With a reported prevalence of 1–20 % of all pregnancies, gestational diabetes mellitus (GDM) accounts for one of the most frequent gestational complications.1 From a pathophysiologic point of view, GDM seems to be related to type 2 diabetes, which most often occurs in elderly, obese women and is caused by pancreatic ß-cell malfunctioning and a peripheral insulin resistance.2
Plasma plasminogen activator inhibitor 1 (PAI-1) is a glycoprotein belonging to the serine protease inhibitors and is synthesized by different cells, such as endothelial cells, liver cells, adipose tissue, and fibroblasts.3–5 Its secretion is regulated by different factors, such as tumor necrosis factor α, insulin-like growth factor 1, and others.6,7 It inactivates tissue-plasminogen activator. The latter protein, together with fibrin, stimulates the transformation of inactive plasminogen to the biologically active plasmin.8
High plasma PAI-1 concentrations have been linked to the development of atherothrombosis, dyslipidemia, hypertension, obesity, insulin resistance, hyperinsulinemia, and diabetes mellitus type 2.9–11 Several polymorphisms of the PAI-1 gene have been described. Among these, the polymorphism at position −675 in the promoter region of the gene exerted the greatest impact on the plasma PAI-1 concentration.12 Plasminogen activator inhibitor 1 activity in plasma decreased according to the genotype (4G/4G, 4G/5G, and 5G/5G), with the lowest activity in carriers of the 5G/5G genotype.13 If PAI concentration is associated with glucose intolerance during pregnancy, then that polymorphism at position 675 which is associated with the lowest activity of the PAI-1 (5G/5G) should be associated with the lowest frequency of glucose intolerance in pregnancy. Therefore, one can hypothesize that the PAI-1 5G/5G polymorphism is associated with normal glucose tolerance in pregnant women.
The influence of PAI-1 on insulin resistance and its potential connection to diabetes mellitus type 2 prompted us to examine the functional −675 4G/5G polymorphism of the PAI-1 gene in pregnant women with and those without GDM.
MATERIALS AND METHODS
The study was conducted from June 2001 to December 2004 with the approval of our institutional review board, and informed consent was obtained from all participants according to Austrian gene technology law. An oral glucose tolerance test (OGTT) was performed between the 24th and 28th weeks of gestation in all women who intended to deliver at the Vienna University Hospital, Department of Obstetrics and Gynecology. Preceded by an 8-hour period of fasting, a standardized 75-g glucose solution (Glucodrink; Unipack, Wr. Neustadt, Austria) was ingested orally. Venous blood samples were drawn before glucose ingestion, then twice again after glucose ingestion: after 1 hour and after 2 hours. On the same day, a measurement for glycosylated hemoglobin (Hb A1C) was performed after venous blood was drawn from the mothers. Before the test a precise medical and obstetric history was taken from the women and entered into a computerized database. We employed the guidelines of the German Society for Diabetes to evaluate the results of the OGTT.14 The upper normal limit for the fasting serum glucose was set at 90 mg/dL and at 180 mg/dL 1 hour and 155 mg/dL 2 hours after glucose ingestion. If at least one value was exceeded, we classified the woman as a gestational diabetic and admitted her to the gestational diabetes program. This program starts with dietary instruction lessons. After the instructions we ask the women to measure their capillary blood glucose concentrations at home daily for 1 week, 1 hour after meals, to find out if insulin therapy is necessary. Upper limits of 90 mg/dL fasting glucose concentration and 130 mg/dL 1 hour postprandial glucose concentration are considered acceptable. If a woman exceeded these limits 5 or more times a week, insulin therapy was initiated. Although the 75-g oral glucose load after an 8-hour fast is different from the test most commonly used in the United States, the physiological principles demonstrated in our study also apply to different methods of glucose tolerance testing.15
Forty white women with normal glucose tolerance and 40 white women with GDM were randomly selected. Randomization was done with a computer-generated table of random numbers for each of the 2 groups separately. On the same day as the OGTT, venous blood samples for DNA analysis were taken from the mothers and stored at −20°C until extraction was performed. Genomic DNA was isolated from anticoagulated blood with the QiAmp Blood Midi Kit (Quiagen, Hilden, Germany) as described by the manufacturer and stored at −20°C. The polymerase chain reaction (PCR) amplification of the PAI-1 promoter region was carried out in a final volume of 50 μL, consisting of 1× PCR buffer, 200 mmol of deoxyribonucleotide triphosphate (dNTP)/L, 1.5 mmol of MgCl2/L, 0.5 U Taq DNA polymerase (Invitrogen, Lofer, Austria), and 500 ng of genomic DNA. The forward and reverse oligonucleotide primers were 5′-CACAGAGAGAGTCTGGCCACGT-3′ and 5′-CCAACAGAGGACTCTTGGTCT-3′. The concentration of each primer was 20 pmol/μL. Polymerase chain reaction conditions were 96°C for 1 minute, 35 cycles (96°C for 1 minute, 53°C for 1 minute, 72°C 1 minute), and 72°C for 10 minutes. The PCR products were purified with the QIAquick PCR purification kit as described by the supplier (Qiagen, Hilden, Germany). Polymerase chain reaction fragments were digested overnight with 25 U Bsl 1 (New England Biolabs, Ipswich, MA) in 1× NEB3 buffer at 55°C. DNA fragments were separated on a 3% agarose gel (50% NuSive-Agarose; FMC BioProducts, Rockland, ME; and 50% LE-agarose (Invitrogen). After digestion the PAI-1 products measured 100 bp (4G/4G), 100bp/78bp/22bp (4G/5G), and 78bp/22bp (5G/5G).
Glucose was determined using an enzymatic in vitro test (Gluco-quant Glucose/HK, Roche Diagnostics GmbH, Mannheim, Germany) on a Roche/Hitachi analyzer (Roche Diagnostics). The assay for glucose did not show any significant cross-reaction with other substances and had intra-assay difference of less than 1.1%. The interassay precision was less than 1.9%. Body mass index (BMI) was calculated as reported prepregnancy weight in kilograms divided by measured height in meters squared.
Data were evaluated with the SPSS 12.0 statistical program (SPSS Inc, Chicago, IL). We performed a Kolmogorov-Smirnov test to verify the use of tests for normally distributed variables. Values are given as mean ± standard deviation (SD). Chi-square test, Fisher exact test, and t test were used accordingly. The one-way analysis of variance on ranks applied for comparisons between the independent groups and corrected by Bonferroni-Holm adjustment.16 We also performed a multiple logistic regression analysis to examine whether women’s age, BMI, and PAI-1 genotype could be used as prognostic factors for GDM. We applied the inclusion method, and odds ratios and 95% confidence intervals were computed as proportional hazards using multiple logistic regression analysis. Our calculation using Fisher exact test with a 5% 2-sided significance level has an 80% power to detect the difference between a control group (women with physiologic OGTT) proportion of 5% and a study group (women with GDM) proportion of 30%. All tests were 2-tailed, and we considered P ≤ .05 to be statistically significant.
We assessed a total of 887 women by oral glucose tolerance testing. Among them, 724 women (81.6%) displayed a physiologic OGTT. We classified 163 women (18.4%) as gestational diabetics. The high rate of women with GDM may be caused by the fact that our department is a tertiary care center. Patient characteristics comparing women in the randomized and nonrandomized groups are shown in Table 1.
When the groups were compared after randomization, women with GDM were significantly older (33.2 ± 4.9 years versus 30.8 ± 5.8 years; P = .03 by t test), had a higher body mass index (26.9 ± 5.8 kg/m2 versus 24.1 ± 5.5 kg/m2; P = .02 by t test), and displayed higher Hb A1C values at the time of the OGTT (5.7 ± 1.3 versus 4.9 ± 0.6; P < .001 by t test). The newborn infants of mothers with GDM had higher cord blood insulin concentrations (14.5 ± 11.3 μU/mL versus 7.4 ± 6.6 μU/mL; P = .01 by t test) than newborns of women with physiologic OGTT results (Table 2).
From the 80 women analyzed for the PAI genotype, the allele distribution was as follows: The frequency of the 4G allele was 89 (55.6%) and that of the 5G allele 71 (44.4%). Forty-seven women (58.8%) were heterozygous (4G/5G), 21 women (26.2%) were homozygous for 4G (4G/4G), and 12 women (15.0%) were homozygous for 5G (5G/5G).
Women with normal glucose tolerance were significantly more often homozygous for 5G allele than women with GDM: 10 (25%) versus 2 (5%); P = .01 by χ2 test (Table 3). Furthermore, low fasting glucose levels in the OGTT were significantly related to homozygosity for 5G. No significant differences were seen with respect to BMI, Hb A1C, or 1- and 2-hour OGTT values in women with this haplotype compared with others (Table 4).
In addition, we performed a multiple logistic regression analysis to estimate whether maternal age, maternal BMI, and PAI-1 genotype were suitable as prognostic factors for GDM. In the logistic regression analysis, the association between GDM and maternal age as well as PAI-1 genotype were significant (odds ratio [OR] 1.11, 95% confidence interval [CI] 1.01–1.22, P = .03; OR 0.16, 95% CI 0.03–0.87, P = .03, respectively) (Table 5).
The present study yielded several findings in pregnant women with GDM. Firstly, homozygosity for the 5G allele of the PAI gene was associated with normal glucose tolerance. Secondly, women homozygous for the 5G allele had the lowest fasting values in the OGTT. Thirdly, BMI was not associated with this PAI-1 genotype in women with GDM in our setting.
Gestational diabetes mellitus and type 2 diabetes are polygenic, multifactorial diseases.17 A pathophysiologic link between GDM and type 2 diabetes seems likely and is substantiated by the fact that women with GDM have up to a 50% chance of developing type 2 diabetes over the next 10 years.18 Physiologically in pregnancy, free fatty acids mediate insulin resistance. In women with GDM, however, this metabolic alteration cannot be compensated because of pancreatic ß-cell malfunctioning.17,19 From 6 controlled follow-up studies, the overall relative risk for developing diabetes after GDM was calculated to be 6.0 (95% CI 4.1–8.8).20
Gene alterations, such as single nucleotide changes, which are associated with type 2 diabetes are likely to be present in women with GDM. Horikawa et al21 and Cassell et al22 have shown that single nucleotide polymorphisms in the gene encoding calpain-10 are associated with an increased risk for type 2 diabetes. In a previous study, our group showed that the Caplain haplotype combination 121/221 is significantly associated with GDM.23 Moreover, it is known that plasma PAI-1 concentrations are positively related to insulin resistance, diabetes mellitus type 2, obesity, and a history of GDM.24–26 The PAI-1 polymorphism at position −675 in the promoter region is correlated with the PAI-1 plasma concentration.12 The present study examined the role of mutations of the PAI-1 −675 polymorphism in the development of GDM. The findings hereby obtained provide further evidence for the hypothesis of a common pathophysiology between diabetes type 2 and GDM.
The linkage of the PAI −675 polymorphism to obesity has been found in several studies.27,28 In a study of Hofstedt et al,27 the risk of being obese is 2-fold in carriers of the 4G allele although the mechanism of how obesity is influenced by PAI is not fully understood. It is known that adipose tissue produces large amounts of PAI-1 and that therefore plasma PAI-1 concentrations are higher in obese patients.9,29 In a knockout mice model (ob/ob mice), disruption of the PAI-1 gene induces weight loss.30 This leads to the assumption that the PAI-1 gene might be involved in the control of the body mass fat. In our study, women who were homozygous for the 5G allele had the lowest BMI compared with others, without reaching statistical significance. Accordingly, in the study of Hoffstedt et al,27 no significant difference between the PAI-1 genotypes and the BMI could be found. Only a separate analysis of lean (< 25 kg/m2) compared with obese (BMI > 30 kg/m2) subjects showed a significant difference between carriers or noncarriers of the 4G allele. Our cohort of obese women (BMI > 30 kg/m2) is too small for an exact analysis comparing the different allele distributions in lean versus obese women.
In conclusion, this study demonstrates a connection between GDM and the functional −675 PAI-1 genotype. Homozygosity for the 5G allele constituted a protecting factor for the development of GDM. Furthermore, fasting values of the OGTT are also linked to the PAI genotype. These findings further support a possible role of PAI in the development of GDM.
1. Sepe SJ, Connell FA, Geiss LS, Teutsch SM. Gestational diabetes: incidence, maternal characteristics, and perinatal outcome. Diabetes 1985;34 suppl 2:13–6.
2. Ferrannini E. Insulin resistance versus insulin deficiency in non-insulin-dependent diabetes mellitus: problems and prospects. Endocr Rev 1998;19:477–90.
3. Noguchi T, Matsuyama S, Akao M, Hagiwara H, Uno S, Seki T, et al. Induction of hepatic tissue-type plasminogen activator and type 1 plasminogen activator-inhibitor gene expressions and appearance of their translation products in the bile following acute liver injury in rats. Thromb Res 2001;104:283–91.
4. Skurk T, Hauner H. Obesity and impaired fibrinolysis: role of adipose production of plasminogen activator inhibitor-1. Int J Obes Relat Metab Disord 2004;28:1357–64.
5. Radha KS, Sugiki M, Harish Kumar, M, Omura S, Maruyama M. Post-transcriptional regulation of plasminogen activator inhibitor-1 by intracellular iron in cultured human lung fibroblasts: interaction of an 81-kDa nuclear protein with the 3′-UTR. J Thromb Haemost 2005;3:1001–8.
6. Hamaguchi E, Takamura T, Shimizu A, Nagai Y. Tumor necrosis factor-alpha and troglitazone regulate plasminogen activator inhibitor type 1 production through extracellular signal-regulated kinase- and nuclear factor-kappaB-dependent pathways in cultured human umbilical vein endothelial cells. J Pharmacol Exp Ther 2003;307:987–94.
7. Anfosso F, Alessi MC, Nalbone G, Chomiki N, Henry M, Juhan-Vague I. Up-regulated expression of plasminogen activator inhibitor-1 in Hep G2 cells: interrelationship between insulin and insulin-like growth factor 1. Thromb Haemost 1995;73:268–74.
8. Collinge JE, Simirskii VN, Duncan MK. Expression of tissue plasminogen activator during eye development. Exp Eye Res 2005;8:90–6.
9. Juhan-Vague I, Alessi MC, Morange PE. Hypofibrinolysis and increased PAI-1 are linked to atherothrombosis via insulin resistance and obesity. Ann Med 2000;32 suppl 1:78–84.
10. Bruckert E, Ankri A, Giral P, Turpin G. Relation between plasminogen activator inhibitor-1 and hepatic enzyme concentrations in hyperlipidemic patients. Thromb Haemost 1994;72:434–7.
11. Herlihy OM, Barrow BA, Grant PJ, Levy JC. Hyperglycaemic siblings of Type II (non-insulin-dependent) diabetic patients have increased PAI-1, central obesity and insulin resistance compared with their paired normoglycaemic sibling. Diabetologia 2002;45:635–41.
12. Nordt TK, Lohrmann J, Bode C. Regulation of PAI-1 expression by genetic polymorphisms. Impact on atherogenesis. Thromb Res 2001;103 suppl 1:S1–5.
13. Eriksson P, Kallin B, van’t Hooft, FM, Båvenholm P, Hamsten A. Allele-specific increase in basal transcription of the plasminogen-activator inhibitor 1 gene is associated with myocardial infarction. Proc Natl Acad Sci U S A 1995;92:1851–5.
14. Deutsche Diabetes-Gesellschaft. Diagnostik und Therapie des Gestationsdiabetes. Richtlinien der Deutschen Diabetes-Gesellschaft [Diagnosis of and therapy for GDM. Guidelines from the German Diabetes Association]. Der Frauenarzt 1993;13–4.
15. Weiss PA, Haeusler M, Kainer F, Purstner P, Haas J. Toward universal criteria for gestational diabetes: relationships between seventy-five and one hundred gram glucose loads and between capillary and venous glucose concentrations. Am J Obstet Gynecol 1998;178:830–5.
16. Holm S. A simple sequentially rejective multiple test procedure. Scand J Statist 1979;6:65–70.
17. Catalano PM, Drago NM, Amini SB. Longitudinal changes in pancreatic beta-cell function and metabolic clearance rate of insulin in pregnant women with normal and abnormal glucose tolerance. Diabetes Care 1998;21:403–8.
18. Henry OA, Beischer NA. Long-term implications of gestational diabetes for the mother. Baillieres Clin Obstet Gynaecol 1991;5:461–83.
19. Homko C, Sivan E, Chen X, Reece EA, Boden G. Insulin secretion during and after pregnancy in patients with gestational diabetes mellitus. J Clin Endocrinol Metab 2001;86:568–73.
20. Cheung NW, Byth K. Population health significance of gestational diabetes. Diabetes Care 2003;26:2005–9.
21. Horikawa Y, Oda N, Cox NJ, Li X, Orho-Melander M, Hara M, et al. Genetic variation in the gene encoding calpain-10 is associated with type 2 diabetes mellitus [published erratum appears in Nat Genet 2000;26:502]. Nat Genet 2000;26:163–75.
22. Cassell PG, Jackson AE, North BV, Evans JC, Syndercombe-Court D, Phillips C, et al. Haplotype combinations of calpain 10 gene polymorphisms associate with increased risk of impaired glucose tolerance and type 2 diabetes in South Indians. Diabetes 2002;51:1622–8.
23. Leipold H, Knofler M, Gruber C, Haslinger P, Bancher-Todesca D, Worda C. Calpain-10 haplotype combination and association with gestational diabetes mellitus. Obstet Gynecol 2004;103:1235–40.
24. Panahloo A, Mohamed-Ali V, Lane A, Green F, Humphries SE, Yudkin JS. Determinants of plasminogen activator inhibitor 1 activity in treated NIDDM and its relation to a polymorphism in the plasminogen activator inhibitor 1 gene. Diabetes 1995;44:37–42.
25. Eriksson P, Van Harmelen, V, Hoffstedt J, Lundquist P, Vidal H, Stemme V, et al. Regional variation in plasminogen activator inhibitor-1 expression in adipose tissue from obese individuals. Thromb Haemost 2000;83:545–8.
26. Winzer C, Wagner O, Festa A, Schneider B, Roden M, Bancher-Todesca D, et al. Plasma adiponectin, insulin sensitivity, and subclinical inflammation in women with prior gestational diabetes mellitus. Diabetes Care 2004;27:1721–7.
27. Hoffstedt J, Andersson IL, Persson L, Isaksson B, Arner P. The common −675 4G/5G polymorphism in the plasminogen activator inhibitor −1 gene is strongly associated with obesity. Diabetologia 2002;45:584–7.
28. Lopes C, Dina C, Durand E, Froguel P. PAI-1 polymorphisms modulate phenotypes associated with the metabolic syndrome in obese and diabetic Caucasian population. Diabetologia 2003;46:1284–90.
29. Yamamoto K, Saito H. A pathological role of increased expression of plasminogen activator inhibitor-1 in human or animal disorders. Int J Hematol 1998;68:371–85.
30. Schafer K, Fujisawa K, Konstantinides S, Loskutoff DJ. Disruption of the plasminogen activator inhibitor 1 gene reduces the adiposity and improves the metabolic profile of genetically obese and diabetic ob/ob mice. FASEB J 2001;15:1840–2.
Figure. No caption available.
© 2006 The American College of Obstetricians and Gynecologists
This article has been cited