'Correction:' Serum transforming growth factor beta-1 (TGF-beta-1) levels in diabetic patients are not associated with pre-existent coronary artery disease
- Beatriz D Schaan†1, 4Email author,
- Alexandre S Quadros†1,
- Rogério Sarmento-Leite†1,
- Giuseppe De LuccaJr†1,
- Alexandra Bender†1 and
- Marcello Bertoluci†2, 3
© Schaan et al; licensee BioMed Central Ltd. 2007
Received: 24 May 2007
Accepted: 25 July 2007
Published: 25 July 2007
The association between TGF-β1 levels and long-term major adverse cardiovascular events (MACE) in patients with coronary artery disease (CAD) is controversial. No study specifically addressed patients with CAD and diabetes mellitus (DM). The association between TGF-β1 levels and long-term major adverse cardiovascular events (MACE) in patients with coronary artery disease (CAD) is controversial. No study specifically addressed patients with CAD and diabetes mellitus (DM).
Patients (n = 135, 30–80 years) referred for coronary angiography were submitted to clinical and laboratory evaluation, and the coronary angiograms were evaluated by two operators blinded to clinical characteristics. CAD was defined as the presence of a 70% stenosis in one major coronary artery, and DM was characterized as a fasting glycemia > 126 mg/dl or known diabetics (personal history of diabetes or previous use of anti-hyperglycemic drugs or insulin). Based on these criteria, study patients were classified into four groups: no DM and no CAD (controls, C n = 61), DM without CAD (D n = 23), CAD without DM (C-CAD n = 28), and CAD with DM (D-CAD n = 23). Baseline differences between the 4 groups were evaluated by the χ2 test for trend (categorical variables) and by ANOVA (continuous variables, post-hoc Tukey). Patients were then followed-up during two years for the occurrence of MACE (cardiac death, stroke, myocardial infarction or myocardial revascularization). The association of candidate variables with the occurrence of 2-year MACE was assessed by univariate analysis.
The mean age was 58.2 ± 0.9 years, and 51% were men. Patients with CAD had a higher mean age (p = 0.011) and a higher percentage were male (p = 0.040). There were no significant baseline differences between the 4 groups regarding hypertension, smoking status, blood pressure levels, lipid levels or inflammatory markers. TGF-β1 was similar between patients with or without CAD or DM (35.1 ×/÷ 1.3, 33.6 ×/÷ 1.6, 33.9 ×/÷ 1.4 and 31.8 ×/÷ 1.4 ng/ml in C, D, C-CAD and D-CAD, respectively, p = 0.547). In the 2-year follow-ip, independent predictors of 2-year MACE were age (p = 0.007), C-reactive protein (p = 0.048) and systolic blood pressure (p = 0.008), but not TGF-β1.
Serum TGF-β1 was not associated with CAD or MACE occurrence in patients with or without DM.
Atherosclerosis is related to lipid abnormalities, platelet activation, thrombosis, endothelial dysfunction, inflammation, oxidative stress and altered matrix metabolism , among other disturbances. Several risk factors are known to play a role in this intricate cascade, and diabetes mellitus (DM) is associated with a 3-fold increase in the risk of death or myocardial infarction due to coronary artery disease (CAD) . There is considerable debate as to whether the pathogenesis of atherosclerosis in diabetics is different, or whether DM simply accelerates this process. The identification of specific pathophysiologic features in diabetics could lead to a targeted approach to control disease progression in this set of patients.
A number of inflammatory cytokines mediate atherosclerosis, and TGF-β1 modulates important events like macrophage and fibroblast chemotaxis, suppression of lymphocyte function, collagen synthesis and stimulation of extracellular matrix synthesis [3–7]. Nikol et al demonstrated increased expression of TGF-β1 in human atherosclerotic plaques, and it was subsequently demonstrated that diabetic patients presenting with an acute myocardial infarction revealed decreased smooth muscle cells (SMCs), increased macrophages and TGF-β1 into the culprit lesion [8, 9].
Other studies showed that reduced plasma levels of TGF-β1 in patients with angiographically proven CAD [10–12], and TGF-β1 levels were also found to be inversely related to the extend of coronary disease . Tashiro et al observed that lower TGF-β1 levels were associated with increased risk of long-term major adverse cardiovascular events (MACE), suggesting that plasma levels of this citokyne may have prognostic significance in patients with CAD . However, other studies failed to demonstrate an association between TGF-β1 and atherosclerotic disease , and no study was specifically addressed to diabetic patients. The aim of the present study is to evaluate the association between serum TGF-β1 and CAD in patients with DM, and its influence in long-term MACE incidence.
Patients included were referred for coronary angiography by their attending physicians between December 2002 and December 2003. Inclusion criteria were age between 30 and 80 years, chest pain and documented myocardial ischemia in the non-invasive evaluation. The exclusion criteria were acute myocardial infarction in the last 60 days, previous heart transplantation or revascularization procedure, use of anticoagulants, known malignant neoplasia and haemodialysis. Each subject gave written informed consent and the study protocol was approved by the Hospital Ethics Committee.
On the day of the coronary angiography, a clinical questionnaire was applied and a physical examination was performed. A 12 h-fasting blood sample collection and a morning spot urine analysis were done before coronary angiography. Blood samples were collected in tubes without conservants, kept at room temperature for 3 hours and then stored at 4°C overnight. Serum aliquots were obtained after centrifugation (1000 G for 40 minutes at 4°C) and stored at -80°C until assay.
Based on the clinical history and on coronary artery disease severity, subjects were divided into four groups: no diabetes and no coronary artery disease (controls, C), diabetes and no coronary artery disease (D), coronary artery disease and no diabetes (C-CAD) and coronary artery disease and diabetes (D-CAD). Diabetes mellitus was defined as serum fasting glucose > 126 mg/dl or known diabetics, as evaluated by a personal history of diabetes or previous use of anti-hyperglycemic drugs or insulin.
Patients were followed-up either by clinical evaluation in an outpatient clinic, by interview with the attending physician or by telephone contact. The incidence of major adverse cardiovascular events (MACE) in the follow-up period was recorded. MACE was defined as the occurrence of cardiac death, ischemic stroke, myocardial infarction (AMI) or the need for a revascularization procedure (either percutaneous or surgical). Myocardial infarction was defined according to the criteria of the American College of Cardiology .
Measurements of TGF-β1 levels
For the measurement of serum active TGF-β1 levels we used a solid phase ELISA designed to measure biologically active TGF-β1 (R&D Systems, Abingdon, UK). On the day of the assay, serum samples (0.5 ml) were acidified to a pH of 2–3 with 100 μl 1N HCL for 10 minutes to activate latent TGF-β, and then re-neutralized to pH 7–8 with 100 μl 1.2N NaOH/0.5 M HEPES. Results were expressed in ng/ml. The mean intra and inter assay coefficients of variation were respectively 2.0% and 13.1%.
Fasting blood samples were obtained for glucose, total cholesterol, high-density cholesterol, triglycerides, creatinine, and laboratory measurements were performed using automated enzymatic commercial kits (Roche, Manhein, GE). Fibrinogen was evaluated on a Fibrintimer II (Dade Behring Inc., Newark, Marburg, Germany) and processed in the auto-analyser (CA-540, Sysmex). Urinary albumin was evaluated in morning spot urine samples by immunoturbidimetry (antiserum to human albumin, Dade Behring Inc., Marburg, Germany) and the absolute concentration was considered. Levels higher than 17 mg/l were considered microalbuminuria. The concentration of low-density cholesterol (LDL-c) was calculated with Friedewald's formula. Serum insulin was determined by enzyme immunoassay commercial kits (Abbot-Murex, Park, IL, USA,), HbA1c by immunoturbidimetry (Roche, Manhein, GE) and ultrasensitive C-reactive protein (CRP) by nephelometry (nephelometer BN100, Dade Behring Inc., Marburg, Germany). Insulin resistance was assessed by the HOMA-IR (homeostasis model assessment of insulin resistance) in individuals who were not on anti-diabetic agents or insulin. The HOMA-IR is the product of the fasting plasma glucose (mmol/l) by plasma insulin (μ/ml), divided by a constant (22.5) .
All coronary angiographies were performed at the Catheterization Laboratory of our institution according to standard routines. The coronary angiograms were evaluated by two experienced interventional cardiologists, unaware of the patients' glucose tolerance status. Angiographic analyses were performed in two orthogonal views with a validated digital quantitative coronary angiography system (Siemens Axiom Artis equipment; Munich, Germany). Significant coronary artery disease was considered to be present if an internal luminal narrowing greater than 70% was detected in at least one major coronary artery.
Categorical variables were expressed as percentiles; continuous variables were expressed as the mean ± standard error of mean (SEM), unless otherwise specified. Differences between the four groups studied were evaluated by the χ2 test for trend (categorical variables) and by ANOVA (continuous variables, post-hoc Tukey). TGF-β1 data were log-transformed before analysis. For all tests, a P value ≤ 0.05 was considered to be statistically significant. The association of candidate variables with the occurrence of 2-year MACE was assessed by univariate analysis, and those with a p ≥ 0.20 were included in a multiple regression logistic model.
Clinical characteristics of the studied patients
Mean age (years)
55.3 ± 1.4
58.2 ± 1.6
61.8 ± 2.0#
61.7 ± 2.0#
28.07 ± 0.7
31.04 ± 1.0*
26.44 ± 0.8
27.27 ± 0.9
Hypertension history (%)
Familial CAD history (%)
Statin users (%)
AAS users (%)
ACEI users (%)
Physically active (%)
141.8 ± 4.7
134.4 ± 3.4
145.8 ± 7.3
140.2 ± 2.9
80.7 ± 2.5
84.3 ± 2.3
86.1 ± 4.5
87.0 ± 1.7
Laboratory characteristics of the studied groups
Plasma glucose (mg/dl)
93.1 ± 1.4
143.6 ± 14.5*
96.2 ± 2.2
141.8 ± 11.7*
5.71 ± 0.05
6.67 ± 0.35*
5.72 ± 0.06
7.08 ± 0.33*
194.4 ± 5.5
201.7 ± 10.2
213.5 ± 7.9
219.6 ± 10.3
111.9 ± 4.3
121.3 ± 8.3
126.2 ± 6.3
129.2 ± 8.2
49.9 ± 1.5
46.6 ± 1.9
44.5 ± 2.2
45.4 ± 1.9
Non HDL-c (mg/dl)
142.7 ± 5.6
155.1 ± 9.3
168.4 ± 7.8#
174.2 ± 9.6#
160.9 ± 13.7
169.1 ± 15.8
216.4 ± 31.4
224.9 ± 30.7
348.2 ± 11.4
335.7 ± 19.5
355.9 ± 17.4
353.2 ± 21.8
7.63 ± 1.27
4.48 ± 1.13
7.21 ± 2.61
8.57 ± 2.00
Insulin levels (μU/ml)
12.92 ± 1.3
21.25 ± 6.9
12.1 ± 1.4
17.6 ± 2.6
3.09 ± 0.34
9.06 ± 3.63*
2.99 ± 0.39
6.42 ± 1.16
TGF B (ng/ml)
35,1 ×/÷ 1,3
33,6 ×/÷ 1,6
33,9 ×/÷ 1,4
31,8 ×/÷ 1,4
Clinical, angiographic and laboratory characteristics of the patient population (n = 123) according to the occurrence of MACE in the 2-year follow-up
57.48 ± 1.5
60.77 ± 1.1
Male sex (%)
33.82 ± 1.5
37.08 ± 1.5
6.31 ± 0.2
5.91 ± 0.09
9.14 ± 1.8
5.51 ± 0.8
144.1 ± 3.4
136.7 ± 2.0
Multivariate analysis of candidate variables associated with 2-year MACE
CAD (presence or absence)
The present study did not demonstrate a significant association between serum TGF-β1 levels and CAD in diabetic patients, or in controls. The independent predictors of 2-year MACE were increased age, C-reactive protein and systolic blood pressure, but not serum TGF-β1.
TGF-β1 biological activities include inhibition of cell proliferation, regulation of cell differentiation, stimulation of cell adhesion, extracellular matrix deposition [18, 19] and inhibition of the growth and proliferation of vascular smooth cells , findings consistent with an anti-atherogenic role for TGF-β1. It also has a role in the regulation of the inflammatory process in the atherosclerotic plaque, being secreted by several cell types, including regulatory Th2 cells. It acts by antagonizing Th1 cell IF-gama production, an inflammatory response inductor . Indeed, TGF-β1 plays a much more important role in the down-regulation of Th1 cell inflammatory response than in the atherosclerotic lesion size. Thus, large plaques do not necessarily mean a more advanced inflammatory process. According to that particular rationale TGF-β1 is a much more important inflammatory marker than an atherosclerotic marker, which agrees with the present data.
Several clinical studies have already demonstrated that patients with advanced atherosclerosis present lower plasma concentrations of TGF-β1 than those without CAD [11–14, 20]. Conversely, Border and Ruoslahti showed that TGF-β1 could enhance atherogenesis by mediating excessive extracellular matrix accumulation  and by down-regulating thrombomodulin, promoting thrombogenesis at the sites of vessel wall injury, where it is released from platelets, smooth muscle cells and monocytes . This is in accordance with clinical studies which showed high levels of active TGF-β1 levels in CAD . However, none of the data mentioned above specifically addressed subjects with diabetes, which was the subject of the pesent work.
Although not classifying patients as diabetics or not, Wang et al demonstrated a positive correlation between active and total TGF-β1 levels and plasma glucose. In vitro studies showed that hyperglycemia induces TGF-β1 production in endothelial cells  and in mesangial cells of the kidney . We suggest that TGF-β1 levels of our diabetic patients are raised in consequence of hyperglycemia, since their plasma glucose levels and HbA1c were higher than those of non-diabetics. Possibly in the diabetic population the hyperglycemia counteracts the atherosclerosis TGF-β1 reducing effect, masking the discriminatory role of this cytokine as found by other studies in non-diabetic subjects. Moreover, as aspirin medication correlates with an increase in active TGF-β1 concentration , the high percentage of subjects using the drug (71%) also could have masked its discriminatory role.
We could not stratify patients according to CAD severity, because of the small number of subjects in each studied group. It is possible that the relevance of TGF-β1 in discriminating CAD could only be observed in severe disease, as shown by other authors in CAD [13, 25, 26] and peripheral artery disease . Using clinical criteria for the diagnosis of significant atherosclerotic vascular disease in hemodialysis patients, Stefoni et al showed reduced survival in a 24 month follow-up of patients with lower TGF-β1 levels at the beginning of the study . Tashiro et al provided the unique evidence in literature that plasma concentrations of TGF-β1 have a prognostic significance in patients with angiographically-proven CAD, in a limited number of patients followed up for a mean period of 979 days . This study did not show plasma TGF-β1 measurements in the patients' follow-up, but these values are supposed to be stable, according to the authors. Although the same drawbacks could be applied to the present study, we showed a different result, and therefore we question the significance of those findings.
A potential limitation of the present study could be the stability of TGF-β1 in peripheral blood. In a recent work, we showed that plasma TGF-β1 is quite stable in patients with diabetes and nephropathy, no changes being detected by strict 12-week blood pressure control despite a significant fall in urinary TGF-β1 levels . Although it is possible that minor changes in TGF-β1 production from endothelial cells could not be detectable in serum where platelet TGF-β1 is abundant, greater decreases, as expected, are unlikely. Other possible limitation of the present study was the number of subjects studied, that did not permit to stratify the sample by age, searching for possible age-related physiological decrease in TGF-β1 levels described by other authors .
In conclusion, serum TGF-β1 was not associated with CAD or MACE occurrence in diabetic patients. The role of TGF-β1 in atherosclerosis associated with diabetes needs further evaluation by means of longer prospective studies with a large number of patients.
- Libby P: Inflammation in atherosclerosis. Nature. 2002, 420 (6917): 868-874. 10.1038/nature01323.View ArticlePubMedGoogle Scholar
- Stamler J, Vaccaro O, Neaton JD, Wentworth D: Diabetes, other risk factors, and 12-yr cardiovascular mortality for men screened in the Multiple Risk Factor Intervention Trial. Diabetes Care. 1993, UNITED STATES , 16 (2): 434-444. 10.2337/diacare.16.2.434.View ArticlePubMedGoogle Scholar
- Maemura K, Kurihara H, Morita T, Oh-hashi Y, Yazaki Y: Production of endothelin-1 in vascular endothelial cells is regulated by factors associated with vascular injury. Gerontology. 1992, 38 Suppl 1: 29-35.View ArticlePubMedGoogle Scholar
- Sarzani R, Brecher P, Chobanian AV: Growth factor expression in aorta of normotensive and hypertensive rats. J Clin Invest. 1989, 83 (4): 1404-1408.PubMed CentralView ArticlePubMedGoogle Scholar
- Zuckerman SH, Panousis C, Evans G: TGF-beta reduced binding of high-density lipoproteins in murine macrophages and macrophage-derived foam cells. Atherosclerosis. 2001, 155 (1): 79-85. 10.1016/S0021-9150(00)00540-2.View ArticlePubMedGoogle Scholar
- Robertson AK, Rudling M, Zhou X, Gorelik L, Flavell RA, Hansson GK: Disruption of TGF-beta signaling in T cells accelerates atherosclerosis. J Clin Invest. 2003, 112 (9): 1342-1350. 10.1172/JCI200318607.PubMed CentralView ArticlePubMedGoogle Scholar
- Mallat Z, Gojova A, Marchiol-Fournigault C, Esposito B, Kamate C, Merval R, Fradelizi D, Tedgui A: Inhibition of transforming growth factor-beta signaling accelerates atherosclerosis and induces an unstable plaque phenotype in mice. Circ Res. 2001, 89 (10): 930-934.View ArticlePubMedGoogle Scholar
- Fukumoto H, Naito Z, Asano G, Aramaki T: Immunohistochemical and morphometric evaluations of coronary atherosclerotic plaques associated with myocardial infarction and diabetes mellitus. J Atheroscler Thromb. 1998, 5 (1): 29-35.View ArticlePubMedGoogle Scholar
- Nikol S, Isner JM, Pickering JG, Kearney M, Leclerc G, Weir L: Expression of transforming growth factor-beta 1 is increased in human vascular restenosis lesions. J Clin Invest. 1992, 90 (4): 1582-1592.PubMed CentralView ArticlePubMedGoogle Scholar
- Crook R: Plasma levels of active transforming growth factor-beta are reduced in patients with three vessel coronary artery disease. Heart. 1996, 75 (5S): 79p.Google Scholar
- Grainger DJ, Kemp PR, Metcalfe JC, Liu AC, Lawn RM, Williams NR, Grace AA, Schofield PM, Chauhan A: The serum concentration of active transforming growth factor-beta is severely depressed in advanced atherosclerosis. Nat Med. 1995, 1 (1): 74-79. 10.1038/nm0195-74.View ArticlePubMedGoogle Scholar
- Tashiro H, Shimokawa H, Yamamoto K, Momohara M, Tada H, Takeshita A: Altered plasma levels of cytokines in patients with ischemic heart disease. Coron Artery Dis. 1997, 8 (3-4): 143-147.View ArticlePubMedGoogle Scholar
- Batuman O, Go D, Clark LT, Smith EL, Clements P, Feit A, Lederer D: Relationship between cytokine levels and coronary artery disease in women. Heart Dis. 2001, 3 (2): 80-84. 10.1097/00132580-200103000-00004.View ArticlePubMedGoogle Scholar
- Tashiro H, Shimokawa H, Sadamatu K, Yamamoto K: Prognostic significance of plasma concentrations of transforming growth factor-beta in patients with coronary artery disease. Coron Artery Dis. 2002, 13 (3): 139-143. 10.1097/00019501-200205000-00001.View ArticlePubMedGoogle Scholar
- Wang XL, Liu SX, Wilcken DE: Circulating transforming growth factor beta 1 and coronary artery disease. Cardiovasc Res. 1997, 34 (2): 404-410. 10.1016/S0008-6363(97)00033-3.View ArticlePubMedGoogle Scholar
- Smith SC, Dove JT, Jacobs AK, Kennedy JW, Kereiakes D, Kern MJ, Kuntz RE, Popma JJ, Schaff HV, Williams DO, Gibbons RJ, Alpert JP, Eagle KA, Faxon DP, Fuster V, Gardner TJ, Gregoratos G, Russell RO: ACC/AHA guidelines for percutaneous coronary intervention (revision of the 1993 PTCA guidelines)-executive summary: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (Committee to revise the 1993 guidelines for percutaneous transluminal coronary angioplasty) endorsed by the Society for Cardiac Angiography and Interventions. Circulation. 2001, 103 (24): 3019-3041.View ArticlePubMedGoogle Scholar
- 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 (7): 412-419. 10.1007/BF00280883.View ArticlePubMedGoogle Scholar
- Merwin JR, Anderson JM, Kocher O, Van Itallie CM, Madri JA: Transforming growth factor beta 1 modulates extracellular matrix organization and cell-cell junctional complex formation during in vitro angiogenesis. J Cell Physiol. 1990, 142 (1): 117-128. 10.1002/jcp.1041420115.View ArticlePubMedGoogle Scholar
- Morisaki N, Kawano M, Koyama N, Koshikawa T, Umemiya K, Saito Y, Yoshida S: Effects of transforming growth factor-beta 1 on growth of aortic smooth muscle cells. Influences of interaction with growth factors, cell state, cell phenotype, and cell cycle. Atherosclerosis. 1991, 88 (2-3): 227-234. 10.1016/0021-9150(91)90085-H.View ArticlePubMedGoogle Scholar
- Li CG, Bethell H, Wilson PB, Bhatnagar D, Walker MG, Kumar S: The significance of CD105, TGFbeta and CD105/TGFbeta complexes in coronary artery disease. Atherosclerosis. 2000, 152 (1): 249-256. 10.1016/S0021-9150(99)00476-1.View ArticlePubMedGoogle Scholar
- Border WA, Ruoslahti E: Transforming growth factor-beta in disease: the dark side of tissue repair. J Clin Invest. 1992, 90 (1): 1-7.PubMed CentralView ArticlePubMedGoogle Scholar
- Ohji T, Urano H, Shirahata A, Yamagishi M, Higashi K, Gotoh S, Karasaki Y: Transforming growth factor beta 1 and beta 2 induce down-modulation of thrombomodulin in human umbilical vein endothelial cells. Thromb Haemost. 1995, 73 (5): 812-818.PubMedGoogle Scholar
- Cagliero E, Roth T, Taylor AW, Lorenzi M: The effects of high glucose on human endothelial cell growth and gene expression are not mediated by transforming growth factor-beta. Lab Invest. 1995, 73 (5): 667-673.PubMedGoogle Scholar
- Ziyadeh FN, Sharma K, Ericksen M, Wolf G: Stimulation of collagen gene expression and protein synthesis in murine mesangial cells by high glucose is mediated by autocrine activation of transforming growth factor-beta. J Clin Invest. 1994, 93 (2): 536-542.PubMed CentralView ArticlePubMedGoogle Scholar
- Stefoni S, Cianciolo G, Donati G, Dormi A, Silvestri MG, Coli L, De Pascalis A, Iannelli S: Low TGF-beta1 serum levels are a risk factor for atherosclerosis disease in ESRD patients. Kidney Int. 2002, 61 (1): 324-335. 10.1046/j.1523-1755.2002.00119.x.View ArticlePubMedGoogle Scholar
- Os I, Djurovic S, Seljeflot I, Berg K: Transforming growth factor (TGF)-beta1 inversely related to vascular cell adhesion molecule-1 in postmenopausal women with coronary artery disease. A possible mechanism for the putative cardioprotective role of TGF-beta1?. J Intern Med. 2002, 251 (3): 223-227. 10.1046/j.1365-2796.2002.00950.x.View ArticlePubMedGoogle Scholar
- Erren M, Reinecke H, Junker R, Fobker M, Schulte H, Schurek JO, Kropf J, Kerber S, Breithardt G, Assmann G, Cullen P: Systemic inflammatory parameters in patients with atherosclerosis of the coronary and peripheral arteries. Arterioscler Thromb Vasc Biol. 1999, 19 (10): 2355-2363.View ArticlePubMedGoogle Scholar
- Bertoluci MC, Uebel D, Schmidt A, Thomazelli FC, Oliveira FR, Schmid H: Urinary TGF-beta1 reduction related to a decrease of systolic blood pressure in patients with type 2 diabetes and clinical diabetic nephropathy. Diabetes Res Clin Pract. 2006, 72 (3): 258-264. 10.1016/j.diabres.2005.10.027.View ArticlePubMedGoogle Scholar
- Okamoto Y, Gotoh Y, Uemura O, Tanaka S, Ando T, Nishida M: Age-dependent decrease in serum transforming growth factor (TGF)-beta 1 in healthy Japanese individuals; population study of serum TGF-beta 1 level in Japanese. Dis Markers. 2005, 21 (2): 71-74.PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.