‘ ,v
inuxfihflum

Irvif.

.h.
J
. 59%.“.
7 IA. 8:
2r:
- . . 5:. 315
£939; 393.1%...

x. . l?
a any!) . :

z i.
Item.

. I:
53.x. «.th
.2. ..
. I ~
:

.
.1

I «S 9
.. v.34!
Raft!

.
11.. . ..
2‘? 4.»: 323...

4.21.3. .
1v§3$£7~€ I a.

1 t 1 1w
3.9.1.23

«

. .23.

a...» i.

. i 4 _ If
. J3"..-

 

 

 

' (HF. 81.

3x

MICHIGAN s

I l I ’l/II/I/I

LIBRARIES

Will/Ill!!!

5416

EUNNE
l
l

l.)

I/

 

l, ll W
2

3 1293 014

l/

 

 

This is to certify that the

dissertation entitled

Mechanisms and Control of Bicarbonate

Secretion in Guinea Pig Pancreatic Ducts

presented by

Jose' de Ondarza

has been accepted towards fulfillment
of the requirements for

 

Ph.D. degree in Physiology

ganja professor

Date 3-21—96

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove thie checkout from your record.
To AVOID FINES return on or before dete due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU IeAnNiitmetive Action/Equal Opportunity Inetitmion
Wm:

MECHANISMS AND CONTROL
OF BICARBONATE SECRETION IN
GUINEA PIG PANCREATIC DUCTS

By

José de Ondarza

A DISSERTATION
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Physiology

1996

ABSTRACT
MECHANISMS AND CONTROL
OF BICARBONATE SECRETION IN
GUINEA PIG PANCREATIC DUCTS
By

José de Ondarza

While elaboration of bicarbonate-rich fluid by the exocrine pancreas is primarily
controlled by the gut hormone secretin, recent studies have identified several other
neurohormonal agents that also evoke ductal secretion. The intracellular signalling
pathways and the molecular transport components involved are, however, poorly defined.
For this reason, preparations of isolated guinea pig pancreatic ducts were utilized to
determine the effects of neurohumoral agents on cyclic AMP (CAMP) in duct epithelial
cells and to establish the presence of transport elements thought to be involved in
bicarbonate secretion.

Secretin caused a 5 to 8-fold elevation of cAMP (EC,o = 150 pM) in isolated
ducts and cultured duct epithelial monolayers. Main duct segments, while responsive,
were less so than segments of interlobular duct. In isolated duct segments, carbachol,
bombesin, cholecystokinin, substance P, calcitonin gene-related peptide, glucagon,
insulin, isoproterenol, neurotensin, and prostaglandin E2 did not significantly alter resting
or secretin-stimulated cAMP levels. By contrast, VIP significantly increased CAMP
levels, while somatostatin significantly attenuated the cAMP elevations produced by a
physiological dose of secretin, suggesting that its effects on pancreatic duct fluid

secretion are mediated by inhibition of adenylyl cyclase.

By examining changes in intracellular pH (pH,) of principal epithelial cells, the
involvement of specific ion transport mechanisms in duct bicarbonate secretion was
determined. In HCO3‘-free medium, pH, fell by 0.32 i 0.06 pH units in the presence of
amiloride and by 0.36 :t 0.08 pH units in the absence of Na”. In the presence of
extracellular HCO3‘, pH, acidified in Na*-free, but not in amiloride-containing, medium.
Superfusion with chloride-free buffers or with buffers containing 4,4'-
diisothiocyanatostilbene-Z,2'-disulfonic acid (DIDS) produced a cytosolic alkalinization
of 0.13-0.22 pH units. These observations demonstrate the presence of Na‘IH+ exchange,
Na“/HCO3' cotransport, and Cl'/HCO3' exchange in guinea pig pancreatic ducts. pHi
recovered significantly from an NH4Cl pulse in HCO3‘-free buffers containing amiloride
and carbachol or secretin. This recovery was blocked by bafilomycin Al and by
preincubation with nocodazole or cytochalasin D, suggesting that a vesicular H*-ATPase
augments Na*-dependent H” extrusion during agonist-stimulated bicarbonate secretion,

and that its activation involves cytoskeletal elements.

T 0 my parents.

You have given me so much.

iv

ACKNOWLEDGMENTS

I want to express my sincere and profound thanks to my mentor, Dr. Seth Hootrnan,
for the years of learning and maturing I have had in his laboratory. He has been not only
an inspiring teacher and researcher, but also a good friend. I also thank the members of
my thesis committee, Drs. James Galligan, Steven Heidemann, William Smith, and Birgit
Zipser, for their advice and guidance throughout the research and writing process. In
addition, I want to acknowledge the much missed advice, enthusiasm, and humor of the
late Dr. Jacob Krier.

I am grateful for the guidance and support of Drs. Thomas Adams and Robert
Stephenson during and beyond the semesters as a Teaching Assistant in PSL 410, 431,
and 501. I am thankfirl also for the friendships of my fellow graduate students - you
have helped make this time better. And lest I forget all the help I have received from the
secretarial staff: Thank you!

To Mary Beth - your encouragement, love, and prayers have meant so much
through the last four years. Thanks for believing in me.

Most of all, I thank God. Through Him, all things are possible.

TABLE OF CONTENTS

LIST OF TABLES ..................................... viii
LIST OF FIGURES ...................................... ix
KEY To ABBREVIATIONS ................................. xii
I. INTRODUCTION ................................... l
A. Physiological Mechanisms .......................... 2

l. Morphological and biochemical hallmarks of pancreatic ducts ...... 2

2. Grth and differentiation ........................ 4

3. Model systems .............................. 5

4. Transport components of pancreatic duct electrolyte secretion ...... 8

5. Physiological regulation of bicarbonate secretion ............ 13

6. Regulation of pancreatic duct mucus secretion .............. 16

B. Pathophysiology ................................ 17

l. Cystic fibrosis ............................... 17

2. Pancreatitis ................................ 18

3. Pancreatic cancer ............................. 19

4. Diabetes .................................. 20

C. Rationale and Specific Aims ......................... 21

11. MATERIALS AND METHODS ............................ 25
A. Materials ................................... 25

B. Isolation of Pancreatic Ducts ......................... 26

vi

III.

IV.

V.

VI.

Regulation of Cyclic AMP Levels in Isolated Pancreatic Ducts

and Duct Epithelial Monolayers ......................... 27
1. Primary culture of duct epithelial cells .................. 27
2. Incubation protocols ............................ 28
3. Cyclic AMP radioimmunoassay ..................... 28
4. Protein microassay ............................ 29
D. Confocal Microscopic Analysis of Isolated Pancreatic Ducts ........ 31
1. Principles of laser scanning confocal microscopy ............ 31
2. Imaging of fixed tissues .......................... 32
3. Regulation of intracellular pH in isolated pancreatic ducts ........ 34
a. SNARF-l loading and superfusion of pancreatic ducts ........ 34
b. Live cell imaging of the duct epithelium ............... 36
c. Determination of pH, ......................... 38
E. Statistics .................................... 41
RESULTS ..................................... ‘. 42
A. Regulation of Cyclic AMP Levels in Guinea Pig Pancreatic
Duct Epithelial Cells ............................. 42
1. Pancreatic duct morphology ....................... 42
2. Effects of secretin and IBMX on pancreatic duct cyclic AMP levels. . . 43
3. Regulation of cyclic AMP levels by other neurohumoral agents ..... 46
B. Electrolyte Transport Components of Guinea Pig Pancreatic Duct
Bicarbonate Secretion ............................. 52
1. Role of NaVH" Exchange in pH regulation in unstimulated
pancreatic duct epithelial cells ...................... 54
2. Contribution of bicarbonate to pH regulation ............... 60
3. Contributions of chloride-bicarbonate exchange to pH regulation . . . . 60
4. Involvement of ATP-dependent H+ transport in proton extrusion
from duct epithelial cells ......................... 63
a. pH regulation in the presence of agonists ............... 63
b. Role of the cytoskeleton in agonist-mediated pH recovery ...... 70
DISCUSSION .................................... 75
SUMMARY AND CONCLUSIONS .......................... 92
LITERATURE CITED ................................ 95

vii

Table 1.

Table 2.

Table 3.

LIST OF TABLES

Composition of physiological buffers used .................. 35

Effects of potential regulators of pancreatic duct secretion on

resting and secretin-stimulated cyclic AMP levels in mixed

suspensions of acutely isolated guinea pig main and interlobular

pancreatic ducts ................................ 51

Effects of secretin and carbachol on recovery of intracellular pH

(pH,) following intracellular acidification in the presence or absence
of amiloride and bafilomycin Al ....................... 56

viii

LIST OF FIGURES

Figure 1. Electron micrograph of a portion of the epithelium of the
main pancreatic duct ............................. 3

Figure 2. Phase contrast light micrograph of a pancreatic duct epithelial
monolayer maintained in primary culture for five days ............ 7

Figure 3. Proposed models for bicarbonate secretion from the principal cells
of the pancreatic duct epithelium ....................... 9

Figure 4. Standard curve of B/Bo vs. concentration of cyclic AMP ........... 30

Figure 5. Confocal microscopic image of the pancreatic duct epithelium

labelled with the F-actin probe, rhodamine-phalloidin ............ 33
Figure 6. Schematic for superfusion of isolated segments of pancreatic ducts

in a F ocht live cell perfusion chamber .................... 37
Figure 7. SNARF -1 chemical structure, emission spectra, and traces of

SNARF-l emissions and their ratio ...................... 39
Figure 8. Schematic for measurement of SNARF -1 emissions ............. 40

Figure 9. Time course of the effect of 0.1 uM secretin on cyclic AMP
levels in mixed suspensions of acutely isolated segments of
guinea pig main and interlobular pancreatic ducts .............. 44

Figure 10. Eflect of IBMX on secretin-stirnulated cyclic AMP levels in
mixed suspensions of acutely isolated segments of guinea pig
main and interlobular pancreatic ducts .................... 45

Figure 11. Dose-dependence of secretin stimulation of cyclic AMP levels

in mixed suspensions of acutely isolated guinea pig main and
interlobular pancreatic ducts ......................... 47

ix

Figure 12.

Figure 13.

Figure 14.

Figure 15.

Figure 16.

Figure 17.

Figure 18.

Figure 19.

Figure 20.

Figure 21.

Figure 22.

Figure 23.

Figure 24.

Figure 25.

Effect of secretin on cyclic AMP levels in acutely isolated guinea
pig main and interlobular ducts and cultured monolayers derived
fi'om these two segments of the duct system ................. 48

Cyclic AMP response to vasoactive intestinal peptide (VIP) in
acutely isolated guinea pig main and interlobular ducts and cultured
monolayers derived from these two segments of the duct system ...... 50

Calibration curves obtained from SNARF -1-loaded principal
epithelial cells (640 nm/580 nm emission ratio vs. pH,) ........... 53

Sodium dependence of intracellular pH regulation in guinea pig
pancreatic duct principal epithelial cells ................... 55

Effect of amiloride on pHi recovery in pancreatic duct cells

following intracellular acidification ...................... 57
Effect of amiloride addition on pH, of pancreatic duct epithelial cells . . . 58
Effect of amiloride addition or sodium substitution on pHi of

pancreatic duct epithelial cells in the absence or presence of

extracellular HCO,‘ .............................. 59
Recovery of pH, in pancreatic duct segments in the presence of

amiloride and Na+ following cytosolic acidification induced by

superfusion in Na*-free buffer ......................... 61
pH recovery in the presence of amiloride or in the absence of

extracellular Na“ following a 2 min NH4C1 pulse, both in the

absence (KRH) and presence (KRB) of extracellular HCO,’ ......... 62
Effect of removal of extracellular Cl' on pHi of pancreatic duct

epithelial cells perfused with a HCO3'-containing buffer (KRB) ....... 64
Effect of exposure to DIDS on the pHi of pancreatic duct epithelial cells . . 65
Effect of secretin on pH, recovery from a 2 min NH4C1 pulse

in pancreatic duct cells in the presence of 0.5 mM amiloride ......... 67
Effect of bafilomycin A, and amiloride on secretin-stimulated pH

recovery in pancreatic duct epithelial cells .................. 68
Effect of bafilomycin A, alone on pH recovery from an acid load

imposed by superfusion with 20 mM NH4C1 ................. 69

Figure 26. Effects of cytoskeletal-disrupting agents on distribution of
microfilament and mcrotubules in pancreatic duct principal
epithelial cells ................................. 72

Figure 27. Effect of preincubation with 10 ug/rnl of cytochalasin D on

pH recovery in pancreatic duct segments following a 2 min
NH4C1 pulse .................................. 73

Figure 28. Recovery of pH, in the presence of amiloride and‘agonists
following a 2 min NH4CI pulse after preincubation with
cytochalasin D or nocodazole ......................... 74

Figure 29. Revised model for bicarbonate secretion from the principal

epithelial cell of the pancreatic duct epithelium, based on
studies in guinea pigs ............................. 89

xi

ATP
ATPase
BSA
CA
cAMP
CCh
CCK
CCK-8
CF TR
CGRP
DIDS
EGF
FITC
GRP
HEPES

IBMX

LSCM
NMG+
PBS
PGE2
PHI
PMT
PKA
PKC
RIA

SNARF -1

TEA
TGF a
TGF [3
VIP

KEY TO ABBREVIATIONS

Adenosine monophosphate

Adenosine triphosphate

Adenosine triphosphatase

Bovine serum albumin

Carbonic anhydrase

Adenosine 3'-5' cyclic monophosphate
Carbamylcholine (carbachol)

Cholecystokinin

Cholecystokinin octapeptide

Cystic fibrosis transmembrane conductance regulator
Calcitonin gene-related peptide
4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid
Epidermal growth factor

Fluorescein isothiocyanate

Gastrin releasing peptide
N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid
HEPES Ringers

3-isobutyl-1 -methylxanthine

Krebs Ringers, bicarbonate and HEPES-buffered
Krebs Ringers, HEPES-buffered

Laser scanning confocal microscope
N-methyl-D-gluconate

Phosphate-buffered saline

Prostaglandin E,

Intracellular pH

Photomultiplier tube

Protein kinase A (cyclic AMP-dependent)
Protein kinase C

Radioirnmunoassay

5-(and-6)-carboxy-SNARF -1 , acetoxymethyl ester
Tetraethylammonium

Transforming growth factor alpha

Transforming growth factor beta

Vasoactive intestinal peptide

xii

I. INTRODUCTION

The highly branched excretory duct system of the mammalian exocrine pancreas
serves both as a passive conduit for delivery of digestive enzymes secreted by pancreatic
acinar cells to the duodenum and as the source of an alkaline, bicarbonate-rich fluid
which neutralizes the acidic chyme that enters the duodenum from the stomach. While it
is well established that postprandial secretion of bicarbonate is stimulated principally by
the release of secretin from duodenal endocrine cells (Case & Argent, 1993), recent
studies using both in vivo and in vitro model systems have suggested that other
gastrointestinal hormones and the autonomic nervous system play important roles as
modulators of secretin's stimulatory effects (Argent et. al., 1992). A recent surge of
interest in pancreatic duct secretory mechanisms has been sparked by the recognition that
the duct system is centrally involved in several pathological disturbances of pancreatic
function, including those associated with pancreatitis, cystic fibrosis, diabetes, and
pancreatic carcinomas. The current studies therefore investigated the regulation of
bicarbonate secretion from the pancreatic duct epithelium of the guinea pig, focusing on
the intracellular messenger pathway utilized by secretin and on the ion transport elements

involved in the secretory process.

A. Physiological Mechanisms

 

The excretory duct system of the mammalian exocrine pancreas comprises less than
10% of the total glandular tissue mass and consists of a branching tree of epithelial
tubules which narrow in diameter from the main pancreatic duct through successive
generations of interlobular, intralobular, and intercalated ducts. The principal cells of the
simple epithelium that lines the duct system possess the morphological hallmarks
associated with active transport; apical microvilli, abundant mitochondria, and
elaboration of the basolateral plasma membrane (Kern, 1993). Goblet cells are also found
in the main and interlobular ducts and are most abundant in the portions of the main duct
proximal to the ampulla (Kodama, 1983; Hootrnan & de Ondarza, 1995). Endocrine cells
have been observed in the pancreatic duct epithelia of some species (Kodama, 1983),
although they constitute only a small percentage of the total cell population (Figure 1).

The principal epithelial cells of the duct system are notable for their high level of
expression of carbonic anhydrase (Spicer et. al., 1982; Buanes ct. al., 1986) and Na“-K*-
ATPase (Madden & Sarras, 1982). Other reported markers of duct epithelial cells include
the cystic fibrosis transmembrane conductance regulator (CFTR) protein (Marino et. al.,
1991, Crawford et. al., 1991), mucins (Githens, 1988), and cytokeratins (Schussler et. al.,
1992). Goblet cells in the pancreatic duct system express mucins that bind wheat germ
agglutinin and other lectins (Geleff & Bock, 1984). A recent study has identified insulin,

glucagon, somatostatin, pancreatic polypeptide, Cholecystokinin, and

 

Figure 1. Electron micrograph of a portion of the epithelium of the main pancreatic duct.
This section of the epithelium features several principal cells (p), a granule-filled goblet cell
(g), and two endocrine cells (e). Goblet cells account for 20-30% of the epithelial cell
population in the main pancreatic duct and less than 10% in the interlobular ducts.
Endocrine cells comprise approximately 0.5% of the cell population in both segments of the
duct system. Duct lumen (L). Scale bar = 5 pm.

4

serotonin-containing endocrine cells in the rat pancreatic duct system (Park & Bendayan,
1992), and somatostatin- and serotonin-containing endocrine cells have been localized by

immunocytochemistry in the guinea pig (Hootrnan et. al., 1996).

2. Wm

The pancreas develops as a pair of evaginations from the embryonic foregut. These
pouches subsequently form branching tubules consisting of protodifferentiated cells
which give rise to acinar cells, endocrine cells, and to mature duct epithelial cells
(Githens, 1988). Although it has been suggested that a pool of these stem cells remains
postnatally, definitive evidence for this assertion is lacking. The terminal differentiation
of protodifferentiated cells in the pancreas may be influenced by extracellular matrix
components (Mai et. al., 1990) and by epithelial-mesenchymal interactions during
development (Dudek et. al., 1991). The pluripotency of differentiated pancreatic
epithelial cells is indicated by the observations that acinar cells in culture may assume a
ductal phenotype (Hall & Lemoine, 1992) and that duct cells can proliferate and give rise
to endocrine cells or hepatocytes following necrosis of acinar tissue (Githens, 1993;
Makino et. al., 1990).

The physiological mechanisms through which pancreatic duct epithelial cell turnover
is regulated have not been extensively investigated. Studies utilizing in viva [3H]-
thymidine labelling demonstrated that exogenous cholecystokinin increased cell division
in small intralobular ducts (Morisset et. al., 1982). By contrast, in vitro investigations

utilizing primary cultures of guinea pig and hamster pancreatic interlobular duct epithelial

5

cells have not supported a mitogenic role for this hormone (V erme & Hootman, 1990;
Mangino et. al., 1992), although trophic effects of epidermal growth factor (EGF),
secretin, bombesin, vasoactive intestinal peptide (VIP), and gastrin were demonstrated.
Insulin and transforming growth factor alpha (TGFu) also were shown to Stimulate, and
transforming growth factor beta (TGFB) to inhibit, growth and cell division in duct
epithelial monolayers in vitro (Bhattacharyya et. al., 1995), reflecting the complexity of

growth regulatory mechanisms.

3. MW

Although exploitation of in vivo model systems has provided a wealth of information
on the systemic control of pancreatic duct secretion, detailed analysis of the mechanisms
of intracellular signal transduction in duct epithelial cells has necessitated isolation of
intact segments of the duct system. These have been obtained from the pancreata of
several mammalian species through a combination of enzymatic dissociation, sieving,
and manual dissection (Githens, 1988; Githens et. al., 1980; Hootrnan & Logsdon, 1988;
Arkle et. al., 1986). Preparations of isolated pancreatic duct segments have now been
exploited by several laboratories as models for characterization of the electrolyte
transport components operating within the ductal epithelium. Electrophysiological and
fluorescence imaging studies have identified a battery of conductive pathways, carrier-
mediated processes, and primary active transport mechanisms, the concerted activity of
which subserves regulated bicarbonate secretion.

The preparation of isolated pancreatic ducts also has made possible the long-terrn

6

maintenance of duct epithelial cells in vitro. Primary cultures have been prepared from
human (Harris & Coleman, 1987), bovine (Sato et. al., 1983), rat and hamster (Githens et.
al., 1987), mouse (Githens, 1992), and guinea pig (Hootman & Logsdon, 1988)
pancreatic ducts (Figure 2). An important advantage of these cultures is the availability
of a pure population of one cell type that is readily accessible for biochemical analysis,
radioisotope flux measurements, and proliferative assays. Pancreatic duct explants have
been cultured as monolayers on solid substrates (Githens, 1988; Hootman & Logsdon,
1988), on permeable collagen gels, and on nitrocellulose filters (Argent et. al., 1992;
Hootman & Logsdon, 1988), and as epithelium-lined vesicles embedded in a collagen or
agarose matrix (Githens et. al., 1987). When cultured on permeable substrates, cells
derived fi'om duct explants retain the morphological hallmarks of duct principal epithelial
cells in vitro (Hootman & Logsdon, 1988). Duct principal cells in primary culture also
have been reported to retain many duct cell markers, including carbonic anhydrase
(Githens et. al., 1992) and cytokeratins (Harris & Coleman, 1988), and to remain
functionally responsive to secretin (Hubchak et. al., 1990). Patch-clamp studies on duct
cells cultured from the human fetal pancreas have identified chloride channels responsive
to secretin, forskolin, and dibutyryl cyclic AMP stimulation (Gray et. al., 1989).
Transformed cell lines also have been developed from several pancreatic
adenocarcinomas. These include BxPC-3 (Lingard et. al., 1992), CFPAC-l
(Schoumacher et. al., 1990), Capan-l (Leviat et. al., 1988), MDAPanc-3 (Frazier et. al.,
1990), Mia Paca-2 (Yunis et. al., 1977), HPAF (Kim et. al., 1989), and Pane-1 (Lieber et.

al., 1975) cells. The physiological responsiveness of these transformed cells has not been

 

Figure 2. Phase contrast light micrograph of a pancreatic duct epithelial monolayer
maintained in primary culture for five days. Epithelial cells spread concentrically from the
site of the original explant (e), forming a confluent monolayer. Goblet cells and endocrine
cells do not proliferate in this culture system. The monolayer thus consists entirely of
principal epithelial cells. Scale bar = 100 um.

8

investigated extensively. In the BxPC-3 and CFPAC-l cell lines, anion transport can be
stimulated by agonists that elevate intracellular calcium, but not by agents that increase
cyclic AMP (Lingard et. al., 1992; Schournacher et. al., 1990), suggesting that normal

signal transduction pathways have been lost or altered during transformation.

4. Iranspgn ngpgnents 9f Pancreatic Dug; Elegmlyle Segretign

Stimulation of the in situ pancreas with secretin results in the production of a large
volume of pancreatic juice, with a bicarbonate concentration which may reach 140-160
mEq/I, far in excess of the normal concentration in plasma and extracellular fluids (Case
& Argent, 1993). Based on recent studies, several models have been proposed to account
for the secretion of bicarbonate ions in this organ (Swanson & Solomon, 1975; Kuijpers
& De Pont, 1987; Novak & Greger, 1988b; Villanger et. al., 1995). Figure 3 illustrates
the major components thought to be involved in this process.

Studies utilizing preparations of isolated, perfused rat pancreatic ductules have
indicated that bicarbonate ions are secreted into the duct lumen through the action of a
chloride/bicarbonate exchange mechanism that is sensitive to disulfonic stilbenes (Novak
& Greger, 1988b; Stuenkel et. al., 1988; Zhao et. al., 1994, Novak & Pahl, 1993). This
anion exchange is driven by the compensatory re-uptake of chloride from the duct lumen
in response to outward chloride flux through low-conductance (5-10 pS) anion-selective
channels in the apical plasma membrane (Gray et. al., 1990a), channels whose electrical
characteristics are in all respects identical to those of the CFTR protein (Welsh et. al.,

1992). The open-state probability of this channel is markedly increased by cyclic AMP,

 

 

 

 

 

   
 
 
 
 

l

I J
Na‘HCO,‘ /
i\%ww} apical

Figure 3. Proposed models for bicarbonate secretion from the principal cells of the
pancreatic duct epithelium. Model A features the major cellular components necessary for
bicarbonate secretion: carbonic anhydrase (CA) generates hydrogen and bicarbonate ions
through creation of carbonic acid. Bicarbonate ions are thought to exit the cytoplasm via a
chloride/bicarbonate exchanger while chloride ions are recycled to the duct lumen through
a cyclic AMP-activated chloride channel. Hydrogen ions which accumulate in the cytoplasm
are thought to be removed via a sodium/hydrogen exchanger. The sodium-potassium pump
removes sodium ions in exchange for potassium ions, while a calcium-activated potassium
conductance completes the current loop. Recent evidence also suggests the presence of
sodium-dependent bicarbonate transporters. Model B resembles model A in all points, but
also includes a vesicular hydrogen ion ATPase, the insertion of which into the basolateral
plasma membrane is regulated by cellular signal transduction pathways. This proton pump
has been suggested as the primary means responsible for removal of hydrogen ions during
active secretion.

A B

10

dibutyryl cyclic AMP, and agents which elevate cyclic AMP levels in cells (Gray et. al.,
1988; Gray et. al., 1993), most likely through activation of the cyclic AMP-dependent
protein kinase (PKA). Other characteristics of this channel include a linear current-
voltage relationship, voltage and time independence, and sensitivity to the chloride
channel blockers NPPB and DCl-HC1(Novak & Greger, 1988b; Gray et. al., 1993).
Studies on Chinese hamster ovary cells transfected with the CF TR gene have shown that
CFTR activation by PKA is potentiated by protein kinase C, and that the channel can be
deactivated by alkaline phosphatase in this in vitro system (Tabcharani et. al., 1991).
These studies are indicative of the complexity of CFTR channel regulation and lend
credence to the supposition that the PKA and PKC binding sites located within the
regulatory domain of the CF TR protein play a functional role (Denning et. al., 1992). A
potential alternate route for chloride efflux during bicarbonate secretion may exist, as a
calcium-activated chloride conductance has been demonstrated in pancreatic duct
epithelial cells in at least one species (Gray et. al., 1994).

Basolateral potassium conductance also appears to play an important role in duct
secretion, an assumption based on the observation that the potassium channel blockers,
barium and tetraethylammonium (TEA), both inhibit secretin-stimulated fluid secretion
by isolated rat pancreatic ducts (Ashton et. al., 1991). The molecular entity responsible
for agonist-regulated potassium conductance in the pancreatic duct principal epithelial
cell appears to be the calcium-activated, voltage-dependent, so-called maxi-K channel
(Argent et. al., 1987), a channel which has a unitary conductance of over 200 p8. Studies

by Gray and coworkers (Gray et. al., 1990b) have shown that the open-state probability of

11

this channel can be increased by cyclic AMP-dependent protein kinase-mediated
phosphorylation, linking its physiological responsiveness to secretin-activated signal
transduction pathways. This phosphorylation step also increases the channel's sensitivity
to calcium. Other studies using perfused pancreatic duct segments have suggested the
presence of high-conductance potassium channels in duct principal cells that are
insensitive to TEA, but regulated by pH (Novak, 1990; Novak & Pahl, 1993), bicarbonate
(Novak & Greger, 1991), or acetylcholine (Plant et. al., 1992). A nonselective cation
channel of low conductance (25 pS) also has been reported (Gray & Argent, 1990).

The involvement of Na*,K*-ATPase and carbonic anhydrase activity in duct
bicarbonate secretion also is well established. As noted above, both enzymes are
expressed at very high levels in duct epithelial cells, and the Na*-K+ pump inhibitor,
ouabain (Kuijpers et. al., 1984; Evans et. al., 1986), and carbonic anhydrase inhibitor
acetazolamide (Rzeder & Mathisen, 1982) each reduce ductal fluid secretion. The
inhibitory effects of ouabain include loss of pump current and a rundown of ionic sodium
and potassium gradients maintained by the pump (N ovak & Greger, 1988a). The
postulated role of these two enzymes in ductal bicarbonate secretion is illustrated in
Figure 3. A carbonic anhydrase-independent pathway for HCO,‘ secretion may also exist,
as suggested by failure of acetazolamide to inhibit bicarbonate secretion in the rabbit
pancreas (Kuijpers et. al., 1984). Recently, a basolateral, Na*-dependent carrier for
HCO,’ also has been proposed to exist in the pancreatic duct epithelial cells of rats and
pigs (V illanger et. al., 1995, Zhao et. al., 1994).

The role of Na”/H* exchange in bicarbonate secretion is at present controversial.

12

Movement of protons across the contraluminal surface of the duct epithelium is necessary
for secretion of HCO,‘ into the duct lumen, and studies with isolated rat (Stuenkel et. al.,
1988; Novak & Greger, 1988a; Zhao et. al., 1994) and pig (V illanger et. al., 1995)
pancreatic duct segments have demonstrated the presence of a carrier-mediated Na*/H+
exchange mechanism sensitive to amiloride. However, in the isolated rabbit pancreas,
amiloride at concentrations as high as 1 mM has little effect on duct bicarbonate secretion
(Kuijpers et. al., 1984). Secretin-stimulated bicarbonate secretion also is not inhibited by
amiloride in pig pancreatic ducts (V eel et. al., 1992). In addition, Raeder (1992) has
suggested that proton pumps sequestered in cytoplasmic tubulovesicles in resting
principal cells of pig pancreatic ducts are inserted into the basolateral plasma membrane
upon secretin stimulation. This hypothesis is based on the observation that numerous
small vesicles evident in the cytoplasm of unstimulated duct principal epithelial cells
disappear during secretin stimulation, paralleled by an increase in basolateral membrane
surface area (Buanes et. al., 1987). These vesicles are acidic in content, as demonstrated
by acridine orange fluorescence (V eel et. al., 1991), and their clearance from the cytosol
can be blocked by colchicine, a microtubule poison (Veel et. al., 1990). Further support
for the involvement of proton pumping in regulated bicarbonate secretion comes from the
observation that bafilomycin A,, a specific inhibitor of vesicular H+ pumps (Bowman et.
al., 1988; Yoshimori et. al., 1991), also blocks secretin-evoked proton flux in isolated pig
(V illanger et. al., 1995) and rat (Zhao et. al., 1994) pancreatic ducts. Similarly, DCCD, a
general inhibitor of proton pumps, reduces bicarbonate secretion by the in vivo pig

pancreas (Grotrnol et. al., 1986b). This process clearly could provide an alternate

13

pathway for proton extrusion during active bicarbonate secretion, although it is not
known whether this mechanism is generally present in pancreatic duct epithelial cells of

mammalian species other than the pig and rat.

S-El'l'lBl' [5'] S'

Secretion of bicarbonate-rich fluid by the in vivo pancreas is markedly stimulated in
most mammalian species by both secretin and vasoactive intestinal peptide (VIP). By
contrast, acetylcholine and cholecystokinin each evoke a chloride-rich secretion that is
thought to emanate from acinar cells (Case & Argent, 1993). Early studies using rats fed
a copper-deficient diet, which selectively destroys acinar tissue, indicated that
cholecystokinin and acetylcholine are not direct agonists of ductal bicarbonate secretion
(Ftilsch & Creutzfeldt, 197 7), although more recent studies have shown that atropine
infusion can block up to 80% of pancreatic juice produced in response to secretin (Magee
& Naruse, 1984; Singer et. al., 1986; Kuvshinoff et. al., 1993a). This observation
suggests that cholinergic innervation may potentiate the stimulatory effects of secretin on
duct epithelial cells. In vivo studies in humans (You et. al., 1983; Adler & Beglinger,
1990) and dogs (Fink et. al., 1994) also support a potentiating role for CCK on pancreatic
fluid and bicarbonate output. A number of other neurohumoral agents have been shown
to stimulate pancreatic bicarbonate output in vivo, including adrenergic agonists, gastrin-
releasing peptide, neurotensin, insulin, PHI, and some prostaglandins, while somatostatin,
pancreatic polypeptide, peptide YY, glucagon, neuropeptide Y, and vasopressin have

inhibitory effects (Case & Argent, 1993).

14

The difficulty of isolating effects of potential secretagogues on ductal epithelial
secretions from acinar-derived secretions has provided impetus for the use of isolated
duct preparations for identification of agonists. However, only a few of the many
possible physiological regulators have been tested for potential effects on levels of
intracellular messengers or fluid and electrolyte secretion in isolated ducts or in primary
cultures of duct epithelial cells.

Recent studies utilizing hamster pancreatic duct cells maintained for up to 26 weeks
in primary culture (Hubchak et. al., 1990) and rat interlobular ducts maintained in Short-
term culture (Argent et. al., 1986) have demonstrated secretin-evoked dose-dependent
increases in intracellular cyclic AMP levels. In rat interlobular ducts, an increase in
secretion of fluid into the duct lumen accompanied the elevation of cyclic AMP levels. In
this study, dibutyryl cyclic AMP also increased duct fluid secretion, although the CCK
analogue caerulein had no effect. More recent studies by Argent and coworkers (Ashton
et. al., 1991; Ashton et. al., 1993) have demonstrated that acetylcholine and bombesin
also stimulate fluid secretion from isolated rat pancreatic ducts, observations which
suggest that agonists that utilize the inositol phospholipid signal transduction pathway
also may directly regulate ductal bicarbonate secretion. The observations that cholinergic
agonists increase intracellular free calcium in acutely isolated rat and guinea pig
interlobular duct principal cells (Evans et. al., 1992; Stuenke1& Hootman, 1990), and the
presence of cholinergic receptors on duct epithelial cells (Hootman et. al., 1993), support
this assumption. Recent secretory and electrophysiological studies also support a role for

VIP, neurotensin, and ATP in regulation of pancreatic duct bicarbonate secretion (Ashton

15

et. al., 1990; Hug et. al., 1994; Pahl & Novak, 1993). These studies indicate that the
effects of ATP, neurotensin, and bombesin are possibly mediated via elevations of
intracellular free Ca2+.

Potentiating interactions of secretin and CCK in eliciting ductal fluid and bicarbonate
secretion have so far not been observed in vitra, and the cellular level at which such
potentiation occurs has not been identified. Nonetheless, CCK-evoked increases in
intracellular calcium have been demonstrated in several studies on pancreatic duct cells
(Smith et. al., 1991; Smith et. al., 1993; Zhao et. al., 1994; Hug et. al., 1994), suggesting
that similar mechanisms may underlie the potentiating effects of carbachol, CCK, and
those shown for neurotensin (Sakamoto et. al., 1988).

Physiological mechanisms of inhibitory control of pancreatic bicarbonate secretion
have been little studied, although Argent and coworkers (Ashton et. al., 1992; Argent et.
al., 1991) recently reported that both substance P and phorbol esters inhibit secretin-
stimulated fluid secretion by isolated rat pancreatic ductules. They conclude that the
observed inhibitory effects resulted from activation of protein kinase C and occurred at a
level distal to the generation of intracellular messengers. An in viva inhibitory effect of
somatostatin on pancreatic duct fluid and bicarbonate secretion (Gullo et. al., 1987)
exists, but its mechanism also is not well understood. The recent findings of
somatostatin-positive endocrine cells in the duct epithelium of several species (Park &
Bendayan, 1992; Hootman et. al., 1996) indicates a possible role of this hormone in

regulating ductal secretion.

16
6.8 l. [E .1: ll 5 .

Goblet cells are present in appreciable numbers in the main and interlobular ducts of
the exocrine pancreas and in the periductal stroma (Kern, 1993; Kodama, 1983; Geleff &
Bock, 1984). Mucins have been purified from human pancreatic juice and appear to be
similar in their overall biochemical characteristics to intestinal mucins (Lee et. al., 1991).
It has been suggested that pancreatic mucus forms a protective barrier against backflux of
bicarbonate ions and degradation of the duct epithelium by digestive enzymes (Reber et.
al., 1979), although these hypotheses have not been tested critically.

Little is currently known of the mechanisms of regulation of pancreatic mucus
secretion. Because they represent a minority cell type within the ducts, pure goblet cell
populations have not been obtained and their scarcity makes them difficult to study.
However, the recent application of morphometric procedures in this laboratory has
allowed us to begin to identify the physiological regulators of mucus secretion in isolated
guinea pig pancreatic ducts. Carbachol, VIP, and bombesin, but not secretin or .
cholecystokinin, cause goblet cell degranulation in this model (Hootman & de Ondarza,
1995). Since acetylcholine, VIP, and gastrin-releasing peptide (the mammalian analogue
of bombesin) are all found in intrapancreatic nerves (De Giorgio et. al., 1992; Holst,
1993), these observations suggest that neural rather than hormonal input regulates ductal
mucus secretion. Degranulation of goblet cells is furthermore evoked by agents which
raise intracellular free Ca2+ and cyclic AMP levels and by activation of protein kinase C
(Wilkins et. al., 1994), suggesting that multiple signal transduction pathways participate

in the regulation of this process.

17

B. Pathophysiology

1.2.13.1 .

Cystic fibrosis results in pancreatic insufficiency in 80-85% of persons affected with
this disease. Histological examination of the pancreata of cystic fibrosis patients
characteristically reveals duct obstruction, atrophy of acinar tissue, and the presence of
extensive interstitial fibrosis (Lebenthal et. al., 1993). While digestive enzyme secretion
from acinar cells is within the normal range in some affected individuals, fluid and
electrolyte secretion is significantly impaired at all levels of digestive enzyme output
(Kopelman et. al., 1988), resulting in luminal protein hyperconcentration and inspissation
of secreted macromolecules.

At the cellular level, cystic fibrosis is manifest as a defect in electrolyte transport
resulting from a mutation in the CF TR protein, which current evidence suggests regulates
chloride conductance of the apical plasma membrane of polarized epithelial cells in
respiratory and gastrointestinal tissues (Quinton, 1990). Hereditary mutations in the
CF TR protein may decrease the ability of cyclic AMP to stimulate opening of the
regulated anion channel or may cause the mutant protein to be misdirected during
biosynthesis, so that very little becomes localized in the apical plasma membrane of the
epithelial cell (Morris et. al., 1993). The consequence of these mutations in pancreatic
ducts is a decrease in the volume of bicarbonate-rich fluid secreted in response to secretin
and other hormonal regulatory molecules. Recent irnmunofluorescence studies of CF TR

protein expression in the pancreas have shown that it is exclusively localized to the apical

18

plasma membrane of principal epithelial cells in intralobular and interlobular ducts
(Marina et. al., 1991). Unpublished studies in our laboratory using the same antibody
have indicated that the CF TR protein also is expressed at high levels in the interlobular
and main pancreatic ducts of the guinea pig. Although electrophysiological studies have
indicated that the CF TR protein is the molecular entity responsible for the cyclic AMP-
regulated chloride conductance of the apical plasma membrane of the pancreatic duct
principal cell (Gray et. al., 1988), it is not clear how this protein is involved in fluid
secretion elicited by other physiological regulatory molecules, including those that utilize

the inositol phospholipid signalling pathway.

2. Pmmatitis

Acute pancreatitis has been described as an autodigestion of the pancreas that is
characterized by edema, inflammation, and cell necrosis (Samer, 1993). Chronic
pancreatitis usually is associated with luminal precipitates or calculi that cause damage to
the ductal epithelium. The etiology of pancreatitis may include duct blockage by
gallstones, hyperlipidemia, excessive alcohol consumption, or acute trauma (Steer, 1989).
In addition, factors that increase the permeability of the duct epithelium, such as reflux of
bile into the main pancreatic duct or an increase in ductal intraluminal pressure, also
cause pancreatitis, probably by allowing the seepage of pancreatic juice into the
interstitium of the gland.

Several model systems have been developed that allow experimental induction of

acute pancreatitis. These include repeated injection of caerulein or CCK, choline-

l9

deficient, ethionine-supplemented dietary regimens, and infusion of bile salts into the

duct lumen (Bettinger & Grendell, 1991). Morphological changes in the duct epithelium
induced by these procedures include blebbing of the apical membranes of duct cells, loss
of microvilli, widening of intercellular spaces, fragmentation of occluding junctions, and

regional epithelial desquamation and detachment (Farmer et. al., 1989).

3. Bancreatisfiancer

Most pancreatic cancers, 75-90% of which are adenocarcinomas. arise within the duct
system (Connolly et. al., 1987). The prognosis for survival after diagnosis is poor, with a
5-year survival rate of 1% (Lack, 1989). Neoplasias may arise most frequently from cells
within the duct system because of the high differentiating potential of these cells, since
they are closely related to the embryonic protodifferentiated state and exhibit the capacity
for transdifferentiation. Determination of the origin of a specific tumor often is
complicated by the fact that acinar cell neoplasms may de-differentiate into ductal
complexes (Scarpelli et. al., 1991).

Although radiation, viral infection, and chemicals such as nitrosamines have been
established as general cancer-causing agents, the detailed roles of these factors in the
initiation and progression of pancreatic cancers are not well understood, nor are the roles
of hormones and growth factors in the promotion and metastasis of pancreatic tumors.
Insulin, EGF, TGF a, interleukin-3, and basic fibroblast growth factor have been reported
to stimulate growth of pancreatic tumor cells in vitra (Dippold et. al., 1992; Beauchamp

et. al., 1990), while bombesin, somatostatin, larninin, chloroquine, and sodium butyrate

20

inhibit pancreatic tumor cell growth (Mai et. al., 1990; Farre et. al., 1992; Szende et. al.,
1990; Zeilhofer et. al., 1989; Bloom et. al., 1989). Cholecystokinin has been reported to
inhibit N-nitro-bis-2-oxopropyl-amine (BOP)-induced carcinogenesis in the hamster
pancreas (Pour et. al., 1988).

Mutations of the c-K-ras oncogene are frequently found in pancreatic neoplasms of
putatively ductal origin (Plinoguera et. al., 1988) and in pre-malignant mucus cell
hyperplasias (Yanagisawa et. al., 1993). Mutations in the tumor-suppressor p53 gene
(Scarpa et. al., 1992) and abnormal expression of an 89-kD heat-shock protein (Gress et.
al., 1992) also have been reported, while analysis of some adenocarcinomas and the
adenocarcinoma-derived cell lines Panc-l and Capan-l has revealed amplification and
overexpression of c-erb-B-Z, a gene coding for an EGF-like receptor tyrosine kinase
(Williams et. al., 1991). Neoplastic pancreatic cells also may express novel antigens not
seen in the cells which gave rise to the tumor, including gastrointestinal and endocrine
markers (Sessa et. al., 1990; Pour & Bell, 1989). The molecular mechanisms through
which mutations in these various genes regulate the progression of growth and metastasis

of tumors of ductal origin remains to be determined.

4. Diabetes

Pancreatic exocrine secretion in patients with diabetes mellitus is frequently abnormal
(Chey et. al., 1963; Bock et. al., 1967). The secretory dysfunctions include a reduction in
both protein output and fluid and bicarbonate secretion, conditions which appear most

frequently among juvenile diabetics (Frier et. al., 1976; Domschke et. al., 1975; Lankisch

21

et. al., 1982). In particular, the secretory response to exogenous CCK and secretin is
depressed significantly in diabetic patients, indicating multiple sites of dysfunction in the
cellular pathways which regulate acinar and ductal pancreatic secretion. The cellular
mechanisms which underlie these defects are poorly understood, a circumstance which
has stimulated the development of animal models that exhibit the exocrine secretory
abnormalities seen in diabetic patients (Balk et. al., 1975; Adler & Kern, 1975). Recent
evidence suggests that insulin does indeed play a role in potentiating hormone-evoked
pancreatic secretion (Lee et. al., 1990; Lee et. al., 1994; Lee et. al., 1995), an effect which
may in part be mediated via somatostatin (Lee et. al., 1995; Matsushita et. al., 1993). The
intracellular pathways affected in the duct epithelium of the pancreas in diabetes mellitus

have not been identified.

C. Rationale and Specific Aims

The functional abnormalities of pancreatic duct fluid and bicarbonate secretion
observed in cystic fibrosis, pancreatitis, and diabetes mellitus point out the need for a
more thorough investigation into the cellular mechanisms through which ductal secretion
is accomplished. In particular, the intracellular signalling cascades involved in regulation
of bicarbonate and fluid secretion are as yet poorly understood, and few studies have
made any attempt at characterizing their interactions within duct principal cells. The

presence of neurotransmitters, such as VIP, gastrin-releasing peptide (GRP), and

22

acetylcholine, in intrapancreatic nerve terminals in close proximity to duct epithelial cells
(De Giorgio et. al., 1992; Holst, 1993), as well as the previously detailed observations
that several neurohumoral agents besides secretin evoke ductal secretion in vitra, support
the possibility of a physiological role for these agents and suggest involvement of
multiple intracellular signalling pathways. It is not clear, however, whether these
modulators of pancreatic duct secretion act at the level of cyclic AMP generation, at
distal intracellular sites, or even at loci outside the ductal system. The present studies
were therefore aimed at elucidating the effects of neurohumoral regulatory molecules on
intracellular levels of cyclic AMP in isolated pancreatic duct segments and in duct
epithelial monolayers, and at determining whether other regulatory molecules exert their
effects at the level of cyclic AMP generation. These protocols also permitted detection of
differential responses in main and interlobular duct segments, a question of interest since,
at least in other species, the interlobular and intralobular ducts are thought to be the
primary source of secreted bicarbonate and fluid, with the main duct serving mainly as a
passive conduit.

The subcellular mechanismswhich subserve bicarbonate secretion by the duct
epithelium of the exocrine pancreas also are still incompletely understood. Recent
studies which have attempted to clarify the bicarbonate secretory process have been
limited to only two species, the pig and rat. The present studies therefore were
undertaken to investigate whether the secretory mechanisms suggested for these two
species are common to other mammals by developing a secretary model for the guinea

pig pancreatic duct system. Other advantages of the guinea pig model are the high

23

bicarbonate concentration elicited (up to 150 mEq/l), the relative ease of isolation of the
ducts, and the suitability of the guinea pig as a model for the study of experimental
diabetes and pancreatitis, both of which have been induced successfully in this laboratory.
The present studies were accordingly carried out to determine the presence or absence of
several proposed ion transporters in the guinea pig duct epithelial cell, and to test the
hypothesis that agonist-evoked proton pumps participate in the bicarbonate secretory
process.

The specific aims addressed by the present studies are therefore as follows:

a. To characterize the cyclic AMP response to secretin.

b. To determine whether main and interlobular pancreatic ducts exhibit differential
responses in cyclic AMP generation to secretin exposure.

c. To determine whether neurohumoral agents other than secretin that are known to
alter pancreatic secretion in viva do so by altering cyclic AMP levels in the
pancreatic duct epithelial cells.

2. 0‘1.l"‘t' o .0'!‘ WV. 0 or "l'l‘lel'fit‘Wi‘i 't
‘_. r A.” . .1.‘.h‘nu.ht-“-._t ... H. .1. .t .
. . E. ll l H l
a. To determine whether NaVH+ exchangers, Cl'lI-ICO,‘ exchangers, and NaVHCO,‘

cotransporters are present in guinea pig pancreatic duct principal cells.

24

b. To determine whether vesicular H+-ATPases are recruited during stimulated

bicarbonate secretion from pancreatic duct principal cells.

11. MATERIALS AND METHODS
A. Materials ‘

Soybean trypsin inhibitor (type 2-S), Eagle's minimum essential medium amino acids
(50x concentrate), L-glutamine, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid
(HEPES), 3-isobutyl-1-methylxanthine (IBMX), antibiotic-antimycotic powder, fetal
bovine serum, calcitonin gene-related peptide (CGRP), somatostatin, insulin, glucagon,
isoproterenol, triethylamine, acetic anhydride, cyclic AMP, rabbit anti-cyclic AMP
antiserum, amiloride hydrochloride, bafilomycin A,, cytochalasin D, methyl-[5-(2-
thienylcarbonyl)-l-H-benzimidazol-2-yl]-carbamate (nocodazole), mouse monoclonal
anti-B-tubulin, fluorescein isothiocyanate (F ITC)-labelled lectin from T riticum vulgaris
(wheat germ agglutinin), 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid (DIDS), and
carbarnylcholine chloride (carbachol) were purchased from Sigma Chemical (St. Louis,
MO). Bovine serum albumin (BSA) was obtained from ICN Biochemicals (Irvine, CA);
chromatographically purified collagenase (693 and 780 units/mg) from Worthington
Biochemical (Freehold, NJ); and secretin, neurotensin, bombesin, vasoactive intestinal
peptide (VIP), and substance P from Bachem California (Torrance, CA). 5-(and-6)-
carboxy-SNARF -1 , acetoxymethyl ester (SNARF -1), FITC-conjugated goat anti-mouse

IgG, rhodamine-conjugated phalloidin, Slow-Fade, and nigericin were purchased

25

26

from Molecular Probes (Eugene, OR), and Triton X-100 from Research Products
International (Elk Grove Village, IL). Epidermal growth factor (EGF), insulin-
transferrin-seleniurn premix, and Cell-Tak were purchased from Collaborative
Biochemical Products (Bedford, MA), CMRL-1066 culture medium from Gibco
(Gaithersburg, MD), and [‘2’I]-cyclic AMP from New England Nuclear (Boston, MA).
Cholecystokinin octapeptide (CCK-8) was a gift fiom the Squibb Institute for Medical
Research (Princeton, NJ). Male outbred Hartley guinea pigs (300-400 gm) were obtained

fiom the Michigan Department of Public Health (Lansing, MI).

B. Isolation of Pancreatic Ducts

Pancreatic ducts were isolated by a digestion/dissociation procedure developed
recently in this laboratory (Hootman & Logsdon, 1988). Guinea pigs were fasted
overnight to reduce pancreatic secretory activity to a minimum, and were euthanized by
exsanguination following carbon dioxide-induced anesthesia. Pancreata were removed
and bathed in a HEPES-buffered Ringers solution consisting of (in mM): 118 NaCl, 4.7
KCl, 1.1 MgClz, 5.5 glucose, 1.0 NazHPO4, 10 HEPES, and 1.3 CaCl, (adjusted to pH 7.4
with 2.0 N NaOH), supplemented with 0.1% BSA, 0.01% soybean trypsin inhibitor (type
2-S), Eagle's minimum essential medium amino acids, and 2 mM L-glutamine. Pancreata
were then trimmed of blood vessels and mesentery, injected with Ringers solution

containing 100 U/ml of collagenase, and incubated in 5 m1 of this enzyme solution for 90

27

min at 37°C. Subsequently, pancreata were dissociated by mechanical shearing and
segments of the pancreatic duct system were manually selected using a dissecting
microscope and incubated for an additional 60 min at 37°C in Ringers containing 100
U/ml of collagenase. Segments of main and interlobular ducts then were separated from
adhering acini and vascular elements using ultrafme forceps, and cut into short (1-2 mm)

segments using iridectomy scissors.

C. Regulation of Cyclic AMP Levels in Isolated Pancreatic Ducts and Duet

Epithelial Monolayers

LE’ Cl [11 15.1““!

Isolated duct segments obtained by the procedure described above were cut into
small pieces (< 0.5 mm), rinsed twice with CMRL-1066 culture medium supplemented
with 0.5 mM IBMX, 0.68 mM L-glutamine, 0.01% soybean trypsin inhibitor (type l-S),
5 ug/ml of insulin, 5 ug/ml of transferrin, 5 ng/ml of sodium selenite, 100 U/ml of
penicillin, 10 ug/ml of streptomycin, 50 ug/ml of amphotericin B, 2.5% fetal bovine
serum, and 10 nM EGF, and distributed into the wells of a 12-well tissue culture plate in
2.0 ml of this culture medium. Duct explants were maintained at 37°C in an atmosphere
of 5% carbon dioxide in air. The culture medium was replaced every 3-4 days, and by 10
days, concentric monolayers of principal epithelial cells had grown around the sites of the

original explants (Figure 2).

28
2. lncuhaticnfmtmnls

Acutely isolated duct segments were distributed into siliconized 20 ml glass vials,
rinsed once with HEPES-buffered Ringers solution (HR), and in all cases preincubated
for 30 min at 37°C in HR. Ducts then were preincubated for a second 30 min interval at
37°C in 1.0 ml of HR with or without the phosphodiesterase inhibitor IBMX, and then
incubated in 0.6 ml of HR (:1: IBMX) with or without potential secretagogues. An IBMX
concentration of 3 mM was used throughout, except when determining the dose-effect
relationship for IBMX. In this case, IBMX concentrations of 30 uM - 3 mM were used.
Incubations with agonists were terminated after 5-30 min by addition of 0.15 ml of 0.2 N
HCl.

Cultured monolayers of duct epithelial cells were rinsed twice with HR prior to
preincubation with or without IBMX and exposure to potential agonists as above.
Secretin was tested at concentrations of 10 pM - 1 uM to determine its potency and
efficacy on inducing cyclic AMP accumulation. Other potential regulators of duct
secretion were tested at the following concentrations: bombesin (10'8 M), carbachol (10'5
M), CCK-8 (10'9 M), CGRP (10‘8 M), glucagon (10'7 M), insulin (10" M), isoproterenol
(10'5 M), neurotensin (10'7 M), PGE, (10'7 M), somatostatin (10'7 M), substance P (10‘7

M), and VIP (107 M).

3.21.,”“3'”

Intracellular cyclic AMP levels were determined by a procedure modified from that of

Harper and Brooker (1975). Samples (750 pl) of acidic cell lysates were neutralized with

29

400 pl of 0.5 M Nazi-IPO4, diluted to 2.0 ml with 0.05 M sodium acetate buffer (pH 6.2),
and sonicated for 20 sec to disperse any remaining tissue fragments. 100 pl of ZOO-fold
dilutions of these samples were transferred into duplicate RIA tubes and acetylated with 5
ul of a 2:1 triethylamine to acetic anhydride mixture prior to overnight incubation at 4°C
with 100 ul of anti-cyclic AMP antiserum and 100 pl of [”51]-cyclic AMP (0.007
uCi/tube). Bound [mu-cyclic AMP was measured by gamma counting following ethanol
precipitation, centrifugation (15 min at 1500 x g), and removal of supernatant. Cyclic
AMP concentrations in cell lysates were determined by comparison to a series of known
cyclic AMP standards (Figure 4), and were normalized to protein concentration in

sonicated samples, as described below.

4. WW):

A modified Coomassie Blue binding assay (Bradford, 1976) was used to measure
protein concentration in each sample. Duplicate 0.5 ml aliquots of sample fluid were
transferred to polyethylene tubes, combined with 0.2 ml of distilled water and 0.1 ml of
0.1 N NaOH, and vortexed. After 10 min, 0.2 ml of the protein assay dye reagent
concentrate was added to each sample tube, and each tube was vortexed again and
allowed to sit for 10 min. Samples then were transferred to a microcuvette (1 ml
volume), and absorbence was read at a wavelength of 595 nm using a DU-62
Spectrophotometer (Beckman Instruments; Fullerton, CA). Protein concentrations were

determined from a standard curve obtained using known amounts of BSA.

30

1.0

 

0.9..

0.8-

0.71

0.6.

0.5.

BIB0

0.4 ..

0.3-1

0.2a

0.1-

 

 

0.0 . .....-., . .......r . ....., - --f.fi.
0.01 0.10 1.00 10.00 100.00

[cydic MP] (WM)

 

Figure 4. Standard curve of B/B0 vs. concentration of cyclic AMP. Known cyclic AMP
concentrations of 0.03, 0.10, 0.30, 1.0, 3.0, 10, and 30 pmol/ml were used for each standard
curve determination, and bound [‘”I]-cyclic AMP ("B") was measured by radioimmunoassay
and expressed as a fraction of the ['251]-cyclic AMP bound in the absence of competing
unlabelled ("cold") cyclic AMP ("Bo").

31

D. Confocal Microscopic Analysis of Isolated Pancreatic Ducts

LB .1 [I S .2 E ”1'

The involvement of several proposed H+ and HCO,‘ transporters in pancreatic duct
bicarbonate secretion was studied through examination of the intracellular pH of viable
duct epithelial cells, a procedure which benefitted greatly from the advent of laser-
scanning confocal microscopy. The major advantages of this technique are improved
lateral resolution, the capability to focus on a section of an intact tissue without
interference from out-of-focus structures, and the digital image storing and processing
features of the equipment. Fluorescent light emitted from the Specimen is passed through
a narrow aperture (confocal slit) which allows only in-focus light to reach the
photodetectors, while a tightly focused laser beam excites only a tiny portion of the
specimen at a time. A complete image is formed when the collected in-focus
fluorescence emission intensities are reconstructed by the Image] software and displayed
in "real-time" (30 frames per second) on an image monitor. The major limitation of this
technique is the limited availability of fluorescent probes which are responsive to the
excitation wavelengths of the laser. However, continuous advances in this field have
increased greatly the number of physiological fluorescent tracers on the market. The
Noran Odyssey laser scanning confocal microscope (LSCM) used in the current studies
contains an argon ion laser with excitation wavelengths of 488 nm, 457 nm, and 529 nm,
suitable with the use of such probes as fluorescein, rhodamine, and the pH-sensitive dye

SNARF-l. By inserting appropriate filters into the emission pathway of fluorescent light,

32

more than one emission wavelength can be monitored at a time, permitting ratiometric

imaging of intracellular ion concentrations.

2.1 . [E if

Initial imaging studies of pancreatic duct segments using the confocal microscope
were done on fixed tissues to gain familiarity with the instrument. In addition, imaging
of fixed tissues was used to localize microfilaments (Figure 5), microtubules, and mucins
in the intact duct epithelium. For labelling of microfilaments and mucins, isolated duct
segments were rinsed twice in phosphate-buffered saline (PBS), fixed for 60 min at room
temperature in PBS containing 2% formaldehyde, rinsed 3-5 times in PBS, and
permeabilized for 30 min in PBS containing 0.5% Triton X-100 and 0.1% BSA. Duct
segments then were rinsed again in PBS and incubated for 90 min at room temperature in
PBS containing 0.5% BSA, 0.5% Triton X-100, and 5 U/ml of rhodamine-conjugated
phalloidin, an F-actin probe, or 10 uM F ITC-conjugated wheat germ agglutinin, a lectin
known to bind mucins of the pancreatic duct system (Geleff & Bock, 1984). To localize
microtubules, isolated duct segments were rinsed once in a microtubule-stabilizing buffer
(MSB; in mM: 100 PIPES, 2 EGTA, and 4% polyethylene glycol, pH 6.9) prior to
fixation and permeabilization for 30 min at 37°C in MSB containing 2% formaldehyde
and 0.5% Triton X-100 and blocking with 10% BSA in PBS. Fixed duct segments were
incubated overnight at 4°C in a 1:100 dilution of a monoclonal mouse anti-B-tubulin
antibody in PBS and for 1 hr at 37°C in a 1:100 dilution of an FITC-conjugated goat anti-

mouse IgG in PBS also containing 5 U/ml of the F-actin probe, rhodamine-phalloidin.

33

 

Figure 5. Confocal microscopic image of the pancreatic duct epithelium labelled with the
F -actin probe, rhodamine-phalloidin. Isolated, fixed pancreatic duct segments were
incubated with 5 U/ml of the fluorescent dye prior to confocal microscopic imaging at an
excitation wavelength of 529 nm. Emitted light was collected and digitally stored, and is
displayed in false color. white and red indicating the most intense regions of fluorescence.

34

Prior to imaging, all duct segments were mounted in Slow-Fade anti-fading reagent
on a glass slide with raised edges to avoid crushing the tissue, and coverslips were glued
into place using clear nailpolish. Images of rhodamine-labelled microfilaments were
obtained at an excitation wavelength of 529 nm, and emitted light was passed through a
550 nm barrier filter and a 590 nm longpass filter before reaching a photomultiplier tube
(PMT). Emission intensities were collected on a pixel-by-pixel basis and stored in
computer memory, from which the scanned image was reconstructed using the Image]
image processing software (Universal Imaging; West Chester, PA). Each single scan of
the selected tissue region was of 33 msec duration, enabling multiple scans of the tissue
in a short time period and reducing photobleaching. In its "live" mode, a continuous scan
of the tissue permits searching the specimen for regions of particular interest. Images of
FITC-labelled mucins and microtubules were obtained by selecting an excitation
wavelength of 488 nm and filtering emitted light through a 515 nm barrier filter and a
520-560 nm bandpass filter to collect the peak F ITC emission at 540 nm. FITC and
rhodamine images were superimposed through the Image] software when working with

double-labelled images.

a. SNARF -1 Loading and Superfusion of Pancreatic Ducts
Segments of isolated interlobular pancreatic ducts were loaded for 60-90 min at 37°C

with 20 pM SNARF -1 in a HEPES-buffered Ringers solution (KRH) (Table 1).

Table l.' Composition of physiological buffers used.

35

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Buffer NaCl NMQ Na Choline 1111:1491 HEPES
Cl 119.03 HQQ:
KRH 118 0 0 0 0 25
KRB 98 0 25 0 0 20
Na-free KRH 0 118 0 0 0 25
Na-free KRB 0 98 0 25 0 20
KRH pulse perfusate 98 0 0 0 20 25
KRB pulse perfusate 78 0 25 0 20 20

 

 

Concentration values are in mM. Each buffer also contained (in mM): 4.8 KCl, 12 MgSO,,,

11 glucose, 1.2 KH2P04, and 1.3 CaCl2 (adjusted to pH 7.4 with 2.0 N NaOH or Trizrna
base).

36

Superfusions were carried out in KRH or in a bicarbonate-containing buffer (KRB),
modified where noted as described in Table l. Dye-loaded duct segments were attached
with Cell-Tak to 40 mm diameter glass coverslips which were mounted in a 0.1 ml
volume Focht Live Cell Perfusion Chamber (Bioptechs, Inc.; Butler, PA) on the stage of
a Nikon Diaphot II inverted microscope (Figure 6). Buffer reservoirs were warmed by a
heating plate and buffers oxygenated by bubbling 100% 02 into the reservoirs. The
chamber and the objective lens were maintained at 37°C, and buffer flow was adjusted to
a rate of 3 mI/min. A rotary teflon valve permitted switching between six separate
perfusing solutions.

Secretagogue effects on pHi were determined by addition of 100 nM secretin or 10
uM carbachol to the superfusing solution in the absence or presence of 0.5 mM amiloride
and/or 30 nM bafilomycin A,. Sodium-free buffers were prepared by equirnolar
replacement of sodium chloride with N-methyl-D-gluconate (NMG*) chloride and of
sodium bicarbonate with choline bicarbonate, and NH4C1 pulse perfiisates by 20 mM
replacement of NaCl with NH4C1 (Table 1). Chloride-free buffers were prepared by

equirnolar replacement of all chloride salts with gluconate salts.

b. Live Cell Imaging of the Duct Epithelium.

SNARF -I -loaded principal epithelial cells of pancreatic duct segments were
identified under brightfield illumination prior to confocal imaging using a 60x oil
immersion objective. The dye was excited at a wavelength of 488 nm by the argon ion

laser of the Noran Odyssey LSCM. When excited at this wavelength, SNARF -1

37

—— Emitted light path

    
 

Reservoir
Flow Rate
_, 3 ml/min Perfusion Chamber

with coverslip

    

 

 

 

 

 

I .11" QEONFQCAL, ., . . . . '

 

Figure 6. Schematic for superfusion of isolated segments of pancreatic ducts in a Focht
live cell perfusion chamber. Perfusate from a reservoir is gravitationally fed through the
perfusion chamber at a flow rate of approximately 3 ml/min. A teflon rotary valve
permits superfusion with up to six different solutions. Pancreatic duct segments are
attached with Cell-Tak, a biopolymer, to a glass coverslip which constitutes the bottom of
the chamber, from which a 60x oil immersion objective collects emitted fluorescent light.
Emitted light is collected by two separate photomultiplier tubes (PMTs) and digitally
processed by the Imagel imaging software of the Noran Odyssey LSCM.

38

fluoresces with peak emissions at 580 nm and 640 nm. The emission intensity at each of
these wavelengths is pH-dependent (Figure 7), thus enabling ratiometric pH
measurements. Emitted light was passed through a 515 nm barrier filter and a 610 nm
beamsplitter, and bandpass filters (630-650 nm and 570-590 nm) were used to filter
emitted light arriving at two photomultiplier tubes in order to collect separate 640 nm and
580 nm emissions (Figure 8). Pairs of images (640 um and 580 nm emission) were stored
on an optical disk at 10 sec intervals for the duration of the perfusion for later intensity

measurements using Imagel image processing software.

a. Determination of pH,

Fluorescence intensity was measured for each cell at each recorded time interval, and
raw data transferred to the GraphPad Prism data analysis software (GraphPad; San Diego,
CA). Following subtraction of background fluorescence intensity, the ratio of
fluorescence emission intensity (640 rim/580 nm) was calculated for each individual cell
at all time intervals, and pH, was determined from a calibration curve obtained by the
high-K+ nigericin method (Thomas et. al., 1979). Duct segments were incubated for 5
min in a high-potassium calibration buffer consisting of (in mM): KCl 140, MgCl, 1.0,
HEPES 20, adjusted to pH values of 6.4, 6.9, 7.4, and 7.9 with 2.0 N KOH, and
supplemented with 20 uM nigericin. Five image pairs (640 nm and 580 nm emissions)
then were collected at each calibration pH, and calculated emission ratios used to
construct a pHi vs. emission ratio calibration curve. pH values reported for respective

experiments constitute the average of observations (mean :1: SE) from 3-6 separate

39

 

 

 

 

 

 

 

i 040 run
>. W
= too.
I
C
3 1
5. no.
r:
2 1
5 to.
E 5.0 INTI
m '1
co.
cu - -ocu o~c “
3 ii 2 u t 3.5
o o ’
'“° e: - us run o
3 3.0
I
i:
70 g
'3 2.5
.2
70 E
III
2.0
00
’0
B , . C is , - , . . fl
we we no mo 0 5 to 15 20
Wavelengthinm) ' time (min)

Figure 7. SNARF-l chemical structure, emission spectra, and traces of SNARF-l emissions
and their ratio. A) Chemical structure of 5-(and-6)-carboxy-SNARF-l , acetoxymethyl ester
(SNARF-l). B) Emission spectra of SNARF -1 when excited at 488 run. As pH of the
SNARF -1 solution increases, fluorescence emission intensity at wavelengths < 610 nm
decreases while emission intensity at wavelengths > 610 nm increases. C) Traces of
fluorescence emissions of SNARF-l at 640 nm (upper trace) and 580 nm (middle trace), and
their calculated ratio (lower trace). While emission intensities at 640 nm change relatively
little with changes in pH, 580 nm emission intensities are more affected.

40

—-— Emittedlight
-——- Excitation beam

Perfusion
Chamber

 

Figure 8. Schematic for Measurement of SNARF-l Emissions. Separate 640 nm and 580
nm peak emissions can be collected by passing emitted light through a series of dichroic
mirrors and filters. A 515 nm barrier filter eliminates any reflected 488 nm light of the
excitation beam, and a 610 nm beamsplitting dichroic mirror directs wavelengths > 610 nm
to one photomultiplier tube (PMT) and wavelengths < 610 nm to a second PMT. Bandpass
filters in place in front of each PMT further restrict received light to the desired emission
wavelengths.

41

perfusions, each yielding measurements from 4-12 individual cells.

E. Statistics

Statistical differences between sample means were evaluated with one-way analysis
of variance (ANOVA) using StatView software (Abacus Concepts; Berkeley, CA), and
with Wilcoxon’s rank-sum test. Differences were considered significant at P values of <

0.05.

III. RESULTS

A. Regulation of Cyclic AMP Levels in Guinea Pig Pancreatic Duct Epithelial Cells.

1. WWW

Isolation of the guinea pig pancreatic duct system yields segments of both the main
duct, which spans the length of the pancreas, and numerous interlobular ducts (Hootman
& Logsdon, 1988). The lining of the smaller interlobular ducts consists of a simple
cuboidal epithelium, which becomes columnar in the main pancreatic duct. This
epithelium is composed of a heterogenous population of cells, of which principal cells
account for 70 - 95%. Principal epithelial cells are characterized by numerous
mitochondria, apical microvilli, and abundant basolateral plasma membrane folds
(Hootman & Logsdon, 1988). The two other cell types present in guinea pig pancreatic
ducts are mucus-secreting goblet cells, which account for 20-30% of the epithelium in the
main duct and 5-10% in interlobular ducts (Hootman & de Ondarza, 1995), and endocrine
cells, which make up approximately 0.5% of the epithelial cell population (Figure 1).

In comparison to the cellular heterogeneity of acutely isolated pancreatic ducts,
cultured monolayers obtained from pancreatic duct explants consist almost entirely of
principal epithelial cells, as goblet cells and endocrine cells do not appear to proliferate

under the culture conditions presently in use (Hootman & Logsdon, 1988). Inclusion of

42

43

IBMX and use of a low serum concentration in the culture medium suppresses growth of
fibroblasts, rendering the duct cell cultures largely free of this cell type for at least two
weeks of culture. The responses of principal epithelial cells to potential regulatory
molecules can thus be distinguished from those of other cell types using this culture
system. Figure 2 illustrates the typical appearance of a monolayer of epithelial cells

derived from an explant of pancreatic duct.

2. i', o ' '14” 7:11,.” '41 r' I. ' .‘u'

To assess the responsiveness of guinea pig pancreatic ducts to secretin stimulation,
mixed suspensions of main and interlobular duct segments were incubated for 0-30 min
in the presence of 0.1 uM secretin following a 30 min preincubation with or without 3.
mM IBMX (Figure 9). Cyclic AMP levels increased significantly over the 30 min
period, from 2.4 :1: 0.6 to 19.7 :t 2.2 pmol/ug protein in the presence of 3 mM IBMX, and
from 1.1 :i: 0.2 to 8.3 :1: 0.3 pmol/ug protein in its absence. In the absence of secretin,
cyclic AMP levels did not change significantly during this 30 min period. Cyclic AMP
accumulation was routinely two to three-fold greater in the presence of the cyclic AMP
phosphodiesterase inhibitor than in its absence. IBMX significantly potentiated cyclic
AMP accumulation at doses greater than 100 uM and was maximally effective at 3 mM,
doubling secretin-stimulated cyclic AMP levels (Figure 10). In the absence of secretin,
IBMX also increased cyclic AMP levels, although the magnitude of the increase was
much smaller than that seen in the presence of the hormone. All subsequent experiments

were carried out in the presence of 3 mM IBMX. The increase in cyclic AMP levels in

44

 

 

25.
+NOIBMX *
+3rrMIBMX
20..
*
E 15‘ *
S a
.9 5’
.5 :i.
5 10..
* *
.' *
5..
‘ 0
A
0 l I r I I T I
0 5 10 15 20 25 30
Titre (min)

Figure 9. Time course of the effect of 0.1 pM secretin on cyclic AMP levels in mixed
suspensions of acutely isolated segments of guinea pig main and interlobular pancreatic
ducts. Duct segments were preincubated for 30 min in HEPES-buffered Ringers solution
with (circles) and without (triangles) 3 mM IBMX, then incubated in the presence of secretin
with or without IBMX for the time periods indicated. Controls (open circle and triangle)
received no secretin. Cyclic AMP levels in secretin-stimulated duct segments were
significantly greater (P < 0.05) in the presence of IBMX than in its absence at all time points
beyond zero, and in all experiments significantly greater (P < 0.05) than controls where
indicated (*). Results represent means :1: SE of four experiments.

45

25.

20.

15.

cyclic AMP

(pmolIP-g protein)

 

.3
o

5.

 

 

 

V V V VVVVV' V V V VVVVV' V V V V—VVVV'

0 30 100 300 1000 3000
iBMX (0M)

Figure 10. Effect of IBMX on secretin-stimulated cyclic AMP levels in mixed suspensions
of acutely isolated segments of guinea pig main and interlobular pancreatic ducts. Duct
segments were preincubated for 30 min at 37°C in HEPES-buffered Ringers solution with
the indicated concentrations of IBMX, then incubated for 30 min with the indicated
concentrations of IBMX with (circles) or without (triangles) 0.1 uM secretin. In the absence
of secretin, IBMX had only a small effect on cyclic AMP levels. In the presence of secretin,
IBMX caused a dose-dependent increase in the cyclic AMP level of up to two-fold. Results
represent means :1: SE of four experiments.

46

the presence of secretin was dose-dependent over a range of 10 pM to 10 nM, with a
calculated EC,0 of 150 pM (Figure 11). Subsequent experiments were performed at
secretin concentrations of 0.1 nM (submaximal) and 0.1 1.1M (maximal).

It has been suggested that secretin-evoked bicarbonate secretion takes place primarily
in the intralobular and interlobular ducts of the exocrine pancreas, and that the main duct
serves mainly as a passive conduit (Case & Argent, 1993). However, because of the
prominence of the main duct in the guinea pig pancreas, we tested the responsiveness of
different portions of the pancreatic duct system to secretin stimulation. Segments of main
and interlobular duet were separated and incubated for 30 min in the presence or absence
of 0.1 uM of secretin (Figure 12). Interlobular ducts segments responded with an 8.8-
fold increase in cyclic AMP levels . While less responsive than interlobular ducts,
segments of main pancreatic duct also exhibited a significant cyclic AMP elevation (4.6-
fold) in response to secretin. This difference in responsiveness was maintained when
explants from main and interlobular ducts were cultured for 10 days. Monolayers derived
fi'om interlobular and main ducts evidenced 4.3- and 3.2-fold increases in cyclic AMP

levels, respectively.

3- RmWMMWAflm

The presence of VIP in intrapancreatic nerve terminals (Holst, 1993) suggests that it
may be involved in regulation of pancreatic duct bicarbonate secretion. We have shown
previously that pancreatic duct goblet cells degranulate in response to VIP exposure

(Hootman & de Ondarza, 1995). Todetermine the effects of VIP on pancreatic

47

20..

.L
0|
1

O

 

 

   

Cyclic AMP
(pmollllg protein)
3

5.

 

 

 

1 I I I

0 -12 -11 -10 -9 -8 -7 -6

Secretin (log molar)

Figure 11. Dose-dependence of secretin stimulation of cyclic AMP levels in mixed
suspensions of acutely isolated guinea pig main and interlobular pancreatic ducts. Duct
segments were preincubated for 30 min at 37°C in HEPES-buffered Ringers solution with
3 mM IBMX, then incubated for 30 min in HR with 3 mM IBMX and the indicated
concentrations of secretin. Cyclic AMP levels were maximal at a secretin concentration of
10 nM, with an EC50 of 150 pM. Results represent means :t SE of four experiments.

48

30-
E] Central

, .
25_ I Secretin

 

Cyclic AMP (pmol/Ilg protein)

.1
5-
I

 

 

 

 

 

 

 

 

 

 

Main Interlobular Main Interlobular

Acutely Isolated Cultured

Figure 12. Effect of secretin on cyclic AMP levels in acutely isolated guinea pig main and
interlobular ducts and cultured monolayers derived from these two segments of the duct
system. Duct segments and monolayers were incubated for 30 min at 37°C in HEPES-
buffered Ringers solution with 3 mM IBMX and with or without 0.1 uM secretin. Secretin
caused a 46-fold increase in cyclic AMP levels in acutely isolated main duct segments and
an 8.8-fold increase in interlobular duct segments. This pattern of responsiveness was
maintained in cultured duct monolayers. Results represent means :1: SE of five acute and six
cultured preparations.

49

duct cyclic AMP levels, we exposed both acutely isolated duct segments and cultured
duct epithelial monolayers to 0.1 uM VIP. Cyclic AMP concentrations in acutely
isolated main duct segments were elevated 4.5-fold over control segments, and in
interlobular duct segments 10.6-fold over controls (Figure 13). The effect of VIP on
cultured duct principal epithelial cells was similar for main and interlobular duct-derived
monolayers, a 6.1-fold elevation of cyclic AMP levels in the former and a 6.7-fold
elevation in the latter. Since goblet cells are not present in appreciable numbers in these
monolayers, the results indicate that principal epithelial cells as well as goblet cells
express VIP receptors.

Mixed main and interlobular duct preparations also were incubated with VIP and
eleven other potential modulators of pancreatic duct secretion in the presence and absence
of maximal (0.1 pM) and submaximal (0.1 nM) secretin concentrations to determine the
impact of these agents on pancreatic duct cyclic AMP metabolism (Table 2). In the
absence of secretin, only VIP significantly elevated cyclic AMP levels from 3.2 d: 0.2 to
11.8 d: 0.4 pmol/ug protein. Calcitonin gene-related peptide and prostaglandin E2 slightly
elevated resting cyclic AMP levels (not significant at P < 0.05) without any effect on
secretin-evoked cyclic AMP increases. VIP also caused a significant increase of cyclic
AMP in the presence of submaximal, but not maximal, doses of secretin, while
somatostatin significantly attenuated secretin's stimulatory effects at the submaximal
secretin concentration. Agents such as carbachol and cholecystokinin that are thought to
operate through the inositol phospholipid signalling pathway did not have a significant

effect on basal or secretin-stimulated cyclic AMP levels. Likewise, the islet hormones

50

351

I D Control
30-1
. I VIP

25.‘
201

15;

Cyclic AMP (pmol/ Mg protein)

10:

 

 

 

 

 

 

 

     

Main Interlobular Interlobular

Acutely Isolated Cultured

Figure 13. Cyclic AMP response to vasoactive intestinal peptide (VIP) in acutely isolated
guinea pig main and interlobular ducts and cultured monolayers derived from these two
segments of the duct system. Duct segments and cultures were incubated for 30 min at 37°C
in HEPES-buffered Ringers solution with 3 mM IBMX and with or without 0.1 uM VIP.
VIP caused a 4.5-fold increase in cyclic AMP levels in acutely isolated main duct segments
and a 10.6-fold increase in interlobular duct segments, as compared to 6.1-fold and 6.7-fold
increases in cultured main and interlobular duct monolayers, respectively. Results represent
means t SE of five acute and nine cultured preparations. '

51

Table 2. Effects of potential regulators of pancreatic duct secretion on resting and secretin-
Stimulated cyclic AMP levels in mixed suspensions of acutely isolated guinea pig main and
interlobular pancreatic ducts.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Agent Control 0.1 nM Secretin 0.1 uM Secretin
- + - + - +
Bombesin 2.0 :1: 0.3 2.6 i 0.2 7.2 i 0.9 8.8 i 0.9 13.1 :t 1.4 16.4 :t 3.5
Carbachol 2.6 i 0.3 3.9 i 0.6 14.8 t 1.1 15.3 :t 1.8 21.4 :t 2.0 23.2 :t 2.4
CCK-8 3.4 :i: 0.5 4.5 d: 0.8 10.6 i 0.9 15.4 i 4.7 23.8 :t 3.8 25.9 i 2.6
CGRP 2.3 a: 0.0 4.2 :t 0.4 7.6 x 1.0 8.5 s 1.2 13.1 :t 0.9 I53 : 0.8
Glucagon 3.2 i 0.4 3.0 :t 0.2 10.3 :1: 1.4 9.0 a: 1.2 18.0 :t 2.9 19.9 i 0.6
Insulin 3.1 t 0.2 3.4 :i: 0.4 8.7 :i: 0.8 11.0 :1: 1.8 16.2 i 0.5 16.9 i 0.6
lsoproterenol 2.4 :t 0.6 3.4 a: 1.0 7.7 d: 0.8 9.4 :t 3.9 16.2 i 1.9 15.5 d: 2.9
Neurotensin 2.4 i 0.3 2.5 :t 0.4 7.5 t 1.0 6.5 :t 1.0 13.8 i 1.7 16.8 :1: 1.5
PCB, 3.5 :t 0.2 6.8 i 0.8 10.3 t 1.3 11.8 :1: 0.9 23.1 :i: 2.4 20.3 :t 1.2
Somatostatin 2.9 d: 0.3 2.9 i 0.3 8.0 i 1.0 4.4 :t 0.3‘ 14.4 i 0.8 13.2 :t 1.1
Substance P 4.0 :t 1.9 2.7 i 0.4 12.9 i 6.1 9.5 i 2.5 20.6 :L- 4.6 20.4 i 3.8
VIP 3.2i0.2 ll.8:t0.4"‘ 8.7i1.0 13.411.1‘ 15.4:1: 1.1 17.3:22

 

 

 

 

 

 

 

 

Ducts were preincubated for 30 min at 37°C in HEPES-buffered Ringers solution (HR) with
3 mM IBMX, and then incubated for 30 min with 3 mM IBMX, with (+) or without (—) the
indicated concentrations of the test substance, in the absence of secretin and at secretin
concentrations of 0.1 nM and 0.1 uM. In the insulin and glucagon experiments, insulin and
glucagon were also included in the 30 min preincubation. Concentrations used were:
bombesin (10" M), carbachol (10'5 M), CCK-8 (10‘9 M), CGRP (10'8 M), glucagon (10'7 M),
insulin (10'7 M), isoproterenol (10" M), neurotensin (10'7 M), PGE, (10‘7 M), somatostatin
(10'7 M), substance P (10'7 M), and VIP (10’7 M). Results represent means :1: SE of four to
six experiments, and are expressed as pmol cyclic AMP/pg protein.

52

insulin and glucagon, the gut hormone neurotensin, the neurotransmitter bombesin, and
the beta-adrenergic agonist isoproterenol did not affect pancreatic duct cyclic AMP levels

in the absence or presence of secretin.

B. Electrolyte Transport Components of Guinea Pig Pancreatic Duct Bicarbonate

Secretion

The subsequent experiments of this project were designed to investigate involvement
of agonist regulation in electrolyte transport by guinea pig pancreatic ducts and to
demonstrate the presence of particular transport elements in duct epithelial cells. In
particular, the hypothesis that agonist-evoked hydrogen pumping is involved in regulated
bicarbonate secretion was tested. To accomplish this aim, changes in the intracellular pH
of duct principal epithelial cells were measured using the pH-sensitive dye. SNARF -1
(Cody et. al., 1993).

Following incubation of isolated segments of interlobular pancreatic ducts for 60-90
min in bicarbonate-free buffer (KRH) containing 20 uM SNARF -1 AM, principal cells of
the duct epithelium that had taken up the dye were visually identified. Initial studies
showed that pH, vs. emission ratio calibration curves from a number of individual cells
were parallel throughout the range of pH values tested (6.4 - 7.9) (Figure 14). Baseline
pH, in unstimulated principal cells superfused with KRH was stable in the absence of

inhibitors, averaging 7.35 :t 0.02 (n = 86).

53

5.

640 nmi580 nm ratio

 

 

 

0 V f 1 V V V T V V V V V I I 1 V V V T V I '

6.0 615 710 7.5 8T0 8.5

Figure 14. Calibration curves obtained from four SNARF -1 loaded principal epithelial cells
(640 urn/580 nm emission ratio vs. pH,). Calibrations were performed at pH values of 6.4,
6.9, 7.4, and 7.9 and the fluorescence emission ratio was calculated for the 640 nm and 580
nm emissions following subtraction of background fluorescence intensity.

54
LEICMYH+EI 'HKl"M'ilE 'D

E . l l' l C [L

To examine pH regulatory mechanisms in unstimulated duct cells, the cytoplasm was
acidified with a 2 min NH4C1 pulse and pH recovery was observed over the next 8 min
(Figure 15). pH, in 20 principal cells tested dropped by 0.20 :i: 0.01 units following the
pulse and then recovered by 0.14 :i: 0.01 units, or to 70.7% of the change in pH induced
by the NILCI pulse (ApH,) within 4 min (Table 3). The rate of pH recovery, measured as
the time required for half of the eventual pH recovery to occur (TV), was 0.92 :t 0.14 min.
Superfusion with Na*-free medium following the NH4CI pulse resulted in a more
profound drop in pHi of 0.36 d: 0.06 units. There was no pH recovery until Na” was
again present in the perfusate (Figure 15).

To further evaluate the role of Na*/I-I* exchange in this recovery process, duct
segments were superfused with KRH containing 0.5 mM amiloride following the NH4C1
pulse (Figure 16). Amiloride significantly inhibited pH recovery in control duct
segments. pH,- increased by only 0.03 :i: 0.02 units over 8 min (Table 3), representing a
recovery of just 6.8% of ApH,. When amiloride was removed from the medium in the
presence of Na*, a recovery of 0. 12 d: 0.02 pH units was observed. In the absence of a
prior NH4C1 pulse, superfusion of duct segments with KRH containing amiloride resulted
in a marked cytosolic acidification of 0.32 i 0.06 pH units, although pH, recovered
rapidly by 0.23 i 0.04 pH units following amiloride wash-out (Figure 17). A similar
drop in pH, of 0.36 :L- 0.08 pH units was observed when duct segments were superfused

with Na“-free KRH (Figure 18).

55

8.0-

7.9- We FHFI

Nos-KRH [Nine-KRH INaJ
7.8

 

7.7-

7.6..

7.5-

 

7.4-

 

PHI

7.3-

 

7.2..

7.1-

7.0-

 

6.9-

 

 

6'8 U f f T I 1* T V f l 1 f V V I T F T f l I I I VT I

time (nin)

Figure 15. Sodium dependence of intracellular pH regulation in guinea pig pancreatic duct
principal epithelial cells. After 5 min superfusion in KRH, cells were acidified with a 2 min
pulse of 20 mM NH,C1. FollOwing the pulse, pH, dropped by 0.2 to 0.3 pH units, but 70.4%
of this drop was recovered rapidly. Following a second NH4C1 pulse, Na* was replaced in
the perfusate by NMG“ for 8 min. pH, did not recover until re-introduction of Na*. Results
represent the mean response of nine principal epithelial cells in a single experiment,
representative of a total of four.

56

Table 3. Effects of secretin and carbachol on recovery of intracellular pH (pH,) following
intracellular acidification in the presence or absence of amiloride and bafilomycin A,.

 

 

 

 

 

 

 

 

 

 

 

 

Azania! Inhihimrts) Mm Messier: InJminl 1!
None 0.14 :i: 0.01 70.7 :1: 7.2 0.92 n 0.14 20

None Amiloride 0.03 :i: 0.02 * 6.8 3: 5.1 * ND 47
Bafilomycin 0.14 a: 0.02 66.5 :h 2.9 0.73 s 0.07 20

none 0.14 :t 0.06 70.9 s 8.1 0.70 s 0.06 33

Secretin Amiloride 0.14 a: 0.02 40.6 :1: 6.6 1.70 i 0.27 * 12
(100 nM) Bafilomycin 0.18 :i: 0.04 73.2 a: 4.0 0.60 d: 0.10 17
Bath 0.05 n 0.03 16.1 n 9.8"“ ND 23

none 0.25 i 0.03 69.7 i 6.1 0.80 i 0.14 39

CCh Amiloride 0.15 :1: 0.02 50.4 s 8.8 1.63 :L- 0.18 r 47
(10 PM) Bafilomycin 0.19 a. 0.05 70.0 3. 3.9 1.10 :i: 0.30 28
Both - 0.02 3. 0.04 * -13.0 :i: 167* ND 23

 

 

 

 

 

 

 

 

pHi recovery values are in pH units. Percent recovery values represent the percentage of the
drop in pH, recovered following a 2 min NH,C1 pulse. Concentrations used were: Amiloride,
0.5 mM; bafilomycin A,, 30 nM. Results represent means :t SE of measurements from N
principal epithelial cells. *, significantly different (P < 0.05) from the mean value in the
absence of inhibitors. ND, not determined.

57

8.1-

8.0., INE'I

KRH | KRH + ArrI'Ioride KRH

 

 

7.9-
7.8-

7.7-

:.- 7.64
D.
7.5-W

7.4-

 

 

7.3-

7.2-

 

 

7.1 Y Y T j T F Y j I I I I I U fir V T l

0 ' 5 10 15 20

time (min)

Figure 16. Effect of amiloride on pHi recovery in pancreatic duct cells following
intracellular acidification. pH recovery following a 2 min NH4C1 pulse was reduced to 0.03
i 0.02 pH units (6.8% of ApH,) in the presence of 0.5 mM amiloride. Results represent the
mean response of five principal cells in a single experiment, representative of a total of Six.

58

 

7.6 KRH l Amiloride KRH J
7.5-
7.4-
7.3-
7.2-

7.1-

 

7.0 . .

time (win)

Figure 17. Effect of amiloride addition on pH, of pancreatic duct epithelial cells. pH,
dropped by 0.32 d: 0.06 pH units following superfusion in the presence of 0.5 mM amiloride,
but recovered almost completely on its removal from the superfusate. Results represent the

mean response of seven principal epithelial cells in a single experiment, representative of a
total of three.

59

 

 

 

 

 

 

 

 

 

 

 

 

0.1 -
I *
-0.0L
.1
-0.1-
I'- -l
a.
.E
8, -0.2.
C .
«I
.r: I
o .
-0.3-
04;
0 5 Amiloride - Sodium-free

 

Figure 18. Effect of amiloride addition or sodium substitution on pHi of pancreatic duct
epithelial cells in the absence or presence of extracellular HCO3'. In the absence of
extracellular HCO,‘ (KRH), addition of amiloride (0.5 mM) or Na+ substitution with NMG"
resulted in marked cytosolic acidification, while in its presence (KRB), only Na“ substitution
produced a cytosolic acidification. In the presence of HCO3', duct cells exposed to amiloride
effectively maintained pH,. *, significantly different (P < 0.05) than change in HCO3'-free
buffer. Results represent the mean :1: SE of measurements from 15-25 principal epithelial
cells in 3-7 separate experiments.

6O
2.: 'l . [3' l HR I'

In the presence of 25 mM extracellular HCO,’ (KRB) an additional Na"-dependent pH
regulatory mechanism was observed. While Na‘ replacement in the superfusate induced a
fall in pH, of 0.29 :t: 0.02 pH units, as seen with Nafifree KRH, superfusion with KRB
containing 0.5 mM amiloride did not result in a reduction of pH, (Figure 18). The
response to amiloride in KRB was therefore significantly difierent (P < 0.05) than the
response in KRH, suggesting the presence of an amiloride-insensitive, HCO3'-dependent
pH regulatory mechanism.

Recovery of intracellular pH from the cytosolic acidification produced by superfusion
in Na*-frec KRH was 96.5 :1: 8.9%, once Na+ was again present in the perfusate. This
recovery was blocked by 0.5 mM amiloride in the absence of extracellular HCO,‘ (Figure
19). In the presence of extracellular HCO3', however, pH,- recovered by 0.27 :t 0.07 pH
units, or 83.8%, even though amiloride was also present.

pHi recovery from cytosolic acidification following a 2 min NI-LCl pulse also was
reduced significantly (P < 0.05) by amiloride in HCO3‘-free KRH to 0.03 i 0.02 pH units,
or 6.8% of ApHi (Figure 20). In HCO3’-containing KRB, pH, recovered by 0.21 :1: 0.03

pH units, or 139.4% of ApHi. pH recovery was reduced significantly by omission of Na‘

from the superfusate in both the absence and presence of HCO3'.

 

When switching from HCO3'-free KRH to KRB, pHi of pancreatic duct epithelial

cells alkalinized slightly by 0.07 i 0.03 pH units. Subsequent superfusion of duct

61

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00- _*_ -100
1
. .
0.75.. -75
>. 0.50- L50
'5 . . .\°
> an
O u
u * o
o o
m . . o
z 0.25- _25 5
a. . ~ 5':
0.00- o
-o.25.. ' .—25
pH Recovery % Recovery

Figure 19. Recovery of pHi in pancreatic duct segments in the presence of amiloride and
Na“ following cytosolic acidification induced by superfusion in Na*-free buffer. In the
absence of extracellular HCO,‘ (KRH), pH, did not recover, while in its presence (KRB), pH,
recovered almost completely (83.8 i 14.4%). *, significantly different (P < 0.05) from pH
recovery in HCO3'-free buffer. Results represent the mean i: SE of measurements from 16-
25 principal epithelial cells in 3-5 separate experiments.

62

175-

 

 

150i

 

 

 

 

 

125$

100$

 

 

75L

 

Virflilrxxnnayr
I
l

 

 

 

 

 

 

 

 
 

 

 

 

   

' *
25.
. "T"
. *
oi . . .
Control Amllonde odlum-free

Figure 20. pH recovery in the presence of amiloride or in the absence of extracellular Na“
following a 2 min NI-LCl pulse, both in the absence (KRH) and presence (KRB) of
extracellular HCO3°. In the presence of Na‘ and absence of amiloride, pH, recovered by 70-
7 5% in both, and omission of extracellular Na+ during the recovery effectively blocked pH
recovery. In the absence of HCO3', addition of 0.5 mM amiloride was also sufficient to
block pH recovery, while in its presence, pH, recovered by 139.4 :I: 27.2%. *, significantly
difi'erent (P < 0.05) from controls. Results represent the mean t SE of measurements from
20-57 principal epithelial cells in 4-7 separate experiments.

63
segments with Cl'-free KRB resulted in a further alkalinization of 0.22 i 0.06 pH units

(Figure 21). This alkalinization was reversible; pH, recovered by 96.7% upon readdition
of C1' to the superfusate. To test whether this alkalinization resulted from inhibition of
Cl’lHCO,’ exchange, duct segments were superfused in KRB containing 0.1 mM DIDS
(Figure 22). Cytosolic alkalinization averaging 0.13 :l: 0.04 pH units was observed in the
presence of the inhibitor, although pH, rapidly returned to its previous level on removal of

DIDS from the superfusate.

4. (mammal gm Ramadan: (1+ Remnant in 13mm 53mm 22m 12m
E 'E l' l 2 It

a. pH Regulation in the Presence of Agonists

Further experiments were carried out to determine whether agonists activate
amiloride-insensitive H+ extrusion mechanisms in the guinea pig pancreatic duct, as has
been reported for rat and pig pancreatic ducts (V illanger et. al., 1995; Zhao et. al., 1994).
These experiments were carried out in a HCO3'-free buffer to eliminate contributions of
HCO3' to pH regulation. The initial pH, of duct principal cells superfused with KRH
containing 100 nM secretin, the primary hormonal regulator of pancreatic HCO,‘
secretion, was 7.31 i 0.04 (n = 33). Superfusion with secretin in the absence of inhibitors
did not significantly alter the extent or rate of pH recovery following the NI-LCl pulse
(Table 3). In the presence of secretin, pH recovery was 0.14 i 0.06 units (70.9% of

ApHi). The rate of pH recovery was not significantly altered in the presence of secretin

64

7.9

 

KRH] KRB | Cf-freeKRB lKRB]

 

7.8..

7.7-

7.6-

7.5..

P";

7.4-

7.3-

7.2-

7.14

 

7.0...-, W
o 5 1o 15 20

 

time (min)

Figure 21. Effect of removal of extracellular Cl’ on pHi of pancreatic duct epithelial cells
perfused with a HCO3'-containing buffer (KRB). Transition to KRB was accompanied by
a cytoplasmic alkalinization of 0.05 to 0.20 pH units. A further alkalinization of 0.22 i 0.06
pH units was observed when extracellular Cl' was replaced with gluconate. pHi returned to
its previous baseline following readdition of Cl'. Results represent the mean response of
seven principal epithelial cells in a single experiment, representative of a total of four.

65

8.01

 

7.9. KRH | KRB | KRB+DIDS | KRB
7.85
7.7a

.16_ L“At/MAJ

7.5-

 

Pfh

7.4-

7.3-

 

7.2

 

I V I Y I I V Y j’ V j

o ' ' ' ' 5 ' ' ' ' 1o 15 20
tune(nin)

Figure 22. Effect of exposure to DIDS on the pHi of pancreatic duct epithelial cells.
Transition to 3 HCO,’ buffer (KRB) was accompanied by cytoplasmic alkalinization of 0.05
to 0.20 pH units. A further alkalinization of 0.13 :t 0.04 pH units was observed when 0.1
mM DIDS was added to the superfusing buffer. pH returned to its previous baseline
following omission of DIDS. Results represent the mean response of seven principal
epithelial cells in a single experiment, representative of a total of three.

66

(TV, = 0.70 i 0.06 min). pHi in duct segments superfused with secretin in the presence of
0.5 mM amiloride recovered by 0.14 i 0.02 units (40.6% of ApHi) after an NH4C1 pulse,
a recovery significantly greater than that in the absence of the hormone (Figure 23 and
Table 3). The rate of pH recovery (TV, = 1.70 i 0.27 min) was significantly slower than
in duct segments exposed to secretin in the absence of amiloride (TV, = 0.70 :t 0.06 min).

To determine whether the observed amiloride-insensitive pH recovery in the presence
of secretin involves operation of an ATP-dependent H“ pump, 30 nM bafilomycin AI was
included in the superfusate together with amiloride during the recovery period (Figure
24). In the presence of secretin and both inhibitors, pH recovery was only 0.05 :1: 0.03
units (16.1% of ApHi) (Table 3). To assure that bafilomycin AI did not interfere with
general pH homeostasis, pH recovery following an NH4C1 pulse also was determined in
the presence of 30 nM bafilomycin AI and secretin, but in the absence of amiloride
(Figure 25). pH recovery under these conditions was 0.18 i 0.04 units (73.2% of ApHi).
The rate of pH recovery (116, = 0.60 i 0.10 min) also was unaffected (Table 3). Addition
of bafilomycin AI alone to the superfusate also did not significantly affect pH recovery,
which in its presence was 0.14 :t 0.02 units, or 66.5% of ApHi (Table 3). The rate of pH,
recovery (Ty, = 0.73 :l: 0.07 min) also was not significantly different than that in control
cells.

Recent studies have shown that acetylcholine and its analogues as well as secretin can
evoke fluid secretion fi'om the pancreatic duct system in some species (Ashton et. al.,
1993). To assess the effects of cholinergic agonists on pH regulation in guinea pig

pancreatic ducts, isolated interlobular duct segments were superfused with 10 pM

67

 

 

7.9-
7.8- W

KRH | KRH+Amiloride | KRH |
7.7- f
7.6-

 

 

 

 

 

"; WWW
7.4-
7.3-
7.2-
Secretin
7'1 ' ' T I 1 I F T u f I
0 5 10 15 20 25

time (min)

Figure 23. Efi‘ect of secretin on pH, recovery from a 2 min NH4Cl pulse in pancreatic duct
cells in the presence of 0.5 mM arrriloride. pH recovery following a 2 min NH4Cl pulse was
0.14 d: 0.03 pH units (40.6% of ApHi) in the presence of 100 nM secretin. Results represent
the mean response of eight principal epithelial cells in a single experiment, representative of
a total of three.

68

8.0 -
IN ”4‘1
7-9~ KRH IAmllorideIBafllomycin KRH

 

 

7.8-
7.7-
7.6-

if 7.5-WV

7.4-

 

7.3-

 

7.2-
Secretin

7'1 I I r I l I I I fifi F I I I [ I f

o §""1o 15 20

 

 

tim (m'n)

Figure 24. Effect of bafilomycin Al and amiloride on secretin-stimulated pH recovery in
pancreatic duct epithelial cells. pH, did not recover significantly following a 2 min NH4C1
pulse in the presence of 0.5 mM amiloride, 30 nM bafilomycin A,, and 100 nM secretin.
Results represent the mean response of eight principal epithelial cells in a single experiment,
representative of a total of three.

69

8.0-

7.9- N H ‘4.
KRH [ KRH + Bafilomycin IKRH ]

 

 

7.8-

7.7-1

pH.

7.6-

7.5-

7.4-
Secretin

 

 

time (min)

Figure 25. Effect of bafilomycin Al alone on pH recovery fi'om an acid load imposed by
superfusion with 20 mM NILCI. pH, recovery following a 2 min NH4C1 pulse was 66.5%
in the presence of bafilomycin Al alone. Results represent the mean response of nine
principal epithelial cells in a single experiment, representative of a total of three.

70

carbachol. pH, in the principal epithelial cells in the presence of the agonist was 7.28 :t
0.03 (n = 39). Following an NI-LCl pulse in the absence of bafilomycin A, and amiloride,
pHi recovered by 0.25 :1: 0.03 units, or 69.7% of ApH, (Table 3). T.,z for this recovery was
0.80 i 0.14 min. In the presence of amiloride, pHi recovered by 0.15 i 0.02 units
(50.4% of ApH,), and in the presence of bafilomycin by 0.19 :t 0.05 units (70.0% of
ApH,). The time course of recovery in the presence of bafilomycin was not different from
that in ducts superfused with carbachol alone (Ty, = 0.80 d: 0.14 vs. 1.10 i 0.30 min), but
was significantly longer in the presence of amiloride (T.,’ = 1.63 :1: 0.18 min). When both
inhibitors were present in the superfusate, pH, recovery after a NH4Cl pulse was 7

completely suppressed (Table 3).

b. Role of the Cytoskeleton in Agonist—mediated pH Recovery

Previous studies on the pig pancreatic duct system have suggested that secretin-
evoked hydrogen pumping is activated by fusion of cytoplasmic vesicles containing the
bafilomycin A,-sensitive H*-ATPase with the basolateral membrane (Buanes et. al., 1987;
Veel et. al., 1990). Vesicular trafficking between the cytoplasm and the plasma
membrane in many cell types, including transporting epithelial cells, is dependent on the
integrity of cytoskeletal elements (Arruda et. al., 1988; Brown & Sabolic, 1993; Dumont
et. al., 1994). To assess the role of the cytoskeleton in the agonist-evoked expression of
bafilomycin A,-sensitive H+ extrusion, we therefore incubated isolated duct segments
with 10 ug/ml of cytochalasin D and nocodazole prior to agonist exposure. Duct

segments also were fixed for localization of microfilaments and microtubules after

71

exposure to the two inhibitors to assess the effectiveness of cytoskeletal disruption
(Figure 26).

As illustrated in Figure 26A, microfilaments in control duct segments were localized
primarily to narrow bands underlying the luminal plasma membranes and the basal poles
of principal epithelial cells. By contrast, microtubules were present throughout the
cytoplasm of principal cells, and were oriented primarily along the longitudinal cell axis
(Figure 26B). Preincubation of duct segments with 10 ug/ml of cytochalasin D
completely disrupted the bands of actin filaments near the apical and basal membranes,
resulting in a diffuse staining of the cytoplasm (Figure 26C). Pretreatment of duct
segments with 10 rig/ml of nocodazole resulted in the disruption of microtubules, as
evidenced by reduced, punctate staining of the epithelial cell cytoplasm, as well as
noticeable swelling of the epithelial cells (Figure 26D).

pHi recovery in segments of guinea pig pancreatic duct that were pre-treated with
cytochalasin D was completely abolished (-0.01 i 0.01 pH units) in the presence of 100
nM secretin and amiloride (Figure 27), a significant change from the pH recovery
. observed previously in the absence of any pretreatment. Cytochalasin D pre-treatrnent
significantly reduced carbachol-evoked pH recovery as well, which was 0.03 :t 0.03 pH
units. Pre-treatment of duct segments with nocodazole similarly attenuated pH recovery
stimulated by secretin to 0.05 i 0.03 pH units, and that stimulated by carbachol to 0.01 :t

0.02 pH units (Figure 28).

C

 

Figure 26. Effects ofcytoskeletal-disrupting agents on distribution of microfilaments and
microtubules in pancreatic duct principal epithelial cells. A) In this cross-sectional view of
a control interlobular duet, microfilaments are stained with rhodamine-conjugated phalloidin
and principally underlie the apical and basolateral plasma membranes. B) Mierotubule
localization in a control interlobular duct incubated with a primary tutti-tubulin antibody and
an FITC-conjugated secondary antibody. Microtubules are abundant and are primarily
oriented parallel to the longitudinal axis ofthe epithelial cells. C) liffect oi‘preincubation
with 10 ug/ml ofcytochalasin-D on the microfilament network The dense staining along
the membrane is replaced by a diffuse cytoplasmic staining. D) Liffect of preincubation with
10 rig/ml of nocodazole on microtubules. Disruption of the microtubule network is ex'idcnt.
as is swelling ofthe epithelial cells. Scale bars 7 10 um.

73

7.9-

 

W}

7.3, KRH KRH + Am'loride |

 

7.7-
7.6-

7.5-

7.4-’\.\-’\’~V‘—l"\4l
7"“ W

7.24

P”:

 

 

Secretin

 

 

 

time (min)

Figure 27. Effect of preincubation with 10 ug/ml of cytochalasin D on pH recovery in
pancreatic duct segments following a 2 min NH4C1 pulse. The microfilament-disrupting
agent completely blocked secretin-stimulated pH recovery in KRH in the presence of
amiloride (0.5 mM). Results represent the mean response of six principal epithelial cells in
a single experiment, representative of a total of five.

74

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

0.20 -
I Secretin
. Carbachol
0.1 5 4
0.1 0 -
i . _*_
0.05 -
*
0.00 .-
*
Control éytochalasin Nocodazole

Figure 28. Recovery of pHi in the presence of amiloride and agonists following a 2 min
NI-LCI pulse afier preincubation with cytochalasin D or nocodazole. Preincubation with 10
ug/ml of cytochalasin D significantly reduced the pH recovery observed in the presence of
either carbachol (10 nM) or secretin (100 nM), as did preincubation with 10 ug/ml of
nocodazole. *, significantly different (P < 0.05) from controls. Results represent the mean
i SE of measurements from 29-40 principal epithelial cells in 4-7 separate experiments.

IV. DISCUSSION

Postprandial stimulation of pancreatic secretion traditionally has been thought of as
being under control of the gut hormones CCK and secretin, with acinar cell digestive
enzyme release regulated primarily by the former and duct cell fluid and bicarbonate
secretion regulated by the latter. However, recent studies have suggested that other
neural and endocrine factors may modulate these secretory responses. Because many of
these potential pancreatic secretagogues or inhibitors may have systemic effects or
actions on acinar cells or the pancreatic vasculature which could influence pancreatic
fluid secretion in viva, preparations of isolated pancreatic ducts have been developed to
identify direct regulatory influences on duct epithelial cells. The current studies have
taken advantage of a method recently devised in our laboratory for isolating main and
interlobular duct segments from the guinea pig pancreas (Hootman & Logsdon, 1988) to
investigate the cellular mechanisms by which bicarbonate secretion is regulated. In this
in vitro system, it is possible to determine the effects of a number of potential regulatory
molecules, including secretin, on cellular cyclic AMP levels of the pancreatic duct, and to
detect the involvement of specific ion transport elements in principal cell bicarbonate
secretion.

Current models of bicarbonate secretion from the pancreatic duct principal epithelial

cell suggest several potential sites in the cellular signalling pathway where extrinsic

75

76

agents might regulate the secretory process, including generation of intracellular
messengers. The primary step initiating pancreatic duct HCO,‘ secretion is thought to
involve a secretin-evoked rise in cellular cyclic AMP levels. We therefore hypothesized
that neurohumoral modulators of bicarbonate secretion might act at this initial step of the
secretin-triggered cascade, the activation of adenylyl cyclase.

The 6 to 8-fold increases in cyclic AMP levels observed when mixed preparations of
isolated main and interlobular pancreatic ducts were stimulated for 30 min with 0.1 uM
secretin (Figures 9-11) compare favorably with previously reported responses in isolated
rat interlobular pancreatic ducts (Argent et. al., 1986; Arkle et. al., 1986; Ashton et. al.,
1993). The secretin-evoked rise in cyclic AMP content in isolated guinea pig pancreatic
ducts was dose-dependent over a concentration range of 10 pM to 10 nM, maximal at
concentrations of 10 nM and above, and had a calculated half-maximal response at a
secretin concentration of 150 pM (Figure 11). These values are lower by an order of
magnitude than those reported previously for rat pancreatic ducts (Arkle et. al., 1986), but
are to the right of the dose-response curve for secretin-evoked fluid secretion (EC50 - 20
pM) (Argent et. al., 1986). These dose-response relationships indicate that even small
changes in intracellular cyclic AMP content may be sufficient to maximally activate
ductal fluid secretion. Correspondingly, at postprandial blood secretin concentrations in
the low picomolar range, ductal cyclic AMP levels would be well below 50% of the
maximal value (Figure l 1).

Secretin significantly increased cyclic AMP content in both main and interlobular

duct segments, although the main duct response was 50% less than that of the interlobular

77

ducts (Figure 11). This difference could reflect larger numbers of secretin receptors in
the interlobular duct cells or differences in coupling stoichiometry between these
receptors, GTP-binding proteins, and adenylyl cyclase in the two regions of the duct
system. Altemately, the difference could reflect the relatively larger percentage of goblet
cells in the main duct, which are unresponsive to secretin (Hootman & de Ondarza,

1995), but which contribute significantly to duct protein content. This second possibility
seems unlikely, however, since the observed difference in responsiveness is maintained in
cultured duct epithelial monolayers, which contain few, if any, goblet cells.

Although evidence from earlier micropuncture studies in cats and rats pointed to the
intralobular and interlobular ducts as the principal sources of agonist-evoked fluid and
bicarbonate secretion in the pancreatic duct system (Rader, 1992), results from the
current studies suggest that the epithelium of the main duct also may play an important
secretory role. Differences in the structure of the pancreatic duct systems - the guinea
pig main duct traverses the distance of the pancreas and measures 60 mm in length,
whereas the rat main duct is quite short - may therefore reflect species-specific functional
differences as well. The presence of all major electrolyte transport elements in principal
epithelial cells of the main and common ducts of the rat pancreas (Zhao et. al., 1994) also
supports the assertion that the larger ducts participate in elaboration of bicarbonate.

The results illustrated in Figures 12 and 13 demonstrate that guinea pig pancreatic
duct epithelial cells grown in primary culture continue to be responsive to secretin and
VIP, but to a reduced degree relative to acutely isolated segments of main and

interlobular ducts. Receptors for these two peptides presumably are located on the

78

basolateral surfaces of principal epithelial cells in intact ducts. It is not clear how VIP
and secretin access their respective receptors on principal epithelial cells grown in
monolayers on an impermeable plastic surface. Conceivably, the receptors may have
translocated to the apical plasma membrane in the cultured duct cells, or junctional
contacts between neighboring cells in the monolayers may be permeable to the two
peptides. While fluid or bicarbonate secretion were not measured in this model system,
an earlier study in our laboratory showed that secretin significantly increases 3‘SCl' efflux
from such duct epithelial monolayers (unpublished observation). Taken together with
previous studies on human duct cell cultures (Gray et. al., 1989), our findings indicate
that the principal epithelial cell retains many of its functional characteristics in culture.
This particular model system therefore appears to be useful in investigating physiological
functions of pancreatic duct principal cells, and has indeed been used to demonstrate
growth regulatory effects of insulin and transforming growth factors on this cell type
(Bhattacharyya et. al., 1995).

In viva control of pancreatic exocrine secretion is complex and involves a number of
neural and endocrine regulators (Adler & Beglinger, 1990; Singer, 1993). Potentiating or
stimulatory effects on in viva pancreatic fluid and bicarbonate secretion have been
reported for acetylcholine, bombesin, VIP, CCK, neurotensin, insulin, substance P, PGEZ,
catecholamines, and CGRP (Case & Argent, 1993). Other reports have indicated
inhibitory influences of glucagon, somatostatin, and substance P. In studies on the
isolated perfused rat pancreas, glucagon and somatostatin inhibited and insulin

potentiated secretin-induced fluid secretion (Hasegawa et. al., 1993). PGE, also has been

79

shown to stimulate pancreatic fluid secretion in vitra, possibly through an effect on the
duct system (Marshall et. al., 1982). Fewer studies have been carried out on isolated
ducts free of the influences of acinar cell secretion and systemic neural, endocrine, and
vascular effects. Acetylcholine (Ashton et. al., 1993; Evans et. al., 1992) and bombesin
(Ashton et. al., 1990) have been shown to stimulate fluid and bicarbonate secretion in
isolated rat pancreatic ducts, although VIP and the CCK analogue caerulein had no effect.
Substance P inhibited fluid secretion stimulated by secretin, forskolin, or dibutyryl cyclic
AMP (Ashton et. al., 1990). These studies imply that duct epithelial cells express
receptors for a number of these physiological regulators and that adenylyl cyclase and
cyclic AMP, through which the effects of secretin are mediated, are a potential site of
modulation by other signalling molecules and intracellular messengers. Our results
indicate that at least two regulators besides secretin, VIP and somatostatin, modulate the
adenylyl cyclase system in the guinea pig pancreatic duct cell. We have observed cyclic
AMP elevations of up to lO-fold in isolated interlobular pancreatic ducts in response to
VIP, levels that are comparable to those achieved with a maximal dose of secretin. VIP
also has been reported to bring about depolarization of the basolateral plasma membrane
in isolated rat pancreatic ducts of a magnitude similar to that evoked by secretin (N ovak
& Pahl, 1993; Pahl & Novak, 1993). These findings are in contrast with the observation
that VIP failed to stimulate fluid secretion in isolated rat pancreatic ducts (Ashton et. al.,
1990). VIP does, however, elevate cyclic AMP levels in rat ducts as well (F olsch et. al.,
1980). It therefore would appear that its mechanisms of action are in most respects

identical to those of secretin, and it would be surprising if VIP turned out not to be a

80

direct agonist of duct fluid secretion.

The islet hormone somatostatin has been shown to inhibit pancreatic exocrine
secretion in viva (Gullo et. al., 1987; Kuvshinoff et. al., 1993b) and in the isolated
pancreas (Hasegawa et. al., 1993; Matsushita et. al., 1993). Results of the present study
suggest that one mechanism through which these effects are mediated is attenuation of
the duct cyclic AMP response to physiological doses of secretin. The 50% inhibition of
secretin-stimulated cyclic AMP elevation noted (Table 2) could be expected to impact
bicarbonate and fluid secretion significantly, given the low blood levels of secretin in
viva, and suggests that the early steps in Signal transduction represent significant control
points for ductal secretory responses. The islets of Langerhans are in close proximity to
pancreatic ducts (Kodama, 1983), and the presence of somatostatin-positive endocrine
cells within the duct epithelium has been reported in rats (Park & Bendayan, 1992) and
guinea pigs (Hootman et. al., 1996). Somatostatin also has been shown to inhibit the
potentiating effects of secretin and VIP on acinar cell amylase secretion stimulated by
CCK, bombesin, and carbachol (Matsushita et. al., 1993). This inhibition is paralleled by
a reduced cyclic AMP response to secretin and VIP, indicating that somatostatin acts to
inhibit the adenylyl cyclase in these cells as well.

Other potential modulators tested did not affect cyclic AMP content in isolated
pancreatic ducts (Table 2), in agreement with previous findings (Ashton et. al., 1993;
Folsch et. al., 1980). The lack of effects may be due either to the absence of functional
receptors for these agents on the pancreatic duct principal cell, or to the involvement of

intracellular signalling pathways which act separately or downstream from cyclic AMP

81

generation. This appears to be the case at least for cholinergic signalling. Acetylcholine
elevates intracellular free calcium in pancreatic duct cells (Stuenkel & Hootman, 1990;
Hug et. al., 1994) and simultaneously stimulates fluid secretion (Ashton et. al., 1993;
Evans et. al., 1992) ), presumably through muscarinic receptor activation (Hootman et.
al., 1993).

The secretin-evoked elevation of cyclic AMP in pancreatic duct principal epithelial
cells is thought to activate an apical membrane chloride channel, known as the CF TR
because of its central involvement in the pathology of cystic fibrosis. CF TR has been
localized immunocytochemically in our laboratory to the apical membrane of main and
interlobular pancreatic duct epithelial cells (unpublished observations). The channel is
thought to contain two nucleotide binding domains and one regulatory domain containing
numerous potential phosphorylation sites for PKA and PKC (Riordan et. al., 1989;
Denning et. al., 1992), including critical sites for cyclic AMP-mediated channel opening.
In addition to the CFTR channel, extrusion of bicarbonate ions from the principal
epithelial cell requires the presence of several other ion transporters. Current models for
the cellular mechanisms responsible for duct HCO,‘ secretion are based largely on studies
on pancreatic ducts fi'om two species, the pig and rat. In both species, evidence has been
obtained for the functional involvement of Na*/H+ and Cl'/HCO3' exchangers, a
NaVI-ICO,‘ cotransporter, and ATP-driven H“ pumping in regulated secretion of
bicarbonate (Stuenkel et. al., 1988; Veel et. al., 1992; Villanger et. al., 1995; Zhao et. al.,
1994)

The present studies were carried out to determine whether these processes also

82

contribute to bicarbonate secretion in pancreatic ducts of the guinea pig, a species that can
secrete pancreatic juice with a HCO,‘ concentration of up to 150 mEq/l (Case & Argent,
1993). Since maximum concentrations of HCO,‘ of only 70-80 mEq/l have been reported
in rat pancreatic juice, such a comparative study is warranted. To assess the possible
presence of the various transport mechanisms of interest, pH, of isolated guinea pig .
interlobular pancreatic duct epithelial cells was determined by laser scanning confocal
microscopy after loading with the pH-sensitive dye, SNARF -1. Analysis of intracellular
pH by confocal microscopy has the advantage over conventional fluorometric techniques
that the responses of individual duct epithelial cells within their native environment can
be examined and that the image analysis software of the confocal microscope allows
ratiometric pH, determination.

Na“/H+ exchange activity was detected through modification of superfusing solutions
by addition of amiloride and by ion substitution. Superfusion of SNARF -I loaded duct
segments with a HCO3'-free HEPES-buffered Ringers solution containing 20 mM NH4C1
resulted in a transient cytoplasmic alkalinization, followed by acidification of the cytosol
upon withdrawal of NH4C1 from the perfusate (Figure 15). The intracellular pH then
recovered rapidly, returning to 70.4% of its baseline value within 2 min. The
mechanisms which could account for pH recovery from such an acid load include Na*/H+
exchange, H+ pumping, and HCO,‘ import, with the latter mechanism discounted when
using HCO3‘-free solutions. pH recovery was blocked by amiloride or by replacement of
Na“ with NMG*, regardless of the cause of cytosolic acidification (Figures 19 and 20).

Moreover, pHi of duct cells acidified without a prior NH4C1 pulse when superfused with

83

amiloride or a Na”-free buffer (Figure 18). Together, these observations indicate that
intracellular pH regulation in unstimulated guinea pig interlobular pancreatic duct
epithelial cells, in the absence of extracellular HCO3‘, is accomplished primarily through
Na‘lI-I“ exchange, as was previously shown for rat pancreatic ducts (Stuenkel et. al.,
1988). Evidence for basolateral Na*/H* exchangers also has been found in the rabbit
pancreas (Swanson & Solomon, 1975) and the pig pancreatic duct (V eel et. al., 1992).
Despite its apparent housekeeping role, several studies have indicated that Na*/H*
exchange is not the sole mechanism responsible for proton extrusion during stimulated
duct HCO,’ secretion. Amiloride did not inhibit secretin-stimulated bicarbonate secretion
by the pig pancreas (Grotmol et. al., l986a; Veel et. al., 1992) or from the isolated,
perfused rabbit pancreas (Kuijpers et. al., 1984). One candidate mechanism that was
proposed by Reder and coworkers to account for amiloride-insensitive HCO,‘ secretion
was the insertion of ATP-dependent proton pumps into the basal cell surface (Rzeder,
1992). Morphological observations indicated that vesicles present in the cytoplasm of
unstimulated pig pancreatic duct cells fused with the basolateral membrane following
secretin exposure (Buanes et. al., 1987). Further in viva studies from this laboratory
(Grotmol et. al., l986b; Veel et. al., 1992) have provided additional support for this
model of stimulated pancreatic bicarbonate secretion in the pig. More recent studies by
Zhao and coworkers (Zhao et. al., 1994) have confirmed the involvement of H*-ATPases
in stimulated bicarbonate secretion in rat pancreatic ducts. This ATPase is sensitive to
the specific inhibitor, bafilomycin A, (V illanger et. al., 1995; Zhao et. al., 1994), and is

also present in other HCO3‘-secreting epithelia (V illanger et. al., 1993).

84

Results of the present studies on guinea pig pancreatic ducts support in most
particulars the model proposed by Rzeder (Rmder et. al., 1992). Unlike the situation in
unstimulated guinea pig pancreatic duct cells, amiloride did not prevent recovery of pH,
after an NILCl pulse when secretin or carbachol were present in the superfusing medium
(Table 3). Inclusion of bafilomycin A, in the medium in addition to amiloride, however,
prevented pH recovery in the presence of either agonist. These results clearly indicate
that principal epithelial cells of the guinea pig pancreatic duct system express proton
pumps that are activated by agonist exposure, as is the case in the rat and the pig.

Veel and colleagues (V eel et. al., 1990) previously demonstrated that the secretin-
stimulated clearance of a prominent vesicular component from the cytoplasm of pig
pancreatic ductules was inhibited by infusion of colchicine, a microtubule-
depolyrnerizing agent, into anesthetized animals. This and similar observations on other
species (Arruda et. al., 1988; Brown & Sabolic, 1993; Dumont et. al., 1994; Schwartz &
Al-Awqati, 1986) led to the proposal that the bafilomycin-sensitive proton pump was
present in the membranes of these vesicles and that microtubules were involved in the
fusion of these vesicles with the basolateral plasma membrane. In guinea pig pancreatic
duct epithelial cells, small electron-lucent vesicles are present in the cytoplasm (Hootman
& Logsdon, 1988), although they are not as numerous or as large as those in pig
pancreatic ductules. It is therefore unclear whether the bafilomycin-sensitive pH
recovery mechanism observed in stimulated guinea pig pancreatic duct cells results from
insertion of vesicular H*-ATPases into the plasma membrane or the activation of pumps

pre-existing in the membrane. To address this question, isolated interlobular pancreatic

85

duct segments were preincubated with cytochalasin D, a microfilament-depolymerizing
agent, and with nocodazole, a microtubule poison, and their effects on the agonist-evoked
appearance of bafilomycin A,-sensitive pH, recovery were determined. The two agents
exerted the expected effects on the cytoarchitecture of pancreatic duct epithelial cells, in
each case depolymerizing the targeted cytoskeletal element (Figure 26). Both
cytochalasin D and nocodazole also significantly inhibited pH, recovery following an acid
pulse in the presence of either secretin or carbachol and amiloride (Figure 28). These
results suggest that the cytoskeleton plays an important role in the appearance of H*-
ATPase activity in the guinea pig pancreatic duct epithelial cell and imply that a vesicular
fiision process is indeed involved in the guinea pig, as it appears to be in the pig.
Although both secretin and carbachol cause appearance of bafilomycin A,-sensitive
proton pumping, this H*-ATPase may be of limited physiological significance in the
guinea pig. After cytoplasmic acidification induced by an NH4Cl pulse, pH, recovered to
approximately 70% of its pre-pulse value in both unstimulated and stimulated duct cells
when neither transport inhibitor was present. These observations indicate that the
capacity of the NaVH“ exchange mechanism is not exceeded in the presence of secretin or
carbachol. Furthermore, inclusion of bafilomycin A, alone in the superfusate did not
alter pH recovery (Table 3). By contrast, addition of amiloride to the superfusing
medium in the presence of either agonist resulted in a slow and significantly attenuated
recovery of pH,. These findings suggest that the amiloride-insensitive mechanism
activated following agonist exposure has a more limited capacity for H+ extrusion than

does the mechanism responsible for Na*/I-I+ exchange. As suggested by Zhao and

86

coworkers (Zhao et. al., 1994), a physiological role for it may exist only to augment
Na“/H+ exchange in the face of a large H+ gradient. The lack of significant morphological
evidence for this mechanism in duct principal cells of rats and guinea pigs serves as an
additional indication of the relatively smaller role of vesicular H” ATPases in these
species, whereas Raeder and colleagues report large numbers of cytoplasmic vesicles in
unstimulated pig pancreatic duct cells. Whether the presence of "smooth vesicles" in the
cytoplasm of human pancreatic duct cells (Kern, 1993) is indicative of a prominent role
for active Hf transport in humans remains to be determined.

A third mechanism involved in pH regulation of pancreatic duct epithelial cells is
Na*-dependent HCO,’ transport, localized to the basolateral membrane of rat pancreatic
duct cells (Zhao et. al., 1994). This mechanism has been considered previously (Case &
Argent, 1993) based on its reported role in electrolyte transport of rat hepatocytes (Fitz et.
al., 1989) and other tissues, and a similar Na*-dependent , disulfonic stilbene-sensitive
Hco,- transport had been proposed for the rabbit pancreas (Kuijpers et. al., 1984).
However, these recent studies in pigs and rats were the first to provide physiological
evidence for Na*/(HCO3'),, cotransport, with the stoichiometry of Na*:HCO3’ not yet
determined (V illanger et. al., 1995; Zhao et. al., 1994). Results of the present studies also
demonstrate a Na” and HCO3'-dependent pH regulatory mechanism which is insensitive
to amiloride (Figures 18-20). Since this mechanism was blocked in the absence of
extracellular Na*, it is unlikely that its activity can be attributed to activation of H+
pumping by the presence of HCO3'. A study by Ishiguro, et al. (1995), recently appearing

in abstract, also has confirmed the presence of Na*-dependent HCO,‘ transport in guinea

87

pig pancreatic duct epithelial cells. Basolateral NaVHCO,‘ cotransport could account for
carbonic anhydrase (CA)-independent pancreatic duct HCO,’ secretion. Several studies
have argued against a significant CA-independent pathway for pancreatic HCO,‘
secretion. Bicarbonate secretion in the pig pancreas is dependent on blood pH and PM),
(Buanes et. al., 1988), and the CA inhibitor acetazolamide inhibited HCO,‘ secretion-by
70% from the pig (Raeder & Mathisen, 1982) and cat pancreas (Case et. al., 1970).
Nonetheless, the observation that acetazolamide only slightly inhibited fluid secretion in
the rabbit (Kuijpers et. al., 1984) indicates that CA-independent pathways may contribute
significantly to ductal HCO,‘ secretion in some species.

Cl'lHCO,‘ exchange also was demonstrated in guinea pig pancreatic ducts through
superfusion with buffers lacking extracellular Cl' or containing DIDS (Figures 21 and
22). In both cases, a significant and reversible cytosolic alkalinization was observed,
presumably due to retention of intracellular HCO3'. These observations are consistent
with previous models wherein Cl'lHCO,‘ exchange provides the major efflux pathway
into the duct lumen for HCO3‘. The discovery of a basolateral Cl’/HCO3' exchanger in rat
ducts (Zhao et. al., 1994) is surprising, yet the present studies do not rule out its presence,
since sidedness of transporters was not addressed. A basolateral Cl'/HCO3' exchanger
could, for example, provide a transcellular pathway for C1' absorption.

Earlier secretory models also hypothesized that basolateral Na*-K*-2Cl‘ cotransporters
may exist in pancreatic ducts, as has been reported for some Cl'—secreting epithelia (Case
& Argent, 1993). However, the lack of an effect of its specific inhibitor furosemide,

either on membrane voltage (Novak & Greger, 1988a) or on fluid and bicarbonate

88

secretion (Grotrnol et. al., 1986a; Kuijpers et. al., 1984), seems to discount this idea.

Results of the present studies, as well as previous studies on other species (V illanger
et. al., 1995; Zhao et. al., 1994), thus suggest a model for HCO,‘ secretion in the guinea
pig pancreatic duct (Figure 29) which includes basolateral K+ channels, Na*-K*-ATPases,
Na*/HCO3‘ cotransporters, NaVI-I" exchangers, H*-ATPases inserted by vesicular
exocytosis, and apical membrane CFTR Cl' channels and Cl'lHCO,’ exchangers. Our
studies also indicate a role for agonist-evoked H+ pumping activity during stimulated
HCO,‘ secretion, although its contribution to pH regulation appears to be minor compared
to that of Na“/I~I+ exchange. Na"-dependent HCO,’ influx contributed significantly to pH
regulation of these ducts, suggesting that it may play an important role during in viva
HCO,‘ secretion in the guinea pig. The ion transport components of this model are
similar to those presented previously (V illanger et. al., 1994) for the pig, although
species-specific variations in the relative importance of each pH regulatory pathway
appear to exist. Our studies, by examining and confirming the validity of elements of this
model in a third species, suggest that it may constitute a general mechanism for
pancreatic duct secretion of HCO,‘ in mammals.

Our current observations, as well as those of Zhao and coworkers and of Rarder and
colleagues, also are of further Significance as they offer an additional potential site for
neurohumoral regulation of fluid and bicarbonate secretion in the pancreatic duct. Other
putative regulatory sites include the cyclic AMP-sensitive CFTR chloride channel (Gray
et. al., 1988; Gray et. al., 1993), Cay-activated basolateral potassium channels (Gray et.

al., 1990; Novak & Pahl, 1993; Novak & Greger, 1988), and Ca2*-activated chloride

89

Basolateral Hi0 Apical
Side C02 Side

 
 

K“ CA

0 HCO;
+

    
 

+ ,
H CFTR

I

Amiloride

Na‘x

K Cytochalasin
Nooodazole

 
 
   
 

Na" ‘\
”CO; DIDS
Channel
K'I'
. Carrier
. ATPase

Figure 29. Model of HCO,‘ secretion by the guinea pig pancreatic duct principal epithelial
cell. Results of the present study demonstrate the presence of amiloride-sensitive Na"/H+
exchange, DIDS-sensitive Cl'lHCO3' exchange, Na‘lHCO,’ cotransport, and a bafilomycin
A,-sensitive H+-ATPase (see text for details). CA, carbonic anhydrase.

90

channels (Plant et. al., 1993; Gray et. al., 1994), while in rats and pigs, an agonist-
sensitive H+ pumping activity is present (Rzeder, 1992; Zhao et. al., 1994). Regulation of
the CFTR channel has been extensively explored in pancreatic duct cell lines (Kopelman
et. al., 1993; Becq et. al., 1993), CFTR-transfected cells (Travis et. al., 1993; Cliff et. al.,
1992; Tabcharani et. al., 1991), and acutely isolated cells of pancreatic duct origin (Gray
et. al., 1988; Gray et. al., 1993). These studies indicate that both cyclic AMP and PKC
pathways are involved in its activation, and that channel inactivation involves
dephosphorylation by alkaline phosphatase. The basolateral potassium conductance also
is regulated by multiple signalling pathways. Calcium-activated K+ channels are known
to be involved in exocrine gland fluid secretion (Peterson, 1986), and the maxi-K+
channel reported in duct cells is Ca2*-sensitive, with its sensitivity to Ca” increased by
cyclic AMP-dependent phosphorylation (Gray et. al., 1990b). Other K” channels are
thought to contribute to the resting potassium conductance of the duct cell and are pH-
sensitive (Novak & Greger, 1988; Novak & Greger, 1993‘; Plant et. al., 1992).

Whereas studies on some tissues indicate that the Na‘lH+ exchanger can be regulated
by agonists (Bastié & Williams, 1990; Sundaram et. al., 1991), no such evidence exists to
date for the pancreatic duct system (Raeder, 1992). NaVH" exchange appeared to be
unaffected by agonist exposure in guinea pig pancreatic ducts over the duration of the
perfusions, as neither the extent nor the rate of pH recovery was altered (Table 3). Our
observation that a bafilomycin A,-sensitive H+ extrusion mechanism is recruited in
guinea pig pancreatic duct cells following exposure to secretin or carbachol also suggests

the involvement of multiple intracellular signals, since secretin acts on pancreatic duct

91

cells through the adenylyl cyclase - cyclic AMP signalling cascade (Figures 9-12), while
carbachol, an analogue of acetylcholine, acts through the inositol phosphate/calcium
pathway (Stuenkel & Hootman, 1990).

It seems certain that these signalling pathways interact to optimize ductal secretion,
by co-regulation of the CFTR chloride channels, by activating parallel ion transport
pathways, and/or by balancing HCO,’ secretion with H* efflux. Whether the effects of
secretin on amiloride-insensitive pH regulation are actually mediated by cyclic AMP or,
as recently suggested, via intracellular free calcium (Zhao et. al., 1994), remains to be
determined. While several in vitra studies have demonstrated that secretin alone is
sufficient to activate fluid and bicarbonate secretion from the pancreatic duct system,
there is now sufficient evidence for potentiation of this process by calcium-mobilizing
agonists such as ACh and CCK, and demonstration of several sites at which calcium may
act (K+ channels, Cl“ channels, PKC). Clearly, additional regulatory mechanisms exist,
since the mechanisms of action of ATP (Hug et. al., 1994), bombesin (Ashton et. al.,
1991), and substance P (Ashton et. al., 1990) differ significantly from those of secretin or

acetylcholine. Their elaboration deserves future attention.

V. SUMMARY AND CONCLUSIONS

The current studies have established a cyclic AMP response profile in guinea pig
pancreatic ducts for stimulation by secretin, and have characterized regulation of this
intracellular signalling molecule by other neurohumoral agents. Using this model system,
we were also able to investigate the differential responsiveness of main and interlobular
pancreatic ducts to secretin stimulation, and to examine these responses in a cell culture
system. In addition, we used laser scanning confocal microscopy to investigate the
involvement of specific electrolyte transport elements in regulated bicarbonate secretion.
This study took advantage of the pH-sensitive fluorescent dye SNARF -1 , permitting the
analysis of intracellular pH changes in response to activation or blockage of electrolyte

transporters in the principal cell membrane. From these studies, it is concluded that:

1. Isolated guinea pig pancreatic ducts are responsive to secretin, evidenced by a 5 to 8-
fold rise in intracellular cyclic AMP levels in mixed main and interlobular duct

preparations. The calculated EC,0 of the response to secretin is 150 pM.

2. Both main and interlobular ducts respond to secretin stimulation with an increase in
cyclic AMP accumulation. While main ducts are somewhat less responsive, cyclic

AMP levels still increased significantly by 3 to 4-fold. These observations suggest

92

93

that the main pancreatic duct is agonist-responsive and more than a passive conduit in

the delivery of pancreatic juice to the duodenum.

. Epithelial cells cultured from isolated pancreatic ducts also respond to secretin
stimulation, indicating that the signal transduction components involved in
bicarbonate secretion (i.e., secretin receptors functionally coupled to the adenylyl

cyclase system) remain intact in the cultured cells.

. Of several other agents tested, only VIP significantly stimulated cyclic AMP
accumulation in isolated preparations of pancreatic ducts, while somatostatin
significantly inhibited cyclic AMP accumulation evoked by submaximal doses of
secretin. These findings suggest that both VIP and somatostatin regulate pancreatic

duct fluid secretion via modulation of adenylyl cyclase.

. Pancreatic duct principal epithelial cells exhibit Na*/H+ exchange, Cl'/HCO3'
exchange, and NaVHCO,‘ cotransport. Documentation of these transporters confirms
earlier models presented from studies in rats and pigs while illustrating species-

specific variations.

. Pancreatic duct principal epithelial cells also contain an agonist-evoked, bafilomycin-
sensitive I-I+ pump, whose activation is dependent on the integrity of the cellular

cytoskeleton.

94

In summary, mechanisms supporting the regulated secretion of bicarbonate ions from
the pancreatic duct epithelium were studied. Secretin, the major hormonal regulator of
bicarbonate secretion, and the neurotransmitter VIP activate adenylyl cyclase in duct
epithelial cells, prompting an elevation of intracellular cyclic AMP levels. Somatostatin
inhibits ductal fluid and bicarbonate secretion by attenuating the secretin-evoked rise in
cyclic AMP levels, possibly through activation of an inhibitory GTP-binding protein.

The ion transport mechanisms involved in the secretion of bicarbonate by the duct
epithelium appear to be common to several species investigated, including the guinea pig
(present study). They include apical membrane Cl'lHCO,‘ exchangers, basolateral
membrane NaVH+ exchangers and NaVI-ICO,‘ cotransporters, and vesicular H*-ATPases
inserted into the basolateral membrane. However, species-specific patterns of electrolyte
secretion exist, and are reflected by differences in the relative contribution of each
mechanism to intracellular pH regulation. In the guinea pig, sodium-dependent
bicarbonate influx appears to play a more prominent role, as does sodium-hydrogen
exchange, whereas the hydrogen ATPase contributes to a lesser extent. Guinea pig
bicarbonate secretion may therefore be less dependent on carbonic anhydrase, as is the
case in the rabbit. Bicarbonate secretion by the pig pancreas, in contrast, appears to be
driven mainly by primary active hydrogen transport and be entirely dependent on

carbonic anhydrase-generated bicarbonate.

LITERATURE CITED

VI. LITERATURE CITED

Adler, G., and HF. Kern. Regulation of exocrine pancreatic secretory process by insulin
in vivo. Harm. Metab. Res. 7: 290-296, 1975.

Adler, G., and C. Beglinger. Hormones as regulators of pancreatic secretion in man. Eur.
J. Clin. Invest. 20: S27-S32, 1990.

Argent, B.E., S. Arkle, MJ. Cullen, and R. Green. Morphological, biochemical and
secretory studies on rat pancreatic ducts maintained in tissue culture. Quart. J. Exp.
Physio]. 71: 633-48, 1986.

Argent, B.E., S. Arkle, M. Gray, and J .R. Greenwell. Two types of calcitun-sensitive
cation channels in isolated rat pancreatic duct cells. J. Physiol. 386: 82P, 1987 (abstract).

Argent, B.E., N. Ashton, R.L. Evans, A.C. Elliott, and R. Green. New insights into the
regulation of fluid secretion by pancreatic duct epithelium. In: The Pancreatic Duct Cell:
Physiology and Pathophysiology. Prac. Natl. Inst. Health USA, 1991, p. 32-33.

Argent, B.E., S. Githens, S. Kalser, D.S. Longnecker, R. Metzgar, and J.A. Williams. The
pancreatic duct cell. Pancreas 7: 403-419, 1992.

Arkle, S., C.M. Lee, M.J. Cullen, and BE. Argent. Isolation of ducts from the pancreas of
copper-deficient rats. Quart. J. Exp. Physio]. 71: 249-265, 1986.

Arruda, J.A.L., Z. Talor, and C. Dytko. Effect of agents that alter cell calcium and
microfilaments on CO2 stimulated H+ secretion in the turtle bladder. Arch. Int. Pharm.
Ther. 293: 273-283, 1988.

Ashton, N., B.E. Argent, and R. Green. Effect of vasoactive intestinal peptide, bombesin
and substance P on fluid secretion by isolated rat pancreatic ducts. J. Physiol. 427:
471-482, 1990.

Ashton, N., B.E. Argent, and R. Green. Characteristics of fluid secretion from isolated rat
pancreatic ducts stimulated with secretin and bombesin. J. Physio]. 435: 533-546, 1991.

95

96

Ashton, N., R.L. Evans, and BE. Argent. Protein kinase C has an inhibitory role in the
control of fluid secretion in isolated rat pancreatic ducts. J. Physiol. 452: 99P, 1992
(abstract).

Ashton, N., R.L. Evans, A.C. Elliott, R Green, and BE. Argent. Regulation of fluid
secretion and intracellular messengers in isolated rat pancreatic ducts by acetylcholine. J.
Physiol. 471: 549-562, 1993.

Balk, M.W., C.M. Lang, W.J. White, and BL. Munger. Exocrine pancreatic dysfunction
in guinea pigs with diabetes mellitus. Lab. Invest. 32: 28-32, 1975.

Bastié, M.-J., and J.A. Williams. Gastrointestinal peptides activate Na*-H" exchanger in
AR42] cells by increasing its affinity for intracellular H*. Am. J. Physio]. 258: G958-
G966, 1990.

Beauchamp, R.D., R.M. Lyons, E.Y. Yang, R.J. Coffey, and H.L. Moses. Expression of
and response to regulatory peptides by two human pancreatic carcinoma cell lines.
Pancreas 5: 369-380, 1990.

Becq, F. M. Fanjul, M. Merten, C. Figarella, E. Hollande, and M. Gola. Possible
regulation of CFTR-chloride channels by membrane-bound phosphatases in pancreatic
duct cells. FEBS Lett. 327: 337-342, 1993.

Bettinger, J.R., and J .H. Grendell. Intracellular events in the pathogenesis of acute
pancreatitis. Pancreas 6: 52-56, 1991.

Bhattacharyya, E., A. Panchal, TJ. Wilkins, J. de Ondarza, and SR. Hootman. Insulin,
transforming growth factors and substrates modulate growth of cultured guinea pig
pancreatic duct epithelial cells. Gastroenterology 109: 944-952, 1995.

Bloom, E.J., B. Siddiqui, J.W. Hicks, and Y.S. Kim. Effects of sodium butyrate, a
differentiating agent, on cell surface glycoconjugates of a human pancreatic cell line.
Pancreas 4:59-64, 1989.

Bock, O.A.A., S. Bank, I.N. Marks, and W.P.U. Jackson. Exocrine pancreatic function in
diabetes mellitus. S. Afr. Med. J. 41: 756-758, 1967.

Bowman, E.J., A. Siebers, and K.H. Altendorf. Bafilomycins: A class of inhibitors of
membrane ATPases from microorganisms, animal cells, and plant cells. Prac. Nat].
Acad. Sci. USA 85: 7972-7976, 1988.

Bradford, M.M. A rapid and sensitive method for the quantitation of microgram
quantities of protein using the principle of protein dye binding. Anal. Biochem. 72:
248-254, 1976.

97

Brown, D., and I. Sabolic. Endosomal pathways for water channel and proton pump
recycling in kidney epithelial cells. J. Cell Sci. (Suppl. ) 17: 45-59, 1993.

Buanes, T., T. Grotrnol, T. Landsverk, Y. Ridderstrale, and MG. Rreder. Histochemical
localization of carbonic anhydrase in the pig's exocrine pancreas. Acta Physio]. Scand.
128: 437-444, 1986.

Buanes, T., T. Grotrnol, T. Landsverk, and MG. Rmder. Ultrastructure of pancreatic duct

cells at secretory rest and during secretin-dependent NaHCO, secretion. Acta Physiol.
Scand 131: 55-62, 1987.

Buanes, T., T. Grotrnol, T. Landsverk, P. Navstad, and MG. Raeder. Effects of arterial
pH and carbon dioxide on pancreatic exocrine HVI-ICO,‘ secretion and secretin-dependent

translocation of cytoplasmic vesicles in pancreatic duct cells. Acta Physiol. Scand. 133:
1-9, 1988.

Case, R.M., T. Scratcherd, and R.D.A. Wynne. The origin and secretion of pancreatic
juice bicarbonate. J. Physio]. 210: 1-15, 1970.

Case, RM. and BE. Argent. Pancreatic duct cell secretion: control and mechanisms of
transport. In: The Pancreas: Biology, Pathobiology, and Disease. 2nd ed. , edited by
V.L.W. Go, E.P. DiMagno, J .D. Gardner, E. Lebenthal, H.A. Reber, and GA. Scheele.
New York: Raven Press, 1993, p. 301-350.

Chey, W.Y., H. Shay, and CR. Shuman. External pancreatic secretion in diabetes
mellitus. Ann. Int. Med. 59: 812-821, 1963.

Cliff, W.H., R.A. Schoumacher, and RA F rizzell. cAMP-activated Cl channels in
CF TR-transfected cystic fibrosis pancreatic epithelial cells. Am. J. Physiol. 262: C1154-
C1160, 1992.

Cody, S.H., P.N. Dubbin, A.D. Beischer, N.D. Duncan, J .S. Hill, A.H. Kaye, and DA.
Williams. Intracellular pH mapping with SNARF-l and confocal microscopy. 1: A
quantitative technique for living tissues and isolated cells. Micron 24: 573-580, 1993.

Connolly, M.M., P.J. Dawson, F. Michelassi, A.R. Mousa, and F. Lowenstein. Survival
of 1,001 patients with carcinoma of the pancreas. Am. Surg. 206: 366-373, 1987.

Crawford, 1., RC. Maloney, P.L. Zeitlin, W.B. Guggino, S.C. Hyde, H. Turley, K.C.
Gatter, A. Harris, and CF. Higgins. Immunocytochemical localization of the cystic
fibrosis gene product CFTR. Prac. Nat]. Acad. Sci. USA 88: 9262-9266, 1991.

98

De Giorgio, R., C. Sternini, L. Anderson, N.C. Brecha, and V.L.W. Go. Tissue
distribution and innervation pattern of peptide irnmunoreactivities in the rat pancreas.
Peptides 13: 91-98, 1992.

Denning, G.M., L.S. Ostegaard, S.H. Cheng, A.E. Smith, and M.J. Welsh. Localization
of cystic fibrosis transmembrane conductance regulator in chloride secretory epithelia. J.
Clin. Invest. 89: 339-349, 1992.

Dippold, W., H. Kaulen, W. Schwable, M. Kerlin, and KM. Meyer zum Buschenfelde.
Gastric factors, cytokines, and adhesion molecules in pancreas carcinoma cells. Digestion
52: 78, 1992.

Domschke, W., F. Tympner, S. Domschke, and L. Demling. Exocrine pancreatic function
in juvenile diabetics. Dig. Dis. 20: 309-312, 1975.

Dudek, R.W., I.E. Lawrence Jr., R.S. Hill, and RC. Johnson. Induction of islet
cytodifferentiation by fetal mesenchyme in adult pancreatic ductal epithelium. Diabetes
40: 1041-1048, 1991.

Dumont, M., C. D'Hont, Y. Lamri, A.M. Durand-Schneider, E. Jacquemin, G. F eldrnan,
and S. Erlinger. Effects of phalloidin and colchicine on diethyl maleate-induced
choleresis and ultrastructural appearance of rat hepatocytes. Liver 14: 308-313, 1994.

Evans, L.A.R., D. Pirani, D.I. Cook, and J .A. Young. Intraepithelial current flow in rat
pancreatic secretory epithelia. Pflugers Arch. 407: SlO7-S111, 1986.

Evans, R.L., N. Ashton, A.C. Elliott, and R. Green. Effects of acetylcholine on
intracellular calcium and fluid secretion in isolated rat pancreatic ducts. J. Physio]. 446:
352P, 1992 (Abstract).

Farmer, R.C., J. Tweedie, S. Maslin, H.A. Reber, G. Adler, and H. Kern. Effects of bile
salts on permeability and morphology of main pancreatic duct in cats. Dig. Dis. Sci. 29:
740-751, 1989.

Farre, A., P. Benikz, N. Nabau, C. Carcasoma, C. Vidal, G. Capella, F. Lluis, and CM.
Townsend. Bombesin inhibits growth of a ductal pancreatic adenocarcinoma in nude
mice but not in hamsters. Digestion 52: 80, 1992.

Fink, A.S., R.J. Brodish, B.W. Kuvshinoff, M. Irving, and AR. DeMar. Potentiation of
canine pancreatic bicarbonate output by oleic acid is not neurally dependent. Pancreas 9:
230-239, 1994.

99

Fitz, J .G., M. Persico, and BF. Scharschmidt. Electrophysiological evidence for Na“-
coupled bicarbonate transport in cultured rat hepatocytes. Am. J. Physiol. 256: G491 -
G500, 1989. '

Folsch, U.R., and W. Creutzfeldt. Pancreatic duct cells in rats: Secretory studies in
response to secretin, cholecystokinin-pancreozyrnin, and gastrin in vivo.
Gastroenterology 73: 1053-1059, 1977.

Folsch, U.R., H. Fischer, H.-D. Soling, and W. Creutzfeldt. Effects of gastrointestinal
hormones and carbamylcholine on cAMP accumulation in isolated pancreatic duct
fragments fiom the rat. Digestion 20: 277-292, 1980.

Frazier, M.L., S. Pathak, Z.W. Wang, K. Cleary, S.E. Singletary, M. Olive, B. Mackay,
P.A. Steck, and B. Lenn. Establishment of a new human pancreatic adenocarcinoma cell
line, MDAPanc-3. Pancreas 5: 8-16, 1990.

Frier, B.M., J.H.B. Saunders, K.G. Wormsley, and I.A.D. Bouchier. Exocrine pancreatic
function in juvenile-onset diabetes mellitus. Gut 17: 685-691 , 1976.

Geleff, S., and P. Bock. Pancreatic duct glands. II. Lectin binding affinities of ductular
epithelium, ductular glands, and Brunner's glands. Histochem. 80: 31-38, 1984.

Githens, S., D.R.G. Holmquist, J .F. Whelan, and J .R. Ruby. Characterization of ducts
isolated from the pancreas of the rat. J. Cell Biol. 85: 122-135, 1980.

Githens, S., J .J . Finley, C.L. Patke, J .A. Schexnayder, K.B. Fallon, and J .R. Ruby.
Biochemical and histochemical characterization of cultured rat and hamster pancreatic
ducts. Pancreas 2: 427-438, 1987.

Githens, S. The pancreatic duct cell: Proliferative capabilities, specific characteristics,
metaplasia, isolation, and culture. J. Pediatr. Gastroentera]. Nutr. 7: 486-506, 1988.

Githens, S., J .A. Schexnayder, and ML. Frazier. Carbonic anhydrase II gene expression
in mouse pancreatic duct cells. Pancreas 7: 556-561 , 1992.

Githens, S. Differentiation and development of the pancreas in animals. In: The
Pancreas: Biology, Pathabiology, and Disease. 2nd ed. , edited by V.L.W. Go, E.P.
DiMagno, J .D. Gardner, E. Lebenthal, H.A. Reber, and GA. Scheele. New York: Raven
Press, 1993, p. 21-55.

Gray, M.A., J .R. Greenwell, and BE. Argent. Secretin-regulated chloride channel on the
apical plasma membrane of pancreatic duct cells. J. Memb. Biol. 105: 131-142, 1988.

100

Gray, M.A., A. Harris, L. Coleman, J .R. Greenwell, and BE. Argent. Two types of
chloride channel on duct cells cultured from human fetal pancreas. Am. J. Physio]. 257:
C240-C251, 1989.

Gray, M.A., and BE. Argent. Non-selective cation channel on pancreatic duct cells.
Biochim. Biophys. Acta 1029: 33-42, 1990.

Gray, M.A., C.E. Pollard, A. Harris, L. Coleman, J .R. Greenwell, and BE. Argent. Anion
selectivity and block of the small conductance chloride channel on pancreatic duct cells.
Am. J. Physiol. 259: C752-C761, 1990a.

Gray, M.A., J .R. Greenwell, A.J. Garton, and BE. Argent. Regulation of maxi-K+
channels on pancreatic duct cells by cyclic AMP-dependent phosphorylation. J. Memb.
Bio]. 115: 203-215, 1990b.

Gray, M.A., S. Plant, and BE. Argent. cAMP-regulated whole cell chloride currents in
pancreatic duct cells. Am. J. Physio]. 264: C591-C602, 1993.

Gray, M.A., J .P. Winpenny, D.J. Porteous, J .R. Dorin, and BE. Argent. CF TR and
calcium-activated chloride currents in pancreatic duct cells of a transgenic CF mouse.
Am. J. Physio]. 266: C213-C221, 1994.

Gress, T., F. Muller-Pillasch, C. Weber, U. Lacher, H. Friess, M. Buchler, and G. Adler.
Overexpression of a 89-kD heat shock protein in human pancreatic carcinoma. Digestion
52: 86, 1992.

Grotrnol, T., T. Buanes, O. Brors, and MG. Raider. Lack of effect of amiloride,
furosemide, bumetanide and triamterene on pancreatic NaHCO3 secretion in pigs. Acta
Physiol. Scand. 126: 593-600, 1986a.

Grotrnol, T., T. Buanes, and MG. Raider. NN'-dicyclohexylcarbodiimide (DCCD)
reduces pancreatic NaHCO, secretion without changing pancreatic tissue ATP levels.
Acta Physiol. Scand. 128: 547-554, 1986b.

Gullo, L., P. Priori, C. Scarpignato, F. Baldoni, G. Mattioli, and L. Barbara. Effect of
somatostatin 14 on pure human pancreatic secretion. Dig. Dis. Sci. 32: 1065-1070, 1987.

Hall, P.A., and NR. Lemoine. Rapid acinar to ductal transdifferentiation in cultured
human exocrine pancreas. J. Pathol. 166: 97-103, 1992.

Harper, J .F., and G. Brooker. F emtomole sensitive radioimmunoassay for cyclic AMP
and cyclic GMP afier 2'O acetylation by acetic anhydride in aqueous solution. J. Cyclic
Nuc. Res. 1: 207-218, 1975.

101

Harris, A., and L. Coleman. Establishment of a tissue culture system for epithelial cells
derived from human pancreas: A model for the study of cystic fibrosis. .1. Cell Sci. 87:
695-703, 1987.

Harris, A., and L. Coleman. Cultured epithelial cells derived from human foetal pancreas
as a model for the study of cystic fibrosis: Further analysis of the origins and nature of the
cell types. J. Cell Sci. 90: 73-77, 1988.

Hasegawa, H., Y. Okabayashi, M. Koide,Y. Kido, T. Okutani, K. Matsushita, M. Otsuki,
and M. Kasuga. Effect of islet hormones on secretin-stimulated exocrine secretion in
isolated perfused rat pancreas. Dig. Dis. Sci. 38: 1278-1283, 1993.

Holst, J .J . Neural regulation of pancreatic exocrine function. In: The Pancreas: Biology,
Pathobiology, and Disease. 2nd ed, edited by V.L.W. Go, E.P. DiMagno, J.D. Gardner,
E. Lebenthal, H.A. Reber, and G.A. Scheele. New York: Raven Press, 1993, p. 381-402.

Hootman, SR. and CD. Logsdon. Isolation and monolayer culture of guinea pig
pancreatic duct epithelial cells. In Vitro Cell. Dev. Biol. 24: 566-574, 1988.

Hootman, S.R., J. Zukerman, and S.A. Kovalcik. Muscarinic receptors in isolated guinea
pig pancreatic ducts. Biochem. Pharmacol. 46: 291-296, 1993.

Hootman, SR, and J. de Ondarza. Regulation of goblet cell degranulation in isolated
pancreatic ducts. Am. J. Physiol. 268: G24-G32, 1995.

Hootman, S.R., A.R. Panchal, K.D. Boviac, and J. de Ondarza. Endocrine cells and their
products in the guinea pig pancreatic excretory duct system. Gastraenterology (in press),
1996 (abstract).

Hubchak, S., M.M. Mangino, M.K. Reddy, and D.G. Scarpelli. Characterization of
differentiated Syrian golden hamster pancreatic duct cells maintained in extended
monolayer culture. In Vitra Cell. Dev. Biol. 26: 889-897, 1990.

Hug, M., C. Pahl, and I. Novak. Effect of ATP, carbachol and other agonists on
intracellular calcium activity and membrane voltage of pancreatic ducts. Pflugers Arch.
426: 412-418, 1994.

Ishiguro, H., A.R.G. Lindsay, M.C. Steward, and RM. Case. Regulation of intracellular
pH in interlobular ducts isolated from the guinea pig pancreas. J. Physiol. 479P: 10P
(abstract).

Kern, H.F . Fine structure of the human exocrine pancreas. In: The Pancreas: Biology,
Pathobialagy, and Disease. 2nd ed, edited by V.L.W. Go, E.P. DiMagno, J .D. Gardner,
E. Lebenthal, H.A. Reber, and G.A. Scheele. New York: Raven Press, 1993, p. 9-19.

102

Kim, Y.W., H.F. Kern, T.D. Mullins, M.J. Koriwchak, and RS. Metzgar.
Characterization of clones of human pancreatic adenocarcinoma cell line representing
different stages of differentiation. Pancreas 4: 353-362, 1989.

Kodama, T. A light and electron microscopic study on the pancreatic ductal system. Acta
Pathol. Jpn. 33: 297-321, 1983.

Kopelman, H., M. Corey, K. Gaskin, P. Durie, Z. Weizrnan, and G. Forstner. Impaired
chloride secretion, as well as bicarbonate secretion, underlying the fluid secretory defect
in the cystic fibrosis pancreas. Gastraenterology 95: 349-355, 1988.

Kopelman, H., C. Gauthier, and M. Bomstein. Antisense oligodeoxynucleotide to the
cystic fibrosis transmembrane conductance regulator inhibits cyclic AMP-activated but
not calcium-activated cell volume reduction in a human pancreatic duct cell line. J. Clin.
Invest. 91: 1253-1257, 1993.

Kuijpers, G.A.J., I.G.P. Van Nooy, J.J.H.H.M. De Pont, and S.L. Bonting. The
mechanism of fluid secretion in the rabbit pancreas studied by means of various
inhibitors. Biachim. Biophys. Acta 778: 324-331, 1984.

Kuijpers, G.A.J., and J .J .H.H.M. De Pont. Role of proton and bicarbonate transport in
pancreatic cell function. Annu. Rev. Physiol. 49: 87-103, 1987.

Kuvshinoff, B.W., A.R. Demar, L. James, D.W. McFadden, and AS. Fink. Effect of
pancreatic denervation and atropine on the pancreatic response to secretin. Pancreas 8:
609-614, 1993a.

Kuvshinoff, B.W., R.J. Brodish, L. James, D.W. McFadden, and AS. Fink. Somatostatin
inhibits secretin-induced canine pancreatic response via a cholinergic mechanism.
Gastroenterology 105: 539-547, 1993b.

Lack, E. Primary tumors of the exocrine pancreas: Classification, overview, and recent
contributions by irnmunohistochemistry and electron microscopy. Am. J. Surg. Pathol.
13: 66-88, 1989.

Lankisch, P.G., G. Manthey, J. Otto, H. Koop, M. Talaulicar, B. Willms, and W.
Creutzfeldt. Exocrine pancreatic function in insulin-dependent diabetes mellitus.
Digestion 25: 211-216, 1982.

Lebenthal, E., A. Lerner, and D.D.K. Rolston. The pancreas in cystic fibrosis. In: The
Pancreas: Biology, Pathobiolagy, and Disease. 2nd ed, edited by V.L.W. Go, E.P.
DiMagno, J .D. Gardner, E. Lebenthal, H.A. Reber, and G.A. Scheele. New York: Raven
Press, 1993, p. 1041-1081.

103

Lee, K.Y., L. Zhou, X.S. Ren, T.-M. Chang, and W.Y. Chey. An important role of
endogenous insulin on exocrine pancreatic secretion in rats. Am. J. Physio]. 258: G268-
6274, 1990.

Lee, K.Y., Y.L. Lee, C.D. Kim, T.-M. Chang, and W.Y. Chey. Mechanism of action of
insulin on pancreatic exocrine secretion in perfused rat pancreas. Am. J. Physiol. 267:
G207-6212, 1994.

Lee, K.Y., D. Krusch, L. Zhou, Y. Song, T.-M. Chang, and W.Y. Chey. Effect of
endogenous insulin on pancreatic exocrine secretion in perfused dog pancreas. Pancreas
11:190-195,1995.

Lee, S.P., Y.S. Choong, and H.Z. Park. Isolation and partial characterization of a soluble
mucin in human pancreatic juice. Biochem. Cell Biol. 69: 566-571, 1991.

Leviat, J .H., C. Palevody, M. Daurnas, G. Ratovo, and E. Hollande. Differentiation of the
human pancreatic adenocarcinoma cell line (Capan-l) in culture and co-culture with
fibroblasts: Dome formation. Int. J. Cancer 42: 615-621, 1988.

Lieber, M., J. Mazzetta, W. Nelson-Rees, M. Kaplan, and G. Todaro. Establishment of a
continuous tumor-cell line (Pane-1) from a human carcinoma of the exocrine pancreas.
Int. J. Cancer 15: 741-747, 1975.

Lingard, J .M., L. Al-Nakkash, N. Taylor, and BE. Argent. 12’1 efflux from a human
pancreatic adenocarcinoma cell line (BxPC-3). J. Physio]. 446: 354P, 1992 (abstract).

Madden, ME, and MP. Sarras Jr. Distribution of Na*,K*-ATPase in rat exocrine
pancreas as monitored by K*-NNPase cytochemistry and [3H]-ouabain binding: A plasma
membrane protein found primarily to be ductal cell associated. J. Histochem. C ytachem.
30: 864-873, 1982.

Magee, D.F., and S. Naruse. Characteristics of secretin-stimulated pancreatic secretion in
dogs. J. Physio]. 356: 391-399, 1984.

Mai, M., K. Brune, B. Jacoby, H.F. Kern, and J. Mollenhauer. Larrrinin interactions with

ductal pancreatic adenocarcinoma cells: Identification of laminin- and collagen-binding
proteins. J. Cell Sci. 95: 65-74, 1990.

Makino, T., N. Usada, S. Rao, J .K. Reddy, and D.G. Scarpelli. Transdifferentiation of
ductular cells into hepatocytes in regenerating hamster pancreas. Lab. Invest. 62: 552-
561, 1990.

104

Mangino, M.M., S. Hubchak, and D.G. Scarpelli. Stimulation of DNA synthesis in
pancreatic duct cells by gastrointestinal hormones: Interaction with other growth factors.
Pancreas 7: 271-279, 1992.

Marino, C.R., L.M. Matovcik, F .S. Gorelick, and J .A. Cohn. Localization of the cystic
fibrosis conductance regulator in pancreas. J. Clin. Invest. 88: 712-716, 1991.

Marshall, P.J., J .F . Dixon, and LE. Hokin. Prostaglandin E2 derived from
phosphatidylinositol breakdown in the exocrine pancreas facilitates secretion by an action
on the ducts. J. Pharmacol. Exp. Therap. 221: 645-649, 1982.

Matsushita, K., Y. Okabayashi, H. Hasegawa, M. Koide, Y. Kido, T. Okutani, Y.
Sugirnoto, and M. Kasuga. In vitro inhibitory effect of somatostatin on secretin action in
exocrine pancreas of rats. Gastroenterolagy 104: 1146-1152, 1993.

Morisset, J ., S. Chamberland, L. Gilbert, A. Lord, and L. Larode. Study of pancreatic
DNA synthesis in vivo and in vitro following caerulein treatment in vivo. Biomed. Res. 3:
151-158, 1982.

Morris, A.P., S.A. Cunningham, and RA. F rizzell. CFTR targeting in. epithelial
cells. J. Bioenerg. Biamembr. 25: 21-26, 1993.

Novak, I., and R. Greger. Electrophysiological study of transport systems in isolated
perfused pancreatic ducts: properties of the basolateral membrane. Pflugers Arch. 411:
58-68, 1988a.

Novak, I., and R. Greger. Properties of the luminal membrane of isolated perfused rat
pancreatic ducts: Effect of cyclic AMP and blockers of chloride transport. Pflugers Arch.
411: 546-553, 1988b.

Novak, I. Electrolyte transport in pancreatic ducts. In: Epithelial Secretion of Water and
Electrolytes, edited by J .A. Young and P.Y. Wong. Berlin: Springer, 1990, p. 239-252.

Novak, 1., and R. Greger. Effect of bicarbonate of potassimn conductance of isolated
perfused rat pancreatic ducts. Pflugers Arch. 419: 76-83, 1991.

Novak, 1., and C. Pahl. Effect of secretin and inhibitors of HCO3'IH+ transport on the
membrane voltage of rat pancreatic duct cells. Pflugers Arch. 425: 272-279, 1993.

Pahl, C., and I. Novak. Effect of vasoactive intestinal peptide, carbachol and other
agonists on the membrane voltage of pancreatic duct cells. Pflugers Arch. 424: 315-320,
1993.

105

Park, I.-S., and M. Bendayan. Characterization of the endocrine cells in the
pancreatic-bile duct system of the rat. Anat. Record 232: 247-256, 1992.

Peterson, O.H. Calcium-activated potassium channels and fluid secretion by exocrine
glands. Am. J. Physiol. 251: G1-G13, 1986.

Plant, S., M.A. Gray, and BE. Argent. K“ channels on rat pancreatic duct cells in vitro. J.
Physiol. 452: 360P, 1992 (abstract).

Plant, S., M.A. Gray, and BE. Argent. Ionomycin-activated chloride conductances in
isolated rat pancreatic duct cells. J. Physio]. 459: 239P, 1993 (abstract).

Plinoguera, C., D. Shibata, K. Forester, J. Martin, N. Amheirn, and M. Perucho. Most
human carcinomas of the exocrine pancreas contain mutant c-K-ras genes. Cell 53: 549-
554, 1988.

Pour, P.M., T. Lawson, S. Helgeson, T. Donnelly, and K. Stepan. Effect of

cholecystokinin on pancreatic carcinogenesis in the hamster model. Carcinogenesis 9:
597-601, 1988.

Pour, P.M., and RH. Bell. Alteration of pancreatic endocrine cell patterns and their

secretion during pancreatic carcinogenesis in the hamster model. Cancer Res. 46: 6396-
6400, 1989.

Quinton, P.M. Cystic fibrosis: A disease in electrolyte transport. FASEB J. 4: 2709-2717,
1990. .

Reader, M., and 0. Mathisen. Abolished relationship between pancreatic HCO,‘ secretion
and arterial pH during carbonic anhydrase inhibition. Acta Physiol. Scand. 114: 97-102,
1982

Rzeder, M.G. The origin of and subcellular mechanisms causing pancreatic bicarbonate
secretion. Gastraenterolagy 103: 1674-1684, 1992.

Reber, H.A., C. Roberts, and L.W. Way. The pancreatic duct mucosal barrier. Am. J.
Surg. 137: 128-134, 1979.

Riordan, J .R., J .M. Rommens, B. Kerem, N. Alon, R. Rozrnahel, Z. Grzelczak, J.
Zielenski, S. Lok, N. Plavsic, and J .L. Chou. Identification of the cystic fibrosis gene:
cloning and characterization of complementary DNA. Science 245: 1066-1073, 1989.

Sakamoto, T., M. Fujirnura, C.M. Townsend, Jr., G.H. Greeley, and J .C. Thompson.
Interaction of neurotensin, secretin and cholecystokinin on pancreatic exocrine secretion
in conscious dogs. Surg. Gynecol. Obstet. 166: 11-16, 1988.

106

Sarner, M. Pancreatitis definitions and classification. In: The Pancreas: Biology,
Pathobiology, and Disease. 2nd ed. , edited by V.L.W. Go, E.P. DiMagno, J .D. Gardner,
E. Lebenthal, H.A. Reber, and G.A. Scheele. New York: Raven Press, 1993, p. 575-579.

Sato, T., M. Sato, E.A. Hudson, and RT Jones. Characterization of bovine pancreatic
ductal cells isolated by a perfusion-digestion technique. In Vitra 19: 651-660, 1983.

Scarpa, A., K. Mukai, G. Zamboni, P. Capelli, and S. Hirobashi. p53 gene mutations in
pancreatic cancer. Digestion 52: 118, 1992.

Scarpelli, D.G., M.S. Rao, and SK. Reddy. Are acinar cells involved in the pathogenesis
of ductal adenocarcinomas of the pancreas? Cancer Cells 3: 27 5-277, 1991.

Schoumacher, R.A., J. Ram, M.C. Ianuzzi, N.A. Bradbury, R.W. Wallace, C.T. Hon,
D.R. Kelly, S.M. Schmid, F.B. Gelder, T.A. Rado, and RA. Frizzell. A cystic fibrosis
pancreatic adenocarcinoma cell line. Proc. Nat]. Acad. Sci. USA 87: 4012-4016, 1990.

Schussler, M.H., A. Skordy, F. Ramaekers, and F.X. Real. Intermediate filaments as
differentiation markers of normal pancreas and pancreas cancer. Am. J. Pathol. 140: 559-
568, 1992.

Schwartz, G.J., and Q. Al-Awqati. Regulation of transepithelial H+ transport by
exocytosis and endocytosis. Ann. Rev. Physiol. 48: 153-161, 1986.

Sessa, F., M. Bonato, B. Frigerio, C. Capella, E. Solcia, M. Prat, J. Bara, and I.M.
Samhoff. Ductal cancers of the pancreas frequently express markers of gastrointestinal
epithelial cells. Gastroenterolagy 98: 1655-1665, 1990.

Singer, M.V., W. Niebel, S. Kniesburges, D. Hoffmeister, and H. Goebell. Action of
atropine on the pancreatic secretory response to secretin before and after cutting the
extrinsic nerves of the pancreas in dogs. Gastraenterolagv 90: 355-361 , 1986.

Singer, M.V. Neurohumoral control of pancreatic enzyme secretion in animals. In: The
Pancreas: Biology, Pathobialogy, and Disease. 2nd ed, edited by V.L.W. Go, E.P.
DiMagno, J .D. Gardner, E. Lebenthal, H.A. Reber, and G.A. Scheele. New York: Raven
Press, 1993, p. 425-448.

Smith, J .P., S.T. Kramer, and J .Y. Cheung. Effects of cholecystokinin on cytosolic
calcium in human pancreatic cancer cells. Reg. Peptides 36: 299-310, 1991.

Smith, J .P., R.V. Yelamarty, S.T. Kramer, and J .Y. Cheung. Effects of cholecystokinin
on cytosolic calcium in pancreatic duct segments and ductal cells. Am. J. Physiol. 264:
G1177-Gll83, 1993.

107

Spicer, S.S., M.A. Seas, and RE. Tashian. Immunocytochemical demonstration of
carbonic anhydrase in human epithelial cells. J. Histochem. Cytochem. 30: 864-873,
1982.

Steer, M.L. Classification and pathogenesis of pancreatitis. Surg. Clin. North Am. 69:
467-480, 1989.

Stuenkel, E.L., T.E. Machen, and J .A. Williams. pH regulatory mechanisms in rat
pancreatic ductal cells. Am. J. Physiol. 254: G925-G930, 1988.

Stuenkel, E.L., and SR. Hootman. Secretagogue effects on intracellular calcium in
pancreatic duct cells. Pflu'gers Archiv 416: 652-658, 1990.

Sundaram, U., R.G. Knickelbein, and J .W. Dobbins. Mechanism of intestinal secretion:
effect of cyclic AMP on rabbit ileal crypt and villus cells. Proc. Natl. Acad. Sci. USA 88:
6249-6253, 1991.

Swanson, CH, and AK. Solomon. Micropuncture analysis of the cellular mechanisms
of electrolyte secretion by the in vitro rabbit pancreas. J. Gen. Physio]. 65: 22-45, 1975.

Szende, B., G. Srkalovic, A.V. Schally, K. Lapis, and K. Croot. Inhibitory effects of
analogs of luteinizing horrnone-releasing hormone and somatostatin on pancreatic

cancers in hamsters: Events that accompany tumor regression. Cancer 65: 2279-2290,
1990. '

Tabcharani, J .A., X.-B. Chang, J .R. Riordan, and J .W. Hanrahan. Phosphorylation-
regulated Cl' channel in CHO cells stably expressing the cystic fibrosis gene. Nature
352: 628-631, 1991.

Thomas, J .A., R.N. Buchsbaum, A. Zimniak, and E. Racker. Intracellular pH
measurements in Ehrlich ascites tumor cells utilizing spectroscopic probes generated in
situ. Biochemistry 18: 2210-221 8, 1979.

Travis, S.M., M.R. Carson, D.R. Ries, and M.J. Welsh. Interaction of nucleotides with
membrane-associated cystic fibrosis transmembrane conductance regulator. J. Biol.
Chem. 268: 15336-15339, 1993.

Veel, T., T. Buanes, E. Engeland, and MG. deer. Colchicine inhibits the effects of
secretin on pancreatic duct cell tubulovesicles and HCO,‘ secretion in the pig. Acta
Physiol. Scand. 138: 487-495, 1990.

Veel, T., T. Buanes, T. Grotrnol, J. Ostensen, and MG. Reader. Secretin dissipates red
acridine orange fluorescence from pancreatic duct epithelium. Acta Physio]. Scand. 141:
221-226, 1991.

108

Veel, T., O. Villanger, M.R. Holthe, E.J. Cragoe Jr., and MG. Raeder. Na*-H+ exchange
is not important for pancreatic HCO,’ secretion in the pig. Acta Physio]. Scand. 144: 23 9-
246, 1992.

Venue, T.B., and SR. Hootman. Regulation of pancreatic duct epithelial grth in vitro.
Am. J. Physiol. 258: G833-G840, 1990.

Villanger, O., T. Veel, and MG. Rzeder. Secretin causes H‘ secretion from intrahepatic
bile ductules by vacuolar-type H*-ATPase. Am. J. Physio]. 265: G719-G724, 1993.

Villanger, O., T.Veel, and MG. Raeder. Secretin causes H*/HCO3' secretion from pig

pancreatic ductules by vacuolar-type H*-adenosine triphosphatase. Gastraenteralagy 108:
850-859, 1995.

Welsh, M.J., M.P. Anderson, D.P. Rich, H.A. Berger, G.M. Denning, L.S. Ostedgaard,
D.N. Sheppard, S.H. Cheng, R.J. Gregory, and A.E. Smith. Cystic fibrosis

transmembrane conductance regulator: A chloride channel with novel regulation. Neuron
8: 821-829, 1992.

Wilkins, T.J., J. de Ondarza, and SR. Hootman. Intracellular mediators of goblet cell
degranulation in isolated pancreatic ducts. Biochem. Biophys. Res. Comm. 205: 423-428,
1994.

Williams, T.M., D.B. Weiner, M.I. Greene, and HG Maguire. Expression of c-erb-B-2 in
human pancreatic adenocarcinomas. Pathobialogy 59: 46-52, 1991.

Yanagisawa, A., K. Ohtake, K. Ohashi, M. Hori, T. Kitagawa, H. Sugano, and Y. Kato.
Frequent c-Ki-ras oncogene activation in mucous cell hyperplasias of pancreas suffering
from chronic inflammation. Cancer Res. 53: 953-956, 1993.

Yoshimori, T., A. Yarnamoto, Y. Moriyarna, M. Futai, and Y. Tashiro. Bafilomycin A,,
a specific inhibitor of vacuolar-type H*-ATPase, inhibits acidification and protein
degradation in lysosomes of cultured cells. J. Biol. Chem. 266: 17707-17712, 1991.

You, C.H., J. Ruminger, and W.Y. Chey. Potentiation effects of CCK-8 on pancreatic
bicarbonate secretion stimulated by a physiological dose of secretin in humans.
Gastroenteralagv 85: 40-45, 1983.

Yunis, A.A., G.K. Arirnura, and DJ. Russin. Human pancreatic carcinoma (Mia Paca-2)
in continuous culture: Sensitivity to asparaginase. Int. J. Cancer 19: 128-135, 1977.

Zeilhofer, H.V., J. Mollenhauer, and K. Byme. Selective growth inhibition of ductal
pancreatic adenocarcinoma cells by the lysosomotropic agent chloroquine. Cancer Lett.
44: 61-66, 1989.

 

109

Zhao, H., R.A. Star, and S. Muallem. Membrane localization of H” and HCO,’
transporters in the rat pancreatic duct. J. Gen. Physio]. 104: 57-84, 1994.

GAN STATE UNIV. LIB

"11111111 11101111 1112111115111111111116111155