`
`Acta Anat 1995;153:210-219
`
`Distribution of Escherichia coli
`Heat-Stable Enterotoxin/Guanylin/
`Uroguanylin Receptors in the
`Avian Intestinal Tract
`
`Abstract
`Pathogenic strains of enteric bacteria secrete small heat-stable toxins (STs) that
`activate membrane guanylyl cyclase receptors found in the intestine. The intes-
`tinal peptide agonists, guanylin and uroguanylin, are structurally related to STs.
`Receptors for l'51-ST were found throughout the entire length of the intestinal
`tract of all the birds examined. These receptors were restricted to intestinal epi-
`thelial cells covering villi and forming intestinal glands and were not observed in
`other strata of the gut wall. The most intense labeling of receptors by '"I-ST
`occurred in the region of the microvillus border of individual enterocytes. There
`appeared to be a decrease in receptor density distally along the length of the
`small intestine. although labeling of receptors by '"I-ST was observed through-
`out the small intestine and colon. Cellular cGMP accumulation responses to
`Escherichia col! ST and rat guanylin in the domestic turkey and duck were
`greater in the proximal small intestine compared to the distal small intestine or
`colon. Brush border membranes (BBM) isolated from the mucosa of proximal
`small intestine of turkeys exhibited agonist-stimulated guanylyl cyclase activity.
`The rank order potency for enzyme activation was E. coil ST> uroguanylin >
`guanylin. Competitive radioligand binding assays using '"I-ST and turkey intes-
`tine BBM revealed a similar rank order affinity for the receptors that was exem-
`plified by the K d values of ST 2.5 nM, uroguanylin 80 nM and guanylin 2.6 pM.
`It may be concluded that functional receptors for the endogenous peptides. gua-
`nylin and uroguanylin, occur in the apical membranes of enterocytes throughout
`the avian intestine. The receptor-guanylyl cyclase(s) of proximal small intestine
`were preferentially activated by uroguanylin relative to guanylin. but both
`endogenous peptides were less potent than their molecular mimic, E. co/i ST.
`
`W.I. Krause'
`K.H. Freeman"
`S.L. Ebert- °
`F.K. Hamra'
`K.F. Foke
`M.G. Currie'
`L.R. Forte"
`
`Departments of
`Pathology and Anatomical Sciences.
`Physiology. and
`Pharmacology. School of Medicine.
`Missouri University. Columbia, Mo.,
`Harry S. Truman Memorial VA
`Medical Center. Columbia. Mo.. and
`Searle Research and Development,
`St. Louis. Mo.. USA
`
`Key Words
`Enterocytes
`Guanylyl cyclase
`Cyclic GMP
`Avian species
`
`Introduction
`
`1-teat-stable enterotoxins (STs) are peptides produced by
`pathogenic strains of bacteria such as Yersinia enterocoh-
`Um or Escherichia co/i that are a major cause of secretory
`
`diarrhea in children as well as in laboratory and domestic
`animals 1Sack et al., 1975: Donta et al., 1977: Burgess et
`al.. 1978; Giannella, 1981; Black et al., 19821. The apical
`plasmalemma (microvillus border) of enterocytes lining the
`intestinal tract and forming the intestinal glands (crypts of
`
`Received:
`November 2%. 1994
`Accepted:
`Ma) 19. 1995
`
`William J. Krause
`Department of Pathology and Anatomical Sciences
`School of Medicine
`University of Missouri
`Columbia, MO 65212 (USA)
`
`(r.:, 1995 S. K tr. r AG. Basel
`
`
`
`Leberkiihn) of man and several other mammalian species
`have been shown to contain specific. high-affinity binding
`sites for these peptides IForte et al.. 1989; Krause et al..
`1990, 19941 . These STs share receptors with two recently
`described endogenous peptides: guanylin and uroguanylin
`'Currie et al., 1992; Hamra et al.. 1993; K ita et al.. 19941.
`The ST/guanylin/uroguanylin receptor has been shown to
`be an intestinal isoform of membrane guanylyl cyclase
`(GC-C) that belongs to a family of cell surface proteins
`which catalyze the production of cyclic 3'.5'guanosine
`monophosphate (cGMP) ISchulz et al., 1990; Forte and
`Currie, 19951. It is by the increase in production of cGMP
`that STs, guanylin or uroguanylin influence cellular func-
`tion. Increased intracellular levels of this second-messenger
`molecule can activate a cAMP-dependent protein kinase
`which stimulates intestinal epithelial cell CI secretion
`'Forte et al.. 1992; Tien et al., 19941. The net effect of bac-
`terial STs, guanylin. and uroguanylin in the mammalian
`intestinal tract is to promote CI secretion and to inhibit Na'
`absorption [Field et al.. 1978; Rao et al.. 1981; Guandalini
`et al, 1982: Currie et al.. 1992: Hamra et al.. 19931. We
`previously reported that 1-51-ST labeled receptors on entero-
`cytes throughout the intestinal tract of man and other mam-
`mats [Krause et al., 19941. Proximal small intestine had
`the greatest apparent density of these receptor-guanylyl
`cyclases that serve as common receptors for bacterial STs
`and
`the endogenous peptide hormones, guanylin and
`uroguanylin.
`Receptors for E. cell ST and ST-stimulated guanylyl
`cyclase activity have been reported to occur in brush border
`membranes isolated from chicken intestine I Katwa and
`White, 19921. The present study examined the intestinal
`tract of several species of birds, to better define the distribu-
`tion of the ST/guanylin receptors in enterocytes along both
`the length and vertical axis (villus/crypt unit) of the avian
`intestinal tract utilizing '''1-ST as a radioligand for these
`receptors and cGMP responses to these peptide agonists in
`vitro IForte et al.. 1988; Krause et al., 1990. 19941. We
`report here that all birds that were examined had ST/gua-
`nylin receptors localized to the enterocytes throughout the
`intestinal tract and that agonist-stimulated cGMP accumu-
`lation responses of intestinal mucosa were greatest in the
`proximal small intestine of turkeys and ducks. Brush-
`border membranes (BBM) isolated from the mucosa of tur-
`key proximal small intestine exhibited guanylyl cyclase
`activity that was stimulated by ST > uroguanylin >guanylin
`in rank order potency. A similar order of relative affinities
`was observed in competitive radioligand binding assays
`using these BBMs suggesting that the receptors were pref-
`erentially activated by ST > uroguanylin> guanylin.
`
`Materials and Methods
`
`Receptor Atamatlio,graphy
`Specimens of small intestine and colon were gathered from a
`variety of avian species (table I ) and frozen in liquid nitrogen as soon
`as possible after death. After freezing. the tissues were stored at
`- 80°C until used. The frozen specimens were sectioned at 14 pm in a
`cryostat maintained at —20°C'. Two cut sections were mounted onto
`opposite ends of gelatin-coated slides. air-dried and stored at —80°C
`until used. Each slide was then incubated with 50 pl of Dulbeccos
`modified Eagle's medium (DMEM). pH 5.5. containing 0.5% bovine
`serum albumin (BSA) at 37°C for 15 min as previously described
`IForte et al.. 1989: Krause et al.. 1990. 19941. To measure the total
`binding for this radioligand. 50 ul DMEM containing I .00(1 cpm '"I-
`ST/p1 was added to one tissue section. The adjacent section was incu-
`bated with the same concentration of radioligand plus I pM unlabeled
`ST to assess nonspecific binding of '"I-ST Additional sections also
`were incubated with 10 pM of rat guanylin to determine if guanylin
`inhibited binding of "I -ST to intestinal receptors. 'I-ST, ,,, was cho-
`sen as the radioligand because iodination of Tyr 9 in guanylin appears
`to interfere with the biological activity of this radioligand resulting in
`poor binding of 1"1-guanylin to receptors on cultured T,„ human intes-
`tinal epithelial cells. For this reason. a radiolabeled form of guanylin
`was not used. Thus. ST peptides currently are the best rad iol igands for
`identifying the tissue location of guanylin/uroguanylin receptors. Fol-
`lowing an incubation of 15 min at 37 °C. the slides were washed with
`a gentle stream of cold phosphate-buffered saline (PBS) and then
`washed 3 additional times by placing them into 50 ml of ice-cold PBS
`for 5 min. The sections were air-dried prior to being coated with
`Kodak NTB-2 or NTB-3 emulsion. dried again. then sealed in light-
`tight boxes and stored at 4°C for 3-4 weeks until developed. Follow-
`ing routine photographic development and fixation, the sections were
`coverslipped and examined by bright and dark field microscopy.
`
`Cyclic GMP Accumulation Bioassay
`Proximal. middle and distal segments of the small intestine as well
`as a segment of midcolon froir domestic turkeys (Nicholas broad-
`breasted strain) and domestic ducks (Peking white strain) were dis-
`sected and mucosa prepared by scraping the intestinal mucosa free.
`and washing the tissue gently once in 0.99k NaCl and twice in DMEM.
`20 mM HEPES. pH 7.4. To measure ST and guanylin stimulation of
`cGMP production, mucosa' suspensions (6(1 mg wet weight) were
`placed in 0.2 ml DMEM (pH 7.4) containing 20 mM HEPES buffer
`( pll 7.41 at 4°C. The tissue was incubated for 40 min at 37°C with
`either I uM E. colt ST. 1(1 pM rat guany I in or vehicle that was added to
`DMEM-HEPES containing I mM isobutylmethylxanthine (IBM X ).
`Perchloric acid was then added to a final concentration of 3.3%. cells
`were centrifuged and the resulting supernatants neutralized with I N
`KOI-l. The supernatant solution was used to measure cGMP concen-
`tration by radioimmunoassay as reported previously IForte et al.,
`19881.
`
`Competitive Rada& gam( and Binding Assays
`BBM were thawed and centrifuged at 16.000 g for 15 min, then
`resuspended in a solution containing 20 mM Tris-HCI. pH 7.5. 15(1
`mM NaCI. and I mM EDTA. The binding assay consisted of 35 pl of
`'"I-ST (50,000- I (X1,000 cpm). 20 pl of peptide
`the same buffer. 20
`ligand in H O. and 25 ill of BBM (15-20 pg protein). Incubation was
`at 37°C for I h. Then 3 ml of cold phosphate-buffered saline (PBS)
`was added and this solution filtered using Whatman GF/F filters
`
`
`
`Table 1. Avian species examined for E. call ST-specific binding
`sites in intestinal epithelial cells
`
`Number of
`animals used'
`
`Species
`
`ST, _,„
`
`3
`I
`
`6
`
`8
`
`1
`1
`1
`3
`
`1
`1
`1
`6
`1
`
`1
`3
`
`Emu IDiemains noraehollandie)
`Blue jay (Cyanocitta :Tishri!! )
`Red-headed woodpecker
`(Mel:mopes erythmeephalus)
`Purple grackle (Quiscalus quiscula)
`Starling (Siurnus I./riga/is)
`Bob-white quail (Colinus virginianus)
`Ring-necked pheasant
`(Phasianus colchicus torquatus)
`Domestic chicken
`(Barred rock/leghorn cross)
`Domestic turkey
`(Nicholas broad breasted)
`Mourning dove (Zenuidura macroura)
`Domestic pigeon (Feral)
`Coot (Fuilea americana)
`Wood duck (Aix sponsa )
`Ring-necked duck (Aythya collaris)
`Northern shoveller (Anus clrpeata)
`Mallard duck (Anus platyrhynchos)
`American green-winged teal (Anus crecca)
`Blue-winged teal (Anus discors)
`Domestic duck (Peking white)
`Giant Canada goose
`airania canadensis maxima/
`Wilson's snipe (Capella gallinago delicata)
`Woodcock ( Philohela minor)
`
`Intensity of silver grains: +++ = strong: ++ = moderate.
`Total number of animals used for both in vitro autoradiography
`and cGMP accumulation bioassays of intestinal mucosa.
`
`(25 mm diameters) and a vacuum manifold. The filters were washed
`twice with 3 nil cold PBS. Each filter had been soaked prior to use with
`0. 19( polyethylenimine and washed with 3 nil PBS before filtering the
`BBM reaction mixture. Each filter was then placed into glass tubes
`and radioactivity measured by gamma scintillation spectrometry. The
`radioligand binding data were analyzed using the Inplot computer
`program to estimate K,, and B„„, values (Graph Pad Software for Sci-
`ence. San Diego. Calif.. USA). The data were fit to a single-site model
`in these experiments.
`
`Isolation of Intestinal BBM
`The proximal one-half of the small intestine from an adult turkey
`was washed in 0.9% NaCI, cut open longitudinally and mucosa
`scraped free with a microscope slide. The mucosa was homogenized
`in 7.5 vol of a buffer containing 300 niM D-mannitol. 5 'JIM EGTA. 12
`mM Tris-(ICI. pl 17.5. per gram of mucosa with a Polytron homoge-
`nizer according to the method of Biber et al. (19811. The isolation of a
`fraction enriched in BBM was carried through the P4 stage of purifi-
`
`cation using a one-precipitation step with 12 mM MgSO,. The BBMs
`were stored frozen at —80°C prior to use. Protein content was mea-
`sured by the method of Bradford 119791.
`
`Measurement of GuanylylCyclase Activity
`Enzyme activity was measured in an assay volume of 100 pl con-
`taining 10 tag BBM protein. 5(1 mM IIEPES. pH 7.6.0.5 mM isohutyl-
`methyl xanthine (IBMX). I mM ATP. 10 mM creatine phosphate.
`5 units creatinc phosphokinase. 1 WO GTP and 5 mM MgCl,. Incu-
`bation was for 15 min at 37 °C. The reaction was stopped by adding
`100 p1 of 6% perchloric acid. Then each reaction mixture was neutral-
`ized with 10 N KOH and centrifuged to remove the potassium per-
`chlorate precipitate. A 25-pi portion was removed to use for (he
`estimation of cGMP by radioimmunoassay as previously described
`[Forte et al.. 19881. The data are expressed as pmol cGMP formed per
`ug protein per IS min. Agonists or vehicle were added at the indicated
`concentration and assayed in duplicate.
`
`Preparation of Intestinal Extracts and Bioassay
`About 200 g of mucosa isolated from the small intestine and colon
`of turkeys was suspended in 2 liters of I M acetic acid. heated at
`100°C for 1(1 min and then homogenized with a Polytron as previ-
`ously described for opossum intestinal mucosa 'Hamra et al.. 19931.
`The homogenate was centrifuged at 10.000g for 2(1 min and the super-
`natant was made in 0.1% trilluoroacetic acid (TFA). Extracts were
`isolated using C18 cartridges as previously described 'Hamra et al..
`19931. Bioactive peptides that eluted with 40% acetonitrile and 0.1%
`TFA in H.O were chromatographed on a 2.5 x 90 cm Sephadex G-25
`column as previously described. The bioassay for guanylin/uro-
`guany I in-like peptides was carried out by removing O.5 ml from each
`10-ml column fraction, drying in a Speed-Vac and resuspending each
`fraction in 200 pl of DMEM. This sample was added to one well of a
`24-well culture plate containing confluent T„,, cells. Incubation was
`for 4(1 min at 37 °C and cellular cGMP was measured by RI A as previ-
`ously described 1Forte et al.. 1984: Hamra et al.. 19931. The conditions
`cells were as previously described 'Hamra et al..
`for culture of
`19931.
`
`Synthesis of E. coli ST 5-17 and Guanylin
`Rat guanylin (PNTCEICAYAACTGC) and E. coli ST5- 17
`(CCELCCNPACAGC) were synthesized by the solid-phase method
`as previously described 'Currie et al.. 19921.
`
`Preparation of '"I-ST
`The iodination of E. call ST ( NSSNYCCELCCNPACTGCY.
`Multiple Peptide Systems. San Diego. Calif.. USA) was carried out
`using the lactoperoxidase procedure described previously 'Forte et al.,
`1988. 19891. Purification of "'I-ST was achieved using high-per-
`formance liquid chromatography with a C I 8 column under reverse-
`phase conditions also as described earlier (Forte et al.. 1988, 19891.
`Na'"I was purchased from DuPont NEN. Wilmington. Del.. USA. as
`the carrier-free radionuclide (14-17 pCi/pg). Lactoperoxidase was
`purchased in a solid-state form from Bionic! Laboratories. Richmond,
`Calif., USA. E. coli ST. GTP. ATP. creatine phosphate. and creatine
`phosphokinase were purchased from the Sigma Chemical Company.
`St. Louis. Mo.. USA. Other reagents and materials were obtained from
`various suppliers.
`
`212
`
`Krause/Freernan/Eber/Hanira/Fok/Currie/Forte
`
`ST/Guanylin Receptors in Avian Intestine
`
`
`
`• -•
`
`littoctxge
`
`•
`
`fy
`
`•
`
`:w'
`
`•
`
`.
`Nap, '
`
`• ilte,L,
`
`:e
`
`r
`kfof,•••
`
`Rci
`
`Fig. 1. A portion of intestinal mucosa from the proximal small intestine of an emu. The enterocytes covering villi
`(V) and forming the intestinal glands (1) express ST/guany lin receptors. Elements of the lamina propria and muscularis
`mucosae (M) show little if any receptor labeling. Note the intense labeling in the region of the microvillus border
`(arrows) by '"I-ST. Dark field. x 297.
`Fig. 2. A region of small intestinal wall from the red-headed woodpecker illustrating the even distribution of recep-
`tors labeled by '-'I-ST in enterocytes of the intestinal mucosa. The lumen of the small intestine (L) is shown at the
`extreme right: the muscularis externa (M) at the far left. Dark field. x 1 19.
`Fig. 3. A portion of small intestine from the purple grackle also shows a relatively even distribution of ' `I-ST-
`labeled receptors among enterocytes (arrows) of the intestinal mucosa. The lumen (L) is shown near the top: the mus-
`cularis externa ( M) near the bottom of the photomicrograph. Dark field. x119.
`Fig. 4. "I -ST-receptor density in the intestinal mucosa of the blue-jay, unlike the red-headed woodpecker and pur-
`ple grackle. appears greater in enterocytes covering villi (V ) than in enterocytes forming intestinal glands (1). The intes-
`tinal lumen (L) is oriented to the right: the muscularis externa ( M ) to the left of the photomicrograph. Dark field. x 1 19.
`Fig. 5. L'I-ST-receptor density in enterocytes lining the small intestine of a domestic chicken. The intestinal lumen
`(L) is to the right; the muscularis externa (M) is to the left. Dark field. x 1 19.
`Fig. 6. A segment of colon taken from a mallard duck illustrates greater '"1-ST-recepior density in enterocytes lin-
`ing the lumina' surface. Note that the most intense labeling occurs in the microvillus border (arrows). Dark field. x 1 19.
`
`213
`
`
`
`500
`
`400 -
`
`al
`
`300
`
`Can 200 -
`
`100
`
`a
`
`▪ Control
`▪
`
`ST 1
`
`Guanylin 10 µM
`
`50 -
`
`40
`
`Od
`
`e
`
`20 -
`
`Proximal
`
`Middle
`
`Distal
`
`Colon
`
`Control
`
`El ST uM
`
`El Guanylin 10 µI44
`
`11
`
`Middle
`
`Distal
`
`Colon
`
`Fig. 7. cGMP production by the intestinal mucosa taken from dif-
`ferent segments of turkey small intestine and colon following stimula-
`tion by I uM E. coil ST and 10 pM rat guanylin. The data are the mean
`of three experiments. Small intestine — proximal, middle and distal
`segments.
`
`Fig. 8. cGMP production by the intestinal mucosa taken from dif-
`ferent segments of Peking duck small intestine and colon following
`stimulation by I uM coli sT and 10 uM rat guanylin. The data are
`the mean of two experiments. Small intestine — proximal. middle and
`distal segments.
`
`Results
`
`Distribution of STIGuanylin Receptors
`Examination of the distribution of receptors labeled with
`'2'I-ST from various segments of the small intestine and
`colon of a number of avian species showed that hie,h-affin-
`ity '"I-ST-binding sites were present and confined to the in-
`testinal epithelium throughout the length of the intestinal
`tract (table I: fig. I —6). '"I-ST binding sites were not ob-
`served in other layers of the gut wall (lamina propria. muscu-
`laris mucosae, submucosa, muscularis external or serosa).
`The greatest intensity of '"I-ST binding occurred along the
`microvillus (striated) border of intestinal epithelial cells.
`Adjacent sections that were incubated with the same concen-
`tration of radioligand plus I pM of either unlabeled ST or I()
`pM of rat guanylin effectively inhibited'-`I-ST binding sug-
`gesting that common receptors were labeled by 9 -ST.
`Along the vertical axis (the villus/crypt unit) of the small
`intestine in the birds that were examined. '"1-ST binding to
`receptors appeared to he evenly distributed among intesti-
`nal epithelial cells (enterocytes) covering villi and forming
`the intestinal glands (fig.2. 3. 5). In some birds, such as the
`blue jay. receptor density appeared greatest in enterocytes
`covering villi and lining the intestinal lumen (fig.4). Recep-
`tor density of enterocytes lining the surface of the colon
`in the mallard duck (fig.6) and Canada goose (data not
`shown) also appeared to be greater than in those entero-
`
`cytes forming the intestinal glands. Receptor density along
`the longitudinal axis of the small intestine following in
`vitro receptor autoradiography appeared to be greater in the
`proximal region in most species.
`To evaluate further the possibility that ST/guanylin re-
`ceptors are more abundant in proximal small intestine, dif-
`ferent segments of the intestinal tract of domestic turkeys
`and ducks were used to measure guanylin and E. coli
`ST-stimulated guanylyl cyclase activity. Agonist-mediated
`activation of the membrane guanylyl cyclase was measured
`by the increased cGMP content of mucosa exposed to these
`peptides. Treatment of the intestinal mucosa with I pM ST
`elicited a large increase in cGMP levels in the proximal
`small intestine of turkeys with a much reduced cGMP accu-
`mulation response to ST occurring in the distal small intes-
`tine and colon (fig.7). ST stimulated cGMP production by
`the intestinal mucosa to much higher levels than did 10 pM
`guanylin in all segments. Likewise. cGMP accumulation
`responses to ST in the intestinal mucosa of the domestic
`(Peking) duck were much greater in the proximal small
`intestine as compared to the middle or distal small intestine
`and colon (fig. 8). In comparison to ST. 10 pM guanylin
`was considerably less effective in stimulating cGMP accu-
`mulation in the mucosa of small intestine or that of the
`colon in ducks. The magnitude of cGMP responses in prox-
`imal intestine of turkeys was substantially greater than that
`observed in ducks.
`
`214
`
`Krause/Freeman/Eberalamra/Fok/Currie/Forte
`
`ST/Guanylin Receptors in Avian Intestine
`
`
`
`■ ST
`fl Uroguanytin
`■ Guanylin
`
`30
`
`20 ti
`
`E
`
`0 10
`
`0
`
`1
`
`0
`
`■-1
`
`100 -
`
`75 -
`
`0
`
`
`
`g 50 -
`
`,op
`co
`t 25 -
`
`•
`
`• ST
`Uroguanylin
`• Guanylin
`
`8
`
`6
`7
`Peptide, -log M
`
`5
`
`4
`
`0
`
`0
`
`F - 1
`
`11
`
`10
`
`8
`7
`9
`Peptide, -log M
`
`5
`
`4
`
`Fig. 9. Guanylyl cyclase activity of brush border membranes iso-
`lated from turkey small intestine. The data shown in this figure are
`representative of two such experiments with each point assayed in
`opossum uroguanylin and rat
`duplicate. Synthetic peptides. ST,
`guanylin were used in these experiments.
`
`Fig. 10. Comparison of the relative affinities of ST. uroguanylin
`and guanylin for receptors in turkey intestine. These data are the mean
`of two experiments with each point assayed in triplicate for each
`experiment. The radioligand was '"I-ST (E. coli STD). Bo = Total
`bound'-`I-ST in the absence of competing ligand.
`
`To characterize further the properties of the guanylyl
`cyclase receptors in turkey intestine, isolated BBM from
`the mucosa of proximal small intestine were used. Agonist-
`stimulated guanylyl cyclase activities were measured to
`compare the relative potencies and efficacies of E. co/i ST,
`rat guanylin and opossum uroguanylin. These peptides
`exhibited a rank order of potencies with ST> uroguany-
`lin >guanylin (fig. 9). A similar rank order of agonist
`potencies has been reported for the human GC-C isoform
`that is expressed in T„ colon carcinoma cells 'Hamra et al.,
`1993: Kita et al., 19941. Thus, turkey small intestine, like
`the human receptor-guanylyl cyclase, has a BBM receptor-
`guanylyl cyclase that appears to prefer uroguanylin relative
`to guanylin. It should be noted that ST is considerably more
`potent than either uroguanylin or guanylin. Our supplies of
`these peptide agonists were insufficient to use concentra-
`tions higher than 10-30 jiM in this assay so that maximal
`stimulation of the guanylyl cyclase was not achieved in
`these experiments.
`Evaluation of the pharmacological properties of the
`BBM receptors was extended using competitive radioli-
`gand binding assays to measure the affinities of the recep-
`tors for E. coil ST, opossum uroguanylin and rat guanylin
`(fig. 10). All three peptides inhibited the binding of '"I-ST
`to an apparently common set of binding sites (B„,„ , - 6.13
`pmol/mg protein) on turkey intestine BBM. The K, values
`for these peptides were: ST 2.5 nM, uroguanylin 80 nM and
`
`guanylin 2.6 uM. Thus, ST had a 40-fold higher apparent
`affinity for these receptors than did uroguanylin, which had
`an approximate 32-fold higher affinity than guanylin.
`These data are consistent with the relative potencies of
`ST > uroguanylin > guanylin as activators of the BBM
`guanylyl cyclase.
`Guanylin and uroguanylin peptides are found in the
`mammalian intestine [Currie et al., 1992: Hamra et al.,
`1993, 19951. To examine whether these bioactive peptides
`also occur in avian intestine, we prepared an extract of tur-
`key intestinal mucosa and subjected this extract to gel fil-
`tration chromatography (fig. 1 1). A broad peak of bioactiv-
`ity eluted in the internal volume of this Sephadex G-25
`column. These fractions activated the human T„ cell intes-
`tinal guanylyl cyclase (GC-C) that was used for the bio-
`assay of turkey intestinal agonists. The dominant peak of
`bioactive fractions eluted before either uroguanylin or gua-
`nylin, suggesting that these putative peptides from turkey
`intestine may he longer forms of guanylin and/or uroguany-
`lin or have different structures from the peptides isolated
`from mammalian intestine or urine 'Currie et al.. 1992;
`Hamra et al., 1993, 19951. Therefore, intestinal mucosa of
`turkeys contains bioactive agonists that stimulate cGMP
`production in human intestinal cells. It is likely that these
`substances are peptides that are structurally similar to uro-
`guanylin. guanylin and ST peptides, the known agonists for
`this class of receptors.
`
`215
`
`
`
`Sephadex G-25
`
`2.5 -
`
`= 2.0 -
`
`E ° 1.5 -
`
`0.5 -
`
`20
`
`25
`Fraction
`
`30
`
`35
`
`Fig.11. Guanylin/uroguanylin-like activity isolated from the
`intestinal mucosa of turkeys. The bioassay is the cGMP accumula-
`tion response of culture T. cells to 2(X) ul of each fraction from a
`2.5 x 90 cm Sephadex G-25 column. vv = Void volume.
`
`Discussion
`
`Receptors for E. coil ST had previously been identified
`in purified BBM isolated from small intestine of the
`chicken IKatwa and White. 19921. The current study
`that
`the
`extends those observations and demonstrates
`ST/guanylin/uroguanylin receptor is localized to entero-
`cytes of the chicken intestine as well as those cells lining
`the intestinal tract of a number of other avian species. Like
`man and other mammalian species examined to date, these
`receptors are localized to the enterocytes lining the intesti-
`nal lumen and forming the intestinal glands. As in the avian
`species investigated in the present study. the greatest recep-
`tor density in enterocytes of mammals occurs in the region
`of the microvillus border [Krause et al.. 1990. 19941. There
`is an apparent decrease in receptor density distally along
`the longitudinal axis of the small intestine of some avian
`species as demonstrated by in vitro receptor autoradio-
`graphy. A similar decrease has been observed in man and
`other mammalian species using this methodology [Krause
`et al.. 19941. E. coil ST/guanylin-stimulated guanylyl cy-
`clase activity, as indicated by an increased accumulation of
`cellular cGMP. also suggests that the receptor-guanylyl
`cyclase density is greatest in the proximal small intestine as
`compared to distal regions of small intestine or colon in the
`turkey and duck. Similar observations have been made
`using these techniques which showed decreasing gradients
`of receptor density along the longitudinal axis of the in-
`
`testinal tract of some mammalian species 1Krause et al.,
`19941. Although substantial interspecies variation may oc-
`cur in birds and mammals. a general conclusion can be
`drawn from these experiments indicating that proximal
`small intestine has the highest levels of ST/guanylin/uro-
`guanylin receptor-guanylyl cyclases relative to other re-
`gions of small or large intestine.
`With regard to the vertical axis (the villus/crypt unit)
`of the avian small intestine. ST/guanylin receptor density
`appeared relatively evenly distributed among enterocytes
`covering villi and in those enterocytes forming intestinal
`glands of the majority of birds that were examined. This
`observation is in contrast to that found in the human small
`intestine as well as several other mammalian species in
`which the receptor density is greatest in enterocytes cover-
`ing the basal one-half of villi and in those lining the proxi-
`mal one-half of the intestinal glands (crypts of Liberkiihn)
`'Cohen et al., 1992; Li and Goy, 1993: Krause et al., 1994J.
`The ST/guanylin/uroguanylin receptor belongs to the
`guanylyl cyclase family of proteins that differ markedly in
`their selectivity and activation by ligands. Included in this
`group are atrial natriuretic peptides and nitric oxide which
`are endogenous activators for two particulate forms (GC-A
`and GC-B) and a soluble (GC-S) form of guanylyl cyclase,
`respectively IDrewett and Garbers, 19941. All these en-
`zymes catalyze the production of cGMP and by this mech-
`anism influence cellular function. The ST/guanylin/uro-
`guanylin receptor, an
`intestinal cytoskeletal-associated
`form of membrane guanylyl cyclase (GC-C) is selectively
`activated by the STs 'Field et al.. 1978; Guerrant et al..
`1980; Schulz et al.. 1990; Forte and Currie. 19951 as well as
`by guanylin [Currie et al., 19921 and the newly discovered
`peptide. uroguanylin [Hamra et al.. 1993; K ita et al., 19941.
`Guanylin and uroguanylin have 15 or V) amino acids, are
`produced in the intestine and appear to serve as endogenous
`regulators for the intestinal form of guanylyl cyclase. GC-
`C. Enterotoxigenic strains of bacteria cause secretory (i.e.
`travelers) diarrhea by producing molecular mimics of guany-
`lin and/or uroguanylin: thereby activating the GC-C recep-
`tors in enterocytes. Guanylin, uroguanylin and E. cell STs
`bind to the N-terminal, extracellular domain of this cell sur-
`face receptor and activate a C-terminal, intracellular catalytic
`domain causing increased cellular levels of cGMR. Although
`the physiological role of guanylin and uroguanylin in the
`avian intestinal tract is unknown at present, both peptides
`stimulate chloride secretion in enterocytes of mammalian
`species 'Currie et al.. 1992; Forte et al.. 1993; Hamra et al..
`1993: Cuthbert et al.. 1994: Kita et al.. 19941. Thus. guany-
`lin and uroguanylin may regulate ion transport and Iluid
`secretion in both the mammalian and avian intestinal tract by
`
`to
`
`Krause/Freeman/F.ber/Hantra/Fok/Currie/Forte
`
`ST/Guanylin Receptors in Avian Intestine
`
`,'J
`
`
`
`this signaling mechanism. Other epithelial tissues may also
`he targets for these peptides. Receptor-guanylyl cyclase re-
`sponsiveness to ST occurs in kidney. liver and testes [Forte
`et al.. 1988. 1989; Krause et al.. 1990: Laney et al.. 19921.
`Thus. in addition to their intestinal paracrinc function, these
`peptides may also function as hormones.
`Receptors for these peptides have now been demon-
`strated in both mammals and birds suggesting that the reg-
`ulation of enterocyte function by guanylin and uroguanylin
`through the cGMP second messenger mechanism is a fun-
`damental signaling pathway that appeared early in verte-
`brate evolution. It is of interest that intense receptor label-
`ing by '"I-ST occurs in the distal small intestine and the
`colon of some birds, yet these regions of the gut had very
`small cGMP responses to either ST or guanylin. The expla-
`nation for the receptors in the colon and distal small intes-
`tine of these species being less responsive to these agonists
`is unclear, particularly in light of the fact that guanylin
`mRNA levels are most abundant in the mammalian colon
`Wiegand et al.. I 992a, h: Li and Goy. 19931. Our observa-
`tion that '- I-ST-labeled receptors appear to he abundant in
`the avian colon. which in the turkey and duck had little or
`no cGMP responses to either ST or guanylin, is consistent
`with the possibility that ST receptors (i.e. binding proteins)
`exist in the avian distal small intestine and colon. which are
`not guanylyl cyclases. A similar conclusion was made from
`our recent studies of these receptors in the mammalian
`colon [Krause et al.. 19941. Similarly. cultured IEC-6 intes-
`tinal cells have been shown to have specific. high affinity
`binding sites for 'I -ST, but these cells express no GC-C
`mRNA and are devoid of ST-stimulated guanylyl cyclase
`activity I Mann et al.. 19931. A 56-kD protein, isolated from
`the small intestine of rats, bound 12. I-Si but exhibited no
`guanylyl cyclase activity II lakki et al.. 19931. This protein
`may be a proteolytic fragment of GC-C with the C-terminal
`catalytic portion of the receptor missing. However. these
`preliminary findings suggest that other receptor proteins
`may occur in the distal small intestine and/or colon of both
`birds and mammals, which hind '"I-ST with high affinity,
`but have no guanylyl cyclase activity. A similar cell-surface
`protein occurs which hinds atrial peptides with high affinity
`but lacks the intracellular catalytic domain found in the