Angiotensin type 1A (AT1A) and 1B (AT1B) receptor deletion (AT1DKO) results in renal microvascular disease, tubulointerstitial injury, and reduced blood pressure. To test the hypothesis that renal preglomerular responses to angiotensin (ANG) II are mediated by AT1A and AT1B receptors, experiments were performed in AT1DKO mice using the in vitro blood perfused juxtamedullary nephron technique. Kidneys were harvested from AT1DKO and wild-type (WT) mice and bathed with ANG II (1–100 nM), norepinephrine (NE; 100–1,000 nM), or acetylcholine (ACh; 10 μM). Baseline diameters of afferent (19.5 ± 0.7 and 13.9 ± 0.7 μm, n = 17 and 16) and efferent (15.5 ± 2.1 and 10.8 ± 1.0 μm, n = 4 and 7) arterioles of AT1DKO were significantly larger than WT. Afferent and efferent arteriolar responses to ANG II, 100, and 300 nM NE were absent in AT1DKO; although significant constriction to 1 μM NE was observed (−17 ± 5 and −23 ± 6%, respectively). Afferent arterioles of WT mice dilated significantly in response to ACh (15.1 ± 0.6 to 17.0 ± 1.2 μm, n = 6); however, arterioles from AT1DKO tended to contract (19.9 ± 1.2 to 17.8 ± 1.6 μm; n = 6, P = 0.06). In summary, loss of ANG II-induced contraction, reduced vasoconstriction to NE, and endothelial cell dysfunction contribute to the renal vascular phenotype of AT1DKO mice. We conclude that ANG II signaling via the AT1 receptor plays a pivotal role in basal renal microvascular tone and effectiveness to respond to vasoconstrictor and vasodilator agonists.
- afferent arteriole
- efferent arteriole
- juxtamedullary nephron
the at1 receptor subtype is the primary receptor responsible for the renal vascular and tubular actions of the renin-angiotensin system (RAS). Humans possess a single angiotensin type 1 (AT1) receptor (1), whereas there are two AT1 receptor isoforms in rodents (AT1A and AT1B) that are products of separate genes (Agtr1a and Agtr1b) (19) and located on different chromosomes (14). The ligand binding signatures of the rodent AT1A, AT1B, and human AT1 receptors are essentially identical (3).
Since physiological and morphological studies in mice lacking only the AT1A receptor have demonstrated that the remaining AT1B receptor partially compensates for the loss of function of the AT1A receptor (6, 7, 15, 18, 26), mice lacking both AT1 receptor genes (AT1 double nullizygotes; AT1DKO) represent an ideal model for the study of the functional impact of the single human AT1 receptor. AT1DKO mice have been characterized by two independent groups (16, 17, 23) and exhibit marked hypotension and impaired growth compared with wild-type (WT) mice (16, 23). Kidneys of AT1DKO mice display hypoplastic papilla, arterial hypertrophy, microvascular disease, and tubulointerstitial injury (16, 17, 23). This phenotype has also been described in models in which the absence of a functional RAS has been accomplished by genetic deletion of renin (2, 25), angiotensinogen (12), and angiotensin-converting enzyme (ACE) (4, 13) in mice. Taken together, these studies indicate that ANG II is essential for achieving normal blood pressure and renal architecture.
The AT1DKO mouse kidney demonstrates increased mRNA expression of renin, endothelin 1, endothelin receptor type A, cyclooxygenase-2 (COX-2), inducible nitric oxide synthase (iNOS), neuronal NOS (nNOS), and monocyte/macrophage inflammatory cytokines and enzymes involved in oxidative stress (17). It is not known how changes in expression of inflammatory, vasoconstrictor, and vasodilator systems affect renal microvascular function in AT1DKO mice, and this is the focus of the present study.
Utilizing immunohistochemical approaches, we demonstrated the localization of renal vascular AT1 receptor protein to afferent and efferent arterioles (9). Also, we found that ANG II-induced afferent arteriolar diameter responses were reduced, while efferent arteriolar diameter responses were absent in kidneys of mice lacking the AT1A receptor compared with WT (6), while responses to norepinephrine (NE) were not altered (7). ANG II responses of afferent and efferent arterioles in kidneys of mice lacking the AT1B receptor were not different from WT (7). Based on these studies performed in kidneys from AT1A−/− (6) or AT1B−/− (7) mice, we concluded that the vasoconstrictor response to ANG II is mediated by the AT1A and AT1B receptors in the afferent arteriole, and by only the AT1A receptor in the efferent arteriole. Characterization of the changes in functional responsiveness of the renal microvasculature of AT1DKO mice has not been previously described. We hypothesized that the renal microvasculature of AT1DKO would not respond to ANG II but would have an intact vasoconstrictor response to NE and a vasodilator response to ACh. Studies were carried out using the mouse in vitro blood-perfused juxtamedullary nephron technique, which allows for direct video microscopic visualization of afferent and efferent arteriolar diameters studied in situ in kidneys obtained from mice with gene-targeted deletion of the AT1A and AT1B receptors.
The procedures used in this study were approved by the Animal Care and Use Committee of Louisiana State University Health Sciences and conducted according to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The establishment of our AT1DKO colony was performed by rederivation (Tulane University Transgenic Mouse Facility) of Agtr1a−/−Agtr1b−/− mice kindly provided by T. M. Coffman. The Agtr1a−/−Agtr1b−/− breeder mice were generated by selective breeding on mixed backgrounds of 129/SvEv and C57BL/6 as previously described (16). C57BL/6J mice (4 wk old, Jackson Labs) were treated with pregnant mare serum gonadotropin and human chorionic gonadotropin and mated with three Agtr1a−/−Agtr1b−/− males. Oocytes (0.5 days postcoitum) were collected, washed 10 times in M2 media, transferred into the barrier facility, and implanted into three pseudopregnant ICR females (Harlan). This mating strategy produced 12 viable double heterozygous Agtr1a +/−Agtr1b+/− pups. These 12 pups were intercrossed and generated the expected nine genotypes. To produce double homozygous deleted Agtr1a−/−Agtr1b−/− (AT1DKO) and double homozygous intact Agtr1a+/+Agtr1b+/+ (WT) animals, littermates were then intercrossed. The final breeding strategy consisted of breeding female Agtr1a+/−Agtr1b−/− and male Agtr1ain−/−Agtr1b−/− to obtain double homozygous null mice as described by Oliverio et al. (16) and Agtr1a+/+Agtr1b+/+ to obtain WT animals. Genotyping of mice bred in our colony was performed by PCR, in duplicate, on DNA obtained by tail biopsy, as previously described in detail (7, 20, 21).
Histological analysis of juxtamedullary afferent and efferent arterioles was assessed in kidneys obtained from adult AT1DKO (male, n = 2; female, n = 2) and WT mice (male, n = 4) that were bred in our colony (n = 5) and obtained from Dr. Coffman (n = 3). Mice were anesthetized with pentobarbital sodium (50 mg/kg ip). Kidneys were prepared by immersion fixation in 10% buffered Formalin overnight as we have previously described (9) or by in vivo perfusion fixation (22). Retrograde perfusion fixation was performed after cannulation of the abdominal aorta using a 25G blunt-end stainless steel needle at a pressure of 100 mmHg (2.3 m). Tissues were perfused with PBS (5 ml) for 5 min, followed by freshly prepared 3% paraformaldehyde in PBS (25 ml, pH 7.3) for 25 min. After perfusion, kidneys were removed, cut coronally at the depth of the papilla, and postfixed in 3% paraformaldehyde at 4°C for 1 h. Tissues were placed in cryoprotectant 2.4% sucrose in PBS and 0.02 sodium azide overnight at 4°C. All kidneys were routinely processed for paraffin embedding and sectioned at a thickness of 3 μm.
Immunohistochemistry for α-smooth muscle actin.
Kidney sections were immunostained by the immunoperoxidase technique as we have previously described (9) with the addition of biotin blocking (4 × 5 min, DAKO) of tissue sections following the methanol peroxide blocking. Tissue sections were incubated with mouse anti-human α-smooth muscle actin monoclonal antibody (1:400; Novocastra Lab) for 90 min. Control experiments were performed by omitting the primary or secondary antibodies. Slides were imaged using an Olympus DP70 Digital Camera System mounted to an Olympus BX51 TRF Microscope.
Mouse in vitro blood-perfused juxtamedullary nephron technique.
Afferent and efferent arteriole diameters were assessed in kidneys obtained from adult AT1DKO (male, n = 9; female, n = 6) and WT mice (male, n = 6) that were bred in our colony, or WT mice (C57BL/6J) purchased from Jackson Labs (male, n = 13). Thirty-four adult male Sprague-Dawley rats (451 ± 10 g body wt, Charles River Laboratories, Raleigh, NC) were used as blood donors. All animals were provided ad libitum access to food and water before the study. Rats and mice were administered an initial dose of pentobarbital sodium (50 mg/kg ip) and supplemental doses as needed to induce anesthesia (cumulative dose in rats 82 ± 4 mg/kg ip and mice 86 ± 3 mg/kg ip). Rats were subjected to bilateral nephrectomy, exsanguinated via a carotid arterial cannula, and the blood was processed for perfusion of the mouse kidney. Experiments were conducted using the mouse in vitro blood-perfused juxtamedullary nephron technique as we have previously reported in detail (6, 7) and described briefly below. The right renal artery was cannulated via the descending abdominal aorta, perfused with albumin-containing (51 g/l) Tyrode buffer, and placed in a perfusion chamber for the dissection procedure. Slight modifications of the dissection procedure were required due to the reduced thickness of the cortex of kidneys from the AT1DKO compared with WT mice as described by Oliverio et al. (16). On completion of the microdissection procedure, the cell-free perfusate was replaced with the reconstituted rat blood, the perfusion chamber was warmed, and the renal arterial perfusion pressure was maintained at 94.4 ± 0.2 mmHg (n = 44). All vasoactive agents were administered by addition to the albumin-containing (10 g/l) Tyrode bathing solution. The chamber was secured to an upright fixed-stage light microscope equipped with a water-immersion objective (×40) and transilluminated. The image of the vessel of interest was focused continuously and stored on a DVD for careful analysis at a later time. Afferent arterioles were studied at a site 25–50% along the length from the glomerulus. Efferent arterioles were studied near the glomerulus before peritubular capillary branching. Experimental protocols were begun after a ≥15-min stabilization period. In some kidneys, another afferent or efferent arteriole was visualized, protocols were repeated, and images were recorded. Only one kidney was studied from each animal.
Afferent and efferent arteriolar responses to ANG II in AT1DKO mice.
We previously observed afferent arteriolar constrictions to ANG II and a lack of an AT2 receptor mediated vasodilatory response to ANG II in the presence of an ANG receptor blocker in AT1A−/−, AT1B−/−, and WT mice (6, 7). Therefore, afferent and efferent arteriole diameters of AT1DKO mice (n = 10) were measured during superfusion with ANG II to document loss of AT1 receptor function and to determine whether the AT2 receptor is activated in AT1DKO mice. After a stabilization period, afferent or efferent arteriolar luminal diameter was monitored under baseline conditions (5 min), and during sequential exposure to increasing concentrations of 1, 10, and 100 nM ANG II (5 min at each concentration) in kidneys of AT1DKO mice. Doses were selected based on our previous work (7). A 10-min recovery period followed.
Afferent and efferent arteriolar responses to NE in AT1DKO mice.
Since we did not expect to observe a vasoconstrictor response to ANG II in AT1DKO mice, afferent and efferent arteriole diameters were measured during superfusion with NE to determine the arteriolar vasoconstrictor potential of these vessels. After recovery from the ANG II protocol, the luminal diameter of the same afferent or efferent arteriole was monitored under baseline conditions (5 min) and during sequential exposure to increasing concentrations of NE (100, 300, and 1,000 nM, 5 min at each concentration). A 10-min recovery period followed. Doses were selected based on our previous work (7). In five kidneys, a second ANG II and NE protocol was performed after the recovery period.
Afferent and efferent arteriolar responses to ACh in WT mice.
Renal microvascular responsiveness to ACh was determined in kidneys of WT mice (n = 13; obtained from Jackson Labs) since we had not studied the vascular effects of this agonist in the mouse kidney previously. After a stabilization period, afferent or efferent arteriolar luminal diameter was monitored under baseline conditions (5 min) and during sequential exposure to increasing concentrations of ACh (1, 10, and 100 μM, 5 min at each concentration) in kidneys of WT mice. A 15-min recovery period followed. In five kidneys, and a second protocol was performed after the recovery period.
Afferent arteriolar responses to ACh in AT1DKO and WT mice.
The afferent arteriolar response to ACh was used to determine the endothelial cell-mediated vasodilatory responsiveness in kidneys from AT1DKO (n = 5) and WT (n = 6) mice whose parents were littermates. After a stabilization period, afferent arteriolar luminal diameter was monitored under baseline conditions (5 min), during exposure to 10 μM ACh (5 min) and during a recovery period (10 min). Only one protocol was performed in each kidney. In one AT1DKO kidney, two afferent arterioles were visible in the same recorded image.
Human ANG II (catalog no. 002-12, Phoenix Pharmaceuticals) and ACh (catalog no. A6625, Sigma) were dissolved in 0.9% NaCl at concentrations of 1 and 100 mM, respectively, stored at −20°C, and diluted on the day of the experiment. NE (Levophed norepinephrine bitartrate injection, Levophed, 1 mg/ml, catalog no. L-680, Abbott Laboratories) was diluted in Tyrode solution on the day of the experiment.
Renal arterial perfusion pressure and vessel diameters were sampled at 1 Hz and converted to a digital form using an analog-to-digital data-acquisition and analysis software as we have previously described (7, 24). Afferent and efferent arteriolar luminal diameters were measured manually and continuously throughout the protocol at a single site along the length of the selected vessel using a digital image-shearing monitor. The average diameter (μm) during the control (5 min), treatment (ANG II or NE, 5 min), and recovery (final 5 min) periods was used for statistical analysis. The average diameter during the plateau phase (final 2 min of each period) of the ACh protocol was utilized for statistical analysis. Statistical analyses were performed on the raw data by one-way repeated-measures or two-way analysis of variance followed by Dunnett's test, Bonferroni's test, or a paired t-test as appropriate. Because of the significant difference in baseline afferent arteriole diameters between AT1DKO and WT mice, two-way analysis of variance was conducted on the percent change from control for the afferent arteriole responses to ACh. Baseline diameters of afferent and efferent arterioles between groups were analyzed by an unpaired t-test. Statistical analysis was performed using a statistical software program. P < 0.05 was considered statistically significant. Values are means ± SE (n = number of arterioles or number of mice as appropriate).
AT1DKO and WT animals.
At the time of the study, the age of AT1DKO was slightly greater than WT mice (4.3 ± 0.4; n = 19 and 3.0 ± 0.4 mo, n = 23, P < 0.05); however, the body weights were not different (25.0 ± 1.1 and 26.5 ± 1.4 g).
Juxtamedullary arteriolar histology in AT1DKO and WT animals.
α-Smooth muscle actin immunostaining was performed in kidney tissue sections to visualize the arteriolar architecture of juxtamedullary nephrons. Figure 1 illustrates the afferent and efferent arteriolar vascular smooth muscle cell structure of kidneys from AT1DKO and WT animals. Images were obtained from the juxtamedullary cortical region of coronally sectioned kidneys. Arterioles are shown in longitudinal section near or touching the juxtamedullary glomerulus. Determination of pre- and postglomerular vessels was accomplished by tracing the afferent arteriole to the associated interlobular arteriole and/or arcuate artery. Basic histological examination of the arterioles shows an intact vascular wall in AT1DKO mice. No positive immunostaining was observed in tissue sections incubated in the absence of the primary or secondary antibody.
Baseline diameters of afferent and efferent arterioles of AT1DKO compared with WT mice.
To determine the influence of loss of expression of AT1 receptors on afferent and efferent arterioles, baseline diameters from all the arterioles of each genotype were pooled as shown in Fig. 2. Diameters of afferent arterioles of AT1DKO were significantly larger than those from WT mice (19.5 ± 0.7 and 13.9 ± 0.7 μm, n = 17 and 16) at baseline. Similarly, efferent arterioles of AT1DKO kidneys were significantly larger than those from WT mice (15.5 ± 2.1 and 10.8 ± 1.0 μm, n = 4 and 7). It is of interest to note that of the 15 kidneys of AT1DKO mice that were studied, efferent arterioles were only visible in 3. In WT mice, multiple efferent arterioles are usually visible in each kidney. This may reflect a change in the architecture of the efferent arteriole in AT1DKO mouse kidneys, in that the vessel does not lie on the juxtamedullary cortical surface. The arteriolar diameters of AT1DKO mice were ∼40% larger than those of the WT mice, indicating a major influence of the loss of AT1 receptor signaling on the basal tone of renal resistance vessels.
Lack of afferent and efferent arteriolar responses to ANG II in AT1DKO mice.
ANG II did not alter afferent (n = 11) or efferent (n = 4) arteriole diameters of kidneys obtained from AT1DKO mice (P > 0.05, Fig. 3). In the presence of 100 nM ANG II, afferent and efferent arteriole diameters were 98 ± 2 and 100 ± 2% of baseline values, respectively (Fig. 3). These data indicate that renal arterioles of AT1DKO mice do not respond to ANG II, even at very high concentrations. The combined actions of AT1A and AT1B receptors mediate the afferent arteriolar responses to ANG II. Also, there was no evidence for ANG II-induced AT2 receptor-mediated vasodilation in renal microvessels of AT1DKO mice.
Afferent and efferent arteriolar vasoconstrictor responses to NE in AT1DKO mice.
Renal microvascular responses to NE were determined in the same vessels that demonstrated a lack of vasoconstrictor responsiveness to ANG II. Because NE responses of the arterioles of kidneys from male and female AT1DKO mice were not statistically different as a function of gender, the data were combined. NE produced a rapid and significant constriction in afferent (n = 11) and efferent (n = 4) arterioles of AT1DKO, but only at the highest dose tested (1,000 nM NE, Fig. 3). Afferent arteriole diameter of AT1DKO mice decreased by 3 ± 2, 4 ± 2, and 17 ± 5%, while efferent arteriole diameter decreased by 2 ± 1, 2 ± 2, and 23 ± 6% in response to 100, 300, and 1,000 nM NE, respectively (P < 0.05, Fig. 3). Therefore, the lack of an effect of ANG II on arteriole diameter of AT1DKO mice does not reflect an inability of vascular smooth muscle cell contraction. However, the contractions in AT1DKO are less than the decreases in afferent and efferent arteriole diameters (51 ± 6 and 38 ± 3%, respectively, P < 0.05, AT1DKO vs. WT) (7) that we have previously observed in response to 1,000 nM NE in WT mice. Diameters of afferent and efferent arterioles of AT1DKO mice returned to values not different from baseline on removal of NE.
Afferent and efferent arteriolar vasodilatory responses to ACh in WT mice.
Baseline afferent and efferent arteriole diameters of WT mice were not different (13.2 ± 1.0 and 10.8 ± 1.0 μm, n = 10 and 7, respectively). ACh produced a significant increase in diameter of afferent and efferent arterioles of WT mice (Fig. 4). Afferent arteriole diameter increased 8 ± 2, 17 ± 5, and 35 ± 6% in response to 1, 10 and 100 μM ACh, respectively (P < 0.05, Fig. 4B). Efferent arteriole diameter increased 2 ± 7, 15 ± 6, and 16 ± 6% in response to 1, 10, and 100 μM ACh, respectively (P < 0.05, Fig. 4B). The afferent arteriolar dilatory response to 100 μM ACh was greater than that to 10 μM ACh; however, the efferent arteriolar dilatory response was maximal at 10 μM ACh. Afferent and efferent arteriolar responses to ACh were significantly different (P < 0.05). Diameters of afferent and efferent arterioles of WT mice returned to values not different from baseline on removal of ACh.
Absence of afferent arteriolar vasodilatory responses to 10 μM ACh in AT1DKO compared with WT mice.
To determine whether afferent arterioles of AT1DKO exhibit intact endothelial dependent vasodilation, responses to ACh were compared with WT mice. As described above, afferent arterioles of AT1DKO and WT mice were significantly different at baseline (19.9 ± 1.2 and 15.1 ± 0.6 μm, respectively, n = 6 and 6, Fig. 5A). ACh produced a significant dilation in afferent arterioles of WT (P < 0.05), but not AT1DKO mice. Afferent arterioles of WT mice dilated significantly in response to ACh (15.1 ± 0.6 to 17.0 ± 1.2 μm); however, arterioles from AT1DKO tended to contract (19.9 ± 1.2 to 17.8 ± 1.6 μm; P = 0.06). Vessel diameter increased by 12 ± 6% in WT mice (Fig. 5B). Diameters of afferent arterioles of AT1DKO and WT mice returned to values not different from baseline on removal of ACh.
Our previous work described the AT1 receptor subtype-specific and renal segmental microvascular responses to ANG II by investigation of the afferent and efferent arteriolar functional responsiveness to ANG II utilizing mice with genetic deletion of the AT1A (6) or AT1B receptor (7) and WT littermates. We concluded that afferent arteriolar vasoconstrictor responses to ANG II are mediated by both the AT1A and AT1B receptors, whereas efferent arteriolar vasoconstrictor responses to ANG II are mediated by only AT1A receptors in the mouse kidney (6, 7). The present study investigates further the renal microvascular responsiveness of afferent and efferent arterioles of mice lacking both the AT1A and AT1B receptor subtypes (AT1DKO).
The renal vascular phenotype of the AT1DKO mouse has been described based on histological examinations of kidney tissue sections. AT1DKO kidneys exhibit severe thickening of arterial walls (16), medial hyperplasia of the interlobular arteries and afferent arterioles (23), with dramatic increases in α-smooth muscle actin expression, and a threefold increase in medial area (17) compared with kidneys of WT mice. Although not described in the peripheral vasculature of the AT1DKO kidney, the overpopulation of vascular smooth muscle cells of AT1A−/− mice was specific to the intrarenal arteries and was not evident in small arteries of the liver and small intestine (11). We have extended these findings with the demonstration of the basic histology of juxtamedullary afferent and efferent arterioles of AT1DKO mice. Baseline diameters of blood-perfused juxtamedullary afferent and efferent arterioles of kidneys obtained from AT1DKO were significantly larger (+40%) compared with those from WT mice. A comparison of baseline afferent arteriole diameters of AT1DKO mice and Sprague-Dawley rats (8) reveals that the vessels are of similar size (19.5 and 20.2 μm, respectively). Also, the baseline afferent arteriole diameter of AT1DKO mice is not different from the maximal diameter of afferent arterioles in the presence of 100 μM ACh (19.5 ± 0.7 μm and 17.5 ± 1.2 μm P = 0.2) in WT mice. We have found that the efferent arterioles of kidneys obtained from AT1A−/− mice were significantly larger than in those from WT mice, averaging 19.6 μm (6). Afferent arteriole diameters of AT1A−/−mice, as well as afferent and efferent arteriole diameters of AT1B−/− mice, are not different from those of WT mice under baseline perfusion conditions (6, 7). In summary, afferent arterioles of AT1DKO mice and efferent arterioles of AT1A−/− and AT1DKO mice exhibit significant dilation at baseline compared with WT mice. Therefore, lack of expression of AT1 receptors on the renal vasculature results in a dilated vessel at baseline and may contribute to the reduced blood pressure reported in AT1DKO mice.
Contributing factors to the enlarged afferent and efferent arterioles at baseline may be a result of structural alterations in the vessels which occur during kidney development, or the loss of AT1 receptor-mediated suppression of vasodilator systems, or overall enhanced vasodilator effects on the kidney of AT1DKO mice. Both nNOS and COX-2 mRNA and protein expressions are increased in AT1DKO mice (17), which is consistent with a vasodilatory input (10). Since arterioles of AT1DKO mice exhibit a larger diameter at baseline, one might expect an increased influence of nitric oxide derived from endothelial nitric oxide synthase (eNOS) ; however, eNOS expression was not altered in renal cortical tissue of AT1DKO mice (17). It is plausible that nitric oxide derived from nNOS may influence arteriolar diameter in AT1DKO mice since an increase in nNOS mRNA and protein expression has been reported in these mice (17). The dilated vessels may also represent a consequence of interrupting the negative feedback of ANG II on the nNOS-COX-2 pathways involved in the tubuloglomerular feedback-control of afferent arterioles (10). The enlarged baseline arteriole diameters of AT1DKO mice do not appear to reflect a generalized hypertrophy of the kidney since the kidneys of AT1DKO mice are growth restricted.
In the kidney, ANG II has powerful effects on hemodynamics that are mediated principally by the AT1 receptor. Tsuchida et al. (23) and Oliverio et al. (16) administered a bolus injection of ANG II into anesthetized AT1DKO mice and observed no change in blood pressure, although blood pressure did increase slightly in AT1A−/− mice (15, 23). Our earlier work supported the hypothesis that AT1A and AT1B receptors are located on the afferent arteriole, since a significant contraction to ANG II was observed in AT1A−/− mice, whereas the AT1A receptor was the only subtype expressed on the efferent arteriole, since the contraction was absent in AT1A−/− mice and recovered in AT1B−/− mice (6, 7). To directly test this hypothesis and to clarify the role of other ANG II receptors in the vasoconstrictor actions of ANG II, we examined renal contractile responses to ANG II in kidneys of AT1DKO and WT mice. We found no discernable vascular effect of ANG II on afferent or efferent arterioles of AT1DKO mice. As a general comparison, we have previously reported that 100 nM ANG II significantly reduces afferent and efferent arteriolar diameters by 30% in kidneys obtained from WT mice (7) and Sprague-Dawley rats (5). These studies verify the importance of AT1 receptors in the vasoconstrictor effects of ANG II. The absence of any renal vascular response to ANG II in AT1DKO mice suggests that there are no other ANG II receptors with significant vasoconstrictor functions. In addition, they confirm the conclusions from our previous study that renal vasoconstrictor responses to ANG II are mediated by the AT1A and AT1B receptors, the only subtypes for the AT1 receptor.
We hypothesized that lifelong loss of AT1A and AT1B receptors and the presence of elevations in circulating ANG II levels would provide an environment for the full expression of AT2 receptor-mediated renal vasodilation. It has been shown that chronic ACE inhibition (10 days) increases systolic blood pressure measured in conscious AT1DKO mice (16), possibly representing an inhibition of vasodilatory AT2 receptor signaling. However, acute AT2 receptor blockade was without effect on blood pressure measured in anesthetized AT1DKO mice (23), indicating that the AT2 receptor, which is thought to transduce the depressor effect of ANG II, does not exert an acute depressor effect in the absence of AT1 receptor signaling. Previously, we found no evidence for afferent or efferent arteriolar vasodilation in response to ANG II in the presence of pharmacological AT1 receptor blockade in WT, AT1A−/−, or AT1B−/− mice (6, 7). In the present study, we did not observe AT2 receptor-mediated dilation in the AT1DKO mice treated with high-dose ANG II. In addition, Ouyang et al. (17) reported that AT2 receptor mRNA expression was unaltered in AT1DKO compared with WT mice. One possible explanation for a lack of an AT2 receptor-mediated vasodilation is that the arterioles of AT1DKO mice may be maximally dilated at baseline. However, we have observed constriction of arterioles of AT1DKO mice kidneys when the cell-free perfusion solution is changed to the red blood cell perfusion (data not shown). Therefore, our results do not support ANG II-induced AT2 receptor-mediated vasodilation in afferent or efferent arterioles in the isolated, perfused mouse kidney.
Bolus administration of epinephrine causes significant and equivalent increases in mean arterial pressure (increase of ∼15 mmHg) in anesthetized AT1DKO and WT mice (16), indicating that loss of AT1 receptor function does not affect increases in total peripheral resistance produced by exogenous epinephrine. In our study, afferent and efferent arterioles of AT1DKO mice did not respond to ANG II, or to 100 and 300 nM NE, but responded significantly to high-dose NE (1,000 nM). The observance of a vasoconstriction of the afferent and efferent arterioles in the AT1DKO mouse indicates that the vascular smooth muscle cells are capable of vasoconstriction in response to another agonist, namely, NE. However, we have previously reported that 300 nM NE produces a significant vasoconstriction in kidneys of AT1A−/− and WT mice (7), suggesting that arterioles of AT1DKO mice have a reduced vasoconstrictor potential. Therefore, we conclude that the absence of a vasoconstrictor response of the pre- and postglomerular vessels to ANG II in AT1DKO mice is not solely limited by vascular smooth muscle cell contractile function, as assessed by NE, but is due to the loss of the AT1 receptor function.
ACh produced significant vasodilation of afferent and efferent arterioles of kidneys of WT mice. The vasodilation to ACh of efferent arterioles was significantly less than that of the afferent arterioles of WT mice. As a means of investigating endothelial-dependent vasodilator responsiveness of afferent arterioles, we tested the effects of superfusion of ACh on afferent arterioles of AT1DKO and WT mice. However, afferent arterioles of AT1DKO mice did not vasodilate to ACh and tended to contract. The response of the afferent arteriole of AT1DKO mice was significantly different than the response of the afferent arteriole of WT mice. This lack of vasodilation to ACh may represent endothelial cell dysfunction, and/or enhanced expression of muscarinic receptors or signaling pathways in the AT1DKO kidney. The mechanisms involved in the reduced response to ACh in AT1DKO mice are the focus of future studies.
In summary, arterioles of AT1DKO mice exhibit larger diameters at baseline, compromised vasoconstrictor responsiveness, and endothelial cell dysfunction. Overall, our functional data on the lack of a vasoconstrictor response to ANG II in the afferent arteriole of AT1DKO mice indicate that the AT1A and AT1B receptors are the subtypes responsible for ANG II-mediated contractions of the preglomerular arterioles. The lack of an ANG II-induced contraction of the efferent arteriole of AT1DKO mice is consistent with our previous findings in AT1A−/− and AT1B−/− mice. Our data confirm and extend our previous conclusion that the AT1A and AT1B receptors mediate the afferent arteriolar vasoconstriction to ANG II, whereas the AT1A receptor alone mediates the efferent arteriolar responses to ANG II. The segmental specific localization of these two AT1 receptor subtypes may contribute to differential regulation of ANG II-induced contractions to ANG II and the regulation of renal plasma flow and glomerular filtration rate in rodents. The study of mice lacking both the AT1A and AT1B receptor subtypes represents an experimental model most closely related to the understanding of the function of the single AT1 receptor in humans.
Support for this study was provided by National Institutes of Health Grants DK-62003 and P20 RR-018766.
The authors acknowledge the excellent technical assistance of Ann Mullins in the rederivation of the AT1DKO mice and Christopher J. Monjure in maintaining the mouse breeding colonies. Portions of this work have been published in abstract form (J Am Soc Nephrol 16: 392A, 2005 and FASEB J 20: A761–A762, 2006).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society