Journal of Immunological Methods, 136 (1991) 85-90 © 1991 Elsevier Science Publishers B.V. (Biomedical Division) 0022-1759/91/$03.50 ADONIS 0022175991000623
85
J1M 05809
Monoclonal antibodies against endometrial protein PP14 and their use for purification and radioimmunoassay of PP14 L. Riittinen 1, O. N~irv~inen z, I. Virtanen 2 a n d M. Sepp~il~i 1 I First Department of Obstetrics and Gynecology, Helsinki University Central Hospital, SF-00290 Helsinki, Finland, and 2 Department of Anatomy, University of Helsinki, SF-O0170 HelsinkL Finland
(Received 24 May 1990, revised received 31 August 1990, accepted 1 October 1990)
Monoclonal antibodies against placental protein 14 (PP14) were generated in mice. One of these (code 105DH1F1) was used for the purification of PP14 from mid-trimester amniotic fluid. The method is a simple one-step procedure, whereby monoclonal antibodies are bound to a solid phase and used for affinity immunoadsorption. 1.2 mg of PP14 was recovered from 25 ml of amniotic fluid. Using this purified PP14 and another monoclonal antibody (code 105AH7G3) a highly specific radioimmunoassay was developed. When the urine of pregnant women (7-10 weeks gestation) was tested little PP14 could be found at the detection level of 10 #g/1. The serum patterns of chorionic gonadotropin and PP14 were very similar in pregnancy. This study suggests that, unlike hCG, little PP14 is secreted into maternal urine. The purification method is gentle and allows quick purification of large amounts of PP14 for studies of its biological action(s). Key words: Placental protein 14; Monoclonal antibody; Protein purification; Chromatography, monoclonal immunoaffinity; Lactoperoxidase iodination; Urine PP14
Introduction
Placental protein 14 (PP14) was originally isolated from human term placenta. It is a glycoprotein containing 17.5% carbohydrate (Bohn et al., 1982). As judged by immunological criteria, PP14 is similar to progestagen-dependent endometrial protein (Julkunen et al., 1986b) and pregnancy-associated endometrial a 2 globulin (Bell and Bohn,
Correspondence to: L. Riittinen, Hormone Laboratory, Departments I and II of Obstetrics and Gynecology, Helsinki University Central Hospital, Haartmaninkatu 2, SF-00290 Helsinki, Finland. Abbreviations: PP14, placental protein 14; PBS, 50 mM sodium phosphate, 0.9% NaC1, pH 7.4; TBS, 50 mM Tris-HCl, 0.9% NaCI, 0.05% NAN3, pH 7.7; RIA, radioimmunoassay.
1986). In contrast to placental proteins, high levels of PP14 are found in amniotic fluid (Julkunen et al., 1985). Structurally PP14 resembles fl lactoglobulins from various species. This similarity was first suggested by its N terminal protein sequence (Huhtala et al., 1987) and subsequently by the complete cDNA sequence and primary structure (Julkunen et al., 1988). Although originally isolated from the placenta, PP14 is not synthesized by the placenta. Instead, secretory and decidualized endometria appear to be capable of incorporating labelled methionine into PP14 (Julkunen, 1986; Julkunen et al., 1986a) and also contain PP14-specific m R N A (Julkunen et al., 1988). Immunohistochemical staining of an immunologically related protein, pregnancy-associated endometrial a 2 globulin, shows maximal staining in endometrial glands during the late luteal phase
86 (Waites et al., 1988). During pregnancy the profile of serum PP14 levels resembles that of hCG, the highest levels occurring at 7-10 weeks. In this study we report the generation of monoclonal antibodies against PP14 and a simple method for their use in the purification of PP14 from amniotic fluid. We also wish to report that, in contrast to serum, PP14 levels in the urine are not elevated at 7-10 weeks pregnancy.
Materials and methods
Purification of PP14 for immunization PP14 was partially purified from human amniotic fluid collected at 15-16 weeks' gestation for fetal chromosome analysis. Amniotic fluid was dialyzed against 50 mM N H 4 H C O 3 buffer, pH 8.3 and then fractionated at room temperature on a DEAE-Sepharose CL-6B column (2.0 × 3.8 cm, Pharmacia) with a linear N H 4 H C O 3 gradient, pH 8.3, containing 2% isopropanol, from 50 to 400 mM. Gel filtration was performed on a Sephadex G-100 column (2.5 × 59 cm, Pharmacia) with 20 mM sodium phosphate buffer, pH 8.0, containing 0.02% NaN 3. The final step was anion exchange chromatography on a Mono Q column (Pharmacia) using a gradient of 20-92 mM sodium phosphate, pH 8.0, containing 0.5% isopropanol. The partially purified protein was dialyzed against 50 mM sodium phosphate, 0.9% NaC1, p H 7.4 (PBS) and used for the generation of monoclonal antibodies (see below). Generation and characterization of monoclonal antibodies Two B A L B / c mice were given intraperitoneal injections of 50/zg partially purified PP14 in PBS. Two booster injections of 50 /~g were given subcutaneously in Freund's complete adjuvant at two-week intervals. 11 days after the second subcutaneous injection the mice received 20 /zg of PP14 in PBS intravenously and were killed 5 days later. Hybridoma cultures were carried out by fusing the spleen cells with the myeloma cell line X63-Ag8.653 (K~Shler and Milstein, 1975). Peritoneal macrophages were used as feeder cells in
cell cultures. Cloning was done under a dissecting microscope by selecting individual cells with a micropipette. Antibody-producing hybridomas were selected using a polyclonal PP14 RIA (Julkunen et al., 1985; Sepp~il~i et al., 1987). Selected hybrids were injected into the peritoneal cavity of B A L B / c mice pretreated with a 0.5 ml intraperitoneal injection of 2,6,10,14-tetramethyl pentadecane (Aldrich-Chemie, Steinheim, F.R.G.). Antibody type was determined by radial inamunodiffusion using reagents from Nordic Immunological Laboratories (Tilburg, The Netherlands).
Use of monoclonal antibodies for purification of PP14 from amniotic fluid Ascitic fluid (12.5 ml) from clone 105DH1F1 was precipitated with 12.5 ml of 36% Na2SO 4, washed with 18% Na2SO 4 and diluted in 0.1 M N a H C O 3, 0.5 M NaC1, p H 8.0. The protein was coupled with CNBr-activated Sepharose 4B (Pharmacia Fine Chemicals, Uppsala, Sweden) and washed according to the manufacturer's instructions. The column (1.5 × 20 cm) was packed with 50 mM Tris-HC1, 0.9% NaCI, 0.05% NaN 3, pH 7.7 (TBS). 20 ml of 1% bovine serum albumin in TBS were then applied to the column. The column was washed with 0.1% trifluoro-acetic acid containing 1 mM CaC1 e until the absorbance at 280 nm was zero. Then the column was washed with TBS. Human amniotic fluid samples containing more than 80 mg/1 PP14 were pooled. 25 ml of pooled amniotic fluid was recirculated through the monoclonal antibody column (4°C, 20 m l / h ) overnight. The monoclonal antibody column was washed (50 m l / h ) until the absorbance at 280 nm was zero using the following buffers: 1, TBS; 2, 1 M NaC1, containing 1% isopropanol; 3, 10 mM ammonium acetate containing 0.005% isopropanol, pH 5.0. PP14 was eluted from the column with 0.1% trifluoro-acetic acid containing 1 mM CaC12. The flow rate was 5 m l / h and the fraction size 5 ml. The tubes into which the fractions were collected contained 200 ~1 of 1 M Tris-HC1 buffer, pH 9.0. The PP14-containing fractions were pooled and dialysed against 100 mM sodium phosphate buffer, pH 8.0. After elution the column was washed with TBS.
87
Characterization of the protein Assays for total protein were performed according to the manufacturer's instructions using the Bio-Rad protein assay (Bio-Rad Laboratories, Richmond, California). SDS-PAGE was performed using the PhastSystem (Pharmacia LKB Biotechnology, Uppsala, Sweden) with a 12.5% gel and silver staining according to the manufacturer's instructions. Immunoblot analysis was done as follows. Proteins were transferred from the SDSPAGE gel to a nitrocellulose filter moistened with 25 mM Tris-HC1, pH 8.3, 192 mM glycine, 20% ( v / v ) methanol by 2 h diffusion at 37°C. The filter was blocked overnight at 4 ° C with 3% BSA in TBS without NaN 3 and then it was treated with a polyclonal anti-PP14 antiserum (Riittinen et al., 1989) (1/100 with TBS without NaN 3 and containing 3% BSA) at 3 7 ° C for 1 h followed by washing with TBS without NaN 3. The filter was then incubated with peroxidase-conjugated second antibody (Dako, Glostrup, Denmark) diluted 1/100 with TBS without NaN 3 and containing 3% BSA, followed by washing with TBS without N a N 3. Peroxidase was detected using 3,3'-diamino-benzidine-tetrahydro chloride (300 mg/1) and H202 (2.5 ml/1) in TBS without NaN 3. Monoclonal PP14 RIA The label. PP14 was radiolabelled using the solid-phase lactoperoxidase method modified from a previously described method (Karonen et al., 1975; Sepp~il~i et al., 1987). To 6 #g of PP14 in 25 /~1 of PBS were added 100/zl lactosorbent solution ( 1 / 3 0 dilution in 0.1 M sodium acetate buffer, pH 6.0) and 1 mCi of Na125I (IMS 300, Amersham Laboratories, Amersham, England). The tube was vortexed and 5/.tl of 0.6% H202 were added. The tube was incubated for 10 min at room temperature. During the incubation it was vortexed twice. Then 3 ~1 of 0.6% H202 were added and the incubation repeated as before. Again 3 ~1 of 0.6% H202 were added and the incubation was repeated. After the incubation the tube was centrifuged. The supernatant was passed through a PD10 Sephadex G-25M column (Pharmacia LKB Biotechnology, Uppsala, Sweden) prewashed with 5 ml of TBS, 2 ml of 1% BSA in TBS, and 50 ml of TBS. Fractions (ten drops each) were collected
into tubes containing 1.5 ml of 1% BSA in TBS. The fractions containing iodinated PP14 were pooled and used as the label. Standards and controls. The standards consisted of purified PP14 prepared as described in this article and diluted in TBS containing 1% BSA. Three samples of pooled pregnancy serum, diluted in the same buffer, were included in each assay to give reference values at various PP14 concentrations. First antibody. Ascitic fluid from clone 105AH7G3 was diluted in TBS containing 1% BSA in order to bind about 20% of the label. This preparation was used as the first antibody. Second antibody. The second antibodies, i.e., rabbit immunoglobulins against mouse immunoglobulins (Dako, Glostrup, Denmark), were bound to a solid phase. 500 ~l of the second antibody diluted with TBS to contain 10 rag/1 were added to Maxisorp Star Tubes (Nunc, Kamstrup, Denmark) and incubated overnight at room temperature. They were then washed with 2 ml of TBS and blocked for 1 h with 1 ml of TBS containing 1% BSA. Finally the tubes were washed with 2 ml of TBS and stored at 4 ° C in a moist chamber. Assay procedure. 50 ~1 of standard, sample, or 1% BSA in TBS (blank) were added in duplicate to the coated tubes. Then 200/zl of appropriately diluted label and the first antibody were added. The tubes were incubated overnight at 4°C, washed with 2 ml of distilled water and the radioactivity counted.
Stability of PP14 in serum as measured by RIA 19 fresh serum samples were divided into seven identical batches and frozen at - 2 0 ° C . After thawing the batches were kept under the following conditions: (1) at room temperature for 16 days; (2) at 4 ° C for 16 days; (3) at room temperature for 8 days; (4) at 4 ° C for 8 days. One batch was frozen and thawed ten times before the assay and another batch five times. One of the batches was thawed just before the assay. P values were calculated for differences using paired Student's t test. hCG assay The h C G concentration was measured using an immunofluorometric assay (Stenman et al., 1987).
88
4000
TABLE I
•
P U R I F I C A T I O N OF PP14 F R O M A M N I O T I C F L U I D
3000 Purification step Amniotic fluid Monoclonal column
Vol. (ml)
PP14 (rag)
Tot. protein (mg)
Recovery (%)
Degree of purity ~
25
2.53
95.0
100
0.027
9
1.22
1.17
48.2
t purified PPI4 amniotic fluid pregnancy serum
• o •
~. 2OOO1000
1.04
0
.
.
.
.
.
]0
PP14 concentration was determined by RIA and protein concentration by Bio-Rad protein assay.
.
.
.
100
.
.
.
1000
.
10000
PP 141ag/l Fig. 2. Dose-response curves of purified PP14, amniotic fluid and pregnancy serum.
Pregnancy urines and sera Pregnancy urines and sera were obtained during the 7 - 1 0 t h week of pregnancy. They were collected as overnight fasting samples and stored at - 2 0 ° C .
Results Both monoclonal antibodies ( 1 0 5 D H I F 1 and 1 0 5 A H 7 G 3 ) were of the IgG1 type. The purification steps are shown in Table I. F r o m 25 ml of amniotic fluid 1.2 mg of PP14 was recovered, corresponding to a yield of 48.2%. S D S - P A G E and i m m u n o b l o t analyses of purified PP14 are shown in Fig. 1. I n RIA, purified PP14, pregnancy serum and amniotic fluid gave parallel dose-re-
sponse curves (Fig. 2). The results obtained with the PP14 R I A described here and with the polyclonal R I A described previously (Seppiil~i et al., 1987; Riittinen et al,., 1989) showed a good correlation ( r = 0.88) (Fig. 3). The detection limit of the PP14 R I A was defined as the PP14 concentration corresponding to the mean radioactivity of 12 blank tubes minus 2 SD. The detection range was 10-1000 /~g/1 with a nearly linear dose-response curve (Fig. 2). The affinity constant was 1.14 × 109 M-~. W h e n 50 /xg/1 of purified PP14 was added to serum containing 82/~g/1 of endogenous PP14, the analytical recovery was 91%. The intra-assay variation (n = 20) was 8.3% at the level of 122 400
94 kDa > 67 kDa 43 kDa >
< e¢
a-
30 kDa--> 20.1 kDa > 14.4 k D a - - >
o
300
2oo
O
-d O
100
O
0
1
2
3
4
Fig. 1. 12.5% SDS-PAGE (lanes 1-2) and immunoblot analysis (lanes 3-4) of PP14. Lanes 1 and 4 : P P I 4 purified by the method described in this article (90 ng of PP14); lanes 2 and 3: PP14 purified from placenta (Bohn et al., 1982) (80 ng of PPI 4). The samples were reduced with mercaptoethanol.
i
i
i
100
200
300
400
polyclonal PPI4 RIA (~g/1) Fig. 3. Correlation of 85 serum samples between PP14 described in this article ('monoclonal PP14 RIA') and the RIA described previously (polyclonal PPI4 RIA) (Sepp~il~i et al., 1987; Riittinen et al., 1989). The positive correlation was described by the equation y = 16.35 +0.65x, r = 0.88.
89 ~g/1, 8.1% at the level of 50/zg/1, and 8.9% at the level of 25 #g/1. The interassay variation (n = 20) was 12.6% at the level of 121 ~g/1, 9.7% at the level of 57 ~g/1 and 11.4% at the level of 31 /~g/1. When serum was kept at room temperature, or at 4 ° C , or when frozen and thawed, the PP14 levels did not change significantly. Urinary PP14 concentrations were below the detection level of the assay (10 ~g/1) in 19 out of 24 specimens and the highest level observed was 15.8 /.tg/1. The hCG concentration in the same urine samples varied from 38 to 7362 /~g/1 (mean 2119 ~g/1). Serum levels of PP14 varied from 183 to 1828 # g / l (mean 879 ~g/1) and of hCG from 41 to 7331 ~ g / l (mean 3480/~g/l).
Discussion The biological action of PP14 is unclear and it is necessary to use a gentle method for the purification of PP14 in order to preserve its activity for such studies. Amniotic fluid PP14 is obviously the same protein as PP14 purified from the placenta (Bohn et al., 1982). This has been shown by immunological criteria (Riittinen et al., 1989) and by its N terminal protein sequence (Huhtala et al., 1987). The same sequence has also been deduced from cDNA sequence analysis (Julkunen et al., 1988). Mid-trimester amniotic fluid contains a high concentration of PP14 (Julkunen et al., 1985) and this appropriate starting material is abundantly available in units carrying out prenatal diagnosis from amniotic fluid. Compared with the previous purification methods from decidua (Sutcliffe et al., 1980; Westwood et al., 1988), endometrium (Bell, 1986) or placenta (Bohn et al., 1982) the purification method presented here, though simple, is much more rapid. The particular advantage of using monoclonal antibodies for the purification of PP14 is the high yield. It is possible to use the same antibody column several times. We have now obtained 5.6 mg of PP14 using the same column, and it is still quite effective. It is possible to purify larger quantities of PP14 by increasing the column size and the volume of the starting material. In addition to the smaller PP14 concentration in tissue, contamination with hCG has been a
problem in PP14 preparations purified from the placenta. The hCG concentration is low in amniotic fluid. Therefore it is not surprising that PP14 purified from amniotic fluid using the monoclonal antibody column contains only 0.00038% ( w / w ) of hCG. The minuscule hCG contamination of this purified PP14 is unlikely to interfere in any studies on the biological action(s) of PP14. Another advantage of using monoclonal antibodies is their specificity and abundance for use in assays, e.g., in RIA such as shown in this study. Unlike hCG, high concentrations of PP14 cannot be found in pregnancy urine at 7-10 weeks gestation in spite of the fact that the circulating levels of both hCG and PP14 are highest during this period. When pure PP14 was added to the urine, after 15 days storage at - 2 0 ° C, the measured value was 64-75% of the added PP14. Since the maximal storage time of the urine samples was 15 days low PP14 levels in urine cannot be explained by a complete loss of antigenicity. These results show that, while both PP14 and hCG are high in serum at the same stage of pregnancy, PP14 is mostly secreted into the amniotic fluid and less is detected in the urine than is the case with hCG.
Acknowledgements This study was supported by grants from Finnish Social Insurance Institution, Sigrid Jus61ius Foundation, and the Academy of Finland. I.V. was supported by a grant from the Finnish Cancer Research Fund and L.R. from Research and Science Foundation of Farmos. We thank Ms. Pirkko Selenius and Ms. Saija Roine for technical assistance.
References Bell, S.C. (19186) Purification of human secretory pregnancyassociated endometrial alpha-2-globulin (alpha-2-PEG) from cytosol of first trimester Pregnancy endometrium. Hum. Reprod. 1, 313-318. Bell, S.C. and Bohn, H. (1986) Immunochemicaland biochemical relationship between human pregnancy-associated secreted endometrial al- and a2-globulins (al- and a2-PEG)
90 and the soluble placental proteins 12 and 14 (PP12 and PP14). Placenta 7, 283-294. Bohn. H., Kraus, W. and Winckler, W. (1982) New soluble placental proteins: Their isolation, characterization, localization and quantification. Placenta 4 (suppl.), 67-81. Huhtala, M.-L, Sepp~l~, M., N~rv~nen, A., Palom~ki, P., Julkunen, M. and Bohn, H. (1987) Amino acid sequence homology between human placental protein 14 and /~lactoglobulins from various species. Endocrinology 120~ 2620-2622. Julkunen, M. (1986) Human decidua syntheses placental protein 14 (PPI4) in vitro. Acta Endocrinol. 112, 271-277. Julkunen, M., Rutanen, E.-M., Koskimies, A., Ranta, T., Bohn, H. and Sepp~.l~, M. (1985) Distribution of placental protein 14 in tissues and body fluids during pregnancy. Br J. Obstet. Gynaecol. 92, 1145-1151. Julkunen, M., Koistinen, R., SjOberg, J., Rutanen, E.-M., WahlstrOm, T. and Sepp~,l~, M. (1986a) Secretory endometrium synthesizes placental protein 14. Endocrinology 118, 1782-1786. Julkunen, M., Raikar, R.S., Joshi, S.G., Bohn, H. and Sepp~l~, M. (1986b) Placental protein 14 and progestagen-dependent endometrial protein are immunologically indistinguishable. Hum. Reprod. 1, 7-8. Julkunen, M., Sepp~l~ and M. J~nne, O.A. (1988) Complete amino acid sequence of human placental protein 14: A progesterone-regulated uterine protein homologous to /~lactoglobulins. Proc. Natl. Acad. Sci. U.S.A. 85, 8845-8849. Karonen, S.-L., M~rsky, P., Siren, M. and Senderling, U. (1975) An enzymatic solid-phase method for trace iodination of proteins and peptides with 12Siodine. Anal. Biochem. 67, 1-10.
K~Shler, G. and Milstein, C. (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256, 495-497., Riittinen, L., Stenman, U.,-H., Alfthan, H., SuikkarL A.-M.. Bohn, H. and Sepp~l~., M. (1989) Time-resolved immunofluorometric assay for placental protein 14. Clin. Chim. Acta 183, 115-124. Sepp~l~, M., ROnnberg, L., Karonen, S.-L. and Kauppila, A. (1987) Micronized oral progesterone increases the circulating level of endometrial secretory PP14//3 lactoglobulin homologue. Hum. Reprod. 2, 453-455. Stenman, U.-H., Alfthan, H., Ranta, T.~ Vartiainen, E., Jalkanen, J. and Sepp~l~, M. (1987) Serum levels of human chorionic gonadotropin in nonpregnant women and men are modulated by gonadotropin releasing hormone and sex steroids. J. Clin. Endocrinol. Metab. 64, 730-736. Sutcliffe, R.G., Bolton, A.E., Sharp, F., Nicholson, L.V.B. and MacKinnon, R. (1980) Purification of human alpha uterine protein. J. Reprod. Fert. 58, 435-442. Waites, G.T., Wood, P i . , Walker, R.A. and Bell, S.C. (1988) |mmunohistological localization of human endometrial secretory protein, 'pregnancy-associated endometrial a2" globulin' (a2-PEG), during the menstrual cycle. J. Reprod. Fert. 82, 665-672. Westwood, O.M.R,, Chapman, M.G., Totty, N., Philp, R., Bolton, A.E. and Lazarus, N.R. (1988) N-terminal sequence analysis of human placental protein 14, purified in high yield from decidual cytosol. J. Reprod. Fert. 82, 493 500.
85
J1M 05809
Monoclonal antibodies against endometrial protein PP14 and their use for purification and radioimmunoassay of PP14 L. Riittinen 1, O. N~irv~inen z, I. Virtanen 2 a n d M. Sepp~il~i 1 I First Department of Obstetrics and Gynecology, Helsinki University Central Hospital, SF-00290 Helsinki, Finland, and 2 Department of Anatomy, University of Helsinki, SF-O0170 HelsinkL Finland
(Received 24 May 1990, revised received 31 August 1990, accepted 1 October 1990)
Monoclonal antibodies against placental protein 14 (PP14) were generated in mice. One of these (code 105DH1F1) was used for the purification of PP14 from mid-trimester amniotic fluid. The method is a simple one-step procedure, whereby monoclonal antibodies are bound to a solid phase and used for affinity immunoadsorption. 1.2 mg of PP14 was recovered from 25 ml of amniotic fluid. Using this purified PP14 and another monoclonal antibody (code 105AH7G3) a highly specific radioimmunoassay was developed. When the urine of pregnant women (7-10 weeks gestation) was tested little PP14 could be found at the detection level of 10 #g/1. The serum patterns of chorionic gonadotropin and PP14 were very similar in pregnancy. This study suggests that, unlike hCG, little PP14 is secreted into maternal urine. The purification method is gentle and allows quick purification of large amounts of PP14 for studies of its biological action(s). Key words: Placental protein 14; Monoclonal antibody; Protein purification; Chromatography, monoclonal immunoaffinity; Lactoperoxidase iodination; Urine PP14
Introduction
Placental protein 14 (PP14) was originally isolated from human term placenta. It is a glycoprotein containing 17.5% carbohydrate (Bohn et al., 1982). As judged by immunological criteria, PP14 is similar to progestagen-dependent endometrial protein (Julkunen et al., 1986b) and pregnancy-associated endometrial a 2 globulin (Bell and Bohn,
Correspondence to: L. Riittinen, Hormone Laboratory, Departments I and II of Obstetrics and Gynecology, Helsinki University Central Hospital, Haartmaninkatu 2, SF-00290 Helsinki, Finland. Abbreviations: PP14, placental protein 14; PBS, 50 mM sodium phosphate, 0.9% NaC1, pH 7.4; TBS, 50 mM Tris-HCl, 0.9% NaCI, 0.05% NAN3, pH 7.7; RIA, radioimmunoassay.
1986). In contrast to placental proteins, high levels of PP14 are found in amniotic fluid (Julkunen et al., 1985). Structurally PP14 resembles fl lactoglobulins from various species. This similarity was first suggested by its N terminal protein sequence (Huhtala et al., 1987) and subsequently by the complete cDNA sequence and primary structure (Julkunen et al., 1988). Although originally isolated from the placenta, PP14 is not synthesized by the placenta. Instead, secretory and decidualized endometria appear to be capable of incorporating labelled methionine into PP14 (Julkunen, 1986; Julkunen et al., 1986a) and also contain PP14-specific m R N A (Julkunen et al., 1988). Immunohistochemical staining of an immunologically related protein, pregnancy-associated endometrial a 2 globulin, shows maximal staining in endometrial glands during the late luteal phase
86 (Waites et al., 1988). During pregnancy the profile of serum PP14 levels resembles that of hCG, the highest levels occurring at 7-10 weeks. In this study we report the generation of monoclonal antibodies against PP14 and a simple method for their use in the purification of PP14 from amniotic fluid. We also wish to report that, in contrast to serum, PP14 levels in the urine are not elevated at 7-10 weeks pregnancy.
Materials and methods
Purification of PP14 for immunization PP14 was partially purified from human amniotic fluid collected at 15-16 weeks' gestation for fetal chromosome analysis. Amniotic fluid was dialyzed against 50 mM N H 4 H C O 3 buffer, pH 8.3 and then fractionated at room temperature on a DEAE-Sepharose CL-6B column (2.0 × 3.8 cm, Pharmacia) with a linear N H 4 H C O 3 gradient, pH 8.3, containing 2% isopropanol, from 50 to 400 mM. Gel filtration was performed on a Sephadex G-100 column (2.5 × 59 cm, Pharmacia) with 20 mM sodium phosphate buffer, pH 8.0, containing 0.02% NaN 3. The final step was anion exchange chromatography on a Mono Q column (Pharmacia) using a gradient of 20-92 mM sodium phosphate, pH 8.0, containing 0.5% isopropanol. The partially purified protein was dialyzed against 50 mM sodium phosphate, 0.9% NaC1, p H 7.4 (PBS) and used for the generation of monoclonal antibodies (see below). Generation and characterization of monoclonal antibodies Two B A L B / c mice were given intraperitoneal injections of 50/zg partially purified PP14 in PBS. Two booster injections of 50 /~g were given subcutaneously in Freund's complete adjuvant at two-week intervals. 11 days after the second subcutaneous injection the mice received 20 /zg of PP14 in PBS intravenously and were killed 5 days later. Hybridoma cultures were carried out by fusing the spleen cells with the myeloma cell line X63-Ag8.653 (K~Shler and Milstein, 1975). Peritoneal macrophages were used as feeder cells in
cell cultures. Cloning was done under a dissecting microscope by selecting individual cells with a micropipette. Antibody-producing hybridomas were selected using a polyclonal PP14 RIA (Julkunen et al., 1985; Sepp~il~i et al., 1987). Selected hybrids were injected into the peritoneal cavity of B A L B / c mice pretreated with a 0.5 ml intraperitoneal injection of 2,6,10,14-tetramethyl pentadecane (Aldrich-Chemie, Steinheim, F.R.G.). Antibody type was determined by radial inamunodiffusion using reagents from Nordic Immunological Laboratories (Tilburg, The Netherlands).
Use of monoclonal antibodies for purification of PP14 from amniotic fluid Ascitic fluid (12.5 ml) from clone 105DH1F1 was precipitated with 12.5 ml of 36% Na2SO 4, washed with 18% Na2SO 4 and diluted in 0.1 M N a H C O 3, 0.5 M NaC1, p H 8.0. The protein was coupled with CNBr-activated Sepharose 4B (Pharmacia Fine Chemicals, Uppsala, Sweden) and washed according to the manufacturer's instructions. The column (1.5 × 20 cm) was packed with 50 mM Tris-HC1, 0.9% NaCI, 0.05% NaN 3, pH 7.7 (TBS). 20 ml of 1% bovine serum albumin in TBS were then applied to the column. The column was washed with 0.1% trifluoro-acetic acid containing 1 mM CaC1 e until the absorbance at 280 nm was zero. Then the column was washed with TBS. Human amniotic fluid samples containing more than 80 mg/1 PP14 were pooled. 25 ml of pooled amniotic fluid was recirculated through the monoclonal antibody column (4°C, 20 m l / h ) overnight. The monoclonal antibody column was washed (50 m l / h ) until the absorbance at 280 nm was zero using the following buffers: 1, TBS; 2, 1 M NaC1, containing 1% isopropanol; 3, 10 mM ammonium acetate containing 0.005% isopropanol, pH 5.0. PP14 was eluted from the column with 0.1% trifluoro-acetic acid containing 1 mM CaC12. The flow rate was 5 m l / h and the fraction size 5 ml. The tubes into which the fractions were collected contained 200 ~1 of 1 M Tris-HC1 buffer, pH 9.0. The PP14-containing fractions were pooled and dialysed against 100 mM sodium phosphate buffer, pH 8.0. After elution the column was washed with TBS.
87
Characterization of the protein Assays for total protein were performed according to the manufacturer's instructions using the Bio-Rad protein assay (Bio-Rad Laboratories, Richmond, California). SDS-PAGE was performed using the PhastSystem (Pharmacia LKB Biotechnology, Uppsala, Sweden) with a 12.5% gel and silver staining according to the manufacturer's instructions. Immunoblot analysis was done as follows. Proteins were transferred from the SDSPAGE gel to a nitrocellulose filter moistened with 25 mM Tris-HC1, pH 8.3, 192 mM glycine, 20% ( v / v ) methanol by 2 h diffusion at 37°C. The filter was blocked overnight at 4 ° C with 3% BSA in TBS without NaN 3 and then it was treated with a polyclonal anti-PP14 antiserum (Riittinen et al., 1989) (1/100 with TBS without NaN 3 and containing 3% BSA) at 3 7 ° C for 1 h followed by washing with TBS without NaN 3. The filter was then incubated with peroxidase-conjugated second antibody (Dako, Glostrup, Denmark) diluted 1/100 with TBS without NaN 3 and containing 3% BSA, followed by washing with TBS without N a N 3. Peroxidase was detected using 3,3'-diamino-benzidine-tetrahydro chloride (300 mg/1) and H202 (2.5 ml/1) in TBS without NaN 3. Monoclonal PP14 RIA The label. PP14 was radiolabelled using the solid-phase lactoperoxidase method modified from a previously described method (Karonen et al., 1975; Sepp~il~i et al., 1987). To 6 #g of PP14 in 25 /~1 of PBS were added 100/zl lactosorbent solution ( 1 / 3 0 dilution in 0.1 M sodium acetate buffer, pH 6.0) and 1 mCi of Na125I (IMS 300, Amersham Laboratories, Amersham, England). The tube was vortexed and 5/.tl of 0.6% H202 were added. The tube was incubated for 10 min at room temperature. During the incubation it was vortexed twice. Then 3 ~1 of 0.6% H202 were added and the incubation repeated as before. Again 3 ~1 of 0.6% H202 were added and the incubation was repeated. After the incubation the tube was centrifuged. The supernatant was passed through a PD10 Sephadex G-25M column (Pharmacia LKB Biotechnology, Uppsala, Sweden) prewashed with 5 ml of TBS, 2 ml of 1% BSA in TBS, and 50 ml of TBS. Fractions (ten drops each) were collected
into tubes containing 1.5 ml of 1% BSA in TBS. The fractions containing iodinated PP14 were pooled and used as the label. Standards and controls. The standards consisted of purified PP14 prepared as described in this article and diluted in TBS containing 1% BSA. Three samples of pooled pregnancy serum, diluted in the same buffer, were included in each assay to give reference values at various PP14 concentrations. First antibody. Ascitic fluid from clone 105AH7G3 was diluted in TBS containing 1% BSA in order to bind about 20% of the label. This preparation was used as the first antibody. Second antibody. The second antibodies, i.e., rabbit immunoglobulins against mouse immunoglobulins (Dako, Glostrup, Denmark), were bound to a solid phase. 500 ~l of the second antibody diluted with TBS to contain 10 rag/1 were added to Maxisorp Star Tubes (Nunc, Kamstrup, Denmark) and incubated overnight at room temperature. They were then washed with 2 ml of TBS and blocked for 1 h with 1 ml of TBS containing 1% BSA. Finally the tubes were washed with 2 ml of TBS and stored at 4 ° C in a moist chamber. Assay procedure. 50 ~1 of standard, sample, or 1% BSA in TBS (blank) were added in duplicate to the coated tubes. Then 200/zl of appropriately diluted label and the first antibody were added. The tubes were incubated overnight at 4°C, washed with 2 ml of distilled water and the radioactivity counted.
Stability of PP14 in serum as measured by RIA 19 fresh serum samples were divided into seven identical batches and frozen at - 2 0 ° C . After thawing the batches were kept under the following conditions: (1) at room temperature for 16 days; (2) at 4 ° C for 16 days; (3) at room temperature for 8 days; (4) at 4 ° C for 8 days. One batch was frozen and thawed ten times before the assay and another batch five times. One of the batches was thawed just before the assay. P values were calculated for differences using paired Student's t test. hCG assay The h C G concentration was measured using an immunofluorometric assay (Stenman et al., 1987).
88
4000
TABLE I
•
P U R I F I C A T I O N OF PP14 F R O M A M N I O T I C F L U I D
3000 Purification step Amniotic fluid Monoclonal column
Vol. (ml)
PP14 (rag)
Tot. protein (mg)
Recovery (%)
Degree of purity ~
25
2.53
95.0
100
0.027
9
1.22
1.17
48.2
t purified PPI4 amniotic fluid pregnancy serum
• o •
~. 2OOO1000
1.04
0
.
.
.
.
.
]0
PP14 concentration was determined by RIA and protein concentration by Bio-Rad protein assay.
.
.
.
100
.
.
.
1000
.
10000
PP 141ag/l Fig. 2. Dose-response curves of purified PP14, amniotic fluid and pregnancy serum.
Pregnancy urines and sera Pregnancy urines and sera were obtained during the 7 - 1 0 t h week of pregnancy. They were collected as overnight fasting samples and stored at - 2 0 ° C .
Results Both monoclonal antibodies ( 1 0 5 D H I F 1 and 1 0 5 A H 7 G 3 ) were of the IgG1 type. The purification steps are shown in Table I. F r o m 25 ml of amniotic fluid 1.2 mg of PP14 was recovered, corresponding to a yield of 48.2%. S D S - P A G E and i m m u n o b l o t analyses of purified PP14 are shown in Fig. 1. I n RIA, purified PP14, pregnancy serum and amniotic fluid gave parallel dose-re-
sponse curves (Fig. 2). The results obtained with the PP14 R I A described here and with the polyclonal R I A described previously (Seppiil~i et al., 1987; Riittinen et al,., 1989) showed a good correlation ( r = 0.88) (Fig. 3). The detection limit of the PP14 R I A was defined as the PP14 concentration corresponding to the mean radioactivity of 12 blank tubes minus 2 SD. The detection range was 10-1000 /~g/1 with a nearly linear dose-response curve (Fig. 2). The affinity constant was 1.14 × 109 M-~. W h e n 50 /xg/1 of purified PP14 was added to serum containing 82/~g/1 of endogenous PP14, the analytical recovery was 91%. The intra-assay variation (n = 20) was 8.3% at the level of 122 400
94 kDa > 67 kDa 43 kDa >
< e¢
a-
30 kDa--> 20.1 kDa > 14.4 k D a - - >
o
300
2oo
O
-d O
100
O
0
1
2
3
4
Fig. 1. 12.5% SDS-PAGE (lanes 1-2) and immunoblot analysis (lanes 3-4) of PP14. Lanes 1 and 4 : P P I 4 purified by the method described in this article (90 ng of PP14); lanes 2 and 3: PP14 purified from placenta (Bohn et al., 1982) (80 ng of PPI 4). The samples were reduced with mercaptoethanol.
i
i
i
100
200
300
400
polyclonal PPI4 RIA (~g/1) Fig. 3. Correlation of 85 serum samples between PP14 described in this article ('monoclonal PP14 RIA') and the RIA described previously (polyclonal PPI4 RIA) (Sepp~il~i et al., 1987; Riittinen et al., 1989). The positive correlation was described by the equation y = 16.35 +0.65x, r = 0.88.
89 ~g/1, 8.1% at the level of 50/zg/1, and 8.9% at the level of 25 #g/1. The interassay variation (n = 20) was 12.6% at the level of 121 ~g/1, 9.7% at the level of 57 ~g/1 and 11.4% at the level of 31 /~g/1. When serum was kept at room temperature, or at 4 ° C , or when frozen and thawed, the PP14 levels did not change significantly. Urinary PP14 concentrations were below the detection level of the assay (10 ~g/1) in 19 out of 24 specimens and the highest level observed was 15.8 /.tg/1. The hCG concentration in the same urine samples varied from 38 to 7362 /~g/1 (mean 2119 ~g/1). Serum levels of PP14 varied from 183 to 1828 # g / l (mean 879 ~g/1) and of hCG from 41 to 7331 ~ g / l (mean 3480/~g/l).
Discussion The biological action of PP14 is unclear and it is necessary to use a gentle method for the purification of PP14 in order to preserve its activity for such studies. Amniotic fluid PP14 is obviously the same protein as PP14 purified from the placenta (Bohn et al., 1982). This has been shown by immunological criteria (Riittinen et al., 1989) and by its N terminal protein sequence (Huhtala et al., 1987). The same sequence has also been deduced from cDNA sequence analysis (Julkunen et al., 1988). Mid-trimester amniotic fluid contains a high concentration of PP14 (Julkunen et al., 1985) and this appropriate starting material is abundantly available in units carrying out prenatal diagnosis from amniotic fluid. Compared with the previous purification methods from decidua (Sutcliffe et al., 1980; Westwood et al., 1988), endometrium (Bell, 1986) or placenta (Bohn et al., 1982) the purification method presented here, though simple, is much more rapid. The particular advantage of using monoclonal antibodies for the purification of PP14 is the high yield. It is possible to use the same antibody column several times. We have now obtained 5.6 mg of PP14 using the same column, and it is still quite effective. It is possible to purify larger quantities of PP14 by increasing the column size and the volume of the starting material. In addition to the smaller PP14 concentration in tissue, contamination with hCG has been a
problem in PP14 preparations purified from the placenta. The hCG concentration is low in amniotic fluid. Therefore it is not surprising that PP14 purified from amniotic fluid using the monoclonal antibody column contains only 0.00038% ( w / w ) of hCG. The minuscule hCG contamination of this purified PP14 is unlikely to interfere in any studies on the biological action(s) of PP14. Another advantage of using monoclonal antibodies is their specificity and abundance for use in assays, e.g., in RIA such as shown in this study. Unlike hCG, high concentrations of PP14 cannot be found in pregnancy urine at 7-10 weeks gestation in spite of the fact that the circulating levels of both hCG and PP14 are highest during this period. When pure PP14 was added to the urine, after 15 days storage at - 2 0 ° C, the measured value was 64-75% of the added PP14. Since the maximal storage time of the urine samples was 15 days low PP14 levels in urine cannot be explained by a complete loss of antigenicity. These results show that, while both PP14 and hCG are high in serum at the same stage of pregnancy, PP14 is mostly secreted into the amniotic fluid and less is detected in the urine than is the case with hCG.
Acknowledgements This study was supported by grants from Finnish Social Insurance Institution, Sigrid Jus61ius Foundation, and the Academy of Finland. I.V. was supported by a grant from the Finnish Cancer Research Fund and L.R. from Research and Science Foundation of Farmos. We thank Ms. Pirkko Selenius and Ms. Saija Roine for technical assistance.
References Bell, S.C. (19186) Purification of human secretory pregnancyassociated endometrial alpha-2-globulin (alpha-2-PEG) from cytosol of first trimester Pregnancy endometrium. Hum. Reprod. 1, 313-318. Bell, S.C. and Bohn, H. (1986) Immunochemicaland biochemical relationship between human pregnancy-associated secreted endometrial al- and a2-globulins (al- and a2-PEG)
90 and the soluble placental proteins 12 and 14 (PP12 and PP14). Placenta 7, 283-294. Bohn. H., Kraus, W. and Winckler, W. (1982) New soluble placental proteins: Their isolation, characterization, localization and quantification. Placenta 4 (suppl.), 67-81. Huhtala, M.-L, Sepp~l~, M., N~rv~nen, A., Palom~ki, P., Julkunen, M. and Bohn, H. (1987) Amino acid sequence homology between human placental protein 14 and /~lactoglobulins from various species. Endocrinology 120~ 2620-2622. Julkunen, M. (1986) Human decidua syntheses placental protein 14 (PPI4) in vitro. Acta Endocrinol. 112, 271-277. Julkunen, M., Rutanen, E.-M., Koskimies, A., Ranta, T., Bohn, H. and Sepp~.l~, M. (1985) Distribution of placental protein 14 in tissues and body fluids during pregnancy. Br J. Obstet. Gynaecol. 92, 1145-1151. Julkunen, M., Koistinen, R., SjOberg, J., Rutanen, E.-M., WahlstrOm, T. and Sepp~,l~, M. (1986a) Secretory endometrium synthesizes placental protein 14. Endocrinology 118, 1782-1786. Julkunen, M., Raikar, R.S., Joshi, S.G., Bohn, H. and Sepp~l~, M. (1986b) Placental protein 14 and progestagen-dependent endometrial protein are immunologically indistinguishable. Hum. Reprod. 1, 7-8. Julkunen, M., Sepp~l~ and M. J~nne, O.A. (1988) Complete amino acid sequence of human placental protein 14: A progesterone-regulated uterine protein homologous to /~lactoglobulins. Proc. Natl. Acad. Sci. U.S.A. 85, 8845-8849. Karonen, S.-L., M~rsky, P., Siren, M. and Senderling, U. (1975) An enzymatic solid-phase method for trace iodination of proteins and peptides with 12Siodine. Anal. Biochem. 67, 1-10.
K~Shler, G. and Milstein, C. (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256, 495-497., Riittinen, L., Stenman, U.,-H., Alfthan, H., SuikkarL A.-M.. Bohn, H. and Sepp~l~., M. (1989) Time-resolved immunofluorometric assay for placental protein 14. Clin. Chim. Acta 183, 115-124. Sepp~l~, M., ROnnberg, L., Karonen, S.-L. and Kauppila, A. (1987) Micronized oral progesterone increases the circulating level of endometrial secretory PP14//3 lactoglobulin homologue. Hum. Reprod. 2, 453-455. Stenman, U.-H., Alfthan, H., Ranta, T.~ Vartiainen, E., Jalkanen, J. and Sepp~l~, M. (1987) Serum levels of human chorionic gonadotropin in nonpregnant women and men are modulated by gonadotropin releasing hormone and sex steroids. J. Clin. Endocrinol. Metab. 64, 730-736. Sutcliffe, R.G., Bolton, A.E., Sharp, F., Nicholson, L.V.B. and MacKinnon, R. (1980) Purification of human alpha uterine protein. J. Reprod. Fert. 58, 435-442. Waites, G.T., Wood, P i . , Walker, R.A. and Bell, S.C. (1988) |mmunohistological localization of human endometrial secretory protein, 'pregnancy-associated endometrial a2" globulin' (a2-PEG), during the menstrual cycle. J. Reprod. Fert. 82, 665-672. Westwood, O.M.R,, Chapman, M.G., Totty, N., Philp, R., Bolton, A.E. and Lazarus, N.R. (1988) N-terminal sequence analysis of human placental protein 14, purified in high yield from decidual cytosol. J. Reprod. Fert. 82, 493 500.
