Cytokine imbalance in pregnancies with fetal chromosomal abnormalities

F. Vesce1,4, C. Scapoli2, G. Giovannini3, L. Tralli1, G. Gotti1, A. Valerio1 and A. Piffanelli3

1 1Department of Biomedical Sciences and Advanced Therapy, Section of Obstetrics and Gynaecology, 2 Department of Biology, Section of Evolutionary Biology and 3 Department of Clinical and Experimental Medicine, Section of Nuclear Medicine, University of Ferrara, Ferrara, Italy


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: The aim of the present study is to investigate the levels of some of the cytokines which may be involved in the mechanisms leading to the impairment of placental perfusion and to the onset of uterine contractions in pregnancies with fetal genetic abnormalities compared with controls. METHODS: The amniotic fluid and maternal plasma levels of interleukin-6, interleukin-8 and tumour necrosis factor-ß in patients with fetal chromosomal abnormalities were measured, as well as in euploid pregnancies in the seventh week of gestation. RESULTS: An increase of interleukin-6 (P = 0.034) and a decrease of interleukin-8 (P <= 0.0001) in amniotic fluid, and a decrease of interleukin-6 in the maternal plasma (P = 0.026) was shown in pregnancies with fetal chromosomal abnormalities. A positive correlation was observed between amniotic interleukin-8 and serum interleukin-6 in the presence of fetal aneuploidy (P < 0.006). CONCLUSION: Further investigations of cytokine imbalance in pregnancies with poor outcome as a consequence of genetic disorders rather that infection is warranted.

Key words: amniotic fluid/aneuploidy/cytokines/fetal chromosomal abnormalities/prenatal diagnosis


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Interleukin-6 (IL-6), interleukin-8 (IL-8) and tumour necrosis factors (TNF) are pro-inflammatory cytokines produced by several tissues (Akira et al., 1993Go; Baggiolini and Dewlad, 1993Go; Elliott et al., 1998Go; Daher et al., 1999Go) upon the stimulus of a number of factors, among which are membrane lipopolisaccarids of gram negative bacteria, viruses, and several cytokines (Akira et al., 1993Go; Baggiolini and Dewlad, 1993Go; Daher et al., 1999Go); their action is directed towards either myeloid or non-myeloid cellular targets (Akira et al., 1993Go; Baggiolini and Dewlad, 1993Go; Daher et al., 1999Go). During pregnancy these cytokines are involved in different ways in the regulation of the mechanisms of implantation and placentation, fetal maturation and uterine contraction (Akira et al., 1993Go; Cherouny and Pankuch, 1993Go; Dudley and Edwin, 1997Go; Rehbock and Chondromatidou, 1997Go; Tanaka and Narahara, 1998Go; Elliott et al., 1998Go; Daher et al., 1999Go). During physiological labour an increase in their amniotic and maternal serum level is related to the activation of the arachidonic acid pathway and prostaglandin formation (Steinborn and Gunes, 1996Go; Greig and Murtha, 1997Go; Daher et al., 1999Go; Furuta and Yamada, 2000Go). Furthermore, their concentration at the level of the amniotic compartment, gestational tissue or maternal serum is significantly higher during premature labour, whether in the presence or in the absence of infection, and is associated with increased fetal morbidity and mortality (Cherouny and Pankuch, 1993Go; Steinborn and Gunes, 1996Go; Greig and Murtha, 1997Go; Dudley and Edwin, 1997Go; Rehbock and Chondromatidou, 1997Go; Tanaka and Narahara, 1998Go; Murtha and Grieg, 1998; Hsu and Meaddough, 1998Go; Daher et al., 1999Go).

It is known that ~15% of human pregnancies end in spontaneous abortion before 12 weeks gestation. However, the percentage of pregnancy loss is much higher in the presence of aneuploidy (Boué et al., 1985Go). It can be thought that the chromosomal abnormality itself represents the cause of abortion, but the pathogenic mechanism is unknown. In our attempts to understand the causes of pregnancy loss, we felt that these are not different in euploidy compared with aneuploidy, they are simply much more frequent in the latter, possibly as a consequence of an imbalance of the mediators of placental perfusion and uterine contraction related to a genetic basis. The levels of the cytokines relevant to the establishment and maintenance of a viable pregnancy in the presence of fetal chromosomal abnormalities has not been reported. However, in the presence of fetal Down's syndrome, an increased amniotic level of TGF-ß has been found and this has been proposed as a possible additional biochemical index for the prenatal detection of the disease (Bromage et al., 2000Go).

Different behaviour of these mediators in the fetal compartment could represent the mechanism by which, in the absence of infection, the human fetus itself participates in its own demise. In this context, aneuploid pregnancies could be considered as a natural study model for a better understanding of the general causes of pregnancy loss.

On the basis of the above considerations, the present study assessed the amniotic and maternal serum levels of IL-6, IL-8 and TNF-ß in euploid and aneuploid pregnancies.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
IL-6, IL-8 and TNF-ß levels were measured in 123 amniotic fluids and corresponding maternal sera obtained from single pregnancies in women undergoing routine genetic amniocentesis (17th gestational week) at the Section of Obstetrics and Gynaecology of the University of Ferrara.

The study population was selected among 1082 amniocentesis performed from 1994 to 1998. All the observed 31 cases with chromosomal abnormalities (2.9%) were included, while 92 control cases were randomly chosen among euploid pregnancies. The pathological sample consisted of 15 aneuploidies, representing 1.4% of the study population, and 16 minor chromosomal abnormalities (1.5%). Average maternal age was 35.1 years in the control group, and 33.4 in the group with chromosomal abnormalities. Indication to amniocentesis in the control group was represented by maternal age over 35 years in 60 cases, positive triple test in 24, history of recurrent abortion in two, previous fetal Down's syndrome in three, assumption of teratogenic agents in two, parental chromosomal translocation in one, previous fetal malformation in two. In the pathological group there were 14 cases over 35 years, six with positive triple test, three with mosaicism at the chorionic biopsy performed for maternal age, three with ultrasound abnormalities, four with history of recurrent abortion and one with previous fetal malformation.

All amniocenteses were performed using ultrasonographic guidance. The amniotic fluid was stored within 6 h at –20°C, in the Laboratory of Nuclear Medicine, after the removal of particulate materials by centrifugation (800 g for 10 min). Cytogenetic analysis on the cells was carried out at the Section of Medical Genetics of the University of Ferrara, according to the International Guidelines Association of Cytogenetics Technologies. At the same time as amniocentesis, a maternal blood sample was collected and centrifuged (800 g for 10 min); the serum was then stored at –20°C and kept frozen without interruption until analysis.

Since it is known that intraamniotic infection can exist early in pregnancy, even with intact membranes, and in most cases without any clinical symptom (Goldstein et al., 1990Go), our standard procedure for amniocentesis included microbiological studies of all the amniotic fluid samples. Two millilitres of amniotic fluid were transferred into two culture bottles, one aerobic and one anaerobic. Positive cultures were identified by means of a computerized Bac T Alert System® (Organon Teknica, Rome, Italy). After Gram stain microscopic examination, further cultures were prepared with the appropriate medium. Final identification of species was obtained by biochemical (Api System®; Montalieu, Verciev, France) or serological methods.

The population was divided into the following three groups (Table IGo): euploid (n = 92), minor chromosomal abnormalities (n =16) and aneuploid (n = 15), in agreement with our previous study (Vesce et al., 2001Go). There were two cases lacking the amniotic fluid sample (aneuploid group) and five cases lacking the maternal serum sample (three patients with aneuploid chromosomal abnormalities and two controls).


View this table:
[in this window]
[in a new window]
 
Table I. Cytogenetic results
 
In each sample of amniotic fluid and serum, human IL-6, human IL-8 neutrophil-activating-peptide-1 and TNF-ß or lymphotoxin were quantitatively determined by commercial enzyme linked immunosorbent assay (ELISA Bender Medsystem Diagnostic®; GMbH, Vienna, Austria) according to the manufacturer's instructions.

The specific monoclonal antibodies of the three different kits were able to detect IL-6, IL-8 and TNF-ß in cell culture supernatants, human serum, plasma, amniotic or other body fluids. For a correct quantitative determination of each factor, all the samples were diluted in the following way: (i) IL-6: dilution 1:6 in assay buffer for amniotic fluids; the sera, instead, were used undiluted; (ii) IL-8: dilution 1:2 in assay buffer for amniotic fluids and sera; and (iii) TNF-ß: no dilution for either amniotic fluids or sera.

Optical density (OD) values were performed using an automated Microplate-Reader Model 550® (Bio-Rad, USA) at 450 nm (620 nm as optional reference wavelength). For calculation of IL-6, IL-8 and TNF-ß levels, linear regressions were established between OD (y-axis) and the corresponding standard concentrations (x-axis). All results were expressed as pg/ml.

In amniotic fluids, TNF-ß levels were distributed normally, whereas IL-6 and IL-8 values were normalized using natural logarithmic transformation. To assay differences in parameter levels between the three different groups of patients, a one-way analysis of variance (ANOVA), with Tukey HSD test for post-hoc comparison, was performed (Vesce et al., 2001Go). For IL-6 and IL-8, values are expressed both as mean ± 95% confidence interval of the antilog of the logarithmic value and as median value and the interquartile range (lower–upper quartile).

In maternal serum, none of the parameters analysed presented a normal distribution, thus the Kruskal–Wallis ANOVA of Ranks test and the Mann–Whitney U-test were applied to assay differences in parameter levels and the median values and the interquartile ranges (lower–upper quartile) are presented.

To estimate whether levels of the components in amniotic fluid depended on levels in maternal serum, a Spearman's correlation analysis was carried out.

Finally, to assess the predictive values of the parameters of interest, i.e. the likelihood that a woman who shows alteration in the level of one of the components studied actually presents a fetus with chromosomal abnormalities, the Pearson {chi}2 test was used (Sokal and Rohlf, 1995Go).

For all tests, significance levels for the rejection of the null hypothesis were set at P value < 0.05. Data is expressed as mean ± standard deviation (SD) of the variable before the logarithmic transformation in tables. In the graphs, the boxes represent the mean of the variable after natural logarithmic transformation (smallest box) ± SEM (large box), while whiskers around the box indicate 95% confidence interval (CI) of the logarithmic mean. Statistical analyses were carried out with the Statistica software package version 4.5® (StatSoft inc., Tulsa, OK, USA).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Microbiological studies gave negative results in all the cases.

In amniotic fluids, the concentrations of TNF-ß did not differ between euploidies, minor chromosomal abnormalities and aneuploidies, whereas for IL-6 and IL-8 levels, significant results were observed (P = 0.034 and < 0.0001) (Table IIGo).


View this table:
[in this window]
[in a new window]
 
Table II. Analysis of differences for interleukin-6 (IL-6), interleukin-8 (IL-8) and tumour necrosis factors (TNF)-ß levels in amniotic fluid for comparison between euploidies, minor chromosomal abnormalities and aneuploidies
 
For the parameter IL-6, the post-hoc comparisons of means show that this result was attributable to the group with minor chromosomal abnormalities, which revealed significantly higher IL-6 values than euploid cases (P = 0.025). For IL-8, pregnancies with aneuploid fetus presented mean levels significantly lower than both euploid cases (P <0.001) and minor chromosomal abnormalities (P < 0.019) (Figure 1Go).



View larger version (18K):
[in this window]
[in a new window]
 
Figure 1. Box and whisker plots of interleukin-8 (IL-8) (A) and interleukin-6 (IL-6) (B) levels in amniotic fluid. The smallest box represents the mean after log transformation, the largest box represents the SEM and the whiskers indicate 95% confidence interval.chr. ab. = chromosomal abnormality.

 
The analysis of IL-6, IL-8 and TNF-ß levels in maternal serum for euploid and aneuploid pregnancies indicated that no variable showed significant differences from one group to another (Table IIIGo). However, since the results concerning IL-6 showed a relatively low P value (P = 0.084), we compared the levels of IL-6 measured in euploid cases with all the aneuploid cases grouped together. For this comparison, the Mann–Whitney test highlighted a significant difference between the median level of IL-6 in aneuploid and euploid cases (euploidies 0.415 ± 1.63 versus aneuploidies 0.000 ± 0.455, P = 0.026).


View this table:
[in this window]
[in a new window]
 
Table III. Analysis of differences for IL-6, IL-8 and TNF-ß levels in maternal serum for comparison between euploidies, minor chromosomal abnormalities and aneuploidies
 
Subsequently, the predictive value of the parameters showing significant values in the previous analyses was assessed. The threshold values were set at 100% sensitivity. The Pearson {chi}2 indicated that only IL-6 and IL-8 in amniotic fluid had significant predictive value, for minor chromosomal abnormalities and aneuplodies respectively (Table IVGo). However, the specificity in predicting aneuploid pregnancies, at 100% sensitivity, was not very high: an IL-8 level below 1011.70 pg/ml in amniotic fluid had a specificity of 26.07%, and IL-6 levels in amniotic fluid above 135.16 pg/ml had a specificity of 20.65% (Table IVGo). The specificity reached higher values, 63% for IL-8 and 60% for IL-6 measured in amniotic fluids, when the level of sensitivity was reduced to 75% (Table IVGo).


View this table:
[in this window]
[in a new window]
 
Table IV. Predictive value of IL-6 and IL-8 for pregnancies with chromosomal abnormalities
 
We then verified whether the levels of these three parameters in amniotic fluid depended on their levels in maternal serum. No correlation was found with TNF-ß. The levels measured for IL-8 in amniotic fluids presented a significant borderline correlation with IL-8 levels in maternal serum (P = 0.0497) and a strong positive correlation with IL-6 levels measured in maternal serum (P = 0.005).

To assess if the correlation observed between IL-8 in amniotic fluid and IL-6 levels in maternal serum was a general trend for all three groups of pregnancies studied, or if a different behaviour was shown by pathological cases, we verified the strength of the correlation within these groups. What we observed was an absence of correlation between these two parameters within both euploid cases (P = NS) and pregnancies presenting minor chromosomal abnormalities (P = NS). Thus, the direct proportionality between IL-8 level in amniotic fluids and IL-6 in maternal serum observed for the whole sample could be attributed almost entirely to the group of pregnancies presenting aneuploidies (P < 0.006).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
It has been hypothesized that successful pregnancy depends upon the activation of a Th-2 response by the maternal immune system under the stimulus of fetal factors, such as IL-3, IL-4, IL-10 and the progesterone-induced blocking factor, released at the placental level (Wegmann et al., 1993Go). Instead, when the scales of the immunoregulatory system are tipped in favour of a Th-1 maternal response, there is an increased incidence of pregnancy loss (Hill et al., 1995Go). This type of response is mediated by the release of TNF-ß, interferon-gamma (INF-{gamma}), and the activation of natural killer, as well as lymphokine-activated cells (Raghupathy, 1997Go). TNF-ß and IFN-{gamma} are the principal stimulators of IL-6 release, which in turn activates the arachidonic acid cascade with production of prostaglandin PGE-2 (Akira et al., 1993Go; Elliott et al., 1998Go). However, the available information on cytokine behaviour even during euploid pregnancy is complex and often contradictory (Rice and Chard, 1998Go).

The results of the present study, which was limited to three cytokines IL-6, IL-8 and TNF-ß, did not show any difference either in maternal serum or in the amniotic levels of TNF-ß between the three groups of patients. A significant increase of amniotic IL-6 level was found in the presence of fetal chromosomal abnormalities. However, the main increase was accounted for by the group of minor chromosomal abnormalities, while for the aneuploidy group, also characterized by a higher level of the cytokine, significance was not reached. Since all the microbiological studies on amniotic fluid gave negative results, the increased level of IL-6 was interpreted as a consequence of a genetic feature rather than infection. In contrast, the amniotic level of IL-8 was significantly lower in aneuploid cases compared with either euploid or minor chromosomal abnormality groups.

As regards maternal serum, the finding of decreased levels of IL-6 concentration in the aneuploid group in the presence of an increased amniotic level of the cytokine represents an apparent discrepancy. However, a strict compartimentalization has been reported for IL-8 and for monocyte chemotactic protein-1 (MCP-1) during the first trimester of pregnancy, with high levels in the extra-embryonic coelomic fluid and very low levels in amniotic fluid and maternal serum (Denison et al., 1998Go). Furthermore, the absence of correlation between maternal plasma and either fetal plasma or amniotic fluid of IL-6 concentration has been recently reported (Romero et al., 1998Go). In Romero's study, the presence of increased fetal plasma concentration of IL-6 was considered as an expression of a fetal inflammatory response to infection. This type of response is supposed to be addressed to trigger uterine contractions, giving rise to labour in order to escape an intrauterine environment that is hostile. It can be confined to the fetus lacking in the amniotic fluid as well as in the maternal plasma, as revealed by lower levels of the cytokine within these compartments. However it is also reported (Gomez et al., 1998Go) that in about 13% of the cases the elevation of the fetal plasma concentration of IL-6 is not related to infection, and therefore it is supposed that there is an unknown infection-independent process able to increase the level of the cytokine. Based on our data such a process could be related to genetically controlled factors.

In the search for possible connections between fetal chromosomal abnormalities and alterations in cytokine levels, it must be considered that IL-6 and IL-8 are coded by genes localized on chromosome 7 (Sehgal et al., 1986Go) and 4 (Modi et al., 1990Go) respectively.

Recently, Hughes et al. (Hughes et al., 2000Go), using whole-genome expression data, discovered an unexpected similarity in transcript profiles for unrelated deletion mutants in Saccharomyces cerevisiae. In fact, in some of the 300 mutants analysed, the observed correlation between gene transcription profiles was mostly due to changes in the expression of genes present on chromosomes other than the aneuploid chromosome, as might be expected when a key transcriptional regulator is affected directly by the aneuploidy. This observation confirms and extends the previous result obtained by Pollack et al. on human transformed cells (Pollack et al., 1999Go).

Thus, considering the complexity of gene expression pattern in eucaryotic cells, it seems credible to form the hypothesis that the connection between the fetal chromosomal abnormalities observed in our data, mainly involving chromosomes X, 21, 18 and 13, and variation of IL-6 and IL-8 levels, could lie in a correlation between the expression of genes present in genomic regions localized on chromosomes other than that which determines aneuploidy.

As regards the obstetrical implications of our findings, it should be considered that IL-8 has been reported to stimulate placental prostacyclin (PGI2) production. During pregnancy PGI2 acts as a potent vasodilator and an inhibitor of platelet aggregation, protecting against the vasoconstrictory and platelet aggregatory action of thromboxane A2. Moreover, it has been found that the placentas of women with pre-eclampsia produce seven times less PGI2 than thromboxane A2 (Walsh, 1985Go), and that decreased PGI2 production correlates with significantly reduced release of IL-8 (Wang et al., 1999Go). Therefore, it can be hypothesized that the significantly reduced amniotic level of IL-8, coupled with the increased amniotic level of IL-6 found in our cases, represents a factor that could impair placental perfusion and promote uterine contractions, thus leading to the adverse outcome of pregnancy.

To explain the high incidence of abortion in aneuploid pregnancies compared with euploid, we have previously reported that the former are characterized by a significantly higher amniotic and lower maternal plasma level of endothelin-1 (Vesce et al., 1996Go), a potent vasoconstrictor peptide, the level of which is also increased in pre-eclampsia (Nova et al., 1991Go). Moreover, we have found that the amniotic and maternal serum levels of the components of the plasminogen system, which are involved in the physiological angiogenic processes of pregnancy, are deeply altered in the presence of fetal aneuploidy (Vesce et al., 2001Go).

In the present study, it is interesting to highlight the significant positive correlation found between IL-6 levels in serum and amniotic IL-8 values.

All these observations support the suggestion that aneuploidy correlates with a disturbance of the release of some mediators of placental perfusion and uterine contraction that might explain the increased incidence of pregnancy loss.

In the cases of abortion, preterm labour, premature rupture of the membranes and fetal inflammatory response syndrome where infection is not found, the imbalanced levels of inflammatory cytokines could be interpreted as a consequence of a genetic feature whereby the fetus participates in the mechanism of its own distress, death and expulsion. However, in this study the cytokine imbalance apparent in this group of patients might represent a chance finding, and it must be confirmed by extensive investigation before any firm conclusion can be reached. Such an approach could be of value for the correct assessment of some cases with poor outcome of pregnancy as a consequence of a genetic disorder rather than infection.


    Notes
 
4 To whom correspondence should be addressed at: Department of Biomedical Sciences and Advanced Therapy, Section of Obstetrics and Gynaecology, University of Ferrara, 44100 Ferrara, Italy. E-mail: vsf{at}dns.unife.it Back

Submitted on May 15, 2001; resubmitted on July 27, 2001


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Akira, S., Taga, T. and Kishimoto, T. (1993) Interleukin-6 in biology and medicine. Adv. Immunol., 54, 1–78.[ISI][Medline]

Baggiolini, M. and Dewlad, B. (1993) Interleukin-8 and related chemotactic cytochines-CXC and CC chemokines. Adv. Immunol., 55, 97–179.[ISI]

Boué, A., Boué, J. and Groop, A. (1985) Cytogenetics of pregnancy wastage. Adv. Hum. Genet., 14–57.

Bromage, S.J., Lang, A.K., Atkinson, I. et al. (2000) Abnormal TGFbeta levels in the amniotic fluid of Down Syndrome pregnancies. Am. J. Reprod. Immunol., 44, 205–210.[ISI][Medline]

Cherouny, P.H. and Pankuch, G.A. (1993) Neutrophil attractant/activating peptide-1/interleukin-8: association with histologic chorioamnionitis, preterm delivery and bioactive amniotic fluid leukoattractans. Am. J. Obstet. Gynecol., 169, 1299–1303.[ISI][Medline]

Daher, S., Fonseca, F. and Ribeiro, O.G. (1999) Tumor necrosis factor during pregnancy and at the onset of labor and spontaneous abortion. Obstet. Gynecol., 83, 77–79.

Denison, F.C., Riley, S.C., Wathen, N.C. et al. (1998) Differential concentrations of monocyte chemotactic protein-1 and interleukin-8 within the fluid compartments present during the first trimester of pregnancy. Hum. Reprod., 13, 2292–2295.[Abstract]

Dudley, D.J. and Edwin, S.S. (1997) Regulation of decidual cell chemokines production by group B streptococci and purified bacterial cell components. Am. J. Obstet. Gynecol., 177, 666–672.[ISI][Medline]

Elliott, C., Kelly, R.W., Critchley, H.O. et al. (1998) Regulation of interleukin 8 production in the term human placenta during labor and by antigestagens. Am. J. Obstet. Gynecol., 179, 215–220.[ISI][Medline]

Furuta, I. and Yamada, H. (2000) Effects of inflammatory cytokines on prostaglandin E (2) production from human amnion cells cultured in serum-free condition. Gynecol. Obstet. Invest., 49, 93–97.[ISI][Medline]

Goldstein, I., Zimmer, E.Z., Merzbach, D. et al. (1990) Intraamniotic infection in the very early phase of the second trimester. Am. J. Obstet. Gynecol., 163, 1261–1263.[ISI][Medline]

Gomez, R., Grezzi, F., Romero, R. et al. (1998) The fetal inflammatory response syndrome. Am. J. Obstet. Ginecol., 179, 194–202.

Greig, P.C. and Murtha, A.P. (1997) Maternal serum interleukin-6 during pregnancy and during term and preterm labor. Obstet. Gynecol., 90, 465–469.[Abstract/Free Full Text]

Hill, J.A., Anderson, D.J. and Polgar, K. (1995) T helper 1-type cellular immunity to trophoblast in women with recurrent spontaneous abortion. J. Am. Med. Assoc., 273, 1933–1958.[Abstract]

Hsu, C.D. and Meaddough, E. (1998) Elevated amniotic fluid levels of leukemia inhibitory factor, interleukin 6 and interleukin 8 in intra-amniotic infection. Am. J. Obstet. Gynecol., 179, 1267–1270.[ISI][Medline]

Hughes, T.R., Roberts, C.J., Dai, H., Jones, A.R. et al. (2000) Widespread aneuploidy revealed by DNA microarray expression profiling. Nature Genet., 25, 333–337.[ISI][Medline]

Modi, W.S., Dean, M., Seuanez, H.N. et al. (1990) Monocyte-derived neutrophil chemotactic factor (MDNCF/IL-8) resides in a gene cluster along with several other members of the platelet factor 4 gene superfamily. Hum. Genet., 84, 185–187.[ISI][Medline]

Murtha, A.P. and Greig, P.C. (1998) Maternal serum interleukin-6 concentration as a marker for impending preterm delivery. Obstet. Gynecol., 91, 161–163.[Abstract/Free Full Text]

Nova, A., Sibai, B.M., Barton, J.R. et al. (1991) Maternal plasma level of endothelin is increased in pre-eclampsia. Am. J. Obstet. Gynecol., 165, 724–727.[ISI][Medline]

Pollack, J.R., Perou, C.M., Alizadeh, A.A. et al. (1999) Genome-wide analysis of DNA copy-number changes using cDNA microarrays. Nat. Genet., 23, 41–46.[ISI][Medline]

Raghupathy, R. (1997) Th1-type immunity is incompatible with successful pregnancy. Immunol. Today, 18, 478–482.[ISI][Medline]

Rehbock, J. and Chondromatidou, A. (1997) Bradykinin stimulates interleukin-6 and interleukin-8 secretion of human decidua derived cells. Br. J. Obstet. Gynaecol., 104, 495–499.[ISI][Medline]

Rice, A. and Chard, T. (1998) Cytokines in implantation. Cytokin Growth Factor Reviews, 9, 287–296.

Romero, R., Gomez, R., Grezzi, F. et al. (1998) A fetal systemic inflammatory response is followed by the spontaneous onset of preterm parturition. Am. J. Obstet. Gynecol., 179, 186–193.[ISI][Medline]

Sehgal, P.B., Zilberstein, A., Ruggieri, R.M. et al. (1986) Human chromosome 7 carries the beta-2 interferon gene. Proc. Natl Acad. Sci., 83, 5219–5222.[Abstract]

Sokal, R.R. and Rohlf, F.J. (eds) (1995) Biometry, 3rd edn. Freeman, New York, pp. 260–265.

Steinborn, A. and Gunes, H. (1996) Elevated placental cytokine release, a process associated with preterm labor in the absence of intrauterine infection. Obstet. Gynecol., 88, 534–539.[Abstract/Free Full Text]

Tanaka, Y. and Narahara, H. (1998) Interleukin 1ß and interleukin-8 in cervicovaginal fluid during pregnancy. Am. J. Obstet. Gynecol., 79, 644–649.

Vesce, F., Farina, A., Jorizzo, G. et al. (1996) Raised level of amniotic endothelin in pregnancies with fetal aneuploidy. Fetal. Diagn. Ther., 11, 94–98.[ISI][Medline]

Vesce, F., Scapoli, C., Giovannini, G. et al. (2001) Plasminogen activator system in serum and amniotic fluid of euploid and aneuploid pregnancies. Obstet. Gynecol., 97, 404–408.[Abstract/Free Full Text]

Walsh, S.W. (1985) Preeclampsia: an imbalance in placental prostacyclin and thromboxane production. Am. J. Obstet. Gynecol., 152, 335–340.[ISI][Medline]

Wang, Y., Baier, J., Adair, C.D. et al. (1999) Interleukin-8 stimulates placental prostacyclin production in preeclampsia. Am. J. Reprod. Immunol., 42, 375–380.[ISI][Medline]

Wegmann, T.G., Lin, H., Guilbert, L. et al. (1993) Bidirectional cytokine intercations in the maternal-fetal relationship: is successful pregnancy a Th2 phenomenon? Immunol. Today, 14, 353–356.[ISI][Medline]

accepted on October 19, 2001.