|Home | About | Journals | Submit | Contact Us | Français|
Multiple lines of investigation have implicated hydrogen peroxide as an important endogenous mediator of cell proliferation in the vessel wall. Heterogeneous nuclear ribonucleoprotein C (hnRNP-C), a nuclear pre-mRNA binding protein that plays roles in vertebrate cell proliferation and differentiation, has been identified as a component of a vascular cell signaling pathway activated by low physiologic levels of hydrogen peroxide. The expression of hnRNP-C in human arteries has not previously been assessed.
Segments of human proximal internal carotid arteries were evaluated for the expression of hnRNP-C by immunohistochemistry.
In normal proximal internal carotid arteries, hnRNP-C is expressed predominantly by the endothelium, with significantly lower expression by medial smooth muscle. In preatherosclerotic intimal hyperplasia, hnRNP-C is upregulated in the artery wall, due to the robust expression by the intimal smooth muscle cells, without upregulation in the medial smooth muscle cells. In arteries with atherosclerotic lesions, there is strong expression of hnRNP-C not only by intimal cells, but also by medial smooth muscle cells.
The hydrogen peroxide responsive pre-mRNA binding protein hnRNP-C is upregulated in atherosclerosis and in pre-atherosclerotic intimal hyperplasia in humans, supporting the hypothesis that hydrogen peroxide is a regulator of vascular cell proliferation in these conditions. These data also suggest that hnRNP-C may be useful as a marker of vascular cell activation.
Hydrogen peroxide (H2O2) has been implicated as playing an important role in cell proliferation in the vessel wall [1-3]. At low physiologic concentrations, H2O2 serves as a mitogen, stimulating vascular cell proliferation . These low concentrations of H2O2 appear to be generated in the vessel wall from the dismutation of superoxide generated by plasma membrane bound NADPH oxidases, in response to cytokine and growth factor ligand / receptor interactions and alterations in blood flow [5,6]. Identification and characterization of signaling pathways activated by these low concentrations of H2O2 has been challenging [6,7]. In a functional proteomic screen, heterogeneous nuclear ribonucleoprotein C (hnRNP-C) was identified as a nuclear protein that is hyperphosphorylated in response to low concentrations of H2O2 in cultured human endothelial cells .
While likely playing an important role in pre-mRNA processing, the precise functions of hnRNP-C remain incompletely understood . The protein possesses two distinct mRNA binding domains, and is largely restricted to the nucleus due to the presence of a nuclear retention sequence . Deletion of the hnRNP-C gene in the mouse is lethal at the egg cylinder stage, prior to the formation of mesoderm or vascular cells . The hnRNP-C deficient stem cells show impaired proliferation and differentiation. Likewise knock-down of hnRNP-C in various cell lines also inhibits proliferation [12,13]. The ability of hnRNP-C to bind mRNA is regulated by phosphorylation, mediated by the nuclear protein kinase CK1αLS in response to H2O2 [14-16]. Based on studies with cultured cells, hnRNP-C is often referred to as one of the most abundant proteins in the nucleus . However, an immunohistochemical study of normal mouse tissues revealed highly variable expression of hnRNP-C in distinct tissues, with the expression being barely detectable in many cell types .
Human atherosclerosis is a complex and largely still poorly understood disease process. It is clear that in humans, atherosclerotic lesions are essentially always preceded by a hyperplastic phase during which there is expansion of the intima by smooth muscle cells [19,20]. It is within this hyperplastic intima that formation of the necrotic / lipid core characteristic of atherosclerosis develops. Since H2O2 has been implicated in stimulating vascular cell proliferation and intimal hyperplasia, and hnRNP-C has been identified as a H2O2-responsive protein in vascular cells, the expression of hnRNP-C was assessed in normal human arteries, as well as arteries with intimal hyperplasia and atherosclerosis.
Arterial segments were obtained from autopsies performed at Massachusetts General Hospital, within 24 hours of death from patients ranging in age from the 4th to 9th decade of life, as described previously . Cases were excluded if there was a history of prior surgery or radiation therapy involving the arterial segments. All activities were approved by the Hospital’s Human Subjects Institutional Review Board. The proximal left internal carotid artery, 0.5 cm distal to the carotid bifurcation was sampled. The cross-section was fixed in 10% buffered formalin for 16 hr, processed with paraffin embedding and used to generate hematoxylin and eosin stained histological slides. For this study, an atherosclerotic lesion was defined as the presence of numerous foam cells or a necrotic/lipid core, which included classic type II lesions and above. For this study, intimal hyperplasia was defined as an intima with multiple layers of smooth muscle cells with or without isolated/focal foam cells, and without features of atherosclerosis. Vessels with only occasional intimal smooth muscle cells were considered normal. Cases were selected for immunohistochemical analysis based on the lesion type present on hematoxylin and eosin stained section, and are summarized in Table 1. The atherosclerotic lesions assessed consisted of one type II fatty streak, two type IV atheromas and one type V fibroatheroma, as defined previously .
Immunohistochemical analyses for the macrophage marker CD68, the endothelial marker CD31, alpha smooth muscle actin (SMA), and hnRNP-C were performed using the Vecastain Elite peroxidase-conjugated avidin-biotin method (Vector), following antigen retrieval with Borg decloaker solution (Biocare Medical). Monoclonal anti-CD68 antibody (DAKO) was used at 1:50 dilution, monoclonal anti-CD31 antibody (DAKO) was used at 1:40 dilution, monoclonal anti-SMA antibody (DAKO) was used at 1:100 dilution, and monoclonal anti-hnRNP-C antibody (clone 4F4, Santa Cruz) was used at 1:200 dilution.
The data were assessed by one-way Analysis of Variance (ANOVA), followed by post-hoc analysis using GraphPad’s post-test calculator, which employs the Bonferroni correction to adjust for multiple comparisons. P values less than 0.05 were considered significant. All values are expressed as mean ± standard deviation.
In normal arteries there was robust staining for hnRNP-C in endothelium and faint staining of fewer numbers of medial smooth muscle cells (Fig. 1). In both cases, the staining was nuclear, consistent with the known nuclear localization of hnRNP-C. For the four cases assessed, 65±10% (mean ± standard deviation) of the intimal nuclei stained for hnRNP-C and only 24±9% of the medial nuclei stained for hnRNP-C.
In vessels with intimal hyperplasia, the intimal cells were mostly SMA+ smooth muscle cells, with only occasional CD68+ macrophages (Fig 2). In these lesions, there was intense staining of the intimal smooth muscle cells for hnRNP-C. Interestingly, this staining was often more pronounced in the inner half of the intimal hyperplasia than in the outer half. As with the normal arteries, fewer numbers of medial smooth muscle cells stained for hnRNP-C, and this staining was typically faint compared to that of the intimal smooth muscle cells. Overall, 62±7% of the intimal nuclei stained for hnRNP-C, and only 21±12% of the medial nuclei stained for hnRNP-C.
In the atherosclerotic lesions, the intimal cells were still predominantly SMA+ smooth muscle cells, but increasing numbers of CD68+ macrophages were present (Fig 3). In these lesions both the intimal smooth muscle cells and the intimal macrophages stained positively for hnRNP-C. Overall, 71±7% of the intimal nuclei stained for hnRNP-C. Surprisingly, in these arteries with atherosclerosis, there was also robust staining of the medial smooth muscle cells for hnRNP-C, such that the media now stained as intensely as the intima (Fig 4). Overall, 74±15% of the medial nuclei stained for hnRNP-C.
Comparison of the three vascular lesion types revealed a statistically significant progressive upregulation of hnRNP-C in the vessel wall from 28±6% of the nuclei in normal arteries, to 40±5 % of the nuclei in arteries with intimal hyperplasia, to 72±6% of the nuclei in arteries with atherosclerosis (Fig. 5). For the progression of normal arteries to intimal hyperplasia, this overall upregulation of hnRNP-C is due to expansion of the hyperplastic intima, with no detectable upregulation in the media. However, for atherosclerotic arteries, there is marked upregulation in the vessel wall overall, due to upregulation of hnRNP-C in the medial smooth muscle cells along with continued expression in the intimal cells.
The upregulation of hnRNP-C in intimal hyperplasia supports the concept that H2O2 generated in the vessel wall stimulates vascular cell proliferation by the activation of signaling pathways involving hnRNP-C [1-4, 6]. Interestingly, the upregulation of hnRNP-C in intimal hyperplasia was most pronounced in the inner half of the intima, adjacent to the endothelium, which may produce the majority of the H2O2 in the normal vessel wall [23,24]. This hnRNP-C staining pattern also correlates with the frequently reported biphasic nature of human intimal hyperplasia, in which the inner half appears more activated and more proteoglycan rich than the outer half .
There is good evidence that intimal smooth muscle cells are biologically distinct from medial smooth muscle cells [19,26,27]. However, interestingly there are relatively few markers that are specific to intimal smooth muscle cells compared to medial smooth muscle cells in human arteries with intimal hyperplasia [28-30]. Certainly, in atherosclerotic lesions and to a lesser extent in human intimal hyperplasia a small number of SMA+/cytokeratin-8+/cytokeratin-18+/desmin- myofibroblasts may be present, but the vast majority of SMA+ cells in routine spontaneous intimal hyperplasia are SMA+/desmin+ smooth muscle cells . The endothelial and stem cell marker CD34 does not differentiate between the two smooth muscle cell groups in human arteries, as typically neither stains for this marker . A recently reported marker that does appear to be upregulated in the intima smooth muscle cells of human intimal hyperplasia compared with medial smooth muscle cells is the calcium binding protein S100A4 . This marker is reported to be relatively specific for smooth muscle cells with a rhomboid phenotype over those with a spindle-shaped phenotype. Overall, there is similarity in the staining patterns of hnRNP-C and S100A4 in regards to the medial and intimal smooth muscle cells in the arteries with intimal hyperplasia, raising the question whether these two markers co-localize to the same smooth muscle cells in these arteries. However, a key difference between these two markers is that hnRNP-C is clearly upregulated diffusely in the medial smooth muscle cells in arteries with atherosclerosis, while S100A4 does not appear to show such diffuse upregulation with disease progression.
It will be important to ascertain precisely what the upregulation of hnRNP-C means. It is unlikely that hnRNP-C upregulation is an indication of actively proliferating cells. Previous studies with human arteries have shown that less than 5% of intimal smooth muscle cells of intimal hyperplasia and atherosclerosis are actively progressing through the cell cycle [34,35], indicating that most of the hnRNP-C+ smooth muscle cells observed here should be in G0 phase. However, hnRNP-C facilitates the expression of gene products positively regulating cell growth and survival including platelet-derived growth factor , c-myc , and the X-chromosome-linked inhibitor of apoptosis . Thus hnRNP-C upregulation may be best considered an indication of vascular cell activation rather than specifically proliferation. This would suggest that vascular smooth muscle cell activation is more pronounced in the intima during the preatherosclerotic intimal hyperplasia phase, with subsequent activation of the medial smooth muscle cells in conjunction with formation of the atherosclerotic lesion. It is unclear if hnRNP-C will be a useful marker for endothelial activation, as the endothelium essentially always expressed hnRNP-C in these arteries, even those arteries with normal histology. However, since the arterial location examined in this study was in fact a post-bifurcation site in a relatively atherosclerosis-prone artery, the endothelium may quite well have been activated in all of these arteries.
This work was supported by NIH Grant HL074324
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.