CLINICAL CHEMISTRY, Vol. 19, No. 4, (1973) 384-386
TurbidimetricProcedure for Determinationof Lipase Activity
Angelo Burlina and Lauro Galzigna
A method is described for turbidimetric assay of Iipase activity. The substrate is a two-phase emulsion
composed of a fairly homogenous dispersion of micelles of either triolein or olive oil. The de-emulsification of the substrate after lipase addition is a true
measure of the enzymatic activity. Turbidity change
is related to amount of the enzyme by use of a commercial standard, purified lipase from hog pancreas.
The quantitative
relationship
between
turbidity
change and the number of micelles provides the
theoretical basis for this type of assay.
Louis, Mo. 63178), was dissolved in 4.9 ml of 2ethoxyethanol (Fluka AG, Buchs, Switzerland) and
1 ml of this solution was mixed with 9.0 ml of “universal” buffer (pH 9.15). The buffer (9) was prepared
by adding 75 ml of 0.2 molar sodium hydroxide
to
100 ml of a mixture 28.57 millimolar
with respect to
boric
acid,
citric
acid monohydrate,
bituric acid, and potassium
(Merck, Darmstadt).
5,5’-diethylbar-
dihydrogeh
phosphate
Olive oil. Highly purified
olive oil (Sigma),
1 ml,
dissolved in 99 ml of 2-ethoxyethanol,
and 0.5
ml of the solution was mixed with 9.5 ml of universal
buffer (pH 9.15). The mixture was emulsified
by mixing with a vortex-type
mixer for 1 mm.
Substrates
so prepared were stored at 5#{176}C,
being
brought to 25#{176}C
before use. Two different concentrations of triolein substrate and olive oil substrate were
used in the standardized
assay in order to have, in
either case, an approximately
similar initial value of
was
Additional Keyphrases: triolein or olive oil as substrates
porcine lipase as standard #{149}normal values
The enzymatic
activity of lipase (glycerol ester hyEC 3.1.1.3) in blood serum has been measured by different authors with colorimetric
(1), fluorometric (2), turbidimetric
(3), and titrimetric
(4)
procedures, and with use of different substrates.
True lipase activity
is considered as a special case
of ester hydrolysis,
which takes place at the interface
of the water-lipid
system (5, 6) constituted by triglyceride molecules dispersed in an aqueous phase.
drolase,
This statement
implies
some uncertainty
as to the
kinetic meaning of the results obtained with methods of lipase assay based solely on the determination
of the final products
of lipase-catalyzed
reaction,
such as measurement
of liberated fatty acid. In such
cases, in fact, the substrate for lipase activity
is considered to be the individual
triglyceride
molecules
and not their multimolecular
aggregates, called “micelles.”
In contrast,
turbidimetric
methods
seem
closer to the physiological
reality,
albeit
their
standardization
and quantitative
reliability
are rather controversial.
We describe here the characteristics
of a simple
turbidimetric
procedure, which, as is also true of another (7, 8), is based on the de-emulsification
of a
water-lipid system. In addition, we attempt to set a
minimal theoretical
background
for the standardization of turbidimetric
assays of lipase, based on a
thorough characterization
of the substrate (8).
Materials and Methods
Substrates
Triolein,
From
the
0.1 ml (grade II; Sigma Chemical Co., St.
General
Hospital
of Conegliano,
31015
Italy; and the Department
of Medical Biochemistry,
Nairobi, P.O. Box 30197, Nairobi, Kenya.
Received Sept. 21, 1972; accepted Jan. 4, 1973.
384
CLINICAL CHEMISTRY, Vol. 19. No. 4, 1973
Conegliano,
University
of
absorbance.
Procedure
Two milliliters
of the emulsified substrate was
mixed with 50 zl of a solution (6 sg/ml) of porcine lipase, and the decrease in absorbance at 546 nm was
followed for 3-5 mm at 25#{176}C
with an Eppendorf recording photometer.
The absorbance change per
minute (M/min)
was calculated from the slope of
the curve taken 1 mm after the initial point.
Standardization
Purified hog pancreas lipase (labeled activity, 190
U/mg; Worthington Biochemical Corp., Freehold, N.
J. 07728) was used for the experiments. It was dissolved in saline [8.5 g of NaC1 and 60 g of albumin
(Fraction V, Sigma) per liter] or in human serum
that had been inactivated
by warming for 3 h at
56#{176}C.
We checked the activity of hog lipase with our
substrate (triolein) by using the titrimetric
method
reported by Varley (10) and found that 1 mg of enzyme released 200 ± 30 smol of fatty acid per minute.
The homogeneity of micelle size of the substrate
was assessed by ultrafiltration
through a “Swinner
13 Filter Unit”
(Millipore
Corp., Bedford, Mass.
01730) and ultra#{235}entrifugationin a Model L Ultracentrifuge
(swinging bucket SW 39L; BeckmanSpinco Div., Palo Alto, Calif. 94304).
The factor for converting the absorbance change
into U/liter is 2230 with triolein as substrate and
2000 with olive oil as substrate.
on the U/liter
value measured
each amount of enzyme.
This factor is based
titrimetrically
for
Table 1. Stability of Substrate Emulsion of Triolein
and Olive Oila
Results
Preliminary
experiments showed that the most
suitable
surface-active
stabilizing
agent for the
emulsion was 2-ethoxyethanol, and activity was optimal in universal buffer. An optimal and convenient
wavelength for measuring the scattering of light is at
546 nm. At lower wavelengths
the measurement
would be disturbed by the absorption of the components of the incubation mixture. The substrate remained stable both during single experiments and in
the course of a week. Moreover, different batches of
a substrate treated with the same enzyme solution
gave similar values for absorbance changes. Table 1
summarizes the evidence regarding the stability of
the substrate.
Ultrafiltration
experiments showed that the substrate is an emulsion of triolein micelles having an
average diameter of 100 ± 20 nm; Figure 1 shows the
ultracentrifugal
behavior of the micelles. Sedimentation and rotation velocity were logarithmically
related, indicating that the size of the micelles is homogeneous. With olive oil as substrate, an analogous
behavior was found.
The pH-activity
curve obtained with porcine lipase dissolved in the albumin-saline
medium shows
the optimal pH of the reaction to be between 9 and
10 for both triolein and olive oil.
Experiments
at different
temperatures
showed
that with triolein substrate, Arrhenius’ law is obeyed
from 19#{176}C
on and a Qio value of 2.3 is obtained from
4#{176}
to 19#{176}C,
whereas a Qio value of 1.3 results at temperatures higher than 19#{176}C.
With olive oil, a Qio of
150
(,)
0
40
too;
N
a.
30
20
50
a
a’
S
(A)
Olive oil
1.777 ± 0.131
1.225 ± 0.066
(CV, 5.3%)
(CV, 7.4%)
Batch stability
(A/min)
0.045 ± 0.005
(CV, 3.1%)
0.060 ± 0.007
(CV, 4.2%)
a As a measure of time stability, the total absorbance (A) was measured at 546 nm during one week at intervals of 6 h. The average values
of all measurements taken in one week are given for one batch of tnolein and for one batch of olive oil, with no enzyme. The batch stability
was measured by reacting hog lipase (6 zg/ml) in saline-albumin solution with 10 different batches of substrate and the change in absorbance
(A/min)
recorded. Each batch contained either triolein, 2 X i0
mol/liter, or olive oil, 1 9/liter.
2.6 was calculated from 10#{176}
to 25#{176}C,
but Arrhenius’
law apparently was not obeyed.
Figure 2 shows that the relation between amount
of enzyme and change in turbidity
is linear for low
concentrations of enzyme. The conversion factor for
calculating the equivalence between standard International Units (U) and the absorbance changes
(M/min)
has been obtained from data in Figure 3.
Figure 31 illustrates the effect of different concentrations of substrate (triolein) on the velocity of the
reaction and the double reciprocal plot from which
an apparent Km value of 9 x 102 mol/liter and a
Vmax
value of 0.05 are calculated. With olive oil as
the substrate, a similar diphasic curve is obtained
which follows a Michaelis-Menten
behavior. The apparent Km value found with olive oil is 2.6 x iO
g/liter and the Vmax is 0.065. The turbidimetric
reaction shows pseudo-first-order
kinetics, but the
initial velocity is easily calculated from the linear
initial portion of the curve relating absorbance and
time.
The specificity of our procedure for lipase has been
proved by experiments with ‘aliesterase (carboxylic
ester hydrolase, EC 3.1.1.1.; Boehringer, Mannheim,
Germany)
and cholinesterase
(acylcholinacyl
bydrolase, EC 3.1.1.8; Worthington),
neither of which
produced any decrease of substrate turbidity. Sodium
desoxycholate (35 g/liter) has a strong activating
effect (3, 8); it increases the enzymatic activity by
about 250%.
The results obtained with plasma from a number
of normal individuals are summarized in Table 2.
,
0
U
0
Time stability
Triolein
E
Discussion
(0
10
Detachment of a fatty acid residue as a result of
lipase activity creates a center capable of electrostatic interaction within the micelle and also forms a
soap-like substance, with consequent effect on the
C
V
U
Absorbanc.
of fluid
sup.rnatant
Fig. 1. Ultracentrifugation
3 ml of emulsion (2.26 X 1O’
mol/liter
at
545
nm
of the substrate
tniolein) is centrifuged
for 8
mm
at different velocities and the amount of micelles sedimented is estimated by measuring the residual turbidity of the supernatant fluid at 546 nm
after each centrifugation
1 Since
lipase activity
is not linear except for the first few minutes (3, 8) then the units (mol/min)
of lipase will depend on the
time chosen for titration;
the shorter
the time is, the higher will
be the units. This consequently
will affect the units given to the
turbidimetric
method
with this kind of approach
to standardization.
CLINICAL
CHEMISTRY,
Vol. 19, No.4.1973
385
Table 2. Lipase Activityof Plasma from 100
Normal Subjects, as Measured
Procedure a
TrIolein
OlIve oil
U/liter
Mean
3
±2.8
0-12
1.9
±1.7
SD
Range
by This
0-8
Activity Is expressed as Mmol of triolein (or olive oil) hydrolyzed/mm
per liter of plasma, at 25#{176}C.
a
LIPASE
(mg)
Fig. 2. Effect of enzyme concentration on the velocity
the reaction
The substrate triolein is 2.26 X i0
25#{176}C
and pH 9.15
mol/liter
of
and olive oil 1 g/liter at
0
#{176}
U
S
a
4
TRI0.IIN
Fig. 3. Effect of substrate (triolein) concentration on the
velocity
ing an average volume of 5 x 10 cm3 for each supposedly spherical particle. If a volume of 10 x 10-i
cm3 is assumed for a molecule of triolein in hydrated
form, we calculate that each micelle
is formed
by
about
50 molecules
of triolein.
This, from a practical viewpoint, should allow turbidity changes (M/min)
to be converted into number of micelles broken down, and consequently into
number of molecules hydrolyzed by the enzyme.
From Table 1 data we can calculate that a change in
concentration of triolein corresponding to 1 zmol/
liter is related to a &4/min of about 0.0009. On the
other hand, from Figure 2 we see that one titrimetric
unit (U) corresponds
to a A/min
of about 0.004, and
this implies a difference of 4-5 fold. This apparent
discrepancy in micromoles hydrolyzed per mm (i.e.,
U) calculated from turbidimetric
assays and from direct conversions of M/min
to zmol/min
can be related to the nonlinearity of the titrimetric
assays of
lipase, which tend to yield a value for the sample that
is less than it should be.
References
of the reaction
A fixed amount of hog pancreas lipase (8 g/m)
in saline-albumin
used at 25#{176}C
and pH 9.15. The apparent Km value is 9 X 10-2 mol/liter
is
1. Popiela,
human
interfacial tension. Both events should lead to a collapse of the micellar architecture, resulting macroscopically in a de-emulsification
of the substrate.
We found that a treatment developed to study gel
flocculation (11) can also be used for our system, and
a relationship between turbidity
changes and number of micelles of the emulsion can be assumed. The
substrate used in the present study is a fairly isodisperse system, i.e., the size of the particles and their
state of aggregation are constant. This permits the
conversion of turbidity
changes to number of mi-
T.,
gastric
Szafran,
H.,
lipase
towards
Chim. Acta 11,283(1965).
2. Guilbault,
ss,
Oxford,
G. G., Enzymatic
1970, p 64.
and
Szafran,
Z.,
para-nitrophenyl
Methods
The activity
esters.
of Analysis,
of
Clin.
Pergamon
3. Vogel, W. C., and Zieve, L., A rapid and sensitive turbidimet-
nc method for serum lipase based upon differences between the
lipases of normal and pancreatitis
serum. Clin. Chem. 9, 168
(1963).
4. Yang, Y. S., and Biggs, H. G., Rapid, reliable method for
suring serum lipase activity. Clin. Chem. 17,512(1971).
5. Desnuelle, P., Rebound, J. P., and Ben Abdeljlil, A., Influence
of the composition of the diet on the enzyme content of rat pancreas. In Ciba Found. Symp. on the Exocrine Pancreas, Churchill, London, 1962, p 90.
6. Benzonana, G., and Desnuelle, P., Action of some effectors on
celles, according
to the expression
(11):
the hydrolysis
of long-chain
triglycerides
by pancreatic
lipase.
Biochim. Biophys. Acta 165,47(1968).
A = k.n ‘V2 (1 + at/T)
(1)
7. Burlina, A., and Gaizigna, L., Preliminary
report at Ill International Symposium on Clinical Enzymology, Conegliano Veneto,
where A is the absorbance, k is a constant,
n is the
November 26-28, 1971, Proceedings, p 516.
number of micelles, v is the volume of one micelle,
8. Shihabi, Z. K., and Bishop, C., Simplified turbidimetnic assay
and (1 + at/T)
is a time constant. The para- 7forlipase
activity.
Clin. Chem. 17, 1150 (1971).
bolic relationship found preliminarily
between substrate (triolein or olive oil) concentration
and absorbance justifies the use of the above expression.
The average micelle diameter is 100 nm, indicat-
388
CLINICAL CHEMISTRY, Vol. 19, No.4,
1973
9. Long, C., Biochemist’s Handbook,
10. Varley,
H.,
Practical
Spon, London, 1961, p 41.
Clinical Biochemistry,
Medical Books Ltd., London, 1969, p 408.
11. Troelstra, S. A., and Kruyt, H. R., In Colloid
Kruyt, Ed.
Elsevier,
Amsterdam,
1952, p 83.
W. Heinemann
Science,
1, H. R.