A study of the effect on nucleophilic hydrolytic activity of pancreatic elastase, trypsin, chymotrypsin, and leucine aminopeptidase by boronic acids in the presence of arabinogalactan: A subsequent study on the hydrolytic activity of chymotrypsin by boronic acids in the presence of mono-, di-, and trisaccharides

Article abstract of DOI:10.1016/j.bioorg.2003.08.004

The hydrolytic activity of trypsin, chymotrypsin, elastase, and leucine aminopeptidase, is inhibited by different boronic acids. However, all the enzymes are inhibited by the compound CbzAla(boro)Gly(OH)₂. Therefore, these additives can control the nucleophilic hydrolytic activity of these enzymes.

Full text of DOI:10.1016/j.bioorg.2003.08.004

BIOORGANIC
CHEMISTRY
Bioorganic Chemistry 31 (2003) 464–474
www.elsevier.com/locate/bioorg
A study of the effect on nucleophilic hydrolytic
activity of pancreatic elastase, trypsin,
chymotrypsin, and leucine aminopeptidase
by boronic acids in the presence
of arabinogalactan: a subsequent study
on the hydrolytic activity of chymotrypsin
by boronic acids in the presence
of mono-, di-, and trisaccharides
Reem Smoum,a Abraham Rubinstein,b,* and Morris Srebnika,*
a
Department of Medicinal Chemistry and Natural Products, Faculty of Medicine,
The Hebrew University in Jerusalem, P.O. Box 12065, Jerusalem, 91120, Israel
b
Department of Pharmaceutics, School of Pharmacy, Faculty of Medicine,
The Hebrew University in Jerusalem, P.O. Box 12065, Jerusalem 91120, Israel
Received 18 May 2003
Abstract
The hydrolytic activity of trypsin, chymotrypsin, elastase, and leucine aminopeptidase, is
inhibited by different boronic acids. However, all the enzymes are inhibited by the compound
CbzAla(boro)Gly(OH)2. Therefore, these additives can control the nucleophilic hydrolytic
activity of these enzymes.
Ó 2003 Elsevier Inc. All rights reserved.
Keywords: Serine proteases; Boronic acids; Inhibition; Hydrolytic activity; Saccharides
* Corresponding authors. Fax: +972-2-675-8201.
E-mail address: msrebni@md.huji.ac.il (M. Srebnik).
0045-2068/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved.
doi:10.1016/j.bioorg.2003.08.004

R. Smoum et al. / Bioorganic Chemistry 31 (2003) 464–474
465
1. Introduction
Ring formation between boronates and hydroxylated compounds such as saccha-
rides [1], amino acids [2], RNA nucleosides [3], and other polar compounds renders
these hydroxylated compounds more lipophilic and facilitates their transport across
cell membranes and lipid bilayers [4,5]. The association constants for such complexes,
in the range of 10–104 Mꢀ1, would suggest that significant amounts of free boronic
acids are available on both sides of the membrane. These two attributes of boronic
acids, namely, facile ring formation with suitable functional groups of polar mole-
cules, and relatively small association constants, would seem to make boron- polysac-
charide complexes good candidates for the delivery of peptides or peptidomimetics.
Examples of polysaccharide drug delivery platforms include the natural polysaccha-
rides guar gum and pectin. Our group has modified and utilized these platforms for
colon specific drug delivery because of their unique characteristic to be specifically de-
graded by colonic bacteria [6,7]. The colon has been discussed as a potential site of
peptidic drug delivery due to apparently lower levels of proteases in this region [8].
While the development of colonic delivery systems for peptides and proteins is our
long term goal, in the short term, we are interested in studying the inhibitory activity
of some boronic acids on various serine proteases found in the GI, such as trypsin,
chymotrypsin, elastase, and leucine aminopeptidase, and the effect of added saccha-
rides on the inhibition ability of these boronic acids. Boronic acids are exceptionally
potent inhibitors of serine proteases. This characteristic is widely believed to derive
from the boronyl groupÕs ability to mimic the transition state of the enzyme-catalyzed
reaction [9]. Suenaga et al. [10] proved that the inhibitory effect of phenylboronic acid
on chymotrypsin was intensified by added diols but weakened by tripodal additives.
Therefore the aim of our study was to assess the effects of saccharides on the inhibi-
tion of GI tract serine proteases by boronic acids. In addition, given the recent ap-
proval of bortezomib [11] and a number of other boronic acid-based compounds in
various stages of research and development [12], our study has implications for other
boronic acid therapeutics that might be given orally.
2. Materials and methods
2.1. Enzymatic analysis
a-Chymotrypsin, Type II: From bovine pancreas, was purchased from Sigma
(MW 25,100). The hydrolytic reaction was carried out according to KouzomaÕs
method (37 °C, standard pH 8.0, with 50 mM phosphate buffer, 0.2 M NaCl,
0.3 vol% methanol plus 0.03 vol% DMSO) and the progress of the reaction was fol-
lowed by monitoring the appearance of the absorption band of p-nitroaniline at
410 nm (Sub ¼ benzoyl-tyrosine-p-nitroanilide).
Trypsin, Type II-S: From porcine pancreas was purchased from Sigma. The hy-
drolytic reaction was carried out at 37 °C, standard pH 7.5, with 50 mM Tris buffer,
0.02 M CaCl2, 10 vol% DMF, and the progress of the reaction was followed by

466
R. Smoum et al. / Bioorganic Chemistry 31 (2003) 464–474
monitoring the appearance of the absorption band of p-nitroaniline at 410 nm
(Sub ¼ benzoyl-arginine-p-nitroanilide).
Elastase, Type II-A: From porcine pancreas was purchased from Sigma. The hy-
drolytic reaction was carried out at 37 °C, standard pH 7.5, with 50 mM phosphate
buffer, 0.2 M NaCl, and the progress of the reaction was followed by monitoring the
appearance of the absorption band of p-nitroaniline at 410 nm (Sub ¼ N-succinyl-
Ala-Ala-Ala-p-nitroanilide).
Leucine aminopeptidase, Type VI: From porcine kidney microsomes, was pur-
chased from Sigma. The hydrolytic reaction was carried out at 37 °C, standard pH
7.5, with 50 mM phosphate buffer, 0.02 M CaCl2, 10 vol% DMF, and the progress
of the reaction was followed by monitoring the appearance of the absorption band
of p-nitroaniline at 410 nm (Sub ¼
L-leucine-p-nitroanilide).
The five kinds of boronic acids used were phenylboronic acid, 3- aminophenylbo-
ronic acid, 4-fluorophenylboronic acid, (Cbz-alanyl)aminomethyl boronic acid, and
(Cbz-phenylalanyl)aminomethyl boronic acid. They were all tested for their ability
to inhibit trypsin, chymotrypsin, elastase, and leucine aminopeptidase.
2.2. Synthesis and identification
1
NMR experiments. H NMR spectra was recorded at 300 MHz (Varian Unity)
instrument. 13C NMR 11B NMR spectras were recorded on a Varian 300 MHz in-
strument at a frequency of 75.9 MHz and 96.29 MHz, respectively. The chemical
shifts were reported with respect to CD3OD for the peptidyl boronic acids.
2.3. Synthesis of the peptidyl boronic acids
A mixture of triisopropylborate (60 ml, 0.25 mol) and dry pinacol (30 g, 0.25 mol)
was heated for 3 h and stirred at 115 °C. The isopropanol was distilled, the residue
1
was cooled, and distilled under vacuo. H NMR (CDCl3): d 1.01 (dd, 6H), 1.10 (s,
12H), 4.20 (sept., 1H). 13C{1H} NMR (CDCl3): d 24.39, 24.63, 67.35, 82.48.
A solution of diiodomethane (4.05 ml, 50.2 mmol) and isopropylpinacolboronate
(10 ml, 46.5 mmol) in THF (200 ml) was cooled to )78 °C and n-BuLi (20 ml of 1.6 M
solution) was dropped during 30 min at )78 °C with stirring. The reaction mixture
was stirred overnight at room temperature. Then 6 N solution of HCl in ether was
added (10 ml) and the solvent was removed. The residue was distilled at 0.1 mmHg.
To a cooled solution of hexamethyldisilazane in THF was added n-BuLi. The
cooling bath was then removed and the mixture was allowed to warm to 0 °C within
0.5 h and then stirred at 0 °C for an additional 0.5 h. The resulting solution of lithi-
ohexamethyldisilizane was cooled again, and the iodomethanedioxaborolane in THF
was added within 10 min. The reaction was then allowed to warm to 20 °C and stir-
red overnight, after which hexane was added and the precipitated LiI was filtered
off. The filtrate was concentrated in vacuo and the residue Kugelrohr distilled to
give 2-[bis(trimethylsilyl)aminomethyl]-4,4,5,5-tetramethyl-1,3,2- dioxaborolane as
a colorless liquid. 1H NMR (CDCl3): d 0.06 (s, 18H), 1.23 (s, 12H), and 2.41 (s, 2H).
13C{1H} NMR (CDCl3): d 1.37, 24.65, 29.08, 83.22.

R. Smoum et al. / Bioorganic Chemistry 31 (2003) 464–474
467
Cesium carbonate (1.21 g, 3.72 mmol) was added at 20 °C to a solution of Cbz-al-
anine/Cbz-phenylalanine (7.4 mmol) in MeOH. The mixture was stirred overnight,
the methanol was then evaporated in vacuo and the resulting cesium salt was dried
over P2O5 at 0.2 mm for 24 h. It was then dissolved in dry DMF (5 ml) under N2, the
solution cooled to )15 °C, and isobutylchloroformate added in one portion and the
mixture stirred at )15 °C for 0.5 h. 2-[Bis(trimethylsilyl)aminomethyl]-4,4,5,5-tetra-
methyl-1,3,2-dioxaborolane (7.43 mmol) was added with stirring followed by water
(0.5 ml, 27.6 mmol). After stirring for 3 h at )15 °C the reaction mixture was
quenched with 1 M aqueous HCl (40 ml), extracted with EtOAc (60 ml), and washed
with water (10 ml). The boronic acid was extracted from EtOAc into 1 M NaOH
(3 ꢁ 6 ml), washed with EtOAc (2 ꢁ 20 ml), and dried on sodium sulfate. After filter-
ing and concentrating in vacuo, the residue was chromatographed (10–25% metha-
nol in chloroform). The resulting boronic dipeptide was obtained as a white solid. 1H
NMR (CD3OD) for CbzPhe(boro)Gly(OH)2: d 2.32 (AB, J ¼ 2.64, 2H), 2.91–3.21
(m, 2H), 4.56 (m, 1H), 5.02 (s, 2H), 7.20–7.34 (m, 10H). 13C NMR (CD3OD): d
31.36 (broad), 38.47, 55.23, 67.65, 127.95, 128.67, 128.94, 129.39, 129.52, 130.23,
137.67, 137.87, 158.00, 178.39. 11B NMR (CD3OD): d 17.46. Anal. Calcd. for
C18H21BN2O5: C, 60.70; H, 5.94; N, 7.86. Found: C, 60.76; H, 5.85; N, 7.92%.
1H NMR (CD3OD) for CbzAla(boro)Gly(OH)2: d 2.81(d, 3H), 2.34 (s, 2H), 4.35
(m, 1H), 5.09 (s, 2H), 7.23–7.35 (m, 5H).13C NMR (CD3OD): d 16.70, 30.85 (broad),
47.83, 66.65, 127.80, 138.84, 156.97, 178.30. 11B NMR (CD3OD): d 14.43. Anal.
Calcd. for C12H17BN2O5: C, 51.46; H, 6.12; N, 10.00. Found: C, 50.80; H, 5.90;
N, 9.71%.
3. Results and discussion
Phenylboronic acid, 3-aminophenylboronic, 4-fluorophenylboronic, CbzAla(boro)-
Gly(OH)2 and CbzPhe(boro)Gly(OH)2(1–5) were all tested for their ability to inhibit
trypsin, chymotrypsin, elastase, and leucine aminopeptidase. The progress of the reac-
tion was followed by monitoring the appearance of the absorption band at 410 nm
(p-nitroaniline).

468
R. Smoum et al. / Bioorganic Chemistry 31 (2003) 464–474
The dissociation constants, Ki, for the inhibitors with all the four enzymes are
shown in Table 1. In addition Lineweaver–Burk plots for a representative grouping
of inhibitors with each enzyme are illustrated in Figs. 1–4. As can be seen in
the plots, the inhibition was competitive and reversible for all the compounds that
Table 1
Inhibition of a-chymotrypsin, trypsin, elastase, and leucine aminopeptidase by boronic acids 1–5
Boronic acid
Chymotrypsin,
Ki (mM)
Trypsin,
Ki (mM)
Elastase,
Ki (mM)
Aminopeptidase,
Ki (mM)
1. Phenylboronic acid
9.70
>0
>0
>0
>0
35.9
>0
1.83
3.61
1.02
0.27
0.60
>0
>0
2. 3-Aminophenylboronic acid.
3. 4-Fluorophenylboronic acid
4. CbzAla(boro)Gly(OH)2
5. CbzPhe(boro)Gly(OH)2
31.01
1.08
1.96
>0
0.35
1.02
Chymotrypsin
2000
1000
0
Control
Phenylboronic acid
4-Fluorophenylboronic acid
3-Aminophenylboronic acid
CbzAla(boro)Gly(OH)2
CbzPhe(boro)Gly(OH)2
0
5
10
15
20
25
1/[S], mM-1
Fig. 1. Lineweaver plot for the inhibition of chymotrypsin without and with inhibitors. Reactions were
initiated by adding the substrate benzoyl-tyrosine-p-nitroanilide to a solution of chymotrypsin and inhib-
itor. The final concentration of the inhibitor was 1.071 mM.
Trypsin
6000
Control
Phenylboronic acid
4-Fluorophenylboronic acid
4000
3-Aminophenylboronic acid
CbzAla(boro)Gly(OH)2
CbzPhe(boro)Gly(OH)2
2000
0
0
10
20
30
1/[S], mM-1
Fig. 2. Lineweaver plot for the inhibition of trypsin without and with inhibitors. Reactions were initiated
by adding the substrate benzoyl-arginine-p-nitroanilide to a solution of trypsin and inhibitor. The final
concentration of the inhibitor was 1.071 mM.

R. Smoum et al. / Bioorganic Chemistry 31 (2003) 464–474
469
Fig. 3. Lineweaver plot for the inhibition of elastase without and with inhibitors. Reactions were initiated
by adding the substrate N-succinyl-Ala-Ala-Ala-p-nitroanilide to a solution of elastase and inhibitor. The
final concentration of the inhibitor was 1.071 mM.
Aminopeptidase
Control
7000
Phenylboronic acid
4-Fluorophenylboroinic acid
5000
3-Aminophenylboronic acid
CbzAla(boro)Gly(OH)2
3000
CbzPhe(boro)Gly(OH)2
1000
-1000
-10
0
10
20
30
1/[S], mM-1
Fig. 4. Lineweaver plot for the inhibition of leucine aminopeptidase without and with inhibitors. Reac-
tions were initiated by adding the substrate leucine-p-nitroanilide to a solution of leucine aminopeptidase
and inhibitor. The final concentration of the inhibitor was 1.071 mM.
inhibited the enzymes except for CbzPhe(boro)Gly(OH)2 that inhibited leucine
aminopeptidase non-competitively and irreversibly.
As can be seen in Table 1, phenylboronic acid and 4-fluorophenylboronic acid re-
vealed moderate inhibition properties for chymotrypsin but 3-aminophenylboronic
acid inhibited the enzyme very weakly. This can be explained by the ability of the
electron donating group (NH2) to weaken the acidity of the boron atom, which,
in turn, prevents further attack on the nucleophilic center. Another reason for the
difference in activity with and without the 3-amino substituent may be due to direct
interactions of this group with chymotrypsin. The best inhibition activity was dem-
onstrated by CbzAla(boro)Gly(OH)2 and CbzPhe(boro)Gly(OH)2 compounds. Dix-
on plots [13] (Fig. 1) showed that both compounds are competitive inhibitors and
that the dissociation constants (Kis) of the enzyme–inhibitor complexes are 1.08
and 1.96 mM for CbzAla(boro)Gly(OH)2 and CbzPhe(boro)Gly(OH)2, respectively

470
R. Smoum et al. / Bioorganic Chemistry 31 (2003) 464–474
at 25 °C and pH 7.5. The competitive nature of the inhibition indicates that both
compounds bind at the active site of the a-chymoytrypsin.
As for trypsin, none of the boronic acids significantly affected the hydrolytic ac-
tivity of the enzyme, except for CbzAla(boro)Gly(OH)2,which inhibited trypsin com-
petitively and reversibly with a Ki value of 35.90 mM (Fig. 2).
All the boronates inhibited elastase, but CbzAla(boro)Gly(OH)2 and CbzPhe(boro)
Gly(OH)2 had the best inhibition abilities with Ki values of 0.27 and 0.60 mM, respec-
tively. These two compounds inhibited the enzyme in a competitive and reversible man-
ner (Fig. 3).
Leucine aminopeptidase was only inhibited by CbzAla(boro)Gly(OH)2 and
CbzPhe(boro)Gly(OH)2 with Ki values of 0.35 and 1.02 mM, respectively.
Whereas CbzAla(boro)Gly(OH)2 inhibited the enzyme competitively and revers-
ibly, CbzPhe(boro)Gly(OH)2 inhibited the enzyme noncompetitively and irrevers-
ibly (Fig. 4).
Bachovchin and co-workers [14,15] have demonstrated that the binding mode
used by a particular boronic acid inhibitor appears to depend on how well the inhib-
itor matches the structure of the physiological or natural substrate of the serine pro-
tease. In cases in which a boronic acid is a good substrate mimic, formation of a
tetrahedral complex with the active site serine is favored. On the other hand, when
boronic acids have structures that are not well related to that of the substrate, they
tend to coordinate with histidine and serine. The serine proteases chymotrypsin and
subtilisin have been shown to be inhibited by aromatic boronic acids with chymo-
trypsin favoring large aromatic hydrophobic side chains (Phe, Tyr, and Trp) while
trypsin favors positively charged groups (Arg and Lys). Consistently, in this study,
all the boronic acids that had aromatic groups were able to inhibit chymotrypsin
to different extents depending on the substituents incorporated. CbzPhe(boro)
Gly(OH)2 and CbzAla(boro)Gly(OH)2 were expected [16] to act as cysteine proteases
inhibitors, but they were proved ineffective up to a concentration of 10 mM. Despite
the fact that inhibitors of serine proteases very often inhibit cysteine proteases [17],
these two compounds were good inhibitors of chymotrypsin since they are peptidyl
boronic acids and it is known that peptidyl boronic acids that are close structural
analogues of good substrates for the enzymes can act as potential transition state
analogue inhibitors of the representative serine proteases.
Only one boronic acid was able to inhibit trypsin slightly, CbzAla(boro)
Gly(OH)2, although it does not contain charged groups like Arg or Lys. One possible
explanation for this behavior is that this compound may be able to fit into the active
site and interact with histidine or serine.
It is known that chymotrypsin prefers an aromatic residue in the P1 site whereas
pancreatic elastase and leukocyte elastase prefer Ala and Val, respectively. Therefore
the peptidyl boronic acids CbzAla(boro)Gly(OH)2 and CbzPhe(boro)Gly(OH)2 were
the most effective inhibitors for elastase with Ki values of 0.27 and 0.60 mM, respec-
tively, while the other boronic acids inhibited the enzyme to a lesser extent.
Aminopeptidases are a family of exopeptidases responsible for cleavage at the
N-termini of proteins and peptides. Their activities have been compared in the
homogenates of duodenum, jejunum, and ileum of albino rabbits with
L-leucine,

R. Smoum et al. / Bioorganic Chemistry 31 (2003) 464–474
471
L
-alanine, and L-arginine-4-methoxy-b-naphthylamide as the substrates [18]. All the
compounds showed very weak inhibitory activity toward leucine aminopeptidase ex-
cept for the compounds CbzAla(boro)Gly(OH)2 and CbzPhe(boro)Gly(OH)2 which
inhibited the enzyme effectively.
3.1. Sugars can increase/decrease the inhibitory effect of boronic acids on the hydrolytic
activity of a-chymotrypsin
There are several possible reasons why saccharides might increase or decrease the
observed inhibition of serine nucleophilic enzymes by boronic acids. These include:
(a) Competition for the boronic acids by the saccharide (reservoir/sink effect), which
should decrease the apparent inhibition. (b) Complexation with the boronic acid and
binding to the enzyme as the complex, which might increase or decrease the apparent
inhibition. (c) Direct effects (allosteric or other) by the saccharide on the enzyme.
Therefore the effect of added mono-, di-, and trisaccharides on the inhibition abil-
ity of some boronic acids in the a-chymotrypsin-catalyzed hydrolysis of N-benzoyl-
L
-tyrosine p-nitroanilide (as a substrate) was studied. As expected, the hydrolytic
activity was affected to different extents by the addition of saccharides to the boron-
ic-acid-inhibited system depending on the boronic acid and on the kind of saccharide
that is used.
Table 2 shows the effects of added saccharides on the activity of chymotrypsin
both in the absence and in the presence of certain inhibitors. The results showed that
saccharide-only (no inhibitor) additives had no effect on the activity of the enzyme,
therefore, the saccharide effects on the inhibition were due to interactions with
Table 2
The effect of the addition of saccharides on the activity of chymotrypsin both in the absence and in the
presence of the inhibitors: phenylboronic acid, CbzPhe(boro)Gly(OH)2, and CbzAla(boro)Gly(OH)2
Compound
% activity no
inhibitor
% activity
phenylboronic
acid
% activity
CbzPhe(boro)
Gly(OH)2
% activity
CbzAla(boro)
Gly(OH)2
No sugar (Inh.)
+ Lactulose
+ Lactose
100.00
104.72
104.72
102.90
96.97
66.50
66.50
80.20
99.25
95.82
89.92
32.96
48.67
59.38
86.16
84.69
73.33
26.04
15.48
23.85
23.48
34.58
28.55
44.96
72.54
61.30
33.02
35.58
-
18.38
100.00
22.02
22.36
30.64
30.04
48.86
41.29
71.56
31.25
39.36
18.95
+ Sucrose
+ Raffinose
+ Cellobiose
+ Glucose
+ Mannose
104.29
101.20
100.91
101.02
111.07
104.33
106.15
+ Galactose
+ Methylglucose
+ Methylgalactose
+ Methylmannose
50 mM phosphate buffer and 0.2 M NaCl, [chymotrypsin] ¼ 7.9 ꢁ 10ꢀ6 mol dmꢀ3, [Sub] ¼ 9.87 ꢁ
10ꢀ5 mol dmꢀ3, [boronic acid] ¼ 1.12 ꢁ 10ꢀ3 mol dmꢀ3, [monosaccharide] ¼ 4.00 ꢁ 10ꢀ3 mol dmꢀ3, [methyl-
monosaccharide] ¼ 3.71 ꢁ 10ꢀ3 mol dmꢀ3
,
[disaccharide] ¼ 2.1 ꢁ 10ꢀ3 mol dmꢀ3
,
[trisaccharide] ¼ 1.21 ꢁ
10ꢀ3 mol dmꢀ3
.

472
R. Smoum et al. / Bioorganic Chemistry 31 (2003) 464–474
boronic acid and not due to direct effect with the enzyme, and possibility (c) that was
mentioned earlier can be ruled out.
For phenylboronic acid, it was found that when
-galactose were added, the rate of the hydrolytic reaction was further suppressed.
The inhibition efficiency for these three saccharides was in the order of (+)-glu-
cose > (+)-mannose > (+)-galactose and this agrees well with the data published
by Suenaga et al. [10] On the other hand, with the addition of other monosaccharides
such as methyl-a- -glucopyranoside, methyl-b- -galactopyranoside, and methyl-a-
-mannopyranoside, the inhibition ability was weakened and some of the enzymatic
D(+)-glucose, D(+)-mannose, and
D
D
D
D
D
D
D
activity was regenerated completely. The presence of disaccharides or trisaccharides
also decreased the inhibitory effect of phenylboronic acid except for lactulose that
did not affect the inhibition at all. It is known that for a-
onates, the sugar ring has a furanose structure with a conformation between T32 and
E3. In addition, methyl-a-
-glucopyranoside reacts with phenylboronic acid (1 mol)
D-glucose mono and bisbor-
D
to give a crystalline 4,6-cyclic ester which, in turn, forms a 2,3-(diphenylpyroboro-
nate) (containing a seven-membered ring) with an excess of the reagent while phen-
ylboronic acid condenses smoothly with methyl-a-D-mannopyranoside, to give the
2,3:4,6-diester. Therefore, this difference in the structure of the final product may
have the effect of either increasing or decreasing the inhibitory activity, i.e., the fu-
ranose structure of the ring of phenylboronic acid with glucose and mannose fits
the active site of chymotrypsin.
The effect of saccharides on the CbzPhe(boro)Gly(OH)2 inhibition of chymotryp-
sin shows that the inhibitory effect is weakened by monosacchaarides. The effect of
decreasing the inhibition varies for different monosaccharides, and their ability to de-
crease the inhibition increases in the order
D
(+)-mannose >
D
(+)-galactose >
D(+)-
-glucopyranoside. On the other
glucose> methyl-b- -galactopyranoside > methyl-a-
D
D
hand, the addition of disaccharides or a trisaccharide did not affect the inhibition
significantly except for the lactulose that removed the inhibition completely (i.e.,
the hydrolytic activity of the enzyme was regenerated completely). One possible
explanation for this is that the association constant of the boronate with lactulose
is so high (we could not detect it by NMR due to instantaneous formation of a
complex) and therefore it was able to form a complex with the boronic acid and
bind to the enzyme as a complex which lead to a decrease in the apparent inhibition.
Another possibility is the competition for the boronic acid by lactulose, which de-
creased the inhibition significantly. This possibility depends on the concentration
of the enzyme, boronic acid, and the saccharide.
For CbzAla(boro)Gly(OH)2, the hydrolytic activity of the enzyme was affected by
almost a similar manner as CbzPhe(boro)Gly(OH)2 by the addition of mono-, di-,
and trisaccharide.
One possible reason for these effects of sugars is the competition for boronic acids by
the saccharides as mentioned earlier, however, this depends upon the relative concen-
trations of the enzyme, boronic acid, and saccharide component. On the other hand,
boronic acids form cyclic esters with saccharides and the reaction occurs reversibly
and rapidly at ambient temperature, therefore, the complexation with the boronic acid
and binding to the enzyme as the complex must be the reason for our results.

R. Smoum et al. / Bioorganic Chemistry 31 (2003) 464–474
473
Table 3
The increase/decrease in inhibitory activity of boronic acids with chymotrypsin, trypsin, elastase, and
leucine aminopeptidase as a result of the addition of arabinogalactan (1.12 mM)
Compound
Chymotrypsin
Trypsin
Elastase
Leucine
aminopeptidase
1. Phenylboronic acid
)4.03
)3.85
)2.10
)26.65
)28.47
+6.50
+6.39
+3.25
)2.60
+0.89
+8.15
)6.70
+6.50
)15.53
)1.43
) 30.67
)21.25
–
+5.13
+2.67
+0.75
)50.60
)40.25
)3.60
2. 3-Aminophenylboronic acid
3. 4-Fluorophenylboronic acid
4. CbzAla(boro)Gly(OH)2
5. CbzPhe(boro)Gly(OH)2
Arabinogalactan only—no
inhibitor
3.2. Effect of a polysaccharide on the inhibition of a-chymotrypsin, trypsin, elastase,
and leucine aminopeptidase by the five boronic acids
Since our interest was with the effect of polysaccharides on the inhibitory activity
of boronic acids, we examined the effect of arabinogalactan on the inhibitory prop-
erties of all boronic acid with the four enzymes as shown in Table 3. With arabino-
galactan only (no inhibitor), the activity of chymotrypsin was increased slightly by
6.80%, while the activity of both trypsin and leucine aminopeptidase was decreased
by 6.70 and 3.60%, respectively, and elastase was not affected at all. These data show
that there is no direct effect by arabinogalactan on the enzymes. However, in the
presence of the inhibitors, arabinogalactan decreased the inhibitory activity of both
CbzAla(boro)Gly(OH)2 and CbzPhe(boro)Gly(OH)2 with chymotrypsin, elastase,
and leucine aminopeptidase. The most significant effect was with leucine aminopep-
tidase, where the inhibition activity was decreased by 50.60 and 40.25%, respectively,
i.e., the Ki values were decreased from 1.02 to 3.95 mM for CbzPhe(boro)Gly(OH)2
and from 0.35 to 1.26 mM for CbzAla(boro)Gly(OH)2. Therefore, it seems that the
boronic acids form complexes with arabinogalactan and binds to the enzyme as the
complex, and this probably decreased the apparent inhibition.
These findings support the view that the enzyme active site ÔrecognizesÕ the mole-
cular structure of the boronic acid–saccharide complexes, and that not all saccharides
act as Ôco-inhibitorsÕ in the boronic acid inhibition system.
4. Conclusion
These preliminary results shows that both compounds, CbzAla(boro)Gly(OH)2
and CbzPhe(boro)Gly(OH)2 are good inhibitors for chymotrypsin, elastase, and
leucine aminopeptidase. For chymotrypsin, this inhibitory activity can be either
unaffected or decreased by the addition of monosaccharides, disaccharides, and
a trisaccharide. The most important result was obtained with the polysac-
charide arabinogalactan that decreased the inhibitory activity of the two inhibi-
tors CbzAla(boro)Gly(OH)2 and CbzPhA(boro)Gly(OH)2 with all the enzymes.

474
R. Smoum et al. / Bioorganic Chemistry 31 (2003) 464–474
This finding is important since it means that the activity of the inhibitor will be
reduced by possible interactions with the polysaccharide. Our study has the po-
tential applicability for boronic acid-based drug delivery systems and the potential
effects of saccharides in the body or diet on new boronic acid-based therapeutic
agents such as bortezomib.
Acknowledgment
This project was supported by a grant from the Israeli Science Foundation Grant
No. 663/99-2.
References
[1] P.R. Westmark, S.T. Gardiner, B.D. Bradley, J. Am. Chem. Soc. 118 (1996) 11093–11100.
[2] M.T. Reetz, J. Huff, J. Rudolph, K. Tolner, A. Deege, R. Goddard, J. Am. Chem. Soc. 116 (1994)
11588–11589.
[3] L.K. Mohler, A.W. Czarnik, J. Am. Chem. Soc. 115 (1993) 2998–2999.
[4] B.D. Bradley, Supramol. Chem. 7 (1996) 55–60.
[5] S.T. Gardiner, B.D. Smith, P.J. Duggan, M.J. Karpa, G.J. Griffin, Tetrahedron 55 (1999) 2857–2864.
[6] I. Gliko-Kabir, B. Yagen, A. Penhasi, A. Rubinstein, Pharm. Res. 15 (1998) 1019–1025.
[7] A. Rubinstein, R. Radi, Eur. J. Pharm. Biopharm. 41 (1995) 291–295.
[8] A. Rubinstein, B. Tirosh, M. Baluom, T. Nassar, A. David, I. Gliko-kabir, R. Radi, M. Friedman,
J. Control. Release 46 (1998) 59–73.
[9] B. Furie, B.C. Furie, Cell. 53 (1988) 505–518.
[10] H. Suenaga, H. Yamamoto, S. Shinkai, Pure Appl. Chem. 68 (1996) 2179–2186.
[11] V.J. Palombella, E.M. Conner, J.W. Fuseler, A. Destree, J.M. Davis, F.S. Larous, R.E. Wolf,
J. Huang, S. Brand, P.J. Elliott, D. Lazarus, T. McCormack, L. Parent, R. Stein, J. Adams, M.B.
Grisham, Proc. Natl. Acad. Sci. USA 95 (1998) 15671–15676.
[12] W. Yang, X. Gao, B. Wang, Med. Res. Rev. 23 (2003) 346–368.
[13] M. Dixon, Biochem. J. 55 (1953) 170–171.
[14] W.W. Bachovchin, W.Y.L. Wong, S. Farr-Jones, A.B. Shenvi, C.A. Kettner, Biochemistry 27 (1988)
7689–7698.
[15] E. Tsilikouras, C.A. Kettner, W.W. Bachouvchin, Biochemistry 31 (1992) 12839–12846.
[16] V. Martichonok, J.B. Jones, Bioorg. Med. Chem. 5 (1997) 679–684.
[17] E. Shaw, Adv. Enzymol. 63 (1990) 271–347.
[18] W. Wang, J. Drug Target. 4 (1996) 195–232.