Fetal-maternal alpha-fetoprotein
distribution
of ovine
P. C. W. LAI, G. J. MEARS, G. R. VAN PETTEN, D. M. HAY, AND F. L. LORSCHEIDER Divisions of Medical Physiology, Pharmacology and Therapeutics, and Obestetrics Gynaecology, Faculty of Medicine, University of Calgary and Alberta Children’s Hospitul Research Centre, Calgary, Alberta, Canada T2N lN4
LAI, P. C. W., G. J. MEARS, G. It. VAN PETTEN, Il. M. HAY, AND F. L. LORSCHEIDER. Fetal-maternal distribution of ovine alpha-fetoprotein. Am. J. Physiol. 235(l): E27-E31, 1978 or Am J. Physiol.: Endocrinol. Metab. Gastrointest, Physiol. 4(l): E27-E31, 1978. -The concentration of ovine alpha-fetoprotein (AFP) was determined in sheep fetal, maternal, and neonatal sera and amniotic fluid by radioimmunoassay. The fetal serum AFP concentration was highest during the first third of pregnancy and continued to decline with fetal and neonatal development. Total fetal synthesis of AFP was highest during the late middle and early latter third of pregnancy. Amniotic fluid AFP concentration was highest during the early middle third of pregnancy. Maternal serum AFP was not elevated above nonpregnant control levels during the first two-thirds of pregnancy, but instead showed a tendency to be elevated only during the last third of pregnancy. The fetal physiological distributions of ovine and human AFP in fetal serum and amniotic fluid appear to be similar, whereas in maternal serum the pattern of AFP levels differs in the two species as a function of gestational age. amniotic fluid; pregnancy; noassay; neonatal serum
fetal serum
proteins;
radioimmu-
(AFP) is a fetus-specific serum protein found in aquatic and terrestrial vkrtebrates (9, -11). This protein is-produced by the human yolk sac and fetal liver (10) and is secreted into the fetal circulation. The synthesis of AFP has been investigated in man (8), rabbits (4), and rats (13). In these species, the fetal synthesis of AFP is curtailed before birth, and only trace amounts of the protein are detectable in adult serum (23, 25). In man, elevated serum AFP levels in the adult may be used clinically as a marker for teratoma and hepatic diseases (1, 2). In addition, abnormally high levels of AFP in human amniotic fluid can be used for the antenatal diagnosis of birth defects (5, 17). Although the clinical diagnostic value of AFP is well established, the specific function of AFP is not known (24). Gitlin and Boesman (8) showed that the fetal synthesis of human AFP reaches maximum levels between 20 and 32 wk of gestation. Recently, highly sensitive radioimmunoassay techniques were employed to measure the concentration of AFP in maternal serum and amniotic fluid in man (18, 26) and rats (13). We reported ALPHA-FETOPROTEIN
0363-6100/78/0000-0000$01.25 Copyright
0
1978the
American
Physiological
and
a difference in the gestational pattern of AFP levels in maternal and fetal vascular compartments between man and rat (13). Thus the rat appeared to be of limited value as an experimental animal model for studying the function of AFP and its relationship to birth defects and abnormal pregnancies. The objective of this experiment was to use the recently developed radioimmunoassay for ovine AFP (15) to quantitate the concentration of AFP in various physiological fluids and to investigate the basic physiology of AFP in the pregnant sheep and fetus. MATERIALS
AND METHODS
Animals, sera, and amniotic fluid. All sheep used in this study were maintained and bred at the Animal Resources Centre at the University of Calgary prior to their transfer to the Medical Vivarium for surgery and experimentation. The day of mating was observed and considered to be day 0 of gestation. Maternal sera were collected from four ewes via the jugular vein at weekly intervals beginning at day 25 of gestation until term (approximate1 y 143 days). Pregnancy was confirmed between 90 and 100 days by fluoroscopy. Under halothane general anesthesia, the femoral artery, saphenous vein, and amniotic sac of the conceptus were chronically cannulated at 110-115 days gestation with silicone rubber catheters, which were exteriorized by surgical procedures previously described (28). In addition, samples were obtained from several other similarly cannulated fetuses from 110 days to term. In the case of twin pregnancies, only one fetus tias cannulated. Weekly samples of 4 ml amniotic fluid and 2.5 ml fetal blood were withdrawn via the catheters. Blood samples were also obtained from two newborn lambs via the jugular vein at 0.5, 5, 9, and 15 days postpartum. Amniotic fluid and fetal sera samples on or before 83 days gestation were obtained either by cesarean section within 5 min after intravenous infusion of sodium pentobarbital (25 mg/kg body wt) to anesthetize the ewe or by laparotomy after halothane anesthesia. Control serum samples were obtained from six normal nonpregnant ewes via the jugular vein. Blood samples and amniotic fluid specimens were centrifuged at 1,500 x g for 15 min to remove cells, and the supernatants were stored at -20°C until assayed. Amniotic fluid specimens Society
E27
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (132.174.255.223) on January 12, 2019.
LA1 ET AL.
E28 contaminated with fetal red blood cells were discarded. The body weight of sheep fetuses was estimated using the standard ovine fetal growth curve established by Stephenson (27) with the exception that actual weights of 36-day-old fetuses and newborn lambs were obtained during our experiments. Radioimmunoassay of ovine AFP. Determination of AFP concentration was carried out using a doubleantibody radioimmunoassay procedure as previously described (15). The assay has a limit of detection of 2 nglml and employs an ovine AFP standard, which was prepared and calibrated as previously described (X,16>. Detection of AFP by electrophoresis. The presence of AFP in fetal sera was also confirmed by analytical disc polyacrylamide gel electrophoresis with a running pH 9.3, 7.5% polyacrylamide gel, and buffers as described (13). A 4.5~1 serum sample was applied to each gel column. After electrophoresis, each gel was stained with 0.75% aniline blue black dissolved in 7% acetic acid for 1 h and destained by washing with three 200-ml aliquots of 7% acetic acid. Estimation of total protein concentration. Concentrations of total protein in sera were determined by the method of Lowry et al. (21) with bovine serum albumin (Calbiochem., La Jolla, Calif.) used as the protein standard for the assay.
6060.
TOTO-
99 c gg
6060-
z 00pp
6060-
E E 22 22 ww $$
.
40403030-
0 20
40
60
DAYS
60
GESTATION
100
120
140 t
Parturition
FIG. 2. Changes in fetal serum protein concentration tion of gestation. Curve was fitted by inspection.
as a func-
RESULTS
Figure 1 illustrates the changes in the concentration of AFP in fetal serum as determined by radioimmunoassay at various stages of gestation. The AFP concentration was highest during the first third of pregnancy (4.5 mglml at 36 days gestation) and continued to decline to less than 100 pg/ml 1 wk before term. Figure 2 illustrates the concentration of total fetal serum protein as a function of gestation. A rapid increase in concentration of serum protein was evident a few weeks before term. Figure 3 shows the analytical disc polyacrylamide gel electrophoresis of proteins in fetal sera at different stages of gestation and compared to that of maternal and adult ram sera. AFP in fetal serum can be visualized as a protein fraction with slower electrophoretic
DAYS GESTATION FIG. 1. Fetal serum AFP concentration Curve was fitted by inspection.
Partthon
as a function of gestation.
FIG. 3. Analytical disc polyacrylamide gel electrophoresis of sheep serum proteins. a, go-day fetal serum; b, 83-day fetal serum; c, l&day fetal serum; d, maternal serum from latter third of pregTf, nancy; e, ram serum. Alb, albumin; AFP, alpha-fetoprotein; transferrin. Electrophoretic mobility of AFP relative to other serum proteins was previously identified (14).
mobility than albumin, which decreases in concentration with gestational age and cannot be detected in maternal or ram sera with this technique. Figure 4 is a graph of fetal serum AFP concentration x fetal weight as a function of gestation. Peak values were attained between 80 and 115 days gestation with decreases thereafter during fetal and neonatal development. Figure 5 illustrates the postpartum catabolic clearance of AFP in a case of twin lambs. The mean half-life for the disannearance of AFP from newborn lamb serum until 15 days postpartum was determined by plotting
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (132.174.255.223) on January 12, 2019.
OVINE
ALPHA-FETOPROTEIN
IN
PREGNANCY
E29
logarithmically transformed AFP values versus time, and the slope was calculated by linear regression. The mean half-life in these two animals during the first 15 days postpartum was approximately 86 h. Figure 6 illustrates the changes in concentration of AFP in amniotic fluid as a function of gestational age. Peak values of approximately 22 pg/ml were reached early in the middle third of pregnancy. Figure 7 illustrates the concentration of AFP in maternal serum at various stages of gestation compared against the mean basal level of AFP t 1 SE (29.9 t 5.8 rig/ml) in six control nonpregnant adult ewes. Maternal serum AFP was not statistically different from control
DAYS FIG.
function
s
6. Changes of gestation
Parturition
GESTATION
in concentration of AFP in amniotic Curve was fitted by inspection.
fluid
as a
4m-
3
2 #
200-
a z
0'
I 20
40
60 DAYS
FIG.
Curve
4. Fetal serum AFP was fitted by inspection.
x
fetal
80
100
120
1 4ot Parturition
GESTATION
weight
as a function
of gestation.
0'
1 20
1 40
I 60 PAYS
1 80 GESTATION
A
1
100
120
1 140 t Plrturitiofl
FIG, 7. Maternal serum AFP concentration as a function of gestation. Each point represents mean serum AFP value for ewes within a lo-day interval. Vertical bars represent standard error. Shaded area represents mean + 1 SE of serum AFP values for 6 control nonpregnant ewes, (*, one-tailed Student t test for comparison of experimental and control values significant at P < 0.025.)
serum values during the first two-thirds of pregnancy. The level of maternal serum AFP increased significantly early in the last third of pregnancy and peaked a few weeks before birth. ,
DISCUSSION
100
POSTPARTUM
FIG.
newborn
5. Decline lambs.
in
serum
300
200;
AFP
400
{hours)
concentration
in
a case
of twin
The concentration of fetal serum AFP is highest during the first third of pregnancy, declines during fetal development, and continues to disappear from neonatal serum. Disc polyacrylamide gel electrophoresis of fetal sera at various stages of gestation not only supports the
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (132.174.255.223) on January 12, 2019.
E30 radioimmunoassay data, but also clearly indicates that AFP is a major serum alpha-globulin in early gestation. Other investigators have demonstrated that ovine AFP is chemically-different from fetuin, a protein not only abundant in fetal serum but also detectable in adult serum by electrophoresis (19, 20). Our data in sheep demonstrate that there are species differences with respect to AFP concentration in fetal serum as a function of gestation. As in man 09, serum AFP is highest duing the first third of pregnancy in fetal sheep, but in rabbits and rats AFP levels peak near term (4, 13). Total serum protein in ovine fetal serum increases with advancing gestation. This increase in total protein concentration, in spite of a reciprocal decrease in AFP concentration with fetal development, can be primarily attributed to increases in serum albumin and transferrin which is corroborated in our disc gel electrophoresis data. Our electrophoresis data also agree with those previously reported by others (3). It is the re f ore advantageous to use fetal sera obtained from early gestation as the source for the isolation and purification of AFP because the ratio of the concentration of AFP to total protein is highest at this time, The concentration of any protein in plasma is not only a function of its rate of synthesis and degradation, but also a function of its compartmental distribution and size of the body fluid spaces. Proteins with molecular weights similar to that of AFP are normally not restricted to the vascular compartment, but are distributed almost equally between interstitial fluids and plasma (6). Because the fetus is growing rapidly, the concentration of AFP in fetal plasma cannot give a meaningful indication of the level of fetal hepatic synthesis of this protein at any stage of fetal development. Gitlin (7, 8) suggested that if the plasma concentration of AFP at a given stage of development is multiplied by the fetal weight at that time, an approximation of the relative amounts of AFP synthesized by the fetus at different points in gestation may be obtained. Applying this same calculation to our data, we conclude that the amount of ovine AFP synthesized is highest between 80 and 115 days gestation and that synthesis is gradually curtailed just a few weeks before birth. At present, the detailed mechanism responsible for the cessation of fetal synthesis of AFP before birth remains to be elucidated. It is interesting to note that the peak synthesis of both human (8) and ovine AFP occurs in the same stage of fetal development, i.e., the late middle and early latter third of pregnancy. AFP can be detected in the amniotic fluid of the sheep conceptus. The source of AFP in amniotic fluid is thought to be fetal rather than maternal because the concentration of AFP in fetal serum is approximately ZOO-300 times higher than that in amniotic fluid
LA1
ET
AL.
throughout gestation. In addition, there is a corresponding decrease in amniotic fluid AFP concentration probably as a result of a decline in fetal serum AFP concentration with advancing gestation. There is a difference in the pattern of appearance of AFP in amniotic fluid among mammalian species. Our data from sheep is similar to human data (22) in that the peak level of AFP is reached in early gestation. In contrast, the concentration of rat amniotic fluid AFP peaks late in gestation (13). Furthermore, the concentration of AFP in rat amniotic fluid is very high, on the order of several hundred micrograms per milliliter (13), but in both man (22) and sheep the concentration is less than 50 pg/ml. AFP can also be detected in the sera of adult nonpregnant ewes (15). In the present study, we showed that maternal serum AFP concentration is not significantly elevated from the basal level during the first two-thirds of pregnancy. Only a moderate elevation in maternal serum AFP of approximately 2 times the basal level can be seen early during the latter third of pregnancy. This is in contrast to the more significant elevation observed in human serum AFP during pregnancy (12). The low level of AFP in ovine maternal serum, approximately O.OOl-0.1% that of fetal serum, can be attributed either to a minimum degree of transplacental and transamniotic transfer of AFP from the conceptus or to a high rate of catabolic clearance of AFP in the ewe, The answer awaits future kinetic and metabolic studies. The sheep should be a suitable animal model for studying AFP because the size of the fetus approximates that of man, and the fetal blood vessels and amniotic cavity can be cannulated to permit serial and simultaneous sampling of fetomaternal compartments. In addition, our data suggest that ovine fetal physiology of AFP is similar to that of man, making the sheep a potential model for the study of the function and pathophysiology of AFP. However, the applicability of the sheep model for studying AFP in complicated pregnancies such as fetal neural tube defects may be limited by our finding that sheep maternal serum AFP levels differ in some respects from man, The technical assistance of Mr. G. Paquette, D. Teichroeb, P, van Muyden, Ms. Y. Brideau, G. Cochrane, P. Hamilton, and H. Mathison is gratefully acknowledged. This study was supported by the Medical Research Council of Canada Grant MA-5292 and a grant from The Alberta Children’s Hospital Research Foundation. P. C. W. Lai is a recipient of a Medical Research Council Studentship. Send requests for reprints to: F. L. Lorscheider, Faculty of Medicine, Div. of Medical Physiology, Univ, of Calgary, 2920 24th Ave., N. W., Calgary, Canada T2N lN4.
Received
31 October
1977; accepted
in final
form
7 March
1978.
REFERENCES 1. ADINOLFI, A., M. ADINOLFI, AND M. H. LESSOF. Alpha-fetoprotein during development and in disease. J. Med. Genetics 12: 138151, 1975. 2. ALPERT, E. Human alphal-fetoprotein (AFP): developmental biology and clinical significance. Progr. Liver Diseases 5: 337349, 1976.
3. BARBORIAK,
J. J., G. MESCHIA, D. H. BARRON, AND Blood plasma proteins in fetal goats and sheep. Sot. Exptl, Biol. Med. 98: 635-637, 1958. 4, BRANCH, W. R. The ontogeny of alpha-foetoprotein in the and neonatal rabbit, and its experimental induction in rabbits. Intern. J. Cancer 10: 451-457, 1972. COWGILL.
R. Proc.
G.
foetal adult
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (132.174.255.223) on January 12, 2019.
OVINE
ALPHA-FETOPROTEIN
IN
PREGNANCY
5. BROCK, D. J. H. Alpha-fetoprotein and the prenatal diagnosis of central nervous system disorders. A review. Cltild’s Brain 2: l23, 1976. 6. GITLIN, D. Distribution dynamics of circulating and extravascular Pa1 plasma proteins. Ann. N. Y. Acad. Sci. 70: 122-136, 1957. 7. GITLIN, D., Normal biology of a-fetoprotein. Ann. N. Y. Acad. Sci. 259: 7-16, 1975. 8. GITLIN, D., AND M. BOESMAN. Serum a-f&protein, albumin, and y-G-globulin in the human conceptus. J. Clin. Invest. 45: 1826-1838,1966. 9. GITLIN, D., AND M. BOESMAN. Fetus-specific serum proteins in several mammals and their relation to human a-fetoprotein. Comp. Biochem. Physiol. 21: 327-336, 1967, 10. GITLIN, D., AND M. BOESMAN. Sites of serum a-fetoprotein synthesis in the human and in the rat. J. CZin. Invest. 46: lOlO1016, 1967. 11. GITLIN, D., A. PERRICELLI, AND J. D. GITLIN. The presence of serum a-fetoprotein in sharks and its synthesis by fetal gastrointestinal tract and liver. Camp. Biochem. PhysioZ. 46B: 207-215, 1973. 12. HAY, D. M., P. I. FORRESTER, R. L. HANCOCK, AND F. L. LORSCHEIDER. Maternal serum alpha-feteprotein in normal pregnancy. Brit. J. Obstet. Gynaecol. 83: 534-538, 1976. 13. LAI, P. C. W., P. I. FORRESTER, R. L. HANCOCK, D. M. HAY, AND F. L. L~R~CHEIDER. Rat alpha-fetoprotein: isolation, radioimmunoassay and fetal-maternal distribution during pregnancy. J. Reprod. Fertility 48: 1-8, 1976. 14. LAI, P. C, W., D. M. HAY, AND F. L, LORSCHEIDER. Purification of ovine alpha-fetoprotein by preparative electrophoresis. Stand. J. Immunol., Suppl. In press. 15. LAI, P. C. W., D. M. HAY, AND F. L. LORSCHEIDER, Radioimmunoassay of ovine alpha-fetoprotein. J. Immunol. Methods. 20: l-10, 1978. 16. LAI, P. C. W., D. M. HAY, E. H. PETERS, AND F. L. LORSCHEIDER. Immunochemical purification and characterization of ovine afetoprotein. Biochim. Biophys. Acta 493: 201-209, 1977. 17. LAU, H. L., AND S. E. LINKINS. Alpha-fetoprotein. Am. J,
E31 Obstet. Gynecol. 124: 533-554, 1976. 18. LEEK, A. E., C. F. Ruoss, M. J. KITAW, AND T. CHARD. Maternal plasma alpha-fetoprotein levels in the second half of normal pregnancy: relationship to fetal weight and maternal age and parity. Brit. J. Obstet. Gynaecol. 82: 669-673, 1975. 19. MARTI, J., S. ALIAW, C. BONFIL~, C. VIGNE, AND J. MORETTI. Preparation, constantes physiques et constitution chimique de la fetuine d’agneau. Biochim. Biophys. Acta 303: 348-359, 1973. 20. MARTI, J., S. ALIAW, AND J. MORETTI. Studies on AFP from ovine and bovine fetuses. In: Protides of Biological Fluids, edited by H. Peeters. Oxford: Pergamon, 1976, p* 259-262. 21. LOWRY, 0. H., N. J. ROSEBROUGH, A. L. FARR, AND R. J. RANDALL. Protein measurement with the folin phenol reagent. J. BioL. Chem. 193: 265-275, 1951. 22. RANDLE, G. H., AND K. N. CUMBERBATCH. Alpha-fetoprotein levels in amniotic fluid in normal pregnancy and in pregnancy complicated by anencephaly. J. Obstet. Gynaecol. Brit. Commonwealth 80: 1054-1058, 1973. 23. RUOSLAHTI, E., AND M. SEPPALA, a-Fetoprotein in normal human serum. Nature 235: 161-162, 1972. 24. SELL, S. Concluding remarks on physiology, metabolism and immune effects of onto-developmental gene products. In: OncoDevelopmental Gene Expression, edited by W. H. Fishman and S. Sell. New York: Academic, 1976, pm 373-374. 25. SELL, S., AND D. GORD. Rat a-fetoprotein. III, Refinement of radioimmunoassay for detection of 1 ng rat cw,F. Immunochemistry 10: 439-442, 1973. 26. SEPPALA, M., AND E. RUOSLAHTI. Alpha-fetoprotein: physiology and pathology during pregnancy and application to antenatal diagnosis. J. Perinat. Med. 1: 104-113, 1973. 27. STEPHENSON, S. K. Wool follicular development in the New Zealand Romney and N-type sheep. IV. Prenatal growth and changes in body proportions. Australian J. Agric. Res. 10: 433452, 1959. 28. WILLES, R. F., G. R. VAN PETTEN, AND J, F. TRUELOVE. Chronic exteriorization of vascular cannulas and ECG electrodes from the ovine fetus. J. Appl. Physiol. 28: 248-250, 1970.
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (132.174.255.223) on January 12, 2019.
distribution
of ovine
P. C. W. LAI, G. J. MEARS, G. R. VAN PETTEN, D. M. HAY, AND F. L. LORSCHEIDER Divisions of Medical Physiology, Pharmacology and Therapeutics, and Obestetrics Gynaecology, Faculty of Medicine, University of Calgary and Alberta Children’s Hospitul Research Centre, Calgary, Alberta, Canada T2N lN4
LAI, P. C. W., G. J. MEARS, G. It. VAN PETTEN, Il. M. HAY, AND F. L. LORSCHEIDER. Fetal-maternal distribution of ovine alpha-fetoprotein. Am. J. Physiol. 235(l): E27-E31, 1978 or Am J. Physiol.: Endocrinol. Metab. Gastrointest, Physiol. 4(l): E27-E31, 1978. -The concentration of ovine alpha-fetoprotein (AFP) was determined in sheep fetal, maternal, and neonatal sera and amniotic fluid by radioimmunoassay. The fetal serum AFP concentration was highest during the first third of pregnancy and continued to decline with fetal and neonatal development. Total fetal synthesis of AFP was highest during the late middle and early latter third of pregnancy. Amniotic fluid AFP concentration was highest during the early middle third of pregnancy. Maternal serum AFP was not elevated above nonpregnant control levels during the first two-thirds of pregnancy, but instead showed a tendency to be elevated only during the last third of pregnancy. The fetal physiological distributions of ovine and human AFP in fetal serum and amniotic fluid appear to be similar, whereas in maternal serum the pattern of AFP levels differs in the two species as a function of gestational age. amniotic fluid; pregnancy; noassay; neonatal serum
fetal serum
proteins;
radioimmu-
(AFP) is a fetus-specific serum protein found in aquatic and terrestrial vkrtebrates (9, -11). This protein is-produced by the human yolk sac and fetal liver (10) and is secreted into the fetal circulation. The synthesis of AFP has been investigated in man (8), rabbits (4), and rats (13). In these species, the fetal synthesis of AFP is curtailed before birth, and only trace amounts of the protein are detectable in adult serum (23, 25). In man, elevated serum AFP levels in the adult may be used clinically as a marker for teratoma and hepatic diseases (1, 2). In addition, abnormally high levels of AFP in human amniotic fluid can be used for the antenatal diagnosis of birth defects (5, 17). Although the clinical diagnostic value of AFP is well established, the specific function of AFP is not known (24). Gitlin and Boesman (8) showed that the fetal synthesis of human AFP reaches maximum levels between 20 and 32 wk of gestation. Recently, highly sensitive radioimmunoassay techniques were employed to measure the concentration of AFP in maternal serum and amniotic fluid in man (18, 26) and rats (13). We reported ALPHA-FETOPROTEIN
0363-6100/78/0000-0000$01.25 Copyright
0
1978the
American
Physiological
and
a difference in the gestational pattern of AFP levels in maternal and fetal vascular compartments between man and rat (13). Thus the rat appeared to be of limited value as an experimental animal model for studying the function of AFP and its relationship to birth defects and abnormal pregnancies. The objective of this experiment was to use the recently developed radioimmunoassay for ovine AFP (15) to quantitate the concentration of AFP in various physiological fluids and to investigate the basic physiology of AFP in the pregnant sheep and fetus. MATERIALS
AND METHODS
Animals, sera, and amniotic fluid. All sheep used in this study were maintained and bred at the Animal Resources Centre at the University of Calgary prior to their transfer to the Medical Vivarium for surgery and experimentation. The day of mating was observed and considered to be day 0 of gestation. Maternal sera were collected from four ewes via the jugular vein at weekly intervals beginning at day 25 of gestation until term (approximate1 y 143 days). Pregnancy was confirmed between 90 and 100 days by fluoroscopy. Under halothane general anesthesia, the femoral artery, saphenous vein, and amniotic sac of the conceptus were chronically cannulated at 110-115 days gestation with silicone rubber catheters, which were exteriorized by surgical procedures previously described (28). In addition, samples were obtained from several other similarly cannulated fetuses from 110 days to term. In the case of twin pregnancies, only one fetus tias cannulated. Weekly samples of 4 ml amniotic fluid and 2.5 ml fetal blood were withdrawn via the catheters. Blood samples were also obtained from two newborn lambs via the jugular vein at 0.5, 5, 9, and 15 days postpartum. Amniotic fluid and fetal sera samples on or before 83 days gestation were obtained either by cesarean section within 5 min after intravenous infusion of sodium pentobarbital (25 mg/kg body wt) to anesthetize the ewe or by laparotomy after halothane anesthesia. Control serum samples were obtained from six normal nonpregnant ewes via the jugular vein. Blood samples and amniotic fluid specimens were centrifuged at 1,500 x g for 15 min to remove cells, and the supernatants were stored at -20°C until assayed. Amniotic fluid specimens Society
E27
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (132.174.255.223) on January 12, 2019.
LA1 ET AL.
E28 contaminated with fetal red blood cells were discarded. The body weight of sheep fetuses was estimated using the standard ovine fetal growth curve established by Stephenson (27) with the exception that actual weights of 36-day-old fetuses and newborn lambs were obtained during our experiments. Radioimmunoassay of ovine AFP. Determination of AFP concentration was carried out using a doubleantibody radioimmunoassay procedure as previously described (15). The assay has a limit of detection of 2 nglml and employs an ovine AFP standard, which was prepared and calibrated as previously described (X,16>. Detection of AFP by electrophoresis. The presence of AFP in fetal sera was also confirmed by analytical disc polyacrylamide gel electrophoresis with a running pH 9.3, 7.5% polyacrylamide gel, and buffers as described (13). A 4.5~1 serum sample was applied to each gel column. After electrophoresis, each gel was stained with 0.75% aniline blue black dissolved in 7% acetic acid for 1 h and destained by washing with three 200-ml aliquots of 7% acetic acid. Estimation of total protein concentration. Concentrations of total protein in sera were determined by the method of Lowry et al. (21) with bovine serum albumin (Calbiochem., La Jolla, Calif.) used as the protein standard for the assay.
6060.
TOTO-
99 c gg
6060-
z 00pp
6060-
E E 22 22 ww $$
.
40403030-
0 20
40
60
DAYS
60
GESTATION
100
120
140 t
Parturition
FIG. 2. Changes in fetal serum protein concentration tion of gestation. Curve was fitted by inspection.
as a func-
RESULTS
Figure 1 illustrates the changes in the concentration of AFP in fetal serum as determined by radioimmunoassay at various stages of gestation. The AFP concentration was highest during the first third of pregnancy (4.5 mglml at 36 days gestation) and continued to decline to less than 100 pg/ml 1 wk before term. Figure 2 illustrates the concentration of total fetal serum protein as a function of gestation. A rapid increase in concentration of serum protein was evident a few weeks before term. Figure 3 shows the analytical disc polyacrylamide gel electrophoresis of proteins in fetal sera at different stages of gestation and compared to that of maternal and adult ram sera. AFP in fetal serum can be visualized as a protein fraction with slower electrophoretic
DAYS GESTATION FIG. 1. Fetal serum AFP concentration Curve was fitted by inspection.
Partthon
as a function of gestation.
FIG. 3. Analytical disc polyacrylamide gel electrophoresis of sheep serum proteins. a, go-day fetal serum; b, 83-day fetal serum; c, l&day fetal serum; d, maternal serum from latter third of pregTf, nancy; e, ram serum. Alb, albumin; AFP, alpha-fetoprotein; transferrin. Electrophoretic mobility of AFP relative to other serum proteins was previously identified (14).
mobility than albumin, which decreases in concentration with gestational age and cannot be detected in maternal or ram sera with this technique. Figure 4 is a graph of fetal serum AFP concentration x fetal weight as a function of gestation. Peak values were attained between 80 and 115 days gestation with decreases thereafter during fetal and neonatal development. Figure 5 illustrates the postpartum catabolic clearance of AFP in a case of twin lambs. The mean half-life for the disannearance of AFP from newborn lamb serum until 15 days postpartum was determined by plotting
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (132.174.255.223) on January 12, 2019.
OVINE
ALPHA-FETOPROTEIN
IN
PREGNANCY
E29
logarithmically transformed AFP values versus time, and the slope was calculated by linear regression. The mean half-life in these two animals during the first 15 days postpartum was approximately 86 h. Figure 6 illustrates the changes in concentration of AFP in amniotic fluid as a function of gestational age. Peak values of approximately 22 pg/ml were reached early in the middle third of pregnancy. Figure 7 illustrates the concentration of AFP in maternal serum at various stages of gestation compared against the mean basal level of AFP t 1 SE (29.9 t 5.8 rig/ml) in six control nonpregnant adult ewes. Maternal serum AFP was not statistically different from control
DAYS FIG.
function
s
6. Changes of gestation
Parturition
GESTATION
in concentration of AFP in amniotic Curve was fitted by inspection.
fluid
as a
4m-
3
2 #
200-
a z
0'
I 20
40
60 DAYS
FIG.
Curve
4. Fetal serum AFP was fitted by inspection.
x
fetal
80
100
120
1 4ot Parturition
GESTATION
weight
as a function
of gestation.
0'
1 20
1 40
I 60 PAYS
1 80 GESTATION
A
1
100
120
1 140 t Plrturitiofl
FIG, 7. Maternal serum AFP concentration as a function of gestation. Each point represents mean serum AFP value for ewes within a lo-day interval. Vertical bars represent standard error. Shaded area represents mean + 1 SE of serum AFP values for 6 control nonpregnant ewes, (*, one-tailed Student t test for comparison of experimental and control values significant at P < 0.025.)
serum values during the first two-thirds of pregnancy. The level of maternal serum AFP increased significantly early in the last third of pregnancy and peaked a few weeks before birth. ,
DISCUSSION
100
POSTPARTUM
FIG.
newborn
5. Decline lambs.
in
serum
300
200;
AFP
400
{hours)
concentration
in
a case
of twin
The concentration of fetal serum AFP is highest during the first third of pregnancy, declines during fetal development, and continues to disappear from neonatal serum. Disc polyacrylamide gel electrophoresis of fetal sera at various stages of gestation not only supports the
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (132.174.255.223) on January 12, 2019.
E30 radioimmunoassay data, but also clearly indicates that AFP is a major serum alpha-globulin in early gestation. Other investigators have demonstrated that ovine AFP is chemically-different from fetuin, a protein not only abundant in fetal serum but also detectable in adult serum by electrophoresis (19, 20). Our data in sheep demonstrate that there are species differences with respect to AFP concentration in fetal serum as a function of gestation. As in man 09, serum AFP is highest duing the first third of pregnancy in fetal sheep, but in rabbits and rats AFP levels peak near term (4, 13). Total serum protein in ovine fetal serum increases with advancing gestation. This increase in total protein concentration, in spite of a reciprocal decrease in AFP concentration with fetal development, can be primarily attributed to increases in serum albumin and transferrin which is corroborated in our disc gel electrophoresis data. Our electrophoresis data also agree with those previously reported by others (3). It is the re f ore advantageous to use fetal sera obtained from early gestation as the source for the isolation and purification of AFP because the ratio of the concentration of AFP to total protein is highest at this time, The concentration of any protein in plasma is not only a function of its rate of synthesis and degradation, but also a function of its compartmental distribution and size of the body fluid spaces. Proteins with molecular weights similar to that of AFP are normally not restricted to the vascular compartment, but are distributed almost equally between interstitial fluids and plasma (6). Because the fetus is growing rapidly, the concentration of AFP in fetal plasma cannot give a meaningful indication of the level of fetal hepatic synthesis of this protein at any stage of fetal development. Gitlin (7, 8) suggested that if the plasma concentration of AFP at a given stage of development is multiplied by the fetal weight at that time, an approximation of the relative amounts of AFP synthesized by the fetus at different points in gestation may be obtained. Applying this same calculation to our data, we conclude that the amount of ovine AFP synthesized is highest between 80 and 115 days gestation and that synthesis is gradually curtailed just a few weeks before birth. At present, the detailed mechanism responsible for the cessation of fetal synthesis of AFP before birth remains to be elucidated. It is interesting to note that the peak synthesis of both human (8) and ovine AFP occurs in the same stage of fetal development, i.e., the late middle and early latter third of pregnancy. AFP can be detected in the amniotic fluid of the sheep conceptus. The source of AFP in amniotic fluid is thought to be fetal rather than maternal because the concentration of AFP in fetal serum is approximately ZOO-300 times higher than that in amniotic fluid
LA1
ET
AL.
throughout gestation. In addition, there is a corresponding decrease in amniotic fluid AFP concentration probably as a result of a decline in fetal serum AFP concentration with advancing gestation. There is a difference in the pattern of appearance of AFP in amniotic fluid among mammalian species. Our data from sheep is similar to human data (22) in that the peak level of AFP is reached in early gestation. In contrast, the concentration of rat amniotic fluid AFP peaks late in gestation (13). Furthermore, the concentration of AFP in rat amniotic fluid is very high, on the order of several hundred micrograms per milliliter (13), but in both man (22) and sheep the concentration is less than 50 pg/ml. AFP can also be detected in the sera of adult nonpregnant ewes (15). In the present study, we showed that maternal serum AFP concentration is not significantly elevated from the basal level during the first two-thirds of pregnancy. Only a moderate elevation in maternal serum AFP of approximately 2 times the basal level can be seen early during the latter third of pregnancy. This is in contrast to the more significant elevation observed in human serum AFP during pregnancy (12). The low level of AFP in ovine maternal serum, approximately O.OOl-0.1% that of fetal serum, can be attributed either to a minimum degree of transplacental and transamniotic transfer of AFP from the conceptus or to a high rate of catabolic clearance of AFP in the ewe, The answer awaits future kinetic and metabolic studies. The sheep should be a suitable animal model for studying AFP because the size of the fetus approximates that of man, and the fetal blood vessels and amniotic cavity can be cannulated to permit serial and simultaneous sampling of fetomaternal compartments. In addition, our data suggest that ovine fetal physiology of AFP is similar to that of man, making the sheep a potential model for the study of the function and pathophysiology of AFP. However, the applicability of the sheep model for studying AFP in complicated pregnancies such as fetal neural tube defects may be limited by our finding that sheep maternal serum AFP levels differ in some respects from man, The technical assistance of Mr. G. Paquette, D. Teichroeb, P, van Muyden, Ms. Y. Brideau, G. Cochrane, P. Hamilton, and H. Mathison is gratefully acknowledged. This study was supported by the Medical Research Council of Canada Grant MA-5292 and a grant from The Alberta Children’s Hospital Research Foundation. P. C. W. Lai is a recipient of a Medical Research Council Studentship. Send requests for reprints to: F. L. Lorscheider, Faculty of Medicine, Div. of Medical Physiology, Univ, of Calgary, 2920 24th Ave., N. W., Calgary, Canada T2N lN4.
Received
31 October
1977; accepted
in final
form
7 March
1978.
REFERENCES 1. ADINOLFI, A., M. ADINOLFI, AND M. H. LESSOF. Alpha-fetoprotein during development and in disease. J. Med. Genetics 12: 138151, 1975. 2. ALPERT, E. Human alphal-fetoprotein (AFP): developmental biology and clinical significance. Progr. Liver Diseases 5: 337349, 1976.
3. BARBORIAK,
J. J., G. MESCHIA, D. H. BARRON, AND Blood plasma proteins in fetal goats and sheep. Sot. Exptl, Biol. Med. 98: 635-637, 1958. 4, BRANCH, W. R. The ontogeny of alpha-foetoprotein in the and neonatal rabbit, and its experimental induction in rabbits. Intern. J. Cancer 10: 451-457, 1972. COWGILL.
R. Proc.
G.
foetal adult
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (132.174.255.223) on January 12, 2019.
OVINE
ALPHA-FETOPROTEIN
IN
PREGNANCY
5. BROCK, D. J. H. Alpha-fetoprotein and the prenatal diagnosis of central nervous system disorders. A review. Cltild’s Brain 2: l23, 1976. 6. GITLIN, D. Distribution dynamics of circulating and extravascular Pa1 plasma proteins. Ann. N. Y. Acad. Sci. 70: 122-136, 1957. 7. GITLIN, D., Normal biology of a-fetoprotein. Ann. N. Y. Acad. Sci. 259: 7-16, 1975. 8. GITLIN, D., AND M. BOESMAN. Serum a-f&protein, albumin, and y-G-globulin in the human conceptus. J. Clin. Invest. 45: 1826-1838,1966. 9. GITLIN, D., AND M. BOESMAN. Fetus-specific serum proteins in several mammals and their relation to human a-fetoprotein. Comp. Biochem. Physiol. 21: 327-336, 1967, 10. GITLIN, D., AND M. BOESMAN. Sites of serum a-fetoprotein synthesis in the human and in the rat. J. CZin. Invest. 46: lOlO1016, 1967. 11. GITLIN, D., A. PERRICELLI, AND J. D. GITLIN. The presence of serum a-fetoprotein in sharks and its synthesis by fetal gastrointestinal tract and liver. Camp. Biochem. PhysioZ. 46B: 207-215, 1973. 12. HAY, D. M., P. I. FORRESTER, R. L. HANCOCK, AND F. L. LORSCHEIDER. Maternal serum alpha-feteprotein in normal pregnancy. Brit. J. Obstet. Gynaecol. 83: 534-538, 1976. 13. LAI, P. C. W., P. I. FORRESTER, R. L. HANCOCK, D. M. HAY, AND F. L. L~R~CHEIDER. Rat alpha-fetoprotein: isolation, radioimmunoassay and fetal-maternal distribution during pregnancy. J. Reprod. Fertility 48: 1-8, 1976. 14. LAI, P. C, W., D. M. HAY, AND F. L, LORSCHEIDER. Purification of ovine alpha-fetoprotein by preparative electrophoresis. Stand. J. Immunol., Suppl. In press. 15. LAI, P. C. W., D. M. HAY, AND F. L. LORSCHEIDER, Radioimmunoassay of ovine alpha-fetoprotein. J. Immunol. Methods. 20: l-10, 1978. 16. LAI, P. C. W., D. M. HAY, E. H. PETERS, AND F. L. LORSCHEIDER. Immunochemical purification and characterization of ovine afetoprotein. Biochim. Biophys. Acta 493: 201-209, 1977. 17. LAU, H. L., AND S. E. LINKINS. Alpha-fetoprotein. Am. J,
E31 Obstet. Gynecol. 124: 533-554, 1976. 18. LEEK, A. E., C. F. Ruoss, M. J. KITAW, AND T. CHARD. Maternal plasma alpha-fetoprotein levels in the second half of normal pregnancy: relationship to fetal weight and maternal age and parity. Brit. J. Obstet. Gynaecol. 82: 669-673, 1975. 19. MARTI, J., S. ALIAW, C. BONFIL~, C. VIGNE, AND J. MORETTI. Preparation, constantes physiques et constitution chimique de la fetuine d’agneau. Biochim. Biophys. Acta 303: 348-359, 1973. 20. MARTI, J., S. ALIAW, AND J. MORETTI. Studies on AFP from ovine and bovine fetuses. In: Protides of Biological Fluids, edited by H. Peeters. Oxford: Pergamon, 1976, p* 259-262. 21. LOWRY, 0. H., N. J. ROSEBROUGH, A. L. FARR, AND R. J. RANDALL. Protein measurement with the folin phenol reagent. J. BioL. Chem. 193: 265-275, 1951. 22. RANDLE, G. H., AND K. N. CUMBERBATCH. Alpha-fetoprotein levels in amniotic fluid in normal pregnancy and in pregnancy complicated by anencephaly. J. Obstet. Gynaecol. Brit. Commonwealth 80: 1054-1058, 1973. 23. RUOSLAHTI, E., AND M. SEPPALA, a-Fetoprotein in normal human serum. Nature 235: 161-162, 1972. 24. SELL, S. Concluding remarks on physiology, metabolism and immune effects of onto-developmental gene products. In: OncoDevelopmental Gene Expression, edited by W. H. Fishman and S. Sell. New York: Academic, 1976, pm 373-374. 25. SELL, S., AND D. GORD. Rat a-fetoprotein. III, Refinement of radioimmunoassay for detection of 1 ng rat cw,F. Immunochemistry 10: 439-442, 1973. 26. SEPPALA, M., AND E. RUOSLAHTI. Alpha-fetoprotein: physiology and pathology during pregnancy and application to antenatal diagnosis. J. Perinat. Med. 1: 104-113, 1973. 27. STEPHENSON, S. K. Wool follicular development in the New Zealand Romney and N-type sheep. IV. Prenatal growth and changes in body proportions. Australian J. Agric. Res. 10: 433452, 1959. 28. WILLES, R. F., G. R. VAN PETTEN, AND J, F. TRUELOVE. Chronic exteriorization of vascular cannulas and ECG electrodes from the ovine fetus. J. Appl. Physiol. 28: 248-250, 1970.
Downloaded from www.physiology.org/journal/ajpendo by ${individualUser.givenNames} ${individualUser.surname} (132.174.255.223) on January 12, 2019.
