Effects of TIMP1 rs4898 Gene Polymorphism on Early-Onset Preeclampsia Development and Placenta Weight

Introduction: Some evidence indicates that the improper trophoblast invasion of maternal spiral arteries could be caused by an imbalance between matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs), leading to preeclampsia (PE) development. This study aimed to assess the potential role of MMP1, MMP9, TIMP1 and TIMP2 gene polymorphisms in the pathogenesis of PE. Materials and methods: A total of 308 Polish women, 115 preeclamptic (55 with early-onset preeclampsia [EOPE], 60 with late-onset preeclampsia [LOPE]) and 193 healthy pregnant women, all of Caucasian origin, were recruited to the study. PE was diagnosed following the ACOG criteria. The polymorphic variants of the MMP-TIMP pathway (MMP1 rs1799750, MMP9 rs17576, MMP9 rs17577, TIMP1 rs4898, TIMP2 rs2277698, TIMP2 rs55743137) were genotyped by polymerase chain reaction and restriction fragment length polymorphism. Results: Analyzing all SNPs in the MMP-TIMP pathway, no significant differences in allele frequencies between preeclamptic women and controls were observed. However, comparing the EOPE and LOPE groups with each other, we observed a statistically significant difference between them for the TIMP1 rs4898 variant. In the whole group of 308 women, the mean placenta weight was the lowest in carriers of the rs4898 CC genotype. Post hoc analysis revealed significant differences between CC-CT (p = 0.0209) and CC-TT (p = 0.0469). Additionally, during allele analysis, a statistically significant difference in the mean placenta weight (for C allele 529.32 ± 157.11 g, for T allele 560.24 ± 162.24 g, p = 0.021) was also observed. Conclusion: Our findings suggest a relationship between TIMP1 rs4898 (372T > C) polymorphism and increased risk of early-onset preeclampsia in a population of pregnant Polish women. Our data suggest that the TIMP1 rs4898 C allele might be associated with increased risk for early-onset, but not for late-onset preeclampsia. To evaluate the role of the TIMP1 polymorphic variants in the etiopathology of preeclampsia, further studies with a larger sample size are needed.


Introduction
Preeclampsia (PE) is known as one of the most threatening complications of pregnancy and can be associated with fetal growth restriction, preterm delivery, placental abruption and increased fetal and maternal mortality. Preeclampsia complicates 2-8% of pregnancies [1]. Despite intensive research, the pathophysiology of preeclampsia is still discussed and probably multifactorial. The main causative factor is improper placentation of the trophoblast, which further leads to endothelial dysfunction and abnormal morphology of spiral arteries. There is some evidence that improper trophoblast invasion of maternal spiral arteries can be caused by an imbalance between matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) [2].
MMPs are a family of 28 endopeptidases that are involved in the degradation of extracellular matrix (ECM) components and therefore play a crucial role in physiological as well as pathophysiological tissue remodeling. The MMPs' activity is suppressed by various endogenous inhibitors, of which the main ones are TIMPs, a family comprising four members (TIMP 1, 2, 3 and 4). MMPs as well as TIMPs play different roles in many physiological processes related to human reproduction, such as ovulation, blastocyst implantation, embryogenesis, and placentation [3][4][5][6]. It is well known that MMP1 and MMP9 are primarily inhibited by TIMP1, and TIMP2 is a crucial inhibitor of MMP2, but it also inhibits, to a lesser extent, MMP9 [7,8].
In early pregnancy, there are activated vasculogenesis, angiogenesis, vascular remodeling and trophoblast invasion, all of these being essential for proper placenta development [9]. The remodeling of the spiral arteries as well as trophoblast invasion may be modulated by proteolysis in the ECM interface and, therefore, these processes may contribute to decidual remodeling at the mother-fetal interface [5]. MMP2 and MMP9 are able to degrade ECM components, especially collagen I and IV, and probably they could be involved in the remodeling of the placenta and uterine artery in healthy pregnancy. Several studies have shown that elevated levels of MMP1, MMP2, and MMP9 are critical for vasodilation, placentation and uterine expansion during healthy pregnancy. Hence, the regulation of the expression of the MMP-TIMP system plays a key role in the development of a normal pregnancy [2].
The abnormal conversion of the spiral arteries may lead to certain complications in pregnancy, including fetal growth restriction and preeclampsia [10]. Decreased vascular MMP1 and MMP9 activity may lead to impaired collagen type I and IV turnover that could result in increased vasoconstriction and decreased blood flow, followed by the development of hypertensive abnormalities in pregnancy [11]. The endothelial cells from preeclamptic placentas produce a smaller amount of MMP1 compared to healthy placentas. MMP imbalance may lead to trophoblast cell apoptosis and reduced uterine perfusion pressure. All these factors are associated with the pathophysiology of preeclampsia. In a few clinical studies, MMP and TIMP level changes were described in women with preeclampsia. Montagnana et al. [12] observed elevated serum levels of TIMP1 in women with preeclampsia compared to healthy pregnant and non-pregnant women, as well as higher levels of TIMP2 in women with pre-eclampsia and healthy pregnant women compared to non-pregnant women [12].
It has also been suggested that MMP and TIMP family genes can play a role in pathophysiology of various cardiovascular diseases such as ischemic stroke, aortic diseases atherosclerosis and preeclampsia [4]. Hence, there is a unique pattern of involvement of MMP and TIMP genes in pathogenetically different phenotypes. Some studies have investigated MMP and TIMP gene polymorphisms as predisposing factors for cardiovascular disorders and pregnancy complications [13]. Increased activity of MMPs was found to be associated with aortic diseases and an association was observed between MMP9 gene polymorphisms (−1562C>T) and risk of ischemic stroke [4]. Several case-control studies have investigated the role of MMP gene polymorphisms in preeclampsia [14][15][16][17][18][19]. The most frequently tested polymorphisms were located in the MMP9 gene (rs3918242, rs2234681) and the MMP2 gene (rs243865, rs2285053). Moreover, two studies were conducted for TIMP1 and TIMP3 gene polymorphisms in preeclampsia and showed the association for TIMP1 rs20705558 and TIMP3 rs80272 polymorphisms [20,21]. This study aimed to assess the potential role of MMP1, MMP9, TIMP1 and TIMP2 gene polymorphisms in the pathogenesis of preeclampsia. Considering the involvement of these genetic variants in abnormal placentation processes typical for hypertensive disorders during pregnancy, we determined them in healthy pregnant as well as in preeclamptic women.

Study Population
A total of 308 Polish women, 115 preeclamptic and 193 healthy pregnant women, all of Caucasian origin, were recruited to the study. The patients were hospitalized over the years 2018-2021 in the Division of Perinatology and Women's Diseases, Poznan University of Medical Sciences, Poland. Preeclampsia was diagnosed following the ACOG criteria (disease occurrence after 20 weeks of gestation, systolic blood pressure equal to or higher than 140 mmHg, diastolic blood pressure equal to or higher than 90 mmHg, possible proteinuria presence, other signs of preeclampsia) [22]. The whole group of preeclamptic patients was divided into two subgroups: early-onset preeclampsia (EOPE) that develops before 34 week of gestation (n = 55) and late-onset preeclampsia (LOPE) that occurs after 34 weeks of gestation (n = 60). We applied the following criteria for exclusion from the study groups: multiple gestations, fetal structural/genetic anomaly, maternal diseases such as diabetes mellitus, chronic hypertension, as well as renal, endocrinological and autoimmune diseases.
In the control group we enrolled healthy normotensive pregnant women without any fetal disorder or multiple pregnancies. All patients were informed about the goal of the study and gave their written consent. The study was approved by the Local Bioethical Committee at Poznan University of Medical Sciences (493/21).

DNA Extraction
Peripheral venous blood samples were collected from case and control groups into sterile EDTA coated tubes. Genomic DNA extraction was performed using the DNA Blood Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol. The amount of the total extracted DNA was determined using the spectrophotometric method (NanoDrop Spectrophotometer, Thermo Scientific, Waltham, MA, USA). The samples were stored at −20 • C until analysis.

Genotyping
The polymorphic variants mentioned in Table 1 were genotyped by polymerase chain reaction and restriction fragment length polymorphism (PCR-RFLP). PCR amplification was carried out in a final volume of 25 µL using a DNA Engine Dyad Thermocycler (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The reaction mixture consisted of 100 ng of genomic DNA, 0.25 µM of each primer (TiBMolBiol, Berlin, Germany), 1 U of Taq DNA polymerase and respective buffer (DreamTaq Green DNA polymerase, Thermo Fisher Scientific, Waltham, MA, USA), 200 µM dNTP Mix (Thermo Fisher Scientific, Waltham, MA, USA) and deionized water. PCR reaction conditions were optimized for each polymorphism. Characteristics of the polymorphisms and the specific primer sequences are shown in Table 2. Amplified fragments were then digested with the appropriate restriction enzyme (Thermo Fisher Scientific, Waltham, MA, USA), according to the manufacturer's instructions, and visualized after electrophoresis on 2% or 3% agarose gel with Midori Green Advanced DNA Stain (Nippon Genetics, Europe GmbH, Düren, Germany).

Statistical Analyses
All statistical analyses were conducted in the R statistical software version 4.0.3 [23] and ggstatsplot package [24]. Quantitative data are presented as mean ± SD, whereas frequency and percentages were used for qualitative variables. Differences were calculated by chi-square and Fisher exact tests (for categorical variables) or the Student t-test (for continuous variables). The Hardy-Weinberg equilibrium (HWE) was assessed using a χ 2 test in each group. Genotype and allele frequency distributions were determined under the different models of inheritance (codominant, dominant, recessive) using the SNPassoc package [25]. ANOVA and the post-hoc Tukey method were carried out for a comparison of genotypes and clinical characteristics. Haplotype analysis was performed using the haplo.stats package [26]. A p value less than 0.05 was considered significant. Additionally, we used the nonparametric multifactor dimensionality reduction (MDR, v3.0.2) approach to detect potential gene-gene interactions. To minimize the chance of obtaining type I errors, the false discovery rate (FDR) approach was used. GAS (Genetic Association Study Power Calculator) (https://csg.sph.umich.edu/abecasis/gas_power_ calculator/index.html accessed on 29 June 2022) was used for post-hoc power calculation. After considering sample size (cases = 115 and controls = 193), preeclampsia prevalence Diagnostics 2022, 12, 1637 5 of 16 2-8% [1]-mean 5%, MAF, and significance level=0.05, the overall study power (average of six tested variants) was calculated at 62.0%.

Clinical Characteristics of the Study Population
The selected characteristics of the study women are presented in Table 3. A total of 115 women at 25-40 weeks of gestation with preeclampsia were enrolled in this study. The differences in clinical parameters between the study and control groups were mostly significant, except for the mother's age and sex of the newborn. The mean maternal age for PE group was 30.32 ± 5.32 years and for healthy subjects it was 31.06 ± 3.89 (p = 0.2019). Among the patients with preeclampsia, 55 (47.83%) were diagnosed before 34 weeks of gestation, and 60 (52.17%) patients were diagnosed at or after 34 weeks of gestation. No significant difference was found in age, diastolic blood pressure, preand post-pregnancy BMI, number of pregnancies or infant sex between early and late onset preeclampsia (p > 0.05 for each variable; Table 3). In the next stage, we compared the laboratory outcomes between early-and late-onset preeclampsia. There were no significant differences between groups in urea, uremic acid, creatinine and alanine and aspartate transaminase concentrations. Women in the EOPE group were characterized by a significantly higher level of proteinuria (369.96 ± 181.79 mg/dL vs. 194.37 ± 164.78 mg/dL, p < 0.001), higher 24-h urine proteinuria (4.86 ± 2.77 g/24 h vs. 2.49 ± 3.42 g/24 h, p = 0.0499) and lower levels of total protein in the blood compared with the LOPE group (5.47 ± 0.57 g/dL vs. 5.85 ± 0.78, p = 0.0049) ( Table 4).

Association Analysis of MMP1, MMP9, TIMP1 and TIMP2 Polymorphisms
For all SNPs, the genotype distribution in PE and controls was in the Hardy-Weinberg equilibrium (p > 0.05). No significant differences were observed in allele frequencies between preeclamptic women and controls ( Table 5). The largest differences in the frequency of alleles were observed for TIMP1 rs4898. Allele C was observed in 47% of LOPE women, 50.0% of controls, 53% of PE and 60.0% of the EOPE group (EOPE vs. control: OR = 1.5, p = 0.063, EOPE vs. LOPE: OR = 1.7, 95%CI = 1.02-2.89, p = 0.042, p cor . = 0.257). Table 6 presents the single-locus analysis for the six investigated SNPs within the MMP1, MMP9, TIMP1 and TIMP2 genes in PE and control women. None of the polymorphisms were associated with the PE risk under the codominant, dominant and recessive models (also after adjustment for pre-pregnancy BMI). Next we evaluated the associations among studied variants when the EOPE and LOPE groups were compared with each other and with the control group. Genotype distributions of polymorphism in EOPE and LOPE subtypes are given in Table 7. Analysis of genotype distributions revealed a non-significant difference for TIMP1 rs4898 polymorphism between the EOPE group and controls in the dominant model (CC vs. CT + TT) (0.665 vs.0.777, OR = 1.84, 95% CI: 0.96-3.53, p = 0.0706). Our data indicated no significant difference in the genotype frequencies of studied polymorphisms between late-onset preeclampsia and control groups. However, comparing EOPE and LOPE groups with each other, we observed a statistically significant difference between them for the TIMP1 rs4898 variant. In the recessive model the CC + CT genotype frequency was 85.5% in the EOPE and 70.0% in the LOPE subgroup (OR = 2.52, 95% CI: 0.99-6.39, p = 0.0452).

Haplotype and Gene-Gene Interaction Analysis
Haplotype Analysis of the MMP9 and TIMP2 Gene Polymorphisms The studied MMP9 and TIMP2 polymorphisms were further characterized using linkage disequilibrium (LD) and haplotype analyses. Interaction of these variants within haplotypes can be more informative than single-locus analysis. A linkage disequilibrium study showed for the MMP9 gene variants D' = 0.93 (r 2 = 0.29) and for TIMP2 variants D' = 0.91 (r 2 = 0.61).
The estimated haplotype frequencies for the two MMP9 and two TIMP2 polymorphisms are shown in Table 8. Similar to the genotype analysis, there were no significant differences between groups in the distributions of haplotype frequencies when PE, EOPE and LOPE were compared with the control group.
To explore epistatic interactions, we assessed all pairwise gene-gene interactions using the SNPassoc R CRAN package and MDR v3.0.2 software. Prominent gene-gene interactions were detected between TIMP1 (rs4898) and MMP9 (rs17577) under codominant (p = 0.0388) and over-dominant (p = 0.0145) models. An interaction was also detected between TIMP1 (rs4898) and MMP9 (rs17576) only under over dominant model (p = 0.0388). However, the p values of interactions were not significant after correction for multiple tests. The entropy-based interaction graph provided with MDR software shows the most significant gene-gene models for the four SNPs (rs17576, rs17577, rs4898, and rs17997500). The graph suggests that these interactions are interdependent, and synergism was observed between the polymorphisms of the TIMP1 and MMP9 genes (Figure 1).   detected between TIMP1 (rs4898) and MMP9 (rs17576) only under over dominant model (p = 0.0388). However, the p values of interactions were not significant after correction for multiple tests. The entropy-based interaction graph provided with MDR software shows the most significant gene-gene models for the four SNPs (rs17576, rs17577, rs4898, and  rs17997500). The graph suggests that these interactions are interdependent, and synergism was observed between the polymorphisms of the TIMP1 and MMP9 genes (Figure 1).

Association between Genotypes and Clinical Characteristics
In the next stage, we examined the association between genotypes and clinical data of the studied women. The TIMP1 gene variant rs4898 was associated in all study women with placenta weight. The mean placenta weight for the PE group was 423.53 ± 162.68 g (range 100.00-900.00 g) and for healthy subjects 611.30 ± 113.43 g (range 330.00-980.00 g) (p < 0.001).
In the whole group of 308 women, the mean weight of the placenta was the lowest in carriers of the rs4898 CC genotype (496.84 ± 145.45 g for CC genotype, 558.62 ± 162.30 g for CT and 562.15 ± 163.42 g for TT, respectively; p = 0.0172 in the codominant model and p = 0.0044 in the dominant model). Post hoc analysis revealed significant differences between CC-CT (p = 0.0209) and CC-TT (p = 0.0469). Additionally, during allele analysis, a statistically significant difference between the mean weights of the placenta (for C allele 529.32 ± 157.11 g, for T allele 560.24 ± 162.24 g, p = 0.021) was observed (Figure 2).

Association between Genotypes and Clinical Characteristics
In the next stage, we examined the association between genotypes and clinical data of the studied women. The TIMP1 gene variant rs4898 was associated in all study women with placenta weight. The mean placenta weight for the PE group was 423.53 ± 162.68 g (range 100.00-900.00 g) and for healthy subjects 611.30 ± 113.43 g (range 330.00-980.00 g) (p < 0.001).
In the whole group of 308 women, the mean weight of the placenta was the lowest in carriers of the rs4898 CC genotype (496.84 ± 145.45 g for CC genotype, 558.62 ± 162.30 g for CT and 562.15 ± 163.42 g for TT, respectively; p = 0.0172 in the codominant model and p = 0.0044 in the dominant model). Post hoc analysis revealed significant differences between CC-CT (p = 0.0209) and CC-TT (p = 0.0469). Additionally, during allele analysis, a statistically significant difference between the mean weights of the placenta (for C allele 529.32 ± 157.11 g, for T allele 560.24 ± 162.24 g, p = 0.021) was observed (Figure 2).

Discussion
Pre-eclampsia is a pregnancy-specific disease in which some genetic aspects may play a significant role in its development [27]. In the literature, various genetic polymorphisms related to pre-eclampsia have been described [28,29]. The MMPs play an important role in morphogenesis, angiogenesis and tissue repair, mainly through tissue

Discussion
Pre-eclampsia is a pregnancy-specific disease in which some genetic aspects may play a significant role in its development [27]. In the literature, various genetic polymorphisms related to pre-eclampsia have been described [28,29]. The MMPs play an important role in morphogenesis, angiogenesis and tissue repair, mainly through tissue remodeling. Tissue inhibitors of metalloproteinases (TIMPs) are endogenous inhibitors of MMPs and are consequently significant regulators of ECM turnover, as well as tissue remodeling. Dysregulation of the balance between MMPs and TIMPs is characteristic for some acute and chronic cardiovascular diseases [30]. Moreover, some authors have reported a relationship between MMP gene polymorphism and hypertensive disorders in pregnancy. However, most of them did not detect an association between different MMP variants and preeclampsia or gestational hypertension development.
The aim of the study by Rahimi et al. [31] was to investigate the possible influence of two polymorphisms-MMP7 −181A>G and MMP9 −1562C>T-on the risk of preeclampsia and lipid peroxidation level. The frequency of the MMP7 −181G allele was not significantly different between the groups with severe and mild preeclampsia as well as the healthy pregnant group. This study did not reveal a direct influence of MMP7 −181A>G polymorphism on the risk of preeclampsia. On the other hand, this polymorphism, through elevation of MDA level and interaction with MMP9 −1562C>T polymorphism, might be associated with the risk of severe preeclampsia [31].
Other research analyzed the relationship between MMP2 polymorphisms (−306C>T and −735C>T) and the response to antihypertensive treatment in a preeclamptic group, patients with gestational hypertension and healthy pregnant women. This study showed that MMP2 polymorphisms do not influence the development of hypertensive disorders of pregnancy and additionally do not affect the responsiveness to antihypertensive therapy in hypertensive disorders of pregnancy, especially preeclampsia [32].
In our study we also did not find any association between genetic polymorphism in genes encoding MMP1 and MMP9 and risk of preeclampsia. The genotype results obtained in our research confirm previous findings which show similar distribution of genetic variants in healthy pregnant and preeclampsia groups.
Fraser et al. reported the lack of a significant association of MMP9 −1562C>T polymorphism with preeclampsia risk and concluded that this MMP9 variant has little value as a screening marker for susceptibility to the syndrome [33].
Coolman et al. [14] reported that the frequency of the MMP9 −1562T allele was 0.06 in women with preeclampsia and 0.12 in healthy pregnant patients. Additionally, women carrying the T allele were less likely to develop preeclampsia than women with the CC genotype. However, they did not detect any association between presence of the T allele (CT + TT genotypes) and severe preeclampsia, eclampsia or HELLP syndrome [14]. Another study reported no statistically significant differences in the frequencies of the MMP1 genotypes between preeclamptic patients and a healthy pregnant control group (p = 0.9137). Allele frequencies did not exhibit any significant differences between study and control groups (p = 0.4210) or among subgroups (mild/severe preeclampsia, eclampsia or chronic hypertension with superimposed preeclampsia) [34].
The lack of significant associations between genotypes for the polymorphisms and preeclampsia suggests no association between susceptibility to this disease and genetic variations in MMP1, MMP9, TIMP1 and TIMP2. It might be suspected that not maternal but fetal genotype affects the susceptibility to preeclampsia, and this issue should be addressed in further studies.
In contrast to the above-mentioned results, Sun et al. [19] found in pregnant Chinese women that MMP9 −1562C>T genetic polymorphism might be associated with the development of pre-eclampsia. Women carrying the T allele were at higher risk of pre-eclampsia compared to women carrying the C allele (OR = 1.62) [19].
TIMP1 is a natural inhibitor of MMPs' activity and limits their proteolytic activities against the ECM. In consequence, it may affect the process of pregnancy vascular remodeling [35,36]. Moreover, genetic polymorphism of TIMP-1 has been linked with higher risk of developing certain diseases such as essential hypertension or intracerebral hemorrhage, which share certain risk factors with preeclampsia [13,37].
Our data did not reveal a significant difference in genotype frequencies of studied polymorphisms between LOPE/EOPE and control groups. However, what is very interesting is that, comparing subgroups EOPE and LOPE with each other, we observed a statistically significant difference between them in frequency of the TIMP1 rs4898 variant. The stratified analysis showed in the recessive model CC + CT genotype frequency 85.5% in the EOPE and 70.0% in the LOPE subgroup (OR = 2.52, p = 0.0452) which suggests that TIMP1 rs4898 could be associated with the form of preeclampsia. This observation might indicate that carriers of the TIMP1 rs4898 C allele represent a select group of patients who need, even in early pregnancy, close monitoring for the first symptoms of preeclampsia.
Additionally, in our analysis the same polymorphic variant TIMP1 rs4898 affected placental weight in all study women. In the whole group of 308 women, the lowest placenta weight was observed in carriers of TIMP1 rs4898 CC genotype (p = 0.0172 in the codominant model and p = 0.0044 in the dominant model). Recently, it was shown that TIMP1 polymorphism may possibly influence TIMP1 expression or modify its biological properties. Moreover, the TIMP1 rs4898 T allele reduces the activity of the TIMP1 protein, resulting in higher MMP activity, and provides normal angiogenesis and a normal placenta in carriers of this variant [13]. This may be associated with higher placental weights in carriers of the CT and TT genotypes compared to those of the CC genotype.
Due to the small number of studies concerning the involvement of TIMP genetic polymorphism in PE development, definite conclusions require future research. Nevertheless, the role of TIMP1 genetic polymorphisms in the etiopathogenesis of PE was demonstrated in the study of Luizon et al. [38]. The authors found that TIMP1 −9830T>G polymorphism is associated with preeclampsia (association with PE in carriers of the G allele as well as the GG genotype). Additionally, TIMP1 levels were higher in preeclamptic patients with the TIMP1 −9830TG genotype. Their data also suggested that this polymorphism may be implicated in poor responses to antihypertensive drugs [38].

Study Limitations
Obviously, there are some limitations of our study. Firstly, there is a small number of patients in the study group, but it should be noted that the sample was collected for a case-control study, which ensured the high quality of analysis. The post-hoc power analysis was performed using the overall study power (average of six tested variants) was underpowered, calculated at 62.0%. However, it is possible that studying a greater number of subjects would allow us to detect other significant associations in the MMP-TIMP pathway. Besides the sample size, the limitations of the research are also the restrictions due to race and region. Secondly, we focused on certain polymorphisms in the MMP-TIMP pathway, but it seems that the selected variants of the MMP1, MMP9, TIMP1, TIMP2 genes could play an important role in extracellular matrix remodeling. Consequently, an important future task is to establish the influence of MMP polymorphisms on enzyme activities as well as the exact interaction between MMPs and TIMPs.
On the other hand, it could be also interesting to determine, besides the analysis of genetic polymorphisms of the MMPs-TIMPs system, the expression level of these genes. Additionally, we have also consciousness that the preeclampsia may be related to the above genes in the fetus, thus it would be advisable to study these genes in the mother-child system as well.
Furthermore, we noted that there are differences between our results and those previously published by other authors. This could be due to the number and ethnicity of patients, the preeclampsia definition, or the methodology of MMP and TIMP detection.

Conclusions
The present results indicate a relationship between TIMP1 rs4898 (372T > C) polymorphism and increased risk of early-onset preeclampsia (EOPE) in a population of pregnant Polish women. Our data suggest that the TIMP1 rs4898 C allele might be associated with increased risk for early-onset, but not late-onset, preeclampsia. This finding could suggest that TIMP polymorphisms actively modulate ECM remodeling in the early stage of trophoblast invasion, and disturbances in this process result in preeclampsia development. The active participation of TIMP polymorphisms in ECM placenta remodeling was also confirmed by our secondary observations, indicating that the TIMP1 rs4898 C allele is associated with a lower placenta weight in the studied population of Polish women.
Although there is quite a lot of evidence for the importance of the MMP-TIMP pathway in trophoblast invasion, the aspects concerning the molecular mechanism remain unclear.
To evaluate the role of the TIMP1 polymorphic variants in the etiopathology of preeclampsia, further studies with larger sample sizes and in other populations are needed. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Written informed consent has been obtained from the patient(s) to publish this paper.