- Email: [email protected]
FGF1 and FGF2 mutations in preeclampsia and related features
Placenta 43 (2016) 81e85
Contents lists available at ScienceDirect
Placenta journal homepage: www.elsevier.com/locate/placenta
FGF1 and FGF2 mutations in preeclampsia and related features Ben Ali Gannoun Marwa a, b, *, Nozha Raguema a, b, Hedia Zitouni a, b, Hachani Ben Ali Feten c, Kacem Olfa c, Raja Elfeleh d, Wassim Almawi e, Touhami Mahjoub a a
Laboratory of Human Genome and Multifactorial Diseases (LR12ES07), Faculty of Pharmacy of Monastir, University of Monastir, Tunisia Faculty of Science of Bizerte, University of Carthage, Tunisia c University Hospital F. Hached, Department of Obstetrics and Gynaecology, Sousse, Tunisia d Centre of Maternity and Neonatology, Monastir, Tunisia e Department of Medical Biochemistry, College of Medicine and Medical Sciences, Arabian Gulf University, P.O. Box 22979, Manama, Bahrain b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 February 2016 Received in revised form 8 May 2016 Accepted 11 May 2016
Background: Fibroblast growth factor (FGF) 1 and FGF2 were previously linked with preeclampsia (PE), possibly through altering decidual and placental FGFR2 expression. Since common variation in FGF1 and FGF2 might influence FGF1 and FGF2 activity, this study evaluated whether common FGF1 and FGF2 variants are linked with PE and associated features. Methods: The association between FGF1 rs34011 and FGF2 rs2922979 SNPs and PE were tested in 300 women with PE, and 300 age-matched control women. Results: The allelic distribution of FGF1 rs34011 (P < 0.001) but not FGF2 rs2922979, variants were significantly different between PE cases and control women. Marginal association of FGF2 rs2922979 was seen after controlling for key covariates. Setting homozygous major allele genotype (1/1) as reference, significantly higher frequencies of heterozygous rs345011, and reduced frequency of heterozygous rs2922979 genotype carriers were seen in PE cases; the distribution of the remaining genotypes were comparable between cases and controls. Carriage of rs2922979 minor allele correlated with fasting glucose (P ¼ 0.02), while the presence of rs34011 minor allele was not correlated with PE-associated features. Conclusions: Our study suggests that the genetic variants of FGF1 rs34011, more so than FGF2 rs2922979, may play a role in PE pathogenesis in Tunisian women. These findings need confirmation in other ethnic populations. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Genotyping Preeclampsia FGF1 FGF2 SNPs Polymorphisms
1. Introduction Preeclampsia (PE), defined by hypertension and proteinuria [1], is a significant cause of maternal and fetal/neonatal morbidity and mortality [2,3], and its susceptibility is influenced by interactions between maternal and fetal factors [4e6]. While its exact physiopathology remains ill-defined, PE is initiated by placental factors that enter the maternal circulation. The pathogenesis of PE involves altered trophoblast cell activity, impaired placental perfusion, vascular endothelial injury, and oxidative stress [4], eventually
* Corresponding author. Laboratory of Human Genome and Multifactorial Diseases (LR12ES07), Faculty of Pharmacy of Monastir, University of Monastir, Street Ibn Sina, 5000, Monastir, Tunisia. E-mail address: [email protected] (B.A.G. Marwa). http://dx.doi.org/10.1016/j.placenta.2016.05.007 0143-4004/© 2016 Elsevier Ltd. All rights reserved.
resulting in endothelial dysfunction [7]. Insofar as angiogenesis is central to normal placental, fetal growth and development [8], several growth factors reportedly regulate angiogenesis [9], and in turn PE pathogenesis. These include acidic fibroblast growth factor (aFGF; FGF1), and basic fibroblast growth factor (bFGF; FGF2), both of which are widely expressed during embryonic development, and act by controlling neovascularization. FGF1 and FGF2 are prototypical cytokines belonging to the FGF family, which contains 22 members of heparin-binding, structurally related cytokines [10]. Although they share several properties, FGF-1 and FGF-2 differ by their mode of action. The expression of FGF1 and FGF2 vary in different tissues, and both play distinct roles during differentiation [10]. FGF1 is a key regulator of angiogenesis and tumorigenesis [11], and variations in FGF1 gene were associated with hypertension, which was attributed to upregulation of
82
B.A.G. Marwa et al. / Placenta 43 (2016) 81e85
FGF1 expression in glomerular mesangial and endothelial cells [12]. FGF1 stimulates the branching of the myocardial arteries, and stimulates the formation of new blood vessels, largely by preventing their regression. This prompted recommending inclusion of FGF1-based therapeutics in ischemic heart disease [13]. On the other hand, FGF2 stimulates angiogenesis in vivo, and is a pleiotropic regulator in vitro of the proliferation, migration, differentiation, and survival of many cell types, including endothelial cells, smooth muscle cells, and pericytes [14,15]. Several polymorphisms in FGF1 and FGF2 gene were identified, of which the FGF1 promoter rs34011 (-1385C/T), and the FGF2 intron-1 rs2922979 (754C/G), were the most investigated. Both variants alter the expression of FGF1 and FGF2 proteins, and were associated with several pathologies, including Alzheimer’s Disease, proliferative diabetic retinopathy (PDR), endometriosis (EM), and adenomyosis [16e18]. While it can be hypothesized that polymorphisms in FGF1 and FGF2 genes might contribute to PE development, the (likely) association of these variants to PE development was not previously investigated. We evaluate the association of FGF1 rs34011 and FGF2 rs2922979 with PE and associated features among Tunisian women with PE. 2. Subjects and methods 2.1. Study subjects This was a retrospective case-control study, involving 300 unrelated women with PE (mean age 31.3 ± 7.0 yr), who were recruited between May 2012 and June 2013 from the outpatient Gynecology service of Farhat Hached University Hospital (Sousse, Central Tunisia), Fattouma Bourguiba University Hospital (Monastir, Central Tunisia), Taher Sfar University Hospital (Mahdia, Eastern Tunisia), and Gafsa Hospital (Southern Tunisia). The inclusion criteria were PE during natural pregnancy, defined as gravidic hypertension, and assessed after 20 weeks of gestation as systolic blood pressure (BP) > 140 mmHg, diastolic BP > 90 mmHg, and/or rise in systolic BP > 30 mm, or diastolic BP > 15 mm Hg on at least two measurements, 6 h apart, along with significant proteinuria (>300 mg/24 h) or proteinuria of >2þ (determined by the dipstick method). While 76 PE cases (25.2%) developed severe early-onset PE form before 34 weeks of gestation according to these criteria, no cases of HELLP syndrome (hemolysis, elevated liver enzymes, low platelets) were recorded. The control group included 300 unrelated women with normal pregnancy (mean age 30.5 ± 5.8 yr), who were recruited from the same geographical area, with no known personal or family history of PE. Local ethics committees approved the study protocol, and both PE cases and control women gave written informed consent for their participation in the study. Demographic data of participants and clinical characteristics of patients are shown in Table 1. 2.2. FGF1/FGF2 genotyping Genomic DNA was extracted from EDTA anti-coagulated peripheral venous blood by the proteinase K/salting-out method. FGF1 -1385A/G (rs34011) and FGF2 754C/G (rs2922979) genotype analysis was performed by PCR-RFLP analysis. For FGF1 -1385A/G, DNA was amplified using the primers: (forward) 50 -TCA AGC AAT TCT CCT GCC TT-30 , and (reverse) 50 -CCA CTT CAA GGG ATT ATG GTG-30 , followed by an overnight digestion with HhaI (New England Biolabs, Ipswich, MA) at 37 C. The FGF2 754C/G polymorphism was assessed by PCR-RFLP analysis of the 334-bp PCR product, using the following primers: (forward) 50 -AAG CGG CTG TAC TGC AAA AAC-30 and (reverse) 50 -GGT ACT GGT TTA CAG GGC
AAA T-30 . After an overnight digestion with HhaI, FGF1 -1385A/G and FGF2 754C/G digested products were separated by electrophoresis on 3% and 4% agarose gel, respectively. For FGF1 -1385A/G, the A allele was represented by 53 and 302 bp DNA bands, while the G allele was identified as 53, 141 and 161 bp bands. FGF2 754C allele yielded 21, 38, 113 and 162 bp bands, whereas the G allele (which is not cleaved at the restriction site) yielded 38, 113 and 183 bp fragments. 2.3. Statistical analysis Data were analyzed using SPSS 20 (SPSS Inc., Chicago, IL, USA). Continuous data were expressed as mean ± SD, while categorical variables were presented as percent of total. Intergroup significance was determined by Student’s t-test (continuous data), and c2 test (categorical variables). Each polymorphism was tested for the HardyeWeinberg equilibrium (HWE) using c2 goodness-of-fit test using SNPStats software (http://bioinfo.iconcologia.net/snpstats/). HWE predicts that comparable allele ratios or rates will be seen in the controls and PE cases unless there is a form of selection for or against this. QUANTO (http://hydra.usc.edu/gxe/) was used in calculating the power, assuming 5% PE prevalence, and 100% genotype call rate. The PE predictors were evaluated by logistic regression analysis, first at the univariate, and later at the multivariate levels. The corresponding crude odds ratio (OR) and 95% confidence interval (95% CI), and adjusted OR (aOR) and 95% CI were calculated; the main covariates that were adjusted for were BMI, region, newborn weight, gestation age, and pregnancy status. Statistical significance was set at P < 0.05. 3. Results 3.1. Study subjects The demographic and clinical features of PE cases and control women are shown in Table 1. Women with PE were matched to controls with respect to age at examination, and delivery methods. Women with PE had significantly higher BMI, elevated systolic and diastolic BP, and gestational age at blood sampling. Significant differences between PE cases and control women were also noted with regards to the newborn weight, and to the pregnancy status. Accordingly BMI, regional origin, newborn weight, gestation age, and pregnancy status were selected as the covariates that were controlled for in subsequent analysis. 3.2. Association studies The association between FGF1 rs34011 and FGF2 rs2922979 and PE in case-control subjects is summarized Table 2. The tested SNPs were in HWE among study subjects. Minor allele frequency (MAF) of FGF1 rs34011, but not FGF2 rs2922979, was significantly different between PE cases and control women. Adjusting for key covariates resulted in marginal, but significant differences in MAF of FGF2 rs2922979 between PE cases and control women. Setting homozygous major allele genotype carrier as reference (OR ¼ 1.00), data from Table 2 demonstrated increased PE risk with FGF1 rs34011 genotypes. On the other hand, reduced frequency of heterozygous rs2922979 genotype carriers were seen in PE cases, thus suggesting a protective nature to this genotype. 3.3. Association analysis Multivariate logistic regression analysis models were used to investigate the relationship between carriage of FGF1 rs34011 and FGF2 rs2922979 minor allele and PE-associated features; separate
B.A.G. Marwa et al. / Placenta 43 (2016) 81e85
83
Table 1 Demographic and clinical characteristics of controls and patients. Characteristic
Controls (N ¼ 300)
Cases (N ¼ 300)
Pa
Age (years)b BMI (kg/m2)b Regionc
30.5 ± 5.8 28.6 ± 4.2 199 (66.3) 7 (2.3) 94 (31.3) 112.3 ± 9.3 68.8 ± 7.9 3253.3 ± 405.2 38.2 ± 3.0 188 (62.7) 112 (37.3) 164 (54.7) 131 (43.7) 5 (1.7)
31.3 ± 7.0 32.2 ± 5.0 234 (78.0) 33 (11.0) 33 (11.0) 155.4 ± 14.9 95.0 ± 8.7 2888.0 ± 755.7 35.8 ± 3.6 165 (55.) 135 (45.0) 75 (25.0) 222 (74.0) 3 (1.00)
0.121 <0.001 <0.001
Sahel region Central Tunisia Southern Tunisia Systolic Diastolic
Blood pressure (mmHg)b Newborn weight (g)b GA at blood samplingb Delivery methodc Pregnancy status
Vaginal delivery Cessarian sections Multiparous Primiparous Nulliparous
c
<0.001 <0.001 <0.001 <0.001 0.068 <0.001
BMI, body mass index; GA, gestational age. a Student t-test for continuous variables, and Pearson’s chi-square for categorical variables. b Mean ± SD. c Number (percent total).
Table 2 Allelic and genotypic frequencies of the studied polymorphisms. SNP rs2922979
rs34011
a b c
Casesa
Allele/genotype G C/C C/G G/G T C/C C/T T/T
112 210 68 22 314 86 98 116
(0.19) (70.0) (22.7) (7.3) (0.52) (28.7) (32.7) (38.7)
c
Controlsa
c2
P
OR (95% CI)
c2
P
aORb (95% CI)
138 185 92 23 293 131 61 108
3.42 4.63 5.81 0.18 9.36 18.23 11.71 4.97
0.064 0.074 0.016 0.587 2.2 103 1.0 104 0.001 0.056
0.77 (0.58e1.02) 1.00 (Reference) 0.65 (0.45e0.94) 0.84 (0.45e1.56) 1.43 (1.14e1.79) 1.00 (Reference) 1.64 (1.12e2.39) 1.54 (1.12e2.38)
6.159 0.46 6.49 0.02 12.50 18.46 12.20 2.86
0.046 0.40 0.56 0.97 0.002 1.9 104 0.019 0.412
0.82 (0.65e1.14) 1.00 (Reference) 0.74 (0.43e1.30) 1.32 (0.56e1.42) 1.90 (1.31e2.75) 1.00 (Reference) 1.59 (0.35e2.34) 4.17 (2.33e7.59)
(0.23) (61.7) (30.7) (7.7) (0.49) (43.7) (20.3) (36.0)
Study subjects included 300 PE cases and 300 control women. aOR ¼ adjusted OR; variables that were controlled for were BMI, region, new born weight, gestation age, and pregnancy status. Number of alleles/subjects (percent total).
4. Discussion
analyses were performed for each of the tested SNPs (Table 3). In the model including FGF2 rs2922979, carriage of the minor allele was positively associated with fasting glucose [P ¼ 0.02; OR (95% CI) ¼ 1.26 (1.05e1.53)]. In the model including FGF1 rs34011, there was no association between the minor allele with any of the examined PE features. Furthermore, in the model involving carriage of FGF1 rs34011 and FGF2 rs2922979, results from Table 4 show that the interaction between both SNPs appear antagonistic. This was based on the finding that carriage of FGF2 rs2922979 abolished the at-risk nature of FGF1 rs34011 (P ¼ 0.084 vs. P < 0.001).
Angiogenesis is controlled by stimulatory and inhibitory angiogenic factors, and disturbing this balance results in pathological changes [19], including PE [20]. While FGF1 and FGF2 are potent angiogenic factors, no study on genetic variability in FGF-1 and FGF-2 genes and PE were previously published. This caseecontrol study is the first to test the association between FGF2 (754C/G) and FGF1 (-1385A/G) gene variants and PE risk. Results obtained expand the contribution of both variants in conditions linked with angiogenesis. MAF of rs2922979 (0.23) and rs34011 (0.49) in Tunisians were
Table 3 Association analysis for the relationship between FGF1/FGF2 SNPs and PE features.a rs2922979
BMI Obesity Family history Systolic BP Diastolic BP Nephropathy Miscarriage C-section delivery Still birth Fasting glucose Creatinine Urea a
rs34011
c2
P
OR (95% CI)
c2
P
OR (95% CI)
0.09 0.17 0.08 1.11 1.14 0.43 0.26 0.02 0.01 6.72 1.22 2.48
0.76 0.70 0.80 0.29 0.28 1.00 0.69 1.00 1.00 0.02 0.30 0.11
1.01 0.90 1.07 1.12 0.83 0.70 1.15 0.97 1.06 1.26 0.99 1.14
0.03 0.05 0.06 0.15 0.80 0.40 0.05 1.22 0.65 0.81 0.23 0.70
0.86 0.90 0.90 0.70 0.37 1.00 0.89 0.31 0.41 0.37 0.63 0.41
1.01 0.95 1.07 1.04 0.86 0.71 1.06 0.75 0.65 1.07 1.00 0.93
Each SNP was analyzed under an (log-) additive genetic model.
(0.96e1.06) (0.54e1.49) (0.65e1.76) (0.91e1.37) (0.60e1.16) (0.65e1.75) (0.68e1.94) (0.59e1.59) (0.36e3.16) (1.05e1.53) (0.98e1.01) (0.97e1.33)
(0.95e1.06) (0.57e1.58) (0.65e1.76) (0.85e1.28) (0.61e1.20) (0.66e1.77) (0.62e1.82) (0.46e1.25) (0.23e1.86) (0.92e1.25) (0.99e1.01) (0.79e1.10)
84
B.A.G. Marwa et al. / Placenta 43 (2016) 81e85
Table 4 Combined effect of FGF1 and FGF2 alleles and PE features.a rs2922979
rs34011
Casesa
e þ
e
107 (0.82)
38 (0.44)
e þ
24 (0.18) 148 (0.69)
48 (0.56) 102 (0.60)
þ
66 (0.31)
67 (0.40)
e þ a b
Controlsa
c2
P
aORb (95% CI)
32.9
<0.001
0.18 (0.10e0.33)
0.084
0.68 (0.45e1.04)
3.23
Number (frequency) of PE cases and control women carrying the specific FGF1 and FGF2 genotypes. aOR ¼ adjusted OR; variables that were controlled for were BMI, region, new born weight, gestation age, and pregnancy status.
intermediate between Europeans and Africans, but higher than the frequencies seen in Asians (www.ncbi.nlm.nih.gov/projects/SNP/ snp). This reflects the genetic make up of present Tunisians, which is admixture of ancestral Africans and later Europeans. It is noteworthy that the association of rs2922979 and rs34011 was seen in heterozygous more so than homozygous carriers, thereby indicating no dose-related association. While the comparable FGF2 rs2922979 genotype and allele frequencies between PE cases and control women suggested no PE association, regression analyses demonstrated marginal association of rs2922979 with PE after controlling for key covariates. Carriage of rs2922979 minor allele positively associated with fasting glucose (the main energy for intrauterine growth), suggesting geneeenvironment interaction in determining PE risk. Although it lies in intron 1, rs2922979 was described as functional SNP, which acts by induction of structural changes in FGF2 mRNA, thus modulating post-transcriptional events. FGF2 rs2922979 was described as atrisk gene variant for proliferative diabetic retinopathy (PDR), as its minor allele was enriched in PDR than in non-diabetic subjects [17], which correlated with increased bFGF plasma level in PDR compared to non-PDR patients [17]. FGF1 is a potent angiogenic factor present in the extracellular matrix in association with heparan sulfate glycosaminoglycans (HSGAGs), and influences activity of several cell types, such as proliferation and differentiation [21]. Over-expression of FGF1 by glomerular endothelial and mesangial cells is a key regulator of glomerular microcirculation and systemic BP [12]. FGF1 gene is localized on chromosome 5q31.3e33.2 [22], and contains three protein-coding exons [23], and its transcription is controlled by at least four distinct promoters expressed in several cell types, including vascular smooth muscle cells [24e26]. It is tempting to speculate that -1385A/G polymorphism may function in controlling FGF-1.C promoter activity [16,18]. Enrichment of FGF1 rs34011 in PE indicates that it constitutes an at-risk marker for PE. While no similar studies were done on rs34011 in PE, carriage of T allele and T/T genotype imparted risk of Alzheimer’s disease in Japanese [16], but not in Chinese Han [27], and with psoriatic arthritis [28], endometriosis and adenomyosis [18]. The inconsistent findings between our results and those reported elsewhere are due to ethnic difference in the distribution of FGF variants, and predisposition to PE. The lack of association of rs2922979 and rs34011 with PE features suggests that both variants serve as genetic markers of PE, but not as predictors of its features. The interaction between FGF1 and FGF2 in PE appear to be antagonistic, since carriage of FGF2 rs2922979 minor allele abolished the at-risk nature of FGF1 rs34011. FGF-1 and FGF-2 possess a distinct, though HSGAGs-binding domain [29], which underscore the differential capacity of HSGAGs in regulating FGF signaling events, largely via cell membrane-bound tyrosine kinase receptors (FGFR). As such, the specificity of FGF-1 or FGF-2 to their unique FGFR appears to be the result of tissue-specific sequences [30,31]. Our study has several strengths, namely the homogeneity of the population tested (only Tunisian Arabs were included), which minimizes the problems of differences in genetic background
inherent in gene association studies, and that potential covariates were controlled for. Some limitations of this caseecontrol study relate to the study design (retrospective caseecontrol study), which questions the causeeeffect relationship, that it was not sufficiently powered, and that the study population was not sufficiently homogeneous in terms of PE manifestations (no severe and late-onset PE). An added limitation is the limited list of FGF SNPs investigated in this study, which raises the possibility that other FGF SNPs may also contribute to PE pathogenesis. Follow-up studies are needed to define the contribution of FGF variants in the PE pathogenesis and associated features. Disclosure of interests None of the authors have any competing conflict of interest to declare. Conflicts of interest statement None of the authors declare any actual or potential competing conflict of interest including financial, personal or other relationships with other people or organizations within three years of beginning the submitted work that could inappropriately influence, or be perceived to influence, their work. Contribution to authorship - Marwa Ben Ali Gannoun: sample recruitment, technical support, drafting of the manuscript. - Nozha Raguema: sample recruitment - Hedia Zitouni: sample recruitment - Feten Hachani Ben Ali: sample recruitment - Olfa Kacem: sample recruitment - Raja Elfeleh: sample recruitment - Wassim Almawi: Statistical analysis, editing of the manuscript. - Touhami Mahjoub: Project leader.
Details of ethics approval The ethics Committee of the hospital Farhed Hached Sousse chaired by Dr. Tasnim Masmoudi approved the study protocol on 2012. References [1] American College of Obstetricians and Gynecologists, Task force on hypertension in pregnancy. Hypertension in pregnancy. Report of the American College of obstetricians and gynecologists’ task force on hypertension in pregnancy, Obstet. Gynecol. 122 (2013) 1122e1131. [2] R.A.K. El Desouki, F.A. Habib, Preeclampsia at molecular level, Appl. Med. Res. 2 (2015) 65e72. [3] M. Emanuel, S. Butt, Frequency and factors leading to recurrent pre-eclampsia, J. Pak. Med. Assoc. 65 (2015) 1173e1177. [4] H.E. Yong, P.E. Melton, M.P. Johnson, K.A. Freed, B. Kalionis, P. Murthi, et al., Genome-wide transcriptome directed pathway analysis of maternal pre-
B.A.G. Marwa et al. / Placenta 43 (2016) 81e85 eclampsia susceptibility genes, PLoS One 10 (2015) 012e230. [5] E.J. Crosley, M.G. Elliot, J.K. Christians, B.J. Crespi, Placental invasion, preeclampsia risk and adaptive molecular evolution at the origin of the great apes: evidence from genome-wide analyses, Placenta 34 (2013) 127e132. [6] H. Salonen Ros, P. Lichtenstein, L. Lipworth, S. Cnattingius, Genetic effects on the liability of developing pre-eclampsia and gestational hypertension, Am. J. Med. Genet. 91 (2000) 256e260. [7] D.V. Antonio, B. Giovanni, G. Domenico, T. Giovanna, C. Francesco, D.A. Rosario, Endoglin, PlGFand sFlt-1as markers for predicting pre-eclampsia, Acta Obstetricia Gynecol. 87 (2008) 837e842. [8] R. Romero, J.K. Nien, J. Espinoza, D. Todem, W. Fu, H. Chung, J.P. Kusanovic, F. Gotsch, O. Erez, S. Mazaki- Tovi, et al., A longitudinal study of angiogenic (placental growth factor) and anti-angiogenic (soluble endoglin and soluble vascular endothelial growth factor receptor-1) factors in normal pregnancy and patients destined to develop preeclampsia and deliver a small for gestational age neonate, J. Matern. Fetal Neonatal Med. 10 (2008) 9e23. [9] H.P. Gerber, N. Ferrara, Angiogenesis and bone growth, Trends cardiovasc. Med. 10 (2000) 223e228. [10] G. Szebenyi, J.F. Fallon, Fibroblast growth factors as multifunctional signaling factors, Int. Rev. Cytol. 185 (1999) 45e106. [11] M. Relf, S. LeJeune, P.A. Scott, S. Fox, K. Smith, R. Leek, A. Moghaddam, R. Whitehouse, R. Bicknell, A.L. Harris, Expression of the angiogenic factors vascular endothelial cell growth factor, acidic and basic fibroblast growth factor, tumor growth factor beta-1, platelet-derived endothelial cell growth factor, placenta growth factor, and pleiotrophin in human primary breast cancer and its relation to angiogenesis, Cancer Res. 57 (1997) 963e969. [12] T. Maciej, J.C. Fadi, D.L. Mark, P. Sandosh, Y.S. William, H.M. William, G.M. Wladyslaw, M. Christine, Z.S. Ewa, F. Anna, M.D. Dominiczak, Fibroblast growth factor 1 gene and hypertension. From the quantitative trait locus to positional analysis, Circulation 116 (2007) 1915e1924. [13] P. Carmeliet, Fibroblast growth factor-1 stimulates branching and survival of myocardial arteries: a goal for therapeutic angiogenesis? Circ. Res. 87 (2000) 176e178. [14] S. Ortega, M. AIttmann, S.H. Tsang, M. Ehrlich, C. Basilico, Neuronal defects and delayed wound healing in mice lacking fibroblast growth factor 2, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 5672e5677. [15] P.K. Bos, G.J. van Osch, D.A. Frenz, J.A. Verhaar, H.L. Verwoerd-Verhoef, Growth factor expression in cartilage wound healing: temporal and spatial immunolocalization in a rabbit auricular cartilage wound model, Osteoarthr. Cartil. 9 (2001) 382e389. [16] H. Yamagata, Y. Chen, H. Akatsu, K. Kamino, J. Ito, S. Yokoyama, T. Yamamoto, K. Kosaka, T. Miki, I. Kondo, Promoter polymorphism in fibroblast growth factor 1 gene increases risk of definite Alzheimer’s disease, Biochem. Biophys. Res. Commun. 321 (2004) 320e323. [17] M. Beranek, P. Kolar, S. Tschoplova, K. Kankova, A. Vasku, Genetic variation and plasma level of the basic fibroblast growth factor in proliferative diabetic
85
retinopathy, Diabetes Res. Clin. Pract. 79 (2008) 362e367. [18] Kang S1, S.Z. Li, N. Wang, R.M. Zhou, T. Wang, D.J. Wang, X.F. Li, J. Bui, Y. Li, Association between genetic polymorphisms in fibroblast growth factor (FGF) 1 and FGF2 and risk of endometriosis and adenomyosis in Chinese women, Hum. Reprod. 25 (2010) 1806e1811. [19] H. Brian, M.D. Annex, Therapeutic angiogenesis: a treatment for the new millennium or passing fad? Cardiol. Rounds 6 (6) (2002) 1e3. [20] A. Bikfalvi, Significance of angiogenesis in tumour progression and metastasis, Eur. J. Cancer 31 (9) (1995) 1101e1104. [21] M. Jaye, J. Schlessinger, C.A. Dionne, Fibroblast growth factor receptor tyrosine kinases: molecular analysis and signal transduction, Biochim. Biophys. Acta 1135 (1992) 185e199. [22] K. Huebner, L. Nagarajan, E. Besa, E. Angert, B.J. Lange, L.A. Cannizzaro, H. van den Berghe, D. Santoli, J. Finan, C.M. Croce, et al., Order of genes on human chromosome 5q with respect to 5q interstitial deletions, Am. J. Hum. Genet. 46 (1990) 26e36. [23] R.A. Payson, M.A. Chotani, I.M. Chiu, Regulation of a promoter of the fibroblast growth factor 1 gene in prostate and breast cancer cells, J. Steroid Biochem. Mol. Biol. 66 (1998) 93e103. [24] R.L. Myers, R.A. Payson, M.A. Chotani, L.L. Deaven, I.M. Chiu, Gene structure and differential expression of acidic fibroblast growth factor mRNA: identification and distribution of four different transcripts, Oncogene 8.6 (1993) 341e349. [25] I.M. Chiu, K. Touhalisky, C. Baran, Multiple controlling mechanisms of FGF1 gene expression through multiple tissue-specific promoters, Prog. Nucleic Acid. Res. Mol. Biol. 70 (2000) 155e174. [26] R.A. Payson, M.A. Chotani, I.M. Chiu, Regulation of a promoter of the fibroblast growth factor 1 gene in prostate and breast cancer cells, J. Steroid Biochem. Mol. Biol. 66 (1998) 93e103. [27] T.B. Jian, L.Z. Hua, X.Z. Zhen, H.B. Xiu, W.Z. Jun, No association of the C > T polymorphism that is located 1385 upstream from initial code of fibroblast growth factor 1 gene with Alzheimer’s disease in Chinese, Brain Res. 1328 (2010) 113e117. [28] C. Butt, S. Lim, C. Greenwood, P. Rahman, VEGF, FGF1, FGF2 and EGF gene polymorphisms and psoriatic arthritis, BMC Musculoskelet. Disord. 4 (2007), 4e1. [29] D.M. Ornitz, N. Itoh, The fibroblast growth factor signaling pathway, Wiley Interdiscip. Rev. Dev. Biol. 4 (2015) 215e266. [30] J. Kreuger, M. Salmivirta, L. Sturiale, G. Gimenez-Gallego, U. Lindahl, Sequence analysis of heparan sulfate epitopes with graded affinities for fibroblast growth factors 1 and 2, J. Biol. Chem. 276 (2001) 30744e30752. [31] A.K. Powell, D.G. Fernig, J.E. Turnbull, Fibroblast growth factor receptors 1 and 2 interact differently with heparin/heparan sulfate. Implications for dynamic assembly of a ternary signaling complex, J. Biol. Chem. 277 (2002) 28554e28563.
Contents lists available at ScienceDirect
Placenta journal homepage: www.elsevier.com/locate/placenta
FGF1 and FGF2 mutations in preeclampsia and related features Ben Ali Gannoun Marwa a, b, *, Nozha Raguema a, b, Hedia Zitouni a, b, Hachani Ben Ali Feten c, Kacem Olfa c, Raja Elfeleh d, Wassim Almawi e, Touhami Mahjoub a a
Laboratory of Human Genome and Multifactorial Diseases (LR12ES07), Faculty of Pharmacy of Monastir, University of Monastir, Tunisia Faculty of Science of Bizerte, University of Carthage, Tunisia c University Hospital F. Hached, Department of Obstetrics and Gynaecology, Sousse, Tunisia d Centre of Maternity and Neonatology, Monastir, Tunisia e Department of Medical Biochemistry, College of Medicine and Medical Sciences, Arabian Gulf University, P.O. Box 22979, Manama, Bahrain b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 February 2016 Received in revised form 8 May 2016 Accepted 11 May 2016
Background: Fibroblast growth factor (FGF) 1 and FGF2 were previously linked with preeclampsia (PE), possibly through altering decidual and placental FGFR2 expression. Since common variation in FGF1 and FGF2 might influence FGF1 and FGF2 activity, this study evaluated whether common FGF1 and FGF2 variants are linked with PE and associated features. Methods: The association between FGF1 rs34011 and FGF2 rs2922979 SNPs and PE were tested in 300 women with PE, and 300 age-matched control women. Results: The allelic distribution of FGF1 rs34011 (P < 0.001) but not FGF2 rs2922979, variants were significantly different between PE cases and control women. Marginal association of FGF2 rs2922979 was seen after controlling for key covariates. Setting homozygous major allele genotype (1/1) as reference, significantly higher frequencies of heterozygous rs345011, and reduced frequency of heterozygous rs2922979 genotype carriers were seen in PE cases; the distribution of the remaining genotypes were comparable between cases and controls. Carriage of rs2922979 minor allele correlated with fasting glucose (P ¼ 0.02), while the presence of rs34011 minor allele was not correlated with PE-associated features. Conclusions: Our study suggests that the genetic variants of FGF1 rs34011, more so than FGF2 rs2922979, may play a role in PE pathogenesis in Tunisian women. These findings need confirmation in other ethnic populations. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Genotyping Preeclampsia FGF1 FGF2 SNPs Polymorphisms
1. Introduction Preeclampsia (PE), defined by hypertension and proteinuria [1], is a significant cause of maternal and fetal/neonatal morbidity and mortality [2,3], and its susceptibility is influenced by interactions between maternal and fetal factors [4e6]. While its exact physiopathology remains ill-defined, PE is initiated by placental factors that enter the maternal circulation. The pathogenesis of PE involves altered trophoblast cell activity, impaired placental perfusion, vascular endothelial injury, and oxidative stress [4], eventually
* Corresponding author. Laboratory of Human Genome and Multifactorial Diseases (LR12ES07), Faculty of Pharmacy of Monastir, University of Monastir, Street Ibn Sina, 5000, Monastir, Tunisia. E-mail address: [email protected] (B.A.G. Marwa). http://dx.doi.org/10.1016/j.placenta.2016.05.007 0143-4004/© 2016 Elsevier Ltd. All rights reserved.
resulting in endothelial dysfunction [7]. Insofar as angiogenesis is central to normal placental, fetal growth and development [8], several growth factors reportedly regulate angiogenesis [9], and in turn PE pathogenesis. These include acidic fibroblast growth factor (aFGF; FGF1), and basic fibroblast growth factor (bFGF; FGF2), both of which are widely expressed during embryonic development, and act by controlling neovascularization. FGF1 and FGF2 are prototypical cytokines belonging to the FGF family, which contains 22 members of heparin-binding, structurally related cytokines [10]. Although they share several properties, FGF-1 and FGF-2 differ by their mode of action. The expression of FGF1 and FGF2 vary in different tissues, and both play distinct roles during differentiation [10]. FGF1 is a key regulator of angiogenesis and tumorigenesis [11], and variations in FGF1 gene were associated with hypertension, which was attributed to upregulation of
82
B.A.G. Marwa et al. / Placenta 43 (2016) 81e85
FGF1 expression in glomerular mesangial and endothelial cells [12]. FGF1 stimulates the branching of the myocardial arteries, and stimulates the formation of new blood vessels, largely by preventing their regression. This prompted recommending inclusion of FGF1-based therapeutics in ischemic heart disease [13]. On the other hand, FGF2 stimulates angiogenesis in vivo, and is a pleiotropic regulator in vitro of the proliferation, migration, differentiation, and survival of many cell types, including endothelial cells, smooth muscle cells, and pericytes [14,15]. Several polymorphisms in FGF1 and FGF2 gene were identified, of which the FGF1 promoter rs34011 (-1385C/T), and the FGF2 intron-1 rs2922979 (754C/G), were the most investigated. Both variants alter the expression of FGF1 and FGF2 proteins, and were associated with several pathologies, including Alzheimer’s Disease, proliferative diabetic retinopathy (PDR), endometriosis (EM), and adenomyosis [16e18]. While it can be hypothesized that polymorphisms in FGF1 and FGF2 genes might contribute to PE development, the (likely) association of these variants to PE development was not previously investigated. We evaluate the association of FGF1 rs34011 and FGF2 rs2922979 with PE and associated features among Tunisian women with PE. 2. Subjects and methods 2.1. Study subjects This was a retrospective case-control study, involving 300 unrelated women with PE (mean age 31.3 ± 7.0 yr), who were recruited between May 2012 and June 2013 from the outpatient Gynecology service of Farhat Hached University Hospital (Sousse, Central Tunisia), Fattouma Bourguiba University Hospital (Monastir, Central Tunisia), Taher Sfar University Hospital (Mahdia, Eastern Tunisia), and Gafsa Hospital (Southern Tunisia). The inclusion criteria were PE during natural pregnancy, defined as gravidic hypertension, and assessed after 20 weeks of gestation as systolic blood pressure (BP) > 140 mmHg, diastolic BP > 90 mmHg, and/or rise in systolic BP > 30 mm, or diastolic BP > 15 mm Hg on at least two measurements, 6 h apart, along with significant proteinuria (>300 mg/24 h) or proteinuria of >2þ (determined by the dipstick method). While 76 PE cases (25.2%) developed severe early-onset PE form before 34 weeks of gestation according to these criteria, no cases of HELLP syndrome (hemolysis, elevated liver enzymes, low platelets) were recorded. The control group included 300 unrelated women with normal pregnancy (mean age 30.5 ± 5.8 yr), who were recruited from the same geographical area, with no known personal or family history of PE. Local ethics committees approved the study protocol, and both PE cases and control women gave written informed consent for their participation in the study. Demographic data of participants and clinical characteristics of patients are shown in Table 1. 2.2. FGF1/FGF2 genotyping Genomic DNA was extracted from EDTA anti-coagulated peripheral venous blood by the proteinase K/salting-out method. FGF1 -1385A/G (rs34011) and FGF2 754C/G (rs2922979) genotype analysis was performed by PCR-RFLP analysis. For FGF1 -1385A/G, DNA was amplified using the primers: (forward) 50 -TCA AGC AAT TCT CCT GCC TT-30 , and (reverse) 50 -CCA CTT CAA GGG ATT ATG GTG-30 , followed by an overnight digestion with HhaI (New England Biolabs, Ipswich, MA) at 37 C. The FGF2 754C/G polymorphism was assessed by PCR-RFLP analysis of the 334-bp PCR product, using the following primers: (forward) 50 -AAG CGG CTG TAC TGC AAA AAC-30 and (reverse) 50 -GGT ACT GGT TTA CAG GGC
AAA T-30 . After an overnight digestion with HhaI, FGF1 -1385A/G and FGF2 754C/G digested products were separated by electrophoresis on 3% and 4% agarose gel, respectively. For FGF1 -1385A/G, the A allele was represented by 53 and 302 bp DNA bands, while the G allele was identified as 53, 141 and 161 bp bands. FGF2 754C allele yielded 21, 38, 113 and 162 bp bands, whereas the G allele (which is not cleaved at the restriction site) yielded 38, 113 and 183 bp fragments. 2.3. Statistical analysis Data were analyzed using SPSS 20 (SPSS Inc., Chicago, IL, USA). Continuous data were expressed as mean ± SD, while categorical variables were presented as percent of total. Intergroup significance was determined by Student’s t-test (continuous data), and c2 test (categorical variables). Each polymorphism was tested for the HardyeWeinberg equilibrium (HWE) using c2 goodness-of-fit test using SNPStats software (http://bioinfo.iconcologia.net/snpstats/). HWE predicts that comparable allele ratios or rates will be seen in the controls and PE cases unless there is a form of selection for or against this. QUANTO (http://hydra.usc.edu/gxe/) was used in calculating the power, assuming 5% PE prevalence, and 100% genotype call rate. The PE predictors were evaluated by logistic regression analysis, first at the univariate, and later at the multivariate levels. The corresponding crude odds ratio (OR) and 95% confidence interval (95% CI), and adjusted OR (aOR) and 95% CI were calculated; the main covariates that were adjusted for were BMI, region, newborn weight, gestation age, and pregnancy status. Statistical significance was set at P < 0.05. 3. Results 3.1. Study subjects The demographic and clinical features of PE cases and control women are shown in Table 1. Women with PE were matched to controls with respect to age at examination, and delivery methods. Women with PE had significantly higher BMI, elevated systolic and diastolic BP, and gestational age at blood sampling. Significant differences between PE cases and control women were also noted with regards to the newborn weight, and to the pregnancy status. Accordingly BMI, regional origin, newborn weight, gestation age, and pregnancy status were selected as the covariates that were controlled for in subsequent analysis. 3.2. Association studies The association between FGF1 rs34011 and FGF2 rs2922979 and PE in case-control subjects is summarized Table 2. The tested SNPs were in HWE among study subjects. Minor allele frequency (MAF) of FGF1 rs34011, but not FGF2 rs2922979, was significantly different between PE cases and control women. Adjusting for key covariates resulted in marginal, but significant differences in MAF of FGF2 rs2922979 between PE cases and control women. Setting homozygous major allele genotype carrier as reference (OR ¼ 1.00), data from Table 2 demonstrated increased PE risk with FGF1 rs34011 genotypes. On the other hand, reduced frequency of heterozygous rs2922979 genotype carriers were seen in PE cases, thus suggesting a protective nature to this genotype. 3.3. Association analysis Multivariate logistic regression analysis models were used to investigate the relationship between carriage of FGF1 rs34011 and FGF2 rs2922979 minor allele and PE-associated features; separate
B.A.G. Marwa et al. / Placenta 43 (2016) 81e85
83
Table 1 Demographic and clinical characteristics of controls and patients. Characteristic
Controls (N ¼ 300)
Cases (N ¼ 300)
Pa
Age (years)b BMI (kg/m2)b Regionc
30.5 ± 5.8 28.6 ± 4.2 199 (66.3) 7 (2.3) 94 (31.3) 112.3 ± 9.3 68.8 ± 7.9 3253.3 ± 405.2 38.2 ± 3.0 188 (62.7) 112 (37.3) 164 (54.7) 131 (43.7) 5 (1.7)
31.3 ± 7.0 32.2 ± 5.0 234 (78.0) 33 (11.0) 33 (11.0) 155.4 ± 14.9 95.0 ± 8.7 2888.0 ± 755.7 35.8 ± 3.6 165 (55.) 135 (45.0) 75 (25.0) 222 (74.0) 3 (1.00)
0.121 <0.001 <0.001
Sahel region Central Tunisia Southern Tunisia Systolic Diastolic
Blood pressure (mmHg)b Newborn weight (g)b GA at blood samplingb Delivery methodc Pregnancy status
Vaginal delivery Cessarian sections Multiparous Primiparous Nulliparous
c
<0.001 <0.001 <0.001 <0.001 0.068 <0.001
BMI, body mass index; GA, gestational age. a Student t-test for continuous variables, and Pearson’s chi-square for categorical variables. b Mean ± SD. c Number (percent total).
Table 2 Allelic and genotypic frequencies of the studied polymorphisms. SNP rs2922979
rs34011
a b c
Casesa
Allele/genotype G C/C C/G G/G T C/C C/T T/T
112 210 68 22 314 86 98 116
(0.19) (70.0) (22.7) (7.3) (0.52) (28.7) (32.7) (38.7)
c
Controlsa
c2
P
OR (95% CI)
c2
P
aORb (95% CI)
138 185 92 23 293 131 61 108
3.42 4.63 5.81 0.18 9.36 18.23 11.71 4.97
0.064 0.074 0.016 0.587 2.2 103 1.0 104 0.001 0.056
0.77 (0.58e1.02) 1.00 (Reference) 0.65 (0.45e0.94) 0.84 (0.45e1.56) 1.43 (1.14e1.79) 1.00 (Reference) 1.64 (1.12e2.39) 1.54 (1.12e2.38)
6.159 0.46 6.49 0.02 12.50 18.46 12.20 2.86
0.046 0.40 0.56 0.97 0.002 1.9 104 0.019 0.412
0.82 (0.65e1.14) 1.00 (Reference) 0.74 (0.43e1.30) 1.32 (0.56e1.42) 1.90 (1.31e2.75) 1.00 (Reference) 1.59 (0.35e2.34) 4.17 (2.33e7.59)
(0.23) (61.7) (30.7) (7.7) (0.49) (43.7) (20.3) (36.0)
Study subjects included 300 PE cases and 300 control women. aOR ¼ adjusted OR; variables that were controlled for were BMI, region, new born weight, gestation age, and pregnancy status. Number of alleles/subjects (percent total).
4. Discussion
analyses were performed for each of the tested SNPs (Table 3). In the model including FGF2 rs2922979, carriage of the minor allele was positively associated with fasting glucose [P ¼ 0.02; OR (95% CI) ¼ 1.26 (1.05e1.53)]. In the model including FGF1 rs34011, there was no association between the minor allele with any of the examined PE features. Furthermore, in the model involving carriage of FGF1 rs34011 and FGF2 rs2922979, results from Table 4 show that the interaction between both SNPs appear antagonistic. This was based on the finding that carriage of FGF2 rs2922979 abolished the at-risk nature of FGF1 rs34011 (P ¼ 0.084 vs. P < 0.001).
Angiogenesis is controlled by stimulatory and inhibitory angiogenic factors, and disturbing this balance results in pathological changes [19], including PE [20]. While FGF1 and FGF2 are potent angiogenic factors, no study on genetic variability in FGF-1 and FGF-2 genes and PE were previously published. This caseecontrol study is the first to test the association between FGF2 (754C/G) and FGF1 (-1385A/G) gene variants and PE risk. Results obtained expand the contribution of both variants in conditions linked with angiogenesis. MAF of rs2922979 (0.23) and rs34011 (0.49) in Tunisians were
Table 3 Association analysis for the relationship between FGF1/FGF2 SNPs and PE features.a rs2922979
BMI Obesity Family history Systolic BP Diastolic BP Nephropathy Miscarriage C-section delivery Still birth Fasting glucose Creatinine Urea a
rs34011
c2
P
OR (95% CI)
c2
P
OR (95% CI)
0.09 0.17 0.08 1.11 1.14 0.43 0.26 0.02 0.01 6.72 1.22 2.48
0.76 0.70 0.80 0.29 0.28 1.00 0.69 1.00 1.00 0.02 0.30 0.11
1.01 0.90 1.07 1.12 0.83 0.70 1.15 0.97 1.06 1.26 0.99 1.14
0.03 0.05 0.06 0.15 0.80 0.40 0.05 1.22 0.65 0.81 0.23 0.70
0.86 0.90 0.90 0.70 0.37 1.00 0.89 0.31 0.41 0.37 0.63 0.41
1.01 0.95 1.07 1.04 0.86 0.71 1.06 0.75 0.65 1.07 1.00 0.93
Each SNP was analyzed under an (log-) additive genetic model.
(0.96e1.06) (0.54e1.49) (0.65e1.76) (0.91e1.37) (0.60e1.16) (0.65e1.75) (0.68e1.94) (0.59e1.59) (0.36e3.16) (1.05e1.53) (0.98e1.01) (0.97e1.33)
(0.95e1.06) (0.57e1.58) (0.65e1.76) (0.85e1.28) (0.61e1.20) (0.66e1.77) (0.62e1.82) (0.46e1.25) (0.23e1.86) (0.92e1.25) (0.99e1.01) (0.79e1.10)
84
B.A.G. Marwa et al. / Placenta 43 (2016) 81e85
Table 4 Combined effect of FGF1 and FGF2 alleles and PE features.a rs2922979
rs34011
Casesa
e þ
e
107 (0.82)
38 (0.44)
e þ
24 (0.18) 148 (0.69)
48 (0.56) 102 (0.60)
þ
66 (0.31)
67 (0.40)
e þ a b
Controlsa
c2
P
aORb (95% CI)
32.9
<0.001
0.18 (0.10e0.33)
0.084
0.68 (0.45e1.04)
3.23
Number (frequency) of PE cases and control women carrying the specific FGF1 and FGF2 genotypes. aOR ¼ adjusted OR; variables that were controlled for were BMI, region, new born weight, gestation age, and pregnancy status.
intermediate between Europeans and Africans, but higher than the frequencies seen in Asians (www.ncbi.nlm.nih.gov/projects/SNP/ snp). This reflects the genetic make up of present Tunisians, which is admixture of ancestral Africans and later Europeans. It is noteworthy that the association of rs2922979 and rs34011 was seen in heterozygous more so than homozygous carriers, thereby indicating no dose-related association. While the comparable FGF2 rs2922979 genotype and allele frequencies between PE cases and control women suggested no PE association, regression analyses demonstrated marginal association of rs2922979 with PE after controlling for key covariates. Carriage of rs2922979 minor allele positively associated with fasting glucose (the main energy for intrauterine growth), suggesting geneeenvironment interaction in determining PE risk. Although it lies in intron 1, rs2922979 was described as functional SNP, which acts by induction of structural changes in FGF2 mRNA, thus modulating post-transcriptional events. FGF2 rs2922979 was described as atrisk gene variant for proliferative diabetic retinopathy (PDR), as its minor allele was enriched in PDR than in non-diabetic subjects [17], which correlated with increased bFGF plasma level in PDR compared to non-PDR patients [17]. FGF1 is a potent angiogenic factor present in the extracellular matrix in association with heparan sulfate glycosaminoglycans (HSGAGs), and influences activity of several cell types, such as proliferation and differentiation [21]. Over-expression of FGF1 by glomerular endothelial and mesangial cells is a key regulator of glomerular microcirculation and systemic BP [12]. FGF1 gene is localized on chromosome 5q31.3e33.2 [22], and contains three protein-coding exons [23], and its transcription is controlled by at least four distinct promoters expressed in several cell types, including vascular smooth muscle cells [24e26]. It is tempting to speculate that -1385A/G polymorphism may function in controlling FGF-1.C promoter activity [16,18]. Enrichment of FGF1 rs34011 in PE indicates that it constitutes an at-risk marker for PE. While no similar studies were done on rs34011 in PE, carriage of T allele and T/T genotype imparted risk of Alzheimer’s disease in Japanese [16], but not in Chinese Han [27], and with psoriatic arthritis [28], endometriosis and adenomyosis [18]. The inconsistent findings between our results and those reported elsewhere are due to ethnic difference in the distribution of FGF variants, and predisposition to PE. The lack of association of rs2922979 and rs34011 with PE features suggests that both variants serve as genetic markers of PE, but not as predictors of its features. The interaction between FGF1 and FGF2 in PE appear to be antagonistic, since carriage of FGF2 rs2922979 minor allele abolished the at-risk nature of FGF1 rs34011. FGF-1 and FGF-2 possess a distinct, though HSGAGs-binding domain [29], which underscore the differential capacity of HSGAGs in regulating FGF signaling events, largely via cell membrane-bound tyrosine kinase receptors (FGFR). As such, the specificity of FGF-1 or FGF-2 to their unique FGFR appears to be the result of tissue-specific sequences [30,31]. Our study has several strengths, namely the homogeneity of the population tested (only Tunisian Arabs were included), which minimizes the problems of differences in genetic background
inherent in gene association studies, and that potential covariates were controlled for. Some limitations of this caseecontrol study relate to the study design (retrospective caseecontrol study), which questions the causeeeffect relationship, that it was not sufficiently powered, and that the study population was not sufficiently homogeneous in terms of PE manifestations (no severe and late-onset PE). An added limitation is the limited list of FGF SNPs investigated in this study, which raises the possibility that other FGF SNPs may also contribute to PE pathogenesis. Follow-up studies are needed to define the contribution of FGF variants in the PE pathogenesis and associated features. Disclosure of interests None of the authors have any competing conflict of interest to declare. Conflicts of interest statement None of the authors declare any actual or potential competing conflict of interest including financial, personal or other relationships with other people or organizations within three years of beginning the submitted work that could inappropriately influence, or be perceived to influence, their work. Contribution to authorship - Marwa Ben Ali Gannoun: sample recruitment, technical support, drafting of the manuscript. - Nozha Raguema: sample recruitment - Hedia Zitouni: sample recruitment - Feten Hachani Ben Ali: sample recruitment - Olfa Kacem: sample recruitment - Raja Elfeleh: sample recruitment - Wassim Almawi: Statistical analysis, editing of the manuscript. - Touhami Mahjoub: Project leader.
Details of ethics approval The ethics Committee of the hospital Farhed Hached Sousse chaired by Dr. Tasnim Masmoudi approved the study protocol on 2012. References [1] American College of Obstetricians and Gynecologists, Task force on hypertension in pregnancy. Hypertension in pregnancy. Report of the American College of obstetricians and gynecologists’ task force on hypertension in pregnancy, Obstet. Gynecol. 122 (2013) 1122e1131. [2] R.A.K. El Desouki, F.A. Habib, Preeclampsia at molecular level, Appl. Med. Res. 2 (2015) 65e72. [3] M. Emanuel, S. Butt, Frequency and factors leading to recurrent pre-eclampsia, J. Pak. Med. Assoc. 65 (2015) 1173e1177. [4] H.E. Yong, P.E. Melton, M.P. Johnson, K.A. Freed, B. Kalionis, P. Murthi, et al., Genome-wide transcriptome directed pathway analysis of maternal pre-
B.A.G. Marwa et al. / Placenta 43 (2016) 81e85 eclampsia susceptibility genes, PLoS One 10 (2015) 012e230. [5] E.J. Crosley, M.G. Elliot, J.K. Christians, B.J. Crespi, Placental invasion, preeclampsia risk and adaptive molecular evolution at the origin of the great apes: evidence from genome-wide analyses, Placenta 34 (2013) 127e132. [6] H. Salonen Ros, P. Lichtenstein, L. Lipworth, S. Cnattingius, Genetic effects on the liability of developing pre-eclampsia and gestational hypertension, Am. J. Med. Genet. 91 (2000) 256e260. [7] D.V. Antonio, B. Giovanni, G. Domenico, T. Giovanna, C. Francesco, D.A. Rosario, Endoglin, PlGFand sFlt-1as markers for predicting pre-eclampsia, Acta Obstetricia Gynecol. 87 (2008) 837e842. [8] R. Romero, J.K. Nien, J. Espinoza, D. Todem, W. Fu, H. Chung, J.P. Kusanovic, F. Gotsch, O. Erez, S. Mazaki- Tovi, et al., A longitudinal study of angiogenic (placental growth factor) and anti-angiogenic (soluble endoglin and soluble vascular endothelial growth factor receptor-1) factors in normal pregnancy and patients destined to develop preeclampsia and deliver a small for gestational age neonate, J. Matern. Fetal Neonatal Med. 10 (2008) 9e23. [9] H.P. Gerber, N. Ferrara, Angiogenesis and bone growth, Trends cardiovasc. Med. 10 (2000) 223e228. [10] G. Szebenyi, J.F. Fallon, Fibroblast growth factors as multifunctional signaling factors, Int. Rev. Cytol. 185 (1999) 45e106. [11] M. Relf, S. LeJeune, P.A. Scott, S. Fox, K. Smith, R. Leek, A. Moghaddam, R. Whitehouse, R. Bicknell, A.L. Harris, Expression of the angiogenic factors vascular endothelial cell growth factor, acidic and basic fibroblast growth factor, tumor growth factor beta-1, platelet-derived endothelial cell growth factor, placenta growth factor, and pleiotrophin in human primary breast cancer and its relation to angiogenesis, Cancer Res. 57 (1997) 963e969. [12] T. Maciej, J.C. Fadi, D.L. Mark, P. Sandosh, Y.S. William, H.M. William, G.M. Wladyslaw, M. Christine, Z.S. Ewa, F. Anna, M.D. Dominiczak, Fibroblast growth factor 1 gene and hypertension. From the quantitative trait locus to positional analysis, Circulation 116 (2007) 1915e1924. [13] P. Carmeliet, Fibroblast growth factor-1 stimulates branching and survival of myocardial arteries: a goal for therapeutic angiogenesis? Circ. Res. 87 (2000) 176e178. [14] S. Ortega, M. AIttmann, S.H. Tsang, M. Ehrlich, C. Basilico, Neuronal defects and delayed wound healing in mice lacking fibroblast growth factor 2, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 5672e5677. [15] P.K. Bos, G.J. van Osch, D.A. Frenz, J.A. Verhaar, H.L. Verwoerd-Verhoef, Growth factor expression in cartilage wound healing: temporal and spatial immunolocalization in a rabbit auricular cartilage wound model, Osteoarthr. Cartil. 9 (2001) 382e389. [16] H. Yamagata, Y. Chen, H. Akatsu, K. Kamino, J. Ito, S. Yokoyama, T. Yamamoto, K. Kosaka, T. Miki, I. Kondo, Promoter polymorphism in fibroblast growth factor 1 gene increases risk of definite Alzheimer’s disease, Biochem. Biophys. Res. Commun. 321 (2004) 320e323. [17] M. Beranek, P. Kolar, S. Tschoplova, K. Kankova, A. Vasku, Genetic variation and plasma level of the basic fibroblast growth factor in proliferative diabetic
85
retinopathy, Diabetes Res. Clin. Pract. 79 (2008) 362e367. [18] Kang S1, S.Z. Li, N. Wang, R.M. Zhou, T. Wang, D.J. Wang, X.F. Li, J. Bui, Y. Li, Association between genetic polymorphisms in fibroblast growth factor (FGF) 1 and FGF2 and risk of endometriosis and adenomyosis in Chinese women, Hum. Reprod. 25 (2010) 1806e1811. [19] H. Brian, M.D. Annex, Therapeutic angiogenesis: a treatment for the new millennium or passing fad? Cardiol. Rounds 6 (6) (2002) 1e3. [20] A. Bikfalvi, Significance of angiogenesis in tumour progression and metastasis, Eur. J. Cancer 31 (9) (1995) 1101e1104. [21] M. Jaye, J. Schlessinger, C.A. Dionne, Fibroblast growth factor receptor tyrosine kinases: molecular analysis and signal transduction, Biochim. Biophys. Acta 1135 (1992) 185e199. [22] K. Huebner, L. Nagarajan, E. Besa, E. Angert, B.J. Lange, L.A. Cannizzaro, H. van den Berghe, D. Santoli, J. Finan, C.M. Croce, et al., Order of genes on human chromosome 5q with respect to 5q interstitial deletions, Am. J. Hum. Genet. 46 (1990) 26e36. [23] R.A. Payson, M.A. Chotani, I.M. Chiu, Regulation of a promoter of the fibroblast growth factor 1 gene in prostate and breast cancer cells, J. Steroid Biochem. Mol. Biol. 66 (1998) 93e103. [24] R.L. Myers, R.A. Payson, M.A. Chotani, L.L. Deaven, I.M. Chiu, Gene structure and differential expression of acidic fibroblast growth factor mRNA: identification and distribution of four different transcripts, Oncogene 8.6 (1993) 341e349. [25] I.M. Chiu, K. Touhalisky, C. Baran, Multiple controlling mechanisms of FGF1 gene expression through multiple tissue-specific promoters, Prog. Nucleic Acid. Res. Mol. Biol. 70 (2000) 155e174. [26] R.A. Payson, M.A. Chotani, I.M. Chiu, Regulation of a promoter of the fibroblast growth factor 1 gene in prostate and breast cancer cells, J. Steroid Biochem. Mol. Biol. 66 (1998) 93e103. [27] T.B. Jian, L.Z. Hua, X.Z. Zhen, H.B. Xiu, W.Z. Jun, No association of the C > T polymorphism that is located 1385 upstream from initial code of fibroblast growth factor 1 gene with Alzheimer’s disease in Chinese, Brain Res. 1328 (2010) 113e117. [28] C. Butt, S. Lim, C. Greenwood, P. Rahman, VEGF, FGF1, FGF2 and EGF gene polymorphisms and psoriatic arthritis, BMC Musculoskelet. Disord. 4 (2007), 4e1. [29] D.M. Ornitz, N. Itoh, The fibroblast growth factor signaling pathway, Wiley Interdiscip. Rev. Dev. Biol. 4 (2015) 215e266. [30] J. Kreuger, M. Salmivirta, L. Sturiale, G. Gimenez-Gallego, U. Lindahl, Sequence analysis of heparan sulfate epitopes with graded affinities for fibroblast growth factors 1 and 2, J. Biol. Chem. 276 (2001) 30744e30752. [31] A.K. Powell, D.G. Fernig, J.E. Turnbull, Fibroblast growth factor receptors 1 and 2 interact differently with heparin/heparan sulfate. Implications for dynamic assembly of a ternary signaling complex, J. Biol. Chem. 277 (2002) 28554e28563.