J C Metcalfe

University of Cambridge, Cambridge, ENG, United Kingdom

Are you J C Metcalfe?

Claim your profile

Publications (194)1222.21 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Consistent and independently replicated laboratory evidence to support a causative relationship between environmental exposure to extremely low-frequency electromagnetic fields (EMFs) at power line frequencies and the associated increase in risk of childhood leukemia has not been obtained. In particular, although gene expression responses have been reported in a wide variety of cells, none has emerged as robust, widely replicated effects. DNA microarrays facilitate comprehensive searches for changes in gene expression without a requirement to select candidate responsive genes. To determine if gene expression changes occur in white blood cells of volunteers exposed to an ELF-EMF, each of 17 pairs of male volunteers age 20-30 was subjected either to a 50 Hz EMF exposure of 62.0 ± 7.1 μT for 2 h or to a sham exposure (0.21 ± 0.05 μT) at the same time (11:00 a.m. to 13:00 p.m.). The alternative regime for each volunteer was repeated on the following day and the two-day sequence was repeated 6 days later, with the exception that a null exposure (0.085 ± 0.01 μT) replaced the sham exposure. Five blood samples (10 ml) were collected at 2 h intervals from 9:00 to 17:00 with five additional samples during the exposure and sham or null exposure periods on each study day. RNA samples were pooled for the same time on each study day for the group of 17 volunteers that were subjected to the ELF-EMF exposure/sham or null exposure sequence and were analyzed on Illumina microarrays. Time courses for 16 mammalian genes previously reported to be responsive to ELF-EMF exposure, including immediate early genes, stress response, cell proliferation and apoptotic genes were examined in detail. No genes or gene sets showed consistent response profiles to repeated ELF-EMF exposures. A stress response was detected as a transient increase in plasma cortisol at the onset of either exposure or sham exposure on the first study day. The cortisol response diminished progressively on subsequent exposures or sham exposures, and was attributable to mild stress associated with the experimental protocol.
    Radiation Research 08/2012; 178(3):138-49. · 2.70 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: TGF-β acts as a suppressor of primary tumor initiation but has been implicated as a promoter of the later malignant stages. Here associations with risk of invasive breast cancer are assessed for single-nucleotide polymorphisms (SNP) tagging 17 genes in the canonical TGF-β ALK5/SMADs 2&3 and ALK1/SMADs 1&5 signaling pathways: LTBP1, LTBP2, LTBP4, TGFB1, TGFB2, TGFB3, TGFBR1(ALK5), ALK1, TGFBR2, Endoglin, SMAD1, SMAD2, SMAD3, SMAD4, SMAD5, SMAD6, and SMAD7 [Approved Human Gene Nomenclature Committee gene names: ACVRL1 (for ALK1) and ENG (for Endoglin)]. Three-hundred-fifty-four tag SNPs (minor allele frequency > 0.05) were selected for genotyping in a staged study design using 6,703 cases and 6,840 controls from the Studies of Epidemiology and Risk Factors in Cancer Heredity (SEARCH) study. Significant associations were meta-analyzed with data from the NCI Polish Breast Cancer Study (PBCS; 1,966 cases and 2,347 controls) and published data from the Breast Cancer Association Consortium (BCAC). Associations of three SNPs, tagging TGFB1 (rs1982073), TGFBR1 (rs10512263), and TGFBR2 (rs4522809), were detected in SEARCH; however, associations became weaker in meta-analyses including data from PBCS and BCAC. Tumor subtype analyses indicated that the TGFB1 rs1982073 association may be confined to increased risk of developing progesterone receptor negative (PR(-)) tumors [1.18 (95% CI: 1.09-1.28), 4.1 × 10(-5) (P value for heterogeneity of ORs by PR status = 2.3 × 10(-4))]. There was no evidence for breast cancer risk associations with SNPs in the endothelial-specific pathway utilizing ALK1/SMADs 1&5 that promotes angiogenesis. Common variation in the TGF-β ALK5/SMADs 2&3 signaling pathway, which initiates signaling at the cell surface to inhibit cell proliferation, might be related to risk of specific tumor subtypes. The subtype specific associations require very large studies to be confirmed.
    Cancer Epidemiology Biomarkers &amp Prevention 06/2011; 20(6):1112-9. · 4.56 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Latent transforming growth factor-β binding protein-1 (LTBP1) has been implicated in the control of secretion, localization, and activation of TGFβ (transforming growth factor-β). We developed a quantitative reverse-transcriptase polymerase chain reaction (Q-RT-PCR) assay using an RNA internal standard to examine the expression of three alternatively spliced isoforms of LTBP1 (LTBP1Δ41, LTBP1Δ53, and LTBP1Δ55) in a variety of human tissues. The assays were also used to determine the expression of LTBP1L and LTBP1S isoforms and total LTBP1. The Q-RT-PCR assays were highly reproducible and showed that in most tissues LTBP1Δ55 and LTBP1L were minor components of LTBP1. The proportion of LTBP1Δ41 ranged from 2% of total LTBP1 mRNA in early coronary atherosclerotic lesions to 54% in advanced lesions.
    Biochemical Genetics 12/2010; 49(3-4):213-25. · 0.94 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Intimal hyperplasia and atherosclerosis are the Achilles' heels of vascular interventions. Many cytokines and growth factors have been shown to mediate these pathological processes. There are conflicting data concerning the expression of transforming growth factor-β1 (TGFβ1) antigen in human intimal hyperplasia and atherosclerotic lesions and conflicting views about whether TGFβ1 is pro- or anti-atherogenic. The presence of TGFβ1 is not sufficient to infer activation of its signaling pathway because TGFβ1 may be present in inactive complexes. A sensitive immuno-fluorescence assay (cyanine-3 tyramide signal amplification system) was used on human coronary artery and aorta sections with early or advanced stage lesions to detect TGFβ1, activin, Smad2-P, a marker of the activated TGFβ1/activin pathway and components of latent TGFβ complexes. All antigens were readily detected in the media and neointima of early stage lesions. The levels were either reduced or undetectable in the media of advanced lesions but were increased in the neointima in areas of high cell density. In marked contrast to activin, TGFβ1 and LAP1 expression levels were closely correlated with Smad2-P throughout the artery wall. Discrepancies in previous data for TGFβ1 expression are probably due to assay sensitivity. TGFβ1, but not activin, expression is consistently correlated with Smad pathway activation in the artery wall. The pattern of Smad2 activation supports a model in which TGFβ/activin signaling is anti-atherogenic in the media of normal artery walls but is equally compatible with an anti-atherogenic or pro-atherogenic response to TGFβ/activin in the neointima of lesions.
    Diagnostic and interventional radiology (Ankara, Turkey) 11/2010; 17(3):290-6. · 1.03 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Transforming growth factor-β (TGFβ) and its receptors have been detected by immunohistochemistry in the normal vessel wall and in atherosclerotic lesions of human coronary arteries. However, TGFβ is normally secreted as an inactive complex associated with a latent TGFβ-binding protein (LTBP). Therefore, detection of TGFβ antigen only in the arterial wall does not imply the activated form of the growth factor. In situ hybridization and immunohistochemistry demonstrated LTBP1 mRNA and protein expression throughout the media and intima of early coronary artery lesions, with the highest levels of protein at the luminal surface. In advanced lesions, LTBP1 mRNA and protein were detected mainly in regions of high cell density, such as the fibrous cap. Assays of the TGFβ signalling pathway will be required to determine the activity associated with TGFβ antigen in the vessel wall.
    Circulation Journal 11/2010; 75(1):196-200. · 3.58 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The Kruppel-like factor, KLF13, is a member of a family of transcription factors shown to be involved in haematopoietic development. Here we show that KLF13 is involved in the development of B and T cells at multiple stages. Expression of KLF13 in the thymus was maximal in the DP population and in KLF13(-/-) deficient mice there was an accumulation of DP thymocytes and reduction of CD4(+)SP cells. Cell-surface expression of CD3(high), CD8, CD5 and HSA were altered on KLF13(-/-) DP cells, consistent with a defect in TCR signalling and at the DP to SP transition in KLF13(-/-) mice. KLF13 is also expressed in peripheral T-cells and peripheral T cell activation was impaired in KLF13(-/-) mice. Analysis of early B cell development in the bone marrow (BM) revealed a partial arrest of B cells at the transition from CD43(+) to CD43(-) pre-B cell, a transition that requires signalling through the pre-BCR. The proportion of IgM(+)/IgD(+) mature B cells was also increased in the BM of the KLF13(-/-) mice. This finding is consistent with a reduction in the strength of BCR signal or an accumulation of recirculating B cells from the periphery. Analysis of splenocytes isolated from KLF13(-/-) mice revealed an increase in the expression of CD21 and CD23 on B220(+) B cells, demonstrating a negative regulatory role for KLF13 in co-regulation of expression of CD21 and CD23. Thus KLF13 is involved at multiple different checkpoints in development that require signalling through the TCR, pre-BCR or mature BCR.
    Cell cycle (Georgetown, Tex.) 08/2008; 7(13):2047-55. · 5.24 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To study the function of the Krüppel-like transcription factor KLF13 in vivo, we generated mice with a disrupted Klf13 allele. Although Klf13(-/-) mice are viable, fewer mice were present at 3 weeks than predicted by Mendelian inheritance. Viable Klf13(-/-) mice had reduced numbers of circulating erythrocytes and a larger spleen. The spleen contained an increased number of Ter119(med)CD71(hi), Ter119(hi)CD71(hi), and Ter119(hi)CD71(med) cells but not Ter119(hi)CD71(-) cells, indicating an increase in less mature erythroblasts. A higher proportion of the Ter119(med)CD71(hi) cells were proliferating, indicating that the mice were under a degree of erythropoietic stress. These data indicate that KLF13 is involved in the normal control of erythropoiesis.
    Journal of Biological Chemistry 06/2008; 283(18):11897-904. · 4.65 Impact Factor
  • Source
    Breast Cancer Research 01/2008; 10:1-1. · 5.33 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Multivariate analysis of 1H-NMR spectra of blood sera was reported previously to predict angiographically defined advanced coronary artery disease (CAD) with >90% accuracy and specificity. The analysis depended mainly on the major lipid regions of the spectra, but many variables, including gender and drug treatment, affect lipid composition and are potential confounders. We have determined the predictive power of the same methodology for angiographically defined CAD using plasma samples from groups of male patients, classified by statin treatment, who had normal coronary arteries (NCAs) or CAD. Predictions for NCA and CAD groups were only 80.3% correct for patients not treated with statins and 61.3% for treated patients, compared with random correct predictions of 50%. A confidence limit of >99% was achieved for 36.2% of predictions for untreated groups and 6.2% for treated groups. Detection of CAD by 1H-NMR with >99% confidence was therefore very weak compared with angiography.
    Nature Medicine 06/2006; 12(6):705-10. · 22.86 Impact Factor
  • 01/2006;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We have previously identified a C2H2 zinc-finger transcription factor [BTEB3 (basal transcription element-binding protein 3)/KLF13 (Krüppel-like factor 13)] that activates the minimal promoter for the smooth muscle-specific SM22alpha gene in other types of cell. We show that recombinant BTEB3 binds to three TGGG motifs in the minimal SM22alpha promoter. By mutation analysis, only one of these boxes is required for BTEB3-dependent promoter activation in P19 cells and BTEB3 activates or inhibits reporter gene expression depending on the TGGG box to which it binds. Transient transfection experiments show that BTEB3 also activates reporter gene expression from the SM22alpha promoter in VSMCs (vascular smooth muscle cells). Similar studies showed that BTEB3 did not activate expression from the promoter regions of the smooth muscle myosin heavy chain or smooth muscle alpha-actin promoters, which contain similar sequences, implying that promoter activation by BTEB3 is selective. The expression of BTEB3 is readily detectable in VSMCs in vitro and is modulated in response to injury in vivo.
    Biochemical Journal 11/2003; 375(Pt 2):457-63. · 4.65 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: There is evidence that transforming growth factor (TGF)beta acts as a suppressor of tumor initiation but also as a promoter of tumor progression when the antiproliferative effect of the TGFbeta signaling pathway has been overridden by other oncogenic mutations. Several somatic mutations that disrupt the TGFbeta-SMAD signaling pathway have been reported in human breast tumors. We have examined the association between single nucleotide polymorphisms (SNPs) in the TGFbeta1 gene and the incidence of invasive breast cancer in three case-control series, with a maximum of 3987 patients and 3867 controls, median age approximately 50 years, and range 22-92 years. The promoter SNP, C-509T, and the T +29C signal-peptide SNP (encoding Leu10Pro) are in strong linkage disequilibrium. They are both significantly associated with increased incidence of invasive breast cancer in a recessive manner [odds ratios: (TT versus C-carrier), 1.25; 95% confidence intervals 1.06-1.48; P = 0.009 and (ProPro versus Leu-carrier), 1.21; 95% confidence intervals 1.05-1.37; P = 0.01]. The G-800A SNP was not significantly associated with incidence of breast cancer. The C-509T SNP is not contained within a known consensus sequence for a promoter regulatory element and therefore unlikely to affect TGFbeta1 expression, whereas the Leu10Pro signal peptide substitution potentially affects TGFbeta1 secretion. Transfections of HeLa cells with constructs encoding either the Pro or Leu forms of TGFbeta1 and driven by the cytomegalovirus promoter indicate that the signal peptide with Pro at residue 10 causes a 2.8-fold increase in secretion compared with the Leu form. These data indicate that the allele encoding Pro10 is associated with increased rates of TGFbeta1 secretion and with increased incidence of invasive breast cancer for the population samples described. It is estimated that 3% of all breast cancer cases may be attributable to Pro10 homozygosity.
    Cancer Research 06/2003; 63(10):2610-5. · 8.65 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: NOV is a member of the CCN family of matricellular proteins. We have shown previously that Nov is strongly expressed by vascular smooth muscle cells (VSMCs) of the rat carotid artery. However, 7 days after injury, Nov expression is down-regulated, except near the luminal surface of the developing intima, where it is strongly expressed. These data suggested that NOV might be involved in the regulation of endothelial cell adhesion. NOV promoted the adhesion of human umbilical vein endothelial cells (HUVECs), which was abolished by anti-NOV antibody. HUVEC adhesion to NOV required divalent cations and was inhibited by GRGDS peptide, implicating integrins in the adhesion mechanism. Monoclonal antibodies (mAbs) against αvβ3 inhibited adhesion of HUVECs to NOV, and NOV was shown to bind to αvβ3. Anti-α5β1 mAbs also inhibited HUVEC adhesion to NOV, but adhesion via α5β1 was mediated by fibronectin. HUVEC adhesion to NOV caused intracellular signalling, as evidenced by increased phosphotyrosine content of focal adhesion kinase. Together with evidence that Nov expression in a variety of tissues is restricted to blood vessels containing VSMCs, these data are consistent with a role for NOV in endothelial cell adhesion in vascular homeostasis and in the response to injury.
    Journal of Vascular Research - J VASC RES. 01/2003; 40(3):234-243.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Recent evidence has implicated CC(A/T(richG))GG (CArG) boxes, binding sites for serum response factor (SRF), in the regulation of expression of a number of genes in response to changes in the actin cytoskeleton. In many cases, the activity of SRF at CArG boxes is modulated by transcription factors binding to overlapping (e.g. Yin Yang 1, YY1) or adjacent (e.g. ets) binding sites. However, the mechanisms by which SRF activity is regulated by the cytoskeleton have not been determined. To investigate these mechanisms, we screened for cells that did or did not increase the activity of a fragment of the promoter for a smooth-muscle (SM)-specific gene SM22alpha, in response to changes in actin cytoskeletal polymerization induced by LIM kinase. These experiments showed that vascular SM cells (VSMCs) and C2C12 cells increased the activity of promoters containing at least one of the SM22alpha CArG boxes (CArG near) in response to LIM kinase, whereas P19 cells did not. Bandshift assays using a probe to CArG near showed that P19 cells lacked detectable YY1 DNA binding to the CArG box in contrast with the other two cell types. Expression of YY1 in P19 cells inhibited SM22alpha promoter activity and conferred responsiveness to LIM kinase. Mutation of the CArG box to inhibit YY1 or SRF binding indicated that both factors were required for the LIM kinase response in VSMCs and C2C12 cells. The data indicate that changes in the actin cytoskeletal organization modify SRF activity at CArG boxes by modulating YY1-dependent inhibition.
    Biochemical Journal 07/2002; 364(Pt 2):547-54. · 4.65 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Retinal neovascularization occurs in a variety of diseases including diabetic retinopathy, the most common cause of blindness in the developed world. There is accordingly considerable incentive to develop drugs that target the aberrant angiogenesis associated with these conditions. Previous studies have shown that a number of anti-angiogenic agents can inhibit retinal neovascularization in a well-characterized murine model of ischemia-induced proliferative retinopathy. Combretastatin-A4 (CA-4) is an anti-vascular tubulin-binding agent currently undergoing clinical evaluation for the treatment of solid tumors. We have recently shown that CA-4 is not tumor-specific but elicits anti-vascular effects in nonneoplastic angiogenic vessels. In this study we have examined the capacity of CA-4 to inhibit retinal neovascularization in vivo. CA-4 caused a dose-dependent inhibition of neovascularization with no apparent side effects. The absence of vascular abnormalities or remnants of disrupted neovessels in retinas of CA-4-treated mice suggests an anti-angiogenic mechanism in this model, in contrast to the anti-vascular effects observed against established tumor vessels. Importantly, histological and immunohistochemical analyses indicated that CA-4 permitted the development of normal retinal vasculature while inhibiting aberrant neovascularization. These data are consistent with CA-4 eliciting tissue-dependent anti-angiogenic effects and suggest that CA-4 has potential in the treatment of nonneoplastic diseases with an angiogenic component.
    American Journal Of Pathology 04/2002; 160(3):1097-103. · 4.60 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The requirement for tumour vascularisation to permit the expansion of solid tumours beyond a threshold size of approximately 1 mm diameter has focussed attention on anti-vascular and anti-angiogenic agents for cancer therapy. Combretastatin-A4 (cis CA-4P) is a tubulin-binding agent that is cytotoxic for proliferating endothelial cells in vitro and causes anti-vascular effects in the established tumour vessels of some primary tumours. Preliminary data from Phase I clinical trials indicate that cis CA-4 may also be effective in targeting the vasculature of human tumours. As metastatic disease is the principal cause of mortality in cancer, we have investigated the effects of cis CA-4 on metastatic development using an in vivo model. We show that bolus or continuous administration of cis CA-4P results in potent inhibition of metastases derived from ectopic primary Lewis lung carcinomas in mice whereas the trans CA-4 isomer is without effect. These data further characterise the activity of CA-4 in vivo and suggest that the drug should be evaluated clinically as an anti-metastatic agent.
    International Journal of Oncology 11/2001; 19(4):821-5. · 2.66 Impact Factor
  • Source
    K M Martin, J C Metcalfe, P R Kemp
    [Show abstract] [Hide abstract]
    ABSTRACT: Klf9 and Klf13 are members of the C(2)H(2) zinc finger family of transcription factors that are thought to be involved in regulating basal transcription. The mRNA localization of Klf9 and Klf13 during development was determined by in situ hybridization of mouse E8, E11, E13 and E16 embryo sections. The data showed that Klf9 and Klf13 are widely expressed at all the mouse embryo stages examined. Whilst the expression patterns of the two genes largely overlap there are differences in the localization or level of expression in some tissues. At E11, both genes are expressed in high levels in the cephalic mesenchyme whilst Klf13 and not Klf9 is expressed at high levels in the developing heart at E8 and E11. In the gut and bladder at E16, Klf13 is expressed in the epithelial cell layer whereas Klf9 is expressed in both the muscle and epithelial layers. Both Klf9 and Klf13 are expressed at high levels in the epidermis at E11, E13 and E16.
    Mechanisms of Development 06/2001; 103(1-2):149-51. · 2.38 Impact Factor
  • Source
    K J Clark, N R Cary, A A Grace, J C Metcalfe
    [Show abstract] [Hide abstract]
    ABSTRACT: A somatic mutation within a microsatellite polyA tract in the coding region of the type II transforming growth factor (TGF)-beta receptor gene was reported to occur in human atherosclerotic and restenotic lesions. This mutation occurs frequently in colorectal cancer with the replication error repair phenotype and results in loss of sensitivity to the growth inhibitory effects of TGF-beta in cells from the tumors. The mutation was proposed to account for the clonal expansion of vascular smooth muscle cells observed in atherosclerotic plaques, through loss of the growth inhibitory effect of TGF-beta. The frequency of the mutation and the extent of clonal expansion of the mutated cells have major implications for the mechanism of atherogenesis and therapeutic strategies. We analyzed a set of 22 coronary arterial and 9 aortic samples containing early to advanced atherosclerotic lesions for the mutation in the type II TGF-beta receptor polyA tract. Only 1 coronary arterial sample from an advanced lesion showed detectable amounts of the mutation, present at a low level (8% of the DNA sample). The data imply that the mutation occurs only at low frequency and is not a major mechanistic contributor to the development of atherosclerosis.
    Arteriosclerosis Thrombosis and Vascular Biology 05/2001; 21(4):555-9. · 6.34 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Combretastatin-A4 phosphate (cis-CA-4) is a tubulin-binding agent currently undergoing clinical trials as an anti-tumour drug. We have investigated whether CA-4 functions as a tumour-specific anti-vascular agent using the hyperplastic thyroid as a novel in vivo model of neovascularization. CA-4 elicited pathological changes in normal tissue, manifested as the induction of multiple, discrete intravascular thrombi. These vascular-damaging effects indicate that CA-4P does not function as a tumour-specific agent but targets neovasculature irrespective of the primary angiogenic stimulus.
    British Journal of Cancer 04/2001; 84(6):832-5. · 5.08 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Tamoxifen and its analogues act as selective estrogen receptor modulators (SERMs) in women, with estrogen-like activities on some plasma cardiovascular risk factors (eg, lipoproteins). Effects of SERMs on men with coronary artery disease (CAD) have not been reported. Thirty-one men with angiographically proven CAD were recruited; 16 were treated with tamoxifen (40 mg/d) for 56 days, and 15 were untreated. All the CAD patients were medicated with aspirin and an HMG-CoA reductase inhibitor for >/=6 weeks before entering the study. Ten men with angina-like symptoms but normal coronary arteries by angiography (NCA group) were also treated with tamoxifen. Blood samples were collected at days -7, 0, 7, 14, 21, 28, and 56 of treatment. Endothelium-dependent flow-mediated dilatation (ED-FMD) of the brachial artery was measured by high-resolution ultrasound at 5 visits. Tamoxifen caused an increase in %ED-FMD maximal at 28 days in the CAD group (2.1+/-0.3% to 7.5+/-0.7%; P<0.0001) and the NCA group (3.8+/-0.4% to 7.9+/-1.0%; P<0.0001), with no significant change in the untreated group. Tamoxifen also caused decreases in several plasma cardiovascular risk factors, including total cholesterol, triglycerides, lipoprotein(a), and fibrinogen. Except for the triglyceride response, these effects were similar to those reported for postmenopausal women treated with tamoxifen. Tamoxifen substantially increased ED-FMD in men with CAD who were taking conventional medication. Together with the effects on risk factors, the data strongly support clinical evaluation of SERMs for the treatment of men with CAD.
    Circulation 03/2001; 103(11):1497-502. · 15.20 Impact Factor

Publication Stats

7k Citations
1,222.21 Total Impact Points

Institutions

  • 1975–2012
    • University of Cambridge
      • Department of Biochemistry
      Cambridge, ENG, United Kingdom
  • 2010
    • Massachusetts General Hospital
      Boston, Massachusetts, United States
  • 1997
    • Stanford University
      • Falk Cardiovascular Research Center
      Stanford, CA, United States
  • 1994
    • Papworth Hospital NHS Foundation Trust
      Papworth, England, United Kingdom
  • 1987
    • National Heart, Lung, and Blood Institute
      Maryland, United States
  • 1974–1984
    • MRC National Institute for Medical Research
      Londinium, England, United Kingdom