Review: Evidence for a functional intracellular angiotensin system in the proximal tubule of the kidney

Abstract

ANG II is the most potent and important member of the classical renin-angiotensin system (RAS). ANG II, once considered to be an endocrine hormone, is now increasingly recognized to also play novel and important paracrine (cell-to-cell) and intracrine (intracellular) roles in cardiovascular and renal physiology and blood pressure regulation. Although an intracrine role of ANG II remains an issue of continuous debates and requires further confirmation, a great deal of research has recently been devoted to uncover the novel actions and elucidate underlying signaling mechanisms of the so-called intracellular ANG II in cardiovascular, neural, and renal systems. The purpose of this article is to provide a comprehensive review of the intracellular actions of ANG II, either administered directly into the cells or expressed as an intracellularly functional fusion protein, and its effects throughout a variety of target tissues susceptible to the impacts of an overactive ANG II, with a particular focus on the proximal tubules of the kidney. While continuously reaffirming the roles of extracellular or circulating ANG II in the proximal tubules, our review will focus on recent evidence obtained for the novel biological roles of intracellular ANG II in cultured proximal tubule cells in vitro and the potential physiological roles of intracellular ANG II in the regulation of proximal tubular reabsorption and blood pressure in rats and mice. It is our hope that the new knowledge on the roles of intracellular ANG II in proximal tubules will serve as a catalyst to stimulate further studies and debates in the field and to help us better understand how extracellular and intracellular ANG II acts independently or interacts with each other, to regulate proximal tubular transport and blood pressure in both physiological and diseased states.

our current understanding of the renin-angiotensin system (RAS) has evolved a great deal since Tigerstedt and Bergman's discovery of renin as a “pressure-elevating substance” from the kidney (9, 37, 69, 76, 115). Although the endocrine role of the RAS is undisputable, there is now increasing evidence that the key member of the RAS, ANG II, may function as a paracrine (cell-to-cell) and intracrine (within the same cell or intracellular) peptide. Indeed, ANG II may be generated in a particular cell type and released extracellularly to exert an endocrine effect in a different cell in a remote tissue via activation of its cell surface G protein-coupled receptors (GPCRs). Alternatively, newly formed ANG II may directly interact with its cytoplasmic and nuclear GPCRs to exert intracellular effects within the same cell (22, 33, 43, 75, 117, 121). A potential role of intracellular ANG II was initially implicated in an early study in the 1970s, in which radiolabeled ANG II was localized in the nuclei of vascular smooth muscle cells (VSMCs) and cardiomyocytes after it was systemically infused into the rats (128). It is Re who has since the 1980s (126) persistently advocated the concept of intracrine ANG II as an effector peptide that is either synthesized within a cell via actions of renin, angiotensinogen, and ANG I-converting enzyme (ACE), or internalized from extracellular ANG II via receptor-mediated endocytosis (121, 124). This effort has led to an explosion of research activities to uncover a functional role of intracellular (or intracrine) ANG II in different tissues or cells in last two decades.

In the 1980s, Re and associates reported that intracellular ANG II induced important nuclear effects in isolated rat hepatocytes (19, 21, 126). In the 1990s, three groups of independent investigators confirmed that rat hepatic nuclei contained the receptor binding sites specific for ANG II (7, 38, 39, 151), and ANG II induced positive renin and angiotensinogen transcription and growth-related effects (38, 39). In the heart, De Mello's group (30–32) performed intracellular dialysis of renin, angiotensinogen, or ANG II directly into hamster cardiomyocytes and demonstrated significantly altered cell-to-cell communications, suggesting an important role for intracellular ANG II to regulate cardiac function. In VSMCs, Haller et al. (54, 56) elegantly microinjected ANG II directly into the cells and demonstrated significant intracellular and nuclear calcium responses, which were blocked by intracellularly administered AT₁ receptor antagonists. In the brain, Sigmund and colleagues (75) recently provided evidence that an intracellular form of renin plays a functional role in the brain. Recent studies from Danser's and Mullins' groups reported that prorenin may be taken up by target cells to induce intracellular signaling or responses that may be independent of ANG II (26, 134). In the renal cortical nuclei of sheep and congenic mRen2 Lewis rats, Crowley et al. (23) and Chappell's team (50, 52, 53) characterized a large number of high-density specific receptors for ANG II and ANG (1–7), activation of which were associated with superoxide and/or nitric oxide production. Cook's team has elegantly designed an intracellular cyan fluorescent fusion of ANG II for its expression and actions only occur intracellularly (19–21). More recently, Redding et al. (127) have reported that transgenic mice expressing this intracellular fluorescent fusion of ANG II induced renal thrombotic microangiopathy and elevated blood pressure in mice. The presence and the potential role(s) of intracrine or intracellular ANG II in tissues other than the kidney have been comprehensively reviewed elsewhere (33, 43, 67, 121), as well as in the current issue of American Journal of Physiology Regulatory, Integrative and Comparative Physiology (5, 17, 51).

Accordingly, the scope of this article will focus on the intracrine and/or intracellular RAS in the kidney with a special emphasis on the proximal tubules. This is primarily because little is known about the presence and biological and physiological roles of intracellular ANG II in the kidney, let alone proximal tubules. The proximal tubules of the kidney reabsorb more than 65% to 70% of the filtered sodium and fluid loads, and increased sodium and fluid reabsorption by a few percentage points in this nephron segment by extracellular and intracellular ANG II may promote sodium and fluid retention and consequently lead to increases in arterial blood pressure. Currently, there is a consensus that extracellular ANG II induces powerful effects on proximal tubule function and tubule-interstitial injury by activating cell surface type 1 (AT₁) receptors (35, 58, 99, 132, 171). However, it is unlikely that an endocrine or a paracrine role may fully explain the long-term hypertensive and genomic effects of ANG II, since repeated exposure or stimulation of the cell surface GPCRs by ANG II is known to induce desensitization of the responses through the so-called receptor-mediated endocytosis of the ANG II/receptor complex (40, 45, 154). Here, we review the evidence for a functional intracellular angiotensin system, specifically in proximal tubules of the kidney. We hope that this brief review may serve as a catalyst to stimulate further studies and constructive debates in this field, and ultimately improve our understanding of how extracellular and intracellular ANG II acts independently or through interactions to regulate proximal tubular sodium and fluid reabsorption and blood pressure in both physiological and diseased states. This information may also be useful for devising novel strategies to prevent or treat ANG II-dependent hypertension and cardiovascular and renal diseases by blocking the effects of both extracellular and intracellular/nuclear ANG II.

Overview of the Classic Role of Endocrine and Paracrine ANG II in the Kidney

ANG II is a powerful vasoactive peptide that is implicated as the major bioactive component of the RAS (11, 41, 72, 73, 102, 181, 182). The classic pathway for its production involves kidney-derived renin, which is the rate-limiting enzyme acting on angiotensinogen synthesized primarily in the liver, to form ANG I. ANG I is converted by ACE to the principle peptide ANG II. ANG II can be further degraded by aminopeptidases into biologically active fragments, including ANG III (des-aspartyl-ANG II), ANG (1–7), and ANG IV [ANG (3–8)] in the circulation and target tissues (11, 41, 72, 73, 102, 181, 182). The endocrine (or circulating) and paracrine (or intrarenal) roles of ANG II in the physiological control of renal hemodynamics and tubular sodium transport have been extensively investigated since the 1970s and 1980s (58, 59, 70, 78, 79, 99, 105, 118). It is now well understood that circulating and intrarenal ANG II activates its cell surface GPCRs in targeted cells, resulting in phosphorylation of the receptor and initiation of downstream signaling responses (28, 47, 63, 93, 154, 171, 181, 182). As in other target tissues, ANG II receptors exist as two major classes in most species in the kidney, designated AT₁ and AT₂, although the rodent AT₁ exists as two further subtypes, AT_1a and AT_1b (10, 28, 66, 145, 171). Activation of AT₁ is responsible for the majority of ANG II's well-recognized actions both in vitro and in vivo, such as renal vasoconstriction or hemodynamic responses and sodium and fluid reabsorption, with AT₂ activation appearing to counter many of those effects induced by AT₁ activation (28, 47, 63, 93, 154, 171, 181, 182). The development of ANG II receptor subtype-selective antagonists, such as losartan and candesartan on AT₁ or PD123319 on AT₂, and generation of transgenic mice with global or cell-specific deletion of a selective receptor have helped demonstrate the importance of ANG II and its cell surface GPCRs in the regulation of blood pressure, cell growth and proliferation, renal hemodynamics, sodium and fluid reabsorption, and cardiovascular homeostasis (28, 47, 63, 93, 154, 155, 171, 181, 182). The endocrine and/or paracrine roles of ANG II in the kidney have been comprehensively reviewed elsewhere (11, 28, 72, 77, 103, 145).

Overview of the Local Renin-Angiotensin System in Proximal Tubules of the Kidney

The proximal tubule is one of major target tissues of the RAS in the kidney, because the proximal tubule function is physiologically and pathophysiologically regulated by both circulating and local paracrine ANG II. All key components of the RAS, including angiotensinogen, renin, ACE, and AT₁ and AT₂ receptors, have been demonstrated in proximal tubules of the kidney (11, 61, 94, 101, 150, 173, 182). The presence of renin, angiotensinogen, and ACE in proximal tubules will ensure the onsite production of ANG II independent of the circulating RAS, whereas the expression of AT₁ and AT₂ receptors is critical for ANG II to induce biological and physiological effects. Up to four major classes of ANG receptors have been found to express in proximal tubules of the kidney, including AT₁ and AT₂ for ANG II (12, 50, 109, 147, 171, 173–175) and ANG III (des-aspartyl-ANG II) (60, 112, 113), the Mas receptors for ANG (1–7) (53, 133), and AT₄ (or IRAP, insulin-regulated aminopeptidase) for ANG IV (Fig. 1) (3, 57, 80). High-affinity AT₁ receptors are localized in both brush border and basolateral membranes (34, 36, 170, 173). AT₁ receptors are GPCRs belonging to the superfamily of seven transmembrane-spanning protein receptors (28, 93, 98, 171). The AT₁ receptor is further classified into two subtypes, AT_1a and AT_1b, in rodents, but the former is the predominant isoform in the kidney, equivalent to the human AT₁ receptor (28). The classical actions of extracellular or exogenously administered ANG II in proximal tubules of the kidney are to stimulate sodium, bicarbonate, and fluid reabsorption physiologically (15, 58, 99, 176), and to induce sodium retention and tubulo-interstitial injury or fibrosis in diseased states (68, 91, 130, 163). AT₁ (AT_1a) receptors appear to mediate the majority of ANG II-induced biological and physiological effects in proximal tubules of the kidney (23, 24, 49), which is coupled to GPCR-mediated multiple signaling pathways, including PLC, PLD, and PLA₂, MAPK, and tyrosine kinase (35, 93, 156, 173). AT₁ receptor activation by ANG II leads to the hydrolysis of phosphoinositide, mobilization of intracellular calcium, and inhibition of adenylyl cyclase (81, 137, 153, 183).

Fig. 1.

Quantitative in vitro autoradiographic mapping of active renin binding in juxtaglomerular apparatus (JGA) in the dog kidney pretreated with sodium depletion using the radiolabeled renin inhibitor, ¹²⁵I-labeled H77 (A), angiotensin I-converting enzyme (ACE) binding in proximal tubules of the rat kidney using ¹²⁵I-labeled 351A (B), AT₁ receptor in the presence of the AT₂ receptor blocker PD123319 to displace AT₂ receptors (C) or AT₂ receptor binding in the presence of the AT₁ receptor blocker losartan to displace AT₁ receptors using ¹²⁵I-labeled [Sar¹, Ile⁸]-ANG II (D), ANG (1–7) receptor binding in the rat kidney using ¹²⁵I-labeled ANG (1–7) as the radioligand (E), and ANG IV receptor binding in the rat kidney using ¹²⁵I-labeled ANG (3–8) (F). The levels of binding are indicated by color calibration bars with red representing the highest, whereas blue showing the lowest levels of enzyme or receptor binding. G, glomerulus; IM, inner medulla; IS, inner stripe of the outer medulla; JGA, juxtaglomerular apparatus; PT, proximal tubule. Reprinted from Peptides, vol. 32, by Zhuo JL and Li XC, New insights and perspectives on intrarenal renin-angiotensin system: focus on intracrine/intracellular angiotensin II, pages 1551–1565, 2011, with permission from Elsevier (182).

AT₂ receptors are also expressed in proximal tubules of the kidney. Although the expression of AT₂ receptor mRNA is low in proximal tubules of the adult rat and mouse kidneys, AT₂ receptor proteins are readily detectable by Western blot analysis (12, 109, 145). In adult human kidney, high levels of the AT₂ receptor are localized in the perivascular or adventitial layers of intrarenal arteries (44, 48, 172). The specific roles and signaling transduction mechanisms of AT₂ receptors in proximal tubules of the kidneys are not as well understood as those of AT₁ receptors. However, Carey and colleagues (110–113, 144, 145) have consistently demonstrated that ANG II and ANG III activate AT₂ receptors to induce significant diuresis and natriuresis via tubule-dependent mechanisms, involving activation of the nitric oxide (NO)/cGMP-dependent signaling pathway. In contrast to AT₁ and AT₂ receptors, the roles and signaling mechanisms of the Mas receptors for ANG (1–7) and the AT₄ receptor for ANG IV in proximal tubules of the kidneys are not well understood (13, 41, 57, 80).

In contrast to the circulating RAS, which is inversely regulated by ANG II and blood pressure, the local RAS in proximal tubules appears to be differentially regulated by the agonist in ANG II-induced hypertension. Major components of the RAS, including angiotensinogen (71, 72, 103), ACE (61), ANG II (85, 90, 180, 184), and AT₁ receptors (61, 90, 180) were significantly increased, rather than decreased, in ANG II-induced hypertension. In spontaneously hypertensive rats, ANG II also increased AT₁ receptors in proximal tubule cells (166). This upregulation of the local proximal tubule RAS by ANG II may play an important role in the pathogenesis of ANG II-dependent hypertension.

Potential Roles for an Intracrine/Intracellular RAS in the Blood Pressure and Renal Regulation

Whether intracrine/intracellular ANG II has an important intracellular role and, if it does, what are the signaling mechanisms are incompletely understood (33, 73, 74, 121, 123, 178, 181, 182). The concept of intracrine/intracellular ANG II has not been well accepted for a number of reasons. Currently, the general view of ANG II-induced biological and physiological effects appears to be that after synthesis, ANG II may have to be released into the extracellular fluid compartments to induce any biological and physiological effects. The second general view is that ANG II receptors, after their expression, may have to appear on the cell surface or plasma membranes to mediate the effects induced by ANG II. The third general view is that cell surface AT₁ (AT_1a) receptors may be continuously desensitized after they are continuously exposed to extracellular ANG II and that internalized ANG II may have to be transported to lysosomes for degradation (40, 92, 93, 156, 167). According to the latter view, internalized ANG II and its receptors are unlikely to be trafficking to other intracellular organelles or the nucleus. However, there is strong evidence suggesting that binding of ANG II to its cell surface AT₁ receptors also induces internalization of both the effector (ANG II) and the receptor into intracellular compartments, including the nucleus. Indeed, at least in the proximal tubules of the kidney, internalized ANG II, along with intracellularly synthesized ANG II, may serve as a major source of intracellular ANG II, which may interact with its cytoplasmic and nuclear AT₁ or AT₂ receptors to induce important intracellular signaling with consequent biological effects (82–88, 179, 183). Here, we review the in vitro and physiological evidence for a biological role of intracellular ANG II in cultured proximal tubule cells and proximal tubules of the kidney, with focuses on the mechanisms of AT₁ (AT_1a) receptor-mediated ANG II uptake (or endocytosis) and its contribution to intracellular ANG II levels, intracellular trafficking pathways, and the potential role of internalized or intracellular ANG II in the regulation of the sodium and hydrogen exchanger 3 (NHE3), proximal tubule reabsorption, and blood pressure.

Evidence that ANG II May Be Intracellularly Formed in Proximal Tubules

The concept that ANG II may be synthesized or formed in proximal tubule cells of the kidney is not quite controversial, as all necessary substrate and enzymes of the entire RAS that are required for ANG II formation are expressed locally (11, 61, 94, 101, 150, 173, 182). Alternatively, angiotensinogen and renin may also be taken up by proximal tubule cells for intracellular ANG II production. In proximal tubule cells, paracrine and intracellular ANG II may be generated by synthesis within the cell with or without secretion into the luminal fluid or interstitial fluid compartments (8, 61, 64, 65, 100, 101, 106, 139, 146, 150). It has been widely cited in the literature that ANG II concentrations may be in the ranges of 20–40 nM (or μg/ml) in intraproximal tubular and peritubular or interstitial fluid compartments, compared with femtomoles of ANG II (or pg/ml) present in the normal plasma samples (8, 139, 146). Although the reported nanomolar levels of ANG II in proximal tubule fluid and cortical interstitial fluid remains highly debatable (158, 159), there is evidence that ANG II appears to be capable of intracellular synthesis under particular conditions in a variety of cells. For instance, high glucose and isoproterenol have been found to stimulate production of intracellular ANG II in cardiac myocytes, renal mesangial cells, and VSMCs (141–143). By using receptor blockers, such as candesartan to inhibit extracellular binding and potential activation of the receptor-mediated endocytic pathway, it has been demonstrated that ANG II can be synthesized intracellularly and, depending on the cell types or conditions, may use an alternate pathway for synthesis from the well-described ACE-mediated pathway (114). For instance, in diabetes, hyperglycemia appears to promote a renin and chymase-dependent mechanism, which is not affected by ACE inhibitors, while isoproterenol stimulation of ANG II synthesis can be blocked with benazepril, an ACE-inhibitor (4, 114, 142, 143). The downstream fate of intracellularly synthesized ANG II is further dependent on the particular stimulus, for reasons not yet fully understood; ANG II synthesized after glucose stimulation tends to be retained within the cell, while isoproterenol-stimulated ANG II is secreted. A variety of components of the RAS, including renin, angiotensinogen, and ANG II, have been found to be dramatically increased intracellularly in high glucose or diabetic conditions in vitro. In vivo studies have further confirmed this intracellular activation within the heart and kidney or proximal tubule cells, with unaffected extracellular or plasma levels (4, 114, 142, 143, 168, 169). Clearly, there is a therapeutic potential for targeting long-term clinical implications of hyperglycemia-induced intracellular ANG II production in such pathologies as diabetic cardiomyopathy and diabetic nephropathy.

In cultured rabbit and mouse proximal tubule cells, we also demonstrated that intracellular ANG II levels are much higher in proximal tubule cell samples than in the medium under basal and ANG II-stimulated conditions, thereby supporting an independent, local RAS (82, 84, 86, 179). We found that high levels of ANG II in proximal tubule cells could be reduced by pretreatments of the cells with an ACE inhibitor or an AT₁ receptor antagonist if the cells were incubated with ANG II. Likewise, intracellular ANG II levels in isolated proximal tubules were much higher than in plasma in the rat and C57BL/6J mice under basal and ANG II-infused conditions, which also appear to be ACE- and/or AT₁ receptor-dependent (85, 87, 90, 107). Our studies suggest that glucose or isoproterenol, however, are not the only means for induction of intracellular synthesis of ANG II. In an interesting study, van Kats et al. (159) investigated whether ANG II generation occurs intracellularly or extracellularly by studying the subcellular localization of ¹²⁵I-labeled ANG I or ¹²⁵I-labeled ANG II in the pig kidney (159). These investigators found high levels ¹²⁵I-labeled ANG I or ¹²⁵I-labeled ANG II in all intracellular organelles with the same distribution profiles and came to a different conclusion that ANG II formation in the kidney occurs predominantly extracellularly, which is followed by rapid AT₁ receptor-mediated endocytosis, thereby leading to high intracellular ANG II levels (157, 158). Furthermore, studies in isolated sheep nuclei from the kidney cortex suggest that ANG II be synthesized in the nucleus itself, as immunoreactive components of the entire RAS—angiotensinogen, renin, and ACE—were identified in the absence of the cytoplasm or extranuclear organelles (50).

Evidence that AT₁ (AT_1a) Receptor-Mediated Uptake of Circulating and/or Extracellular ANG II in Proximal Tubule Cells

The early evidence that the kidney takes up the circulating ANG II was provided by Navar and colleagues, who infused exogenous ANG II to raise the plasma ANG II level to that seen in two-kidney, one-clip (2K1C) renal hypertensive rats (160). These authors showed that while renin content was markedly suppressed by 80%, intrarenal ANG II content of the contralateral kidneys of the 2K1C and the ANG II-infused rats was greater than that of 2K1C control or uninephrectomized control animals. These results were subsequently confirmed by several other groups (140, 157, 158, 184, 185). These studies suggest that with high affinity and high density of AT₁ receptors localized in apical and basolateral membranes of the proximal tubule cells, intraluminal and interstitial ANG II may be continuously taken up by proximal tubule cells in vitro and in vivo to serve as an important source of intracellular ANG II. However, the site(s) or cell type(s) responsible for intrarenal accumulation of ANG II is less well understood with the whole-kidney approach.

We first sought to determine whether the systemically infused ANG II is taken up into renal cortical endosomes during ANG II-induced hypertension, and if so, whether the AT₁ receptor antagonist, candesartan, prevents this accumulation (180). We infused [Val⁵]-ANG II continuously for 2 wk rather than for hours via an osmotic minipump and isolated and purified heavy (intermicrovillar clefts) or light endosomes to determine endosomal ANG II levels. The levels of ANG II in light endosomes were almost 10-fold higher in ANG II-infused rats, where ANG II was colocalized with fluorescein-labeled AT_1a receptor antibody binding and an endosomal marker (Fig. 2) (180). This finding provides strong evidence that the circulating and/or paracrine ANG II is taken up by the superficial cortex of the kidney, probably proximal tubules. The endocytosis of circulating and paracrine ANG II may also be taken up by proximal tubules via the scavenger receptor megalin-mediated mechanism (46, 119).

Fig. 2.

Colocalization of entrapped endosomal fluorescein-labeled dextran and AT_1a receptor antibody binding using flow cytometry in light endosomes freshly isolated from the control rats, ANG II-infused rats, and rats treated concurrently with ANG II and the AT₁ receptor blocker candesartan. Each panel displays data on 2,000 individual endosomes with axes on a log scale. For the entrapped fluorescence, the origin is at the top left corner. A: autofluorescence without entrapped endosomal fluorescein-labeled dextran and with minimal AT_1a receptor antibody binding. B: colocalization of entrapped endosomal fluorescein-labeled dextran and AT_1a receptor antibody binding in a control rat. C: colocalization of entrapped endosomal fluorescein dextran and AT_1a receptor antibody binding in an ANG II-infused rat. D: colocalization of entrapped endosomal fluorescein dextran and AT_1a receptor antibody binding in a rat treated concurrently with ANG II and candesartan. Note a marked right shift in three-dimensional frequency histogram in ANG II-infused rats, which was prevented by candesartan, indicating an increase in AT_1a receptor antibody binding during long-term ANG II infusion and a return to control with concurrent candesartan treatment. From Zhuo JL, Imig JD, Hammond TG, Orengo S, Benes E, Navar LG. Ang II accumulation in rat renal endosomes during Ang II-induced hypertension: role of AT₁ receptor. Hypertension 39: 116–121, 2002 (180).

The endocytic pathway has commonly and long been thought to result primarily in degradation of the internalized molecules by directing them to the lysosomes for destruction or deactivation. However, recent research indicates that endosomes can serve as an intracellular platform for continuous receptor signaling, for example, the MAP kinase ERK1/2 signaling (25, 40, 92, 95, 97, 167). Moreover, internalized ANG II and AT₁ receptors may also be targeted to the nucleus (14, 19, 84, 88, 96, 182). ANG II along with its AT₁ receptors can be taken up, in part, through the classical clathrin-coated pits in VSMCs or through cytoskeleton microtubule-dependent mechanisms in proximal tubule cells (40, 62, 82, 84, 85, 87, 92, 154). The current dogma holds that the receptor is then recycled back to the plasma membrane for reactivation by circulating and paracrine ANG II. However, recent research indicates that the receptor might be trafficked to the nucleus, in addition to the membranes, thereby possibly facilitating downstream long-term genetic or transcriptional effects by intracellular ANG II, as ANG II receptors have been identified both within endosomes (Fig. 2) and the nucleus of proximal tubules or renal cortical cells (Fig. 3) (50, 82, 84, 88, 116, 180). Our in vitro studies in proximal tubule cells supports a microtubule-dependent pathway, as opposed to the canonical clathrin-coated pit pathway (82, 84); additionally, a nuclear localization sequence on the AT_1a receptor has recently been discovered and may be involved in intracellular nuclear trafficking, of both the receptor and of ANG II (14, 96). AT₁ receptors are present on both the basolateral and the apical membranes of proximal tubule cells, and both are capable of undergoing endocytosis. Indeed, much of ANG II signaling on the apical membranes is at least, in part, dependent on endocytosis, and blockade of internalization inhibits sodium flux and PLC activation (135–137, 152). However, blockade of the AT_1a receptor endocytosis on the basolateral membranes appears to have no effect (152). AT₂ receptors, however, are not thought to undergo internalization and trafficking upon agonist stimulation and the source and purpose of intracellular AT₂ receptors are not yet fully elucidated (16, 50, 62). Both receptor types have been found on the nuclear membranes of different cell types, and it is likely that both receptors have a function in gene transcription (16, 50, 62, 88, 89, 149). So far, we have evidence that circulating and/or paracrine ANG II is accumulated within renal endosomes or isolated proximal tubules of ANG II-induced hypertensive rodent models (85, 87, 90, 180). This accumulation of ANG II appears to be predominantly mediated by the AT₁ (AT_1a) receptor in rodent kidneys, as demonstrated in AT_1a receptor-knockout mice (Fig. 4) (85, 87).

Fig. 3.

Effects of the knockdown of AT₁ receptor (AT₁R) expression with AT₁ siRNAs on AT₁R-mediated uptake of FITC-labeled ANG II in rabbit proximal tubule (PT) cells. A–C: FITC-ANG II- (green, white arrows) and 4,6-diamidino-2-phenylindole (DAPI)-stained (blue in B or red in merged image C for optimal visualization) nuclei in PT cells incubated with FITC-ANG II (1 nM) at 4°C for 3 h. FITC-ANG II was restricted to the cell surface (white arrows). D–F: AT₁-mediated uptake of FITC-ANG II in PT cells pretransfected with control siRNAs for 48 h and incubated at 37°C for 1 h. Control siRNAs had no effect on FITC-ANG II uptake; thus, FITC-ANG II was seen in the supranuclear regions and nuclei (F, yellow arrows). G–I: AT₁R-mediated uptake of FITC-ANG II was blocked in PT cells pretransfected with AT₁R siRNAs for 48 h. J: semiquantitated levels of FITC-ANG II uptake. Magnification: ×40. **P < 0.01 vs. control PT cells incubated with FITC-ANG II at 4°C for 3 h. ⁺⁺P < 0.01, untransfected (FITC-ANG II) vs. specific siRNA-transfected PT cells incubated with FITC-ANG II at 37°C for 1 h. [From Li et al. (84).]

Fig. 4.

Visualization of intracellular uptake of ¹²⁵I-labeled Val⁵-ANG II in kidneys of wild-type (C57BL/6J) and AT_1a receptor knockout mice by quantitative in vivo autoradiography (A and B). Intracellular ¹²⁵I-labeled Val⁵-ANG II in the kidney was determined after kidneys were perfused with an acidic buffer to clear blood and cell surface receptor-bound ¹²⁵I-labeled Val⁵-ANG II. Levels of ¹²⁵I-labeled Val⁵-ANG II uptake are keyed to the color calibration bar, with red representing the highest (H) and blue the background (L) level of uptake. C: quantitated intracellular ¹²⁵I-labeled Val⁵-ANG II. **P < 0.01 vs. wild-type. [From Li and Zhuo (87).]

In Vitro Evidence that Intracellular ANG II Induces Intracellular and Nuclear Effects in Proximal Tubule Cells

ANG II is a well-recognized vasoactive peptide hormone in the physiological regulation of proximal tubular sodium and fluid transport and blood pressure homeostasis (15, 58, 99, 138, 176). ANG II is also a well-documented growth and transcriptional factor, cytokine, or chemokine in the development and progression of chronic pathologies, including hypertensive kidney diseases and diabetic nephropathy through its proinflammatory and growth-promoting effects (2, 89, 130, 131, 148, 161, 162, 164, 177, 179). Although the best-recognized effects are mediated by extracellular ANG II acting on cell surface receptors, we and others have gathered evidence during last few years that intracellular ANG II or its major metabolite ANG (1–7) may induce important intracellular and nuclear effects in the proximal tubule cells or the freshly isolated renal cortical nuclei (50, 52, 53, 86, 88, 89, 116, 181, 183). The demonstrated internalization of extracellular ANG II along with AT₁ receptors, as previously discussed, and the identification of AT₁ and AT₂ receptors on in freshly isolated intact nuclei of kidney cells strongly suggest downstream effects at the genetic or transcriptional level (50, 52, 53, 86, 88, 89, 116, 181, 183). For instance, intracellular ANG II has been found to upregulate transcription of precursor components of the RAS itself, such as renin and ACE, and in this way, can increase its own production (103, 104). Recent in vitro studies in a tubular epithelial cell line indicate that chronic exposure to ANG II increased NHE3 promoter activity, apparently through the interplay of several intracellular signaling mechanisms (120). Unfortunately, intracellular activation of AT₁ receptor compared with cell membrane activation was not determined in this study (120).

We recently tested the hypothesis that intracellular ANG II directly induces transcriptional effects in the nucleus of rat renal cortical cells by stimulating AT_1a receptors (88). In our study, the nuclei with an intact nuclear membrane, but devoid of cell surface membranes or other intracellular organelles, were freshly isolated from the rat renal cortex. Both FITC-labeled ANG II and ¹²⁵I-labeled Val⁵-ANG II bound the nuclei with a predominance of AT₁ receptors (Fig. 5). RT-PCR confirmed that AT_1a mRNAs predominated in these nuclei. In freshly isolated nuclei stimulated by ANG II, in vitro transcription of mRNAs for transforming growth factor-β1, macrophage chemoattractant protein-1, and NHE3 were all significantly increased (88). Because these transcriptional responses to ANG II were induced by the agonist in isolated nuclei, devoid of cell surface receptor-mediated signaling and completely blocked by losartan, we suggested that ANG II may directly stimulate nuclear AT_1a receptors to induce transcriptional responses that may be associated with proximal tubular sodium transport, cellular growth and hypertrophy, and proinflammatory responses, especially in chronically elevated ANG II states. Interestingly, in a recent study, Tadevosyan et al. (149) confirmed that cardiomyocyte nuclear membranes also possess ANG II receptors that couple to nuclear signaling pathways and regulate transcription. Furthermore, Chappell and colleagues (50, 53, 116) have demonstrated that the renal cortical nuclei have receptor binding sites specific for AT₁, AT₂, and ANG (1–7), stimulation of which led to increased production of superoxide or nitric oxide. These studies, along with ours, provide further support for a nuclear role for intracellular ANG II.

Fig. 5.

Localization of ANG II receptors in freshly isolated rat renal cortical nuclei using FITC-labeled ANG II. A, D, and G: total FITC-ANG II receptor binding. B, E, and H: DAPI-labeled nucleic acids in the same isolated nuclei. C, F, and I: merged images of FITC-ANG II binding and DAPI-labeled nucleic acids in the same nucleus. J: levels of AT₁ (>90%) and AT₂ receptor binding (<10%) as a percentage of total FITC-ANG II binding. Magnification: ×40. AU, arbitrary fluorescence intensity unit. **P < 0.01 vs. AT₁ binding. [From Li and Zhuo (88).]

The signal mechanisms by which ANG II induces increases in transcriptional responses at the nuclear level are currently not yet fully understood. However, the effects of ANG II on intracellular calcium, a major downstream intracellular signaling for hormone-mediated actions, might, at least in part, play a significant role (29, 33, 43, 54, 55, 121, 181–183). Indeed, Haller et al. (54, 55) reported that intracellular ANG II-induced increases in cytoplasmic [Ca²⁺] were rapidly followed by a rise in intranuclear [Ca²⁺], suggesting an important role of cytoplasmic [Ca²⁺] response. This may be applied to proximal tubule cells as well. In single rabbit proximal tubules cells, we demonstrated that in vitro microinjection of ANG II directly into single rabbit proximal tubules also significantly induced intracellular [Ca²⁺] responses that could be observed in the nucleus (Fig. 6) and that comicroinjection of the AT₁ receptor blocker losartan with ANG II effectively blocked the intracellular ANG II-induced cytoplasmic and nuclear [Ca²⁺] responses (181–183). In contrast, studies by Baker and Kumar (6), using a nonsecreted form of recombinant ANG II, demonstrated that the AT₁ receptor or an AT₁-like receptor is not necessary for ANG II to induce intracrine mitogenic effects in Chinese hamster ovary (CHO) cells. One explanation for this activation in the absence of a known receptor is that ANG II could interact directly with chromatin and induce gene expression in a manner similar to PDGF and insulin (6, 22, 122). Studies in A7r5 VSMCs, which do not express extracellular ANG II receptors and, therefore, do not normally respond to extracellular ANG II, have shown an influx of Ca²⁺ with the introduction of intracellular ANG II (42). Unlike in proximal tubule cells, intracellular ANG II-dependent Ca²⁺ response could not be blocked with extracellular or intracellular addition of AT₁ or AT₂ receptor blockers, indicating activation of the Ca²⁺ responses proceeded by a mechanism that does not rely on any known receptors (6, 142). However, intracellular administration of ANG II in these cells also stimulated cell proliferation, which, unlike Ca²⁺ responses, could be partially blocked both by AT₁ and AT₂ receptor antagonists (73, 143). Additionally, activation of AT₂ or ANG (1–7) receptors in the nuclei of sheep kidney cells has been found to promote nitric oxide signaling and inhibits production of renin (50, 53). It is not yet known whether AT₂ and ANG (1–7) receptors play a role at the genetic level in a similar fashion to AT₁ receptors.

Fig. 6.

Effect of intracellular microinjection of ANG II (1 nM, ∼70–100 fl) on [Ca²⁺]_i responses in single PT cells at baseline (0 s; A) and 15 (B), 30 (C), 60 (D), and 120 (E) s after microinjection of ANG II in the cells. F: relative levels of [Ca²⁺]_i signaling before and after microinjection of ANG II. ROD, relative optical density. Red represents the highest level of [Ca²⁺]_i responses, whereas black is the background. **P < 0.01 vs. basal. [From Zhuo et al. (183).]

Evidence that Intracellular ANG II May Have Mitochondrial Effects

Whether intracellular ANG II has mitochondria-specific effects remains to be further studied (1, 125). Mitochondria are indispensible organelles in eukaryotic cell survival, responsible for energy production, as well as for the generation of reactive oxygen species (125). The mitochondria have been implicated in a number of age-related pathologies, including hypertension, renal disease, and atherosclerosis (27, 125). AT₁ and AT₂ receptors have been localized not only in the nuclei, but also in the mitochondria of a variety of cell types (1, 27, 125). A number of observations indicate ANG II elicits some of its pathophysiological effects directly through its interactions with AT₁ on the mitochondria; many of these interactions may lead to increased autophagy and decreased mitochondria number in the affected cells (27, 125, 165). Conversely, functional AT₂ receptors have been recently localized in the mitochondria of human monocytes, cardiac myocytes, and renal tubular cells, where ANG II stimulated AT₂ receptors to produce nitric oxide (1). This is an exciting area of research that should be explored further for studying long-term genomic and/or transcriptional effects of intracellular/mitochondrial ANG II mediated by AT₁ or AT₂ receptors.

In Vivo Evidence for a Physiological Role of Intracellular ANG II in Proximal Tubular Reabsorption and Blood Pressure Regulation

Whether intracellular ANG II has a physiological role in the regulation of proximal tubular sodium and fluid reabsorption and blood pressure was only recently investigated but still remains poorly understood. As previously discussed, ANG II exerts diverse biological effects in cardiovascular and proximal tubule cells of the kidney in vitro, which include increases in intracellular and/or nuclear [Ca²⁺] (33, 54, 55, 183), increased production of superoxide or nitric oxide (50, 53, 116), increases in proximal tubular cellular NHE3 expression (84, 86), and increased nuclear transcriptional responses in growth factor(s) or proliferative cytokines (18, 20, 88, 89, 179). These in vitro findings strongly suggest a physiological role for intracellular ANG II in proximal tubules of the kidney. Indeed, ANG II is widely recognized to play an important physiological role in blood pressure homeostasis, as well as in the development of hypertension and hypertension-induced target organ injury. At least some of these effects are mediated through its extracellular and intracellular actions on proximal tubules of the kidney. Chronic infusion of ANG II is known to downregulate its AT₁ receptors in many target cells but not proximal tubule cells, where AT₁ receptor expression remains upregulated in response to ANG II (103, 104, 182). Continuous upregulation of AT₁ receptors in proximal tubules of the kidney in the presence of high intratubular and interstitial ANG II levels may have important implications in hypertension.

ACE inhibitors and ARBs have become the first lines of treatments for hypertension and diabetic nephropathy. The studies of the intracellular RAS may have important implications for the future of hypertension treatment. Many ARBs are less effective at blocking intracellular ANG II-induced effects due to limited cell permeability. As previously discussed, glucose-induced intracellular ANG II synthesis is not dependent on ACE, rendering ACE-inhibitors less effective (73, 142, 143). However, it has been difficult to discover the effects of intracellular ANG II because of a lack of approaches that can differentiate the effects induced by intracellular ANG II via cytoplasmic and nuclear receptors from those induced by extracellular ANG II acting on the cell surface receptors. Recently, Cook and colleagues (127) developed a transgenic mouse model for the first time that expresses an intracellular ANG II protein that fuses to the downstream of enhanced cyan fluorescent protein (ECFP)/ANG II. These mice expressing ECFP/ANG II have their basal blood pressure significantly higher than in nontransgenic control counterparts. Because the transgene was expressed in all tested tissue with no circulating fusion proteins discovered in the plasma, the blood pressure-increasing effect was entirely dependent on the intracellular ANG II fusion protein. Additionally, VSMCs were found to proliferate despite normal levels of plasma ANG II (127). This is the first study to show that the global expression of an intracellular ANG II fusion protein may increase blood pressure and induce intrarenal microangiopathy in mice. However, this study does not explore the effects of intracellular expression of this fusion protein in proximal tubules of the kidney.

We have recently sought to test the hypothesis that proximal tubule-specific expression of an intracellular cyan fluorescent fusion of ANG II via intrarenal adenoviral gene transfer selectively in proximal tubules of the kidney may alter proximal tubular reabsorption and blood pressure in rats and mice (83). We developed an adenoviral construct encoding a nonsecreted form of ANG II, ECFP/ANG II, and a proximal tubule cell-specific promoter, sglt2 (129), which was used to drive ECFP/ANG II expression selectively in proximal tubules of the kidney. In both rats and C57BL/6J mice, time-dependent expression of ECFP/ANG II selectively in proximal tubules of the kidney (Fig. 7) and increases in blood pressure were demonstrated, accompanied by 24-h decreases in urinary sodium and lithium excretion. Because the blood pressure-elevating effect of ECFP/ANG II was blocked by concurrent losartan treatment and in AT_1a receptor-knockout mice, we suggest this effect is mediated by AT₁ (AT_1a) receptors (Fig. 8) (83). This proximal tubule-specific expression of intracellular ANG II resulted in blood pressure increases comparable to those seen in transgenic mice globally expressing intracellular ANG II (127). However, blood pressure increased by proximal tubule-specific expression of intracellular ANG II were still much smaller than in rodents chronically infused with exogenous ANG II, as we reported previously (85, 87, 90, 180). The mechanisms in which AT_1a receptors are involved in this blood pressure increase induced by ECFP/ANG II is not fully understood, but it is likely that increases in sodium and fluid reabsorption as a result of increased NHE3 expression may be involved. In vitro studies have shown that ANG II stimulation of AT_1a receptors on proximal tubule cells or freshly isolated renal cortical cell nuclei could induce NHE3 transcriptional responses (84, 86, 182). More studies should be directed to further study the signaling mechanisms by which intracellular ANG II induced physiological responses in vivo, including proximal tubular transport responses and blood pressure.

Fig. 7.

Proximal tubule-selective expression of an intracellular enhanced cyan fluorescent fusion of ANG II (ECFP/ANG II) in the rat kidney 2 wk after intrarenal adenoviral transfer in the superficial cortex. A: specific ECFP/ANG II expression in proximal tubules of the left kidney, which received ECFP/ANG II transfer. B: lack of ECFP/ANG II expression in the left inner medulla. C: lack of ECFP/ANG II expression in the right contralateral renal cortex. D: lack of ECFP/ANG II expression in the right contralateral inner medulla. Please note that ECFP/ANG II proteins are shown as blue-green (cyan), whereas nuclear DNA is stained with DAPI and converted into the red color for better visualization. G, glomerulus. PT, proximal tubules. Scale bar = 100 μm. [From Li et al. (83).]

Fig. 8.

Effects of proximal tubule-specific transfer of an intracellular cyan fluorescent fusion of ANG II (ECFP/ANG II) or its scrambled control, ECFP/ANG IIc, with or without losartan treatment on systolic blood pressure in rats (SBP; A) or ECFP/ANG II transfer in wild-type or AT_1a-KO mice (B). **P < 0.01 vs. basal SBP. ⁺⁺P < 0.01 vs. SBP in ECFP/ANG II-transferred rats. [From Li et al. (83).]

Perspectives and Significance

In summary, we have reviewed the progress recently made on the novel roles and signaling pathways of intracrine and/or intracellular ANG II in cardiovascular, renal, and blood pressure regulation with a focus on proximal tubules of the kidney (Fig. 9). Although it still remains incompletely understood and often debatable, recent studies generally support an intracellular and/or nuclear role for ANG II and other members of the RAS in cardiovascular, renal, and neural systems. In proximal tubules of the kidney, we now have functional evidence that circulating and paracrine ANG II is internalized via AT₁ (AT_1a) receptor-mediated endocytosis (or uptake), which contributes significantly to high levels of intracellular ANG II in this tubular segment. Upon internalization, not all ANG II and AT₁ receptors are transported to lysosomes for destruction or for desensitization as widely expected, since internalized ANG II and its receptors can be readily visualized in other intracellular organelles and the nucleus, and importantly, long-term infusion of ANG II leads to persistent hypertension and target organ injury. We also have the physiological evidence that ANG II, when administered or expressed intracellularly in proximal tubular cells, may induce significant intracellular and nuclear responses. In cultured proximal tubule cells, intracellular ANG II may directly stimulate cytoplasmic and nuclear AT₁ (AT_1a) receptors to mobilize intracellular [Ca²⁺] and activates nuclear transcription factor NF-κB, leading to increased expression or transcription of NHE3 and proinflammatory cytokines and growth factors in proximal tubule cells. Indeed, augmented intrarenal NF-κB activity appears to play an important role in the development of ANG II-induced hypertension renal injury (2, 108). Under physiological conditions, intrarenal adenoviral transfer of an intracellular cyan fluorescent fusion of ANG II in proximal tubules of the kidney selectively increases proximal tubular reabsorption and blood pressure in rats and mice. Our studies suggest that increased expression and/or activity of NHE3 and proinflammatory cytokines and growth factors by intracellular ANG II may at least, in part, contribute to sodium retention and tubulo-interstitial injury in ANG II-induced hypertension. However, more studies are required to further elucidate the precise roles and signaling mechanisms of the intracellular RAS in proximal tubules of the kidney, especially underlying the long-term genomic or transcriptional effects of intracellular ANG II and its nuclear receptors and how their interactions affect transcriptional responses. Furthermore, it may be necessary to generate various transgenic animals, which express the so-called intracellular ANG II in a cell- or tissue-specific manner with or without the enzyme renin or ACE or a particular receptor for ANG II, to facilitate the studies on long-term physiological roles of intracellular and/or intracrine ANG II or other biologically active ANG fragments. Clearly, in the future, improved understanding of the intracellular RAS may have important therapeutic and clinical implications in the development of new strategies or drugs to block the long-term genomic or transcriptional effects of intracellular ANG II and its nuclear receptors in diabetes, progressive renal diseases, and hypertension.

Fig. 9.

A schematic representation showing that extracellular ANG II may induce short-term physiological responses in proximal tubule cells of the kidney through the classic cell surface GPCR-mediated signaling pathways, or long-term genomic and/or transcriptional responses mediated by internalized ANG II and AT_1a receptors under pathophysiological conditions. The classic GPCR-mediated signaling pathways play an essential role in the physiological regulation of proximal tubular sodium, solute, and fluid transport by acutely altering the activities of NHE3 and other sodium transporters. Alternatively, the persistent elevation in the circulating and tissue ANG II levels, such as during ANG II-dependent hypertension, continuously increases AT₁ (AT_1a) receptor-mediated uptake of ANG II by proximal tubules of the kidney. Internalized ANG II may act as an intracellular peptide to activate cytoplasmic and nuclear receptors to induce long-term genomic or transcriptional effects on the expression of NHE3 and other sodium transporters, growth factors, and cytokines, with the consequences of increased sodium retention and development of hypertension and tubulointerstitial injury in the kidney.

GRANTS

The work as described in this article was supported in part by the National Institute of Diabetes and Digestive and Kidney Disease Grants (5RO1DK067299, 2R56DK067299-06, and 2RO1DK067299-07), the National Kidney Foundation of Michigan, the American Heart Association Grant-in-Aids, and the American Society of Nephrology M. James Scherbenske Grant to Dr. Jia L. Zhuo.