`0019-9567/93/114710-06$02.00/0
`Copyright© 1993, American Society for Microbiology
`
`Vol. 61, No. 11
`
`The Escherichia coli Heat-Stable Enterotoxin Is a Long-Lived
`Superagonist of Guanylin
`BRUCE W. CARPICK AND JEAN GARIEPY*
`Department of Medical Biophysics, University of Toronto, and The Ontario Cancer Institute,
`500 Sherbourne Street, Toronto, Ontario, Canada M4X JK9
`
`Received 7 July 1993/Retumed for modification 18 August 1993/Accepted 31 August 1993
`
`The mechanism by which bacterial heat-stable enterotoxins (ST I, ST A) cause diarrhea in humans and
`animals has been linked to the activation of an intestinal membrane-bound guanylate cyclase. Guanylin, a
`recently discovered rat intestinal peptide, is homologous in structure to ST I and can activate guanylate cyclase
`present on the human colonic carcinoma cell line T84. To directly test the mechanistic association of guanylate
`cyclase activation with diarrhea, we synthesized guanylin and a guanylin analog termed N9P10 guanylin and
`compared their biological activities with those of a synthetic ST I analog, termed ST lb(6-18). We report that
`guanylin is able to inhibit the binding of a radiolabeled ST I analog to rat intestinal cells but causes diarrhea
`in infant mice only at doses at least 4 orders of magnitude higher than that of ST lb(6-18). In contrast, N9p10
`guanylin was enterotoxic in mice at much lower doses than guanylin but proved to be a weaker inhibitor of
`radiolabeled ST I than guanylin in the receptor binding assay. The pattern of guanylate cyclase activation
`obse"ed for ST lb(6-18) and the two guanylin analogs parallels the results obse"ed in the receptor binding
`assay rather than those obse"ed in the diarrheal assay. Treatment of guanylin with chymotrypsin or lumenal
`fluid derived from newborn mouse intestines resulted in a rapid loss of binding activity. Together, these results
`suggest that ST I enterotoxins may represent a class of long-lived superagonists of guanylin.
`
`The heat-stable enterotoxins are a group of small homol(cid:173)
`ogous peptides elaborated by enterotoxigenic strains of
`bacteria (23, 29, 35). They are collectively responsible for a
`large proportion of worldwide cases of secretory diarrhea in
`humans and animals. These enterotoxins, abbreviated ST I
`(or STA), are known to bind to receptors located on the
`brush border surface of intestinal cells (13, 15, 17, 20, 22) and
`to cause an elevation of intracellular cyclic GMP (cGMP)
`levels (11, 16, 19, 26). It is generally believed that the binding
`of the enterotoxin to its receptor is coupled to the activation
`of a guanylate cyclase and that cGMP acts as the intracellu(cid:173)
`lar second messenger causing the eventual onset of diarrhea.
`Although more than one class of ST I receptors may exist
`(20, 24), research efforts in this field have focused on the
`cloning of one class of guanylate cyclases acting as ST I
`receptors (7, 8, 31, 33). The evidence linking the ST I-in(cid:173)
`duced elevation of cGMP levels to secretory diarrhea has
`centered around the administration of a nonhydrolyzable
`analog of cGMP termed 8-bromo-cGMP to ligated rabbit
`intestines and to infant mice (16, 19). This treatment resulted
`in fluid accumulation in both animal models and suggested
`that an increased production of intracellular cGMP is a
`necessary step leading to watery diarrhea. Recently, a
`naturally occurring peptide termed guanylin was isolated
`from rat jejunum and was found to activate a particulate
`form of guanylate cyclase present on the human colonic
`carcinoma cell line T84 (6). This peptide was able to displace
`the binding of 1251-labeled ST I to receptors on the surface of
`T84 cells (6). Guanylin is a 15-amino-acid peptide that is
`highly homologous in sequence to a region of ST I, abbre(cid:173)
`viated ST lb(6-18), that codes for its receptor binding and
`enterotoxigenic properties (4, 14, 32, 39). In particular,
`identical residues are found at eight positions within the
`13-amino-acid sequence of ST lb(6-18) (Fig. 1). A major
`
`• Corresponding author.
`
`difference between the two peptides is that ST Ib(6-18) has
`six cysteine residues participating in three intramolecular
`disulfide bridges within its sequence (14, 32) while guanylin
`has only four cysteines and two disulfide bridges (6) (Fig. 1).
`As a consequence of structural and functional similarities
`between guanylin and ST I, one would expect guanylin to
`cause diarrhea in mammals at a concentration relative to ST
`I that parallels its ability to inhibit the binding of radiolabeled
`ST I to intestinal cells. In this study, we report the synthesis
`of two analogs of guanylin and their ability to (i) inhibit the
`binding to rat cells of a radiolabeled ST I analog (4, 14), (ii)
`cause a diarrheal response in infant mice (18), and (iii)
`stimulate rat intestinal brush border guanylate cyclase (22).
`
`MATERIALS AND MEfflODS
`
`Preparation of analogs. Peptides were synthesized on an
`Applied Biosystems Model 430A automated peptide synthe(cid:173)
`sizer using terl-butoxycarbonyl-protected amino acids cou(cid:173)
`pled to a phenylacetamidomethyl resin support by classical
`solid phase methods (34). The peptides were cleaved from
`the resin by using anhydrous HF in the presence of anisole,
`dimethyl sulfide, and p-thiocresol. The reduced peptides
`were dissolved in a dilute aqueous solution (50 µM), pH 8.5,
`for 5 days to allow disulfide pairing to occur. The oxidized
`peptides were concentrated on a preparative C18 column,
`eluted with 40% (vol/vol) acetonitrile in water, and Iyophi(cid:173)
`Iized. The peptides were then purified by high-performance
`liquid chromatography (HPLC) on a semipreparative C18
`column with a linear gradient going from 10 to 40% (vol/vol)
`acetonitrile-0.08% (vol/vol) trifluoroacetic acid in water-
`0.1 % (vol/vol) trifluoroacetic acid over a 60-min period.
`HPLC peaks were initially screened for their ability to
`inhibit the binding of a radiolabeled ST lb analog, 125I-Y4ST
`Ib(4-18), to rat villus cells (4). Only those peaks which
`exhibited inhibitory activity in this experiment were further
`analyzed. The homogeneity, amino acid composition, and
`
`4710
`
`Bausch Health Ireland Exhibit 2060, Page 1 of 6
`Mylan v. Bausch Health Ireland - IPR2022-00722
`
`
`
`VOL. 61, 1993
`
`BIOLOGICAL ACTIVITY OF GUANYLIN ANALOGS
`
`4711
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9 10 11 12 13 14 15
`
`16 17 18 19
`
`ST lb Asn Ser Ser Asn Tyr Cys Cys Glu Leu Cys Cys Asn Pro Ala Cys Thr Gly Cys Tyr
`
`ST1b(6-18)
`Guanylln
`
`N1P10guanylln
`Pro Asn Thr ys Glu
`lie Cys Ala Asn Pro Ala Cys Thr Gly Cys
`FIG. 1. Amino acid sequences of the E. coli heat-stable entero(cid:173)
`toxin ST lb (1, 25) and the three peptides prepared for this study: ST
`lb(6-18), guanylin, and N9P10 guanylin. The numbering refers to the
`position of the amino acids in relation to the ST lb sequence. ST
`lb(6-18) corresponds to the biologically active domain of ST lb (4,
`14, 32, 39). Regions of sequence homology are boxed. The substi(cid:173)
`tution Leu-9 ---+ Ile represents a conversed substitution observed in
`other ST I sequences (35). Amino acids are identified by their
`three-letter codes. All peptides were synthesized as described in
`Materials and Methods.
`
`molecular weight of each analog were respectively con(cid:173)
`firmed by thin-layer chromatography, amino acid analysis,
`and fast atom bombardment mass spectrometry.
`Displacement assay of 125I-Y4ST lb(4-18) binding to rat
`villus cells. Y4ST Ib(4-18), a peptide consisting of residues 6
`to 18 of ST lb, in addition to two N-terminal tyrosine
`residues, was radioiodinated and purified as described pre(cid:173)
`viously (4, 14). Displacement assays were performed as
`follows. Twenty-microliter aliquots of unlabeled test peptide
`dilutions were combined with 1()5 cpm of 125I-Y4ST lb(4-18),
`20 µ.l of phosphate-buffered saline (PBS (0.15 M NaCl, 10
`mM NaH2PO4 , pH 7.41) containing 5 mM EDTA and 0.02%
`(wt/vol) sodium azide, and 5 x Hf' villus cells isolated from
`the small intestines of adult female Sprague-Dawley rats in
`polypropylene test tubes. Samples were incubated at 37°C
`for 30 min, treated with 50 µ.l of 1 % (wt/vol) bovine serum
`albumin in PBS, and placed on ice for 10 min. The samples
`were filtered onto Whatman GF/A filters, washed four times
`with PBS, and counted in a 'Y counter. Displacement curves
`were constructed from averaged values of experiments per(cid:173)
`formed in triplicate.
`Suckling mouse assay. One hundred microliters of each
`analog dilution prepared in PBS containing 0.04% (wt/vol)
`Evans blue dye was administered orally to 4-day-old Swiss(cid:173)
`Webster suckling mice (18). After 3 h, the mice were
`sacrificed and their intestines were excised. The guts and
`remaining carcasses were weighed, and the resulting gut/
`carcass ratio (G:C) was calculated. Each G:C represents an
`average of measurements performed for three infant mice.
`Two regimens were performed in order to evaluate the
`ability of guanylin to inhibit the enterotoxic effect of ST
`Ib(6-18) in infant mice. In the first regimen, a 100-µ.l aliquot
`of ST lb(6-18) (10- 11 mol (see Fig. 3)) sufficient to cause
`diarrhea in mice was mixed with increasing amounts of
`guanylin (10-9 to 10-11 mol) prepared in PBS containing
`0.04% (wt/vol) Evans blue dye. The resulting solutions were
`administered to suckling mice according to the protocol
`descnbed previously. In the second regimen, 50-µ.l doses
`representing increasing amounts of guanylin were adminis(cid:173)
`tered to the mice, followed 30 min later by a 50-µ.l dose of ST
`lb(6-18) (10-11 mol). After 3 h, the G:C was determined as
`described above.
`Guanylate cyclase assay. Aliquots (10 µ.l) of each peptide
`dilution were combined on ice with 42 µ.l of 50 mM Tris-HCI
`(pH 7.6), 8 µ.l of rat intestinal brush border preparation
`(protein content, 28 µ.g) (22), and 20 µ.l of a mixture contain(cid:173)
`ing 2.5 mM 3-isobutyl-1-methylxanthine, 38 mM phospho(cid:173)
`creatine, and 3.2 U of phosphocreatine kinase. The reaction
`was initiated by the addition of 20 µ.l of a substrate solution
`
`consisting of 5 mM GTP and 20 mM MgC12• The samples
`were incubated at 37°C for 5 min, and the reaction was
`quenched by the addition of 400 µ.l of cold sodium acetate
`buffer (50 mM, pH 4.0), followed by immersion of the sample
`in a boiling water bath for 3 min (22). The samples were
`centrifuged at 4°C for 15 min, and the supernatants were
`immediately removed and stored on ice. cGMP levels in
`100-µ.l aliquots of the supernatants were determined with a
`commercial radioimmunoassay kit (NEN Du Pont, Missis(cid:173)
`sauga, Ontario, Canada).
`Protease digestion experiments. Chymotrypsin was pur(cid:173)
`chased from Sigma (St. Louis, Mo.), and its activity was
`calibrated with the substrate benzoyl-L-tyrosine ethyl ester
`(Sigma) (38). Infant mouse lumenal fluid was recovered by
`excising the small intestines of euthanized newborn suckling
`mice and flushing their contents with cold PBS. The resulting
`solution was filtered, and the filtrate was immediately stored
`at - 700C. The remaining flushed small intestine segments
`were also collected and stored at - 700C without processing.
`A homogenate of infant mouse intestinal tissue was prepared
`from the intestinal segments as follows. The frozen tissue
`was further cooled with liquid nitrogen and pulverized with
`a precooled mortar and pestle. The pulverized tissue from
`one intestine was suspended in 400 µ.l of 0.1 M Tris-HCI (pH
`8.1) and vortexed. The suspension was then spun at 12,000 x
`g in a microcentrifuge for 5 min. The supernatant was
`removed, filtered, and stored on ice until used. Both intes(cid:173)
`tinal preparations were found to contain significant chymo(cid:173)
`trypsin-like activity by benzoyl-L-tyrosine ethyl ester assay.
`Samples of each of the three peptides were prepared in Tris
`buffer at the following concentrations: ST Ib(6-18), 0.02
`mg/ml; guanylin, 0.5 mg/ml; N9P10 guanylin, 1 mg/ml. At
`zero time, the peptide solutions were treated with either
`chymotrypsin, lumenal fluid, or filtrate of intestinal homoge(cid:173)
`nate to an enzyme/substrate equivalence ratio of 1:50 (on the
`basis of the benzoyl-L-tyrosine ethyl ester assay) and were
`incubated at 37°C. Aliquots were withdrawn from each
`peptide solution prior to the addition of either protease or
`intestinal extracts. The digests were then sampled after 15,
`60, and 180 min of incubation. Each aliquot was immersed in
`a boiling water bath for 3 min and treated with the inhibitor
`phenylmethylsulfonyl fluoride (Sigma) to a final concentra(cid:173)
`tion of 1 mM phenylmethylsulfonyl fluoride. Twenty-micro(cid:173)
`liter aliquots of each sample, corresponding to the minimal
`peptide dose able to completely inhibit the binding of 1()5
`cpm of 125I-Y4ST Ib(4-18) to 5 x lOS rat villus cells, were
`then tested with the displacement assay described above.
`
`RESULTS AND DISCUSSION
`Structure of guanylin analogs. The rat intestinal peptide
`guanylin is structurally homologous to heat-stable enterotox(cid:173)
`ins (ST I), sharing identical residues at eight positions within
`the 13-amino-acid-long enterotoxic domain (ST lb(6-18)]
`(Fig. 1). More importantly, four of the six cysteines of ST I
`are conserved in guanylin, suggesting that the intestinal
`peptide may possess functional properties associated with
`these enterotoxins (6). Guanylin was initially isolated on the
`basis of its ability to stimulate guanylate cyclase activity in
`T84 cells and to bind to receptors on the surface of these
`cells, two properties which it shares with ST I (6). Guanylin
`represents the C-terminal region of a larger inactive precur(cid:173)
`sor (a 94-amino-acid-long proguanylin form (9)). Bacterial
`strains may thus produce ST I enterotoxins in order to
`exploit a mechanistic pathway associated with the naturally
`occurring intestinal peptide guanylin. Because diarrhea is
`
`Bausch Health Ireland Exhibit 2060, Page 2 of 6
`Mylan v. Bausch Health Ireland - IPR2022-00722
`
`
`
`4712
`
`CARPICK AND GARIEPY
`
`INFECT. IMMUN.
`
`TABLE 1. Biological properties of ST lb(6-18), guanylin, and
`N9P10 guanylin
`
`Peptide
`
`ST lb(6-18)
`Guanvlin
`N9Pld' guanyJin
`
`ICso (M)°
`3.0 X 10-s
`10-6
`6.0 X 10-s
`
`ED50 (mot)"
`2.5 X 10-lZ
`>3.0 X 10-s
`3.2 X 10-10
`
`GCso(Mf
`10-s
`7.0 X 10-7
`6.3 X 10-6
`
`0 lCso, 50% inhtbitory concentration (peptide concentration required to
`inhibit 50% of the specific binding of radiolabeled Y"ST lb(4-18) to rat villus
`cells).
`b EDso, 50% effective dose (peptide dose required to cause a half-maximal
`increase in G:.C after oral administration of the analog to infant mice).
`• GCso, 50% guanylate cyclase activation (peptide concentration required to
`cause a half-maximal activation of intestinal brush-border guanylate cyclase).
`
`not a normal physiological event, it was projected that the
`oral administration of guanylin may not cause a diarrheal
`response in experimental animals. We thus synthesized
`guanylin and a second guanylin analog, termed N9P10 gua(cid:173)
`nylin, which is a closer homolog of ST I (Fig. 1), particularly
`within the central region of the ST I molecule (residues 11 to
`14). This tum region was previously found to be the most
`important region of ST lb in terms of biological activity and
`is particularly sensitive to amino acid substitutions (4, 27).
`The objective of this study was to quantitatively compare the
`biological activities of guanylin and ST Ib(6-18), using estab(cid:173)
`lished in vitro and in vivo assays in order to define how
`differences in the primary structure of guanylin and ST I may
`result in enterotoxicity.
`The two guanylin analogs bind to the most abundant ST I
`receptor on rat enterocytes. The binding of guanylin analogs
`to the ST I receptor on rat small intestinal villus cells was
`assessed by measuring their ability to compete with radiola(cid:173)
`beled ST I for receptor sites on enterocytes (4, 14). Binding
`experiments were performed at pH 7 .4 in accordance with
`the slightly alkaline environment observed in the small
`intestine of both rat and suckling mice (measured pH of
`lumenal fluid, 7.5 ± 0.5 pH units [this study]). As shown in
`Fig. 1, synthetic guanylin displaced, in a concentration(cid:173)
`dependent manner, the binding of 125I-Y4ST Ib(4-18) to
`receptors on rat intestinal cells with a 50% inhibitory con(cid:173)
`centration 30-fold higher than that of ST Ib(6-18) (Table 1).
`Thus, guanylin is an ~onist of the ST I receptor on rat villus
`cells. The peptide N9P10 guanylin, on the other hand, was
`approximately 2,000-fold less active than ST lb(6-18) in the
`binding assay (Fig. 2 and Table 1).
`The enterotoxicity of guanylin in suckling mice is low. It is
`widely believed that binding of ST I to intestinal villus cells
`and the subsequent elevation of intracellular cGMP levels
`are necessary and sufficient events leading to the eventual
`macroscopic episode of diarrhea (11, 16, 19, 26). The capac(cid:173)
`ity of ST I analogs to cause intestinal fluid accumulation in
`infant mice (18) correlated well with the ability of these
`peptides to bind to rat villus cells (4, 14). Guanylin was
`unable to cause diarrhea in mice except at concentrations 4
`orders of magnitude higher than that of ST lb(6-18) (Fig. 3
`and Table 1). In contrast, N9P10 guanylin was only 125-fold
`less active than ST lb(6-18) in this experiment, despite the
`fact that it was 2,000-fold less active than ST lb(6-18) in the
`binding assay (Fig. 2 and 3 and Table 1). If a single class of
`ST I receptors existed that was mechanistically linked to the
`onset of diarrhea, one would expect guanylin to act as an
`antagonist and block diarrhea induced by ST I. To test this
`hypothesis, doses of guanylin between 10-9 and 10-11 mot
`were administered orally to infant mice in conjunction with a
`dose of ST lb(6-18) (10- 11 mot) sufficient to cause diarrhea.
`
`30000
`
`QC~
`
`s
`"Cl-= e
`j~
`-~ 20000
`":4 CJ ""--"
`.... -E-t r.l
`.c =
`... fl.l :§ 10000
`~e
`...
`
`~
`
`0
`-12
`10
`
`-10
`10
`
`-8
`10
`
`-6
`10
`
`-4
`10
`
`-2
`10
`
`[peptide] (M)
`
`FIG. 2. Displacement curves of the binding of 1251-Y'ST lb(4-18i
`to rat villus cells by ST Ib(6-18) (0), guanylin (□), and N9P1
`guanylin (b.). The relative affinity of each analog for the ST I
`receptor was determined by its concentration-dependent capacity to
`inhibit the binding of 125I-Y4ST lb(4-18) to rat enterocytes. Each
`datum point represents the average of experiments performed in
`triplicate.
`
`Guanylin was unable to act as an antagonist, because diar(cid:173)
`rhea was observed in all suckling mice even at a molar ratio
`of g11anylin to ST lb(6-18) of llf [10-9 mol of guanylin to
`10-11 mot of ST lb(6-18)]. The results were identical whether
`guanylin was administered simultaneously to ST lb(6-18) or
`30 min prior to ST lb(6-18) (data not shown). These results
`suggest that guanylin lacks the enterotoxic potential of ST I
`but binds predominantly to the most prevalent class of ST I
`receptors which are coupled to a guanylate cyclase.
`Both guanylin analogs are able to activate a rat intestinal
`guanylate cyclase. Both guanylin analogs were assayed for
`their ability to activate rat intestinal brush border guanylate
`
`0 . 1 6 - - - - - - - - - - - - - - ,
`
`0.12
`
`0.08
`
`0.04
`
`~ e
`
`Ill
`
`~ a I i C,
`
`0.00 J----.....-........ -.--...----.--,-...... -1
`-10
`-6
`-8
`-12
`-14
`-16
`10
`10
`10
`10
`10
`10
`
`Oral peptide dose (mol)
`FIG. 3. Enterotoxicity of ST lb(6-18) (0), guanylin (□), and
`N9P10 guanylin (b.) in suckling mice. The capacity of each analog to
`cause a diarrheal response in infant mice is quantified according to
`the increase in G:C resulting from an oral dose of a peptide (18).
`Each datum point represents the average of experiments performed
`in triplicate.
`
`Bausch Health Ireland Exhibit 2060, Page 3 of 6
`Mylan v. Bausch Health Ireland - IPR2022-00722
`
`
`
`VOL. 61, 1993
`
`BIOLOGICAL ACTIVITY OF GUANYLIN ANALOGS
`
`4713
`
`14
`
`12
`
`8
`
`=
`....
`I 10
`c:i.
`~ .@
`~
`~ 6
`~ 4
`2
`
`-~ !
`
`0
`
`10•14 10-12 10·10 10·8
`
`10-6
`
`10-4 102
`
`[peptide] (M)
`
`FIG. 4. Concentration-dependent activation of rat brush border
`membrane guanylate cyclase by ST lb(6-18) (0), guanylin (□), and
`N9P10 guanylin (D.). Each datum point represents the average
`quantity of cGMP produced per milligram of protein in experiments
`performed in duplicate.
`
`cyclase in relation to ST lb(6-18). A comparison of Fig. 2 and
`4 clearly shows that the pattern of activation of guanylate
`cyclase for the three peptides mirrors their pattern of binding
`affinities for the ST I receptor. Guanylin is a 30-fold weaker
`ligand than ST Ib(6-18) in the receptor binding assay and is
`70-fold less active in activating membrane-bound guanylate
`cyclase. In contrast, N9P10 guanylin is 2,000-fold weaker
`than ST lb(6-18) in the binding assay and 630-fold less active
`in the guanylate cyclase assay (Table 1). The binding of ST
`I to its receptor has been correlated with the elevation of
`intestinal cGMP levels (10). Recent studies have indicated
`that the toxin binding site and the cyclase domain reside
`within the same intestinal membrane protein (7, 31, 33). If
`the activation of a specific guanylate cyclase was linked to
`diarrhea, one would have expected the relative pattern of
`potency for the concentration-dependent activation of gua(cid:173)
`nylate cyclase observed for all three analogs to parallel their
`dose dependency in causing diarrhea in mice, with the
`relative order of decreasing potency being ST Ib(6-18) >
`N9P10 guanylin > guanylin.
`Structural and mechanistic implications. Two models can
`be proposed that reconcile past and present findings in
`relation to the mechanism of action of ST I. The first model
`would view ST I as a long-lived superagonist of guanylin and
`that processing or clearance of ST I from the guanylin
`receptor (previously referred to as the ST I receptor) is not
`as efficient as in the case of guanylin causing a chronic
`activation of the guanylate cyclase. The rapid proteolysis of
`guanylin would represent the most probable event. The
`tyrosine-alanine segment present in the sequence of guanylin
`is the only site absent in ST I enterotoxins which represents
`a segment potentially susceptible to protease digestion (pu(cid:173)
`tative chymotryptic site [Fig. 1 ]). Figure SA shows that
`guanylin is rapidly inactivated by chymotrypsin compared
`with N9P10 guanylin and ST lb(6-18). Similar results in terms
`of loss of receptor binding activity were observed when
`guanylin was exposed to lumenal fluid recovered for small
`intestines of suckling mice (Fig. SB) or treated with filtrates
`of infant mouse intestinal tissues (Fig. SC). The replacement
`of Tyr-Ala with Asn-Pro in N9P10 guanylin restored both
`properties of enterotoxicity and protease resistance. The
`
`100
`
`80 A
`
`60
`
`40
`
`20
`
`,e,
`"i
`
`Cl.
`
`0
`
`100
`
`80
`
`60
`
`40
`
`0
`
`100
`
`€1 ..
`:8
`I ..
`t
`~
`.!! = 20
`.a
`ii, s
`C f!
`&!
`
`Control
`
`15
`
`60
`
`180
`
`B
`
`Control
`
`15
`
`60
`
`180
`
`80 C
`
`60
`
`40
`
`20
`
`0
`
`Control
`
`15
`
`60
`
`180
`
`Incubation time (minutes)
`FIG. 5. Susceptibility of ST lb(6-18) (shaded bars), N9P10 gua(cid:173)
`nylin (hatched bars), and guanylin (open bars) to protease degrada(cid:173)
`tion. The ability of treated analogs to inhibit the binding of radiola(cid:173)
`beled toxin to rat intestinal cells was measured as a function of time
`of exposure to either chymotrypsin (A), suckling mouse intestinal
`fluid (B), or filtrates of infant mouse intestinal tissue homogenates
`(C). Each histogram bar represents the average of experiments
`performed in duplicate.
`
`long-lived nature of ST I enterotoxins may thus be partially
`responsible for their ability to chronically activate the intes(cid:173)
`tinal guanylate cyclase, causing an ensuing diarrheal re(cid:173)
`sponse. Interestingly, an analog of ST I with a substitution of
`Asn-12 by tyrosine was shown to be enterotoxic (40). It
`would thus appear that the presence of a praline residue at
`position 13 of ST lb is sufficient to reduce the protease
`susceptibility of the Tyr-Pro peptide bond. Cohen and Gi(cid:173)
`annella (5) have reported that the susceptibility of immature
`rat jejunum to ST I action is higher than that of adult rat
`intestinal tissue. They concluded that the rapid clearance of
`ST I observed in adult rats and the slow disappearance of the
`enterotoxin in young rats are predominantly linked to bio(cid:173)
`chemical events other than ST l's exposure to intestinal
`proteases present in lumenal fluid or to brush border mem(cid:173)
`brane-associated proteases (5). Finally, the recent observa(cid:173)
`tions that guanylin and ST Ia(S-17) bind to the same receptor
`and are able to produce a similar elevation in intracellular
`cGMP and stimulate a- secretion in T84 cells support the
`notion that one receptor common to both peptides is prob(cid:173)
`ably involved in the diarrheal response (12).
`An alternate mechanism of action for ST I enterotoxins
`
`Bausch Health Ireland Exhibit 2060, Page 4 of 6
`Mylan v. Bausch Health Ireland - IPR2022-00722
`
`
`
`4714
`
`CARPICK AND GARIEPY
`
`INFECT. IMMUN.
`
`would be to invoke the existence of a low-abundance class of
`ST I receptors not linked to the activation of guanylate
`cyclase but potentially acting in conjunction with the acti(cid:173)
`vated cyclase, resulting in the net efflux of water, sodium,
`and chloride ions. The recent observation that a class of
`high-affinity, low-occupancy receptors for ST I which are
`not coupled to guanylate cyclase may exist on rat intestinal
`membranes would provide some support for this hypothesis
`(20). Similarly, a novel receptor for ST I enterotoxins has
`been identified on IEC-6 cells (24). The key to the ligand
`specificity of the two (or more) receptor classes may lie in
`the central region of the ST I sequence, which is closely
`homologous to the same region of N9P10 guanylin but is less
`so to guanylin. Experiments involving amino acid substitu(cid:173)
`tions within the ST I sequence (4, 21, 27) have pointed to the
`functional importance of the central structural region of ST
`I. Interestingly, the sequence of the recently discovered
`Escherichia coli enteroaggregative heat-stable enterotoxin is
`more divergent than that of guanylin when compared with
`other ST I enterotoxins. Its enterotoxicity may thus be
`dependent on the structural context of Ala-14 and Cys-15 in
`particular rather than the entire tum region encompassing
`residues Asn-12 to Cys-15 (30). A consequence of this
`hypothesis is that residues such as Asn-12 and Pro-13 may
`only play an indirect role in the binding and/or signalling
`event leading to the diarrheal effect. In addition, several
`signal transduction pathways (linked or not to cGMP as a
`second messenger) have been implicated in the induction of
`intestinal secretion by ST I enterotoxins. Specifically, other
`groups have provided evidence for the activation of a protein
`kinase C (37) and the production of inositol polyphosphates
`(2, 3). It is unclear whether these responses are essential
`steps leading to fluid accumulation or are a consequence of
`triggering other intracellular second messengers. In this
`model, the ST I receptor guanylate cyclase remains occupied
`with either guanylin or ST I, implying that the elevation of
`intracellular cGMP is an ongoing event that may still play a
`supportive if not essential role in the onset of diarrhea.
`Because the overall increase in cGMP is similar between ST
`lb(6-18) and the guanylin analogs (Fig. 4), it may be that the
`potency of ST I causes a downregulation in the expression of
`the ST I-receptor guanylate cyclase and thus a short-term
`insensitivity of these cells to the presence of guanylin. One
`would thus project a drop in intracellular cGMP levels after
`the initial stimulatory event. This event may also be a key to
`the action of ST I and was originally observed by Giannella
`and Drake (16). The physiological role of guanylin remains to
`be understood.
`In summary, it was found that native guanylin competes
`with radiolabeled ST lb analog for binding to rat cells but
`does not cause a diarrheal response in suckling mice except
`at elevated doses. In contrast, N9P10 guanylin is a weaker
`competitor than guanylin in displacing radiolabeled ST lb
`bound to rat cells, yet in mice it causes diarrhea at much
`lower concentrations. The pattern of activation of rat intes(cid:173)
`tinal guanylate cyclase for the two guanylin analogs parallels
`the results of the receptor binding assay but not those of the
`infant mouse assay. Guanylin is rapidly degraded by com(cid:173)
`ponents present in the infant mouse intestinal tract, suggest(cid:173)
`ing that ST I enterotoxins may be acting as long-lived
`superagonists of guanylin.
`
`ACKNOWLEDGMENTS
`This work was supported by grants from the National Institutes of
`Health (AI26152) and the Medical Research Council of Canada
`(MT-12204).
`
`We thank Jim Ferguson and Nancy Stokoe for technical assis(cid:173)
`tance and Henrianna Pang for performing the mass spectrometry
`analysis.
`
`REFERENCES
`1. Aimoto, S., T. Takao, Y. Shimonishi, S. Hara, T. Takeda, Y.
`Takeda, and T. Miwatani. 1982. Amino acid sequence of a
`heat-stable enterotoxin produced by human enterotoxigenic
`Escherichia coli. Eur. J. Biochem. 129:257-263.
`2. Banik, N., and U. Ganply. 1988. Stimulation of phosphoinosi(cid:173)
`tides breakdown by the heat-stable E. coli enterotoxin in rat
`intestinal epithelial cells. FEBS Lett. 236:489--492.
`3. Banik, N., and U. Ganguly. 1989. Diacylglycerol breakdown in
`plasma membrane of rat intestinal epithelial cells: effect of E.
`coli heat-stable toxin. FEBS Lett. 250:201-204.
`4. Carpick, 8. W., and J. Gariepy. 1991. Structural characteriza(cid:173)
`tion of functionally important regions of the Escherichia coli
`heat-stable enterotoxin ST lb. Biochemistry 30:4803--4809.
`5. Cohen, M. B., and R. A. Giannella. 1992. Jejunal toxin inacti(cid:173)
`vation regulates susceptibility of the immature rat to ST •.
`Gastroenterology 102:1988-1996.
`6. Cnrrie, M. G., K. F. Fok, J. Kato, R. J. Moore, F. K. Hamra,
`K. L Dulin, and C. E. Smith. 1992. Guanylin: an endogenous
`activator of intestinal guanylate cyclase. Proc. Natl. Acad. Sci.
`USA 89:947-951.
`7. de Sauvage, F. J., T. R. Camerato, and D. V. Goedde!. 1991.
`Primary structure and functional expression of the human
`receptor for Escherichia coli heat-stable enterotoxin. J. Biol.
`Chem. 266:17912-17918.
`8. de Sanvage, F. J., R. Borok, G. Bennett, C. Quan, J.P. Rumler,
`and D. V. Goedde!. 1992. Characterization of the recombinant
`human receptor for Escherichia coli heat-stable enterotoxin. J.
`Biol. Chem. 267:6479--6482.
`9. de Sauvage, F. J., s. Keshav, W.-J. Kuang, N. Gillett, W.
`Henzel, and D. V. Goedde!. 1992. Precursor structure, expres(cid:173)
`sion, and tissue distribution of human guanylin. Proc. Natl.
`Acad. Sci. USA 89:9089-9093.
`10. Dreyfus, LA., L. Jaso-Friedmann, and D. C. Robertson. 1984.
`Characterization of the mechanism of action of Escherichia coli
`heat-stable enterotoxin. Infect. Immun. 44:493--501.
`11. Field, M., L H. Graf, Jr., W. J. Laird, and P. L Smith. 1978.
`Heat-stable enterotoxin of Escherichia coli: in vitro effects on
`guanylate cyclase activity, cGMP concentration, and ion trans(cid:173)
`port in small intestine. Proc. Natl. Acad. Sci. USA 7S:2800-
`2804.
`12. Forte, LR., S. L Eber, J. T. Turner, R.H. Freeman, K. F. Fok,
`and M. G. Cnrrie. 1993. Guanylin stimulation of a- secretion in
`human intestinal T84 cells via cyclic guanosine monophosphate.
`J. Clin. Invest. 91:2423--2428.
`13. Frantz, J. C., L. Jaso-Friedman, and D. C. Robertson. 1984.
`Binding of Escherichia coli heat-stable enterotoxin to rat intes(cid:173)
`tinal cells and brush border membranes. Infect. Immun. 43:622-
`630.
`14. Gariepy, J., A. K. Jndd, and G. K. Schoolnlk. 1987. Importance
`of disulfide bridges in the structure and activity of Escherichia
`coli enterotoxin ST lb. Proc. Natl. Acad. Sci. USA 84:8907-
`8911.
`15. Gariepy, J., and G. K. Schoolnik. 1986. Design of a photoreac(cid:173)
`tive analogue of the Escherichia coli heat-stable enterotoxin ST
`lb: use in identifying its receptor on rat brush border mem(cid:173)
`branes. Proc. Natl. Acad. Sci. USA 83:483--487.
`16. Glannella, R. A., and K. W. Drake. 1979. Effect of purified
`Escherichia coli heat-stable enterotoxin on intestinal cyclic
`nucleotide metabolism and fluid secretion. Infect. Immun. 24:
`19-23.
`17. Giannella, R. A., M. Luttrell, and M. Thompson. 1983. Binding
`of Escherichia coli heat-stable enterotoxin to recept