Water Immersion Stress Prevents Caerulein-induced Pancreatic Acinar Cell NF- k B Activation by Attenuating Caerulein-induced Intracellular Ca 2 1 Changes*

Prior stress ameliorates caerulein-induced pancreatitis in rats. NF- k B is a proinflammatory transcription factor activated during caerulein pancreatitis. How-ever, the effects of prior stress on pancreatic NF- k B activation are unknown. In the current study, the effect of prior water immersion stress on caerulein and tumor necrosis factor- a (TNF- a )-induced NF- k B activation in the pancreas was evaluated. Water immersion of rats for up to 6 h prevents supramaximal caerulein-induced pancreatic I k B- a degradation and NF- k B activation in vivo. NF- k B activity is also inhibited in vitro in pancreatic acini prepared from water-immersed animals. TNF- a -induced NF- k B activation in pancreas or in pancreatic acini is unaffected by prior water immersion. Chelation of intracellular Ca 2 1 by 1,2-bis(2-aminophenoxy)ethane-N , N , N * , N * -tetraacetate/acetoxymethyl ester has similar effects to water immersion in preventing caerulein but not TNF- a -induced NF- k B activation in pancreas. Both the spike response and the sustained rise in [Ca 2 1 ] i in response to supramaximal caerulein p50/p50 and p50/p65 NF- k B dimers into the nucleus. in activation of I B kinase and phosphorylation of I k a to dissociation of I k a from NF- k B and degradation of I k B- a are for caerulein- and TNF- a -induced activation of NF- k B.

Doses of caerulein, the decapeptide analog of cholecystokinin, in excess of those that elicit a maximal rate of digestive enzyme secretion from the rat pancreas, elicit a reversible form of acute interstitial pancreatitis. This model of pancreatitis, referred to as secretagogue-induced pancreatitis, is associated with and possibly brought about by intra-acinar cell activation of digestive enzyme zymogens including trypsinogen (1). Supramaximal stimulation of freshly prepared but otherwise nor-mal pancreatic acini with caerulein also results in intra-acinar cell activation of trypsinogen and, subsequently, evidence of acinar cell injury in vitro (2).
We have shown recently that prior water immersion stress, under conditions that result in expression of heat shock protein 60 (HSP60), 1 prevents caerulein-induced in vivo activation of trypsinogen in acinar cells and protects against this form of secretagogue-induced pancreatitis (3). Pancreatic acini prepared from prior water-immersed rats, which contain increased amounts of HSP60, are also protected against caerulein-induced in vitro activation of trypsinogen as well as caeruleininduced in vitro injury. The mechanism(s) responsible for this stress-induced protection against caerulein-induced injury has not been established.
Nuclear factor-B (NF-B) is a family of widely expressed transcription factors that acts to modulate inflammatory processes. Activation of NF-B involves an intracytoplasmic kinase cascade that culminates in the phosphorylation of IB proteins leading to their dissociation from NF-B (4). As a result, the now activated NF-B, which is comprised of dimers made up of various combinations of Rel homology-sharing subunits, can translocate to the nucleus and act to regulate expression of genes coding for various proinflammatory factors (5). The phosphorylated IBs are, in parallel, degraded by proteasome.
Pancreatic acinar cell NF-B activation has been reported to occur during the very early stages of caerulein-induced pancreatitis (6,7) and evidence has been presented that suggests that prevention of caerulein-induced NF-B activation in the pancreas can ameliorate the severity of caerulein-induced pancreatitis (6). Supramaximal stimulation of acini in vitro with caerulein also causes activation of NF-B by a mechanism that has been shown to involve a rise in cytoplasmic free calcium levels ([Ca 2ϩ ] i ) and activation of protein kinase C (8 -9).
In the present communication, we report studies that have evaluated the effects of prior water immersion stress, under conditions associated with up-regulated HSP60 expression, on the activation of pancreatic NF-B. We show that prior water immersion stress prevents caerulein-induced in vivo NF-B activation in the rat pancreas. Similarly, NF-B activation in vitro is not observed when acini prepared from prior waterimmersed animals are exposed to a supramaximally stimulating dose of caerulein. In contrast, NF-B activation, both in vivo and in vitro, in response to TNF-␣ is not altered by prior water immersion. Finally, we show that prior water immersion stress markedly attenuates the acinar cell [Ca 2ϩ ] i changes that follow exposure to a supramaximally stimulating dose of caerulein, and in acini prepared from animals not stressed by prior water immersion, chelation of cytoplasmic Ca 2ϩ with BAPTA/AM can prevent caerulein-induced NF-B activation in those acini. Taken together, these observations lead us to conclude that water immersion stress prevents caerulein-induced NF-B activation by interfering with caerulein-induced changes in [Ca 2ϩ ] i . This phenomenon may contribute to the protective effects of water immersion on secretagogue-induced pancreatitis.

MATERIALS AND METHODS
Male Wistar rats weighing 100 -200 g (Charles River Laboratories, Wilmington, MA) were used in all experiments. The animals were housed in temperature-controlled (23 Ϯ 2°C) rooms with a 12-h light/ dark cycle, fed standard laboratory chow, and allowed to drink ad libitum. Caerulein was purchased from Research Plus (Bayonne, NJ), rat recombinant TNF-␣ was from BIOSOURCE, collagenase (CLS-4) was from Worthington, and both Fura-2/AM and BAPTA/AM were from Molecular Probes (Ridgefield, CT). All other reagents and chemicals were purchased from Sigma. All experimental protocols were approved by the Institutional Animal Use Committee of the Beth Israel Deaconess Medical Center (Boston, MA).
In Vivo Experiments-Rats were exposed to water immersion stress as described previously (3). In brief, the animals were placed in restriction cages and vertically immersed to the xiphoid process. The duration of water immersion was 6 h in all experiments unless otherwise specified. Caerulein (50 g/kg) and rat TNF-␣ (10 g/kg) were dissolved in saline and administered as a single intraperitoneal dose to normal and water-immersed animals. Control animals received no injection. Animals were sacrificed in a CO 2 chamber at various times after the injections, and pancreatic tissue samples were collected immediately.
In Vitro Experiments-Dispersed rat pancreatic acini were prepared by collagenase digestion from normal rats or rats after 6 h of water immersion and studied as described previously (10 -12). The acini were allowed to equilibrate for 5 min at 37°C and then were stimulated with either caerulein (0.1 M) or TNF-␣ (200 ng/ml). Other acini were preincubated with BAPTA/AM (50 M) for 30 min at 37°C before adding caerulein or TNF-␣. After an additional 30 or 90 min of incubation, the acini were washed in ice-cold HEPES-Ringer buffer, and nuclear and cytoplasmic protein extracts were prepared. Viability of acini after washing, as assessed by trypan blue exclusion, was Ͼ95%.
Assays-For evaluation of NF-B activation and IB-␣ degradation, nuclear and cytoplasmic protein extracts were prepared as described by Dyer and Herzog (13). Protein concentrations were determined by the method of Bradford (14). For [Ca 2ϩ ] i measurements, pancreatic acini from normal and water-immersed rats were loaded with 2 M Fura-2/AM for 30 min at 37°C and then washed extensively. [Ca 2ϩ ] i was quantitated using a Spex dual excitation spectrofluorometer as de-scribed previously (11,15). Excitation was at 340 and 380 nm, and emission was measured at 505 nm. The results are expressed as the fluorescence ratio.
Electrophoretic Mobility Shift Assay (EMSA)-Reaction mixtures (25 l, pH 7.5) contained 7.5-10.0 g of nuclear protein, 5 mM Tris, 100 mM NaCl, 1 mM dithiothreitol, 1 mM EDTA, 4% (v/v) glycerol, and 0.08 mg/ml salmon sperm DNA. The oligonucleotide probe (5Ј-AGT TGA GGG GAC TTT CCC AGG C-3Ј, Promega, Madison, WI) containing the B binding motif was end-labeled with [␥-32 P]ATP using T 4 polynucleotide kinase and purified over two successive 1-ml G-50 columns (Amersham Pharmacia Biotech). 1 ϫ 10 6 cpm of the probe was added to the reaction mixture, and the binding reaction was allowed to proceed for 20 min at room temperature. For supershift assays, 2 l of specific antibody against NF-B subunits p50, p52, p65, or c-Rel (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) was added to the reaction mixture and the resulting solution was allowed to incubate for 30 min at 4°C before adding the oligonucleotide probe. DNA-protein complexes were resolved in a 6% nondenaturing polyacrylamide gel in a TBE buffer (22.5 mM Tris, 22.5 mM boric acid, and 0.5 mM EDTA, pH 8.3) at 140 V for 2-3 h. Gels were dried and exposed to Kodak Bio Max MR films at Ϫ70°C. NF-B bands from films were quantitated by using an HP Scanjet 4100 scanner and a Scion image analysis program.
Western Blot Analysis-Equal amounts of cytoplasmic protein extracts (5-10 g) were diluted in Laemmli sample buffer with 5% mercaptoethanol. After boiling, the samples were resolved in 10% polyacrylamide gels in Tris-glycine-SDS buffer. The gels were transferred to nitrocellulose membranes, blocked in 5% nonfat dry milk in phosphatebuffered saline (PBS), pH 7.5, containing 0.1% (v/v) Tween 20 (PBSTmilk). Blots then were incubated with polyclonal rabbit anti-IB-␣ antibody (sc-371, Santa Cruz Biotechnology, Inc., Santa Cruz, CA) at 1:1000 (v/v) dilution in PBST-milk at 4°C overnight. The membranes then were washed in PBST and incubated with horseradish peroxidaseconjugated anti-rabbit IgG at 1:5000 (v/v) dilution in PBST-milk for 1 h. After washing, IB-␣ protein bands in the membranes were visualized by enhanced chemiluminescence (PerkinElmer Life Sciences).
Analysis of Data-The results reported in this communication represent means Ϯ S.E. of the mean values obtained from three or more separate experiments. In all figures, vertical bars denote S.E. values. Statistical evaluation of data was accomplished by analysis of variance, and p values of less than 0.05 were considered significant. All EMSA and Western blot gels shown are representative of at least three such gels prepared from independent experiments.

Effects of Water Immersion Stress on NF-B Activation in
Vivo-In preliminary experiments, rats were exposed to a supramaximally stimulating dose of caerulein (50 g/kg) and sacrificed at various times thereafter. Activation of pancreatic NF-B and degradation of IB-␣ were observed to occur in a biphasic manner with an initial peak of 30 min (11.1 Ϯ 1.8-fold control) after caerulein administration, a subsequent decline, and then a second phase of increase that reached a maximal level between 90 min (9.4 Ϯ 1.5-fold control) and 180 min (8.7 Ϯ 1.7-fold control) after caerulein administration. These observations are in accord with those reported recently by Gukovsky et al. (6). In our subsequent experiments, NF-B activation and the effects of water immersion stress on this process were evaluated at the selected times of 30 and 90 min after caerulein administration.
As shown in Fig. 1, prior water immersion profoundly inhibits the NF-B activation and IB-␣ degradation observed 30 as well as 90 min after caerulein administration. The time course of the water immersion-induced blockade of caerulein-induced NF-B activation is shown in Fig. 2. Half-maximal inhibition is observed after ϳ3 h of water immersion, and at that time water immersion-induced HSP60 expression is also approximately half-maximal (Fig. 2B). In contrast to these effects on caerulein-induced NF-B activation, TNF-␣-induced NF-B activation and IB-␣ degradation are not altered by prior water immersion (Fig. 3).
The observation that caerulein-induced NF-B activation, but not TNF-␣-induced NF-B activation, is prevented by prior water immersion suggested the possibility that caerulein and

FIG. 3. Effect of water immersion on TNF-␣-induced NF-B activation.
A, NF-B DNA binding activity and IB-␣ protein levels. Nuclear and cytoplasmic protein extracts were prepared from water-immersed and non-water-immersed animals 90 min after supramaximal stimulation with caerulein (C) or after stimulation with TNF-␣ (T, 10 g/kg). Nonstimulated animals served as controls (Co). The upper panel represents EMSA, and the lower panel represents the corresponding IB-␣ Western blot. B, densitometric quantitation of NF-B binding activity. Waterimmersed and non-water-immersed animals before (black bars) and after 90 min of either supramaximal caerulein stimulation (white bars) or TNF-␣ (10 g/kg) stimulation (gray bars) were evaluated. The values represent fold increase over nonstimulated control animals. *, p Ͻ 0.05 when the water-immersed group was compared with the corresponding non-water-immersed group.

FIG. 4. The subunits involved in caerulein-and TNF-␣-induced activation of NF-B.
Nuclear protein extracts were prepared from non-water-immersed animals subjected to supramaximal stimulation with caerulein for 90 min or with TNF-␣ (10 g/kg) for 90 min. Nuclear extracts were incubated for 30 min at 4°C in the presence of 2 l of either anti-p50, anti-p52, anti-p65, or anti-c-Rel antibodies before adding the labeled oligonucleotide probe. TNF-␣ might activate different species of NF-B. To evaluate this possibility, supershift assays were performed using antibodies to the p50, p52, p65, and c-Rel subunits of the NF-B. As shown in Fig. 4 and in accord with results reported by others (6,16), the caerulein-or cholecystokinin-induced activation of NF-B involves p50/p50 and p50/p65 dimers. p50/p50 and p50/ p65 dimers are involved also in the TNF-␣-induced activation of NF-B. These observations indicate that caerulein and TNF-␣ activate the same species of NF-B in rat pancreas.
Effects of Prior Water Immersion on NF-B Activation in Vitro-Acini were prepared from control rats and from rats immediately after 6 h of water immersion stress. Those acini then were exposed to a supramaximally stimulating concentration of caerulein or to TNF-␣ in vitro, and NF-B activation was evaluated 30 or 90 min later. As shown in Fig. 5, caeruleininduced NF-B activation and IB-␣ degradation are not observed in acini prepared from prior water-immersed animals, but water immersion does not interfere with either TNF-␣induced degradation of IB-␣ or activation of NF-B.

Effects of Chelating [Ca 2ϩ ] i on Caerulein-and TNF-␣-induced NF-B Activation-Freshly
prepared acini obtained from control (i.e. not water-immersed) animals were incubated with the Ca 2ϩ chelator BAPTA/AM and then exposed to either caerulein or TNF-␣. As shown in Fig. 6, chelation of intracellular Ca 2ϩ with BAPTA/AM, which prevents the caeruleininduced rise in [Ca 2ϩ ] i (11), prevents caerulein-but not TNF-␣-induced NF-B activation.

Effects of Prior Water Immersion on Caerulein-induced Changes in [Ca 2ϩ ] i -Freshly prepared acini were loaded with
Fura-2/AM, washed, and then incubated with a supramaximally stimulating concentration of caerulein. As shown in Fig.  7, when those acini were prepared from control animals, caerulein causes a large but transient rise in [Ca 2ϩ ] i , which is followed by a sustained but lesser elevation of [Ca 2ϩ ] i that persists throughout the period of observation. The resting [Ca 2ϩ ] i in acini prepared from water-immersed animals is lower than that observed in acini prepared from the control group, and both the peak and the sustained increases in [Ca 2ϩ ] i noted after caerulein addition are attenuated profoundly. DISCUSSION Otaka et al. (17), in 1994, reported the results of studies that indicated that HSP60 expression in the rat pancreas was upregulated by prior water immersion stress and that water immersion stress protected those animals from subsequent caerulein-induced pancreatitis. Recently, in studies designed to explore the mechanism(s) responsible for the protective effect of water immersion on pancreatitis, we found that caeruleininduced intrapancreatic trypsinogen activation, an early event in secretagogue-induced pancreatitis, was prevented by prior water immersion (3). We noted that this prevention of trypsinogen activation was correlated temporally with the up-regulation of HSP60 expression. Furthermore, we found that the effects of prior water immersion could be detected in acini studied in vitro. That is, intra-acinar cell activation of trypsinogen and cell injury were observed when acini from non-waterimmersed animals were exposed to a supramaximally stimulating concentration of caerulein in vitro, but neither trypsinogen activation nor cell injury was observed when acini prepared from water-immersed animals were exposed in vitro to a supramaximally stimulating concentration of caerulein (3). The currently reported studies were designed to examine further the mechanisms responsible for the protective effect of water immersion on secretagogue-induced pancreatitis.
NF-B is a widely expressed transcription factor that in many systems has been shown to play a critical role in regulating inflammatory processes by modulating the expression of genes coding for inflammatory mediators including cytokines, chemokines, and adhesion molecules (5,18). Studies from several groups have indicated that activation of pancreatic acinar cell NF-B is an early event in secretagogue-induced pancreatitis. In a recent study we showed that although caeruleininduced trypsinogen activation and NF-B activation are closely related temporally, they are independent events in pancreatic acinar cells (19). Although still somewhat controversial (6,7), the preponderance of evidence suggests that NF-B activation is a proinflammatory event in this model of pancreatitis and that interventions that interfere with acinar cell NF-B activation reduce the severity of secretagogue-induced pancreatitis (6,16,20).
Pancreatic acinar cell NF-B activation by supramaximally stimulating doses of caerulein or cholecystokinin has been examined recently by several groups, and the results of their studies have shown that (a) activation depends on a rise in acinar cell [Ca 2ϩ ] i and activation of protein kinase C (8,9), (b) activation is accompanied by degradation of IB-␣ (8,9), and (c) the NF-B dimers activated in this process are composed of p50 and p65 subunits (6,16). Pancreatic acinar cell NF-B also can be activated by TNF-␣ (21), but this process has not been studied as extensively. Our own results (Figs. 3-5) indicate that TNF-␣ also promotes IB-␣ degradation and translocation of The currently reported studies indicate that the activation of pancreatic acinar cell NF-B by supramaximally stimulating doses of caerulein both in vivo (Figs. [1][2][3] and in vitro (Fig. 5) is inhibited by prior water immersion. Under these conditions, HSP60 expression is up-regulated and the time course of HSP60 expression after water immersion stress roughly correlates with that of prevention of NF-B activation (Fig. 2). It is tempting, therefore, to speculate and simplify these observations by concluding that HSP60 mediates the process by which NF-B activation is inhibited, but this conclusion can only be tentative because water immersion stress also might set in motion other as yet unidentified events that prevent NF-B activation beside HSP60 expression. Further studies will be needed before the relationship between HSP60 expression and prevention of caerulein-induced NF-B activation can be defined more clearly and unambiguously.
Although the role of HSP60 in the events triggered by water immersion stress remains uncertain, the studies reported in this communication still provide some insights into the mechanisms by which water immersion stress affects NF-B activation. Our studies indicate that water immersion stress interferes with caerulein-induced NF-B activation, but it does not alter TNF-␣-induced NF-B activation either in vivo (Figs. 1 and 3) or in vitro (Fig. 5). This finding suggests that water immersion stress affects the activation process at a step that occurs before IB kinase activation and IB-␣ phosphorylation because, as noted above, these steps seem to be shared by both the TNF-␣-and caerulein-induced NF-B activation process. The finding that caerulein-induced NF-B activation but not TNF-␣-induced NF-B activation can be blocked by chelation of cytoplasmic Ca 2ϩ with BAPTA/AM (Fig. 6) indicates that the two pathways for NF-B activation differ in their requirement for a rise in [Ca 2ϩ ] i , and this observation suggested to us that prior water immersion stress might interfere with caeruleininduced NF-B activation by interfering with caerulein-in-duced [Ca 2ϩ ] i changes in pancreatic acinar cells. To examine this possibility, [Ca 2ϩ ] i changes in acini prepared from control and water-immersed animals were evaluated (Fig. 7). Prior water immersion was found to reduce the resting [Ca 2ϩ ] i level in acini and to attenuate markedly the caerulein-induced rise in [Ca 2ϩ ] i . This observation leads us to conclude that prior water immersion prevents caerulein-induced NF-B activation by interfering with caerulein-induced [Ca 2ϩ ] i changes and to suggest that this may at least in part explain the protection against caerulein-induced pancreatitis that is afforded by prior water immersion.
In summary, our studies indicate that supramaximally stimulating doses of caerulein and TNF-␣ activate the same species of NF-B in pancreatic acinar cells but that they do so by different mechanisms. Caerulein-induced activation is a Ca 2ϩdependent process, whereas TNF-␣-induced activation is independent of a rise in [Ca 2ϩ ] i . Prior water immersion stress induces HSP60 expression and also prevents caerulein-induced NF-B activation. We suggest that prior water immersion stress prevents caerulein-induced NF-B activation by inter- . The values shown represent fold increases over nonstimulated controls (black bars). *, p ϭ Ͻ0.05 when the group exposed to BAPTA/AM was compared with the corresponding group not exposed to BAPTA/AM.

FIG. 7. Effects of prior water immersion on [Ca 2؉ ] i changes induced by supramaximal stimulation with caerulein.
Freshly prepared pancreatic acini from water-immersed (open circles) and nonwater-immersed (closed circles) rats were loaded with Fura-2/AM for 30 min, washed, and then exposed to a supramaximally stimulating dose of caerulein (0.1 M), and fluorescence was monitored as described in the text. Results shown are representative of three separate experiments.
fering with the caerulein-induced rise in [Ca 2ϩ ] i , which is critical to that event. Chelation of cytoplasmic Ca 2ϩ with BAPTA/AM can bring about the same effect. The reduction in caerulein-induced [Ca 2ϩ ] i rise and the activation of NF-B that follows water immersion stress is correlated with the rise in HSP60 expression that also follows water immersion stress, but whether HSP60 actually mediates the effects of water immersion stress on Ca 2ϩ dynamics and NF-B activation will require further studies.