INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

IN ADDITION to the systemic renin-angiotensin system (RAS), which plays an important role in regulating blood pressure and body fluid homeostasis, increasing evidence suggests the existence of a number of intrinsic tissue RAS. Intrinsic RAS have been proposed to exist in tissues capable of de novo synthesis of all RAS components and, although their function remains poorly understood, are thought to operate, in part or entirely, independent of the systemic RAS. By virtue of the actions of angiotensin II (ANG II) produced and delivered locally in a tissue, tissue RAS may act to regulate cell growth and proliferation and therefore may be important developmentally.

Clinical observations have demonstrated an association between treatment of hypertensive pregnant women with angiotensin-converting enzyme (ACE) inhibitors and fetal oligohydramnios, neonatal renal failure, and hypocalvaria (2, 29). The finding that fetal or neonatal treatment of rats with ACE inhibitors caused renal abnormalities suggests that ANG II plays an important role in renal development (29). Moreover, the observation that angiotensinogen-deficient (MAGT /) mice die between birth and 3 wk of age and have marked renal abnormalities strongly supports the importance of ANG II in continued renal development after birth (17). Our recent observation that the lethality and renal abnormalities observed in MAGT / mice can be rescued by the human angiotensinogen (HAGT) (and human renin) suggests that the temporal and spatial expression of HAGT in those mice is sufficient to prevent the developmental defects from becoming apparent (5). Although it is not clear whether the systemic or the tissue RAS is involved in fetal development, the autocrine and paracrine actions of tissue ANG II as a growth factor appear to be a major contributor to the developmental regulation of the cardiovascular, renal, and central nervous systems (9, 12). Therefore, a key issue to understanding the developmental role of the RAS is to delineate the expression pattern exhibited by components of the RAS during growth of the fetus.

Although developmental regulation of rat and sheep AGT has been reported (7, 26), little information is available regarding the regulation of HAGT gene during prenatal and neonatal life. We therefore examined the developmental expression of HAGT, using a transgenic mouse model, which we previously reported to exhibit appropriate cell- and tissue-specific expression, to generate physiologically functional HAGT protein and to cause chronic hypertension in the presence of human renin or the human renin gene (21, 34).

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Timed breeding of transgenic mice. Male HAGT transgenic mice from transgenic line 204/1 were crossed with female C57BL/6J mice, and daily observation for the presence of vaginal plugs was undertaken to determine the initiation of pregnancy. Mouse fetuses from separate litters were removed for gestational or embryonic days 6.5 (E6.5) through E18.5. Newborn and 2-wk-old mice were also collected. Positive transgenic fetuses, newborn, and 2-wk-old mice were identified by PCR analysis of genomic DNA purified from placenta or tail biopsies, as described previously (34). In some experiments, the sex of the mouse fetuses was determined by PCR analysis of genomic DNA from tail or placental biopsies, amplifying a fragment from the mouse sry gene (sex determining region Y), as described previously (10). The primer set employed the oligonucleotides 5' GAGAGCATGGAGGGCCAT 3' and 5' CCACTCCTCTGTGACACT 3'.

Developmental time course of HAGT expression. RNase protection assay was performed to compare the time course of expression of the HAGT transgene with that of the endogenous mouse AGT (MAGT) gene during gestation. Total RNA was isolated from whole fetuses pooled from pregnancies from E7.5 to E13.5 (n = 5). Fetuses were separated from all membranes prior to pooling and RNA extraction. Preparation of total tissue RNA and labeling of an antisense RNA HAGT probe was performed as described previously (34). An antisense RNA probe to detect MAGT mRNA was transcribed from a partial cDNA clone that was derived from exon 2 of MAGT at nucleotide coordinates 302-811, relative to the transcription start site of the MAGT cDNA. This MAGT cDNA fragment was PCR amplified and subcloned into the pCR2.1 vector (Invitrogen), using the oligonucleotide primers 5' GACACACAGAAGCAAATGCAC 3' and 5' TCTCCCTCCTTCACAGGGAC 3'. The SP6 and T7 RNA promoters were identified by DNA sequencing as the antisense transcript promoters for HAGT and MAGT, respectively. Additionally, a mouse glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antisense RNA probe generated from a cDNA template (Ambion) was included as an internal control for sample loading. Full-length probes were purified on a 5% acrylamide gel and eluted after in vitro transcription into buffer containing 2 mol/l ammonium acetate and 1% SDS. For the hybridization reaction, 20 µg of total RNA, full-length labeled RNA probe (1 × 10⁵ cpm), and 30 µl hybridization solution [80% deionized formamide, 0.4 mol/l NaCl, 40 mmol/l PIPES (pH 6.4), and 1 mmol/l EDTA (pH 8.0)] were heated to 80°C for 10 min and then incubated at 50°C for 12-16 h. After hybridization, nonhybridized and nonspecifically bound antisense RNAs were removed by RNase T1 for 30 min at room temperature [10 mmol/l Tris (pH 7.5), 5 mmol/l EDTA (pH 8.0), 200 mmol/l sodium acetate (pH 7.5), and 50 U/ml RNase T1; GIBCO-BRL, Life Technologies]. RNase T1 was inactivated by adding 0.1 mg proteinase K into the reaction and incubating at 37°C for 20 min. After phenol-chloroform extraction, the protected RNA was ethanol precipitated and resuspended in RNA running buffer. Finally, protected RNA was separated on a 5% acrylamide gel and exposed to X-ray film overnight. Protected fragments of 539, 509, and 300 nucleotides in length were indicative of HAGT, MAGT and GAPDH, respectively. HAGT and MAGT mRNAs were assayed on separate gels. There was no cross-reactivity between HAGT mRNA and the MAGT probe and vice versa.

Northern blot analysis was employed to examine the tissue-specific expression of HAGT during development. Tissues including liver, kidney, heart, lung, and brain were removed from positive transgenic mice and negative littermates at E16.5-E18.5, newborn, and 2 wk of age. To obtain a sufficient amount of total RNA, kidneys and hearts from fetuses of the same gestational age were pooled together (n = 5), whereas liver, lung, and brain from individual fetuses were used. Total RNA was isolated from the different tissues as described above. RNA blots were then hybridized to a ³²P-labeled antisense RNA probe transcribed as described above. Total RNA from the corresponding tissues of nontransgenic mice were used as a control.

Localization of the HAGT gene expression during development. In situ hybridization analysis was performed to detect HAGT mRNA in both whole fetus and kidney during development. Whole fetuses from E9.5, E11.5, and E13.5 and kidneys from mice of gestation days E15.5-E18.5, newborn, and 2 wk of age were removed and immediately placed in 30% dextrose at 4°C overnight before freezing in dry ice and storage at 80°C. Only male fetuses and neonates were used for in situ hybridization analysis. Sex of fetuses was determined as described above. Frozen sections were cut on a Reichert-Jung cryostat at 8 µm, mounted onto slides, and hybridized to an antisense RNA probe colabeled with [³H]UTP and [³H]CTP, as described previously (34). After hybridization, sections were washed and covered with liquid emulsion (Kodak) for autoradiography. Sections were then stained by hematoxylin and eosin for histological examination. Bright- and dark-field images were photographed after 1, 3, or more weeks of exposure. Specificity of the probe was confirmed by applying the antisense RNA probe to kidney or whole fetus sections from nontransgenic mice and by applying sense RNA probe to alternating kidney or whole fetus sections from transgenic mice.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Expression of HAGT during development. To determine the earliest time of HAGT expression during development, RNase protection assay was performed on total RNA isolated from whole mouse fetuses obtained at E7.5-E11.5. Low-level expression of HAGT was first detected at E8.5, then it was increased and was similar in E9.5 and E11.5 embryos (Fig. 1A). Mouse AGT expression was first clearly detected at E10.5, and its level was markedly increased at E11.5 (Fig. 1B). No MAGT mRNA was detected before E10.5, indicating an earlier temporal expression of HAGT mRNA. Mouse and rat AGT exhibit a similar temporal pattern of expression (19). The specificity of the probes is demonstrated by the absence of HAGT signal in the nontransgenic fetus (Fig. 1A) and the absence of a MAGT signal in the E8.5 and E9.5 transgenic samples (Fig. 1B), a time when HAGT mRNA is detectable.

View larger version (42K): [in this window] [in a new window]   Fig. 1.   Time course of HAGT and MAGT expression in transgenic mice. RNase protection assay analysis of total RNA isolated from whole mouse fetuses is shown. Antisense RNA probes for HAGT mRNA and MAGT mRNA were generated as described in MATERIALS AND METHODS. A: HAGT probe. B: MAGT probe. , Nontransgenic mice; +, transgenic mice; E, gestation day; Tg, transgenic mice; HAGT, human angiotensinogen; MAGT, mouse angiotensinogen; mGAPDH, mouse glyceraldehyde-3-phosphate dehydrogenase antisense control probe.

Tissue- and developmental-specific expression of HAGT in transgenic mouse fetuses obtained at E16.5-E18.5 and in newborn and 2-wk-old transgenic mice was assessed by Northern blot (Fig. 2A). These analyses revealed that the HAGT transgene was expressed in liver, kidney, heart, and brain throughout late gestation and in neonates. Expression was generally higher in liver and kidney than in heart and brain, and the same results were obtained in three independent experiments, using different samples. Extremely low levels of HAGT mRNA were detected in some but not all lung samples, indicating it to be below the limit of detection by our Northern blots. No HAGT expression was detected in the nontransgenic littermate controls. We observed a substantial decrease in hepatic (but not renal) HAGT mRNA in newborn transgenic mice, compared with samples taken from E16.5-E18.5 embryos and 2-wk-old mice (Fig. 2A). Because these initial samples were not typed to determine the sex of the embryos and since we previously reported some sexually dimorphic expression of HAGT (34), we repeated this experiment using total liver RNA obtained only from male fetuses identified by genotyping each fetal sample at the sry locus (see MATERIALS AND METHODS). A consistent pattern of decreased expression was observed in the male newborns (Fig. 2B) and was reproducible in two different sets of timed breedings. Equivalent loading of RNA in each lane was confirmed with a mouse 18S ribosomal RNA internal control (Fig. 2B, bottom).

View larger version (66K): [in this window] [in a new window]   Fig. 2.   Tissue-specific expression of HAGT gene in transgenic mice during development. A: Northern blots of total RNA from liver, kidney, heart, brain, and lung of transgenic (+) and nontransgenic () mice at different gestation stages, as indicated. Kidney and heart samples were pooled (n = 5) from fetuses from the same pregnancy. Adult male (M), female (F), and nontransgenic () mice were included as controls. An antisense RNA probe for HAGT mRNA was used. Closed and open arrows (left) show position of the 28S and 18S rRNAs, respectively, determined by staining the nitrocellulose blots with methylene blue prior to hybridization. Exposure time was 24 h for liver, kidney, and heart blots and 4 days for brain and lung. B: Northern blots of total RNA from liver during gestation and neonatal period are shown. Northern blots were hybridized with antisense HAGT RNA probes. Methylene blue-stained 18S rRNA is shown for confirmation of sample loading (bottom). All samples were obtained from male fetuses typed as described in MATERIALS AND METHODS. NB, newborn; 2W, 2 wk of age; A, adult; HAGT, human angiotensinogen.

Cell specificity of HAGT expression during development. In situ hybridization analysis was performed to examine the cell and tissue localization of HAGT expression more closely. In the first set of experiments, HAGT mRNA localization was examined in sections of whole fetuses removed at E9.5, E11.5, and E13.5. There was no detectable expression in E11.5 (Fig. 3A) or in E9.5 or E13.5 (data not shown) embryos, when a sense orientation control probe was used. Expression of HAGT was evident in the heart and liver of E9.5 embryos (Fig. 3B). HAGT expression in both tissues markedly increased both in intensity and cell number at E11.5 and E13.5 (Fig. 3, C and D). Hybridization was found equally in all of the compartments of the heart (ventricle, atrium, and septum) and appeared in both developing myocardium and endocardium (Fig. 4, A-C). HAGT expression was also observed in the descending aorta (Fig. 3C), choroid plexus, primitive gut, fetal adrenal gland, and primordium of vertebra and tail (Fig. 3D). There was no detectable HAGT expression in lung or parenchyma of the brain (Fig. 3, C and D).

View larger version (104K): [in this window] [in a new window]   Fig. 3.   Cell type-specific expression of HAGT during development. In situ hybridization analysis was performed on sections of whole HAGT transgenic fetuses (male) from different gestational stages as described in MATERIALS AND METHODS. Frozen whole fetus sections were hybridized with a 3H-labeled sense (A and A') and antisense (B and B', C and C', and D and D') HAGT RNA probes. A-D: bright-field photomicrographs; A'-D': corresponding dark-field photomicrographs. A and A', embryonic day 11.5 (E11.5); B and B', E9.5; C and C', E11.5; D and D', E13.5. B, brain; H, heart; Lg, lung; Lv, liver; K, kidney; Ad, adrenal gland; T, tail; A, atrium; V, ventricle; G, intestine; Ub, umbilici; cp, choroid plexus; a, aorta.

View larger version (93K): [in this window] [in a new window]   Fig. 4.   Cell type-specific expression of HAGT during development. An enlargement of inset regions from Fig. 3 are shown. A is from Fig. 3B (E9.5), B is from Fig. 3C (E11.5), C is top inset from Fig. 3D (E13.5), and D is bottom inset from Fig. 3D (E13.5) (all bright-field micrographs; A'-D' are dark-field micrographs). Lg, lung; Lv, liver; K, kidney; Ad, adrenal gland; A, atrium; V, ventricle; G, intestine.

There was no detectable HAGT mRNA in the mesonephric kidney at E11.5 and only low levels in the metanephric kidney at E13.5 (Figs. 3D and 4D). To localize the expression of HAGT in kidney more precisely, a second set of experiments were performed on kidneys taken from E15.5-E18.5 embryos, neonates, and mice at 2 wk of age. No hybridization signal was detected at any stage when a sense orientation control probe was used (Fig. 5, A-F). At E15.5, low levels of HAGT expression were observed in some cordlike structures scattered in the metanephric mesenchyme of the primitive kidney (Fig. 5G). The levels of HAGT transgene expression markedly increased in the medial region of the kidney at E16.5 (Fig. 5H). From E17.5 through 2 wk of age (Fig. 5, I-L), HAGT expression in the kidney became completely restricted to tubule cells in the cortex, but not medulla, consistent with cell-specific expression of the HAGT transgene we previously reported in adult transgenic mice and the cell specificity of AGT reported previously in rats (13).

View larger version (123K): [in this window] [in a new window]   View larger version (135K): [in this window] [in a new window]   Fig. 5.   Cell-specific expression of HAGT in kidney during development. In situ hybridization analysis was performed on kidney sections of HAGT transgenic fetuses, as described in MATERIALS AND METHODS. Frozen kidney sections were hybridized with a 3H-labeled sense (A and A' through F and F', above) and antisense (G and G' through L and L', facing page) HAGT RNA probes. A-L are bright-field photomicrographs, whereas A'-L' are the corresponding dark-field photomicrographs. A and A' and G and G', E15.5; B and B' and H and H', E16.5; C and C' and I and I', E17.5; D and D' and J and J', E18.5; E and E' and K and K', newborn; F and F', L and L', 2 wk of age.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

We demonstrated that HAGT is developmentally regulated in transgenic mice by examining the timing of HAGT expression and by in situ hybridization. These results are consistent with the tissue-specific pattern of expression exhibited by adult transgenic mice containing HAGT (34), the developmental pattern of AGT expression in rodents (7), and the developmental pattern of AGT minigenes in other transgenic mice (4), strongly suggesting that our transgene contains the necessary regulatory elements needed to target appropriate tissue- and cell-specific expression of HAGT during fetal development.

Developmental expression and hormonal regulation of HAGT. We determined the earliest expression of mouse and human AGT expression during fetal development in whole mouse embryos. We found that MAGT mRNA is first expressed around E10.5, which is similar to the appearance of rat AGT mRNA (19). On the other hand, low-level HAGT expression was detected at E8.5 and was highly expressed by E9.5. The reason for the earlier activation of transgene expression remains unclear. In these studies, we employed a transgene consisting of all exons, introns, and flanking sequences extending 1.2 kb upstream and 1.4 kb downstream of the gene. We reported that the transgene exhibited appropriate tissue- and cell-specific expression and that the level of HAGT mRNA in this transgenic line (204/1) was markedly higher than expression of endogenous MAGT mRNA (34). Moreover, we recently reported that the level of HAGT expression in thirteen independent transgenic lines was proportional to transgene copy number (35) and that the copy number in the line used for these studies is several times greater than the endogenous gene. Therefore, the detection of HAGT at E8.5 may merely reflect a higher level of expression. Copy number proportional expression of transgenes generally indicates the presence of cis-acting elements, which can manipulate chromatin at the local level (6). In general, such chromatin-modulating sequences may function as locus control regions, which act either as enhancers or in concert with enhancers controlling the transcription of genes. Indeed, when examined in total, our data are consistent with the conclusion that all sequences necessary to confer appropriate temporal (developmental) and spatial (tissue- and cell-specific) expression of HAGT are present within our transgene.

It is of interest to note that there was a substantial decrease in HAGT gene expression in liver in newborn transgenic mice, which returned to late gestation levels by 2 wk of age. Although the significance of this drop in expression in newborns is not well understood, we speculate that it may result from changes in glucocorticoid hormone levels at birth. The level of hepatic AGT mRNA and AGT transgene mRNA was reported to be responsive to glucocorticoids (4). Moreover, rats (and presumably mice) undergo dramatic changes in the level of plasma corticosterone, the physiologically active glucocorticoid, around birth (reviewed in Ref. 28). High levels of the hormone are present throughout late gestation (E15-E20) but markedly decrease (>90% reduction) around birth. Hormone levels then recover back to prebirth levels during the 2nd wk of life. The period between birth and 2 wk of age is referred to the stress-hyporesponsive period, during which rats are poorly responsive to stresses that efficiently activate glucocorticoids in adults. Our data showing a decrease in HAGT mRNA around birth with recovery in mice 2 wk of age suggest a glucocorticoid-hyporesponsive period may also occur in neonatal mice. Similar results have also been reported in sheep, where it was proposed that this change in expression around birth may reflect a "switch" in the function of the RAS from its role as a regulator of growth and hemodynamics during fetal development to its primary role as a regulator of hemodynamics postnatally (26, 27).

RAS and kidney development. There was no detectable HAGT expression during the mesonephric stage of kidney development (E11.5). As the fetal kidney developed to the next stage with the appearance of a poorly differentiated metanephros and some collecting tubules around E13.5, very low but detectable levels of the transgene were found. At E15.5, when primitive glomeruli and tubules start appearing in kidney, increased HAGT expression was observed in cordlike structures surrounding the mesenchyme. E16.5 is a critical time in kidney development, when cortex and medulla begin to become distinct and functional proximal tubule cells start differentiating. It was at this stage, when epithelialization and vascularization of the kidneys begin (15) and later, that high-level expression of the transgene was found to be localized to the proximal tubule.

The RAS has been shown to play an important role in the development of the kidney (9, 32). Studies on AGT-deficient mice revealed that the loss of endogenous AGT does not cause apparent renal abnormalities in mice during late gestation (which may receive ANG II from the maternal circulation) but leads to a rapid progression of severe renal lesions between the neonatal and weanling period (3 wk of age) (17). These data are consistent with the observations that ACE inhibitors affect renal development during late gestation (29). Interestingly, the presence of lesions coincided with an upregulation of growth factor expression in the kidney (25). Although de novo ANG II production is clearly critical for the maintenance of appropriate renal morphology after birth, the exact source of ANG II remains unclear. It must be established whether ANG II produced locally in the kidney (from AGT produced locally in proximal tubules) or derived from the systemic circulation (from hepatic-derived AGT) is the important mediator of renal development. These results, along with the findings of developmentally regulated renin and ANG II receptor mRNA in the kidney, support an important role for the RAS during renal development both in utero and after birth (8, 31).

RAS and cardiac development. We found that HAGT was expressed in myocardium and endocardium of developing heart as early as E9.5 and at a high level by E11.5. HAGT expression was also detected in aorta after E11.5. Primordial cardiac tissue is apparent as early as E8.5, and cardiac and large vessel asymmetry occurs between gestation days E10.5 and E13.5 in the normal mouse (15). It is possible that ANG II present in heart during these stages may be critically involved in regulating heart developmental processes. The RAS has been suggested to play a role in regulating cardiovascular development (16), and several lines of evidence point to the effects of ANG II-induced growth factors (3, 14, 20). Indeed, growth factors in endocardium, as well as in myocardium, have been shown to play an important role in heart development (24).

Finally, ANG II has trophic or mitogenic effects on a variety of target tissues (11, 30) and has been shown to induce expression of transcription factors (23) and growth factors (22), to have angiogenic effects (1, 18), and has been reported to play an important role in mediating programmed cell death (33). Taken together, there appears to be a clear, yet undetermined, role for the RAS in embryogenesis.