Abstract
The natriuretic peptides (NPs) ANF, BNP, and CNP have potent anti-proliferative and anti-migratory effects on vascular smooth muscle cells (SMCs). These properties make NPs relevant to the study of human coronary atherosclerosis because vascular cell proliferation and migration are central to the pathophysiology of atherosclerosis. However, the existence and cytological distribution of NPs and their receptors in human coronary arteries remain undetermined. This has hampered the development of hypotheses regarding the possible role of NPs in human coronary disease. We determined the pattern of expression of NPs and their receptors (NPRs) in human coronary arteries with atherosclerotic lesions classified by standard histopathological criteria as fatty streak/early atherosclerotic lesions, intermediate plaques, or advanced lesions. The investigation was carried out using a combination of immunocytochemistry (ICC), in situ hybridization (ISH), and semiquantitative polymerase chain reaction (PCR). Both by ICC and ISH, ANF was found in the intimal and medial layers of all lesions. BNP was highly expressed in advanced lesions where it was particularly evident by a strong ISH signal but weak ICC staining. CNP was demonstrable in all types of lesions, giving a strong signal by ISH and ICC. This peptide was particularly demonstrable in the endothelium, as well as in the SMCs of the intima, media, and vasa vasorum of the adventitia and in macrophages. By ISH, NPR-A was not detectable in any of the lesions but both NPR-B and NPR-C were found in the intimal and the inner medial layers. By RT-PCR, mRNA levels of all NPs tended to be increased in macroscopically diseased arteries, but only the values for BNP were significantly so. No significant changes in NPR mRNA levels were detected by PCR. In general, the signal intensity given by the NPs and their receptors by ICC or ISH appeared dependent on the type of lesion, being strongest in intermediate plaques and decreasing with increasing severity of the lesion. This study constitutes the first demonstration of NPs and NPR mRNAs in human coronary arteries and supports the existence of an autocrine/paracrine NP system that is actively modulated during the progression of atherosclerotic coronary disease. This suggests that the coronary NP system is involved in the pathobiology of intimal plaque formation in humans and may be involved in vascular remodeling.
Keywords
T
Three receptors (NPRs) have been described for the NPs. Types A and B are guanylyl cyclases, through which the ligands induce the production of cGMP. The type C receptor is a clearance receptor.
Because all three NP have potent anti-proliferative and anti-migratory effects on vascular smooth muscle cells (SMCs) (de Bold and Bruneau 2000) they have a potential role in the pathogenesis of the coronary atherosclerotic plaque. However, the existence and cytological distribution of NPs and their receptors in human coronary arteries remain largely undetermined. This has hampered the development of hypotheses regarding the possible role of NPs in human disease, with specific reference to coronary artery disease. To fill this gap, the present study was designed to determine the existence and distribution of NPs and their receptors in human coronary arteries with variable degrees of atherosclerosis, using immunocytochemistry (ICC) and in situ hybridization (ISH). We also used a semi-quantitative polymerase chain reaction (PCR) to quantify the transcript abundance for all three NPs and their receptors in coronary artery segments with different degrees of atherosclerotic disease. We identified the existence of an NP system that is actively modulated during the progression of atherosclerotic coronary disease. It is expected that these findings will contribute to the development of etiopathological concepts regarding the role of the coronary NP system in the development of atherosclerotic plaque in humans.
Materials and Methods
Patients
Hearts were obtained from patients undergoing orthotopic cardiac transplantation. There were six males and three females. The mean patient age was 51.5 years (range 33–64 years). Seven patients were transplanted for severe ischemic heart disease with multiple myocardial infarcts and two had dilated cardiomyopathy.
In all cases, the hearts were removed at surgery with excision at the atrioventricular valve rings. Immediately after excision, small lengths (0.5–1 cm) of right coronary artery, left anterior descending artery, and left circumflex artery were dissected and immediately fixed in fresh 4% paraformaldehyde in 0.1 M PBS, pH 7.2, at 4C for 2–4 hr. Tissue samples were obtained in compliance with institutional guidelines.
After fixation, the tissues were washed three times in PBS for 3.5 hr and dehydrated in a graded series of ethanol, cleared in toluene, and embedded in paraffin.
Four- to 5-μm sections were placed on Superfrost Plus slides for ICC or ISH. These vessels were classified as follows: (a) no atherosclerotic involvement; (b) early atherosclerotic coronary arteries (only minor intimal thickening with scattered CD68-positive macrophages just under the luminal surface); (c) intermediate atherosclerotic coronary arteries (fibrocalcific plaques with little lipid or necrosis); or (d) advanced atherosclerotic coronary arteries (thickened intima with many CD68-positive macrophages, regions with necrosis, and cholesterol deposits).
To determine the extent of medial changes and the possible participation of the myofibroblasts of the adventitia in neointimal development, the tissue sections were stained for elastic tissue using the Verhoeffvan Gieson's stain (VVG) or Masson's trichrome stain.
Immunocytochemistry for Cell Identification
Serial tissue sections were immunostained with the following antibodies: anti-rat PCNA for proliferating cells (mouse MAb IgG2a, clone PC10, dilution 1:100; Novo Castra Laboratories, Newcastle, UK); anti-human macrophage CD68 (MAb IgG3k, clone PG-M1 dilution 1:50; Dakopatts, Glostrup, Denmark; anti-smooth muscle actin HHF 35 (mouse MAb IgG1, dilution 1:20; Enzo Diagnostic, New York, NY); anti-desmin (rabbit anti-chicken smooth muscle desmin, dilution 1:50; Sigma, St Louis, MO); anti-vimentin (mouse MAb IgG1, dilution 1:50; Amersham, Arlington Heights, IL). These were applied at the indicated dilutions in 1.0% BSA in PBS and incubated in a humidified chamber for 60 min at room temperature (RT). The sections were washed in PBS and then incubated with a biotinylated secondary antibody [horse anti-mouse IgG at a 1:400 dilution or goat anti-rabbit IgG at a 1:200 dilution for the monoclonal or polyclonal antibodies, respectively (Vector Laboratories; Burlingame, CA)] in PBS containing 1.0% BSA and 2.0% normal serum (goat or horse). This was followed by washing the sections in PBS and incubation with the avidin-biotin enzyme complex and the chromogenic substrates according to the manufacturer's instructions. Control sections stained with non-immune serum from the appropriate species and stained with only secondary antibody were always run concurrently with the test sections. Positive controls consisted of atrial tissue sections for ANF and BNP and myofibroblasts and endothelium of microvessels in wound tissue sections for CNP.
Immunohistochemistry for NPs
The immunolocalization of ANF, BNP, and CNP was carried out as described above using polyclonal rabbit anti-ANF 99–126 (dilution 1:1000–1:2000) developed in our laboratory and previously characterized by immunocytochemistry and solution binding (de Bold and de Bold 1985; Sarda et al. 1989) and polyclonal anti-CNP22 and anti-BNP32 (Peninsula Laboratories, Belmont, CA; dilution 1:1000). In addition to the immunostaining controls described above, these antibodies were also pre-adsorbed with 1 μg/ml ANF 99–126, BNP32, and CNP22 (Peninsula Laboratories). Sections were also incubated with cross antigen pre-adsorbed antibodies and whole horse non-immune antiserum.
In Situ Hybridization
Rehydration and Blocking. The slides were deparaffinized with three 15-min changes in toluene and rehydrated through a graded ethanol series. After rehydration, the sections were denatured in 0.2 N HCl for 20 min at RT, heat-denatured 15 min at 70C in 2 × SSC, and washed in 1 × PBS for 2 min.
The sections were digested in Pronase for 15 min at 37C (100 μg/ml), rinsed in 2 mg/ml glycine in 1 × PBS for 30 sec at RT, and postfixed for 5 min at RT in freshly prepared 4% paraformaldehyde fixative. Postfixing was followed by treatment with 3 × PBS for 5 min at RT and two washes in 1 × PBS for 30 sec. The sections were then equilibrated in 10 mM DTT, prepared in preheated 1 × PBS for 10 min at 45C in a water bath, and immediately in blocking solution for 30 min at 45C. The sections were washed twice in 1 × PBS for 2 min at RT and equilibrated in freshly prepared TEA buffer for 2 min. To block free reactive groups, the sections were acetylated in 0.5% acetic anhydride in TEA buffer for 10 min at RT. The blocking steps were stopped by immersion in 2 × SSC for 5 min at RT. The sections were dehydrated through graded alcohols, dried under vacuum, treated with UV for 30 min, and stored in a slide box with desiccant at −70C overnight.
Probe Generation. The general experimental strategy for the generation of PCR-derived riboprobe self templates for ISH (Sitzmann and LeMotte 1993) to detect ANF, BNP, C-type NP, NPR-A, NPR-B, and NPR-C consisted of designing four overlapping oligonucleotides (Table 1) corresponding to human sequence (NCBI nucleotide database) portions of the gene-coding region of each peptide or receptor flanked by restriction sites and T3 or T7 phage RNA polymerase promoters used in PCR reactions as a self-priming template. After generation of free 3′ and 5′ ends by restriction endonucleases, the riboprobes, which in all cases were 161 bp long, were isolated and purified as described below.
The quantity of internal primers was highly limited and the resultant reaction causes an asymmetric single-stranded amplification of the two halves of the total sequence due to an excess of the two flanking primers. In subsequent PCR cycles, these dual asymmetrically amplified fragments, which overlap each other, yield a double-stranded full-length product. All external (1 and 4) (200 nmol) and internal (2 and 3) (2 nmol) primers were placed in six identical 100-μl PCR mixtures. Two initial cycles were carried out at 92C for 1 min 30 sec, 68C for 1 min, and 74C for 40 sec, followed by 25 cycles at 92C for 1 min 30 sec, 70C for 30 sec, and 74C for 40 sec, using a PCR cycler.
The PCR products were analyzed by electrophoresis in a 1.5% agarose gel, pooled, and precipitated with 1:10 volumes 3 M sodium acetate and 2.5 volumes 100% ethanol at −80C for at least 30 min. After precipitation the mix was centrifuged at 13,200 × g for 5 min and the supernatant discarded. After washing with 75% ethanol, the pellet was air-dried and resuspended in TE buffer, subdivided into two separate tubes, and digested with the corresponding enzymes (Table 1). After separation on a 1.5% agarose gel, both enzyme-cut cDNAs were OD read to calculate the final concentration to be used in the transcription step.
Primers for riboprobe construction a
P, primers; T3, T7, RNA polymerase promoters.
Indicates cut within recognition sequence (RS) (underlined) for the indicated restriction enzymes.
PCR primers
Transcription. For 35S-radiolabeled antisense or sense probe generation, a Riboprobe in vitro Transcription System (Promega; Madison, WI) was used according to the technical manual. The following components were added at RT in the order indicated: 5 × Transcription Optimized Buffer, 100 mM DTT, 20 U RNasin ribonuclease inhibitor, 2.5 mM (each) ATP, CTP, and GTP, 100 mM UTP, 1 μg cDNA sense or antisense in TE buffer, 20 μCi/μl [35S]-UTP (Amersham), and 15–20 U T3 or T7 RNA polymerase to obtain the sense or antisense riboprobe, respectively. The mix was centrifuged for 5 sec and incubated for 1 hr at 37–40C. After incubation, 1 μl from each reaction was removed to determine the percent incorporation into the TCA precipitate. The specific activity of the probes was expressed as the total incorporated cpm/total μg of RNA synthesized.
The cDNA template was removed by digestion with DNase I after the transcription reaction. In each reaction, RQ1 RNase-free DNase was added to a concentration of 1 U/μg of template cDNA used and incubated for 15 min at 37C. After this digestion the probes were extracted by addition of 1 volume of TE-saturated phenol, vortexed for 1 min, and centrifuged at 12,000 × g for 2 min. The upper phase was transferred to a fresh tube and 1 volume of 24:1 chloroform:isoamyl alcohol was added. After mixing for another minute, the mix was centrifuged at 12,000 × g for 2 min. The upper phase was transferred into a fresh tube and 0.5 volumes of 7.5 M ammonium acetate and 2.5 volumes of 100% ethanol were added. The mix was placed at −80C for at least 30 min and centrifuged at 13,200 × g for 20 min. The supernatant was discarded and the pellet was washed with 1 ml of 70 % ethanol and vacuum-dried.
In each case the transcripts were visualized by standard denaturing gel electrophoresis using a pGEM system (Promega) as positive control, following the manufacturer's instructions. The probes were dissolved in 50 mM DTT and heat-treated to 100C for 1 min in a water bath. The hybridization solution (50% formamide, 0.3 M sterile NaCl, 10 mM Tris-HCl, pH 8.0, 1 mM EDTA, 1 × Denhardt solution, 500 μg/ml yeast tRNA, 500 μg/ml poly(A) (Pharmacia; Uppsala, Sweden), 50 mM DTT, and 10% polyethylene glycol (MW 6000 EM grade) was added to obtain a 0.3 μg/ml final probe concentration. This mix was vortexed for 1 min, centrifuged at 12,000 × g for 5 min, and counted (~1 × 105 cpm/μl).
Hybridization. The hybridization mixture was spread on the slides (approximately 20 μl/mm2) and these were placed in a moist chamber containing 50% formamide, 0.3 M NaCl, 10 mM Tris-HCl, pH 8.0, and 1 mM EDTA and incubated at 45C for 4 hr in a water bath. After hybridization the sections were washed twice at 55C with wash Solution A containing 50% formamide, 2 × SSC, and 20 mM mercap-toethanol for 15 min each; twice at 55C with wash Solution B containing 50% formamide, 2 × SSC, 20 mM mercapto-ethanol, and 0.5% Triton X-100, and twice for 2 min each with wash Solution C containing 2 × SSC and 20 mM mercaptoethanol at RT. After these washing steps the slides were treated with 500 μl of RNase digestion solution containing 40 μg/ml RNase A, 2 μg/ml RNase T1, 10 mM Tris-HCl, pH 7.5, 5 mM EDTA, and 0.3 M NaCl for 15 min at RT. After the RNase incubation, the slides were washed twice with gentle shaking for 30 min each with wash Solution C at 50C, twice for 30 min each with wash Solution A at 50C, and twice for 5 min each in 2 × SSC at RT.
The slides were dehydrated through 50%, 70%, and 90% ethanol/0.3 M ammonium acetate, 100% ethanol, and air-dried for 30 min.
For detection of hybridized probes the slides were dipped into a dipper chamber containing a 1:1 dilution of NTB2 Kodak emulsion and placed vertically to dry for 2 hr. The slides were placed in a light-tight slide box with desiccant for 1–2 weeks. The slides were developed in Kodak D19 at 15C, fixed, stained, and mounted for microscopic analysis.
Riboprobe specificity for ANF and BNP was established using atrial (positive) and ventricular (negative) tissue sections as well as Northern blotting of RNA to show the appropriate size of the transcripts and their absence in liver. CNP riboprobe performance was determined by Northern blotting and by its ability to demonstrate CNP mRNA in wound tissue section myofibroblasts and in endothelial cells of small vessels. NPR positive controls were carried out on normal rat kidney tissues. All exhibited a moderate ISH signal, which was suppressed using RNase treatment as described.
Total RNA Extraction
Tissue samples were extracted using Trizol (GIBCO BRL; Burlington, ON, Canada). To remove contaminating genomic DNA, RNA samples were incubated with 1 U of RNase-free DNase (Promega) for 10 min at 37C in 50 μl of a buffer containing 40 mM Tris-HCl (pH 7.9), 10 mM NaCl, 6 mM MgCl2, and 10 mM CaCl2. After DNase treatment, the samples were re-extracted with 10 volumes of Trizol and the RNA was quantified by spectrophotometry.
Semi-quantitative RT-PCR
RNA samples were reverse-transcribed using Super Script II RNase H-Reverse Transcriptase and oligo(dT)12–18 primer of a reverse transcription kit (GIBCO BRL). One fifth of the cDNA product was used for PCR amplification using the primers shown in Table 2. For primer design and calculation of optimal annealing temperatures, Oligo software (National Biosciences; Plymouth, MI) (Rychlik and Rhoads 1989; Rychlik et al. 1990) was employed. PCR reactions were conducted in a final volume of 50 μl containing PCR buffer (10 mM Tris-HCl, pH 8.3, and 50 mM KCl), 0.2 mM of each deoxynucleotide triphosphate, 0.4 μM primers, 1.25 U of Ampli Taq DNA Polymerase (Perkin Elmer; Mississauga, ON, Canada) followed by HotWax Mg2+ beads (Invitrogen; San Diego, CA). The reactions were heated at 95C for 2 min and cycled 40 times through a 60-sec denaturing step at 95C, a 45-sec annealing step at 58C, and a 90-sec extension step at 72C. After the final cycle, a 10-min extension step at 72C was included. These conditions, except for the annealing temperature, were used for all amplifications. Aliquots (5 μl) of the PCR product were electrophoresed on a 2% agarose gel and were visualized by ethidium bromide staining. Negative photographs were taken using Polaroid 55 film (Cambridge, MA). The negatives were scanned using an Ultrascan XL laser densitometer and Gelscan XL 2000 software package. Pilot experiments (not shown) were carried out to determine time and concentration linear range of the reactions. Results were normalized to the β-actin signal for each sample.
Results
A summary of findings for the different atherosclerotic lesions with respect to the ICC and ISH for the NPs and their receptors, as well as for cellular markers, is provided in Table 3.
Fatty Streaks/Early Atherosclerotic Lesions
These lesions consisted primarily of foamy macrophages, mononuclear inflammatory cells, and fibrointimal tissue intermingled (Figure 1A). In this degree of lesion, only the intimal cells were moderately immunopositive for the anti-PCNA antibody. Anti-smooth muscle actin (SMA)-positive cells were found in the medial and intimal regions. Both desmin and vimentin were highly positive, both in the intimal myofibroblasts and adventitia, vasa vasorum, and fibroblasts. Using the antibody to CD68, we were able to identify macrophages in the subendothelial region of the intima. Moderate immunolabeling for ANF was found in the intima and, at a very low level, in the media (Figure 1B). The adventitia demonstrated ANF-immunopositive cells in the vasa vasora. By ISH, no ANF or BNP mRNA was detected in these early lesions, nor was there BNP immunostaining. Anti-CNP22 antibody demonstrated strong CNP peptide expression in the luminal endothelium and less intense labeling in the medial SMCs and intimal cells (Figure 1C). CNP mRNA signal was very strong in the medial SMCs (Figures 1D and 1E). The CD68-positive macrophages and monocytes also displayed strong CNP immunolabeling. NPR-A, −B, and −C were not detectable by ISHR.
Summary findings for NPs and their receptors in coronary artery lesions a
Signal strength: +++++, very strong; ++++, strong; +++, moderate; ++, low; +, very low; −, not detectable.
(
Intermediate Plaques
These lesions are atheromatous plaques (Stary et al. 1995) and are characterized by accumulation of extracellular lipids that occupy an extensive proportion of the intimal mass. This type of extracellular lipid accumulation, known as the lipid core, may cause architectural reorganization of the vascular wall, with thinning of the media and remodeling of the artery (Figure 1F).
High proliferative activity was detected surrounding the lipid core in the intima. About half the nuclei showed positive immunoreactivity for the anti-PCNA antibody in such regions. A few medial SMCs in these specimens were localized near the internal elastica lamina and were PCNA-positive. At this stage we could not detect PCNA immunoreactivity in the endothelium or the adventitia. Overall, the fibrofatty plaques showed strong immunoreactivity for SMA, desmin, and vimentin in the intimal myofibroblasts, and the medial SMCs exhibited moderate staining. The vasa vasora walls in the adventitia also stained with these antibodies.
CD68-positive macrophages were identified surrounding and also inside of the lipid core. CD68 staining appeared to correspond to that of ANF immunoreactivity at the intimal level, in the lipid core regions, and in the adjacent media. ANF immunoreactivity had a very similar distribution to that of ANF mRNA by ISH (Figures 1G and 1H). BNP immunoreactivy or mRNA by ISH were not evident in these lesions. The endothelium, including that of the lumen, plaque microvessels, and adventitial vasa vasorum, exhibited strong immunopositivity for CNP (Figure 1I). The luminal endothelial cells and plaque macrophages gave a positive signal with the CNP antisense riboprobe. At the intimal level, the CNP mRNA and peptides were localized mainly in the microvessels next to the lipid cores, which are postulated to play an important role in plaque biology, including plaque progression and remodeling. The medial SMA-positive cells exhibited moderate immunoreactivity for CNP and a relatively low signal of CNP mRNA.
NPR-A mRNA by ISH was not detectable but NPR-B was demonstrable in the intima of the plaques. NPR-C mRNA hybridization signal was very strong at the intimal level and less so in the medial and endothelial layers.
Advanced Atherosclerotic Lesions
These vessels had a variable appearance but had well-developed lipid cores, with necrosis and calcification that were more marked than in the less advanced vascular lesions studied (Figure 2A). PCNA-positive cells were mainly observed surrounding the lipid core and/or the calcified plaque, and in the intima. SMA, desmin, and vimentin were also positive in the intimal cells and, less conspicuously, in the SMCs of the media. The macrophages of the plaque were immunolabeled in the subendothelial region and around the calcified areas of the intima. Clusters of macrophages were also identified in the adventitia.
ANF immunostaining was found in the intimal layer, both in intimal myofibroblasts and macrophages. Lesser staining was found in the medial SMCs. ISH for ANF mRNA had a similar distribution but reached a very pronounced level in the adventitial layer, mainly in the vasa vasorum (Figure 2B). BNP immunoreactivity was weak overall, despite a very strong signal detected by ISH in the intima and adventitia (Figure 2C). The immunoreactivity observed in the cells of the intima at high power corresponded to the hybridization signal (Figures 2D and 2E). CNP immunoreactivity was observed in endothelial cells of these lesions and of the adventitial vasa vasora and intimal microvessels of the plaque. In advanced atherosclerotic lesions, a moderate hybridization level for NPR-B and NPR-C was observed.
(
Semi-quantitative RT-PCR in Coronary Arteries
Figure 3 shows semi-quantitative PCR values obtained for mRNAs encoding for NPs in samples grouped as either normal or diseased vessels, as assessed by gross appearance and microscopy of the area adjacent to sampling. Average values for NP mRNAs were increased in diseased arteries, although a statistically significant increase was demonstrable for BNP only. All three receptor types were detected by RT-PCR but no significant changes were found in their expression compared to that in normal vessels (data not shown).
Discussion
Atherosclerosis is a complex process involving the interaction of many cell types and growth factors. Increasingly, it is realized that this process is active, with vascular wall remodeling and plaque progression and regression, based on the activity of the plaque constituent cells (Fuster et al. 1992a, b).
The NPs ANF, BNP, and CNP possess biological properties compatible with a role in growth regulation of the cardiovascular system (for review see de Bold and Bruneau 2000), but there is no knowledge concerning the location and cell type(s) associated with the NPs and their receptors in the human arterial wall. In the present work, for the first time to our knowledge we demonstrate by ISH and ICC the existence of an NP system in atherosclerotic human coronary artery vessels.
It can be anticipated that because NPs antagonize the action of the renin-angiotensin system (RAS) (de Bold and Bruneau 2000), circulating or locally expressed NPs would modulate the cardiovascular growth processes that are dependent on locally expressed or circulating systems, such as the RAS, and that are present in the vasculature (Linz et al. 1989). For example, ANF is capable of inhibiting thymidine incorporation by rat mesangial cells (Appel 1988; Vanneste et al. 1988), and vasodilators that are capable of elevating intracellular cGMP levels, such as sodium nitroprusside and ANF, inhibit the proliferation of cultured SMCs (Kariya et al. 1989). In addition, ANF can inhibit, in a dose-dependent manner, PDGF-stimulated thymidine incorporation by SMCs (Abell et al. 1989) and may act as an anti-growth factor for endothelial cells (Itoh et al. 1992).
Semi-quantitative PCR values obtained for mRNAs encoding for NPs in samples grouped as either normal vessels or as diseased vessels, as assessed by gross appearance and microscopy of the area adjacent to sampling (n = 9 each). ∗ p<0.05.
We found immunolabelling for ANF increasing from early atherosclerotic lesions to the intimal or medial layers of intermediate plaques. However, in advanced atherosclerotic lesions both peptide and mRNA were restricted to intimal myofibroblasts and microvessels of the intima and vasa vasorum, suggesting that disease progression entails the downregulation of locally expressed ANF. The microvessels of the vascular wall and the cells of the adventitia are believed to play a major role in vascular plaque progression and arterial remodeling. The localization of NPs in these advanced lesions would support the view that they are involved in these processes.
BNP was found in the vascular wall intima, giving a strong signal by ISH and a weak one by ICC, suggesting that the production of BNP in vascular lesions is mainly constitutive in nature. BNP mRNA levels by RT-PCR were significantly increased in vessels with macroscopic evidence of atherosclerotic lesions. In fact, all three NP mRNA levels, as determined by RT-PCR, tended to be increased (Figure 3), and it is therefore possible that a larger study would have demonstrated an association of the level of expression of NP with the nature of the lesion. However, our samples for RT-PCR studies contained a mix of lesions, from intermediate to advanced lesions. Only in the advanced lesions did BNP by ISH appear strongly upregulated (Table 3). Therefore, the sample population for RT-PCR of diseased arteries had an inherent large statistical variance.
Because both ANF and BNP signal through the same NPR-A receptor and their biological properties are similar (de Bold and Bruneau 2000), differential roles for these peptides in atheromatous plaque formation are not obvious. However, there is evidence that BNP may act locally as an anti-fibrotic factor (Ogawa et al. 2001) and that some of the actions of BNP may be mediated by an NPR preferential to BNP over ANF (Goy et al. 2001).
As discussed above for ANF and BNP, a similar postulation for the involvement of CNP vascular growth can be made. CNP is expressed in the vascular endothelium (Heublein et al. 1992; Kohno et al. 1992; Porter et al. 1992; Stingo et al. 1992a, b; Suga et al. 1992) and is a potent growth inhibitor of SMCs. In addition, the production of CNP by aortic bovine endothelial cells in culture is potently stimulated by TGF-β (Suga et al. 1992), which is highly expressed at the site of vascular injury. Furuya et al. (1993) reported that continuous infusion of CNP inhibits intimal proliferation induced in rats after injury to the common carotid artery. These authors found that the type B NP receptor, which preferentially binds CNP, is expressed at the site of the vascular injury, and therefore constitutes a phenotypic change that appears to favor interaction with endothelium-derived CNP. More recently Brown et al. (1997), using ICC, RIA, and RT-PCR, reported a new source of CNP in the neointima of rat carotid arteries 14 days after balloon angioplasty. These authors also demonstrated that the neointima expresses the type C receptor concurrently with the synthesis of CNP. The data presented in the present work generally support these reports by demonstrating that, in humans, CNP and the receptor types B and C do undergo changes in expression in relation to the progression of the atherosclerotic lesion. Furthermore, by ICC and ISH, CNP was the most prominent among all the NPs in the tissues studied, even though CNP levels by RT-PCR relative to the housekeeping gene β-actin are lower than those of ANF and BNP. The prominence of CNP visualization may be due to the fact that CNP peptide and mRNA are highly localized in endothelial cells. Plaque micro-vessels and adventitial vasa vasora were also clearly demonstrable sites of CNP expression. This finding strongly suggests a role for this peptide in vascular remodeling and in progression/regression of the plaque of the diseased arteries.
By ISH, NPR-A was not found in diseased vessels, but both NPR-B and NPR-C were present in intermediate and advanced atherosclerotic lesions. However, transcripts for all three types of receptors were detected after RT-PCR. The NPR-B transcript was less evident by ISH as the plaque became more advanced. This finding is consistent with the fact that NPR-B is associated with the secretory non-contractile pheno-type of vascular smooth muscle, which would be more prominent in atherosclerotic lesions. This type of intimal cell is a significant component responsible for the formation of the plaque extracellular matrix components which, in turn, is of relevance for plaque changes including progression and regression and complications including rupture, hemorrhage, erosion, and calcification. The decrease in demonstrable NPs and their receptors in advanced lesions may underlie a failure of the local NP system to exert an inhibitory effect on the progression of plaque development.
In summary, the present study shows the presence of a natriuretic peptide system in the human coronary atherosclerotic plaque as evidenced by the presence of both mRNA and its translation products for all three NPs and their receptors. Furthermore, strong evidence was found to indicate the active modulation of NPs and their receptors in the vascular wall in accordance with the degree of the lesion. This suggests that the coronary NP system is involved in regulation of intimal plaque formation in humans and may be involved in vascular remodeling. It can be anticipated that these concepts are applicable to the processes of vascular re-stenosis, including in-stent re-stenosis and angioplasty injury, and in transplant-associated graft vascular disease. The NPs and their receptors thus form a part of a complex group of vascular wall growth mediators such as TGF-β, PDGF, endothelin, and angiotensin II. Manipulation or therapy targeting the NPs, their receptors, or their clearance pathways would be of interest in many vascular disorders.
Footnotes
Acknowledgments
Supported by the Canadian Institutes for Health Research and the Heart and Stroke Foundation of Ontario. VHC is a postdoctoral fellow of the Consejo Nacional de In-vestigaciones Científicas y Técnicas (CONICET) and National University of Entre Ríos, Argentina.
The technical support of Carole Frost and Amalia Ponce is gratefully acknowledged.