Physicochemical studies of pregnancy-specific beta 1-glycoprotein: unusual ultracentrifugal and circular dichroic properties

The molecular properties of native and modified pregnancy-specific beta 1-glycoprotein (SP1) from human placenta were evaluated by sedimentation equilibrium, gel electrophoresis, and circular dichroic measurements. Native SP1 contained 6.2% N-acetylneuraminic acid (NANA), 5.8% galactose (Gal), 13% N-acetylglucosamine, 6.5% mannose, and 1.1% fucose but no detectable N-acetylgalactosamine. Treatment with mixed exoglycosidases and alpha-mannosidase removed 79% of the carbohydrate including all of the NANA and Gal. The intensity of the circular dichroic spectrum of SP1 in the far ultraviolet was quite low with a positive maximum at 235 nm and a negative maximum at 215 nm. The 235-nm band was lost upon treatment with reducing agents or with guanidinium chloride (GdmCl), but not by treatment with neuraminidase. Treatment of SP1 with neuraminidase, or with mixed exoglycosidases and alpha-mannosidase, resulted in decreases of the apparent molecular weight obtained by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Neither exposure of SP1 to GdmCl nor its reduction and alkylation resulted in the appearance of subunits on sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The partial specific volume of SP1 determined experimentally by comparing sedimentation equilibrium profiles in H2O and D2O was 0.695 +/- 0.007 mL/g. The molecular weight of SP1 in 6 M GdmCl (in the presence or absence of reducing agents) by equilibrium sedimentation was 42 300 +/- 400. In the absence of denaturing agents, SP1 existed in the form of aggregates (at least as high as trimeric SP1) that dissociated only slowly upon dilution. The presence of these aggregates may contribute to the reported molecular heterogeneity of SP1.

Radioimmunoassay of placental protein 12: levels in amniotic fluid, cord blood, and serum of healthy adults, pregnant women, and patients with trophoblastic disease

A radioimmunoassay for placental protein 12 (PP12) is described and the levels in amniotic fluid, cord blood, and serum of nonpregnant individuals, pregnant women, and patients with trophoblastic disease are presented. During pregnancy, the highest PP12 levels were found at 22 to 23 weeks (mean +/- SD, 169 +/- 123 ng/ml), and there was a transient decline at 32 to 33 weeks (63 +/- 23 ng/ml). In amniotic fluid, the levels were 100 to 1,000 times higher than in maternal serum. In cord blood at birth, the values were of the same magnitude as in maternal serum. Also healthy nonpregnant women and men had PP12-like immunoreactivity in serum. Nonpregnant women (9 to 47 ng/ml) had higher levels than men (undetectable to 21 ng/ml). Elevated levels up to 84 ng/ml were occasionally observed in trophoblastic disease, both hydatidiform mole and choriocarcinoma, but they bore no correlation with the human chorionic gonadotropin levels. On the basis of these results PP12 is not a suitable marker for trophoblastic disease. PP12 values in normal pregnancy provide the basis for the evaluation of PP12 levels in abnormal pregnancy.

Placental protein 5 is related to blood coagulation and fibrinolytic systems

Previous studies have shown that placental protein 5 (PP5) forms complexes with heparin. In order to further elucidate the biological role of PP5 we studied the effect of plasmin and thrombin on the immunoreactivity of PP5, and the possible functional antiplasmin and antithrombin effects of purified PP5. Varying concentrations of plasmin and thrombin were added to pregnancy plasma, and the PP5 levels, measured by radioimmunoassay, were found to be elevated by 558% (plasmin and 48-87% (thrombin). Incubation of radiolabeled PP5 with plasmin resulted in the formation of radioactive fragments with smaller molecular weights. Functional studies using a chromogenic substrate confirmed that purified PP5 has an antiplasmin activity. An average increase of 15% was observed in the antiplasmin activity when 200 ng purified PP5 was added to 150 microliters of pregnancy serum. Thus, there are certain similarities between PP5 and antihrombin III. Both form complexes with heparin and have antiplasmin properties, and both were found to be heat labile. But, functional studies utilizing a chromogenic substrate failed to demonstrate any antithrombin III-like activity in the purified PP5 preparation that had antiplasmin activity. Our results show that the function of PP5 is related to the blood coagulation and fibrinolytic systems, at least through its inhibitory action on plasmin.

Placental protein 5 (PP5) in pregnancy and malignant disease: the influence of heparin binding

The molecular mass of immunoreactive placental protein 5 (PP5) in late pregnancy blood was studied by gel filtration. The predominant form of PP5 in serum was of molecular mass 37000, whereas in heparinised plasma the molecular mass ranged upwards from 50000, with the majority of material in high molecular mass forms, thought to arise by binding of PP5 to heparin. The effect of heparin on PP5 measurements appeared to be enhanced by a serum component related to pregnancy, since addition of heparin decreased the measured level of PP5 to a greater extent in pregnancy serum than in male serum to which purified PP5 had been added. The radioimmunoassay dose-response curve was shallower for PP5 in heparinised plasma than for purified PP5 and PP5 in serum, and this may explain previous low measurements and high recoveries after chromatography. PP5 produced by a hydatidiform mole behaved in a similar manner to the placental form; PP5 form a malignant breast tumour cytosol preparation did not.

Placental protein 5 (PP5)-like immunoreactivity in human urine

Placental protein 5 (PP5)-like immunoreactive material was detected in human male urine in small concentrations (17-60 pg/ml). Part of the PP5 immunoreactivity in the urine had a molecular weight of 36 000-42 500, which is the molecular weight of purified PP5 from the human placenta. In radioimmunoassay, serial dilutions of the 36 000-42 500 molecular weight material gave an inhibition curve parallel to that of the PP5 standard. The source of PP5-like immunoreactivity in the urine is not known. It may be related to urokinase, which coeluted with PP5-immunoreactive material in gel filtration.

Monitoring therapy in trophoblastic diseases by radioimmunoassay of pregnancy-specific beta 1-glycoprotein and the beta subunit of human chorionic gonadotropin

The beta-subunit of human chorionic gonadotrophin (hCG) and pregnancy-specific beta 1-glycoprotein (SP1) were measured simultaneously by specific radioimmunoassays in 473 serum samples from 43 patients with trophoblastic tumors. Both SP1 and hCG values were high in 16 untreated patients. 27 patients had been treated previously elsewhere, and their further treatment and monitoring were started simultaneously in our institute. Raised SP2 and hCG levels were found in 22 and 21 cases, respectively. Treatment was followed by a parallel decrease of both SP1 and hCG, but the hCG decline was more marked. Seronegativity was reached only after several months; SP1 was present in the serum for 60-150 days, and hCG could be found even after this time. Both parameters were sensitive indicators of residual tumor activity, calling for further therapy months before any clinical manifestation of recurrence. In three cases only one of the tumor makers could be found, which underlines the importance of the simultaneous measurement of hCG and SP1.

Ectopic production of placenta-"specific" tissue proteins (PP5 and PP11) by malignant breast tumors

With use of an enzyme-bridge immunoperoxidase (PAP) technique, an immunohistochemical localization of the two new placenta-"specific" tissue proteins, PP5 and PP11, was investigated in 16 cases of breast cancer. PP5 was localized in the cytoplasm of breast cancer cells in 62.5% of cases examined and PP11, in 46.7%. Thus PP5 and PP11 may be markers for breast cancer and the detection of these proteins in sera from breast cancer patients may be more reliable and useful in the detection and monitoring of breast cancer patients than the determination of SP1, PP10, or PP12, other pregnancy "specific" proteins.

Placental proteins (SP1, hCG, PP5) and alpha 2-PAG in trophoblastic diseases

455 serial serum samples from 41 patients with trophoblastic tumour were analysed. Pregnancy-specific beta-1 glycoprotein (SP1), human chorionic gonadotropin (hCG), and placental protein 5 (PP5) were measured by specific radioimmunoassay. Pregnancy-associated alpha 2-glycoprotein (alpha 2-PAG) was measured by electroimmunoassay. SP1 and hCG levels were high in 36 and 35 patients respectively; alpha 2-PAG concentration was also high in 29 patients. PP5 was found only in patients with a hydatidiform mole and it disappeared from serum within a few hours of evacuation. SP1 and hCG levels decreased during the course of the treatment but seronegativity was reached only after some months: hCG was still detectable after that and disappeared from the serum more slowly. Both parameters were sensitive indicators of residual tumour activity and indicated the need more effective treatment some months before clinical manifestations of recurrence became evident. In some patients only one of the tumour markers was present during follow-up, indicating the importance of careful monitoring including the measurement of serum SP1 and hCG levels. A positive correlation between alpha 2-PAG and hCG, SP1 and the clinical course was found only in 80% of patients a finding which demonstrates the limited value of alpha 2-PAG measurements.

Human seminal plasma contains a protein that shares physicochemical and immunochemical properties with placental protein 5 from the human placenta

Placental protein 5 (PP5)-like immunoreactive material was detected in human seminal plasma at high concentrations (32-1000 ng/ml). This was not due to interference with proteases or binding to seminal plasma proteins, since immunoreactivity was not affected by treatment with protease inhibitors, and incubation with seminal plasma of [125I]PP5 did not bring about any significant change in the elution pattern in gel filtration. Part of the PP5 immunoreactivity in seminal plasma had a molecular weight of 36,000-42,500 which is the molecular weight of purified PP5 from the human placenta. In RIa, serial dilutions of the 36,000-42,500 molecular weight material gave an inhibition curve parallel to that of the PP5 standard. The source of seminal plasma PP5-like material is not from the testes, as the levels in vasectomized men were similar to those in nonvasectomized men.

Immunohistochemical location of proteins related to pregnancy-specific beta 1-glycoprotein (SP1) and placental tissue proteins (PP5, PP8, PP9, PP10, PP11, PP12) in the afterbirth of Cynomolgus monkeys (Macaca fascicularis)

The occurrence and location of proteins antigenically related to human pregnancy specific beta 1-glycoprotein (SP1) and to human placental tissue proteins (PP5, PP8, PP9, PP10, PP11, PP12) were studied in the afterbirth (placenta, membranes, decidua, and umbilical cord) of cynomolgus monkeys with use of an immunoglobulin enzyme bridge (PAP) technique. The results were basically the same throughout pregnancy. The syncytiotrophoblast of villi showed clear positive stainings for SP1, PP8, PP9, and PP11, whereas the cytotrophoblast (villi) was found to show clear positive staining for PP9 only. The main sources of proteins related to SP1, PP9, and PP11 were the syncytiotrophoblast cells of the villi and PP8, the syncytiotrophoblast cells of villi as well as the chorionic trophoblast cells. PP12 was mainly located in histiocytes found in the decidua, umbilical cord, amnion, and chorion. There were only traces of PP5 and PP10 in the placentae of cynomolgus monkeys throughout gestation. As in human placentae some granulocyte-like blood cells in the intervillous space were strongly positive for PP8, PP10, and PP12. The findings in the afterbirth of cynomolgus monkeys were similar to those in the human; the monkey could thus serve as a model for the investigation of the new placental proteins.

Placental protein 5 (PP5) in complicated pregnancies

Maternal serum PP5 concentrations were assayed in pregnancies complicated by intrauterine growth retardation, pre-eclampsia, diabetes mellitus, placental abruption and twins. Elevated PP5 levels were noted in twin pregnancies and to a lesser extent in diabetes and pre-eclampsia. Measurement of PP5 was of no value in the detection of fetal growth retardation. The great majority of PP5 levels above the normal range were associated with a placental abnormality.

Placental protein 5 (PP5): development of a radioimmunoassay and measurement of circulating levels in normal pregnancy

A specific and sensitive, double antibody, solid phase radioimmunoassay was developed for measurement of placental protein 5 (PP5) in serum. A sensitivity limit of 0.8 microgram/1 was achieved by pre-incubation with antiserum and use of a small assay volume. Single serum samples from 244 retrospectively proven normal pregnancies were used to construct a normal range of PP5 concentrations. PP5 was detectable at week 7 and rose to reach a maximum mean value of 54 microgram/1 at weeks 36-37. Diurnal variation of PP5 concentration was similar to that encountered from day to day, and slightly greater than variation due to the radioimmunoassay. The decline of serum PP5 concentration post partum was initially rapid (half-life 5-14 minutes), but slowed after 0.5-1 hours; immunoreactive PP5 was still detectable seven days after delivery.

Cultured human syncytiotrophoblasts synthesize pregnancy-specific beta-1-glycoprotein (SP1)

The synthesis of pregnancy-specific beta-1-glycoprotein (SP1) was studied in human syncytiotrophoblast cultures from first trimester placentas. Incorporation of 35S-methionine into SP1 was demonstrated by immunoprecipitation and polyacrylamide gel electrophoresis. In double diffusion the precipitation line of neosynthesized SP1 gave a reaction of complete immunological identity with maternal serum SP1. By indirect immunofluorescence intracellular SP1 was localised in the perinuclear region of the cells. The average rate of SP1 release into culture fluid was 600 ng/mg protein in 24 hours. On a molar basis the release of SP1 was 10 times less than that of human chorionic gonadotrophin (hCG). Our results demonstrate, for the first time, that cultured human syncytiotrophoblasts synthesis SP1 immunologically indistinguishable from placental SP1.

The effect of protamine on serum levels of placental protein 5 (PP5) in normal and abnormal pregnancy: a possible relation to coagulation abnormalities

The levels of placental protein 5 (PP5) in pregnancy serum show an apparent increase after the addition of protamine sulphate. The difference of the apparent concentrations of serum PP5 in the presence and absence of protamine (delta PP5) was determined in 345 normal and abnormal pregnancies. A positive delta PP5 was less frequent in diabetic pregnancy (45 per cent) and pre-eclampsia (57 per cent) than in normal pregnancy (80 per cent). These findings provide further evidence that PP5 may be involved in the coagulation system, and in particular may relate to abnormal coagulation processes at the placental site.

Possible synthesis of pregnancy-'specific' beta-1-glycoprotein (sp1) and placenta-'specific' tissue proteins (PP10,PP12) by human and cynomolgus monkey leukocytes

Using an enzyme-bridge immunoperoxidase (PAP) technique, it was revealed that pregnancy- and/or placenta-specific proteins, SP1, PP10 and PP12 were localized in human and cynomolgus monkey (Macaca fascicularis) polymorphonuclear neutrophils, and PP10, in mononuclear phagocytes. There was no difference in staining results, irrespective of whether the specimens were from human or monkey males, or pregnant or nonpregnant females. The results suggest that SP1, PP10 and PP12 may be synthesized by polymorphonuclear neutrophils and/or mononuclear phagocytes and could represent markers not only for trophoblasts and some kinds of malignant neoplasms but also for normal neutrophils and/or monocytes.

Radioimmunoassay of SP1 (pregnancy-specific beta1-glycoprotein) in maternal blood and in amniotic fluid normal and pathologic pregnancies

Sp1, the pregnancy-specific beta 1-glycoprotein, was studied in normal and pathologic pregnancies. We developed a highly specific and sensitive double-antibody-radioimmunoassay by radioiodination of purified placental SP1. This RIA allowed the estimation of SP1 concentrations as low as 2 ng/ml. In a collective of 227 women with normal pregnancies we established the normal distribution curve in maternal plasma from the fifth week of gestation to term. The median value rose steadily from 3 microgram/ml in the 8th week to 140 microgram/ml in the 36th week when a plateau was formed. In more than 400 patients with pregnancies complicated by a variety of pathologic disorders the SP1 levels were controlled by either single assays or serial estimations throughout pregnancy and were compared with the normal distribution range. SP1 was also determined in about 200 samples of amniotic fluid gained by amniocentesis and during parturition of normal pregnant women from the 13th gestational week until term. The normal range was established up to the 20th w.o.p. The concentrations rose from below 0.2 microgram/ml in early pregnancy to 3 microgram/ml and generally amounted to approximately 1% of the respective serum value. Pathologic cases with diverse chromosomal anomalies, Rh-incompatibility, anencephaly, hydramnios and other abnormal conditions were examined. From these only twin-pregnancies with slightly elevated levels and cases with fetal trisomies with reduced SP1 concentrations showed aberrations from the normal distribution. The estimation of serum concentrations in mothers with diabetes or Rh-incompatibility were not significantly different from the normal collective. In diabetes a characteristic course of the follow-up curves was observed. Abortion in early pregnancy was frequently but not always indicated by reduced SP1 values. Threatened abortion with subsequent continuation of pregnancy exhibited SP1 values scattered within the normal range. Since the radioimmunological determination of SP1 is possible in the early stage of gestation (from week 8) it may serve as a useful tool for prediction at times when the determination of placental lactogen is not yet possible. In pregnancies with "small-for-date babies" the correlation between SP1 in maternal plasma and fetal growth retardation was reflected in a pronounced tendency to low SP1 levels. Serial determinations of SP1 in the serum of women with EPH-gestosis were compared with the corresponding HPL determinations and showed the equality of SP1 concerning the assessment of the placental function.

[New placental protein (PP15) with immunosuppressive properties (author's transl)]

The isolation and characterization of a new placental protein (PP15) which has immunosuppressive properties are described. The protein was purified from extracts of human term placentae by rivanol and ammonium sulfate fractionation, gel filtration, ion exchange chromatography, preparative zone electrophoresis, and chromatography on hydroxyapatite. PP15 was found to have a sedimentation coèfficient of 2.9 S and a molecular weight of 30,700 daltons as determined by ultracentrifugation; its molecules apparently are composed of two identical subunits which are held together by non-covalent bonds. The electrophorectic mobility of PP15 corresponds to that of albumin. PP15 is a glycoprotein and contains 3.3% carbohydrates (hexoses 2.8%, hexosamines 0.3%, sialic acid 0.2%). The amino acid composition of this protein was also determined: the most abundant amino acids in the peptide chain were found to be glutamic acid, aspartic acid, isoleucine, and leucine. PP15 has immunosuprressive properties: on testing its effect on the lymphoycte transformation in the MLC-test in vitro a significant inhibitory activity could be demonstrated.

Immunohistochemical location of placental proteins (PP8, 9, 10, 11, 12) in human term placentae

By using an immunoglobulin enzyme bridge (PAP) technique, the location of two ubiquitous tissue proteins (PP8, PP9) and three placental specific tissue proteins (PP10, PP11, PP12) was investigated in ethanol/acetic acid-fixed paraffin-embedded human term placentae. PP8 was found mainly in both the cytoplasm and the nucleus of trophoblast cells (chorion) and syncytiotrophoblast (villi). PP9 was found in the cytoplasm of trophoblast cells (chorion), the fibrous part of interstitial connective tissues (villi), and the cytoplasm of histiocytes (villi, amnion, and decidua). PP10 was found in the cytoplasm of epithelium (amnion) and in both the cytoplasm and the nucleus of syncytiotrophoblast (villi). PP12 was found in the cytoplasm of histiocytes in decidua, amnion, and chorion and the cytoplasm of syncytiotrophoblast (villi). In addition, stains for PP8, PP10, and PP12 were taken up by the cytoplasm of histiocytes found in amnion, villi, the intervillous space, and decidua, in contrast to PP11, which was found to be specific to the cytoplasm of trophoblast cells (chorion and villi). Possible clinical applications of these findings are discussed.

Ectopic synthesis of pregnancy specific beta 1-glycoprotein (SP1) and placental specific tissue proteins (PP5, PP10, PP11, PP12) in nontrophoblastic malignant tumours. Possible markers in oncology

By using an enzyme-bridge immunoperoxidase (PAP) technique, localization of so-called pregnancy-specific beta 1-glycoprotein (SP1) and placental-specific tissue proteins (PP5, PP10, PP11, PP12) was investigated in 19 cases of breast cancer, 24 cases of testicular malignant tumours, 12 cases of gastric cancer and some cases of other malignant tumours. These five glycoproteins isolated from human term placentae and characterized by Bohn are localised mainly in the cytoplasm, or nucleus of syncytiotrophoblast and appear to be specific for the trophoblast. But to a certain percentage these proteins could also be detected in the cytoplasma of malignant cells: In breast cancer, SP1 was present in 52.6% of cases, PP5 in 63.2%, PP10 in 68.4%, PP11 in 55.6% and PP12 in 31.6%. In testicular malignant tumours the detection rates of these proteins varied from 20.8 to 75.0% and in gastric cancer from 41.7 to 66.7%. In total 50% of the tumours tested were SP1-positive, 58.3% PP5-positive, 55.6% PP10-positive, 38.0% PP11-positive and 31.9% PP12-positive. In addition, it was found that mononuclear histiocytes showed a strong cytoplasmic staining for PP10 and PP12 and that a few other non-cancerous cells (i.e. striated cells in gastric metaplasia, gastric polyps etc.) stained for PP5, PP10, PP11 and/or PP12. All control sections were negative in malignant cells as well as in other tissue cells. This study confirms the reports that some kinds of malignant tumours produce SP1 and indicates that the ectopic production of proteins normally produced by the trophoblast is a common finding in malignant tumours. It furthermore suggests that such proteins may be useful as markers in monitoring patients with malignant diseases. The possible mechanisms of ectopic production of these inappropriate proteins in malignant tumours are discussed.