Intracellular Transport of Pancreatic Enzymes

L. J. COOK, 0. A. MUSA & R. M. CASE

School of Biological Sciences, University of Manchester, Manchester, UK

Cook JJ,Musa OA, Case RM. Intracellular transport of pancreatic enzymes. Scand J Gastroenterol
1996;31 Suppl 219:l-5.
Most pancreatic secretory proteins are packaged within the trans-Golgi network into zymogen granules,
which are secreted in a regulated manner by exocytosis. But others enter alternative, constitutive-like
pathways directed towards both apical and basolateral membranes. Our in-vivo studies suggest that
secretion via the latter type of pathway, which may be responsible for the appearance of pancreatic
enzymes in the circulation, can be increased by stimulation, especially supramaximal stimulation. This
may partly explain the increased concentration of pancreatic enzymes in the circulation in the early stages
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Auckland on 01/06/15

of pancreatitis. The mechanisms by which secretory proteins are sorted into zymogen granules remain
vague. However, dissipation of the normally acidic gradient across the trans-Golgi network in vitro (eg.
with NH4Cl) inhibits the process by which newly synthesized proteins reach zymogen granules.
However, secretion via the constitutive-like pathways is apparently not increased under these conditions.
Thus, although the acidic milieu of the trans-Golgi network plays a role in pancreatic protein sorting, it
may not be the mechanism by which constitutive-like secretion of pancreatic enzymes is increased.
Key words: Regulated secretion; constitutive secretion; trans-Golgi network; protein sorting;
experimental pancreatitis
Dr R. M.Case, School of Biological Sciences, G.38 Stopford Building, University of Manchester, Oxford
Road, Manchester, MI3 9PT, UK
For personal use only.

secretion (i.e. one which bypasses zymogen granules) was

ROUTES OF PROTEIN SECRETION
Exportable proteins are synthesized on mRNA molecules
bound to the endoplasmic reticulum in such a way that the
nascent proteins are directed into the cisternae of the
endoplasmic reticulum. Following post-translational proces-
sing of the nascent proteins in the endoplasmic reticulum and
Golgi complex, they reach the trans-Golgi network where
they are directed to their various destinations, namely
secretory granules, lysosomes and small vesicles secreted
constitutively across the plasma membrane. It would seem
likely that sorting in the trans-Golgi network is determined
by a signalheceptor mechanism. This is true for lysosomal
enzymes, each of which possesses a mannose 6-phosphate
residue. This residue results in its recognition by a receptor
molecule in the Golgi complex and its direction thereafter
towards a prelysosomal compartment. The acidic pH in this
compartment causes detachment of the lysosomal enzyme
from the receptor, which is recycled (1). However, in the
case of secretory proteins, no common structural feature or
sequence homology has yet been identified which might act
as a sorting signal. One proposal is that proteins destined for
secretory granules aggregate into a core which excludes
proteins secreted constitutively (2).
The dominance of the regulated secretory pathway in the
pancreatic acinar cell has perhaps delayed recognition of the
coexistence of constitutive pathways, which are the main
route of secretion in cells such as fibroblasts. The occurrence
in the pancreatic acinar cell of a specific constitutive route of
first convincingly demonstrated by in-vitro pulse-labelling
studies in fetal (3) and adult (4)rat pancreas. These studies
showed that release of labelled proteins occurred in two
phases. The first rapid phase (approximately 15% of labelled
proteins) was unaffected by stimulation (though physiological
regulation of this phase was not excluded), while the second
slower phase could be greatly accelerated by stimulation (4).
Using an in-vivo method, this latter study further demon-
strated that the constitutive-like secretion occurred not only
across the apical cell membrane but also across basolateral
membranes (Fig. 1). This may be of some importance. For
unknown reasons, pancreatic enzymes are present in the
circulation at all times and their concentrations increase
during pancreatic disease. Could constitutive-like secretion
be responsible for this phenomenon?
Our studies in the anaesthetized rat summarized in Table I
support this idea (5). We measured the release of amylase into
pancreatic juice, intestinal lymph, and portal blood using
radioactive labelling techniques to distinguish between
enzymes stored in zymogen granules (labelled amino acids
given 24 h before the experiment) and newly synthesized
enzymes (labelled amino acids given immediately before the
experiment).
In an attempt to summarize a large body of work, Table I
only presents data before stimulation (0-30 min) and at one of
our time-points after stimulation (60-90 min). In some cases,
effects were more pronounced at other time-points. For
pancreatic juice and lymph, the data represent output over the
 2 L. J. Cook et al.

30-min period; for serum, they represent concentrations at 30

and 90 min, respectively. All data are mean f SEM. Stored
protein represents proteins labelled by injecting 100 pCi
tritiated mixed amino acids 24 h before the experiment.
Newly synthesized protein represents proteins labelled
following a similar injection at time 0. Labelled protein
represents total protein; in lymph and serum not all of this will
be pancreatic secretory proteins.
Following maximal stimulation with cholecystokinin-
octapeptide (CCK-8), 1 p a@ , there was a significant
increase in the release of amylase into pancreatic juice,
which was not observed following supramaximal stimulation
(CCK-8, 10 pa@).Maximal CCK-8 stimulation also
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Auckland on 01/06/15

increased the release of both stored and newly synthesized

radiolabelled protein into pancreatic juice, reflecting the
exocytosis of zymogen granules coupled with the release of
newly synthesized protein. Supramaximal stimulation
+I +I
blocked the release of both stored radiolabelled protein and
newly synthesized radiolabelled protein into pancreatic juice.
In lymph and serum, there was no increase in stored
radiolabelled protein but an increase in the release of newly
synthesized protein. This occurred with either maximal or
supramaximal stimulation, suggesting that this is the
intracellular source for the elevated amylase output in lymph
For personal use only.

and serum.
Our data suggest that the increase evoked by CCK is
preferentially caused by an enhanced secretion of newly

+I +I

Fig. 1. Diagrammatic model of the routes and rates of protein

discharge from the pancreatic acinar cell based on kinetic studies of
secretion in the unstimulated rat pancreas. The half-time (tin) fof
secretion along each route is indicated. Abbreviations: G = golgi
complex; CV = condensing vacuole; IG = immature granule;
ZG = mature zymogen granule. (Adapted with permission from
Arvan P, Castle JD. J Biol Chem 1987; 104: 243-252.)
 Intracellular Transport of Pancreatic Enzymes 3

0 Control

-
NH4CI (10 mol)

r ACh (10" mol)

ACh (10" mol)

"t
A
I
u
t9 9
h

1st chase
T

2nd chase
Incubation period
3rd chase
6M
3

0
1st chase 2nd cha3e
Incubation period
3rd
:hase
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Auckland on 01/06/15

Fig. 2. The effect of NH4CI (10 mmol) on the release of newly synthesized protein and amylase from
pancreatic pieces. Newly synthesized proteins were labelled using a 10-min pulse period and their release
was monitored for two consecutive 60-min periods (1st and 2nd chase) and during a subsequent period of
stimulation (solid bar) with acetylcholine (ACh) ( mol). Samples obtained after N&CI incubation
are shown by the solid column (m) compared to control samples indicated by the open columns (0). The
data represented are the mean ? SEM.For the newly synthesized protein data, n = 6 (test and control),
and for the amylase data, n = 5 . The effect of stimulation was compared to the corresponding value after
the 2nd chase period using a paired Student's t test (*p< 0.05; * * p < 0.01).

synthesized enzymes. Thus, our data support the concept of a completely efficient so that, for example, some lysosomal
direct, constitutive-like pathway from the trans-Golgi net- enzymes are secreted into pancreatic juice following
For personal use only.

work to the basolateral membrane, by which newly
synthesized enzymes reach the circulation, and suggest that
this pathway is enhanced by stimulation, especially supra-
maximal stimulation.
INTRACELLULAR SORTING OF PROTEINS
As mentioned above, the trans-Golgi network is responsible
for sorting proteins between a number of routes (Fig. 1). In
the pancreatic acinar cell, this sorting mechanism is not
stimulation (6,7). There has been speculation, but rather
little evidence, that co-localization of lysosomal and
secretory enzymes may be part of the mechanism which
triggers pancreatitis (7,8). However, it is certainly true that
experimental pancreatitis causes abnormal structural features
to develop in the acinar cell, most notably the formation of
large vacuolar structures in the trans-Golgi region (9).
Similar structures have been observed in the human pancreas
during acute pancreatitis (10). These fragile structures
contain both secretory and lysosomal enzymes (11) suggest-

0 Control Chloroquine(200 w o l )

CCK-8 (10-9 mol)

-
CCK-8 (lo-' mol)

i
7

1st chase 2nd chase

Incubation period
li
3rd chase
1st chase 2nd cha3e
Incubation period
3rd chase

Fig. 3. The effect of chloroquine (200 kmol) on the release of newly synthesized protein and amylase
from pancreatic pieces. The protocol was as described in Fig. 2. Samples obtained after chloroquine
incubation are shown by the solid bars (m)compared to control samples indicated by the open bars (0).
The data represented are at the mean f SEM (n = 8). The effect of stimulation was compared to the
corresponding value after the 2nd chase period using a paired Student's t test (*p< 0.05). Abbreviation:
CCK-8 = cholecystokinin-octapeptide.
 4 L. J. Cook et al.

0 Control NH4C1(10 mmol)

125 r

c
100 -

h
4-

-.--
9
8 7s -
x

ll
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Auckland on 01/06/15

2s I

0 L PCB
L1
PEL L2 CHYM AMY 'I2
PCA

Enzyme
Fig. 4. The effect of N&CI (10 mmol) on the intracellular transport of newly synthesized pancreatic
enzymes to zymogen granules. Newly synthesized proteins were labelled by using a 10-min pulse period
followed by two consecutive 60-min chase periods. Zymogen granules were then isolated and the
pancreatic enzymes present were separated by hydrophobic interaction chromatography (14). Samples
obtained after NH4CI incubation are shown by the solid bars (m) compared to control samples indicated
For personal use only.

by the open bars (0). The data are represented as the mean t SEM (n = 5). Newly synthesized
radioactivity in each enzyme after NH4CI treatment was compared to the corresponding enzyme in
control samples using a paired Student's t test (*p< 0.05). Abbreviations: PCAPCB = procarboxy-
peptidase AB; L = lipase; T = trypsinogen; PEL = proelastase; CHYM = chymotrypsinogen; AMY =
amylase.

ing that mis-sorting of protein in the trans-Golgi network
may increase in pancreatitis (9).
In many systems, a reduced pH in the trans-Golgi network
is required for targeting proteins to the secretory pathway. In
the pancreatic acinar cell, the pH of immature secretory
granules is low (12). This acidic pH may be necessary for the
aggregation of secretory proteins which is proposed to be part
of the sorting mechanisms. If so, might the diversion of
secretory proteins to the constitutive-like pathways described
above, and which has been suggested to occur by others (9),
reflect changes in intraorganelle pH?
To assess this possibility, we have studied the effect of
disrupting intraorganelle pH using weak bases. Pieces of
pancreas obtained from male Sprague-Dawley rats were
incubated with or without either NH&l (10mmol) or
chloroquine (200 pmol) for 60 min. The pieces were labelled
for 10 min with a mixture of tritiated amino acids, washed and
then incubated for two 60-min chase periods in media
containing an excess of non-radioactive amino acids,
followed by a 30-min chase period of stimulation with a
maximal concentration of either acetylcholine (ACh) or
CCK-8. Samples of incubation media from all three chase
periods were assayed for amylase and radiolabel1ed.protein.
Both N b C l and chloroquine caused the dissipation of
intraarganelle pH gradients, as assessed by visualizing
acridine orange fluorescence of zymogen granules (data not
shown).
The effects of NH4Cl are shown in Fig. 2. Incubation with
this weak base did not affect the unstimulated release of
newly synthesized protein (i.e. labelled protein) or pre-
existing proteins (i.e. amylase). Fallowing stimulation with
ACh, there was a significant increase in the release of newly
synthesized protein under control conditions, but not in the
presence of NH&l. By contrast, N K C l did not influence the
secretion of amylase. The data suggest that NH4Cl inhibits or
reduces the incorporation of newly synthesized protein into a
regulated store (e.g. the zymogen granules). However, NH4Cl
did not increase the unstimulated release of newly synthesized
protein, suggesting it does not divert pancreatic enzymes from
the regulated to a constitutive pathway. Similar, though less
clear-cut, results were obtained using chloroquine (Fig. 3). In
these experiments, CCK-8 was used as a stimulus, as
chloroquine acts as a muscarinic antagonist (13). An increase
in the release of newly synthesized protein was observed
under control conditions, but not in the presence of
chloroquine.
The inhibitory effects of N b C l and chloroquine were
further defined by measuring the content of individual, newly
synthesized proteins in zymogen granules, after pulse-
labelling pancreatic tissue for 10 min, followed by 2 h

incubation in the presence of NH&l (10mmol) or chloro-
quine (200 pmol). While the content of lipase and chymo-
trypsinogen were significantly reduced, the content of some
other enzymes was not changed by NH&l (Fig. 4). This
suggests that intraorganelle pH may influence the sorting of
some secretory enzymes more than others. If this is so, such
selective sorting may contribute to the link between
pancreatic stimulation and the existence of non-parallel
enzyme secretion (15). By contrast, chloroquine did not
influence the content of radiolabelled proteins in zymogen
granules (data not illustrated). The inability of chloroquine (as
compared with methylamine) to affect processing has
previously been observed in mouse pancreas (16) and has
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Auckland on 01/06/15
been correlated with the different effects that these agents
have on morphology: methylamine affects both Golgi and
lysosomal vesicles, whereas chloroquine primarily affects
lysosomal structures (16).
In conclusion, NH4C1, which dissipates intraorganelle pH
gradients, reduced the intracellular transport of pancreatic
enzymes to the zymogen granules and their subsequent
secretion following stimulation. Thus, the maintenance of
appropriate pH gradients across the trans-Golgi network may
play a significant role in the sorting and transport of
pancreatic enzymes to zymogen granules. However, it
For personal use only.
appears from our studies that the maintenance of intraorga-
nelle pH gradients is not the sole factor which determines
protein sorting in the pancreatic acinar cell, as no elevation of
a ‘constitutive-like’ pathway was observed. At present, the
control mechanisms responsible for the enhanced release of
newly synthesized pancreatic enzymes into the circulation,
especially during pancreatitis, remain to be determined.
REFERENCES
1. Kornfeld S. Trafficking of lysosomal enzymes. FASEB Journal
1987;1:462-8.
Intracellular Transport of Pancreatic Enzymes 5
2. Tooze J, Kern HF, Fuller SD, Howell KE. Condensation-sorting
events in the rough endoplasmic reticulum of exocrine
pancreatic cells. J Cell Biol 1989;109:35-50.
3. Arvan P, Chang A. Constitutive protein secretion from the
exocrine pancreas of fetal rats. J Biol Chem 1987;262:3886-90.
4. Arvan P, Castle JD. Phasic release of newly synthesized
secretory proteins in the unstimulated rat exocrine pancreas. J
Cell Biol 1987;104:243-52.
5. Fletcher LI,Seehra H, Hadlev H, Case RM. Routes of pancreatic
enzyme secretion in the anaesthetized rat. Digestion 1992;52:82.
6. Rinderknecht H, Renner IG, Koyama HH. Lysosomal enzymes
in pure pancreatic juice from normal healthy volunteers and
chronic alcoholics. Dig Dis Sci 1979;24:180-6.
7. Hirano T, Saluja A, Ramarao P, Lerch MM, Saluja M, Steer ML.
Apical secretion of lysosomal enzymes in rabbit pancreas occurs
via a secretagogue related pathway and is increased after
pancreatic duct obstruction. J Clin Invest 1991;87:865-9.
8. Steer ML, Meldolesi J, Figarella C. Pancreatitis: the role of
lysosomes. Dig Dis Sci 1984;29:934-8.
9. Gorelick FS, Adler F, Kern HF. Caerulein-induced pancreatitis.
In: Go LVW, DiMagno EP, Gardner JD, Lebenthal E, Reber
HA, Scheele GA, eds. The pancreas: biology, pathobiology and
disease. Raven Press, New York, 1993:501-26.
10. Kloppel G, Dreyer T, Willemer S, Kern HF, Adler G. Human
acute pancreatitis: its pathogenesis in the light of immunocyto-
chemistry and ultrastructural findings in acinar cells. Virchows
Arch 1986;407:791-803.
11. Watanabe 0, Baccino FM, Steer ML, Meldolesi J. Supramaxi-
ma1 caerulein stimulation and ultrastructure of rat pancreatic
acinar cell: early morphological changes during development of
experimental pancreatitis. Am J Physiol 1984;246:G457-67.
12. Orci L, Ravazzola M, Anderson RGW. The condensing vacuole
of exocrine cells is more acidic than the mature secretory
vesicle. Nature 1987;32677-9.
13. Habara Y, Williams JA, Hootman SR. Antimuscarinic effects of
chloroquine in rat pancreatic acini. Biochem Biophys Res
Comm 1986;137:664-9.
14. Padfield PJ, Case RM. Separation of proteins present in
pancreatic juice using hydrophobic interaction chromatography.
Anal Biochem 1988;171:294-9.
15. Rothman S, Liebow C, Grendell J. Non-parallel transport and
mechanisms of secretion. Biochem Biophys Acta 1991;1071:
159-73.
16. DeLisle RC, Williams JA. Zymogen granule acidity is not
required for stimulated pancreatic protein secretion. Am J
Physiol 1987;253:G711-19.