INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

REGULATION OF BLOOD pressure and sodium homeostasis are well-recognized functions of the renin-angiotensin system (RAS). Several lines of evidence suggest that the RAS may also play a role in fetal development and growth. For example, circulating levels of renin and angiotensin II are elevated in newborn animals (42), expression of angiotensinogen and renin are enhanced in a number of fetal tissues (8, 9, 17), and angiotensin receptors are expressed in a tightly regulated program during embryonic development (10, 11, 18, 39). In support of a physiological role for the RAS in development are observations that administration of RAS antagonists to weanling rats causes structural abnormalities in the kidney (7, 41) and clinical reports describing teratogenicity of angiotensin converting enzyme (ACE) inhibitors in humans (35).

Angiotensin II stimulates growth and proliferation of various cell types including vascular smooth muscle cells (31), glomerular mesangial cells (33), and renal tubular epithelial cells (43). The growth-promoting activities of angiotensin II are mediated by the type I angiotensin receptor (AT₁) (33, 38) and may involve JAK/STAT kinase signaling cascades (26). Thus, in addition to potential effects of the RAS on organogenesis and vascular development, angiotensin II has been suggested to be an important factor in promoting growth. In this regard, Tufro-McReddie and associates (40) found that treatment of weanling rats with an AT₁ receptor blocker inhibited both renal and somatic growth (40).

Gene targeting using homologous recombination in embryonic stem cells has been widely used to assign developmental functions to individual genes (3). This technique provides a specificity and efficacy of inhibition that cannot be achieved pharmacologically. Moreover, it allows for examination of the role of the targeted gene beginning from the earliest stages of development. Recently, several mouse lines with targeted mutations in genes encoding components of the RAS have been produced, and analysis of these animals further supports a role for the RAS in development and early postnatal life (19, 21, 29, 37). Most mice that are homozygous for a targeted disruption of the angiotensinogen gene (Agt) do not survive to weaning (19). Those that survive to adulthood develop abnormalities of their kidneys including vascular hypertrophy and focal tubular dropout with interstitial inflammatory infiltrates (19, 29). Similar lesions are seen in mice with targeted mutations of the ACE (Ace) gene (6, 21), suggesting that the absence of angiotensin II peptide is critical to the pathogenesis of the phenotype.

To examine the role of the AT_1A receptor in mediating functions related to organogenesis and development, we evaluated the survival, growth, and renal structural development of mice lacking AT_1A receptors for angiotensin II [Agtr1A-(/)] compared with their wild-type littermates. Specifically, we were interested in determining whether 1) the absence of AT_1A receptors impairs growth and development and 2) the lack of AT_1A signaling causes structural abnormalities in the kidney similar to those previously observed in Agt-(/) and Ace-(/) mice or in neonatal rats given RAS inhibitors.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Mouse breeding, genotype, and survival analyses. Mice lacking AT_1A receptors for angiotensin II were produced using homologous recombination in embryonic stem cells as described previously (15). Animals were bred and maintained in the Animal Facility at the Durham Veterans Affairs Medical Center under National Institutes of Health guidelines. Adult mice were fed standard chow containing 0.4% sodium chloride and were allowed free access to water. Agtr1A genotypes were determined by Southern blot analysis of DNA obtained from tail biopsies as described (15). Mice used in these studies were all progeny of matings between C57BL/6 × 129/J, F1 Agtr1A-(+/) heterozygote animals.

To determine the effect of the Agtr1A mutation on survival, we analyzed the genotypes of 612 consecutive F2 progeny of C57BL/6 × 129/J, F1 Agtr1A-(+/) crosses. Using ² analysis, we compared the observed distribution of Agtr1A genotypes [(+/+), (+/), and (/)] in 21-day-old animals with the distribution predicted by Mendelian genetics.

Measurement of protein and DNA content in hearts and kidneys. At 21 (n = 28) and 60 days (n = 12) of age, Agtr1A-(+/+) and Agtr1A-(/) mice were weighed and euthanized. The kidneys and hearts were removed, surrounding fat and tissue were carefully dissected away, and the organs were weighed. Individual organs were then homogenized for 1 min in 1.0 ml of phosphate-buffered saline (PBS) containing 2 mM EDTA using a Polytron homogenizer. The homogenates were stored at 70°C until protein and DNA content were measured as described below.

Left kidneys were used for the protein measurements, and right kidneys were used for the DNA determinations. Protein concentration was determined using the Coomassie blue technique (2). Protein content is expressed as milligram per organ. To measure DNA content, homogenates were first disrupted by brief sonication (Heat Systems ultrasonic model W-220F sonicator) to ensure a very fine suspension. DNA concentrations were determined using a fluorometer based on the enhancement of fluorescence seen when bisbenzimidazole dye binds to DNA (model TKO 100; Hoefer Scientific, San Francisco, CA) (22). DNA content is presented as milligrams per organ.

Light and transmission electron microscopy. To remove circulating blood elements, kidneys from adult Agtr1A-(+/+) and Agtr1A-(/) mice (n = 6) were perfused in vivo through the heart with PBS. To preserve the kidneys for morphological analysis, this was immediately followed by perfusion with 1% glutaraldehyde in Tyrode buffer or PBS for 4 min. After the perfusion, the tissue was immersed in the same fixative for an additional 2 h and then was rinsed in buffer. Tissue blocks were sampled from the cortex and both outer and inner medulla. These blocks were postfixed in 2% osmium tetrachloride for 1 h, dehydrated in a graded series of ethanol washes, and embedded in TAAB 812 epoxy resin. One-micrometer sections were cut from different regions of the kidney and stained with toluidine blue. The sections were than examined by light microscopy. Thin sections were stained with uranyl acetate and lead citrate and were examined and photographed on a Zeiss model 10A electron microscope.

Preservation of tissue for immunohistochemistry. The kidneys from five Agtr1A-(+/+) and five Agtr1A-(/) mice were preserved for immunohistochemistry by in vivo perfusion through the abdominal aorta with 2% paraformaldehyde in PBS for 10 min. After perfusion, the kidneys were bivalved and immersed in the same fixative at 4°C. Tissue slices were cut through the entire kidney, dehydrated in a graded series of ethanol washes, and embedded in wax (polyethylene glycol 400 distearate; Polysciences, Warrington, PA).

Light microscopic immunohistochemistry. Four-micrometer wax sections were processed for immunohistochemistry using the avidin-biotin-horseradish peroxidase technique (Vectastain ABC kit; Vector Laboratories, Burlington, CA). The sections were dewaxed, rehydrated, and incubated with 3% H₂O₂ for 30 min to eliminate endogenous peroxidase activity. After treatment with blocking serum for 15 min, the sections were incubated overnight at 4°C with the primary antibody against renin, diluted 1:8,000. Sections incubated without primary antibody served as negative controls. The sections were rinsed in PBS, incubated with secondary antibody against mouse immunoglobulin G for 30 min and subsequently with the Vectastain ABC reagent for 60 min. After being rinsed with PBS, the sections were incubated with the peroxidase substrate solution and then with diaminobenzidine, counterstained with hematoxylin, and examined by light microscopy. The antibody against renin was kindly provided by Dr. Jean Sealey (Cornell University Medical College, New York, NY); it is a rabbit polyclonal antibody directed against recombinant human renin. This antibody has been characterized in detail in previous studies and is known to recognize rat renin and prorenin (4).

Quantitation of renin immunoreactive profiles. To obtain a quantitative estimate of renin immunoreactivity in Agtr1A-(+/+) and Agtr1A-(/) animals, the number of renin-positive profiles were counted in the renal cortex of five animals in each of the two groups of animals. When a profile consisted of several arteriolar branches (e.g., an interlobular artery with several afferent arterioles), each branch was counted separately. The number of labeled profiles and the total number of glomeruli were counted on two sections from each animal, and the number of labeled profiles was expressed per 100 glomeruli. These data are expressed as means ± SD.

Blood pressure measurement. In a subset of weanling mice, blood pressures were measured by intra-arterial catheterization. Animals were anesthetized with isofluorane, and a flexible plastic catheter (0.015 mm ID; Norton, Akron, OH) was placed in the carotid artery. Arterial pressures were monitored and recorded over a period of 10 min, and a mean value was calculated using the Windaq software package (Dataq Instruments, Akron, OH).

Data analysis. Data are presented as the means ± SE. Statistical significance of differences between the two experimental groups was determined using an unpaired t-test.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

We analyzed Agtr1A genotypes of 612 F2 progeny of consecutive C57BL/6 × 129/J, F1 Agtr1A-(+/) crosses. Among this group, 194 (32%) were homozygous for the wild-type allele Agtr1A-(+/+), 299 (49%) were heterozygous Agtr1A-(+/), and 119 (19%) were homozygous for the mutant allele Agtr1A-(/). The observed distribution of genotypes differed significantly from the predicted 25:50:25 Mendelian distribution (P = 0.01 by ² analysis). Similar skewing of genotype distribution was observed among the 60 F2 weanlings from 7 consecutive litters that were used here for the analysis of renal and somatic growth in 21-day-old animals. In this group, 19 (32%) were Agtr1A-(+/+), 32 (53%) were Agtr1A-(+/), and 9 (15%) were Agtr1A-(/).

We have previously found that adult mice lacking AT_1A receptors have a significant reduction in blood pressure compared with controls. To assess the effects of the mutation on blood pressures in weanling mice, intra-arterial pressures were measured in a subset of 21-day-old Agtr1A-(/) (n = 7) and Agtr1A-(+/+) (n = 7) mice. Mean arterial blood pressure in the wild-type Agtr1A-(+/+) mice was 91 ± 4 mmHg. In 21-day-old Agtr1A-(/) mice, the mean arterial pressure was 78 ± 3 mmHg [P < 0.02 vs. Agtr1A-(+/+)].

As shown in Table 1, the absence of AT_1A receptors had no significant effect on somatic growth prior to weaning. Thus, in 21-day-old Agtr1A-(/) and Agtr1A-(+/+) littermates, body weights did not differ significantly. Between 21 and 60 days of age, body weights increased significantly in both groups (P < 0.001), and there was no significant difference between the body weights of 60-day-old Agtr1A-(+/+) and Agtr1A-(/) F2 littermates. Likewise, as can be seen in Table 1, the absence of AT_1A receptors had no effect on kidney growth. There were no significant differences in kidney weights in 21-day-old Agtr1A-Agtr1A-(/) compared with Agtr1A-(+/+) mice, whether expressed as total kidney weight or when normalized to body weight. Kidney weight increased with age in both groups between 21 and 60 days, and there was no difference in kidney weights in 60-day-old Agtr1A-(+/+) and Agtr1A-(/) mice. A similar pattern was seen with heart weights, although by 100 days, heart weight tended to be lower in the Agtr1A-(/) group (P = 0.057).

View this table: [in this window] [in a new window]   Table 1.   Body, kidney, and heart weights in Agtr1A-( +/+) and -(/) F2 littermates

To further assess the role of AT_1A receptors on renal and cardiac growth, we measured protein and DNA content of kidneys and hearts at weaning (21 days) and in mature animals (60 days), comparing Agtr1A-(/) and Agtr1A-(+/+) littermates. As shown in the Table 2, protein and DNA content did not differ significantly in kidneys and hearts from 21-day-old Agtr1A-(/) and Agtr1A-(+/+) mice. Between 21 and 60 days, renal protein and DNA contents increased significantly, reflecting the increases in cell number and growth that occur during this time. But again, there were no significant differences in protein or DNA content in kidneys or hearts from 60-day-old Agtr1A-(/) mice compared with their Agtr1A-(+/+) wild-type littermates.

View this table: [in this window] [in a new window]   Table 2.   Total DNA and protein content of kidneys and hearts from Agtr1A-(+/+) and -(/) F2 littermates

Macroscopically, no abnormalities were noted in the kidneys of Agtr1A-(/) mice, except for an occasional slight dilatation of the renal pelvis and associated mild compression of the renal papillae in the Agtr1A-(/) animals. Such changes can be seen in states of increased urine flow, and in other studies we have found that Agtr1A-(/) mice have very high urine volumes (30). By light microscopy, there were no major alterations in the overall structure of the kidneys of Agtr1A-(/) mice, and there were no signs of damage to any components of the kidney. Specifically, there was no evidence of the papillary and medullary atrophy or the thickened and abnormal intrarenal arteries and arterioles that have been reported in both the Agt-(/) and Ace-(/) mouse lines (6, 19, 21, 29). As shown in Fig. 1, numerous strongly stained granules, most likely representing renin granules, were observed in the juxtaglomerular apparatus (JGA) in both Agtr1A-(/) and Agtr1A-(+/+) animals. However, these granules were much more prominent in Agtr1A-(/) animals (Fig. 1B) than in their Agtr1A-(+/+) littermates (Fig. 1A) and were often seen in cross sections of the arterioles. In addition, as demonstrated in Fig. 1B, some glomeruli in Agtr1A-(/) mice exhibited evidence of modest mesangial cell expansion. No abnormalities in the morphology of renal tubular cells were detected in any segments of the nephron or in the collecting duct.

View larger version (147K): [in this window] [in a new window]   Fig. 1.   Light micrographs of renal cortex from Agtr1A-(+/+) and Agtr1A-(/) mice. A: Agtr1A-(+/+), with cells containing renin granules (arrow) in the juxtaglomerular apparatus (JGA). B: Agtr1A-(/), with increased numbers of cells containing renin granules in cross sections of the afferent arterioles (arrows). Glomerulus in the Agtr1A-(/) animal also exhibits mesangial expansion (asterisk). Magnification, ×570.

To confirm the identity of the granules observed by light microcopy and to quantitatively assess the degree of JGA expansion, we performed immunostaining of kidney sections using an anti-renin antibody. As shown in Fig. 2, there was a striking increase in the intensity of immunostaining and in the number of renin-positive sites in the renal cortex of Agtr1A-(/) mice when compared with Agtr1A-(+/+) mice. This was reflected by an increase in the number of immunoreactive renin profiles from 49 ± 8 positive profiles per 100 glomeruli in Agtr1A-(+/+) mice to 93 ± 32 positive profiles per 100 glomeruli in Agtr1A-(/) mice. This increase in immunoreactive profiles represented mainly an expansion of staining proximally along the afferent arterioles and interlobular arteries.

View larger version (146K): [in this window] [in a new window]   Fig. 2.   Immunostaining of kidney sections from Agtr1A-(+/+) and Agtr1A-(/) mice. A: Agtr1A-(+/+), with positive immunostaining localized to a small segment at the JGA. B: Agtr1A-(/), with a substantial increase in the intensity and the number of labeled profiles compared with the Agtr1A-(+/+) animal. Magnification, ×190.

Transmission electron microscopy of the kidneys likewise showed the presence of tightly packed granules in the afferent arteriole of the JGA in both the Agtr1A-(+/+) and Agtr1A-(/) mice (Fig. 3). Again, the cells containing renin granules were much more prominent in the Agtr1A-(/) mice (Fig. 3B) than in their Agtr1A-(+/+) littermates (Fig 3A). Furthermore, as shown Fig. 3C, these granules could be observed outside of the JGA in cells surrounding the afferent arterioles of Agtr1A-(/) but not Agtr1A-(+/+) mice. As shown in Fig. 4B, some glomeruli in Agtr1A-(/) mice had expanded mesangial regions with increased amounts of mesangial matrix. These abnormalities were not seen in the Agtr1A-(+/+) animals (Fig. 4A). No ultrastructural abnormalities were detected in epithelial cells in proximal or distal tubule or in the collecting duct.

View larger version (146K): [in this window] [in a new window]   View larger version (171K): [in this window] [in a new window]   View larger version (205K): [in this window] [in a new window]   Fig. 3.   Transmission electron micrographs showing renin granules in afferent arterioles of Agtr1A-(+/+) and Agtr1A-(/) mice. A: densely packed granules containing renin, in afferent arteriole of JGA from an Agtr1A-(+/+) mouse. G, glomerulus. Magnification, ×3,100. B: transmission electron micrograph of afferent arteriole from JGA of an Agtr1A-(/) animal; note that cells containing renin granules are more frequent and more prominent than in Agtr1A-(+/+) mice. Two erythrocytes can be seen in the lumen of afferent arteriole. Magnification, ×3,000. C: cross section of an afferent arteriole from an Agtr1A-(/) mouse; note presence of numerous renin granules in cells surrounding the arteriolar wall outside of JGA. Magnification, ×2,600.

View larger version (165K): [in this window] [in a new window]   View larger version (189K): [in this window] [in a new window]   Fig. 4.   Transmission electron micrographs of glomeruli from Agtr1A-(+/+) and Agtr1A-(/) mice. A: mesangial region (M) appears normal in the Agtr1A-(+/+) animal. B: in glomerulus of the Agtr1A-(/) mouse, the mesangium (M) is expanded with large amounts of mesangial matrix. Magnification, ×4,100.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The RAS is a major regulator of blood pressure and fluid balance (32). Alterations in the activity of the RAS can cause hypertension (16) and may contribute to chronic kidney injury (1). Regulation of fetal development and growth has been suggested to be another important role of the RAS (8, 9, 28, 39). Virtually all of the recognized physiological functions of the RAS, including stimulation of cell growth and proliferation, are thought to be mediated by the type 1 (AT₁) angiotensin receptor (33, 38). In mice and rats, there are two AT₁ receptor isoforms, AT_1A and AT_1B (38). These receptors are products of separate genes, but they have very high levels of sequence homology. Their pharmacological characteristics and signaling mechanisms are virtually identical, and there are no known antagonists that discriminate between the two AT₁ receptors. In most tissues, including the kidney, the AT_1A receptor is the predominant receptor. However, we show here that the complete absence of AT_1A receptors does not significantly alter growth and developmental processes in the kidney. Our studies consequently demonstrate that a functional Agtr1A gene is not essential for normal renal development and growth.

In an analysis of a smaller group of animals, we and others have previously reported that the absence of AT_1A receptors did not affect survival in mice (15, 27). In the current analysis of more than 600 progeny of a series of F1 matings, we now find a modest skewing of the expected Mendelian distribution of genotypes (P < 0.01). Sugaya and associates (36) have reported a similar trend toward reduced numbers of Agtr1A-(/) mice in an analysis of the genotypes of a group of 396 progeny of Agtr1A-(+/) matings, although in their study this trend was not statistically significant. Thus the complete absence of a functional Agtr1A gene results in a mild survival disadvantage before or during early postnatal life. The reason for reduced survival is not clear, but it is not due to gross developmental abnormalities in the kidney or other major organ systems. We speculate that, in the absence of AT_1A receptors, young neonates may have a reduced ability to compete as a result of their reduced blood pressures, inability to conserve sodium normally, and their need for a high water intake consequent to abnormalities in their urinary concentrating functions (30).

Our finding of a mildly diminished survival of Agtr1A-(/) mice contrasts with the much more substantial reduction in survival of mice that are homozygous for targeted disruptions for the angiotensinogen (Agt) or ACE (Ace) genes. The few Ace-(/) or Agt-(/) mice that survive to adulthood develop characteristic abnormalities of renal histology and renal vasculature (6, 19, 21, 29); these abnormalities consist of marked thickening of arteriolar walls with medial hypertrophy confined exclusively to the renal vasculature. We do not find these structural abnormalities in the kidneys of adult Agtr1A-(/) mice. This suggests that the absence of signaling through angiotensin receptor pathways other than AT_1A must be involved in producing the abnormal vascular lesions observed in Agt-(/) and Ace-(/) mice. Reports of normal development and survival of mice lacking AT₂ (12, 14) and AT_1B receptors (5) raise the possibility that there may be an unidentified angiotensin II receptor whose functions include regulation of vascular growth and integrity. Alternatively, the combined absence of signaling through at least two of the known receptor subtypes may be required to reproduce this vascular phenotype.

Our results differ from previous studies in which the effects of pharmacological AT₁ receptor blockade have been examined in weanling rats. For example, Tufro-McReddie and associates (40, 41) reported that administration of the AT₁ receptor antagonist losartan to neonatal rats impairs growth and arrests nephrovascular development. There are several potential explanations for these differences. First are the obvious inherent differences between the genetic and pharmacological approaches that were used in the studies. With gene targeting, a specific and complete absence of AT_1A receptors is achieved from conception onward without direct effects on any other genes. In experiments using pharmacological antagonists such as losartan, the observed effects might be caused by actions of the agent that are unrelated to AT₁ receptor antagonism. In addition, since losartan blocks signaling through both AT_1A and AT_1B receptors, some effects of losartan may reflect concomitant inhibition of both AT₁ receptor isoforms. Alternatively, there may be significant species differences between rat and mouse in the expression and/or function of angiotensin receptors in kidney and elsewhere. Indeed, some differences in angiotensin receptor expression between rat and mouse have been reported. For example, only low levels of AT₂ receptors are expressed in adult rat kidney (34), whereas AT₂ receptors are easily detected in adult mouse kidney (15). However, the pattern and levels of expression of AT₁ receptors appear to be quite similar in rat and mouse kidneys where AT_1A is the predominant AT₁ receptor isoform (15, 20, 24).

Differences in the activity of compensatory systems might also explain the differences in the outcome of experiments using gene targeting compared with pharmacological RAS inhibitors. For example, because the AT_1A receptor is absent through all stages of development in the genetic experiments compared with a more limited period of pharmacological inhibition, the effects on compensatory systems might be quite different. However, using renin as an example of one such compensatory system, the absence of a functional Agtr1A gene and treatment with losartan have very similar effects. Thus, in the present study, we observed an increase in the number of renin-containing granules in the juxtaglomerular cells in Agtr1A-(/) mice. Furthermore, Sugaya and associates (36) have reported elevated plasma renin levels in Agtr1A-(/) mice. Similar enhancements of renin expression and an increase in renin-producing cells in the kidney are seen during losartan treatment of weanling rats (41).

Activation of the RAS plays a role in the progression of a variety of kidney diseases, including diabetic nephropathy (44) and glomerulosclerosis associated with partial renal ablation (1). In these disorders, angiotensin II promotes the development of glomerular injury and fibrosis through mechanisms that are independent of its effects on systemic blood pressure (13). These effects are mediated by AT₁ receptors (23). Furthermore, AT₁ receptors stimulate growth and proliferation of mesangial cells in vitro (33). Thus our finding of mesangial expansion in adult mice that lack AT_1A receptors was quite unexpected. The specific cause of these mesangial changes is not clear. Although the expansion of mesangial matrix may result directly from the lack of AT_1A signaling in mesangial cells, it is also possible the decreased blood pressure of the Agtr1A-(/) mice might lead to stimulation of mesangial cell growth mediated by sympathetic nerves or non-AT_1A angiotensin receptors. A cautionary note is required, namely that C57BL/6 strain mice have a propensity to develop spontaneous mesangiopathic changes and glomerulopathy (25). Because the genetic background of our Agtr1A-(/) mice consists of an F2 mixture of the C57BL/6 and 129/J strains, we cannot rule out the possibility that the abnormal mesangium seen in some Agtr1A-(/) mice may result from the chance assortment of certain background genes. It will be possible to test that possibility as we breed the Agtr1A mutation onto a series of different inbred backgrounds.

In summary, the genetic absence of AT_1A receptors in the mouse produces a mild but significant survival disadvantage. In surviving F2 Agtr1A-(/) mice, somatic and renal growth and development proceed normally, although some of the adult Agtr1A-(/) mice develop modest mesangial expansion. We conclude that the AT_1A receptor is not essential for the normal organogenesis and growth of the kidney.