br Materials and methods br Results

Materials and methods
Results
Discussion To our knowledge, this is the first large-scale longitudinal measurement of more than 20 cytokines in normal pregnancy. Simultaneous measurements of several cytokines with multiplex magnetic bead-based immunoassays are becoming increasingly common in scientific research. Serum cytokine reference values for normal pregnancy as presented here are required to explore the clinical potential of cytokine profiling in pregnancy. Without standardized methods, a reference table must ideally be presented for each commercially available type of multiplex analysis. The absolute cytokine values in this study are valid for a Bio-Plex Assay measured on the Luminex 200 system which is commonly used for broad cytokine profiling [8], [28]. As individual basal cytokine levels vary and must be taken into account [7], [20], our reference table of relative individual changes in cytokine levels provides information that has a wider application. With caution, these normalized reference values can be used for comparison of cytokine development in samples analysed with different cytokine profiling assays. The gestational variation of chemokines reported here has not been revealed previously, but a significant decrease of MCP-1 from first to third trimester is consistent with our results [31]. The common function of chemokines is directed movement of leukocytes. The first trimester is characterized by increased immune activity and leukocyte recruitment, as part of NS-398 synthesis and placentation [25]. The decreasing chemokine levels in second trimester may accompany a physiological reduction in immune activity and leukocyte migration. Early pregnancy disease-specific differences in chemokine levels has been reported. Increased serum MIP-1α and decreased serum MCP-1 in first trimester has been associated with later development of preeclampsia [34] or giving birth to small for gestational age babies [15]. In second trimester, elevated levels of MCP-1 in women experiencing preeclampsia have been detected [13]. These seemingly contradictory MCP-1 observations related to preeclampsia and our findings of substantial gestational variation of chemokines in first half of normal pregnancies, points to caution for measuring a single chemokine when investigating disease-specific patterns. This study of gestational changes in cytokine patterns in normal pregnancies with multivariate methods is novel. In studies of single immune biomarkers, inflammatory cytokines and CRP appear most frequent but with contradictory results. The gestational IL-2 and IL-8 development observed here is supported by others [11], [14]. We showed increasing levels of CRP from first to second trimester, earlier this has been shown from early to late pregnancy [22], while others have reported stable CRP levels [6], [35]. Our results reported decreasing levels of IL-6 throughout the first half of pregnancy, while others have shown either increasing [16], [40] or stable levels of IL-6 with increasing gestational age in this period [2], [11], [14]. Likewise, we found decreasing levels of IL-12p70, while an increase between the first and second trimester has been reported [14]. The exclusion of women with previous or current disease or pregnancy complication may differentiate this from some other studies where such women were included or the medical history was unknown [2], [6], [14], [22], [35], [40]. Diverging results may also be expected from small study populations [11], [16]. Serum inflammatory cytokines may reflect a wide number of infectious and non-infectious diseases, highlighting the importance of including well characterized study groups. The nine inflammatory cytokines were strongly correlated with each other at all gestational ages, indicating a common regulation of this cytokine group during pregnancy. IL-1β, IL-6 and TNF-α has been shown to correlate positively at all trimesters [11]. In our study, the two cytokines with the highest mean correlation were the inflammatory cytokines IL-2 and IL-15. They are both members of the 4α-helix bundle family and have similar functions [18]. To our knowledge, their specific role in human pregnancy has not yet been explored. The high intra-group correlation may reflect that different inflammatory cytokines have a powerful potentiating effect on each other in normal pregnancy, and highlight the detrimental role of inflammatory cytokines when dysregulated. Surprisingly, CRP did not correlate significantly with the inflammatory cytokines during first half of pregnancy, even though the synthesis of CRP is regulated primarily by inflammatory cytokines [30]. A positive correlation between CRP and inflammatory cytokines in pregnancy has previously been indicated in a study group with higher mean BMI [11], possibly affecting the results [29]. We find that the inflammatory cytokine patterns in pregnancy may be more sensitive and informative than CRP in characterising the early maternal systemic immune status. Inflammatory cytokines has been connected to development of pregnancy complications. Preeclampsia is associated with increased levels of IL-6, IL-8, IL-12p70, IFN-γ and TNF-α compared to healthy pregnant controls [21], [23], [24], [37]. Still, the conflicting reports on predictive value of single cytokine measurements from early pregnancy clearly underlines the need for more sensitive methods [13], [26], [34], [39]. The first half of pregnancy forms the basis of many common complications later in pregnancy such as intrauterine fetal growth restriction and preeclampsia [32], [43], and maternal inflammation during pregnancy may further have impact on cognitive development in infants [33]. This underlines the relevance of cytokine analyses in pregnancy, not only for a successful pregnancy outcome, but also for future development of the child. The stable intra-group correlation of inflammatory cytokines in normal pregnancies shown here, combined with the inflammatory dysregulation in many diseases, provides further evidence for the clinical usefulness of combined profiling of inflammatory cytokines in maternal serum. This must be explored in disease-specific populations.