Introduction

Top Abstract Introduction Methods Results Discussion References

PYY and its neural homolog NPY are located in gut endocrine cells and neurons, respectively. The distribution of PYY-containing L-cells is primarily in ileum and colon, and direct release of PYY coincides with increased amounts of unabsorbed fat and, to a lesser extent, complex carbohydrate reaching the distal small bowel (Walsh, 1994). Immunohistochemical studies have shown that these endocrine cells have long cytoplasmic processes that extend to neighboring cells, suggesting that PYY may have both an endocrine and paracrine function (Adrian et al., 1985). PYY has recently been shown to have a physiological role in enhancing postprandial small bowel absorption of water and electrolytes in both canine and human species (Bilchik et al., 1993, 1994). Furthermore, PYY has been shown to inhibit PGE₂- and VIP-induced intestinal secretion (Okuno et al., 1992; Playford et al., 1990; Saria et al., 1985).

Because of the powerful (picomolar) effects of PYY on intestinal electrolyte transport, we hypothesized that PYY would antagonize the altered intestinal absorption observed in piglet cryptosporidiosis; a complex malabsorption/secretory diarrheal disease that primarily affects the ileum (Argenzio et al., 1990). Although this disease is characterized by extensive villous atrophy and impaired glucose-coupled Na⁺ absorption, the neutral NaCl absorptive mechanism remains largely intact. However, this neutral mechanism is strongly inhibited, and active anion secretion is enhanced, by increased PG production in this infection (Argenzio et al., 1993). Furthermore, both PGE₂, which alters epithelial electrolyte transport directly, and PGI₂, which indirectly alters epithelial transport via stimulation of enteric nerves, are elevated at the peak of infection (Argenzio et al., 1993, 1996). The present study was conducted to determine whether (1) PYY could correct the altered NaCl transport in the infected tissue and (2) such a PYY effect was mediated through inhibition of PG and/or neural pathways.

	Methods

Top Abstract Introduction Methods Results Discussion References

Experimental animals were 1-day-old cross-bred pigs obtained from the College of Agriculture at North Carolina State University. All procedures were approved by the University Animal Care and Use Committee. Piglets were placed in one of two isolation units and fed a synthetic liquid diet by an automated delivery system.

Cryptosporidium parvum oocysts were obtained and treated as described previously(Argenzio et al., 1990). An inoculum of 5 × 10⁸ oocysts was administered to the piglets by orogastric tube on day 3 of life. Control or infected piglets were studied on days 3 to 4 after inoculation, a time shown previously to be at the peak of infection (Argenzio et al., 1990). Piglets were killed by intracardiac sodium pentobarbital, and 20-cm sections of ileum beginning 10 cm above the ileocecal junction were taken for in vitro transport studies and histological analysis. Formalin-fixed segments were embedded in paraffin, cut into 7-um-thick sections, and stained with hematoxylin and eosin for examination by light microscopy. All infected pigs in the study showed evidence of villous atrophy and oocysts adherent to the villus, whereas control piglets showed normal villous architecture with no evidence of infection. In addition, histological analysis of tissues stripped of their muscle layers for mounting in Ussing chambers demonstrated the presence of an intact submucosal plexus situated just beneath the muscularis mucosa.

Methods used in this laboratory for in vitro Ussing chamber studies have been described in detail (Argenzio et al., 1993). Briefly, a piece of ileum was removed, opened along the antimesenteric border and pinned flat in a tray containing oxygenated Ringer's solution or Ringer's containing 10⁶ M indomethacin (to inhibit endogenous PG production). The external muscle layers were removed by blunt dissection, and the tissue was mounted in Ussing chambers with an aperture of 1.13 cm². Tissues were bathed on both surfaces with 10 ml of Ringer and oxygenated with 95% O₂/5% CO₂ at 39°C. Tissues stripped in indomethacin were also maintained in 10⁶ M indomethacin in the mucosal and serosal chamber baths. The basic Ringer's solution contained (mmol/l): Na⁺ 142, K⁺ 5, Ca⁺⁺ 1.25, Mg⁺⁺ 1.1, Cl 124, HCO₃ 25, HPO₄⁼ 1.65 and H₂PO₄ 0.3. In all studies, 10 mM serosal glucose was balanced with 10 mM mucosal mannitol.

The spontaneous PD was measured with Ringer-agar bridges connected to matched calomel electrodes and was short circuited with Ag-AgCl electrodes, using an automated voltage-clamp that corrected for fluid resistance. Tissues were maintained in the short-circuited state, except for brief intervals to record the open-circuit PD. G was calculated from the spontaneous PD and I_sc using Ohm's law. In ion flux studies, I_sc was recorded at 10-min intervals, averaged for the 30-min period and converted to µEq/cm² using the Faraday constant.

In NaCl flux studies, tissues bathed in Ringer's, 10⁶ M serosal plus mucosal indomethacin-Ringer's or 10⁷ M serosal TTX were given a 30-min period to equilibrate; then 2 µCi each of the isotopes ²²Na⁺ and ³⁶Cl were added to the mucosal or serosal solutions bathing paired tissues, which were matched according to their conductance. After allowing 20 min for isotopic equilibration, standards and zero time samples were removed from the reservoirs. Samples then were removed at the end of 30 min (flux period 1), and a given treatment (PYY, 10⁷ M; carbacyclin, 10⁶ M; PGE₂, 10⁶ M or a combination and simultaneous addition of PYY and a PG) was added to the serosal bath. After an additional 20-min equilibration period, a second zero time sample followed by a second 30-min flux period (flux period 2) was obtained.

Samples were counted for ²²Na⁺ in a crystal scintillation counter and for ³⁶Cl in a liquid scintillation counter. The contribution of the ²²Na⁺ counts to the Cl channel was determined and subtracted from the total counts. Unidirectional Na⁺ and Cl fluxes from mucosa to serosa (J_ms) and from serosa to mucosa (J_sm) were calculated using standard equations (Schultz and Zalusky, 1964).

In dose-response studies, PYY at concentrations ranging from 5 × 10¹⁰ to 10⁷ M was added to the serosal side of tissues 10 min after the serosal addition of 10⁶ M carbacyclin, and the inhibition of the carbacyclin-stimulated I_sc was examined. In other experiments, tissues were pretreated with 10⁷ M PYY for 20 min, and the maximal I_sc responses to 10⁶ M carbacyclin, 10⁶ M PGE₂ or 10⁷ M VIP were recorded.

Unpaired statistical comparisons were made between litter-matched control and infected piglets studied on the same day and for the same flux interval. Paired comparisons within pigs were obtained using the first and second flux interval in the same tissue. This approach was found to be more reliable than comparing two different tissues for the same flux interval. Previous studies had shown no statistical differences in time controls run for the two flux periods consecutively (Argenzio et al., 1994). Results are reported as mean ± S.E.M. of n animals, and statistical comparisons were performed with Student's t test for paired or unpaired observations as appropriate.

The PGI₂ analog carbacyclin, PGE₂, VIP and PYY were obtained from Sigma Chemical (St. Louis, MO).

	Results

Top Abstract Introduction Methods Results Discussion References

Effect of PYY on basal ion transport. The effect of PYY on basal fluxes of Na⁺ and Cl, I_sc and G in control and infected piglet ileum bathed in normal Ringer's is shown in figure 1 and table 1. The infection reduced net Na⁺ and Cl absorption and G but increased I_sc. The addition of PYY to control tissue had no significant effect on unidirectional or net fluxes of Na⁺ and Cl but modestly reduced I_sc. The latter reflects Cl and HCO₃ secretion in this tissue (Argenzio et al., 1993, 1994). Although PYY did not significantly affect Na⁺ fluxes in the infected tissue, net Cl secretion was abolished. This was primarily a result of a decrease in the secretory (J_sm) Cl flux; J_ms was not significantly affected. Furthermore, the I_sc was strongly inhibited by the addition of PYY to the infected tissue.

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Effect of PYY on Isc in control and infected piglet ileum showing the standard protocol for obtaining two ion flux periods (see table 1). Tissues were bathed in normal Ringer's and allowed 30 min to equilibrate. Then, the isotopes 22Na+ and 36Cl were added; 20 min later, a control flux period was obtained, followed by 107 M PYY at 85 min and a second PYY flux period from 105 to 135 min (n = 11 control and 8 infected pigs ± S.E.M.).

View this table: [in this window] [in a new window]   TABLE 1 Effect of PYY on basal ion transport in control and infected ileum (107 M) PYY was added to tissues 20 min before second flux period (see fig. 1).

Effect of PYY in indomethacin- and TTX-blocked tissue. Previous studies with the Cryptosporidium-infected piglet (Argenzio et al., 1993, 1996) showed that inhibition of PG synthesis with indomethacin or inhibition of enteric neural activity with TTX completely or partially restored net NaCl fluxes to normal, respectively. Therefore, to determine whether the antisecretory effects of PYY in the infected ileum noted above were related to PG or neural pathways, infected tissues were studied in the presence of indomethacin or TTX. In contrast to infected tissue bathed in normal Ringer's, the addition of PYY to infected tissue pretreated with either indomethacin or TTX had no significant effect on unidirectional or net fluxes of Na⁺ or Cl or I_sc (fig. 2) or G (not shown). As expected, PYY also had no effect in control tissue pretreated with these agents (data not shown).

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Effect of PYY on indomethacin (INDO)- or TTX-treated infected tissues. Tissues were bathed in 106 M indomethacin in both mucosal and serosal baths or 107 M TTX in the serosal bath during the first flux period; then, PYY (107 M) was added to the serosal bath for the second flux period. PYY had no significant (paired t statistic) effect on unidirectional (Jms or Jsm) or net fluxes or on Isc in tissues treated with indomethacin (n = 5 piglets) or TTX (n = 4 piglets).

Effect of PYY on prostacyclin-induced alterations in ion transport. Recently, we showed that both the neural-modulator prostacyclin and PGE₂ were elevated in infected tissue and could account for the altered Na⁺ and Cl fluxes (Argenzio et al., 1993, 1996). To determine whether PYY interfered with prostacyclin-mediated alterations, we performed dose-response studies with PYY after the addition of the prostacyclin analog carbacyclin to control tissue. Tissues were stripped and bathed in indomethacin to eliminate endogenous PG generation because endogenous PGs partially desensitize this tissue to exogenous PG action (Argenzio et al., 1993). As shown in figure 3, the addition of 10⁶ M carbacyclin to the serosal bath resulted in a prompt and biphasic increase in I_sc. The addition of PYY in concentrations of 5 × 10⁹ M abolished the carbacyclin-mediated increase in the I_sc, whereas 5 × 10¹⁰ M PYY had no effect (data not shown for 10⁸ and 10⁷ M PYY).

View larger version (25K): [in this window] [in a new window]   Fig. 3.   Effect of PYY on the Isc response to carbacyclin. Carbacyclin was added to indomethacin-treated control tissues at 30 min; then, PYY at concentrations of 0 to 107 M were added at 40 min. Concentrations of PYY of >5 × 1010 M were effective in abolishing the Isc response (results only shown for 5 × 109 M; however, 108 M and 107 M gave essentially identical results; n = 4 or 5 piglets ± S.E.M.).

7

View this table: [in this window] [in a new window]   TABLE 2 Effect of PYY on carbacyclin-treated control and infected piglet ileum Carbacyclin (106 M) or carbacyclin (106 M) plus PYY (107 M) were added 20 min before second flux period.

Effect of PYY on cAMP-mediated secretory events. Studies with rat small intestinal epithelial cells have shown that PYY mediates its absorptive effects via receptors that are negatively coupled to the cAMP production system (Voisin et al., 1990). Because PGE₂ mediates its effects via cAMP, we examined this possible PYY mode of action in control piglet ileum bathed in indomethacin-Ringer's. Figure 4 shows that PYY pretreatment had no effect on the increased I_sc mediated by PGE₂, whereas it totally abolished I_sc induced by carbacyclin. A similar lack of effect of PYY on another cAMP agonist, VIP, is also shown in figure 4. These I_sc responses to VIP and PGE₂ are associated with an inhibition of net NaCl absorption and anion (Cl and HCO₃) secretion in this tissue (Argenzio et al., 1993).²

View larger version (36K): [in this window] [in a new window]   Fig. 4.   Isc response to carbacyclin (106 M), PGE2 (106 M) and VIP (107 M) in the presence or absence of 107 M PYY. Agonists were placed on the serosal side of indomethacin-treated control tissue. Paired tissues were pretreated for 20 min with serosal PYY. The maximum change in Isc in response to agonists was recorded (n = 6 piglets ± S.E.M.). *P < .05 (paired t statistic).

View this table: [in this window] [in a new window]   TABLE 3 Effect of PYY on PGE2 -treated control and infected ileum PGE2 (106 M) or PGE2 (106 M) plus PYY (107 M) were added 20 min before second flux period.

	Discussion

Top Abstract Introduction Methods Results Discussion References

The present results indicate that PYY interrupts the Cl secretory process in the Cryptosporidium-infected ileum. This action of PYY was nullified with the PG synthesis inhibitor indomethacin and with the nerve conduction blocker TTX, suggesting that PYY operates through both PG and neural pathways. Furthermore, PYY inhibited the effects of the PGI₂ analog carbacyclin, which blocks neutral NaCl absorption and elicits net Cl secretion in this tissue. In contrast, PYY had no inhibitory effect on PGE₂, which also blocks neutral NaCl absorption and elicits net Cl or HCO₃ secretion in piglet ileum (Argenzio et al., 1993). Thus, the inhibitory effects of PYY on PG action are selective, suggesting an intermediate level of control proximal to the transport mechanisms.

Although both prostanoids, PGE₂ and carbacyclin, are capable of inhibiting Na⁺ and Cl absorptive fluxes and stimulating I_sc with equal potency, carbacyclin clearly increases the secretory Cl flux, whereas PGE₂ has no significant effect on unidirectional Cl secretion (e.g., tables 2 and 3, respectively). These differing effects on Cl J_sm cannot be explained by changes in conductance because the transepithelial conductance is diminished equally with both agents. These results confirm earlier studies of PGE₂ and carbacyclin on unidirectional Cl fluxes in infected piglet ileum (Argenzio et al., 1993, 1996). A similar lack of effect on J_sm by 8-bromo-cAMP has also been demonstrated in pig ileum by Brown et al.(1990). The I_sc response to PGE₂ thus likely represents a Cl-dependent HCO₃ secretion (Minhas et al., 1993) that is resistant to PYY, whereas the I_sc response to carbacyclin may represent a Cl secretory pathway that is inhibited by PYY (i.e., the increased I_sc in the presence of carbacyclin is roughly equivalent to the increased J_sm and net Cl secretion). The inhibition of Cl J_sm by PYY in infected ileum bathed in normal Ringer's is thus consistent with a PYY inhibition of prostacyclin-stimulated pathways.

In the present study, both the antiabsorptive and secretory effects of carbacyclin were interrupted by PYY in indomethacin-treated control or infected tissue (i.e., Na⁺ and Cl absorptive fluxes were increased and Cl secretion was decreased by PYY). However, in the Cryptosporidium-infected tissue bathed in normal Ringer's, PYY selectively inhibited the secretory Cl flux; no stimulation by PYY of the neutral NaCl absorptive mechanism was detectable. These differing effects of PYY in normal and indomethacin-Ringer's may be explained by the fact that the PYY-resistant PGE₂ is also elevated in infected tissue bathed in normal Ringer's and is capable of directly inhibiting neutral NaCl absorption in villous epithelium.

Pathways controlling the transport effects of PGE₂ and carbacyclin also differ in this tissue. The effects of carbacyclin are abolished by the nerve conduction blocker TTX, whereas TTX has no effect on the PGE₂ secretory (I_sc) response (Argenzio et al., 1996). Likewise, carbacyclin effects are also partially or totally inhibited by atropine, hexamethonium, VIP receptor antagonists and alpha-2 adrenergic agonists, whereas these agents have no effect on PGE₂-induced I_sc (Argenzio et al., 1996). Therefore, it is likely that PGE₂ exerts its effects on epithelial cell receptors, whereas PGI₂ interacts with enteric neurons, as has also been demonstrated in rat, rabbit and guinea pig colon (Bern et al., 1989; Diener et al., 1988; Frieling et al., 1995). Accordingly, the inhibitory actions of PYY on carbacyclin-mediated events are more than likely related to modulation of carbacyclin-stimulated neural pathways.

Our results suggesting a neural site of PYY action differ from rat and rabbit intestine and colon, which show an inhibition of PGE- and VIP-induced secretion (Okuno et al., 1992; Saria et al., 1985), and the actions of PYY are resistant to TTX in these species (Cox et al., 1988; Hubel and Renquist, 1986). Similarly, studies with isolated rat crypt epithelial cells show PYY receptors are directly coupled to epithelial cAMP (Servin et al., 1989; Voisin et al., 1990). However, studies of guinea pig and porcine ileum and of mouse jejunum are consistent with PYY or NPY having neural sites of action (Brown et al., 1990; Riviere et al., 1993; Zafirov et al., 1992). Furthermore, a recent study in humans, in which intestinal perfusion techniques were used, has shown that PYY inhibits PGE-induced intestinal secretion via a neural mechanism (Roze et al., 1997). Thus, there appear to be variations in the sites of PYY action depending on the species examined.

During the past several years, there has been a great deal of emphasis placed on the signaling pathways linking inflammatory cells, nerves and epithelium in the regulation of intestinal ion transport (reviewed by Powell, 1991; see also Castro and Powell, 1994). Through the use of stimulants of phagocytes and mast cells in normal tissue or antigen stimulation of parasite-infected tissue and the effects of specific immune cell products on epithelial Na⁺ and Cl transport, it has been possible to provide a conceptual framework for the interpretation of acute inflammatory events that culminate in intestinal secretion and diarrhea. Key elements of this paradigm illustrate that immune system agonists alter electrolyte transport through the release of PGs from cells in the lamina propria with 50% of the response being due to PG activation of the enteric nervous system. Recent studies of enteric infections with Entamoeba histolytica (McGowan et al., 1990), Trichinella spiralis (Castro and Russell, 1987), Clostridium difficile (Triadafilopoulous et al., 1987, 1989) and enterohemorrhagic Escherichia coli (Elliot et al., 1994; Li et al., 1993) are consistent with this concept of inflammatory cell signaling and amplification through PG and neural pathways.

The present results suggest that the endocrine/paracrine mediator PYY may provide an important control point for PG-nerve communication in intestinal secretion associated with cryptosporidial infection. Increased release of PYY, as measured by its increased plasma concentration during malabsorptive states and in infectious diarrhea has been demonstrated (Adrian et al., 1986). These studies suggest a pathophysiological role of this peptide in diarrheal disease, but further study will be necessary to demonstrate this conclusively. Attenuation of the powerful Cl secretory response induced by PGI₂-activated enteric nerves by PYY may moderate a severe secretory diarrhea, such as that seen in AIDS patients infected with Cryptosporidium (Fayer and Ungar, 1986). The efficacy of PYY also suggests potential avenues for pharmacological treatment of secretory diarrhea associated with acute enteric infections.