Microsatellite Mutation of Type II Transforming Growth Factor-β Receptor Is Rare in Atherosclerotic Plaques
Abstract—A somatic mutation within a microsatellite polyA tract in the coding region of the type II transforming growth factor (TGF)-β 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-β 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-β. 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-β 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.
- transforming growth factor-β
- type II transforming growth factor-β receptors
- microsatellite mutation
- Received March 7, 2000.
- Accepted December 7, 2000.
Arecent study of the transforming growth factor (TGF)-β type II receptor (TβRII) in human coronary atherosclerotic and restenotic lesions identified a somatic mutation that generates a form of the receptor lacking cytosolic and transmembrane domains that is not expressed in the plasma membrane.1 TGF-β1 signaling normally occurs through heterodimer complexes of the TβRII and TGF-β type I receptors; therefore, loss of expression of TβRII in the plasma membrane inhibits TGF-β1 signaling.2 3 The mutation occurs through reduction of a 10A tract to either a 9A or 8A tract, which introduces a stop codon at 161 or 129, respectively, within the coding region of TβRII.4 5 6 Both mutations were first described in human colorectal cancer and were found to occur frequently in replication error repair–positive tumors but were uncommon in other colorectal tumors or other tumor types.5 7 8 9 In replication error repair–positive colorectal cancer, the TβRII mutation is correlated with a mutation of mismatch repair enzymes4 10 ; however, no mutation was detected in the hMSH3 gene associated with the receptor mutation in coronary atherosclerotic plaques.
The association of a TGF-β receptor mutation with atherosclerosis is of interest in several respects. It was the first demonstration of a frequent somatic cell mutation in atherosclerosis. The mutation occurs in a signaling pathway known to regulate in vitro vascular smooth muscle cell (VSMC) proliferation.11 12 It was also noted that inhibition of the TGF-β signaling pathway in mutated cells provided an explanation for the well-characterized clonal expansion of VSMCs in atherosclerotic plaques, which is reported to occur in a high proportion of lesions (in 24 of 3013 and 12 of 1514 lesions). If the mutation contributes significantly to the clonal expansion of VSMCs in plaques, it would be expected to occur in a substantial proportion of lesions. The total number of plaques examined and the frequency with which the mutation was detected in the plaques were not stated, although of the 20 plaques described, 14 were reported to contain the mutation.1 However, in a subsequent report in which only 6 atherosclerotic plaques were screened for the mutation, no mutations were detected.15 The frequency of the mutation is important in terms of the molecular pathology of plaque development and the genetic and environmental factors that determine susceptibility to the mutation, if it is a major contributory mechanism to atherogenesis. Therefore, we have screened DNA for the mutation in 41 sections of coronary and aortic tissue of which 31 contained atherosclerotic plaques from early to advanced stages.
Human Tissue Samples
Eleven coronary artery samples were obtained from 10 patients and prepared as paraffin-embedded sections. The status of the artery in the sections analyzed is summarized in Table 1⇓. A further series of 19 coronary artery and 11 thoracic aortic samples was obtained from hearts removed from transplant recipients in the heart and heart-lung transplantation program at Papworth Hospital, Cambridge, UK (Table 2⇓). The present study was approved by an institutional review committee, and informed consent was obtained from each patient. Tissue samples were immersed in Cryo-M-Bed embedding compound (Bright Instruments), snap-frozen in liquid nitrogen, and stored at −80°C. Embedded blocks were equilibrated at −20°C in an OFT motor-driven microtome, and 10- and 20-μm transverse sections were cut. Microdissection was performed by manipulating 20-μm tissue sections with needles. Staining with oil red O as previously described16 was used before staging the lesions according to the method of Stary and colleagues.17 18 19
Various methods of DNA extraction were compared to determine whether the proportion of slippage observed in control tissue was affected by the method of extraction. A sample of donor aortic tissue was extracted by (1) the method used in the previous study of proteinase K/SDS digestion, phenol/chloroform extraction, and RNase digestion; (2) proteinase K digestion overnight, followed by centrifugation (10 000g, 15 minutes at room temperature) to remove insoluble material; and (3) the QIAamp tissue kit (Qiagen Ltd) according to the manufacturer’s instructions with or without RNase treatment. The method of extraction had no significant effect on the proportion of slippage (9A product), which was (1) 9.9±3.0%, (2) 8.6±0.8%, and (3) 8.1±0.7% with RNase and 8.0±1.0% without RNase. Therefore, except where indicated, the QIAamp kit was used in subsequent experiments because it gave greater reproducibility of amplification from the DNA extracted from the tissue sections.
PCR and Strand-Length Polymorphisms
The TβRII polyadenine tract was amplified according to the method of McCaffrey et al.1 Briefly, 2 oligonucleotide primers were synthesized: a forward primer, 5′-CAGTTTGCCATGACCCCAAG-3′, and a reverse primer, 5′-CATTGCACTCATCAGAGCTACAGG-3′. For polymerase chain reaction (PCR) amplification, 0.2 to 1 μg DNA was mixed with 200 μmol/L dNTPs and 12.5 pmol primer 1 in Pfu polymerase buffer (Stratagene Inc), heated to 95°C for 1 minute, cooled to 80°C, and hot-started with 1.25 U Pfu polymerase, 12.5 pmol primer 2, and 5 μCi [α-33P]dATP. The thermal cycling profile was 1-minute denaturation at 95°C, 1-minute annealing at 60°C, and then 1-minute extension at 72°C for 30 cycles. Radiolabeled PCR products were blunt-ended by digestion with 5 U AluI (Pharmacia Biotech) for 1.5 hours at 37°C. The digested PCR products were separated on 10% polyacrylamide/8 mol/L urea sequencing gels run at 40 W for 4 to 6 hours, fixed in 10% methanol/10% acetic acid for 20 minutes, dried, and exposed to a phosphorimage screen (Molecular Dynamics Ltd). Radioactive products were quantified by using ImageQuant software. PCR reactions were performed on 2 DNA concentrations for each sample. At least 2 separate experiments were performed on all samples with the same result, except for samples C1, C2, C12, C19, and C24, for which only 1 experiment was performed. Control reactions in the absence of genomic DNA were negative in all experiments.
Quantification of PolyA Tract Products
Control plasmids containing either wild-type TβRII (10A tract) or mutated TβRII (9A tract) were made by cloning PCR fragments of the genomic DNA from colorectal cancer cell lines, provided by Drs L. Myeroff and S. Markowitz (Ireland Cancer Center, Case Western Reserve University, Cleveland, Ohio) into pCR II vector, and the sequences were confirmed. The plasmids were used as control samples for quantification of PCR products in all experiments.
Pfu polymerase slippage for the wild-type receptor plasmid (10A tract) was determined as the area of the 100-bp (9A) product peak relative to the total area of both the 101-bp (10A) and 100-bp (9A) tract product peaks (see below). The amount of mutation in experimental samples was expressed as the increase in the proportion of 9A tract product observed above the level of 9A tract slippage obtained from wild-type plasmid controls.
The accuracy in determining the proportion of 9A tract product depends on the signal-to-noise ratio for the intensity of the 9A product band, inasmuch as this band was of lower intensity than the corresponding 10A band in all experiments. The signal-to-noise ratio for each polyA tract product was determined by integrating the intensity of the band compared with the integrated intensity of the same area of background within the same lane. By use of serial dilution of the wild-type plasmid control, the signal-to-noise ratio was determined for each dilution after 24-hour exposure on the phosphorimager screen (Figure 1a⇓ and 1b⇓). Constant values for the proportion of 9A tract product from the wild-type plasmid control were obtained when the signal-to-noise ratio of the 9A band (ie, the weaker signal in wild-type plasmid controls) was >3 (Figure 1c⇓). Therefore, data were excluded if the signal-to-noise ratio of the 9A band was <5.
The proportion of 9A tract obtained from the control 10A plasmid in 22 determinations was 9.1±2.2% (mean±SD). The amount of 9A band detected was linear with the proportion of 9A in mixtures of the control plasmids (Figure 1d⇑). The amount of 9A observed was significantly greater than the amount of 9A obtained by slippage from 10A control plasmid for additions of >4% of 9A plasmid (P<0.002, 2-tailed unpaired Student t test). Therefore, under the assay conditions used, the lower limit to the proportion of mutated receptor that could be detected in the patient samples was taken as 4%.
Coronary and Aortic Sections
To obtain a preliminary estimate of the proportion of plaques that contained mutations, a study of 11 transverse sections from 10 coronary arteries was made. The sections varied from normal artery3 to early atherosclerotic plaques4 and then to advanced plaques4 and were examined without knowledge of the status of the samples.
DNA was extracted (proteinase K digestion; see Methods) from paraffin-embedded samples of undissected coronary artery sections (C1 to 11, Table 1⇑) and analyzed for polyA tract mutation (Figure 2⇓). Of the 11 samples examined, 10 had proportions of 9A tract between 9.5% and 11.3%, which was well within the SD for all wild-type receptor forms examined (9.1±2.2%, data not shown). Only 1 sample (C3) showed an increased proportion of 9A tract (to 17.0±1.1%, 4 separate determinations), which was 2 SD above the mean wild-type value.
Because the proportion of plaques containing mutated receptor was low, a further set of 19 coronary arteries and 11 pieces of aorta from explanted hearts was collected. None of the 15 coronary sections or 11 aortic sections examined (C12 to C26 and A1 to A11, Table 2⇑) contained amounts of mutated receptor gene >1 SD above the mean wild-type value.
Microdissected Coronary Sections
To determine whether the inability to detect a significant proportion of atherosclerotic plaques containing mutated receptor gene (1 of 18 plaques) was due to the presence in the sections of large proportions of wild-type receptor gene, the plaques were microdissected. One set of coronary arterial sections with advanced atherosclerotic plaques (9 from 8 arteries) was microdissected. Regions of the plaque were separated from the underlying media and adventitia and were collected, together with a region of the fibrous cap where this was well defined (4 of 9 plaques). Representative sections (lesion stages IV, Va, Vb, and Vc) stained with hematoxylin/eosin (with the regions microdissected indicated) are shown in Figure 3⇓. None of the separated components of the plaque sections contained detectable mutated receptor gene.
The description by McCaffrey et al1 of a mutation in atherosclerotic plaques that would inhibit TGF-β signaling was consistent with a role for TGF-β as a major antiatherogenic cytokine.16 20 However, in the 22 coronary arterial and 9 aortic samples examined, which contained lesions from early to advanced stages, only 1 coronary arterial sample showed a detectable mutation of the TβRII polyA tract. The level of mutation detected was low (at 8% above background) but well above the threshold for assay significance (4% above background). Therefore, a key issue to be addressed in future studies is the origin of the different frequencies of mutation observed in the study of McCaffrey et al compared with the present study and that of Bobik et al.15 We have taken great care to ensure that the experimental conditions used to extract and process the DNA do not have any effect on the ratio of 10A to 9A forms detected after PCR. Furthermore, sufficient DNA was obtained from most plaques to repeat the duplicate assays in 2 to 4 separate experiments for each sample. We conclude that the reason for the difference in the observed frequency of mutation is very unlikely to be due to differences in the experimental techniques used.
A potential contribution to the difference in mutation frequency observed between the studies is the type of lesion material analyzed. McCaffrey et al1 reported that a significant level of TβRII mutation is present in lesions of a subset of surgical endarterectomy vascular specimens, but no samples of this type were examined in the present study or in the study of Bobik et al.15 However, sections from coronary arteries containing plaques were examined in all 3 studies, and the lesions in the plaques that we analyzed ranged from stages II to V according to the classification of Stary et al,17 indicating that the extent of lesion development is unlikely to account for the difference in frequency of mutation observed. An alternative explanation is that there are major differences in the frequency of the mutation in lesions obtained from different populations of patients, eg, differences due to sex or gene pool (eg, associated with ethnic groups). A more plausible explanation lies in the possibility that an environmental cofactor contributes to those mutations that arise in blood vessels.21
Monoclonality of atherosclerotic plaques is well established, although the mechanism by which it occurs remains unclear.14 However, it has been suggested that the clonality observed in lesions may represent the expansion of preexisting clones from within uninjured vessel beds.22 In contrast, McCaffrey et al1 have suggested that the TβRII mutation might bring about the monoclonal expansion of cells within atherosclerotic lesions by allowing the mutated cells to escape the inhibition of proliferation that is usually exerted by TGF-β. This suggestion implies that the mutation frequency would approach the frequency of the monoclonal phenotype. In the first study of monoclonal cells in plaques (Benditt and Benditt13 ), data in their Table 1⇑ indicated that 24 of 30 fibrous caps of atherosclerotic plaques showed monoclonal phenotypes. In a more recent study, plaque smooth muscle cells showed a single pattern of X inactivation, indicating that the smooth muscle cells were monoclonal in 3 of 4 aortic plaques and in 9 of 11 coronary plaques.14 Therefore, the present data and the data of Bobik et al15 imply that it is unlikely that the TβRII mutation is a major cause of the monoclonal regions detected in a high proportion of plaques. However, it should be noted that any deficit in TGF-β signaling, either through decreased levels TGF-β activity or through altered expression of the signal transduction proteins (receptors and/or Sma- and Mad-related proteins), might promote expansion of VSMCs within plaques. It is of interest that mutations throughout the TGF-β signaling pathway are estimated to occur in 80% of human colorectal tumors.23
This study was supported by The British Heart Foundation (program grant to J.C.M.). A.A.G. is a British Heart Foundation Senior Research Fellow. The authors are grateful to Dr P.D. Ellis for transplant tissue collection and Drs P.R. Kemp and T.R. Hesketh for helpful discussions.
Markowitz S, Wang J, Myeroff L, Parsons R, Sun LZ, Lutterbaugh J, Fan RS, Zborowska E, Kinzler KW, Vogelstein B, et al. Inactivation of the type-II TGF-beta receptor in colon-cancer cells with microsatellite instability. Science. 1995;268:1336–1338.
Myeroff LL, Parsons R, Kim SJ, Hedrick L, Cho KR, Orth K, Mathis M, Kinzler KW, Lutterbaugh J, Park K, et al. A transforming growth-factor-beta receptor-type-II gene mutation common in colon and gastric but rare in endometrial cancers with microsatellite instability. Cancer Res. 1995;55:5545–5547.
Park KC, Kim SJ, Bang YJ, Park JG, Kim NK, Roberts AB, Sporn MB. Genetic changes in the transforming growth-factor-beta (TGF-beta) type-II receptor gene in human gastric-cancer cells: correlation with sensitivity to growth-inhibition by TGF-beta. Proc Natl Acad Sci U S A. 1994;91:8772–8776.
Parsons R, Myeroff LL, Liu B, Willson JKV, Markowitz SD, Kinzler KW, Vogelstein B. Microsatellite instability and mutations of the transforming growth-factor-beta type-II receptor gene in colorectal-cancer. Cancer Res. 1995;55:5548–5550.
Iwaya T, Maesawa C, Nishizuka S, Suzuki Y, Sakata K, Sato N, Ikeda K, Koeda K, Ogasawara S, Otsuka K, et al. Infrequent frameshift mutations of polynucleotide repeats in multiple primary cancers affecting the esophagus and other organs. Genes Chromosomes Cancer. 1998;23:317–322.
Tani M, Takenoshita S, Kohno T, Hagiwara K, Nagamachi Y, Harris CC, Yokota J. Infrequent mutations of the transforming growth factor beta-type II receptor gene at chromosome 3p22 in human lung cancers with chromosome 3p deletions. Carcinogenesis. 1997;18:1119–1121.
Owens GK, Geisterfer AAT, Yang YWH, Komoriya A. Transforming growth factor-beta-induced growth-inhibition and cellular hypertrophy in cultured vascular smooth-muscle cells. J Cell Biol. 1988;107:771–780.
Benditt E, Benditt JM. Evidence for a monoclonal origin of human atherosclerotic plaques. Proc Natl Acad Sci U S A. 1973;70:1753–1756.
Bobik A, Agrotis A, Kanellakis P, Dilley R, Krushinsky A, Smirnov V, Tararak E, Condron M, Kostolias G. Distinct patterns of transforming growth factor-beta isoform and receptor expression in human atherosclerotic lesions: Colocalization implicates TGF-beta in fibrofatty lesion development. Circulation. 1999;99:2883–2891.
Stary HC, Chandler AB, Dinsmore RE, Fuster V, Glagov S, Insull W, Rosenfeld ME, Schwartz CJ, Wagner WD, Wissler RW. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis: a report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Circulation. 1995;92:1355–1374.
Stary HC, Chandler AB, Glagov S, Guyton JR, Insull W, Rosenfeld ME, Schaffer SA, Schwartz CJ, Wagner WD, Wissler RW. A definition of initial, fatty streak, and intermediate lesions of atherosclerosis: a report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Arterioscler Thromb. 1994;14:840–856.
Stary HC, Blankenhorn DH, Chandler AB, Glagov S, Insull W, Richardson M, Rosenfeld ME, Schaffer SA, Schwartz CJ, Wagner WD, et al. A definition of the intima of human arteries and of its atherosclerosis-prone regions: a report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Circulation. 1992;85:391–405.
Grady WM, Myeroff LL, Swinler SE, Rajput A, Thiagalingam S, Lutterbaugh JD, Neumann A, Brattain MG, Chang J, Kim S-J, et al. Mutational inactivation of transforming growth factor β receptor type II in microsatellite stable colon cancers. Cancer Res. 1999;59:320–324.