Amniotic Prolactin Control Over Amniotic and Fetal Extracellular Fluid Water and Electrolytes in the Rhesus Monkey1'2 JOHN B. JOSIMOVICH, KAREN MERISKO, AND LOUISE BOCCELLA Department of Obstetrics and Gynecology, University of Pittsburgh, Magee-Womens Hospital, Pittsburgh, Pennsylvania 15213 fluid (ECF) volume in the face of hypertonic amniotic fluid. Efflux of these substances from the fetal ECF in the face of hypotonic amniotic fluid was similarly prevented or reversed by intraamniotic prolactin injection. Ovine PRL had no effect on fetal ECF water and electrolytes in the face of isotonic amniotic fluid. Possible sites of these oPRL effects were amnion, placenta, fetal lung and/or fetal gastrointestinal tract. (Endocrinology 100: 564, 1977)
ABSTRACT. Administration of 1-10 mg ovine pituitary prolactin (oPRL) into the amniotic fluid of 10 rhesus monkeys in the last third of gestation consistently caused a decrease in amniotic fluid volume not seen when saline, vasopressin, or bovine serum albumin were injected into 9 other monkeys. The effects lasted for 24 h. Intraamniotic injection of 10 mg oPRL prevented or reversed a doubling of water and electrolyte content of the fetal extracellular
L
ARGE amounts of pituitary prolactin may be found in the amniotic fluid of the rhesus monkey (1,2). The source of the large reservoir has been shown to derive at least in part from the maternal circulation (2). Although amniotic prolactin may be bound in large quantities to the amnion, certain fetal tissues such as lung, and placenta whose cell membranes appear to bind this hormone with high affinity may also bind lesser amounts of amniotic prolactin (3). Because of the well known effects of prolactin on water and sodium conservation when salt water fish or larval amphibians enter fresh water (1,4), the present series of studies were undertaken to define possible roles of altered amniotic prolactin levels on fetal electrolyte and water conservation in the face of different amniotic fluid tonicities. Materials and Methods Thirteen rhesus monkeys bred at the MageeWomens Hospital facility, underwent phencyclidine and pentobarbital medication at 115-156 days gestation (2). Amniotic fluid (AF) volumes were determined by the Congo Red dye dilution technique (5)
Received December 22, 1975. 1 This work was supported by NIH Grant HD-06929. 2 Reprint requests should be sent to the first author at Magee-Womens Hospital.
whose precision was confirmed to be within 10% of actual volumes directly measured at the time of cesarean section in two instances. Volumes were determined before, and at 2, 24, and 48 h after injection of 1 ml normal saline, 1 ml saline containing 1 or 10 mg ovine prolactin (P-S-9, obtained from the Hormone Distribution Officer, National Institutes of Health, Bethesda, Md.), 10 mg bovine serum albumin, or 1 or 100 mU vasopressin (Pitressin, Parke, Davis & Co., Detroit, Mich.). In 6 of the monkeys, additional halothane anesthesia was performed and indwelling catheters placed in the maternal saphenous vein, amniotic cavity and fetal interplacental vein (2) where sampling was made at 3-20 min intervals over two 2-h periods. The two periods were distinguished by differences in the amniotic fluid composition: after 60-87% replacement of the amniotic fluid with hypertonic, isotonic or hypotonic amniotic fluid with no added prolactin; or after similar replacement of amniotic fluid with 10 mg ovine prolactin. The order of the 2-h experiments under the influence of lowered endogenous amniotic fluid prolactin ("low prolactin period") or high prolactin levels ("high prolactin period") was reversed in one instance without altering the effects seen under the usual sequence. The hypertonic amniotic fluid consisted of 9 parts Ringer's-lactate solution and 1 part adult rhesus serum, with 165 meq/1 extra of both sodium and chloride ions; the isotonic fluid consisted of 10% rhesus serum in mammalian Ringer's-lactate solution, while the
564 The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 February 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.
AMNIOTIC PROLACTIN—FETAL FLUID BALANCE hypotonic amniotic fluid was 10% rhesus serum in Ringer's-lactate diluted 1:1 with sterile distilled water. All of these fluids prepared for intraamniotic injection were maintained at 37 C prior to use. To ascertain the actual levels of rhesus prolactin (rhPRL) after the amniotic fluid had been partially replaced with isotonic fluid, the hormone was measured by radioimmunoassay (2) in several samples of the fluid drawn from the amniotic cavity over a 2-h period, in 2 pregnant animals. In each of the 6 amniotic fluid replacement experiments, 0.6 ml heparinized blood samples were drawn from indwelling catheters in the maternal saphenous vein or from a fetal interplacental vein, as well as 0.6 ml amniotic fluid from an indwelling amniotic catheter. After each fetal blood sample was drawn, 0.6 ml heparinized maternal blood drawn the morning of anesthesia and kept at 37 C was infused back into the fetal vessel catheter, which would assure a gradual decrease in fetal hematocrit of only 3-6%over a 2-h period because the maternal hematocrit was 20% lower, on the average, than the initial fetal hematocrit. Blood samples were analyzed for hematocrit. After centrifugation plasma was analyzed for sodium and potassium ion concentrations by flame photometry, for osmolarity in a freezing-point depression apparatus, and 2800*
40
60
80
TIME(min) FIG. 1. Amniotic fluid PRL in the 2 h after 60-87% replacement of the fluid with PRL-free fluid. Points designated by crosses, at 10 min, signify calculated nadir after amniotic fluid (AF) replacement. Closed circles represent measured values of animal no. 90. Open circles represent no. 86.
565
TABLE 1. Effects of intraamniotic injection of ovine prolactin and other substances on rhesus amniotic fluid volume (AF Vol) (115-156 days gestation) Mean % AAF
No. monkeys
Hours postinj.
1 ml saline
7
2
10 mg NIH oPRL
9 2
2 24
1
48
LOmgNIHoPRL
1
2
-66
10 mg bovine serum albumin
1
2
0
Vasopressin, 1 mU
1
2
+8
Vasopressin, 100 mU
1
2
-10
Substance injected
Vol ± SEM*
-4 ± 5 -49 ± 8t -56(-50 and -63) 0
* SEM = standard error of the mean, f P < .001 in t test for difference from saline-injected control experiments.
for total protein concentration by the method of Lowry et al. (6). Amniotic fluid samples were similarly subjected to all analyses excepting hematocrit. Amniotic fluid volumes were determined by the Congo Red dye dilution method before and after each 2-h period. Calculations of fetal extracellular fluid (ECF) volume were made on the assumption that it was equivalent to 35% of the body weight [human newborn data (7)]. The fetal body weight was determined at delivery within 40 h of the experiment or, in rare instances, derived from published fetal weight-gestational age data (8) to which our rhesus colony has been found to conform in the past. Statistical significance of differences in fetal ECF water and Na+ content at low vs. high prolactin concentrations in the AF was evaluated by t tests based on the variance of differences between pairs of samples drawn at equivalent times under the two conditions in each animal. Thus, the t tests evaluated a null hypothesis that the mean difference between pairs of samples over two hours from the two treatment periods did not differ from zero.
Results Figure 1 demonstrates the reaccumulation of rhPRL in the amniotic fluid of 2 monkeys after replacement of 60 and 87% of
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 February 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.
JOSIMOVICH, MERISKO AND BOCCELLA
566
Endo i 1977 Vo! 100 i No 2
9" —9-
FIG. 2. Changes in fetal hematocrit ( • • ) and in serum protein concentration (O O) in the face of hypertonic amniotic fluid replacement with low amniotic fluid prolactin (on left) and with high amniotic fluid prolactin (on right).
-60
I
LOW PROLACTIN I I
HIGH PROLACTIN I I
I
TIME (mm)
the amniotic fluid volume. A rapid return from a theoretical low of 13% of the original level in 1 monkey, and of 40% of the original level in the other animals, at 10 min, to 65-72% of the original levels had occurred by 35-40 min. A more gradual further increase toward original values was seen in the ensuing 1-2 h. It can be calculated that in the rapid reaccumulation phase, 70-160 fig rhPRL were quickly released into the amniotic fluid. Table 1 shows changes in amniotic fluid volume at differing times after the injection of various substances into the amniotic cavity. Saline (0.9%), bovine serum albumin, and vasopressin failed to cause a significant change in volume, while 1.0 or 10 mg NIH ovine prolactin consistently caused marked decreases by 2 and 24 h, although a return to normal amniotic volume was evident by 48 h in the one study carried out for that long a period. No differences were seen in either the reduction of volume caused by oPRL or in the failure of saline alone to cause such reductions, when the original amniotic fluid was drained and replaced with hypertonic, isotonic, or hypotonic fluid. Figure 2 shows the similarity in the changes in fetal plasma protein concentrations and blood hematocrit in one of the studies in which amniotic fluid was partially
replaced with hypertonic fluid in two 2-h periods, the second time with 10 mg ovine prolactin. Calculated changes in fetal ECF volume were based on such changes in blood hematocrit as described in Materials and Methods. Thisfigureshows that changes in calculated ECF volume based on hematocrit are supported by parallel changes in plasma protein concentration, such that the marked changes seen in the left-hand panel were greatly attenuated for both red cell volume and protein concentration by the presence of high levels of ovine prolactin, as seen in the right-hand panels. The changes in ECF calculated from the hematocrit change data in Fig. 2 are displayed in the upper left-hand panel of Fig. 3 and in the upper panel of Fig. 4. Figure 3 shows the effects of altered tonicity of amniotic fluid with lowered prolactin levels at different amniotic electrolyte tonicities on fetal ECF and amniotic volume on the left, and on fetal ECF and amniotic sodium ion content on the right. In the face of hypertonic amniotic fluid (one study shown in the upper panels), there is an influx of water and sodium ion into the ECF, at times doubling that originally present. In the amniotic fluid, on the other hand, a temporary loss of lesser amounts of water and sodium is evident.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 February 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.
AMNIOTIC PROLACTIN—FETAL FLUID BALANCE
567
HIGH PROLACTIN LOW PROLACTIN After partial replacement of amniotic fluid with isotonic fluid (one study in the middle panels), there is no significant change in A either amniotic or fetal ECF water volume or Na+ content. In the face of hypotonic amniotic fluid replacement (one study in the lower panels), there is an egress from the ECF of water and sodium ion of about 25% of original values, accompanied by a minimal decrease in amniotic water and sodium ion content. Figure 4 shows all three studies in which the amniotic fluid was largely replaced with hypertonic amniotic solution, when the endogenous rhPRL was low (left-hand panels) there was a consistent ingress of fluid into the fetal ECF. In the presence of 10 mg additional oPRL (right-hand panels) the ingress of water was reversed or prevented to a statistically significant degree. Figure 5 shows similar patterns of ingress of Na+, FIG. 4. Changes in fetal ECF water content in three
studies employing hypertonic AF accompanied by low endogenous PRL levels (left-hand panels) or after augmentation of AF PRL with 10 mg oPRL (righthand panels). Original calculated values given in figure. P values of significance between left and righthand panels, calculated as described in Materials and Methods, are all ./ 1.0-10 mg ovine prolactin, an effect which o persisted for 24 h but was no longer visible in the one monkey examined at 48 h. In O 30 contrast, no significant change was caused in any other monkeys injected with saline cr l0 or bovine serum albumin (Table 1), regardE less of the tonicity of the amniotic fluid. < o That ovine prolactin might exert biologic effects in the rhesus is supported by its 20 ft 22m.q lactogenic effects in the non-pregnant ani10 mal (9). It is also obvious that the effects .—. on amniotic fluid volume were not attribut0 40 80 120 0 40 80 120 able to contamination of the NIH oPRL TIME (min) with vasopressin (10), since 1.0 and 100 mU FIG. 5. Changes in fetal ECF sodium ion content vasopressin failed to affect amniotic fluid in same three studies shown in Fig. 4, employing volume, even though the larger dose was hypertonic AF accompanied by low endogenous PRL absorbed into the maternal circulation at levels (left-hand panels) or after augmentation of AF PRL with 10 mg oPRL (right-hand panels). Original high enough levels to cause marked oliguria. calculated values given in figure. P values (calculations Artificial lowering of the endogenous prodescribed in Materials and Methods) of significance lactin by largely replacing the amniotic between left and right-hand panels are (from top to fluid with prolactin-free fluid in 6 of the 10 bottom experiment):
ABSTRACT. Administration of 1-10 mg ovine pituitary prolactin (oPRL) into the amniotic fluid of 10 rhesus monkeys in the last third of gestation consistently caused a decrease in amniotic fluid volume not seen when saline, vasopressin, or bovine serum albumin were injected into 9 other monkeys. The effects lasted for 24 h. Intraamniotic injection of 10 mg oPRL prevented or reversed a doubling of water and electrolyte content of the fetal extracellular
L
ARGE amounts of pituitary prolactin may be found in the amniotic fluid of the rhesus monkey (1,2). The source of the large reservoir has been shown to derive at least in part from the maternal circulation (2). Although amniotic prolactin may be bound in large quantities to the amnion, certain fetal tissues such as lung, and placenta whose cell membranes appear to bind this hormone with high affinity may also bind lesser amounts of amniotic prolactin (3). Because of the well known effects of prolactin on water and sodium conservation when salt water fish or larval amphibians enter fresh water (1,4), the present series of studies were undertaken to define possible roles of altered amniotic prolactin levels on fetal electrolyte and water conservation in the face of different amniotic fluid tonicities. Materials and Methods Thirteen rhesus monkeys bred at the MageeWomens Hospital facility, underwent phencyclidine and pentobarbital medication at 115-156 days gestation (2). Amniotic fluid (AF) volumes were determined by the Congo Red dye dilution technique (5)
Received December 22, 1975. 1 This work was supported by NIH Grant HD-06929. 2 Reprint requests should be sent to the first author at Magee-Womens Hospital.
whose precision was confirmed to be within 10% of actual volumes directly measured at the time of cesarean section in two instances. Volumes were determined before, and at 2, 24, and 48 h after injection of 1 ml normal saline, 1 ml saline containing 1 or 10 mg ovine prolactin (P-S-9, obtained from the Hormone Distribution Officer, National Institutes of Health, Bethesda, Md.), 10 mg bovine serum albumin, or 1 or 100 mU vasopressin (Pitressin, Parke, Davis & Co., Detroit, Mich.). In 6 of the monkeys, additional halothane anesthesia was performed and indwelling catheters placed in the maternal saphenous vein, amniotic cavity and fetal interplacental vein (2) where sampling was made at 3-20 min intervals over two 2-h periods. The two periods were distinguished by differences in the amniotic fluid composition: after 60-87% replacement of the amniotic fluid with hypertonic, isotonic or hypotonic amniotic fluid with no added prolactin; or after similar replacement of amniotic fluid with 10 mg ovine prolactin. The order of the 2-h experiments under the influence of lowered endogenous amniotic fluid prolactin ("low prolactin period") or high prolactin levels ("high prolactin period") was reversed in one instance without altering the effects seen under the usual sequence. The hypertonic amniotic fluid consisted of 9 parts Ringer's-lactate solution and 1 part adult rhesus serum, with 165 meq/1 extra of both sodium and chloride ions; the isotonic fluid consisted of 10% rhesus serum in mammalian Ringer's-lactate solution, while the
564 The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 February 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.
AMNIOTIC PROLACTIN—FETAL FLUID BALANCE hypotonic amniotic fluid was 10% rhesus serum in Ringer's-lactate diluted 1:1 with sterile distilled water. All of these fluids prepared for intraamniotic injection were maintained at 37 C prior to use. To ascertain the actual levels of rhesus prolactin (rhPRL) after the amniotic fluid had been partially replaced with isotonic fluid, the hormone was measured by radioimmunoassay (2) in several samples of the fluid drawn from the amniotic cavity over a 2-h period, in 2 pregnant animals. In each of the 6 amniotic fluid replacement experiments, 0.6 ml heparinized blood samples were drawn from indwelling catheters in the maternal saphenous vein or from a fetal interplacental vein, as well as 0.6 ml amniotic fluid from an indwelling amniotic catheter. After each fetal blood sample was drawn, 0.6 ml heparinized maternal blood drawn the morning of anesthesia and kept at 37 C was infused back into the fetal vessel catheter, which would assure a gradual decrease in fetal hematocrit of only 3-6%over a 2-h period because the maternal hematocrit was 20% lower, on the average, than the initial fetal hematocrit. Blood samples were analyzed for hematocrit. After centrifugation plasma was analyzed for sodium and potassium ion concentrations by flame photometry, for osmolarity in a freezing-point depression apparatus, and 2800*
40
60
80
TIME(min) FIG. 1. Amniotic fluid PRL in the 2 h after 60-87% replacement of the fluid with PRL-free fluid. Points designated by crosses, at 10 min, signify calculated nadir after amniotic fluid (AF) replacement. Closed circles represent measured values of animal no. 90. Open circles represent no. 86.
565
TABLE 1. Effects of intraamniotic injection of ovine prolactin and other substances on rhesus amniotic fluid volume (AF Vol) (115-156 days gestation) Mean % AAF
No. monkeys
Hours postinj.
1 ml saline
7
2
10 mg NIH oPRL
9 2
2 24
1
48
LOmgNIHoPRL
1
2
-66
10 mg bovine serum albumin
1
2
0
Vasopressin, 1 mU
1
2
+8
Vasopressin, 100 mU
1
2
-10
Substance injected
Vol ± SEM*
-4 ± 5 -49 ± 8t -56(-50 and -63) 0
* SEM = standard error of the mean, f P < .001 in t test for difference from saline-injected control experiments.
for total protein concentration by the method of Lowry et al. (6). Amniotic fluid samples were similarly subjected to all analyses excepting hematocrit. Amniotic fluid volumes were determined by the Congo Red dye dilution method before and after each 2-h period. Calculations of fetal extracellular fluid (ECF) volume were made on the assumption that it was equivalent to 35% of the body weight [human newborn data (7)]. The fetal body weight was determined at delivery within 40 h of the experiment or, in rare instances, derived from published fetal weight-gestational age data (8) to which our rhesus colony has been found to conform in the past. Statistical significance of differences in fetal ECF water and Na+ content at low vs. high prolactin concentrations in the AF was evaluated by t tests based on the variance of differences between pairs of samples drawn at equivalent times under the two conditions in each animal. Thus, the t tests evaluated a null hypothesis that the mean difference between pairs of samples over two hours from the two treatment periods did not differ from zero.
Results Figure 1 demonstrates the reaccumulation of rhPRL in the amniotic fluid of 2 monkeys after replacement of 60 and 87% of
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 February 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.
JOSIMOVICH, MERISKO AND BOCCELLA
566
Endo i 1977 Vo! 100 i No 2
9" —9-
FIG. 2. Changes in fetal hematocrit ( • • ) and in serum protein concentration (O O) in the face of hypertonic amniotic fluid replacement with low amniotic fluid prolactin (on left) and with high amniotic fluid prolactin (on right).
-60
I
LOW PROLACTIN I I
HIGH PROLACTIN I I
I
TIME (mm)
the amniotic fluid volume. A rapid return from a theoretical low of 13% of the original level in 1 monkey, and of 40% of the original level in the other animals, at 10 min, to 65-72% of the original levels had occurred by 35-40 min. A more gradual further increase toward original values was seen in the ensuing 1-2 h. It can be calculated that in the rapid reaccumulation phase, 70-160 fig rhPRL were quickly released into the amniotic fluid. Table 1 shows changes in amniotic fluid volume at differing times after the injection of various substances into the amniotic cavity. Saline (0.9%), bovine serum albumin, and vasopressin failed to cause a significant change in volume, while 1.0 or 10 mg NIH ovine prolactin consistently caused marked decreases by 2 and 24 h, although a return to normal amniotic volume was evident by 48 h in the one study carried out for that long a period. No differences were seen in either the reduction of volume caused by oPRL or in the failure of saline alone to cause such reductions, when the original amniotic fluid was drained and replaced with hypertonic, isotonic, or hypotonic fluid. Figure 2 shows the similarity in the changes in fetal plasma protein concentrations and blood hematocrit in one of the studies in which amniotic fluid was partially
replaced with hypertonic fluid in two 2-h periods, the second time with 10 mg ovine prolactin. Calculated changes in fetal ECF volume were based on such changes in blood hematocrit as described in Materials and Methods. Thisfigureshows that changes in calculated ECF volume based on hematocrit are supported by parallel changes in plasma protein concentration, such that the marked changes seen in the left-hand panel were greatly attenuated for both red cell volume and protein concentration by the presence of high levels of ovine prolactin, as seen in the right-hand panels. The changes in ECF calculated from the hematocrit change data in Fig. 2 are displayed in the upper left-hand panel of Fig. 3 and in the upper panel of Fig. 4. Figure 3 shows the effects of altered tonicity of amniotic fluid with lowered prolactin levels at different amniotic electrolyte tonicities on fetal ECF and amniotic volume on the left, and on fetal ECF and amniotic sodium ion content on the right. In the face of hypertonic amniotic fluid (one study shown in the upper panels), there is an influx of water and sodium ion into the ECF, at times doubling that originally present. In the amniotic fluid, on the other hand, a temporary loss of lesser amounts of water and sodium is evident.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 25 February 2015. at 17:28 For personal use only. No other uses without permission. . All rights reserved.
AMNIOTIC PROLACTIN—FETAL FLUID BALANCE
567
HIGH PROLACTIN LOW PROLACTIN After partial replacement of amniotic fluid with isotonic fluid (one study in the middle panels), there is no significant change in A either amniotic or fetal ECF water volume or Na+ content. In the face of hypotonic amniotic fluid replacement (one study in the lower panels), there is an egress from the ECF of water and sodium ion of about 25% of original values, accompanied by a minimal decrease in amniotic water and sodium ion content. Figure 4 shows all three studies in which the amniotic fluid was largely replaced with hypertonic amniotic solution, when the endogenous rhPRL was low (left-hand panels) there was a consistent ingress of fluid into the fetal ECF. In the presence of 10 mg additional oPRL (right-hand panels) the ingress of water was reversed or prevented to a statistically significant degree. Figure 5 shows similar patterns of ingress of Na+, FIG. 4. Changes in fetal ECF water content in three
studies employing hypertonic AF accompanied by low endogenous PRL levels (left-hand panels) or after augmentation of AF PRL with 10 mg oPRL (righthand panels). Original calculated values given in figure. P values of significance between left and righthand panels, calculated as described in Materials and Methods, are all ./ 1.0-10 mg ovine prolactin, an effect which o persisted for 24 h but was no longer visible in the one monkey examined at 48 h. In O 30 contrast, no significant change was caused in any other monkeys injected with saline cr l0 or bovine serum albumin (Table 1), regardE less of the tonicity of the amniotic fluid. < o That ovine prolactin might exert biologic effects in the rhesus is supported by its 20 ft 22m.q lactogenic effects in the non-pregnant ani10 mal (9). It is also obvious that the effects .—. on amniotic fluid volume were not attribut0 40 80 120 0 40 80 120 able to contamination of the NIH oPRL TIME (min) with vasopressin (10), since 1.0 and 100 mU FIG. 5. Changes in fetal ECF sodium ion content vasopressin failed to affect amniotic fluid in same three studies shown in Fig. 4, employing volume, even though the larger dose was hypertonic AF accompanied by low endogenous PRL absorbed into the maternal circulation at levels (left-hand panels) or after augmentation of AF PRL with 10 mg oPRL (right-hand panels). Original high enough levels to cause marked oliguria. calculated values given in figure. P values (calculations Artificial lowering of the endogenous prodescribed in Materials and Methods) of significance lactin by largely replacing the amniotic between left and right-hand panels are (from top to fluid with prolactin-free fluid in 6 of the 10 bottom experiment):
