`
`
`
`
`
`
`
`INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PCT)
`
`WORLD INTELLECTUAL PROPERTY ORGANIZATION
`
`International
`Bureau
`
`
`
`
`
`(51) International Patent Oassification 5 :
`
`
`
`(11) lntematiqnal Pnblication Number:
`
`WO 91/11457
`
`
`
`Al
`
`
`C07K 7 /34, 7 /10, A61K 37 /02
`A61K37/28
`
`( 43)International Publication Date:
`
`
`
`8 August 1991 (08.08.91)
`
`(2l)lntemationalApplicationNumber: PCT/US91/00500 (74)Agents: MURASHIGE, Kate, H. et al.; Irell & Manella,
`
`
`
`
`
`
`
`545 Middlefield Road, Suite 200, Menlo Park, CA 94025
`(US).
`1991 (24.01.91) (22)International Filing Date:24 January
`
`
`
`
`
`
`
`(81)Designated States: AT (European
`
`
`
`patent), BE (European
`(30)Priority data:
`
`
`
`
`
`patent), CA, CH (European patent), DE (European pa
`468,136 24 January 1990 (24.01.90)
`us
`
`
`
`
`
`tent), DK (European patent), ES (European patent), FR
`
`
`
`
`(European patent), GB (European patent), GR (Euro
`
`
`
`pean patent), IT (European patent), JP, LU (European
`(60)Parent Application or Grant
`
`
`
`
`
`
`patent), NL (European patent), SE (European patent),
`
`(63)Related by Continuation
`468,736 (CIP)
`us.
`usFiled on
`
`24 January 1990 (24.01.90)
`
`Published
`
`
`
`
`(71)(72) Applicants and Inventors: BUCKLEY, Douglas, I. [US/
`
`With intemational search report.
`
`
`US]; 215 Brookwood Road, Woodside, CA 94062 (US).
`
`
`
`HABENER, Joel, F. [US/US]; 217 Plymouth Road,
`
`Newton Highlands, MA 02161 (US). MALLORY,
`
`
`
`Joanne, B. [US/US]; 243 Acalanes, Apt. 9, Sunnyvale,
`
`
`
`CA 94086 (US). MOJSOV, Svetlana [YU/YU]; 504 East
`
`63rd Street, New York, NY 10021 (US).
`
`
`
`(54)Title: GLP-1 ANALOGS USEFUL FOR DIABETES TREATMENT
`
`
`
`(57) Abstract
`
`The invention provides effective analogs of the active GLP-1 peptides, 7-34, 7-35, 7-36, and 7-37, which have improved
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`characteristics for treatment of diabetes Type IL These analogs have amino acid substitutions at positions 7-10 and/or are trun
`
`
`
`
`
`
`
`
`cated at the C-terminus and/ or contain various other amino acid substitutions in the basic peptide. The analogs may either have
`
`
`
`
`
`
`
`
`
`
`
`an enhanced capacity to stimulate insulin production as conipared to glucagon or may exhibit enhanced stability in plasma as
`
`
`
`
`
`
`
`
`
`compared to GLP-1 (7-37) or both. Either of these properties will enhance the potency of the analog as a therapeutic. Analogs
`
`
`
`
`
`
`
`
`having D-amino acid substitutions in the 7 and 8 positions and/or N-alkylated or N-acylated amino acids in the 7 position are
`
`particularly resistant to degradation
`in vivo.
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 1 of 50
`
`
`
`FOR THE PURPOSES OF INFORMATION ONLY
`
`
`
`
`
`Codes used to identify States party to the PCT on the front pages of pamphlets publishing international
`
`
`
`
`applications under the PCT.
`
`AT Austria
`ES Spain
`AU Australia
`Fl Finland
`Barbados
`FR France
`BB
`Belgium
`GA Gabon
`BE
`
`Burkina Faso
`United Kingdom
`BF
`GB
`BG Bulgaria
`Guinea
`GN
`BJ Benin
`Greece
`GR
`Brazil
`HU Hungary
`BR
`Canada
`IT Italy
`Republic . JP Japan
`CA
`CF Central African
`CG Congo
`
`
`
`Switzerland
`of Korlllj
`CH
`Cote d'Ivoire
`KR Republic of Korea
`
`Cl
`CM Cameroon
`Liechtenstein
`LI
`Sri Lanka
`cs Czech0&lovakia
`LK
`DE Germany
`LU Luxembourg
`DK Denmarll.
`MC Monaco
`
`Madagascar
`MG
`ML Mali
`MN Mongolia
`Mauritania
`Malawi
`Netherlands
`NL
`NO Norway
`Poland
`PL
`RO -Romania
`Sudan
`SD
`Sweden
`KP Democratic People's Republic
`SE
`SN Senegal
`SU Soviet Union
`Chad
`TD
`TC Togo
`us United States of America
`
`
`MW
`
`MR
`
`II!
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 2 of 50
`
`
`
`WO 91/1457
`
`oo
`
`|
`
`PCT/US91/00500
`
`e
`
`-1-
`
`GLP-1 ANALOGS USEFUL FOR DIABETES TREATMENT
`“a
`
`This is a continuation<-in=part of U.S.
`Application Serial No. 468,736, filed 24 January 1990.
`
`10
`
`Technical Field
`
`The invention relates to the field of improved
`pharmaceutical compositions. Specifically,
`the invention
`concerns analogs of the glucagon-like peptide I fragment
`7-36 or 7-37 with improvéd pharmacological properties.
`
`15
`
`20
`
`25
`
`BackgroundArt
`Glucose metabolism is regulated by a number of
`peptide hormones,
`including insulin, glucagon, and
`gastric inhibitory peptide (GIP).
`’ The complex mechanism
`by which these peptide hormones regulate this metabolism
`and the manner in which they affecteach other is at
`least partially elucidated.
`For;vexample, glucagon binds
`to receptors on the surface of the pancreatic beta cells
`which produce insulin, and: stimulates insulin secretion.
`Glucagon-like peptide I has been suggested to stimulate
`insulin secretion but this has not-“been confirmed.
`Several of thege- hormones’originate from a mam-
`malian glucagon precursor “proglucagon": which is a 180
`amino acid peptide. - Proteolysis. and processing of this
`peptide results ina number of these protein hormones;
`“the results of the processing depend on the origin of the
`cells in which thisodcurs.
`For example,
`in the pig and
`rat pancreas, proglucagor is ‘processed to form glucagon
`and glicentin-related pancreatic ‘peptide, a large peptide
`
`30
`
`35
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 3 of 50
`
`
`
`WO 91/11457
`
`mc
`
`,
`
`PCT/US91/00500
`
`-2-
`
`which contains both GLP-1 and GLP-2 Sequences. ‘In
`porcine small intestine,
`the secreted products are the 69
`amino acid glucagon~-containing peptide glicentin and the
`two glucagon-like sequences; GLP-1 and GLP-2 as separate
`peptides.
`:
`
`the overall sequence of
`In any event, however,
`proglucagon contains the 29 amino acid sequence of
`glucagon,
`the 36 or 37 amino acid sequence of GLP-1 and
`the 34 amino acid sequence of GLP-2, separated by amino
`acid spacer sequences.
`Early attempts to assign a pattern of activity
`to GLP-1 gave ambiguous results, and it was subsequently
`concluded that truncated forms of this peptide are bio-
`logically active. Mojsov, S., et al. J Clin Invest
`(1987) 79:616-619 disclose that only the 31 amino acid
`peptide GLP-1 (7-37) strongly stimulates the release of
`insulin from pancreas; although both the truncated and
`full length 37 amino acid form had earlier been found in
`pancreas and intestine.
`It has been demonstrated that
`GLP-1 (7-36), possibly with the carboxy terminus
`- amidated, is also a potent mediator of insulin release.
`
`(See, e.g., Holst, J.J., et al. FEBS Letters (1987)
`
`211:169-174).
`The invention described below concerns analogs
`of these truncated forms of GLP-1, which have desirable
`combinations of characteristics as they relate to potency
`in potentiating glucose-induced insulin secretion and
`glucose-induced inhibition of glucagon secretion and to
`circulating half-life.
`The physiological effects of the
`truncated forms in potentiating glucose-induced insulin
`secretion have been shown as described above by Holst,
`J.J., et al. and Mojsov, S., et al.
`(supra). The
`_activity of the truncated hormones in inhibiting glucagon
`release has been shown by Orskov, C., et al. Endocrinol
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 4 of 50
`
`
`
`AREOUR yaa MIRE
`
`WO 91/11457
`
`,
`
`;
`
`PCT/US91/00500
`
`-3-.
`
`The
`
`(1988) 123:2009-2013; Suzuki, S., et al. Diabetes
`Research: Clinical Practice (1988) 5 (Supp. 1):S30.
`circulating half-life of these truncated forms is
`short--approximately four minutes as shown by Kreymann et
`al. The Lancet
`(December 5, 1987) 1300-1303.
`The
`modified forms of these truncated GLP-1 peptides provide
`the opportunity to optimize these properties.
`There is some literature relating to the study
`of degradation of peptidehormones in the liver and in
`plasma and the half-life .of such,,normones in vivo
`generally.
`An early paper by McDonald, J.K. et al., 7
`Biol Chem (1969) 244: 6199~6208 showed that a dipeptiaase
`was responsible for the degradation of glucagon in rat
`liver. Studies on the growth hormone releasing factor, a
`member of the general glucagon,GLP-1, GLP-2 family, was
`
`shown to be rapidly degraded in plasma in vitro and also
`
`in vivo by a dipeptidase,
`(Frohman, L.A. et al., J Clin
`Invest
`(1986) 78:906-913). Murphy, W.A. et al.,
`in
`Peptide Research (1988) 1236-41, showed that some but not
`all alkylated growth hormone releasing factor peptides
`
`had higher potency in vivo.
`In particular, for example,
`the triisopropylated GRF-29 was found to be 106 times
`more active than GRF-29 itself.
`On the other hand, GRF-—
`29 which. was in methylated at the N-terminus was only 40%
`as potent as the parent.
`It was also shown that
`substitution of D-Ala position 2 of this hormone enhanced
`its potency.
`It was, of course, not certain to what
`effect on properties the enhancement of potency could be
`attributed.
`|
`:
`.
`Others have attempted some modifications of
`GLP-1 (7-37).
`It has been shown that deletion of the
`histidine residue at position 7 greatly diminishes the
`activity of the hormone. (Suzuki, S., et al.
`(supra);
`Hendrick, G.K.,; et al. Abstract: Endocrine Society
`
`.
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 5 of 50
`
`
`
`WO 91/11457
`
`‘
`
`PCT/US91/00500
`
`_ 74-
`
`Meeting, New Orleans, LA (1988)). There have been
`conflicting reports concerning the effect of one or more
`C-terminal deletions (Suzuki, s., et al.
`(supra) ;
`Yanaihara, C., et al. Abstract for A Glucagon and Related
`Peptides Satellite Symposium, 8th International Congress
`of Endocrinology, July 15-16, 1988, Osaka, Japan).
`However,
`there is an extensive literature with regard to
`modifications of other members of this peptide hormone
`family, such as GIP, glucagon releasing factor (GRF),
`secretin and vasoactive intestinal peptide (VIP).
`
`Disclosure of the Invention
`The invention provides modified forms of the
`GLP-1 (7-34);
`(7-35);
`(7-36) or (7-37) human peptide or
`the C-terminal amidated forms thereof.
`The native
`peptides have the amino acid sequence:
`
`25
`20
`15
`10
`7
`H-A-E-G-T-F~T-S-D-V-S-S-Y-L-E-G-Q-A-A-
`
`37
`30
`K-E~F-I~A-W-L-V-K-(G)-(R)-(G)
`
`(R), and (G) are present or absent depending
`wherein (G),
`on indicated chain length.
`The modified forms contain
`one or more alterations of the native structure and are
`of improved ability for therapeutic use. Either the
`modified forms have greater potency than glucagon to
`potentiate insulin secretion or enhanced stability in
`plasma or both. This potency and enhanced stability can
`be assessed as described below.
`The standard one letter abbreviation code for amino
`_acids is used.
`The analogs of the invention which show enhanced ©
`insulin stimulating properties have the foregoing
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`°
`
`.
`
`%
`
`-
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 6 of 50
`
`
`
`WO 91/11457
`
`,
`
`PCT/US91/00500
`
`-5-
`
`sequence, or the C-terminal amide thereof, with at least
`one modification selected from the group consisting of:
`(a) substitution of a neutral amino acid, arginine,
`or a D form of lysine for lysine atposition 26 and/or 34
`and/or a neutral amino acid,
`lysine, or a D form of
`arginine for arginine at position 36;
`(b) substitution of an oxidation-resistant amino
`acid for tryptophan at position31;
`(c) substitution according to at least one of:
`Y for V at position 16;
`K for S at position 18;
`D for E at position 21;
`S for G at position 22;
`R for Q at position 23;
`R for A at position 24; and
`Q for K at position 26;
`(a) a substitution comprising at least one of:
`an alternative’small neutral amino acid for A
`at position 8;.
`an alternative acidic amino acid or neutral
`amino acid for E at’ position 9;
`an alternative neutral amino acid for G at
`position 10; and
`an alternative acidic amino acid for D at
`position 15; and
`(e) substitution of an alternative neutral amino
`acid or the D or N-acylated or alkylated form of
`histidine for histidine at position 7.
`(d) and
`(b),
`With respect to modifications (a),
`the substitutedamino acids may be in the D form, as
`(e),
`indicated by a superscript f, e.g., cl,
`The amino acids
`substituted at position 7 can also be inthe N-acylated
`or N-alkylated. forms.
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 7 of 50
`
`
`
`WO 91/11457
`
`.
`
`PCT/US91/00500
`
`-6-
`
`Thus, one aspect of the invention is directed to
`peptides having enhanced insulin stimulating properties
`analogous to the above-mentioned truncated forms of GLP-1
`
`(7-34)
`
`to GLP-1 (7-37), as described above.
`In another aspect,
`the invention is directed to
`peptides which show enhanced degradation resistance in
`
`plasma as compared to GLP-1 (7-37) wherein this enhanced
`resistance to degradation is defined as set forth below.
`In these analogs, any of the above-mentioned truncated
`
`to GLP-1 (7-37) or their C-terminal
`forms of GLP-1 (7-34)
`amidated forms is modified by
`(a) substitution of a D-neutral or D-acidic amino
`acid for H at position 7, or
`(b) substitution of a D-amino acid for A at
`position 8, or
`
`(c) both, or
`(d) substitution of an N-acylated or N-alkylated
`form of any naturally occurring amino acid for H at
`position 7.
`
`Thus, analogs of the invention which are resistant
`to degradation include (N-acyl
`(1-6C) aay? GLP=1 (7-37)
`and (N-alkyl
`(1-6C) AA)’ GLP-1 (7-37) wherein when AA is
`a lysyl residue, one or both nitrogens may be alkylated
`or acylated.
`AA symbolizes any amino acid consistent
`with retention of insulin stimulating activity.
`For substitutions of D-amino acids in the 7 and 8
`
`the D residue of any acidic or neutral amino
`positions,
`acid can be used at position 7 and of any amino acid at
`position 8, again consistent with insulin stimulating
`activity. Either or both of position 7 and 8 can be
`substituted by a D-amino acid; the D-amino acid at
`position 7 can also beacylated or alkylated as set forth
`above. These modified forms are applicable not only to
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 8 of 50
`
`
`
`To mee. Se
`
`WO 91/11457
`
`-
`
`PCT/US91/00500
`
`-J=-
`
`GLP-1 (7-37) but also the: shorter truncated analogs - as:
`set forth above.
`
`the invention is directed to
`In other aspects,
`pharmaceutical compositions containing one or more of
`these peptides as active ‘ingredients and to methods to
`treat Type II diabetes using these peptides or
`compositions thereof.
`
`Brief Description of the Drawings’
`Figure 1 schematically outlines the classification
`of amino acids as used herein.
`,
`
`Figure 2 gives a list of various compounds of the
`invention.
`;
`Figure 3 shows the results of ‘radiolabel sequencing
`analysis for degradation, 6f two analogs in plasma.
`Figure 4 shows the restlts of various GLP-1 (7-37)
`analogs with changes in the amino terminal region,
`to
`displace 1257_¢rp-1 (7-39). from amino terminal specific
`antiserum.
`me
`
`Modes of Carrying Out theInvention
`The analogs of the invention, which are modified
`forms of the GLP-1(7-34), (7-35),
`(7-36) or (7-37) are
`characterized by showing.greater potency than glucagon in
`
`an in vitro assay measuring insulin release from isolated
`rat islets in culture, or’by enhanced stability in plasma
`or both.
`
`Assays for Analogs with Enhanced Insulin Release
`Stimulating Properties =”
`One group of analogs of the invention is more
`potent than glucagon in stimulating: insulin release from
`islet cells.
`By being "more potent than glucagon in
`stimulating insulin release from. islet cells" is meant
`that the analog referred: to shows greater potency in an
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 9 of 50
`
`
`
`ee
`
`ct
`
`‘
`
`.“
`
`e
`
`WO 91/11457
`
`,
`
`PCT/US91/00500
`
`~B=—
`
`
`in vitro assay selected from the group consisting of the
`following: Rat islets for these assays are isolated by
`the method of Sutton, R. et al., Transplantation (1986)
`42:689-691,
`incorporated herein by reference. Briefly,
`Sprague-Dawley male rats are anesthetized and the lower
`end of the common bile duct is cannulated with a 2 FG
`cannula tied in place.
`The left and right hepatic ducts
`are then ligated separately above the region of the entry
`of pancreatic ducts into the biliary tree. The rats are
`killed by exsanguination and 3 mL Hank's solution
`containing 7.5 mM cacl., 20 mM HEPES buffér and 1-6 mg/mL
`Type I collagenase are run into the cannula to uniformly
`distend the pancreas.
`The pancreas is then excised and
`placed in a beaker on ice prior to incubation in Hank's
`solution containing 20 mM HEPES buffer at 37°C.
`After 13-25 min of incubation, the pancreas is
`removed and placed in Hank's solution containing 5 g/l
`bovine serum albumin and 20 mM HEPES buffer at 4°c. All
`
`of the pancreatic tissue is then gently syringed through
`a 14 FG needle, suspended in further Hank's solution
`containing HEPES as above, centrifuged at 50 g for 10 sec
`and the supernatant is discarded.
`Thetissue pellet is
`resuspended and again gently syringed, followed by
`another wash, after which the dispersed tissue is passed
`through a nylon mesh filter of 500 u pore size.
`The
`filtered tissue is centrifuged at 350 g for 5 sec,
`the
`supernatant discarded, and the tissue is then suspended
`in 25% Ficoll made up in Hank's with HEPES as above, on
`which was layered a discontinuous density gradient of
`23%, 20%, and 11% Ficoll solutions. This density
`gradient was spun at 750 g for 10 min at 4°C, and the .
`tissue obtained from the. upper two interfaces was washed
`three times in Hank's solution and viewed through a
`dissecting microscope for hand picking of islets.
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 10 of 50
`
`
`
`WO 91/11457
`
`se
`
`.
`
`PCT/US91/00500
`
`-9-
`
`In one approach the ability of the GLP-1 analog to
`potentiate secretion from these Islets is then determined
`according to the method of Schatz, H. et al.,
`in "Methods
`in Diabetes Research" (1984) Volume 1, Part C: pages
`291-307,
`incorporatedherein by reference.
`In this
`method, 5-10 islets per test tube are incubated in 1 mL
`Krebs-Ringer-bicarbonate buffer (KRB buffer) . For
`testing, glucagon or the modified analog of the invention
`is added at 5-10 wq/mL.
`The level of insulin released
`may be measured by the method of Jensen, S.L. et al., MJ
`Physiol
`(1978) .235:E381-E386,
`incorporated herein by
`reference.
`
`The following protocol is a preferred method to
`measure stimulation of insulin secretion. After
`collagenase digestion, the islets are allowed to recover
`overnight by incubation in DMEM (Dulbecco's Modified
`Eagle Medium 16 w/o glucose) , 2.8 mM glucose, 10% fetal
`bovine serum (FBS), at 37°C, 5% co,.
`The next day, islets to-be used for the experiment
`are transferred to DMEM, no glucose, 0.2% BSA (Armour,
`clinical grade, made at 5% stock) for a 60 min
`Islets
`preincubation in serum-free, glucose-free medium.
`are picked up by Eppendorf pipette and transferred to 60
`mm TC plates containing 8.0 mL medium and returned to the
`incubator for 60 min.
`Islets are counted during this
`transfer.
`(Note:
`each data point is 5 islets,
`experiments are usually performed in quadruplicate;
`therefore, 20 islets are used per data point.)
`Typically, recoveries are 150-200 islets per pancreas.
`Any suspect islets--too ragged or falling apart-~-are not
`used.
`
`the experiment is
`During the 60 min preincubation,
`set up, so that all that is needed at the end of the
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 11 of 50
`
`
`
`WO 91/11457
`
`,
`
`PCT/US91/00500
`
`-10-
`
`preincubation is to transfer islets in groups of 5 to.
`experimental conditions.
`The experiment is set up in 48
`well TC plates with 0.5 mL medium per well.
`To DMEM-0.2%
`BSA is added glucose to desired concentration (usually
`2.8 mM for hypoglycemic conditions, 5.6 mM glucose for
`euglycemic, or 16.7 mM glucose for hyperglycemic) and
`test compound at various dose ranges (typically, 1 pM to
`100 nM). Test compound is diluted from stock stored at
`-80°C and at ~0.3 mM serially into phosphate buffered
`saline (PBS) 0.2% BSA to prevent loss on sides of tubes.
`After medium plus test compound is mixed, 0.5 mL each is
`added to 4 wells for quadruplicate data points.
`After the preincubation period,
`5 islets are added
`per well.
`Islets are picked up by eppendorf pipette in
`25 ul volume.
`Incubation continues another 60 min, at
`which time 0.3 mL is harvested per well with care taken
`not to pick up islets. Wells are then rechecked for
`islet number. Medium is then assayed for insulin content
`using an insulin RIA.
`If medium is not immediately
`assayed, it is stored at -20°C until assay. Dose
`response curves for insulin secretion are plotted and
`ED. is calculated from the curves.
`Higher potency as compared to glucagon is defined
`as either higher levels of insulin released by the analog
`using the same concentrations of glucagon and analog or,
`alternatively,
`the same level of insulin release but
`using a lower concentration of analog than glucagon.
`While the foregoing assays form specific criteria
`for judging enhanced potency, alternative assays can also
`be used as substitutes for those set forth above.
`An additional test for potency of the compounds of
`the invention measures their ability to stimulate cAMP
`production in RIN 1046-38 cells.. This assay can be.
`conducted as follows:
`,
`
`16
`
`15
`
`20
`
`25
`
`30
`
`35
`
`z
`
`°
`
`‘
`
`¥
`
`
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 12 of 50
`
`
`
`
`
`WO 91/11457
`
`*:
`
`.
`
`PCT/US91/00500
`
`-ii-
`
`On day 1, 5 ¥ 105 pry 1046-38 cells (Drucker, D.J.,
`et al., Proc Natl Acad Sci USA (1987) 84:3434-3438) are
`seeded into individual wells of 6-well dishes with 2.5 mL
`M199 culture medium.
`On day4 ' cells are re-fed with
`fresh medium and on day’5 the assay is performed. At
`this time there are ~2.0-2.5 x 10° cells per well.
`Assays are only performedon cell passage $24.
`At time -60 min, monolayers are washed twice with
`2.5 mL PBS, and medium is changed to 1.0 mL of DMEM
`medium plus 4.5 g/l glucose and 0.1% BSA (assay medium).
`At 0 time, medium is aspirated and fresh assay medium,
`1.0 mL, containing test compound is added. Test compound
`is added in 50 ul volume of PBS plus 0.1% BSA; controls
`are added in vehicle alone.
`Incubation is continued for
`0 to 60 min.
`me
`At termination, conditioned medium and monolayer
`are harvested to measure both extra- and intracellular
`cAMP content.
`For extracellular measurement, medium is
`removed and centrifuged to remove any cellular debris.
`For intracellular determination, after medium removal,
`1.0 mL of ice cold 95% ethanol is. added to monolayer.
`Cells are collected by scraping, ‘lysed by two cycles of
`quick freeze/thawing using liquid No, and cell debris
`then removed by centrifugation. Aliquots (1/40th well
`content) of conditioned medium and ethanol cell extract
`are measured in duplicate for cAMP levels using an RIA
`kit by the acetylated protocol.
`As above, higher potency as compared to glucagon is
`defined either as higher cAMP stimulation by both the
`analog and glucagon at the same concentration, or the
`same cAMP stimulation by |the analog at a lower
`concentration.
`a
`.Still other assays for measurement of. enhanced
`potency to mediate insulin release can be used.
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 13 of 50
`
`
`
`WO 91/11457
`
`,
`
`PCT/US91/00500 .
`
`-12-
`
`' The ability of the compounds to potentiate the
`release of insulin can be tested both in vitro and in
`
`vivo.
`Insulin released can be detected using a standard
`antibody assay both in analyzing plasma in in vivo
`studies and in analyzing media or perfusion liquid in
`vitro.
`
`For example, a useful in vitro assay uses the
`pancreatic infusion assay method of Penhos, J.C., et al.
`Diabetes (1969) 18:733-738, as employed in the method of
`Weir, G.C., et al. J Clin Investigat (1974) 54:1403-1412.
`Insulin secretion can also be measured by the method
`described by Holst, J.7:, et al. FEBS Letters (1987)
`211:169-174 (supra). Also useful as an assay for
`insulinotropic effect is the measurement of stimulation
`of adenylate cyclase in the RIN 1046-38 cell line.
`Drucker, D.J. et al., Proc Natl Acad sci USA (1987)
`
`84:3434-3438 (supra).
`Inhibition of glucagon release can be shown as
`described by Orstov, C., et al. Endocrinol
`(1988)
`
`123:2009-2013; Suzuki, S., et al. Diabetes Research:
`
`Clinical Practice (1988) 5(Supp. 1):S30 (both supra).
`Assays for Enhanced Stability to Deqradation
`The therapeutic efficiency of the GLP-1 analogs of
`the invention can also be enhanced by providing analogs
`with increased half-lives in vivo.
`By "enhanced half-
`life in vivo" is meant a demonstrated ability to resist
`degradation in the presence of plasma according to an
`assay selected from the group consisting of the
`following.
`In all assays,
`the plasma is prepared by
`collecting blood into heparinized tubes, placing the
`tubes on ice and centrifuging at about 3,000 rpm for 10
`minutes in.a tabletop centrifuge. The.separated plasma
`. is stored at 4°C,
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 14 of 50
`
`
`
`WO 91/11457
`
`,
`
`PCT/US91/00500
`
`* =13-
`
`A. RadiolabelSequencing:
`The GLP analog is labeled, by radio-iodination in
`position 19 using standard radiolabeling methods. After
`exchange into RIA buffer (50.mM NaHPO, pH 7.4,. 0.25% BSA
`(Armour insulin and FFA free), 0.5% BME, 0.002%
`polylysine (Sigma 15,000 mw), 0.05% Tween 20, 0.1% NaN),
`the radioiodinated peptide (about 10° cpm/50 mL) and cold
`uniodinated peptide (20 41 100 nM) are added into 2 ml of:
`plasma to a final concentration of 1 nM and incubated in
`a circulating water bathfor presettimes. Total RIA
`buffer added to plasma never exceeds 5% of total volume.
`At the end of incubation, 10% bacitracin (w/v)
`in water
`is added to a final concentration of 0.1% to stop the
`reaction.
`.
`os
`.
`|
`The plasma is then extracted using C18 Sep-Pak to
`separate the analog and any fragments from the bulk of
`the plasma proteins.
`Sep-Pak cartridges (Waters) are
`washed with 2 mL of 1-propanol, followed by 2 mL of water
`and then equilibrated with 2 mL of 20% CH,CN containing
`0.1% trifluoroaceticacid.(TFA)
`(Buffer A).
`The bacitracin-treatedplasma is made 20% CH,CN
`with CH,CN containing 0.1% TFA and is, expressed slowly
`through a 3 mL plastic syringe through the cartridge.
`The cartridge is then washed with two 1 mL Buffer A
`washes and eluted with asingle 2 mL wash of 50% CH.,CN
`containing 0.1%: TFA (Buffer B)
`into“a siliconized 12 x 75
`
`glass tube. Recovery of the analog or fragments is more
`than 90%.
`:
`:
`:
`
`The eluates are concentrated to 100 ul in a Speed
`vac and transferred to a 1.5 mL Eppendorf tube to which a
`1 mL RIA buffer rinse of the original tube had been
`_ added.
`
`To purify any analog or its fragments when the
`analogs of GLP-1 (7-37) are used, the concentrates are
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 15 of 50
`
`
`
`WO 91/11457
`
`.
`
`.
`
`PCT/US91/00500
`
`-14-
`
`treated with 5 wl of antiserum prepared to a synthetic
`peptide corresponding to residues 24-37 which recognizes
`GLP-1, GLP-1 (7-37) but not GLP-1 (7-36). When the
`shorter forms of analogs are used, alternate carboxy
`terminal-specific antisera (prepared in the same manner
`but using a peptide corresponding to residues 24-34, 24-
`35 or 24-36 as immunogen) are used.
`To this is added 100
`Ml of a 10% (w/v) solution of protein A-Sepharose
`(Pharmacia)
`in PBS, and the mixture is incubated
`overnight at 4°C with gentle rocking.
`The Sepharose is
`then pelleted with a 5 second spin in an Eppendorf
`centrifuge at 4°C after which the pellet is washed two
`times with cold RIA buffer and four times with cold PBS.
`Polyclonal antisera were raised in New Zealand
`White rabbits against a synthetic peptide fragment
`corresponding to residues 24 to 37 of GLP-1 (7-37) using
`the method of Mosjoy, S. et al., J Biol Chem (1986)
`261:11880-11889.
`Initial immunizations were into the
`inguinal lymph nodes and used Freund's complete adjuvant.
`Two subcutaneous boosts were performed at 1 week
`‘intervals after the initial immunization and used
`Freund's incomplete adjuvant.
`For a single immunization
`or boost 100 ug peptide and 100 ug methylated BSA
`dissolved in 0.3 mL phosphate-buffered saline (PBS) were
`emulsified with 0.9 mL adjuvant. Bleeds (50 mL) began at
`week 6 after the initial immunization and continued at 1
`month intervals thereafter. Repeat boosts were performed
`as above when titers dropped noticeably from the level of
`the previous bleed.
`
`.
`
`Serum was prepared by allowing the blood to clot
`overnight at 4°C. The clot was pelleted by~
`centrifugation at 2000 g for 15 minutes and the serum
`removed.
`Serum is stored in aliquots at -20 or -80°Cc.
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 16 of 50
`
`
`
`Fl
`
`Re pee GTPTEN I
`ng
`gt
`ceiae
`ee
`=
`
`WO 91/11487
`
`mC
`
`PCT/US91/00500
`
`-15~
`
`10
`
`20
`
`25
`
`The peptides are then eluted from the antibody
`protein-A sepharose complex with three 100 wl washes of
`Buffer B.
`The combined 300 ul of wash are then applied
`directly to an ABI model 477A sequencer used according to
`the manufacturer's instructions. Fractions from each
`cycle are then: diverted for counting. Counting can be
`effected in 4 mL aqueous:‘scintillant (ACS, Amersham).
`The cycle at which label appears indicates the
`extent of degradation from the N-terminus.
`If no
`degradation from the N-terminus has occurred in the GLP-
`1 (7-37) analog, all of the label will appear in the 13th
`cycle, corresponding to’ the tyrosine at position 19; if
`degradation has occurred, the label will appear in
`earlier cycles.
`»
`B.AssaybyRP-HPLc: ngs
`
`15
`While the foregoing’ method is'a clear criterion for
`exhibiting a longer half-life in plasma, alternative
`forms of the assay forthis property can also be used.
`In one convenient assay, the analog can be assessed for
`degradation into fragments using: reverse phase-HPLC,
`since the fragments have different retention times from
`the analog per se.
`In this assay,
`the analog is added to
`Plasma for various times and recovered similarly to the
`method described above: for. radiolabel: sequencing
`analysis. Specifically, the analog at a concentration of
`100 nM in RIA buffer is spiked into 1 mL plasma to a
`final concentration of 1 nM andafter incubation in 37°C
`circulating water bath for various preset times,
`the
`reaction is stopped Py bringing the plasma to 0.1% (w/v)
`The peptides are ean puesby Sep-Pak
`extraction as’ described above. Theeluates are
`concentrated to about 1 mL on a Spéed-vac, diluted with 1
`mL distilled water, frozen at 80°C and lyophilized
`
`30
`
`in bacitracin.
`
`|
`
`eo
`
`7
`
`35
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 17 of 50
`
`
`
`WO 91/11457
`
`.
`
`,
`
`PCT/US91/00500
`
`-16-"
`
`The powder is resuspended in 0.5 nb Buffer c
`overnight.
`(0.1% TFA in water) per mL starting plasma and 0.25 mL
`
`are injected on a Hewlett-Packard 1090L liquid
`chromatograph using an Alltech C18 column (0.45 x 25 cm;
`10 pm particle size) with a Brownlee 2 cm C18 guard
`column.
`The extraction is monitored at OD5,4 throughout
`the run and the solvent flow rate was 1 mL/minute.
`A
`gradient between Buffer c and Buffer D (0.1% TFA in
`acetonitrile) is set up over a 40 minute run time. The
`gradient starts at 35% D is held for the first 2 minutes
`after injection and then increased to 42% D over 24
`minutes.
`The gradient is then increased to 60% D over
`the next two minutes, held at this level for 2 minutes
`and returned to 35% D over the next 2 minutes.
`The 4D
`
`10
`
`15
`
`remains at 35% for the remaining 8 minutes of the run.
`
`Fractions are collected at 0.5 minute intervals for the
`
`first 30 minutes of each run and dried in a Speed-vac.
`The samples can be assayed for the presence of analog or
`fragment using RIA (measuring competition with labeled
`GLP-1 (7-37) for binding to C-terminal specific
`
`antiserum) or by any conventional or convenient
`
`alternative method.
`
`Radioimmunoassays for the amino or carboxyl
`terminus: of GLP-1 (7-37) use a single antibody
`¢ 125,-GLp-1 (7-37)
`displacement format. Binding o
`antibody is incrementally displaced by increasing
`concentrations of unlabeled peptide insolution.
`Antibody bound iodinated peptide is separated from free
`iodinated peptide in solution by precipitation of the
`antibody-peptide complex with Pansorbin™ (Boheringer
`Mannheim).
`The resulting pellet is then counted on a
`gamma counter.
`o
`-
`
`to
`
`20
`
`25
`
`30
`
`35
`
` PFIZER, INC. v. NOVO NORDISK A/S - IPR2020-01252, Ex. 1041, p. 18 of 50
`
`
`
`WO 91/11457
`
`.
`
`PCT/US91/00500
`
`_-17-
`
`|
`
`“C. Loss of Binding. to N-Terminal Specific
`Antibodies:
`oe
`'
`,
`A third approach ‘to assessment of half-life in
`plasma utilizes polyclonal ‘or monoclonal antibodies
`specifically prepared tothe N-terminus which will fail
`to bind degraded analog. These aritisera were raised
`against a synthetic peptide correspérding to GLP-1 (7-
`22) which contains an additional cysteine residue at the
`carboxyl terminus and is spetifically coupled to KLH via
`the cysteine using mal+sac-HSNA as described by Aldwin,
`L. et al. Analytical Biochem (1987) 164:494-501.
`Polyclonal antibodies were generated in New Zealand white
`rabbits by giving