Post on 15-Jul-2016
Effect of Human Serum Albumin on the Kineticsof 4-Methylumbelliferyl-b-D-N-N0-N00 TriacetylchitotriosideHydrolysis Catalyzed by Hen Egg White Lysozyme
Cristian Calderon • Elsa Abuin • Eduardo Lissi •
Rodrigo Montecinos
Published online: 12 July 2011
� Springer Science+Business Media, LLC 2011
Abstract The effect of human serum albumin (HSA)
addition on the rate of hydrolysis of the synthetic substrate
4-methylumbelliferyl-b-D-N-N0-N00 triacetylchitotrioside
((NAG)3-MUF) catalyzed by hen egg white lysozyme has
been measured in aqueous solution (citrate buffer 50 mM
pH = 5.2 at 37 �C). The presence of HSA leads to a decrease
in the rate of the process. The reaction follows a Michaelis–
Menten mechanism under all the conditions employed. The
catalytic rate constant decreases tenfold when the albu-
min concentration increases, while the Michaelis constant
remains almost constant in the albumin concentration range
employed. Ultracentrifugation experiments indicate that the
main origin of the observed variation in the kinetic behavior
is related to the existence of an HSA–lysozyme interaction.
Interestingly, the dependence of the catalytic rate constant
with albumin concentration parallels the decrease of the free
enzyme concentration. We interpret these results in terms of
the presence in the system of two enzyme populations;
namely, the HSA associated enzyme which does not react
and the free enzyme reacting as in the absence of albumin.
Other factors such as association of the substrate to albumin
or macromolecular crowding effects due to the presence of
albumin are discarded. Theoretical modeling of the structure
of the HSA–lysozyme complex shows that the Glu35 and
Asp52 residues located in the active site of lysozyme are
oriented toward the HSA surface. This conformation will
inactivate lysozyme molecules bound to HSA.
Keywords Enzyme kinetics � Lysozyme � Human serum
albumin
Abbreviations
DS Dansylsarcosine
DNSA Dansylsulfonamide
GPNA N-glutaryl-L-phenylalanine
p-nitroanilide
HSA Human serum albumin
kcat Catalytic rate constant
KM Michaelis constant
((NAG)3-MUF) 4-Methylumbelliferyl-b-D-N-N0-N00
triacetylchitotrioside
Glu Glutamic acid
Asp Aspartic acid
1 Introduction
Serum albumin is the most abundant protein in blood
plasma, (accounting for ca. 60% of the total protein; its
concentration amounts to ca. 800 lM [11, 16]. Albumins,
and in particular HSA are able to bind and transport a wide
range of organic compounds, including drugs and fatty
acids [1, 2, 10, 12, 13, 17, 19, 20, 24, 27]. This capacity of
albumin to bind hydrophobic drugs in plasma modulates
their delivery to cells in vivo and reduces the free drug
concentration, a determinant parameter of its physiological
activity [14].
The presence of albumin in the reaction medium can
affect the activity of enzymes through different factors,
namely, (a) the adsorption of the substrate onto the protein,
(b) the so called ‘‘crowding effect’’ which would affect the
enzyme conformation, and, hence, its activity, and (c) an
C. Calderon (&) � E. Abuin � E. Lissi
Facultad de Quımica y Biologıa, Universidad de Santiago de
Chile, Av. Bernardo O’Higgins 3363, Santiago, Chile
e-mail: cristian.calderon@usach.cl
R. Montecinos
Facultad de Quımica, Pontificia Universidad Catolica de Chile,
Santiago, Chile
123
Protein J (2011) 30:367–373
DOI 10.1007/s10930-011-9339-8
interaction between the enzyme and the protein which
would also alter the enzyme conformation even at low
albumin concentrations. There are several reports that take
into account factors (a) and (b), but, to the best of our
knowledge, factor (c) has not been yet considered in kinetic
studies.
Awad-Elkarim et al. [4] demonstrated that p-nitrophenyl
phosphate, a substrate for Lipases, readily binds to bovine
and human serum albumin. Wang et al. [32] studied the
effect of addition of albumin on the acyl chain specificity
of a lipoprotein lipase using triacylglycerols of various
acyl-chain lengths as substrates. An unexpected finding of
this work was that the albumin ligand binding site is
accessible not only to long-chain fatty acids, but also to
short and medium chain monoacid triacylglycerol sub-
strates. It was concluded that the observed inhibitory effect
of albumin on the lipase-catalyzed hydrolysis of trihexa-
noylglycerol is probably the result of the high affinity
interaction of albumin with this substrate, but no inde-
pendent measurements of the binding affinity were per-
formed. The effect of substrate depletion due to its
adsorption onto albumin was also demonstrated in a study
of the effect of human serum albumin (HSA) on the ‘‘in
vitro’’ enzyme kinetics of the formation of hydroxytolbu-
tamide on the tolbutamide hydroxylation examined using
human liver microsomes [33]. In this system, the addition
of HSA greatly decreased the unbound concentration of
tolbutamide in the incubation medium. The value of the
Michaelis constant KM for tolbutamide, even when the
unbound concentration of the substrate is considered,
decreased from 123 lM without HSA to 73 lM in the
presence of HSA at a concentration of 5 mg/mL. In this
study it was concluded that the addition of HSA to
microsomal incubation media may yield enzyme kinetic
estimates more comparable with ‘‘in vivo’’ results than
studies carried out in its absence. A similar study on the
effect of albumin on the phenytoin hydroxylation catalyzed
by human liver microsomes has been performed by Rowland
et al. [29]. In this study, both, either bovine serum albumin
or essentially fatty acid-free HSA reduced the KM values
(based on unbound substrate concentration) with only a
minor effect on the catalytic rate constant. A discussion is
also made on the albumin effect in the reaction under study
and in vitro–in vivo extrapolations.
At relatively high concentrations of albumin the ‘‘space
filling’’ or ‘‘crowding’’ effect must be also considered
[9, 26]. This effect could alter the enzyme conformation
leading to changes in its catalytic activity. Bergman and
Winzor [9] studied the effect of albumin on the reduction
of pyruvate by rabbit muscle lactate dehydrogenase. They
found that in the presence of albumin the catalysis was
enhanced as a consequence of an increase in the catalytic
rate constant with no appreciable effect on the Michaelis
constant for pyruvate or its co-factor. These authors
explained the results in terms of the ‘‘crowding effect’’
which provokes a change in the conformation of the
enzyme. Olsen [26] studied the kinetic properties of
hexokinase in concentrated protein solutions (BSA), (up to
4 mM). All results could be accounted for by a Michaelis–
Menten’s approach and both, KM and kcat decreased with
increasing albumin concentration. The decrease in KM with
increasing protein concentration was ascribed to an
increase in the ratio of activity coefficients between the
native enzyme and the enzyme–substrate complex. The
decrease in kcat with increasing protein concentrations was
also explained in terms of the ‘‘crowding effect’’, which in
this case leads to conformational changes of the enzyme’s
that disfavor the catalytic step. Recently, we have reported
that human serum albumin (up to 0.2 mM) do not affect the
kinetic parameters ((KM) and kcat) of the reaction of
hydrolysis of N-glutaryl-L-phenylalanine p-nitroanilide
(GPNA) catalyzed by a-chymotrypsin when they are cal-
culated in terms of the unbound substrate concentration [3].
In the present work we report the results of a study on
the effect of human serum albumin on the hydrolysis of the
synthetic substrate ((NAG)3-MUF); Scheme 1) catalyzed
by egg white lysozyme. The possible occurrence of HSA-
substrate and HSA–lysozyme interactions have also been
investigated, as well as the ‘‘crowding effect’’, were taken
into consideration.
2 Experimental Section
2.1 Chemicals and Equipments
Hen egg white lysozyme (Sigma), human serum albumin
(HSA), essentially fatty acid free (Sigma), 4-methylumbel-
liferyl-b-D-N-N0-N00 triacetylchitotrioside ((NAG)3-MUF)
(Sigma), 4-methylumbelliferone (Sigma) and dansyl deriv-
atives, dansylsarcosine (DS) and dansylsulfonamide (DNSA)
(Sigma) were used as received. Ultrapure water obtained from a
Modulab Type II equipment was employed to prepare all the
solutions.
Scheme 1 Structure of 4-methylumbelliferyl-b-D-N-N0-N00 triacetyl-
chitotrioside ((NAG)3-MUF)
368 C. Calderon et al.
123
Absorption spectra and absorbances were recorded in a
Hewlett–Packard UV–visible 8453 spectrometer. Fluores-
cence measurements were recorded in an Aminco-Bowman
fluorescence spectrometer. Light scattering measurements
were done using a Zetasizer Nano S-590 (Malvern Instru-
ments). Microcentrifugations were carried out using Mil-
lipore Ultrafree-MC centrifugal filter units (molecular
weight cut off equal to 30 kDa).
2.2 Lysozyme Activity Assays
The enzyme activity was measured at pH 5.2 (50 mM
citrate buffer) and 37 �C, employing the synthetic substrate
((NAG)3-MUF). The reaction was stopped at different
times by the addition of glycine buffer (pH 11). The rate of
the catalyzed reaction was determined by following the
formation of the fluorescent product 4-methylumbellyfer-
one (kex = 360 nm; kem = 445 nm) as a function of time.
The amount of product liberated was determined using a
linear calibration curve.
2.3 Dansyl Derivatives Displacement Experiments
The possible association of (NAG)3-MUF with HSA was
evaluated by the microcentrifugation experiments (see
Sect. 2.4) an from displacement of dansyl derivatives
(dansylsulfonamide and dansylsarcosine) which are spe-
cific markers of the main binding sites present in HSA
(sites I and II, respectively). Dansyl derivatives are highly
fluorescent when they are incorporated in the hydrophobic
sites of HSA and are almost non-fluorescent in an aqueous
environment. The possible displacement of the probes was
then determined by registering the fluorescence intensity
(kex = 360 nm; kem = 475 nm) of solutions of HSA
incubated in the presence of the considered dansyl deriv-
ative, before and after addition of (NAG)3-MUF.
2.4 Microcentrifugation Experiments
Microcentrifugation experiments were carried out at 37 �C,
pH 5.2 (50 mM citrate buffer), using Millipore Ultrafree-
MC centrifugal filter units. The molecular weight cut off of
the filter employed was 30,000 Dalton. Calibrations with
respect to filtrate volume passed through the membrane
were made as a function of the centrifugation time and
speed. Control measurements were carried out to ensure
that the molecule being monitored (the substrate or the
enzyme) was not retained by the membrane.
To examine the modes of interaction between HSA and
lysozyme the HEX v.6.3 [28] molecular docking program
was used. The structures 1E78 and 1UCO, both obtained
from Protein data Bank, were used for HSA and Lysozyme,
respectively [5, 25]. The protein surface shape and
electrostatic charges were modeled using 3D spherical
polar basis functions. This allows performing global rota-
tional and translational space scans by using fast Fourier
transformations. Structures of 500 complexes HSA–Lyso-
zyme were ranked by energy criteria according to steric
and electrostatic surface interactions.
2.5 Molecular Dynamic Calculations
To evaluate the protein–protein interaction as a function of
time, the five HSA–Lysozyme complexes with the lowest
energies obtained from molecular docking were optimized
by molecular dynamics. Each HSA–Lysozyme complex
was solvated with 22,853 water molecules and six sodium
ions to neutralize the charge, and introduced in a box
10.0 9 8.0 9 10.0 nm3 with periodic boundary conditions.
The simulations were performed using the software pack-
age GROMACS v 4.0 [8, 19, 22, 31]. For the visualization
of molecular graphics and trajectories, the program VMD
[21] was employed. HSA and lysozyme were represented
using the GROMOS96 43a2 force field [30], with SPC
water [7]. LINCS [18] was used to constrain the bond
lengths of the proteins and SETTLE [23] to restrict the
structure of the water molecules. A 1 nm cut-off was used
for the Lennard-Jones potential and the real space elec-
trostatic interactions. Long range electrostatic interactions
were calculated using PME [15]. The neighbor list was
updated every 10 time steps. All simulations were per-
formed in an isothermic-isobaric ensemble. To maintain
the temperature and pressure constant at 300 K, and 1 bar,
the weak coupling algorithm of Berendsen [6] was used,
with time constants of 0.1 and 1 ps, respectively. The time
step size in all simulations was 3 fs.
3 Results and Discussion
Figure 1a shows the results obtained for the effect of HSA
on the rate of reaction versus (NAG)3-MUF profiles. The
data shown in this figure indicate that, at all substrate
concentrations considered, the presence of HSA decreases
the rate of the lysozyme catalyzed process and that, the
decrease depends on HSA concentration. An important
decrease in activity is observed at low concentrations of
HSA (20 lM), reaching an almost complete inactivation of
the enzyme at 100 lM HSA.
The curves shown in Fig. 1a correspond to the fitting of
the data to the Michaleis–Menten equation, from which the
values of the Michaelis constant (KM) and the catalytic rate
constant (kcat) were obtained. The results obtained show that
KM remains almost unmodified in the presence of albumin
(1 9 10-5 ± 2.5 9 10-6 M-1 in the absence of albumin
and 8.0 9 10-6 ± 1.2 9 10-6 at [HSA] = 100 lM) while
Effect of Human Serum Albumin on the Kinetics 369
123
(kcat) decreases with albumin concentration as shown in
Fig. 1b (by a factor ca. 10).
The possible interaction of the substrate with albumin
was assessed by microcentrifugation experiments. Results
obtained are shown in Fig. 2a. As can be seen in this figure,
the concentration of (NAG)3-MUF in the filtrand and the
filtrate remains almost the same, irrespective of HSA
concentration, indicating a lack of adsorption of the sub-
strate onto albumin.
In line with this result, dansyl derivatives displacement
experiments (Fig. 2b) shows no incorporation of (NAG)3-
MUF to the main binding sites present on HSA. It clearly
turns out the absence of substrate binding to albumin.
These results are quite satisfactory with the insensitivity of
KM upon the concentration of albumin and clearly indicate
that the effect of albumin on the activity of lysozyme is not
related to a deprivation of the substrate to the enzyme in
the presence of HSA.
The possible interaction between HSA and lysozyme is
an interesting alternative worth to be explored in order to
interpret the origin of the lysozyme inactivation by HSA.
Microcentrifugation experiments (Fig. 3) show a decrease
in the concentration of free lysozyme with the increase of
albumin concentration (being the enzyme almost totally
associated at 100 lM). The formation of a HSA–lysozyme
complex could induce conformational changes leading to
distortion of the active site of the enzyme leading a less
active conformation.
Fig. 1 a Effect of HSA on the reaction rate for the hydrolysis of
(NAG)3-MUF catalyzed by lysozyme (2 lM). Kinetic data fitted
according to the Michaelis–Menten equation. No HSA (filled circle);
20 lM (filled square); 50 lM (filled triangle); 80 lM (filled invertedtriangle); 100 lM (filled diamond). b Dependence of the catalytic rate
constant with the concentration of HSA for the hydrolysis of (NAG)3-
MUF catalyzed by hen egg white lysozyme
Fig. 2 a Concentration of ((NAG)3-MUF) in the filtrand (black bars)
and the filtrate (white bars), measured in microcentrifugation
experiments. b Effect of the addition of ((NAG)3-MUF) on the
fluorescence intensity of dansyl derivatives. ([Dansyl deriva-
tive] = 20 lM, [HSA] = 2 lM); DS as probe (open square); DNSA
as probe (filled circle)
370 C. Calderon et al.
123
The more relevant result of the present study is that the
decrease of free lysozyme and the catalytic rate constant
show the same dependence with albumin concentration as
shown in Fig. 3. We interpret this finding as indicative of
the presence in the system of two lysozyme populations:
the free enzyme catalyzing the reaction as in the absence of
albumin and the HSA-associated lysozyme which would be
completely inactive.
3.1 Theoretical Modeling
In order to obtain further insight on the molecular basis of the
lysozyme inactivation by HSA, theoretical modeling tools
were used. First, molecular docking calculations were per-
formed to model the formation of the complex between
lysozyme and the synthetic substrate (NAG)3-MUF (Fig. 4).
The modeling shows the incorporation of the substrate to the
enzyme across a cleft that runs through the enzyme molecule,
cleft in which the catalytic active site is located.
The HSA–lysozyme complexes, obtained by molecular
dynamics simulations, do not alter significantly their struc-
tures after 50 ns. MD energy profile indicates that the
interaction protein–protein is mainly electrostatic. Figure 5
shows the HSA–lysozyme complex with the lowest potential
energy after 49.5 ns of simulation. The structure of this
complex shows that the Glu35 and Asp52 residues located in
the active site of lysozyme are oriented toward the HSA
surface. This conformation would restrict the binding of the
substrate to the active site of lysozyme, leading to an inactive
enzyme population, in line with the experimental results.
Molecular docking calculations were performed
between (NAG)3-MUF and the solution for the HSA/
Lysozime complex obtained by molecular dynamics sim-
ulations. The results obtained (Fig. 6) demonstrate that the
incorporation of the substrate to the active site of the
enzyme is hindered by formation of the HSA/Lysozyme
complex, mainly due to an occlusion of the cleft in which
the active site is located.
Finally, to take into account the possibility of the
operation of a ‘‘crowding effect’’, we calculated the volume
0 20 40 60 80 1000,0
0,5
1,0
1,5
2,0
[HSA] (µM)
[Lys
ozym
e] ( µ
M)
0,0
3,0x10-4
6,0x10-4
9,0x10-4
1,2x10-3
1,5x10-3
1,8x10-3
kC
AT ( s
-1)
Fig. 3 Comparison between free lysozyme concentration (filledcircle) and the decrease on kcat (filled square) as a function of HSA
concentrationFig. 4 Complex formation between lysozyme and the synthetic
substrate (NAG)3-MUF, obtained by molecular docking calculations.
Front view (left panel); side view (right panel). The aminoacidic
residues Glu35 and Asp52 present in the active site of lysozyme are
shown as space filling models
Fig. 5 HSA (green)/Lysozyme
(orange) complex obtained by
molecular dynamic simulations.
Glu35 (purple) and Asp52
(blue) located at the active site
of lysozyme are represented
indicating their van der Waals
radii. For simplicity water
molecules and ions were
removed from the visualization.
Side view (left panel); upperview (right panel) (Color figure
online)
Effect of Human Serum Albumin on the Kinetics 371
123
occupied by albumin molecules. To do this calculation of
the hydrodynamic radius of HSA (6 nm approx.) was
determined employing dynamic light scattering measure-
ments (Fig. 7). The results obtained shows that even at the
highest albumin concentration used in the lysozyme
activity studies (100 lM HSA), the volume occupied by
the albumin molecules slightly exceeds 1% of the total
solution volume. This is by far lower than those presented
in several works on the subject were macromolecular
crowding effect is operative. Furthermore, the distances
between albumin molecules are 3–4 times larger than the
lysozyme molecules diameter (approx. 3 nm), ruling out a
physical effect on the enzyme structure due to an excluded
volume effect. According with these results, the effect of
macromolecular crowding can be neglected under the
conditions employed in the precedent work.
4 Conclusions
The presence of HSA (up to 100 lM) leads to a decrease in
the rate of hydrolysis of 4-methylumbelliferyl-b-D-N-N0-N00
triacetylchitotrioside catalyzed by hen egg white lysozyme.
The reaction follows a Michaelis–Menten mechanism under
all the conditions employed. The decrease in the rate of
hydrolysis is related to a decrease in the catalytic rate
constant (kcat) with albumin concentration rather than a
modification of the Michaelis constant (KM). The decrease
of kcat arises from the occurrence of HSA–lysozyme inter-
actions leading to an inactivation of the enzyme, so
decreasing the concentration of active enzyme available for
catalysis. These results are relevant, since as far as we
know, the effect on the kinetic parameters of complex for-
mation between albumin and enzymes has not been
explicitly demonstrated. No depletion of the substrate by
albumin or macromolecular crowding effects is observed.
Acknowledgments Thanks are given to Dicyt (USACH) and
Fondecyt (Grant No. 1095036) for financial support. C. Calderon
acknowledges CONICYT for a doctoral fellowship.
References
1. Abuin E, Aspee A, Lissi E, Leon L (2007) J Chil Chem Soc
52:1196–1197
2. Abuin E, Calderon C, Lissi E (2008) J Photochem Photobiol A
Chem 195:295–300
3. Abuin E, Lissi E, Ahumada M, Calderon C (2011) Protein J
30:143–147
4. Awad–Elkarim A, Means GE (1988) Comp Biochem Physiol
91B:262–272
5. Bhattacharya AA, Curry S, Franks NP (2000) J Biol Chem
275:38731–38738
6. Berendsen HJC, Postma JPM, van Gunsteren WF, Dinola A,
Haak JR (1984) J. Chem Phys 81:3684–3690
7. Berendsen HJC, Postma JPM, van Gunsteren WE, Hermans J
(1981) Interaction models for water in relation to protein
hydration. In: Pullman B (ed) Intermolecular Forces. Reidel,
Dordrecht, pp 331–342
Fig. 6 Interaction of the
synthetic substrate (NAG)3-
MUF with the HSA/Lysozyme
complex obtained by molecular
docking calculations. Global
view of the complex (leftpanel); Zoomed view of the
complex, showing the
surroundings of the occluded
active site of the enzyme (rightpanel)
0,1 1 10 1000
5
10
15
20
25
30
35
Inte
nsi
ty (
%)
Diameter (nm)
Fig. 7 Size distribution by intensity obtained by dynamic light
scattering measurements. HSA (black bars); lysozyme (white bars).
The measurements of each protein were preformed independently, but
for simplicity the distributions are shown in the same graph
372 C. Calderon et al.
123
8. Berendsen HJC, van der Spoel D, van Drunen R (1995) Comp
Phys Comm 91:43–56
9. Bergman DA, Winzor DJ (1989) Eur J Biochem 185:91–97
10. Bordbar A-K, Taheri-Kafrani A (2007) Colloid Surf B Bioin-
terfaces 55:84–89
11. Carter D, Ho JX (1994) Adv Protein Chem 45:153–203
12. de Wolf FA, Brett GM (2000) Pharm Review 52:207–236
13. Dıaz X, Abuin E, Lissi E (2003) J Photochem Photobiol A Chem
155:157–162
14. Epps DE, Raub TJ, Caiolfa V, Chiari A, Zamai M (1999) J Pharm
Pharmacol 51:41–48
15. Essman U, Perela L, Berkowitz ML, Darden T, Lee H, Pedersen
LG (1995) J Chem Phys 103:8577–8592
16. Gelamo EL, Silva CHTP, Imasato H, Tabak M (2002) Biochim
Biophys Acta 1594:84–99
17. Hajduk PJ, Mendoza R, Petros AM, Huth JR, Bures M, Fesik SW,
Martin YC (2003) J Comput Aided Mol Des 17:93–102 (and
references cited therein)
18. Hess B, Bekker H, Berendsen HJC, Fraaije JGEM (1997)
J Comput Chem 18:1463–1472
19. Hess B, Kutzner C, van der Spoel D, Lindahl E (2008) J Chem
Theory Comput 4:435–447
20. Hongwei Z, Min G, Zhaoxia Z, Wenfeng W, Guozhong W (2006)
Spectrochim Acta Part A 65:811–817
21. Humphrey W, Dalke A, Schulten K (1996) J. Mol Graph 14:33–38
22. Lindahl E, Hess B, Van der Spoel E (2001) J Mol Model
7:306–317
23. Miyamoto S, Kollman PA (1992) J Comput Chem 13:952–962
24. Moosavi-Movahedi Z, Safarian S, Zahedi M, Sadeghi M, Saboury
AA, Chamani J, Bahrami H, Ashraf-Modarres A, Moosavi-
Movahedi AA (2006) Protein J. 25:193–201
25. Nagendra HG, Sudarsanakumar C, Vijayan M (1996) Acta
Crystall Sect D52:1067–1074
26. Olsen SN (2006) Thermochim Acta 448:12–18
27. Peters TJ (1995) All about albumin: biochemistry, genetics and
medical applications. Academic Press, New York
28. Ritchie DW, Venkatraman V (2010) Bioinformatics 26:
2398–2405
29. Rowland A, Elliot DJ, Knights KM, Mackenzie PI, Miners JO
(2008) Drug Metab Dispos 36:870–877
30. Scott WRP, Hunenberger PH, Tironi IG, Mark AE, Billeter SR,
Fennen J, Torda JAE, Huber T, Kruger P, van Gunsteren WF
(1999) J Phys Chem A 103:3596–3607
31. Van der Spoel D, Lindahl E, Hess B, Groenhof G, Mark AE,
Berendsen H (2005) J Comput Chem 26:1701–1718
32. Wang C-S, Bass H, Whitmer R, McConathy WJ (1993) J Lipid
Res 34:2091–2095
33. Wang J-S, Wen X, Backman JT, Neuvonen PJ (2002) J Phar-
macol Exp Ther 302:43–49
Effect of Human Serum Albumin on the Kinetics 373
123