The Journal of Thoracic and Cardiovascular Surgery
Volume 130, Issue 6 , Pages 1531-1536, December 2005

Ubiquitin fusion degradation 1–like gene dysregulation in bicuspid aortic valve

  • Salah Ali Mohamed, PhD

      Affiliations

    • Department of Cardiac Surgery, University of Schleswig-Holstein Campus Luebeck, Luebeck, Germany
    • ,
  • Thorsten Hanke, MD

      Affiliations

    • Department of Cardiac Surgery, University of Schleswig-Holstein Campus Luebeck, Luebeck, Germany
    • ,
  • Claudia Schlueter, PhD

      Affiliations

    • the Center for Human Genetics, University of Bremen, Bremen, Germany
    • ,
  • Joern Bullerdiek, PhD

      Affiliations

    • the Center for Human Genetics, University of Bremen, Bremen, Germany
    • ,
  • Hans-Hinrich Sievers, MD

      Affiliations

    • Department of Cardiac Surgery, University of Schleswig-Holstein Campus Luebeck, Luebeck, Germany
    • Corresponding Author InformationAddress for reprints: Hans-H. Sievers, MD, Department of Cardiac Surgery, University of Schleswig-Holstein Campus Luebeck, Ratzeburger Allee 160, 23538 Luebeck, Germany

Received 19 April 2005; received in revised form 17 June 2005; accepted 8 August 2005.

Article Outline

Objective

Bicuspid aortic valve, the most common congenital cardiac malformation, is caused by fusion of valve cushions at the onset of valvulogenesis. Although its exact pathogenesis is still unclear, a genetic basis is appearing more and more likely. Search for a potential candidate gene by reviewing semilunar valve morphogenesis led us to the ubiquitin fusion degradation 1–like gene (UFD1L), which is highly expressed in the cardiac outflow tract during embryogenesis.

Methods

Aortic valves were collected during surgery from 39 patients with bicuspid aortic valve (mean age 56.8 ± 18.1 years) and from 38 patients with tricuspid aortic valve (mean age 61.7 ± 16.1 years). Fluorescence in situ hybridization was performed for detection of microdeletion, quantitative reverse transcriptase–polymerase chain reaction to measure gene expression, and Western blotting to analyze the amount of UFD1L gene product.

Results

No microdeletion was found in either group in the critical region of chromosome 22 containing the UFD1L gene. UFD1L gene expression, however, was significantly reduced in bicuspid aortic valve samples (median 787-fold) relative to tricuspid aortic valve samples (median 10,887-fold, P = .001). The amount of UFD1L gene product was also significantly diminished in bicuspid aortic valve samples (3.9 ± 2.6 vs 8.4 ± 4.8 optical density units, P < .05).

Conclusion

Bicuspid aortic valve was associated with downregulation of UFD1L gene expression, supporting the hypothesis that bicuspid aortic valve is a genetic disorder, with the UFD1L gene as a potential candidate gene.

Abbreviations and Acronyms:  AP-2, activating enhancer-binding protein 2, BAV, bicuspid aortic valve, CATCH, cardiac defect, abnormal facies, thymic hypoplasia or aplasia and T-cell deficiency, cleft palate, hypoparathyroidism, and hypocalcemia, dHAND, deciduum, heart, autonomic nervous system, neural crest-derived, GAPDH, reduced glyceraldehyde-phosphate dehydrogenase gene, NF-ATc, nuclear factor of activated T cells, SRY, sex determining region Y–box 4 (SOX4), TAV, tricuspid aortic valve, TBX1, T-box 1 gene, UFD1L, ubiquitin fusion degradation 1–like gene

CTSNet classification:  23, 29, 35

 

Bicuspid aortic valve (BAV) is the most common congenital cardiac malformation, with an estimated incidence of 1% to 2% in the general population.1 BAV not only contributes significantly to the cardiovascular health burden but also poses a challenge to valve surgeons. Knowledge of the pathogenetic mechanisms underlying BAV could help in stratification of treatment modalities. Several family studies have found that anomalies of the left ventricular outflow tract, including BAV, are inherited disorders2; their genetic basis, however, remains obscure. Some recent reports suggest that BAV reflects a developmental defect during embryogenesis.3, 4, 5 At the beginning of valvulogenesis, a population of cells called neural crest cells migrate away from the neural fold and spread throughout the embryo. These cells seem to play a crucial role in normal development of the cardiac outflow tract and semilunar valves.6, 7, 8 The basic helix-loop-helix transcription factor dHAND⁎ is essential for survival of cells in neural crest–derived ventricular structures and aortic arch arteries.9, 10, 11 One of the dHAND-dependent genes is the ubiquitin fusion degradation 1–like gene (UFD1L), which is expressed at high levels during embryogenesis, maps to human chromosome 22, an area called the CATCH region,⁎ and encodes a protein involved in degradation of ubiquinated proteins. These findings suggest that UFD1L gene insufficiency is involved in the development of congenital heart defects, especially malformations in the conotruncal region. We hypothesized that the UFD1L gene might also be dysregulated in BAV, which has not been investigated before.

Back to Article Outline

Materials and Methods 

Study Protocol 

Specimens from 39 patients with BAV (27 male and 12 female, aged 56.9 ± 18.1 years) and 38 patients with tricuspid atrial valve (TAV, 25 male and 13 females, aged 61.7 ± 16.1 years) were analyzed (Table 1). Valve cusps were collected during surgery and immediately frozen in liquid nitrogen. Only fused leaflets were used for analysis of gene expression and protein analysis.

TABLE 1. Patient hemodynamic characteristics, underlying systemic diseases, and valve morphologic types
BAVTAV
Aortic valve disease
Predominant aortic insufficiency169
Predominant aortic stenosis615
Balanced1714
Systemic disease
Hypertension1717
Endocarditis31
Chronic obstructive pulmonary disease13
Diabetes37
Aneurysm of ascending aorta159
Dissection01
Morphologic types of BAV
Raphe between left-right coronary sinus38—
Raphe between right noncoronary sinus1—

Entries represent numbers of patients. BAV, Bicuspid aortic valve; TAV, tricuspid aortic valve.

None of the patients showed clinical signs characteristic of 22q11.2 syndrome. Fluorescence in situ hybridization analysis was performed for detection of microdeletion, reverse transcriptase–polymerase chain reaction for gene expression, and Western blotting for protein analysis.

The study protocol was approved by the institutional ethics committee. Written, informed consent was obtained from each patient.

Microdeletion Analysis 

For fluorescence in situ hybridization analysis, 3-mL samples of venous blood collected from 10 of the patients with BAV were compared with 3-mL samples of venous blood from 10 patients with TAV. Fluorescence in situ hybridization for the DiGeorge syndrome critical region on chromosome 22q11.2 was performed with a dual color probe (N25 LSI/ARSA; Vysis Inc, Downers Grove, Ill). Slides were prepared according to standard cytogenetic procedures. The dual color probe allows simultaneous identification of both number 22 chromosomes. For each patient, at least 10 metaphases were scored for possible deletions in the DiGeorge region. Aberrations of less than 10% in the normal signal pattern were regarded as artifacts.

UFD1L Gene Expression Analysis 

For UFD1L gene expression studies, tissues from 27 patients with BAV were compared with tissues from 20 patients with TAV. Total RNA was prepared from 200-mg snap-frozen tissue with TRIzol (Invitrogen, Karlsruhe, Germany). Reverse transcription was carried out with 1 μg total RNA with the GeneAmp PCR system 2700 (Applied Biosystems, Darmstadt, Germany) in a reaction volume of 25 μl containing 7.5 μmol/L random hexamers, 1× reverse transcription buffer, 220 μmol/L of each deoxynucleotide triphosphate, 20 U ribonuclease inhibitor, and 50 U reverse transcriptase. A 2-μL portion of complementary DNA was used for final application to quantify UFD1L messenger RNA. TaqMan reverse transcriptase–polymerase chain reaction was performed with fluorogenic probe and Universal Master mix on ABI PRISM 7000 SDS (Applied Biosystems) according to manufacturer instructions. Polymerase chain reaction products were measured at the threshold cycle at which fluorescence became detectable above the baseline. Levels of UFD1L messenger RNA were normalized to those of the reduced glyceraldehyde-phosphate dehydrogenase gene (GAPDH) and relative to a calibrator. The relative changes in the gene expression are given by 2−ΔΔCT, where ΔΔCT = (ΔCTSample) − (ΔCTCalibrator). Amplification primers and detection probes were designed for the UFD1L gene in our laboratory with PrimerExpress software version 2.0 (Applied Biosystems) to cross intron/exon boundaries. Oligosequences are based on those retrieved from the GenBank database (U64444 for hUFD1L and BC029618 for hGAPDH). Forward primer 5′-TCC ACA CAG TAC CGC TGC TTC T-3′, reverse primer 5′-TCA GAT GTG GAG AAA GGA GGG AA-3′, and TaqMan FAM-MGB probe 5′-TAG CAT GGC CTA ATG ACA-3′ were used to amplify and detect a 77–base pair UFD1L fragment. The human GAPDH predeveloped TaqMan assay reagent VIC-MGB probe (part number 4326317E; Applied Biosystems) was used as endogenous control.

UFD1L Gene Product Analysis 

Protein analysis was performed on 12 patients in each group. For the Western blot, 500 mg valve tissue was homogenized in 300 μl lysis buffer (Cell Signaling Technology, Inc, Beverly, Mass). sodium dodecylsulfate–polyacrylamide gel electrophoresis was performed according to the method of Laemmli11a on an 18% polyacrylamide gel with 50 μg tissue-extracted protein loaded per lane. After electrophoresis, separated proteins were blotted onto an Immobilon membrane (Millipore [UK] Limited, Watford, United Kingdom). Nonspecific sites were blocked by a solution containing 5% fat-free milk powder in phosphate-buffered saline solution. Membranes were incubated with antibody solution containing 1 μg/mL anti–human ubiquitin overnight at 37°C (R&D Systems, Minneapolis, Minn). After five washes with blocking buffer, membranes were incubated with the secondary antibody solution containing a 1:2000 dilution of horseradish peroxidase–conjugated goat antimouse antibodies. For detection of protein signals, peroxidase reaction was carried out by enzyme-linked chemiluminescence (Amersham Biosciences Europe GmbH, Freiburg, Germany). The polypeptide bands were semiquantified by computerized blot scanning and measurement of optical density. A positive control of UFD1L gene product protein was used (Sigma, Taufkirchen, Germany). In addition we applied the β-actin antibody (Abcam Ltd, Cambridge, United Kingdom) as a loading control antibody.

Statistical Analysis 

Statistical analysis was performed with the Statistical Package for Social Sciences version 9.0 (SPSS GmbH Software, Munich, Germany). Differences among groups were assessed with the Mann-Whitney U test. The changes in gene expression in the two groups were compared with analysis of variance. Bonferroni adjustment was applied for multiple tests.

Back to Article Outline

Results 

Microdeletion on Chromosome 22 

In all investigated cases, no microdeletion of the critical CATCH region on chromosome 22, including the UFD1L gene, was detected (Figure 1, a). Figure 1, b, shows representative metaphase and interphase cells from a patient with BAV. Figure 1, c, shows a positive control sample of the CATCH region deletion (obtained from the Center for Human Genetics, University of Bremen, Germany).

  • View full-size image.
  • Figure 1. 

    a, CATCH region on chromosome 22 and its important genes, including UFD1L. b, Fluorescence in situ hybridization study of blood with 22q13.3 (LSI/ARSA) and DiGeorge/VCFS probe. All signals were present in metaphase and interphase cells of patient with BAV and showed no microdeletion in CATCH region. Arrows indicate that the loci are present on both chromosomes in metaphase and interphase cells. c, Positive control with 22q11.2 microdeletion. Arrow indicates one chromosome 22 with only one signal of LSI ARSA control probe in metaphase cell. The second signal is missed, indicating deletion of the 22q11.2 (obtained from Center for Human Genetics, University of Bremen, Germany).

UFD1L Gene Expression 

UFD1L gene expression was significantly decreased in BAV. The median expression levels in BAV and TAV, respectively, were 787-fold (maximum 16,845-fold, minimum 7-fold) and 10,887-fold (maximum 52,136-fold, minimum 24-fold, P = .001; Figure 2).

  • View full-size image.
  • Figure 2. 

    Relative UFD1L gene expressions (box plots) in aortic leaflets from patients with BAV (median 787-fold) and in leaflets from patients with TAV (median 10,887-fold). Numbers were calculated with 2−ΔΔCT method. Significantly higher level of UFD1L gene expression was seen in TAV than in BAV (P = .001).

UFD1L Gene Product Quantification 

The UFD1L gene product protein level was significantly reduced in BAV (3.9 ± 2.6 optical density units) relative to TAV (8.4 ± 4.8 optical density units, P < .05; Figure 3).

  • View full-size image.
  • Figure 3. 

    a, UFD1L gene product levels (box plots) in aortic leaflets from patients with BAV and patients with TAV. b, Levels of UFD1L gene product protein on Western blot (arrow) in aortic leaflets from 3 patients with BAV (lanes 2-4) and from 3 patients with TAV (lanes 5-7), in 1 positive pure control (lane 8), in water (lane 9), and in two protein markers (lanes 1 and 10). c, Measurement of amount of protein analyzed by Western blotting, with β-actin as loading control antibody. Lanes 1 and 8 are positive control, lane 9 is water, and lane 10 is protein marker.

Back to Article Outline

Discussion 

This study shows that BAV is associated with a reduced UFD1L gene expression. It supports the hypothesis of BAV as a genetic disorder and also indicates UFD1L as among the candidate genes warranting further investigation.

BAV is the most common congenital cardiac malformation, affecting 1% to 2% of the general population.1 BAV contributes significantly to the cardiovascular health burden. Comprehensive knowledge of the pathogenesis of BAV may be important for treatment strategies.

A number of mechanisms may be involved, either alone or combined, in the pathogenesis of BAV. Theoretically, epigenetic factors during cardiac morphogenesis, such as fluid forces, may contribute to BAV development in a manner similar to that of factors resulting in impaired valve function, valve stiffness, and opening abnormalities.12 Several recent studies have demonstrated that BAV is inheritable as part of the whole spectrum of anomalies of the left ventricular outflow tract. Wessels and colleagues2 even suggest that all left ventricular outflow tract anomalies, including BAV, may be caused by a single gene defect.

A number of studies have shown that 22q11.2 deletions cause a variety of cardiac outflow tract defects.13, 14, 15 These cardiac defects are also seen after neural crest ablation, suggesting that certain genes controlling neural crest cells may be involved in the development of cardiac outflow tract malformations and may map to chromosome 22. The critical CATCH region of this chromosome consists of many genes, including the UFD1L and T-box 1 (TBX1) genes. Yamagishi and associates10 detected the UFD1L gene in all 182 patients with 22q11.2 syndrome in their investigation.

During embryogenesis, the UFD1L gene is highly expressed in certain tissues. It encodes a component of a multienzyme complex involved in the degradation of ubiquitin fusion proteins. A downregulation of the UFD1L gene, hypothetically resulting from an anomalous behavior of neural crest cells, may lead to reduced degradation activities and thus explain in part the hypotheses of Sans-Coma and associates,16 who demonstrated in Syrian hamsters the fusion of valve cushions as a key factor in the development of congenital BAV. Furthermore, mouse homolog Ufd1 gene expression was detected in the cardiac outflow tract at the same time as neural crest cell condensation, before these cells transformed into ectomesenchymal cells.

These observations highlight the important role of the UFD1L gene in the development of ectoderm-derived structures, including neural crest cells.10, 17 These cells have also been found in developing aortic leaflets,18 pointing to the UFD1L gene as a candidate for the pathogenesis of BAV. In our study, no deletion of the critical CATCH region on chromosome 22, including the UFD1L gene, was found in any of the patients with BAV. This was not entirely unexpected, because a microdeletion would theoretically lead to more phenotypic abnormalities. We did find, however, downregulation of the UFD1L gene and consequently diminished amounts of its gene product. Our results accord with the finding that a functional attenuation of chick UFD1L in cardiac neural crest cells results in an increased incidence of conotruncal septation defects.14 The exact mode of downregulation of the UFD1L gene and how it induces BAV in human beings remain unclear. Some findings implicate the transcription factor AP-2,⁎ which is active as a regulator of gene expression during the differentiation and development of neural crest cells.19, 20 It is also involved in the regulation of transforming growth factor α,21 estrogen receptor,22 type IV collagenase,23 insulinlike growth factor binding 5,24 and many other genes. A systematic investigation is needed to elucidate the important biologic functions of AP-2 in the regulation of the UFD1L gene.

Another gene that maps to the CATCH region involved in 22q11.2 syndrome and may be involved in the pathogenesis of BAV is the TBX1 gene. We measured TBX1 gene expressions in 10 patients in each group (data not shown), but we did not find a significant difference between those with BAV and those with TAV. Whether this is because TBX1, unlike UFD1L, is not expressed in neural crest cells25 remains to be studied.

Ranger and colleagues26 and de la Pompa and coworkers27 have reported that the targeted dysregulation of the transcription factor NF-ATc † resulted in selective absence of the aortic valves, leading to death in utero. A similar phenotype was observed in SRY (sex determining region Y)–box 4 (SOX4)–deficient mice.28 Disorders of the latter transcription factors, however, do not lead to the characteristic phenotype of fused leaflets seen in BAV.

A limitation of this study is the fact that it demonstrates only an association, not a causal relationship, between BAV and UFD1L downregulation. Whether the relationship is causal must be determined by in vivo experiments targeting downregulation of the UFD1L gene in an animal model. Also, we did not show pedigree analysis of familial clustering explaining a genetic cause of BAV. Furthermore, only a few neural crest cells could be identified in late fetal aortic semilunar valves as a result of a reduction in their numbers caused by death or failure to divide.29 This may have influenced our results, although the ages of our two groups were comparable, and we found no correlation between the age of patients with BAV and the degree of UFD1L gene expression. In addition, the two groups differed with regard to the underlying valvular etiology, with more aortic insufficiencies in the BAV group. Whether this influenced the gene expression remains unclear, but it seems rather unlikely because leaflet fusion was the basic pathology underlying both stenotic and insufficient valves. Although the genetic pathomechanism of BAV is probably far more complicated (regarding, for example, accompanying defects such as coarctation), our findings appear to take the first steps toward an understanding of the genetic basis of BAV.

Back to Article Outline

 

We thank Sven Hauke and Norbert Drieschner for their valuable help with the fluorescence in situ hybridization analyses.

Back to Article Outline

References 

  1. Roberts WC . The congenitally bicuspid aortic valve . Am J Cardiol . 1970;26:72–82
  2. Wessels MW , Berger RM , Frohn-Mulder IM , Roos-Hesselink JW , Hoogeboom JJ , Mancini GS , et al.  Autosomal dominant inheritance of left ventricular outflow tract obstruction . Am J Med Genet A . 2005;134:171–179
  3. McKusick VA . Association of congenital bicuspid aortic valve and Erdheim's cystic medial necrosis . Lancet . 1972;1:1026–1027
  4. Nataatmadja M, West M, West J, Summers K, Walker P, Nagata M, et al. Abnormal extracellular matrix protein transport associated with increased apoptosis of vascular smooth muscle cells in Marfan syndrome and bicuspid aortic valve thoracic aortic aneurysm Circulation. 2003;108 Suppl 1:II329-34.
  5. Bonderman D , Gharehbaghi-Schnell E , Wollenek G , Maurer G , Baumgartner H , Lang IM . Mechanisms underlying aortic dilatation in congenital aortic valve malformation . Circulation . 1999;99:2138–2143
  6. Zigmond MJ , Bloom FE , Landis SC , Roberts JL , Squire LR . Neurogenesis and migration . In:  Zigmond MJ ,  Bloom FE ,  Landis SC ,  Roberts JL ,  Squire LR editor. Fundamental neuroscience . San Diego: Academic Press; 1998; p. 451-80
  7. Le Douarin NM , Teillet MA . Experimental analysis of the migration and differentiation of neurectodernam mesenchymal derivatives, using a biological cell making technique . Dev Biol . 1974;41:162–184
  8. Kirby ML , Waldo KL . Role of neural crest in congenital heart disease . Circulation . 1990;82:332–340
  9. Srivastava D , Cserjesi P , Olson EN . A subclass of bHLH proteins required for cardiac morphogenesis . Science . 1995;270:1995–1999
  10. Thomas T , Yamagishi H , Overbeek PA , Olson EN , Srivastava D . The bHLH factors, dHAND and eHAND, specify pulmonary and systemic cardiac ventricles independent of left-right sidedness . Dev Biol . 1998;196:228–236
  11. Yamagishi H , Garg V , Matsuoka R , Thomas T , Srivastava D . A molecular pathway revealing a genetic basis for human cardiac and craniofacial defects . Science . 1999;283:1158–1161
  12. Laemmli UK . Cleavage of structural proteins during the assembly of the head of bacteriophage T4 . Nature . 1970;227:680–685
  13. Hove JR , Koster RW , Forouhar AS , Acevedo-Bolton G , Fraser SE , Gharib M . Intracardiac fluid forces are an essential epigenetic factor for embryonic cardiogenesis . Nature . 2003;421:172–177
  14. Momma K , Kondo C , Ando M , Matsuoka R , Takao A . Tetralogy of Fallot associated with chromosome 22q11 deletion . Am J Cardiol . 1995;76:618–621
  15. Ryan AK , Goodship JA , Wilson DI , Philip N , Levy A , Seidel H , et al.  Spectrum of clinical features associated with interstitial chromosome 22q11 deletions: a European collaborative study . J Med Genet . 1997;34:798–804
  16. Yamagishi C, Hierck BP, Gittenberger-De Groot AC, Yamagishi H, Srivastava D. Functional attenuation of UFD1L, a 22q11.2 deletion syndrome candidate gene, leads to cardiac outflow septation defects in chicken embryos. Pediatr Res. 2003;53:546-53.
  17. Sans-Coma V , Fernandez B , Duran AC , Thiene G , Arque JM , Munoz-Chapuli R , et al.  Fusion of valve cushions as a key factor in the formation of congenital bicuspid aortic valves in Syrian hamsters . Anat Rec . 1996;244:490–498
  18. Wilson DI , Goodship JA , Burn J , Cross IE , Scambler PJ . Deletions within chromosome 22q11 in familial congenital heart disease . Lancet . 1992;340:573–575
  19. Jiang X , Rowitch DH , Soriano P , McMahon AP , Sucov HM . Fate of the mammalian cardiac neural crest . Development . 2000;127:1607–1616
  20. Mitchell PJ , Timmons PM , Hebert JM , Rigby PW , Tjian R . Transcription factor AP-2 is expressed in neural crest cell lineages during mouse embryogenesis . Genes Dev . 1991;5:105–119
  21. Schorle H , Meier P , Buchert M , Jaenisch R , Mitchell PJ . Transcription factor AP-2 essential for cranial closure and craniofacial development . Nature . 1996;381:235–238
  22. Wang D , Shin TH , Kudlow JE . Transcription factor AP-2 controls transcription of the human transforming growth factor-alpha gene . J Biol Chem . 1997;272:14244–14250
  23. McPherson LA , Baichwal VR , Weigel RJ . Identification of ERF-1 as a member of the AP2 transcription factor family . Proc Natl Acad Sci U S A . 1997;94:4342–4347
  24. Frisch SM , Morisaki JH . Positive and negative transcriptional elements of the human type IV collagenase gene . Mol Cell Biol . 1990;10:6524–6532
  25. Duan C , Clemmons DR . Transcription factor AP-2 regulates human insulin-like growth factor binding protein-5 gene expression . J Biol Chem . 1995;270:24844–24851
  26. Hierck BP, Molin DG, Boot MJ, Poelmann RE, Gittenberger-De Groot AC. A chicken model for DGCR6 as a modifier gene in the DiGeorge critical region. Pediatr Res. 2004;56:440-8.
  27. Ranger AM , Grusby MJ , Hodge MR , Gravallese EM , de la Brousse FC , Hoey T , et al.  The transcription factor NF-Atc is essential for cardiac valve formation . Nature . 1998;12:186–190
  28. de la Pompa JL , Timmerman LA , Takimoto H , Yoshida H , Elia AJ , Samper E , et al.  Role of the NF-Atc transcription factor in morphogenesis of cardiac valves and septum . Nature . 1998;12:182–186
  29. Schilham MW , Oosterwegel MA , Moerer P , Ya J , de Boer PA , van de Wetering M , et al.  Defects in cardiac outflow tract formation and pro-B-lymphocyte expansion in mice lacking Sox-4 . Nature . 1996;380:711–714
  30. Yacoub MH , Cohn LH , et al.  Novel approaches to cardiac valve repair from structure to function: part I . Circulation . 2004;109:942–950
  • ⁎ dHAND = deciduum, heart, autonomic nervous system, neural crest-derived.
  • ⁎ CATCH = cardiac defect, abnormal facies, thymic hypoplasia or aplasia and T-cell deficiency, cleft palate, hypoparathyroidism, and hypocalcemia.
  • ⁎ AP-2 = activating enhancer-binding protein 2.
  • † NF-ATc = activating enhancer-binding protein 2

PII: S0022-5223(05)01382-6

doi:10.1016/j.jtcvs.2005.08.017

The Journal of Thoracic and Cardiovascular Surgery
Volume 130, Issue 6 , Pages 1531-1536, December 2005

Article Tools

* Image Usage & Resolution