Aortic valve stenosis progresses due to cuspal calcification,

Transcription

1 Progression of Aortic Valve Stenosis: TGF- 1 is Present in Calcified Aortic Valve Cusps and Promotes Aortic Valve Interstitial Cell Calcification Via Apoptosis Bo Jian, MD, PhD, Navneet Narula, MD, Quan-yi Li, PhD, Emile R. Mohler III, MD, and Robert J. Levy, MD Cardiology Research Laboratory, Children s Hospital of Philadelphia, and Department of Medicine, University of Pennsylvania Health System, Philadelphia, Pennsylvania Background. Aortic valve stenosis characteristically progresses due to cuspal calcification, often necessitating valve replacement surgery. The present study investigated the hypothesis that TGF- 1, a cytokine that causes calcification of vascular smooth muscle cells in culture, initiates apoptosis of valvular interstitial cells as a mechanistic event in cuspal calcification. Methods. Noncalcified and calcified human aortic valve cusps were obtained at autopsy or at the time of cardiac surgery. The distributions within cusps of TGF- 1, latent-tgf- 1-associated peptide, and TGF- receptors were studied using immunohistochemistry. The effects of TGF- 1 on mechanistic events contributing to aortic valve calcification were also investigated using sheep aortic valve interstitial cell (SAVIC) cultures. Results. Immunohistochemistry studies revealed that calcific aortic stenosis cusps characteristically contained within the extracellular matrix qualitatively higher levels of TGF- 1 than noncalcified cusps. Noncalcified normal valves demonstrated only focal intracellular TGF- 1. Addition of TGF- 1 to SAVIC cultures led to a cascade of events, including: cellular migration, aggregation, formation of apoptotic-alkaline phosphatase enriched nodules, and calcification of these nodules. The time course of these events in the SAVIC culture system was rapid with nodule formation with apoptosis by 72 hours, and calcification after 7 days. Furthermore, ZVAD-FMK, an antiapoptosis agent (caspase inhibitor), significantly inhibited calcification and apoptosis induced by TGF- 1, but had no effect on nodule formation. However, cytochalasin D, an actin-depolymerizing agent, inhibited nodule formation, but not calcification. Conclusions. TGF- 1 is characteristically present within calcific aortic stenosis cusps, and mediates the calcification of aortic valve interstitial cells in culture through mechanisms involving apoptosis. (Ann Thorac Surg 2003;75:457 66) 2003 by The Society of Thoracic Surgeons Aortic valve stenosis progresses due to cuspal calcification, often requiring valve replacement surgery. The mechanisms responsible for cuspal mineralization are incompletely understood. Previous investigations have identified a number of unique features that characterize calcific aortic valvular disease including the apparent early involvement of apoptotic vesicles [1], the presence of extracellular matrix proteins typically found in bone [2 4], and a change in cellular phenotype involving aortic valve interstitial cell expression of smooth muscle actin in calcified valve cusps, but not normals [5]. In addition, scattered inflammatory infiltrates have also been reported to be often present within the spongiosa of calcified aortic valve cusps, but not in normal cusps [5]. TGF- 1 is a cytokine that has been demonstrated to promote the calcification of aortic smooth muscle cells in culture [6], as well as canine and human aortic valve Accepted for publication Aug 19, Address reprint requests to Dr Levy, Children s Hospital of Philadelphia, Abramson Research Building, 3416 Civic Center Boulevard, Philadelphia, PA ; levyr@ .chop.edu. interstitial cells [7]. However, the presence of TGF- 1 in calcified human aortic valve cusps has not been previously described. TGF- 1 is a member of the same gene super family as the bone morphogenic proteins [8], and has also been demonstrated to have osteogenic activity [8, 9], as well as pro-apoptotic activity [10 12]. The present studies investigated the hypothesis that TGF- 1 initiates apoptosis of valvular interstitial cells as a mechanistic event in cuspal calcification. Thus, in this study, we investigated a series of calcified human aortic valves for the immunohistochemical presence of TGF- 1 and TGF- receptors, compared with noncalcified normal valves. Furthermore, using a cell culture calcification model, we investigated the mechanisms responsible for TGF- 1 induced aortic valve interstitial cell calcification. Material and Methods Human Pathology Specimens Human aortic valves were obtained at the time of cardiac surgery or at autopsy, under exemptions granted by the 2003 by The Society of Thoracic Surgeons /03/$30.00 Published by Elsevier Science Inc PII S (02)

2 458 JIAN ET AL Ann Thorac Surg TGF- 1 IN AORTIC STENOSIS 2003;75: Institutional Review Boards of the Hospital of the University of Pennsylvania and the Children s Hospital of Philadelphia. Calcified human aortic valve cusps were obtained from five cases of calcific aortic stenosis following valve removal and prosthetic valve replacement; the age range was 70 to 91, with 4 males and 1 female. Normal, noncalcified human aortic valves were obtained at autopsy from human subjects ranging in age from 40 to 77, including 3 males and 3 females. All valve specimens were fixed in neutral buffered formalin. Calcified valves were decalcified before embedding. Paraffin thin sections were prepared for studies confirming the presence or absence of calcification with Alizarin Red S [10], and for routine microscopy using hematoxylin and eosin. Immunohistochemistry Immunohistochemical studies were performed as previously described [4]. Sections of immunostained specimens were evaluated by three independent observers using a semiquantitative rating system: Negative ( ); Focal, weak staining ( ); Present, strong staining ( ); and very strong, widespread staining ( ). Primary antibodies (rabbit polyclonal IgG), anti-tgf- 1, anti- TGF- RI, anti-tgf- RII, were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Antihuman (rabbit polyclonal) latent associated peptide (LAP-TGF- 1) was obtained from R&D Systems, Inc. (Minneapolis, MN). Paraffin sections of human high-grade osteosarcoma were used as positive controls [13], and a nonspecific IgG was used as a negative control. Other antibodies including mouse monoclonal antibodies (IgG) against antihuman Von Willebrand Factor, antilymphocyte common antigen (LCA), CD68, and nonimmune mouse IgG were purchased from DAKO Corporation (Carpinteria, CA). Cell Culture Aortic valve cusps were obtained from mature female sheep (Western Cross from Thomas Morris, Reisterstown, MD) and cultured on bovine dermal type I collagen (Vitrogen; Cohesion Technologies, Inc, Palo Alto, CA) coated culture plates as described [4]. For calcification studies, 10 mmol/l -glycerophosphate (Sigma, St. Louis, MO) and 1.5 mmol/l CaCl 2 (Sigma) were added to the M199 (Gibco, Grand Island, NY) culture medium. from passages 3 to10 were used in all experiments. The fetal bovine serum (FBS; Hyclone, Logan, UT) used was screened for active TGF- 1 content by ELISA (R&D Systems, Inc), and found to contain less than 35 pg/ml. TGF- 1 Cell Culture Studies Cell suspensions in the presence of 10 ng/ml TGF- 1 (R&D Systems) or a specific anti-tgf- 1 antibody (R&D Systems, Inc) were plated on type I bovine collagen gels at a cell density of 750 cells/mm 2. Serum content in the culture medium was reduced to 0.5% for all TGF- 1 protocols to minimize the effects of TGF- 1 present in fetal bovine serum. In some experiments, TGF- 1 was preincubated with 50 g/ml TGF- 1 neutralizing antibody (R&D Systems, Inc) for 1 hour before adding to the cultures. A caspase inhibitor, Z-VAD (Biomol; Research Laboratories Inc, Plymouth Meeting, PA), was also used in the study. Alkaline Phosphatase Histochemistry and Calcification Measurement Alkaline phosphatase (AP) histochemical activity was detected using nitroblue tetrazolium and X-phosphate according to the manufacturers directions (Boeheringer Mannheim, Mannheim, Germany). 45 Ca accumulation in SAVIC culture was quantitated as described previously [14] except cells were cultured in type I collagen coated 24-well plates in the presence or absence of test reagents. Calcium deposits were stained using Alizarin Red S (Sigma) as described previously [7]. Analysis of Apoptosis by Fluorescent Microscopy Apoptosis related cell-surface changes were detected using a TACS Annexin V-FITC apoptosis detection kit (R&D Systems, Inc) and mounted with Vectashield mounting medium for fluorescence with DAPI (Vector Laboratories, Burlingame, CA). The nuclear chromatin morphology changes due to apoptosis were detected using a fluorescent DNA-binding dye (Hoechst 33258, Aldrich, Milwaukee, WI). The overall pattern of viability and cell death of the cultures was assayed by the LIVE/ DEAD eukolight viability/cytotoxicity kit (Molecular Probes Inc, Eugene, OR). Statistical Analysis Triplicate samples were analyzed for each experiment, and two to four independent experiments were performed for each treatment. All data were calculated as mean standard error of the mean. Student s t-test, or one-way analysis of variance (ANOVA), was used to analyze the data. Value of p less than 0.05 (*) was considered statistically significant. Results Presence of TGF- 1 and TGF- Receptors in Calcific Aortic Stenosis Cusps Compared With Noncalcified Normals TGF- 1 immunohistochemistry studies of five specimens of calcific aortic stenosis removed at cardiac surgery, revealed that all calcified cusps (Alizarin positive, Fig 1) studied were strongly positive (Fig 1A) for extracellular matrix TGF- 1 both in calcific sites and noncalcified extracellular matrix. Noncalcified autopsied human aortic valve cusps showed either no, or only cell-localized, focal TGF- 1 immuno-positivity (Fig 1B, Table 1). Occasional cell-localized TGF- 1 immunostaining was seen in both normal and calcified valves (Table 1). Some lymphocytes, identified by positive LCA staining (not shown), were also immunopositive for TGF- 1 (data not shown). A positive control consisting of sections of human highgrade osteosarcoma, demonstrated cytoplasmic immunoreactive TGF- 1 in neoplastic cells and osteoid matrix produced by neoplastic cells (data not shown) as previously reported [13]. Immunohistochemistry results with

3 Ann Thorac Surg JIAN ET AL 2003;75: TGF- 1 IN AORTIC STENOSIS 459 TGF- 1 Stimulates Aortic Valve Interstitial Cell Aggregation and Calcification When treated with TGF- 1, sheep aortic valve interstitial cells (SAVIC) cultivated on a collagen gel begin to migrate and form aggregates by 48 hrs. After 72 hrs, these initial aggregates formed nodules, which typically consisted of clusters of 10 or more cells with multilayer overgrowth (Fig 2A). When TGF- 1 preincubated with a TGF- 1 function blocking antibody and then added to the SAVIC cultures, nodule formation did not occur (data not shown), thus confirming the specificity of the TGF- 1 effect. No nodules formed in the absence of TGF- 1 treatment (Fig 2B) under the conditions used. The progression of SAVIC nodules to calcification was studied in cultures for up to 14 days in the presence or absence of TGF- 1. Calcification was documented with Alizarin Red S staining and 45 Ca accumulation assay. In the presence of TGF- 1, 45 Ca accumulation was increased about 10% over controls after 7 days in culture; this was further increased to more than 30% over control after 14 days in culture. (Fig 2C). The presence of calcific deposits in the nodules were confirmed by positive Alizarin Red S staining (Fig 2D). Fig 1. The presence of tumor growth factor- 1 (TGF- 1) in human calcified aortic valves and noncalcified control valves. (A) In calcified valves, intense immunopositive TGF- 1 staining (brown) was observed both in the calcified sites (* denotes calcified region) and in the surrounding noncalcified extracellular matrix and occasional cells. (B) Normal noncalcified valves show no extracellular matrix staining, but occasional cell oriented immunostaining. Nonspecific rabbit IgG shows an absence of peroxidase reactivity (data not shown). Peroxidase immunohistochemistry. (Original magnification, 200.) antihuman LAP-TGF- 1 revealed variable, often intense immunopositive staining in both calcified and noncalcified valves in the and interstitial cells (Table 1). Both TGF- RI and RII were present in the interstitial cells of all noncalcified control valves and 4 of the 5 calcified valves (Table 1). Apoptosis and Alkaline Phosphatase Activity Associated With the Development of Calcified Nodules Aortic valve calcification has been noted to be associated with apoptotic vesicles [1] resembling the alkaline phosphatase enriched vesicles present in mineralizing bone [1]. Thus the present studies sought to learn if similar apoptotic mechanisms were operative in aortic valve interstitial cell calcification due to TGF- 1. In SAVIC cultures with added TGF- 1, AP positive staining was characteristically noted in rare isolated individual cells after 24 hrs in culture, and was comparable to the level of AP positivity seen in cultures without TGF- 1 (data not shown). However, 72 hrs studies revealed marked observed differences, with TGF- 1 treated cells demonstrating nodules that were intensely, but not uniformly positive for AP (Fig 3A). In comparison, the control SAVIC cultures without the addition of TGF- 1 demonstrated qualitatively low levels of AP staining (Fig 3B). The presence of apoptotic nodules was observed only in SAVIC treated with TGF- 1, per results of studies with Annexin V and Hoechst staining (Fig 3C 3E). TGF- 1 treated SAVIC demonstrated intense binding of FITC-labeled Annexin V onto 72 hrs nodules (Fig 3C), while most other cells in these cultures were predominantly negative for the occurrence of apoptosis (Fig 3E). The apoptotic index (per Hoechst 33258) was significantly higher in 72-hour TGF- 1 treated cultures ( ) than that of the control cultures ( ; p 0.05). Seven-day calcified nodules in TGF- 1 treated cultures demonstrated strong Annexin V positive staining (Fig 3D), while most other cells in these cultures (not included in nodules) were negative. Hoechst staining confirmed these results, demonstrating relatively intact nuclear DNA in control cells without the addition of TGF- 1 (data not shown). Apoptosis Inhibitor ZVAD Inhibits SAVIC Apoptosis and Calcification, But Not Nodule Formation Induced by TGF- 1 The caspases are a family of enzymes that are integrally involved in the progression of apoptosis [15]. Thus, because our studies (above) indicated that apoptotic

4 460 JIAN ET AL Ann Thorac Surg TGF- 1 IN AORTIC STENOSIS 2003;75: Table 1. Distribution of TGF- 1, LAP-TGF- 1, TGF- RI, and TGF- RII on Noncalcified Aortic Valves and Calcified Aortic Valves Retrieved From Patients Who Underwent Aortic Valve Replacement Case Number Age (years) M/F Interstitial TGF- 1 LAP-TGF- 1 TGF- RI TGF- RII Interstitial Interstitial Interstitial Noncalcified aortic valves from autopsy 1 63 F 2 77 M b 3 71 M a a 4 56 M 5 40 F 6 60 F Interestitial Calcific Deposit Calcified aortic valves from patients who underwent valve replacement surgery 7 73 M 8 83 M 9 91 F M M a Spongiosa layer; b fibrosa layer. negative; focal, faint staining; present, strong staining; very strong staining; extracellular matrix; LAP latent-associated peptide; TGF tumor growth factor. mechanisms were operative in aortic valve interstitial cell calcification, we sought to learn if inhibiting apoptosis by using a caspase inhibitor would reduce the amount of calcification. A caspase inhibitor, Z-VAD-FMK (ZVAD), was used in these studies and had no effect on TGF- 1 stimulated nodule formation. However, this agent significantly inhibited apoptosis, calcification, and cell death induced by TGF- 1 (Fig 4). By adding ZVAD to the cell cultures with TGF- 1, the apoptotic index at 72 hours was decreased significantly compared with TGF- 1 alone, from to (p ), which was not significantly different than the control level ( ) (Fig 4A). The 45 Ca incorporation was also decreased significantly by day 7 and day 14 (p 0.01) (Fig 4B). Live/Dead staining of the culture qualitatively confirmed that apoptosis and nodule formation are independent events, with nodule formation evident even though apoptosis is blocked (Fig 4C 4E). The Effects of an Actin Depolymerization Agent, Cytochalasin D, on TGF- 1 Induced SAVIC Calcification Since SAVIC nodule formation was associated with calcification in the above TGF- 1 experiments, we sought to learn if the presence of nodules was a necessary component of the mineralization mechanism. This may be especially relevant since calcified aortic valves do not contain microscopic pathology findings that resemble these nodules seen in cell culture. Therefore, we used an actin depolymerization agent, cytochalasin D, that is known to interfere with cell motility [16] and thus would hypothetically prevent nodule formation, allowing us to learn if nodules are an essential step in the cell culture calcification process. Cytochalasin D treatment of SAVIC cultures prevented TGF- 1 induced aggregation and nodule formation (Fig 5), but not calcification. In fact, 45 Ca incorporation (Fig 5A) and cell death were significantly increased by the administration of cytochalasin D alone and even more so with TGF- 1 plus cytochalasin D (Fig 5B and 5C). Cytochalasin D alone also stimulated SAVIC apoptosis compared to control ( vs , p ). Thus, these observations and the ZVAD data support the view that although both apoptotic and cytoskeletal-motility mechanisms are operative in this cell culture calcification system due to TGF- 1 administration to SAVIC, only apoptosis contributes to calcification, which can occur in the absence of nodule formation. Comment The present study has provided new insights concerning mechanisms governing the progression of aortic valve stenosis by pathologic calcification. The presence of TGF- 1 in the of calcified aortic cusps was not previously recognized. Furthermore, the TGF- 1 immunopositive regions were not specifically associated with the sporadic inflammatory infiltrates that have been described by others [5], thus suggesting the long standing presence of TGF- 1 in the extracellular matrix rather than an association with more recent inflammatory events. TGF- 1 has been demonstrated previously to potentiate the calcification in cell cultures of vascular smooth muscle cells (SMC) [6], and preliminary studies by our group demonstrated TGF- 1 induced calcification in primary cultures of canine and human aortic valve

5 Ann Thorac Surg JIAN ET AL 2003;75: TGF- 1 IN AORTIC STENOSIS 461 Fig 2. Tumor growth factor- 1 (TGF- 1) causes sheep aortic valve interstitial cell aggregation and calcification, as demonstrated in (A) nodules that have formed after 72 hrs in response to TGF- 1 compared to (B) a comparable culture without TGF- 1 treatment showing no nodule formation. (C) Increased 45 Ca accumulation in TGF- 1 treated cultures compared to control cultures in the absence of TGF- 1 (*p 0.05, **p 0.001, student s t-test). (D) After 14 days in culture in the presence of TGF- 1. All nodules are intensely positive for Coomassie Blue (A and B) and Alizarin Red S (D). Arrows indicate nodules. (Original magnification, 100.)

6 462 JIAN ET AL Ann Thorac Surg TGF- 1 IN AORTIC STENOSIS 2003;75: Fig 3. Increased sheep aortic valve interstitial cell alkaline phosphatase (arrows) per histochemistry, and apoptosis, following tumor growth factor- 1 (TGF- 1) treatment, demonstrating nodule formation after 72 hrs associated with apoptosis. (A) Alkaline phosphatase positive staining (deep purple) in the nodules after 72 hrs, compared to (B) cultures not treated with TGF- 1 that reveal isolated cells that are positive for alkaline phosphatase. (C and D) Annexin V positive (FITC) cells localized in DAPI-stained nodules after 3 days (C) and 7 days (D) in culture in the presence of TGF- 1. (E) Hoechst positive cells in control cultures after 72 hrs (apoptotic nuclei indicated by arrows). (A and B: alkaline phosphatase histochemistry, counterstained with eosin, 100; C and D: Annexin V-FITC, mounted with Vectashield mounting medium with DAPI, 200; E: Hoechst 33258, 200.) interstitial cells [7]. However, the TGF- 1 mechanisms explored in the present SAVIC studies, involving qualitatively increased AP per enzyme-histochemistry and apoptosis, have not been previously described. The presence of TGF- 1 in the extracellular matrix of calcified human aortic valves is very likely in association with TGF- -latent binding proteins (LTBP s), that are covalently bound to the extracellular matrix [14]. TGF- 1 activation and release from LTBP s can be due to protease activity, including matrix metalloproteinase involvement, or other cellular interactions [17]. The presence of intense extracellular TGF- 1-LAP complexes in calcified cusps could also represent residual accumulation reflecting long standing TGF- 1 activity during the course of disease progression (Table 1). SAVIC cell culture studies delineated a number of mechanistic steps following TGF- 1 exposure, which could contribute to the pathogenesis and progression of

7 Ann Thorac Surg JIAN ET AL 2003;75: TGF- 1 IN AORTIC STENOSIS 463 Fig 4. ZVAD, a caspase inhibitor that thereby inhibits apoptosis, prevented sheep aortic valve interstitial cell calcification and cell death induced by tumor growth factor- 1 (TGF- 1). (A) Apoptotic index (measured by counting apoptotic nuclei stained with Hoechst and expressed as percentage of the total cell number counted) showing ZVAD inhibited TGF- 1 induced apoptosis at 72 hrs compared to control (**p 0.01 TGF- 1 treated versus control; *p 0.01 TGF- 1 treated versus ZVAD treated, ANOVA). (B) 45 Ca assay showed decreased 45 Ca accumulation following ZVAD treatment (p 0.01, ANOVA). (C E) Live/dead assay revealed nodule formation with TGF- 1 plus caspase inhibitor, but qualitatively fewer dead cells following ZVAD treatment in 7 days culture, in which (C) is control, (D) is TGF- 1 treated, and (E) is TGF- 1 plus ZVAD treated cultures. (C E) Green fluorescence represents live cells and red fluorescence represents dead cells, with intense overlap indicated by yellow (live/dead assay, 100). calcific aortic stenosis. TGF- 1 is known to be proapoptotic in many cell types [10 12], and thus the TGF- 1 stimulated apoptosis observed in the present studies is in agreement with previous research. However, the strong association of a TGF- 1 effect with both multi-cell aggregates and increased alkaline phosphatase activity has not

8 464 JIAN ET AL Ann Thorac Surg TGF- 1 IN AORTIC STENOSIS 2003;75: Fig 5. Cytochalasin D (cytod) prevented sheep aortic valve interstitial cell aggregation induced by tumor growth factor- 1 (TGF- 1), but potentiated TGF- 1 induced cell-oriented calcification and cell death. (A) 45 Ca accumulation was significantly increased following cytochalasin D treatment alone (p 0.01), and even more so in the presence of TGF- 1 (p 0.01, ANOVA). Live/dead assay showed scattered red fluorescent dead cells in cytochalasin D treated culture (B) as compared to an increased proportion of dead cells following cytochalasin D treatment in the presence of TGF- 1 (C) (live/dead assay, 100). been observed in other investigations of TGF- 1 induced apoptosis. Furthermore, our studies demonstrated that a caspase inhibitor, ZVAD, significantly inhibited calcification and apoptosis of SAVIC in culture, but had no effect on nodule formation. Similar results were observed in aortic SMC [18]. TGF- 1 is a bone inducing cytokine [9, 19], and is part of the same gene super family that includes the bone morphogenic proteins [8]. Interestingly, while other bone related proteins have been described in calcific valvular and vascular disease [20 22], TGF- 1 has not been previously shown to be present in calcified human valvular lesions, until the present studies. Our observations that cytochalasin D promotes calcification are unique. However, studies by others [15, 23] have demonstrated that cytochalasin D is pro-apoptotic and this undoubtedly explains the increased calcification associated with this agent in our cell culture studies. Thus, our pathology observations and experimental results can be integrated into a mechanistic scheme concerning the role of TGF- 1 in the progression of aortic stenosis by calcification. The presence of this cytokine in diseased cardiac valves could hypothetically be due to ongoing endothelial injury with platelet and inflammatory cell infiltration, leading to TGF- 1 accumulation in the extracellular matrix. Our in vitro results clearly show that aortic valve interstitial cells are profoundly influenced by TGF- 1, which triggers their calcification by an apoptotic mechanism. Others have observed the strong association of apoptotic vesicles with calcific aortic stenosis [1]. Furthermore, our studies demonstrated that SAVIC calcification in culture can be prevented in vitro through the use of an antiapoptosis agent, a caspase inhibitor, thereby further emphasizing the importance of the TGF- 1-induced apoptosis as a potential mechanistic step leading to calcific degeneration of diseased aortic valves. Our working hypothesis is that the accumulation of TGF- 1 within aortic stenosis cusps initiates apoptosis of valvular interstitial cells leading to their calcification as an early initial event. This view is supported by our immunohistochemistry results (Fig 1, Table 1) demonstrating that the bulk of the calcified cusp TGF 1 is either present in the extracellular matrix or present in its latent form. Thus, the interactions of active TGF 1 in cusps with heart valve interstitial cells are far more limited than those that can be created in cell culture. Our cell culture results demonstrate an accelerated calcification response due to TGF 1, with initial events occurring in 72 hours, and significant calcification after two weeks. These results are in contrast to the much longer clinical course, typically decades, for the development of significant obstructive valvular calcifications. Nevertheless, the presence of TGF 1 in increased amounts in calcified heart valves compared to lesser amounts in noncalcified cusps supports the view that it may be in part responsible for the progression of pathologic calcification based on the mechanisms investigated in the present studies. In conclusion, TGF- 1 is characteristically present in the extracellular matrix and mineral deposits in calcified aortic stenosis cusps. The results of our cell culture studies support the view that TGF- 1 may contribute to

9 Ann Thorac Surg JIAN ET AL 2003;75: TGF- 1 IN AORTIC STENOSIS 465 the progression of calcific aortic stenosis by initiating the apoptosis associated mineralization of aortic valve interstitial cells. The authors thank Jennifer LeBold for her assistance in preparing the manuscript and Jeanne Connolly, Zhibin Lu, Siobhan Haney, and Suzanne DeFelice for their technical assistance. This research was supported in part by grants from the NHLBI, RO1 HL38118, T32 HL07915, KO8 HL03974, and the William J. Rashkind Endowment of the Children s Hospital of Philadelphia. References 1. Kim KM. Apoptosis and calcification. Scanning Microsc 1995;9: Srivatsa SS, Harrity PJ, Maercklein PB, et al. Increased cellular expression of matrix proteins that regulate mineralization is associated with calcification of native human and porcine xenograft bioprosthetic heart valves. J Clin Invest 1997;99: Mohler ER III, Gannon F, Reynolds C, Zimmerman R, Heane MG, Kaplan FS. Bone formation and inflammation in cardiac valves. Circulation 2001;103: Jian B, Jones PL, Li Q, Mohler ER 3rd, Schoen FJ, Levy RJ. Matrix metalloproteinase-2 is associated with tenascin-c in calcific aortic stenosis. Am J Pathol 2001;159: Otto CM, Kuusisto J, Reichenbach DD, Gown AM, O Brien KD. Characterization of the early lesion of degenerative valvular aortic stenosis histological and immunohistochemical studies. Circulation 1994;90: Watson KE, Bostrom K, Ravindranath R, Lam T, Norton B, Demer LL. TGF- 1, and 25-hydroxycholesterol stimulate osteoblast-like vascular cells to calcify. J Clin Invest 1994;93: Mohler ER III, Chawla MK, Chang AW, et al. Identification and characterization of calcifying valve cells from human and canine aortic valves. J Heart Valve Dis 1999;8: Miyazono K, Kusanagi K, Inoue H. Divergence and convergence of TGF-beta/BMP signaling. J Cell Physiol 2001;187: Bonewald LF, Dallas SL. Role of active and latent transforming growth factor beta in bone formation. J Cell Biochem 1994;55: Hishikawa K, Nakaki T, Fujii T. Transforming growth factorbeta(1) induces apoptosis via connective tissue growth factor in human aortic smooth muscle cells. Eur J Pharmacol 1999;385: Pollman MJ, Naumovski L, Gibbons GH. Vascular cell apoptosis: cell type-specific modulation by transforming growth factor-beta1 in endothelial cells versus smooth muscle cells. Circulation 1999;99: Mattey DL, Dawes PT, Nixon NB, Slater H. Transforming growth factor beta 1 and interleukin 4 induced alpha smooth muscle actin expression and myofibroblast-like differentiation in human synovial fibroblasts in vitro: modulation by basic fibroblast growth factor. Ann Rheum Dis 1997;56: Franchi A, Arganini L, Baroni G, et al. Expression of transforming growth factor beta isoforms in osteosarcoma variants: association of TGF beta 1 with high-grade osteosarcomas. J Pathol 1998;185: Parhami F, Morrow AD, Balucan J, et al. Lipid oxidation products have opposite effects on calcifying vascular cell and bone cell differentiation: a possible explanation for the paradox of arterial calcification in osteoporotic patients. Arterioscler Thromb Vasc Biol 1997;17: Suria H, Chau LA, Negrou E, Kelvin DJ, Madrenas J. Cytoskeletal disruption induces T cell apoptosis by a caspase-3 mediated mechanism. Life Sciences 1999;65: Pedersen SF, Hoffmann EK, Mills JW. The cytoskeleton and cell volume regulation. Comp Biochem Physiol A Mol Integr Physiol 2001;130: Munger JS, Harpel JG, Gleizes PE, Mazzieri R, Nunes I, Rifkin DB. Latent transforming growth factor-beta: structural features and mechanisms of activation. Kidney Int 1997;51: Proudfoot D, Skepper J, Hegyi L, Bennett M, Shanahan C, Weissberg P. Apoptosis regulates human vascular calcification in vitro: evidence for initiation of vascular calcification by apoptotic bodies. Circ Res 2000;87: Bonewald LF. Regulation and regulatory activities of transforming growth factor. Crit Rev Eukaryot Gene Expr 1999;9: Fitzpatrick L, Severson A, Edwards W, Ingram R. Diffuse calcification in human coronary arteries: association of osteopontin with atherosclerosis. J Clin Invest 1994;94: Willette RN, Gu JL, Lysko PG, Anderson KM, Minehart H, Yue T-L. BMP-2 gene expression, and effects on human vascular smooth muscle cells. J Vasc Res 1999;36: Bostrom K, Watson KE, Horn S, Wortham C, Herman IM, Demer LL. Bone morphogenetic protein expression in human atherosclerotic lesions. J Clin Invest 1993;91: White SR, Williams P, Wojcik KR, et al. Initiation of apoptosis by actin cytoskeletal derangement in human airway epithelial cells. Am J Respir Cell Mol Biol 2001;24: INVITED COMMENTARY Although cardiac surgeons encounter calcific degenerative aortic valve disease on a frequent basis, little is known about the molecular causes of this disease. The study by Dr Jian and colleagues presents convincing evidence that transforming growth factor- 1 (TGF- 1) may be an important piece in the puzzle of the etiology of this disease. Clues as to the causes of calcific aortic valve degeneration are revealed by pathologic analysis of diseased cusps removed at the time of valve replacement. In normal circumstances, aortic valve leaflets are composed of three principal layers: the ventricularis at the inflow surface, the spongiosa in the center, and the fibrosa at the outflow surface. The extracellular matrix components of these layers have been well characterized: the ventricularis contains radially aligned collagen and is rich in elastin; the spongiosa is primarily composed of glycosaminoglycans with some collagen; and the fibrosa contains densely packed collagen fibers aligned in parallel with the free edge of the valve cusp [1]. In calcific degeneration, five basic processes are observed: (1) focal endothelial damage of the fibrosa or outflow tract side; (2) inflammatory cellular infiltration of damaged sites with macrophages; (3) vascular smooth muscle transformation into an osteogenic phenotype making these cells capable of initiating mineralization; (4) vascular smooth muscle cell apoptosis with vesicle formation; (5) apatite nucleation of these sites with calcification of devitalized cells [2]. Thus the calcified valve nodules we encounter as surgeons represent focal areas of cell death, petrification 2003 by The Society of Thoracic Surgeons /03/$30.00 Published by Elsevier Science Inc PII S (02)