`Proc. Natl. Acad. Sci. USA
`Vol. 89, pp. 947-951, February 1992
`Vol. 89, pp. 947-951, February 1992
`Pharmacology
`Pharmacology
`
`Guanylin: An endogenous activator of intestinal guanylate cyclase
`Guanylin: An endogenous activator of intestinal guanylate cyclase
`(intestine/cyclic GMP/heat-stable enterotoxin/diarrhea/peptide)
`(intestine/cyclic GMP/heat-stable enterotoxin/diarrhea/peptide)
`MARK G. CURRIE*t, KAM F. FOKt, JowI KATO*, ROSALYN J. MOORE*, FRANKLIN K. HAMRA*,
`MARK G. CURRIE*t, KAM F. FOO, JOHJI KATO*, ROSALYN J. MOORE*, FRANKLIN K. HAMRA*,
`KEVIN L. DUFFIN§, AND CHRISTINE E. SMITHS
`KEVIN L. DUFFIN§, AND CHRISTINE E. SMITH¶
`Departments of *Molecular Pharmacology, tBiological Chemistry, §Physical Sciences, and lProtein Biochemistry, Monsanto Corporate Research, Monsanto
`Departments of *Molecular Pharmacology, *Biological Chemistry, *Physical Sciences, and 1Protein Biochemistry, Monsanto Corporate Research, Monsanto
`Company, St. Louis, MO 63167
`Company, St. Louis, MO 63167
`
`Communicated by Philip Needleman, October 11, 1991
`Communicated by Philip Needleman, October 11, 1991
`
`Intestinal guanylate cyclase mediates the ac-
`ABSTRACT
`Intestinal guanylate cyclase mediates the ac-
`ABSTRACT
`tion of the heat-stable enterotoxin to cause a decrease in
`tion of the heat-stable enterotoxin to cause a decrease in
`intestinal fluid absorption and to increase chloride secretion,
`intestinal fluid absorption and to increase chloride secretion,
`ultimately causing diarrhea. An endogenous ligand that acts on
`ultimately causing diarrhea. An endogenous ligand that acts on
`this guanylate cyclase has not previously been found. To search
`this guanylate cyclase has not previously been found. To search
`for a potential endogenous ligand, we utilized T84 cells, a
`for a potential endogenous ligand, we utilized T84 cells, a
`human colon carcinoma-derived cell line, in culture as a
`human colon carcinoma-derived cell line, in culture as a
`bioassay. This cell line selectively responds to the toxin in a very
`bioassay. This cell line selectively responds to the toxin in a very
`sensitive manner with an increase in intracellular cyclic GMP.
`sensitive manner with an increase in intracellular cyclic GMP.
`In the present study, we describe the purification and structure
`In the present study, we describe the purification and structure
`of a peptide from rat jejunum that activates this enzyme. This
`of a peptide from rat jejunum that activates this enzyme. This
`peptide, which we have termed guanylin, is composed of 15
`peptide, which we have termed guanylin, is composed of 15
`amino acids and has the following amino acid sequence,
`amino acids and has the following amino acid sequence,
`PNTCEICAYAACTGC, as determined by automated Edman
`PNTCEICAYAACTGC, as determined by automated Edman
`degradation sequence analysis and electrospray mass spec-
`degradation sequence analysis and electrospray mass spec-
`trometry. Analysis of the amino acid sequence of this peptide
`trometry. Analysis of the amino acid sequence of this peptide
`reveals a high degree of homology with heat-stable enterotox-
`reveals a high degree of homology with heat-stable enterotox-
`ins. Solid-phase synthesis of this peptide confirmed that it
`ins. Solid-phase synthesis of this peptide confirmed that it
`stimulates increases in T84 cyclic GMP levels. Guanylin re-
`stimulates increases in T84 cyclic GMP levels. Guanylin re-
`quired oxidation for expression of bioactivity and subsequent
`quired oxidation for expression of bioactivity and subsequent
`reduction of the oxidized peptide eliminated the effect on cyclic
`reduction of the oxidized peptide eliminated the effect on cyclic
`GMP, indicating a requirement for cysteine disulfide bond
`GMP, indicating a requirement for cysteine disulfide bond
`formation. Synthetic guanylin also displaces heat-stable enter-
`formation. Synthetic guanylin also displaces heat-stable enter-
`otoxin binding to cultured T84 cells. Based on these data, we
`otoxin binding to cultured T84 cells. Based on these data, we
`propose that guanylin is an activator of intestinal guanylate
`propose that guanylin is an activator of intestinal guanylate
`cyclase and that it stimulates this enzyme through the same
`cyclase and that it stimulates this enzyme through the same
`receptor binding region as the heat-stable enterotoxins.
`receptor binding region as the heat-stable enterotoxins.
`
`Guanylate cyclase is comprised of a group of proteins that
`Guanylate cyclase is comprised of a group of proteins that
`share a common enzymatic function ofproducing cyclic GMP
`share a common enzymatic function of producing cyclic GMP
`but differ quite remarkably in their selectivity toward acti-
`but differ quite remarkably in their selectivity toward acti-
`vation by ligands. The three.major types of guanylate cyclase
`vation by ligands. The three major types of guanylate cyclase
`are the soluble, particulate, and intestinal [cytoskeletal-
`are the soluble, particulate, and intestinal [cytoskeletal-
`associated particulate or heat-stable enterotoxin (STa)-
`associated particulate or heat-stable enterotoxin (STa)-
`sensitive] forms and each is regulated by different ligands (1,
`sensitive] forms and each is regulated by different ligands (1,
`2). Activation of the soluble guanylate cyclase occurs in
`2). Activation of the soluble guanylate cyclase occurs in
`response to nitric oxide, whereas activation of the particulate
`response to nitric oxide, whereas activation of the particulate
`enzyme occurs in response to the natriuretic peptides (atrial
`enzyme occurs in response to the natriuretic peptides (atrial
`natriuretic peptide, brain natriuretic peptide, and C-type
`natriuretic peptide, brain natriuretic peptide, and C-type
`natriuretic peptide) (1, 2). An endogenous activator of the
`natriuretic peptide) (1, 2). An endogenous activator of the
`intestinal guanylate cyclase has not previously been identi-
`intestinal guanylate cyclase has not previously been identi-
`fied. However, the STa from Escherichia coli is known to
`fied. However, the STa from Escherichia coli is known to
`selectively activate this form of the enzyme (3, 4). The
`selectively activate this form of the enzyme (3, 4). The
`intestinal form is predominantly found in the intestinal epi-
`intestinal form is predominantly found in the intestinal epi-
`thelial cells with the largest number of receptors oriented
`thelial cells with the largest number of receptors oriented
`toward the lumen (1, 2). Recently, it has been cloned and
`toward the lumen (1, 2). Recently, it has been cloned and
`expressed from rat small intestinal mucosa (5). This enzyme
`expressed from rat small intestinal mucosa (5). This enzyme
`is characterized by an extracellular receptor binding region,
`is characterized by an extracellular receptor binding region,
`a transmembrane region, an intracellular protein kinase-like
`a transmembrane region, an intracellular protein kinase-like
`region, and a cyclase catalytic domain (5).
`region, and a cyclase catalytic domain (5).
`
`The publication costs of this article were defrayed in part by page charge
`The publication costs of this article were defrayed in part by page charge
`payment. This article must therefore be hereby marked "advertisement"
`payment. This article must therefore be hereby marked "advertisement"
`in accordance with 18 U.S.C. §1734 solely to indicate this fact.
`in accordance with 18 U.S.C. §1734 solely to indicate this fact.
`
`Pathogenic strains of E. coli and other bacteria produce a
`Pathogenic strains of E. coli and other bacteria produce a
`family of heat-stable enterotoxins (STs) that activate intes-
`family of heat-stable enterotoxins (STs) that activate intes-
`tinal guanylate cyclase. STs are acidic peptides that contain
`tinal guanylate cyclase. STs are acidic peptides that contain
`18 or 19 amino acids with six cysteines and three disulfide
`18 or 19 amino acids with six cysteines and three disulfide
`bridges that are required for full expression of bioactivity (6).
`bridges that are required for full expression of bioactivity (6).
`The increase of intestinal epithelial cyclic GMP elicited by
`The increase of intestinal epithelial cyclic GMP elicited by
`STs is thought to cause a decrease in water and sodium
`STs is thought to cause a decrease in water and sodium
`absorption and an increase in chloride secretion (7, 8). These
`absorption and an increase in chloride secretion (7, 8). These
`changes in intestinal fluid and electrolyte transport then act
`changes in intestinal fluid and electrolyte transport then act
`to cause secretory diarrhea. In developing countries, the
`to cause secretory diarrhea. In developing countries, the
`diarrhea resulting from STs causes many deaths, particularly
`diarrhea resulting from STs causes many deaths, particularly
`in the infant population (9). STs are also considered a major
`in the infant population (9). STs are also considered a major
`cause of traveler's diarrhea in developed countries (10). They
`cause of traveler's diarrhea in developed countries (10). They
`have also been reported to be a leading cause of morbidity
`have also been reported to be a leading cause of morbidity
`and death in domestic animals (11).
`and death in domestic animals (11).
`In the present study, we designed a bioassay to search for
`In the present study, we designed a bioassay to search for
`a potential endogenous ligand that activates the intestinal
`a potential endogenous ligand that activates the intestinal
`guanylate cyclase. This bioassay is based on the demonstra-
`guanylate cyclase. This bioassay is based on the demonstra-
`tion that T84 cells in culture respond to ST in a selective and
`tion that T84 cells in culture respond to ST in a selective and
`sensitive manner with graded increases of intracellular cyclic
`sensitive manner with graded increases of intracellular cyclic
`GMP. This bioassay revealed that the intestine as well as the
`GMP. This bioassay revealed that the intestine as well as the
`kidney possessed an active material. Purification of this
`kidney possessed an active material. Purification of this
`material from the rat intestine was accomplished and the
`material from the rat intestine was accomplished and the
`structure was determined to be a 15-amino acid peptide with
`structure was determined to be a 15-amino acid peptide with
`4 cysteines that must be disulfide-linked for bioactivity. The
`4 cysteines that must be disulfide-linked for bioactivity. The
`peptide, termed guanylin, also possesses a high degree of
`peptide, termed guanylin, also possesses a high degree of
`homology with STs.
`homology with STs.
`
`MATERIALS AND METHODS
`MATERIALS AND METHODS
`Cell Culture. A cultured human colon carcinoma cell line
`Cell Culture. A cultured human colon carcinoma cell line
`(T84) was obtained from the American Type Culture Collec-
`(T84) was obtained from the American Type Culture Collec-
`tion at passage 52. Cells were grown to confluency in 24-well
`tion at passage 52. Cells were grown to confluency in 24-well
`culture plates with a 1:1 mixture of Ham's F12 medium and
`culture plates with a 1:1 mixture of Ham's F12 medium and
`Dulbecco's modified Eagle's medium (DMEM) supple-
`Dulbecco's modified Eagle's medium (DMEM) supple-
`mented with 10% fetal calf serum, 100 international units of
`mented with 10% fetal calf serum, 100 international units of
`penicillin per ml, and 100 ,g of streptomycin per ml. Cells
`penicillin per ml, and 100 µg of streptomycin per ml. Cells
`were used at passages 54-60.
`were used at passages 54-60.
`Cyclic GMP Determination. Monolayers of T84 cells in
`Cyclic GMP Determination. Monolayers of T84 cells in
`24-well plates were washed twice with 1 ml of DMEM per ml
`24-well plates were washed twice with 1 ml of DMEM per ml
`and then incubated at 37°C for 10 min with 0.5 ml of DMEM
`and then incubated at 37°C for 10 min with 0.5 ml of DMEM
`containing 1 mM isobutylmethylxanthine (IBMX), a cyclic
`containing 1 mM isobutylmethylxanthine (IBMX), a cyclic
`nucleotide phosphodiesterase inhibitor. Agents and fractions
`nucleotide phosphodiesterase inhibitor. Agents and fractions
`were then added for the indicated time as described in
`were then added for the indicated time as described in
`Results. The media was then aspirated and the reaction was
`Results. The media was then aspirated and the reaction was
`terminated by the addition of ice-cold 0.5 ml of 0.1 M HCL.
`terminated by the addition of ice-cold 0.5 ml of 0.1 M HCI.
`Aliquots were then evaporated to dryness under nitrogen and
`Aliquots were then evaporated to dryness under nitrogen and
`resuspended in 5 mM sodium acetate buffer (pH 6.4). The
`resuspended in 5 mM sodium acetate buffer (pH 6.4). The
`
`Abbreviations: IBMX, isobutylmethylxanthine; TFA, trifluoroacetic
`Abbreviations: IBMX, isobutylmethylxanthine; TFA, trifluoroacetic
`acid; C18, octadecasilyl; STa, heat-stable enterotoxin; ST, heat-
`acid; C18, octadecasilyl; STa, heat-stable enterotoxin; ST, heat-
`stable enterotoxin; DPI, dithiothreitol; PTH, phenylthiohydantoin.
`stable enterotoxin; DTT, dithiothreitol; PTH, phenylthiohydantoin.
`tTo whom reprint requests should be addressed at: Monsanto
`tTo whom reprint requests should be addressed at: Monsanto
`Company, Mail Zone T3P, 800 North Lindbergh Boulevard, St.
`Company, Mail Zone T3P, 800 North Lindbergh Boulevard, St.
`Louis, MO 63167.
`Louis, MO 63167.
`
`947
`947
`
`MYLAN EXHIBIT - 1031
`Mylan Pharmaceuticals, Inc. v. Bausch Health Ireland, Ltd. - IPR2022-00722
`
`
`
`948
`
`Pharmacology: Currie et al.
`
`Proc. Natl. Acad. Sci. USA 89 (1992)
`
`samples were subsequently measured for cyclic GMP by RIA
`as described by Steiner et al. (12).
`Purification of Guanylin. Rat jejunums flushed of luminal
`contents with 50 ml of saline and immediately placed on dry
`ice were obtained from Bioproducts for Science (Indianap-
`olis). The jejunums were thawed, minced, and boiled for 10
`min in 1 M acetic acid. The extract was centrifuged at 20,000
`x g for 20 min at 4°C. The resulting supernatant was filtered
`and applied to an octadecasilyl (C18) Sep-Pak (Waters). The
`column was washed with 10% acetonitrile/0.1% trifluoroace-
`tic acid (TFA)/H2O and eluted with 60% acetonitrile/0.1%
`TFA/H2O. The eluted peptide fraction was lyophilized and
`resuspended in 50 ml of distilled H2O containing 0.8% am-
`pholytes, pH range 3-10, and applied to a preparative iso-
`electric focusing cell (Rotofor, Bio-Rad). The sample was
`focused for 150 min at 12 W constant power. The fractions
`were harvested, pH determined, and bioassayed. The active
`fractions, which focused around pH 3.8, were then refocused
`under similar conditions and the resulting active fractions
`were lyophilized. The sample was then resuspended in 1 ml
`of 10% acetonitrile/0.1% TFA/H2O, applied to a C18 semi-
`preparative HPLC column (Vydac, Hesperia, CA), and
`eluted at a flow rate of 3 ml/min. The following gradient was
`used to fractionate the sample: 10% acetonitrile, 0.1% TFA
`to 30% acetonitrile, 0.1% TFA in 180 min. The active fraction
`was then determined by bioassay and lyophilized. This
`sample was resuspended in 1 ml of 10% acetonitrile/0.1%
`TFA/H2O and applied to a phenyl analytical HPLC column
`(Vydac, Hesperia, CA). The conditions for elution were
`similar to that described above for the semipreparative col-
`umn except the flow was 1 ml/min. The active fraction was
`lyophilized and then resuspended in 1 ml of 10% acetonitrile/
`0.1% TFA/H2O. The sample was then applied to a C18
`analytical HPLC column (Vydac) and eluted according to the
`above description for the phenyl column. The active fraction
`was identified by bioassay and lyophilized. The sample was
`reconstituted in 1 ml of 10% acetonitrile/0.1% TFA/H2O,
`reapplied to the analytical C18 column, and eluted by a
`gradient of 10% acetonitrile/10 mM ammonium acetate/H2O,
`pH 6.2, to 30% acetonitrile/10 mM ammonium acetate/H2O,
`pH 6.2, in 180 min. The active fraction was lyophilized and
`reconstituted in 0.05 ml of 0.1% TFA/H2O. The sample was
`then applied to a C8 microbore column and eluted by an
`increasing gradient of 0.33%/min of acetonitrile/0.1% TFA/
`H2O. Two separate batches of purified peptide were then
`subjected to sequence analysis.
`N-Terminal Protein Sequence Analysis. Automated Edman
`degradation chemistry was used to determine the NH2-
`terminal protein sequence. An Applied Biosystems model
`470A gas-phase sequencer was employed for the degrada-
`tions (13) using the standard sequencer cycle 03RPTH. The
`respective phenylthiohydantoin (PTH) amino acid deriva-
`tives were identified by reverse-phase HPLC analysis in an
`on-line fashion employing an Applied Biosystems model
`120A PTH analyzer fitted with a Brownlee 2.1-mm i.d. PTH
`C18 column. On-sequencer pyridylethylation was performed
`as outlined by Kruft et al. (14). The PTH derivative of
`pyridylethylcysteine was identified by HPLC as eluting
`slightly prior to the PTH derivative of methionine.
`Electrospray Mass Spectrometry. Individual samples of
`native and synthetic guanylin were purified by microbore C8
`reverse-phase HPLC (Brownlee Aquapore RP-300 7-µm col-
`umn, P. J. Cobert, St. Louis, MO) and eluting fractions of the
`pmol/µ1 for
`peptides were collected and concentrated to
`mass analysis. Sample solutions were introduced to the mass
`spectrometer via injection into a stream of acetonitrile/H2O/
`TFA, 1000:1000:1, vol/vol/vol, which continuously flowed
`to the mass spectrometer at a flow of 10 µ1/min. Three
`microliters of each of the concentrated guanylin samples was
`injected to obtain the results that are presented in this paper.
`
`A Sciex API III triple-quadrupole mass spectrometer
`(Thornhill, Ontario, Canada) equipped with an atmospheric
`pressure ion source was used to sample positive ions pro-
`duced from an electrospray interface (15). Mass analysis of
`sample ions was accomplished by scanning the first quadru-
`pole in 1 atomic mass unit increments from 1000 to 2400
`atomic mass units in
`s and passing mass-selected ions
`through the second and third quadrupoles operated in the
`rf-only mode to the multiplier. For maximum sensitivity, the
`mass resolution of the quadrupole mass analyzer was set so
`that ion signals were "---2 atomic mass units wide at half peak
`height, but the centroid of the ion signal still represented the
`correct mass of the ion. Comparison of the oxidized and
`reduced guanylin molecular ion region was made by scanning
`the quadrupole mass analyzer in 0.1 atomic mass unit steps
`from 1510 to 1525 atomic mass units in 2 s. Mass spectra of
`the guanylin samples were averaged over all of the scans that
`were acquired during elution of the 3-µI sample solution.
`Binding Assay. 125I-labeled STa (125I-STa) was prepared by
`the Iodo-Gen method (16). T84 cell monolayers were washed
`twice with 1 ml of DMEM and then incubated for 30 min at
`37°C in 0.5 ml of DMEM with 125I-STa (amino acids 5-18)
`(100,000 cpm per well) and either guanylin or 100 nM STa.
`The cells were then washed four times with 1 ml of DMEM
`and solubilized with 0.5 ml of 1 M NaOH per well. This
`volume was transferred to tubes and assayed for radioactivity
`by a y counter. Results are expressed as the percentage
`specifically bound.
`Chemical Synthesis of Guanylin. Guanylin was synthesized
`by the solid-phase method (17) with an Applied Biosystems
`430A peptide synthesizer on Cys(4-CH3Bzl)-OCH2-phenyl-
`acetamidomethyl resin using double coupling cycles to en-
`sure complete coupling at each step. Coupling was effected
`with preformed symmetrical anhydride of tert-butoxycar-
`bonyl amino acids (Applied Biosystems), and peptides were
`cleaved from the solid support in hydrogen fluoride/
`dimethylsulfide/anisole/p-thiocresol, 8:1:1:0.5, vol/vol/
`vol/wt, at 0°C for 60 min. Peptides were cyclized using
`dimethylsulfoxide as described by Tam et al. (18). Peptides
`were purified by successive reverse-phase chromatography
`on a 45 x 300 mm Vydac C18 column and on a 19 x 150 mm
`µBondapak C18 column, using a gradient of 10-30% aceto-
`nitrile in 0.5% TFA. The structures and purity of the syn-
`thetic peptides were verified by fast atom bombardment mass
`spectrometry, amino acid analysis, and gas-phase sequence
`analysis.
`
`RESULTS
`Initial characterization of the T84 cell response indicated that
`these cells were very sensitive to STa with a concentration of
`10-1° M eliciting a 4-fold increase in cyclic GMP. The cells
`also displayed a remarkable range, with a maximal response
`of STa (10-7 M) eliciting a >1000-fold increase in cyclic
`GMP. Furthermore, we failed to detect an effect of either
`sodium nitroprusside (10-3 M) or atrial natriuretic peptide
`(10-6 M) on cyclic GMP levels, suggesting that the T84 serves
`as a selective bioassay for agents that activate the intestinal
`guanylate cyclase. Various rat tissues were surveyed as
`sources for T84 cell cyclic GMP agonist activity, and jejunum
`and kidney were found to possess activity while liver, brain,
`pancreas, spleen, lung, and testes lacked detectable activity
`(Fig. 1). We also observed that rat embryonic intestine
`possessed a similar activity. Treatment of the T84 cells with
`10% of embryonic intestinal extract increased the cyclic GMP
`from a basal level of 120 ± 10 fmol per well to 270 ± 10 fmol
`per well (mean ± SE).
`Purification of the adult rat jejunal bioactivity was accom-
`plished by the processing scheme described in Materials and
`Methods. Briefly, following acid boiling and extraction by a
`
`
`
`Pharmacology: Currie et al.
`
`Proc. Natl. Acad. Sci. USA 89 (1992)
`
`949
`
`3000
`
`a)
`
`2000
`
`0
`E
`2a_ 1000
`
`Gds
`c)./
`
`•d
`/?
`
`<5,-
`i ro
`o
`Pd S
`
`47 4'
`
`"12
`
`FIG. 1. Effect of extracts from various tissues on TM cyclic GMP
`levels. Acid extracts were prepared from 1 g of tissue and 10% of each
`extract was applied to IBMX-treated cells. Values are means ± SE
`(n = 3).
`
`C18 reverse-phase matrix, the material was fractionated on a
`preparative isoelectric focusing cell, which resulted in a
`200-fold purification and indicated that the isoelectric point
`was about 3.8. Refocusing of the active fraction resulted in a
`further 5- to 10-fold purification. The active fraction was then
`purified to homogeneity by a series of reverse-phase HPLC
`steps, including a semipreparative C18 column, a phenyl
`column, two separations on a C18 column utilizing different
`ion-pairing reagents, and final purification on a microbore C8
`column (Fig. 2).
`Preliminary experiments suggested that the material was a
`low molecular weight peptide; therefore the material was
`subjected to N-terminal protein sequence analysis and to
`electrospray mass spectrometric analysis. The combination
`of the data derived from these two techniques yielded the
`complete sequence for guanylin: Pro-Asn-Thr-Cys-Glu-Ile-
`Cys-Ala-Tyr-Ala-Ala-Cys-Thr-Gly-Cys. The N-terminal se-
`quence through 14 places was determined by two indepen-
`dent gas-phase sequencing experiments. The C-terminal
`amino acid was deduced from data obtained by electrospray
`mass spectroscopy. The initial results yielded a sequence in
`which no PTH amino acid derivative was observed at posi-
`tions 4, 7, and 12. Since cysteine residues cannot be posi-
`tively identified during gas-phase sequencing without reduc-
`tion and alkylation, the lack of a PTH amino acid derivative
`at these positions suggested the presence of cysteine resi-
`dues. For complete verification, the putative cysteine resi-
`dues of guanylin were pyridylethylated and the peptide was
`resequenced. The subsequent N-terminal gas-phase se-
`quence analysis verified cysteine residues at positions 4, 7,
`and 12. Further primary structure information was obtained
`by electrospray mass spectrometry. The electrospray mass
`spectrum of native guanylin (Fig. 3A) contains an ion signal
`at m/z 1516 that corresponds to the protonated peptide. This
`assignment is 103 atomic mass units higher than the mass
`expected for the peptide whose sequence was obtained by
`gas-phase sequence analysis. This difference of 103 atomic
`
`FIG. 2. Final purification of guany-
`lin by C8 reverse-phase microbore
`HPLC. Chromatographic peaks (A214,
`0.1 absorbance unit full scale) were
`collected and measured for activity.
`The active peak (55% of a full-scale
`response) is indicated by shading, with
`1% of the fraction giving a 10-fold in-
`crease in cyclic GMP.
`
`I
`10
`
`I
`20
`
`I
`30
`
`Time (min)
`
`mass units is consistent with an additional disulfide-linked
`cysteine or with a threonine at the C terminus of the peptide.
`Reduction of the disulfide bonds of guanylin with dithiothrei-
`tol (DTT) resulted in a 4 atomic mass unit increase in
`molecular weight of the peptide (Fig. 3 B and C), indicating
`that it contains two disulfide bonds. Therefore, since there
`are only three cysteines in the original 14 N-terminal amino
`acids, the 103 atomic mass unit difference must result from an
`additional C-terminal cysteine that is disulfide-linked to one
`of the other three cysteines in the guanylin sequence.
`The resulting full amino acid sequence of the peptide was
`compared with all other proteins in the GenBank, National
`Biomedical Research Foundation, and SwissProt databases
`by a computer-based search. This search revealed that gua-
`nylin has homology with the STs, with the greatest homology
`identified in the cysteine-rich regions of the molecules (19,
`20). The distinctive difference between guanylin and the STs
`is that guanylin possesses four cysteines with two disulfide-
`linked bridges while all of the known STs have six cysteines
`with three disulfide-linked bridges (Fig. 4).
`Chemical synthesis of guanylin based on the experimentally
`derived sequence resulted in three different HPLC fractions
`following oxidation in air. Each of these fractions contained a
`peptide with the same molecular weight as native guanylin
`(1516 atomic mass units) as determined by mass spectrometric
`analysis. However, only one of these fractions exhibited
`potent bioactivity in the T84 cell bioassay consistent with
`guanylin. This fraction also exhibited a similar HPLC reten-
`tion time to that of native guanylin. Since guanylin has four
`cysteine residues, the three fractions of synthetic guanylin
`probably represented the three possible different disulfide
`bridge alignments. Bioactive synthetic guanylin stimulated
`increases in cyclic GMP levels of T84 cells that were time and
`concentration dependent. Guanylin (10-8 M) caused a marked
`elevation of cyclic GMP after 1 min, which progressively
`increased through 30 min (Fig. 5A). Examination of the
`concentration—response curve shows that guanylin elicited an
`increase in cyclic GMP at 10-10 M and this response increased
`through the range of concentrations tested (Fig. 5B). To
`characterize the effect of treatment of reducing agents on the
`bioactivity of guanylin, we pretreated the peptide for 30 min
`with 1 mM DTT. The basal level of cyclic GMP for this
`experiment was 160 ± 50 fmol per well, which increased to
`2820 ± 500 fmol per well after a 30-min treatment with guanylin
`(10' M). However, following the pretreatment of the peptide
`with DTT, the effect of the 30-min treatment with the peptide
`on cyclic GMP was almost completely abolished (250 ± 50
`fmol per well). The action of DTT does not appear to be a
`direct effect of DTT on guanylate cyclase since treatment of
`the cells with 10 µM DTT (final concentration of DTT that the
`cells were exposed to in the experiment) failed to affect their
`responsiveness to STa treatment (data not shown). Finally, we
`examined in preliminary experiments the ability of guanylin to
`displace specifically bound 125I-STa from T84 cells. In this
`experiment, guanylin caused a concentration-dependent dis-
`placement of labeled STa from the T84 cells (Fig. 6) with an
`IC50 of 5 x 10-8 M.
`
`DISCUSSION
`In the present study, we describe the purification and se-
`quence of a rat intestinal peptide that possesses the proper-
`ties consistent with an endogenous ligand for the intestinal
`guanylate cyclase. We have termed this peptide guanylin
`because of its ability to stimulate intestinal guanylate cyclase.
`Synthetic guanylin was found to increase cyclic GMP levels
`in T84 cells in a time- and concentration-dependent manner.
`Guanylin was also found to displace the specific binding of
`' 25I-STa from T84 cells. Therefore, these data support our
`proposal that guanylin is an endogenous activator of the
`
`
`
`950
`
`Pharmacology: Currie et al.
`
`Proc. Natl. Acad. Sci. USA 89 (1992)
`
`100
`
`A
`
`1 5 1 6
`
`1000
`
`1250
`
`1500
`
`1750
`
`2000
`
`2250
`
`100
`
`1515.6
`
`m/z
`
`C
`
`100
`
`cc
`
`1519.6
`
`1510
`
`1515
`
`1520
`
`1525
`
`1510
`
`1515
`
`1520
`
`1525
`
`m/z
`
`m/z
`
`FIG. 3. Electrospray mass spectra of native guanylin. (A) Mass spectral analysis of native guanylin in a range of 1000-2400 atomic mass units
`shows the molecular weight to be 1516 atomic mass units. (B and C) Comparison of the mass spectra of oxidized (B) and reduced (C) native
`guanylin in the range 1510-1525 atomic mass units. R.A., relative abundance.
`
`peaks, each of which contained a peptide of the same
`molecular weight as native guanylin. Since guanylin has four
`cysteine residues, the three fractions of synthetic guanylin
`probably represent the three possible alignments of the
`disulfide bridges. It is likely that the two inactive fractions
`represent improperly folded peptide. The alignment of the
`
`3000
`
`a
`
`2000 -
`
`A
`
`1000 -
`
`20
`10
`Time (min)
`
`30
`
`10 5
`
`1'
`
`
`
`4
`
`10
`
`1'
`
`
`
`3
`
`10
`
`2
`
`10
`
`10
`
`-7 -6
`-8
`-10 -9
`0 10 10 10 10 10
`Guanylin (M)
`
`cGMP (fmol/well)
`
`FIG. 5. Time course (A) and concentration—response (B) effect of
`synthetic guanylin on cyclic GMP levels in T84 cells. In the time
`course experiment, TM cells were treated with 10-8 M guanylin for
`the indicated times. For the concentration—response, the cells were
`incubated with various concentrations of guanylin for 30 min. Cells
`for both experiments were treated with 1 mM IBMX. Values
`represent means ± SE (n = 4).
`
`intestinal guanylate cyclase and suggest that this peptide may
`influence intestinal fluid and electrolyte transport.
`Purification of guanylin was accomplished by capitalizing
`on the stable nature of this peptide, its acidic isoelectric
`point, and its characteristic elution on reverse-phase HPLC.
`Initially, we were concerned that the activity may result from
`bacterial contamination. To limit this possibility the purifi-
`cation was limited to jejunum, which in normal animals is
`considered unlikely to contain considerable bacterial con-
`tamination. Furthermore, we found in preliminary experi-
`ments that every individual rat intestine that we extracted
`possessed bioactivity. We also found that embryonic intes-
`tine, which is considered free of bacteria, exhibits similar
`activity, strongly suggesting that the intestine indeed pos-
`sesses a unique ligand. The structure of guanylin further
`strengthens this proposal. However, definitive proof of the
`intestinal source of guanylin must await a thorough analysis
`of the tissue by immunological and molecular methods.
`The unique structure of the 15-amino acid peptide guanylin
`is characterized by an N-terminal proline, a C-terminal
`cysteine, a glutamic acid, a total of four cysteines, and the
`absence of basic amino acids. The conditions under which
`guanylin was isolated may have given rise to a truncated form
`that possesses the properties required for bioactivity but is
`derived from a larger precursor. The synthetic peptide re-
`quired cyclization for expression of bioactivity and this
`activity was abolished by treatment with the reducing agent
`DTT. Interestingly, during the purification of oxidized syn-
`thetic guanylin, we observed three major chromatographic
`
`GUANYLIN
`
`PNTCE ICAYAACTGC
`
`STa
`
`NT F Y C C ELCCN PA C AGCY
`I
`I
`I
`
`
`FIG. 4. Comparison of the structures of guanylin and STa.
`Identical amino acids are indicated by the dotted lines. The reported
`disulfide alignment for STa (20) is represented by the solid lines.
`
`
`
`Pharmacology: Currie et al.
`
`Proc. Natl. Acad. Sci. USA 89 (1992)
`
`951
`
`-10 -9 -8 -7 -6 -5
`0 10 10 10 10 10 10
`Guanylin (M)
`
`FIG. 6. Displacement of 125I-STa specific binding from T84 cells
`by guanylin. Cells were incubated for 30 min at 37°C with labeled STa
`and various concentrations of guanylin. Specific binding (%) was
`determined by dividing the specifically bound 125I-STa at each
`guanylin concentration by the specifically bound 125I-STa in the
`absence of guanylin. Each point represents the mean of triplicates.
`
`disulfide bridges at this time remains undetermined; how-
`ever, a comparison of the conserved cysteines in guanylin
`with the