0013-7227/9l/1283-1655$03.00/0 Endocrinology Copyright © 1991 by The Endocrine Society
Vol. 128, No. 3 Printed in U.S.A.
Plasma Atrial Natriurc^tic Factor Responses to Blood Volume Changes in the Pekin Duck DAVID A. GRAY, HELMUTH SCHUTZ, AND RUDIGER GERSTBERGER Department of Physiology I, Max-Planck Institute, W. G. Kerckhoff Institute for Physiological and Clinical Research, D-6350 Bad Nauheim, Germany
ABSTRACT. A RIA was developed for the measurement of immunoreactive atrial natriuretic factor (irANF) in avian plasma and was used to investigate the relationship between vascular volume and plasma irANF concentrations in conscious Pekin ducks. In normally hydrated birds, the mean plasma irANF concentration was 78.5 ± 6.8 pg/ml. Blood volume expansion by iv infusion of isotonic saline at 1 ml/min elevated irANF concentrations by 132% after 1 h and by 233% after 2 h. Reduction of the vascular volume by nonhypotensive hem-
T
HERE is now abundant evidence to show that the heart functions as an endocrine organ to process and secrete a peptide that plays a vital role in osmoregulation. This hormone, atrial natriuretic factor (ANF), has been most extensively characterized by studies in mammals (1-3); however, ANF-like peptides appear to be present in other vertebrates too (4, 5). For birds, in particular, a data base is accumulating to support the concept that ANF is important in the maintenance of their salt and fluid balance. Proceeding on from the demonstration that chicken hearts contain a diuretic agent (6), a 29-amino acid peptide has been isolated from the same species that has significant structural homology with mammalian ANF (7) and is stored in atrial and ventricular cardiocytes (8). Initial investigations in the Pekin duck with a synthetic form of this avian ANF have shown that in addition to promoting diuresis and natriuresis, ANF antagonizes the reninangiotensin-aldosterone system (9). Furthermore, ANFbinding sites have been identified in duck renal and adrenal tissue (10). To further characterize the physiology of ANF in birds, factors important in the regulation of this peptide need to be evaluated. Analogy with mammalian studies (11) suggests that vascular volume is likely to have a primary Received October 24,1990. Address requests for reprints to: Dr David A. Gray, Max-Planck Institute, W. G. Kerckhoff Institute, Parkstrasse 1, D-6350 Bad Nauheim, Germany.
orrhage of 10% and then 20% of the total blood volume reduced irANF levels by 16% and 42%, respectively, 5 min after blood removal was completed. Similarly, in birds deprived of water for 24 h and with blood volume reduced by 4.8%, plasma irANF concentrations were 50% lower than in the same euhydrated animals. The observed correlation between plasma irANF concentrations and vascular volume is consistent with the concept that ANF has a physiological role in avian volume homeostasis. (Endocrinology 128: 1655-1660,1991)
influence on avian endogenous plasma ANF concentrations, so the present studies were undertaken. A sensitive RIA for the measurement of ANF-like material has been used to monitor the response of circulating hormone levels to blood volume (BV) changes in conscious Pekin ducks.
Materials and Methods Animals The studies were carried out with adult male and female Pekin ducks [Anas platyrhynchos), within a body weight range of 2-3 kg, housed in flocks in a thermoneutral environment under the natural day-night cycle. The animals were maintained on tap water ad libitum and chickenfeed enriched with vitamins and minerals. All experiments were performed in conscious birds accustomed to the experimental set-up and procedures. Preparation of animals The animals were starved overnight, and on the morning of the experimental day each bird was weighed and placed in a canvas sling that comfortably supported its trunk, but prevented it from turning around. The birds remained without further restraint, except for a loose cotton sling around the legs to prevent their retraction underneath the trunk. A flexible aseptic cannula, as is used in pediatrics (Braun, Melsungen, Germany), was quickly inserted into one leg vein for the administration of various sterile solutions with an infusion pump (Perfusor E, Braun) or for phlebotomy. For collecting free flowing blood samples for hormone analysis, a second cannula 1655
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 20:17 For personal use only. No other uses without permission. . All rights reserved.
CONTROL OF PLASMA ANF IN DUCKS
1656
was placed in a vein of the other leg and kept patent by the infusion of heparinized isotonic saline (5 U/ml) at a rate of 0.1 ml/min. Venipuncture was performed quickly and did not cause any apparent sustained discomfort. One day before the experiment, some of the birds were anesthetized by inhalation anesthesia (halothane, N2O, O2), and a catheter (PP60, Portex, Hythe, United Kingdom) was placed into a brachial artery. In some cases this catheter was used to monitor mean arterial blood pressure (MABP) via a pressure transducer (Endevco N8510, 5PS1, San Juan Capistrano, CA), and in others was used to remove arterial blood samples for hormone analysis. Experimental procedures A screen was placed around the bird, and after an equilibration period of 45-60 min, and subsequently when required, blood samples were taken via the indwelling cannula. Blood for ANF measurement (2 ml) was collected into polypropylene tubes on ice, with 100 pi of an enzyme inhibitor containing 2 mg EDTA and 1000 kallikrein inhibitory units aprotinin (Trasylol, Bayer, Leverkusen, Germany). The total bleeding time was 1-2 min, and after each bleeding, the total BV removed was replaced by an equal volume of saline. After mixing and centrifugation at 4000 X g for 10 min at 4 C, plasma was separated and stored at -20 C before assay. Small blood samples (0.5 ml) were also taken without enzyme inhibitor and used for hematocrit, osmolality, and electrolyte determination. After control blood samples had been taken, the responses of immunoreactive ANF (irANF) to maneuvers producing BV changes were examined in the following experimental series. Series I. Vascular volume expansion was produced by the iv infusion of isotonic saline at 1 ml/min for 2 h. Venous blood samples were taken after 1 and 2 h. Series II. BV reduction was achieved in two ways: 1) acute shrinkage, by the nonhypotensive hemorrhage of 10% and then 20% of the total BV, estimated as 10% body weight (12); venous blood samples for hormone analysis were taken 5 min after the end of phlebotomy; or 2) more chronic BV reduction, by water deprivation for 24 h. Series III. To characterize the ANF immunoreactivity measured in plasma extracts (see below), the BV of six ducks was expanded by iv administration of 4% dextran given at 0.7 ml/ min for 60 min. Simultaneous arterial and venous blood samples were taken before and after 30 and 60 min of infusion. Aliquots of the extracted plasma were pooled and subjected to reverse phase HPLC, and the fractions were analyzed by RIA. In a further set of experiments, synthetic chicken (ch) ANF was infused iv into five normally hydrated ducks at a dose of 100 ng/kg-min for 30 min. Venous blood samples were taken 5 min before the end of chANF administration and 2 min after termination of the infusion. Aliquots of the extracted plasma were pooled and subjected to HPLC/RIA analysis Analytical methods Osmolality of plasma was measured by vapour pressure osmometry (Wescor 5100C, Logan, UT). Sodium and potassium were measured by flame photometry (IL943, Instrumentation
Endo • 1991 Vol 128 • No 3
Laboratories, Lexington, MA), and chloride by titration (Coming-Eel, Halstead, United Kingdom). Hematocrit (Hct) was measured in duplicate with microhematocrit tubes (Hawkesley, Lancing, United Kingdom) and used to estimate changes in BV using the equation BV2/BV1 = Hctl/Hct2 (13). RIA for ANF The antiserum was raised in a rabbit against synthetic chANF [Peninsula Laboratories (Belmont, CA) and Bachem (Torrence, CA)] using repeated intradermal injections of antigen coupled to porcine thyroglobulin (14). Synthetic chANF was used as standard and was also labeled with [125I]BoltonHunter reagent (chANF contains no tyrosine residue) for use as tracer after purification by HPLC. irANF was measured in plasma after acetone precipitation of the proteins. One milliliter of thawed plasma was mixed with 2 ml chilled acetone (—20 C) and centrifuged at 4000 X g for 10 min at 4 C. The supernatant was extracted twice with petroleum benzine and left at room temperature for 30 min. After discarding the ether phase, the aqueous phase was evaporated to dryness (SpeedVac, Savant, Munich, Germany). The dried extracts were redissolved in 500 fA assay buffer (see below), and two 200-/il aliquots were transferred to assay tubes for RIA. All assay procedures were carried out on ice. The buffer used throughout was 0.1 M Tris (hydroxymethyl) methylamine adjusted to pH 7.4 with HC1 and containing 3 mg/ml BSA (fraction V, Sigma, St. Louis, MO), 2 mg/ml neomycin sulfate (Merck, Darmstadt, Germany), and 0.1% Triton-X 100 (Sigma). To optimize assay sensitivity, a nonequilibrium procedure was used; the unlabeled peptide and antiserum were incubated for 24 h at 4 C before the addition of labeled tracer. Standard curves were obtained by adding 200 fil doubling dilutions of standard chANF (1.5-200 pg/tube) and 200 MI antiserum (code Gll/7), giving a final dilution of 1:120,000. Samples were assayed in place of standard. Tracer (3,000-4,000 cpm) was added in 100 ^1» and incubation was continued for 24 h at 4 C. Separation of bound from free tracer was achieved by the addition of 1 ml absolute alcohol, followed by rapid mixing and centrifugation at 4000 x g for 10 min at 4 C. The supernatants were removed by aspiration, and radioactivity in the pellets was estimated by 7-counting (Berthold, Wildbad, Germany). Figure 1 shows a typical standard curve for chANF together with the lack of cross-reactivity with arginine vasotocin, mesotocin, angiotensin-II, or frog ANF. In addition, the antiserum failed to cross-react with human or rat ANF, but did show 10% cross-reactivity with porcine brain natriuretic peptide, supporting the idea that chANF has a greater homology to mammalian brain natriuretic peptide-type rather than mammalian ANF-type peptides (7, 8). On the average, the specific binding was 44.7%, and the nonspecific binding was 4.4%. Displacement of 50% bound tracer occurred at 14.8 pg/tube, and the slope of the standard curve at this point was 1.25. The limit of detection of the assay has been defined as the standard concentration producing a depression from the initial binding by 2 SD and was approximately 3 pg/tube. Extraction recovery, measured over a range of 20-200 pg/ml chANF added to a plasma pool, averaged 71% (n = 20). The results were corrected for the recovery. Serial dilutions of pooled plasma extracts gave binding curves that were completely superimposable upon the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 20:17 For personal use only. No other uses without permission. . All rights reserved.
1657
CONTROL OF PLASMA ANF IN DUCKS B/Bo
100
Frog ANF MESO
100 r
r
Plasma irANF 80
[pgml-1]
60
»T
50 40
«
20
0 10
Control
100
1000 10000 Peptide per tube [ p g ]
FIG. 1. A typical standard curve for chANF, also showing the crossreactivity of the antiserum with arginine vasotocin (AVT), angiotensinII (All), mesotocin (MESO), and frog ANF. B/Bo, Ratio of binding in standard to binding in zero standard. 300
L
r
-10% blood
-20% blood
FIG. 3. Changes in plasma irANF concentrations induced by the removal of 10% and 20% total BV. Blood samples were taken 5 min after the end of hemorrhage. • , P < 0.05; • * , P < 0.02 (n = 8).
the extracted materials behave in the assay in the same way as the standard preparation. The intraassay coefficient of variation was 7.4% (n = 16), and the interassay coefficient of variation was 9.1% (n = 10). HPLC
Plasma irANF [pgml-1] +18.5%
blood volume
200
Plasma extracts were subjected to reverse phase HPLC using a linear acetonitrile gradient from 18-30% over 30 min at a flow rate of 2 ml/min. Solvent A was 10% acetonitrile in water with 0.1% triflouroacetic acid, and solvent B was 90% acetonitrile in water with 0.1% triflouroacetic acid. The separation was carried out with a two-pump system (Waters, Milford, MA) and a 25 X 0.4-cm Nucleosil C18 column (Chromatography Service, Langerwehe, Germany), which was repeatedly callibrated with a small amount (500 pg) of chANF. Collected fractions (2 ml) were lyophilized, reconstituted in buffer, and assayed for irANF by RIA. Statistics The results are presented as means with their SEs. Statistical significance of differences was evaluated by the Wilcoxon matched pairs signed ranks test. The relationships between variables were analyzed by linear regression.
100
Results Plasma irANF levels
0
L
Control
1 hr
2 hr
FIG. 2. Plasma irANF concentrations measured after 1 and 2 h of iv loading with isotonic saline. Indicated BV changes were calculated from hematocrit values. • * , P < 0.02; • • • , P < 0.01 (n = 8).
In the three series of experiments, basal venous plasma irANF concentrations were measured in 20 individual birds. The average hormone concentration in these normally hydrated animals was 78.5 ± 6.8 pg/ml, with a range from 46.7-147.2 pg/ml. Effect of BV expansion
standard curve. Furthermore, a linear relationship exists between the amount of irANF measured and the volume of extract assayed, with the regression line passing through zero. Thus,
The iv infusion of isotonic saline for 2 h produced no significant change in plasma osmotic or sodium concen-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 20:17 For personal use only. No other uses without permission. . All rights reserved.
CONTROL OF PLASMA ANF IN DUCKS
1658
Endo • 1991 Vol 128 • No 3
TABLE 1. Osmotic parameters in ducks under conditions of normal hydration and after 24 h dehydration BW (kg) 2.24 ± 0.08 2.06 ± 0.08"
Normal hydration 24-h dehydration
Hct (%)
pOsm (mosol/kg)
pNa (meq/liter)
38.7 ± 1.1 39.9 ± 1.1°
293.4 ± 1.1 339.9 ± 7.3*
148.4 ±1.1 174.8 ± 2.6*
pK (meq/liter) 2.7 ± 0.2 3.5 ± 0.1°
pCl (meq/liter)
pirANF (pg/ml)
119.8 ± 1.2 145.7 ± 2.6*
72.3 ± 8.1 37.1 ± 5.6*
Values are the mean ± SE for 10 birds. pOsm, Plasma osmolality; pNa, plasma sodium; pK, plasma potassium; pCl, plasma chloride; pirANF, plasma irANF. P< 0.02 vs. normal. * P < 0.01 us. normal. a
400 r Plasma irANF
Effect of BV reduction r = 0.65
Hemorrhage. The response of plasma irANF concentrations to the removal of 10% and then 20% of the BV, measured 5 min after each bleeding (which usually took less than 4 min), is shown in Fig. 3. Compared to the control value, plasma irANF concentrations were reduced by 16% and 42%, respectively. No significant changes in MABP or heart rate were observed during the experiment.
P < 0.001
[pg ml-1] 300
-
n - 58
200
100
j
-20
-10
I
I
I
10 20 A Blood volume
30
FIG. 4. Relationship between plasma irANF concentrations and BV in conscious Pekin ducks. TABLE 2. Changes in arterial and venous plasma irANF concentrations in ducks given an iv infusion of 4% dextran at 0.7 ml/min Plasma irANF (pg/ml)
Control 30 min dextran
Arterial
Venous
81.1 ± 7.1 191.5 ± 36.2
53.8 ± 3.3° 133.4 ± 24.10-6
ABV (%) 0 +14.7
± 60 min dextran
306.6 ± 38.7
217.4 ± 23.60-6
2.9 +23.2 3.3
Values are the mean ± SE for six birds. Changes in BV (ABV) were calculated from hematocrit measurements. " P < 0.05 vs. arterial. * P < 0.01 vs. control.
trations, which had control values of 296.3 ± 1.3 mosmol/ kg and 139.4 ± 0.9 meq/liter, respectively. Hematocrit values were, however, decreased from the control level of 38.1 ± 1.4% to 34.1 ± 1.3% after 1 h and 32.0 ± 1.1% after 2 h, indicating vascular expansions of 11.7 ± 1.7% and 18.5 ± 2.4%, respectively. Figure 2 shows that associated with this BV expansion, irANF concentrations were elevated by 132% after 1 h and 223% after 2 h. Throughout the experiment, MABP and heart rate did not change significantly from the control levels of 138.4 ± 6.3 mm Hg and 172.2 ± 9.0 beats/min.
Dehydration. A comparison of the osmotic parameters measured in normally hydrated birds and in the same animals after 24 h of water deprivation is given in Table 1. Dehydration increased plasma electrolyte concentrations and raised the osmolality by 44.5 ± 7.2 mosmol/kg. There was a decrease in body weight by 188.3 ± 26 g, equivalent to 8.3 ± 1.1%. Hematocrit increased and indicated a reduction of BV by 4.8 ± 0.7%. Plasma irANF concentrations were 50% lower in dehydrated birds than in normally hydrated animals. In five of the birds, MABP and heart rate were also measured before and after dehydration. No significant changes in these parameters were detected. Combining the results from series I and II showed that plasma irANF concentrations were significantly correlated with the changes in BV (Fig. 4). Characterization of irANF The levels of irANF measured in series I and II represent the concentrations found in venous plasma. To evaluate the relationship between venous and arterial systemic irANF levels, simultaneous blood samples of both types were taken from birds in which vascular volume had been expanded by dextran infusion. Table 2 shows that arterial plasma levels of irANF were consistently 30% higher than those in venous plasma; however, the relative increases in irANF concentrations induced by vascular expansion were the same in both venous and arterial blood. Analysis of venous and arterial plasma extracts by HPLC/RIA (Fig. 5) showed that the immunoreactive profiles were the same for both types of plasma. Blood taken under basal conditions produced several immunoreactive peaks of similar size; however, when ANF secretion was stimulated by BV expansion, the relative proportions of the immunoreactive peaks changed so that
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 20:17 For personal use only. No other uses without permission. . All rights reserved.
CONTROL OF PLASMA ANF IN DUCKS
1659
VENOUS BLOOD 80
ARTERIAL BLOOD 80
chANF
ChANF
30
Q
40
10
0
80
80
30
Ol
40
s
0
(EV 20
30
in
/f j \
o.
O
20 I
40
10
0
ive ANF
FIG. 5. Immunoreactivity profiles of extracted arterial and venous plasma taken under control conditions (A) and after 30 min (B) and 60 min (C) of dextran infusion. Aliquots of extracts from six birds were pooled and subjected to analysis. The given irANF values represent the mean levels measured per ml plasma. Acn, Acetonitrile.
per frac
20
30
40
20
Vol. 128, No. 3 Printed in U.S.A.
Plasma Atrial Natriurc^tic Factor Responses to Blood Volume Changes in the Pekin Duck DAVID A. GRAY, HELMUTH SCHUTZ, AND RUDIGER GERSTBERGER Department of Physiology I, Max-Planck Institute, W. G. Kerckhoff Institute for Physiological and Clinical Research, D-6350 Bad Nauheim, Germany
ABSTRACT. A RIA was developed for the measurement of immunoreactive atrial natriuretic factor (irANF) in avian plasma and was used to investigate the relationship between vascular volume and plasma irANF concentrations in conscious Pekin ducks. In normally hydrated birds, the mean plasma irANF concentration was 78.5 ± 6.8 pg/ml. Blood volume expansion by iv infusion of isotonic saline at 1 ml/min elevated irANF concentrations by 132% after 1 h and by 233% after 2 h. Reduction of the vascular volume by nonhypotensive hem-
T
HERE is now abundant evidence to show that the heart functions as an endocrine organ to process and secrete a peptide that plays a vital role in osmoregulation. This hormone, atrial natriuretic factor (ANF), has been most extensively characterized by studies in mammals (1-3); however, ANF-like peptides appear to be present in other vertebrates too (4, 5). For birds, in particular, a data base is accumulating to support the concept that ANF is important in the maintenance of their salt and fluid balance. Proceeding on from the demonstration that chicken hearts contain a diuretic agent (6), a 29-amino acid peptide has been isolated from the same species that has significant structural homology with mammalian ANF (7) and is stored in atrial and ventricular cardiocytes (8). Initial investigations in the Pekin duck with a synthetic form of this avian ANF have shown that in addition to promoting diuresis and natriuresis, ANF antagonizes the reninangiotensin-aldosterone system (9). Furthermore, ANFbinding sites have been identified in duck renal and adrenal tissue (10). To further characterize the physiology of ANF in birds, factors important in the regulation of this peptide need to be evaluated. Analogy with mammalian studies (11) suggests that vascular volume is likely to have a primary Received October 24,1990. Address requests for reprints to: Dr David A. Gray, Max-Planck Institute, W. G. Kerckhoff Institute, Parkstrasse 1, D-6350 Bad Nauheim, Germany.
orrhage of 10% and then 20% of the total blood volume reduced irANF levels by 16% and 42%, respectively, 5 min after blood removal was completed. Similarly, in birds deprived of water for 24 h and with blood volume reduced by 4.8%, plasma irANF concentrations were 50% lower than in the same euhydrated animals. The observed correlation between plasma irANF concentrations and vascular volume is consistent with the concept that ANF has a physiological role in avian volume homeostasis. (Endocrinology 128: 1655-1660,1991)
influence on avian endogenous plasma ANF concentrations, so the present studies were undertaken. A sensitive RIA for the measurement of ANF-like material has been used to monitor the response of circulating hormone levels to blood volume (BV) changes in conscious Pekin ducks.
Materials and Methods Animals The studies were carried out with adult male and female Pekin ducks [Anas platyrhynchos), within a body weight range of 2-3 kg, housed in flocks in a thermoneutral environment under the natural day-night cycle. The animals were maintained on tap water ad libitum and chickenfeed enriched with vitamins and minerals. All experiments were performed in conscious birds accustomed to the experimental set-up and procedures. Preparation of animals The animals were starved overnight, and on the morning of the experimental day each bird was weighed and placed in a canvas sling that comfortably supported its trunk, but prevented it from turning around. The birds remained without further restraint, except for a loose cotton sling around the legs to prevent their retraction underneath the trunk. A flexible aseptic cannula, as is used in pediatrics (Braun, Melsungen, Germany), was quickly inserted into one leg vein for the administration of various sterile solutions with an infusion pump (Perfusor E, Braun) or for phlebotomy. For collecting free flowing blood samples for hormone analysis, a second cannula 1655
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 20:17 For personal use only. No other uses without permission. . All rights reserved.
CONTROL OF PLASMA ANF IN DUCKS
1656
was placed in a vein of the other leg and kept patent by the infusion of heparinized isotonic saline (5 U/ml) at a rate of 0.1 ml/min. Venipuncture was performed quickly and did not cause any apparent sustained discomfort. One day before the experiment, some of the birds were anesthetized by inhalation anesthesia (halothane, N2O, O2), and a catheter (PP60, Portex, Hythe, United Kingdom) was placed into a brachial artery. In some cases this catheter was used to monitor mean arterial blood pressure (MABP) via a pressure transducer (Endevco N8510, 5PS1, San Juan Capistrano, CA), and in others was used to remove arterial blood samples for hormone analysis. Experimental procedures A screen was placed around the bird, and after an equilibration period of 45-60 min, and subsequently when required, blood samples were taken via the indwelling cannula. Blood for ANF measurement (2 ml) was collected into polypropylene tubes on ice, with 100 pi of an enzyme inhibitor containing 2 mg EDTA and 1000 kallikrein inhibitory units aprotinin (Trasylol, Bayer, Leverkusen, Germany). The total bleeding time was 1-2 min, and after each bleeding, the total BV removed was replaced by an equal volume of saline. After mixing and centrifugation at 4000 X g for 10 min at 4 C, plasma was separated and stored at -20 C before assay. Small blood samples (0.5 ml) were also taken without enzyme inhibitor and used for hematocrit, osmolality, and electrolyte determination. After control blood samples had been taken, the responses of immunoreactive ANF (irANF) to maneuvers producing BV changes were examined in the following experimental series. Series I. Vascular volume expansion was produced by the iv infusion of isotonic saline at 1 ml/min for 2 h. Venous blood samples were taken after 1 and 2 h. Series II. BV reduction was achieved in two ways: 1) acute shrinkage, by the nonhypotensive hemorrhage of 10% and then 20% of the total BV, estimated as 10% body weight (12); venous blood samples for hormone analysis were taken 5 min after the end of phlebotomy; or 2) more chronic BV reduction, by water deprivation for 24 h. Series III. To characterize the ANF immunoreactivity measured in plasma extracts (see below), the BV of six ducks was expanded by iv administration of 4% dextran given at 0.7 ml/ min for 60 min. Simultaneous arterial and venous blood samples were taken before and after 30 and 60 min of infusion. Aliquots of the extracted plasma were pooled and subjected to reverse phase HPLC, and the fractions were analyzed by RIA. In a further set of experiments, synthetic chicken (ch) ANF was infused iv into five normally hydrated ducks at a dose of 100 ng/kg-min for 30 min. Venous blood samples were taken 5 min before the end of chANF administration and 2 min after termination of the infusion. Aliquots of the extracted plasma were pooled and subjected to HPLC/RIA analysis Analytical methods Osmolality of plasma was measured by vapour pressure osmometry (Wescor 5100C, Logan, UT). Sodium and potassium were measured by flame photometry (IL943, Instrumentation
Endo • 1991 Vol 128 • No 3
Laboratories, Lexington, MA), and chloride by titration (Coming-Eel, Halstead, United Kingdom). Hematocrit (Hct) was measured in duplicate with microhematocrit tubes (Hawkesley, Lancing, United Kingdom) and used to estimate changes in BV using the equation BV2/BV1 = Hctl/Hct2 (13). RIA for ANF The antiserum was raised in a rabbit against synthetic chANF [Peninsula Laboratories (Belmont, CA) and Bachem (Torrence, CA)] using repeated intradermal injections of antigen coupled to porcine thyroglobulin (14). Synthetic chANF was used as standard and was also labeled with [125I]BoltonHunter reagent (chANF contains no tyrosine residue) for use as tracer after purification by HPLC. irANF was measured in plasma after acetone precipitation of the proteins. One milliliter of thawed plasma was mixed with 2 ml chilled acetone (—20 C) and centrifuged at 4000 X g for 10 min at 4 C. The supernatant was extracted twice with petroleum benzine and left at room temperature for 30 min. After discarding the ether phase, the aqueous phase was evaporated to dryness (SpeedVac, Savant, Munich, Germany). The dried extracts were redissolved in 500 fA assay buffer (see below), and two 200-/il aliquots were transferred to assay tubes for RIA. All assay procedures were carried out on ice. The buffer used throughout was 0.1 M Tris (hydroxymethyl) methylamine adjusted to pH 7.4 with HC1 and containing 3 mg/ml BSA (fraction V, Sigma, St. Louis, MO), 2 mg/ml neomycin sulfate (Merck, Darmstadt, Germany), and 0.1% Triton-X 100 (Sigma). To optimize assay sensitivity, a nonequilibrium procedure was used; the unlabeled peptide and antiserum were incubated for 24 h at 4 C before the addition of labeled tracer. Standard curves were obtained by adding 200 fil doubling dilutions of standard chANF (1.5-200 pg/tube) and 200 MI antiserum (code Gll/7), giving a final dilution of 1:120,000. Samples were assayed in place of standard. Tracer (3,000-4,000 cpm) was added in 100 ^1» and incubation was continued for 24 h at 4 C. Separation of bound from free tracer was achieved by the addition of 1 ml absolute alcohol, followed by rapid mixing and centrifugation at 4000 x g for 10 min at 4 C. The supernatants were removed by aspiration, and radioactivity in the pellets was estimated by 7-counting (Berthold, Wildbad, Germany). Figure 1 shows a typical standard curve for chANF together with the lack of cross-reactivity with arginine vasotocin, mesotocin, angiotensin-II, or frog ANF. In addition, the antiserum failed to cross-react with human or rat ANF, but did show 10% cross-reactivity with porcine brain natriuretic peptide, supporting the idea that chANF has a greater homology to mammalian brain natriuretic peptide-type rather than mammalian ANF-type peptides (7, 8). On the average, the specific binding was 44.7%, and the nonspecific binding was 4.4%. Displacement of 50% bound tracer occurred at 14.8 pg/tube, and the slope of the standard curve at this point was 1.25. The limit of detection of the assay has been defined as the standard concentration producing a depression from the initial binding by 2 SD and was approximately 3 pg/tube. Extraction recovery, measured over a range of 20-200 pg/ml chANF added to a plasma pool, averaged 71% (n = 20). The results were corrected for the recovery. Serial dilutions of pooled plasma extracts gave binding curves that were completely superimposable upon the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 20:17 For personal use only. No other uses without permission. . All rights reserved.
1657
CONTROL OF PLASMA ANF IN DUCKS B/Bo
100
Frog ANF MESO
100 r
r
Plasma irANF 80
[pgml-1]
60
»T
50 40
«
20
0 10
Control
100
1000 10000 Peptide per tube [ p g ]
FIG. 1. A typical standard curve for chANF, also showing the crossreactivity of the antiserum with arginine vasotocin (AVT), angiotensinII (All), mesotocin (MESO), and frog ANF. B/Bo, Ratio of binding in standard to binding in zero standard. 300
L
r
-10% blood
-20% blood
FIG. 3. Changes in plasma irANF concentrations induced by the removal of 10% and 20% total BV. Blood samples were taken 5 min after the end of hemorrhage. • , P < 0.05; • * , P < 0.02 (n = 8).
the extracted materials behave in the assay in the same way as the standard preparation. The intraassay coefficient of variation was 7.4% (n = 16), and the interassay coefficient of variation was 9.1% (n = 10). HPLC
Plasma irANF [pgml-1] +18.5%
blood volume
200
Plasma extracts were subjected to reverse phase HPLC using a linear acetonitrile gradient from 18-30% over 30 min at a flow rate of 2 ml/min. Solvent A was 10% acetonitrile in water with 0.1% triflouroacetic acid, and solvent B was 90% acetonitrile in water with 0.1% triflouroacetic acid. The separation was carried out with a two-pump system (Waters, Milford, MA) and a 25 X 0.4-cm Nucleosil C18 column (Chromatography Service, Langerwehe, Germany), which was repeatedly callibrated with a small amount (500 pg) of chANF. Collected fractions (2 ml) were lyophilized, reconstituted in buffer, and assayed for irANF by RIA. Statistics The results are presented as means with their SEs. Statistical significance of differences was evaluated by the Wilcoxon matched pairs signed ranks test. The relationships between variables were analyzed by linear regression.
100
Results Plasma irANF levels
0
L
Control
1 hr
2 hr
FIG. 2. Plasma irANF concentrations measured after 1 and 2 h of iv loading with isotonic saline. Indicated BV changes were calculated from hematocrit values. • * , P < 0.02; • • • , P < 0.01 (n = 8).
In the three series of experiments, basal venous plasma irANF concentrations were measured in 20 individual birds. The average hormone concentration in these normally hydrated animals was 78.5 ± 6.8 pg/ml, with a range from 46.7-147.2 pg/ml. Effect of BV expansion
standard curve. Furthermore, a linear relationship exists between the amount of irANF measured and the volume of extract assayed, with the regression line passing through zero. Thus,
The iv infusion of isotonic saline for 2 h produced no significant change in plasma osmotic or sodium concen-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 20:17 For personal use only. No other uses without permission. . All rights reserved.
CONTROL OF PLASMA ANF IN DUCKS
1658
Endo • 1991 Vol 128 • No 3
TABLE 1. Osmotic parameters in ducks under conditions of normal hydration and after 24 h dehydration BW (kg) 2.24 ± 0.08 2.06 ± 0.08"
Normal hydration 24-h dehydration
Hct (%)
pOsm (mosol/kg)
pNa (meq/liter)
38.7 ± 1.1 39.9 ± 1.1°
293.4 ± 1.1 339.9 ± 7.3*
148.4 ±1.1 174.8 ± 2.6*
pK (meq/liter) 2.7 ± 0.2 3.5 ± 0.1°
pCl (meq/liter)
pirANF (pg/ml)
119.8 ± 1.2 145.7 ± 2.6*
72.3 ± 8.1 37.1 ± 5.6*
Values are the mean ± SE for 10 birds. pOsm, Plasma osmolality; pNa, plasma sodium; pK, plasma potassium; pCl, plasma chloride; pirANF, plasma irANF. P< 0.02 vs. normal. * P < 0.01 us. normal. a
400 r Plasma irANF
Effect of BV reduction r = 0.65
Hemorrhage. The response of plasma irANF concentrations to the removal of 10% and then 20% of the BV, measured 5 min after each bleeding (which usually took less than 4 min), is shown in Fig. 3. Compared to the control value, plasma irANF concentrations were reduced by 16% and 42%, respectively. No significant changes in MABP or heart rate were observed during the experiment.
P < 0.001
[pg ml-1] 300
-
n - 58
200
100
j
-20
-10
I
I
I
10 20 A Blood volume
30
FIG. 4. Relationship between plasma irANF concentrations and BV in conscious Pekin ducks. TABLE 2. Changes in arterial and venous plasma irANF concentrations in ducks given an iv infusion of 4% dextran at 0.7 ml/min Plasma irANF (pg/ml)
Control 30 min dextran
Arterial
Venous
81.1 ± 7.1 191.5 ± 36.2
53.8 ± 3.3° 133.4 ± 24.10-6
ABV (%) 0 +14.7
± 60 min dextran
306.6 ± 38.7
217.4 ± 23.60-6
2.9 +23.2 3.3
Values are the mean ± SE for six birds. Changes in BV (ABV) were calculated from hematocrit measurements. " P < 0.05 vs. arterial. * P < 0.01 vs. control.
trations, which had control values of 296.3 ± 1.3 mosmol/ kg and 139.4 ± 0.9 meq/liter, respectively. Hematocrit values were, however, decreased from the control level of 38.1 ± 1.4% to 34.1 ± 1.3% after 1 h and 32.0 ± 1.1% after 2 h, indicating vascular expansions of 11.7 ± 1.7% and 18.5 ± 2.4%, respectively. Figure 2 shows that associated with this BV expansion, irANF concentrations were elevated by 132% after 1 h and 223% after 2 h. Throughout the experiment, MABP and heart rate did not change significantly from the control levels of 138.4 ± 6.3 mm Hg and 172.2 ± 9.0 beats/min.
Dehydration. A comparison of the osmotic parameters measured in normally hydrated birds and in the same animals after 24 h of water deprivation is given in Table 1. Dehydration increased plasma electrolyte concentrations and raised the osmolality by 44.5 ± 7.2 mosmol/kg. There was a decrease in body weight by 188.3 ± 26 g, equivalent to 8.3 ± 1.1%. Hematocrit increased and indicated a reduction of BV by 4.8 ± 0.7%. Plasma irANF concentrations were 50% lower in dehydrated birds than in normally hydrated animals. In five of the birds, MABP and heart rate were also measured before and after dehydration. No significant changes in these parameters were detected. Combining the results from series I and II showed that plasma irANF concentrations were significantly correlated with the changes in BV (Fig. 4). Characterization of irANF The levels of irANF measured in series I and II represent the concentrations found in venous plasma. To evaluate the relationship between venous and arterial systemic irANF levels, simultaneous blood samples of both types were taken from birds in which vascular volume had been expanded by dextran infusion. Table 2 shows that arterial plasma levels of irANF were consistently 30% higher than those in venous plasma; however, the relative increases in irANF concentrations induced by vascular expansion were the same in both venous and arterial blood. Analysis of venous and arterial plasma extracts by HPLC/RIA (Fig. 5) showed that the immunoreactive profiles were the same for both types of plasma. Blood taken under basal conditions produced several immunoreactive peaks of similar size; however, when ANF secretion was stimulated by BV expansion, the relative proportions of the immunoreactive peaks changed so that
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 20:17 For personal use only. No other uses without permission. . All rights reserved.
CONTROL OF PLASMA ANF IN DUCKS
1659
VENOUS BLOOD 80
ARTERIAL BLOOD 80
chANF
ChANF
30
Q
40
10
0
80
80
30
Ol
40
s
0
(EV 20
30
in
/f j \
o.
O
20 I
40
10
0
ive ANF
FIG. 5. Immunoreactivity profiles of extracted arterial and venous plasma taken under control conditions (A) and after 30 min (B) and 60 min (C) of dextran infusion. Aliquots of extracts from six birds were pooled and subjected to analysis. The given irANF values represent the mean levels measured per ml plasma. Acn, Acetonitrile.
per frac
20
30
40
20
