T-kininogen inhibits kinin-mediated activation of ERK in
endothelial cells

ELIAS LEIVA-SALCEDO, VIVIANA PEREZ, CLAUDIO ACUÑA-CASTILLO,
ROBIN WALTER AND FELIPE SIERRA*

Instituto de Ciencias Biomédicas, Facultad de Medicina, Universidad de Chile
Independencia 1027, Santiago, Chile

ABSTRACT

Serum levels of T-kininogen increase dramatically as rats approach the end of their lifespan. Stable expression of the protein in Balb/c 3T3 fibroblasts leads to a dramatic inhibition of cell proliferation, as well as inhibition of the ERK signaling pathway. T-kininogen is a potent inhibitor of cysteine proteinases, and we have described that the inhibition of ERK activity occurs, at least in part, via stabilization of the MAP kinase phosphatase, MKP-1. Since fibroblasts are not a physiological target of T-kininogen, we have now purified the protein from rat serum, and used it to assess the effect of T-kininogen on endothelial cells. Adding purified T-kininogen to EAhy 926 hybridoma cells resulted in inhibition of basal ERK activity levels, as estimated using appropriate anti-phospho ERK antibodies. Furthermore, exogenously added T-kininogen inhibited the activation of the ERK pathway induced by either bradykinin or T-kinin. We conclude that the age-related increase in hepatic T-kininogen gene expression and serum levels of the protein could have dramatic consequences on endothelial cell physiology, both under steady state conditions, and after activation by cell-specific stimuli. Our results are consistent with T-kininogen being an important modulator of the senescent phenotype in vivo.

Key terms: aging, endothelial cells, ERK pathway, kininogen, kinins

ABBREVIATIONS: ERK: Extracellular Regulated Kinase, HMW-kininogen: High Molecular Weight kininogen, P-ERK: Phosphorylated ERK, T-KG: T-kininogen

INTRODUCTION

Aging is characterized primarily by a decline in functionality and responsiveness to the environment. At the cellular level, this is manifested by a decreased capacity to respond to external cues, as evidenced by a diminished capacity to activate signal transduction cascades, leading to a decreased ability to induce changes in gene expression necessary to adapt to the changes in the external milieu (Jazwinski, 1996). We previously described that hepatic expression of T-kininogen (T-KG) increases dramatically as rats approach the end of their lives (Sierra et al., 1989). Serum levels of the protein increase between 2 and 4-fold approximately 3 months before their time of death, and this is true for both sexes and in several strains of rats (Sierra et al., 1992, and unpublished observations). Furthermore, dietary restriction, the only known non-genetic method of extending lifespan in mammals, delays but does not abolish the increase in serum T-KG levels (Sierra et al., 1992). We have therefore suggested that T-KG is a good and reliable biomarker of aging, at least in rats (Walter et al., 1998). We have also identified an age-related increase in the serum levels of a closely related protein, HMW-kininogen (Sierra, 1995), and others have published results describing an age-related increase in the human ortholog, also called HMW-kininogen (Kleniewski & Czokalo, 1991). At least in rats, the increase in the levels of the two proteins occurs via distinct mechanisms, being transcriptional in the case of T-KG and post-translational in the case of HMW-kininogen (Sierra, 1995). The fact that functionally related proteins increase with age in two different species, and by at least two different mechanisms suggests that the increase in serum kininogens might be an invariant feature of aging in mammals.

Like all kininogens, T-KG is a multifunctional protein. Its best known functions are those as a strong and specific inhibitor of cysteine proteinases (Sueyoshi et al., 1985) and as a precursor to vasoactive peptides called kinins (Bhoola et al., 1992). In an effort to clarify the role of T-KG in the aged animal, we have expressed the protein in Balb/c 3T3 fibroblasts. T-KG expression in these cells leads to a strong inhibition of cell proliferation (Torres et al., 2001). In synchronized cells, proliferative arrest occurs at the end of G1, and before entering the S phase of the cell cycle. Mechanistically, we have established that T-KG leads to a dramatic inhibition of the basal activity of ERK. This in turn seems to be the result of inhibition by T-KG of proteases that control the half-life of phosphatases of the MKP family (Torres et al., 2000). These phosphatases act as specific inactivators (by dephosphorylation) of MAP kinases, preferentially the MAP kinases ERK1 and ERK2. Thus, their stabilization in the presence of T-KG results in inhibition of the ERK pathway.

While these results implicate T-KG as a potential mediator of the decline in cellular responsiveness to the environment, characteristic of aged individuals, there are several important unresolved issues. Foremost among them is the nature of the target cells, and whether or not they would be affected by T-KG in a fashion comparable to what we have observed in fibroblasts. T-KG is produced primarily in the liver, from where it is secreted to the blood circulation (Chao et al., 1988). Therefore, relevant cells include hepatocytes, which produce T-KG and as a consequence are exposed to it from both endogenous and exogenous sources (auto- paracrine model), and blood-borne cells, which are exposed to exogenous T-KG but do not produce the molecule to any significant extent. Among the latter, we have focused our attention on endothelial cells, because they contain the highest number of receptors for both kininogens and their metabolites, kinins (McEachern et al., 1991, Hasan et al., 1995).

To assess the effect of T-KG on endothelial cells, we have purified T-KG from the serum of Brown Norway Katholiek rats, and we have used this material to test the human endothelial hybridoma cell line EAhy 926. Our results indicate that, as we observed in fibroblasts, T-KG can inhibit the basal level of ERK activity in these cells. Furthermore, we demonstrate that T-KG inhibits the induction of ERK in response to kinin treatment. Our results suggest that the age-related increase in T-KG could play a significant role in the decreased responsiveness of endothelial cells to environmental cues.

METHODS

Purification of T-KG.

Brown Norway rats (Katholiek strain) were injected i.p. with 2 mg/Kg of bacterial LPS (Sigma), and blood was collected 24 hr later. Serum prepared from this blood was fractionated on carboxymethylated papain-agarose (Calbiochem). For this, 750 µl of serum were incubated overnight with the resin in a buffer containing 40 mM benzamidine (Sigma), 40 µg/ml polybrene (Sigma), 2 mM EDTA (Sigma), 0.2 mM PMSF (Sigma), 0.2 mg/ml trypsin inhibitor (Sigma) in 2 M NaCl (Merck). The slurry was then packed into a column, and washed with equilibration buffer (50 mM Na phosphate (Merck), pH 6.5, 0.5 M NaCl) until the A₂₈₀ reached zero. The column was washed successively with 10 ml of phosphate buffer at different pH (6.5, 7.5 and 8.5), and elution of bound proteins was performed in 50 mM Na phosphate, pH 11.5 containing 10% glycerol. Both washes and elution were at a rate of 2 ml/min. The eluate (2 ml aliquots) was analyzed by A₂₈₀ and fractions in the peak were tested for cysteine proteinase inhibitory activity (see below).

Measurement of cysteine proteinase inhibitory activity.

Samples were pre-incubated for 10 min at 37 °C in a final volume of 500 µl with 10 nM papain (Sigma) in 0.2 M sodium phosphate buffer, pH 6.8, containing 0.05% (v/v) Triton X-100 (Sigma) , 2 mM EDTA and 5 mM DTT (Sigma). Then, 10 µl of 50 nM Z-Arg-Arg-NHMec (Sigma: 1 nM final) was added and incubation continued for 10 min at 37 °C. The enzymatic reaction was terminated with acetic acid (Merck; 1 mM final), and fluorescence intensities read using 360 nm-excitation and 460 nm-emission filters (Anastasi et al., 1983). Specificity of the assay was estimated using E-64, a stoichiometric and irreversible inhibitor of most papain-like cysteine peptidases (Barrett et al., 1982).

Cell culture and treatments.

EAhy 926 cells are a hybridoma of human umbilical vein endothelial cells (Edgell et al., 1983). Cells were seeded at 3x10⁴ cells/cm², and they were grown for 48 hr in MEM supplemented with HAT, 10 U/ml gentamycin (Gibco BRL) and 10% fetal calf serum (Cellgro, USA) in a CO2 incubator. When subconfluent, cells were washed in PBS and incubated for further 24 hr in the absence of serum before experimentation.

To test the effect of T-KG on basal levels of ERK activity, cells were incubated in the absence of serum with 10 nM T-KG for times ranging from 5 min to 24 hr. For kinin induction, cells were treated with either bradykinin or T-kinin (1-500 nM) for 10 min, and in some experiments, cells were pre-exposed to 100 nM T-KG for 60 min before induction with 100 nM kinin. Samples were collected following 3 washes with ice-cold PBS and lysis in Laemmli buffer.

Western blot analysis

Total cellular protein extracts were boiled for 5 min in Laemmli buffer containing 100 mM ß-mercaptoethanol (final concentration), and samples were loaded onto 10% (w/v) SDS-polyacrylamide gels. Electrophoresis, transfer to nitrocellulose filters and western blotting were by standard methods using ECL (Amersham). Images were captured in BioMax MR-100 film (Kodak, USA), and the data was analyzed using Scion Image software (Scion Corp., USA).

All data is expressed as means + standard error of at least 3 independent measurements. Statistical significance was considered at p<0.05 by the non-parametric test of Mann-Whitney, and the data was processed with the Prism 3.0 software (GraphPad Inc., USA).

RESULTS

Affinity purification of T-KG from rat serum.

Serum T-KG was purified by affinity chromatography on carboxymethylated papain-agarose columns from the serum of acutely inflamed (LPS) Katholiek Brown Norway rats. These rats have a point mutation that prevents the secretion of K-KG, thus eliminating the major source of contamination in such preparations (Damas 1996). Figure 1A indicates a peak in A₂₈₀ at pH 11.5, which corresponds to four fractions containing most of the T-KG, as determined by western blot analysis (data not shown). Albumin, the major protein component of serum, has a MW similar to T-KG, and could therefore be a significant contaminant in the preparations. This was tested by proteolytic digestion with trypsin and chymotrypsin, and the patterns of peptide digestion were compared to both purified bovine albumin and total rat serum (whose main protein component is albumin). The tryptic and chymotrypsin peptides of rat serum and bovine albumin were similar, but different from that obtained with purified T-KG, suggesting that the preparation did not contain significant amounts of rat albumin (data not shown). Finally, the purified T-KG was tested functionally for its ability to inhibit papain, a model cysteine proteinase. Figure 1B shows that purified T-KG inhibited papain at a 1:1 molar ratio, further confirming that the purified fractions were composed primarily of T-KG.

Fig. 1. Purification of T-KG from rat serum. Rat serum was prepared and fractionated on carboxymethyl-papaine agarose as described in Methods. A. Elution profile of proteins (A280, dotted line, right margin) as a function of pH in the wash/elution buffers (pH, solid line, left margin). B. Characterization of purified T-KG as an inhibitor of cysteine proteinases. A titrated amount of enzymatically active papain (10 nM) was pre-incubated with increasing amounts of purified T-KG, and then the mixture was assayed for remaining papain proteolytic activity against a synthetic fluorescent substrate. (FAU= fluorescent arbitrary units).

Purified T-KG inhibits basal levels of ERK activity in EAhy 926 cells.

Cells were seeded and serum deprived as described in the Methods section. They were then treated with 10 nM T-KG for different periods of time as shown in Figure 2, and analyzed for ERK activity by western blot. Data was standardized against total ERK, which did not show any changes throughout the experiment, as determined by comparing the level of total ERK with ß-actin (data not shown). The data indicates that the presence of T-KG reduced basal ERK activity levels in EAhy 926 cells. This effect was most pronounced within 3 hr of treatment (75% inhibition). Subsequently, a minor recovery of ERK activity was observed, but levels remained at 50% or lower even after 24 hr of treatment. We do not know currently whether attenuation of the T-kininogen effect as a function of time was due to inactivation of T-KG in the medium, or a compensatory response of the cell.

Fig. 2. T-KG inhibits basal ERK activity in EAhy 926 endothelial cells. EAhy 926 cells were maintained in the absence of serum for 24 hr, and were then treated with 10 nM purified T-KG for times indicated. At the end of each treatment, cells were lysed directly in SDS loading buffer, and proteins were separated by SDS-PAGE as described in Methods. The membranes were probed in succession for P-ERK, total ERK and ß-actin (in that order), and the resulting radiograms were quantified and analyzed as described in Methods. Total ERK levels did not change relative to ß-actin (not shown), and therefore the figure was prepared by plotting P-ERK relative to total ERK.

Kinins activate ERK in EAhy 926 cells.

While EAhy 926 cells have been described as having characteristics of endothelial cells, it is not clear from the literature whether or not they contain kinin receptors. For this reason, and in order to measure the effect of T-KG on inducible ERK activities, we proceeded to test whether or not EAhy 926 cells respond to exogenous addition of kinins. It has been shown that bradykinin induces proliferation in smooth muscle cells, and that this activity is mediated by ERK (Velarde et al., 1999).

However, while it is generally accepted that both bradykinin and T-kinin work via the same B2 receptors, it is not clear whether a similar mechanism is responsible for activation of ERK in endothelial cells, or whether T-kinin can also activate the ERK pathway. Figure 3 shows that EAhy 926 cells robustly activated ERK within 10 min in response to either bradykinin or T-kinin. The effect was maximal at 100 nM of either kinin, and T-kinin was slightly more effective than bradykinin.

Fig. 3. ERK phosphorylation in response to kinin treatment. EAhy 926 cells were maintained in the absence of serum for 24 hr, and were then treated with different concentrations of either bradykinin (BK) or T-kinin (TK) for 10 min. Cells were collected and relative levels of P-ERK were measured as described in Figure 2. Top panel: A representative western blot showing P-ERK levels in response to different concentrations of either BK (top) or TK (bottom). Bottom panel: Average and standard errors obtained by quantification of 3 independent experiments are shown.

Purified T-KG inhibits kinin-mediated induction of ERK activity.

Next we tested whether the induction of ERK activity by kinins could be inhibited by addition of purified T-KG to the medium of EAhy 926 cells. For this, cells kept in the absence of serum for 6 hr were pre-treated with T-KG (100 nM) for 1 additional hr, as described in Figure 2. Control cells were left untreated during this time. Then cells were induced by addition of 100 nM of either bradykinin or T-kinin for 10 minutes, and ERK activity was measured as described in Figure 3. The results in Figure 4 indicate that kinins were able to activate ERK, but induction was significantly hampered by the presence of purified T-KG. This inhibitory effect was more pronounced for T-kinin than for bradykinin (24 % vs 48 %, respectively).

Fig. 4. Purified T-KG inhibits kinin-dependent induction of ERK activity. The experiment is identical to the one described in Figure 3, except that when appropriate, cells were exposed to 100 nM T-KG for 60 min before induction with kinins. Control: Cells not treated either with T-KG or with kinins. TK and BK: Cells treated with either T-kinin or Bradykinin alone, TKG-TK and TKG-BK: Cells pre-treated with T-KG before induction with either T-kinin or Bradykinin. T-KG: Cells pre-treated with T-KG, but not exposed to kinin. All values are shown as a percentage of the value obtained for maximal induction, obtained by treating cells with TK alone. The asterisks represent statistically significant differences (p<0.05) between samples either treated or not with T-KG.

DISCUSSION

The data presented here shows that both bradykinin and T-kinin were similarly capable of activating the ERK pathway in human endothelial cells. In addition, we show that affinity-purified T-KG inhibited both basal and kinin-induced levels of ERK activation in EAhy 926 cells. Furthermore, the data indicated that this inhibition was independent of the kinin used to stimulate the cells. The experiments were prompted by our previous observation that T-KG expression in Balb/c 3T3 fibroblasts led to inhibition of both cell proliferation and basal ERK activity. We previously postulated that inhibition of the ERK pathway is the mechanism by which cell proliferation is inhibited in that model (Torres et al., 2001). Similar inhibition of cell proliferation was also observed in our laboratory in Jurkat lymphoma cells induced to proliferate with Con A (Acuña-Castillo et al, unpublished observations). However, we have also found that expression of T-KG in NIH 3T3 cells, a different fibroblast cell line, had no effect on either basal levels of signal transduction or proliferation (Nishimura and Sierra, unpublished data). Moreover, recent data in rat aorta endothelial cells indicated that T-KG not only did not inhibit proliferation, but on the contrary, reproducibly induced proliferation approximately 4-fold, even in the absence of serum (Perez et al., unpublished data). Thus, it appears likely that T-KG has pleiotropic effects on cell proliferation and signal transduction, and that these effects are cell-type specific (Perez et al., 2000). Furthermore, in light of all the available data, our previous hypothesis suggesting that inhibition of cell proliferation is linked to inhibition of ERK activity no longer seems tenable. Indeed, while we have yet to measure both parameters simultaneously in different cell lines, we have shown that T-KG inhibited ERK activity (EAhy 926 cells, this report), but promoted cell proliferation in rat aorta endothelial cells (not shown). We are currently in the process of measuring the missing parameters in each of these cell lines (i.e., proliferation in EAhy 926 cells, and ERK activity in rat aorta cells).

Our results (Figure 2 and data not shown) also indicate that inhibition of the basal level of ERK activity required pre-incubation of the cells with T-KG, the maximal effect being observed after 3 hr of pre-incubation. This data suggests that the seffect of T-KG is not mediated by simple steric competition with free kinins for available kinin receptors, and is consistent with our own observations in lymphocytes and macrophages (data not shown). The presence of high affinity receptors for both kinins (primarily B2 receptors, McEachern et al., 1991) and kininogens (see below) on the surface of endothelial cells complicates data interpretation and the delineation of molecular mechanisms by which T-KG leads to inhibition of ERK activity in these cells. Cytokeratins (Shariat-Madar et al.,1999), Mac-1 (Sheng et al., 2000; Jiang et al., 1992), uPAR (Colman et al., 1997), the gC1q receptor (Herwald et al., 1996), thrombospondin-1 (De la Cadena 1994) and glycoproteins GP 1b (Bradford et al., 1997) have all been described as kininogen receptors. However, it should be mentioned that in all these cases, kininogen binding was assayed using HMW-kininogen. Thus, the assumption that T-kininogen binds to any or all of these molecules remains a speculation and is based exclusively on the high degree of homology between these two molecules. Nevertheless, since we have never observed either internalization of exogenous T-KG (by confocal microscopy of tagged molecules), we assume that the biological effects observed are most likely mediated through binding of T-KG to kininogen receptors. In this scenario, T-KG might elicit an intracellular response that takes several hours to develop and leads indirectly to either a decrease in ERK phosphorylation or an increase in ERK de-phosphorylation. Based on our published results on Balb/c 3T3 fibroblasts, we currently favor the second possibility.

Since serum levels of T-KG are significantly increased during aging, our results suggest a role for T-KG in the well known decreased responsiveness of cells from old individuals to external stimuli. At the least, our results predict that old animals will have a decreased responsiveness to kinins, and that this effect will be mediated by T-KG. We are currently testing this hypothesis. On the other hand, recent results with LPS-stimulated macrophages or Con A- stimulated lymphocytes suggest that the effect of T-KG might not be limited to kinins, and T-KG can inhibit ERK activation in response to a variety of stimuli. In that sense, these results point towards a global effect of T-KG on the physiology of aging rats.

ACKNOWLEDGEMENTS

We sincerely thank all members of the Sierra group for encouraging discussions and helpful comments. This work was supported by Fondecyt grants 1981064 and 1010615, as well as FONDAP 15010006.

REFERENCES

ANASTASI A, BROWN N, KEMBHAVI A, NICKLIN M, SAYERS C, SUNTER D, BARRETT A (1983) Cystatin, a protein inhibitor of cysteine proteinases. Biochem J 211: 129-138

BARRETT AJ, KEMBHAVI AA, BROWN MA, KIRSHKE H, KNIGHT CG, TAMAI M, HANADA K (1982) L-trans-Epoxysuccinyl-leucylamido(4-guanidino)butane (E-64) and its analogues as inhibitors of cysteine proteinases including cathepsins B, H and L. Biochem J 201, 189-198

BHOOLA KD, FIGUEROA CD, WORTHY K (1992) Bioregulation of kinins: kallikreins, kininogens and kininases. Pharmac Rev 44:1-80

BRADFORD H, DE LA CADENA R, KUNAPULI S, DONG J, LOPEZ J, COLMAN R (1997) Human Kininogen regulate thrombin binding to platelets through the Glycoprotein Ib-IX-V complex. Blood 90: 1508-1515

COLMAN R, PIXLEY R, NAJAMUNNISA Y, WANG J, MAZAR A, MCCRAE H (1997) Binding of high molecular weight Kininogen to endothelial cells is mediated via a site within domains 2 and 3 of the urokinase receptor. J Clin Invest 100: 1481-1487

CHAO J, SWAIN C, CHAO S, XIONG W, CHAO L (1988) Tissue distribution and kininogen gene expression after acute-phase inflammation. Bioch Biophys Acta 964:329-339

DAMAS J (1996) The Brown Norway rat and the kinin system. Peptides 17: 859-872

DE LA CADENA R, WYSHOCK E, KUNAPULI S, SCHULTZE R, MULLER M, WALTZ D, COLMAN R (1994) Platelet thrombospondin interaction with human high and low molecular weight kininogens. Thromb Haemost 72: 125-131

EDGELL CJS, McDONALD CC, GRAHAM JB (1983) Permanent Cell Line Expressing Human Factor VIII-Related Antigen Established by Hybridization. Proc Natl Acad Sci USA 80: 3734-3737

HASAN AAK, CINES DB, NGAIZA JR, JAFFE EA, SCHMAIER AH (1995) High-molecular-weight-kininogen is exclusively membrane bound on endothelial cells to influence activation of vascular endothelium. Blood 85: 3134-3143

HERWALD H, DEDIO J, KELLNER R, LOSS M, MULLER-ESTERL W (1996) Isolation and characterization of the Kininogen-binding protein p33 from endothelial cells. J Biol Chem 271: 13040-13047

JAZWINSKI SM (1996) Longevity, Genes, and Aging. Science 273: 54-58

JIANG Y, MULLER-ESTERL W, SCHMAIER A (1992) Domain 3 of Kininogen contains a cell-binding site and a site that modifies thrombin activation of platelets. J Biol Chem 267: 3712-3717

KLENIEWSKI J CZOKALO M (1991) Plasma kininogen concentration: the low level in cord blood plasma and age dependence in adults. Eur J Haematol 46:257-262

MCEACHERN AE, SHELTON ER, BHAKTA S, OBERNOLTE R, BACH C, ZUPPAN P, FUJISAKI J, ALDRICH RW, JARNAGIN K (1991) Expression cloning of a rat B2 bradykinin receptor. Proc Natl Acad Sci USA 88:7724-7728

PEREZ V, ACUNA-CASTILLO C, SIERRA F (2000) Effect of exogenous T-KG on proliferation and ERK activity in several cell lines. Biol Res 33:R70

SHARIAT-MADAR Z, MAHDI F, SCHMAIER A (1999) Mapping binding domains of Kininogen on endothelial cell cytokeratin 1. J Biol Chem 274: 7137-7145

SHENG N, FAIRBANKS M, HEINRIKSON R, CANZIANI G, CHAIKEN I, MOSSER D, ZHANG H, COLMAN R (2000) Cleaved high molecular weight kininogen binds directly to integrin CD11b/CD18 (Mac-1) and blocks adhesion to fibrinogen and ICAM-1. Blood 95: 3788-3795

SIERRA F, FEY GH, GUIGOZ Y (1989) T-kininogen gene expression is induced during aging. Mol Cell Biol 9:5610-5616

SIERRA F, COEYTAUX S, JUILLERAT M, RUFFIEUX C, GAULDIE J, GUIGOZ Y (1992) Serum T- kininogen levels increase two to four months before death. J Biol Chem 267:10665-10669

SIERRA F (1995) Both T- and K-kininogens increase in the serum of old rats, but by different mechanisms. Mech Ageing Dev 84:127-137

SUEYOSHI T, ENJYOII K, SHIMADA T, KATO H, IWANAGA S, BANDO Y, KOMINAMI E, KATUNUMA N (1985) A new function of kininogens as thiol-proteinase inhibitors: inhibition of papain and cathepsins B, H and L by bovine, rat and human plasma kininogens. FEBS Lett 182:193-195

TORRES C, LI M, WALTER R, SIERRA F (2000) Modulation of the ERK pathway of signal transduction by cysteine proteinase inhibitors. J Cell Biochem 80:11-23

TORRES C, LI M, WALTER R, SIERRA F (2001) T-kininogen inhibits fibroblast proliferation in the G1 phase of the cell cycle. Exp Cell Res 269: 171-179

VELARDE V, ULLIAN M, MAYFIELD R, JAFFA A (1999) Mechanism of MAPK activation by bradykinin in vascular smooth muscle cells. Am J Physiol 277: C 253-261

VELARDE V, ULLIAN M, MAYFIELD R, JAFFA A (1999) Mechanism of MAPK activation by bradykinin in vascular smooth muscle cells. Am J Physiol 277: C 253-261

WALTER R, MURASKO DM, SIERRA F (1998). T-kininogen is a biomarker of senescence in rats. Mech Ageing Dev 106:129-144

To whom correspondence should be addressed: Dr. Felipe Sierra, Instituto de Ciencias Biomédicas. Facultad de Medicina. Universidad de Chile. Independencia 1027, Santiago, Chile. Phone: (56 2) 678 6059, E-mail: Fsierra@machi.med.uchile.cl

Received: June 27, 2002. In revised form: July 11, 2002. Accepted: July 14, 2002