Changes in Serum Salivary Isoamylases
in Sjogren's Syndrome
ROBERT
O.
WOLF,
D.D.S.,
MICHAEL
E.
ROSS,
M.D.,
AND THOMAS
M.
TARPLEY,
D.D.S.
National Institutes of Health, National Institute of Dental Research, Laboratory of
Oral Medicine, Bethesda, Maryland
ABSTRACT
Wolf, Robert O., Ross, Michael E., and Tarpley, Thomas M.: Changes in
serum salivary isoamylases in Sjogren's syndrome. Am J Clin Pathol 65:
1022-1025, 1976. Quantitative polyacrylamide disk gel electrophoresis of sera
from 27 patients with Sjogren's syndrome (SS) and 12 comparable normal
subjects revealed that serum amylase activity in patients with SS varies
due to changes in the salivary isoenzyme, while pancreatic isoamylase
remains normal. T h e SS group can be divided into those patients with
markedly increased salivary isoamylase and those with normal or low
salivary isoamylase. At this time we cannot be certain whether this reflects
different stages in a progressive disorder, or differences in the underlying pathologic processes.
Analysis of amylase isoenzymes in serum previously has been shown to
be of value in pancreatic disorders, and we have now demonstrated that
changes in the salivary glands may also be reflected in serum amylase isoenzymes. Study of patients with other salivary and pancreatic disorders
will be needed to define the clinical utility of amylase isoenzyme analysis.
(Key words: Amylase; Serum, human; Sjogren's syndrome; Electrophoresis.)
HUMAN SERUM AMYLASE is derived primarily, if not exclusively, from the pancreas and salivary glands. 3 T h e levels of
activity of pancreatic and salivary amylase
in serum have been shown to be helpful
in the diagnosis of pancreatic insufficiency
in children with cystic fibrosis of the pancreas. 6,7 It was found that the serum pancreatic amylase was much lower than
Received J u n e 9, 1975; received revised manuscript
August 25, 1975; accepted for publication August
25, 1975.
Presented at the 1975 General Session of the American Association for Dental Research, April 5 - 8 ,
1975. T h e New York Hilton, New York, New York.
Address reprint requests to Dr. Wolf: Laboratory
of Oral Medicine, National Institute of Dental
Research, National Institutes of Health, Bethesda,
Maryland 20014.
normal in patients who had pancreatic
insufficiency, while the serum salivary
isoamylase level remained essentially
normal. Thus, changes in total serum
amylase in patients who have cystic fibrosis
are a function of pancreatic isoamylase
levels.
Sjogren's syndrome (SS) is a relatively
rare disease, predominantly affecting
women. Xerostomia, keratoconjunctivitis
sicca, and rheumatoid arthritis or another
collagen disease constitute the diagnostic
triad of SS. Serum pancreatic and salivary
amylase activities in SS patients were examined to determine whether characteristic changes in the serum salivary isoenzyme
levels might be present.
1022
1023
SJOGREN'S SYNDROME—SERUM ISOAMYLASE
June 1976
Materials and Methods
Subjects
T h e sera of 27 patients, mostly women,
studied at the NIH and diagnosed as
having SS were compared with the sera
of 12 normal women. The patients ranged
in age from 24 to 75 years, while the normal subjects were 30 to 62 years old.
each being eluted with 4 ml distilled water
for 18 hours at 4 C. The resulting eluates
were incubated at 36 C. for 16 hours with
the Phadebas substrate to assay the amylase
activity of each disk, thereby yielding the
ratio of pancreatic to salivary amylase in
the serum. A zymogram of the duplicate
column was made to serve as a linear control of isoamylase band position, as previously described. 7
Assay
Results
For every sample, total serum amylase
was obtained using the Phadebas blue
starch substrate and dye release method. 4
Amylase activity was expressed in international units per liter (I.U. per 1.).
Electrophoresis
The serum isoamylases were separated
on standard 7% polyacrylamide gels at pH
9.5 with a tris-glycine reservoir buffer
(pH 8.3) in the Canalco Model 1200 system.
Thirty microliters of sera were applied to
each gel. Electrophoresis was carried out at
5 milliamperes of constant current per
gel for 75 minutes at an ambient temperature of 0 C. as previously described. 9
Each sample was run in duplicate. One
column was sectioned into 1.6-mm disks,
Serum hyperamylasemia was observed in
five of 27 (18.5%) patients who had
Sjogren's syndrome. These patients, all of
whom had serum amylase values more
than three standard deviations above the
mean for the controls, have been considered as a separate class, referred to as
"high-amylase." The 22 other SS patients
are designated "low-amylase" (Table 1).
T h e manufacturer's published normal
range for this method is 7 0 - 3 0 0 I.U. per
1., and all control sera were within this
normal range.
No statistically significant difference in
the levels of the pancreatic isoamylase was
found between the control group and
either of the SS groups (Table 1). A highly
significant increase in serum salivary iso-
Table 1. Serum Amylase and Isoamylase Levels in Patients with Sjogren's Syndrome
Serum Amylase I.U. peir l .
No.
Total
Pancreatic
Salivary
Normal
12
218 ±
48
103 ± 31
(47%)*
1 1 5 ± 31
(53%)*
Sjogren's syndrome (SS)
(all patients)
27
240 ± 139
99 ± 36
(41%)
141 ± 132
(59%)
SS low amylase
22
183 ±
57t
98 ± 38
(54%)
85 ± 47t
(46%)
SS high amylase
5
491 ± 105:t
104 ± 29
(21%)
387 ± 87t
(79%)
* Percentage o f total amylase.
t 0.05 < j) < 0.0 I, patients TAV. normals (Student's t test).
X p < .001, patients vs. normals (Student's t test).
1024
WOLF, ROSS, AND TARPLEY
-
Discussion
>
z
<
o
<n
>|
200
in
i
i
100 -
±
f:
T
50 -
0L
65
The assumption that the recoveries of
pancreatic and salivary isoamylases are
equal is strongly supported by the observed ratio of pancreatic isoamylase to
salivary isoamylase in our controls. T h e
ratio obtained by our method of quantitation is the same as that found by others
using different technics for isoamylase
separation. 1,2
400 -
£300
A.J.C.P.—Vol.
<>
Normal
Sjtigrens
SjSgrens
Low
High Amylase
Amylase
Fie. 1. Histogram of serum salivary amylase, l.U.
per I., for normal, high and low serum amylase
values. Boxes indicate the mean and one standard
deviation unit.
amylase (p < .001) was observed in the SS
high-amylase group. In the SS low-amylase
group the salivary isoamylase level was
slightly decreased from normal, but this
was not statistically significant (0.05 < p
<0.1).
The serum salivary isoamylase values
(l.U. per 1.) of the SS low- and highamylase patients, compared with those of
the normal individuals, are plotted in
Figure 1, with class mean and one standard
deviation indicated. Although patients
with hyperamylasemia are completely
separable from controls, SS patients without hyperamylasemia overlap extensively
with controls.
Generally, about 60% of the total amylase
activity applied to the gel was recovered.
We have shown that in SS, a disease
involving the salivary glands, serum
salivary amylase activity varies considerably, while serum pancreatic amylase
activity remains essentially normal. The
patients who had SS could be divided into
two groups. The patients in the smaller of
these subcategories had hyperamylasemia,
due exclusively to three- to four-fold increases in the levels of serum salivary
isoamylase. In the larger group of patients with SS, salivary amylase was slightly
lower than the normal, while total amylase
was in the normal range.
We have little reason to believe that the
two groups of patients reflect different
pathologic processes within the spectrum
of Sjogren's syndrome. All five of the SS
patients who had hyperamylasemia were
women, as were 21 of 22 patients with
normal total serum amylase. T h e hyperamylasemia group ranged in age from 24
to 61 years, while the normal-amylase
group ranged from 29 to 75 years old.
The presence or absence of anti-salivary
duct antibody, the diagnosis of associated
collagen disease, and the salivary flow rates
were all similar in the two groups. When
the severity of pathologic change was
graded in labial biopsies, no difference
was discernible between the two groups.
However, these biopsies were of the minor
salivary glands, and may not accurately
reflect the changes in the major salivary
glands, where most amylase is presumably
produced.
When the salivary glands or the pancreas
June 1976
SJOGREN'S SYNDROME—SERUM ISOAMYLASE
are acutely damaged (e.g., by infection,
irradiation, trauma), there is an increase in
the total serum amylase, which can be
attributed to the organ damaged. 8 In SS
there often is acute, recurrent swelling of
the salivary glands, especially the parotid. 5
Presumably, the hyperamylasemia we
have observed in five patients is a reflection
of the salivary gland damage. As salivary
gland destruction progresses, the ability
of the salivary gland to produce amylase
and to release enzyme into the serum may
decrease, accounting for the normal or
depressed levels of salivary isoamylase seen
in the majority of the patients. A study of
serum amylase and its isoamylase components in patients with SS followed for
many years may contribute to our understanding of the pathophysiology of SS.
Conclusions
Serum amylase activity in patients who
have SS varies due to changes in the salivary
isoenzyme, while pancreatic isoamylase
remains normal. The SS group can be
divided into those patients who have
markedly increased salivary isoamylase
and those with normal or low salivary
isoamylase. Although we cannot be certain,
this probably reflects different stages of
salivary gland damage, rather than two
different pathologic processes.
Analysis of amylase isoenzymes in serum
1025
previously has been shown to be of value
in pancreatic disorders, and we have now
demonstrated that changes in the salivary
glands may also be reflected in serum
amylase isoenzymes. Study of patients with
other salivary and pancreatic disorders will
be needed to define better the ultimate
clinical utility of amylase isoenzyme
analysis.
References
1. Fridhandler L, Berk JE, Ueda M: Isolation and
m e a s u r e m e n t of pancreatic amylase in
human serum and urine. Clin Chem 18:14931497, 1972
2. Kamaryt J, Brunecky A. Laxova R: Amylasemia.
amylase heterogeneity in blood serum and
activity of amylolytic enzyme in sweat in
members of families afflicted with cystic
fibrosis. Cesk Pediatr 2 5 : 7 8 - 8 1 . 1970
3. Meites S, Rogals S: Amylase isoenzymes. CRC
Crit Rev Clin Lab Sci f: 103- 138. 1971
4. Melnychuk A: Serum amylase: A comparison of a
saccharogenic and four commercial dye substrate methods. Can ] Med Technol 3 5 : 9 18, 1973
5. Shearn MA: Sjogren's syndrome. Major Probl
Intern Med 2:1-262, 1971
6. Taussig LM, Wolf RO, Woods RE. et al: Use of
serum amylase isoenzymes in evaluation of
pancreatic function. Pediatrics 54:229-235,
1974
7. Wolf RO, Taussig LM. Ross ME. et al: Quantitative evaluation of serum pancreatic isoamylase in cvstic fibrosis. J Lab Clin Med
87:164-168,1976
8. Wolf RO, Taylor LL, Brace K: Effects of irradiation of the parotid gland and pancreas in
human isoamvlases. Am J Clin Pathol 54:
214-218, 1970
9. Wolf RO, Taylor LL: Isoenzyme demonstration
technique. Am J Clin Pathol 49:871-876, 1968