Role of Collectrin, an ACE2 Homologue, in Blood Pressure Homeostasis

Abstract

Collectrin (Tmem27) is a transmembrane glycoprotein that is highly expressed in the kidney and vascular endothelium. It is a homolog of the angiotensin converting enzyme 2 (ACE2) but harbors no catalytic domain. In the extravascular tissues of the kidney, collectrin is localized to the proximal tubule and collecting duct. Collectrin-deficient mice are featured with hypertension and exaggerated salt sensitivity. These phenotypes are associated with impaired uptake of the nitric oxide precursor L-arginine and the expression of its amino acid transporters, CAT-1 and y+LAT1, in endothelial cells. In addition, collectrin-deficient mice display decreased dimerization of nitric oxide synthase and decreased nitric oxide synthesis, but enhanced superoxide generation, suggesting that deletion of collectrin leads to a state of nitric oxide synthase uncoupling. These findings suggest collectrin plays a protective role against hypertension. The collectrin knockout mouse represents a unique model for hypertension research. Furthermore, collectrin may serve as a novel therapeutic target in the treatment of hypertension.

Introduction

Hypertension is a major risk factor for cardiovascular and renal diseases worldwide. Its pathogenesis is complex, and involves several key pathways, including the sympathetic adrenergic system, renin-angiotensin-aldosterone System (RAAS), and nitric oxide (NO) and superoxide (O₂•−) balance. While several pathways are commonly targeted for treatment of hypertension, these existing therapies are only partially effective [1]. Currently, no treatment modality targeting the NO and/or O₂•− pathways has been established clinically for the treatment of hypertension. Recently, a novel gene, collectrin (Tmem27), has been identified as a key regulator of arterial pressure, which may modulate the balance of NO and/or O₂•−. This review will summarize current knowledge about collectrin and its role in blood pressure homeostasis.

Collectrin is a member of the Angiotensin Converting Enzyme Family

Collectrin gene (Tmem27) encodes a 222-amino acid transmembrane glycoprotein that shares 47.8% homology with angiotensin converting enzyme 2 (ACE2) [2]. Both ACE2 and collectrin are located on the X chromosome. By genomic sequence alignment, ACE2 has been proposed as a chimeric duplication of ACE and collectrin [3]. However, unlike ACE and ACE2, collectrin lacks enzymatic activity because of the absence of an active dipeptidyl carboxypeptidase catalytic domain [2]. Collectrin is highly conserved among species, sharing >80% identity between mouse, rat, and human [2]. Gene expression database and published studies demonstrate that collectrin is most abundantly expressed in the kidney [2]. The renal expression of collectrin is upregulated during salt-sensitive hypertension (SSH) [4] and after 5/6 nephrectomy [5]. Taken together, these observations raised the question whether collectrin plays a role in blood pressure regulation.

Function of Collectrin may be tissue specific

During mouse kidney development, collectrin expression is first detected at the ureteric bud branches on day 13 of gestation [2], followed by the apical brush border of proximal tubule [6], specifically in the S1 segment [7], as well as collecting duct [2]. In addition to the kidney, collectrin is highly expressed in pancreatic β-islet cells [8, 9], followed by intestine, liver, heart, as well as stomach to a lesser extent [10]. Importantly, collectrin is also expressed in the vascular endothelium of conduit and resistance vessels [11]. Because of its varied localization, collectrin may have distinct effects specific to its tissue localization. Of significance, collectrin is a downstream target of the hepatocyte nuclear factor-1α (HNF-1α) [8, 9], the mutations of which account for up to 73% of maturity onset diabetes of the young (MODY) [12, 13]. The deletion of HNF-1α in the mouse leads to Fanconi syndrome that is featured with polyuria, glucosuria, phosphaturia, and aminoaciduria [14]. We and others reported that deletion of collectrin in the mouse results in severe generalized urinary amino acid wasting due to reduced expression of neutral and cationic amino acid transporters in the plasma membrane of the proximal tubule brush border [6, 10]. Collectrin knockout mice also display polyuria, likely from a urinary concentrating defect, but the other features of Fanconi syndrome are absent [6, 10]. It is likely that collectrin acts as a chaperone for the trafficking of amino acid transporters in the proximal tubule [6]. In the vascular endothelium described in detail below, collectrin influences the uptake of L-Arginine (L-Arg). In the pancreas, gene overexpression and deletion studies in the mouse reveal that collectrin is involved in insulin secretion [8] and islet cell mass [9]. Whether there is a common pathway through which collectrin mediates its biological effect in different tissues remains to be determined. It is possible that a common pathway may be through its binding with the SNARE complex. In the colleting duct, collectrin interacts with snapin, SNAP-23, syntaxin-4 and VAMP-2 [4]. In the pancreas, collectrin facilitates SNARE complex formation through interacting with snapin, a synaptosomal-associated protein [8].

In addition to the full-length 222-amino acid collectrin, a cleaved soluble N-terminal (extracellular) fragment of collectrin has also been identified in both the pancreas [15] and the kidney [16]. The cleavage is mediated by beta site amyloid precursor protein cleaving enzyme 2 (Bace2) [15]. Currently, the biological function of the cleaved collectrin is still unknown.

Collectrin Links Amino Acid Transport with Blood Pressure Regulation

Its homology to ACE2 and its upregulation in the kidney during high salt feeding and after 5/6 nephrectomy raised the central question whether collectrin might play a role in blood pressure homeostasis. It was thus surprising that the key phenotype observed in collectrin knockout (KO) mice is massive urinary amino acid wasting [6, 10]. In our initial characterization of collectrin KO mice, on a mixed genetic background of 129/SvEv × C57BL/6J, we did not detect a statistically significant difference in arterial pressure compared to wild-type (WT) mice [4]. Due to the possibility of genetic background masking any arterial pressure effect commonly observed in many mouse models of cardiovascular diseases and hypertension (HTN), including Ace2 [17], we therefore backcrossed the collectrin null mutation on the 129/SvEV background for > 12 generations to generate inbred 129/SvEv KO mice. This 129 strain is more susceptible to the development of HTN and kidney disease than the C57BL/6J [18, 19]. Furthermore, it is well recognized that the 129/SvEv strain is a salt-sensitive strain [20-22], and collectrin’s expression is upregulated during SSH in rat models [4]. On this salt sensitive background, we found that collectrin KO mice have hypertension at baseline and augmented salt sensitivity that is associated with impaired pressure natriuresis [11]. These phenotypes are associated with increased O₂•− generation and decreased NO production in the kidney in vivo, and impaired endothelium-dependent relaxation of resistance arteries ex vivo. These effects resulting from deletion of collectrin are independent of activation of the RAS, since there were no detectable significant differences in mRNA levels of ACE-2, Mas receptor [11], and AT_1A receptor (unpublished data). There was even a trend towards lower plasma renin activity in KO mice [11]. These observations suggest that collectrin KO mice have impaired arterial pressure control as a result of abnormal NO synthesis and O₂•− generation, an imbalance that may involve the uncoupling of nitric oxide synthase (NOS). Furthermore, our findings raised a key question whether there is a causal link between amino acid transport defect and blood pressure regulation in collectrin KO mice.

Collectrin and the L-arginine-nitric oxide pathway

The NO signaling pathway is recognized as a key second messenger systems involved in the regulation of vascular tone and arterial pressure [23]. NO is synthesized from L-Arg, a semiessential cationic AA, through a reaction catalyzed by NOS in the presence of the cofactor tetrahydrobiopterin (BH4) [24]. There are three NOS isoforms: endothelial (eNOS) and neuronal (nNOS) isoforms are constitutively expressed, and the inducible isoform (iNOS) mainly expressed by activated inflammatory cells. Both eNOS and nNOS are only fully functional in a dimeric form, and the stabilization of their dimeric forms is dependent on several factors, including the availability of L-Arg substrate and BH4 co-factor, and S-nitrosylation [25]. In the absence of L-Arg and/or BH4, NOS cannot generate NO, but is capable of generating O₂•−, a phenomenon commonly referred to as “uncoupling” [25, 26] that can lead to the development of endothelial dysfunction (ED) or impaired ability of the vascular endothelium to stimulate vasodilation, and hence the maintenance of HTN. To evaluate for possible altered eNOS expression in collectrin KO mice, we compared baseline levels of eNOS dimer (active eNOS) to monomer (inactive eNOS) in the aorta, kidney cortex and medulla from WT and collectrin KO mice. By protein analysis in all three tissues, expression of eNOS dimer was significantly lower in KO mice compared to WT [11]. In addition, nNOS dimer levels were also significantly lower in renal tissue in collectrin KO mice (unpublished data). Our data suggest a state of NOS uncoupling that favors increased O₂•− production in collectrin KO mice.

The altered balance of NO and O₂•−, diminished NOS dimerization or “uncoupling”, and the known functional property of collectrin as a chaperone of amino acid transporter in the proximal tubule raised the possibility that the central defect in blood pressure homeostasis could be due to impaired cellular uptake of L-Arg in collectrin KO mice. Indeed, we showed that primary endothelial cells from collectrin KO mice were shown to have decreased [³H]L-Arg uptake [11]. In addition, overexpression of collectrin in primary human coronary endothelial cells resulted in increased [³H]L-Arg uptake [11].

There is emerging evidence that there are several different L-Arg sources or pools for NO synthesis by NOS: 1) plasma membrane transport of L-Arg from the extracellular space [27], 2) intracellular L-citrulline to L-Arg recycling [28], and 3) lysosomal and proteasomal protein degradation that in turn also generates the L-Arg analogue asymmetric dimethyl-L-Arg (ADMA), an endogenous inhibitor of NOS [29]. Certain pathophysiologic conditions may determine which L-Arg pool is the rate limiting source [29]. In several cell types, including the endothelium and renal epithelium, synthesis of NO requires the de novo import of extracellular L-Arg [30-32]. Under normal physiologic conditions, intracellular concentration of L-Arg is in far excess (20 to 760 fold) of the Km for both eNOS and nNOS, and normal plasma L-Arg concentration in humans and animals ranges 5 to 10 fold the Km for NOS [33, 34]. However, further elevation of extracellular L-Arg levels by L-Arg administration directly enhances endothelial NO production. The dependence of endogenous NO generation on the uptake of extracellular L-Arg has led to the concept of “L-arginine paradox” [33, 34]. It follows that a defect in the import of extracellular L-Arg can cause endothelial dysfunction and HTN. In support, Schaich et al. demonstrated that impaired L-Arg transport, both in vitro and in vivo, occurs in patients with essential HTN, as well as in normotensive individuals with a positive family history of essential HTN [35]. The effect of L-Arg administration may be mediated both by endothelial and renal epithelial cells. In this regard, hypertensive humans and animals respond to L-Arg administration with reduced arterial pressure and improved endothelial vasodilator function [36, 37]. Renal medullary interstitial infusion of L-Arg to Dahl salt-sensitive rats prevents the development of HTN during high-salt feeding [38]. We reported that treatment with L-Arg partially lowered blood pressure in collectrin KO mice [11]. Taken together, these data suggest that collectrin may also chaperone the proper localization and/or function of amino acid transporters for L-Arg.

L-Arg transport into cells is mediated by different classes of amino acid transporters that are defined by their ion dependency, substrate specificity, and relative affinity [39, 34]. In endothelial cells (ECs), the system y⁺, a sodium-independent system, accounts for ~60% of L-Arg transport, and system y⁺L, a sodium-dependent system, accounts for ~40% of L-Arg transport [40, 41, 30]. System y⁺ selectively mediates the cellular transport of cationic amino acids, including L-Arg, whereas system y⁺L transports both cationic and neutral amino acids [30, 32]. Cloning studies have identified at least 3 cationic transporters in system y⁺, namely, the cationic amino acid transporter (CAT) proteins: CAT1, CAT2A and CAT2B isoforms, and CAT3 [39]. System y⁺L is a glycoprotein-associated amino acid heterodimeric transporter whose heavy chain is the glycoprotein 4F2hc/CD98, and the hydrophobic “light chain”, y⁺LAT1 or y⁺LAT2, is responsible for the recognition and operational features of the transport process [42]. Both y⁺ and y⁺L systems are also expressed in renal endothelial and epithelial cells, and play an important role in influencing L-Arg uptake, NO production, medullary blood flow, and arterial pressure [31, 43, 44, 38].

Analogous amino acid transport systems, encoded by different genes, exist in epithelial cells such as the kidney and intestine. In the kidney, in addition to y⁺ and the y⁺L systems which have been well described in the renal endothelium and inner medullary CD [43, 45], the broad-scope transporters also have been well characterized: B⁰, system 1, is sodium-dependent and is a low affinity transporter for almost all neutral amino acids [46] but also has preference for the cationic substrates lysine and arginine [47, 48]; b^0,+, system 2, is sodium-independent, and is an antiporter that takes up cationic amino acid in exchange for neutral amino acids [46]. These two systems are expressed along the brush border of the proximal tubule [48]. We and others reported that collectrin KO mice have severe aminoaciduria due to decreased protein expression of B⁰AT1 (B⁰ system), and b^0,+AT/rBAT heteromeric complex (b^0,+ system) in the plasma membrane fractions of the renal proximal tubule [6]. Furthermore, the expression of the y⁺L system (y⁺LAT1 and y⁺LAT2) in kidney plasma membrane fractions are also decreased in collectrin KO mice [11]. Moreover, through gene knockdown and overexpression studies, we found that collectrin directly mediates the plasma membrane expression of the y⁺ and y⁺L transporter systems in ECs [11]. Collectively, our studies suggest that collectrin is a genetic factor that regulates arterial pressure and salt-sensitivity, likely by modulating the L-Arg/NO/O₂•− pathway through its role in regulating expression of amino acid transporters and cellular L-Arg uptake.

Our finding provides a corollary to a published report that a patient with compound heterozygosity for 2 mutations in the SLC7A7 gene that encodes the y+LAT1 transporter exhibited vascular endothelial dysfunction that was corrected with L-Arg infusion [49]. Similarly, the diminished CAT1 (SLC7A1) expression caused by a single nucleotide polymorphism (SNP) in the 3′ UTR of SLC7A1 has been linked to hypertension and endothelial dysfunction [50]. Thus, collectrin may modulate endothelial function and blood pressure through its central role in chaperoning the L-Arg transporters CAT1 and y+L. However, while decreased L-Arg availability due to impaired uptake may be the cause for decreased eNOS dimerization in the KO mice, we cannot rule out the possibility that collectrin may influence eNOS dimerization independently of its effect on L-Arg uptake by interacting with other proteins that play a role in stabilizing eNOS. For example, caveolae, which are important in determining eNOS intracellular localization and activity [25], contain some members of the SNARE complex [51] with which collectrin binds and interacts [8, 4].

Although the functional role of collectrin in L-Arg uptake has been delineated primarily via endothelial cells in vitro, collectrin was initially identified in the collecting duct [2]. We have since confirmed this finding [6]; and, specifically, while collectrin is expressed on the brush border of the proximal tubule, it is localized to the basolateral side of the collecting duct. Of particular relevance, the cellular uptake of L-Arg by CAT1 in the renal medulla is critical for NO production, maintenance of medullary blood flow, and regulation of Na⁺ excretion and blood pressure [31, 43]. The source of NO in the renal medulla could be from either eNOS and/or nNOS. Neuronal NOS is highly expressed in the collecting duct, and is significantly increased in the early phase of high salt intake, where it is thought to generate NO as an adaptive response to salt intake [10]. The renal expression of collectrin is also increased during high salt diet [4], suggesting that collectrin is involved in this pathway. We queried whether loss of collectrin also affects nNOS dimerization and its activity. Western analysis of renal medullary tissue revealed that nNOS dimerization is significantly diminished in KO mice compared to WT, despite no changes in mRNA levels (unpublished data), consistent with our findings for eNOS. This suggests that collectrin influences the dimerization of both eNOS and nNOS post-transcriptionally, likely through its role mediating L-Arg uptake.

Data have emerged that collecting duct-derived NO plays a key role in the regulation of arterial pressure and sodium excretion. In particular, Kohan and associates demonstrated that the effect of collecting duct-derived endothelin in arterial pressure regulation and Na⁺ excretion is mediated through NO [52]. Recently, Pollock and associates showed that collecting duct nNOS regulates arterial pressure and fluid-electrolyte homeostasis [53]. Our findings that collectrin KO mice have impaired pressure-natriuresis relationship that is associated with decreased nNOS activity/dimerization is consistent with the notion that collectrin in the collecting duct regulates arterial pressure through its effect on nNOS and Na⁺ excretion.

Future Considerations

Guyton and colleagues hypothesized that the kidney’s substantial capacity for Na⁺ excretion provides a compensatory system of virtually infinite gain to oppose processes causing elevation in arterial pressure, including increases in peripheral vascular resistance (PVR). It follows that, in collectrin KO mice, even if the initial cause of HTN is due to increased PVR from reduced vasorelaxation, a defect in renal excretory function would be a pre-requisite for the sustained chronic increase in arterial pressure. The defect in Na⁺ handling and altered pressure-natriuresis observed during high salt diet feeding in collectrin KO mice suggests that altered kidney function could result from a defect in renal epithelial function and/or renal hemodynamics. Future efforts including generation of mouse lines in which collectrin is deleted specifically in the endothelium, proximal tubule and collecting duct, will be crucial to determine the relative contribution of collectrin in each of these tissues to blood pressure homeostasis.

The dependence on a proper genetic background for the measureable effect of collectrin on arterial pressure suggests that there is a genetic modifier that facilitates the development of hypertension in the context of loss of function of collectrin. Gene mapping studies using 129/SvEv × C57BL/6J intercrosses may enable identification of this genetic modifier.

Collectrin is located on the X chromosome, and is harbored by a locus that modulates blood pressure in female rats in the Sabra rat model of SSH [54], raising the possibility that collectrin may be a genetic factor in HTN. Although it is not known whether there are any loci on the X chromosome that contribute to HTN in humans - since genome wide association studies (GWAS) have largely excluded X chromosome variants from analyses - there are known variants altering mRNA transcript structures (splicing, frameshift and stop gained SNPs) and amino acid composition (missense/non-synonymous SNPs) in the human collectrin gene (http://ensembl.org/Homo_sapiens/Gene/Variation_Gene/Table?db=core;g=ENSG00000147003;r=X:15645441-15683154;t=ENST00000380342). Our eventual goal would be test for the association of these human SNPs in a cohort of patients with hypertension and salt-sensitivity.

Potential Clinical Application

Collectively, the evidence demonstrates that collectrin has a protective role in HTN and salt sensitivity, at least by maintaining a balance between NO and O₂•−, likely through its central function in chaperoning L-Arg transport in endothelial cells and possibly in renal epithelial cells. Theoretically, modulation of the expression of collectrin, either by increasing its expression or decreasing its metabolism, might serve as a novel therapeutic approach in the treatment of HTN. In this regard, the beta site amyloid precursor protein cleaving enzyme 2 (Bace2), a transmembrane aspartic protease that is expressed in multiple tissues including the pancreas and kidney [16], might be a therapeutic target of this pathway. It has been demonstrated that Bace2 is able to cleave collectrin and releases the soluble N-terminal fragment of collectrin [15], resulting in rapid degradation of the remaining membrane-bound C-terminal collectrin fragment [9]. The deletion of Bace2 causes an increase in collectrin protein expression [15]. Currently, the cleavage of collectrin has only been reported in beta cells in the pancreas [15] and primary proximal tubular cells isolated from mouse kidneys [55], but the biological significance of the cleavage as well as the soluble N-terminal fragment are unknown.

Pharmacological inhibition of Bace2 is able to increase β-islet cell mass, increase insulin secretion and improve glycemic control in obesity mouse models (ob/ob and C57BKS^db/db) [15]. These effects are thought to be through inhibition of the cleavage of collectrin. However, whether Bace2 inhibitor is able to modulate blood pressure remains to be determined. Currently, the available Bace2 inhibitors are not selective and also exhibit off-target effect on other BACE enzymes [56].

An alternative therapeutic approach might involve direct modification of collectrin dimerization, deglycosylation, plasma membrane depletion and its extracellular Phe-Phe motif to enable collectrin to be resistant to cleavage of Bace2 [56]. However, these modifications might inevitably affect the physiological characteristics of collectrin such as its plasma membrane localization [56]. Future research focusing on selective Bace2 inhibition or direct collectrin stabilization may yield novel therapeutic strategies for the treatment of HTN, particularly in conditions where the balance of NO and O₂•− is altered.

Conclusion

Arguably a distant member of the Renin-Angiotensin System, collectrin is a highly conserved gene that encodes a 222-amino acid transmembrane glycoprotein that is a member of the angiotensin-converting enzyme (ACE) family, sharing ~50% homology with ACE2, a carboxypeptidase that directly converts the vasoconstrictive peptide angiotensin II (Ang II) to the vasodilatory peptide angiotensin 1-7. Unlike ACE2, collectrin lacks a catalytic domain. In the context of conceptual advancement, the newly discovered role of collectrin as a mediator of L-Arg uptake in endothelial cells directly links L-Arg uptake together with “eNOS uncoupling” in blood pressure regulation, by modulating NO and O₂•− balance. These findings may challenge the existing paradigm of the Renin-Angiotensin System as a simple cascade of enzymatic events. Collectrin may also influence other physiologic and pathological conditions that are dependent on nitric oxide and/or superoxide signaling.