Intra-acinar cell activation of trypsinogen during
caerulein-induced pancreatitis in rats
B. HOFBAUER,1 A. K. SALUJA,1 M. M. LERCH,2 L. BHAGAT,1 M. BHATIA,1 H. S. LEE,1
J. L. FROSSARD,1 G. ADLER,3 AND M. L. STEER1
1Department of Surgery, Beth Israel Deaconess Medical Center and Harvard Medical School,
and Harvard Digestive Diseases Center, Boston, Massachusetts 02215; 2Department
of Medicine B, Westfaelische Wilhelms-Universitaet, 48129 Muenster; and 3Department
of Gastroenterology, University of Ulm, 89070 Ulm, Germany
pancreas; immunolocalization; trypsinogen activation peptide; trypsin; subcellular fractionation; cathepsin B
is generally believed to be an
autodigestive disease in which the pancreas is injured
by enzymes that are normally secreted by acinar cells.
Under physiological conditions, most of these potentially harmful digestive enzymes are synthesized and
secreted by acinar cells as inactive proenzymes or
zymogens with activation occurring in the duodenum.
There, the brush-border enzyme enteropeptidase cleaves
the NH2-terminal trypsinogen-activation peptide (TAP)
from trypsinogen, releasing active trypsin that can
then catalyze the activation of the other zymogens. In
contrast, during the early stages of pancreatitis, digestive enzyme zymogens, including trypsinogen, are beACUTE PANCREATITIS
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
G352
lieved to be prematurely activated in the pancreas (1, 7,
17, 24, 25, 30, 35).
The location in the pancreas of premature trypsinogen activation and the fate of activated trypsin during
the early stages of pancreatitis are issues of considerable importance and interest, but a clear understanding of these phenomena has not yet been reached.
Studies exploring these issues in clinical pancreatitis
have not been possible, because pancreatic tissue is
generally not available for examination during the
early stages of pancreatitis in humans. To overcome
this limitation, investigators have employed a number
of experimental models of pancreatitis, which permit
examination of pancreas samples during the very early
stages of pancreatitis.
Secretagogue (caerulein)-induced pancreatitis is, perhaps, the most frequently employed model of experimental pancreatitis (6). Caerulein is an analog of the
pancreatic secretagogue CCK. Rats infused with a dose
of caerulein exceeding that required for maximal digestive enzyme secretion develop acute edematous pancreatitis (16). In time-dependence studies performed by
our laboratory (7), we have recently noted that activated trypsinogen and elevated TAP levels can be
detected in the pancreas within 15–30 min of supramaximal secretagogue stimulation. We have suggested
(32) that the intrapancreatic activation of trypsinogen
is catalyzed by the lysosomal hydrolase cathepsin B,
which, during the early stages of secretagogue-induced
pancreatitis, as well as in other models of pancreatitis,
is colocalized with digestive enzyme zymogens.
In the present study, we report our findings using the
model of secretagogue-induced pancreatitis to identify
the intrapancreatic site of trypsinogen activation and
the fate of activated trypsin. To address these questions, we have used the techniques of subcellular
fractionation and both light and electron microscopy
immunolocalization. Our studies were performed under both in vivo and in vitro conditions. We report that,
in response to supramaximal secretagogue stimulation,
trypsinogen is activated in a particulate intracellular
compartment that is, most likely, composed of cytoplasmic vacuoles. Subsequent to its activation, trypsin
moves from this compartment to a soluble intracellular
compartment that, we suggest, is the acinar cell cytosol.
MATERIALS AND METHODS
Male Wistar rats (125–200 g), obtained from Charles River
Laboratories (Wilmington, MA), were housed in temperature-
0193-1857/98 $5.00 Copyright r 1998 the American Physiological Society
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.6 on June 16, 2017
Hofbauer, B., A. K. Saluja, M. M. Lerch, L. Bhagat, M.
Bhatia, H. S. Lee, J. L. Frossard, G. Adler, and M. L.
Steer. Intra-acinar cell activation of trypsinogen during
caerulein-induced pancreatitis in rats. Am. J. Physiol. 275
(Gastrointest. Liver Physiol. 38): G352–G362, 1998.—Supramaximal stimulation of the pancreas with the CCK analog
caerulein causes acute edematous pancreatitis. In this model,
active trypsin can be detected in the pancreas shortly after
the start of supramaximal stimulation. Incubation of pancreatic acini in vitro with a supramaximally stimulating caerulein concentration also results in rapid activation of trypsinogen. In the current study, we have used the techniques of
subcellular fractionation and both light and electron microscopy immunolocalization to identify the site of trypsinogen
activation and the subsequent fate of trypsin during caeruleininduced pancreatitis. We report that trypsin activity and
trypsinogen-activation peptide (TAP), which is released on
activation of trypsinogen, are first detectable in a heavy
subcellular fraction. This fraction is enriched in digestive
enzyme zymogens and lysosomal hydrolases. Subsequent to
trypsinogen activation, both trypsin activity and TAP move to
a soluble compartment. Immunolocalization studies indicate
that trypsinogen activation occurs in cytoplasmic vacuoles
that contain the lysosomal hydrolase cathepsin B. These
observations suggest that, during the early stages of pancreatitis, trypsinogen is activated in subcellular organelles containing colocalized digestive enzyme zymogens and lysosomal
hydrolases and that, subsequent to its activation, trypsin is
released into the cytosol.
TRYPSINOGEN ACTIVATION IN PANCREATITIS
(2%) and glutaraldehyde (0.01%) for 60 min. For preparation
of ultrathin cryosections, the blocks were immersed in PBS,
infiltrated with increasing concentrations of sucrose (up to
2.3 M) at pH 7.4, and frozen in liquid nitrogen. Sections were
prepared as previously described (11). In brief, after blocking
with horse serum and BSA, we incubated thin sections for 60
min with primary antibody (affinity-purified rabbit anti-TAPantibody, a kind gift from D. McNally, Abbott Laboratories).
These antibodies were raised against TAP. They do not bind to
either the zymogen (trypsinogen) or the active enzyme (trypsin) (7, 10). They were diluted (1:100) in PBS (containing 0.5%
BSA) at 4°C. After repeated rinsing in PBS, incubation with
goat-anti-rabbit IgG conjugated with 15-nm gold particles
(Amersham, Buckinghamshire, UK) followed. Sections were
finally stained with 0.05% uranylacetate and embedded in 2%
methylcellulose before viewing on a Zeiss 109 electron microscope. For controls, pancreatic tissue from animals infused
with saline instead of caerulein was used. A second control
consisted of cryosections incubated without primary TAP
antibody but under otherwise identical conditions. A third
control consisted of cryosections incubated with TAP antibody
that had been kept overnight at 4°C with a 50-fold excess of
antigen (TAP peptide). Micrographs were taken at calibrated
magnifications of 320,000, and prints were coded for morphometry. To determine the relative concentration of TAP
label in the different subcellular compartments, 100 coded
prints from each experiment were morphometrically evaluated in a blinded fashion (20) by placing a graticule with
random fields over the areas occupied by cytoplasmic vacuoles, zymogen granules, cytosol (including endoplasmic reticulum), and the nucleus (as an internal control). The relative
concentration of TAP label was expressed as the number of
15-nm gold grains/0.04 µm2.
Light microscopy immunolocalization of TAP and cathepsin B. For these studies, pancreatitis was induced by infusion
of a supramaximally stimulating dose of caerulein, as described above, for 30 min. Pancreata were fixed in neutralbuffered Formalin for 12 h and after processing were embedded in paraffin wax. Immunocytochemical localization of
cathepsin B and TAP, at the light microscopic level, was
carried out on consecutive sections. Sections 4 µm thick were
cut and mounted on Superfrost/Plus slides. They were deparaffinized by passing them through two changes of xylene and
a graded series of alcohol followed by rinses in tap water and
PBS (0.01 M phosphate buffer, pH 7.4, 0.154 M NaCl),
respectively. Nonspecific binding was blocked by incubating
sections in 1% BSA in PBS for 30 min. Sections were then
incubated for 1 h in the two primary antibodies applied
together (anti-TAP, 1:120 or anti-cathepsin B, 1:100), diluted
in BSA-PBS. Control sections were incubated in BSA-PBS.
After rinsing in PBS, the sections were incubated for 30 min
with the corresponding fluorescent secondary antibodies, i.e.,
FITC-conjugated anti-rabbit IgG and Texas red anti-sheep
IgG. After three rinses in PBS, the sections were mounted in
Vectashield mounting medium.
Assays. Amylase activity was determined according to the
method of Pierre et al. (22), cathepsin B by the method of
McDonald and Ellis (21) with modifications (27), and trypsin
by the method of Kawabata et al. (12). In the trypsin assay,
the slope of rising fluorescence emission was calculated as
arbitrary units and expressed per milligram of DNA in the
homogenate of that particular tissue or acini sample. To allow
for pooling of data from different experiments, we expressed
the trypsin activity in each fraction as a percentage of the
total trypsin activity determined by combining the activity
found in all of the fractions. We measured DNA, using
Hoechst dye 33258, according to the method of Labarca and
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.6 on June 16, 2017
controlled (23 6 2°C) rooms with a 12:12-h light-dark cycle.
They were fed standard laboratory chow and given water ad
libitum. Caerulein was purchased from Research Plus (Bayonne, NJ), pentobarbital from Veterinary Laboratories (Lenexa, KS), V/3 tubing from Biolab Products (Lake Havasu, AZ),
collagenase from Worthington Biochemical (Freehold, NJ),
trypsin substrate (t-butyloxycarbonyl-L-glutaminyl-L-alanylL-arginine 4-methyl-coumaryl-7-amide) from Peptides International (Louisville, KY), Superfrost/Plus slides from Fisher
Scientific (Pittsburgh, PA), Vectashield mounting medium,
fluorescein anti-rabbit IgG, and Texas red anti-sheep IgG
from Vector Laboratories (Burlingame, CA), and sheep antihuman cathepsin B from ICN Pharmaceuticals (Costa Mesa,
CA). All other chemicals and reagents were purchased from
Sigma Chemical (St. Louis, MO). All experimental protocols
were approved by the Institutional Animal Use Committee of
the Beth Israel Deaconess Medical Center.
Animal and tissue preparation. For in vivo studies, animals
were prepared and pancreatic tissue was processed as described previously (7). Similarly, for in vitro studies pancreatic acini were freshly prepared, using collagenase digestion
according to the method of Powers et al. (23). The acini were
incubated in 10 mM HEPES buffer (pH 7.4) containing 0.1
µM caerulein for 15–120 min. They were separated from the
extracellular fluid by centrifugation (60 g for 30 s), and the
pellet, containing the acini, was used for subcellular fractionation (see below).
Subcellular fractionation. For subcellular fractionation of
in vivo-derived samples taken from whole pancreas, portions
of the pancreas were minced in ice-cold 5 mM MOPS buffer
(pH 6.5) containing 250 mM sucrose and 1 mM MgSO4. The
minced pancreas was homogenized with a motorized glassTeflon grinder (0.10- to 0.15-mm clearance) using three up
and down strokes at a motor speed of 3,000 rpm. For
subcellular fractionation of freshly prepared acini studied in
vitro, the acini were washed twice by repeated centrifugation
and resuspension in the same MOPS buffer and homogenized
using the same tissue grinder but with seven up and down
strokes. In both cases, a sample of the homogenate was used
to measure DNA content. The remaining portion was subjected to low-speed centrifugation (150 g for 10 min) to
remove nuclei and unbroken cells. The resulting supernatant
(postnuclear supernatant) was used for subcellular fractionation, according to the method of Tartakoff and Jamieson (33)
as slightly modified by Saluja et al. (28). Three subcellular
fractions were obtained as follows: a zymogen granuleenriched fraction (1,300 g for 15 min pellet), a lysosome/
mitochondria-enriched fraction (12,000 g for 13 min pellet),
and a microsome/cytosol-soluble fraction (12,000 g for 13 min
supernatant). The identity of these fractions was confirmed
by assay of marker enzymes as previously described (28). In
preliminary experiments, the possibility that either trypsin
or TAP was located within microsomes as opposed to being in
the nonsedimentable soluble portion of the 12,000 g supernatant fraction was evaluated by subjecting the 12,000 g
supernatant to high-speed centrifugation (105,000 g for 60
min). All of the active trypsin and TAP was subsequently
recovered in the resulting supernatant, indicating that they
are soluble and not contained in microsomes.
Electron microscopy immunolocalization of TAP. For these
studies, pancreatitis was induced by infusion of a supramaximally stimulating caerulein dose as described above for 30
min or 3.5 h. Control animals received an equal amount of
saline. Animals were killed at the respective time intervals,
the pancreas was quickly removed, and 2-mm tissue blocks
from the head, body, and tail of the organ were fixed by
incubation in PBS (pH 7.4, at 4°C) containing formaldehyde
G353
G354
TRYPSINOGEN ACTIVATION IN PANCREATITIS
Paigen (15), with calf thymus DNA as the standard. TAP
levels were determined using a direct competitive ELISA
according to the method of Gudgeon et al. (9).
Analysis of data. Results represent means 6 SE obtained
from multiple determinations in three or more separate
experiments. In all figures, vertical bars denote SE and the
absence of such bars indicates that the SE is too small to
depict. The significance of changes was evaluated using
Student’s t-test when the data consisted of only two groups or
by ANOVA when comparing three or more groups. If ANOVA
indicated a significant difference, the data were analyzed
using Tukey’s method as a post hoc test for the difference
between groups. P # 0.05 was considered to indicate a
significant difference.
RESULTS
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.6 on June 16, 2017
Subcellular distribution of trypsin activity. We have
previously noted that supramaximal stimulation with
caerulein for 15–30 min causes trypsinogen activation
in the pancreas, but that in the absence of supramaximal stimulation trypsin activity cannot be detected in
the gland (7). A small portion (20–30% of total) of the
trypsin activity detected after 30 min of supramaximal
stimulation is recovered in the lysosome/mitochondriaenriched fraction, and this component does not change
with continued supramaximal stimulation for 210 min
(data not shown). On the other hand, the major portion
(60–65%) of the trypsin activity detected after 30 min of
supramaximal stimulation is recovered from the heavier
zymogen granule-enriched (1,300 g pellet) fraction (Fig.
1). With further supramaximal stimulation for up to
210 min, trypsin activity gradually falls in the 1,300 g
pellet and, during the same period, trypsin activity in
the soluble (12,000 g supernatant) fraction rises (Fig. 1).
A similar pattern for the subcellular distribution of
trypsin activity is detected when freshly prepared
pancreatic acini are exposed to in vitro supramaximal
stimulation with caerulein. Once again, trypsin activity in the acini is only detected when acini are exposed
to a supramaximally stimulating concentration of the
secretagogue, and this trypsin activity can be detected
within 10 min of supramaximal stimulation (30). A
small portion of the activated trypsin is detected in the
lysosome/mitochondria fraction (not shown), but the
greatest portion appears in the heavier, zymogen granule-enriched fraction (Fig. 1). With continued supramaximal stimulation, trypsin activity in the heavy
particulate fraction declines, while that in the 12,000 g
supernatant (soluble fraction) increases. In contrast to
the in vivo situation in which intrapancreatic trypsin
activity gradually increases with prolonged supramaximal stimulation (7), the total trypsin activity in acini
exposed to supramaximal secretagogue stimulation
declines with time (Fig. 2A). Some of this loss of trypsin
activity from the acini can be accounted for by an
increase in trypsin activity in the suspending medium
(Fig. 2B), suggesting that it reflects discharge of trypsin from the acini. Most of the fall in trypsin activity,
however, is not accounted for by discharge into the
medium, and this loss of total trypsin activity appears
to be the result of either trypsin inhibition or autoinactivation.
Subcellular distribution of cathepsin B. We have
previously noted that under in vivo conditions supramaximal secretagogue stimulation changes the sedimentation characteristics of lysosomal hydrolases such
as cathepsin B and, as a result, a greater portion is
recovered in the zymogen granule-enriched fraction
and a lesser portion is recovered in the lysosome/
mitochondria-enriched fraction (27). This shift in cathepsin B distribution can be detected within 15–30
min of the start of supramaximal stimulation and is
believed to be accounted for by the formation of cytoplasmic vacuoles that contain colocalized digestive zymogens and lysosomal hydrolases (27, 29). We now report
that a similar change in the subcellular distribution of
cathepsin B is noted when freshly prepared acini are
incubated in vitro with a supramaximally stimulating
concentration of caerulein. The cathepsin B activity in
the 12,000 g pellet falls and the activity in the 1,300 g
pellet rises. As a result, the ratio of cathepsin B in the
1,300 g pellet to that in the 12,000 g pellet more than
doubles (Fig. 3). This subcellular redistribution of
cathepsin B is detected within 15 min of the start of in
vitro supramaximal stimulation (Fig. 3).
Subcellular distribution of TAP. TAP is the NH2terminal peptide of trypsinogen that is cleaved from the
zymogen during activation (9). Using the techniques of
subcellular fractionation and light and electron microscopy immunolocalization, we have evaluated the distribution of TAP in the pancreas after in vivo supramaximal stimulation. In control samples, a small amount of
TAP can be detected in the absence of supramaximal
secretagogue stimulation, but this level of TAP increases markedly within 30 min of supramaximal
stimulation (7). As shown in Fig. 4, roughly 50% of the
TAP can be recovered and measured, by ELISA, in the
zymogen granule-enriched 1,300 g pellet within 15 min
of supramaximal stimulation. With more prolonged
supramaximal stimulation, TAP levels in this heavy
subcellular fraction fall. Roughly 35% of the TAP can
also be detected in the soluble 12,000 g supernatant
within 15 min of supramaximal stimulation. However,
with more prolonged stimulation, this value gradually
increases and by 210 min nearly all of the TAP measured by ELISA is in the soluble compartment (Fig. 4).
Immunolocalization of TAP at electron microscopic
level. Immunogold labeling of pancreatic cryosections
from control animals with anti-TAP antibody results in
only a weak background gold decoration (Fig. 5A). This
background label does not appear to indicate that TAP
is present in acinar cells under physiological conditions, because it is also found on sections that are
incubated without primary antibody (Fig. 5B). When
the primary antibody is preincubated with a 50-fold
excess of antigen (TAP peptide), no specific gold decoration beyond background levels results, regardless of
whether or not the cryosections are from caeruleininfused or control animals. Even the cytoplasmic vacuoles that are characteristic of experimental pancreatitis remain negative under these conditions (Fig. 5C),
but they are strongly TAP positive (Fig. 5D, inset) when
the primary antibody is not preincubated with antigen.
TRYPSINOGEN ACTIVATION IN PANCREATITIS
G355
After 30 min of supramaximal secretagogue stimulation, these vacuoles represent the principal TAPcontaining compartment and neither zymogen granules nor the acinar lumen (Fig. 5D) is found to be
significantly labeled with anti-TAP antibodies. Even
the cytosol in the immediate vicinity of TAP-positive
cytoplasmic vacuoles (Fig. 6A) and the adjacent endoplasmic reticulum (Fig. 6B) contain only background
gold decoration at 30 min of supramaximal stimulation.
After 3.5 h of caerulein infusion, the TAP-specific gold
label becomes much less distinct. The surrounding
cytoplasm near the cytoplasmic vacuoles becomes
strongly TAP positive (Fig. 6C) and only the nucleus,
mitochondria, and zymogen granules (Fig. 6C, inset)
remain without gold decoration beyond background
levels. At this time, the endoplasmic reticulum appears
progressively dilated (Fig. 6D) and its TAP label is
comparable to that of other cytoplasmic structures.
To quantitate the concentration of TAP in the respective subcellular compartments, we performed a morphometric evaluation of coded micrographs taken at a
calibrated magnification. The results of these experiments (Fig. 7) confirm our suggestion that, after 30 min
of supramaximal secretagogue stimulation, the cytoplas-
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.6 on June 16, 2017
Fig. 1. In vivo and in vitro subcellular distribution of trypsin activity. For in vivo studies,
rats were infused with caerulein (10
µg · kg21 · h21 ) for up to 210 min. Subcellular
fractions were prepared by differential centrifugation, and trypsin activity was measured in each fraction (A and B) as described
in the text. A: 1,300 g pellet. B: 12,000 g
supernatant. For in vitro studies, pancreatic
acini were prepared by collagenase digestion
and stimulated with 0.1 µM caerulein for up
to 60 min. Acini were homogenized, and subcellular fractions were prepared as described in
the text. Trypsin activity was measured in
each fraction (C and D). C: 1,300 g pellet. D:
12,000 g supernatant. Trypsin activity in each
fraction was calculated as % of that noted in
the postnuclear supernatant. Results are
means 6 SE obtained from 4 different experiments. * P # 0.05 compared with 30-min
values.
G356
TRYPSINOGEN ACTIVATION IN PANCREATITIS
mic vacuoles represent the only compartment in which
the concentration of TAP-specific gold label is elevated
above background levels. At this time, the TAP concentration does not appear to be increased in either the
cytosol or zymogen granules and, in those areas, it is
not found to be different from the label over the nuclei
that were studied as an internal control. After 3.5 h of
supramaximal stimulation with caerulein, cytoplasmic
vacuoles still represent the principal TAP-containing
compartment, but they are not the only compartment
with significant TAP label. Although the density of gold
decoration in vacuoles has increased above that noted
at 30 min, the cytosol now contains an equally significant concentration of TAP. Because the volume of the
cytosolic compartment greatly exceeds that of the vacuolar compartment, these observations indicate that
most of the TAP is in the cytosolic space after 3.5 h of
caerulein infusion. The nuclei and zymogen granules at
this time point still remain at background levels.
Light microscopy immunolocalization of TAP and
cathepsin B. To determine whether the vacuoles that
contain TAP shortly after the onset of supramaximal
stimulation also contain the lysosomal hydrolase cathepsin B, we performed immunolocalization studies at the
light microscopic level of resolution using antibody
directed against either TAP or cathepsin B. As shown in
Fig. 8, both TAP (Fig. 8A) and cathepsin B (Fig. 8B) are
localized to cytoplasmic vacuoles. Figure 8C, which is
the merged image obtained when both anti-TAP and
anti-cathepsin B antibodies are used, indicates that
both enzymes are localized in the same vacuoles.
DISCUSSION
Fig. 3. In vitro subcellular distribution of cathepsin B. Pancreatic
acini were prepared by collagenase digestion and incubated with
either buffer alone or buffer containing 0.1 µM caerulein for up to 120
min at 37°C. Acini were homogenized, and subcellular fractions were
prepared by differential centrifugation as described in text. Cathepsin B activity was measured in both the 1,300 and 12,000 g pellets
and expressed as a ratio between these activities in these 2 fractions.
Results are means 6 SE obtained from 3 or more separate experiments. * P # 0.05, caerulein vs. control.
The pathogenesis of acute pancreatitis has been the
recent subject of intensive investigation and considerable controversy. Although most investigators agree
that the disease is initiated by digestive enzyme activation, the location at which activation occurs and the
fate of the enzymes, once activated, have not been
unequivocably established. Some have suggested that
enzyme activation occurs at an extracellular site in the
gland parenchyma (2) and that pancreatic injury be-
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.6 on June 16, 2017
Fig. 2. In vitro trypsin activity in acini
homogenate and incubation medium. Acini
were incubated with either buffer alone or
buffer containing 0.1 µM caerulein for up
to 120 min at 37°C. Acini were separated
from incubation medium by centrifugation, resuspended in 5 mM MOPS buffer,
and homogenized. Trypsin activity was
measured in acini homogenate (A) and
incubation medium (B). Crosshatched bars,
samples incubated with buffer only (control). Solid bars, samples incubated in
buffer containing caerulein. Results are
means 6 SE obtained from 3 or more
separate experiments. * P # 0.05 compared with control.
TRYPSINOGEN ACTIVATION IN PANCREATITIS
G357
Fig. 4. In vivo subcellular distribution of trypsinogen-activated peptide (TAP). Rats were infused
with caerulein (10 µg · kg21 · h21 ) for up to 210
min. Subcellular fractions were prepared by differential centrifugation, and TAP was measured
in each fraction as described in text. A: 1,300 g
pellet. B: 12,000 g supernatant. Results are
means 6 SE obtained from 3 separate experiments. * P # 0.05 compared with 15-min values.
can be found in an acinar cell particulate fraction 2 h
after the onset of supramaximal secretagogue stimulation. We reasoned that measurement of trypsin activity
in subcellular fractions prepared at varying but earlier
times after the start of supramaximal stimulation
would allow us to define the subcellular compartment
in which activation occurs and to track the fate of
trypsin after it becomes activated. Furthermore, we
reasoned that immunolocalization of the activation
peptide released during trypsinogen activation (i.e.,
TAP) would enhance our ability to define the location of
trypsin activation. In designing these studies, we recognized the importance of making determinations at the
earliest times possible after enzyme activation, because
intracellular transport or organelle disruption could
result in altered localization of both trypsin and TAP at
later times.
As shown in Fig. 1, the trypsin activity detected in
the pancreas after in vivo supramaximal stimulation
and in acinar cells after in vitro supramaximal stimulation is first seen in a heavy subcellular fraction that
sediments at 1,300 g. This fraction is normally enriched
in zymogen granules and, after either in vivo or in vitro
supramaximal stimulation, it is the fraction that is also
enriched in the lysosomal hydrolase cathepsin B (32)
(Fig. 3). This 1,300 g pellet is also the fraction that
contains most of the measurable TAP after 30 min of
supramaximal stimulation (Fig. 4). With prolonged
supramaximal stimulation, both trypsin activity and
measurable TAP decrease in this heavy subcellular
fraction and a corresponding increase in both trypsin
and TAP is noted in the soluble compartment (Figs. 1
and 4). With in vivo supramaximal stimulation, this
shift in trypsin activity from the particulate to the
soluble fraction occurs after 60 min, whereas it is first
seen after 30 min of supramaximal stimulation in vitro
(Fig. 1). This difference may reflect the influence of
studying the same phenomenon under different experimental conditions. The value of the soluble compart-
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.6 on June 16, 2017
gins in the periductal or perilobular areas (13, 25). We
(32), on the other hand, have argued that digestive
enzyme activation occurs in acinar cells as a result of
the colocalization of digestive enzyme zymogens with
lysosomal hydrolases. Evidence supporting this socalled ‘‘colocalization hypothesis’’ includes the following
observations: 1) the earliest morphological changes in
acute biliary pancreatitis are seen in acinar cells (18);
2) digestive enzyme zymogens and lysosomal hydrolases are colocalized in cytoplasmic vacuoles that appear, in several experimental models of pancreatitis,
before other evidence of cell injury can be detected (14,
26, 27, 29); 3) under appropriate in vitro conditions, the
lysosomal hydrolase cathepsin B can activate trypsinogen (3, 8, 19); 4) activated digestive zymogens, including trypsin, can be detected in the pancreas in clinical
pancreatitis (5) and, in several models of experimental
pancreatitis, activated trypsin can be detected before
other evidence of cell injury (7, 24); 5) hereditary
pancreatitis results from mutation of the trypsinogen
gene with expression of trypsinogen that, once activated, is resistant to inactivation (34); and 6) supramaximal secretagogue stimulation, which is known to induce acute pancreatitis in animals, can cause activation
of trypsinogen in isolated pancreatic acini studied in
vitro and this activation can be prevented by the
cell-permeant, specific cathepsin B inhibitor E64D (30).
The present study was undertaken to elucidate 1) the
location of trypsinogen activation in acinar cells during
the early stages of pancreatitis and 2) the fate of
trypsin after its activation. We have chosen to explore
these issues using the secretagogue model of pancreatitis, because it is a model that evolves in a rapid as well
as highly reproducible fashion and because we have
recently noted (7) that activation of trypsinogen along
with release of TAP can be detected in the pancreas as
early as 15–30 min after the start of supramaximal
stimulation. Bialek et al. (1) have already shown, using
enzyme-blotting techniques, that activated zymogens
G358
TRYPSINOGEN ACTIVATION IN PANCREATITIS
ment may also differ between the in vivo and in vitro
studies. In samples prepared from portions of whole
pancreas (i.e., in vivo experiments), this soluble compartment would be expected to include both cytosol and
extracellular interstitial edema fluid, but in samples
prepared from suspended acini, the soluble compartment should include only the cytosol.
Taken together, these results suggest that trypsinogen activation during supramaximal secretagogue
stimulation occurs in a particulate compartment that
includes zymogen granules, and, during pancreatitis,
also contains lysosomal hydrolases. Furthermore, these
results suggest that, once activated, trypsin moves
from this particulate location to the acinar cell cytosol.
In general, results of studies performed using the
technique of subcellular fractionation must be interpreted with caution because of the possibility that
enzyme redistribution during or after the time of tissue
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.6 on June 16, 2017
Fig. 5. Electron microscopy localization of TAP control studies. Immunogold labeling of pancreatic cryosections
with purified anti-TAP antibody was
carried out as described in text. Animals were infused with either saline (A
and B) or caerulein (10 µg · kg21 · h21 for
30 min; C and D). Micrographs are
representative of 3 or more experiments. Background gold decoration obtained when the pancreas of a control
animal is labeled for TAP is shown in A.
This background is comparable to that
obtained without primary antibody (B).
When regular labeling is performed
after prior incubation of TAP antibody
with an excess of antigen (C), even the
characteristic cytoplasmic vacuoles associated with pancreatitis (arrows) remain unlabeled. The acinar lumen (D;
p) remains free of TAP regardless of
whether or not caerulein has been infused; in contrast, a strongly TAPpositive cytoplasmic vacuole is shown
(D, inset) from the same animal. Calibration bar, 250 nm.
processing may alter the sedimentation characteristics
of many components (31). Furthermore, subcellular
fractionation experiments do not permit discrimination
between different organelles that have similar sedimentation characteristics. Thus, using only subcellular
fractionation, we would have been unable to establish
whether trypsin is activated within zymogen granules,
lysosomes, or cytoplasmic vacuoles, which cosediment
along with the zymogen granules after supramaximal
secretagogue stimulation.
We have used the techniques of light and electron
microscopy immunolocalization to resolve these areas
of uncertainty. At present, probes for pancreatic proteolytic activity at the ultrastructural level are not available and antibodies to digestive enzymes would not
discriminate between the zymogens and their activated
enzymes. For this reason, we have chosen to use
purified antibodies raised against TAP. Those antibod-
TRYPSINOGEN ACTIVATION IN PANCREATITIS
G359
ies are specific for TAP and have little or no cross
reactivity with either trypsin or trypsinogen. It could be
anticipated, therefore, that the site of highest TAP
content at early times would correspond to the initial
site of trypsinogen activation.1
Our initial studies were performed using the technique of electron microscopy immunogold localization
and anti-TAP antibodies. When we labeled the pancreas of saline-infused control animals for TAP, we
found a sparse but even distribution of gold grains over
1 It should be noted that morphometric studies quantitating immunogold anti-TAP labeling allow one to define the concentration of TAP
in specified compartments. The actual content of TAP in a compartment would depend on the volume of that compartment. However, it
is highly unlikely that a mechanism exists in acinar cells that results
in the taking up of TAP against a concentration gradient. Thus the
site of greatest TAP concentration is most likely to correspond with
the site of initial TAP release (i.e., trypsinogen activation).
the entire acinar cell. Although this could indicate that
a small amount of trypsinogen activation takes place
under physiological conditions, we reject this interpretation for two reasons. First, we would have expected to
see a greater concentration of the TAP-specific label in
compartments that are known to contain large amounts
of trypsinogen (e.g., endoplasmic reticulum and zymogen granules), but these compartments were not labeled at a greater density than, for example, the
nucleus. The second reason for rejecting this interpretation is the fact that the same background gold decoration was also found when no primary TAP antibody was
used. Thus we conclude that trypsinogen activation
either does not occur under physiological conditions or,
alternatively, that it occurs at a level impossible to
detect by immunogold labeling.
As early as 30 min after the start of caerulein
infusion, the distribution of TAP-specific gold label
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.6 on June 16, 2017
Fig. 6. Electron microscopy localization of TAP pancreatitis studies. Immunogold labeling of pancreatic cryosections with purified anti-TAP antibody
was performed as described in text.
Animals were infused with caerulein
(10 µg · kg21 · h21 ) for either 30 (A and
B) or 210 min (C and D). Micrographs
are representative of 3 or more experiments. Early in experimental pancreatitis, cytoplasmic vacuoles (A; arrows)
are the most prominent localization of
TAP-specific gold label, whereas the
surrounding cytoplasm and the endoplasmic reticulum (B) show only background gold decoration. After 210 min
of pancreatitis, not only the cytoplasmic vacuoles but also the cytoplasm
around them (C) and the dilated endoplasmic reticulum (D) contain TAP label. Mitochondria, nuclei, and zymogen
granules (ZG; inset in C) still show only
background gold decoration. Calibration bar, 250 nm.
G360
TRYPSINOGEN ACTIVATION IN PANCREATITIS
changes remarkably. At that time cytoplasmic vacuoles
have formed and these pancreatitis-associated structures represent the principal and exclusive compartment in which significant concentrations of gold label
are consistently found. Although occasional gold label
is observed in other areas of the cell (zymogen granules,
cytosol, nucleus), the density of that label is not significantly above the background level. This observation
strongly suggests that the initial conversion of trypsinogen to active trypsin occurs in the cytoplasmic vacuoles.
Furthermore, our findings all but exclude some other
potential or additional sites of premature protease
activation, such as zymogen granules, the acinar or
ductal lumena, or the interstitial space between acinar
cells. After 3.5 h of caerulein infusion, the density of
gold label associated with cytoplasmic vacuoles has still
further increased, but the vacuoles no longer represent
Fig. 8. Light microscopy colocalization of TAP and cathepsin B in rat pancreatic acini, 30 min after infusion of
caerulein (10 µg · kg21 · h21 ). Double-labeling fluorescence microscopy of paraffin-embedded 4-µm pancreas sections
was carried out with purified TAP antibody/FITC-conjugated anti-rabbit IgG (A) and with anti-cathepsin B/Texas
red-conjugated anti-sheep IgG (B). On merging the 2 images in A and B, the colocalized TAP and cathepsin B appear
as yellow dots in C. Arrows in A, B, and C indicate some of the areas in the acinar cells that are positive for TAP (A),
cathepsin B (B), or both (C). Original magnification, 3600.
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.6 on June 16, 2017
Fig. 7. Morphometric quantification of TAP. Animals were infused
with either saline (hatched bars) or caerulein (10 µg · kg21 · h21 ) for 30
min (solid bars) or 210 min (crosshatched bars). Immunogold-labeled
cryosections were prepared and photographed, and the micrographic
prints were morphometrically evaluated as described in text. TAP
label in the respective subcellular compartment was expressed as
gold grains/0.04 µm2. Values are means 6 SE from 3 or more
experiments and a minimum of 100 evaluations/group. * P # 0.05
compared with background. No morphometric data for cytoplasmic
vacuoles in saline-infused animals are given, because these structures are generally not seen in control animals.
the exclusive site of TAP localization. Large areas of the
acinar cell cytosol, including the endoplasmic reticulum, are now found to be gold labeled, but this redistribution of TAP still excludes zymogen granules and
nuclei. From these data we conclude that, very early
after supramaximal stimulation, a premature and intrapancreatic activation of proteases occurs and this event
is initially confined to the cytoplasmic vacuoles in
acinar cells. Later, TAP can be found over large areas of
the cytosol in addition to the vacuoles. This subsequent
redistribution of TAP could, theoretically, be explained
in a number of ways. The explanation we consider most
likely is that the cytoplasmic vacuoles, which are
known to be unstable cellular organelles in vitro (27,
29), disintegrate in vivo and their content, which
includes TAP and by inference active trypsin, is released into the cellular cytosol. An alternative explanation for our observed redistribution of TAP would be the
passage of this small peptide, as opposed to the much
larger trypsin, through the membrane of intact cytoplasmic vacuoles and, thus, into the cytosol. Unfortunately,
probes that discriminate between activated digestive
enzymes and their inactive zymogens are not currently
available and, for this reason, we cannot unequivocably
conclude that trypsin is released into the cytoplasm
along with TAP.
The final issue we have addressed is the mechanism
by which trypsinogen becomes activated during the
early stages of pancreatitis. In previous reports (27), we
have suggested that trypsinogen may be activated by
the lysosomal hydrolase cathepsin B during pancreatitis. Indeed, cosedimentation of TAP, active trypsin, and
cathepsin B in the same subcellular fraction during the
early stages of caerulein-induced pancreatitis and after
in vitro supramaximal stimulation with caerulein (Figs.
1, 3, and 4; Ref. 27) is compatible with this conclusion.
To further evaluate this issue, we have performed light
microscopy studies to immunolocalize cathepsin B and
TAP 30 min after the start of in vivo supramaximal
stimulation with caerulein. As shown in Fig. 8, cytoplasmic vacuoles located primarily on the luminal side of
the nucleus are apparent under these conditions and
these vacuoles contain both cathepsin B and TAP.
Based on their location, one might conclude that these
TRYPSINOGEN ACTIVATION IN PANCREATITIS
We thank Dan Brown, E. Wolff-Hieber for expert technical assistance with the immunolocalization experiments, and L. Edelmann
and E. Morgenstern for electron microscopy support.
This study was supported by National Institute of Diabetes and
Digestive and Kidney Diseases Grant DK-31396, Deutsche Forschungsgemeinschaft Grant Le 325/4-1, and by Boehringer Ingelheim.
Address for reprint requests: M. Steer, Dept. of Surgery, Beth
Israel Deaconess Medical Center, 330 Brookline Ave., Boston, MA
02215.
Received 13 March 1998; accepted in final form 7 April 1998.
REFERENCES
1. Bialek, R., S. Willemer, R. Arnold, and G. Adler. Evidence of
intracellular activation of serine proteases in acute caeruleininduced pancreatitis in rats. Scand. J. Gastroenterol. 26: 190–
196, 1991.
2. Fernandez del Castillo, C., J. Schmidt, A. L. Warshaw, and
D. W. Rattner. Interstitial protease activation is the central
event in progression to necrotizing pancreatitis. Surgery 116:
497–504, 1994.
3. Figarella, C., B. Miszczuk-Jamska, and A. J. Barrett. Possible lysosomal activation of pancreatic zymogens: activation of
both human trypsinogens by cathepsin B and spontaneous acid
activation of human trypsinogen 1. Biol. Chem. Hoppe Seyler
369, Suppl.: 293–298, 1988.
4. Foulis, A. K. Histological evidence of initiating factors in acute
necrotizing pancreatitis in man. J. Clin. Pathol. 33: 1123–1131,
1980.
5. Geokas, M. C., H. Rinderknecht, V. Swanson, and B. J.
Haverback. Role of elastase in acute pancreatitis. Lab. Invest.
19: 235–239, 1969.
6. Gorelick, F. S., G. Adler, and H. F. Kern. Caerulein-induced
pancreatitis. 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,
1993, p. 501–526.
7. Grady, T., A. Saluja, A. Kaiser, and M. Steer. Edema and
intrapancreatic trypsinogen activation precede glutathione depletion during caerulein pancreatitis. Am. J. Physiol. 271 (Gastrointest. Liver Physiol. 34): G20–G26, 1996.
8. Greenbaum, L. A., and A. Hirschkowitz. Endogenous cathepsin activates trypsinogen in extracts of dog pancreas. Proc. Soc.
Exp. Biol. Med. 107: 74–76, 1961.
9. Gudgeon, A. M., D. I. Heath, P. Hurley, A. Jehanli, G. Patel,
C. Wilson, A. Shenkin, B. M. Austen, C. W. Imrie, and J.
Hermon-Taylor. Trypsinogen activation peptides assay in the
early prediction of severity of acute pancreatitis. Lancet 335:
4–8, 1990.
10. Hurley, P. R., A. Cook, A. Jehanli, B. M. Austen, and J.
Hermon-Taylor. Development of radioimmunoassays for free
tetra-L-aspartyl-L-lysine trypsinogen activation peptides (TAP).
J. Immunol. Methods 111: 195–203, 1988.
11. Jungermann, J., M. M. Lerch, H. Weidenbach, M. P. Lutz, B.
Kruger, and G. Adler. Disassembly of the rat pancreatic acinar
cytoskeleton during supramaximal secretagogue stimulation.
Am. J. Physiol. 268 (Gastrointest. Liver Physiol. 31): G328–G338,
1995.
12. Kawabata, S., T. Miura, T. Morita, H. Kato, K. Fujikawa, K.
Takada, and S. Sakakibara. Highly sensitive peptide-4methycoumaryl-7-amide substrates for blood-clotting proteases
and trypsin. Eur. J. Biochem. 172: 17–25, 1988.
13. Kloppel, G., T. Dreyer, S. Willemer, H. F. Kern, and G. Adler.
Human acute pancreatitis: its pathogenesis in the light of
immunocytochemical and ultrastructural findings in acinar cells.
Virchows Arch. 409: 791–803, 1989.
14. Koike, H., M. L. Steer, and L. Meldolesi. Pancreatic effects of
ethionine: blockade of exocytosis and appearance of crinophagy
and autophagy precede cellular necrosis. Am. J. Physiol. 242
(Gastrointest. Liver Physiol. 5): G297–G307, 1982.
15. Labarca, C., and K. Paigen. A simple, rapid and sensitive DNA
assay procedure. Anal. Biochem. 102: 344–352, 1980.
16. Lampel, M., and H. F. Kern. Acute interstitial pancreatitis in
the rat induced by excessive doses of pancreatic secretagogue.
Virchows Arch. 373: 97–117, 1977.
17. Leach, S. D., I. M. Modlin, G. A. Scheele and F. S. Gorelick.
Intracellular activation of digestive enzymes in rat pancreatic
acini: stimulation by a high dose of cholecystokinin. J. Clin.
Invest. 87: 362–366, 1991.
18. Lerch, M. M., A. K. Saluja, R. Dawra, P. Ramarao, M.
Saluja, and M. L. Steer. Acute necrotizing pancreatitis in the
opossum: earliest morphological changes involve acinar cells.
Gastroenterology 103: 205–213, 1992.
19. Lerch, M. M., A. K. Saluja, R. Dawra, M. Saluja, and M. L.
Steer. The effect of chloroquine administration on two experimental models of acute pancreatitis. Gastroenterology 104: 1768–
1779, 1993.
20. Lerch, M. M., A. K. Saluja, M. Rünzi, R. Dawra, and M. L.
Steer. Luminal endocytosis and intracellular targeting by acinar
cells during early biliary pancreatitis in the opossum. J. Clin.
Invest. 95: 2222–2231, 1995.
21. McDonald, J. K., and S. Ellis. On the substrate specificity of
cathepsin B1 and cathepsin B2 including a new fluorogenic
substrate for cathepsin B1. Life Sci. 17: 1269–1276, 1975.
22. Pierre, K. J., K. K. Tung and H. Nadj. A new enzymatic kinetic
method for determination of amylase. Clin. Chem. 22: 1219,
1976.
23. Powers, R. E., A. K. Saluja, M. J. Houlihan, and M. L. Steer.
Diminished agonist-stimulated inositol triphosphate generation
blocks stimulus-secretion coupling in mouse pancreatic acini
during diet-induced experimental pancreatitis. J. Clin. Invest.
77: 1668–1674, 1986.
24. Rao, K. N., J. Tuma and B. Lombardi. Acute hemorrhagic
pancreatic necrosis in mice. Intraparenchymal activation of
zymogens, and other enzyme changes in pancreas and serum.
Gastroenterology 70: 720–726, 1976.
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.6 on June 16, 2017
vacuoles are derived from the Golgi apparatus, but
further studies will be needed to more clearly define
their genesis. The colocalization of cathepsin B and
TAP within these vacuoles is consistent with the conclusion that cathepsin B catalyzes trypsinogen activation
in these organelles.
In summary, our findings lead us to conclude that
intrapancreatic activation of trypsinogen occurs at a
very early stage in the evolution of secretagogueinduced pancreatitis and that that activation occurs in
cytosolic vacuoles containing both digestive enzymes
and the lysosomal hydrolase cathepsin B. On the basis
of these and previously reported observations (30), we
conclude that intra-acinar cell activation of trypsinogen in the cytoplasmic vacuoles is catalyzed by cathepsin B. At later times, trypsin appears to be released
from the cytoplasmic vacuoles into the cytoplasm. We
suggest that the presence of activated trypsin in the
cytoplasmic space is a critical event that eventually
leads to cell injury, but further studies evaluating the
relationship between cytoplasmic trypsin and cell injury will be needed to unequivocally establish this
relationship. In the present study, we found that the
increase in soluble (i.e., cytoplasmic) trypsin activity
and TAP levels was not apparent before 60–120 min of
caerulein infusion, yet, in earlier studies (6), we found
that hyperamylasemia and pancreatic edema were
present within 30–60 min of supramaximal stimulation. Although these time-dependent differences may
merely reflect the inherent technical difficulty encountered in monitoring small but potentially critical
changes in trypsin or TAP distribution, these differences may also indicate that other, as yet unidentified,
phenomena contribute to cell injury in this model.
G361
G362
TRYPSINOGEN ACTIVATION IN PANCREATITIS
31.
32.
33.
34.
35.
tion of trypsinogen in rat pancreatic acini is mediated by
cathepsin B. Gastroenterology 113: 304–310, 1997.
Scheele, G., G. E. Palade, and A. M. Tartakoff. Cell fractionation studies on guinea pig pancreas. Redistribution of exocrine
proteins during tissue homogenization. J. Cell Biol. 78: 110–130,
1978.
Steer, M. L., and A. K. Saluja. Experimental acute pancreatitis: studies of the early events that lead to cell injury. 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, 1993, p. 489–499.
Tartakoff, A., and J. E. Jamieson. Fractionation of a guinea
pig pancreas. Methods Enzymol. 31: 41–59, 1974.
Whitcomb, D. C., M. C. Gorry, R. A. Preston, W. Furey, M. J.
Sossenheimer, C. D. Ulrich, S. P. Martin, L. K. Gates, Jr.,
S. T. Amann, P. P. Toskes, R. Liddle, K. McGrath, G. Uomo,
J. C. Post and G. D. Ehrlich. Hereditary pancreatitis is caused
by a mutation in the cationic trypsinogen gene. Nat. Genet. 14:
141–145, 1996.
Yamaguchi, H. T., K. Kimura, and H. Nawata. Activation of
proteases in caerulein induced pancreatitis. Pancreas 4: 565–
571, 1989.
Downloaded from http://ajpgi.physiology.org/ by 10.220.33.6 on June 16, 2017
25. Rinderknecht, H. Activation of pancreatic zymogens. Normal
activation, premature intrapancreatic activation, protective
mechanisms against inappropriate activation. Dig. Dis. Sci. 31:
314–321, 1986.
26. Saito, I., S. Hashimoto, A. Saluja, M. L. Steer, and J.
Meldolesi. Intracellular transport of pancreatic zymogens during caerulein supramaximal stimulation. Am. J. Physiol. 253
(Gastrointest. Liver Physiol. 16): G517–G526, 1987.
27. Saluja, A., H. Sadamitsu, M. Saluja, R. E. Powers, J.
Meldolesi, and M. L. Steer. Subcellular redistribution of
lysosomal enzymes during caerulein-induced pancreatitis. Am. J.
Physiol. 253 (Gastrointest. Liver Physiol. 16): G508–G516, 1987.
28. Saluja, A., I. Saito, M. Saluja, M. J. Houlihan, R. Powers, J.
Meldolesi, and M. L. Steer. In vivo rat pancreatic acinar cell
function during supramaximal stimulation with caerulein. Am. J.
Physiol. 249 (Gastrointest. Liver Physiol. 12): G702–G710, 1985.
29. Saluja, A., M. Saluja, A. Villa, U. Leli, P. Rutledge, J.
Meldolesi, and M. L. Steer. Pancreatic duct obstruction in
rabbits causes digestive zymogen and lysosomal enzyme colocalization. J. Clin. Invest. 84: 1260–1266, 1989.
30. Saluja, A. K., E. A. Donovan, K. Yamanaka, Y. Yamaguchi, B.
Hofbauer, and M. L. Steer. Caerulein-induced in vitro activa-