Protocol design was approved by the Ethical Committee from University of Colima (No. 2006/54); this protocol is in compliance with the Helsinki Declaration.

One hundred and fifty healthy non-smoking Mexican male subjects were recruited by advertisement. Volunteers had no clinical history of hypertension, diabetes or inflammatory disorders. Haematological and biochemical tests, including white blood cell count (WBC) and C-reactive protein (CRP) were determined to reject subjects with sub clinical inflammation; written informed consent was obtained from each participant after the procedures had been fully explained. Thus, samples from subjects having a BMI lower than 18.5 Kg/m² or higher than 24.9 Kg/m² were excluded from this study. Blood samples of volunteers were taken in the morning to avoid normal circadian variation in proteins concentrations 11.

PAI-1 concentration was determined by using HPAIKT kit (Stratech Scientific, Suffolk, England) and following the manufacturer instructions.

Explants of dermis (2 mm³) with attached epidermis were aseptically removed from donors. Subsequently, tissue was enzymatically dissociated by 5-min incubation with 0.05% trypsin, prepared in phosphate buffer saline (PBS). Fibroblasts were harvested, then plated and maintained in DMEM (Life Technologies, Grand Island, NY) plus 10% fetal calf serum (FCS; Life Technologies, Grand Island, NY), at 37°C in 5% CO₂ atmosphere. In order to obtain fibroblasts with their cellular cycles synchronized, cultures were grown to confluence and then serum-starved for 72 h (culture medium containing 0.5% serum). Thereafter, cells were re-fed (DMEM + 10% FCS) and twenty four hours later total RNA and DNA were isolated from cells.

C1367R and L1074F WRN polymorphisms and PAI-1 promotor polymorphism -675 4G/5G were determined by automatic sequencing. Polymorphic fragments of WRN and PAI-1 were amplified from genomic DNA by polymerase chain reaction (PCR) in a cycling instrument iCycler (Bio-Rad, Hercules, CA), with the following primers: CR1367F: GCCTAATCAGAATGTTAGTT and CR1367R: CCTCAGTATTGATGCCTACTTC; LF1074F: CTTGTGAAGAGGCCTATAAACTGG and LF1074R: GCCTCATCCTTCAAGCTAATGCAG; PAIF: TACCATGGTAACCCCTGGTC and PAIR: GGATCTGTTAGTGCACCGAC. Sequencing reactions were performed with a Big Dye Terminator V3.1 Sequencing kit (Applied Biosystems, Foster City, CA) following the manufacturer instructions, purified by DyeEx Spin kit (QIAGEN, Valencia, CA) and automatically sequenced by ABI PRISM 310 Genetic Analyzer (Applied Biosystems, Foster City, CA).

Confluent layers of skin fibroblasts were used for RNA isolation. Total RNA was isolated by Trizol Reagent (Gibco BRL, New York, NY). RNA was used to perform a semiquantitative RT-PCR analysis for PAI-1 and WRN expression For each sample 1.0 μg RNA was reverse transcribed and amplified by Superscript One-Step RT-PCR kit (Invitrogen, Carlsbad, CA) following the manufacturer's instructions using the following set of intron spanning primers, annealed into the coding sequence in order to determine splicing variants 12: ActinF: CTCGTCGTCGACAACCGGCT, ActinR: GCCAATGGTGATGACCTGGC, PAIF: CAAGGATGAGATCAGCACCACAG, PAIR: GGGTTCCATCACTTGGCCCATG WRNF: TGACTTCAATCTCTGAGGAAG and WRNR: CAGCTCTACCAATCTCCTGA. PCR products were separated in an agarose gel (2%) electrophoresis and stained with ethidium bromide. The intensity of the fluorescence was automatically measured and integrated by gel data acquisition "Gel Doc" (Bio-Rad, Hercules, CA) software. All RT-PCR experiments were done in triplicate and 26 amplification cycles were used.

The iRNA sequences were as follows: sense UAGUGGCACGGUAGAACCAdTT, antisense UUGUUCUACCGUGCCACUdTT. A scrambled sequence was designed as a mismatch control: sense GUCACACGAUAGCGAGGUAdTT and antisense UACCUCGCUAUCGUGUGACdTT. These RNA oligonucleotides were synthesized individually, deprotected and purified by RNase-free HPLC (Invitrogen, Carlsbad, CA). iRNA and mismatch RNA (scrambled) duplexes were formed following the manufacturer instructions.

Five fibroblast cultures presenting heterozygosis for C1367R WRN and for F1074L WRN (1367CR/1074FL genotype) were synchronized and re-fed (DMEM/10%SFC), to be immediately used for iRNA treatment. Cells were transfected with 2 μg siRNA/i-Fect complexes (Invitrogen, Carlsbad, CA) in a 1:4 ratio (w/v) per plate. Twenty four hours after transfection, cells were treated with TNFα (150U/mL; Genzyme. Cambridge, MA), TGFβ (18U/mL; Genzyme. Cambridge, MA) or insulin (1 mg/mL; Sigma. St. Louis, MO). Twenty four hours later, cells were harvested and total RNA was isolated. PAI-1 and WRN expression was evaluated by RT-PCR. The same procedure was followed for scrambled RNA and control.

Fibroblast cultures with genotypes 1367CC/1074FF and 1367RR/1074FL were treated equally, except for iRNA treatment. Five cultures for each genotype were used.

Cultured fibroblasts treated and untreated with iRNA or scrambled molecule were lysed with sodium dodecyl sulfate (SDS) sample buffer. Proteins were quantified using the Bradford assay reagent. Equal amounts of protein (50 μg) were separated under denaturing conditions, by SDS-PAGE on 10–20% polyacrylamide Tris-glycine gels, and then electroblotted to polyvinylidene difluoride (PVDF) membrane. Non-specific sites were blocked with 5% non fat dry milk in 0.2% Tween-20 in Tris-buffered saline (TBS-T) for 1 hour at room temperature then reacted with the primaries anti-WRN and anti PAI-1 antibodies for 1 hour, followed by horseradish peroxidase (HRP)-labeled anti-rabbit IgG (Jackson Laboratories) for 1 hour. The bands were visualized using an ECL Plus kit (Amersham Pharmacia Biotech, Amersham, UK). β-actin was detected as a protein loading control in the same membrane.

Values are presented as mean ± SD or mean ± SEM. The difference of PAI-1 levels between genotypes was analyzed by paired Student's test. One-way ANOVA was used to test differences in PAI-1 expression trough WRN genotypes and in WRN knocking down experiments across treatments (control, scrambled and iRNA). Significant results from ANOVA were further analyzed by Tukey's post hoc test. P < 0.05 was considered significant.

PAI-1 is commonly present at low levels in plasma form healthy humans; however, acute and chronic diseases are strongly associated with increased PAI-1 expression and release 13. Therefore, to avoid that PAI-1 quantification was not influenced by sub-clinical inflammation, different serum markers of inflammation were measured as well, thereby trying to avoid high or low levels of PAI-1 independently of WRN genotype. Thus, plasmatic concentration values of C-reactive protein (CRP), an acute phase protein, fibrinogen, glucose, creatinine (through others markers) are shown in Table 1. Such normal values demonstrate that the studied population (n = 150) did not have evidence of inflammatory processes.

In order to know whether the amount of PAI-1 in plasma is related of WRN genotype, PAI-1 concentration was determined as described in methods. Subsequently, WRN polymorphisms (rs1346044 and rs2725362) were evaluated from genomic DNA. Table 2 summarizes the amount of PAI-1 from a Mexican male population endowed with the following WRN genotypes: 1367CC/1074FF, a genotype whose alleles are not related to cardiovascular disease protection and 1367 CR/1074LL, 1367 RR/1074FL or 1367 RR/1074LL, whose alleles are associated with protection against cardiovascular disease 67. It was found that the individuals with 1367CC/1074FF genotype systematically revealed higher levels of PAI-1 as compared with those with genotypes whose alleles were associated with protection to cardiovascular disease (P = 0.031).

To address whether WRN polymorphisms can modulate PAI-1 expression, a semiquantitative RT-PCR analysis of PAI-1 gene expression was done in an array of primary cultured fibroblasts obtained from the same Mexican male population. Constant amounts of reverse-transcribed mRNA were amplified together with the constitutive gene β-actin to normalize PAI-1 expression. Figure 1 shows that fibroblasts from carriers 1367RR and/or 1074LL expressed the lowest PAI-1 levels, whereas those carrying the 1367CC and/or 1074FF alleles expressed the highest ones. These results suggest that WRN may modulate PAI-1 expression or, alternatively, its differential expression may depend on PAI-1-promoter polymorphisms such as -675 4G/5G 14. To evaluate the latter possibility, fibroblasts used for the above experiments were further genotyped for -675 4G/5G. There were no significant differences in the steady-state level of PAI-1 among groups with different PAI-1 genotypes (data no shown).

To search for the potential regulatory role of WRN in PAI-1 expression, five cultures of heterozygous fibroblasts carrying the WRN genotype, 1367CR/1074LF, were transfected with a specific iRNA against WRN mRNA to transiently block its expression. The effectiveness of WRN silencing was evaluated by western blot analysis for each cell culture, as shown in the autoradiography in Figure 2A. All of those cell cultures treated with iRNA against WRN messenger, were not reactive (or scarcely reactive) to the antibody against WRN. In contrast, cell cultures treated or not with the scrambled molecule presented a high reactive signal (data no shown). Thereafter, mRNA levels of WRN and PAI-1 were measured in those knocked-down cultures. As expected, fibroblasts showed significant reduced levels of WRN mRNA at 24 and 48 h post-transfection (open bars in Figure 2D). During the same period there was a parallel large enhancement of PAI-1 expression that reached a maximum of ~400 % 48 h upon transfection with the iRNA (filled bars in Figure 2D). The effect of the iRNA on levels of WRN mRNA and PAI-1 mRNA were reversed after 72 h post-transfection. In contrast, the expression of neither WRN nor PAI-1 was modified by the scrambled control RNA molecule (Figure 2C). Moreover, there were no significant differences in WRN and PAI-1 expression between treated and untreated fibroblasts with the scrambled RNA molecule (data not shown). Taken together, these results suggest that WRN tightly regulates PAI-1 expression.

It is well known that signaling molecules including TNFα, TGFβ and insulin are inducers of PAI-1 expression 1516. To search for the potential cross-talk between the WRN-dependent regulation of PAI-1 expression and the signaling pathways used by those physiological PAI-1-inducers, TNFα (TGFβ, or insulin) was added 24 hours after heterozygous fibroblasts (1367 CR/1074LF) were transfected with iRNA against WRN. PAI-1 and WRN expression was measured 24 hours after the addition of inducers to coincide their maximum effect (24 h) with that of iRNA (48 h). In cells transfected with the scrambled molecule, TNFα (TGFβ, or insulin) produced a large increase of PAI-1 expression (Figure 3A, filled bars) without any noticeable effect on expression of WRN (Figure 3A, open bars). As expected, in fibroblasts transfected with the iRNA, there was a reduction in WRN mRNA (Figure 3B, open bars) and in that condition, TNFα, TGFβ, or insulin produced a further overexpression of PAI-1 (Figure 3B, filled bars). WRN expression showed no significant differences among treatments, neither in the scrambled nor in iRNA transfected cells (Figure 3A and 3B). Moreover, there were no significant differences in WRN and PAI-1 expression between treated and untreated fibroblasts with the scrambled RNA molecule (data not shown).

These results confirm that WRN controls PAI-1 overexpression and suggest that when WRN is knocked down, PAI-1 escapes this control and can be over-induced by TNFα, TGFβ, or insulin (Figure 3B). Experimental inducer concentrations (TNFα, TGFβ and insulin) were previously standardized at 150 U/ml, 18 U/ml and 1 mg/ml, respectively. Higher concentrations did not significantly augment PAI-1 expression when they were added to the culture medium in normal fibroblast cultures (data not shown).

Given that WRN polymorphisms may modulate constitutive PAI-1 expression (Figure 1), we examined whether WRN polymorphisms could affect PAI-1 overexpression induced by TNFα, TGFβ, or insulin. Therefore, we used fibroblasts from individuals with the following genotypes: 1367CC/1074FF, 1367CR/1074FL and 1367RR/1074FL (we used the last genotype instead of 1367RR/1074LL, because we had just one culture from an individual with that genotype). Cultures were synchronized and re-fed and they were immediately treated with PAI-1 inducers and 24 hours later, cells were used to determine WRN and PAI-1 expression levels. TNFα, TGFβ, or insulin elicited significant lower expression of PAI-1 in fibroblasts with the 1367RR/1074FL genotype compared with that found in cells with the 1367CC/1074FF genotype (Figure 4). Insulin-dependent overexpression PAI-1 gave also significant differences in fibroblasts with the1367CR/1074FL genotype. We did not find significant differences when comparing other groups.