Vol. 90, pp. 10464-10468, November 1993
`
`Proc. Natl. Acad. Sci. USA
`
`
`Medical Sciences
`
`Uroguanylin: Structure and activity of a second endogenous peptide
`
`
`
`
`
`
`
`that stimulates intestinal guanylate cyclase
`(guanylin/urlne/ cyclic GMP / chloride secretion)
`
`F. KENT HAMRA*t, LEONARD R. FORTE*t*, SAMMY L. EBER*t, NYKOLAI V. PIDHORODECKYJ*t,
`
`
`
`
`
`WILLIAM J. KRAusE§, RoNALD H. FREEMAN11, DAVID T. CHINII, JAY A. ToMPKINsll, KAM F. FoK**,
`
`
`
`
`CHRISTINE E. SMITH**, KEVIN L. DUFFIN**, NED R. SIEGEL**, AND MARK G. CURRIE**
`
`*The Truman Veterans Affairs Medical Center and Departments of tPharmacology, fAnatomy, fPhysiology, and ii Biochemistry, School of Medicine, Missouri
`University, Columbia, MO 65212; and **Monsanto Corporate Research, St. Louis, MO 63167
`
`
`
`Communicated by James 0. Davis, August 11, 1993
`
`
`
`(12). Guanylin/ST-like bioactivity was also found in extracts
`
`
`
`ABSTRACT The intestinal hormone guanylin and bacte•
`
`
`
`
`
`of rat kidney (12). Molecular cloning of intestinal cDNAs
`
`rial beat-stable enterotoxins (STs) are members of a peptide
`
`
`
`from rats, humans, and mice revealed that the purified form
`family that activates intestinal membrane guanylate cyclase.
`
`
`
`
`of guanylin was a 15-residue carboxyl-terminal peptide of a
`
`Two different peptides that activate the human intestinal T84
`
`
`
`115-residue precursor polypeptide (13-16). Guanylin shares
`
`cell guanylate cyclase have been purified from urine and
`
`
`either 7 or 8 amino acid residues with the STs secreted by
`intestinal mucosa of opossums (Didelphis virginiana). The
`
`
`
`diarrhea. different strains of E. coli that cause secretory
`
`
`blgbJy acidic peptide, QEOCELCINVACTGC, was named
`
`
`
`
`Moreover, synthetic preparations of guanylin or ST-(5-17)
`uroguanylin because it was isolated from urine and shares 53%
`
`
`
`stimulated c1secretion and cGMP accumulation responses
`
`
`identity with guanylin. A second peptide, SHTCEICAF AA­
`
`
`
`
`in human T84 intestinal cells. Gnanylin also competed with
`
`CAGC, was purified from urine and intestinal mucosa. This
`
`
`
`ST (1251-ST) for high-affinity binding sites on T84
`1251-labeled
`
`
`alanine-rich peptide was 47% identical to uroguanylin and
`
`
`cells (12, 17). It was suggested that guanylin may be an
`
`
`73 % identical to human guanylin, suggesting that it may be an
`
`
`
`
`
`endogenous ST-like agonist that serves as a paracrine hor­
`
`
`opossum homologue of guanylin. Synthetic uroguanylin-(2-15)
`
`
`
`mone for regulation of intestinal salt and water transport (12).
`
`(i.e., EOCELCINV ACTGC) was IO-fold more potent than
`
`
`
`the phosphoryla­Guanylin may function in vivo to regulate
`
`
`synthetic rat guanylin, but both peptides were less potent than
`
`tion state of the CFTR in c1-secreting cells by cGMP­
`Escherichia coli ST in the T84 cell cGMP bioassay. Urogua­
`
`
`
`
`
`mediated control of intestinal protein kinases (18).
`
`
`nylin-(2-15) and guanylin inhibited
`125 1-ST binding to T84 cell
`
`
`
`
`Because receptors for ST are abundant in the kidney of the
`
`
`receptors in competitive radioligand binding assays. Transepi­
`
`
`
`opossum (19, 20) and guanylin-like bioactivity was observed
`
`thelial a-secretion was stimulated by 1 µM uroguanylin,
`
`
`
`in extracts of rat kidney (12), it was suggested that guanylin
`
`indicated by an increase in the short circuit current of T84 cells.
`
`
`may also be a renal hormone. The receptors for ST occur on
`
`Thus, uroguanylin is another paracrine hormone in the emerg­
`
`
`
`
`
`the apical membranes of proximal tubule cells in the opossum
`
`
`
`ing peptide family that activates intestinal membrane guanylate
`
`
`
`
`(20-22), thus an endogenous ST-like agonist could be se­
`
`cyclase. The second peptide may be the opossum form of
`
`
`
`
`creted from renal cells into the proximal tubule filtrate and
`
`
`guanylin, or perhaps it is still another member of this peptide
`
`
`
`appear in urine. In support of this hypothesis, two distinctly
`
`family. The presence of uroguanylin and guanylin in urine and
`
`
`
`
`
`different peptides were purified from opossum urine, and
`
`
`receptors in proximal tubules suggests that these peptides may
`
`
`
`both peptides were found to activate the guanylate cyclase of
`also originate from renal tissue and may regulate kidney
`
`
`
`human intestinal T84 cells. One of these peptides may be the
`function.
`
`
`
`opossum homologue of guanylin, since it has a
`putative
`
`
`
`
`similar amino acid sequence. The second, a highly acidic
`Secretory diarrhea caused by intestinal microorganisms in
`
`
`
`
`
`
`
`
`peptide, is 53% identical to guanylin and 67% identical to E.
`
`
`
`humans and domestic animals is a major public health prob­
`
`
`
`because it was coli ST. This peptide was named uroguanylin
`
`lem (1). Enteric bacteria, including
`Escherichia coli, Yersinia
`
`
`
`
`isolated from opossum urine. Preliminary elements of this
`
`enterocolitica, and Vibrio cholerae, cause diarrhea by se­
`
`
`work have been reported (23).
`
`
`
`creting small heat-stable enterotoxins (STs) that bind to and
`
`
`
`
`activate an intestinal isoform of membrane guanylate cyclase
`
`
`
`
`(2-4). Guanosine 3',5'-cyclic monophosphate (cGMP) is the
`MATERIALS AND METHODS
`
`
`
`
`second messenger molecule that mediates the parallel stim­
`
`
`
`
`Puriilcation of Guanylin Peptides. Urinary and intestinal
`
`ulation of c1secretion and inhibition of Na+ absorption
`
`
`
`
`
`peptides were purified separately. Urine was collected daily
`
`
`
`elicited by STs (5, 6). The stimulation of intestinal c1
`
`from opossums (Didelphis virginiana) housed in metabolism
`
`
`
`secretion occurs via cGMP-regulated phosphorylation of the
`
`
`cages and was stored at -20"C. After thawing, the urine was
`
`
`
`cystic fibrosis transmembrane conductance regulator
`
`centrifuged at 10,000 x g for 20 min. The supernatant was
`
`
`
`
`(CFfR) protein located in the apical membrane of crypt cells
`
`
`made 0.1% in trifluoroacetic acid (TFA) before further use.
`
`
`
`
`(7, 8). Cystic fibrosis patients have no ion transport response
`
`
`
`Full-length intestines were removed from six adult opos­
`
`to E. coli ST (9). Refractoriness to ST occurs because the
`
`sums. The mucosa was scraped from intestinal muscle by
`
`
`
`
`mutant CFfR either functions improperly or is not localized
`
`
`using a glass microscope slide, suspended in 10 vol of 1 M
`
`
`
`
`to apical membranes where it serves as a c1channel (10, 11).
`
`and acetic acid, heated for 10 min at 100°C, homogenized
`
`
`
`
`
`
`Guanylin is a recently discovered endogenous ST-like
`
`
`
`peptide hormone that was originally isolated from rat jejunum
`Abbreviations: ST, heat-stable enterotoxin; CFfR, cystic fbrosis
`
`
`
`
`
`
`
`
`transmembrane conductance regulator; TFA, trifluoroacetic acid;
`
`RP-HPLC, reverse-phase HPLC.
`Toe publication costs of th.is article were defrayed in part by page charge
`payment. Th.is article must therefore be hereby marked "advertisement"
`
`
`
`
`+To whom reprint requests should be addressed at: Department of
`
`
`
`
`Phannacology, University of Missouri, Columbia, MO 65212.
`in accordanc e with 18 U.S.C. §1734 solely to indicate this fact.
`
`10464
`
`Bausch Health Ireland Exhibit 2012, Page 1 of 5
`Mylan v. Bausch Health Ireland - IPR2022-00722
`
`

`

`
`
`
`
`Medical Sciences: Hamra et al.
`
`Proc. Natl. Acad. Sci. USA 90 (1993) 10465
`
`was thawed and centri­stored at -20°C. The homogenate
`
`
`with 1 ml of ice-cold medium. The cells were then solubilized
`
`
`
`fuged at 10,000 x g for 20 min, and the supernatant was made
`
`with 1 M NaOH for measurement of radioactivity.
`
`
`0.1% in TFA. These supematants were processed with
`
`
`Rat guanylin (PNTCEICA Y AACTGC)
`
`Peptide Synthesis.
`
`
`
`Waters Sep-Pak cartridges [octadecylsilane (ODS) car­
`
`
`were prepared and E.coli ST-(5-17) (CCELCCNPACAGC)
`
`tridges; C18] that had been washed with 20 ml of 100%
`
`
`
`as previously described by the solid-phase method with an
`
`
`acetonitrile/0.1% TFA followed by 20 ml of 0.1% TFA in
`
`
`
`
`Applied Biosystems 430A peptide synthesizer on Cys(4-
`H2O before
`
`
`
`50 ml of extract was applied to the cartridge. The
`CH3Bzl)-OCH
`
`
`
`
`to coupling cycles 2-Pam resin, using double
`cartridge was then washed with 10 ml of 0.1% TFA, and
`
`
`
`ensure complete coupling at each step (12). Peptides were
`
`
`
`bioactive peptides were eluted with 8 ml of 40% (vol/vol)
`
`
`
`
`cyclized by using dimethyl sulfoxide as described by Tam et
`
`
`
`acetonitrile/0.1% TFA in H2O. The eluted fractions were
`
`
`
`by electrospray al. (24). Peptide structures were verified
`
`
`
`
`mass spectrometry, gas-phase sequence analysis, and amino
`
`
`lyophilized, resuspended in 16 ml of 50 mM ammonium
`at 500 x g for 10 min,
`
`acetate, pH 5.0, and then centrifuged
`
`acid composition analysis.
`
`
`and 7.5 ml was applied to a 2.5 cm x 90 cm Sephadex G-25
`Synthetic uroguanylin-(2-15), EDCELCINV ACTGC, was
`
`
`
`
`
`synthesized by the solid-phase method with an Applied
`
`
`
`
`column. Ten-milliliter fractions were collected, and those
`
`
`with activity in the T84 cell cGMP bioassay were combined,
`
`
`
`Biosystems 431A peptide synthesizer. Fmoc (9-fluorenyl­
`
`
`lyophilized, and resuspended in 50 ml of 0.1% TFA and
`
`
`methoxycarbonyl)-protected amino acids activated with
`
`
`
`applied to a C18 cartridge. A step gradient with 6 ml each of
`
`2-(lH-benzotriazol-1-yl)-l, 1,3 ,3-tetramethyluronium
`5%, 10%, 15%, 20%, 25%, 30%, 35%, and 60% acetonitrile,
`
`hexafluorophosphate were added to Fmoc-Cys(trityl)-Wang
`in H2O containing 0.1 % TF A, was used to elute the
`
`
`
`
`resin (Nova Biochem). Coupling efficiencies were monitored
`all vol/vol
`
`
`
`
`
`peptides. The active fractions eluted primarily at 20% and
`
`
`by the UV absorbance of the released Fmoc groups. The
`
`
`
`peptide was cleaved from the resin and the side chains were
`
`
`25% acetonitrile/0.1 % TF A, and these fractions were pooled
`
`
`
`
`and lyophilized. Intestinal extracts were resuspended in 50 ml
`
`
`deprotected, except for the acetamidomethyl groups on
`
`of distilled H2O containing
`
`0.8% ampholytes, pH range 3-10
`
`
`
`
`Cys-3 and Cys-11, by incubation in TFA, ethane dithiol, and
`
`
`
`(Bio-Rad), and applied to a preparative isoelectric focusing
`
`
`
`water (95:2.5:2.5, vol/vol) for 2 hr at room temperature. The
`
`
`
`cell (Rotofor; Bio-Rad). Each extract was focused by using 12
`
`
`peptide was cyclized by using air oxidation. The acetami­
`
`
`W of constant power for 60 min, and fractions were collected
`
`
`
`domethyl groups on Cys-3 and Cys-11 were removed with
`
`
`as described by the manufacturer. Two peaks of bioactivity
`
`
`
`iodine. The peptide was desalted with a 12-ml Whatman
`
`
`
`ODS-3 solid-phase extraction device and purified to a single
`
`
`were observed, at pl ""' 3.0 and pl ""' 5.2. Urinary extracts
`
`
`
`were not purified by isoelectric focusing because the extracts
`
`
`peak by C18 RP-HPLC (acetonitrile/ammonium acetate). The
`
`
`
`were highly acidic and disrupted the pH gradient.
`
`
`
`
`sequence of uroguanylin was confirmed by protein sequenc­
`
`
`
`Final purification of the active peptides was accomplished
`
`
`
`
`ing on an Applied Biosystems 470A gas-phase protein se­
`
`
`
`by a series of four reverse-phase (RP)-HPLC steps as pre­
`quencer.
`
`
`viously described (12).
`DMEM and Ham's F12 media and Sephadex
`Materials.
`Cell Culture. T84 cells (passage 21 obtained from Jim
`
`
`G-25 gel were obtained
`
`
`
`McRoberts, Harbor-University of California, Los Angeles,
`from DuPont/NEN. Fetal bovine serum was obtained from
`
`
`
`
`
`
`
`Medical Center, Torrance, CA) were cultured in Dulbecco's
`
`
`GIBCO. Other reagent-grade chemicals were purchased from
`
`modified Eagle's medium (DMEM) and Ham's F12 medium
`
`Sigma or Fisher Scientific.
`
`
`
`(1:1) containing serum, and 60 µ,g of penicillin 5% fetal bovine
`
`
`
`and 100 µ,g of streptomycin per ml as previously described (17,
`RESULTS AND DISCUSSION
`
`
`
`
`
`18). Permeable membranes (Falcon cell culture inserts of
`
`
`Cyclopore membranes, 25 mm diameter, 0.45 µm pore size,
`Isolation of ST-Like Peptides from Urine. Three liters of
`
`
`
`
`Fisher Scientific) were first coated with 0.25 ml of collagen
`
`
`urine was used as starting material for the first purification
`
`(bovine type I, 1.3 mg/ml; Sigma) for 16 hr, while the filters
`
`
`
`
`experiments. When the bioactive fractions were subjected to
`
`
`
`were being sterilized by UV irradiation. T84 cells were seeded
`
`the RP-HPLC step using a semipreparative C18 column
`
`
`
`
`at 2.5-3.0 x 106 cells per filter as previously described (17, 18).
`
`
`
`(acetonitrile/TFA), a single broad peak ofbioactivity eluted
`
`from the C18 column at 21-22% acetonitrile (data not shown).
`Measurement of Short Circuit Current in T84 Monolayers.
`
`
`
`Further separation of this fraction on RP-HPLC using an
`
`
`
`
`T84 cells raised on permeable fllters were mounted in a
`
`
`
`custom-made Ussing chamber for measurement of c1se­
`
`analytical-scale C18 column (acetonitrile/ammonium
`acetate)
`
`
`
`cretion as previously described (17, 18).
`
`
`
`
`resolved two distinct bioactive components (Fig. 1). Each of
`Assay of cGMP in T84 Cells. T84 cells were cultured in
`
`
`
`
`
`these components was subjected separately to further puri­
`
`
`
`
`
`24-well plastic dishes, and the cGMP levels were measured in
`
`
`fication by RP-HPLC using a C18 analytical column (aceto­
`
`
`
`
`control or agonist-stimulated cells by radioimmunoassay (17,
`
`
`
`nitrile/TFA) and a C8 microbore (Vydac) column (acetoni­
`
`
`18). In brief, confluent monolayers of T84 cells in DMEM
`
`
`
`trile/TFA). The two bioactive fractions from the microbore
`
`
`
`containing 20 mM N-(2-hydroxyethyl)piperazine-N'-(2-
`
`
`
`RP-HPLC were subjected to gas-phase amino-terminal se­
`
`ethanesulfonic acid) at pH 7 .4 and 1 mM isobutylmethylxan­
`
`
`quence analysis. The component contained in peak 2 (see
`
`
`
`or vehicle for 40 min. thine were treated at 37°C with agonists
`
`
`
`
`Fig. 1) yielded a 13-residue peptide with the sequence DCEL­
`
`
`
`Then cells were treated with 200 µ.I of 30% perchloric acid to
`
`
`
`CINV ACTGC. Electrospray mass spectrometry revealed [M
`
`
`
`extract cGMP. The pH of the extract was adjusted to 7 .0, the
`+ H]+ at m/z 1339.5,
`
`which was consistent with the calcu­
`
`
`
`extract was centrifuged, and the supernatant was used to
`
`
`
`
`lated monoisotopic molecular weight of this peptide (1338.5).
`
`measure cGMP.
`
`
`
`Peak 1 contained two peptides, 14 and 15 residues in length.
`
`Binding Experiments. Iodination of synthetic
`
`
`
`The structure of the 15-residue peptide was QEDCELCIN­
`Radioligand
`
`
`Multiple E.coli ST-(1-19) (NSSNYCCELCCNPACTGCY;
`
`
`
`V ACTGC and the 14-residue form of this peptide lacked the
`
`
`
`Peptide Systems, San Diego) was accomplished by using
`
`
`
`glutamine residue at the amino terminus. The mass spectrum
`of 1251-ST to receptors
`
`lactoperoxidase, and the product was purified as previously
`
`
`
`and of this fraction contained [M + H]+ signals at m/z 1488.4
`
`
`described (17). The binding
`on T84
`
`
`1596.6, which was consistent with the calculated monoiso­
`cells was measured using 50,000 cpm of 1251-ST per well of
`
`
`
`
`
`
`topic molecular weights (1487.4 and 1595.6) of the 14-and
`
`
`
`
`15-residue peptides predicted from the sequence analysis.
`
`
`
`T84 cells cultured in 24-well dishes. The medium was 0.2 ml
`
`ofDMEM containing 15 mM 2-(N-morpholino)ethanesulfon­
`
`
`
`
`Either glutamine or lysine at the amino terminus of the
`ic acid at pH 5.5. After incubation for 60 min at 37°C, the
`
`
`
`
`15-residue peptide would be consistent with the data from
`
`medium was aspirated and the cells were washed two times
`
`
`
`mass spectrometry. The peptide that was isolated from
`
`from Sigma and Na1251 was purchased
`
`Bausch Health Ireland Exhibit 2012, Page 2 of 5
`Mylan v. Bausch Health Ireland - IPR2022-00722
`
`

`

`
`
`
`
`10466 Medical Sciences: Hamra et al.
`
`Proc. Natl. Acad. Sci. USA 90 (1993)
`
`10�------------
`
`Peak 1
`
`----�
`GMP
`Acetonitrile
`19
`
`-c
`
`15
`
`14
`
`QJ
`
`13 ?J2.
`-....
`b 12•2
`0 ....
`1 1 �
`<
`10
`
`Peak 3
`
`100
`-
`
`cGMP
`Acetonitrile
`
`80
`QJ
`=
`�
`'""
`� 60
`p..
`�
`C., 40
`I.I
`
`-0 e 20
`
`C.
`
`oi-��������-�::!!!�!___L9
`40 50
`80
`90
`60 70
`Fraction
`
`0-1-- �--�--�---.-::-�-��-�
`-1. 7
`110 125
`35 50
`65 80
`95
`Fraction
`
`FIG. 1. Purification of uroguanylin peptides from urine by RP­
`
`
`
`FIG. 3. Purification of the opossum homologue of guanylin from
`
`
`
`
`
`
`HPLC. Urine extracts were purified through the semipreparative
`
`
`
`
`
`urine. An 8-liter batch of urine was purified through the third step and
`
`
`
`RP-HPLC step (no. 4) and the active fractions were combined, dried,
`on a 7.8 mm x 30 cm C1s column
`
`
`the active fractions were separated
`onto a 3.9 mm x 30 crri C1s
`
`
`and dissolved in 5% acetonitrile to inject
`
`
`
`
`
`
`(semipreparative scale, µBondapak; Waters). A linear gradient of
`
`
`
`
`column (µBondapak; Waters). The elution was isocratic for 5 min
`
`
`
`10 30% acetonitrile containing 0.1 % TFA was developed at 3 ml/min
`
`
`
`before a linear gradient of 5 25% acetonitrile in 10 mM ammonium
`
`
`
`over 3 hr. A total of 180 fractions were collected and assayed for
`
`
`
`acetate, pH 6.2, was used to elute peptides at a rate of 1 ml/min over
`
`
`
`activity in the T84 cell cGMP bioassay. The fractions in the second
`
`
`
`a 3-hr period, and 185 fractions were coliected. This figure shows
`
`
`
`peak, which was eluted at 22% acetonitrile/0.1% TFA, were com­
`
`
`
`only the region of the chromatogram where biological activity, as
`scale C18 column
`
`bined and chromatographed on ail analytical
`
`
`
`assessed in the T84 cell cGMP accumulation bioassay, was observed.
`
`
`
`
`exactly as described in the legend to Fig. 1 and the bioactivity of
`
`
`
`
`eluted fractions plotted here was assessed in the T84 cell cGMP
`
`
`opossum urine was designated as uroguanylin (Fig. 2).
`bioassay.
`to E. coli ST and 53% identical
`
`Uroguanylin is 67% identical
`an opossum homologue of guanylin. It was 79% identical to
`
`
`
`
`
`
`
`
`to rat guanylin (underlined amino acids). Conserved regions
`
`
`rat or human guanylin but only 50% identical to uroguanylin.
`
`
`
`
`between uroguanylin and ST or guanylin include the car­
`
`
`Thus, opossum urine contains at least two different ST-like
`
`
`
`boxyl-terminal ACTGC motif, the relative positions of the
`
`
`
`
`peptides that activate T84 human intestinal guanylate cy­
`
`
`
`
`four cysteine residues, and the glutamate residue following
`clase.
`
`
`
`the first cysteine. Most of the ST peptides, however, have six
`Isolation of ST-Like Peptides from Intestine. Bioactive
`
`
`
`
`
`
`
`
`
`cysteine residues and a uniformly conserved proline residue,
`
`
`
`
`peptides were then purified from opossum intestinal mucosa.
`
`
`
`
`whereas the guanylin peptides have four cysteines and an
`
`
`
`
`In these experiments, preparative isoelectric focusing was
`
`
`
`
`
`alanine residue instead of proline (2-4, 12-14). The possibil­
`
`added after the third step. Interestingly, two peaks of bio­
`
`
`
`ity exists that uroguanylin may have been derived from
`
`
`
`activity were resolved by this method, with pl = 3.0 and pl
`
`
`bacteria, which could have contaminated the urine samples.
`
`
`
`= 5.2 (Fig. 4). The pl= 5.2 fraction of active peptides was
`
`
`
`We believe this to be unlikely in view of the distinct structural
`
`
`purified to homogeneity by using the RP-HPLC methods
`
`
`
`
`differences found in uroguanylin compared with bacterial ST
`
`
`utilized above. Upon sequencing, the pl = 5.2 fraction
`
`
`
`
`peptides. The presence ofreceptor/guanylate cyclases in the
`
`
`
`yielded the 15-residue peptide SHTCEICAFAACAGC,
`
`
`
`brush border membranes of opossum kidney cortex also
`
`
`which is the same as the guanylin-like peptide above that was
`
`
`
`
`argues for the existence of an endogenous ligand, such as
`
`
`
`purified from urine except that serine occurs at the amino
`
`
`
`
`uroguanylin, to activate those receptors (19-22). However,
`
`
`
`the proof that uroguanylin is derived from opossum tissue
`
`
`
`terminus (Fig. 2). The pl= 3.0 fraction isolated from intestine
`
`
`will require isolation of the cDNA for this peptide.
`
`
`
`may be uroguanylin. This fraction has not been purified to
`
`
`
`When 8 liters of urine was used as a starting batch, a second
`
`
`homogeneity and subjected to sequence analysis. Thus,
`
`
`
`
`bioactive fraction was isolated by semipreparative RP­
`
`
`
`
`intestinal mucosa contains a guanylin-like peptide and addi­
`
`
`
`HPLC. This fraction, which was eluted following the main
`
`
`
`tional peptides with pl = 3.0 that stimulate the intestinal
`
`
`peak that had yielded the uroguanylin peptides described
`guanylate cyclase.
`
`
`
`above (data not shown), was subjected to further purification
`Since the final purification of uroguanylin required C8
`
`
`
`by RP-HPLC using a C18 analytical
`column (acetonitrile/
`
`
`
`microbore RP-HPLC to provide homogenous peaks of UV­
`
`
`
`
`ammonium acetate). Three separate fractions were resolved
`
`
`
`
`absorbing peptides and all of the purified peptides were used
`
`
`
`as shown by their activity in the T84 cell cGMP bioassay (Fig.
`
`
`
`
`in the analyses, a peptide was synthesized corresponding to
`
`
`3). Peak 3 was subsequently purified and the amino acid
`
`
`
`the linear sequence EDCELCINV ACTGC [uroguanylin-(2-
`
`
`
`sequence was determined. Surprisingly, this alanine-rich
`
`15)] to test its potency and efficacy when T84 cells are used
`
`
`
`
`peptide was distinctly different from uroguanylin. This 14-
`
`
`
`
`as a model bioassay. Sequential oxidation methods were used
`
`amino acid peptide, HTCEICAFAACAGC, appeared to be
`
`
`
`to provide disulfide bonds from Cys-3 to Cys-11 and Cys-6 to
`
`Uroguanylin
`D £
`Q E
`li 1' £ I N V A £ I .Q £
`°"
`C £
`li 1' £ C N p A £ I .Q £ y
`E.coli ST
`N s s N y
`li ! £ A F A A £ A .Q £
`s H
`I £
`li ! £ A y A A £ T .Q £
`p N
`I £
`
`
`
`
`
`
`
`
`
`of E. coli ST and rat guanylin were taken from refs. 2 and 12, respectively.
`
`FIG. 2. Comparison of the primary structures of opossum uroguanylin and guanylin to E. coli ST and rat guanylin. The primary structures
`
`"Putative" Guanylin
`
`..,
`,..,
`Guanylin
`
`Bausch Health Ireland Exhibit 2012, Page 3 of 5
`Mylan v. Bausch Health Ireland - IPR2022-00722
`
`

`

`Proc. Natl. Acad. Sci. USA 90 (1993) 10467
`
`t;.
`
`� 100 cl
`._
`e ST-(5-17)
`0
`� 75
`& Guanylin
`-o· C
`;:;
`0 50
`.0
`E--�
`§" 25
`
`121110 9 8 7
`6
`
`Peptide, -log10M
`5
`
`of 1251-ST binding
`1251-ST for binding
`
`
`
`
`
`Medical Sciences: Hamra et al.
`
`�10
`
`8
`
`6 ::c:
`
`Q.,
`
`4
`
`cGMP
`pH
`
`----0--
`
`sr--
`
`10,000.-- --------------,
`
`e ST-(5-17)
`•------ ----t;.
`& Guanylin
` / h/ &
`• 1/
`t;. Uroguanylin
`1//
`&
`./
`�/
`-��A-----
`
`1,000
`" �
`100
`::8
`0
`u
`
`10
`
`.... ., c.. �
`0 E
`
`•
`
`10
`
`1
`
`
`
`Peptide, -log10M
`
`35 45
`Time, min
`55
`
`65
`
`FIG. 5. Stimulation of cGMP accumulation in T84 cells by
`
`
`E. coli ST-(5-17). Synthetic preparations
`
`FIG. 7. Stimulation of short circuit current in T84 cells by
`
`
`uroguanylin, guanylin, and
`
`
`
`and E. coli ST-(5-17). The peptides were added at 1 µ.M
`uroguanylin
`
`
`
`were of opossum uroguanylin, rat guanylin, and E. coli ST-(5-17)
`
`
`
`tested for their potency and efficacy in activation of cGMP accu­
`
`
`
`to the apical reservoir at the time designated by the arrow. Data are
`
`
`
`
`
`
`
`
`mulation in T84 cells. This is a representative experiment that has the mean of three experiments with uroguanylin and two experiments
`
`been repeated with the same results.
`with ST.
`
`FIG. 6. Inhibition
`to receptors on T84 cells by
`
`
`
`
`and E. coli ST-(5-17). Synthetic preparations
`
`uroguanylin, guanylin,
`
`
`
`
`of these peptides were tested for their affinity as competitors with
`1 3 5 7 9 ll U B V H
`
`
`
`of to receptors. bound in the absence Bo, amount
`Fraction
`
`competitor. These data are the mean of two experiments.
`
`
`
`FIG. 4. Separation of guanylin/ST-like peptides by preparative
`
`
`
`
`
`isoelectric focusing. The intestinal mucosa of six opossums was used
`of these bioactive peptides may be made by the kidney as
`
`
`
`
`
`
`
`
`to purify active peptides through step 3. The active peptides were
`
`
`
`well, since both were isolated from urine. The acidic peptide
`
`
`
`fractionated by preparative isoelectric focusing and each fraction
`
`
`
`was named uroguanylin because it was originally isolated
`
`was assayed by using the T84 cell cGMP bioassay.
`
`
`
`
`from urine and is similar in primary structure to guanylin. It
`
`
`
`Cys-14. After purification of the active polypeptide to a single
`
`
`
`is also apparent that uroguanylin and the putative opossum
`
`
`
`
`peak by RP-HPLC, the bioactivity ofuroguanylin-(2-15) was
`
`
`
`
`homologue of guanylin are different gene products. The
`
`
`and rat guanylin. compared with that of E. coli ST-(5-17)
`
`
`
`
`discovery of uroguanylin suggests that further diversity may
`
`
`
`Uroguanylin-(2-15) stimulated cGMP accumulation in T84
`
`
`
`
`exist in peptide ligands as well as in their receptors. It is likely
`
`
`
`
`
`cells (Fig. 5). Its potency appeared to be about 10-fold greater
`
`
`
`that, as with atrial peptides and their receptors (25), at least
`
`
`than that of guanylin, and the concentration-response curves
`
`
`
`
`two different selective receptors exist for uroguanylin and
`
`
`
`
`were parallel, indicating that uroguanylin-(2-15) was a full
`guanylin.
`
`
`
`more potent than was about 10-fold agonist. E. coli ST-(5-17)
`The alanine-rich peptide (pl= 5.2) that was isolated from
`
`
`
`
`uroguanylin-(2-15) and 100-fold more potent than guanylin.
`
`
`
`urine and intestine shares 11 amino acids (i.e., 73% identity)
`
`
`
`
`Both uroguanylin-(2-15) and guanylin inhibited the binding of
`
`
`with rat or human guanylin peptides (12-16). Other differ­
`
`125 1-ST to receptor sites on T84 cells (Fig. 6). Uroguanylin­
`
`
`ences between these putative forms of guanylin represent
`
`
`
`(2-15) had an =10-fold higher affinity for these binding sites
`
`
`
`
`conservative changes, since the pl 5.2 opossum peptide is
`
`
`93% similar to rat or human guanylin. Thus the bioactive
`
`
`
`than did guanylin, but ST-(5-17) had a higher affinity than
`
`
`
`
`
`domains of guanylin hormones appear to be highly conserved
`
`
`
`
`either of the endogenous ligands. When T84 cells were raised
`
`
`
`
`on semipermeable ftlters and mounted in modified Ussing
`
`
`
`
`between these species. Another peptide could exist that is
`
`
`
`
`chambers, it was observed that uroguanylin stimulated short
`
`
`
`more closely related to guanylin than is the pl= 5.2 peptide,
`
`
`
`
`circuit current, lsc, which in these cells is proportional to the
`
`
`
`which has been tentatively classified as the opossum form of
`
`
`rate oftransepithelial c1secretion (Fig. 7). The lsc response
`
`
`
`
`guanylin. Uroguanylin contains two additional acidic amino
`
`
`acids and does not contain the aromatic amino acids that
`to 1 µ,M uroguanylin-(2-15) was less than that found with 1
`
`
`
`
`occur in all guanylin peptides examined thus far (12-16).
`
`
`ofurogua­µ,M ST-(5-17), consistent with the lower potency
`
`
`
`
`nylin compared with ST-(5-17) observed in both the cGMP
`
`
`
`These differences in structure between uroguanylin and
`
`
`
`guanylin suggest that different receptors may exist, which
`and radioligand-binding bioassays.
`
`
`
`
`have selectivity for uroguanylin versus guanylin. The recep-
`
`
`
`
`It appears that intestinal guanylate cyclase may be regu­
`
`
`
`lated by at least two different peptide hormones, which could
`
`
`
`be made in the intestine and secreted luminally. One or both
`
`
`
`30 - E. coli ST-(5-17)
`
`Uroguanylin
`
`--0-
`
`Bausch Health Ireland Exhibit 2012, Page 4 of 5
`Mylan v. Bausch Health Ireland - IPR2022-00722
`
`

`

`
`
`
`
`10468 Medical Sciences: Hamra et al.
`
`Proc. Natl. Acad. Sci. USA 90 (1993)
`
`tor/ guanylate cyclase in T84 cells appeared to be selective for
`
`
`
`
`
`1. Field, M., Rao, M. C. & Chang, E. B. (1989)
`N. Engl. J. Med.
`321, 800 806 and 879 883.
`
`uroguanylin, because this peptide was about 10-fold more
`2. Aimoto, S., Takao, T., Shimonishi, Y., Hara, S., Takeda, T.,
`
`
`
`
`
`
`potent than guanylin in stimulating cGMP accumulation or in
`Eur. J. Biochem. 129,
`Takeda, Y. & Miwatani, T. (1982)
`inhibiting
`
`
`
`1251-ST binding. Moreover, rat kidney has guany­
`257263.
`
`
`
`lin-like bioactivity (12) and an mRNA that hybridizes to
`3. Takao, T., Tominaga, N. & Shimonishi, Y. (1984)
`
`
`Biochem.
`
`
`
`guanylin cDNA (15). These findings and the occurrence of
`Biophys. Res. Commun. 125, 845 851.
`4. Takao, T., Shimonishi, Y . , Kobayashi, M., Nishimura, 0.,
`
`
`
`
`
`
`guanylin and uroguanylin in opossum urine imply that one or
`
`
`Arita, M., Takeda, T., Honda, T. & Miwatani, T. (1985)
`FEBS
`
`
`
`both of these peptides are made by the kidney and secreted
`Lett. 193, 250 254.
`
`
`
`into the filtrate, where they appear in urine. Thus, urogua­
`
`
`P. L. (1978) 5. Field, M., Graf, L. H., Laird, W. J. & Smith,
`
`
`nylin and/or guanylin may be renal hormones that could
`Proc. Natl. Acad. Sci. USA 15, 2800 2804.
`
`
`
`
`regulate kidney function through activation of receptor/
`
`6. Hughes, J.M., Murad, F., Chang, B. & Guerrant, R. L. (1978)
`Nature (London) 271, 755 756.
`
`
`
`
`
`guanylate cyclases localized on the apical membranes of
`7. Crawford, I., Maloney, P. C., Seitlin, P. L., Guggino, W. B.,
`
`
`
`
`tubular cells (19-22).
`
`Hyda, S. C . , Turley, H., Gatter, K. C., Harris, A. & Higgins,
`
`
`
`In summary, the opossum has two distinctly different
`
`C. F. (1991) Proc. Natl. Acad. Sci. USA 88, 92629266.
`
`
`
`
`peptides, uroguanylin and a peptide that is similar to guanyl­
`8. Denning, G. M., Ostedgaard, L. S., Cheng, S. H., Smith,
`
`
`
`in. A reasonable assumption is that these closely related
`
`
`A. E. & Welsh, M. J. (1991) J. Clin. Invest.
`89, 339 349.
`
`
`
`9. Baxter, P. S., Goldhill, J., Hardcastle, J., Hardcastle, P. T. &
`
`
`
`
`peptides play a role in the regulation of intestinal and/or renal
`335, 211.
`Taylor, C. J. (1988)
`
`
`
`
`electrolyte transport. The determination of the cellular dis­
`Nature (London)
`
`
`10. Anderson, M. P., Sheppard, D. N., Berger, H. A. & Welsh,
`
`
`
`
`tribution of these two peptides should prove exciting, with
`263, Ll L14.
`M. J. (1992) Am. J. Physiol.
`
`
`the possibility that in some instances they share sites of
`11. Denning, G. M., Anderson, M. P., Amara, J. F., Marshall, J.,
`
`
`
`
`production and in other instances the pattern of their distri­
`
`Smith, A. E. & Welsh, M. J. (1992) Nature (London) 358,
`761 764.
`
`
`
`bution may be independent. It will be particularly interesting
`12. Currie, M. G., Fok, K. F., Kato, J., Moore, R. J., Hamra,
`
`
`to examine their distribution in the different regions of the
`
`F. K., Duffin, K. L. & Smith, C. E. (1992)
`Proc. Natl. Acad.
`
`
`
`intestine and along the crypt-villus axis. Similarly, the dis­
`Sci. USA 89, 947 951.
`
`
`
`tribution of the two peptides along the length of the nephron
`13. Wiegand, R. C., Kato, J. & Currie, M. G. (1992) Biochem.
`
`
`
`
`
`should yield valuable information important to understanding
`Biophys. Res. Commun. 185, 812 817.
`14. Wiegand, R. C., Kato, J., Huang, M. D., Fok, K. F., Kachur,
`
`
`
`
`their function. Furthermore, it appears that pathogenic en­
`
`J. F. & Currie, M. G. (1992) FEBS Lett. 311, 150 154.
`
`
`
`
`
`teric bacteria cause secretory diarrhea by releasing a molec­
`
`
`15. Schulz, S., Chrisman, T. D. & Garbers, D. L. (1992) J. Biol.
`
`
`
`ular mimic of guanylin and uroguanylin which activates a
`Chem. 267, 16019-16021.
`
`
`common set of receptor/guanylate cyclases. The biochemi­
`16. De Sauvage, F. J., Keshav, S., Kuang, W.-J., Gillett, N.,
`
`
`
`
`
`
`cal result of this molecular mimicry is an activation of
`
`Henze, W. & Goedde), D. V. (1992) Proc. Natl. Acad. Sci.
`
`
`
`
`guanylate cyclase, followed by an increase in intracellular
`USA 89, 9089 9093.
`17. Forte, L. R., Eber, S. L., Turner, J. T., Freeman, R. H., Fok,
`
`
`
`
`cGMP, which serves as a second messenger that regulates the
`
`K. F. & Currie, M. G. (1993)
`91, 2423 2428.
`J. Clin. Invest.
`
`
`
`phosphorylation of CFTR molecules, thus activating c1-
`18. Forte, L. R., Thorne, P. K., Eber, S. L., Krause, W. J.,
`
`
`
`secretion. Evidence has been presented that cGMP may
`
`
`Freeman, R. H., Francis, S. & Corbin, J. D. (1992) Am. J.
`
`
`
`activate a cAMP-dependent protein kinase as one possible
`263, C607-C615.
`Physiol.
`
`
`
`
`mechanism for regulation of this transport pathway (18). The
`
`19. Forte, L. R., Krause, W. J. & Freeman, R.H. (1988)
`Am. J.
`
`
`
`
`discovery of a second member of the guanylin peptide family,
`
`Physiol. 255, F1040 Fl046.
`
`termed uroguanylin, indicates that this paracrine/endocrine
`
`20. Forte, L. R., Krause, W. J. & Freeman, R. H. (1989)
`Am. J.
`Physiol. 257, F874 F881.
`
`
`
`system may have a significant role in the control of epithelial
`21. White, A. A., Krause, W. J., Turner, J. T. & Forte, L. R.
`
`
`
`
`cell function in the kidney as well as the intestine and opens
`(1989)
`
`Biochem. Biophys. Res. Commun. 159, 363 367.
`
`
`the field for further studies of this peptide family in other
`22. Krause, W. J., Freeman, R.H. & Forte, L. R. (1990) Cell
`epithelia.
`Tissue Res. 260, 387 394.
`23. Hamra, F. K., Eber, S. L., Krause, W. J., Freeman, R. H.,
`We express our appreciation to Judy Richey for typing this
`
`
`
`Smith, C. E., Duffin, K. L., Siegel, N. R., Currie, M. G. &
`
`
`manuscript and to Matthew Buss and Helena Hillman for their
`Forte, L. R. (1993) FASEB J. 7, A215 (abstr.).
`
`
`
`assistance. This