Dependence of Renal Microvessel Density on Angiotensin II: Only in the Fetus?

During renal development, coordinate positioning and lengthening of the postglomerular microvasculature takes place as the tubular compartment expands and lengthens to form the renal papilla.¹ The three-dimensional relationship of peritubular capillaries and vasa rectae to their specific nephron segments^1,2 is essential for effective water and solute reabsorption in balance with glomerular filtration,³ for regulated salt excretion through the pressure natriuresis mechanism,^4,5 and for formation of hypertonic urine.^6,7 Conversely, rarefaction of the peritubular vasculature associated with tubulointerstitial fibrosis is a hallmark of chronic allograft nephropathy⁸ and probably most other progressive renal diseases.^9,10 Hence, there is an obvious need to understand mechanisms that control the formation and maintenance of the peritubular vasculature.

In this issue of JASN, Madsen et al.¹¹ describe the critical importance of signaling through angiotensin receptor 1 (ATR1) for development of a normal density and position of the postglomerular renal microvasculature. They administered the angiotensin II (AngII) receptor antagonist candesartan during the first 2 postnatal weeks in rat pups, a window that corresponds roughly to gestational weeks 24 through 36 for human fetal kidney development, and found a dramatic reduction in peritubular microvessel density. The reduced microvessel density accompanies a dramatic and very specific reduction of inner medulla size and an accumulation of interstitial myofibroblasts in the outer medulla. Consequently, the diffusion distance from a nephron to the nearest capillary is significantly greater in rat pups treated with candesartan than in controls. Moreover, the medullary fibrosis persists and renal blood flow continues markedly reduced 2 weeks after removal of the ATR1 inhibitor. In rodents, ATR1 is expressed as two isoforms, ATR1A and ATR1B, and both are blocked by candesartan. Previous work using ATR1A/ATR1B double-null mice had already shown that the inner medulla does not develop in the absence of ATR1 signaling.¹² Madsen et al.¹¹ now extend their findings to ATR1A/ATR1B-deficient mice and show that endothelial-specific transcripts Tie-2 and Flk-1, measures of endothelial cell density, are reduced in kidneys from double-null mice, as in the medulla of cadesartan-treated rat pups. Their findings that ATR1A and the vascular endothelial growth factor (VEGF) are prominently expressed in the thick ascending loop of Henle and collecting duct and that VEGF expression in these nephron segments is profoundly reduced during candesartan treatment lead them to conclude that coordinate development of the medullary vasculature during the period of collecting duct and loop of Henle lengthening is very likely due to specific activation of ATR1 in these epithelia, stimulating VEGF production, which in turn drives lengthening and proper positioning of peritubular microvessels and vasa rectae.

It is already well established that inhibition of angiotensin-converting enzyme during pregnancy results in renal dysplasia.^13–15 Furthermore, that deletion of various components of the renin-angiotensin system (RAS), including ATR1A/ATR1B,¹² ACE,¹⁶ and angiotensinogen but not ATR2,¹⁷ are associated with renal medullary dysplasia leaves little room for doubt that signaling through ATR1 is necessary for development of the renal medulla. Teratogenic effects of RAS inhibition during pregnancy¹⁸ were initially ascribed to inhibition of uterine prostaglandin synthesis and a consequent reduction in uterine blood flow¹⁹ or to significant maternal hypotension.¹² Of course, in the work of Madsen et al.,¹¹ maternal and uterine factors do not play a role.

The notion that ATR1 signaling regulates development of the renal vasculature through more direct, local effects has received scant attention thus far. Nonetheless, a marked reduction in the density of the peritubular vasculature, profoundly limited formation of vasa recta bundles, and interstitial fibrosis in a fetus of a mother who continued angiotensin receptor blocker therapy during pregnancy has been reported²⁰ and is entirely consistent with the observation by Madsen et al.¹¹ Also, Tufro-McReddie et al.²¹ previously reported abnormalities in afferent arteriolar and glomerular development in neonatal rat pups treated with ATR1 inhibitors.

It could still be argued that abnormal renal vascular development in the study by Madsen et al.¹¹ is accounted for by an indirect effect resulting from systemic inhibition or inactivation of ATR1. In fact, the BP in ATR1A/ATR1B-null mice is approximately 46 mmHg lower than in wild-type controls,¹² so renal hypoperfusion could potentially produce this effect. Madsen et al.¹¹ argue that hypoperfusion is not a likely mechanism leading to failure of renal microvascular development because other classes of antihypertensive agents are not associated with renal dysplasia.¹¹ There also is mounting evidence that AngII stimulates VEGF expression in a wide variety of tissues,^22,23 including proximal tubular epithelium,²⁴ all in keeping with the hypothesis that ATR1 blockade inhibits local ATR1-dependent VEGF synthesis. Still, proof that local angiotensin-stimulated VEGF synthesis regulates coordinate vasa recta development during expansion and growth of the renal medulla will require conditional deletion of ATR1A/ATR1B specifically in the ascending limb of Henle and/or collecting duct. Despite this qualification, this work adds substantial weight to the evidence that AngII, acting on ATR1, plays a critical role in the normal development of the renal vasculature.

It is also of note that quantification of renal microvessel density as a measure of progressive parenchymal damage is challenging, given that other compartments, for instance the interstitium and nephron mass, tend to change simultaneously. The stereologic technique used by Madsen et al.¹¹ is the most robust approach to establish the fractional volume and length of the renal microvasculature and sets a new bar for the evaluation of renal microvessel density.

Finally, we use inhibitors of the RAS with abandon for the treatment of hypertension and prevention of vascular dysfunction, among them progressive renal failure associated with proteinuria, and diabetic nephropathy in particular. The study by Madsen et al.¹¹ reemphasizes that utmost care must be exercised in the use of these agents in women of child-bearing age. Furthermore, mechanisms that stimulate renal vascular development in the fetus also reemerge in response to injury, for instance diabetic nephropathy²⁵ and acute kidney injury. Although angiotensin-driven VEGF synthesis may be detrimental in diabetic nephropathy,²⁵ this article raises the important question of whether RAS inhibition is potentially detrimental in other forms of renal injury repair.

Disclosures

None.