Heat Shock Protein 27 Is Associated With Freedom From Graft Vasculopathy After Human Cardiac Transplantation
http://www.100md.com
循环研究杂志 2005年第7期
Transplant Immunology (A.I.D.S., M.L.R.), National Heart and Lung Institute, Imperial College London, Harefield Hospital, Harefield, Middlesex, UK
Kennedy Institute of Rheumatology Division (R.W.), Faculty of Medicine, Imperial College London, Hammersmith, London, UK
the Department of Cardiology (A.G.M., N.R.B.), Royal Brompton and Harefield NHS Trust, Harefield Hospital, Harefield, Middlesex, UK
the Proteome Research Centre (M.J.D.), Conway Institute of Biomolecular and Biomedical Research, University College Dublin, Belfield, Ireland.
Abstract
Experimental studies have suggested that protective genes protect allografts from cardiac allograft vasculopathy (CAV), the major complication after cardiac transplantation. Here we have sought to confirm this hypothesis using long-term heart transplant recipients. Twenty-two patients that were 9 years or older after transplant were investigated; 11 of these were without angiographic evidence of CAV; 11 had developed early CAV at 1 to 3 years after transplant. To identify proteins that may act as protectors from CAV, a global proteomic approach was used comparing cardiac biopsies from 12 patients taken within the first 2 weeks after transplant and those taken after 9 years from the same patient. Proteins were separated by 2-D gel-electrophoresis, detected by silver staining, and analyzed using Progenesis software. A particular protein spot was found in 4/6 biopsies from patients without CAV, but absent from 5/6 biopsies from those with CAV (P=0.24); however, quantitative analysis of spot intensity showed a significant difference (0.061±0.05 versus 0.003±0.01, P=0.04). This spot was identified by mass spectrometry and a combination of techniques as a diphosphorylated form of HSP27. Immunohistochemistry of further biopsies not only validated that HSP27 was more abundantly expressed on biopsies without CAV but also showed it to be localized to blood vessels. In contrast, vessels from patients with CAV did not express HSP27 (P=0.028x10eC4). Immunohistochemistry of 12 further early biopsies and nontransplanted heart showed HSP27 to be present in normal blood vessels. These findings suggest that expression of a specific diphosphorylated form of HSP27 is associated with healthy blood vessels; it appears to be lost from vessels of patients with graft vasculopathy.
Key Words: cardiac allograft vasculopathy heat shock protein 27 protection proteomics and transplantation
Introduction
Accelerated graft atherosclerosis or transplant associated coronary artery disease, often referred to as chronic rejection or cardiac allograft vasculopathy (CAV), is one of the most serious long-term complications of heart transplantation; the incidence increases every year after transplant and is currently 43% at 7 years.1 Histologically, CAV consists of a diffuse concentric intimal proliferation comprised of smooth muscle cells and fibroblasts. The disease is not confined to the large epicardial arteries, but also affects the small vessels penetrating the myocardium.2
The endothelial response to injury hypothesis proposed by Ross in 19933 to explain nontransplant atherosclerosis is equally applicable to transplant atherosclerosis. Essentially, the hypothesis proposes that an initial insult results in endothelial cell activation and upregulation of cytokines, chemokines, and adhesion molecules. The latter events allow transendothelial migration of monocytes which may become further activated in the vessel wall by modified low-density lipoproteins, and subsequently release cytokines and growth factors causing smooth muscle cell proliferation. There are a number of possible insults to the endothelium that could initiate such a cascade of events culminating in CAV; these include brain stem death, preservation injury, reperfusion injury, and the alloimmune attack after transplantation.4
In recent years there has emerged the important concept of cytoprotection. It is known that the endothelial response to injury can be protective as well as inflammatory. This may explain why some patients remain free of CAV for many years after their transplant whereas others succumb at a very early stage. For example, upregulation of HO-1, A20, and Bcl-xL genes have been described as protecting against antibody-mediated graft vasculopathy in hamster/mouse to rat cardiac xenografts models5,6 and in a mouse allograft model.7
The aim of this study was to use a global proteomic approach to investigate the hypothesis that long-term survivors of heart transplantation, without signs of CAV, show preferential expression of potential protective proteins. Two-D gel electrophoresis has been used to compare quantitative protein expression by cardiac biopsies of transplant patients with and without angiographic evidence of cardiac allograft vasculopathy.
Materials and Methods
Patients
This is a retrospective study of patients transplanted at this center between 1990 and 1993. The aim of the study was to compare biopsies from long-term survivors (9 years) who were free of CAV to long-term survivors who had developed CAV within the first 3 years of their transplant. Patients at this center receive routine surveillance endomyocardial biopsy procedures at weekly intervals for the first month after transplant; in that era (1990eC1993), in all cases, extra pieces of biopsy were taken, they were snap frozen in liquid nitrogen, and stored in liquid nitrogen for future research. The study population consisted of those patients where biopsy material was available. Surveillance angiography for CAV is performed routinely at 1 to 2 year intervals, CAV detected by angiography was defined as any new irregularity or stenosis not present in patients’ previous angiograms 25% luminal diameter. At the time of angiography, patients do not normally undergo the biopsy procedure. Thus, patients were selected if they were >9 years after transplant, had early biopsies available (taken within 21 days of their transplants), and if they consented to an additional biopsy procedure as part of their annual angiography assessment, many years after transplantation. Twenty-two heart transplant recipients (all male) fulfilled these requirements, and their ages ranged from 17 to 68 years. Indications for transplant were ischemic heart disease (n=14), dilated cardiomyopathy (n=7), and congenital heart disease (n=1). Eleven patients had developed early CAV (25% to 50% stenosis, first detected 1 to 3 years); their late biopsies were taken at 133.1±10.7 months (mean±SEM); early biopsies from 6 of these patients were available and obtained at 17.6±3.3 days. Eleven patients had not developed CAV at the time of biopsy (126.5±5.7 months); early biopsies from 6 of these patients had been taken at 15.5±3.3 days. Two samples of endomyocardial biopsy were taken at the time of angiography for this study; 1 for proteomics and 1 for immunohistochemistry. Right ventricular endomyocardial biopsies were embedded in OCT compound, frozen in isopentane, and then stored in liquid nitrogen until required. In addition to the biopsies collected specifically for this study, 12 early biopsies (taken within 3 months of transplant) and 9 pieces of explanted heart were used for immunohistochemistry. Local ethical permission (Royal Brompton and Harefield Research Ethics Committee) was obtained in accordance with the Helsinki Declaration of 1975 and written informed consent obtained from each patient.
Sample Preparation and Quantification
Total tissue proteins were extracted as previously described8 from 24 biopsies (n=6 patients with CAV and 6 without CAV, 1 early and 1 late biopsy from each patient). The resulting supernatant was harvested and protein concentration determined using the 2D-Quant kit (Amersham Biosciences). The amount of total protein extracted from each biopsy was 256±21 e (mean±SEM), ranging from 143 to 453 e. Solubilized protein samples were stored at eC80°C until required.
Proteomic Analysis
Protein extracts (100 e) of right ventricular endomyocardial biopsies (n =24) were separated by 2-D gel electrophoresis (2DE) as described by De Souza et al.8 Briefly, the first dimension was nonlinear pH 3 to 10 IPG isoelectric focusing, and the second dimension was 12% SDS-PAGE. Gels were silver-stained and computer-assisted image analysis was performed as described previously.9 Protein spot matching and intensities were assessed using the Progenesis Workstation, Version 2003.02 (Nonlinear Dynamics). Four comparisons were made: (1) early versus late without CAV; (2) early versus late with CAV; (3) early versus early; (4) late versus late. All results are expressed as arithmetic means (±SD). Paired and unpaired Student t tests were performed, with P<0.05 required for statistical significance. The first 2 comparisons were intrapatient comparisons (paired t tests); the second 2 were interpatient comparisons (unpaired t tests). Protein spots showing a statistically significant difference in intensity (changed in expression by 1.5-fold or more, P<0.05) were manually excised from micropreparative 2D gels (400 e protein) for identification by tandem mass spectrometry. A detailed methodology is provided in the online data supplement available at http://circres.ahajournals.org.
Phosphoprotein Identification
Phosphoprotein identification was examined on micropreparative (400 e protein) 2-D gels of left ventricular proteins (n=2). Total protein extracts were separated on pH 3 to 10 NL IPG strips, followed by a 12% SDS polyacrylamide gel. Pro-Q Diamond (Molecular Probes) phosphoprotein gel stain provides a method for selectively staining phosphoproteins in polyacrylamide gels,10 and was used according to the manufacturer’s instruction and scanned on the Typhoon 9400 scanner (Amersham Biosciences) at 532 nm, emission filter at 560 LP, and the photomultiplier tube set at 600 V. Subsequently, Pro-Q diamond stained gels were washed in MQ water and then fixed overnight in a methanol: acetic acid: water solution (4:1:5 v/v/v), before staining with a mass spectrometry compatible silver stain.
Western Blottting
Immediately after 2DE, the separated proteins were transferred by semidry blotting onto nitrocellulose membranes for 2 hours at 0.8 mA per cm2. HSP27 proteins were identified by immunostaining with a monoclonal anti-HSP27 antibody (SPA-800, Stressgen) and polyclonal antibodies specific for phosphorylated HSP27 at Serine residues 78 and 82 (SPA-523 and SPA-524 respectively, both from Stressgen).
Immunohistochemistry
Frozen sections (5 e) of explanted cardiac tissue (n=9), early biopsies (taken within 3 months of transplant, n=12), and late biopsies from patients with and without CAV (n=11 for each group) were stained using the immunoperoxidase method as described by Taylor et al.11 Briefly, tissue sections were stained with PBS as a negative control, and monoclonal antibodies to CD31, a marker of endothelial cells (DakoCytomation), smooth muscle -actin (Sigma), or HSP27 (SPA-800, Stressgen).
Results
Progenesis analysis of protein spots separated and displayed by 2DE revealed that 69 proteins showed time-dependent changes in abundance in comparison 1 (early versus late without CAV), and 38 proteins changed in biopsies from comparison 2 (early versus late with CAV). Twenty-three proteins changed in comparison 3 (early CAV versus early non-CAV), and 21 in comparison 4 (late CAV versus late non-CAV). Proteins differences in the 4 comparisons outlined above changed by 1.5-fold or more, P<0.05. The proteins of interest have been identified by tandem mass spectrometry. In summary in comparison 1, 37 of the 69 proteins have been identified, whereas 14 of 38, 14 of 23, and 16 of 21 have been identified from comparisons 2, 3, and 4 respectively (see online Figure I and online Table I).
HSP27 Is Associated With Freedom From CAV
Figure 1 shows representative 2D images from both groups of patients. The spots indicated with asterisks were identified as HSP27 by mass spectrometry, but no difference was observed in their expression between the groups. However, an additional spot in this region (Figure 1, spot 3306, arrowed), was significantly differentially expressed between the late groups. Eight tryptic peptides from this spot were sequenced by mass spectrometry, all of which mapped onto the sequence of HSP27. Figure 2A shows a representative example of one of these spectra. This protein was present in the 2-D patterns obtained from 4/6 late biopsies from subjects without disease, but was not detected in 5/6 late biopsies from patients with CAV (P=0.24, Fisher’s exact test). Quantitative analysis demonstrated a 20-fold increase in spot intensity in biopsies free of CAV compared with those with CAV (spot intensities were 0.061±0.05 versus 0.003±0.01, P=0.04). The trend was similar in the early groups (spot intensities 0.47±0.38 versus 0.11±0.14; without CAV versus with CAV), but were not statistically significant (P=0.10).
Phosphorylation Status of HSP27 Protein Spot 3306
The most basic form of HSP27 (spot 2079, highlighted on Figure 1) was observed at a pI of 5.9, which is in good agreement with previously published values for the unphosphorylated protein,12 and is consistent with the Figure of 5.98 calculated from the amino acid sequence of HSP27. The observed pI of spot 3306 however was significantly lower, at 5.42, which would be consistent with the presence of acidic modifications such as phosphorylation. Support for this hypothesis was provided by staining with Pro-Q Diamond, which is specific for phosphoproteins. Figure 3A through 3D shows overlays of the same gel, visualized with either silver or with Pro-Q Diamond, and indicates that both stains colocalize to spot 3306.
Tandem mass spectrometry provided unequivocal evidence for phosphorylation of Serine 82. In unphosphorylated HSP27 (spot 2079) a doubly charged ion was observed at m/z 538.3, which was shown by collisional fragmentation and MS/MS to correspond to the sequence QLSSGVSEIR, representing residues 80 to 89 of HSP27 (Figure 2B). This peptide is clearly unphosphorylated, because the mass difference between the C-terminaleCcontaining fragments y8 and y7, y7 and y6, and y4 and y3, is 87, corresponding to the residue mass of serine. By contrast, in the tryptic digest of spot 3306, a doubly charged ion was observed at m/z 578.3, which would be consistent with the addition of a single phosphate group to the peptide QLSSGVSEIR. Phosphorylation is further indicated by a facile losses of H3PO4 (mass 98) from the protonated molecule (Figure 2C). The fragmentation pattern establishes Serine 82 as the site of phosphorylation, because the mass difference between the y8 (m/z 914) and y7 (m/z 747) fragments is 167 rather than 87, and there is a loss of dehydroalanine (69) between the dephosphorylated y8 ion (m/z 816) and y7 (m/z 747). The other 2 serines in the sequence are unphosphorylated, because the residue mass between the respective contiguous y ions is 87 (Figure 2C). The observed pI of spot 3306 (5.42) is consistent with diphosphorylated HSP27, both from calculation (predicted value 5.6) and by agreement with published data in which bis-phospho HSP27 was observed to migrate at a pI of 5.45.12 There are 3 experimentally verified phosphorylation sites in human HSP27, Serine 15, 78, and 82. The present data suggest that Serine 15 is not phosphorylated, because we observed a doubly charged ion at m/z 481.2, which was shown by MS/MS (Figure 2D) to correspond to unphosphorylated GPSWDPFR (residues 13 to 20), whereas the phosphorylated form of this peptide (m/z 521.2) was undetectable. It was unfortunately impossible to verify the phosphorylation state of Serine 78, as the tryptic peptide which includes it (ALSR) is too small to be easily detected in a conventional ESI LC MS/MS experiment.
Further confirmation was provided by Western blotting with antibodies specific for the phosphorylation sites Ser-78 and Ser-82 and with an antibody that recognizes all forms of HSP27 whether phosphorylated or unphosphorylated. These results (Figure 3E through 3G; spot 3306 highlighted) confirm the presence of phosphorylation on Serine residues 78 and 82.
Localization of HSP27 by Immunohistochemistry
Having found by 2DE that HSP27 is expressed in late biopsies from patients who are angiographically free of CAV, but is absent from late biopsies form those patients who develop CAV, immunohistochemistry was used to validate this finding in paired late cardiac biopsies (n=12) and 10 further late biopsies from patients in this study. HSP27 was found expressed diffusely on all myocytes, and within endothelial and smooth muscle cells of medium to large blood vessels in all 11 biopsies examined from patients without disease. Positive staining of the endothelial and smooth muscle cells of blood vessels was confirmed by staining serial sections with CD31 and SMA (see Materials and Methods for details). In contrast, expression of HSP27 on all biopsies from patients with CAV (n=11) was confined to myocytes only, and absent from the endothelial and smooth muscle cells (Figure 4). Using the Fisher exact test, a P value of 0.028x10eC4 was calculated for the presence/absence of HSP27. Immunohistochemistry thus supported and extended the proteomic results, by showing less expression of HSP27 on blood vessels in late biopsies from patients with CAV. Early biopsies from this group of patients were not available for immunohistochemistry, but it was possible to examine 12 early biopsies from a different group of patients transplanted within the same era. All of these biopsies showed HSP27 to be expressed on all myocytes and blood vessels. In addition, to examine normal heart, 9 pieces of explanted heart were investigated and all showed diffuse expression of HSP27 on cardiac myocytes and blood vessels.
Discussion
In this study a global proteomics approach has been used to examine protein expression in endomyocardial biopsies from human transplant patients with and without angiographic evidence of cardiac allograft vasculopathy. We postulated that differential expression of protective proteins could happen at early or late times after transplantation, and it would therefore be ideal to be able to examine biopsies from both time periods. At this center, it has been possible to retrospectively identify patients who were at least 9 years after transplant for whom early biopsies were available for research. In total, 141 proteins were significantly altered (by 1.5-fold or more) between the groups. Of these, 81 were identified by mass spectrometry (see online Table I). The most significant finding was of a particular protein spot, found only in gel preparations from late biopsies from patients without CAV; this was identified by mass spectrometry to be a specific form of HSP27. The position of the HSP27 protein spot and the pI shift from the theoretical value suggested that this protein was phosphorylated. This was confirmed using a stain specific for phosphoproteins (Pro-Q Diamond), mass spectrometry, and Western blotting, and it was demonstrated to be diphosphorylated at Serine residues 78 and 82. We have focused on this particular protein because it showed the greatest fold change, namely 20-fold increase in expression in biopsies from patients without disease (spot intensity 0.061±0.05) than in biopsies from patients with disease (0.003±0.01). This difference was statistically significant (P=0.04). It should be noted that when comparing biopsies for presence or absence of HSP27, the difference between the groups was not significantly different (P=0.24, Fisher exact test). The discrepant results reflect the skewed distribution of HSP27, being highly expressed in 4/6 biopsies without CAV and below detection in 2 biopsies without CAV. A possible explanation why only 4/6 biopsies without CAV were positive may be caused by a relative lack of blood vessels in the 2 negative biopsies. Thus immunohistochemistry of further late biopsies not only confirmed that HSP27 was more strongly expressed in late biopsies which were free of CAV, but it showed vascular localization, something not found in late biopsies from patients with CAV (P=0.028x10eC4). Taken together, these results strongly suggest vascular expression of HSP27 is associated with freedom from CAV. It has been previously reported that HSP27 is diffusely expressed by cardiac myocytes of transplanted hearts,13 but blood vessels were not investigated. Little is known about expression of HSP27 in normal and diseased human blood vessels. Martin-Ventura et al14 have recently reported that atherosclerotic vessels maintained in culture release very little HSP27 protein into the medium compared with control vessels; they also showed smooth muscle cells of mammary arteries to be positive for HSP27. These findings therefore support our findings of vascular expression of HSP27 in biopsies from patients that did not develop CAV. In an attempt to investigate nondiseased cardiac blood vessels, early biopsies from another group of 12 transplant patients and 9 pieces of explanted heart were examined; all showed HSP27 positive blood vessels. The current results together with those of Martin-Ventura et al14 suggest that HSP27 is expressed normally by blood vessels, and lack of expression is associated with both nontransplant and transplant atherosclerosis. Endothelial cells and smooth muscle cells of all medium-large size vessels were positive for HSP27 in early biopsies and explanted heart. Interestingly, staining with antibody to CD31, a marker of endothelial cells, demonstrated many tiny capillary endothelial cells which appeared not to be HSP27 positive. The functional significance of HSP27 may therefore be caused more by its association with smooth muscle cells rather than endothelial cells.
HSP27 is a small heat-shock protein, the expression of which correlates with increased survival following cytotoxic stimuli.15 Like all heat-shock proteins, it functions as a molecular chaperone in protein biosynthesis to facilitate protein folding and prevent accumulation of misfolded proteins in the cell. Other means by which it is cytoprotective include protection from apoptosis,16eC19 modulation of oxidative stress,20 and regulation of the cytoskeleton.21 Several studies have shown that overexpression of HSP27 increases the stability of F-actin microfilaments during exposure to stress such as hyperthermia,22 oxidants,23 and cytochalsin D.21 This activity may depend on the degree of phosphorylation and its structural organization. Use of transgenic mice overexpressing human HSP27 has demonstrated that hearts from these mice are protected from ischemia reperfusion injury24 and nerves are protected from chemically induce apoptosis.25
The most likely explanation for the protective effect of HSP27 on pathology of blood vessels is its antiapoptopic properties16 and ability to stabilize the actin cytoskeleton. Apoptosis has been found in transplanted hearts at all times after transplantation,26 and has been shown to be involved in the pathogenesis of chronic rejection in cardiac allografts.27 As well as binding to actin, HSP27 also binds to the intermediate filament vimentin.28eC30 Caspase dependent cleavage of vimentin is said to be a prerequisite for apoptosis in all cell types,31 suggesting that stabilization of vimentin microfilaments by HSP27 may be one of the ways it inhibits apoptosis. Interestingly, our previous studies have shown that patients with CAV make antibodies reactive with vimentin, suggesting exposure of vimentin in the transplant setting.32 HSP27 has been shown to increase in vascular smooth muscle cells when subjected to hemodynamic stress,12 so it may play a part in preventing smooth muscle cell proliferation. To further explore the role of phosphorylated HSP27, laser microdissection will be used, microdissecting blood vessels from cardiac biopsies,8 to enrich for proteins from blood vessels.
A potential weakness of this study was the use of angiography to diagnose CAV; angiography is not as sensitive as intravascular ultrasound and it is possible that our patients, said to be disease-free, do indeed have some degree of intimal thickening. On the other hand, the study has used the same technique to investigate all patients, we are therefore comparing like with like. The second weakness is the relatively small numbers of late biopsies (n=22) that could be studied; the 10-year survival of patients transplanted at this center from 1990 to 1993 was 53.9%; obviously this is a small and unusual group of patients. The third weakness is that the criteria for selection of patients was availability of early biopsies, consent for late biopsies, and the possibility that these criteria could introduce bias into the results. Availability means biopsies were available unless they had been used for other research projects. The consent rate for late biopsies (at the time of angiography) was 75%. It is impossible to know whether these factors could influence the present results. This study raises the issue of the dynamics of expression of HSP27 in relation to disease activity. It is not clear whether HSP27 is constantly being induced after transplantation as a result of chronic attack by the immune system (effect of circulating antibodies and cytokines) or whether levels are normally low but are induced as a result of an acute immune attack by recipient T lymphocytes. It may be that early expression of HSP27 is important for late protection, indeed these studies demonstrated there was more HSP27 in the early biopsies of patients without CAV than patients with disease, but this difference was not statistically significant. Further work needs to be done on early biopsies. Further work also needs to be done to understand basal level of expression of HSP27 in different sized blood vessels from normal and transplanted hearts, the relative significance of smooth muscle cell and endothelial cell expression; the relationship to various insults associated with the process of transplantation.
In conclusion, using a proteomic approach a protein has been identified which is associated with freedom from CAV after cardiac transplantation. This protein has been identified as a phosphorylated form of HSP27 and is thus an after translational modification. As an aside, using a genomic approach, such as microarrays, this change would not have been identified. The significance of this finding is not limited to transplant-induced graft vasculopathy; the vascular response to stress includes responses to diverse factors such as hemodynamic forces, hypertension, oxidized LDL, glycosylated proteins, infections, and the alloimmune response. The current results taken together with results of Martin-Ventura et al14 suggest a fundamental role of HSP27, possibly in conjunction with smooth muscle cells, in protecting against atherosclerosis.
Acknowledgments
This work was supported by a grant from the British Heart Foundation (RG/2001005). M.J.D. is the recipient of a Science Foundation Ireland Research Professorship. The help of Drs Manuel Mayr (St. George’s Hospital Medical School, London, UK), Emma McGregor (Proteome Sciences Plc, London, UK), and Elizabeth Sage (Kings College, London, UK) are gratefully acknowledged.
References
Taylor DO, Edwards LB, Boucek MM, Trulock EP, Keck BM, Hertz MI. The registry of the international society for heart and lung transplantation: Twenty-first official adult heart transplant report eC2004. J Heart Lung Transplant. 2004; 23: 796eC803.
Schoeder J, Gao S. Accelerated graft coronary artery disease in heart-transplant recipients. Coron Artery Dis. 1995; 6: 226eC233.
Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature. 1993; 362: 801eC809.
Waller J, Brook NR, Nicholson ML. Cardiac allograft vasculopathy: current concepts and treatment. Transpl Int. 2003; 16: 367eC375.
Bach FH, Ferran C, Hechenleitner P, Mark W, Koyamada N, Miyatake T, Winkler H, Badrichani A, Candinas D, Hancock WW. Accommodation of vascularized xenografts: expression of "protective genes" by donor endothelial cells in a host Th2 cytokine environment. Nat Med. 1997; 3: 196eC204.
Soares MP, Anrather LJ, Takigami CK, Sato K, Grey ST, Colvin RB, Choi AM, Poss KD, Bach FH. Expression of heme oxygenase-1 can determine cardiac xenograft survival. Nat Med. 1998; 4: 1073eC1077.
Hancock WW, Buelow R, Sayegh MH, Turk LA. Antibody-induced transplant arteriosclerosis is prevented by graft expression of anti-oxidant and anti-apoptotic genes. Nat Med. 1998; 4: 1392eC1396.
De Souza AI, McGregor E, Dunn MJ, Rose ML. Preparation of human heart for laser microdissection and proteomics. Proteomics. 2004; 4: 578eC586.
McGregor E, Kempster L, Wait R, Welson SY, Gosling M, Dunn MJ, Powell JT. Identification and mapping of human saphenous vein medial smooth muscle proteins by two-dimensional polyacrylamide gel electrophoresis. Proteomics. 2001; 1: 1405eC1414.
Steinberg TH, Agnew BJ, Gee KR, Leung W-Y, Goodman T, Schulenberg B, Hendrikson J, Beechem JM, Patton WF. Global quantitative phosphoprotein analysis using Multiplexed Proteomics technology. Proteomics. 2003; 3: 1128eC1144.
Taylor PM, Rose ML, Yacoub MH. Coronary artery immunogenicity: a comparison between explanted recipient or donor hearts and transplanted hearts. Transplant Immunol. 1993; 1: 294eC301.
McGregor E, Kempster L, Wait R, Gosling M, Dunn M, Powell J. F-actin capping (CapZ) and other contractile saphenous vein smooth muscle proteins are altered by hemodynamic stress: a proteonomic approach. Mol Cell Proteomics. 2004; 3: 115eC124.
Schimke L, Lutsch G, Schernes U, Kruse I, Dubel H-P, Pregla R, Hummel M, Meyer R, Stahl J. Increased level of HSP27 but not of HSP72 in human heart allografts in relation to acute rejection. Transplantation. 2000; 27: 1694eC1697.
Martin-Ventura JL, Duran MC, Blanco-Colio LM, Meilhac O, Leclercq A, Michel JB, Jensen ON, Hermandex-Merida S, Tunon J, Vivanco F, Egido J. Identification by a Differential Proteomic Approach of Heat Shock Protein 27 as a Potential Marker of Atherosclerosis. Circulation. 2004; 110: 2216eC2219.
Concannon CG, Gorman AM, Samali A. On the role of Hsp27 in regulating apoptosis. Apoptosis. 2003; 8: 61eC70.
Garrido C, Gurbuxani S, Ravagnan L, Kroemer G. Heat shock proteins: endogenous modulators of apoptotic cell death. Biochem Biophys Res Commun. 2001; 286: 433eC442.
Mehlen P, Schulze-Osthoff K, Arrigo AP. Small stress proteins as novel regulators of apoptosis. Heat shock protein 27 blocks Fas/APO-1- and staurosporine-induced cell death. J Biol Chem. 1996; 271: 16510eC16514.
Mosser DD, Martin LH. Induced thermotolerance to apoptosis in a human T lymphocyte cell line. J Cell Physiol. 1992; 151: 561eC570.
Samali A, Cotter TG. Heat shock proteins increase resistance to apoptosis. Exp Cell Res. 1996; 223: 163eC170.
Merendino AM, Paul C, Vignola AM, Costa MA, Melis M, Chiappara G, Izzo V, Bousquet J, Arrigo AP. Heat shock protein-27 protects human bronchial epithelial cells against oxidative stress-mediated apoptosis: possible implication in asthma. Cell Stress Chaperones. 2002; 7: 269eC280.
Guay J, Lambert H, Gingras-Breton G, Lavoie JN, Huot J, Landry J. Regulation of actin filament dynamics by p38 map kinase-mediated phosphorylation of heat shock protein 27. J Cell Sci. 1997; 110: 357eC368.
Lavoie JN, Hickey E, Weber LA, Landry J. Modulation of actin microfilament dynamics and fluid phase pinocytosis by phosphorylation of heat shock protein 27. J Biol Chem. 1993; 268: 24210eC24214.
Huot J, Lambert H, Lavoie JN, Guimond A, Houle F, Landry J. Characterization of 45-kDa/54-kDa HSP27 kinase, a stress-sensitive kinase which may activate the phosphorylation-dependent protective function of mammalian 27-kDa heat-shock protein HSP27. Eur J Biochem. 1995; 227: 416eC427.
Efthymiou CA, Mocanu MM, de Belleroche J, Wells DJ, Latchman DS, Yellon DM. Heat shock protein 27 protects the heart against myocardial infarction. Basic Res Cardiol. 2004; 99: 392eC394.
Akbar MT, Lundberg AMC, Liu K, Vidyadaran S, Wells KE, Dolatshad H, Wynn S, Wells DJ, Latchman DS, de Belleroche J. The neuroprotective effects of heat shock protein 27 overexpression in transgenic animals against kainate-induced seizures and hippocampal cell death. J Biol Chem. 2003; 278: 19956eC19965.
Miller LW, Granville DJ, Narula J, McManus BM. Apoptosis in cardiac transplant rejection. Cardiol Clin. 2001; 19: 141eC154.
Xu B, Sakkas LI, Slachta CA, Goldman BI, Jeevanandam V, Oleszak EL, Platsoucas CD. Apoptosis in chronic rejection of human cardiac allografts. Transplantation. 2001; 71: 1137eC1146.
Coulombe PA, Wong P. Cytoplasmic intermediate filaments revealed as dynamic and multipurpose scaffolds. Nat Med. 2004; 6: 699eC706.
Perng M, Cairns L, van den IJssel P, Prescott A, Hutcheson AM, Quinlan RA. Intermediate filament interactions can be altered by HSP27 and alphaB-crystallin. J Cell Sci. 1999; 112: 2099eC2112.
Salvador-Silva M, Ricard CS, Agapova OA, Yang P, Hernandez MR. Expression of small heat shock proteins and intermediate filaments in the human optic nerve head astrocytes exposed to elevated hydrostatic pressure in vitro. J Neurosci Res. 2001; 66: 59eC73.
Morishima N. Changes in nuclear morphology during apoptosis correlate with vimentin cleavage by different caspases located either upstream or downstream of Bcl-2 action. Genes Cells. 1999; 4: 401eC414.
Jurcevic S, Ainsworth ME, Pomerance A, Smith JD, Robinson DR, Dunn MJ, Yacoub MH, Rose ML. Antivimentin antibodies are an independent predictor of transplant-associated coronary artery disease after cardiac transplantation. Transplantation. 2001; 71: 886eC892.(A.I. De Souza, R. Wait, A)
Kennedy Institute of Rheumatology Division (R.W.), Faculty of Medicine, Imperial College London, Hammersmith, London, UK
the Department of Cardiology (A.G.M., N.R.B.), Royal Brompton and Harefield NHS Trust, Harefield Hospital, Harefield, Middlesex, UK
the Proteome Research Centre (M.J.D.), Conway Institute of Biomolecular and Biomedical Research, University College Dublin, Belfield, Ireland.
Abstract
Experimental studies have suggested that protective genes protect allografts from cardiac allograft vasculopathy (CAV), the major complication after cardiac transplantation. Here we have sought to confirm this hypothesis using long-term heart transplant recipients. Twenty-two patients that were 9 years or older after transplant were investigated; 11 of these were without angiographic evidence of CAV; 11 had developed early CAV at 1 to 3 years after transplant. To identify proteins that may act as protectors from CAV, a global proteomic approach was used comparing cardiac biopsies from 12 patients taken within the first 2 weeks after transplant and those taken after 9 years from the same patient. Proteins were separated by 2-D gel-electrophoresis, detected by silver staining, and analyzed using Progenesis software. A particular protein spot was found in 4/6 biopsies from patients without CAV, but absent from 5/6 biopsies from those with CAV (P=0.24); however, quantitative analysis of spot intensity showed a significant difference (0.061±0.05 versus 0.003±0.01, P=0.04). This spot was identified by mass spectrometry and a combination of techniques as a diphosphorylated form of HSP27. Immunohistochemistry of further biopsies not only validated that HSP27 was more abundantly expressed on biopsies without CAV but also showed it to be localized to blood vessels. In contrast, vessels from patients with CAV did not express HSP27 (P=0.028x10eC4). Immunohistochemistry of 12 further early biopsies and nontransplanted heart showed HSP27 to be present in normal blood vessels. These findings suggest that expression of a specific diphosphorylated form of HSP27 is associated with healthy blood vessels; it appears to be lost from vessels of patients with graft vasculopathy.
Key Words: cardiac allograft vasculopathy heat shock protein 27 protection proteomics and transplantation
Introduction
Accelerated graft atherosclerosis or transplant associated coronary artery disease, often referred to as chronic rejection or cardiac allograft vasculopathy (CAV), is one of the most serious long-term complications of heart transplantation; the incidence increases every year after transplant and is currently 43% at 7 years.1 Histologically, CAV consists of a diffuse concentric intimal proliferation comprised of smooth muscle cells and fibroblasts. The disease is not confined to the large epicardial arteries, but also affects the small vessels penetrating the myocardium.2
The endothelial response to injury hypothesis proposed by Ross in 19933 to explain nontransplant atherosclerosis is equally applicable to transplant atherosclerosis. Essentially, the hypothesis proposes that an initial insult results in endothelial cell activation and upregulation of cytokines, chemokines, and adhesion molecules. The latter events allow transendothelial migration of monocytes which may become further activated in the vessel wall by modified low-density lipoproteins, and subsequently release cytokines and growth factors causing smooth muscle cell proliferation. There are a number of possible insults to the endothelium that could initiate such a cascade of events culminating in CAV; these include brain stem death, preservation injury, reperfusion injury, and the alloimmune attack after transplantation.4
In recent years there has emerged the important concept of cytoprotection. It is known that the endothelial response to injury can be protective as well as inflammatory. This may explain why some patients remain free of CAV for many years after their transplant whereas others succumb at a very early stage. For example, upregulation of HO-1, A20, and Bcl-xL genes have been described as protecting against antibody-mediated graft vasculopathy in hamster/mouse to rat cardiac xenografts models5,6 and in a mouse allograft model.7
The aim of this study was to use a global proteomic approach to investigate the hypothesis that long-term survivors of heart transplantation, without signs of CAV, show preferential expression of potential protective proteins. Two-D gel electrophoresis has been used to compare quantitative protein expression by cardiac biopsies of transplant patients with and without angiographic evidence of cardiac allograft vasculopathy.
Materials and Methods
Patients
This is a retrospective study of patients transplanted at this center between 1990 and 1993. The aim of the study was to compare biopsies from long-term survivors (9 years) who were free of CAV to long-term survivors who had developed CAV within the first 3 years of their transplant. Patients at this center receive routine surveillance endomyocardial biopsy procedures at weekly intervals for the first month after transplant; in that era (1990eC1993), in all cases, extra pieces of biopsy were taken, they were snap frozen in liquid nitrogen, and stored in liquid nitrogen for future research. The study population consisted of those patients where biopsy material was available. Surveillance angiography for CAV is performed routinely at 1 to 2 year intervals, CAV detected by angiography was defined as any new irregularity or stenosis not present in patients’ previous angiograms 25% luminal diameter. At the time of angiography, patients do not normally undergo the biopsy procedure. Thus, patients were selected if they were >9 years after transplant, had early biopsies available (taken within 21 days of their transplants), and if they consented to an additional biopsy procedure as part of their annual angiography assessment, many years after transplantation. Twenty-two heart transplant recipients (all male) fulfilled these requirements, and their ages ranged from 17 to 68 years. Indications for transplant were ischemic heart disease (n=14), dilated cardiomyopathy (n=7), and congenital heart disease (n=1). Eleven patients had developed early CAV (25% to 50% stenosis, first detected 1 to 3 years); their late biopsies were taken at 133.1±10.7 months (mean±SEM); early biopsies from 6 of these patients were available and obtained at 17.6±3.3 days. Eleven patients had not developed CAV at the time of biopsy (126.5±5.7 months); early biopsies from 6 of these patients had been taken at 15.5±3.3 days. Two samples of endomyocardial biopsy were taken at the time of angiography for this study; 1 for proteomics and 1 for immunohistochemistry. Right ventricular endomyocardial biopsies were embedded in OCT compound, frozen in isopentane, and then stored in liquid nitrogen until required. In addition to the biopsies collected specifically for this study, 12 early biopsies (taken within 3 months of transplant) and 9 pieces of explanted heart were used for immunohistochemistry. Local ethical permission (Royal Brompton and Harefield Research Ethics Committee) was obtained in accordance with the Helsinki Declaration of 1975 and written informed consent obtained from each patient.
Sample Preparation and Quantification
Total tissue proteins were extracted as previously described8 from 24 biopsies (n=6 patients with CAV and 6 without CAV, 1 early and 1 late biopsy from each patient). The resulting supernatant was harvested and protein concentration determined using the 2D-Quant kit (Amersham Biosciences). The amount of total protein extracted from each biopsy was 256±21 e (mean±SEM), ranging from 143 to 453 e. Solubilized protein samples were stored at eC80°C until required.
Proteomic Analysis
Protein extracts (100 e) of right ventricular endomyocardial biopsies (n =24) were separated by 2-D gel electrophoresis (2DE) as described by De Souza et al.8 Briefly, the first dimension was nonlinear pH 3 to 10 IPG isoelectric focusing, and the second dimension was 12% SDS-PAGE. Gels were silver-stained and computer-assisted image analysis was performed as described previously.9 Protein spot matching and intensities were assessed using the Progenesis Workstation, Version 2003.02 (Nonlinear Dynamics). Four comparisons were made: (1) early versus late without CAV; (2) early versus late with CAV; (3) early versus early; (4) late versus late. All results are expressed as arithmetic means (±SD). Paired and unpaired Student t tests were performed, with P<0.05 required for statistical significance. The first 2 comparisons were intrapatient comparisons (paired t tests); the second 2 were interpatient comparisons (unpaired t tests). Protein spots showing a statistically significant difference in intensity (changed in expression by 1.5-fold or more, P<0.05) were manually excised from micropreparative 2D gels (400 e protein) for identification by tandem mass spectrometry. A detailed methodology is provided in the online data supplement available at http://circres.ahajournals.org.
Phosphoprotein Identification
Phosphoprotein identification was examined on micropreparative (400 e protein) 2-D gels of left ventricular proteins (n=2). Total protein extracts were separated on pH 3 to 10 NL IPG strips, followed by a 12% SDS polyacrylamide gel. Pro-Q Diamond (Molecular Probes) phosphoprotein gel stain provides a method for selectively staining phosphoproteins in polyacrylamide gels,10 and was used according to the manufacturer’s instruction and scanned on the Typhoon 9400 scanner (Amersham Biosciences) at 532 nm, emission filter at 560 LP, and the photomultiplier tube set at 600 V. Subsequently, Pro-Q diamond stained gels were washed in MQ water and then fixed overnight in a methanol: acetic acid: water solution (4:1:5 v/v/v), before staining with a mass spectrometry compatible silver stain.
Western Blottting
Immediately after 2DE, the separated proteins were transferred by semidry blotting onto nitrocellulose membranes for 2 hours at 0.8 mA per cm2. HSP27 proteins were identified by immunostaining with a monoclonal anti-HSP27 antibody (SPA-800, Stressgen) and polyclonal antibodies specific for phosphorylated HSP27 at Serine residues 78 and 82 (SPA-523 and SPA-524 respectively, both from Stressgen).
Immunohistochemistry
Frozen sections (5 e) of explanted cardiac tissue (n=9), early biopsies (taken within 3 months of transplant, n=12), and late biopsies from patients with and without CAV (n=11 for each group) were stained using the immunoperoxidase method as described by Taylor et al.11 Briefly, tissue sections were stained with PBS as a negative control, and monoclonal antibodies to CD31, a marker of endothelial cells (DakoCytomation), smooth muscle -actin (Sigma), or HSP27 (SPA-800, Stressgen).
Results
Progenesis analysis of protein spots separated and displayed by 2DE revealed that 69 proteins showed time-dependent changes in abundance in comparison 1 (early versus late without CAV), and 38 proteins changed in biopsies from comparison 2 (early versus late with CAV). Twenty-three proteins changed in comparison 3 (early CAV versus early non-CAV), and 21 in comparison 4 (late CAV versus late non-CAV). Proteins differences in the 4 comparisons outlined above changed by 1.5-fold or more, P<0.05. The proteins of interest have been identified by tandem mass spectrometry. In summary in comparison 1, 37 of the 69 proteins have been identified, whereas 14 of 38, 14 of 23, and 16 of 21 have been identified from comparisons 2, 3, and 4 respectively (see online Figure I and online Table I).
HSP27 Is Associated With Freedom From CAV
Figure 1 shows representative 2D images from both groups of patients. The spots indicated with asterisks were identified as HSP27 by mass spectrometry, but no difference was observed in their expression between the groups. However, an additional spot in this region (Figure 1, spot 3306, arrowed), was significantly differentially expressed between the late groups. Eight tryptic peptides from this spot were sequenced by mass spectrometry, all of which mapped onto the sequence of HSP27. Figure 2A shows a representative example of one of these spectra. This protein was present in the 2-D patterns obtained from 4/6 late biopsies from subjects without disease, but was not detected in 5/6 late biopsies from patients with CAV (P=0.24, Fisher’s exact test). Quantitative analysis demonstrated a 20-fold increase in spot intensity in biopsies free of CAV compared with those with CAV (spot intensities were 0.061±0.05 versus 0.003±0.01, P=0.04). The trend was similar in the early groups (spot intensities 0.47±0.38 versus 0.11±0.14; without CAV versus with CAV), but were not statistically significant (P=0.10).
Phosphorylation Status of HSP27 Protein Spot 3306
The most basic form of HSP27 (spot 2079, highlighted on Figure 1) was observed at a pI of 5.9, which is in good agreement with previously published values for the unphosphorylated protein,12 and is consistent with the Figure of 5.98 calculated from the amino acid sequence of HSP27. The observed pI of spot 3306 however was significantly lower, at 5.42, which would be consistent with the presence of acidic modifications such as phosphorylation. Support for this hypothesis was provided by staining with Pro-Q Diamond, which is specific for phosphoproteins. Figure 3A through 3D shows overlays of the same gel, visualized with either silver or with Pro-Q Diamond, and indicates that both stains colocalize to spot 3306.
Tandem mass spectrometry provided unequivocal evidence for phosphorylation of Serine 82. In unphosphorylated HSP27 (spot 2079) a doubly charged ion was observed at m/z 538.3, which was shown by collisional fragmentation and MS/MS to correspond to the sequence QLSSGVSEIR, representing residues 80 to 89 of HSP27 (Figure 2B). This peptide is clearly unphosphorylated, because the mass difference between the C-terminaleCcontaining fragments y8 and y7, y7 and y6, and y4 and y3, is 87, corresponding to the residue mass of serine. By contrast, in the tryptic digest of spot 3306, a doubly charged ion was observed at m/z 578.3, which would be consistent with the addition of a single phosphate group to the peptide QLSSGVSEIR. Phosphorylation is further indicated by a facile losses of H3PO4 (mass 98) from the protonated molecule (Figure 2C). The fragmentation pattern establishes Serine 82 as the site of phosphorylation, because the mass difference between the y8 (m/z 914) and y7 (m/z 747) fragments is 167 rather than 87, and there is a loss of dehydroalanine (69) between the dephosphorylated y8 ion (m/z 816) and y7 (m/z 747). The other 2 serines in the sequence are unphosphorylated, because the residue mass between the respective contiguous y ions is 87 (Figure 2C). The observed pI of spot 3306 (5.42) is consistent with diphosphorylated HSP27, both from calculation (predicted value 5.6) and by agreement with published data in which bis-phospho HSP27 was observed to migrate at a pI of 5.45.12 There are 3 experimentally verified phosphorylation sites in human HSP27, Serine 15, 78, and 82. The present data suggest that Serine 15 is not phosphorylated, because we observed a doubly charged ion at m/z 481.2, which was shown by MS/MS (Figure 2D) to correspond to unphosphorylated GPSWDPFR (residues 13 to 20), whereas the phosphorylated form of this peptide (m/z 521.2) was undetectable. It was unfortunately impossible to verify the phosphorylation state of Serine 78, as the tryptic peptide which includes it (ALSR) is too small to be easily detected in a conventional ESI LC MS/MS experiment.
Further confirmation was provided by Western blotting with antibodies specific for the phosphorylation sites Ser-78 and Ser-82 and with an antibody that recognizes all forms of HSP27 whether phosphorylated or unphosphorylated. These results (Figure 3E through 3G; spot 3306 highlighted) confirm the presence of phosphorylation on Serine residues 78 and 82.
Localization of HSP27 by Immunohistochemistry
Having found by 2DE that HSP27 is expressed in late biopsies from patients who are angiographically free of CAV, but is absent from late biopsies form those patients who develop CAV, immunohistochemistry was used to validate this finding in paired late cardiac biopsies (n=12) and 10 further late biopsies from patients in this study. HSP27 was found expressed diffusely on all myocytes, and within endothelial and smooth muscle cells of medium to large blood vessels in all 11 biopsies examined from patients without disease. Positive staining of the endothelial and smooth muscle cells of blood vessels was confirmed by staining serial sections with CD31 and SMA (see Materials and Methods for details). In contrast, expression of HSP27 on all biopsies from patients with CAV (n=11) was confined to myocytes only, and absent from the endothelial and smooth muscle cells (Figure 4). Using the Fisher exact test, a P value of 0.028x10eC4 was calculated for the presence/absence of HSP27. Immunohistochemistry thus supported and extended the proteomic results, by showing less expression of HSP27 on blood vessels in late biopsies from patients with CAV. Early biopsies from this group of patients were not available for immunohistochemistry, but it was possible to examine 12 early biopsies from a different group of patients transplanted within the same era. All of these biopsies showed HSP27 to be expressed on all myocytes and blood vessels. In addition, to examine normal heart, 9 pieces of explanted heart were investigated and all showed diffuse expression of HSP27 on cardiac myocytes and blood vessels.
Discussion
In this study a global proteomics approach has been used to examine protein expression in endomyocardial biopsies from human transplant patients with and without angiographic evidence of cardiac allograft vasculopathy. We postulated that differential expression of protective proteins could happen at early or late times after transplantation, and it would therefore be ideal to be able to examine biopsies from both time periods. At this center, it has been possible to retrospectively identify patients who were at least 9 years after transplant for whom early biopsies were available for research. In total, 141 proteins were significantly altered (by 1.5-fold or more) between the groups. Of these, 81 were identified by mass spectrometry (see online Table I). The most significant finding was of a particular protein spot, found only in gel preparations from late biopsies from patients without CAV; this was identified by mass spectrometry to be a specific form of HSP27. The position of the HSP27 protein spot and the pI shift from the theoretical value suggested that this protein was phosphorylated. This was confirmed using a stain specific for phosphoproteins (Pro-Q Diamond), mass spectrometry, and Western blotting, and it was demonstrated to be diphosphorylated at Serine residues 78 and 82. We have focused on this particular protein because it showed the greatest fold change, namely 20-fold increase in expression in biopsies from patients without disease (spot intensity 0.061±0.05) than in biopsies from patients with disease (0.003±0.01). This difference was statistically significant (P=0.04). It should be noted that when comparing biopsies for presence or absence of HSP27, the difference between the groups was not significantly different (P=0.24, Fisher exact test). The discrepant results reflect the skewed distribution of HSP27, being highly expressed in 4/6 biopsies without CAV and below detection in 2 biopsies without CAV. A possible explanation why only 4/6 biopsies without CAV were positive may be caused by a relative lack of blood vessels in the 2 negative biopsies. Thus immunohistochemistry of further late biopsies not only confirmed that HSP27 was more strongly expressed in late biopsies which were free of CAV, but it showed vascular localization, something not found in late biopsies from patients with CAV (P=0.028x10eC4). Taken together, these results strongly suggest vascular expression of HSP27 is associated with freedom from CAV. It has been previously reported that HSP27 is diffusely expressed by cardiac myocytes of transplanted hearts,13 but blood vessels were not investigated. Little is known about expression of HSP27 in normal and diseased human blood vessels. Martin-Ventura et al14 have recently reported that atherosclerotic vessels maintained in culture release very little HSP27 protein into the medium compared with control vessels; they also showed smooth muscle cells of mammary arteries to be positive for HSP27. These findings therefore support our findings of vascular expression of HSP27 in biopsies from patients that did not develop CAV. In an attempt to investigate nondiseased cardiac blood vessels, early biopsies from another group of 12 transplant patients and 9 pieces of explanted heart were examined; all showed HSP27 positive blood vessels. The current results together with those of Martin-Ventura et al14 suggest that HSP27 is expressed normally by blood vessels, and lack of expression is associated with both nontransplant and transplant atherosclerosis. Endothelial cells and smooth muscle cells of all medium-large size vessels were positive for HSP27 in early biopsies and explanted heart. Interestingly, staining with antibody to CD31, a marker of endothelial cells, demonstrated many tiny capillary endothelial cells which appeared not to be HSP27 positive. The functional significance of HSP27 may therefore be caused more by its association with smooth muscle cells rather than endothelial cells.
HSP27 is a small heat-shock protein, the expression of which correlates with increased survival following cytotoxic stimuli.15 Like all heat-shock proteins, it functions as a molecular chaperone in protein biosynthesis to facilitate protein folding and prevent accumulation of misfolded proteins in the cell. Other means by which it is cytoprotective include protection from apoptosis,16eC19 modulation of oxidative stress,20 and regulation of the cytoskeleton.21 Several studies have shown that overexpression of HSP27 increases the stability of F-actin microfilaments during exposure to stress such as hyperthermia,22 oxidants,23 and cytochalsin D.21 This activity may depend on the degree of phosphorylation and its structural organization. Use of transgenic mice overexpressing human HSP27 has demonstrated that hearts from these mice are protected from ischemia reperfusion injury24 and nerves are protected from chemically induce apoptosis.25
The most likely explanation for the protective effect of HSP27 on pathology of blood vessels is its antiapoptopic properties16 and ability to stabilize the actin cytoskeleton. Apoptosis has been found in transplanted hearts at all times after transplantation,26 and has been shown to be involved in the pathogenesis of chronic rejection in cardiac allografts.27 As well as binding to actin, HSP27 also binds to the intermediate filament vimentin.28eC30 Caspase dependent cleavage of vimentin is said to be a prerequisite for apoptosis in all cell types,31 suggesting that stabilization of vimentin microfilaments by HSP27 may be one of the ways it inhibits apoptosis. Interestingly, our previous studies have shown that patients with CAV make antibodies reactive with vimentin, suggesting exposure of vimentin in the transplant setting.32 HSP27 has been shown to increase in vascular smooth muscle cells when subjected to hemodynamic stress,12 so it may play a part in preventing smooth muscle cell proliferation. To further explore the role of phosphorylated HSP27, laser microdissection will be used, microdissecting blood vessels from cardiac biopsies,8 to enrich for proteins from blood vessels.
A potential weakness of this study was the use of angiography to diagnose CAV; angiography is not as sensitive as intravascular ultrasound and it is possible that our patients, said to be disease-free, do indeed have some degree of intimal thickening. On the other hand, the study has used the same technique to investigate all patients, we are therefore comparing like with like. The second weakness is the relatively small numbers of late biopsies (n=22) that could be studied; the 10-year survival of patients transplanted at this center from 1990 to 1993 was 53.9%; obviously this is a small and unusual group of patients. The third weakness is that the criteria for selection of patients was availability of early biopsies, consent for late biopsies, and the possibility that these criteria could introduce bias into the results. Availability means biopsies were available unless they had been used for other research projects. The consent rate for late biopsies (at the time of angiography) was 75%. It is impossible to know whether these factors could influence the present results. This study raises the issue of the dynamics of expression of HSP27 in relation to disease activity. It is not clear whether HSP27 is constantly being induced after transplantation as a result of chronic attack by the immune system (effect of circulating antibodies and cytokines) or whether levels are normally low but are induced as a result of an acute immune attack by recipient T lymphocytes. It may be that early expression of HSP27 is important for late protection, indeed these studies demonstrated there was more HSP27 in the early biopsies of patients without CAV than patients with disease, but this difference was not statistically significant. Further work needs to be done on early biopsies. Further work also needs to be done to understand basal level of expression of HSP27 in different sized blood vessels from normal and transplanted hearts, the relative significance of smooth muscle cell and endothelial cell expression; the relationship to various insults associated with the process of transplantation.
In conclusion, using a proteomic approach a protein has been identified which is associated with freedom from CAV after cardiac transplantation. This protein has been identified as a phosphorylated form of HSP27 and is thus an after translational modification. As an aside, using a genomic approach, such as microarrays, this change would not have been identified. The significance of this finding is not limited to transplant-induced graft vasculopathy; the vascular response to stress includes responses to diverse factors such as hemodynamic forces, hypertension, oxidized LDL, glycosylated proteins, infections, and the alloimmune response. The current results taken together with results of Martin-Ventura et al14 suggest a fundamental role of HSP27, possibly in conjunction with smooth muscle cells, in protecting against atherosclerosis.
Acknowledgments
This work was supported by a grant from the British Heart Foundation (RG/2001005). M.J.D. is the recipient of a Science Foundation Ireland Research Professorship. The help of Drs Manuel Mayr (St. George’s Hospital Medical School, London, UK), Emma McGregor (Proteome Sciences Plc, London, UK), and Elizabeth Sage (Kings College, London, UK) are gratefully acknowledged.
References
Taylor DO, Edwards LB, Boucek MM, Trulock EP, Keck BM, Hertz MI. The registry of the international society for heart and lung transplantation: Twenty-first official adult heart transplant report eC2004. J Heart Lung Transplant. 2004; 23: 796eC803.
Schoeder J, Gao S. Accelerated graft coronary artery disease in heart-transplant recipients. Coron Artery Dis. 1995; 6: 226eC233.
Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature. 1993; 362: 801eC809.
Waller J, Brook NR, Nicholson ML. Cardiac allograft vasculopathy: current concepts and treatment. Transpl Int. 2003; 16: 367eC375.
Bach FH, Ferran C, Hechenleitner P, Mark W, Koyamada N, Miyatake T, Winkler H, Badrichani A, Candinas D, Hancock WW. Accommodation of vascularized xenografts: expression of "protective genes" by donor endothelial cells in a host Th2 cytokine environment. Nat Med. 1997; 3: 196eC204.
Soares MP, Anrather LJ, Takigami CK, Sato K, Grey ST, Colvin RB, Choi AM, Poss KD, Bach FH. Expression of heme oxygenase-1 can determine cardiac xenograft survival. Nat Med. 1998; 4: 1073eC1077.
Hancock WW, Buelow R, Sayegh MH, Turk LA. Antibody-induced transplant arteriosclerosis is prevented by graft expression of anti-oxidant and anti-apoptotic genes. Nat Med. 1998; 4: 1392eC1396.
De Souza AI, McGregor E, Dunn MJ, Rose ML. Preparation of human heart for laser microdissection and proteomics. Proteomics. 2004; 4: 578eC586.
McGregor E, Kempster L, Wait R, Welson SY, Gosling M, Dunn MJ, Powell JT. Identification and mapping of human saphenous vein medial smooth muscle proteins by two-dimensional polyacrylamide gel electrophoresis. Proteomics. 2001; 1: 1405eC1414.
Steinberg TH, Agnew BJ, Gee KR, Leung W-Y, Goodman T, Schulenberg B, Hendrikson J, Beechem JM, Patton WF. Global quantitative phosphoprotein analysis using Multiplexed Proteomics technology. Proteomics. 2003; 3: 1128eC1144.
Taylor PM, Rose ML, Yacoub MH. Coronary artery immunogenicity: a comparison between explanted recipient or donor hearts and transplanted hearts. Transplant Immunol. 1993; 1: 294eC301.
McGregor E, Kempster L, Wait R, Gosling M, Dunn M, Powell J. F-actin capping (CapZ) and other contractile saphenous vein smooth muscle proteins are altered by hemodynamic stress: a proteonomic approach. Mol Cell Proteomics. 2004; 3: 115eC124.
Schimke L, Lutsch G, Schernes U, Kruse I, Dubel H-P, Pregla R, Hummel M, Meyer R, Stahl J. Increased level of HSP27 but not of HSP72 in human heart allografts in relation to acute rejection. Transplantation. 2000; 27: 1694eC1697.
Martin-Ventura JL, Duran MC, Blanco-Colio LM, Meilhac O, Leclercq A, Michel JB, Jensen ON, Hermandex-Merida S, Tunon J, Vivanco F, Egido J. Identification by a Differential Proteomic Approach of Heat Shock Protein 27 as a Potential Marker of Atherosclerosis. Circulation. 2004; 110: 2216eC2219.
Concannon CG, Gorman AM, Samali A. On the role of Hsp27 in regulating apoptosis. Apoptosis. 2003; 8: 61eC70.
Garrido C, Gurbuxani S, Ravagnan L, Kroemer G. Heat shock proteins: endogenous modulators of apoptotic cell death. Biochem Biophys Res Commun. 2001; 286: 433eC442.
Mehlen P, Schulze-Osthoff K, Arrigo AP. Small stress proteins as novel regulators of apoptosis. Heat shock protein 27 blocks Fas/APO-1- and staurosporine-induced cell death. J Biol Chem. 1996; 271: 16510eC16514.
Mosser DD, Martin LH. Induced thermotolerance to apoptosis in a human T lymphocyte cell line. J Cell Physiol. 1992; 151: 561eC570.
Samali A, Cotter TG. Heat shock proteins increase resistance to apoptosis. Exp Cell Res. 1996; 223: 163eC170.
Merendino AM, Paul C, Vignola AM, Costa MA, Melis M, Chiappara G, Izzo V, Bousquet J, Arrigo AP. Heat shock protein-27 protects human bronchial epithelial cells against oxidative stress-mediated apoptosis: possible implication in asthma. Cell Stress Chaperones. 2002; 7: 269eC280.
Guay J, Lambert H, Gingras-Breton G, Lavoie JN, Huot J, Landry J. Regulation of actin filament dynamics by p38 map kinase-mediated phosphorylation of heat shock protein 27. J Cell Sci. 1997; 110: 357eC368.
Lavoie JN, Hickey E, Weber LA, Landry J. Modulation of actin microfilament dynamics and fluid phase pinocytosis by phosphorylation of heat shock protein 27. J Biol Chem. 1993; 268: 24210eC24214.
Huot J, Lambert H, Lavoie JN, Guimond A, Houle F, Landry J. Characterization of 45-kDa/54-kDa HSP27 kinase, a stress-sensitive kinase which may activate the phosphorylation-dependent protective function of mammalian 27-kDa heat-shock protein HSP27. Eur J Biochem. 1995; 227: 416eC427.
Efthymiou CA, Mocanu MM, de Belleroche J, Wells DJ, Latchman DS, Yellon DM. Heat shock protein 27 protects the heart against myocardial infarction. Basic Res Cardiol. 2004; 99: 392eC394.
Akbar MT, Lundberg AMC, Liu K, Vidyadaran S, Wells KE, Dolatshad H, Wynn S, Wells DJ, Latchman DS, de Belleroche J. The neuroprotective effects of heat shock protein 27 overexpression in transgenic animals against kainate-induced seizures and hippocampal cell death. J Biol Chem. 2003; 278: 19956eC19965.
Miller LW, Granville DJ, Narula J, McManus BM. Apoptosis in cardiac transplant rejection. Cardiol Clin. 2001; 19: 141eC154.
Xu B, Sakkas LI, Slachta CA, Goldman BI, Jeevanandam V, Oleszak EL, Platsoucas CD. Apoptosis in chronic rejection of human cardiac allografts. Transplantation. 2001; 71: 1137eC1146.
Coulombe PA, Wong P. Cytoplasmic intermediate filaments revealed as dynamic and multipurpose scaffolds. Nat Med. 2004; 6: 699eC706.
Perng M, Cairns L, van den IJssel P, Prescott A, Hutcheson AM, Quinlan RA. Intermediate filament interactions can be altered by HSP27 and alphaB-crystallin. J Cell Sci. 1999; 112: 2099eC2112.
Salvador-Silva M, Ricard CS, Agapova OA, Yang P, Hernandez MR. Expression of small heat shock proteins and intermediate filaments in the human optic nerve head astrocytes exposed to elevated hydrostatic pressure in vitro. J Neurosci Res. 2001; 66: 59eC73.
Morishima N. Changes in nuclear morphology during apoptosis correlate with vimentin cleavage by different caspases located either upstream or downstream of Bcl-2 action. Genes Cells. 1999; 4: 401eC414.
Jurcevic S, Ainsworth ME, Pomerance A, Smith JD, Robinson DR, Dunn MJ, Yacoub MH, Rose ML. Antivimentin antibodies are an independent predictor of transplant-associated coronary artery disease after cardiac transplantation. Transplantation. 2001; 71: 886eC892.(A.I. De Souza, R. Wait, A)