plasma, then the rele-
`vance of this substitution to human therapy was less
`clear since the amount of postproline-cleaving enzyme
`present in human plasma was reported to be 2-5 times
`less than in rat plasma (75, 76). Does the statement of
`Nestor also imply that any C-terminal amide modifica-
`tions generally would not confer any advantage over the
`parent Gly1°-NH2 function in humans? On the basis of
`the human data available on [D-Trp6]GnRH and [D-
`Trp6,Pro9-NEt]GnRH (60, 63, 64), the answer would
`appear to be yes, although systematic comparisons would
`have to be made.
`
`Ser(But) groups at position 6 and 7 not only failed to
`enhance, but actually decreased, the ovulation-inducing
`potency of Buserelin.
`
`Conformationally constrained and backbone modifications
`With the recognition that the biological activity of [D-
`Ala6, N-Me-LeulGnRH was consistent with that of a /3-
`turn conformation for residues 5-8 of GnRH, Freidinger
`et al. (78) introduced a y-lactam as a conformational
`constraint into the 6,7 position of GnRH and found that
`the resulting analog was 9 times more potent than GnRH
`in vitro and, by iv injection in ovariectomized rats, 2.4
`times more potent than GnRH. Further exploration of
`the 1, -lactam modification has not been made with ago-
`nists, per se, but this modification has been tried with
`GnRH antagonists. Various attempts to obtain a confor-
`mationally restricted agonist through cyclic analog de-
`sign have yielded inactive analogs (79) or agonists with
`low biological potency (22, 80). Spatola (81) has reviewed
`the effect of peptide backbone modification on structure-
`activity relationships. Backbone modifications, as new
`approaches to GnRH agonist design, resulted in rela-
`tively little in vitro potency in the cases of peptide bond
`reversals at the 5-6 or 5-6 and 6-7 position (retro-inverso
`analogs) (82) or in the cases of substitution of a thiom-
`ethylene (-CH2S-) group for the peptide linkage at the
`5-6, 6-7, or 9-10 position of GnRH. The latter substi-
`tution at 9-10 had 10% of the in vitro potency compared
`to that of GnRH, indicating the necessity of more precise
`conformational requirements for residues 5-8 than for
`residues 9-10 (23).
`Before closing the discussion on the current stage of
`development of the GnRH agonists, it is necessary to
`comment upon efforts to increase the potency of the
`agonists by modifying other amino acid residues.
`
`Position 7 modifications
`
`Miscellaneous modifications
`
`[N-Me-LeuiGnRH was found to be equipotent with
`GnRH and [D-Ala6,N-Me-LeulGnRH was found to be
`at least as active as [n-Ala6]GnRH in vitro (77). In fact,
`the N-Me-Leu7 modification has been incorporated into
`[D-Trp6,Pro9-NEt]GnRH
`yielding
`[D-Trp6,N-Me-
`Leu7,Pro9-NEt]GnRH, an analog currently undergoing
`clinical development (63). Generally, the effect of the N-
`Me-Leu" modification in enhancing the potency of the
`parent peptide depends on the D-amino acid at position
`6 and the bioassay used to compare their potencies (45).
`The introduction of the bulky alkyl