- Email: [email protected]
Elevated plasma activator inhibitor 1 is not related to insulin resistance and to gene polymorphism in healthy centenarians
Atherosclerosis 160 (2002) 385– 390 www.elsevier.com/locate/atherosclerosis
Elevated plasma activator inhibitor 1 is not related to insulin resistance and to gene polymorphism in healthy centenarians Maria Rosaria Rizzo a, Emilia Ragno a, Michelangela Barbieri a, Domenico De Lucia b, Daniela Manzella a, Maria Rosaria Tagliamonte a, Donatella Colaizzo c, Maurizio Margaglione c, Giuseppe Paolisso a,* a
Department of Geriatric Medicine and Metabolic Diseases, IV Di6isione di Medicina Interna, Second Uni6ersity of Naples, Piazza Miraglia, 2, 80138 Naples, Italy b Institute of General Pathology and Oncology, Second Uni6ersity of Naples, Naples, Italy c Atherosclerosis and Thrombosis Unit, Casa Sollie6o della Sofferenza, San Gio6anni Rotondo, Italy Received 21 February 2001; received in revised form 11 June 2001; accepted 15 June 2001
Abstract Previous studies demonstrated a relationship between the degree of insulin resistance and plasma plasminogen activator inhibitor type-1 (PAI-1) levels. We aim at investigating the relationship between the degree of insulin resistance and plasma PAI-1 levels in aged subjects (n=83) and in healthy centenarians (n= 42). In all subjects the degree of insulin resistance was assessed by HOMA method. Our data demonstrated that healthy centenarians have higher plasma PAI-1 levels (73.1 9 13.9 vs 23.7 914.7 ng/ml, PB 0.001) and lower degree of insulin resistance (1.4 9 0.5 vs 3.3 9 1.3, P B0.001) than aged subjects. In aged subjects plasma PAI-1 levels correlated with the degree of insulin resistance (r= 0.61, P B 0.001), fasting plasma triglycerides (r=0.74, PB 0.001) and age (r=0.33, PB0.001). All such associations were lost in centenarians. Plasma PAI-1 Ag levels were also similar in aged subjects and centenarians even after categorization for PAI gene polymorphism. In multivariate analysis, a model made by age, sex, body mass index, fasting plasma triglycerides, HOMA and PAI-1 gene explained 65 and 50% of plasma PAI-1 level variations in aged subjects and centenarians, respectively. Nevertheless, HOMA (P B 0.001) was significantly and independently associated with plasma PAI-1 levels only in aged subjects. In conclusion, our data demonstrates that in healthy centenarians, plasma PAI-1 were not associated with the degree of insulin resistance as in aged subjects. Frequency of PAI-1 genotype does not provide an explanation for such differences between aged subjects and centenarians. © 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Plasma plasminogen activator inhibitor type-1 levels; Insulin resistance; Centenarians; Aged subjects; PAI-1 gene polymorphism
1. Introduction An unbalanced ratio between plasma levels of coagulation and fibrinolytic factors are associated with an increased risk of arterial thrombosis in cardiovascular [1 – 5] and cerebrovascular [6] diseases. Several factors have been shown to affect plasminogen activator inhibitor type-1 (PAI-1) synthesis and secretion. Genotype might have a role since subjects carrying the
* Corresponding author. Tel.: + 39-081-566-5016; fax: + 39-081566-5051. E-mail address: [email protected] (G. Paolisso).
4G/4G polymorphism (single-base-pair guanine insertion/deletion polymorphism (4G/5G) within the promoter region of the PAI-1 gene) showed the highest plasma PAI-1 levels [7]. In addition, metabolic determinants such as insulin resistance [8,9], body mass index (BMI) [10], plasma triglycerides [11] and small dense LDL [12] should be also taken into account. The relationship between plasma PAI-1 antigen levels and insulin resistance is of particular interest in human aging. In fact, advancing age has been associated with an increased insulin resistance [13] and elevated risk of atherothrombotic events [14,15]. Thus, aged subjects may be more prone to atherothrombotic events due to the impact of insulin resistance on PAI-1 levels.
0021-9150/02/$ - see front matter © 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 2 1 - 9 1 5 0 ( 0 1 ) 0 0 5 7 5 - 5
386
M.R. Rizzo et al. / Atherosclerosis 160 (2002) 385–390
A special group of aged subjects are the centenarians who represent the best model of successful aging; in fact, these subjects have escaped age-related atherothrombotic and metabolic diseases and reached the extreme limits of lifespan. The possibility that particular environmental, genetic, and/or haemostatic profiles prevent major diseases in these individuals cannot be ruled out and it needs to be clarified. Thus, our study aims at evaluating the possible relationship among plasma PAI-1 levels, insulin resistance and PAI-1 gene polymorphism in healthy aged subjects and centenarians.
2. Materials and methods
the Ethical Committee of our Institution and each subject, as well as their relatives for centenarians, provided a written informed consent before being enrolled into the study.
2.2. Metabolic determination Anthropometric determinations were made as previously reported [16]. At enrolment all subjects underwent an oral glucose tolerance test for assessing the degree of glucose intolerance [17]. Degree of insulin resistance was estimated by homeostatic (HOMA) method [18]. This method correlates well with insulin sensitivity measured by euglycemic clamp (r=0.88, PB 0.0001) [18,19].
2.1. Subjects Eighty-three aged subjects (mean age: 74.59 7.0 years) and 42 centenarians (age\100 years) living in the five districts of the Campania region, a geographically and administratively well-defined Mediterranean area located on the West Coast of Southern Italy, volunteered for the study. The electoral roll for a referendum held on 18 April 1999, was used as a population source and the subjects were examined between July and December 1999. The subjects were contacted at home or in their institution and examined by physicians previously trained to administer a questionnaire that included cognitive and depression tests. All subjects were in good health, as determined by a medical history, physical examination, and routine screening laboratory analyses, and were free from any drug known to interfere with glucose metabolism. More detailed information concerning the two study populations is given in Table 1. The study was approved by
2.3. Determination of PAI-1 genotype by polymerase chain reaction The PAI-1 4G/5G polymorphism was evaluated as previously reported [20]. Briefly, a mutated oligonucleotide was synthesized that inserts a site for the Bsl I enzyme within the amplification product. Polymerase chain reaction was carried out in 50-ml samples in a Perkin–Elmer – Cetus thermal cycler. Each sample contained 0.5 mg of genomic DNA, 15 pmol of each primer, 100 mmol/l of NTP, 10 mmol/l Tris –HCl, pH 8.3, 50 mmol/l KCl, 1.5 mol/l MgCl2, and 1 U thermostable Taq polymerase. The 30 cycles consisted of steps at 95 °C for 1 min, 60 °C for 1 min, and 72 °C for 2 min. Then, 20-ml volumes of the amplification products were digested for 2.5 h at 55 °C with 5 U of the Bsl I restriction enzyme. The fragments were fractionated by 4% agarose gel electrophoresis and visualized under UV light.
Table 1 Clinical characteristics of the two study groups
Age range (years) Gender (M/F) BMI (kg/m2) WHR FP glucose (mmol/l) 2 h plasma glucose (mmol/l) FP insulin (pmol/l) FP cholesterol (mmol/l) FP triglycerides (mmol/l) HOMA index PAI-1 (ng/ml) PAP (ng/ml) TAT (ng/ml)
Aged subjects (n = 83)
Centenarians (n = 42)
P
62–90 34/49 26.5 91.4 0.84 90.04 6.29 1.1 7.29 1.3 11.79 3.4 5.49 0.8 1.7 90.5 3.39 1.3 23.7914 320 994 1.5 90.8
100–105 15/27 21.6 91.2 0.83 9 0.03 4.2 9 0.9 5.3 9 1.1 7.4 9 1.6 4.6 9 0.9 1.1 9 0.4 1.4 9 0.5 73.1 913 575 9 87 4.1 9 1.0
B0.001 0.02* B0.001 0.13 B0.001 B0.001 B0.001 B0.001 B0.001 B0.001 B0.001 B0.001 B0.001
All results are expressed as mean 9S.D. For testing differences between the two groups unpaired Student’s t-test was used. For investigating gender differences 2 test was used; d.f. = 1. BMI, Body mass index; WHR, waist/hip ratio; FP, fasting plasma; PAI-1, plasminigen activator inhibitor type-1; PAP, plasmin–antiplasmin complex; TAT, thrombin–antithrombin complex.
M.R. Rizzo et al. / Atherosclerosis 160 (2002) 385–390
387
2.4. Analytical methods After an overnight fast, blood samples were drawn for measurements of plasma metabolite concentrations. Plasma glucose level was immediately determined by enzymatic method (Autoanalyzer, Beckman). Blood samples for insulin measurements were collected in heparinized tubes. After centrifugation, plasma insulin concentration was determined by a commercially available radioimmunoassay kit (coefficient of variation (c.v.) 3.29 0.3%; cross-reactivity with proinsulin 0.3%; Sorin, Biomedical). Commercial enzymatic methods were used for plasma total cholesterol (Monotest, Boehringer Mannheim, Milan, Italy; c.v.= 3.8 9 0.9%) and triglyceride (Peridecrome, Boehringer Mannheim; c.v.=4.19 0.6%) levels determination. PAI-1 Ag levels were measured by enzyme-linked immunoabsorbent assay (ELISA) technique (Byk Gulden, Milan, Italy; c.v.=14 9 3.5%). Thrombin–antithrombin complex (T– AT), a measure of thrombin generation and neutralisation by antithrombin, was assayed by ELISA (Behring Werke, Marburg, Germany). In a subgroup of 63 aged subjects and 32 centenarians plasma plasmin–antiplasmin complexes (PAP) were determined by ELISA (Behring Werke).
2.5. Calculation and statistical analyses To approximate normal distribution, plasma insulin, triglycerides and PAI-1 Ag levels were log transformed for use in all calculations, then back-transformed for presentation of results. All results are expressed as mean 9standard deviation (S.D.). For testing differences between the two study groups, unpaired Student’s t test was used for continuous variables. The significance of the difference in gene frequency between groups was analysed by a 2 test. ANOVA with the Scheffe` test was used for comparing the PAI-1 Ag levels among the genotype groups. Simple correlation by the Pearson method allowed us to assess univariate relations. Linear multiple regression analysis was used for investigating the different contribution of each covariate to the variability of PAI-1 Ag levels. In this latter analysis, gender difference was calculated by considering female=0 and male = 1. In a same way, PAI-1 genotypes were categorised as follows: 4G/4G =0, 4G/ 5G = 1, and 5G/5G =2. Data were analysed using the Statistical Package for Social Science (SPSS).
3. Experimental results Clinical characteristics of the two study groups are reported in Table 1. Gender ratio demonstrated a prevalence of female in centenarians. This latter group
Fig. 1. Simple correlations between HOMA index and plasma PAI-1 levels in aged subjects (top) and centenarians (bottom).
of subjects had also a significantly lower BMI, fasting plasma glucose, insulin, cholesterol, and triglycerides, and HOMA index than aged subjects. In contrast, plasma PAI-1 Ag, T–AT, and PAP levels were significantly higher in centenarians than in aged subjects. In particular, plasma PAI-1 Ag levels were three times higher in centenarians than in the aged subjects. Plasma PAI-1 Ag levels correlated with HOMA in aged subjects but not in centenarians (Fig. 1). In aged subjects, plasma PAI-1 Ag levels correlated also with age, fasting plasma glucose, insulin, triglycerides, T– AT and PAP levels. In contrast, centenarians had plasma PAI-1 Ag levels correlated only with age, fasting plasma triglycerides and PAP concentrations (Table 2). In aged subjects, plasma T–AT levels correlated with age, BMI, fasting plasma glucose, insulin, triglycerides levels and HOMA index, while plasma PAP correlates with fasting plasma glucose, insulin, triglycerides levels and HOMA index. In contrast, all these correlations were not significant in centenarians (Table 2). As far as PAI-1 gene polymorphism is concerned, the frequency of the 4G allele was 0.532 and the frequency
M.R. Rizzo et al. / Atherosclerosis 160 (2002) 385–390
388
Table 2 Simple correlations among PAI-1 Ag, TAT, PAP levels and anthropometric and metabolic parameters Aged subjects (n= 83)
Age BMI WHR Plasma glucose Insulin Total cholesterol Triglycerides HOMAa TAT PAP
Centenarians (n =42)
PAI-1 Aga
P
T–AT
P
PAP
P
PAI-1 Aga
P
T–AT
P
PAP
P
0.33 0.55 0.04 0.41 0.56 0.21 0.74 0.61 0.56 0.83
0.002 0.625 0.71 B0.001 B0.001 B0.05 B0.001 B0.001 B0.001 B0.001
0.46 0.22 0.11 0.34 0.26 0.20 0.34 0.35 – 0.64
B0.001 0.05 0.30 B0.001 0.01 0.06 B0.001 B0.001 – B0.001
0.22 0.09 0.004 0.30 0.49 0.25 0.60 0.52 0.64 –
0.09 0.94 0.97 B0.01 B0.001 B0.05 B0.001 B0.01 B0.001 –
0.47 −0.12 0.05 0.11 0.03 0.25 0.69 0.12 0.07 0.47
B0.001 0.40 0.71 0.46 0.80 0.09 B0.001 0.42 0.64 B0.005
0.22 −0.08 0.17 0.06 0.01 −0.01 −0.16 −0.15 – 0.31
0.15 0.60 0.28 0.90 0.92 0.94 0.28 0.30 – 0.07
0.36 −0.03 0.07 −0.13 0.15 −0.05 0.21 −0.01 0.32 –
0.05 0.86 0.68 0.47 0.42 0.77 0.21 0.94 0.09 –
a
Back log transformed; BMI, body mass index; WHR, waist/hip ratio; PAI-1, plasminigen activator inhibitor type-1; TAT, thrombin–antithrombin complex; PAP, plasmin–antiplasmin complex.
of the 5G allele was 0.468; the frequency of 4G/4G, 4G/5G and 5G/5G genotypes was 0.325, 0.410 and 0.265, respectively. Genotype and allele distributions were in Hardy–Weinberg equilibrium. Furthermore, the allele frequency and the genotype distribution did not differ between aged subjects and centenarians (Table 3). No differences in plasma PAI-1 Ag levels between aged subjects and centenarians were found even after categorization for PAI gene polymorphism (Table 4). Due to the occurrence of several determinants of plasma PAI-1 Ag levels variability, multiple stepwise linear regression analysis investigates the independent contribution of main covariates in the two study groups. A model made by age, sex, BMI, fasting plasma triglycerides, HOMA index, and PAI-1 gene polymorphism explained 65% of the variability of plasma PAI-1 Ag levels in aged subjects with age (PB 0.001), sex (P B0.04), BMI (PB 0.03), fasting plasma triglycerides (P B 0.001), and HOMA index (PB 0.001) significantly and independently associated with plasma PAI-1 Ag levels. The same model in centenarians explained 50% of the variability of the plasma PAI-1 Ag levels with only age (PB 0.04) and fasting plasma triglycerides (P B0.001) significantly and independently associated with the dependent variable.
PAI-1 is a potent inhibitor of fibrinolysis by binding and rapidly inactivating both tissue- and urokinase-type plasminogen activator. Increased levels of PAI-1 and decreased plasma fibrinolytic activity have been demonstrated in survivor of acute myocardial infarction and in patients with coronary artery stenosis [1–4]. Furthermore, PAI-1 has prognostic value in predicting recurrence of myocardial infarction [4]. A positive association between fasting plasma insulin levels and PAI-1 has been demonstrated by several investigations [21]. In addition, PAI-1 has been linked to separate feature of syndrome X including abdominal obesity, hypertension, hypertriglyceridemia, glucose intolerance and type 2 diabetes mellitus [21]. The association of PAI-1 with features of syndrome X supports the link between plasma PAI-1 levels and insulin resistance. In our study, we confirmed that in aged subjects there is a strong relationship between plasma PAI-1 levels and degree of insulin resistance (measured by HOMA) and/ or plasma triglycerides. In contrast, an unexpected finding was the lack of a relationship between plasma PAI-1 levels and degree of insulin resistance in healthy centenarians. In fact, several previous studies have shown the absence in healthy centenarians of cardiovascular risk factors such as obesity [16], hypertension [22], Table 3 PAI-1 gene polymorphism distribution in the two populations
4. Discussion Our study confirms that healthy centenarians have higher levels of PAI-1 Ag levels, T– AT, and PAP than aged subjects; furthermore, we also demonstrate that (a) plasma PAI-1 Ag levels were strongly correlated with HOMA index in aged subjects but not in centenarians, (b) analysis of PAI-1 genotype did not explain plasma PAI-1 levels either in aged subjects or in centenarians.
Aged (n =83)
Centenarians (n = 42)
4G/4G 4G/5G 5G/5G
27 (32.5%) 34 (41%) 22 (26.5%)
15 (35.7%) 15 (35.7%) 12 (28.6%)
Allele frequency 4G 5G
88 (53%) 78 (47%)
45 (53.6%) 39 (46.4%)
For investigating difference in gene frequency x2 test was used; d.f. = 2; P= NS for genotype distribution.
M.R. Rizzo et al. / Atherosclerosis 160 (2002) 385–390 Table 4 PAI-1Ag levels in different PAI-1 gene polymorphisms
Aged (n = 83) Centenarians (n =42)
4G/4G
4G/5G
5G/5G
21.2912.3 72.2913.7
25.4 916.2 69.3 9 12.1
24.6 916.3a 78.3 917.3b
ANOVA with the Scheffe` test was used for comparing the PAI-1 Ag levels among the genotype groups d.f. = 2. a P = 0.43. b P = 0.27.
dyslipidemia [23], and insulin resistance or impaired glucose tolerance [24]. Thus, it was hypothesized that age-remodelling occurs in healthy centenarians and that such a process is also responsible for the achievement of extreme life span [25]. Indeed, the evidence that healthy centenarians have higher plasma PAI-1, PAP and T– AT levels than aged subjects seems in contrast either with the age-remodelling hypothesis [25] or with the well-known role of PAI-1 and insulin resistance in the pathogenesis of atherosclerosis [1– 9]. Why plasma PAI1 levels are not related to the degree of insulin resistance in healthy centenarians is still unknown. Nevertheless, a very recent study [26] has shown also in nonobese normal glucose tolerant offsprings of type 2 diabetic patients the lack of association between plasma insulin levels — a proxy of insulin resistance—and plasma PAI1 concentrations. Indeed, controversy exists with respect to the relationship between plasma insulin and PAI-1 activity. In vitro, data have shown that both insulin [27] and its precursor [28] stimulate PAI-1 synthesis from hepatocytes. Accordingly, a relationship between insulin and/or its precursors and PAI-1 activity has been shown in a large, healthy normal glucose tolerant population for northern Sweden [29]. In that study, however, neither proinsulin nor insulin concentrations predicted the serum PAI-1 activity in a multivariate linear regression model. In the Insulin Resistance Atherosclerosis Study [30], a strong and independent relationship between PAI-1 antigen and insulin has been found consistently across varying degrees of glucose tolerance. In vivo, insulin infusion has been shown to decrease [31] or to have no affect on [32] PAI-1 activity. Furthermore, it has been suggested that in vitro, the stimulatory effect of insulin on PAI-1 synthesis in hepatocytes may be counteracted by an inhibitory effect of insulin on PAI-1 synthesis and release in extrahepatic tissues, or by stimulation of PAI-1 degradation by insulin [33]. These complicated interactions between insulin and PAI-1 may explain why we found an association between insulin and PAI-1 activity in aged subjects but not in the special group of centenarian subjects, and thus, we hypothesize that in healthy centenarians elevated plasma PAI-1 levels might be determined by or related to other factors than plasma insulin levels/degree of insulin resistance.
389
Additional findings of our study were (a) healthy centenarians have plasma PAI-1, PAP and T– AT levels more elevated than in aged subjects; (b) lack of difference in genotype distribution between centenarians and aged subjects despite the evidence that centenarians have higher plasma PAI-1 Ag levels than aged subjects. As far as plasma PAI-1, PAP and T–AT levels are concerned, our data are in agreement with a previous report showing hypercoagulabity in centenarians [34]. With regard to the lack of difference in genotype distribution between aged subjects and centenarians, due to the limited number of centenarians studied, caution should be used before generalizing such results. Nevertheless, it should be pointed out that our data are in agreement with Bladbjerg et al. [35] showing comparable frequency of PAI-1 Ag genotype in unselected Danish centenarians and adult blood donors. Contrasting results were described by Mannucci et al. [36] showing the 4G/4G genotype, the corresponding genetic marker of high plasma PAI-1 levels, more frequent in centenarians than in young healthy individuals. Notwithstanding, in both these latter studies the authors did not provide plasma PAI-1 levels and, in addition, our control group had an age range different from that reported by Mannucci et al. [36]. Finally, the discrepancy between genotype distribution and plasma PAI-1 levels in the two populations is only apparent since it is widely accepted that PAI-1 4G/5G gene polymorphism only poorly contributes to the variability of plasma PAI-1 concentrations compared to BMI and plasma triglycerides [9–11,34,36]. A potential limitation of our study might be due to the fact that the observed group differences might depend on confounder factors. Our hypothesis is that age remodelling may occur for healthy centenarians. An alternative hypothesis could be that subject ‘destined’ to become healthy centenarians constitute a subgroup with distinctive metabolic features among younger age groups. In conclusion, our data demonstrate that in healthy centenarians plasma PAI-1, T–AT and PAP levels were not associated with the degree of insulin resistance as in aged subjects; further studies will needed to clarify if the lack of association between hypofribinolytic activity and degree of insulin resistance contributes to the low prevalence of cardiovascular diseases in healthy centenarians.
References [1] Ernst E, Resch KL. Fibrinogen as a cardiovascular risk factor: a meta-analysis and review of the literature. Ann Intern Med 1993;118:956 – 63. [2] Meade TW, Mellows S, Brozovic M, Miller GJ, Chakrabarti RR, North WRS, et al. Haemostatic function and ischemic heart disease: principal results of the Northwick Park Study. Lancet 1986;2:533 – 7.
390
M.R. Rizzo et al. / Atherosclerosis 160 (2002) 385–390
[3] Ridker PM, Vaughan DE, Stamptfer MJ, Manson JE, Hennekes CH. Endogenous tissue-type plasminogen activator and risk of myocardial infarction. Lancet 1993;341:1165 –8. [4] Hamsten A, De Faire U, Walldius G, Dahlen G, Szamosi A, Landou C, et al. Plasminogen activator inhibitor in plasma: risk factor for recurrent myocardial infarction. Lancet 1987;2:3 – 9. [5] Di Minno G, Mancini FP, Margaglione M. Haemostatic variables and ischemic cardiovascular disease: do we need a concerted effort for more profitable future clinical investigation? Cardiovasc Drugs Ther 1996;10:743 –9. [6] Ridker PM, Hennekes CH, Stamptfer MJ, Manson JE, Vaughan DE. Prospective study of endogenous tissue plasminogen activator and risk of stroke. Lancet 1994;343:940 – 3. [7] Mansfield MW, Stickland MH, Grant PJ. Environmental and genetic factors in relation to elevated circulating levels plasminogen activator inhibitor-1 in Caucasian patients with non insulin dependent diabetes mellitus. Thromb Haemost 1995;74:842 – 7. [8] Bastard JP, Peironi L, Hainque B. Relationship between plasma plasminogen activator inhibitor 1 and insulin resistance. Diabetes Metab Res Rev 2000;16:192 –201. [9] Meigs JB, Mittleman MA, Nathan DM, Tofler GH, Singer DE, Murphy-Sheely PM, et al. Hyperinsulinemia, hyperglycemia and impaired hemostasis: the Framingham Offspring Study. JAMA 2000;283:221 –8. [10] Landin K, Stigendal L, Eriksson E, Krotkiewski M, Risberg B, Tengborn L, et al. Abdominal obesity is associated with impaired fibrinolytic activity and elevated plasminogen activator inhibitor-1. Metabolism 1990;39:1044 – 8. [11] Asplund-Carlson A, Hamsten A, Wiman B, Carlson LA. Relationship between plasma plasminogen activator inhibitor – 1 activity and VLDL triglyceride concentration, insulin levels and sensitivity: studies randomly selected normo- and hypertriglyceridaemic med. Diabetologia 1993;36:817 –25. [12] Festa A, D’Agostino R, Mykkanen L, Tracy L, Howard BV, Haffner SM. Low density lipoprotein particle size is inversely related to plasminogen activator inhibitor-1 levels. The Insulin Resistance Atherosclerosis Study. Arterioscler Thromb Vasc Biol 1999;19:605 – 10. [13] Paolisso G, Tagliamonte MR, Rizzo MR, Giugliano D. Advancing age and insulin resistance: new facts about an ancient history. Eur J Clin Invest 1999;29:758 –69. [14] Javorschi S, Richard-Harston S, Labrouche S, Manciet G, Freyburger G. Relative influence of age and thrombotic history on hemostatic parameters. Thromb Res 1998;91:241 – 8. [15] Byberg L, Siegbahn A, Berglund L, McKeigue P, Reneland R, Lithell H. Plasminogen activator inhibitor-1 activity is independently related to both insulin sensitivity and serum triglycerides in 70-years old men. Arterioscler Thromb Vasc Biol 1998;18:258 – 64. [16] Paolisso G, Gambardella A, Balbi V, Ammendola S, D’Amore A, Varricchio G. Body composition, body fat distribution and resting metabolic rate in healthy centenarians. Am J Clin Nutr 1995;62:746 – 50. [17] Shaw JE, McCarty D, Zimmet P, De Courten M. Type 2 diabetes worldwide according to the new classification and criteria. Diab Care 2000;23(Suppl 2):B5 – B10. [18] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412 – 9. [19] Bonora E, Targher G, Alberiche M, Bonadonna R, Saggiani F, Zenere MB, et al. Homeostasis model assessment closely mirrors the glucose clamp technique in the assessment of insulin sensitivity. Study in subjects with various degree of glucose tolerance and insulin sensitivity. Diab Care 2000;23:57 –63.
[20] Margaglione M, Cappucci G, D’Addedda M, Colaizzo D, Giuliani N, Vecchione G, et al. PAI-1 plasma levels in a general population without clinical evidence of atherosclerosis: relation to environmental and genetic determinants. Arterioscler Thromb Vascu Biol 1998;18:562 – 7. [21] Potter van Loon BJ, Kluft C, Radder JK, Blankenstein MA, Mienders AE. The cardiovascular risk factor plasminogen activator inhibitor type 1 is related to insulin resistance. Metabolism 1993;42:945 – 9. [22] Paolisso G, Tagliamonte MR, Rizzo MR, Rotondi M, Gualdiero P, Gambardella A, et al. Mean arterial blood pressure and serum levels of the molar ratio of insulin-like growth factor-1 to its binding protein-3 in healthy centenarians. J Hypertension 1999;17:67 – 73. [23] Paolisso G, Gambardella A, Ammendola S, Tagliamonte MR, Rizzo MR, Capurso A, et al. Preserved antilipolytic insulin action is associated with a less atherogenic plasma lipid profile in healthy centenarians. J Am Geriatr Soc 1997;45:1504 –9. [24] Paolisso G, Gambardella A, Ammendola S, D’Amore A, Balbi V, Varricchio M, et al. Glucose tolerance and insulin action in healthy centenarians. Am J Physiol 1996;270:E890 – 4. [25] Paolisso G, Barbieri M, Bonafe` M, Franceschi C. Metabolic age remodelling: the lesson from centenarians. Eur J Clin Invest 2000;30:888 – 94. [26] Gurlek A, Bayraktar M, Kirazli S. Increased plasminogen activator inhibitor-1 activity in offspring of type 2 diabetic patients. Lack of association with plasma insulin levels. Diab Care 2000;23:88 – 92. [27] Alessi MC, Juhan-Vague I, Kooistra T, Declerck PJ, Collen D. Insulin stimulates the synthesis of plasminogen activator inhibitor by human hepatocelluar cell line Hep G2. Thromb Haemost 1988;60:491 – 4. [28] Nordt TK, Schneider DJ, Sobel BE. Augmentation of the synthesis of plasminogen activator inhibitor type-1 by precursor of insulin. Circulation 1994;89:321 – 30. [29] Eliasson M, Roder ME, Dinesen B, Evrin PE, Lindhal B. Proinsulin, intact insulin and fibrinolytic variables and fibrinogen in healthy subjects: a population study. Diab Care 1997;20:1252 – 5. [30] Festa A, D’Agostino R Jr, Mykkanen L, Tracy RP, Zaccaro DJ, Hales CN, et al. Relative contribution of insulin and its precursor to fibrinogen and PAI-1 in a large population with different states of glucose tolerance: the Insulin Resistance Atherosclerosis Study. Arterioscler Thromb Vasc Biol 1999;19:562 – 8. [31] Landin K, Tengborn L, Chmielewska J, Von Schenck H, Smith U. The acute effect of insulin on tissue plasminogen activator inhibitor in man. Thromb Haemost 1991;65:130 – 3. [32] Grant PJ, Kruithof EKO, Felley CP, Felber JP, Bachmann F. Short term infusion of insulin, triacylglycerol and glucose do not cause acute increase in plasminogen actiavtor-1 concentrations in man. Clin Sci 1990;79:513 – 6. [33] Vuorinen-Markkola H, Puhakainen I, Yki-Jarvinen H. No evidence for short-term regulation of plasminogen activator inhibitor activity by insulin in man. Thromb Haemost 1992;67:117 – 20. [34] Mari D, Mannucci PM, Coppola R, Bottasso B, Bauer KA, Rosenberg RD. Hypercoagulabity in centenarians: the paradox of successful aging. Blood 1995;85:3144 – 9. [35] Bladbjerg EM, Andersen Ranberg K, De Maat MP, Kristensen SR, Jeune B, Gram J, et al. Longevity is independent of common variation in genes associated with cardiovascular risk. Thromb Haemost 1999;82:1100 – 5. [36] Mannucci PM, Mari D, Peyvandi L, Tagliabue E, Sacchi E, Taioli E, et al. Gene polymophisms predicting high plasma levels of coagulation and fibrinolysis. A study in centenarians. Arterioscler Thromb Vasc Biol 1997;17:755 – 9.
Elevated plasma activator inhibitor 1 is not related to insulin resistance and to gene polymorphism in healthy centenarians Maria Rosaria Rizzo a, Emilia Ragno a, Michelangela Barbieri a, Domenico De Lucia b, Daniela Manzella a, Maria Rosaria Tagliamonte a, Donatella Colaizzo c, Maurizio Margaglione c, Giuseppe Paolisso a,* a
Department of Geriatric Medicine and Metabolic Diseases, IV Di6isione di Medicina Interna, Second Uni6ersity of Naples, Piazza Miraglia, 2, 80138 Naples, Italy b Institute of General Pathology and Oncology, Second Uni6ersity of Naples, Naples, Italy c Atherosclerosis and Thrombosis Unit, Casa Sollie6o della Sofferenza, San Gio6anni Rotondo, Italy Received 21 February 2001; received in revised form 11 June 2001; accepted 15 June 2001
Abstract Previous studies demonstrated a relationship between the degree of insulin resistance and plasma plasminogen activator inhibitor type-1 (PAI-1) levels. We aim at investigating the relationship between the degree of insulin resistance and plasma PAI-1 levels in aged subjects (n=83) and in healthy centenarians (n= 42). In all subjects the degree of insulin resistance was assessed by HOMA method. Our data demonstrated that healthy centenarians have higher plasma PAI-1 levels (73.1 9 13.9 vs 23.7 914.7 ng/ml, PB 0.001) and lower degree of insulin resistance (1.4 9 0.5 vs 3.3 9 1.3, P B0.001) than aged subjects. In aged subjects plasma PAI-1 levels correlated with the degree of insulin resistance (r= 0.61, P B 0.001), fasting plasma triglycerides (r=0.74, PB 0.001) and age (r=0.33, PB0.001). All such associations were lost in centenarians. Plasma PAI-1 Ag levels were also similar in aged subjects and centenarians even after categorization for PAI gene polymorphism. In multivariate analysis, a model made by age, sex, body mass index, fasting plasma triglycerides, HOMA and PAI-1 gene explained 65 and 50% of plasma PAI-1 level variations in aged subjects and centenarians, respectively. Nevertheless, HOMA (P B 0.001) was significantly and independently associated with plasma PAI-1 levels only in aged subjects. In conclusion, our data demonstrates that in healthy centenarians, plasma PAI-1 were not associated with the degree of insulin resistance as in aged subjects. Frequency of PAI-1 genotype does not provide an explanation for such differences between aged subjects and centenarians. © 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Plasma plasminogen activator inhibitor type-1 levels; Insulin resistance; Centenarians; Aged subjects; PAI-1 gene polymorphism
1. Introduction An unbalanced ratio between plasma levels of coagulation and fibrinolytic factors are associated with an increased risk of arterial thrombosis in cardiovascular [1 – 5] and cerebrovascular [6] diseases. Several factors have been shown to affect plasminogen activator inhibitor type-1 (PAI-1) synthesis and secretion. Genotype might have a role since subjects carrying the
* Corresponding author. Tel.: + 39-081-566-5016; fax: + 39-081566-5051. E-mail address: [email protected] (G. Paolisso).
4G/4G polymorphism (single-base-pair guanine insertion/deletion polymorphism (4G/5G) within the promoter region of the PAI-1 gene) showed the highest plasma PAI-1 levels [7]. In addition, metabolic determinants such as insulin resistance [8,9], body mass index (BMI) [10], plasma triglycerides [11] and small dense LDL [12] should be also taken into account. The relationship between plasma PAI-1 antigen levels and insulin resistance is of particular interest in human aging. In fact, advancing age has been associated with an increased insulin resistance [13] and elevated risk of atherothrombotic events [14,15]. Thus, aged subjects may be more prone to atherothrombotic events due to the impact of insulin resistance on PAI-1 levels.
0021-9150/02/$ - see front matter © 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 2 1 - 9 1 5 0 ( 0 1 ) 0 0 5 7 5 - 5
386
M.R. Rizzo et al. / Atherosclerosis 160 (2002) 385–390
A special group of aged subjects are the centenarians who represent the best model of successful aging; in fact, these subjects have escaped age-related atherothrombotic and metabolic diseases and reached the extreme limits of lifespan. The possibility that particular environmental, genetic, and/or haemostatic profiles prevent major diseases in these individuals cannot be ruled out and it needs to be clarified. Thus, our study aims at evaluating the possible relationship among plasma PAI-1 levels, insulin resistance and PAI-1 gene polymorphism in healthy aged subjects and centenarians.
2. Materials and methods
the Ethical Committee of our Institution and each subject, as well as their relatives for centenarians, provided a written informed consent before being enrolled into the study.
2.2. Metabolic determination Anthropometric determinations were made as previously reported [16]. At enrolment all subjects underwent an oral glucose tolerance test for assessing the degree of glucose intolerance [17]. Degree of insulin resistance was estimated by homeostatic (HOMA) method [18]. This method correlates well with insulin sensitivity measured by euglycemic clamp (r=0.88, PB 0.0001) [18,19].
2.1. Subjects Eighty-three aged subjects (mean age: 74.59 7.0 years) and 42 centenarians (age\100 years) living in the five districts of the Campania region, a geographically and administratively well-defined Mediterranean area located on the West Coast of Southern Italy, volunteered for the study. The electoral roll for a referendum held on 18 April 1999, was used as a population source and the subjects were examined between July and December 1999. The subjects were contacted at home or in their institution and examined by physicians previously trained to administer a questionnaire that included cognitive and depression tests. All subjects were in good health, as determined by a medical history, physical examination, and routine screening laboratory analyses, and were free from any drug known to interfere with glucose metabolism. More detailed information concerning the two study populations is given in Table 1. The study was approved by
2.3. Determination of PAI-1 genotype by polymerase chain reaction The PAI-1 4G/5G polymorphism was evaluated as previously reported [20]. Briefly, a mutated oligonucleotide was synthesized that inserts a site for the Bsl I enzyme within the amplification product. Polymerase chain reaction was carried out in 50-ml samples in a Perkin–Elmer – Cetus thermal cycler. Each sample contained 0.5 mg of genomic DNA, 15 pmol of each primer, 100 mmol/l of NTP, 10 mmol/l Tris –HCl, pH 8.3, 50 mmol/l KCl, 1.5 mol/l MgCl2, and 1 U thermostable Taq polymerase. The 30 cycles consisted of steps at 95 °C for 1 min, 60 °C for 1 min, and 72 °C for 2 min. Then, 20-ml volumes of the amplification products were digested for 2.5 h at 55 °C with 5 U of the Bsl I restriction enzyme. The fragments were fractionated by 4% agarose gel electrophoresis and visualized under UV light.
Table 1 Clinical characteristics of the two study groups
Age range (years) Gender (M/F) BMI (kg/m2) WHR FP glucose (mmol/l) 2 h plasma glucose (mmol/l) FP insulin (pmol/l) FP cholesterol (mmol/l) FP triglycerides (mmol/l) HOMA index PAI-1 (ng/ml) PAP (ng/ml) TAT (ng/ml)
Aged subjects (n = 83)
Centenarians (n = 42)
P
62–90 34/49 26.5 91.4 0.84 90.04 6.29 1.1 7.29 1.3 11.79 3.4 5.49 0.8 1.7 90.5 3.39 1.3 23.7914 320 994 1.5 90.8
100–105 15/27 21.6 91.2 0.83 9 0.03 4.2 9 0.9 5.3 9 1.1 7.4 9 1.6 4.6 9 0.9 1.1 9 0.4 1.4 9 0.5 73.1 913 575 9 87 4.1 9 1.0
B0.001 0.02* B0.001 0.13 B0.001 B0.001 B0.001 B0.001 B0.001 B0.001 B0.001 B0.001 B0.001
All results are expressed as mean 9S.D. For testing differences between the two groups unpaired Student’s t-test was used. For investigating gender differences 2 test was used; d.f. = 1. BMI, Body mass index; WHR, waist/hip ratio; FP, fasting plasma; PAI-1, plasminigen activator inhibitor type-1; PAP, plasmin–antiplasmin complex; TAT, thrombin–antithrombin complex.
M.R. Rizzo et al. / Atherosclerosis 160 (2002) 385–390
387
2.4. Analytical methods After an overnight fast, blood samples were drawn for measurements of plasma metabolite concentrations. Plasma glucose level was immediately determined by enzymatic method (Autoanalyzer, Beckman). Blood samples for insulin measurements were collected in heparinized tubes. After centrifugation, plasma insulin concentration was determined by a commercially available radioimmunoassay kit (coefficient of variation (c.v.) 3.29 0.3%; cross-reactivity with proinsulin 0.3%; Sorin, Biomedical). Commercial enzymatic methods were used for plasma total cholesterol (Monotest, Boehringer Mannheim, Milan, Italy; c.v.= 3.8 9 0.9%) and triglyceride (Peridecrome, Boehringer Mannheim; c.v.=4.19 0.6%) levels determination. PAI-1 Ag levels were measured by enzyme-linked immunoabsorbent assay (ELISA) technique (Byk Gulden, Milan, Italy; c.v.=14 9 3.5%). Thrombin–antithrombin complex (T– AT), a measure of thrombin generation and neutralisation by antithrombin, was assayed by ELISA (Behring Werke, Marburg, Germany). In a subgroup of 63 aged subjects and 32 centenarians plasma plasmin–antiplasmin complexes (PAP) were determined by ELISA (Behring Werke).
2.5. Calculation and statistical analyses To approximate normal distribution, plasma insulin, triglycerides and PAI-1 Ag levels were log transformed for use in all calculations, then back-transformed for presentation of results. All results are expressed as mean 9standard deviation (S.D.). For testing differences between the two study groups, unpaired Student’s t test was used for continuous variables. The significance of the difference in gene frequency between groups was analysed by a 2 test. ANOVA with the Scheffe` test was used for comparing the PAI-1 Ag levels among the genotype groups. Simple correlation by the Pearson method allowed us to assess univariate relations. Linear multiple regression analysis was used for investigating the different contribution of each covariate to the variability of PAI-1 Ag levels. In this latter analysis, gender difference was calculated by considering female=0 and male = 1. In a same way, PAI-1 genotypes were categorised as follows: 4G/4G =0, 4G/ 5G = 1, and 5G/5G =2. Data were analysed using the Statistical Package for Social Science (SPSS).
3. Experimental results Clinical characteristics of the two study groups are reported in Table 1. Gender ratio demonstrated a prevalence of female in centenarians. This latter group
Fig. 1. Simple correlations between HOMA index and plasma PAI-1 levels in aged subjects (top) and centenarians (bottom).
of subjects had also a significantly lower BMI, fasting plasma glucose, insulin, cholesterol, and triglycerides, and HOMA index than aged subjects. In contrast, plasma PAI-1 Ag, T–AT, and PAP levels were significantly higher in centenarians than in aged subjects. In particular, plasma PAI-1 Ag levels were three times higher in centenarians than in the aged subjects. Plasma PAI-1 Ag levels correlated with HOMA in aged subjects but not in centenarians (Fig. 1). In aged subjects, plasma PAI-1 Ag levels correlated also with age, fasting plasma glucose, insulin, triglycerides, T– AT and PAP levels. In contrast, centenarians had plasma PAI-1 Ag levels correlated only with age, fasting plasma triglycerides and PAP concentrations (Table 2). In aged subjects, plasma T–AT levels correlated with age, BMI, fasting plasma glucose, insulin, triglycerides levels and HOMA index, while plasma PAP correlates with fasting plasma glucose, insulin, triglycerides levels and HOMA index. In contrast, all these correlations were not significant in centenarians (Table 2). As far as PAI-1 gene polymorphism is concerned, the frequency of the 4G allele was 0.532 and the frequency
M.R. Rizzo et al. / Atherosclerosis 160 (2002) 385–390
388
Table 2 Simple correlations among PAI-1 Ag, TAT, PAP levels and anthropometric and metabolic parameters Aged subjects (n= 83)
Age BMI WHR Plasma glucose Insulin Total cholesterol Triglycerides HOMAa TAT PAP
Centenarians (n =42)
PAI-1 Aga
P
T–AT
P
PAP
P
PAI-1 Aga
P
T–AT
P
PAP
P
0.33 0.55 0.04 0.41 0.56 0.21 0.74 0.61 0.56 0.83
0.002 0.625 0.71 B0.001 B0.001 B0.05 B0.001 B0.001 B0.001 B0.001
0.46 0.22 0.11 0.34 0.26 0.20 0.34 0.35 – 0.64
B0.001 0.05 0.30 B0.001 0.01 0.06 B0.001 B0.001 – B0.001
0.22 0.09 0.004 0.30 0.49 0.25 0.60 0.52 0.64 –
0.09 0.94 0.97 B0.01 B0.001 B0.05 B0.001 B0.01 B0.001 –
0.47 −0.12 0.05 0.11 0.03 0.25 0.69 0.12 0.07 0.47
B0.001 0.40 0.71 0.46 0.80 0.09 B0.001 0.42 0.64 B0.005
0.22 −0.08 0.17 0.06 0.01 −0.01 −0.16 −0.15 – 0.31
0.15 0.60 0.28 0.90 0.92 0.94 0.28 0.30 – 0.07
0.36 −0.03 0.07 −0.13 0.15 −0.05 0.21 −0.01 0.32 –
0.05 0.86 0.68 0.47 0.42 0.77 0.21 0.94 0.09 –
a
Back log transformed; BMI, body mass index; WHR, waist/hip ratio; PAI-1, plasminigen activator inhibitor type-1; TAT, thrombin–antithrombin complex; PAP, plasmin–antiplasmin complex.
of the 5G allele was 0.468; the frequency of 4G/4G, 4G/5G and 5G/5G genotypes was 0.325, 0.410 and 0.265, respectively. Genotype and allele distributions were in Hardy–Weinberg equilibrium. Furthermore, the allele frequency and the genotype distribution did not differ between aged subjects and centenarians (Table 3). No differences in plasma PAI-1 Ag levels between aged subjects and centenarians were found even after categorization for PAI gene polymorphism (Table 4). Due to the occurrence of several determinants of plasma PAI-1 Ag levels variability, multiple stepwise linear regression analysis investigates the independent contribution of main covariates in the two study groups. A model made by age, sex, BMI, fasting plasma triglycerides, HOMA index, and PAI-1 gene polymorphism explained 65% of the variability of plasma PAI-1 Ag levels in aged subjects with age (PB 0.001), sex (P B0.04), BMI (PB 0.03), fasting plasma triglycerides (P B 0.001), and HOMA index (PB 0.001) significantly and independently associated with plasma PAI-1 Ag levels. The same model in centenarians explained 50% of the variability of the plasma PAI-1 Ag levels with only age (PB 0.04) and fasting plasma triglycerides (P B0.001) significantly and independently associated with the dependent variable.
PAI-1 is a potent inhibitor of fibrinolysis by binding and rapidly inactivating both tissue- and urokinase-type plasminogen activator. Increased levels of PAI-1 and decreased plasma fibrinolytic activity have been demonstrated in survivor of acute myocardial infarction and in patients with coronary artery stenosis [1–4]. Furthermore, PAI-1 has prognostic value in predicting recurrence of myocardial infarction [4]. A positive association between fasting plasma insulin levels and PAI-1 has been demonstrated by several investigations [21]. In addition, PAI-1 has been linked to separate feature of syndrome X including abdominal obesity, hypertension, hypertriglyceridemia, glucose intolerance and type 2 diabetes mellitus [21]. The association of PAI-1 with features of syndrome X supports the link between plasma PAI-1 levels and insulin resistance. In our study, we confirmed that in aged subjects there is a strong relationship between plasma PAI-1 levels and degree of insulin resistance (measured by HOMA) and/ or plasma triglycerides. In contrast, an unexpected finding was the lack of a relationship between plasma PAI-1 levels and degree of insulin resistance in healthy centenarians. In fact, several previous studies have shown the absence in healthy centenarians of cardiovascular risk factors such as obesity [16], hypertension [22], Table 3 PAI-1 gene polymorphism distribution in the two populations
4. Discussion Our study confirms that healthy centenarians have higher levels of PAI-1 Ag levels, T– AT, and PAP than aged subjects; furthermore, we also demonstrate that (a) plasma PAI-1 Ag levels were strongly correlated with HOMA index in aged subjects but not in centenarians, (b) analysis of PAI-1 genotype did not explain plasma PAI-1 levels either in aged subjects or in centenarians.
Aged (n =83)
Centenarians (n = 42)
4G/4G 4G/5G 5G/5G
27 (32.5%) 34 (41%) 22 (26.5%)
15 (35.7%) 15 (35.7%) 12 (28.6%)
Allele frequency 4G 5G
88 (53%) 78 (47%)
45 (53.6%) 39 (46.4%)
For investigating difference in gene frequency x2 test was used; d.f. = 2; P= NS for genotype distribution.
M.R. Rizzo et al. / Atherosclerosis 160 (2002) 385–390 Table 4 PAI-1Ag levels in different PAI-1 gene polymorphisms
Aged (n = 83) Centenarians (n =42)
4G/4G
4G/5G
5G/5G
21.2912.3 72.2913.7
25.4 916.2 69.3 9 12.1
24.6 916.3a 78.3 917.3b
ANOVA with the Scheffe` test was used for comparing the PAI-1 Ag levels among the genotype groups d.f. = 2. a P = 0.43. b P = 0.27.
dyslipidemia [23], and insulin resistance or impaired glucose tolerance [24]. Thus, it was hypothesized that age-remodelling occurs in healthy centenarians and that such a process is also responsible for the achievement of extreme life span [25]. Indeed, the evidence that healthy centenarians have higher plasma PAI-1, PAP and T– AT levels than aged subjects seems in contrast either with the age-remodelling hypothesis [25] or with the well-known role of PAI-1 and insulin resistance in the pathogenesis of atherosclerosis [1– 9]. Why plasma PAI1 levels are not related to the degree of insulin resistance in healthy centenarians is still unknown. Nevertheless, a very recent study [26] has shown also in nonobese normal glucose tolerant offsprings of type 2 diabetic patients the lack of association between plasma insulin levels — a proxy of insulin resistance—and plasma PAI1 concentrations. Indeed, controversy exists with respect to the relationship between plasma insulin and PAI-1 activity. In vitro, data have shown that both insulin [27] and its precursor [28] stimulate PAI-1 synthesis from hepatocytes. Accordingly, a relationship between insulin and/or its precursors and PAI-1 activity has been shown in a large, healthy normal glucose tolerant population for northern Sweden [29]. In that study, however, neither proinsulin nor insulin concentrations predicted the serum PAI-1 activity in a multivariate linear regression model. In the Insulin Resistance Atherosclerosis Study [30], a strong and independent relationship between PAI-1 antigen and insulin has been found consistently across varying degrees of glucose tolerance. In vivo, insulin infusion has been shown to decrease [31] or to have no affect on [32] PAI-1 activity. Furthermore, it has been suggested that in vitro, the stimulatory effect of insulin on PAI-1 synthesis in hepatocytes may be counteracted by an inhibitory effect of insulin on PAI-1 synthesis and release in extrahepatic tissues, or by stimulation of PAI-1 degradation by insulin [33]. These complicated interactions between insulin and PAI-1 may explain why we found an association between insulin and PAI-1 activity in aged subjects but not in the special group of centenarian subjects, and thus, we hypothesize that in healthy centenarians elevated plasma PAI-1 levels might be determined by or related to other factors than plasma insulin levels/degree of insulin resistance.
389
Additional findings of our study were (a) healthy centenarians have plasma PAI-1, PAP and T– AT levels more elevated than in aged subjects; (b) lack of difference in genotype distribution between centenarians and aged subjects despite the evidence that centenarians have higher plasma PAI-1 Ag levels than aged subjects. As far as plasma PAI-1, PAP and T–AT levels are concerned, our data are in agreement with a previous report showing hypercoagulabity in centenarians [34]. With regard to the lack of difference in genotype distribution between aged subjects and centenarians, due to the limited number of centenarians studied, caution should be used before generalizing such results. Nevertheless, it should be pointed out that our data are in agreement with Bladbjerg et al. [35] showing comparable frequency of PAI-1 Ag genotype in unselected Danish centenarians and adult blood donors. Contrasting results were described by Mannucci et al. [36] showing the 4G/4G genotype, the corresponding genetic marker of high plasma PAI-1 levels, more frequent in centenarians than in young healthy individuals. Notwithstanding, in both these latter studies the authors did not provide plasma PAI-1 levels and, in addition, our control group had an age range different from that reported by Mannucci et al. [36]. Finally, the discrepancy between genotype distribution and plasma PAI-1 levels in the two populations is only apparent since it is widely accepted that PAI-1 4G/5G gene polymorphism only poorly contributes to the variability of plasma PAI-1 concentrations compared to BMI and plasma triglycerides [9–11,34,36]. A potential limitation of our study might be due to the fact that the observed group differences might depend on confounder factors. Our hypothesis is that age remodelling may occur for healthy centenarians. An alternative hypothesis could be that subject ‘destined’ to become healthy centenarians constitute a subgroup with distinctive metabolic features among younger age groups. In conclusion, our data demonstrate that in healthy centenarians plasma PAI-1, T–AT and PAP levels were not associated with the degree of insulin resistance as in aged subjects; further studies will needed to clarify if the lack of association between hypofribinolytic activity and degree of insulin resistance contributes to the low prevalence of cardiovascular diseases in healthy centenarians.
References [1] Ernst E, Resch KL. Fibrinogen as a cardiovascular risk factor: a meta-analysis and review of the literature. Ann Intern Med 1993;118:956 – 63. [2] Meade TW, Mellows S, Brozovic M, Miller GJ, Chakrabarti RR, North WRS, et al. Haemostatic function and ischemic heart disease: principal results of the Northwick Park Study. Lancet 1986;2:533 – 7.
390
M.R. Rizzo et al. / Atherosclerosis 160 (2002) 385–390
[3] Ridker PM, Vaughan DE, Stamptfer MJ, Manson JE, Hennekes CH. Endogenous tissue-type plasminogen activator and risk of myocardial infarction. Lancet 1993;341:1165 –8. [4] Hamsten A, De Faire U, Walldius G, Dahlen G, Szamosi A, Landou C, et al. Plasminogen activator inhibitor in plasma: risk factor for recurrent myocardial infarction. Lancet 1987;2:3 – 9. [5] Di Minno G, Mancini FP, Margaglione M. Haemostatic variables and ischemic cardiovascular disease: do we need a concerted effort for more profitable future clinical investigation? Cardiovasc Drugs Ther 1996;10:743 –9. [6] Ridker PM, Hennekes CH, Stamptfer MJ, Manson JE, Vaughan DE. Prospective study of endogenous tissue plasminogen activator and risk of stroke. Lancet 1994;343:940 – 3. [7] Mansfield MW, Stickland MH, Grant PJ. Environmental and genetic factors in relation to elevated circulating levels plasminogen activator inhibitor-1 in Caucasian patients with non insulin dependent diabetes mellitus. Thromb Haemost 1995;74:842 – 7. [8] Bastard JP, Peironi L, Hainque B. Relationship between plasma plasminogen activator inhibitor 1 and insulin resistance. Diabetes Metab Res Rev 2000;16:192 –201. [9] Meigs JB, Mittleman MA, Nathan DM, Tofler GH, Singer DE, Murphy-Sheely PM, et al. Hyperinsulinemia, hyperglycemia and impaired hemostasis: the Framingham Offspring Study. JAMA 2000;283:221 –8. [10] Landin K, Stigendal L, Eriksson E, Krotkiewski M, Risberg B, Tengborn L, et al. Abdominal obesity is associated with impaired fibrinolytic activity and elevated plasminogen activator inhibitor-1. Metabolism 1990;39:1044 – 8. [11] Asplund-Carlson A, Hamsten A, Wiman B, Carlson LA. Relationship between plasma plasminogen activator inhibitor – 1 activity and VLDL triglyceride concentration, insulin levels and sensitivity: studies randomly selected normo- and hypertriglyceridaemic med. Diabetologia 1993;36:817 –25. [12] Festa A, D’Agostino R, Mykkanen L, Tracy L, Howard BV, Haffner SM. Low density lipoprotein particle size is inversely related to plasminogen activator inhibitor-1 levels. The Insulin Resistance Atherosclerosis Study. Arterioscler Thromb Vasc Biol 1999;19:605 – 10. [13] Paolisso G, Tagliamonte MR, Rizzo MR, Giugliano D. Advancing age and insulin resistance: new facts about an ancient history. Eur J Clin Invest 1999;29:758 –69. [14] Javorschi S, Richard-Harston S, Labrouche S, Manciet G, Freyburger G. Relative influence of age and thrombotic history on hemostatic parameters. Thromb Res 1998;91:241 – 8. [15] Byberg L, Siegbahn A, Berglund L, McKeigue P, Reneland R, Lithell H. Plasminogen activator inhibitor-1 activity is independently related to both insulin sensitivity and serum triglycerides in 70-years old men. Arterioscler Thromb Vasc Biol 1998;18:258 – 64. [16] Paolisso G, Gambardella A, Balbi V, Ammendola S, D’Amore A, Varricchio G. Body composition, body fat distribution and resting metabolic rate in healthy centenarians. Am J Clin Nutr 1995;62:746 – 50. [17] Shaw JE, McCarty D, Zimmet P, De Courten M. Type 2 diabetes worldwide according to the new classification and criteria. Diab Care 2000;23(Suppl 2):B5 – B10. [18] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412 – 9. [19] Bonora E, Targher G, Alberiche M, Bonadonna R, Saggiani F, Zenere MB, et al. Homeostasis model assessment closely mirrors the glucose clamp technique in the assessment of insulin sensitivity. Study in subjects with various degree of glucose tolerance and insulin sensitivity. Diab Care 2000;23:57 –63.
[20] Margaglione M, Cappucci G, D’Addedda M, Colaizzo D, Giuliani N, Vecchione G, et al. PAI-1 plasma levels in a general population without clinical evidence of atherosclerosis: relation to environmental and genetic determinants. Arterioscler Thromb Vascu Biol 1998;18:562 – 7. [21] Potter van Loon BJ, Kluft C, Radder JK, Blankenstein MA, Mienders AE. The cardiovascular risk factor plasminogen activator inhibitor type 1 is related to insulin resistance. Metabolism 1993;42:945 – 9. [22] Paolisso G, Tagliamonte MR, Rizzo MR, Rotondi M, Gualdiero P, Gambardella A, et al. Mean arterial blood pressure and serum levels of the molar ratio of insulin-like growth factor-1 to its binding protein-3 in healthy centenarians. J Hypertension 1999;17:67 – 73. [23] Paolisso G, Gambardella A, Ammendola S, Tagliamonte MR, Rizzo MR, Capurso A, et al. Preserved antilipolytic insulin action is associated with a less atherogenic plasma lipid profile in healthy centenarians. J Am Geriatr Soc 1997;45:1504 –9. [24] Paolisso G, Gambardella A, Ammendola S, D’Amore A, Balbi V, Varricchio M, et al. Glucose tolerance and insulin action in healthy centenarians. Am J Physiol 1996;270:E890 – 4. [25] Paolisso G, Barbieri M, Bonafe` M, Franceschi C. Metabolic age remodelling: the lesson from centenarians. Eur J Clin Invest 2000;30:888 – 94. [26] Gurlek A, Bayraktar M, Kirazli S. Increased plasminogen activator inhibitor-1 activity in offspring of type 2 diabetic patients. Lack of association with plasma insulin levels. Diab Care 2000;23:88 – 92. [27] Alessi MC, Juhan-Vague I, Kooistra T, Declerck PJ, Collen D. Insulin stimulates the synthesis of plasminogen activator inhibitor by human hepatocelluar cell line Hep G2. Thromb Haemost 1988;60:491 – 4. [28] Nordt TK, Schneider DJ, Sobel BE. Augmentation of the synthesis of plasminogen activator inhibitor type-1 by precursor of insulin. Circulation 1994;89:321 – 30. [29] Eliasson M, Roder ME, Dinesen B, Evrin PE, Lindhal B. Proinsulin, intact insulin and fibrinolytic variables and fibrinogen in healthy subjects: a population study. Diab Care 1997;20:1252 – 5. [30] Festa A, D’Agostino R Jr, Mykkanen L, Tracy RP, Zaccaro DJ, Hales CN, et al. Relative contribution of insulin and its precursor to fibrinogen and PAI-1 in a large population with different states of glucose tolerance: the Insulin Resistance Atherosclerosis Study. Arterioscler Thromb Vasc Biol 1999;19:562 – 8. [31] Landin K, Tengborn L, Chmielewska J, Von Schenck H, Smith U. The acute effect of insulin on tissue plasminogen activator inhibitor in man. Thromb Haemost 1991;65:130 – 3. [32] Grant PJ, Kruithof EKO, Felley CP, Felber JP, Bachmann F. Short term infusion of insulin, triacylglycerol and glucose do not cause acute increase in plasminogen actiavtor-1 concentrations in man. Clin Sci 1990;79:513 – 6. [33] Vuorinen-Markkola H, Puhakainen I, Yki-Jarvinen H. No evidence for short-term regulation of plasminogen activator inhibitor activity by insulin in man. Thromb Haemost 1992;67:117 – 20. [34] Mari D, Mannucci PM, Coppola R, Bottasso B, Bauer KA, Rosenberg RD. Hypercoagulabity in centenarians: the paradox of successful aging. Blood 1995;85:3144 – 9. [35] Bladbjerg EM, Andersen Ranberg K, De Maat MP, Kristensen SR, Jeune B, Gram J, et al. Longevity is independent of common variation in genes associated with cardiovascular risk. Thromb Haemost 1999;82:1100 – 5. [36] Mannucci PM, Mari D, Peyvandi L, Tagliabue E, Sacchi E, Taioli E, et al. Gene polymophisms predicting high plasma levels of coagulation and fibrinolysis. A study in centenarians. Arterioscler Thromb Vasc Biol 1997;17:755 – 9.