Angiotensin II Type 1 and 2 Receptors in Conduit Arteries of Normal Developing Microswine
Objective— To identify vascular cells capable of responding to angiotensin II (Ang II) generated in conduit arteries, we examined the Ang II type 1 receptor (AT1R) and Ang II type 2 receptor (AT2R) in the thoracic aorta (TA) and abdominal aorta (AA) and branches in 90-day fetal, 3-week postnatal, and 6-month adult microswine.
Methods and Results— By autoradiography (125I-[Sar1Ile8]-Ang II with or without AT1R- or AT2R-selective analogues or 125I-CGP 42112), there were striking rostrocaudal differences in (1) AT2R binding at all ages (prominent in AA wall and branches, sparse in TA wall and branches) and (2) a non-AT2R binding site for CGP 42112 (consistently evident in postnatal TA and branches but absent in AA and branches). Furthermore, patterns of AT2R distribution in infradiaphragmatic arteries were developmentally distinct. In fetal AAs, high-density AT2Rs occupied the inner 60% of the medial-endothelial wall. In postnatal AAs, AT2Rs were sparse in the medial-endothelial wall but prominent in a circumferential smooth muscle α-actin–negative cell layer at the medial-adventitial border, occupying ≈20% to 25% of the AA cross-sectional area. AT1R density in the TA and AA medial-endothelial wall increased with age, whereas AT2R density decreased after birth.
Conclusions— A novel AT2R-positive cell layer confined to postnatal infradiaphragmatic arteries physically links adventitial and medial layers, appears optimally positioned to transduce AT2R-dependent functions of local Ang II, and suggests that adventitial Ang II may elicit regionally distinct vascular responses.
The physiological and pathophysiological functions of large conduit arteries have received increasing attention as modulators of cardiovascular risk.1 Considerable evidence points to a local vascular renin-angiotensin system (RAS) regulated independently of the classic circulating RAS.2,3⇓ Emerging evidence also supports an effect of the Ang II type 2 (AT2) receptor (AT2R) on normal vasomotor tone via an NO/cGMP vasodilator pathway.4 However, there is little information on the identity of the cell type(s) that expresses AT2R in the vascular tree, how that cell is regulated, or how it, in turn, modulates vascular functions in health and/or disease states.
Large conduit vessels, particularly at their branch points,5,6⇓ are the sites of the atherosclerotic process endemic in westernized societies. Increasing data suggest that vascular injury and/or inflammation is accompanied by increased expression of the Ang II type 1 (AT1) receptor (AT1R) and AT2R.7,8⇓ Furthermore, Ang II has been implicated in the progression of these lesions,9 whereas ACE inhibition or AT1R blockade has been linked to reduced cardiovascular risk in clinical trials.10
Swine provide a uniquely suitable model for the study of human cardiovascular diseases because they exhibit similar responses to a wide variety of physiological and pathological stimuli.11 Accordingly, to initiate the study of RAS contributions to large-artery function in a swine model, we have quantified the developmental changes in the density and distribution of AT1Rs and AT2Rs in the thoracic aorta (TA) and abdominal aorta (AA) and their major branches in normal fetal, neonatal, and adult microswine. Our findings indicate major developmental differences in the density of AT1R and in the density and rostrocaudal distribution of AT2R and a novel circumferential AT2R-positive/smooth-muscle α-actin–negative cell layer at the medial-adventitial (ME) border in infradiaphragmatic arteries.
Please refer to http://atvb.ahajournals.org for online Methods supplement.
Time-bred normal microswine were obtained from Charles River Laboratories, Wilmington, Mass, and studied according to protocols approved by the Oregon Health and Science University (OHSU) Institutional Animal Care and Use Committee (No. A439). Piglets at 90 days of gestation, sequentially delivered by cesarean section in anesthetized sows (pretreated with 8 mg/kg Telazol [tiletamine HCl/zolazepam], intubated, and maintained on 3% to 4% isoflurane), were immediately euthanized, and TAs and AAs plus their branches were quickly removed and immersed in ice-cold PBS. In 3-week-old pigs (n=5) and 6-month-old pigs (n=10 in initial series, 5 in second series), which were anesthetized via masking with 5% isoflurane for induction and 3% to 4% for maintenance, aortas were removed via midline thoracoabdominal incision. Tissue for radioligand binding and for frozen sections was kept in ice-cold PBS for 3 to 4 hours before processing.
Film Autoradiography for Ang II Receptors
Autoradiography for AT1R and AT2R binding was performed within 24 hours of tissue harvest; this time frame was based on preliminary observations of decreased AT1R, but not AT2R, binding after storage (S. Bagby, unpublished data, 2002) and on reported studies.12 Tissue was frozen in OCT (Miles Laboratories), and 20-μm frozen sections were prepared by standard methods. Sections were thawed, preincubated for 30 minutes at room temperature in isotonic buffer (150 mmol/L NaCl, 50 mmol/L Tris, 50 μmol/L Plummer’s inhibitor [carboxypeptidase inhibitor], 20 μmol/L bestatin [aminopeptidase inhibitor], 5 mmol/L EDTA, 1.5 mmol/L 1,10-phenanthroline, and 0.1% heat-treated protease-free BSA, pH 7.4), and then incubated for 25 minutes in 1 of 4 solutions in isotonic buffer: 0.4 nmol/L 125I-[Sar1Ile8]-Ang II (*SIAII) for total Ang II receptor binding, *SIAII plus 10−6 mol/L [Sar1]-Ang II (nonselective analogue) for nonspecific binding, *SIAII plus 10−6 mol/L valsartan (Val, an AT1-selective antagonist) to demonstrate AT2R, and *SIAII plus 10−6 mol/L CGP 42112 (AT2 selective) to demonstrate AT1R. To independently examine AT2R, we also exposed a second set of 4 serial sections to the following: 0.4 nmol/L 125I-CGP 42112 (*CGP) for total AT2R, *CGP+[Sar1]-Ang II for nonspecific binding, *CGP+Val to confirm the absence of AT1R binding, and *CGP+PD123319 (also AT2R selective) to confirm AT2R specificity of CGP 42112 binding in microswine. (All radioligands were purchased from Peptide Radioiodination Service Center.) Sections were rinsed, dried, placed on Kodak Biomax MR1 x-ray film, and maintained at −80°C for 4 weeks. A standard slide with calibrated concentrations of 125I (1 to 600 nCi/mg 125I, Microscales, Amersham Biosciences) was included in each cassette. Film was developed with the use of an automated Kodak film processor.
Images were captured by use of an AIS image analysis system (Imaging Research Inc) via an analog video camera. The density of Ang II receptors was referenced to the 125I standards processed with each film. For diffusely distributed AT1Rs, density was averaged over the total ME wall cross-sectional area. For AT2Rs, which occupied only a portion of the cross-sectional area, a threshold setting was used to assess the average density within the aggregate AT2-positive area and to also estimate percentage of the cross-sectional area occupied by AT2Rs (“fractional area”). Because results with *SIAII+Val were qualitatively identical but of lower resolution compared with results with *CGP alone, formal analysis in infradiaphragmatic arteries was based on *CGP for quantification of AT2Rs and *SIAII+CGP for quantification of AT1Rs, each assessed with subtraction of relevant nonspecific binding. Because of a non-AT2R *CGP binding site in the medial wall of postnatal supradiaphragmatic conduit arteries (see Results), AT2R quantification in neonatal and adult TAs and their branches was based on *SIAII+Val autoradiography. Image analysis data were statistically evaluated (SigmaStat, version1.01, SPSS Inc) via 2-way repeated-measures ANOVA, with age and AA vascular region as factors.
Ang II Receptor Binding in TA Membranes
In our initial series, adult TAs were opened longitudinally and pinned, the endothelium was removed by gentle scraping, and the underlying media was stripped from the adventitia (yielding “medial strips”). In pooled fetal TAs (male versus female, 2 to 4 TAs per pool), medial stripping was not technically feasible; therefore, intact segments of aortas were used after loose adventitia was carefully removed. Additional studies in adults used paired segments of intact TAs versus AAs from 5 pigs. All aortic tissues were homogenized in homogenizing buffer (10 mmol/L Tris-HCl [pH 7.4], 1 mmol/L EDTA, and 0.1 mmol/L bacitracin), and a crude membrane fraction was made by 100 000g ultracentrifugation. Protein was assayed (Bio-Rad), and samples were diluted with assay buffer (50 mmol/L Tris-Cl [pH 7.4], 150 mmol/L NaCl, 1 mmol/L EDTA, and 0.1 mmol/L bacitracin) to adjust the concentration. Protein added per assay well was 20 to 50 μg for fetal samples and 50 to 80 μg for adult samples. The binding assay was based on 150 μL final volume: 15 μL 1.5% BSA, 15 μL radioligand (added at ×10 final concentration), 15 μL competing ligand, 15 μL [Sar1]-Ang II (final 200 nmol/L) in nonspecific wells, assay buffer to equalize well volumes, and 25 μL protein (added last to initiate the reaction). Plates were incubated for 90 minutes at 37°C and harvested in a Brandel Cell Harvester onto Whatman GF/C filter paper presoaked in 2% BSA. Each plate was rinsed with 3 vol of ice-cold 0.9% NaCl plus an additional three 200-mL NaCl (0.9%) rinses of each filter. Filters were counted in a Beckman γ-counter.
Each binding assay included the following: (1) an 8-point *SIAII saturation curve (range 0.07 to 0.55 nmol/L *SIAII, performed in triplicate with a single *SIAII+[Sar1]-Ang II well for nonspecific binding at each concentration), (2) a *CGP saturation curve (range 0.10 to 2.0 nmol/L *CGP with nonspecific wells), and (3) *SIAII displacement curves with the use of either CGP 42112 (AT2 displacement), Val (AT1 displacement), or Ang II (AT1 and AT2 displacement), with each performed in duplicate with a singlet nonspecific binding well. Specific binding curves were generated in PRISM (GraphPad Software), and the following parameters were statistically analyzed (Sigma Stat) for the effects of age (fetal versus adult): (1) maximum binding (Bmax) for *SIAII (AT1+AT2), for *CGP (AT2R), and for AT1R by difference; (2) Kd for each radioligand; (3) percent AT1 with the use of Val displacement curves; (4) percent AT2 with the use of CGP 42112 displacement curves; and (5) −log IC50 for each displacing Ang II analogue.
Immunohistochemistry was performed on 20-μm cryostat sections adjacent to sections processed for autoradiography. Sections were fixed in freshly prepared 1% paraformaldehyde/PBS for 10 minutes at room temperature. Staining was performed by using a standard avidin-biotin protocol as previously described,13 except that all incubations and washes were carried out on an automated immunostainer (Dako Corp). Mouse monoclonal IgG to smooth muscle α-actin (SMα-actin, Dako Corp, M0851) was used at 1:125 dilution; anti-factor VIII–related antigen antibody (Dako Corp) was used at 1:4000 dilution; and anti-neurofilament (low molecular mass [68-kDa] subunit, Zymed) was used at 1:4000 dilution.
AT1Rs were undetectable or very low in fetal TA and AA walls. In 3-week-old neonates, AT1R density was increased ≈3-fold over fetal levels (Table 1), was similar along the entire length of the aorta, and was typically localized to the inner two thirds of the ME wall. Of note, in neonates and adults with clearly visible binding of AT1R and AT2R, the respective binding sites defined mutually exclusive layers of the AA wall (Figure 1). Adult AT1R density (Table 1) in TAs and AAs was higher than fetal and neonatal density levels. In the fetus, perivascular mesenchyme was strongly and diffusely positive for AT1R, although true vascular adventitia (immediately adjacent to the media of AA and major branches) was typically devoid of AT1Rs (Figure 1). In postnatal tissues, AT1R binding in perivascular tissues and true adventitia was sparse and was confined to discrete microstructures within the connective tissue.
Autoradiography: Rostrocaudal Differences in a Non-AT2R *CGP Binding Site in Postnatal Conduit Arteries
In postnatal (but not fetal) TA (at 3 weeks and at 6 months of age), *CGP exposure resulted in dense binding, which was diffusely present in the TA ME wall but absent in the AA wall (Figure 2). This *CGP binding site was not displaced by cold [Sar1]-Ang II (Figure 2), indicating a non-AT2R/non-AT1R interaction. Similarly, all primary branches of the TA tested (subclavian, carotid, and internal mammary arteries) exhibited an identical non-AT2R *CGP binding site (Figure 2), whereas all AA branches lacked this site (Figures 1 and 2⇑), indicating a striking regional difference in supradiaphragmatic versus infradiaphragmatic conduit arteries. A non-AT2R binding site for *CGP has also been described by Viswanathan et al.14 As a consequence, AT2R evaluation in postnatal TAs and their branches is based solely on *SIAII+Val autoradiography.
Fetal Aortic AT2Rs
The ME wall of the fetal TA lacked AT2Rs except as isolated punctate foci of high AT2R density that were confined to the outer third of the medial wall (Figure 1) and occupied only a tiny fraction (average 1.7%) of the TA wall cross-sectional area (Table 2). On the basis of immunohistochemical localization of vascular and neural elements, the AT2R-positive microfoci in the TA ME wall colocalized with the vasa vasorum and/or the associated neural microfibers. Fetal TA adventitia and surrounding loose connective tissue/mesenchyme were diffusely positive for AT2R binding (Figure 1).
In contrast, the fetal AAs exhibited high-density AT2R binding in the inner ME wall throughout its length: AT2R density averaged 168±72 fmol/g and occupied the inner 60±10% of the ME wall cross-sectional area (Figure 1, Table 2). At fetal AA branch points, AT2R distribution was radially asymmetric, with intensification of AT2R binding at sites of branch-artery takeoff (fetal AA, Figure 1). Perivascular mesenchyme of fetal AAs exhibited AT2R binding in discrete structures, unlike the diffuse AT2R binding in the TA periaortic mesenchyme.
AT2Rs in Adult Aortas
In adult TAs (via *SIAII+Val autoradiography), minimal AT2R binding was detectable in the ME layers of the vessel wall (Figure 1, Table 2). In surrounding connective tissue, AT2R binding was noted in discrete microstructures, which included nerves (see online Figure I, which can be accessed at http://atvb.ahajournals.org). In contrast, adult AAs showed clear AT2R binding (Figures 1 and 2⇑). AT2Rs in adult AAs differed from fetal AAs in that binding occurred in an anatomically distinct, continuous circumferential layer appearing on cross section as a band at the medial-adventitial interface with a typically “beaded” appearance (Figures 1 and 2⇑). In interbranch AA segments, the AT2R-positive cell layer occupied 23±6% of the adult AA wall cross-sectional area, with an average density of 88±49 fmol/g, which was lower than AT2R density in the fetal AA medial wall (P<0.001, Table 2). At adult AA branch points, there was (as in fetal AAs) a distinct increase in fractional area occupied by the AT2R-positive layer as it extended into the branch-artery wall, maintaining the same circumferential pattern at the medial-adventitial boundary in the wall of the branch vessel (Figures 2 and 3⇓).
To determine whether AT2Rs at the medial-adventitial-boundary in adult AAs localized to typical medial wall vascular smooth muscle cells (VSMCs) or to perivascular nerves, we compared adjacent frozen sections that were immunostained for SMα-actin or neurofilaments or processed for AT2R autoradiography. On the basis of identically enlarged and overlaid images, the AT2R-positive cells at the medial-adventitial border were negative for neural elements (not shown) and for SMα-actin (Figure 3). However, the AT2R-positive cells appeared to extend into and occupy the “interstitial” spaces located between the longitudinal muscle bundles in the outermost medial wall layer, as described by Frid et al.15 The resulting interdigitating pattern appears to account for the beaded appearance (see Figures 1 and 3⇑ for examples).
AT2Rs in Neonatal Aortas
To ascertain timing of the shift from the fetal to adult pattern of AT2R binding, we examined AT2Rs by autoradiography in 3-week-old piglets. As in fetal (but not adult) TAs, AT2Rs were present within the ME wall of neonatal TAs only as scattered punctate foci, localized by immunostaining to vasa vasorum/neural complexes. In neonatal TA periaortic connective tissue, AT2R binding was also prominent in nerve trunks (neonatal TA, Figure 1; also see online Figure I). There was no residual evidence of the fetal pattern of inner ME wall AT2Rs in neonatal AAs; the adult pattern of AT2R distribution was present but was less well developed (Figure 1). Thus, the transition from the fetal to the neonatal/adult pattern of AA AT2R distribution occurs before 3 weeks of age in developing microswine. By image analysis, AT2R density in neonatal AAs was lower than that of fetal and adult AAs (P<0.004), whereas the fractional area occupied was less than that for fetal AAs (P<0.001) but was not significantly different from the adult pattern (Table 2). Of note, in neonates and adults with clearly visible binding of AT1Rs and AT2Rs, the respective binding sites define a mutually exclusive layer of the AA wall (neonatal AA and adult AA, Figure 1).
AT2Rs in Branch Arteries of TAs Versus AAs
The AT2R pattern in aortic branch arteries consistently mimicked the parent aortic pattern in each age group. Thus, there were minimal AT2Rs detectable in the ME wall of TA primary branches (subclavian, carotid, or internal mammary artery) at any age, whereas AA branches (celiac, superior mesenteric, renal, and iliac arteries) exhibited abundant AT2Rs in the inner ME wall in the fetus (Figure 1, Table 2) or showed discrete AT2R binding in a circumferential medial-adventitial cell layer in 3-week-old and adult pigs (Figures 1 and 2⇑). Accordingly, as with expression of the non-AT2R CGP binding site, the unique rostrocaudal differences in AT2R binding patterns appeared to be a regional phenomenon differentiating supradiaphragmatic versus infradiaphragmatic conduit arteries.
Radioligand Binding in Aortic Membranes
Results of Ang II receptor binding in membrane fractions from fetal TAs and from adult TAs and AAs are shown in Table 3 (see online Figures II and III, which can be accessed at http://atvb.ahajournals.org, for representative binding curves). Results can be misleading unless they are interpreted in conjunction with autoradiographic findings. Thus, abundant AT1Rs and AT2Rs in membranes of intact fetal TAs (Table 3) in fact are derived primarily from the periaortic mesenchyme rather than from the ME wall proper (Figure 1). In adult TAs, comparison of medial wall strips (showing only AT1Rs) with intact TAs (showing both types of receptors, Table 3) supported autoradiographic findings of AT1Rs localized largely to the medial wall and AT2Rs confined to extramedial areas (Figure 1). On the other hand, comparison of intact adult TAs versus AAs with the use of radioligand binding (Table 3) failed to detect the striking rostrocaudal differences in AT2R distribution (Figures 1 and 2⇑, Table 2). However, taken in conjunction with autoradiography, radioligand binding provided additional and independent quantitative information. Thus, total Ang II receptors in adherent adventitia of fetal TAs (Bmax *SIAII) averaged 52.4±20.6 fmol/mg protein, of which 40±13% was AT2Rs (Table 3). In adult pig TA and AA medial walls, AT1R density was homogenous (Table 3), ranging from 13 to 18 fmol/mg protein. Overall, adult medial AT1R density was at or below detection limits for radioligand binding: only 7 of 11 adult TA medial strip membranes contained sufficient Ang II receptors to detect binding in competition curves, and only 5 of 11 exhibited sufficient Ang II receptors to determine *SIAII Bmax. In general, results of competition curves agreed well with estimates derived from saturation curves. The Kd value for *SIAII in adult TA membranes was significantly higher than that in fetal TA (P=0.037); log IC50 values for Ang II and its analogues were not significantly different between fetal and adult tissues (Table 3).
To identify the vascular cell types potentially responsive to a functionally active tissue RAS in conduit arteries, we quantitatively examined the developmental evolution of AT1Rs and AT2Rs in large conduit arteries of microswine, a species with a cardiovascular physiology similar to that in humans. The most important findings of the present study include developmental differences in Ang II receptor binding and novel rostrocaudal differences in the swine macrovascular tree. These are as follows: (1) developmental differences in the density of AT1Rs (increasing with age) and AT2Rs (decreasing with maturity), (2) differential rostrocaudal expression of AT2Rs at all ages (abundant in AAs and its branches, sparse or absent in the TA and its branches), (3) a developmental shift in AT2R localization in infradiaphragmatic conduit arteries (inner medial wall prenatally versus a unique SMα-actin–negative/non–neural cell layer occupying the adventitial-medial boundary postnatally), (4) a novel age-dependent rostrocaudal difference in vascular expression of a non-AT2 CGP 42112 binding site (diffusely present in medial wall of postnatal TA and its branches, absent in AA and its branches at all ages), and (5) a unique pattern of AT2R distribution at AA branch points (characterized by radially asymmetric increase in density and fractional area at the site of branch-artery takeoff). Finally, the results document the specificity of *CGP for the porcine AT2R, demonstrate its superior resolution for autoradiographic AT2R localization in tissues that lack the non-AT2R CGP binding site, and emphasize the advantages of quantitative autoradiography over whole-vessel radioligand binding for localizing Ang II receptors in specific vascular wall compartments.
Developmental Changes in Ang II Receptor Density
As described in other species,16–18⇓⇓ diffusely distributed ME wall AT1Rs were significantly increased in swine TAs and AAs by 3 weeks after birth and were higher still in adult aortas (Table 1). Also, as predicted by prior observations in other species,19,20⇓ AT2Rs were dominant in fetal microswine AAs (Figure 1) and declined after birth in density and fractional area occupied. Despite an increase with age, AT1R density in adult TAs remained very low; it was detectable in only ≈60% of the animals. Unlike AT2R binding (see below), there were no significant regional differences in AT1R binding. AT2R binding in peripheral nerves has not, to our knowledge, been previously described. Gallinat et al21 reported upregulation of AT2R mRNA in transected sciatic nerve, and Bleuel et al22 described AT2Rs in cultured Schwann cells.
Rostrocaudal Differences in AT2R Binding
One novel finding of the present study is the striking regional vascular differences in the abundance and the distribution pattern of the AT2R. On the basis of reported studies of AT2R mRNA in the developing rat,8,16,23⇓⇓ the sparse AT2R binding in the fetal TA ME wall was unexpected. Yet AT2Rs were densely abundant in fetal AAs, occupying the inner 60% of the ME wall. This rostrocaudal difference (few AT2Rs in isolated microclusters in TAs versus abundant AT2Rs in AAs) persisted after birth. In all 3 age groups, the AT2R microclusters in the TA ME wall, according to neurofilament and SMα-actin immunochemistry, coincided with combined microvessel/neural units in the outer medial wall; whether one or both of these elements accounted for the punctate AT2R binding could not be resolved by the methods used. In the postnatal period, along with persistence of the rostrocaudal differences in AT2R abundance, 2 striking developmental differences emerged. First, the inner ME wall site of AT2R binding in fetal AAs shifted postnatally to a unique circumferential cell layer forming a band at the medial-adventitial boundary in infradiaphragmatic arteries. The cell type was neither neural nor typical vascular smooth muscle in nature. Second, the supradiaphragmatic conduit arteries developed, in the postnatal period, a non-AT2R binding site for CGP 42112 in the medial wall, a feature consistently absent in the infradiaphragmatic postnatal vasculature.
The significance of the developmental change in AT2R binding from the inner medial wall of the fetal AA to the outer medial-adventitial boundary in the postnatal AA and its branches is not known. Frid and colleagues15,24⇓ have documented 3 distinct layers in the walls of large elastic conduit vessels, each containing VSMCs of unique phenotypes as judged by immunochemistry and by proliferative behaviors in culture. Whereas the middle compact layer consists of classic VSMCs between layers of elastin, the third (outermost medial) layer is composed of 2 distinct types of VSMC: the first, a “nonmuscle” phenotype arranged in longitudinal bundles; the second, a typical VSMC phenotype arranged circumferentially and interspersed among the longitudinal bundles.15 An “interstitium” located between the longitudinally oriented fibers is also a part of the outer medial layer. The AT2R-positive/SMα-actin–negative cell layer in pig AAs appears to extend into this interstitium within the outer medial layer (Figure 3), suggesting continuity of the closely adherent adventitial layer and the medial interstitial compartment. Allen et al25 and Zhuo et al26 have reported adventitial AT2Rs in the human renal artery25 and in intrarenal preglomerular arterioles,25,26⇓ as assessed by emulsion autoradiography. We have observed an identical autoradiographic pattern of AT2R binding in human renal arteries (S. Bagby, R.C. Speth, unpublished data, 2002). We propose, given the highly organized and continuous nature of the AT2R-positive cell layer in postnatal infradiaphragmatic conduit vessels, that this layer represents a cellular interface that physically and functionally links the media and adventitia.
The AT2R-laden cell layer is notable for proximity to what has been recently characterized as a perivascular/adventitial RAS.27 Chymase-dependent interstitial Ang II generation has been shown to be increased in vascular disease states, including atherosclerosis,9,28⇓ and after balloon injury.29 The AT2R-positive cell layer may mediate the initial physiological response to interstitial Ang II, serving as the source of vasodilatory compounds (bradykinin, nitric NO) generated by AT2R activation. Similarly, dysfunction of these cells may augment Ang II vasoconstrictive effects via loss of the AT2R-mediated counterregulatory actions. The rostrocaudal differences in the presence of the AT2R-positive cell layer raises the possibility that interstitial Ang II may elicit regionally distinct vascular effects.
The significance of the striking rostrocaudal differences (both in AT2R abundance and in the mirror-image rostrocaudal differences in the non-AT2R CGP 42112 binding site) is also not known. Of potential relevance, studies of the embryological origins of aortic smooth muscle cells (SMCs) have shown 2 distinct sources: (1) neural crest cells for SMCs in the proximal aorta and (2) lateral mesenchyme for SMCs of the distal aorta.30 Thus, it seems plausible that the distinct differences in AT2R binding and in non-AT2R CGP binding between supradiaphragmatic versus infradiaphragmatic conduit arteries may reflect unique embryological origins. Viswanathan et al14 also described a non–Ang II CGP 42112 binding site in the rat carotid artery that increased after balloon injury, suggesting the possibility that activated cell phenotypes and/or their unique matrix products may be important. Neural crest–derived postnatal SMCs may synthesize a CGP-reactive matrix product not associated with the mesenchyme-derived SMCs of the infradiaphragmatic vasculature. Resolution of this unexplained finding awaits further study.
The AT2R-positive layer in postnatal AAs (as in fetal AAs) exhibited continuity with the respective layer in the walls of branching arteries and an unusual spatial organization at sites of branching. Symmetrically distributed around the circumference of the AA in interbranch segments, the AT2R-positive cell layer becomes asymmetric at branch points: thicker on the branching side and thinner opposite the branch. The concentration of AT2Rs at arterial branch points may reflect and/or play a role in response to shear stress–related hemodynamic forces. The unique predilection of branch-artery takeoff sites to atherosclerotic lesions is well established.31 AT2Rs concentrated at these sites may be important in preventing this outcome in the healthy vascular tree via AT2R-induced bradykinin and/or NO32 release at sites of disturbed laminar shear stress.6
In summary, we describe developmental changes in AT1R and AT2R conduit arteries and report novel changes in conduit artery Ang II receptor binding in normal microswine, including (1) age-independent rostrocaudal differences in supradiaphragmatic (sparse) versus infradiaphragmatic (abundant) AT2R binding and (2) a unique postnatal shift in the locus of AT2R binding in infradiaphragmatic arteries from the inner half of the medial wall before birth to a medial-adventitial boundary layer after birth. This AT2R-positive circumferential cell layer in postnatal infradiaphragmatic arteries appears optimally positioned to transduce AT2R-mediated vascular effects of locally generated interstitial Ang II. We speculate that given the apparent absence of an AT2R-rich cell layer in the supradiaphragmatic vasculature, interstitial renin/angiotensin activation may elicit regionally distinct vascular responses.
This study was supported by funding from the National Institutes of Health (National Institute of Child Health and Human Development grant PO1 HD-34430). The authors wish to acknowledge the outstanding assistance of animal technician Vicki Feldman of the OHSU Department of Comparative Medicine, the superb support and skilled technical contributions of Carolyn Gendron and Linda Jauron-Mills of the OHSU Heart Research Center Imaging Core Laboratory, the technical skills of photographer Michael Moody of the Portland VAMC Medical Media Service, and the administrative contributions of Kathleen Beebe in manuscript preparation.
Received April 17, 2002; revision accepted May 1, 2002.
- ↵Franklin SS, Khan SA, Wong ND, Larson MG, Levy D. Is pulse pressure useful in predicting risk for coronary heart disease?: the Framingham Heart Study. Circulation. 1999; 100: 354–360.
- ↵Hollenberg NK, Fisher ND, Price DA. Pathways for angiotensin II generation in intact human tissue: evidence from comparative pharmacological interruption of the renin system. Hypertension. 1998; 32: 387–392.
- ↵Gallinat S, Busche S, Raizada MK, Sumners C. The angiotensin II type 2 receptor: an enigma with multiple variations. Am J Physiol Endocrinol Metab. 2000; 278: E357–E374.
- ↵Langille BL. Fluid dynamics in vascular pathology: adaptations of the arterial wall to chronic changes in blood flow.In: The Critical Role of Mechanical Forces in Blood Vessel Development, Physiology and Pathology. Toronto, Canada: The Toronto Hospital Research Institute; 1999.
- ↵Nagel TE, Resnick N, Dewey CF, Gimbrone MA. Signal transduction of physical forces differential responses of endothelial cells of uniform and disturbed laminar shear stress.In: The Critical Role of Mechanical Forces in Blood Vessel Development, Physiology and Pathology. Istanbul, Turkey/Houston, Tex: Gladsone Institute/Baylor College of Medicine; 1999.
- ↵Akishita M, Horiuchi M, Yamada H, Zhang L, Shirakami G, Tamura K, Ouchi Y, Dzau VJ. Inflammation influences vascular remodeling through AT2 receptor expression and signaling. Physiol Genomics. 2000; 2: 13–20.
- ↵Hutchinson HG, Hein L, Fujinaga M, Pratt RE. Modulation of vascular development and injury by angiotensin II. Cardiovasc Res. 1999; 41: 689–700.
- ↵Ihara M, Urata H, Kinoshita A, Suzumiya J, Sasaguri M, Kikuchi M, Ideishi M, Arakawa K. Increased chymase-dependent angiotensin II formation in human atherosclerotic aorta. Hypertension. 1999; 33: 1399–1405.
- ↵Yusuf S, Sleight P, Pogue J, Bosch J, Davies R, Dagenais G. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients: the Heart Outcomes Prevention Evaluation Study Investigators [published corrections appear in N Engl J Med. 2000;342:748 and 2000;342:1376]. N Engl J Med. 2000; 342: 145–153.
- ↵Lee KT. Swine as animal models in cardiovascular research.In: Tumbleson ME, ed. Swine in Biomedical Research. New York, NY: Plenum Press; 1986: 1481–1496.
- ↵Moulik S, Speth RC, Rowe BP. Differential loss in function of angiotensin II receptor subtypes during tissue storage. Life Sci. 2000; 66: L233–L237.
- ↵Viswanathan M, Johren O, de Oliveira M, Saavedra JM. Increased non-angiotensin II CGP 42112 binding in rat carotid artery after balloon injury. Peptides. 2001; 17: 695–699.
- ↵Frid MG, Aldashev AA, Dempsey EC, Stenmark KR. Smooth muscle cells isolated from discrete compartments of the mature vascular media exhibit unique phenotypes and distinct growth capabilities. Circ Res. 1997; 81: 940–952.
- ↵Tufro-McReddie A, Harrison JK, Everett AD, Gomez RA. Ontogeny of type 1 angiotensin II receptor gene expression in the rat. J Clin Invest. 1993; 91: 530–537.
- ↵Grady EF, Sechi LA, Griffin CA, Schambelan M, Kalinyak JE. Expression of AT2 receptors in the developing rat fetus. J Clin Invest. 1991; 88: 921–933.
- ↵Shanmugam S, Monnot C, Corvol P, Gasc J. Distribution of type 1 angiotensin II receptor subtype messenger RNAs in the rat fetus. Hypertension. 1994; 23: 137–141.
- ↵Bleuel A, DeGasparo M, Whitebread S, Puttner I, Monard D. Regulation of protease nexin-1 expression in cultured Schwann cells is mediated by angiotensin II receptors. J Neurosci. 1995; 15: 750–761.
- ↵Shanmugam S, Corvol P, Gasc JM. Angiotensin II type 2 receptor mRNA expression in the developing cardiopulmonary system of the rat. Hypertension. 1996; 28: 91–97.
- ↵Frid MG, Dempsey EC, Durmowicz AG, Stenmark KR. Smooth muscle cell heterogeneity in pulmonary and systemic vessels: importance in vascular disease. Arterioscler Thromb Vasc Biol. 1997; 17: 1203–1209.
- ↵Arakawa K, Urata H. Hypothesis regarding the pathophysiological role of alternative pathways of angiotensin II formation in atherosclerosis. Hypertension. 2000; 36: 638–641.
- ↵Uehara Y, Urata H, Sasaguri M, Ideishi M, Sakata N, Tashiro T, Kimura M, Arakawa K. Increased chymase activity in internal thoracic artery of patients with hypercholesterolemia. Hypertension. 2000; 35: 55–60.
- ↵Higashiura K, Ura N, Miyazaki Y, Shimamoto K. Effect of an angiotensin II receptor antagonist, candesartan, on insulin resistance and pressor mechanisms in essential hypertension. J Hum Hypertens. 1999; 13 (suppl 1): S71–S74.
- ↵Majesky MW, Landerholm TE, Lu J. Developmental origins of vascular smooth muscle diversity.In: The Critical Role of Mechanical Forces in Blood Vessel Development, Physiology and Pathology. Houston, Tex: Baylor College of Medicine; 1999.
- ↵Zarins CK, Giddens DP, Bharadvaj BK, Sottiurai VS, Mabon RF, Glagov S. Carotid bifurcation atherosclerosis: quantitative correlation of plaque localization with flow velocity profiles and wall shear stress. Circ Res. 1983; 53: 502–514.