Premature rupture of membrane (PROM) is characterized by membrane rupture prior to the onset of full-term labor. PROM complicates approximately 8% of pregnancies at term.1 The main clinical concern for PROM is a high risk of infectious morbidity and mortality in newborns.2 To reduce the risk of fetal and maternal complications, it is required to find specific biomolecules associated with development of PROM.

The membrane rupture is related to collagen destruction and programmed cell death in fetal membranes. The risk factors for PROM involve reproductive tract infection, cigarette smoking, and uterine overdistention, and various mediators of PROM, including matrix-degrading enzymes, prostaglandins and inflammatory cytokines, can be produced in the fetal membranes and deciduas.1-3 Recent studies suggest that reactive oxygen species (ROS) might damage connective tissues of chorioamnion and induce PROM.3-5 However, the exact mechanism of PROM is still unclear.

Haptoglobin (Hp) is an acute phase protein which plays roles in the systemic host defense in response to inflammation.6 The protein binds hemoglobin (Hb), thereby protecting cells from oxidative damage that occurs during hemolytic injury.7 In addition, Hp exerts an inhibitory effect on prostaglandin synthesis8 and a stimulatory effect on differentiation of endothelial cells.9 Therefore, Hp is considered to have functions as an antioxidant, immunomodulator and angiogenic factor.

Although it has been well-known that the main site of Hp biosynthesis is liver, several recent studies demonstrated that the Hp protein is locally expressed in female reproductive organs such as the ovary, endometrium, and decidua.10,11 The Hp can also be detected in amniotic fluid and vaginal fluid.12 In addition, the plasma Hp level is elevated during pregnancy in a biphasic pattern with peaks in the first and third trimester.13 These findings suggest that Hp may play a role in human reproduction and its activity may be involved in obstetric and gynecologic diseases.

The human Hp gene has 2 major alleles, Hp¹ and Hp², which are expressed as 3 major phenotypes (Hp 1-1, Hp 2-2 and Hp 2-1). The Hp 1-1 and Hp 2-2 correspond to homozygote of the Hp¹ and Hp² alleles, respectively, while the Hp 2-1 is heterozygous for the two alleles (Hp¹/Hp²).14 It has been reported that the allelic polymorphism of Hp is associated with susceptibility and severity of specific diseases such as atherosclerosis, Parkinson's disease, gestational diabetes and preeclampsia. 15-18

To date, the association of Hp with PROM has not been reported. To elucidate whether the Hp polymorphism is related to a high risk of PROM, we investigated the Hp phenotypes/genotypes of pregnant Korean women without or with PROM in the present study.

To analyze Hp phenotypes, 34 pregnant women who suffered from PROM and 187 pregnant women who underwent normal delivery were recruited. There were no significant differences between the PROM and the normal delivery (control) groups in the clinical characteristics such as nullipara, maternal age, gestational age at delivery, neonatal birth weight and Apgar score at 5 minutes (Table 1).

As for Hp polymorphism, the pregnant women with normal delivery (control group) showed Hp 1-1, 2-1, and 2-2 phenotypes at 8.0%, 43.3%, and 48.7%, respectively (Table 2). The Hp phenotypic distribution in these women was found to be statistically similar to ratios in healthy Koreans.20 The distribution rate of Hp phenotype for total population enrolled in this study (n = 221) was also similar; 10.4%, 42.5%, and 47.1% for Hp 1-1, 2-1 and 2-2, respectively. In contrast, however, patients with PROM demonstrated the Hp 1-1 type at 23.5%, which is a significantly higher rate than 8.0% of control group. The percentage to express the Hp 2-2 type was lower in PROM cohort (38.2%) than in the control group (48.7%). By Pearson Chi-Square test, Hp 1-1 type was significantly prevalent in PROM group as compared with control (23.5% vs 8.0%, p = 0.006) with odds ratio of 3.528 (95% CI: 1.362 - 9.141) (Table 2).

The Hp allele frequencies were calculated from the distribution data for Hp phenotypes (Table 2) and evaluated using Pearson Chi-Square test. As shown in Table 3, the frequency of Hp¹ allele in the PROM group was significantly higher than that of the control group (0.426 vs 0.297, p = 0.034) with odds ratio of 1.762 (95% CI: 1.038 - 2.991). These findings suggest that term pregnant women with the Hp¹ allele have higher incidence of PROM than women with the Hp² allele.

Several studies have demonstrated the relation of Hp polymorphism with susceptibility and severity of clinical symptoms in specific disorders, including angiopathies, diabetes, Parkinson's disease and preeclampsia.15-18 However, to the best of our knowledge, the correlation between PROM and Hp has not yet been studied. In the present study, we found that the frequency of individuals afflicted with PROM was significantly higher in women expressing the Hp 1-1 than in women expressing the Hp 2-2, suggesting that individuals with the Hp¹ allele are more susceptible to PROM as compared to those with the Hp² allele (Tables 2 and 3).

Hp proteins with the 3 phenotypes show phenotype-dependent functional differences in anti-oxidant, anti-inflammatory and angiogenic activities; 7,9,21 Hp 1-1 has a higher anti-inflammatory activity than Hp 2-2. On the contrary, Hp 2-2 possesses more potent angiogenic activity compared to Hp 1-1 and plays a role more efficiently during angiogenic process, while Hp 2-1 shows an intermediate activity. Although the exact mechanism involved in higher incidence of PROM in individuals with Hp 1-1 is not known yet, it is quite plausible to suggest that the lower angiogenic potential in carriers of Hp 1-1 may contribute less effectively to the repair of damaged connective tissues of chorioamnion and may develop PROM more easily.

Hp forms a stable Hp-Hb complex by irreversible binding with Hb. The Hp-Hb ligand binds to CD163, which is known as a Hb scavenger receptor, and is endocytosed. Recent studies demonstrated that the binding of Hp-Hb complex to CD163 receptor increases the expression of anti-inflammatory modulators such as interleukin-10 and heme oxygenase-1.22 It was also reported that the CD163 surface ligation induces tyrosine kinase-dependent calcium mobilization as well as casein kinase II- and protein kinase C-dependent secretion of proinflammatory cytokines (interleukin-6 and interleukin-1β).23 Kristiansen et al. (2001) reported that affinity of the Hp-Hb complex for CD163 was Hp phenotype-dependent;24 the complex of Hp 2-2 and Hb exhibits higher functional affinity for CD163 than the complex of Hp 1-1 and Hb, although both Hp 1-1 and Hp 2-2 can efficiently remove Hb through CD163 receptor. In addition, the role of Hp in inflammatory response is phenotype-dependent, showing that Hp 1-1 and Hp 2-2 make dominant Th2 response and Th1 response, respectively.21 Therefore, to understand higher incidence of PROM in the individuals with Hp 1-1, because of the reason that PROM is an inflammation-associated gestational disease, further studies are required, focusing mainly on the expression of inflammatory regulators via CD 163 signaling in amniotic tissues or decidua.

Preterm PROM is the cause of nearly half of all preterm delivery and a high risk factor of perinatal morbidity and mortality.2 Because the pathophysiologic processes involved in subclinical infection are considered to be similar in both term and preterm PROM, the evidences for term PROM in this study might be applied similarly to preterm PROM. Nevertheless, it is necessary to confirm exact relationship of Hp to preterm PROM.

In conclusion, pregnant women with Hp 1-1 type have higher incidence of PROM than those with other Hp types. To our best knowledge, this is the first study that evaluated the significance of Hp phenotype with respect to the development of PROM. The findings suggested that the Hp polymorphism may be potentially applied to the prevention of PROM during pregnancy.