Angiotensin deficient FVB/N mice are normotensive

All previous rodent models lacking the peptide hormone angiotensin II (Ang II) were hypotensive. A mixed background strain with global deletion of the angiotensinogen gene was backcrossed to the FVB/N background (Agt‐KO), a strain preferred for transgenic generation. Surprisingly, the resulting line turned out to be normotensive. Therefore, this study aimed to understand the unique blood pressure regulation of FVB/N mice without angiotensin peptides.


| INTRODUCTION
The classical renin angiotensin system (RAS or RAAS, renin angiotensin aldosterone system) committing the production of the peptide hormone angiotensin II (Ang II) in the circulation is a pivotal system controlling cardiovascular homeostasis as well as salt and water balance. Among the mechanisms by which Ang II regulates blood pressure (BP) are vascular smooth muscle contraction, aldosterone production and secretion, cardiac contractility, vasopressin release and sympathetic nerve activity (SNA). These physiological processes are in their majority mediated by the binding of Ang II to the angiotensin type 1 receptor (AT 1 ) expressed by different cell types in cardiovascular organs (Bader, 2010;Bekassy et al., 2022;. The broad modulatory effects of Ang II on BP render angiotensin-converting enzyme (ACE) inhibitors and AT 1 antagonists (inhibiting Ang II/AT 1 axis) to be one of the most successful drug-based therapies to lower BP in human hypertension (Carey & Whelton, 2018). Humans have one gene encoding the AT 1 receptor, whereas rodents have two (Agtr1a and Agtr1b; AT 1a and AT 1b , respectively). Selective targeted deletion of these genes demonstrated that the AT 1a receptor is the major protein controlling cardiovascular homeostasis, because mice lacking the AT 1a receptor are hypotensive contrary to AT 1b deficient mice (Sparks et al., 2014).
All genes encoding proteins required for Ang II formation (the precursor protein angiotensinogen [Agt] and the two enzymes, renin and ACE) have been previously deleted in mouse models. All three mouse lines displayed hypotension, polyuria, polydipsia and post-natal renal morphological alterations including hydronephrosis (Krege et al., 1995;Takahashi et al., 2005;Tanimoto et al., 1994). These alterations were similar in double knockout mice lacking both AT 1 receptor subtypes demonstrating a fundamental role of the Ang II/AT 1 axis on BP control and renal development (Gembardt et al., 2008;Tsuchida et al., 1998).
Here, we present a puzzling finding considering all the knowledge acquired during the last three decades in animal model with gene targeted deletion of RAS proteins leading to the elimination of the Ang II/AT 1a axis in rodents. Surprisingly, FVB/N mice completely lacking Agt (Agt-KO) are normotensive. Comprehensive phenotyping of this line indicated that Agt-KO develop a strong neurogenic pressor activity compensating for the RAS loss.
Originally, Agt-KO mice were obtained in a mixed background strain and the line was backcrossed for more than 10 generations with the FVB/N (FVB/NCrl; Charles Rivers #207, RRID:IMSR_CRL:207) strain at the animal facility of the Max Delbrück Center, Berlin. The FVB/N background strain is ideal for transgene mouse generation due to the large pronucleus that facilitates the microinjection of DNA constructs (Popova et al., 2005). Our group has generated and backcrossed several mouse lines to the FVB/N background strain to study the role of the RAS in cardiovascular control (Cardoso et al., 2010;Rodrigues et al., 2021;Rosendahl et al., 2014;Xu et al., 2008). Interestingly, by studying components of the protective arm of the RAS, our group has previously detected differences in cardiovascular control among FVB/N and C57BL/6 strains (Rabelo et al., 2016). Therefore, we decided to generate a RAS-deficient FVB/N line by means of genetically deleting the RAS precursor protein Agt.

| Animal experiments
All experiments involving animals (mice) are reported in compliance with the ARRIVE guidelines (Percie du Sert et al., 2020) as well as recommendations of the British Journal of Pharmacology (Lilley et al., 2020). Mice were kept in individually ventilated cages at a maximum number of 6 in a room with controlled temperature of 22 ± 1 C, under a standard light/dark cycle of 12 h each. All mice received commercial standard mouse chow and water ad libitum. For experiments, mice were transferred to experimental rooms at least 5 days prior to the experiment and were kept in conventional open cages covered with a polyester filter sheet. Food, water, room temperature and light schedule were kept as described above. All experiments were performed with adult male mice at 12 to 15 weeks of age. Surgical procedures were performed under anaesthesia with ketamine and xylazine (100 and 10 mgÁKg À1 i.p., respectively) and isoflurane inhalation (1%-2%).

What is already known
• Rodents lacking components of the angiotensin II/AT 1a axis are life-long hypotensive.

What does this study add
• The loss of the renin-angiotensin system in FVB/N mice has no effect on baseline blood pressure.
• Increased vascular tone due to exaggerated vascular sympathetic activity and reduced nitric oxide production prevent the blood pressure fall.
What is the clinical significance During surgeries mice breath spontaneously and body temperature was maintained performing the procedure on a thermo-controlled plate that is connected to a rectal temperature sensor. Pain was managed providing 200 mgÁkg À1 Áday À1 metamizole in drinking water before and after the surgeries. After any surgical procedure, mice were individually monitored at least once a day. In vivo experiments were performed at the animal facility of the Max Delbrück Center for Molecular Medicine, Berlin, following the EU directive 2010/63/EU for animal experiments. All in vivo procedures were previously approved by the Berlin State Office for Health and Social Affairs (Landesamt für Gesundheit und Soziales).

| Cardiovascular phenotyping
Acute BP recording with heparin-filled catheter: Baseline acute BP and heart rate (HR) were recorded in freely moving mice using self-made catheters as previously described (Rodrigues et al., 2021;Todiras et al., 2017). Briefly, femoral catheters were implanted in the femoral artery and vein to reach the abdominal aorta and vena cava before the renal bifurcations. The catheters were exteriorized in the interscapular region and fixed with a silk-suture (3/0). Recordings of BP and HR were initiated 2 days after surgery in freely moving mice. Cardiovascular parameters were recorded constantly (beat-by-beat) at 200 Hz connecting the arterial catheter to a pressure transducer (AD Instruments #MLT0699) connected to a PowerLab/4sp via bridge amplifiers (AD Instruments #ML110). Baseline mean arterial pressure (MAP) and HR were obtained before any drug administration averaging the recorded data for $1 h after a habituation period of the approximate same length. MAP was automatically calculated with the formula MAP = ((systolic + [2 x diastolic])/3) and the cardiac frequency (HR) was deduced from the oscillatory pressure waveform. The vascular response to several drugs was investigated in vivo using intravenous bolus injection of the substances and calculating the MAP response (Table 1). The acute maximal pressor or depressor responses to vasoactive drugs (ΔMAP) were calculated by the differences between the maximal MAP (peak or trough) response and the averaged MAP $1 min before drug infusion (Todiras et al., 2017). Data analysis was carried out using the LabChart v5 software (AD Instruments). For quantifying the baroreflex control of HR, the ΔHR was calculated from doses of phenylephrine (1-10 μgÁkg À1 , i.v.) that triggered equi-pressor ($20 mmHg) MAP changes. The data is reported as the baroreflex sensitivity index calculated by the formula (BRS = ΔPI/ ΔMAP, msÁmmHg À1 ). HR values where previously converted to pulse interval (PI, ms; 60,000/HR),and ΔPI = (peak response PI À baseline PI).
All acute recordings of cardiovascular parameters and drug-induced pressure alterations were performed between ZT3 and ZT7.
Chronic BP recording with radiotelemetry: Telemetry probes (PhysioTel #PA-C10; DSI) were used to chronically record BP, HR and locomotor activity. The catheter of the pressure transmitter was inserted into the abdominal aorta via the femoral artery and the body of the transmitter was allocated in a subcutaneous pouch on the back of the mouse. Data acquisition started 12-14 days after surgery and data was recorded 24 h for 10 s every 5 min. Data were analysed with the software Dataquest ART.

| Cardiovascular response to chemical peripheral sympathetic ablation
Unspecific peripheral sympathectomy was used to assess the effect of sympathetic ablation on baseline cardiovascular parameters. After the establishment of the baseline cardiovascular parameters for 5 days mice received 6-hydroxydopamine hydrobromide (6-OHDA), a drug that destroys sympathetic terminals but does not penetrate the blood-brain barrier in adult rodents (Clark et al., 1972). 6-OHDA solutions were daily prepared in sterile physiological saline supplied with 0.01% ascorbic acid. Mice received five daily injections of 100 mgÁkg À1 6-OHDA i.p. Cardiovascular parameters were constantly recorded by radiotelemetry for 15 days (5 days before, 5 days during and 5 days after the 6-OHDA treatment).

| Renal denervation
Mice were previously instrumented with radiotelemetry probes as described above, but for this particular experiment the recording of baseline parameters was started 12-16 h after surgery. After baseline cardiovascular parameters recording for 48 h the animals underwent a total (afferent and efferent) renal denervation protocol. Renal denervation was performed exposing kidney and renal vessels by flank incisions (skin and muscle layers) under recovery anaesthesia. The kidney was gently retracted and all nerves visualized under a stereo T A B L E 1 List of vasoactive substances used in this study. Yohimbine hydrochloride Sigma-Aldrich Y3125 Note: All drugs were diluted in sterile saline except for candesartan and nifedipine for which sterile saline was supplied with Na 2 CO 3 (0,01%, w/v) and DMSO (5%, v/v), respectively. microscope were mechanically destroyed using fine surgical tools. To assure complete renal denervation, renal vessels were painted ($5 min) with a phenol solution (10% phenol in 95% ethanol). To prevent phenol spreading into the abdominal cavity parafilm was placed underneath the vessels and the excess of phenol was cleared with sterile physiological saline and cotton. Muscle and skin layers were closed separately with discontinuous sutures (silk 4/0) and the procedure repeated on the contralateral side. Animals recovered overnight ($12-16 h) in their home cage placed on a warm plate at 37 C. Postdenervation cardiovascular recordings were started immediately after recovery and lasted 1 week. The renal denervation success was evaluated by quantification of noradrenaline in each denervated kidney after 1 week as described below.

| Sample (organs, blood and urine) collection
Sampling was performed between ZT4 and ZT6 (4-6 after the onset of the light). The sample collection was done within these time points to match with the time that the acute cardiovascular phenotyping was undertaken.
Timed spontaneous urine was collected by holding the mouse over an empty tube. The procedure was repeated on the following 2 days in case additional urine was desired. After sampling, urine was snap frozen in dry ice and stored at À80 C until use.
Mice were killed by inhalation of an isoflurane overdose and blood was obtained using a 23G needle coupled to a 1-ml syringe from a cardiac puncture. EDTA-Plasma for western blot was prepared by collecting blood into MiniCollect ® EDTA-K3 (Greiner #450531) coated tubes and centrifugation at 2000g for 10 min at 4 C. Plasma used for biochemical quantification of metabolites was performed as described above but collecting the blood into lithium-heparin coated tubes (Greiner #450537). Plasma samples were stored at À80 C until usage.
Kidneys used for histological analyses were collected into a 15 ml tube filled with buffered 4% paraformaldehyde and kept at room temperature until processing. Kidneys and liver used for molecular biology or biochemical measurements were harvested, shortly rinsed in ice cold PBS buffer (pH 7.4), transferred to a sterile tube and snap frozen in dry-ice. Thoracic aorta and mesenteric arteries were collected into a tube containing RNAlater solution (Sigma #R0901) and stored at 4 C for a maximum of 7 days. Fat and connective tissue were cleared from aorta and second/third branch mesenteric arteries placed under a stereomicroscope with fine surgical tools.

| Gene expression analyses
RNA extraction and cDNA synthesis: The trizol/chloroform method was used to extract RNA from kidney, liver, aorta and mesenteric arteries. Tissues were homogenized in 1 ml of Trizol (Invitrogen #15596018) using a FastPrep homogenization system with dedicated tubes and ceramic beads (MPI #116004500). To assure complete genomic DNA clearance, 1-5 μg of RNA was incubated with DNase I (Sigma, #04716728001). A cDNA library was generated from 0.

| Western blot
Immuno-related procedures used comply with the recommendations made by the British Journal of Pharmacology (Alexander et al., 2018).
The absence of Agt protein in Agt-KO was verified using western blots with liver and plasma protein. Livers were homogenized in radioimmunoprecipitation assay (RIPA) buffer (100 mgÁml À1 , cell signalling #9806) supplemented with protease and phosphatase inhibitor cocktails using a benchtop homogenizer, FastPrep-24 (MP Biomedical #6004500). Samples were frozen and thawed three times, and centrifuged at 13.000 g for 10 min at 4 C. Finally, the supernatant was transferred to an empty tube. EDTA-plasma was diluted (1:10, v/v) in distilled water. Total proteins of liver extracts and diluted plasma were quantified using the bicinchoninic acid assay with a commercial kit (Sigma #BCA1-1KT). Samples were mixed with a 4X concentrated reducing buffer that is based on the Laemmli original formulation (Carl Roth #K929.1) and heated at 95 C for 5 min. Proteins were separated by molecular weight using 10% sodium dodecyl sulphate (SDS) polyacrylamide gels by electrophoresis. After electrophoresis proteins were transferred from the SDS gels to nitrocellulose membranes using Membranes were washed with PBS twice and scanned using an Odyssey infrared imaging system (LI-COR). The signals were visualized using the Image Studio Lite Software (LI-COR). The primary antibody procedure was repeated in the membrane containing liver protein extract by incubating the same membrane with a primary anti glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody T A B L E 2 List of primer pairs used for RT-qPCR. (1:2000, Cell Signaling #2118, RRID:AB_561053) as described above.

| Plasma angiotensin II quantification
The absence of Ang II (angiotensin II) was verified in plasma of Agt-KO using a radioimmunoassay assay as previously described (Schelling et al., 1980).

| Total nitrates and nitrites quantification
The overall nitric oxide (NO) production was estimated by quantifying the products of NO metabolism (nitrates and nitrites) in timed urine. Urine samples were diluted 1:40 (v/v) in ultrapure water, nitrate was enzymatically converted to nitrite, and nitrite levels were quantified in urine with a commercial colorimetric assay kit following manufacturer's instructions (Cayman #780001). Nitrates and nitrites concentration was calculated using a nitrite standard curve and the concentration normalized by the urinary creatinine concentration.

| Plasma and urine clinical biochemistry
Timed urine and lithium-heparin plasma were used for this purpose.
All measurements were performed by trained personnel at the animal phenotyping facility of the Max Delbrück Center for Molecular Medicine, Berlin, using an automated chemistry analyser (Beckman Coulter #AU480). The following parameters were measured in urine: sodium, potassium, albumin and creatinine, and plasma: sodium, potassium, chloride, albumin, total proteins, urea, creatinine, glucose.
Blood urea nitrogen (BUN) was calculated using the formula BUN = (Urea*0.467). The relative urine volume between the two lines was calculated using inverse proportion of the creatinine concentration in the urine.

| Data and statistical analyses
Experimental design as well as data and statistical analyses were performed in agreement with the guidelines of the British Journal of Pharmacology (Curtis et al., 2022). For all experiments randomization was applied and blinding whenever possible. The number of biological replicates is represented in the bar graphs by scatter plot whenever possible or indicated in the figure. To calculate the minimum number of replicates (equal number per group) to achieve statistical power, power analysis was applied using the software G Power 3. and unaltered HR (Figure 1c) confirming the findings of the acute recordings. Additionally, locomotor activity was equal between Agt-KO and controls (Figure 1c). Further telemetric measurements revealed that systolic pressure and pulse pressure were even increased in Agt-KO but diastolic pressure was not different from controls ( Figure 1c). Finally, heart weight and cardiomyocyte width, parameters that normally vary according to baseline BP levels, were found increased in Agt-KO ( Figure 1b). Additionally, heart sections containing the left ventricle were stained with picrosirius red for fibrosis quantification ( Figure S3a). Figure S3c shows that Agt-KO present increased interstitial left ventricle collagen accumulation.
Body weight as well as kidney, lungs, spleen and adrenal glands wet weights were not different among adult FVB/N wildtype and Agt-KO (Table S2).
For the purpose of better visualization, the data recorded by radiotelemetry was hourly averaged and plotted (Figure 1d). Agt-KO presented a circadian rhythmicity in all measured parameters similar to wildtype mice. A differential (day and night) calculation of the cardiovascular parameters is displayed in the Figure S2. Curiously, MAP of Agt-KO was increased in the dark phase (active phase) and this effect was not paralleled by HR or locomotor activity changes discarding an involvement of both parameters in the BP circadian rhythm (Figures 1d and S2). Additionally, Agt-KO presented increased systolic pressure and pulse pressure especially during the dark phase ( Figures   1c and S2b,d). The high nocturnal systolic pressure likely contributes to the increased nocturnal MAP levels found in Agt-KO ( Figure S2a).
Altogether, Agt-KO on the FVB/N background are normotensive during the day (resting phase) and even slightly hypertensive in the night (active phase).
To ascertain that Agt is not expressed in our backcrossed FVB/N Agt-KO line, RT-qPCRs and western blots were used. Quantification of Agt mRNA in liver samples of Agt-KO demonstrated absence of Agt expression ( Figure S1a,b). Confirming this finding plasma and liver of Agt-KO mice lack immunoreactivity for the $53 kDa Agt band present in controls ( Figure S1c,d). In addition, we demonstrated the presence of a functional response to Ang II in our FVB/N mice showing that the peptide triggered an increase in MAP similarly in wildtypes and Agt-KO ( Figure S1e). In agreement, a sub-chronic treatment with the AT 1 antagonist candesartan reduced MAP in control mice but not in Agt-KO ( Figure S1F). These results combined demonstrated functional AT 1 receptors as well as the absence of Ang II. We further verified the lack of Ang II in plasma of Agt-KO by radioimmunoassay and thereby also excluded the existence of any other precursor protein for angiotensins ( Figure S1g).

| Neurogenic pressor activity
We hypothesized that the sympathetic nervous system is hyperactive in Agt-KO to compensate for the lack of the RAS. To test this hypothesis, the mice used for cardiovascular phenotyping by radiotelemetry a stronger BP dependence on sympathetic nerve activity (SNA). Interestingly, MAP recovered to levels close to baseline even before the end of the 6-OHDA treatment in both controls and Agt-KO ( Figure   2a). After the cessation of the 6-OHDA treatment, BP and vascular sympathetic innervation are known to be restored in rats. Moreover, other mechanisms as augmented vascular adrenergic reactivity and adrenal catecholamine production as well as activation of the RAS were previously described in sympathectomized rats (Finch et al., 1973;Obayashi et al., 2000;Porlier et al., 1977;Vavřínová et al., 2019). 6-OHDA causes a long-term generalized peripheral sympathetic ablation. Therefore, the immediate MAP response to intrave-  Figure   3f). Finally, noradrenaline was quantified in mesenteric arteries and urine by ELISA and found to be unaltered in Agt-KO (Figure 3d).

| Endothelial function
Endothelial-derived nitric oxide (NO) is a major factor counteracting vasoconstriction. NO homeostasis was assessed in vivo using the changes in MAP triggered upon the administration of the NO synthase blocker L-NAME and the NO donor sodium nitroprusside.
Agt-KO presented a decreased MAP response to L-NAME ( Figure 4a) and increased response to sodium nitroprusside (Figure 4b)

| Renal damage and function
Mice with global deletion of the RAS components (Agt, renin and angiotensin converting enzyme) develop a typical renal morphology marked by atrophy of the renal papilla, hydronephrosis, fibrosis and concentric vascular hypertrophy of the renal arterial tree (Ding et al., 2001;Kihara, 1998). FVB/N Agt-KO also presented these common Renal renin expression is regulated by multiple control mechanisms such as BP, Ang II levels and SNA. Therefore, it was interesting to quantify renin gene expression by RT-qPCR. Strikingly, the normotensive FVB/N Agt-KO showed similar renin mRNA levels in comparison to controls (Figure 5d) despite the absence of Ang II ( Figure S1g) and likely increased renal SNA, as estimated by measurements of kidney noradrenaline (Figure 3d).
Plasma urea and creatinine were increased in Agt-KO suggesting a reduced glomerular filtration rate (Table 3). Agt-KO presented increased plasma sodium (Table 3) as well as total protein and albumin concentrations indicating dehydration (Table 3). Moreover, urine concentration ability was impaired in FVB/N Agt-KO as described in previous reports on other Agt-deficient mouse strains (Kihara, 1998;Okubo et al., 1998) ($3X higher volume, Table 3). As consequence of a high urinary output, sodium and potassium concentrations were strongly diluted in the urine of Agt-KO (Table 3). Despite the morphological alterations, the renal damage marker albumin normalized by the creatinine concentration was only marginally increased in Agt-KO (Table 3). Pax8 (Tong et al., 2009) was not different ( Figure S5e). Finally, the mRNA levels of the water channel aquaporin-2 (AQP2) were quantified by RT-qPCR in kidney but no differences were detected ( Figure   S5f).
To verify if renal afferent and/or efferent nerve activity influence the baseline BP in Agt-KO, a bilateral total renal denervation was performed and the changes in MAP monitored. MAP was immediately reduced in wildtype mice by $15 mmHg after renal denervation and this effect was sustained over a week (Figures 6a and S6a). Interestingly, Agt-KO presented only a slight and gradual MAP reduction after denervation peaking around 4 to 5 days after surgery (Figures 6a and   S6a). The HR increased the day after surgery in controls and Agt-KO, and gradually returned to baseline levels in controls, but it remained elevated ($40 bpm) in Agt-KO for over a week (Figures 6b and S6b).
Before renal denervation Agt-KOs were normotensive but their HR was reduced by $40 bpm ( Figure S6a,b). The increased HR after denervation is not preventing MAP to fall in Agt-KO because it was not different between the lines during the first 2 days after denervation, whereas MAP in controls already markedly dropped. Possible mechanisms behind the renal denervation-induced tachycardia in Agt-KO are altered sympatho-vagal balance or dehydration caused by reduced renal urine concentrating ability. The success of the renal denervation procedure was verified by measuring $90% reduced renal noradrenaline at the end of the in vivo recordings ( Figure S7).
Altogether, the renal denervation revealed that the renal nerves and the increased renal SNA are not involved in the unexpectedly high levels of baseline BP in FVB/N Agt-KO. Moreover, it showed that the BP lowering effect of renal denervation in wildtype mice relies on an intact RAS.

| DISCUSSION
We have backcrossed a previously described hypotensive Agt-KO line (Tanimoto et al., 1994) [Takahashi et al., 2005;Yanai et al., 2000] and angiotensin-converting enzyme [Cole et al., 2000;Esther et al., 1996;Krege et al., 1995]) as well as the AT 1a (Ito et al., 1995;Sugaya et al., 1995) or both AT 1a and AT 1b doubleknockout (Gembardt et al., 2008;Oliverio et al., 1998;Tsuchida et al., 1998) resulted in hypotension in mice. More recently, renin and AT 1a knockout rats were generated and reported as hypotensive (Exner et al., 2020;Moreno et al., 2011). The difference of baseline BP (systolic pressure or MAP) values between wildtypes and Agt-KOs varies from $15-50 mmHg across the different studies and the previously published FVB/N Agt-KO presented the smallest difference (14 mmHg). These studies used different methodologies and conditions to measure baseline BP (involving anaesthesia or not) and mice of different age and background strains, thus, differences are not surprising. We detected normotension in FVB/N Agt-KO employing two methodologies in freely-moving mice including radiotelemetry chronic recordings, which is the current gold standard to measure baseline BP in rodents (Carnevale et al., 2021;Kurtz et al., 2005). In contrast, the previously described slight hypotension in FVB/N Agt-KO was detected by tail-cuff plethysmography, which includes light anaesthesia and may therefore have moderated SNA (Lochard et al., 2003).
Interestingly, one study already observed an effect of anaesthesia on baseline BP in Agt-KO: MAP in Agt-KO on a mixed genetic background differed from wildtypes by $35 and $50 mmHg in recordings performed in freely-moving and anaesthetised mice, respectively (Table S1) (Tsuchida et al., 1998).
Renin release and expression are largely controlled by baseline BP levels (Ehmke et al., 1987;Holmer et al., 1994;Watanabe et al., 2021). Consistently, hypotension in Agt-KO from previous studies caused marked increases in kidney renin mRNA (Ishida et al., 1998;Nakamori et al., 2018;Tamura et al., 1998;Tanimoto et al., 1994). In contrast, the normotensive Agt-KO presented unaltered kidney renin mRNA again confirming that the Ang II feedback on renin expressing cells is not a determinant of renal renin expression (Matsusaka et al., 1996;Neubauer et al., 2018). Furthermore, most RAS deleted mice had lighter hearts due to life-long hypotension (Gembardt et al., 2008;Kang et al., 2002;Tamura et al., 1998). Normotensive Agt-KO has an increased cardiac mass, which may reflect the increased active phase MAP and systolic pressure.
The neural control of the vascular tone is an important player in baseline BP control. Increased neurogenic tone is commonly found in association with primary hypertension and several models of hypertension and the mechanisms associated to BP increase are direct vascular constriction and vascular remodelling (Aalkjaer et al., 2021;Asirvatham-Jeyaraj et al., 2021;DeLalio et al., 2020;Jama et al., 2022). The neurogenic pressor activity was not previously measured in Agt-KO using sympatholytic drugs, only plasma noradrenaline levels were measured but they were not altered (Sun et al., 2003). Another study found increased renal noradrenaline consistent to our findings in FVB/N Agt-KO (Nakamori et al., 2018). We found increased response to long-term and short-term sympatholytic drugs. However, acutely administered sympathomimetic agents exerted diminished responses. This phenotype was accompanied by decreased expression of the major α 1 and α 2 adrenoceptors in the vasculature. In vitro studies indicated that Ang II induces α-adrenoceptor expression in cultured rat smooth muscle cells (Hu et al., 1995;Lázaro-Suárez et al., 2011). Additionally, the reduced receptor expression might be a mechanism to mitigate increased SNA as observed during heart failure (Lohse et al., 2003). Taken together, the increased hypotensive response to generalized sympathectomy, different sympatholytic and unspecific vasorelaxants confirmed an exaggerated neurogenic tone in Agt-KO resulting in increased vasoconstriction powerfully preventing baseline BP to fall in the absence of Ang II.
On top of the increased neurogenic tone, the NO production was reduced in our Agt-KO line. Endothelial NO is essential to counteract diverse vasoconstrictor stimuli by increasing smooth-muscle cGMP (Farah et al., 2018). Therefore, the reduced NO in Agt-KO likely contributes to sustain the normotensive baseline BP. Increased Ang II has been previously associated with impaired vascular NO synthesis and increased BP (Laursen et al., 1997;Rabelo et al., 2016). In agreement with that a previous study with severely hypotensive Agt-KO demonstrated increased NO metabolites in urine and plasma contrary to our findings (Sun et al., 2003). Altogether, changes in the neurogenic tone and endothelial NO production that resemble human primary hypertension are maintaining normal BP in FVB/N Agt-KO.
The lack of Ang II/AT 1 axis during nephrogenesis leads to postnatal renal morphological alterations in rodents with genetic deletion of RAS components (Hilgers et al., 1997;Ishida et al., 1998;Pentz et al., 2004;Takahashi et al., 2005) or treated with RAS blockers after birth (Deluque et al., 2020;Madsen et al., 2010). At birth the renal morphology of Agt-KO is comparable to controls, but in the first weeks of life alterations including papillary atrophy, hydronephrosis and vascular concentric hypertrophy develop (Nagata et al., 1996;Niimura et al., 1995;Tsuchida et al., 1998). In addition, fibrosis and infiltration of monocytes were previously demonstrated in Agt-KO and renin knockout mice in agreement with our findings (Kang et al., 2002;Nakamori et al., 2018;Niimura et al., 1995;Takahashi et al., 2005). Our study demonstrated that these alterations are BP independent but solely dependent on the lack of Ang II in early postnatal life. Increased plasma markers of renal and muscular function, urea and creatinine, suggest reduced glomerular filtration rate consistent with previous data from renin knockout mice (Takahashi et al., 2005). The same study reported no albuminuria corresponding to the minimal albuminuria found in our study. Interestingly, the gene expression of major tubular sodium transporters NKCC2, NCC and ENaC was decreased in kidneys of FVB/N Agt-KO despite a greater urinary volume. Hydronephrosis is likely the cause because a similar phenomenon was described in ureter ligationinduced experimental hydronephrosis (Li et al., 2003;Zhang et al., 2015).
Renal sympathetic efferents increases BP by distinct mechanisms involving renin expression/release, renal vascular resistance control and salt and water reabsorption Osborn & Foss, 2017). Additionally renal pathologies with increased immune cell infiltration and intrarenal pressure may activate the renal afferents and invert the reno-renal-reflex causing neurogenic hypertension (Banek et al., 2019;Lopes et al., 2020;Osborn et al., 2021). However, these mechanisms are not involved in the FVB/N Agt-KO normotension because renal denervation minimally reduced its baseline BP. In wildtype mice, BP decreased strongly and immediately after denervation compared to the small, slow and gradual reduction in Agt-KO, which is probably caused by increased natriuresis (Osborn & Foss, 2017). The strong BP reduction in FVB/N wildtype mice is compatible with previous observations in denervated wildtype Sprague-Dawley rats that supports a role of the renal nerves in baseline BP maintenance. Selective afferent renal denervation has no effect on baseline BP in these rats, thus, renal efferents control baseline BP Foss et al., 2015;Jacob et al., 2003;Osborn & Foss, 2017). However, other studies denervated the same rat strain and found no baseline BP alterations, which is compatible to findings from other rat strains and mice Foss et al., 2018;Gueguen et al., 2019;Hohl et al., 2022;Moreira et al., 2020;Xiao et al., 2015). Further studies have to be performed to test if the species and/or background strain or rather technical issues like denervation efficacy are decisive. Because Agt-KO differ from wildtype FVB/N mice by the lack of the RAS precursor Agt, a possible mechanism behind the exclusive BP fall in wildtype mice after renal denervation is reduced renal renin production/release and consequently Ang II formation.
Altogether, we presented a completely unexpected BP phenotype in FVB/N mice with germline Agt deletion when compared to all previous studies with Agt-KO in the literature (Table S1). As compensatory mechanism for the lack of the RAS, FVB/N Agt-KO exhibit increased SNA to the vasculature accompanied by reduced NO production. The renal morphology and physiology are similar to what was previously described and the renal nerves are not contributing to the normotension in this line.

| LIMITATIONS AND FUTURE DIRECTIONS
Described below are limitations of this study that might be further addressed in follow up studies on the model. The data presented is essentially from young adult male mice, sex-and-age differences would be interesting to be evaluated, especially to further understand the BP regulation in FVB/N Agt-KO. Some conclusions drawn from this study rely on mRNA quantifications rather than direct protein quantifications, which might provide a more robust analysis. Urine volume was estimated by creatinine levels in timed spotted urine. The 24-h total urine collection would allow a more precise quantification of volume and metabolite/mineral output. Moreover, direct recordings of renal sympathetic nerves should be performed to provide a better evaluation of the actual nerve activity than noradrenaline measurements. The study shows reduced vascular eNOS expression with a vascular effect in Agt-KO. However, the mechanism downregulating eNOS is unknown and future studies may address whether also increased oxidative stress and eNOS uncoupling contributes to low NO generation in FVB/N Agt-KO. Finally, the FVB/N Agt-KO may have increased vascular stiffness because pulse pressure is increased.
Pulse wave velocity analyses should be performed to confirm if these mice indeed present increased vascular stiffness.

RIGOUR
This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research as stated in the BJP guidelines for Design & Analysis, Immunoblotting and Immunochemistry and Animal Experimentation, and as recommended by funding agencies, publishers and other organizations engaged with supporting research.

AUTHOR CONTRIBUTIONS
The study was designed by A. R., M. T. and M. B. Data collection was performed by A. R., M. T., F. Q. and N. A., whereas data analysis was performed by A. R. The study was supervised by M. B. A. R. and M. B.
drafted the manuscript. All authors reviewed and approved the final version of the manuscript.