Dehydration elevates osmotic threshold salt gland secretion in the duck R. KAUL
AND
for
H. T. HAMMEL
Physiological Research Laboratory, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, California 92093 KAUL, R., AND H. T. HAMMEL. Dehydration elevates osmotic threshold for salt gland secretion in the duck. Am. J. Physiol. 237(5): R355R359, 1979 or Am. J. Physiol.: Regulatory Integrative Comp. Physiol. 6(3): R3554359,1979.-Acute salt and water balance measurements were made in two conscious salt water-acclimated Pekin ducks at and-.-above their osmotic threshold for salt gland secretion. Intravenous infusion of 1,000 mosmol/kg Hz0 NaCl at 0.350 ml/min increased plasma tonicity less than 0.5% and increased secretion from nearly zero to a rate matching the infusion. Continuous secretion at a similar submaximal rate was driven by 5,600 mosmol/kg Hz0 NaCl infused at 0.070 ml/min. Osmolality of secreted fluid was constant for any secretion rate, so that net water loss occurred when the concentration of infusate exceeded that of secreted fluid. Threshold plasma osmolality increased by 9 mosmol/kg Hz0 after the loss of 77 g water (3% body wt). Solutes were always secreted at the infusion rate, even when body fluid osmolality increased while body water decreased: We conclude that the salt gland controller is sensitive to more than just extracellular fluid (ECF) tonicity, and we suggest that elevation of the osmotic threshold may occur in response to decreased ECF volume. sodium excretion; osmoregulation; osmoreceptor; volume receptor
extracellular
fluid
volume;
tors are not required to mediate the secretory response. A problem in all of these studies is that the simple occurrence or absence of activity is a very limited basis from which to infer the nature of control. Consequently, we have investigated the effects of altered ECF volume and tonicity by employing submaximal rates of secretion to indicate the output of the salt gland controller. In the following experiment ducks are challenged with continuous intravenous infusion of hypertonic NaCl solutions, known to drive continuous matching salt gland secretion (8,9), at rates far below the maximum secretory capacity of the glands. Assessment of salt and water balance allows calculation of extracellular osmolality and volume, while the concurrent rate of secretion presumably reflects the magnitude of the secretomotor signal. The concentration of the secreted fluid is approximately constant for any secretory rate and the rate of solute secretion always matches that of solute infusion. Thus, absolute dehydration occurs when the infused concentration exceeds that of the secreted fluid. Decreased ECF volume is accompanied by an increased plasma osmotic threshold for secretion, indicating that the output of the salt gland controller requires reciprocal integration of both osmotic and volumetric input.
MORE THAN TWENTY YEARS AGO Schmidt-Nielsen,Jorgensen, and Osaki (16) discovered that hypertonic fluid
MATERIALS
secreted by cranial salt glands is a major route of electrolyte excretion in marine birds. The properties of the body fluids transduced in the control of secretion are still not well understood, and the role of extracellular fluid (ECF) volume remains particularly controversial (13). The original studies conducted by Schmidt-Nielsen’s group clearly showed that salt glands are activated by secretomotor efferents from the central nervous system when ECF tonicity is increased by intravenous infusion of hypertonic solutions (3). McFarland (12), though, noted the large increase in blood volume produced by such infusions and Holmes (11) proposed that volume expansion, rather than elevated tonicity per se, is the stimulus that triggers secretion. Isotonic volume expansion has been variously reported to induce secretion (11, 19), to not induce secretion (7, lo), and to induce transient secretion (5). Stewart (17) dissociated volume expansion from hypertonic stress by withholding water from ducks. Evaporative water loss removed water from the ECF, increasing concentration while decreasing volume. He concluded that because secretion occurred volume recep-
The experiment was performed three times on each of two adult Pekin ducks (Anas platyrhynchos). The birds were maintained outdoors with free access to breeders lay-pellets and to water. The water provided for drinking and grooming was seawater diluted 1:l with tap water525 mosmol/kg HZO, [Na] = 236 meq/l and [K] = 5.2 meq/l. The ducks were held on this regimen at least 30 days before the first experiment and at least 1 wk elapsed between experiments on the same bird. Before the series of experiments was started the weight of duck 1 had stabilized at 2.76 kg and that of duck 2 at 2.42 kg. During the month in which experiments were performed the weight of each duck varied less than 2% both above and below their means. At about 9 A.M. on the day of an experiment a duck was brought inside, where a percutaneous venipuncture of the right jugular vein was made with a 16 gauge Deseret Angiocath. A sterile PE-50 catheter was inserted through the lumen, advanced 10 to 15 cm toward the heart, and the Angiocath was withdrawn. The ducks tolerated this procedure with no obvious distress. The
0363-6119/79/oooO-~$o1.25
Copyright
@ 1979 the American
Physiological
Society
AND
METHODS
R355
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
R356
R.
external nares were then sealed with waterproof adhesive tape, the tape over each nostril being penetrated about a millimeter by l-mm ID soft plastic tubing. The two pieces of tubing joined at a Y attached to a stopper that fitted tared collection vials. Pressure in the vials was held at -200 Torr so that all salt gland secretion was aspirated and collected. Vials were changed at 15min intervals. During an experiment the conscious and alert duck was free to sit or stand in a small oval cage with a large-mesh wire floor. The cage was fitted to a large oval fiberglass funnel so that any cloacal discharge was collected in tared beakers. Within 15 min of voiding, cloacal discharge was weighed, centrifuged, and the fluid portion decanted and saved for analysis. Because blood samples were drawn from the same catheter through which infusions were made, 1 ml of rinsing blood was drawn before each sample. After drawing a Z-ml sample the rinsing blood was reinfused. Blood samples were immediately centrifuged and the plasma osmolality measured. Plasma samples were stored frozen for later determination of [Na] and [K]. Osmolality of all samples was determined by freezing point depression (Knauer Halbmikro-Osmometer), and the sodium and potassium concentrations by flame photometry (Instrumentation Labotatory model 143). Before each experiment the duck was induced to secrete by infusion of 1,000 mosmol/kg Hz0 NaCl at 0.350 or 0.700 ml/min for 30 to 90 min. After the establishment of secretion the infusion was stopped, and when the secretion rate fell to 0.02 mosmol/min or less, a blood sample was drawn for determination of threshold plasma osmolality and sodium concentration. To begin the experiment 1,000 mosmol/kg Hz0 NaCl was infused at 0.350 ml/min for 90 min and the interval from the start of this infusion until the rate of salt gland secretion again fell to or below 0.02 mosmol/min is called the initial control period. Earlier experiments had shown these birds able to secrete salt and water at a rate greater than twice this control infusion rate. Following the initial control period another blood sample was drawn and 5,600 mosmol/kg Hz0 NaCl was infused at 0.070 ml/min for 240 min. When the secretion rate fell to below 0.02 mosmol/min after this infusion, another blood sample was drawn and a postdehydration control infusion, identical to the first, was administered and a final blood sample drawn. The differences between the amounts of salt and water infused and the amounts secreted by the salt gland were used to calculate changes in body fluid osmolality and TABLE I. Excretion of solute and Hz0 by salt glands infusion of hypertonic saline
__~
mosmol mosmol
solute solute
g water g water
-_--Values
for
solute
secreted infused
excreted infused and water
secreted
are means
AND
H.
T. HAMMEL
ECF volume during each 15,min collection period. Because the solutes in the secreted fluid were nearly all sodium and chloride (see RESULTS) the reasonable assumption was made that the net solute load was restricted to the extracellular space. It was further assumed that the net water load distributed freely throughout the total body water (TBW) to maintain osmotic equilibrium between the intracellular and extracellular compartments. In beginning these calculations the osmolality of the body fluids was set at the measured threshold plasma osmolality of the primed and ready-to-secrete duck, and the TBW and ECF volumes were assumed to be, respectively, 64.0 and 26.4% of the body weight, as determined by Ruth and Hughes (15) from the 8”Br and ‘Hz0 spaces in salt water-acclimated Pekin ducks. RESULTS
The infusion of hypertonic NaCl solutions into ducks already at the threshold of secretion stimulates the salt glands to quantitatively secrete the amount of solute with which the ducks are loaded. While the total amount of solute secreted usually slightly exceeded the amount of NaCl infused, Table 1 shows that for each of the three hypertonic challenges solute secreted was within about 10% of solute infused. Because the osmolality of the control infusate approximated that of the secreted fluid, a similar correspondence was achieved between the 2. Effect of dehydration on plasma threshold osmolality for salt gland secretion ______~ --
TABLE
Plasma
Threshold
Osmolality,
Before
dehydration
mosmol/kg
After ~~--_
H,O
dehydration -__--
-
Start control
End control
Start control
End cont rol
Duck I 4-24 5-01 5-31
333 317 306
332 318 304
344 326 314
340 324 310
Duck 2 4-26 5-04 5-26
331 320 317
335 318 318
344 327 328
344 11 328 9 329 11 --~.-. ________ Plasma osmolality measured when secretion rate 0.02 mosmol/min or less before and after control infusions (1,000 mosmol/kg Hz0 NaCl at 0.350 ml/min for 90 min). Acute absolute dehydration produced by secretory response to 5,600 mosmol/kg Hz0 NaCl at 0.070 ml/min for 240 min.
in response to intravenous
Initial Control, 1,000 mosmol/kg H1O at 0.350 ml/min for 90 min
-p___--__-_-.--
KAUL
__._- .__.-._~___...~~__ --- _ . Dehydrating Infusion, !5,6W mosmol/ kg H,O at 0.070 ml/min for 240 min
~-~__-_-----
_ _ ~~-
-_...
Dehydrated Control, 1,000 mosmol/kg Hz0 at 0.350 ml/min for 90 min
_~--~_.-~_- _ _._-_._-_ ____
33.5 t 1.0 =1.08 31.1
98.6 t 2.7 =l.ll 88.9
34.0 +, 1.6 =1.09 31.1
33.4 +, 1.7 =1.08 31.1
93.2 +_ 4.3 =5.86 15.9 __-_____. -_____
32.1 +, 2.3 _.--_
31.1
=1.03
t SE (n = 6).
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
OSMOTIC
THRESHOLD
FOR
SALT
GLAND
R357
SECRETION
amounts of water infused and excreted during the control periods. When a similar rate of solute secretion was driven by infusion of 5,600 mosmol/kg Hz0 NaCl at an appropriately lower volume infusion rate, the salt glands excreted nearly six times as much water as was infused. The net loss of 77.3 t 4.3 g water represents about 5% of the TBW of a 2.6-kg duck. Table 2 shows the effects of the net solute and water exchanges on the threshold plasma osmolality directly measured before and after each infusion. Because of the large variability, even among thresholds for the same duck on different days, individual values rather than means are presented. Despite variability in threshold osmolality from day to day, paired t testing shows that there is no significant difference between values before and after the control infusions (P > 0.1). Following dehydrating infusion/secretion the osmotic threshold is elevated by 9.1 t 0.7 mosmol/kg Hz0 (mean t SE, significant at 0.0005 level by paired t test). In each experiment the sodium concentration (meq/l) of the salt gland secretion could be expressed to within 1% as osmolality (mosmol/kg HZO) divided by 1.883. This divisor is just 2.4% larger than that for pure NaCl solutions of similar concentrations, and the Na-to-K ratio of the secreted &id (constant over the duration of the experiments) was 38.8 t 0.9 (mean t SE, n = 19), i.e., only 2.6% of the measured cation is K+. It therefore seems that the fluid secreted is, to within 2 or 3%, a pure NaCl solution. This means that solute balance is equivalent to sodium balance, and Fig. 1 shows that changes in plasma osmolality accurately reflect changes in plasma sodium concentration. For these plasma samples meq Na/l = 0.381 (mosmol/kg HZO) + 32.0, r” = 0.93. This very strong correlation shows that increased threshold osmolality reflects a true increase in threshold tonicity, and is not an artifact attributable to a stress-induced elevation of plasma glucose or some other permeant solute not effective in driving secretion. Plasma osmolalities and ECF volumes for each experiment have been expressed as 100% at their values in the primed and ready-to-secrete condition (before fust control), and in Fig. 2 the mean values are plotted over the concurrent mean rates of salt gland secretion. The mean values of the directly measured plasma osmolalities are also plotted to show the close agreement between calculated and directly measured osmolality. Figure 2 clearly shows that 1) the loss of extracellular water produced by the secretory response to 5,600 mosmol/kg Hz0 NaCl is associated with an elevation of the threshold osmolality for secretion in response to a control challenge; 2) salt gland activity occurs at an appropriate submaximal rate during dehydration, though plasma osmolality is continually rising as ECF volume is decreasing; and 3) the second control response is indistinguishable from the first, though the threshold plasma osmolality has evidently been reset at a higher level. A final result worthy of note is that during these experiments one duck never produced any cloacal discharge while the other voided only small amounts of fluid with a sodium concentration always much less than that of the plasma. It is clear that under these conditions the
r
170
e-31 A5-I
OUCK
I 310
140 I 300
I
I 320
I 330
I 340
I 350
mOsm /kg H20 FIG.
1. Plasma
regression
equation
renal-cloacal the excretion
[Na] plotted as a function of osmolality and coefficient of determination).
(see text
system did not contribute significantly if the hypertonic salt load.
for
to
DISCUSSION
Because avian salt glands are not constantly active it may be presumed that for any controlling signal there is a threshold level that must be exceeded before the glands are activated. When secretory activity is the only data from which receptor stimulation is inferred, it is clear that the properties of receptors are best revealed by investigations conducted at and above the threshold of activity. Figure 2 shows, for example, that control infusions that increase plasma osmolality less than 0.5% are sufficient to increase salt gland activity from nearly zero to a rate of secretion that would clear the entire plasma sodium content in about 2 h if it were not being continuously replaced by infusion. This implies that the osmotic sensitivity of the receptors is at least an order of magnitude greater than indicated by measurement of changes in plasma solute concentration associated with bringing ducks from somewhere below threshold to somewhere above threshold (14). Of much greater significance than this high order of osmotic sensitivity is our determination that the osmotic threshold is variable and that its value is increased when ECF volume is reduced. We perceive that the patterns of secretion observed might be explained by postulating that the salt glands are driven by a proportional controller sensitive to ECF tonicity and that the reference tonicity is inversely proportional to ECF volume. In this way a constant error signal could be generated during the dehydrating infusion/secretion to drive the constant rate of secretion, and the identity of the initial and dehydrated control responses would result from the defense of plasma tonicity with an elevated set point for secretion. Using an experimental protocol similar to that described here, Hammel, Simon-Oppermann, and Simon (unpublished observations) have found that infusion of isotonic saline or autologous blood into ducks already at threshold always induces secretion. Loss of hypertonic fluid in response to isotonic infusion results in dilution of the plasma and expansion of ECF volume. The osmotic
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
R358
R. KAUL
105
AND
H.
T. HAMMEL
r
i..,.“., i...: ~--z...,“~.“.( L.., ““L....... i i...
i
1.., ..... ... EC
V
FIG. 2. Body fluid osmolality and extracellular fluid volume (expressed as 96 of initial values) during salt gland secretion. Top: solid line, body fluid osmolality (BFO); dotted line, extracellular volume (ECV); circles, directly measured plasma osmolality. Bottom: solid line, solute secretion rate + SE, shaded areas, infused solute loads. All values plotted in each half of this figure are means of 6 trials. See text for method used to calculate extracellular volume and osmolality.
h
TIME 1 hours) threshold is apparently lowered because the salt glands still quantitatively excrete a hypertonic control challenge just as they do before ECF dilution and expansion. This demonstration that multiple modalities must be combined in the control of secretion leads us to suggest the following model for salt gland control. Considering the substantial body of evidence since Verney (18) for cranial osmoreceptors subserving the control of antidiuretic and drinking activities in mammalian and submammalian species, as well as the strong evidence for cardiac mechanoreceptors effecting control of ECF volume in mammalian species (2, 4); we hypothesize that the salt gland secretomotor signal is proportional to the difference between the tonicity of some ECF compartment perfusing cranial osmoreceptors,and an adjustable reference tonicity, the value of which is inversely proportional to ECF volume sensed in the heart or great veins. Hanwell, Linzell, and Peaker (10) have obtained convincing evidence that vagal afferents of cardiac origin are important for the induction of secretion in salt loaded geese. These workers believed the vagal signal to be of osmotic origin, but their published results are completely consistent with a lowering of threshold tonicity in response to ECF volume expansion. Further support is the finding by Gilmore et al. (6) that an “attenuated” secretory response to salt loading sometimes occurs in geese following chronic bilateral vagotomy. Our model predicts elevation of the reference tonicity in the absence of a
vagal volume signal. This would reduce the difference between cranial ECF tonicity and the reference value, so that a diminished osmotic error signal drives the attenuated secretory response. Combination of sensitivities to both sodium concentration and to the size of the principal volume of sodium distribution allows the salt gland controller to regulate the amount of sodium in the body. Osmotic release of arginine vasotocin (AVT), the avian antidiuretic hormone, can provide a mechanism by which sodium excretion is partitioned between the renal and cranial routes according to the availability of osmotically free water. Dantzler (1) and his colleagues have shown that this compound is able to effect a massive reduction in glomerular filtration. When hypertonic stress exceeds the renal concentrating capacity, as in the experiments reported here, AVT probably circulates in high concentration so that little or no sodium is filtered for renal excretion. When ducks at the threshold of salt gland activity are challenged with hypotonic NaCl (200 mosmol/kg H20 at 1.75 ml/min for 240 min) circulating AVT levels are evidently lowered and most of the sodium load is renally excreted as dilute urine. However, the expansion of ECF volume evidently lowers the secretory threshold because, throughout the hypotonic infusion, salt gland activity is driven at a rate such that the amount of the sodium load not appearing in the cloacal discharge is secreted by the salt glands (unpublished observations).
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
OSMOTIC
THRESHOLD
FOR
SALT
GLAND
R359
SECRETION
We are indebted to Yvonne Coleman and Carole Mayo for secretarial aid; to Donald Ward for expert technical assistance; and to Dr. Lee Henderson of the San Diego Veterans Administration Hospital, in whose laboratory the flame photometry was performed.
This Grants Received
work was supported in part BNS-76-24053 and GB40176X. 7 August
1978; accepted
by National
in final
form
Science June
Foundation
1979.
REFERENCES 1. DANTZLER, W. H. Some renal glomerular and tubular mechanisms involved in osmotic and volume regulation in reptiles and birds. In: Proc. AZfied Benzon Symposium XI, edited by C. B. Jorgensen and E. Skadhauge. New York: Academic, 1978, p. 187-201. 2. DIRKS, J., J. SEELY, AND M. LEVEY. Control of extracellular fluid volume and the pathophysiology of edema formation. In: The Kidney, edited by B. M. Brenner and F. C. Rector. Philadelphia, PA: Saunders, 1976, p. 495-552. 3. FANGE, R., K. SCHMIDT-NIELSEN, AND M. ROBINSON. Control of secretion from the avian salt gland. Am. J. Physiol. 195: 321-326, 1958. 4. GAUER, 0. H., 3. P. HENRY, AND C. BEHN. The regulation of extracellular fluid volume. Annu. Reu. Physiol. 32: 547-596, 1970. 5. GILMORE, J. P., J. DIETZ, C. GILMORE, AND I. TUCKER. Evidence for a chloride pump in the salt gland of the goose. Comp. Biochem. Physiol. A 56: 121-126, 1976. 6. GILMORE, J. P., C. GILMORE, J. DIETZ, AND I. H. TUCKER. Influence of chronic cervical vagotomy on salt gland secretion in the goose. Comp. Biochem. Physiol. A. 57: 119-121, 1977. 7. HAJJAR, R., F. SATTLER, B. G. ANDERSON, AND G. GWINUP. Definition of the stimulus to secretion of the nasal salt gland of the seagull. Horn. Metab. Res. 2: 35-37, 1970. 8. HAMMEL, H. T., R. KAUL, C. SIMON-OPPERMANN, AND E. SIMON. Sensitivity of osmoreceptors above the threshold for nasal gland secretion in the duck (Abstract). Proc. Znt. Union Physiol. Sci. 13: 302, 1977. 9. HAMMEL, H. T., J. E. MAGGERT, E. SIMON, L. I. CRAWSHAW, AND R. KAUL. Thermoand osmoregulatory responses induced by heating and cooling the rostra1 brainstem of the Adelie penguin. In: Proc. 3rd Symp. Antarctic Biol., edited by G. A. Llano. Houston,
TX: Gulf, 1977, p. 489-500. 10. HANWELL, A., J. L. LINZELL, AND M. PEAKER. The location and nature of the receptors for salt-gland secretion in the goose. J. Physiol. London 226: 453-472, 1972. II. HOLMES, W. N. Some aspects of osmoregulation in reptiles and birds. Arch. Anat. Micros. Morphol. Exp. 54: 491-513, 1965. 12. MCFARLAND, L. 2. Observations on the haematology and blood volume of captive western gulls (Abstract). Proc. 26th Znt. Congr. Zool. 2: 86, 1963. 13. PEAKER, M. Do osmoreceptors or blood volume receptors initiate salt-gland secretion in birds? J. Physiol. London 276: 66P-67P, 1978. 14. PEAKER, M., S. J. PEAKER, A. HANWELL, AND J. L. LINZELL. Sensitivity of receptors for salt gland secretion in the domestic duck and goose. Comp. Biochem. Physiol. A 44: 41-46, 1973. 15. RUCH, F. E., AND M. R. HUGHES. The effects of hypertonic sodium chloride injection on body water distribution in ducks (Anas platyrhynchos), gulls (Larus glaucescens), and roosters (Gallus domesticus). Comp. Biochem. Physiol. A 52: 21-28, 1975. 16. SCHMIDT-NIELSEN, K., C. B. JORGENSEN, AND H. OSAKI. Extrarenal salt excretion in birds. Am. J. Physiol. 193: 101-107, 1958. 17. STEWART, D. J. Secretion by salt gland during water deprivation in the duck. Am. J. Physiol. 223: 384-386, 1972. 18. VERNEY, E. B. The anti-diuretic hormone and the factors which determine its release. Proc. Roy. Sot. London Ser. B 135: 25-106, 1947. 19. ZUCKER, K. H., C. GILMORE, J. DIETZ, AND J. P. GILMORE. Effect of volume expansion and veratrine on salt gland secretion in the goose. Am. J. Physiol. 232: Rl85-189, 1977 or Am. J. Physiol.: Regulatory Integrative Comp. Physiol. 1: Rl85-R189, 1977.
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
AND
for
H. T. HAMMEL
Physiological Research Laboratory, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, California 92093 KAUL, R., AND H. T. HAMMEL. Dehydration elevates osmotic threshold for salt gland secretion in the duck. Am. J. Physiol. 237(5): R355R359, 1979 or Am. J. Physiol.: Regulatory Integrative Comp. Physiol. 6(3): R3554359,1979.-Acute salt and water balance measurements were made in two conscious salt water-acclimated Pekin ducks at and-.-above their osmotic threshold for salt gland secretion. Intravenous infusion of 1,000 mosmol/kg Hz0 NaCl at 0.350 ml/min increased plasma tonicity less than 0.5% and increased secretion from nearly zero to a rate matching the infusion. Continuous secretion at a similar submaximal rate was driven by 5,600 mosmol/kg Hz0 NaCl infused at 0.070 ml/min. Osmolality of secreted fluid was constant for any secretion rate, so that net water loss occurred when the concentration of infusate exceeded that of secreted fluid. Threshold plasma osmolality increased by 9 mosmol/kg Hz0 after the loss of 77 g water (3% body wt). Solutes were always secreted at the infusion rate, even when body fluid osmolality increased while body water decreased: We conclude that the salt gland controller is sensitive to more than just extracellular fluid (ECF) tonicity, and we suggest that elevation of the osmotic threshold may occur in response to decreased ECF volume. sodium excretion; osmoregulation; osmoreceptor; volume receptor
extracellular
fluid
volume;
tors are not required to mediate the secretory response. A problem in all of these studies is that the simple occurrence or absence of activity is a very limited basis from which to infer the nature of control. Consequently, we have investigated the effects of altered ECF volume and tonicity by employing submaximal rates of secretion to indicate the output of the salt gland controller. In the following experiment ducks are challenged with continuous intravenous infusion of hypertonic NaCl solutions, known to drive continuous matching salt gland secretion (8,9), at rates far below the maximum secretory capacity of the glands. Assessment of salt and water balance allows calculation of extracellular osmolality and volume, while the concurrent rate of secretion presumably reflects the magnitude of the secretomotor signal. The concentration of the secreted fluid is approximately constant for any secretory rate and the rate of solute secretion always matches that of solute infusion. Thus, absolute dehydration occurs when the infused concentration exceeds that of the secreted fluid. Decreased ECF volume is accompanied by an increased plasma osmotic threshold for secretion, indicating that the output of the salt gland controller requires reciprocal integration of both osmotic and volumetric input.
MORE THAN TWENTY YEARS AGO Schmidt-Nielsen,Jorgensen, and Osaki (16) discovered that hypertonic fluid
MATERIALS
secreted by cranial salt glands is a major route of electrolyte excretion in marine birds. The properties of the body fluids transduced in the control of secretion are still not well understood, and the role of extracellular fluid (ECF) volume remains particularly controversial (13). The original studies conducted by Schmidt-Nielsen’s group clearly showed that salt glands are activated by secretomotor efferents from the central nervous system when ECF tonicity is increased by intravenous infusion of hypertonic solutions (3). McFarland (12), though, noted the large increase in blood volume produced by such infusions and Holmes (11) proposed that volume expansion, rather than elevated tonicity per se, is the stimulus that triggers secretion. Isotonic volume expansion has been variously reported to induce secretion (11, 19), to not induce secretion (7, lo), and to induce transient secretion (5). Stewart (17) dissociated volume expansion from hypertonic stress by withholding water from ducks. Evaporative water loss removed water from the ECF, increasing concentration while decreasing volume. He concluded that because secretion occurred volume recep-
The experiment was performed three times on each of two adult Pekin ducks (Anas platyrhynchos). The birds were maintained outdoors with free access to breeders lay-pellets and to water. The water provided for drinking and grooming was seawater diluted 1:l with tap water525 mosmol/kg HZO, [Na] = 236 meq/l and [K] = 5.2 meq/l. The ducks were held on this regimen at least 30 days before the first experiment and at least 1 wk elapsed between experiments on the same bird. Before the series of experiments was started the weight of duck 1 had stabilized at 2.76 kg and that of duck 2 at 2.42 kg. During the month in which experiments were performed the weight of each duck varied less than 2% both above and below their means. At about 9 A.M. on the day of an experiment a duck was brought inside, where a percutaneous venipuncture of the right jugular vein was made with a 16 gauge Deseret Angiocath. A sterile PE-50 catheter was inserted through the lumen, advanced 10 to 15 cm toward the heart, and the Angiocath was withdrawn. The ducks tolerated this procedure with no obvious distress. The
0363-6119/79/oooO-~$o1.25
Copyright
@ 1979 the American
Physiological
Society
AND
METHODS
R355
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
R356
R.
external nares were then sealed with waterproof adhesive tape, the tape over each nostril being penetrated about a millimeter by l-mm ID soft plastic tubing. The two pieces of tubing joined at a Y attached to a stopper that fitted tared collection vials. Pressure in the vials was held at -200 Torr so that all salt gland secretion was aspirated and collected. Vials were changed at 15min intervals. During an experiment the conscious and alert duck was free to sit or stand in a small oval cage with a large-mesh wire floor. The cage was fitted to a large oval fiberglass funnel so that any cloacal discharge was collected in tared beakers. Within 15 min of voiding, cloacal discharge was weighed, centrifuged, and the fluid portion decanted and saved for analysis. Because blood samples were drawn from the same catheter through which infusions were made, 1 ml of rinsing blood was drawn before each sample. After drawing a Z-ml sample the rinsing blood was reinfused. Blood samples were immediately centrifuged and the plasma osmolality measured. Plasma samples were stored frozen for later determination of [Na] and [K]. Osmolality of all samples was determined by freezing point depression (Knauer Halbmikro-Osmometer), and the sodium and potassium concentrations by flame photometry (Instrumentation Labotatory model 143). Before each experiment the duck was induced to secrete by infusion of 1,000 mosmol/kg Hz0 NaCl at 0.350 or 0.700 ml/min for 30 to 90 min. After the establishment of secretion the infusion was stopped, and when the secretion rate fell to 0.02 mosmol/min or less, a blood sample was drawn for determination of threshold plasma osmolality and sodium concentration. To begin the experiment 1,000 mosmol/kg Hz0 NaCl was infused at 0.350 ml/min for 90 min and the interval from the start of this infusion until the rate of salt gland secretion again fell to or below 0.02 mosmol/min is called the initial control period. Earlier experiments had shown these birds able to secrete salt and water at a rate greater than twice this control infusion rate. Following the initial control period another blood sample was drawn and 5,600 mosmol/kg Hz0 NaCl was infused at 0.070 ml/min for 240 min. When the secretion rate fell to below 0.02 mosmol/min after this infusion, another blood sample was drawn and a postdehydration control infusion, identical to the first, was administered and a final blood sample drawn. The differences between the amounts of salt and water infused and the amounts secreted by the salt gland were used to calculate changes in body fluid osmolality and TABLE I. Excretion of solute and Hz0 by salt glands infusion of hypertonic saline
__~
mosmol mosmol
solute solute
g water g water
-_--Values
for
solute
secreted infused
excreted infused and water
secreted
are means
AND
H.
T. HAMMEL
ECF volume during each 15,min collection period. Because the solutes in the secreted fluid were nearly all sodium and chloride (see RESULTS) the reasonable assumption was made that the net solute load was restricted to the extracellular space. It was further assumed that the net water load distributed freely throughout the total body water (TBW) to maintain osmotic equilibrium between the intracellular and extracellular compartments. In beginning these calculations the osmolality of the body fluids was set at the measured threshold plasma osmolality of the primed and ready-to-secrete duck, and the TBW and ECF volumes were assumed to be, respectively, 64.0 and 26.4% of the body weight, as determined by Ruth and Hughes (15) from the 8”Br and ‘Hz0 spaces in salt water-acclimated Pekin ducks. RESULTS
The infusion of hypertonic NaCl solutions into ducks already at the threshold of secretion stimulates the salt glands to quantitatively secrete the amount of solute with which the ducks are loaded. While the total amount of solute secreted usually slightly exceeded the amount of NaCl infused, Table 1 shows that for each of the three hypertonic challenges solute secreted was within about 10% of solute infused. Because the osmolality of the control infusate approximated that of the secreted fluid, a similar correspondence was achieved between the 2. Effect of dehydration on plasma threshold osmolality for salt gland secretion ______~ --
TABLE
Plasma
Threshold
Osmolality,
Before
dehydration
mosmol/kg
After ~~--_
H,O
dehydration -__--
-
Start control
End control
Start control
End cont rol
Duck I 4-24 5-01 5-31
333 317 306
332 318 304
344 326 314
340 324 310
Duck 2 4-26 5-04 5-26
331 320 317
335 318 318
344 327 328
344 11 328 9 329 11 --~.-. ________ Plasma osmolality measured when secretion rate 0.02 mosmol/min or less before and after control infusions (1,000 mosmol/kg Hz0 NaCl at 0.350 ml/min for 90 min). Acute absolute dehydration produced by secretory response to 5,600 mosmol/kg Hz0 NaCl at 0.070 ml/min for 240 min.
in response to intravenous
Initial Control, 1,000 mosmol/kg H1O at 0.350 ml/min for 90 min
-p___--__-_-.--
KAUL
__._- .__.-._~___...~~__ --- _ . Dehydrating Infusion, !5,6W mosmol/ kg H,O at 0.070 ml/min for 240 min
~-~__-_-----
_ _ ~~-
-_...
Dehydrated Control, 1,000 mosmol/kg Hz0 at 0.350 ml/min for 90 min
_~--~_.-~_- _ _._-_._-_ ____
33.5 t 1.0 =1.08 31.1
98.6 t 2.7 =l.ll 88.9
34.0 +, 1.6 =1.09 31.1
33.4 +, 1.7 =1.08 31.1
93.2 +_ 4.3 =5.86 15.9 __-_____. -_____
32.1 +, 2.3 _.--_
31.1
=1.03
t SE (n = 6).
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
OSMOTIC
THRESHOLD
FOR
SALT
GLAND
R357
SECRETION
amounts of water infused and excreted during the control periods. When a similar rate of solute secretion was driven by infusion of 5,600 mosmol/kg Hz0 NaCl at an appropriately lower volume infusion rate, the salt glands excreted nearly six times as much water as was infused. The net loss of 77.3 t 4.3 g water represents about 5% of the TBW of a 2.6-kg duck. Table 2 shows the effects of the net solute and water exchanges on the threshold plasma osmolality directly measured before and after each infusion. Because of the large variability, even among thresholds for the same duck on different days, individual values rather than means are presented. Despite variability in threshold osmolality from day to day, paired t testing shows that there is no significant difference between values before and after the control infusions (P > 0.1). Following dehydrating infusion/secretion the osmotic threshold is elevated by 9.1 t 0.7 mosmol/kg Hz0 (mean t SE, significant at 0.0005 level by paired t test). In each experiment the sodium concentration (meq/l) of the salt gland secretion could be expressed to within 1% as osmolality (mosmol/kg HZO) divided by 1.883. This divisor is just 2.4% larger than that for pure NaCl solutions of similar concentrations, and the Na-to-K ratio of the secreted &id (constant over the duration of the experiments) was 38.8 t 0.9 (mean t SE, n = 19), i.e., only 2.6% of the measured cation is K+. It therefore seems that the fluid secreted is, to within 2 or 3%, a pure NaCl solution. This means that solute balance is equivalent to sodium balance, and Fig. 1 shows that changes in plasma osmolality accurately reflect changes in plasma sodium concentration. For these plasma samples meq Na/l = 0.381 (mosmol/kg HZO) + 32.0, r” = 0.93. This very strong correlation shows that increased threshold osmolality reflects a true increase in threshold tonicity, and is not an artifact attributable to a stress-induced elevation of plasma glucose or some other permeant solute not effective in driving secretion. Plasma osmolalities and ECF volumes for each experiment have been expressed as 100% at their values in the primed and ready-to-secrete condition (before fust control), and in Fig. 2 the mean values are plotted over the concurrent mean rates of salt gland secretion. The mean values of the directly measured plasma osmolalities are also plotted to show the close agreement between calculated and directly measured osmolality. Figure 2 clearly shows that 1) the loss of extracellular water produced by the secretory response to 5,600 mosmol/kg Hz0 NaCl is associated with an elevation of the threshold osmolality for secretion in response to a control challenge; 2) salt gland activity occurs at an appropriate submaximal rate during dehydration, though plasma osmolality is continually rising as ECF volume is decreasing; and 3) the second control response is indistinguishable from the first, though the threshold plasma osmolality has evidently been reset at a higher level. A final result worthy of note is that during these experiments one duck never produced any cloacal discharge while the other voided only small amounts of fluid with a sodium concentration always much less than that of the plasma. It is clear that under these conditions the
r
170
e-31 A5-I
OUCK
I 310
140 I 300
I
I 320
I 330
I 340
I 350
mOsm /kg H20 FIG.
1. Plasma
regression
equation
renal-cloacal the excretion
[Na] plotted as a function of osmolality and coefficient of determination).
(see text
system did not contribute significantly if the hypertonic salt load.
for
to
DISCUSSION
Because avian salt glands are not constantly active it may be presumed that for any controlling signal there is a threshold level that must be exceeded before the glands are activated. When secretory activity is the only data from which receptor stimulation is inferred, it is clear that the properties of receptors are best revealed by investigations conducted at and above the threshold of activity. Figure 2 shows, for example, that control infusions that increase plasma osmolality less than 0.5% are sufficient to increase salt gland activity from nearly zero to a rate of secretion that would clear the entire plasma sodium content in about 2 h if it were not being continuously replaced by infusion. This implies that the osmotic sensitivity of the receptors is at least an order of magnitude greater than indicated by measurement of changes in plasma solute concentration associated with bringing ducks from somewhere below threshold to somewhere above threshold (14). Of much greater significance than this high order of osmotic sensitivity is our determination that the osmotic threshold is variable and that its value is increased when ECF volume is reduced. We perceive that the patterns of secretion observed might be explained by postulating that the salt glands are driven by a proportional controller sensitive to ECF tonicity and that the reference tonicity is inversely proportional to ECF volume. In this way a constant error signal could be generated during the dehydrating infusion/secretion to drive the constant rate of secretion, and the identity of the initial and dehydrated control responses would result from the defense of plasma tonicity with an elevated set point for secretion. Using an experimental protocol similar to that described here, Hammel, Simon-Oppermann, and Simon (unpublished observations) have found that infusion of isotonic saline or autologous blood into ducks already at threshold always induces secretion. Loss of hypertonic fluid in response to isotonic infusion results in dilution of the plasma and expansion of ECF volume. The osmotic
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
R358
R. KAUL
105
AND
H.
T. HAMMEL
r
i..,.“., i...: ~--z...,“~.“.( L.., ““L....... i i...
i
1.., ..... ... EC
V
FIG. 2. Body fluid osmolality and extracellular fluid volume (expressed as 96 of initial values) during salt gland secretion. Top: solid line, body fluid osmolality (BFO); dotted line, extracellular volume (ECV); circles, directly measured plasma osmolality. Bottom: solid line, solute secretion rate + SE, shaded areas, infused solute loads. All values plotted in each half of this figure are means of 6 trials. See text for method used to calculate extracellular volume and osmolality.
h
TIME 1 hours) threshold is apparently lowered because the salt glands still quantitatively excrete a hypertonic control challenge just as they do before ECF dilution and expansion. This demonstration that multiple modalities must be combined in the control of secretion leads us to suggest the following model for salt gland control. Considering the substantial body of evidence since Verney (18) for cranial osmoreceptors subserving the control of antidiuretic and drinking activities in mammalian and submammalian species, as well as the strong evidence for cardiac mechanoreceptors effecting control of ECF volume in mammalian species (2, 4); we hypothesize that the salt gland secretomotor signal is proportional to the difference between the tonicity of some ECF compartment perfusing cranial osmoreceptors,and an adjustable reference tonicity, the value of which is inversely proportional to ECF volume sensed in the heart or great veins. Hanwell, Linzell, and Peaker (10) have obtained convincing evidence that vagal afferents of cardiac origin are important for the induction of secretion in salt loaded geese. These workers believed the vagal signal to be of osmotic origin, but their published results are completely consistent with a lowering of threshold tonicity in response to ECF volume expansion. Further support is the finding by Gilmore et al. (6) that an “attenuated” secretory response to salt loading sometimes occurs in geese following chronic bilateral vagotomy. Our model predicts elevation of the reference tonicity in the absence of a
vagal volume signal. This would reduce the difference between cranial ECF tonicity and the reference value, so that a diminished osmotic error signal drives the attenuated secretory response. Combination of sensitivities to both sodium concentration and to the size of the principal volume of sodium distribution allows the salt gland controller to regulate the amount of sodium in the body. Osmotic release of arginine vasotocin (AVT), the avian antidiuretic hormone, can provide a mechanism by which sodium excretion is partitioned between the renal and cranial routes according to the availability of osmotically free water. Dantzler (1) and his colleagues have shown that this compound is able to effect a massive reduction in glomerular filtration. When hypertonic stress exceeds the renal concentrating capacity, as in the experiments reported here, AVT probably circulates in high concentration so that little or no sodium is filtered for renal excretion. When ducks at the threshold of salt gland activity are challenged with hypotonic NaCl (200 mosmol/kg H20 at 1.75 ml/min for 240 min) circulating AVT levels are evidently lowered and most of the sodium load is renally excreted as dilute urine. However, the expansion of ECF volume evidently lowers the secretory threshold because, throughout the hypotonic infusion, salt gland activity is driven at a rate such that the amount of the sodium load not appearing in the cloacal discharge is secreted by the salt glands (unpublished observations).
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
OSMOTIC
THRESHOLD
FOR
SALT
GLAND
R359
SECRETION
We are indebted to Yvonne Coleman and Carole Mayo for secretarial aid; to Donald Ward for expert technical assistance; and to Dr. Lee Henderson of the San Diego Veterans Administration Hospital, in whose laboratory the flame photometry was performed.
This Grants Received
work was supported in part BNS-76-24053 and GB40176X. 7 August
1978; accepted
by National
in final
form
Science June
Foundation
1979.
REFERENCES 1. DANTZLER, W. H. Some renal glomerular and tubular mechanisms involved in osmotic and volume regulation in reptiles and birds. In: Proc. AZfied Benzon Symposium XI, edited by C. B. Jorgensen and E. Skadhauge. New York: Academic, 1978, p. 187-201. 2. DIRKS, J., J. SEELY, AND M. LEVEY. Control of extracellular fluid volume and the pathophysiology of edema formation. In: The Kidney, edited by B. M. Brenner and F. C. Rector. Philadelphia, PA: Saunders, 1976, p. 495-552. 3. FANGE, R., K. SCHMIDT-NIELSEN, AND M. ROBINSON. Control of secretion from the avian salt gland. Am. J. Physiol. 195: 321-326, 1958. 4. GAUER, 0. H., 3. P. HENRY, AND C. BEHN. The regulation of extracellular fluid volume. Annu. Reu. Physiol. 32: 547-596, 1970. 5. GILMORE, J. P., J. DIETZ, C. GILMORE, AND I. TUCKER. Evidence for a chloride pump in the salt gland of the goose. Comp. Biochem. Physiol. A 56: 121-126, 1976. 6. GILMORE, J. P., C. GILMORE, J. DIETZ, AND I. H. TUCKER. Influence of chronic cervical vagotomy on salt gland secretion in the goose. Comp. Biochem. Physiol. A. 57: 119-121, 1977. 7. HAJJAR, R., F. SATTLER, B. G. ANDERSON, AND G. GWINUP. Definition of the stimulus to secretion of the nasal salt gland of the seagull. Horn. Metab. Res. 2: 35-37, 1970. 8. HAMMEL, H. T., R. KAUL, C. SIMON-OPPERMANN, AND E. SIMON. Sensitivity of osmoreceptors above the threshold for nasal gland secretion in the duck (Abstract). Proc. Znt. Union Physiol. Sci. 13: 302, 1977. 9. HAMMEL, H. T., J. E. MAGGERT, E. SIMON, L. I. CRAWSHAW, AND R. KAUL. Thermoand osmoregulatory responses induced by heating and cooling the rostra1 brainstem of the Adelie penguin. In: Proc. 3rd Symp. Antarctic Biol., edited by G. A. Llano. Houston,
TX: Gulf, 1977, p. 489-500. 10. HANWELL, A., J. L. LINZELL, AND M. PEAKER. The location and nature of the receptors for salt-gland secretion in the goose. J. Physiol. London 226: 453-472, 1972. II. HOLMES, W. N. Some aspects of osmoregulation in reptiles and birds. Arch. Anat. Micros. Morphol. Exp. 54: 491-513, 1965. 12. MCFARLAND, L. 2. Observations on the haematology and blood volume of captive western gulls (Abstract). Proc. 26th Znt. Congr. Zool. 2: 86, 1963. 13. PEAKER, M. Do osmoreceptors or blood volume receptors initiate salt-gland secretion in birds? J. Physiol. London 276: 66P-67P, 1978. 14. PEAKER, M., S. J. PEAKER, A. HANWELL, AND J. L. LINZELL. Sensitivity of receptors for salt gland secretion in the domestic duck and goose. Comp. Biochem. Physiol. A 44: 41-46, 1973. 15. RUCH, F. E., AND M. R. HUGHES. The effects of hypertonic sodium chloride injection on body water distribution in ducks (Anas platyrhynchos), gulls (Larus glaucescens), and roosters (Gallus domesticus). Comp. Biochem. Physiol. A 52: 21-28, 1975. 16. SCHMIDT-NIELSEN, K., C. B. JORGENSEN, AND H. OSAKI. Extrarenal salt excretion in birds. Am. J. Physiol. 193: 101-107, 1958. 17. STEWART, D. J. Secretion by salt gland during water deprivation in the duck. Am. J. Physiol. 223: 384-386, 1972. 18. VERNEY, E. B. The anti-diuretic hormone and the factors which determine its release. Proc. Roy. Sot. London Ser. B 135: 25-106, 1947. 19. ZUCKER, K. H., C. GILMORE, J. DIETZ, AND J. P. GILMORE. Effect of volume expansion and veratrine on salt gland secretion in the goose. Am. J. Physiol. 232: Rl85-189, 1977 or Am. J. Physiol.: Regulatory Integrative Comp. Physiol. 1: Rl85-R189, 1977.
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 16, 2019.
