Acta physiol. scand. 1976. 98. 400-406 From the Department of Physiology and Biophysics, University of Lund, Sweden
Tissue Hyperosmolality as a Causal Factor in Vasodilatation Following Sympathetic Stimulation of the Submandibular Gland BY
JANHOLMBERC and JANLUNDVALL Received 8 March 1976
Abstract HOLMBERG, J. and J. LUNDVALL. Tissue hyperosmolality as a causal factor in vasodilatation following sympathetic stimulation of the submandibular gland. Acta physiol. scand. 1976. 98. W 0 6 . In a previous investigation, parasympathetic activation of the submandibular gland in the cat was found to cause a considerable increase of regional tissue osmolality, the degree of which was related to the evoked functional hyperernia; intra-arterial hypertonic infusion t o the resting gland producing tissue hyperosmolality of similar magnitudes caused graded and marked dilatations (Lundvall and Holmberg 1974). It was concluded that hyperosmolality contributes significantly to the functional hyperemia response. In the present study evidence is presented to indicate that tissue hyperosmolality is a mediator of the dilatation associated with sympathetic activation as well. An increase of tissue hyperosmolality, as traced in the venous effluent, was found at all frequencies of sympathetic stimulation (2-16 Hz). At high stimulation rates it sometimes exceeded the resting control level by more than 20 mOsm/kg H,O. There was a direct relation between the degree of venous hyperosmolality and the hyperemia response observed immediately after cessation of stimulation. Comparison of the dilator effects evoked by sympathetic stimulation and by hypertonic infusion to the resting gland indicated that tissue hyperosmolality is an important causal factor for the nerve induced dilatation, especially at low and moderate stimulation rates.
Activation of the sympathetic nerves to the cat submandibular gland causes vasoconstriction and a scant salivary secretion. The vasoconstrictor response is most pronounced in the early period of excitation and then usually subsides gradually. Upon cessation of stimulation quite a marked vasodilatation is observed, a phenomenon which only to a minor extent can be explained in terms of a reactive hyperemia response (e.g. Hilton and Lewis 1955). Several investigations have tried to elucidate the mechanismts) responsible for the gradual abolition of the constrictor response during stimulation and for the poststimulatory vasodilatation. The problem may thus not only be of academic interest, but may have general implications with regard to blood flow control in glandular tissue. The dilator effect was attributed to specific sympathetic vasodilator nerves already by Carlson (1907), and Bhoola et al. (1965) obtained evidence to indicate that such a direct
OSMOLALITY IN GLANDULAR VASODILATATION
40 1
neurogenic inhibition of vascular tone was mediated via /3-adrenoceptor activation. On the other hand, Barcroft and Piper (1912) suggested that the vasodilatation was caused by metabolites released during the sympathetic activation of the glandular cells. Hilton and Lewis (1956) and later Gautvik and associates (1972, 1973) presented experimental support for the hypothesis that a “specific” metabolite, the vasoactive polypeptide bradykinin, was causally involved in the reaction. A “non-specific” metabolic factor, regional tissue hyperosmolality, was recently found to play a significant role in the vasodilatation caused by parasympathetic stimulation of the cat submandibular gland (Lundvall and Holmberg 1974). Briefly, it was shown that parasympathetic activation was associated with an increased tissue osmolality as reflected in the venous effluent, and the degree of this osmolar change was related to the magnitude of the evoked functional hyperemia. Intra-arterial infusion of hypertonic solutions to the resting gland producing comparable levels of venous hyperosmolality caused marked vasodilatation. The present study shows that also sympathetic activation of the submandibular gland is associated with an increased tissue osmolality. A comparison with the hypertonic infusion data from the aforementioned report indicates that this osmolar environmental change contributes significantly to the dilator phenomena during and after syrnpathetic nerve stimulation.
Methods 9 cats of both sexes, weighing 2.4-3.6 kg were used. The animals were anesthetized i.v. with chloralose (SO mg/kg b.wt.) and urethane (100 mg/kg b.wt.) after induction with ether. The submandibular gland was exposed unilaterally and its salivary duct cannulated. The chorda-lingual nerve was cut. The main vein of the gland, emptying into the external jugular vein, was carefully isolated and other, minute, veins draining the gland were ligated. After heparinization (3-5 mg/kg b.wt.), the external jugular vein was cannulated and all its tributaries except that from the gland were ligated. Venous outflow from the gland was diverted through an optical drop-recorder unit for continuous registration of blood flow and then returned to the animal via a funnel connected to the right femoral vein. Venous outflow pressure was adjusted to heart level. Arterial inflow pressure was monitored from the right femoral artery. The distal end of the severed cervical sympathetic trunk was stimulated proximal to the superior cervical ganglion for 2-5 min with bipolar ring electrodes. Supramaximal square wave shocks (5-7 V, 2 ms) at frequencies of 2-16 Hz were used. Venous samples were collected at intervals for determination of plasma osmolality. In some animals, arterial samples were taken as well from a cannula in the brachial artery. The “dead space” fluid volume of the venous outflow tubing and of the arterial cannula was allowed for during sampling. Plasma osmolality was determined by thermistor cryoscopy (Osmometer 3 1 LAS, Advanced Instruments, Inc.). Each sample was measured twice and if, occasionally,different readings were obtained the mean value was used. Repetitive measurements o n osmolar standards deviated at most by 1.5 mOsm/kg H 2 0 from the true value. Arterial blood pressure, glandular blood flow, and submandibular secretion [the latter registered a s drops of known volume) were recorded on a Grass Polygraph. The following drugs were given i.v.: tolazoline (PriscoP), chlorpromazine (Hibemalt@),dihydroergotamine (Orstanorm@’), and propranolol (InderaP).
Results Sympathetic activation of the submandibular gland was invariably found to be associated with an increase of plasma osmolality in the venous effluent from the gland. The increased osmolality was present both during and immediately after stimulation in the great majority
402
JAN HOLMBERG AND JAN LUNDVALL
B
A ARTERIAL BLOOD PRESSURE mmHg
150 1001
dOsec
-6
TIME DROPS OF SALIVA
SIGNAL
INCREASE OF VENOUS OSMOLALITY mOsm/kg H 2 0
S Y M P STlM 1 6 H Z
S Y M P STIM 16 t i 1
0
7
9
2
0
0
4
8
11
BLOOD FLOW ml/minx100g
69.9 6.8
24.5 240.0
60.1
6.9
36.5
96.4
Fig. 1. Changes of blood flow and regional venous osmolality in the cat submandibular gland evoked by short (Panel A) and more prolonged (Panel B) sympathetic nerve activation (16 Hz). The induced saliva secretion (each drop 0.06 ml) is also shown.
of the expts. In 4 of the stimulation experiments (total number =43), however, it appeared only after cessation of the nerve activation. The venous hyperosmolality, reflecting an increased osmolality in the active gland, was caused by local osmolar production since arterial osmolality was concomitantly unchanged. The degree of hyperosmolality was graded in relation to the rate of sympathetic stimulation and at 16 Hz it sometimes could amount to more than 20 mOsm. Fig. 1 shows original records illustrating the relation in time between the changes of blood flow and venous osmolality evoked by sympathetic stimulation. The induced salivary secretion is also shown (drop volume 0.06 ml). In panel A, stimulation of relatively short duration (16 Hz, 2 min) caused a rapid and pronounced decrease of blood flow, which was fairly well maintained during the period of stimulation. The gland was active during stimulation producing a scanty secretion of saliva. Cessation of the stimulation evoked a short-lasting but quite marked hyperemia response, corresponding to a 227 % increase of vascular conductance above the resting, control level. Venous osmolality rose by 7 mOsm in the later phase of stimulation and a peak increase of 9 mOsm was observed at the time of the poststimulatory hyperemia. It then rapidly returned to the control level. The initial pronounced constriction in the resistance vessels was often found to decline quite significantly in a gradual manner despite continuous stimulation, and this effect was most marked in response to more prolonged sympathetic activation as illustrated in Fig. 1, Panel B (16 Hz, 5 min). The constriction was here completely abolished after about 3 min of stimulation and was reversed to vasodilatation towards the end of the activation period. The abolition of the constrictor response apparently cannot be ascribed to a failing transmission of the nerve impulses since the salivation was well maintained during the period of stirnulation. It can be seen that venous osmolality, in parallel to the decline of the constrictor response, rose gradually to a level 11 mOsm above the prestimulatory, control value. The majority of the sympathetic stimulations were performed for a period of about 2 min and the results obtained in these experiments are summarized in Fig. 2. The diagram shows the peak hyperemia response observed upon cessation of nerve activation, expressed as the per cent increase in vascular conductance above resting, control level, plotted versus the corresponding increase of venous hyperosmolality (closed circles). The data are classed with regard to the rate of sympathetic stimulation and given as mean values 2S.E. Resting,
403
OSMOLALITY IN GLANDULAR VASODILATATION
w
M
z
o--e
0
Symp. stimulation Hypertonic infusion
600
4 I-
0 3 0
z
8
500 A
Fig. 2. Diagram showing mean values2S.E. of the dilator responses in the submandibular gland resistance vessels observed immediately after sympathetic nerve stimulation (closed circles) and during La. hypertonic infusion (open circles). The dilator responses are plotted versus concomitantly observed increases of venous osmolality. Stimulation data classed with regard to rate of nerve activation (2-4 Hz, 6 Hz and 16 Hz) and infusion data with regard to evoked increase of venous osmolality.
a w a> -lw
3 J
0
UJ-4
400
4 0
>U
b5 W
8
300
0
eY EO 0:
I
200
I-
z w
0
100
a w
a 0 l'0 2'0 INCREASE OF VENOUS OSMOLALITY ABOVE CONTROL LEVEL mOsm/kg H 0
50
control vascular conductance was quite similar for the 3 groups of data and averaged 39.5 ml/min 100 g x 100 mmHg for the whole material. Control venous osmolality averaged 3 I8 mOsm. It can be seen that the magnitude of the poststimulatory vasodilatation showed a relation to the degree of hyperosmolality. Both parameters were roughly related to the nerve activation rate. After stimulation at 2 4 Hz, the increase of vascular conductance was 177?, and the concomitant osmolar increase about 5 mOsm. At 16 Hz the corresponding figures were 519% and 11 mOsm. It should be mentioned that the mean value for the maximal decrease of vascular conductance evoked during stimulation was 55 yo at 2 4 Hz, 68 Yo at 6 Hz, and 74 Yo at 16 Hz. The corresponding average decrease of vascular conductance during the period of stimulation was less pronounced due to the gradual decline of the constrictor response (cf. Fig. l), the mean value being 41, 54, and 5 1 %, respectively. The latter values, especially, do not differ much implying that the induced blood flow debt was quite similar at the different nerve activation rates. Yet, the poststimulatory dilator response increased considerably with the stimulation frequency (Fig. 2 ) supporting the conclusion that the dilator phenomenon cannot be explained in terms of a reactive hyperemia response ccf. Introduction). The results presented above show that sympathetic activation of the submandibular gland causes a regional hyperosmolar change related in time (Fig. 1, Panel A and B) as well as in magnitude (Fig. 2) to the dilator effect in the resistance vessels associated with the stimulation. Since tissue hyperosmolality per se causes an inhibition of vascular tone in the gland (Lundvall and Holmberg 1974), these experimental findings strongly suggest that the osmolar change is causally related to the dilator response. In a n attempt to obtain a rough evaluation of the quantitative importance of the nerve induced hyperosmolality 26 - 765882
404
JAN HOLMBERG AND JAN LUNDVALL
for the resistance vessel dilatation, data on the resistance effects of intra-arterial hypertonic infusion to the resting gland from our previous investigation have been recalculated so as to be included in Fig. 2 (open circles). These data refer to the initial, peak flow response evoked by hypertonic infusion, which has a similar duration and much the same appearance in general as the hyperemia observed after sympathetic stimulation. A comparison of the two curves in the figure suggests that tissue hyperosmolality can account for a substantial part of the poststimulatory dilatation. Apparently, the factor plays a greater role at lower than at higher nerve activation rates. In additional stimulation experiments (16 Hz), the nerve induced vascular, osmolar and secretory responses were studied before and after i.v. administration of various agents with a-adrenergic blocking properties. The drugs chosen for these experiments have been shown to diminish or abolish either the vasoconstriction, the vasodilatation, or both these vascular responses and to interfere to a varying extent with the secretion evoked by sympathetic activation of the submandibular gland (Emmelin 1955). These findings were confirmed in the present experiments, which further showed that none of the agents used caused a dissociation of the dilator and hyperosmolar responses. After tolazoline (0.1-0.3 mg/kg b.wt.), sympathetic stimulation caused n o vasoconstriction but instead a clear-cut dilatation in the presence of essentially unchanged secretory and hyperosmolar effects (observations made on 2 of the animals). Chlorpromazine (0.14.3 mg/kg b.wt.), on the other hand, abolished the secretory, dilator and hyperosmolar responses but augmented, if anything, the vasoconstriction during stimulation (2 animals). Dihydroergotamine (300 pg/kg b.wt.) abolished the vascular constrictor and dilator effects, the hyperosmolar change, as well as the salivary secretion (3 animals; the effects of p-blockade had been investigated before administration of dihydroergotamine). 8-adrenoceptor blockade (propranolol300 pg/kg b.wt. i.v.), finally, caused some reduction of both the vasodilator and the hyperosmolar responses associated with sympathetic activation (16 Hz; 3 animals).
Discussion The present experiments showed that sympathetic activation of the submandibular gland causes a locally produced tissue hyperosmolality, the magnitude of which was related to the nerve excitation rate. The presented data, when taken together, strongly suggest that this hyperosmolality is an important mediator of the dilator response in the resistance vessels associated with sympathetic stimulation. An increased osmolality causing vasodilatation is present also during parasympathetic activation of the gland, but the degree of hyperosmolality, like the hyperemia response, is then more pronounced (Lundvall and Holmberg 1974). The hyperosmolality, in all probability, is primarily created in the intracellular compartment during both sympathetic and parasympathetic nerve activation as a result of the raised glandular metabolism. A secondary hyperosmolality then develops in the interstitial space, from where it can affect the tone of the vascular smooth muscle, and is later reflected in the venous effluent. Experimental evidence for such a train of events was presented with regard to the hyperosmolality in active skeletal muscle (Lundvall 1972). Besides the direct neurogenic increase of metabolism, the magnitude of the regional blood flow may also
OSMOLALITY IN GLANDULAR VASODILATATION
405
affect the osmolar production in the gland. During sympathetic activation the flow reduction may thus tend to increase anaerobic metabolism and thereby the hyperosmolality. Such an effect is apparently small, however, since the hyperosmolality evoked during stimulation was clear-cut also after blockade of the constrictor response with tolazoline. The relative importance of tissue hyperosmolality for the dilator response in the resistance vessels evoked by sympathetic stimulation was estimated by comparing, at equal levels of venous hyperosmolality, the increase of vascular conductance observed upon cessation of nerve stimulation with that evoked by intra-arterial hypertonic infusion to the resting gland (Fig. 2, cf. Lundvall 1972, Lundvall and Holmberg 1974). The data indicated that the hyperosmolality contributes significantly to the vasodilatation at all nerve activation rates; at low to moderate stimulation rates it apparently may be the dominating causal factor. One special methodological difficulty met with in the present experiments should be discussed briefly in this context since it is of relevance for this estimation of the quantitative role of hyperosmolality in the dilatation. The blood sample volume required for each osmolality determination was thus fairly large (about 1.0 ml) in relation to the absolute blood flow levels encountered in the submandibular gland. Since the maximal poststimulatory hyperemia was quite short, the hyperosmolar values depicted in the diagram of Fig. 2 therefore often had to represent measurements on blood sampled not only during the peak hyperemia but also during the disappearance of the vasodilatation and sometimes after its abolition. This, in all probability, must have led to a sometimes serious underestimation of the hyperosmolality present during the peak hyperemia response because of the rapid post-stimulatory osmolar wash-out (cf. Fig. 1, panel A). Accordingly, the comparison of the sympathetic stimulation and hypertonic infusion curves in Fig. 2 may tend to underestimate the role of hyperosmolality for the nerve induced vasodilatation. Nevertheless, tissue hyperosmolality in all likelihood is not the sole causal factor for the hyperemia response. Apparently, several dilator factors contribute to the hyperemia by mutual, synergistic actions. Besides tissue hyperosmolality, the previously proposed bradykinin and B-adrenergic vasodilator fibre mechanisms seem to be involved as indicated by several investigations. It appears possible, however, that the /?-adrenergic vasodilatation, at least partly, could be metabolically linked, since recent findings have indicated the presence of a previously unknown P-adrenoceptor mediated secretion in the cat submandibular gland (Emmelin and Gjorstrup 1975). The present observation of a reduction of both the sympathetic vasodilator and hyperosmolar responses after ,&blockade may support such a hypothesis. This study was supported by grants from the Swedish Medical Research Council (B76-14X-2210-10B) and from the Faculty of Medicine, University of Lund.
References BARCROFT, J. and H. PIPER,The gaseous metabolism of the submaxillary gland with reference especially to the effect of adrenalin and the time relation of the stimulus of the oxidation process. J . Physiol. (Lond.) 1912. 44. 359-373. BHOOLA,K. D., J. MORLEY, M. SCHACHTER and L. H. SMAJE, Vasodilatation in the submaxillary gland of the cat. J. Physiol. (Lond.) 1965. 179. 172-184.
406
JAN HOLMBERG AND JAN LUNDVALL
CARLSON, A. J., Vaso-dilator fibres to the submaxillary gland in the cervical sympathetic of the cat. Amer. J. Physiol. 1907. 19. 408416. EMMELIN, N., Sympathicolytic agents used to separate secretory and vascular effects of sympatheticstimulation in the submaxillary gland. Acta physiol. scand. 1955. 34. 29-37. EMMELIN, N. and P. GIORSTRUP, Secretory responses to sympathetic stimulation of the cat's salivary glands in a state of resting secretion. Quart. J. exp. Physiol. 1975. 60. 325-332. GAUTVIK, K. M., M. KRITZ and K. LUND-LARSEN, Plasma-kinins and adrenergic vasodilatation in the submandibular salivary gland of the cat. Acta physiol. scand. 1972 a. 86. 419-426. GAUTVIK, K., M. KRIZ,K. LUND-LARSEN and B. A. WAALER, Sympathetic vasodilatation, kallikrein release and adrenergic receptors in the cat submandibular salivary gland. Acta physiol. scand. 1974. 90.438444. HILTON, S . M. and G. P. LEWIS,The cause of the vasodilatation accompanying activity in the submandibular salivary gland. J. Physiol. (Lond.) 1955. 128. 235-248. HILTON,S. M. and G. P. LEWIS,The relationship between glandular activity, bradykinin formation and functional vasodilatation in the submandibular salivary gland. J. Physiol. (Lond.) 1956. I34. 471-483. LUNDVALL, J., Tissue hyperosmolality as a mediator of vasodilatation and transcapillary fluid flux in exercising skeletal muscle. Acta physiol. scand. 1972. suppl. 379. 1-142. LUNDVALL, J. and J. HOLMBERG, Role of tissue hyperosmolality in functional vasodilatation in the submandibular gland. Acta physiol. scand. 1974. 92. 165-174.
Tissue Hyperosmolality as a Causal Factor in Vasodilatation Following Sympathetic Stimulation of the Submandibular Gland BY
JANHOLMBERC and JANLUNDVALL Received 8 March 1976
Abstract HOLMBERG, J. and J. LUNDVALL. Tissue hyperosmolality as a causal factor in vasodilatation following sympathetic stimulation of the submandibular gland. Acta physiol. scand. 1976. 98. W 0 6 . In a previous investigation, parasympathetic activation of the submandibular gland in the cat was found to cause a considerable increase of regional tissue osmolality, the degree of which was related to the evoked functional hyperernia; intra-arterial hypertonic infusion t o the resting gland producing tissue hyperosmolality of similar magnitudes caused graded and marked dilatations (Lundvall and Holmberg 1974). It was concluded that hyperosmolality contributes significantly to the functional hyperemia response. In the present study evidence is presented to indicate that tissue hyperosmolality is a mediator of the dilatation associated with sympathetic activation as well. An increase of tissue hyperosmolality, as traced in the venous effluent, was found at all frequencies of sympathetic stimulation (2-16 Hz). At high stimulation rates it sometimes exceeded the resting control level by more than 20 mOsm/kg H,O. There was a direct relation between the degree of venous hyperosmolality and the hyperemia response observed immediately after cessation of stimulation. Comparison of the dilator effects evoked by sympathetic stimulation and by hypertonic infusion to the resting gland indicated that tissue hyperosmolality is an important causal factor for the nerve induced dilatation, especially at low and moderate stimulation rates.
Activation of the sympathetic nerves to the cat submandibular gland causes vasoconstriction and a scant salivary secretion. The vasoconstrictor response is most pronounced in the early period of excitation and then usually subsides gradually. Upon cessation of stimulation quite a marked vasodilatation is observed, a phenomenon which only to a minor extent can be explained in terms of a reactive hyperemia response (e.g. Hilton and Lewis 1955). Several investigations have tried to elucidate the mechanismts) responsible for the gradual abolition of the constrictor response during stimulation and for the poststimulatory vasodilatation. The problem may thus not only be of academic interest, but may have general implications with regard to blood flow control in glandular tissue. The dilator effect was attributed to specific sympathetic vasodilator nerves already by Carlson (1907), and Bhoola et al. (1965) obtained evidence to indicate that such a direct
OSMOLALITY IN GLANDULAR VASODILATATION
40 1
neurogenic inhibition of vascular tone was mediated via /3-adrenoceptor activation. On the other hand, Barcroft and Piper (1912) suggested that the vasodilatation was caused by metabolites released during the sympathetic activation of the glandular cells. Hilton and Lewis (1956) and later Gautvik and associates (1972, 1973) presented experimental support for the hypothesis that a “specific” metabolite, the vasoactive polypeptide bradykinin, was causally involved in the reaction. A “non-specific” metabolic factor, regional tissue hyperosmolality, was recently found to play a significant role in the vasodilatation caused by parasympathetic stimulation of the cat submandibular gland (Lundvall and Holmberg 1974). Briefly, it was shown that parasympathetic activation was associated with an increased tissue osmolality as reflected in the venous effluent, and the degree of this osmolar change was related to the magnitude of the evoked functional hyperemia. Intra-arterial infusion of hypertonic solutions to the resting gland producing comparable levels of venous hyperosmolality caused marked vasodilatation. The present study shows that also sympathetic activation of the submandibular gland is associated with an increased tissue osmolality. A comparison with the hypertonic infusion data from the aforementioned report indicates that this osmolar environmental change contributes significantly to the dilator phenomena during and after syrnpathetic nerve stimulation.
Methods 9 cats of both sexes, weighing 2.4-3.6 kg were used. The animals were anesthetized i.v. with chloralose (SO mg/kg b.wt.) and urethane (100 mg/kg b.wt.) after induction with ether. The submandibular gland was exposed unilaterally and its salivary duct cannulated. The chorda-lingual nerve was cut. The main vein of the gland, emptying into the external jugular vein, was carefully isolated and other, minute, veins draining the gland were ligated. After heparinization (3-5 mg/kg b.wt.), the external jugular vein was cannulated and all its tributaries except that from the gland were ligated. Venous outflow from the gland was diverted through an optical drop-recorder unit for continuous registration of blood flow and then returned to the animal via a funnel connected to the right femoral vein. Venous outflow pressure was adjusted to heart level. Arterial inflow pressure was monitored from the right femoral artery. The distal end of the severed cervical sympathetic trunk was stimulated proximal to the superior cervical ganglion for 2-5 min with bipolar ring electrodes. Supramaximal square wave shocks (5-7 V, 2 ms) at frequencies of 2-16 Hz were used. Venous samples were collected at intervals for determination of plasma osmolality. In some animals, arterial samples were taken as well from a cannula in the brachial artery. The “dead space” fluid volume of the venous outflow tubing and of the arterial cannula was allowed for during sampling. Plasma osmolality was determined by thermistor cryoscopy (Osmometer 3 1 LAS, Advanced Instruments, Inc.). Each sample was measured twice and if, occasionally,different readings were obtained the mean value was used. Repetitive measurements o n osmolar standards deviated at most by 1.5 mOsm/kg H 2 0 from the true value. Arterial blood pressure, glandular blood flow, and submandibular secretion [the latter registered a s drops of known volume) were recorded on a Grass Polygraph. The following drugs were given i.v.: tolazoline (PriscoP), chlorpromazine (Hibemalt@),dihydroergotamine (Orstanorm@’), and propranolol (InderaP).
Results Sympathetic activation of the submandibular gland was invariably found to be associated with an increase of plasma osmolality in the venous effluent from the gland. The increased osmolality was present both during and immediately after stimulation in the great majority
402
JAN HOLMBERG AND JAN LUNDVALL
B
A ARTERIAL BLOOD PRESSURE mmHg
150 1001
dOsec
-6
TIME DROPS OF SALIVA
SIGNAL
INCREASE OF VENOUS OSMOLALITY mOsm/kg H 2 0
S Y M P STlM 1 6 H Z
S Y M P STIM 16 t i 1
0
7
9
2
0
0
4
8
11
BLOOD FLOW ml/minx100g
69.9 6.8
24.5 240.0
60.1
6.9
36.5
96.4
Fig. 1. Changes of blood flow and regional venous osmolality in the cat submandibular gland evoked by short (Panel A) and more prolonged (Panel B) sympathetic nerve activation (16 Hz). The induced saliva secretion (each drop 0.06 ml) is also shown.
of the expts. In 4 of the stimulation experiments (total number =43), however, it appeared only after cessation of the nerve activation. The venous hyperosmolality, reflecting an increased osmolality in the active gland, was caused by local osmolar production since arterial osmolality was concomitantly unchanged. The degree of hyperosmolality was graded in relation to the rate of sympathetic stimulation and at 16 Hz it sometimes could amount to more than 20 mOsm. Fig. 1 shows original records illustrating the relation in time between the changes of blood flow and venous osmolality evoked by sympathetic stimulation. The induced salivary secretion is also shown (drop volume 0.06 ml). In panel A, stimulation of relatively short duration (16 Hz, 2 min) caused a rapid and pronounced decrease of blood flow, which was fairly well maintained during the period of stimulation. The gland was active during stimulation producing a scanty secretion of saliva. Cessation of the stimulation evoked a short-lasting but quite marked hyperemia response, corresponding to a 227 % increase of vascular conductance above the resting, control level. Venous osmolality rose by 7 mOsm in the later phase of stimulation and a peak increase of 9 mOsm was observed at the time of the poststimulatory hyperemia. It then rapidly returned to the control level. The initial pronounced constriction in the resistance vessels was often found to decline quite significantly in a gradual manner despite continuous stimulation, and this effect was most marked in response to more prolonged sympathetic activation as illustrated in Fig. 1, Panel B (16 Hz, 5 min). The constriction was here completely abolished after about 3 min of stimulation and was reversed to vasodilatation towards the end of the activation period. The abolition of the constrictor response apparently cannot be ascribed to a failing transmission of the nerve impulses since the salivation was well maintained during the period of stirnulation. It can be seen that venous osmolality, in parallel to the decline of the constrictor response, rose gradually to a level 11 mOsm above the prestimulatory, control value. The majority of the sympathetic stimulations were performed for a period of about 2 min and the results obtained in these experiments are summarized in Fig. 2. The diagram shows the peak hyperemia response observed upon cessation of nerve activation, expressed as the per cent increase in vascular conductance above resting, control level, plotted versus the corresponding increase of venous hyperosmolality (closed circles). The data are classed with regard to the rate of sympathetic stimulation and given as mean values 2S.E. Resting,
403
OSMOLALITY IN GLANDULAR VASODILATATION
w
M
z
o--e
0
Symp. stimulation Hypertonic infusion
600
4 I-
0 3 0
z
8
500 A
Fig. 2. Diagram showing mean values2S.E. of the dilator responses in the submandibular gland resistance vessels observed immediately after sympathetic nerve stimulation (closed circles) and during La. hypertonic infusion (open circles). The dilator responses are plotted versus concomitantly observed increases of venous osmolality. Stimulation data classed with regard to rate of nerve activation (2-4 Hz, 6 Hz and 16 Hz) and infusion data with regard to evoked increase of venous osmolality.
a w a> -lw
3 J
0
UJ-4
400
4 0
>U
b5 W
8
300
0
eY EO 0:
I
200
I-
z w
0
100
a w
a 0 l'0 2'0 INCREASE OF VENOUS OSMOLALITY ABOVE CONTROL LEVEL mOsm/kg H 0
50
control vascular conductance was quite similar for the 3 groups of data and averaged 39.5 ml/min 100 g x 100 mmHg for the whole material. Control venous osmolality averaged 3 I8 mOsm. It can be seen that the magnitude of the poststimulatory vasodilatation showed a relation to the degree of hyperosmolality. Both parameters were roughly related to the nerve activation rate. After stimulation at 2 4 Hz, the increase of vascular conductance was 177?, and the concomitant osmolar increase about 5 mOsm. At 16 Hz the corresponding figures were 519% and 11 mOsm. It should be mentioned that the mean value for the maximal decrease of vascular conductance evoked during stimulation was 55 yo at 2 4 Hz, 68 Yo at 6 Hz, and 74 Yo at 16 Hz. The corresponding average decrease of vascular conductance during the period of stimulation was less pronounced due to the gradual decline of the constrictor response (cf. Fig. l), the mean value being 41, 54, and 5 1 %, respectively. The latter values, especially, do not differ much implying that the induced blood flow debt was quite similar at the different nerve activation rates. Yet, the poststimulatory dilator response increased considerably with the stimulation frequency (Fig. 2 ) supporting the conclusion that the dilator phenomenon cannot be explained in terms of a reactive hyperemia response ccf. Introduction). The results presented above show that sympathetic activation of the submandibular gland causes a regional hyperosmolar change related in time (Fig. 1, Panel A and B) as well as in magnitude (Fig. 2) to the dilator effect in the resistance vessels associated with the stimulation. Since tissue hyperosmolality per se causes an inhibition of vascular tone in the gland (Lundvall and Holmberg 1974), these experimental findings strongly suggest that the osmolar change is causally related to the dilator response. In a n attempt to obtain a rough evaluation of the quantitative importance of the nerve induced hyperosmolality 26 - 765882
404
JAN HOLMBERG AND JAN LUNDVALL
for the resistance vessel dilatation, data on the resistance effects of intra-arterial hypertonic infusion to the resting gland from our previous investigation have been recalculated so as to be included in Fig. 2 (open circles). These data refer to the initial, peak flow response evoked by hypertonic infusion, which has a similar duration and much the same appearance in general as the hyperemia observed after sympathetic stimulation. A comparison of the two curves in the figure suggests that tissue hyperosmolality can account for a substantial part of the poststimulatory dilatation. Apparently, the factor plays a greater role at lower than at higher nerve activation rates. In additional stimulation experiments (16 Hz), the nerve induced vascular, osmolar and secretory responses were studied before and after i.v. administration of various agents with a-adrenergic blocking properties. The drugs chosen for these experiments have been shown to diminish or abolish either the vasoconstriction, the vasodilatation, or both these vascular responses and to interfere to a varying extent with the secretion evoked by sympathetic activation of the submandibular gland (Emmelin 1955). These findings were confirmed in the present experiments, which further showed that none of the agents used caused a dissociation of the dilator and hyperosmolar responses. After tolazoline (0.1-0.3 mg/kg b.wt.), sympathetic stimulation caused n o vasoconstriction but instead a clear-cut dilatation in the presence of essentially unchanged secretory and hyperosmolar effects (observations made on 2 of the animals). Chlorpromazine (0.14.3 mg/kg b.wt.), on the other hand, abolished the secretory, dilator and hyperosmolar responses but augmented, if anything, the vasoconstriction during stimulation (2 animals). Dihydroergotamine (300 pg/kg b.wt.) abolished the vascular constrictor and dilator effects, the hyperosmolar change, as well as the salivary secretion (3 animals; the effects of p-blockade had been investigated before administration of dihydroergotamine). 8-adrenoceptor blockade (propranolol300 pg/kg b.wt. i.v.), finally, caused some reduction of both the vasodilator and the hyperosmolar responses associated with sympathetic activation (16 Hz; 3 animals).
Discussion The present experiments showed that sympathetic activation of the submandibular gland causes a locally produced tissue hyperosmolality, the magnitude of which was related to the nerve excitation rate. The presented data, when taken together, strongly suggest that this hyperosmolality is an important mediator of the dilator response in the resistance vessels associated with sympathetic stimulation. An increased osmolality causing vasodilatation is present also during parasympathetic activation of the gland, but the degree of hyperosmolality, like the hyperemia response, is then more pronounced (Lundvall and Holmberg 1974). The hyperosmolality, in all probability, is primarily created in the intracellular compartment during both sympathetic and parasympathetic nerve activation as a result of the raised glandular metabolism. A secondary hyperosmolality then develops in the interstitial space, from where it can affect the tone of the vascular smooth muscle, and is later reflected in the venous effluent. Experimental evidence for such a train of events was presented with regard to the hyperosmolality in active skeletal muscle (Lundvall 1972). Besides the direct neurogenic increase of metabolism, the magnitude of the regional blood flow may also
OSMOLALITY IN GLANDULAR VASODILATATION
405
affect the osmolar production in the gland. During sympathetic activation the flow reduction may thus tend to increase anaerobic metabolism and thereby the hyperosmolality. Such an effect is apparently small, however, since the hyperosmolality evoked during stimulation was clear-cut also after blockade of the constrictor response with tolazoline. The relative importance of tissue hyperosmolality for the dilator response in the resistance vessels evoked by sympathetic stimulation was estimated by comparing, at equal levels of venous hyperosmolality, the increase of vascular conductance observed upon cessation of nerve stimulation with that evoked by intra-arterial hypertonic infusion to the resting gland (Fig. 2, cf. Lundvall 1972, Lundvall and Holmberg 1974). The data indicated that the hyperosmolality contributes significantly to the vasodilatation at all nerve activation rates; at low to moderate stimulation rates it apparently may be the dominating causal factor. One special methodological difficulty met with in the present experiments should be discussed briefly in this context since it is of relevance for this estimation of the quantitative role of hyperosmolality in the dilatation. The blood sample volume required for each osmolality determination was thus fairly large (about 1.0 ml) in relation to the absolute blood flow levels encountered in the submandibular gland. Since the maximal poststimulatory hyperemia was quite short, the hyperosmolar values depicted in the diagram of Fig. 2 therefore often had to represent measurements on blood sampled not only during the peak hyperemia but also during the disappearance of the vasodilatation and sometimes after its abolition. This, in all probability, must have led to a sometimes serious underestimation of the hyperosmolality present during the peak hyperemia response because of the rapid post-stimulatory osmolar wash-out (cf. Fig. 1, panel A). Accordingly, the comparison of the sympathetic stimulation and hypertonic infusion curves in Fig. 2 may tend to underestimate the role of hyperosmolality for the nerve induced vasodilatation. Nevertheless, tissue hyperosmolality in all likelihood is not the sole causal factor for the hyperemia response. Apparently, several dilator factors contribute to the hyperemia by mutual, synergistic actions. Besides tissue hyperosmolality, the previously proposed bradykinin and B-adrenergic vasodilator fibre mechanisms seem to be involved as indicated by several investigations. It appears possible, however, that the /?-adrenergic vasodilatation, at least partly, could be metabolically linked, since recent findings have indicated the presence of a previously unknown P-adrenoceptor mediated secretion in the cat submandibular gland (Emmelin and Gjorstrup 1975). The present observation of a reduction of both the sympathetic vasodilator and hyperosmolar responses after ,&blockade may support such a hypothesis. This study was supported by grants from the Swedish Medical Research Council (B76-14X-2210-10B) and from the Faculty of Medicine, University of Lund.
References BARCROFT, J. and H. PIPER,The gaseous metabolism of the submaxillary gland with reference especially to the effect of adrenalin and the time relation of the stimulus of the oxidation process. J . Physiol. (Lond.) 1912. 44. 359-373. BHOOLA,K. D., J. MORLEY, M. SCHACHTER and L. H. SMAJE, Vasodilatation in the submaxillary gland of the cat. J. Physiol. (Lond.) 1965. 179. 172-184.
406
JAN HOLMBERG AND JAN LUNDVALL
CARLSON, A. J., Vaso-dilator fibres to the submaxillary gland in the cervical sympathetic of the cat. Amer. J. Physiol. 1907. 19. 408416. EMMELIN, N., Sympathicolytic agents used to separate secretory and vascular effects of sympatheticstimulation in the submaxillary gland. Acta physiol. scand. 1955. 34. 29-37. EMMELIN, N. and P. GIORSTRUP, Secretory responses to sympathetic stimulation of the cat's salivary glands in a state of resting secretion. Quart. J. exp. Physiol. 1975. 60. 325-332. GAUTVIK, K. M., M. KRITZ and K. LUND-LARSEN, Plasma-kinins and adrenergic vasodilatation in the submandibular salivary gland of the cat. Acta physiol. scand. 1972 a. 86. 419-426. GAUTVIK, K., M. KRIZ,K. LUND-LARSEN and B. A. WAALER, Sympathetic vasodilatation, kallikrein release and adrenergic receptors in the cat submandibular salivary gland. Acta physiol. scand. 1974. 90.438444. HILTON, S . M. and G. P. LEWIS,The cause of the vasodilatation accompanying activity in the submandibular salivary gland. J. Physiol. (Lond.) 1955. 128. 235-248. HILTON,S. M. and G. P. LEWIS,The relationship between glandular activity, bradykinin formation and functional vasodilatation in the submandibular salivary gland. J. Physiol. (Lond.) 1956. I34. 471-483. LUNDVALL, J., Tissue hyperosmolality as a mediator of vasodilatation and transcapillary fluid flux in exercising skeletal muscle. Acta physiol. scand. 1972. suppl. 379. 1-142. LUNDVALL, J. and J. HOLMBERG, Role of tissue hyperosmolality in functional vasodilatation in the submandibular gland. Acta physiol. scand. 1974. 92. 165-174.
