| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Immunology |
Department of Medical Microbiology and Immunology,3 Medical School, Pécs University, H-7624 Pécs, Hungary
Department of Obstetrics and Gynecology,4 County Hospital, H-7623 Pécs, Hungary
Reproductive and Tumor Immunology Research Group of the Hungarian Academy of Sciences,5 H-7643 Pécs, Hungary
INTERCELL AG,6 A-1030 Vienna, Austria
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
diagnostic test, immunology, PIBF, pregnancy, progesterone
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
PIBF plays a role in the maintenance of pregnancy, most likely by inhibiting NK lymphocytes [5]. By manipulating the intracellular concentration of PIBF in vitro, one can modulate the killing activity of human peripheral blood NK cells [6]. Neutralization of endogenous PIBF activity in pregnant mice by anti-PIBF antibody results in a 70% reduction in the number of viable fetuses, and this is associated with an increased splenic NK activity [5]. Ninety percent of pregnancy loss induced by anti-PIBF administration is corrected by treatment of the pregnant animals with anti-NK antibodies [5]. These data suggest that, at least in mice, PIBF contributes to the success of pregnancy and that the major part of its pregnancy-protective effect is due to its NK inhibitory activity.
The second main mechanism of action of PIBF during pregnancy is the induction of TH2 dominant cytokine response [7]. The secreted PIBF facilitates the production of IL-3, IL-4, and IL-10, while it suppresses TH1 cytokines, such as IL-12 and IFN-
both in vitro and in vivo [6, 7]. Neutralization of PIBF by specific antibodies results in a shift toward TH1 in vivo, which is also a characteristic of failed pregnancies [6, 8]. The effect of PIBF on humoral immune responses includes the induction of asymmetric antibody production [9, 10]. This population of antibodies, owing to the presence of a mannose-rich oligosaccharide residue on one of the Fab arms of the molecule, does not precipitate; however, these antibodies might have a blocking effect. The percent of asymmetric IgG was significantly higher in supernatants of hybridoma cells cultured in the presence of PIBF than in those cultured in the absence of PIBF, and there was a positive relationship between asymmetric antibody content of pregnancy sera and PIBF expression on lymphocytes from the same women. Furthermore, blocking of progesterone receptors by RU 486 or neutralizing endogenous PIBF activity by specific anti-PIBF antibodies significantly reduced the production of asymmetric antibodies in pregnant mice [11].
The abovementioned biological effects of PIBF suggest that it might contribute to the maintenance of a normal pregnancy. Therefore, alterations of PIBF concentrations in biological fluids might indicate the well being of the fetus and the prognosis of pregnancy. Because PIBF is a secreted molecule, it might be present in biological fluids. PIBF has been cloned and sequenced, and now the recombinant human PIBF is available [12].
The aim of this study was to develop an ELISA test for the determination of PIBF concentration in biological samples of healthy pregnant women as well as in those of women with pathologic pregnancies.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A total of 582 human urine samples were tested. The study was approved by the local ethical committee and an informed consent was obtained in each case. Fifty milliliters of urine samples were collected in the morning and immediately transferred to the laboratory, where they were either immediately tested or aliquoted and frozen. Samples from 319 healthy pregnant and 177 pathological pregnant women (16–42 years old) between the 7th and 41st week of gestation were tested for PIBF concentration by ELISA. Distribution of patients with pathological pregnancies are described in Table 1. The diagnosis of threatened abortion was based on the presence of regular uterine contractions and vaginal bleeding before the 24th week, whereas that of threatened preterm delivery on regular uterine contractions and progressing cervical dilatation between the 24th and 36th weeks of gestation. Preeclampsia was defined by repeatedly high blood pressure (>140/90 mm Hg) with or without proteinuria (>0.3 g protein/24 h) after the 20th week of gestation. Dysmaturity was diagnosed by biometrical ultrasound, based on measurements of biparietal diameter, abdominal circumference, and femur length (<10 percentile of the corresponding gestational age). The diagnosis was verified and corrected for sex by neonatological examination after birth. Polyhydramnion was diagnosed if the largest diameter of the amniotic fluid window was found to be larger than 4 cm by ultrasonography.
|
The mean + 2SD of PIBF concentrations in urine samples of 86 healthy nonpregnant volunteers was considered as a threshold level for control (nonpregnant) values.
Antibodies Recognizing Different Forms of the Recombinant Human PIBF
Polyclonal antibodies were generated in our laboratory by immunizing rabbits with a secreted 34-kDa lymphocyte PIBF and the 48-kDa N-terminal part of the human recombinant PIBF [12]. The antibody titers were determined by ELISA using the recombinant human PIBF protein as the antigen and the IgG from immune sera was affinity purified on protein-A or protein-G columns (AP Hungary Ltd., Budapest, Hungary). An aliquot of the antibody was absorbed with the 48-kDa N-terminal part of the recombinant PIBF to be used for testing the specificity of binding on Western blots.
Placental Lysate
One hundred milligrams of fresh human placental tissue was homogenized in 1 ml ice-cold lysis buffer with a glass-potter homogenizer. The homogenate was filtered, lysed on ice (in 100 mM HEPES, 2% Triton-X, 300 mM NaCl, 4 mM EDTA, 0.4 mM phenylmethylsufonylfluoride) for 30 min, and pulse-sonicated three times (3 x 30 sec on ice). The solution was centrifuged at 15 000 rpm for 15 min at 4°C, and then the supernatant was collected, immediately frozen, and kept at –80°C until used.
Western Blotting
For the detection of PIBF in lysates of placental cells, amniotic fluid, and urine of pregnant women, 10 µg of total protein per sample was separated by 12% SDS-PAGE and transferred to a Hybond ECL membrane (Amersham Biosciences, Budapest, Hungary). The blots were blocked with 5% nonfat dry milk in PBS for 1 h and incubated with anti-human 48-kDa recombinant PIBF antibody (1:500) in PBS containing 1% nonfat dry milk. After the membrane was washed three times with PBS, horseradish peroxidase-coupled anti-rabbit IgG (1:5000; DAKO, Budapest, Hungary) was added and the signal was detected with the ECL Plus Western blotting detection system (Izinta, Budapest, Hungary) according to the manufacturer's instructions.
ELISA Test
Urine samples were either used immediately or stored at –80°C until tested in a competitive ELISA assay, which was performed according to the following. During overnight incubation at 4°C, 96-well microtiter plates were coated either with anti-human 48-kDa recombinant PIBF IgG (100 µl/well of 2 µg/ml) in 50 mM carbonate buffer, pH 9.6 (plate 1), or with human 48-kDa recombinant PIBF (100 µl of 0.5 µg/ml) in 0.5 M Tris buffer, pH 6.5 (plate 2). For generating a standard curve, recombinant PIBF (1000 to 0.1 ng/ml) in logarithmic dilutions in 0.5 M phosphate buffer, pH 7.3–7.4, was incubated with a standard amount of biotin-labeled antirecombinant PIBF IgG (400 ng) for 60 min at 37°C. Urine samples were diluted 1:2.5 and 1:5 and incubated with 400 ng biotin-labeled antirecombinant PIBF IgG in 0.5 M PBS for 60 min at 37°C before being added to ELISA plate 1. During the 1-h incubation at 37°C, nonspecific binding sites on plate 2 (coated with human recombinant PIBF) were blocked with 200 µl of 0.1% BSA, 0.5% gelatin in PBS-Tween. After this incubation step, 100 µl standard solutions or the urine samples were transferred from plate 1 to plate 2 and incubated for 1 h at 37°C. After washing the plates three times with PBS Tween, 100 µl of 1:1000 diluted HRPO (horse radish peroxidase)-conjugated Streptavidine (AP Hungary Ltd., Budapest, Hungary) in 0.1% BSA PBS-Tween was added and plates were incubated for 30 min at 37°C. The reaction was developed by adding the substrate OPD (orthophenylene-diamine; Sigma, Hungary) and measured at 495 nm. The reproducibility of the test was evaluated in 10 independent experiments and the results are shown in Table 2.
|
Statistics
The two-tailed Student t-test was used if two groups were compared, while analysis of variance was used for multiple comparisons over time or among several treatment groups for statistical evaluation of the data. Differences were considered significant if the P value was equal to or less than 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Urine samples from healthy pregnant women and from women at risk for premature pregnancy termination were analyzed by Western blotting using polyclonal rabbit IgG generated with human 48-kDa recombinant PIBF. Several PIBF forms with different electrophoretic mobility were detected, with estimated molecular sizes of 65, 55, 34, 22, 18, and 10 kDa. The 34-kDa form and the occasional smaller forms were present in urine of healthy pregnant women (37th week of gestation) and to a lesser degree in samples of women postterm. Notably, these isoforms were missing or underrepresented in samples of women in labor, of those women showing symptoms of threatened abortion or threatened preterm delivery, toxemia, or women bearing small-for-date babies (Fig. 1, A and B). The specificity of detection was demonstrated by the lack of signal with the immune serum after depletion of the specific antibodies with recombinant PIBF (Fig. 1C). Based on this experiment, the 55-kDa band invariably detected with all samples was due to nonspecific reaction.
|
To determine whether similar PIBF forms were present in pregnancy tissues, term placenta and amniotic fluid were also analyzed by Western blotting. Placental lysates and amniotic fluid samples displayed anti-PIBF reactivities that were similar to those found in urine. Interestingly, the 34-kDa form was present only in the placenta and was absent from the amniotic fluid (Fig. 2).
|
PIBF Concentrations in Urine Samples from Nonpregnant Individuals and Pregnant Women
To define normal, nonpregnancy PIBF levels, urinary PIBF concentrations of 86 healthy, nonpregnant individuals were determined by ELISA. It was assumed the concentrations follow a log normal distribution. Due to the unreliability of the assay in the low concentration ranges, values less than 0.5 were neglected. The analysis resulted in a mean of 4.43 ng/ml with a SD of 2.23, SEM 1.32, and 95% confidence interval = (2.84 ng/ml, 6.89 ng/ml). To reduce the number of false-positive cases, 14 ng/ml, a value higher than the 95% confidence interval, was considered as a threshold level between control and normal pregnancy values.
The ELISA assay was reproducible, based on the intra- and interassay variation that was within 10% in the medium- and high-concentration ranges. At very low concentrations (less than 1 ng/ml), the coefficient of variation % was high (Table 2).
Altogether, 496 urine samples from pregnant women were tested. Three hundred nineteen of the women had normal pregnancies, while the others showed clinical symptoms of threatened abortion, threatened preterm delivery, or preeclampsia. We found that, during normal uneventful pregnancy, the concentration of PIBF continuously increased from the 7th to the 37th gestational weeks until reaching a maximum concentration of 120 ng/ml. After the 41st week of pregnancy, PIBF concentrations dramatically decreased. In patients with a diagnosis of threatened premature pregnancy termination, urinary PIBF levels failed to increase during pregnancy (Fig. 3).
|
The onset of labor was also predictable on the basis of this test; however, the predictive value of PIBF measurement depended on the interval between sampling and the onset of labor. In samples that were taken within 2 days before labor started, PIBF concentrations were significantly lower than in those obtained 7–16 days before labor (Fig. 4).
|
In 12 women with term deliveries of small-for-date babies, PIBF concentrations were significantly (P < 0.05) lower than in normal pregnancy (Fig. 5). In four cases of polyhydramnion, we detected very high PIBF levels (983.6 ng/ml); however, due to the small number of cases and the high individual variations, the difference from normal pregnancy values was not statistically significant (data not shown).
|
PIBF Concentrations in Urine Samples from Pregnant Women at Risk for Premature Pregnancy Termination
To assess the predictive value of the test, we determined PIBF concentrations in samples from women who were at risk for premature pregnancy termination based on several clinical signs, such as bleeding, regular uterine contractions, and toxemia)—data were analyzed both prospectively (with regard to the clinical diagnosis) and retrospectively (in view of delivery data). We found that PIBF concentrations were significantly higher in urine samples of women with clinical diagnoses of normal pregnancy compared with women with diagnoses of pathological pregnancies. The percentage of samples with higher than nonpregnant control PIBF concentrations was also considerably higher in the former group (Fig. 6, A and B). Seventy percent of women with a diagnosis of normal pregnancy and 80% of those who delivered at term demonstrated higher than threshold PIBF levels. On the other hand, only 4% of patients with a diagnosis of threatened premature pregnancy termination and 10% of those who in fact delivered preterm had higher than normal urinary PIBF concentrations (Fig. 6).
|
PIBF Concentrations in Urine Samples from Pregnant Women with Preeclampsia
In urine of patient with toxemia, the PIBF concentrations were significantly reduced compared with healthy pregnant women. However, PIBF concentrations depended on the number of symptoms presented. If hypertension was the only symptom, urinary PIBF concentrations were similar to those in normal pregnancies of similar gestational ages. It was in contrast with patients with two or more symptoms of toxemia who displayed urinary PIBF levels significantly lower than in healthy pregnant women of the corresponding gestational ages (Fig. 7).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In urine samples of pregnant women, we detected several PIBF forms of different molecular weights, among them a 34-kDa protein specifically reacting with antibodies generated with recombinant human PIBF. Notably, this isoform was only present in urine of healthy pregnant women but not in those of women showing symptoms of threatened abortion or threatened preterm delivery.
A search in the cDNA databases revealed that PIBF is present in embryonic tissues as well as in the placenta; thus, the placenta and/or the embryo might express some PIBF isoforms. Indeed, our Western blot analysis showed that both term placental lysates and amniotic fluid contain a 65-kDa PIBF form, whereas the 34-kDa molecule was absent from the amniotic fluid. This might be due to differences in processing or transport of the protein to the various compartments or, alternatively, the 34-kDa form PIBF protein might be secreted by the maternal, but not by the fetal, side. This latter possibility is in line with previous findings demonstrating that PIBF is secreted by activated maternal lymphocytes [1, 2].
The origin of the different PIBF forms in urine is not defined yet. Recently, we reported that the PIBF cDNA that we identified during cloning of the gene encodes a 90-kDa protein. In resting cells, the full length PIBF has perinuclear localization. Following activation, several smaller molecular weight forms are secreted [12]; thus, PIBF might appear in the sera and, considering the small molecular mass of certain isoforms, also in the urine. Moreover, this heterogeneity can be the result of degradation or processing of nascent PIBF protein, as well as the reflection of the complex alternative mRNA splicing pattern detected in various cells and tissues [20].
To detect PIBF quantitatively, we designed an ELISA test using the 48-kDa N-terminal part of the recombinant human PIBF as an antigen and polyclonal anti-human 48-kDa recombinant PIBF antibodies. The PIBF isoform detected by the ELISA has not been identified.
Samples from healthy nonpregnant individuals were used for determining the range of the normal nonpregnancy levels, and the mean + 2SD was considered as the threshold level between control and pregnancy values. To evaluate the predictive potential of PIBF measurement, PIBF concentrations from women with clinically diagnosed normal and pathological pregnancies were analyzed retrospectively, in view of pregnancy outcome. Eighty percent of the women with a normal, uneventful pregnancy and 10% of those whose pregnancies ended in miscarriage or preterm labor had higher PIBF concentrations than control threshold. The sensitivity (defined as the ability to correctly identify those who will deliver preterm) and specificity (the ability to correctly identify those who will not deliver preterm) of the test for predicting pregnancy failure is 90% and 80%, respectively. These data suggest that low PIBF values indicate the onset of spontaneous pregnancy termination. PIBF concentrations in urine of women who were close to labor were analyzed with regard to the interval between sampling and the onset of labor. The percentage of cases with higher than nonpregnancy urinary PIBF levels increased with the interval between sampling and the onset of labor.
Women with toxemia had lower PIBF values than healthy pregnant women. Because PIBF favors a TH2 cytokine response, these women should have a relative TH1 dominance. Rein et al. [21] reported that trophoblast cells from preeclamptic women produce significantly less IL-10 in the third trimester of pregnancy than those from healthy pregnant women, and an excessive TH1 activity has been associated with toxemia [22]. Several studies have shown that the clinical severity of preeclampsia is related to the severity of cytokine abnormalities [23]. PIBF concentrations in urine of toxemic women were related to the clinical symptoms. PIBF levels of women demonstrating hypertension only did not differ from those of healthy pregnant women. In contrast with this, only 33% of women with two or more symptoms had higher than threshold levels. This is in line with earlier observations of Varga et al. [24], who could not demonstrate an increased peripheral NK activity in the group of preeclamptic patients with a single symptom (hypertension), whereas lymphocytes of preeclamptic women with at least two symptoms showed a significantly increased NK activity. PIBF inhibits NK activity both in vitro and in vivo [2, 5], and the lack of PIBF results in an increased NK activity [25].
All the women bearing small-for-date babies had low PIBF values. This suggests that PIBF-related mechanisms relevant for threatened abortion and preeclampsia might play a role in the development of intrauterine growth retardation (IUGR). Bartha et al. [26] have shown an association between increased TNF
levels and IUGR. In our hands, treatment of pregnant mice with neutralizing anti-PIBF antibodies resulted in high NK activity of spleen cells and elevated serum and placental TNF levels that was associated with increased resorption rates of embryos. Both TNF levels and resorption rates were corrected by simultaneous PIBF administration. In vitro data from studies with lymphocyte cultures suggest that PIBF counteracts the cytotoxic action of TNF but does not interfere with its production [27].
To test whether the low values in samples from women with threatened abortion or threatened preterm delivery or those of preeclamptic women could have been due to interference with substance(s) in these samples, high- and low-concentration samples were mixed and PIBF concentration in these mixtures was determined. It is unlikely that high protein concentration or potential pathology-specific substances interfere with the test because mixing high-concentration samples with low-concentration ones still gave high values.
An interesting extension of the use of this assay is its application for cancer diagnosis. Based on our discovery that PIBF is overexpressed by certain human primary tumors and transformed cell lines and, moreover, is secreted by MCF-7 human breast cancer cells [12, 20], we hypothesized that high PIBF levels in body fluids can be indicative for the presence of malignant tissues. Indeed, this assay described here detected elevated PIBF levels in cancer patients that were correlated with tumor mass and corrected by the removal of the tumor (unpublished results).
In conclusion, the assay developed and used in this study has not been designed to serve as an alternative for hCG or progesterone determination. Insufficient information on PIBF levels in early pregnancy does not allow a conclusion of whether the test is sensitive enough to indicate the establishment of pregnancy. On the other hand, our data suggest that this ELISA test is a useful tool for sensitive determination of PIBF concentration in body fluids. The concentration of PIBF correlates with the positive or negative outcome of pregnancy; furthermore, premature pregnancy termination is predictable by lower than normal pregnancy PIBF values, suggesting that this method predicts problems of immunological origin and can contribute to the diagnosis of preterm pregnancy termination due to immune pathology.
|
FOOTNOTES |
|---|
2 Correspondence: Júlia Szekeres-Barthó, Department of Medical Microbiology and Immunology, Medical School, Pécs University, 12 Szigeti Str. H-7624 Pécs. Hungary. FAX: 003672 536253, szjuli{at}main.pote.hu
Received: 31 March 2004.
First decision: 29 April 2004.
Accepted: 15 July 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. G Lea and O. Sandra Immunoendocrine aspects of endometrial function and implantation Reproduction, September 1, 2007; 134(3): 389 - 404. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
