Journal of Reproductive Immunology, 21 (1992) 47-56
47
Elsevier Scientific Publishers Ireland Ltd.
JRI 00744
Pregnancy-associated major basic protein in amniotic fluid K a r e n K. V e r n o f a, Steven J. O r y a a n d G e r a l d J. G l e i c h b'c aDepartment of Obstetrics and Gynecology. hDepartment ~[ Immunology and "Tile Division ~[" AIh'rgic Diseases and Internal Medicine, Mayo Clinic and Mayo Foundation, and the Mayo Medical School. Rochester, MN 55905 (U.S.A,) (Accepted for publication 14 August. 19911
Summary The pregnancy-associated major basic protein, a protein elevated in the sera of all pregnant women, is virtually identical to the eosinophil granule major basic protein. To determine whether pregnancy-associated major basic protein is present in amniotic fluid, we examined samples from both early and late gestation by a double antibody radioimmunoassay. A total of 112 amniotic fluids were tested and all but three contained levels of pregnancy-associated major basic protein greater than 400 ng/ml. Amniotic fluid pregnancy-associated major basic protein antigenic activity was immunochemically identical to that of the eosinophil granule major basic protein and also had identical physicochemical properties such as heat stability and the need for reduction and alkylation. Although the majority of amniotic fluid samples (90 of 112) were obtained from healthy women with normal gestations, the remaining 21 amniotic fluid samples were from women with Rh sensitization and from one gestation complicated by intrauterine growth retardation.
Key words: major basic protein," amniotic fluid
Correspondence to." Gerald J. Gleich, M.D., Department of Immunology, Mayo Clinic, Rochester. MN 55905, U.S.A, 0165-0378/91/$03,50 © 1991 Elsevier Scientific Publishers Ireland Ltd. Published and Printed in Ireland
48
Introduction
The pregnancy-associated major basic protein (pMBP) is an unusual and novel protein recently described in the sera of pregnant women (Maddox et al., 1983). pMBP was so named because it is immunochemically and physicochemically identical to the eosinophil granule major basic protein (gMBP) (Maddox et al., 1983; Wasmoen et al., 1989) whose biologic functions include cytotoxicity to parasites and mammalian cells, and activation of basophils and mast cells (Gleich and Adolphson, 1986). gMBP likely functions as a mediator of tissue damage in hypersensitivity diseases (Frigas and Gleich, 1986; Leiferman et al., 1986; Gleich, 1990). Elevated plasma levels of pMBP were identified in pregnant women in the absence of peripheral blood eosinophilia (Maddox et al., 1983) and pMBP has been localized by immunofluorescence to the placental X-cells, placental site giant cells and septal cysts (Maddox et al., 1984). Serum levels of pMBP rise early in pregnancy, reach a plateau about 20 weeks, and then show a late rise which can be related to the onset of spontaneous labor (Wasmoen et al., 1987~ Coulam et al., 1987). Although pMBP is present in maternal serum and changes of pMBP during gestation have been described, pMBP has not been characterized in amniotic fluid. This study was performed to identify and describe the levels of amniotic fluid pMBP throughout gestation in comparison to serum pMBP. Materials and Methods
Clinical specimens Amniotic fluid was collected from patients either at the time of a medically indicated amniocentesis, or after rupture of membranes at the time of delivery. Amniotic fluid samples with visible blood contamination were discarded. A total of 112 amniotic fluid and 77 paired serum samples were obtained from 96 women. Both serum and amniotic fluid samples were centrifuged and stored at -20°C until assayed. Of the 112 total amniotic fluid samples, 49 were obtained at the time of delivery and all had paired serum samples. Sixty-three amniotic fluid samples were obtained at the time of amniocentesis of which 28 had paired serum samples. The indications for amniocentesis were: (1) 39.7% genetic analysis, (2) 33.3% fetal-maternal blood group incompatibilities, (3) 32% fetal lung maturity, and (4) 1.3% alpha-fetoprotein determination. Five patients had more than one indication for amniocentesis; hence the total is greater than 100%. Two women had two amniotic fluid/serum pairs, one each at mid gestation and at term. Four women with Rh sensitization had multiple amniotic fluid samples because of repeat amniocentesis for AOD~50 determinations.
,19
Reduction and alkylation of test samples MBP antigenic activity cannot be fully measured in serum without reduction and alkylation of the sample of pMBP before immunoassay (Wassom et al., 1981; Maddox et al., 1983). Therefore, unless noted otherwise, amniotic fluid and serum samples were reduced and alkylated as previously described (Wassom et al., 1981; Maddox et al., 1983). Briefly, after 50 tzl of sample was diluted with 130/zl 0.1 M Tris buffer, 0.075 M dithiothreitol solution was added to a final concentration of 0.0075 M. After a 60-min incubation at room temperature, 0.15 M iodoacetamide was added to a final concentration of 0.015 M. The samples were then assayed after a 20-min incubation at room temperature.
Double antibody radioimmunoassay for MBP ( DARIA ) The double antibody competition radioimmunoassay used for determination of MBP was performed as previously described (Wassom et al., 1981). Briefly, both serum and amniotic fluid were reduced and alkylated and then diluted in PPF-E (0.086 M Na2PO4, 0.014 M KH2PO4, pH 7.4; 0.5% fetal calf serum, 0.1% NAN3, 0.01 M EDTA, and 1% protamine sulfate at pH 7.4). Five hundred microliters of PPF-E, 100 tA rabbit antihuman MBP (1 : 10,000 in PPF-E) and 50/zl of sample were mixed and incubated at 37°C for 30 min. gMBP was prepared from human eosinophils as previously described (Ackerman et al., 1985; Gleich et al., 1986) and was radioiodinated by the chloramine-T method (McConahey and Dixon, 1966). One hundred microliters of 125I-gMBP (10 ng/ml in PPF-E) was added and incubated overnight at 4°C. Dilutions of antiserum were tested with ~25I-gMBP to yield a 50% binding of ~25I-gMBP. Fifty to 100/~1 of normal rabbit serum (1:20 in PPF-E) and 100-250 Izl of burro antirabbit IgG were added to precipitate the immune complexes. Incubation time was 3 h at room temperature, after which the tubes were centrifuged in a Sorvall H2B at 2750 rev./min (1800 × g) for 30 min. The supernatants were discarded and the precipitates counted. Maximum binding, non-specific binding, and total count tubes were determined in each assay. Sera from a normal male and from a patient with hypereosinophilic syndrome were utilized as internal controls. All samples and controls were analyzed in duplicate. All serum/amniotic fluid pairs were analyzed in the same assay. Data were analyzed using a radioimmunoassay program (Hewlett Packard 9845-14254). Statistical programs by Hewlett Packard (9845-15030, 9845-15010 and 9845-15020) and SAS (version FF, release 5.16) were also employed. Results
Specificity of the immunochemical detection of MBP pMBP activity was detected in all amniotic fluid samples tested. To deter-
50
mine whether pMBP in amniotic fluid was immunochemically similar to purified gMBP, we titrated three samples in the DARIA. Figure 1 shows that the slopes of the regression lines for the amniotic fluid pMBP and eosinophil granule MBP were strikingly similar, though not identical, and that amniotic fluid completely inhibits the binding of ~25I-gMBP to antibody. Table 1 shows that amniotic fluid pMBP activity required reduction and alkylation to demonstrate maximal MBP immunoreactivity, as is seen with gMBP detection in sera of patients with the hypereosinophilic syndrome (HES) (Wassom et al., 1981) and with pMBP in the sera of pregnant women (Maddox et al., 1983). Table 1 also shows that the heat lability of pMBP in amniotic fluid depended upon antecedent reduction and alkylation, as is required for sera of patients with HES (Wassom et al., 1981) and for pMBP in the sera of pregnant women (Maddox et al., 1983). Heating of amniotic fluid without prior reduction and alkylation did not alter activity, in keeping with prior findings (Wassom et al., 1981; Maddox et al., 1983).
Changes of amniotic fluid pMBP levels with advancing gestation Comparison of samples from early (~ = 16.8 weeks) and late (~ = 38.9 weeks) gestation demonstrated a rise in p M B P levels in both maternal serum and amniotic fluid (Table 2). In the vast majority of cases, we obtained only one amniotic fluid sample from each woman, either early or late gestation as outlined in Materials and Methods. However, there were two women who had both early and late amniocenteses for genetic screening and fetal lung maturity, respectively (Table 3). In both, amniotic fluid pMBP rose during
100
o~
!}noc.u0s
8O
0
h. Standard MBP
-
•
0
~
m
~E
-2 -3
I,-
q
20:
I 2
I 4
I 6
Dose, ng or/JJ
I I~
10
-1.2
-0.8
-04
0
04
08
12
Log dose, ng or/.LI
Fig. 1. Comparison of dose response curves for amniotic fluids and purified MBP in the DARIA. (A) Untransformed data, (B) the same data after logit-log transformation. Linear regression analysis showed r 2 = 0.988 for purified MBP and a mean r 2 of 0.984 (range 0.978-0.999) for the amniotic fluid. One-way analysis of covariance indicated that the slopes for titrations of amniotic fluid differ from that of pure MBP (F5,81 =3.09, P < 0.05).
51
TABLE
1
Effect o f r e d u c t i o n a n d a l k y l a t i o n , a n d h e a t i n g o n M B P c o n c e n t r a t i o n s in a m n i o t i c fluid a n d sera. All s a m p l e s w e r e a n a l y z e d b y the D A R I A . S a m p l e s t h a t were h e a t e d were p l a c e d in a 56°C w a t e r b a t h f o r 2 h. P e r c e n t a g e s s h o w n i n d i c a t e the a m o u n t r e c o v e r e d c o m p a r e d to the total as m e a s u r e d w h e n the s a m p l e is r e d u c e d a n d a l k y l a t e d ( R & A ) . Specimen
Treatment
M B P ng/ml
A m n i o t i c fluid ( 173/7 weeks)
R&A No R&A
3362 675 (20.1'7,,)
R&A then heat Heat then R&A
223 (6.6%) 3835 ( > 100%)
R&A
6894
No R&A R&A then heat Heat then R&A
1,234 (17.9%) 302 (4.4'7,,) 6716 (97.4%)
Hypereosinophilic syndrome, serum
R&A No R&A R&A then heat Heat then R&A
19,951 2351 (11.8%) 431 (2.2%) 13,053 (65%)
Normal male serum
R&A No R&A R & A t h e n heat Heat then R&A
328 122 (37.2%) 138 (42%) 326 (99.4%)
A m n i o t i c fluid (38 weeks)
TABLE 2 p M B P (ng/ml) levels in a m n i o t l c fluid a n d s e r u m in e a r l y a n d late g e s t a t i o n .
E a r l y g e s t a t i o n , n = 28 = 16.8 w e e k s
Late gestation, n =64 = 38.9 w e e k s
Blood group isoimmunizat i o n , n = 20 ,~- = 32.4 w e e k s
A m n i o t i c fluid
Serum
n = 28 ~ = 1438
n = 16 ~ = 3596
M e d i a n = 971 Range = 160-4802
M e d i a n = 3596 Range 1568-6719
n =64
n = 52
~ = 2567 M e d i a n = 1851 Range = 173-8700
~ = 4630 M e d i a n = 4782 Range = 1439-9386
n = 20 = 5649 M e d i a n = 4852
n--9 .7" = 8559 M e d i a n = 4980 Range = 3321-16,186
Range = 1528-11,791
52
TABLE 3 pMBP levels in amniotic fluid obtained at mid-gestation and at term in two patients.
Patient No. l Patient No. 2
Gestational age (days)
Amniotic fluid pMBP (ng/ml)
116 267 117 267
499 470t 2929 3135
gestation, although in patient 1 the rise was almost 10-fold in contrast to only a modest change in patient 2. Both had normal karyotypes and normal pregnancies. The samples obtained from four women with blood group isoimmunization presented an opportunity to evaluate serial amniotic fluid samples from a single patient (Fig. 2). The first samples obtained from these women were at mid-gestation (;~ = 28.6 weeks). There were no samples from the normal pregnancy group at mid-gestation for comparison. Interestingly, in all four women pMBP levels were exceedingly high at mid-gestation and actually decreased with increasing gestational age. The high levels seen at midgestation raise the overall mean level of this group. Actually, the last samples (2 = 36.4 weeks) had pMBP levels comparable to those seen in the late gestation group (Fig. 2, Table 2).
Relationship of serum and amniotic jluid pMBP The above results established that pMBP immunoreactivity is present in the amniotic fluid. Figure 3 and Table 2 show pMBP levels in paired sera and
10,950
O) C
8,950
~"
6,950
._(2 4,950
-_o c E
2,951:) 951) 175
I 200
I 225
250
275
Gestational age, days Fig. 2, Serial amniotic fluid samples from four women with blood group isoimmunization, pMBP levels fell from very high levels at mid gestation to levels at term that are within the range seen in late gestation (See Table 2).
I
Serum
I
Amniotic fluid
100
1,500
~.
2,2s0
250
2,900
"~ J~
4,300
E "~.
4,250
6,250
5.700
E ~)
7,100
Amnioficfluid
B
Serum
I
~
~
"E
0.
1,200
3.2oo
5,200
9.200
O~ C 13,200 E '~0 11,200
15,200
17,200
I
J Amniotic fluid
C
I
Serum
Fig. 3. Amniotic fluid/serum pairs. With few exceptions, serum pMBP levels are higher than paired amniotic fluid. This is shown in each of the three groups: (A) late gestation, (B) early gestation, and (C) blood group isoimmunization.
E
"(~ J::l
¢1 O
"~
8,250
10,250
54
amniotic fluid. In 68 of the 77 (88"/,,) paired samples, the serum levels were higher than the amniotic fluid levels (2-4 times higher). Although protein levels rise in both fluids with advancing gestational age and there is a trend for higher amniotic fluid levels to be associated with higher serum levels, neither early (r = +0.33, NS) or late (r = +0.1, NS) amniotic fluid/serum pairs are statistically correlated. Discussion
A protein has been identified in the sera and amniotic fluid of all pregnant women that is immunochemically similar to the core granule protein of the eosinophil, the major basic protein. The role of this toxic protein in normal reproductive physiology remains an enigma. Here, pMBP levels were determined in maternal serum and amniotic fluid in a large number of normal pregnancies. Forty-nine percent of all samples were obtained at the time of ruptured membranes. Twenty-two percent of all samples were obtained from genetic amniocenteses for increased maternal age. There were no abnormal karyotypes. There was one group comprised from pregnancies complicated by maternal blood group isoimmunization (19% of all samples), and for purposes of analysis, these were kept in a separate group. pMBP was detected in all amniotic fluid samples and had immunochemical properties similar to the granule protein of the eosinophil. Prior studies have established that there is no maternal eosinophilia to account for the increased levels of the protein (Maddox et al., 1983). pMBP has been localized to the placenta, the likely source for both maternal serum and amniotic fluid pMBP. In particular, immunofluorescent techniques have localized pMBP to the X-cell, a non-villous trophoblast found predominantly at the maternal-fetal junction of the basal plate, and in the placental septa and lining the septal cysts (Maddox et al., 1984). As with other placental proteins, pMBP levels are higher in maternal serum than in either fetal serum or amniotic fluid (Klopper, 1980). The non-villous trophoblast is in closer proximity with the maternal intervillous space than with the fetal capillaries which are surrounded by the four-layered placental membrane. Umbilical cord serum has pMBP levels equivalent to normal adult male levels (Maddox et al., 1983). The source o f p M B P in amniotic fluid could either be indirectly through the fetus, or more likely, directly by diffusion through the semipermeable amnion-chorion membrane. Interestingly, the amniotic fluid obtained from pregnancies complicated by maternal blood group isoimmunization showed higher mean levels of pMBP compared to the mean levels from normal pregnancies (P < 0.001, two-sided t-test). Hemolytic disease is unlikely to affect eosinophils and is, therefore, an unlikely source for the higher values. The higher mean pMBP levels in the
55
isoimmunization group may simply be the result of the timing of the amniotic fluid samples (2 = 32.4 weeks). A plausible hypothesis is that pMBP production plateaus at mid-gestation and the increasing amniotic fluid volume dilutes the level of the protein measured. This is a difficult hypothesis to test as serial samples of amniotic fluid from normal gestations cannot be obtained. The data from the group of women with serial amniocenteses for determination of hemolysis in blood group isoimmunization suggests that amniotic fluid levels o f p M B P initially rise to very high levels (Fig. 2) but then decrease with increasing gestational age. This would correspond with increasing amniotic fluid volume which consists mainly of fetal urine (Brace, 1989). Neonatal urine has undetectable amounts of pMBP (unpublished data), and thus the increasing volume of fetal urine could act to dilute the levels of pMBP. The protein level in late gestation is still significantly higher than that found in early gestation. One pregnancy in this study was complicated by intrauterine growth retardation (IUGR). Only amniotic fluid was obtained from this patient and demonstrated one of the highest amniotic fluid values measured in our laboratory (8700 ng/ml). Of note, this patient had normal amniotic fluid volume measured by ultrasound prior to the amniocentesis. It has previously been reported that massive proliferation of the X-cell occurs in the placentas from infants with some types of growth retardation (Ermocilla and Altshuler, 1973).
Acknowledgements Supported in part by the National Institutes of Health, HD 22924, AI 07047 and AI 09728, and the Mayo Foundation. We thank Terri L. Wasmoen, Ph.D. for assistance in the inception of this study and Mrs. Linda H. Arneson for secretarial help.
References Ackerman, S.J., Gleich, G.J., Loegering, D.A., Richardson, B.A. and Butterworth, A.E. (1985) Comparative toxicity of purified human eosinophil granule cationic proteins for schistosomula of Schistosoma mansoni. Am. J. Trop. Med. Hyg. 34, 735-745. Brace, R.A. (1989)Amniotic fluid dynamics. In: Maternal-Fetal Medicine: Principles and Practice (Creasy, R.K. and Resnik, R., eds.), W.B. Saunders Co., Philadelphia, p. 128. Coulam, C.B., Wasmoen, T., Creasy, R., Siiteri, P. and Gleich, G.J. (1987) Major basic protein as a predictor of preterm labor: A preliminary report. Am. J. Obstet. Gynecol. 156, 790-796. Ermocilla, R. and Altshuler, G. (1973) The origin of "X cells" of the human placenta and their possible relationship to intrauterine growth retardation: An enigma. Am. J. Obstet. Gynecol. 117, 1137-1140. Frigas, E. and Gleich, G.J. (1986) The eosinophil and the pathophysiology of asthma. J. Allergy Clin. Immunol. 77, 527-537. Gleich, G.J. and Adolphson, C.R. (1986) The eosinophilic leukocyte: Structure and function. Adv. lmmunol. 39, 177-253.
56 Gleich, G.J., Loegering, D.A., Mann, K.G. and Maldonado, i.E. (1986) Comparative properties of the Charcot-Leyden crystal protein and the major basic protein from human eosinophils. J. Clin. Invest. 57, 633-640. Gleich, G.J. (1990) The cosinophil and bronchial asthma: Current understanding. Review article. J. Allergy Clin. Immunol. 85, 422-436. Klopper, A. (1980) The new placental proteins. Placenta 1, 77-89. Leiferman, K.M., Peters, M.S. and Gleich, G.J. (1986) The eosinophil and cutaneous edema. J. Am. Acad. Dermatol. 15, 513-517. Maddox, D.E., Butterfield, J.H., Ackerman, S.J., Coulam, C.B. and Gleich, G.J. (1983) Elevated serum levels in human pregnancy of a molecule immunochemically similar to eosinophil granule major basic protein. J. Exp. Med. 158, 1211-1226. Maddox, D.E., Kephart, G.M., Coulam, C.B., Butterfield, J.H., Benirschke, K. and Gleich, G.J. (1984) Localization of a molecule immunochemically similar to eosinophil major basic protein in human placenta. J. Exp. Med. 160, 29-41. McConahey, P.J. and Dixon, F.J. (1966) A method of trace iodination of proteins for immunologic studies. Int. Arch. Allergy Appl. Immun. 29, 185-189. Wasmoen, T.L., Coulam, C.B., Leiferman, K.M. and Gleich, G.J. (1987) Increases of plasma eosinophil major basic protein levels late in pregnancy predict onset of labor. Proc. Natl. Acad. Sci. U.S.A. 84, 3029-3032. Wasmoen, T.L., McKean, D.J., Benirschke, K., Coulam, C.B. and Gleich, G.J. (1989) Evidence for eosinophil granule major basic protein in human placenta. J. Exp. Med. 170, 2051-2063. Wassom, D.L., Loegering, D.A., Solley, G.O. et al. (1981) Elevated serum levels of the eosinophil granule major basic protein in patients with eosinophilia. J. Clin. Invest. 67, 651-661.
47
Elsevier Scientific Publishers Ireland Ltd.
JRI 00744
Pregnancy-associated major basic protein in amniotic fluid K a r e n K. V e r n o f a, Steven J. O r y a a n d G e r a l d J. G l e i c h b'c aDepartment of Obstetrics and Gynecology. hDepartment ~[ Immunology and "Tile Division ~[" AIh'rgic Diseases and Internal Medicine, Mayo Clinic and Mayo Foundation, and the Mayo Medical School. Rochester, MN 55905 (U.S.A,) (Accepted for publication 14 August. 19911
Summary The pregnancy-associated major basic protein, a protein elevated in the sera of all pregnant women, is virtually identical to the eosinophil granule major basic protein. To determine whether pregnancy-associated major basic protein is present in amniotic fluid, we examined samples from both early and late gestation by a double antibody radioimmunoassay. A total of 112 amniotic fluids were tested and all but three contained levels of pregnancy-associated major basic protein greater than 400 ng/ml. Amniotic fluid pregnancy-associated major basic protein antigenic activity was immunochemically identical to that of the eosinophil granule major basic protein and also had identical physicochemical properties such as heat stability and the need for reduction and alkylation. Although the majority of amniotic fluid samples (90 of 112) were obtained from healthy women with normal gestations, the remaining 21 amniotic fluid samples were from women with Rh sensitization and from one gestation complicated by intrauterine growth retardation.
Key words: major basic protein," amniotic fluid
Correspondence to." Gerald J. Gleich, M.D., Department of Immunology, Mayo Clinic, Rochester. MN 55905, U.S.A, 0165-0378/91/$03,50 © 1991 Elsevier Scientific Publishers Ireland Ltd. Published and Printed in Ireland
48
Introduction
The pregnancy-associated major basic protein (pMBP) is an unusual and novel protein recently described in the sera of pregnant women (Maddox et al., 1983). pMBP was so named because it is immunochemically and physicochemically identical to the eosinophil granule major basic protein (gMBP) (Maddox et al., 1983; Wasmoen et al., 1989) whose biologic functions include cytotoxicity to parasites and mammalian cells, and activation of basophils and mast cells (Gleich and Adolphson, 1986). gMBP likely functions as a mediator of tissue damage in hypersensitivity diseases (Frigas and Gleich, 1986; Leiferman et al., 1986; Gleich, 1990). Elevated plasma levels of pMBP were identified in pregnant women in the absence of peripheral blood eosinophilia (Maddox et al., 1983) and pMBP has been localized by immunofluorescence to the placental X-cells, placental site giant cells and septal cysts (Maddox et al., 1984). Serum levels of pMBP rise early in pregnancy, reach a plateau about 20 weeks, and then show a late rise which can be related to the onset of spontaneous labor (Wasmoen et al., 1987~ Coulam et al., 1987). Although pMBP is present in maternal serum and changes of pMBP during gestation have been described, pMBP has not been characterized in amniotic fluid. This study was performed to identify and describe the levels of amniotic fluid pMBP throughout gestation in comparison to serum pMBP. Materials and Methods
Clinical specimens Amniotic fluid was collected from patients either at the time of a medically indicated amniocentesis, or after rupture of membranes at the time of delivery. Amniotic fluid samples with visible blood contamination were discarded. A total of 112 amniotic fluid and 77 paired serum samples were obtained from 96 women. Both serum and amniotic fluid samples were centrifuged and stored at -20°C until assayed. Of the 112 total amniotic fluid samples, 49 were obtained at the time of delivery and all had paired serum samples. Sixty-three amniotic fluid samples were obtained at the time of amniocentesis of which 28 had paired serum samples. The indications for amniocentesis were: (1) 39.7% genetic analysis, (2) 33.3% fetal-maternal blood group incompatibilities, (3) 32% fetal lung maturity, and (4) 1.3% alpha-fetoprotein determination. Five patients had more than one indication for amniocentesis; hence the total is greater than 100%. Two women had two amniotic fluid/serum pairs, one each at mid gestation and at term. Four women with Rh sensitization had multiple amniotic fluid samples because of repeat amniocentesis for AOD~50 determinations.
,19
Reduction and alkylation of test samples MBP antigenic activity cannot be fully measured in serum without reduction and alkylation of the sample of pMBP before immunoassay (Wassom et al., 1981; Maddox et al., 1983). Therefore, unless noted otherwise, amniotic fluid and serum samples were reduced and alkylated as previously described (Wassom et al., 1981; Maddox et al., 1983). Briefly, after 50 tzl of sample was diluted with 130/zl 0.1 M Tris buffer, 0.075 M dithiothreitol solution was added to a final concentration of 0.0075 M. After a 60-min incubation at room temperature, 0.15 M iodoacetamide was added to a final concentration of 0.015 M. The samples were then assayed after a 20-min incubation at room temperature.
Double antibody radioimmunoassay for MBP ( DARIA ) The double antibody competition radioimmunoassay used for determination of MBP was performed as previously described (Wassom et al., 1981). Briefly, both serum and amniotic fluid were reduced and alkylated and then diluted in PPF-E (0.086 M Na2PO4, 0.014 M KH2PO4, pH 7.4; 0.5% fetal calf serum, 0.1% NAN3, 0.01 M EDTA, and 1% protamine sulfate at pH 7.4). Five hundred microliters of PPF-E, 100 tA rabbit antihuman MBP (1 : 10,000 in PPF-E) and 50/zl of sample were mixed and incubated at 37°C for 30 min. gMBP was prepared from human eosinophils as previously described (Ackerman et al., 1985; Gleich et al., 1986) and was radioiodinated by the chloramine-T method (McConahey and Dixon, 1966). One hundred microliters of 125I-gMBP (10 ng/ml in PPF-E) was added and incubated overnight at 4°C. Dilutions of antiserum were tested with ~25I-gMBP to yield a 50% binding of ~25I-gMBP. Fifty to 100/~1 of normal rabbit serum (1:20 in PPF-E) and 100-250 Izl of burro antirabbit IgG were added to precipitate the immune complexes. Incubation time was 3 h at room temperature, after which the tubes were centrifuged in a Sorvall H2B at 2750 rev./min (1800 × g) for 30 min. The supernatants were discarded and the precipitates counted. Maximum binding, non-specific binding, and total count tubes were determined in each assay. Sera from a normal male and from a patient with hypereosinophilic syndrome were utilized as internal controls. All samples and controls were analyzed in duplicate. All serum/amniotic fluid pairs were analyzed in the same assay. Data were analyzed using a radioimmunoassay program (Hewlett Packard 9845-14254). Statistical programs by Hewlett Packard (9845-15030, 9845-15010 and 9845-15020) and SAS (version FF, release 5.16) were also employed. Results
Specificity of the immunochemical detection of MBP pMBP activity was detected in all amniotic fluid samples tested. To deter-
50
mine whether pMBP in amniotic fluid was immunochemically similar to purified gMBP, we titrated three samples in the DARIA. Figure 1 shows that the slopes of the regression lines for the amniotic fluid pMBP and eosinophil granule MBP were strikingly similar, though not identical, and that amniotic fluid completely inhibits the binding of ~25I-gMBP to antibody. Table 1 shows that amniotic fluid pMBP activity required reduction and alkylation to demonstrate maximal MBP immunoreactivity, as is seen with gMBP detection in sera of patients with the hypereosinophilic syndrome (HES) (Wassom et al., 1981) and with pMBP in the sera of pregnant women (Maddox et al., 1983). Table 1 also shows that the heat lability of pMBP in amniotic fluid depended upon antecedent reduction and alkylation, as is required for sera of patients with HES (Wassom et al., 1981) and for pMBP in the sera of pregnant women (Maddox et al., 1983). Heating of amniotic fluid without prior reduction and alkylation did not alter activity, in keeping with prior findings (Wassom et al., 1981; Maddox et al., 1983).
Changes of amniotic fluid pMBP levels with advancing gestation Comparison of samples from early (~ = 16.8 weeks) and late (~ = 38.9 weeks) gestation demonstrated a rise in p M B P levels in both maternal serum and amniotic fluid (Table 2). In the vast majority of cases, we obtained only one amniotic fluid sample from each woman, either early or late gestation as outlined in Materials and Methods. However, there were two women who had both early and late amniocenteses for genetic screening and fetal lung maturity, respectively (Table 3). In both, amniotic fluid pMBP rose during
100
o~
!}noc.u0s
8O
0
h. Standard MBP
-
•
0
~
m
~E
-2 -3
I,-
q
20:
I 2
I 4
I 6
Dose, ng or/JJ
I I~
10
-1.2
-0.8
-04
0
04
08
12
Log dose, ng or/.LI
Fig. 1. Comparison of dose response curves for amniotic fluids and purified MBP in the DARIA. (A) Untransformed data, (B) the same data after logit-log transformation. Linear regression analysis showed r 2 = 0.988 for purified MBP and a mean r 2 of 0.984 (range 0.978-0.999) for the amniotic fluid. One-way analysis of covariance indicated that the slopes for titrations of amniotic fluid differ from that of pure MBP (F5,81 =3.09, P < 0.05).
51
TABLE
1
Effect o f r e d u c t i o n a n d a l k y l a t i o n , a n d h e a t i n g o n M B P c o n c e n t r a t i o n s in a m n i o t i c fluid a n d sera. All s a m p l e s w e r e a n a l y z e d b y the D A R I A . S a m p l e s t h a t were h e a t e d were p l a c e d in a 56°C w a t e r b a t h f o r 2 h. P e r c e n t a g e s s h o w n i n d i c a t e the a m o u n t r e c o v e r e d c o m p a r e d to the total as m e a s u r e d w h e n the s a m p l e is r e d u c e d a n d a l k y l a t e d ( R & A ) . Specimen
Treatment
M B P ng/ml
A m n i o t i c fluid ( 173/7 weeks)
R&A No R&A
3362 675 (20.1'7,,)
R&A then heat Heat then R&A
223 (6.6%) 3835 ( > 100%)
R&A
6894
No R&A R&A then heat Heat then R&A
1,234 (17.9%) 302 (4.4'7,,) 6716 (97.4%)
Hypereosinophilic syndrome, serum
R&A No R&A R&A then heat Heat then R&A
19,951 2351 (11.8%) 431 (2.2%) 13,053 (65%)
Normal male serum
R&A No R&A R & A t h e n heat Heat then R&A
328 122 (37.2%) 138 (42%) 326 (99.4%)
A m n i o t i c fluid (38 weeks)
TABLE 2 p M B P (ng/ml) levels in a m n i o t l c fluid a n d s e r u m in e a r l y a n d late g e s t a t i o n .
E a r l y g e s t a t i o n , n = 28 = 16.8 w e e k s
Late gestation, n =64 = 38.9 w e e k s
Blood group isoimmunizat i o n , n = 20 ,~- = 32.4 w e e k s
A m n i o t i c fluid
Serum
n = 28 ~ = 1438
n = 16 ~ = 3596
M e d i a n = 971 Range = 160-4802
M e d i a n = 3596 Range 1568-6719
n =64
n = 52
~ = 2567 M e d i a n = 1851 Range = 173-8700
~ = 4630 M e d i a n = 4782 Range = 1439-9386
n = 20 = 5649 M e d i a n = 4852
n--9 .7" = 8559 M e d i a n = 4980 Range = 3321-16,186
Range = 1528-11,791
52
TABLE 3 pMBP levels in amniotic fluid obtained at mid-gestation and at term in two patients.
Patient No. l Patient No. 2
Gestational age (days)
Amniotic fluid pMBP (ng/ml)
116 267 117 267
499 470t 2929 3135
gestation, although in patient 1 the rise was almost 10-fold in contrast to only a modest change in patient 2. Both had normal karyotypes and normal pregnancies. The samples obtained from four women with blood group isoimmunization presented an opportunity to evaluate serial amniotic fluid samples from a single patient (Fig. 2). The first samples obtained from these women were at mid-gestation (;~ = 28.6 weeks). There were no samples from the normal pregnancy group at mid-gestation for comparison. Interestingly, in all four women pMBP levels were exceedingly high at mid-gestation and actually decreased with increasing gestational age. The high levels seen at midgestation raise the overall mean level of this group. Actually, the last samples (2 = 36.4 weeks) had pMBP levels comparable to those seen in the late gestation group (Fig. 2, Table 2).
Relationship of serum and amniotic jluid pMBP The above results established that pMBP immunoreactivity is present in the amniotic fluid. Figure 3 and Table 2 show pMBP levels in paired sera and
10,950
O) C
8,950
~"
6,950
._(2 4,950
-_o c E
2,951:) 951) 175
I 200
I 225
250
275
Gestational age, days Fig. 2, Serial amniotic fluid samples from four women with blood group isoimmunization, pMBP levels fell from very high levels at mid gestation to levels at term that are within the range seen in late gestation (See Table 2).
I
Serum
I
Amniotic fluid
100
1,500
~.
2,2s0
250
2,900
"~ J~
4,300
E "~.
4,250
6,250
5.700
E ~)
7,100
Amnioficfluid
B
Serum
I
~
~
"E
0.
1,200
3.2oo
5,200
9.200
O~ C 13,200 E '~0 11,200
15,200
17,200
I
J Amniotic fluid
C
I
Serum
Fig. 3. Amniotic fluid/serum pairs. With few exceptions, serum pMBP levels are higher than paired amniotic fluid. This is shown in each of the three groups: (A) late gestation, (B) early gestation, and (C) blood group isoimmunization.
E
"(~ J::l
¢1 O
"~
8,250
10,250
54
amniotic fluid. In 68 of the 77 (88"/,,) paired samples, the serum levels were higher than the amniotic fluid levels (2-4 times higher). Although protein levels rise in both fluids with advancing gestational age and there is a trend for higher amniotic fluid levels to be associated with higher serum levels, neither early (r = +0.33, NS) or late (r = +0.1, NS) amniotic fluid/serum pairs are statistically correlated. Discussion
A protein has been identified in the sera and amniotic fluid of all pregnant women that is immunochemically similar to the core granule protein of the eosinophil, the major basic protein. The role of this toxic protein in normal reproductive physiology remains an enigma. Here, pMBP levels were determined in maternal serum and amniotic fluid in a large number of normal pregnancies. Forty-nine percent of all samples were obtained at the time of ruptured membranes. Twenty-two percent of all samples were obtained from genetic amniocenteses for increased maternal age. There were no abnormal karyotypes. There was one group comprised from pregnancies complicated by maternal blood group isoimmunization (19% of all samples), and for purposes of analysis, these were kept in a separate group. pMBP was detected in all amniotic fluid samples and had immunochemical properties similar to the granule protein of the eosinophil. Prior studies have established that there is no maternal eosinophilia to account for the increased levels of the protein (Maddox et al., 1983). pMBP has been localized to the placenta, the likely source for both maternal serum and amniotic fluid pMBP. In particular, immunofluorescent techniques have localized pMBP to the X-cell, a non-villous trophoblast found predominantly at the maternal-fetal junction of the basal plate, and in the placental septa and lining the septal cysts (Maddox et al., 1984). As with other placental proteins, pMBP levels are higher in maternal serum than in either fetal serum or amniotic fluid (Klopper, 1980). The non-villous trophoblast is in closer proximity with the maternal intervillous space than with the fetal capillaries which are surrounded by the four-layered placental membrane. Umbilical cord serum has pMBP levels equivalent to normal adult male levels (Maddox et al., 1983). The source o f p M B P in amniotic fluid could either be indirectly through the fetus, or more likely, directly by diffusion through the semipermeable amnion-chorion membrane. Interestingly, the amniotic fluid obtained from pregnancies complicated by maternal blood group isoimmunization showed higher mean levels of pMBP compared to the mean levels from normal pregnancies (P < 0.001, two-sided t-test). Hemolytic disease is unlikely to affect eosinophils and is, therefore, an unlikely source for the higher values. The higher mean pMBP levels in the
55
isoimmunization group may simply be the result of the timing of the amniotic fluid samples (2 = 32.4 weeks). A plausible hypothesis is that pMBP production plateaus at mid-gestation and the increasing amniotic fluid volume dilutes the level of the protein measured. This is a difficult hypothesis to test as serial samples of amniotic fluid from normal gestations cannot be obtained. The data from the group of women with serial amniocenteses for determination of hemolysis in blood group isoimmunization suggests that amniotic fluid levels o f p M B P initially rise to very high levels (Fig. 2) but then decrease with increasing gestational age. This would correspond with increasing amniotic fluid volume which consists mainly of fetal urine (Brace, 1989). Neonatal urine has undetectable amounts of pMBP (unpublished data), and thus the increasing volume of fetal urine could act to dilute the levels of pMBP. The protein level in late gestation is still significantly higher than that found in early gestation. One pregnancy in this study was complicated by intrauterine growth retardation (IUGR). Only amniotic fluid was obtained from this patient and demonstrated one of the highest amniotic fluid values measured in our laboratory (8700 ng/ml). Of note, this patient had normal amniotic fluid volume measured by ultrasound prior to the amniocentesis. It has previously been reported that massive proliferation of the X-cell occurs in the placentas from infants with some types of growth retardation (Ermocilla and Altshuler, 1973).
Acknowledgements Supported in part by the National Institutes of Health, HD 22924, AI 07047 and AI 09728, and the Mayo Foundation. We thank Terri L. Wasmoen, Ph.D. for assistance in the inception of this study and Mrs. Linda H. Arneson for secretarial help.
References Ackerman, S.J., Gleich, G.J., Loegering, D.A., Richardson, B.A. and Butterworth, A.E. (1985) Comparative toxicity of purified human eosinophil granule cationic proteins for schistosomula of Schistosoma mansoni. Am. J. Trop. Med. Hyg. 34, 735-745. Brace, R.A. (1989)Amniotic fluid dynamics. In: Maternal-Fetal Medicine: Principles and Practice (Creasy, R.K. and Resnik, R., eds.), W.B. Saunders Co., Philadelphia, p. 128. Coulam, C.B., Wasmoen, T., Creasy, R., Siiteri, P. and Gleich, G.J. (1987) Major basic protein as a predictor of preterm labor: A preliminary report. Am. J. Obstet. Gynecol. 156, 790-796. Ermocilla, R. and Altshuler, G. (1973) The origin of "X cells" of the human placenta and their possible relationship to intrauterine growth retardation: An enigma. Am. J. Obstet. Gynecol. 117, 1137-1140. Frigas, E. and Gleich, G.J. (1986) The eosinophil and the pathophysiology of asthma. J. Allergy Clin. Immunol. 77, 527-537. Gleich, G.J. and Adolphson, C.R. (1986) The eosinophilic leukocyte: Structure and function. Adv. lmmunol. 39, 177-253.
56 Gleich, G.J., Loegering, D.A., Mann, K.G. and Maldonado, i.E. (1986) Comparative properties of the Charcot-Leyden crystal protein and the major basic protein from human eosinophils. J. Clin. Invest. 57, 633-640. Gleich, G.J. (1990) The cosinophil and bronchial asthma: Current understanding. Review article. J. Allergy Clin. Immunol. 85, 422-436. Klopper, A. (1980) The new placental proteins. Placenta 1, 77-89. Leiferman, K.M., Peters, M.S. and Gleich, G.J. (1986) The eosinophil and cutaneous edema. J. Am. Acad. Dermatol. 15, 513-517. Maddox, D.E., Butterfield, J.H., Ackerman, S.J., Coulam, C.B. and Gleich, G.J. (1983) Elevated serum levels in human pregnancy of a molecule immunochemically similar to eosinophil granule major basic protein. J. Exp. Med. 158, 1211-1226. Maddox, D.E., Kephart, G.M., Coulam, C.B., Butterfield, J.H., Benirschke, K. and Gleich, G.J. (1984) Localization of a molecule immunochemically similar to eosinophil major basic protein in human placenta. J. Exp. Med. 160, 29-41. McConahey, P.J. and Dixon, F.J. (1966) A method of trace iodination of proteins for immunologic studies. Int. Arch. Allergy Appl. Immun. 29, 185-189. Wasmoen, T.L., Coulam, C.B., Leiferman, K.M. and Gleich, G.J. (1987) Increases of plasma eosinophil major basic protein levels late in pregnancy predict onset of labor. Proc. Natl. Acad. Sci. U.S.A. 84, 3029-3032. Wasmoen, T.L., McKean, D.J., Benirschke, K., Coulam, C.B. and Gleich, G.J. (1989) Evidence for eosinophil granule major basic protein in human placenta. J. Exp. Med. 170, 2051-2063. Wassom, D.L., Loegering, D.A., Solley, G.O. et al. (1981) Elevated serum levels of the eosinophil granule major basic protein in patients with eosinophilia. J. Clin. Invest. 67, 651-661.
