Clinical Science and Molecular Medicine (1973) 45,213-224.
THE INACTIVATION BY CARBENOXOLONE OF
INDIVIDUAL H U M A N PEPSINOGENS A N D PEPSINS
N . B. ROBERTS
AND
W . H. TAYLOR
Department of Chemical Pathology, The United Liverpool Hospitals
(Received 28 November 1972)
SUMMARY
1. Carbenoxolone, in suspension at pH 4.0, inhibits swine pepsin A, and human
pepsins 1,3 and 5. Human pepsin 5 is the most readily inhibited, and human pepsin 1
the least.
2. Inhibition occurs by a process which is time-dependent, temperature-dependent
and proportional to the quantity of carbenoxolone suspended.
3. Carbenoxolone, in solution at pH 7.4 and pH 8.0,inhibits the activation of the
total pepsinogens of human gastric mucosal extracts and of the individual pepsinogens 1,3 and 5. Pepsinogen 1 was the most readily inhibited, pepsinogen 5 the least.
4. Chymotrypsin was readily inhibited by carbenoxolone at pH 7-4and 8.0.Trypsin
was not inhibited at pH 7.4 but was inhibited, relatively weakly, at pH 8.0. Pronase
was weakly inhibited at pH 7-4 and 8.0 but papain was weakly activated.
5. Carbenoxolone is therefore not a general enzyme inhibitor but shows specificity
for enzymes (pepsins and chymotrypsin) which split proteins at the same bonds,
rather than for enzymes with similar active centres (chymotrypsin and trypsin).
6. The results suggest that, in vivo, carbenoxolone might diminish peptic activity in
three ways: by inactivating pepsinogens irreversibly in the mucosal cells or at some
point before their activation to pepsins; by inhibiting pepsins irreversibly in the gastric
lumen; and by binding pepsins in the lumen without destroying their activity but decreasing their effective concentration.
7. These results are compatible with the hypothesis that pepsins, and pepsin 1 particularly, are factors in the aetiology of peptic ulcer.
Key words : pepsinogens, pepsins, carbenoxolone, peptic ulcer.
Taylor (1959) showed that the proteolytic pH-activity curves of the histamine-stimulated
gastric juices from patients with gastric and duodenal ulcers differed from those of the juices
of normal subjects. Fernandez Costa (1970) found an identical difference in the gastric proCorrespondence: N. B. Roberts, Department of Chemical Pathology, Ashton Street, Liverpool L3 5RT.
F
213
214
N. B. Roberts and W. H . Taylor
teolytic activity of normal dogs and of the same dogs after induction of gastric ulceration by
cincophen; when the dogs recovered, the normal proteolytic pattern was restored. During
further investigation of this difference Taylor (1970a) observed an increased frequency and
raised concentration of pepsin 1 in the gastric juice of patients with gastric and duodenal
ulcers as compared with normal subjects (Etherington & Taylor, 1969). The effect was more
marked with gastric ulcer than with duodenal ulcer.
Carbenoxolone (sodium salt) is a triterpene, the di-sodium salt of 3-O-(B-carboxypropionyl)1l-oxo-18/3-olean-12-en-30-oic
acid. It has been shown by Doll, Hill Hutton & Underwood
(1962), and subsequently by several other groups, to increase the rate of healing of gastric
ulcers, but the effect upon duodenal ulceration has been either absent or less marked. It was
therefore of interest to study the effect of carbenoxolone upon individual purified human
pepsins and pepsinogens and particularly upon pepsin 1. Henman (1970) has already observed
that carbenoxolone decreases the total peptic activity of gastric juice obtained from the anaesthetized pylorusligated rat. During the course of the present investigation, Berstad (1972) has
reported the inhibition of total peptic activity of normal human gastric juice by carbenoxolone
both in vitro and in vivo; in one patient with a duodenal ulcer, carbenoxolone did not decrease
the peptic activity of the gastric content. A preliminary account of the present work has been
given (Roberts & Taylor, 1973).
MATERIALS A N D M E T H O D S
Gastric juice was collected by pernasal intragastric tube from patients undergoing augmentedhistamine tests (Kay, 1953) or pentagastrin-stimulation tests, using 6 pg/kg body weight
intramuscularly. Pepsins were prepared from human gastric juice and pepsinogens from human
gastric mucosal extracts as described previously (Etherington & Taylor, 1969, 1970).
Crystalline swine pepsin, chymotrypsin (bovine pancreas) and bovine haemoglobin
substrate powder were obtained from Armour Laboratories, Eastbourne, England; trypsin
from Worthington Biochemical Corp., New Jersey, U.S.A. ; Pronase (from Streptomyces
griseus) from Koch-Light Laboratories, Colnbrook, England; papain (twice-crystallized)from
the Sigma Chemical Co. St Louis, U.S.A. Carbenoxolone (sodium salt) was kindly donated by
Dr S . Gottfreid of Biorex Laboratories, London, N.l.
Pepsins and pepsinogens were examined for purity on agar-gel electrophoresis at pH 5.0
(Etherington & Taylor, 1969). To prevent activation of the zymogens to pepsins, application
of the pepsinogen preparations was in phosphate buffer (0.025 mol/l; pH 7.4). Activated
pepsinogen fractions and pepsins were applied in acetate buffer (0.025 mol/l; pH 5.0). Measurements of pH were carried out with thevibron pH meter (model 39A, Electronic Instruments
Ltd, Richmond, Surrey).
Inhibition of enzymic activity
Solutions or suspensions of carbenoxolone (sodium salt; containing 0-6.5 mmol/l) in
sodium phosphate buffers (0.1 mol/l; pH 7.4 or pH %O), phosphate buffer (0.05 mol/l; pH 6.0),
or sodium acetate buffer (0.05 mol/l; pH 4.0),were added in a volume of 0.05 ml to 0.05 ml of
enzyme solutions in the same respective buffers and at the same pH. Incubation for 30 min at
37°C (unless otherwise stated) was carried out and the residual enzymic activity then determined
Inhibition of pepsins by carbenoxolone
215
by adding 1.9 ml of a haemoglobin solution (3.3 g/l) dissolved in the appropriate buffer and at
the appropriate pH for the ensuing proteolytic reaction. Control observations were carried out
either on enzyme solutions incubated alone, or by rapidly adding the inhibitor and then the
substrate to the enzyme solution (zero-time incubation) or by both methods, as appropriate.
Determination of proteolytic activity
The proteolytic activity of pepsins, pepsinogens and their various fractions was determined
at pH 2.0 (Etherington & Taylor, 1969). Zymogens were not activated until they were added
to the digestion mixture, which at pH 2.0 gives a rapid conversion of the pepsinogens into
pepsins (Seijffers, Segal & Miller, 1963). The proteolytic activity of chymotrypsin, trypsin,
01
60
I
I
6.4
I
I
I
6.8
7 *2
PH
1
I
m l / c
0325
(mg/mll
(02)
I
l
l
I
I
l
l
0-81 I.30228 325 390 650
(0.5) W ( l . 4 ) (20) (24) (40)
Concn. of c a t m o x d a w
FIG.1. pH;inactivation curves of human pepsins :o and A , human pepsins 1 and 3 respectively in
sodium phosphate buffer (0.05 mol/l) without carbenoxolone: and A , human pepsins 1 and 3
respectively in aqueous solution of carbenoxolone (sodium salt). Enzyme solutions were preincubated for 20 min at 20°C in a volume of 0.1 ml at the indicated pH.
Pronase and papain was measured at pH 7.4 and at pH 8.0 in sodium phosphate buffer (0.1
mol/l). For all these determinations the substrate was bovine haemoglobin (3.3 g/l) and the
method that of Anson & Mirsky (1933), as modified by Hanley Boyer & Naughton (1966).
216
N. B. Roberts and W. H . Taylor
Dialysis experiments
Portions (0.3 ml) of solutions of pepsinogens and 0.3 ml of the carbenoxolone solutions at the
different concentrations were incubated for 30 min at 37°C and then dialysed at 4°C against a
total of 1.5 litres of sodium phosphate buffer (0.1 mol/l) at pH 7.4 or at pH 8.0 for 17 h.
Proteolytic activity was measured on 0.1 ml aliquots after dialysis. These solutions now contained so little carbenoxolone that there was no opalescence on bringing the pH to 2.0, at
which value carbenoxolone is insoluble.
RESULTS
Pepsin studies
Carbenoxolone (sodium salt) is readily soluble in water to give a mildly alkaline solution,
but in buffered solutions below pH 7-0 it is only poorly soluble. Preliminary experiments
indicated that the very small amounts of carbenoxolone that are dissolved at pH 4.0 (the upper
mmoC/l
(rnghl)
065
(0.4)
(.
30
1.95
2.60
3.25
(08)
(1.2)
(1.6)
( 20)
Concn. of corbenoxolone
FIG.2. Effect of increasing concentrations of carbenoxolone on the proteolytic activity of human
and swine pepsins: 0 , human pepsin 1 ; 0 , human pepsin 3; A , human pepsin 5, preincubated for
30 min at 37°C in sodium acetate (0.05 mol/l; pH 4.0); 0 , swine pepsin in sodium acetate (0.05
mol/l; pH 4.0); m, swine pepsin in sodium phosphate buffer (0.05 mol/l; pH 6.0).
end of the pH-activity range of the pepsins) did not inhibit human pepsins 1 and 3. Incubation
of each of these pepsins with increasingconcentrations of carbenoxolone (sodium salt) in water,
up to 6.5 mmol/l, gives increasing inhibition of proteolytic activity (Fig. l), but this inhibition
results from the increasing pH, for the curve follows closely that of alkaline inactivation in the
Inhibition of pepsins by carbenoxolone
217
absence of carbenoxolone. It is possible that the in vitro inhibition by carbenoxolone of swine
pepsin, reported by Henman (1970), can be partly explained in this way, since the pH of the
final reaction mixtures is not stated.
Fig. 2 records the inhibition at pH 4.0 of human pepsins 1, 3 and 5, and at pH 4.0 and pH
6.0 of swine pepsin A after incubation with increasing concentrations of carbenoxolone
(sodium salt), which is in non-flocculant suspension at pH 6.0 but is flocculant at pH 4-0.
Under these experimental conditions, pepsin 1 retains activity better than the other pepsins.
Inhibition of swine pepsin A, which is analogous to human pepsin 3, was more marked at pH
4.0 than at pH 6.0.
I
t
2o
0
I
\
A
1
I
I
I
1
10
20
30
40
50
1
60
T i m (mh)
FIG.3. Effect of time on inhibition of human pepsin 3 by different concentrationsof carbenoxolone
(sodium salt): 0 , none; q 0 . 4 1 mmol/l(O*25mg/ml); A , 3.25 mmol/l(2.0 mg/ml); A , 6.50 mmol/l
(40 mg/ml) at pH 4 0 in sodium acetate (0.05 mol/l).
In a subsequent series of experiments, human pepsins 1 and 3 were preincubated with
suspensions of carbenoxolone at pH 4.0 for 20 min at 37°C. After centrifuging at lo00 g for
2 min, the pellet was washed with 0.1 ml of sodium acetate buffer (0.05 mol/l; pH 4-0), recentrifuged, and the washings added to the first supernatant. The proteolytic activities of the pellet
and the combined supernatant were then examined separately. For human pepsin 3, the lost
activity of the supernatant could be recovered partially from the pellet, but little of the human
pepsin 1 was recovered in this way. For example, at an inhibitor concentration of 1.61 mmol/l
(1-0 mg/ml) the supernatant containing pepsin 3 showed only 0.5% of the activity of the uninhibited enzyme, but 38.7% of the lost activity was in the pellet; for pepsin 1, 53.1% of the
N . B. Roberts and W. H . Taylor
218
original activity remained in the supernatant and only 4.8% was in the pellet. Once again the
degree of inhibition of pepsin 3 in pellet plus supernatant was greater than that of pepsin 1.
Although the pellet of carbenoxolone was of greater mass in the more-concentrated solutions,
the quantity of recovered pepsin did not increase with mass. Indeed, with pepsin 3 the value fell.
The observations with pepsins would be most readily explained if carbenoxolone inhibits
irreversibly some of the pepsin molecules, but binds with others without inhibiting them.
The extent of the irreversible inhibition is greater with pepsin 3 than with pepsin 1.
6ot
80
8
P
E
2
40
logo t m m d / l I
(log10Cmg/mlf)
T.21
(T.00)
log (
0.21
(0)
1.21
(1.00)
concn of cab en ox^)
FIG.4. Effect of carbenoxolone on the proteolytic activity of a gastric mucosal extract from a
patient with a duodenal ulcer. Solutions of carbenoxolone were preincubated with the extract for:
0 , 40 min at 37°C in sodium phosphate buffer (0.05 mol/l; pH 7.4); 0 , 30 min at 37°C in sodium
phosphate (0.05 mol/l; pH 7.4), then followed by 10 min at 37°C at pH 2-1.
Fig. 3 shows the time-course of inhibition of human pepsin 3 by different concentrations
(in suspension) of carbenoxolone. The loss of activity is not produced by an instantaneous
precipitation, adsorption or denaturation of the enzyme, but by a time-dependent process. The
effect of increasing concentrations of carbenoxoloneis again confirmed.
The inhibitory process is temperature-dependent.Thus on incubation of carbenoxolonewith
human pepsin 3 at T C , no inhibition was observed at inhibitor concentrations of up to 6.5
mmol/l (4 mglml); at 23"C, 28% inhibition was obtained at a concentration of 6.5 mmol/l,
whereas at 37"C,71% inhibition was achieved at an inhibitor concentration of only 1.61 mmol/l.
Pepsinogen studies
Working with suspensions of an inhibitor is unsatisfactory. It is possible, however, to study
Inhibition of pepsins by carbenoxolone
219
the effect of carbenoxolone upon gastric mucosal pepsinogens at pH 7.4 and pH 8.0, at which
values it is much more stable. Fig. 4 shows that incubation at pH 7.4 of an unactivated gastric
mucosal extract for 40 min with increasing concentrations of carbenoxolone gives increasing
inhibition of the subsequent activation of total pepsinogens. In this experiment the unaffected
pepsinogen molecules are activated by the addition of the haemoglobin substrate at pH 2.0.
If the pepsinogens are incubated for 30 min at pH 7.4 and then activated by incubation at
pH 2.0 for a further 10 min (the carbenoxolone being now in suspension), the resulting inhibition of total peptic activity is much greater. This experiment indicates that carbenoxolone
in solution can inhibit or prevent the conversion of pepsinogens to pepsins, but that the
I
0 65
I
I
I
2.60
1.95
1.30
Concn. of cabenoxobm (rnmol/L 1
I
3.25
FIG.6. Effect of carbenoxolone on the proteolytic activity of human pepsinogens, after preincubation at pH 7-4 in sodium phosphate buffer ( 0 1 mol/l) for 30 min at 37°C.0 , Fraction I
(pepsinogen 5); 0 , fraction I11 (pepsinogens 3 and 2); 0 , fraction V (pepsinogen 1).
inhibition of pepsins is additional to this. Fig. 5 shows the behaviour of the individual pepsinogen and pepsin fractions during a similar experiment, visualized on agar gel. The precursors
of pepsins 1 and 3 are decreased by incubation with increasing amounts of carbenoxolone.
Pepsins 1,3 and 5 are progessively decreased in amount, when incubated after activation, but
much of the activity is accumulating at the point of application. This is caused by uninhibited
pepsins attached to the precipitated carbenoxolone; the pellet of the earlier experiments.
Three pepsinogen fractions purified by DEAE-cellulose chromatography were next investigated. Fraction I contained the precursors of pepsin 5 and of the gastric cathepsins (‘pepsin’ 7
of Etherington & Taylor, 1967), fraction 111 the precursor of pepsin 3, and fraction V that of
pepsin 1. The fractions are numbered and the pepsinogens identified as described by Etherington
& Taylor (1970). Fig. 6 shows the effect of increasing amounts of carbenoxolone on approxi-
220
N . B. Roberts and W. H . Taylor
mately equal quantities, in terms of activity, of each precursor at pH 7-4 for 30 min at 37°C.
The precursors of pepsins 1 and 3 were much more readily inactivated than that of pepsin 5.
At pH 8.0 a similar result was obtained.
In a further series of experiments the incubated mixtures of pepsinogens and carbenoxolone
were dialysed against phosphate buffer (0.1 mol/l; pH 7-4) in order to remove excess carbenoxolone. There was no significant change in the degree of inactivation. This suggeststhe reaction
of the pepsinogens with carbenoxolone is not readily reversible.
Concn. of carbenoxolone ( t n t n O l / l )
FIG.7. Effect of carbenoxolone on the proteolytic activity of chymotrypsin and trypsin. 0 , Chymotrypsin, pH 7.4; A , trypsin, pH 7.4; 0 , chymotrypsin, pH 8.0; A , trypsin, pH 8.0.
Experiments with other proteolytic enzymes
A series of enzymes acting over the pH range 7.0-8.0 was chosen, so that carbenoxolone
would remain in solution. Fig. 7 shows that chymotrypsin is at least as sensitive to carbenoxolone in solution as is pepsin 3 to carbenoxolone in suspension. Trypsin, by contrast, was only
slightly inhibited at pH 7.4. At pH 8.0, slight activation occurred at low concentrations of
carbenoxolone followed by inhibition at higher concentrations. Pronase, from Streptomyces
griseus, was slightly inhibited at both pH 7.4 and 8-0;thus at a carbenoxolone concentration of
6.5 mmol/l, 20.5% (pH 7-4) and 20.0% (pH 8.0) inhibition was found. Papain, on the other
hand, showed increasing activation with increasing concentration of carbenoxolone, so that at
a concentration of 6-5 mmol/l the activity of the enzyme was increased by 23.7% at pH 7.4
and by 32.5% at pH 8.0.
Inhibition of pepsins by carbenoxolone
221
Enzymelinhibitor ratios
Table 1 shows the approximate molecular proportions which are needed to give 50% inhibition of each enzyme. Comparisons between the enzymes are difficult because the relatively
low specific activity of chymotrypsin, trypsin, Pronase and papain for their substrates meant
that much larger amounts of them had to be used than with the pepsins or pepsinogens. The
ratios show clearly, however, that for none of the enzymes is there likely to be a specific
inhibition at the active site. They re-emphasize the relative insensitivity of human pepsin 1 to
carbenoxolone as compared with its zymogen and the increased sensitivity of trypsin at pH
8.0 as compared with pH 7.4.
TABLE
1. Molecular ratios of carbenoxolone to pepsin, pepsinogen, trypsin and chymotrypsin
at 50% inactivation. Molecular weights used are: carbenoxolone (sodium salt), 615; human
pepsinogens, 42 000 (Taylor, 1968); human pepsin 5, 34 600; human pepsin 3, 37 000;
human pepsin 1, 43 800 (Roberts & Taylor, 1972); chymotrypsin, 24000; trypsin, 23 300;
swine pepsin, 35000 (Weber & Osborn, 1969)
Enzyme
Enzyme
concentration
(mmol/l)
Mucosal extract
i.i9x 10-4
Human pepsinogen 5
1.19 x 10-4
Human pepsinogen 3
8.33 x 10-5
pH
7.4
2-1
7.4
8.0
7.4
8.0
Human pepsinogen 1
7 . 1 4 ~10-5
7.4
Human pepsin 5
Human pepsin 3
Human pepsin 1
Swine pepsin
Chymotrypsin
2.89 x lo-'
1.29~
10-4
1.14 x 10-4
1.29 x 10-4
2-08x 10-3
4.0
4.0
4.0
4.0
7.4
8.0
8.0
Trypsin
2.146 x
8.0
Carbenoxolone
concentration
(mmol/l)
5.14
0.16
>3.25
>3*25
0.618
047
0504
0.406
0.439
0.764
2.85
0926
1.04
0488
2.93
Inhibitorlenzyme
ratio
(mol :mol)
4 . 3 2 ~10" : 1
1 . 3 4 lo3
~ :1
DISCUSSION
Carbenoxolone is known to be absorbed partly from the stomach (Parke, 1968). It might thus
come into contact with the intracellular pepsinogens during its absorption, or by subsequent
entry into the mucosal cells from the blood stream. Assuming an intracellular pH of about 7.0,
our experiments would suggest that the intracellular pepsinogens might be partly inactivated by
carbenoxolone and that pepsinogen 1 (fraction V) would be particularly sensitive to this effect.
Our experiments suggest also that oral carbenoxolone inhibits pepsins directly when in
222
N . B. Roberts and W. H . Taylor
suspension in the lumen of the stomach. Pepsin 3, the principal pepsin, would be particularly
sensitive to this type of inhibition. Our data thus provide two possible complementary
mechanisms whereby the pepsins or their precursor might be inhibited by carbenoxolone in v i m .
It must next be considered whether either of these mechanisms is likely to be of quantitative
significance.No data are available to enable a quantitative assessment of intracellular inactivation to be made. As far as intraluminal inhibition is concerned, we have illustrative data on a
single male patient with a duodenal ulcer. His gastric juice (histamine-stimulated) contained
a total pepsin activity of 40 mg/lOO ml (as swine pepsin equivalent) at pH 4.0, and 50 mg/lOO ml
at pH 2.0. When incubated for 30 min at pH 4.0 with carbenoxolone, it was found that 12.5
mg of carbenoxolone (sodium salt) were required to inhibit, by 50%, 1 mg of ‘total pepsin’,
which was contained in 2.5 ml of histamine-stimulated gastricjuice. At pH 2.0, 37% inhibition
was achieved with 8 mg of carbenoxolone (sodium salt) and 1 mg of ‘total pepsin’, which was
contained in 2 ml of the juice. Since the daily dose of carbenoxolone (sodium salt) may be as
much as 300 mg, it is apparent that significant, though partial, inhibition of the pepsins of
60-75 ml of gastric juice (histamine-stimulated) might thus be achieved. Berstad (1972) describes a significant decrease (about 3 0 4 0 % ) of peptic activity of normal human gastric juice
in vivo after instilling 100 mg of carbenoxolone (sodium salt) into the stomach. This degree of
inhibition is probably of the same order as in our experiment.
Variations in the protein content of gastric juices will affect the inhibition of pepsins, as
carbenoxolone binding to constituents such as albumin, for example, is known to occur (Parke,
Pollock & Williams, 1963; Whitehouse, Dean & Halsall, 1967). The non-pepsin proteins will
therefore decrease the effective intragastric inhibition of the pepsins, and might make intracellular inactivation of pepsinogens virtually impossible, since the intracellular protein content
is so high. Much will depend upon the localization of carbenoxolone intracellularly. It should be
noted, however, from a comparison of Figs. 4 and 6, that carbenoxolone inhibits the activation of the pepsinogens of a gastric mucosal extract, containing many proteins, over a similar
concentration range to that for the individual pepsinogens. This observation raises the possibility that the pepsinogens may have a high affinity for carbenoxolone.
The mechanism by which carbenoxolone inactivates the pepsins and pepsinogens must take
into account the following facts.
1. The molar ratios of enzyme to inhibitor (Table 1) indicate the absence of specific inhibition of the active site, such as is achieved with pepstatin with a pepsin/pepstatin molar ratio
of 1 : 1 (Aoyagi, Kunimoto, Morishima, Takeuchi & Umezawa, 1971).
2. In several instances, notably with pepsin 1 and with pepsinogen 5, increasing the concentration of carbenoxolone does not increase further the degree of inactivation. For these enzymes there is a maximal, though partial, degree of inactivation.
3. Carbenoxolone may bind to some of the pepsin and pepsinogen molecules without inactivating them (Fig. 5).
4. Carbenoxolone has the general property of binding to proteins.
5. Carbenoxolone activates papain.
It may be that each pepsin and pepsinogen reacts with or is denatured by carbenoxolone in
its own individual way. Alternatively, a general hypothesis to account for the experimental
facts may be advanced. Carbenoxolone may bind to proteolytic enzymes, as it binds to proteins generally. The binding sites would be to some extent specific and become fully saturated in
the case of those enzymes for which a partial inhibition is the most that can be achieved. The
Inhibition of pepsins by carbenoxolone
223
configurational changes brought about by the binding would decrease the activity of each
enzyme, perhaps by hindering access of the substrate to the active site, by an amount which is
typical for each individual enzyme. In the case of papain, the hypothesis would be extended
to suggest that the active centre is becoming more accessible as a result of the configurational
changes. In addition to inhibition there is, with pepsin 3 for example, some form of aggregation
of the enzyme with the suspended particles of carbenoxolone, without inactivation of the
enzyme.
The similar response of the pepsinogens, pepsins and chymotrypsin to carbenoxolone is of
interest because these enzymes split the B chain of oxidized insulin at similar sites (Taylor,
1962, 1968; Roberts & Taylor, 1972) involving the hydrophobic residues, phenylalanine,
tyrosine, leucine and valine. Trypsin splits the B chain quite differently, acting only on bonds
involving lysine and arginine, and Pronase is described as tryptic-like (Trop & Birk, 1970).
Yet chymotrypsin and trypsin have similar peptide sequences at the active centre, being
serine peptidases, but differ from swine pepsin (Knowles & Wybrandt, 1968; Meitner, 1971),
which is not a serine peptidase and which, unlike the serine peptidases, is not inhibited by diisopropylfluorophosphate (Etherington, 1967). The similar specificity of the various pepsins
and chymotrypsin is thus independent of apparent differences of peptide sequence at the active
centre, but is associated with the changes, possibly configurational, induced by binding with
carbenoxolone.
The healing effect of carbenoxolone in patients with gastric ulcer is well established, but the
mechanism of its action is unclear. It does not appear to influence the secretion of hydrogen
ions by the gastric mucosa (Berstad, Peterson & Myren, 1970) and more recent work has
focussed attention upon its possible action on mucus (Dean, 1968)and on the turnover rate of
gastric mucosal cells (Lipkin, 1970). Taylor (1959, 1970a) has, however, demonstrated an
abnormal pattern of peptic activity in the gastricjuice of patients with both gastric and duodenal
ulcer, and has suggested (Taylor, 1970b) that pepsin 1 should be regarded as one of the aetiological factors involved in gastric ulceration. The ability of carbenoxolone readily to inhibit
the activation of pepsinogen 1 lends some support to the hypothesis that pepsin 1 is a factor
in gastric ulceration, but the ready inhibition of pepsin 3 and the relative resistance of pepsin
1 itself to the inhibitor also emphasizes that the total amount of peptic activity may be
important.
ACKNOWLEDGMENT
We are grateful to Mrs Lesley Waft for skilled technical assistance.
REFER E N C ES
A.E. (1933) Estimation of pepsin with haemoglobin.Journafof Genera[Physiofogy, 16,
ANSON,M.L. & MIRSKY,
59-63.
AOYAGI,
T., KUNIMOTO,
S., MORISHIMA,
H., TAKEUCHI,
T. & UMEZAWA,
H. (1971) Effect of pepstatin on acid
proteases. Journal of Antibiotics, 24,687-694.
BERSTAD,
A. (1972) Inhibition of peptic activity in man by carbenoxolone sodium. Scandinavian Journal of
Gastroenterology, 7,129-135.
BERSTAD,
A., PETERSON,
H. & MYREN,
J. (1970) The effect of intraduodenalcarbenoxolonesodium on gastric and
duodenal secretion in man. Carbenoxolone Sodium, pp. 69-83. Ed. J. H. Baron & F. M. Sullivan. Butterworths,London.
224
N . B. Roberts and W. H . Taylor
DEAN,A.C.B. (1968) Protective effect of carbenoxolone sodium in drug-induced lesions of the stomach. A
Symposium on Carbenoxolone Sodium, pp. 33-39. Ed. J. M. Robson & F. M. Sullivan. Butterworths,
London.
DOLL,R., HILL,J.D., HUTTON,
C. &UNDERWOOD,
D.J. (1962) Clinical trial of a triperpenoid liquorice compound
in gastric and duodenal ulcer. Lancet, ii, 793-796.
ETHERINGTON,
D.J. (1967) Observations upon the proteinases of normal and neoplastic mucosa from the gastrointestinal tract. Ph.D. thesis, University of Liverpool.
ETHERINGTON,
D.J. &TAYLOR,
W.H. (1967) Nomenclature of the pepsins. Nature, 216,279-280.
ETHERINGTON,
D.J. & TAYLOR,
W.H. (1969) The pepsins of normal human gastric juice. Biochemical Journal,
113, 663-668.
ETHERINGTON,
D.J. & TAYLOR,
W.H. (1970) The pepsins from human gastric mucosal extracts. Biochemical
Journal, 118,587-594.
FERNANDEZ
COSTA,J. (1970) Cambios en la actividad proteolitica del jug0 gastric0 durante la formacion de
ulcera experimental. Doctoral thesis, University of Havana. (Copies of the author’s translation, into
English, of the summary of the thesis are available from W. H. Taylor.)
HANLEY,
W.B., BOYER,S.H. & NAUGHTON,
M.A. (1966) Electrophoretic and functional heterogeneity of pepsinogen in several species. Nature, 209,9961002.
HENMAN,
F.D. (1970) Inhibition of peptic activity by carbenoxolone and glycyrrhetinic acid. Gut, 11,344-351.
KAY,
A.W. (1953) Effect of large doses of histamine on gastric secretion of HCI. An augmented histamine test.
British Medical Journal, ii, 77-80.
G.B. (1968) The sequence around an active-site aspartyl residue in pepsin.
KNOWLES,
J.R. & WYBRANDT,
Federation of European Biochemical Societies Letters, 1,211-212.
LIPKIN,M. (1970) Carbenoxolone sodium and the rate of extrusion of gastric epithelial cells. Carbenoxolone
Sodium, pp. 11-17. Ed. J. H. Baron & F. M. Sullivan. Butterworths, London.
MEITNER,
P.A. (1971) Bovine pepsinogens and pepsins; the sequence around a reactive aspartyl residue. Biochemical Journal, 124,673-676.
PARKE,
D.V. (1968) Metabolic studies with carbenoxolone in man and animals. A Symposium on Carbenoxolone
Sodium, pp. 15-23. Ed. J. M. Robson & F. M. Sullivan. Butterworths, London.
PARKE,D.V.,POLLOCK,
S. & WILLIAMS,R.T. (1963) The fate of tritium-labelled B-glycyrrhetic acid in the rat.
Journal of Pharmacy and Pharmacology, 15,500-506.
ROBERTS,
N.B. & TAYLOR,
W.H. (1972) A comparison of the properties of pure human pepsins 1,2, 3 and 5.
Biochemical Journal, 128,103~.
ROBERTS,
N.B. & TAYLOR,
W.H. (1973) The inhibition by carbenoxolone sodium of individual human pepsinogens and pepsins. Clinical Science, 4 4 , 6 ~ - 7 ~ .
SEIJFFERS,
M.J., SEGAL,H.L. & MILLER,L.L. (1963) Separation of pepsin I, pepsin IIa, pepsin IIb and pepsin
I11 from human gastric mucosa. American Journal of Physiology, 205,1099-1 105.
TAYLOR,
W.H. (1959) The proteolytic activity of gastric juice and gastric mucosal extracts from patients with
chronic gastric and duodenal ulcer. Journal of Clinical Pathology, 12,338-343.
TAYLOR,
W.H. (1962) Proteinases of the stomach in health and disease. Physiological Reviews, 42,519-553.
TAYLOR,
W.H. (1968) Biochemistry of pepsins. Handbook of Physiology-Alimentary Canal, Vol. V, pp. 25672587. Ed. C. F.Code & W. Heide.
TAYLOR,
W.H. (1970a) Pepsins of patients with peptic ulcer. Nature, 227,76-77.
TAYLOR,
W.H. (1970b) The pepsins of patients with peptic ulcer. Journalof Clinical Pathology, 23,378.
TROP,M. & BIRK,Y. (1970) The specificity of proteinases from Streptomyces griseus (Pronase). Biochemical
Journal, 116, 19-25.
WEBER,
K. & OSBORN,
M. (1969) The reliability of molecular weight determinations by dodecyl sulphate-polyacrylamide gel electrophoresis. Journal of Biological Chemistry, 244,4406-4412.
WHITEHOUSE,
M.W., DEAN,P.D.G. & HALSALL,
T.G. (1967) Uncoupling of oxidative phosphorylation by
glycyrrhetic acid, fusidic acid and some related tirterpenoid acids. Journal of Pharmacy and Pharmacology, 19, 533-544.