Mechanisms of Ageing and Development
119 (2000) 131 – 147
www.elsevier.com/locate/mechagedev

An analysis of submandibular salivary gland
function with desipramine and age in female
NIA Fischer 344 rats
Markus M. Koller a,*, Richard A. Cowman b,
Michael G. Humphreys-Beher c,d, Philip J. Scarpace d,e
a
Department of Oral and Maxillofacial Surgery,
Claude Denson Pepper Center for Research on Oral Health in Aging, Uni6ersity of Florida, Gaines6ille,
FL 32610, USA
b
Dental Research Unit, Department of Veterans Affairs Medical Center, Miami, FL 33125, USA
c
Department of Oral Biology, Uni6ersity of Florida, Gaines6ille, FL 32610, USA
d
Department of Pharmacology and Therapeutics, Uni6ersity of Florida, Gaines6ille, FL 32610, USA
e
Geriatric Research, Education and Clinical Center, Department of Veterans Affairs Medical Center,
Gaines6ille, FL 32608, USA

Received 5 May 2147; received in revised form 30 July 2000; accepted 6 August 2000

Abstract
Dry mouth is one of the major side effects of cyclic antidepressants, which are still a
dominating group of psychotherapeutic drugs used in the treatment of depression. In this
study we analyzed the effects of 28 day tricyclic antidepressant administration and the
reversibility of this treatment following a 15 day washout period on different parameters in
submandibular gland function in aging rats. We postulated that desipramine would decrease
gland function, accented with age, and delay recovery in senescent animals. In contrast to
body weight, desipramine had no effect on glandular wet weight. While glandular DNA
synthesis was changed with age and treatment, the concentration of total membrane and
soluble proteins was not affected. Flow rate was significantly changed with age, but
desipramine increased salivary flow in the youngest animals only. Neither age nor treatment
influenced salivary protein concentrations, but the total amount of proteins secreted, revealed

* Present address: Center for Dental and Oral Medicine, Clinic for Geriatric and Special Care
Dentistry, University of Zurich, P.O. Box 322, CH-8028 Zurich, Switzerland. Tel.: + 41-1-6343341; fax:
+ 41-1-6344319.
E-mail address: mmkoller@zzmk.unizh.ch (M.M. Koller).
0047-6374/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved.
PII: S 0 0 4 7 - 6 3 7 4 ( 0 0 ) 0 0 1 7 6 - 7

132

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

perturbation with age. SDS- polyacrylamide gel analysis revealed changes in protein expression with treatment and age. Desipramine decreased epidermal growth factor (EGF) levels in
all age groups, but increased the secretion of peroxidase and lysozyme. Analysis of total
RNA showed a pronounced decrease with age. These data indicate that desipramine has
profound effects on submandibular salivary gland function. © 2000 Elsevier Science Ireland
Ltd. All rights reserved.
Keywords: Desipramine; EGF; Lysozyme; Peroxidase; Submandibular gland; Saliva

1. Introduction
The primary function of salivary glands is to secrete saliva, a fluid composed of
water, electrolytes, and various multifunctional proteins. The autonomic sympathetic and parasympathetic receptor systems regulate salivary gland secretion via a
co-ordinated sequence of signal transduction and intracellular signalling events
(Looms et al., 1998; Ambudkar, 2000). Saliva is an essential fluid for the health of
human teeth and oral mucosal surfaces, maintains microbial balance, and supports
other oral functions. The basic protective mechanism mediated by saliva is bacterial
clearance. But saliva also contains many antibacterial, antifungal and antiviral
substances. Salivary glands and saliva are a part of the mucosal immune system
(Tomasi and Plaut, 1985). Besides the predominant immunoglobulin secretory IgA,
the oral cavity is protected, as well by many salivary nonimmunoglobulin antimicrobial molecules such as lysozyme, lactoferrin, and peroxidases (Mandel and
Ellison, 1985). Many of these antimicrobial factors interact with each other and
depend on salivary flow rate (Mandel and Ellison, 1985).
Pharmaceutical side effects, systemic diseases, and radiation therapy are reported
to be the most common causes of acquired salivary gland dysfunction (Mandel,
1980), leading to many subjective complaints (Fox et al., 1987) and dysfunction
related objective findings (Grahn et al., 1988; Atkinson and Fox, 1992). With age,
˚
an increasing number of people have medical conditions and/or use therapeutics
that may significantly affect salivary gland physiology and the oral ecology.
Tricyclic antidepressants, the basic group of amine uptake inhibitors, have a wide
pharmacological spectrum displaying a1-adrenoceptor antagonistic, antiserotonin,
antimuscarinic, and antihistaminic properties. Unwanted side effects are common
(Trindade et al., 1998), and become more common with age (Tumer et al., 1992).
¨
One of the most reported ailments of tricyclic antidepressant therapy is dry mouth
(Fox et al., 1985; Remick, 1988). Decreased salivary flow rate (Wu and Ship, 1993),
increased salivary proteins and electrolytes (Blackwell et al., 1980; von Knorring
and Mornstad, 1981; Mornstad et al., 1986), and changes in oral microflora were
¨
¨
found in humans (Parvinen et al., 1984). This may compromise oral and systemic
health, and overall quality of life. Salivary gland derived health problems pose a
significant disease burden for the host.
As in humans, submandibular flow rate seems not to be reduced in aging rats
(Koller et al., 1992), while a progressive age-related decline in the rates of secretory

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

133

protein synthesis in submandibular glands has been observed (Baum et al., 1983).
Epidermal growth factor (EGF) was found to decrease with age, as well (Koller et
al., 1992). Catecholamine responsiveness declines with age and is primarily due to
a decrease in b-adrenergic signal transduction (Scarpace et al., 1991). In peripheral
tissue, for the most part, the density of b-adrenoceptors is maintained with age,
whereas the ability to activate adenyl cyclase declines (Scarpace et al., 1991).
Tricyclic antidepressants may indirectly (central nervous system), and/or directly
(salivary glands) reduce salivary gland function as a result of their action on
modulating neural transmission. It has been shown (Scarpace et al., 1992, 1993),
that signal transduction in submandibular glands was perturbed after chronic
administration of desipramine. This drug down-regulated b-adrenergic receptors at
all ages, but receptor recovery after drug withdrawal was prolonged in the brain of
aged animals (Greenberg et al., 1985). In addition, we recently reported that
desipramine led to changes in protein secretion, what may have negatively influenced oral health in aging rats (Koller et al., 2000).
In this study, we investigated physiological and biochemical changes in submandibular gland secretory function in vivo with age and the long-term administration of the tricyclic antidepressant desipramine. This drug is a relatively selective
inhibitor of the reuptake of norepinephrine and has weak effects on serotonergic
neuronal activity. Therefore we expected (a) that submandibular flow rate would
decrease temporarily with treatment; (b) that total protein secretion, as well as the
content of specific proteins (EGF, peroxidase, lysozyme) in submandibular saliva
would be depressed transitorily, and (c) that age would modulate the treatment-related changes.

2. Materials and methods

2.1. Animals
Pathogen-free female, Fischer 344 NIA rats of 3, 12 and 24 months of age were
purchased from Harlan Sprague-Dawley (Indianapolis, USA) under contract with
the National Institute of Aging (NIA). Animals were housed individually in 7× 10
in. stainless steel micro-isolated cages, in a room maintained at 25°C. No clinical
evidence of infectious illness was detected. All animals were maintained on Purina
rat chow and water ad libitum and exposed to 12-h light–12-h dark cycle starting
at 07:00 h. Animals were acclimatized for a minimum of 10 day before starting the
experiments. At the beginning of the experimental period the young animals
weighed 176 – 200 g, the middle age group rats 216–261 g, and the oldest ones
310 – 352 g.

2.2. Experimental design
Animals received a daily intraperitoneal (i.p. between 07:00 and 09:00 h) injection
of desipramine or saline (10 mg/kg body weight) for 28 days. One half of each

134

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

group was sacrificed on the third day after the last dose (DMI and SAL groups),
the other half after a washout (WO) period of 15 days (DMI/WO and SAL/WO
groups). Body weight was measured weekly, the final weight was monitored at the
day of the experiment.

2.3. Sali6a collection and flow rate
The same person between 09:00 and 11:00 h, using two animals per day, always
carried out saliva collections. The rats were deprived of food the day before the
experiment at 16:00 h. Animals were anesthetized i.p. (sodium pentobarbital, 50
mg/kg) and tracheotomized. The submandibular duct of the animal’s left side was
canulated with calibrated (lumen and length) drawn PE-10 plastic tubes. To
stimulate salivary flow, pilocarpine HCl (Sigma Chemical, USA; 5 mg/kg, i.p.) was
administered; to induce discharge of stored proteins isoproterenol (Sigma Chemical,
USA; 5 mg/kg, i.p.) was chosen. These agents were dissolved in 37°C isotonic
saline. The initial drop of saliva was discarded and subsequent flow was collected
for 30 min into pre-weighed plastic tubes kept on ice. The sample tubes were
reweighed to obtain a gravimetric estimate of the volume of saliva, assuming that
1 ml of saliva equals 1 mg. Flow rate was calculated as ml/min per gland and
corrected for glandular wet weight and/or body weight. Saliva was stored at
− 80°C until analyzed. Sodium ethylene diamine tetraacetate (EDTA 1%, pH 8.6)
was added to submandibular saliva (100 ml/ml saliva) to prevent the formation of
a calcium-protein precipitate (Abe et al., 1980).

2.4. Sali6ary proteins
Samples were aliquoted to avoid repetitive freezing and thawing, assayed at one
time, and in duplicate, to minimize error. Proteins were analyzed at one time by the
method of Bradford (1976), using bovine plasma gamma globulin as the standard.
For EGF estimation the volume of total saliva was brought up to 1 ml with
sterile water. The glacial acetic acid (58 ml) was added, the solution cooled on ice
for 30 min and then centrifuged at 4°C for 30 min in an Eppendorf centrifuge. The
supernatant was transferred to a new tube, frozen at − 80°C and then lyophilized.
The remaining proteins were resuspended in 200 ml sterile water. Then 100 ml were
added to 100 ml of human placental microvilli membranes and [125I]-labeled human
EGF. Following incubation at 37°C for 2 h the membrane was diluted in phosphate
buffered saline (PBS), centrifuged for 20 min at 7000×g and the radiolabel
associated with the membrane determined in a gamma counter. This method of
membrane binding competition is independent of species origin for the EGF source
(Booth et al., 1980).
Salivary peroxidase was assayed by the ABTS [2,2%-azinobis (3-ethylbenzothiozaline-6-sulfonic acid)] method of Shindler et al. (1976). The mmol substrate oxidized
per min at 20°C was calculated from the absorbance change at 412 nm using the
extinction coefficient for the radical cation produced. The salivary activity values
were compared with a plot of activity versus ng protein prepared using purified

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

135

bovine milk lactoperoxidase (Sigma Chemical, USA) as the standard. Salivary
peroxidase concentrations were expressed as ng peroxidase per ml saliva.
Lysozyme was assayed by the lysoplate method of Greenwald and Moy (1976)
using low sulfur content agarose, and Micrococcus lysodeikticus as the indicator
bacterium. Plates were incubated in a humidity chamber at 35°C for 48 h. The
diameter of zone clearing was measured to the nearest 0.1 mm in both a vertical
and horizontal direction. The mean zonal diameter values were compared with a
plot of zone clearing versus ng protein prepared using grade I hen egg-white
lysozyme (Sigma Chemical, USA) as the standard. Lysozyme concentrations were
expressed as ng lysozyme per ml saliva.

2.5. Membrane preparation, cellular proteins, RNA content
At sacrifice, the circulatory system was perfused with 20 ml of cold saline. The two
submandibular glands were excised and dissected free of lymph nodes, fat, and fascia,
and the sublingual glands removed. The wet weight was determined on an analytical
balance, and the tissues immersed in 0.9% NaCl at 4°C. All subsequent steps were
carried out at 4°C. The glands were finely minced in 20 volumes of 0.25 M sucrose,
1 mM MgCl2, and 5 mM Tris – HCl (pH 7.4). To avoid protein degradation, protease
inhibitors were added to the mixture (1 mM leupeptin, 100 mM benzamidine, and 100
mM phenylmethylsulfonylfluoride-PMSF final concentrations). Preparations were
disrupted by a Tekmar Tissuemizer (Cincinnati, USA) for 20 s and subsequently
homogenized with 10 strokes of a motor-driven, teflon-tipped pestle at moderate
speed. The homogenate was passed through two layers of cheese cloth, and
centrifuged at 48 000×g for 15 min. The resultant pellet was resuspended in 20
volumes of 8 mM MgCl2, 0.08 mM ascorbic acid, the above mentioned protease
inhibitors, and 50 mM HEPES (N-hydroxyethyl piperazine-N%-2% ethane sulfonic
acid, pH 7.4). Protein assays for total membrane and supernatant were determined
by the above mentioned method of Bradford (1976).
Submandibular gland RNA was isolated as described by Chomczynski and Sacchi
(1987). One gland was homogenized in 2 ml of denaturing solution consisting of 4
M guanidinium thiocyanate. 25 mM sodium citrate, pH 7, 0.5% sarcosyl, 0.1 M
2-mercaptoethanol were added followed by 0.35 ml of 3 M sodium acetate, pH 5.2,
5 ml of phenol (water saturated), 1 ml of chloroform-isoamyl alcohol mixture (49:1)
to extract the proteins from nucleic acid. The suspension was mixed and cooled on
ice for 15 min and centrifuged at 10 000×g for 20 min at 4°C. The aqueous phase
containing the RNA was transferred to a new tube, mixed with 5 ml of isopropanol,
and then placed at − 20°C for 1 h to precipitate RNA. Sedimentation at 10 000× g
for 20 min at 4°C was again performed. The resulting RNA pellet was dissolved in
0.3 ml of the denaturing solution and precipitated with 0.1 ml 3 M sodium acetate
(pH 5.2) and 0.4 ml isopropanol at − 20°C for 1 h. After centrifugation in an
Eppendorf centrifuge for 10 min at 4°C the RNA pellet was resuspended in 1 ml of
99% ethanol, sedimented, vacuum dried (10 min), and dissolved in 100 ml DEPtreated water and 0.5% SDS. The RNA content of all samples was estimated at one
time by absorption at a wavelength of 260 nm and expressed as mg/g gland.

136

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

2.6. DNA synthesis
In vivo DNA synthesis was determined in submandibular glands by the incorporation of [3H]-thymidine into DNA by the injection of radiolabel (50 mCi per 100
g) 1 h before pentobarbital anesthesia and 4 h prior to sacrificing. Samples (100 ml)
for [3H]-thymidine incorporation were removed prior to membrane centrifugation,
and analyzed by scintillation counting using 10 ml of Fisher premixed nonaqueous
scintillation cocktail (Sci Vers 5X20-4).

2.7. Polyacrylamide gel electrophoresis
Submandibular saliva was subjected to electrophoresis on a 1.5 mm thick 10%
sodium dodecyl-sulfate polyacrylamide gel (SDS/PAGE) using a modified Tris–
glycine system of Laemmli, as described by Pugsley and Schnaitman (1979). Gels
were fixed and stained by a modification of the method of Fairbanks et al. (1971)
as described by Humphreys-Beher and Wells (1984). Samples for gels were made up
at 1.0 mg/ml of sample buffer. 15 mg of protein per well was used for gel
electrophoresis.

2.8. Statistical analysis
The null hypotheses, no change with age and desipramine treatment, were tested
by two way analysis of variance (ANOVA). When the main effect was significant,
subgroups were examined by one-way ANOVA. Body weight was analyzed by
repeated measures ANOVA. The Bonferroni/Dunn multiple comparison procedure
was carried out to determine significant differences among the means following a
significant one- and two-way ANOVA. No significant differences between the two
saline groups were detected for all data. Therefore, we did not differentiate between
the two saline groups in the final analysis.

3. Results

3.1. Body weight, organ weight and DNA-synthesis
Long-term desipramine administration significantly reduced body weight
throughout the experimental period between control and drug treated animals in all
age groups (P B 0.0001), with the greatest loss occurring in senescent rats (around
80 g). The animals gradually regained weight after the cessation of treatment. By
day 15 post-treatment, body weight was comparable to that of control animals in
the youngest age group only. A significant interaction between the parameters age
and treatment was observed (PB 0.0001).
Comparing the control groups, the wet gland weight significantly increased
between 3 and 12 m (P B0.0001), while desipramine administration had no major
effect on glandular weight as on cortex weight in any age group. The histosomatic

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

137

index, the organ to body weight ratio, was significantly decreased with treatment in
the oldest animals only (P B 0.0001).
With regard to the general state of the glandular cells we found, that the DNA
synthesis, as measured by the incorporation of tritiated thymidine (counts per min/g
gland), was significantly influenced by treatment (PB 0.0001) and age (PB 0.0001),
with no significant interaction (Fig. 1). While DNA synthesis was significantly
decreased with desipramine in the two younger age groups (P5 0.0010), the control
animals showed a significant increase between 3 and 12 month of age (P= 0.0002).
Recovery from drug effects was significantly delayed in senescent rats (P= 0.0076).

3.2. Flow rate
Submandibular flow rate (Fig. 2), expressed as ml/min per gland, was significantly
affected by age (P = 0.0013). Correction for differences in glandular wet weight
(ml/min/g gland) and/or for body weight (ml/min/g gland per kg, ml/min per gland
per kg) eliminated the age related effect (P= 0.5099, 0.0435, 0.7598, respectively).
Correction for body weight only revealed a treatment related effect (P= 0.0004).
With desipramine, significant changes in submandibular flow rate were observed in
the youngest animals only. Comparing the control groups, submandibular flow rate
significantly increased with age (P= 0.0017) when expressed as volume per time,
but significantly decreased with age when correcting for glandular wet weight and
body weight (P =0.0118).

Fig. 1. Box-plot of the incorporation of tritiated thymidine into submandibular glands in vivo with age
and treatment. Results are expressed as counts per min per g gland of six animals. Effect of treatment:
significant differences (5%) compared with WO and SAL are marked with (*), compared with SAL with
(+ ) by one-way ANOVA; effect of age: significant differences (5%) compared with 3 m are marked with
(†) by one-way ANOVA.

138

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

Fig. 2. Box-plots of submandibular flow rates with age and treatment. Results are expressed as ml/min
per gland (a) and ml/min/g gland (b). Values represent six to eight animals in each group. Effect of
treatment: significant differences (5%) compared with WO and SAL are marked with (*), compared with
SAL with (+ ) by one-way ANOVA. Effect of age: significant differences (5%) compared with 3 m are
marked with (†), compared with 12 m with (Y) by one-way ANOVA.

3.3. Proteins
No significant age- and treatment-related changes in protein concentrations
(mg/ml) were detected for total membrane (structural proteins) and supernatant
(soluble proteins). For the ratio structural/soluble proteins, a significant age-related
effect was noted between 3 and 12 month old animals (P=0.0099). Subgroup
analysis revealed no significant differences.
Salivary protein concentrations (mg/ml) were not significantly changed with age
and treatment, but the three age groups reacted differently to the drug regimen
(Fig. 3). While in young and adolescent animals a decrease with desipramine
occurred (27 and 40%, respectively), senescent rats showed a treatment-related
increase of 45%. Recovery during the 15 day washout period was quite complete in
all age groups (108, 85 and 102%). Correction for the observed changes in flow rate
(total proteins secreted in 30 min) demonstrated significant age-related changes
(P= 0.0007). Again, the three age groups showed different effects with treatment.
While 3 and 12 month old animals showed depressed protein secretion with the
administration of the antidepressant (30 and 27%), total proteins secreted were
elevated with treatment in senescent rats (30%). Protein concentrations in control
animals revealed a steady increase with age, being significant when comparing 3
and 24 month old animals (P =0.0073), while total proteins secreted showed a
steady, but not significant increase with age.
On SDS/PAGE gels, loaded with a representative sample of each of the three
treatment and age groups (15 mg of salivary proteins), protein expression differed
with treatment and age (Fig. 4). In the Coomassie blue-stained gel, changes were
detected in the bands around 36 kDa. The synthesis of these proteins was most

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

139

Fig. 3. Box-plots of (a) the protein concentration (mg/ml), and (b) of the total amount of protein secreted
(mg) with age and treatment. Results represent the values of six to eight animals in each group. Effect
of age: significant differences (5%) compared with 3 m are marked with (†) by one-way ANOVA.

pronounced in the treatment groups. An unusual protein (around 49 kDa) was
detected in the desipramine group of 24 month old animals.
Estimating the concentration (mg/ml) of EGF (Fig. 5) tested the secretory
capacity of the submandibular granular tubule cells. Both, age and treatment

Fig. 4. Submandibular saliva protein analyzed on 10% SDS polyacrylamide gels with a representative
sample (15 mg of protein) of each of the three age and treatment groups (Coomassie blue staining).
Pre-stained molecular standards (Std) were; phosphorylase B, 130 kDa; bovine serum albumin, (BSA) 80
kDa; ovalbumin, 49.5 kDa; carbonic anhydrase, 32.5 kDa; soybean trypsin inhibitor, 27.5 kDa;
lysozyme, 18.5 kDa.

140

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

Fig. 5. Box-plots of (a) the salivary EGF-concentration (mg/ml), and of (b) the salivary EGF content per
mg protein secreted (mg/mg). The results express the values of six animals in each group. Effect of
treatment: significant differences (5%) compared with SAL are marked with ( + ), compared with WO
with ( ") by one-way ANOVA; effect of age; significant differences (5%) compared with 3 and 12 m are
marked with (c ), compared with 3 m with (†), compared with 12 m with (Y) by one-way ANOVA.

revealed significant changes in the salivary EGF concentration (PB 0.0001), and
the interaction age× treatment was significant, as well (PB 0.0211). Comparing the
control groups, submandibular saliva of 12 month old animals contained 40% more
EGF per ml saliva than the saliva of the youngest animals (P= 0.1282), 24 month
old rats showed a decrease of 81% compared with the adolescent rats (P=0.0004).
All age groups showed decreased EGF concentrations with desipramine, significant
in the middle age group only (P B 0.0001). Basing the content of EGF on 1 mg of
total protein in the sample (mg/mg) revealed a quite similar pattern for age
(P B 0.0001) and treatment (P B 0.0140), and the interaction was still significant
(P =0.0367).
The content of peroxidase in submandibular saliva (Fig. 6), expressed as ng/mg
protein secreted, was significantly affected by age (P=0.0237) and treatment
(P= 0.0206), with a significant interaction between the two parameters (P =
0.0439). With desipramine administration, an increase was noted in all age groups,
significant in 12 month old animals (P= 0.0007). Recovery was incomplete in this
age group only (P =0.0036). Comparing the controls, saliva of adolescent rats
showed a significant lower peroxidase content than in the youngest age groups
tested (P 50.0054).
The immunoprotein lysozyme showed similar effects with age (P= 0.0002) and
treatment (P =0.0021) as peroxidase (Fig. 6). The increase with desipramine was
significant in the oldest age group only (P= 0.0021). The control animals showed
no age-related effect.
Total RNA (Fig. 7), expressed as mg/g gland, was significantly affected by age
(P =0.0213), but not by treatment (P= 0.3440). In control animals, RNA decreased from 3 to 12 month of age by 17% and by 16% from 12 to 24 month old
animals.

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

141

Fig. 6. Box-plot of the salivary peroxidase (a) and lysozyme (b) content per mg protein secreted (ng/mg)
with age and treatment. The results represent the values of six animals in each group. Effect of
treatment: significant differences (5%) compared with WO and SAL are marked with (*), compared with
SAL with ( + ) by one-way ANOVA; effect of age: significant differences (5%) compared with 3 m are
marked with (†), compared with 12 m with (Y) by one-way ANOVA.

Fig. 7. Box-plot of the RNA content in submandibular glands with age and treatment. The results are
expressed as mg per g gland of six to nine animals in each group. Effect of age; significant differences
(5%) compared with 3 m are marked with (†) by one-way ANOVA.

142

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

4. Discussion
In the brain, desipramine blocks the reuptake of norepinephrine, leading to
desensitization of the postjunctional b-adrenergic receptors, involving both a rapid
attenuation of the responsiveness of adenylate cyclase and a slower downregulation
of receptor number (Hausdorff et al., 1990). In submandibular glands of young
rats, we observed desensitization of isoproterenol-stimulated activity, unchanged
post-receptor signal transduction, and complete recovery after a 15 day washout
period (Scarpace et al., 1992). With age, desipramine led to down-regulation in
receptor number and attenuated receptor stimulated adenylate cyclase activity in
submandibular glands (Scarpace et al., 1993). In addition, drug attenuated Gprotein linked post receptor adenylate cyclase activity persisted through the 15 day
washout period. In 12 month old rats, submandibular receptor up-regulation and
adenylate cyclase supersensitivity were observed. Therefore, the animals in the
present study probably also experienced receptor downregulation and adenylate
cyclase desensitization, and the revealed alterations in submandibular protein
secretion are, at least partially, the consequences of these changes in the signal
transduction cascade changes present in all ages.
It is well known, that most exocrine protein secretion occurs subsequent to
b-adrenoreceptor activation. In this study we confirmed, that long-term administration of aging rats to the tricyclic antidepressant desipramine results in changes in
salivary protein synthesis. A stimulating effect on the biosynthesis and secretion of
different specific salivary proteins (e.g. bands around 36 kDa, lysozyme, peroxidase)
was observed in all age groups. To maintain stable levels of total proteins secreted,
the excretion rate of other proteins must have been depressed with desipramine.
This was the case for the secretion of EGF in all the age groups. In a parallel study
with the same drug regimen we showed unaffected salivary lactoferrin contents with
treatment (Koller et al., 2000). Thyroid hormone is part of the regulatory mechanisms in salivary gland function (Tumilasci et al., 1986; Johnson and Kalu, 1988;
Rousseau et al., 1998), e.g. it regulates the synthesis of EGF (Rall et al., 1985;
Gubits et al., 1986; Johnson et al., 1987). In addition, serum levels of tetra- and
triiodothyronine are influenced by treatment with antidepressants (Dagogo-Jack,
1995). Therefore, the detected changes in EGF secretion with desipramine may have
indirectly influenced submandibular gland physiology in this study. As growth
factor and cytokine receptors, many steroid hormones and their receptors found in
salivary cells play a major regulatory role in salivary gland proliferation and
function (Baum and Wellner, 1999; Purushotham and Humphreys-Beher, 1995), the
decline in serum levels of thyroid hormones may have influenced cellular DNA
synthesis and specific protein synthesis in this study. In the brain, chronic antidepressant therapy has been shown to induce changes in the function of protein
kinase C (PKC), cyclic AMP-dependent protein kinase, and calcium/calmodulin-dependent protein kinase (Popoli et al., 2000). Therefore, as protein phosphorylation
is an obligate step for most signaling pathways, desipramine may directly affect
salivary gland proliferation and inactivate certain enzymes through the inactivation
of important second messenger signaling pathway components (Purushotham and
Humphreys-Beher, 1995).

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

143

Differences in the induction of specific proteins with desipramine may be due to
either nonspecific interaction with b-adrenergic signal transduction or differences in
the induction of protein synthesis in different glands (Bedi, 1993). Additionally,
there may be a differential control of secretion of proteins synthesized by ductal
and acinar cells. While the secretion of many ductal proteins, such as EGF, appears
to be initiated by a-adrenergic receptors (Wallace and Partlow, 1976; Orstavik and
Gautvik, 1977; Hirata and Orth, 1980), the synthesis and secretion of acinar cell
products is regulated by b-adrenergic receptors (Johnson and Cortez, 1988). Therefore, in submandibular glands, desipramine may affect not only b-adrenergic signal
transduction (Scarpace et al., 1992, 1993), but b-adrenergic and muscarinic signaling, as well. Hanft and Gross (1990) showed, that rat cortical a1a- and a1b-adrenoceptors are regulated in a subtype-selective manner by desipramine (Hanft and
Gross, 1990). This mechanism may influence secretory functioning of peripheral
organs, as well. Additionally, desipramine administration may alter other neurotransmitter receptors involved in salivary secretory events as vasoactive intestinal
polypeptide (VIP), tachykinin and purine receptors.
The appearance of structurally altered proteins (Danner and Holbrook, 1990) is
a prominent age-related change observed in various organs and tissues. These
changes affect cellular contents, as well as functional properties of glandular
proteins. Inanaga et al. (1988) observed unusual proteins in parotid saliva of 8- and
15-month-old rats similar to those we detected in submandibular saliva in senescent
rats treated with the antidepressant desipramine.
Saliva, having antimicrobial and growth-stimulating factors simultaneously, and
the oral microbiota are the main factors determining oral health. Salivary antibody
and non-immunoglobulin antimicrobial factors (e.g. lysozyme, peroxidase) maintain
ecological balance (Grahn et al., 1988). Peroxidase and lysozyme are only minor
˚
constituents of submandibular saliva. The changes in their content with treatment
may have only a minor influence on oral microbiota in rats, but may partially
reflect the reported increased incidence of gingivitis with desipramine treatment
(Koller et al., 2000).
Contrary to our expectations, considering the antagonistic action of desipramine
on a1-adrenergic and muscarinic-cholinergic receptors in the brain (Potter et al.,
1991), significant increased submandibular flow rates were detected with desipramine administration in young animals. One can speculate, that the detected
increase in flow rate was caused by a positive feedback mechanism through the
muscarinic-cholinergic, a1-adrenergic and substance P signal transduction system
due to the observed decreased protein concentration.
In contrast to human studies (Fernstrom and Kupfer, 1988), long-term treatment
of rats with tricyclic antidepressants caused decreased food intake and a related
weight loss (Nobrega and Coscina, 1987; Orthen-Gambill and Salomon, 1990).
Long-term desipramine administration altered food and water intake (Durcan et
al., 1988). After an initial decrease, body weight progressively returned towards
pretreatment levels. This was explained by a rapid suppressive effect (acute central
action of desipramine), and an adaptive effect during treatment (changes in feeding
associated brain regions). This confirms our results, as body weight, after an initial

144

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

decrease (1 – 2 weeks), was quite stable during the drug regimen in the two younger
age groups. The decrease in the oldest animals lasted until the end of treatment and
recovery was incomplete. As mentioned before, food intake, mastication, and
long-term desipramine administration influenced serum levels of tetra- and triiodothyronine. Therefore, the loss in body weight and related changes in food and
water intake with treatment may have indirectly influenced submandibular gland
function in this study.
In conclusion, the complexity of psychotropic drug actions, and variations in the
absorption and excretion of these compounds and their metabolites, make precise
predictions of their effects on salivary gland function difficult. The correlation
between receptor activation and intracellular responses is often not predictable.
Various signaling pathways seem to intersect and crosstalk, to modify and influence
the biological outcome of the extracellular signal. Therefore, interactions among
various transmitters must be considered with desipramine (Hill, 1998; Looms et al.,
1998).

Acknowledgements
This work was supported in part by PHS Grants DE 08845 (Claude Denson
Pepper Center for Research on Oral Health in Aging, P.J.Scarpace and
R.A.Cowman.), and DE 08778 (M.H.B.) from the National Institutes for Dental
Research, USA, the Medical Research Service of the Department of Veterans
Affairs (P.J.Scarpace. and R.A.Cowman.), USA, and the University of Zurich
(M.M.K.), Switzerland.

References
Abe, K., Yoneda, K., Fujita, R., Yokota, Y., Dawes, C., 1980. The effects of epinephrine, norepinephrine, and pilocarpine on the types of proteins secreted by rat salivary glands. J. Dent. Res. 59,
1627–1634.
Ambudkar, I.S., 2000. Regulation of calcium in salivary gland secretion. Crit. Rev. Oral. Biol. Med. 11,
4–25.
Atkinson, J.C., Fox, P.C., 1992. Salivary gland dysfunction. Clin. Geriatr. Med. 8, 499 – 511.
Baum, B.J., Wellner, R.B., 1999. Receptors in salivary glands. In: Garrett, J.R., Ekstrom, J., Andersen,
¨
L.C. (Eds.), Neural mechanisms of salivary gland secretion. Front. Oral. Biol., Vol. 11. Karger,
Basel, pp. 44–48.
Baum, B.J., Kuyatt, B.L., Humphreys, S., 1983. Protein production and processing in young adult and
aged rat submandibular gland cells in vitro. Mech. Ageing. Dev. 23, 123 – 136.
Bedi, G.S., 1993. The effect of adrenergic agonists and antagonists on the expression of proteins in rat
submandibular and parotid glands. Crit. Rev. Oral. Biol. Med. 4, 565 – 571.
Blackwell, B., Peterson, G.R., Kuzma, R.J., Hostetler, R.M., Adolphe, A.B., 1980. The effect of five
tricyclic antidepressants on salivary flow and mood in healthy volunteers. Commun. Psychopharmacol. 4, 255–261.
Booth, A.G., Olaniyan, R.O., Vanderpuye, O.A., 1980. An improved method for the preparation of
human placental syncytiotrophoblast microvilli. Placenta 1, 327 – 336.

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

145

Bradford, M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of
protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248 – 254.
Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156 – 159.
Dagogo-Jack, S., 1995. Salivary epidermal growth factor concentration in thyrotoxicosis. Nutrition 11
(suppl 5), S643–S645.
Danner, D.B., Holbrook, N.J., 1990. Alterations in gene expression with aging. In: Schneider, E.L.,
Rowe, J.W. (Eds.), Handbook of the Biology of Aging, Part I. van Nostrand Reinhold, New York,
pp. 97–115.
Durcan, M.J., McWilliam, J.R., Campbell, I.C., Neale, M.C., Dunn, G., 1988. Chronic antidepressant
drug regimes and food and water intake in rats. Pharmacol. Biochem. Behav. 30, 299 – 302.
Fairbanks, G., Steck, T.L., Wallach, D.F.H., 1971. Electrophoretic analysis of the major polypeptides of
the human erythrocyte membrane. Biochemistry 10, 2606 – 2617.
Fernstrom, M.H., Kupfer, D.J., 1988. Antidepressant-induced weight gain: a comparison study of four
medications. Psychiatry Res. 26, 265– 271.
Fox, P.C., van der Ven, P.F., Sonies, B.C., Weiffenbach, J.M., Baum, B.J., 1985. Xerostomia: evaluation
of a symptom with increasing significance. J. Am. Dent. Assoc. 110, 519 – 525.
Fox, P.C., Busch, K.A., Baum, B.J., 1987. Subjective reports of xerostomia and objective measures of
salivary gland performance. J. Am. Dent. Assoc. 115, 581 – 584.
Grahn, E., Tenovuo, J., Lehtonen, O.-P., Eerola, E., Vilja, P., 1988. Antimicrobial systems of human
˚
whole saliva in relation to dental caries cariogenic bacteria and gingival inflammation in young
adults. Acta. Odontol. Scand. 46, 67– 74.
Greenberg, L.H., Brunswick, D.J., Weiss, B., 1985. Effect of age on the rate of recovery of b-adrenergic
receptors in rat brain following desmethylimipramine-induced subsensitivity. Brain Res. 328, 81 – 88.
Greenwald, R.A., Moy, W.W., 1976. Effect of agarose variability on the measurement of lysozyme
activity. Clin. Chim. Acta. 73, 299–305.
Gubits, R.M., Shaw, P.A., Gresik, E., Onetti-Muda, A., Barka, T., 1986. Epidermal growth factor gene
expression is regulated differently in mouse kidney and submandibular gland. Endocrinology 119,
1382–1387.
Hanft, G., Gross, G., 1990. The effect of reserpine desipramine and thyroid hormone on a1a- and
a1b-adrenoceptor binding sites: evidence for a subtype-specific regulation. Br. J. Clin. Pharmacol. 30
(Suppl 1), S125–S127.
Hausdorff, W.P., Caron, M.G., Lefkowitz, R.J., 1990. Turning of the signal, desensitization of
b-adrenergic receptor function. FASEB J. 4, 2881 – 2889.
Hill, S.M., 1998. Receptor crosstalk: communication through cell signaling pathways. Anat. Rec. 253,
42–48.
Hirata, Y., Orth, D.N., 1980. Secretion of epidermal growth factor, nerve growth factor, and renin-like
enzymes by dispersed male mouse submandibular gland cells in vitro. Endocrinology 107, 92 – 97.
Humphreys-Beher, M.G., Wells, D.J., 1984. Metachromatic staining patterns of basic proline rich
proteins from rat and human saliva in sodium dodecyl sulfate-polyacrylamide gels. Appl. Biochem.
6, 353–360.
Inanaga, A., Habu, T., Tanaka, E., Taniguchi, T., Nishiura, T., Ishibashi, K., Abe, K., 1988. Age
changes in secretory function of male and female rat parotid glands in response to methoxamine and
pilocarpine. J. Dent. Res. 67, 565–573.
Johnson, D.A., Cortez, J.E., 1988. Chronic treatment with b-adrenergic agonists and antagonists alters
the composition of proteins in rat parotid saliva. J. Dent. Res. 6, 1103 – 1108.
Johnson, D.A., Kalu, D.N., 1988. Influence of thyroxine in the regulation of rat parotid salivary protein
composition. J. Dent. Res. 67, 812–816.
Johnson, D.A., Alvares, O.F., Etzel, K.R., Kalu, D.N., 1987. Regulation of salivary proteins. J. Dent.
Res. 66, 576–582.
Koller, M.M., Maeda, N., Purushotham, K.R., Scarpace, P.J., Humphreys-Beher, M.G., 1992. A
biochemical analysis of parotid and submandibular gland function with age after simultaneous
stimulation with pilocarpine and isoproterenol in female NIA Fischer 344 rats. Archs. Oral. Biol. 37,
219–230.

146

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

Koller, M.M., Purushotham K.M., Maeda N., Scarpace P.J., Humphreys-Beher, M.G., 2000. Desipramine induced changes in salivary proteins, cultivable oral microbiota and gingival health in
aging female NIA Fischer 344 rats. Life Sciences, accepted for publication.
Looms, D.K., Tritsaris, K., Dissing, S., Jørgensen, T.D., Nauntofte, B., 1998. Regulation of salivary
secretion: interactions between signalling pathways. In: Guggenheim, B., Shapiro, S. (Eds.), Oral
Biology at the Turn of the Century. Karger, Basel, pp. 79 – 88.
Mandel, I., Ellison, S.A., 1985. The biological significance of the nonimmunoglobulin defense factors.
In: Pruitt, K.M., Tenovuo, J. (Eds.), The Lactoperoxidase System: Chemistry and Biological
Significance. Marcel Dekker, New York, pp. 1 – 14.
Mandel, I.D., 1980. Sialochemistry in diseases and clinical situations affecting salivary glands. Crit. Rev.
Clin. Lab. Sci. 12, 321–366.
Mornstad, H., von Knorring, L., Forsgren, L., Holmgren, S., 1986. Long-term effects of two principally
¨
different antidepressant drugs on saliva secretion and composition. Scand. J. Dent. Res. 94, 461 – 470.
Nobrega, J.N., Coscina, D.V., 1987. Effects of chronic amitriptyline and desipramine on food intake and
body weight in rats. Pharmacol. Biochem. Behav. 27, 105 – 112.
Orstavik, T.B., Gautvik, K.M., 1977. Regulation of salivary kallikrein secretion in submandibular gland.
Acta Physiol. Scand. 100, 33–44.
Orthen-Gambill, N., Salomon, M., 1990. Differential effects of psychotropic drugs on feeding in rats: is
histamine blockade involved? Pharmacol. Biochem. Behav. 36, 837 – 841.
Parvinen, T., Parvinen, I., Larmas, M., 1984. Stimulated salivary flow rate pH and lactobacillus and
yeast concentrations in medicated persons. Scand. J. Dent. Res. 92, 524 – 532.
Popoli, M., Brunello, N., Perez, J., Racagni, G., 2000. Second messenger-regulated protein kinases in the
brain: their functional role and the action of antidepressant drugs. J. Neurochem. 74, 21 – 33.
Potter, W.Z., Rudorfer, M.V., Manji, H., 1991. The pharmacologic treatment of depression. New Engl.
J. Med. 325, 633–642.
Pugsley, A.P., Schnaitman, C.A., 1979. Factors affecting the electrophoretic mobility of the major outer
membrane proteins of Escherichia coli in polyacrylamide gels. Biochim. Biophys. Acta 581, 163 – 178.
Purushotham, K.R., Humphreys-Beher, M.G., 1995. The role of phosphotyrosine signaling pathway in
parotid gland proliferation and function. Crit. Rev. Oral. Biol. Med. 6, 119 – 131.
Rall, L., Scott, J., Bell, G., Crawford, R., Penschow, J., Niall, H., Coghlan, J., 1985. Mouse
prepro-epidermal growth factor synthesis by the kidney and other tissues. Nature 313, 228 – 231.
Remick, R.A., 1988. Anticholinergic side effects of tricyclic antidepressants and their management. Prog.
Neuro-Psychopharmacol. Biol. Psychiat. 12, 225 – 231.
Rousseau, A., Marquet, P., Lagorce, J.F., Sauvage, M.F., Buxeraud, J., Lachatre, G., Raby, C., 1998.
Thyroid accumulation and adverse effects of imipramine and desipramine in rats after long-term
administration. Pharmacology 57, 242 – 248.
Scarpace, P.J., Tumer, N., Mader, S.L., 1991. a-adrenergic function in aging. Basic mechanisms and
¨
clinical implications. Drugs Aging 1, 116 – 129.
Scarpace, P.J., Koller, M.M., Rajakumar, G., 1992. Desipramine desensitizes b-adrenergic signal
transduction in rat salivary glands. Neuropharmacology 31, 1305 – 1309.
Scarpace, P.J., Koller, M.M., Rajakumar, G., 1993. Desipramine desensitizes b-adrenergic signal
transduction in salivary glands, differential regulation with age. Eur. J. Pharmacol. 247, 65 – 72.
Shindler, J.S., Childs, R.E., Bardsley, W.G., 1976. Peroxidase from human cervical mucus. The isolation
and characterization. Eur. J. Biochem. 65, 325 – 331.
Tomasi, T.B., Plaut, A.G., 1985. Humoral aspects of mucosal immunity. In: Gallin, J.I., Fauci, A.S.
(Eds.), Advances in Host Defense Mechanisms, vol. 4. Raven Press, New York, pp. 31 – 61.
Tumer, N., Scarpace, P.J., Lowenthal, D.T., 1992. Geriatric pharmacology: basic and clinical consider¨
ations. Ann. Rev. Pharmacol. Toxicol. 32, 271 – 302.
Tumilasci, O.R., Houssay, A.B., Paz, C., del, V., Sosto, N.E., Varela, V., 1986. Influence of thyroid
function upon ‘substance P’ induced secretion of saliva by submaxillary glands. Horm. Metab. Res.
18, 234–237.
Trindade, E., Menon, D., Topfer, L.A., Coloma, C., 1998. Adverse effects associated with selective
serotonin reuptake inhibitors and tricyclic antidepressants: a meta-analysis. Can. Med. Ass. J. 159,
1245–1252.

M.M. Koller et al. / Mechanisms of Ageing and De6elopment 119 (2000) 131–147

147

von Knorring, L., Mornstad, H., 1981. Qualitative changes in saliva composition after short-term
¨
administration of imipramine and zimelidine in healthy volunteers. Scand. J. Dent. Res. 89, 313 – 320.
Wallace, L.J., Partlow, L.M., 1976. a-adrenergic regulation of secretion of mouse saliva rich in nerve
growth factor. Proc. Natl. Acad. Sci. USA 73, 4210 – 4214.
Wu, A.J., Ship, J.A., 1993. A characterization of major salivary gland flow rates in the presence of
medications and systemic disease. Oral. Surg. Oral. Med. Oral. Pathol. 76, 301 – 306.

.