Connection between elastin haploinsufficiency and increased cell proliferation in patients with supravalvular aortic stenosis and Williams-Beuren syndrome

Abstract

To elucidate the pathomechanism leading to obstructive vascular disease in patients with elastin deficiency, we compared both elastogenesis and proliferation rate of cultured aortic smooth-muscle cells (SMCs) and skin fibroblasts from five healthy control subjects, four patients with isolated supravalvular aortic stenosis (SVAS), and five patients with Williams-Beuren syndrome (WBS). Mutations were determined in each patient with SVAS and in each patient with WBS. Three mutations found in patients with SVAS were shown to result in null alleles. RNA blot hybridization, immunostaining, and metabolic labeling experiments demonstrated that SVAS cells and WBS cells have reduced elastin mRNA levels and that they consequently deposit low amounts of insoluble elastin. Although SVAS cells laid down approximately 50% of the elastin made by normal cells, WBS cells deposited only 15% of the elastin made by normal cells. The observed difference in elastin-gene expression was not caused by a difference in the stability of elastin mRNA in SVAS cells compared with WBS cells, but it did indicate that gene-interaction effects may contribute to the complex phenotype observed in patients with WBS. Abnormally low levels of elastin deposition in SVAS cells and in WBS cells were found to coincide with an increase in proliferation rate, which could be reversed by addition of exogenous insoluble elastin. We conclude that insoluble elastin is an important regulator of cellular proliferation. Thus, the reduced net deposition of insoluble elastin in arterial walls of patients with either SVAS or WBS leads to the increased proliferation of arterial SMCs. This results in the formation of multilayer thickening of the tunica media of large arteries and, consequently, in the development of hyperplastic intimal lesions leading to segmental arterial occlusion.

Figure 1

A and B, Representative photomicrographs comparing a transverse section of the ascending aorta of a healthy 2-year-old child (A) and the nonstenotic region of the ascending aorta of 2-year-old patient with WBS (B) (staining is with Movat's pentachrome; original magnification ×100). Nonstenotic regions of the WBS aorta demonstrate a larger-than-normal aortic-wall diameter, resulting from an increase in the number of SMC layers surrounded by abnormally thin elastic lamellae. C–E, Higher-power images of sections stained with Movat's pentachrome (magnification ×400), comparing the tunica media of the aorta of a healthy 2-year-old child (C) and the medial (D) and neointimal (E) layers of the stenotic region of the aorta of a 2-year-old patient with WBS. In contrast to the normal aortic tunica media composed of parallel elastic lamellae separated by single layers of SMC, in the stenotic region the lamellar architecture of the tunica media is disorganized and contains multiple clusters of SMC separated by thin, fragmented elastic fibers (D). The intimal thickening of the WBS aorta is composed of numerous SMC surrounded by elastin-poor ECM (E).

Figure 2

Identification of a new elastin-gene mutation in patient 2250 from family SVAS-31, and reduced elastin mRNA expression in SVAS SMCs and WBS SMCs. A, CSGE of ELN exon 20, which detects a mobility shift in patient 2250 (arrow). Samples from families SVAS-21 and SVAS-22 are included as negative controls, and a sample from family SVAS-26 is included as a positive control. B, DNA sequence analysis of exon 20, which identifies a guanosine deletion at position 1238 (1238delG), which is predicted to result in premature termination of translation. C, RT-PCR analysis of exon 20 in skin fibroblasts (Fib) and aortic SMCs (Ao) in patient 2250, which indicates that elastin mRNA containing mutation 1238delG is degraded by the NMD mechanism. Inhibition of NMD by cycloheximide (CHX) restores the expression of the mutant transcript. Samples from a normal skin-fibroblast line (9001) are included as controls. D, RNA blot analysis of elastin mRNA in aortic SMC from a patient with WBS (1001), a patient with nonsyndromic SVAS (2250), and a healthy control subject (9006), which show reduced total elastin mRNA in WBS and SVAS samples. Note that elastin mRNA abundance is less in WBS than in SVAS. Ethidium bromide staining of 28S rRNA is included as a control.

Figure 3

Phosphorimager analysis of elastin mRNA abundance in normal, SVAS, and WBS fibroblasts. RNA samples were analyzed by blot hybridization using elastin cDNA probe H-11. The intensity of the hybridization signal (gels “ELN” [inset]) was normalized to the abundance of 28S rRNA (gels “28S” [inset]) in the corresponding gel stained with ethidium bromide, and the resulting values are shown, in the bar graph, in arbitrary units. The order of the samples in the inset is the same as in that in the bar graph.

Figure 4

A, Representative photomicrographs of 10-d-old cultures of human aortic SMC immunostained with specific antibodies to elastin (magnification ×100). Nuclei were counterstained with propidium iodide. B and C, Representative higher-power images (magnification ×400) comparing expression of the EBP and LO in 3-d-old cultures of aortic SMC. D, Representative photomicrographs of 10-d-old cultures of human aortic SMC metabolically labeled with BrdU and immunostained with anti-BrdU antibodies (brown nuclei), demonstrating higher-than-normal proliferating rates of SVAS cells and of WBS cells (magnification ×40). E and F, Results (mean ± SD) of quantitative analysis of a typical experiment using 3-d-long metabolic labeling of quadruplicate cultures with radioactive valine, followed by biochemical isolation of insoluble elastin, which demonstrates that aortic SMC from patients with SVAS and patients with WBS deposit, respectively, low and very low amounts of insoluble elastin (E), and of measurements of [³H]-thymidine incorporation into the parallel quadruplicate cultures, which indicate that abnormally low levels of insoluble elastin are associated with an inversely proportional increase in the proliferation rate of SMC in patients with SVAS and patients with WBS (F) (* [P<.001]).

Figure 5

Results (mean ± SD) of quantitative assay of insoluble elastin (metabolically labeled with [³H]-valine) deposited in 3-d-old cultures of human fibroblasts (A) and assessment of cellular proliferation rate in parallel cultures, measured by incorporation of [³H]-thymidine for the last 24 h (B). Fibroblasts from all four patients with SVAS deposited consistently lower amounts of insoluble elastin than did those from four healthy control subjects. Fibroblasts from all four patients with WBS deposited even lower amounts of insoluble elastin than did those from patients with isolated SVAS. These abnormally low levels are associated with an inversely proportional increase in proliferation rates. Combined results from three separate experiments, with quadruplicate cultures of fibroblasts derived from each patient (initially plated 100,000 cells /dish), indicate that differences between experimental groups, normal versus SVAS, normal versus WBS (* [P<.002]), and SVAS versus WBS (** [P<.05]) are statistically significant. Differences between cultures derived from different patients within each experimental groups are not statistically significant.

Figure 6

Results of a typical experiment when quantitative assays of metabolically labeled insoluble elastin (A) and assessments of cellular proliferation (B), measured by total DNA assay and cell counting, were performed in the time course. Results consistently showed that the abnormally high proliferation rate of cultured SVAS fibroblasts and of WBS fibroblasts is inversely proportional to their deposition of insoluble elastin during the entire 10-d culture period. At each time point, quadruplicate cultures of fibroblasts derived from each patient (initially plated 100,000 cells/dish) were assessed. Statistical evaluation showed that at 3, 6, and 10 d, the time-point differences between normal versus SVAS, normal versus WBS, and SVAS versus WBS were statistically significant.

Figure 7

Results of typical experiment comparing [³H]-thymidine incorporation into quadruplicate cultures of aortic SMC (A) and skin fibroblasts (B), derived from healthy control subjects, patients with SVAS, and patients with WBS, which were maintained for 3 d in the presence or absence of either 1 mg of exogenous insoluble elastin/ml or media that were preincubated for only 1 h with the 1 mg of exogenous insoluble elastin/ml. Statistical evaluations indicate that higher-than-normal proliferation of SMC and fibroblasts derived from patients with SVAS and patients with WBS was significantly reduced either in cultures treated with exogenous insoluble elastin or when maintained in media preincubated with insoluble elastin (* [P<.002]). Both treatments also decreased the proliferation rate of normal fibroblasts (** [P<.05]).