CLIN.CHEM.25/8, 1406-1410(1979)
Human Amylase Isoenzymes Separated on Concanavalin
A-Sepharose
Toshiyuki Takeuchi
Human salivary amylase and pancreatic amylase were
purified and characterized. These amylases gave two
bands and one band, respectively, each staining for both
protein and sugar, after electrophoresis on sodium dodecyl
sulfate-polyacrylamide gel. The relative molecular mass
(Mr) of pancreatic amylase was calculated to be 60 000;
for the two components (A and B) of salivary amylase the
bohydrate contents. The characteristics of the purified
preparations are also compared with those reported by Keller’s group (6, 7), who reported two forms of salivary amyl.ase,
with relative molecular masses (Mi) of 56000 and 62 000, and
one form of pancreatic amylase, with an Mr of 54000. My data
indicate that the salivary amylase that binds to concanavalin
A is that with the higher Mr.
Mr were 61 000 and 64 000. The two salivary amylases
were separated by chromatography
on concanavalin
A-Sepharose;
only component B bound to concanavalin
Materials and Methods
A. The carbohydrate content of pancreatic amylase was
1.61 ± 1.02% (SD),andofsalivaryamylasesAandB2.18
± 0.71% and 8.77 ± 2.28%, respectively. The salivary
and pancreatic amylases had completely identical antigenicities against antibody to either. On isoelectric fo-
cusing, pancreatic amylase showed one peak at pH 7.0,
salivary amylase A showed a major peak at pH 6.4 with a
trace of material at pH 5.9, and salivary amylase B a major
peak at pH 5.9 and one minor peak at pH 6.4. Serum
amylase was separated into two major peaks with isoelectric points (p1) of 6.4 and 7.0, respectively, and one
minor peak, with a p1of 5.9. Only a small part of the serum
amylase with a p1 of 5.9 combined with concanavalin A;
the two other serum amylases did not.
AddItIonal Keyphrases:
isoelectric
focusing
electrophoresis
immunodiffusion
specific
activities
diagnosis of hyperamylasemia
characterization of isoenzymes
Michaelis constants
.
Human salivary amylase and pancreatic amylase have been
separated by electrophoresis on a cellulose acetate membrane
(1), by polyacrylamide gel isoelectric focusing (2), by chromatography on DEAE-cellulose (3), and by a method in which
a salivary amylase inhibitor
from wheat is used (4). Human
serum contains two types of amylase, corresponding
in electrophoretic mobility to salivary and pancreatic amylase (1-4),
and their measurement in serum has been used clinically in
differential
diagnosis of hyperamylasemia
(1-4), but their
determination
is not always helpful in differential
diagnosis
of this condition
(5). Markers other than the ionic charge
difference of the two amylase isoenzymes are required to
identify the origin of amylase, and to find suitable markers
it is essential to determine the molecular properties of purified
human salivary and pancreatic amylase.
This paper reports comparative studies on human salivary
and pancreatic amylase with respect to their specific enzyme
activities, Michaelis constants, molecular weights, isoelectric
points, antigenicities,
affinities
for concanavalin A, and car-
Biochemistry Division, National Cancer Center Research Institute,
Tsukiji, 5-chome, Chuo-ku, Tokyo 104, Japan.
Received May 7, 1979; accepted May 30, 1979.
1406 CLINICALCHEMISTRY,Vol. 25, No. 8, 1979
Samples
We collected
specimens
of serum and saliva from appar-
ently healthy volunteers in our laboratory, and pancreatic
juice was collected by cannulation of the pancreatic duct of
three patients who were undergoing pancreaticoduodenectomy at Keio University Hospital. Specimens of parotid gland
and pancreas were obtained at operation or autopsy.
Assays
Amylase activity was measured as described by Ceska et al.
(8), with Blue Starch (Pharmacia Co., Uppsala, Sweden) as
a substrate, and expressed in international units (mt.units),
as described in the brochure supplied with the Blue Starch.
Protein was measured by the method of Lowry et a!. (9), with
bovine serum albumin as the standard.
Measurement
of Michaelis
Constants
For kinetic studies, seven different substrate concentrations
were used and the Michaelis constants determined by the
method of Lineweaver and Burk. The molarity of the Blue
Starch, a polymer of glucose, could not be determined, so its
concentration is expressed in grams per liter.
Isoelectric
Focusing
For isoelectric focusing we used a 110-mL column (LKB
8100-10; LKB Instruments, Stockholm, Sweden) with pH 5-8
carrier ampholyte (LKB Instruments) as described previously
(10). The density gradient was formed with 0-50% (by vol)
glycerol and 0.5-1.5% (by vol) carrier ampholyte. Specimens
were inserted into the column near the middle layer. Isoelectric focusing was done at 4 #{176}C
for 48 h at 400-900 V.
Preparation
of Human Amylases
There are many reports on the purification of human salivary and pancreatic amylases (7, 11, 12), but we devised a new
method for their purification. For salivary amylase from the
parotid gland and pancreatic amylase from the pancreas, the
steps used were centrifugation of the tissue homogenate at
105 000 X g for 60 mm, fractionation in 20 to 45% saturated
ammonium sulfate, affinity chromatography on Sephadex
G-100, anion-exchange
and affinity
chromatography
on DEAE-cellulose,
chromatography on concanavalin A-Sepharose
4B. Chromatographies
on Sephadex G-100 and Sepharose 4B
Table 1. PurIfication of Human Amylases
Protein,
Acty,
bit, units
From human
parotki
SpecIfic acty,
mt. unltslmg
mg
gland
Homogenate
18300
673
27.2
105 000 X 9 supemate
(NH4)2S0420-45%
Sephadex 6-100
14560
410
35.5
12020
5840
167
20.0
14.2
3720
790
8.7
1.8
6120
DEAE-cellulose
72.0
306
411
Concanavalin A-Sepharose
pass-through fraction
adsorbed fraction
From human pancreas
13200
10400
6360
3800
3010
Homogenate
105 000 X g supernate
(NH4)2SO420-45%
Sephadex
6-100
DEAE-cellulose
Concanavalin
416
981
577
240
13.5
15.0
18.0
26.5
163
201
10.6
231
23.4
A-Sepharose
pass-through fraction
act as affinity
428
2450
chromatographies,
not as gel filtrations,
for
amylases (13). Elution of amylase was retarded
on Sephadex 0-100 column chromatography. On chroma-
Results
mammalian
tography on a column of concanavalin A-Sepharose 4B the
unadsorbed fraction of amylase was also retarded. For preparation of salivary amylase from saliva, the saliva was first
acidified to pH 4.0 with dilute acetic acid (1 mol/L), to reduce
the amount of mucin-like material; this step was indispensable
for concentration of the saliva. The concentrated saliva was
then applied to Sephadex G-100 as described above. For
preparation
of pancreatic
amylase from pancreatic juice,
fractionation with ammonium sulfate was performed before
Sephadex G-100 chromatography.
For partial purification of
serum arnylase, the preparation was dialyzed against distilled
water to reduce the amount of immunoglobulin,
and material
precipitated with ammonium sulfate was subjected to isoelectric focusing. The fractions of amylase separated were then
subjected
to chromatography
on a column
Purification
of Salivary and Pancreatic
Amylases
Amylases were purified from saliva (three cases), parotid
gland (15 cases), pancreatic juice (three cases), and pancreas
(12 cases). Table 1 summarizes the purification of salivary
amylase from parotid gland and of pancreatic amylase from
a pancreas. Before the step of concanavalin A-Sepharose
chromatography, salivary and pancreatic amylases gave two
bands and one band, respectively, on SDS-polyacrylamide
gel
electrophoresis, and these bands all stained both with Coomassie Brilliant Blue and periodic acid-Schiff
reagent (Figure
I
of concanavalin
A-Sepharose. Protein adsorbed by the concanavalin ASepharose was eluted with a linear gradient of 0-0.1 mol/L
a-methylmannoside
(Sigma Chemical Co., St. Louis, MO
63178).
Preparation
of Antisera
Antisera against purified human salivary and pancreatic
amylases (obtained from chromatography
on DEAE-cellulose)
were prepared in rabbits as described previously (14). The IgG
fractions of antisera were separated by ammonium sulfate
precipitation
and DEAE-cellulose
column chromatography.
Double diffusion was performed by the method of Ouchterlony (15).
Sodium Dodecyl Sulfate-Polyacrylamide
Electrophoresis
Slab
gel electrophoresis
on sodium
Gel
dodecyl
L
sulfate
(SDS)-polyacrylamide
(2 mm X 12 cm) containing
7.5%
acrylamide was done by the method of Laemmli (16). After
electrophoresis, gels were stained for protein with Coomassie
Brilliant
Blue or for carbohydrate
with periodic acid-Schiff
reagents (17).
Carbohydrate Determination
Neutral carbohydrates were determined by the phenolsulfuric acid method (18), with glucose as standard.
Fig. 1. SDS-polyacrylamide
gel electrophoresis
of human am-
ylases
Human amylases obtained from DEAE-cellulose were subjectedto electrophoresis. About 10-and 2O0-tg samplesof amylaseswereusedin gels stained
with Coomassie BrilliantBlueandperiodicacld-Schlff reagent(lTj, respectively.
Electrophoresis was In 7.5% polyacrylamide running gel by the method of Laemmii (16). 1, 3: salIvary amylase. 2, 4: pancreatic amylase. 1, 2: Coomassle
Brilliant Blue staIning. 3, 4: perIodic acid-Schlff staining
CLINICAL CHEMISTRY, Vol. 25, No. 8, 1979
1407
I
I’
U
I
Fraction number
Fig. 2. Affinity chromatography on concanavalin A-Sepharose
and SDS-polyacrylamide gel electrophoresis
E
40-
Salivaryamylasesobtainedfrom DEAE-celluiose.84 mt. units in a total volume
of 1 mL, were appliedto a 1.0 X 5.0cm columnof concanavalln A-Sepharose.
The elution buffer was Tris#{149}HCI
(10 mmoi/L, pH 7.4) containing, per liter, 0.15
mel of NaCI, 1 mmoi of CaCI2 and 1 mmol of MgCI2. The adsorbedamylasewas
eiuted with a linear gradientof 0-0.1 mol/L a-methylmannoside In the above
buffer. The unadsorbed and adsorbed fractions of amylasewereelectrophoresed
in SDS-poiyacrylamide gel as described for Figra’e 1
JNSNUS
55
60
6.5
7.0
7.5
pH
1). The salivary amylase with the larger relative molecular
mass was more strongly stained than were the other two am-
Fig. 3. Isoelectric focusing of human amylases
Isoelectric focusing was done as described in the Materials and Methods.
ylases with periodic acid-Schiff reagent, perhaps owing to its
higher carbohydrate content. Relative molecular masses of
60 000, 61 000, and 64 000 were obtained for the pancreatic
amylase and the two salivary amylases, respectively.
In the next step of concanavalin A-Sepharose chromatography the salivary amylases were separated into two fractions.
The unadsorbed and adsorbed fractions of salivary amylase
each gave a single band on SDS-polyacrylamide
gel electrophoresis, corresponding Mr of 61 000 and 64 000, respectively
(Figure 2). The ratio of unadsorbed amylase (salivary amylase
A) to adsorbed amylase (salivary amylase B) was approximately 4, and the ratios in saliva and the parotid gland were
similar. Incubation of a homogenate of parotid gland for 3 h
at 37 #{176}C
did not change the ratio of the two salivary amylases.
Pancreatic amylase did not show any affinity for concanavalin
A-Sepharose, but this step was indispensable for removing
serum amylase; B, pancreatic amytase; C. salivary amytase; 0, salivary amylase
passed through concanavalin A-Sepharose; E. salivary amylase adsorbed to
concanavalin A-Sepharose
contaminating
carbohydrates.
The specific activities of the two salivary amylases were
similar,
about
Michaelis
twice
that of the pancreatic
Constants
amylase.
of the Three AmylaseS
With Blue Starch as substrate the Michaelis constants for
ainylases A and B (19.3 g/L and 16.6 g/L, respectively)
salivary
Table 2. Carbohydrate Content of Human
Amylase
Neutral sugars
g/100 g
Pancreatic amylase
1.61 ± 1.02
and pancreatic
amylase (12.5 g/L) were similar; the Vmax
values for salivary amylases A and B, and pancreatic ainylase
were calculated to be 421,433, and 228 mt.units/mg, respectively.
Carbohydrate
A
2.18±0.71
B
8.77 ± 2.28
1408 CLINICAL CHEMISTRY, Vol. 25, No. 8, 1979
Content
The carbohydrate contents of salivary amylase A and B, and
pancreatic amylase (Table 2) varied slightly in different
samples of the same tissues, but there was a marked difference
in the carbohydrate contents of the amylases with (salivary
amylase B) and without affinity
(salivary amylase A and
pancreatic amylase) for concanavalin A.
Isoelectric
Focusing
focusing,
the serum amylases were separated
into two major isoenzymes with p1 values of 6.4 and 7.0, respectively, and one minor isoenzyme with apT of 5.9, as shown
in Figure 3A. Purified pancreatic amylase showed one major
On isoelectric
peak at pH 7.0, with a trace of material
at pH 6.4 (Figure
3B).
Salivary amylase at the step of DEAE-cellulose chromatography gave one major peak at pH 6.4 and one smaller peak at
pH 5.9 (Figure 3C). Salivary amylase A with no affinity for
concanavalin A gave one major peak at pH 6.4, with a very
small peak at pH 5.9 (Figure 3D). On re-chromatography,
material in this small peak with a p1 of 5.9 did not show any
affinity
for concanavalin
A-Sepharose.
Salivary
amylase
B
affinity for concanavalin A gave one major peak at pH
5.9 and one minor peak at pH 6.4 (Figure 3E), which again
showed affinity for concanavalin A on re-chromatography.
SDS-polyacrylamide
gel electrophoresis
of the salivary amwith
Salivary amylase
A,
2
1
4
3
L
Fig. 5. Double-diffusion
patterns of human amylases
Double diffusion on agar (12 g/L) was done with use of antibodyto human salivary
amylase (obtained from DEAE-celluiose)(A) and antibody to human pancreatic
amylase (B). 1, pancreatic amylase; 2, salivary amyiase passed through concanavalin A-Sepharose; 3. salivary amyiase adsorbed to concanavalin ASepharose; As, antibody to salivary amylase; Ap, antibody to pancreatic amy-
lase
(salivary amylase B) are both produced in the parotid gland,
Fig. 4. SDS-poiyacrylamide
gel electrophoresis
of salivary
amylase with p1values of 5.9, 6.2, and 6.4
Salivary amylase obtained from DEAE-ceiiulose was separated by isoelectric
focusing. Amylase fractions at pH 5.9, 8.2, and 6.4 were subjectedto electrophoresis in polyacrylamide gel (75 g/L) asdescribedin Fig. 1. 1. salivaryamylase
at pH 5:9; 2. salivary amytaseat pH 6.2; 3, salivaryamylaseat pH 6.4; 4, pancreatic amyiase at pH 7.0
ylases with p1 values of 5.9 revealed the presence of predominantly
salivary
amylase B, with a small amount
of salivary
amylase A. The salivary amylases with p1 values of 6.2 and 6.4
consisted mainly of the A form with small amounts of the B
form (Figure
it is unknown whether they are produced in the same cells or
in different cells in the parotid gland. Cells producing these
two isoenzymes cannot be identified immunohistochemically,
because the antigenicities of the two isoenzyrnes are completely identical, as shown in Figure 5A and B.
Another interesting problem about the presence of two
isoenzymes with different carbohydrate
contents in the parotid gland is whether one is a precursor or degradation product
of the other (19). The parotid gland probably contains many
endogenous sugar transferases and glycosidases (20), so we
thought that the proportions of the two isoenzymes might
change during incubation of a parotid gland homogenate, but
4).
Antigenicities
of the Three Amylases
Double diffusion on agarose gel was done with use of antibody to salivary
to pancreatic
amylase (from DEAE-cellulose)
to salivary
amylase. Antibody
reacted completely with salivary amylases A and B and pancreatic amylase (Figute 5A). Antibody to pancreatic amylase
also cross reacted completely with all three amylases (Figure
5B). Thus salivary amylases A and B and pancreatic amylase
were not distinguished with amylase antibodies.
Concanavalin
Amylase
A Binding Properties
I.0
and antibody
amylase cross
0.5
E
of Serum
Serum amylase was separated into two major peaks with
p1 values of 6.4 and 7.0 and one minor peak with apI of 5.9 by
isoelectric focusing (Figure 3A). The material in each peak was
subjected to concanavalin
A-Sepharose
column chromatography (Figure 6). Most of the amylase with a p1 of 5.9 had no
affinity for concanavalin A-Sepharose (Figure 6A). The am-
ylases with p1 values of 6.4 and 7.0 also had no affinity for
concanavalin A-Sepharose (Figure 6B and C). Because most
of the salivary amylase with apI of 5.9, and a little of the salivary amylase with apI of 6.4,showed affinity for concanavalin
A, serum amylase,
duced affinity
and even that with a p1 of 5.9 showed re-
for concanavalin A.
ELO
4
0.5
Discussion
Human serum amylase was separated into three isoenzymes
by isoelectric focusing (10). The isoenzymes with p1 values of
7.0 and 6.4 are the main components
of pancreatic
and parotid
gland amylase, respectively, and have been analyzed for
clinical purposes. The parotid gland also produces another
isoenzyme
with a p1 of 5.9, which has a much higher carbohydrate content (8.77 ± 2.28%) than do the other two isoenzymes (1.61 ± 1.02%and 2.18± 0.71%).Most of this isoenzyme
(p1 5.9) has an affinity for concanavalin A.
Although the isoenzymes without affinity for concanavalin
A (salivary amylase A) and with affinity for concanavalin A
0
10
20
30
40
50
Fraction number
Fig. 6. Affinity chromatography on concanavalin A-Sepharose
Serum amylase was separated by isoelectric focusing. Amylase fractions with
p1 values of 5.9, 6.4, and 7.0 were chromatographed on concanavaiin ASepharose as described in FIg. 2. Adsorbed amylase was eluted with a linear
gradient of 0-0.1 mol/L a-methylmannoside in Tris-HCI (10 mmol/L, pH 7.4)
containing 1 mmol of CaCI2 and 1 mmol of MgCl2 per liter. A, serum amylase
of p1 = 7.0; B. serum amylase of p1 = 6.4; C. serum amylase of p1 5.9
CLINICAL CHEMISTRY, Vol. 25, No. 8, 1979
1409
no change was observed. Moreover, the proportions of the two
isoenzymes in the parotid gland and saliva were almost the
same. These findings make it unlikely that one isoenzyrne is
a precursor or degradation product of the other.
The three isoenzymes seemed to have similar protein
moieties, becausetheir antigenicities were completely identical. Stiefel et al. (7) reported that the amino acid compositions of pancreatic and salivary ainylases are similar, but they
detected a slight difference in the peptide maps
and salivary amylases obtained after digestion
In spite of the similarities in the protein moieties
isoenzymes, definite differences were found in
of pancreatic
with pepsin.
of these three
their specific
which corresponded to their Vmax values:
enzyme activities,
salivary amylases A and B (428 and 416 mt. units/mg, respectively) and pancreatic amylase (231 mt.units/mg).
The relative molecular masses of pancreatic amylase, salivary amylase A, and salivary amylase B were calculated to be
60000, 61 000, and 64 000, respectively, from the results of
SDS-polyacrylamide
gel electrophoresis. Since the relative
molecular masses of glycoproteins with high carbohydrate
contents
are overestimated
by SDS-polyacrylamide
gel elec-
trophoresis, the relative molecular mass of salivary amylase
B with about 9% carbohydrate may be overestimated. The
relative molecular masses of pancreatic amylase and salivary
aniylase A, the main amylase in the parotid gland, were
slightly different. The carbohydrate contents of salivary
amylase A (2.18 ± 0.71) and pancreatic arnylase (1.61 ± 1.02%)
were similar, although the contents varied in different
tissues.
Keller et al. (6) reported the presence of two sizes of aniylase
in parotid saliva. They separated two isoenzymes by Bio-Gel
P-100 column chromatography,
one containing 8 mol of
neutral sugar and 4 mol of glucosamine per mole of enzyme,
and the other containing no detectable carbohydrate. They
observed that these amylases were retarded more than ovalbumin on Bio-Gel P-100 column chromatography. Although.
it is unknown whether the amylases are separated by size
difference or by affinity to polyacrylamide beads or Bio-Gel
P-100 column chromatography, or both, it seems likely that
the salivary amylase with 12 mol of carbohydrate corresponds
to that with affinity to concanavalin A.
Most of the serum amylase with a p1 of 5.9 had no affinity
for concanavalin A, unlike the salivary amylase with a p1 of
5.9. Thus carbohydrates in human amylase with affinity to
concanavalin A do not seem to be responsible for the isoelectric point of amylase. The affinity to concanavalin A may
decrease when the amylase is released from the parotid cells.
Sudo and Kanno (21) reported the appearance of amylase
with affinity for concanavalin A in the serum of a patient with
adenocarcinoma
of the lung. Appearance
of this form may be
and pancreas, and to Dr. A. Tsuyuki, Department of Surgery, Keio
University Hospital, for samples of pancreatic juice.
References
1. Takeuchi, T., Matsushima, T., Sugimura, T., et al., A rapid, new
method for quantitative analysis of human amylase isoenzymes. Clin.
Chim. Acta 54, 137 (1974).
2. Long, W. B., and Grider, J. R., Amylase isoenzyme clearances in
normal subjects and in patients with acute pancreatitis. Gastroenterology
71, 589 (1976).
3. Berk, J. E., and Fridhandler, L., Clinical application of amylase
isoenzyme analysis. Am. J. Gastroenterol.
63, 457 (1975).
4. O’Donnell, M.D., FitzGerald, 0., and McGeeney, K. F., Differential
serum amylase determination by use of an inhibitor, and design of a
routine procedure. Clin. Chem. 23, 560 (1977).
5. Benjamin, D. R., and Kenny, M. A., Clinical value of amylase isoenzyme determinations.
Am. J. Clin. Pathol. 62, 752 (1974).
6. Keller, P. J., Kauffman, D. L., Allan, B. J., et al., Further stjdies
on the structural differences between the isoenzymes of human parotid a-amylase. Biochemistry
10, 4867 (1971).
7. Stiefel, D. J., and Keller, P. J., Preparation and some properties
of human pancreatic amylase including a comparison with human
parotid amylase. Biochim. Biophys. Acta 302, 345 (1973).
8. Ceska, M., Birath, K., and Brown, B., A new and rapid method for
the clinical determination of a-amylase activities in human serum
and urine: Optimal conditions. Clin. Chim. Acta 26, 437 (1969).
9. Lowry, 0. H., Rosebrough, N. J., and Farr, A. J., et al., Protein
measurement
with the Folin phenol reagent. J. Biol. Chem. 193,265
(1951).
10. Takeuchi, T., Matsushima, T., and Sugimura, T., Separation of
human a-amylase isozymes by electrofocusing and their immunological properties. Clin. Chim. Acta 60, 207 (1975).
11. Bernfeld, P., Amylases, a and i3. In Methods in Enzymology,
1,
S. P. Colowick and N. 0. Kaplan, Eds., Academic Press, New York,
NY, 1955, pp 149-158.
12. Kauffman, D. L., Zager, N. I., Cohen, E., et al., The isoenzymes
of human parotid amylase. Arch. Biochem.
Biophys.
137, 325
(1970).
13. Takeuchi, T., Nakagawa, Y., Ogawa, M., et al., Immunoreactivities
of a-amylase of humans and rats. Clin. Chim. Acta 77, 203 (1977).
14. Takeuchi, T., Mura, M., Sasaki, R., et al., Comparative studies
on electrophoretic mobility and immunogenicity of pancreatic and
parotid amylases of rat. Biochim. Biophys. Acta 403, 456 (1975).
15. Ouchterlony, O., Antigen-antibody reactions in gels. IV. Types
of reactions in coordinated systems of diffusion. Acta Pat hol. Microbiol. Scand. 30, 231 (1953).
16. Laemmli, U. K., Cleavageof structural proteins during the assembly of the head of bacteriophage
T4. Nature
227, 680 (1970).
17. Zacharius, R. M., Zell, T., Morrison,
J. H., et al., Glycoprotein
staining following electrophoresis
on acrylamide gels. Anal. Biochem.
30, 148 (1969).
18. Dubois, M,, Gilles, K. A., Hamilton,
J. K., et al., Colorimetric
method for determination
of sugars and related substances. Anal.
Chem. 28, 350 (1956).
caused by abnormal release of amylase from amylase-producing cells.
19. Takeuchi,
of multiforms
This work was supported by a Grant-in-Aid for Cancer Research
from the Ministry of Education, Science and Culture, Japan.
I am grateful to Dr. T. Kameya, Pathology Division of the National
Cancer Center Research Institute, for preparations of parotid gland
20. Salegui, M. D., and Plonska, H., Preparation and properties of
porcine submaxillary
mucins. Arch. Biochem. Biophys.
129, 49
(1969).
21. Sudo, K., and Kanno, T., Properties of the amylase produced in
carcinoma of the lung. Clin. Chim. Acta 73, 1(1976).
1410 CLINICAL CHEMISTRY, Vol. 25, No. 8, 1979
T., Matsushima,
T., Sugimura, T., et al., Occurrence
of a-amylase,
new isozymes or autodigested forms?
Clin. Chim. Acta 60, 205 (1975).