Maternal Supplementation With Citrulline Increases Renal Nitric Oxide in Young Spontaneously Hypertensive Rats and Has Long-Term Antihypertensive Effects

Methods

Renal NO Content at 2 Weeks

To determine tissue NO content, we modified a previously described protocol.¹³ Figure 1 represents a schematic cartoon of the protocol. We injected 2-week-old female SHRs, SHRcitrs, and WKY rats with a mixture of 37.5 mg/kg of FeSO₄·7H₂O and 190 mg/kg of sodium citrate SC and with 500 mg/kg of Na-DETC IP. The hydrophobic FE2⁺-(DETC)₂ complex associates with cell membranes and traps NO trafficking into or out of cells resulting in the formation of paramagnetic mononitrosyl-iron complexes with DETC. After 30 minutes, rats were anesthetized with sodium pentobarbital (60 mg/kg) and whole-body perfused with saline for 3 minutes. Kidney and heart samples were halved, placed in a plastic syringe with HEPES buffer (150 mmol/L; pH 7.4), snap-frozen in liquid nitrogen, and subjected to electron paramagnetic resonance (EPR) spectroscopy.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Cartoon of the protocol used to measure tissue NO content. Rats are injected with FeSO₄+Na-Citrate SC and Na-DETC IP. The hydrophobic FE2⁺-(DETC)₂ complex traps NO by forming paramagnetic NO-Fe-DETC complexes (MNIC). After 30 minutes, rats are anesthetized and whole-body saline perfused. Kidney and heart samples are snap frozen and subjected to EPR spectroscopy. After reduction with dithionite, the NO-Fe-DETC complexes are clearly visible.

EPR spectra were recorded at 77 K on a modified X-band ESP 300 radiospectrometer (Bruker BioSpin) operating near 9.54 GHz with 20 mW of power. Frozen samples were placed in a quartz liquid finger dewar at the center of a Burker ER4103TM cavity. Field modulation was 0.5 mT, gain 2.10⁵, time constant, and apparent diffusion coefficient conversion time 82 ms. Nine consecutive scans were accumulated to reduce instrumental noise. Spin densities were calibrated with frozen reference solutions of NO-Fe²⁺-(MGD)₂ in PBS.¹³ The signals from Cu²⁺-DETC complexes, formed when DETC ligands chelate endogenous Cu²⁺, completely dominate the NO-Fe-DETC complexes and were removed by a reduction with dithionite (100 mmol/L for 15 minutes). After this reduction, NO-Fe-DETC complexes (mononitrosyl-iron complexes) were clearly visible and could be quantified. NO assessed by this approach is equivalent to the mononitrosyl-iron complex yield in tissue after 30 minutes of in vivo trapping. Because some NO will be shunted into other metabolic pathways before being trapped, this method will underestimate the actual tissue NO content. Renal perfusion, essential for iron removal, is not possible in 2-day-old rats.

Results

Renal NO Content at 2 Weeks

In both the left and right kidneys, NO content was markedly higher in 2-week WKY rats than in age-matched SHRs (P<0.01 and P<0.001, respectively; Figure 2a). This renal NO deficiency in very young female SHRs seems to be organ specific, because in the heart, NO yields were similar (Figure 2b). Perinatal supplementation with citrulline in female SHRs resulted in significantly higher renal NO (P<0.05; Figure 2a) without an effect on NO yields in the heart (Figure 2b). Kidney weight was similar in SHRs and WKY rats (153±11 mg versus 150±8 mg). Perinatal citrulline had no effect on organ weights.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. a, NO yields determined by EPR in the left and right kidney in 2-week-old female SHRs (open circles and bars; n/litters= 10/5), SHRcitrs (closed circles and diagonally striped bars; n/litters=9/4), and WKY rats (triangles and horizontally striped bars; n/litters=10/5). b, NO yields in the heart determined by EPR in female SHRs (n/litters=7/2), SHRcitrs (n/litters=9/4), and WKY rats (n/litters=8/3). Both raw and average values are displayed. *P<0.05, †P<0.01, and ‡P<0.001 vs SHRs.

Long-Term Effects of Perinatal Citrulline in SHRs

In SHRcitrs, 24-hour citrulline intake was 416±11 and 281±21 mg/kg of body weight per day at 4 and 6 weeks of age, respectively. (Micro)nutrient intake during lactation could not be measured. Perinatal supplementation of citrulline increased body weight in both females and males up to 24 to 28 weeks of age (Figure S1). Systolic blood pressure (SBP) was reduced in female SHRcitrs at 8 weeks of age, ie, 2 weeks after cessation of supplementation, and this reduction in SBP persisted for a further 8 months (P<0.001; Figure 3). In male SHRcitrs, SBP was already reduced at 4 weeks, ie, during supplementation. Thereafter, this difference gradually diminished and was similar to control SHRs from 24 weeks onward.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3. SBP in control female (top left panel) and male (bottom left panel) SHRs (open circles; n=24 and 16, respectively) and SHRcitrs (closed circles; n=16 and 10, respectively) and WKY rats (triangles; n=9 and 10, respectively). WKY rat was lower than SHR from 8 weeks onward (P<0.001). MAP at 50 weeks of age in female (top right panel) and male (bottom right panel) SHRs (open bars), SHRcitrs (diagonally striped bars), and WKY rats (striped bars). *P<0.05 and ‡P<0.001 vs SHRs.

Terminal Arterial Pressure and Renal Function at 28 to 30 and 48 to 50 Weeks

Perinatal citrulline significantly reduced mean arterial pressure (MAP) at 28 to 30 weeks in both female and, unexpectedly, considering SBP, male SHRs (P<0.001 and P<0.05, respectively; Table 2). Renal hemodynamics was not affected by perinatal citrulline in female SHRs. Glomular filtration rate (GFR), effective renal plasma flow (ERPF), and renal blood flow (RBF) measured by flow probe were all significantly reduced in male SHRcitrs (P<0.05). Perinatal citrulline had no effect on filtration fraction (FF), para-aminohippurate (PAH) extraction, or fractional electrolyte excretion in male or female SHRs. Unexpectedly, irrespective of perinatal treatment, female SHRs displayed a much lower PAH extraction (0.60±0.03; n=24) than male SHRs (0.87±0.06; n=13; P<0.0001).

View this table:

• View inline
• View popup

Table 2. Renal Function in Male and Female SHRs and SHRcitrs at 28 to 30 Weeks

During infusion of the superoxide dismutase mimetic Tempol, MAP decreased significantly in all of the groups. However, in female SHRcitrs, this decrease was significantly less than in SHR controls (Figure 4). Although in male SHRs and SHRcitrs RBF increased during Tempol, time-control experiments in control SHRs indicated that this was an aspecific effect. Tempol significantly decreased renal vascular resistance in all of the groups except female SHRcitrs. During Tempol, GFR, ERPF, PAH extraction, FF, and fractional electrolyte excretions were not different between SHR, and SHRcitr (Table S2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4. MAP (left), RBF (middle), and renal vascular resistance (RVR) (right) before (baseline) and during Tempol in 28- to 30-week–old male and female SHRs (n=7 and 7, respectively) and SHRcitrs (n=4 and 5, respectively). *P<0.05, †P<0.01, and ‡P<0.001 vs SHRs; §P<0.05, ∥P<0.01, and ¶P<0.001 vs baseline.

At 48 to 50 weeks, female SHRcitrs had significantly lower MAP than SHR controls (P<0.05), but MAP was no different in males (Figure 3). As expected, in aged male SHRs, ERPF was lower and FF higher than in aged-matched WKY rats,¹⁴ and here we document that this is also the case in aged females (Table S3). These strain differences were not affected in either males or females by supplementing SHRs perinatally with citrulline.

Discussion

The aim of our study was to investigate whether a defect in the renal citrulline-arginine production pathway or a decrease in transporters involved in arginine reabsorption reduces renal NO preceding the development of hypertension in SHRs. Gene expression of ASS and a transporter responsible for arginine reabsorption were both decreased in young SHRs. This did not coincide with clear abnormalities in renal amino acid pools. Young SHRs displayed organ-specific renal NO deficiency, because the cardiac NO pool was normal. Furthermore, supporting renal NO by perinatal supplementation of citrulline persistently ameliorated hypertension in female SHRs. Administration of Tempol, a superoxide dismutase mimetic, did not reduce renal vascular resistance in adult female SHRs that had been perinatally exposed to citrulline, suggesting a persistent shift in the renal constrictor-dilator balance toward relaxation as compared with control female SHRs.

Although Jones⁸ observed a reduced arginine pool in plasma and skeletal muscle in SHRs, we did not find a difference in the SHR kidney compared with WKY rats. In addition, brief perinatal supplementation with citrulline, which increased renal NO, did not affect amino acid pools in the kidney. However, arginine concentrations were consistently increased in the heart tissue of SHRs perinatally supplemented with citrulline, suggesting that citrulline can indeed increase arginine levels and that there is clearly an organ-specific defect in the renal citrulline-arginine pathway. Therefore, it is likely that a defect in the tubular citrulline-arginine synthesis pathway and/or arginine reabsorption has a compartmentalized effect on arginine pools in the kidneys of 2-week-old SHRs.¹⁵ A defect in the citrulline-arginine synthesis pathway could impair the synthesis of NO, despite unrestricted or even excess amounts of NOS, as observed previously in 3-week–old SHRs.⁵ Indeed, at 2 days and 2 weeks of age, we observed decreased renal gene expression of ASS, and ASL was decreased at 2 weeks. Although not directly assessed, this suggests an impaired ability to convert citrulline via argininosuccinate to arginine, thereby decreasing arginine synthesis. How citrulline increases renal NO in young SHRs requires further investigation. Both ASS and ASL gene expression were decreased but not absent. Citrulline delivery to the proximal tubule cells is rate limiting for renal arginine synthesis.^6,7 Therefore, we speculate that increasing delivery of citrulline to the juvenile SHR kidney could result in some increase in arginine synthesis in specific renal compartments resulting in a net increase of renal NO.

In the absence of enhanced citrulline delivery, the kidney may compensate for such arginine shortage by increasing arginine reabsorption. Thus, most likely, renal NO deficiency because of substrate shortage would only occur when both renal arginine synthesis and reabsorption are impaired. Indeed, expression of the cationic amino acid Y-transporter was also reduced at both 2 days and 2 weeks in SHRs. This transporter is basolateral in renal and intestinal epithelial cells and exchanges neutral for cationic amino acids.⁹ Because arginine is a cationic amino acid, Y-transporters may regulate the rate of renal NO synthesis by controlling the uptake of arginine. Welch et al¹⁶ observed an increase in renal neuronal NOS expression in adult SHRs versus WKY rats, and we found that this increase was already present at 2 weeks, thus clearly preceding the onset of hypertension. However, we did not observe any difference in the renal expression of endothelial NOS. Taken together, substrate deficiency because of reduced synthesis and impaired delivery to NOS could very well explain the reduction in renal NO content.

To determine renal NO content in renal tissue, we trapped NO in paramagnetic NO-Fe-DETC complexes, which were quantified by EPR spectroscopy. This new and direct determination of organ NO content in young rats revealed a markedly reduced NO yield in the kidney, but not in the heart, in SHRs compared with WKY rats. Cosentino et al¹⁷ could increase aortic NO production in vitro with tetrahydrobiopterin in 4-week–old prehypertensive SHRs, and they ascribed the enhanced vascular superoxide production, measured in the aorta, to uncoupled endothelial NOS. The present study clearly shows that, already in 2-week-old prehypertensive SHRs, renal NO production is markedly decreased. Interestingly, this deficiency was not found in the heart, suggesting that uncoupling of endothelial NOS in the vascular bed in prehypertensive SHRs¹⁷ is already a secondary effect. Although administration of tetrahydrobiopterin in young SHRs has antihypertensive effects,¹⁸ perinatal tetrahydrobiopterin or inducible NOS inhibition in SHRs had no persistent reproducible effects on blood pressure (M.P. Koeners, unpublished data, 2006). Outspoken defects in gene expression relating to renal arginine availability were found at 2 days and 2 weeks, and decreased renal NO was documented at 2 weeks. Technical limitations preclude the measurement of renal amino acids and NO in 2d old SHR.

Following up on our observations on defects in the citrulline-arginine pathway and arginine reabsorption in 2-day and 2-week-old SHRs, we postulated that perinatal supplementation of citrulline would correct renal NO deficiency, because intestinal citrulline supply determines renal arginine production in a master-slave relation, and citrulline is a neutral amino acid the reabsorption of which does not involve the y⁺L transporter.⁷ Indeed, this is exactly what we found, suggesting that dietary citrulline may be a novel way of supporting renal arginine availability, especially when arginine transport is disturbed. In 1991, Chen and Sanders¹⁹ showed that in 3-week-old Dahl/Rapp rats, daily IP l-citrulline injections over 5 days could prevent salt-sensitive hypertension as effectively as daily IP l-arginine injections. The present study shows that, in SHRs, this concept is also applicable to the development of hypertension in a much earlier phase, namely, during nephrogenesis. Supporting NO availability in the developing SHR kidney appears to be able to permanently reprogram blood pressure regulation in adult life, leading to a lower blood pressure. Intra-arterial MAP measured at the end of the protocol generally reproduced differences between groups found by external SBP measurements. These findings suggest that maternal amino acid supplements during pregnancy could be a feasible approach to prevent or ameliorate the development of hypertension in genetically or environmentally predisposed children.

Previously, we treated SHRs perinatally up to 4 or 8 weeks with a mixture of l-arginine, taurine, vitamin C, and vitamin E and observed a similar effect on blood pressure regulation in the offspring.¹¹ Perinatal vitamin C and vitamin E alone had no persistent antihypertensive effects.²⁰ Our new data suggest that, because reduced expression of the renal cationic amino acid transporter could have limited the effect of l-arginine supplements in our previous study, antioxidative effects of taurine could have contributed substantially to the persistent effects.¹¹ Offspring of rat dams exposed to food restriction during pregnancy develop mild hypertension, and supplementation of l-arginine after weaning can reduce this blood pressure increase²¹ and eventually even protect the kidneys.²² Supplementing a low-protein diet during pregnancy with glycine can also reverse the development of hypertension in the offspring,²³ and this has been postulated to be because of correction of reduced vascular endothelial NOS activity.²⁴ However, perinatal supplementation of l-arginine, let alone l-citrulline, with the aim of protecting the offspring in experimental models of perturbed pregnancy, has to our knowledge not yet been tested.

Hypertension is characterized by a self-perpetuating cycle of NO deficiency and increased reactive oxygen species and a disturbed renal constrictor-dilator balance.¹ To investigate whether the perinatal citrulline supplement had a persistent effect on renal vasoreactivity, we scavenged superoxide with a superoxide dismutase mimetic (Tempol) during acute experiments. Tempol had similar effects on all of the groups with the notable exception of female SHRcitrs. Female SHRcitrs displayed a smaller decrease in MAP and practically no decrease in renal vascular resistance during Tempol, suggesting a persistent decrease in renal superoxide-dependent vasoconstriction. This is compatible with the idea that, in SHRcitrs, a persistent shift in the renal vasoconstrictor-vasodilator balance toward relaxation underlies the antihypertensive effect of perinatal citrulline. Note that a response to Tempol does not prove that there is a specific defect in superoxide dismutase activity. Tempol and other superoxide dismutase mimetics have been shown to correct enhanced peripheral vasoconstriction in models of perinatal malnutrition,^25,26 and we have observed that perinatal Tempol can also persistently reduce SBP in SHRs.¹²

Judging by the higher body weight, it is possible that perinatal citrulline has long-term stimulatory effects on muscle protein synthesis. Such an effect has been observed previously during refeeding in old malnourished rats.²⁷ However, because monitoring growth was not a primary objective, we did not measure terminal organ weights, with the exception of the kidney and heart (ventricles). Thus, we cannot provide an explanation for this effect on body growth. Note that the increase was no longer present in the aged SHRcitrs.

In multiple studies of developmental plasticity in different species, including humans, gender differences have been observed.²⁸ Although not a primary focus of our study, we also observed a marked gender difference in the response to perinatal citrulline supplementation in SHRs, with a more consistent and pronounced effect in females than in males. Because we did not measure renal gene expression and NO content in young male SHRs, we cannot speculate on the basis for this difference.

In conclusion, there is clearly reduced expression of both the renal citrulline-arginine synthesis pathway, as well as the arginine reabsorption pathway, during nephrogenesis in the female SHRs. We propose that reduced cationic amino acid transporter disables the kidney to use arginine reabsorption to compensate for reduced arginine synthesis. It seems plausible that this results in renal NO deficiency that precedes the development of hypertension in SHRs. This NO deficiency and its adverse sequels can be persistently corrected with perinatal citrulline supplementation.