Identification of a novel non-AT1, non-AT2 angiotensin binding site in the rat brain

BR A IN RE S E A RCH 1 1 43 ( 20 0 7 ) 8 3 –9 1

a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m

w w w. e l s e v i e r. c o m / l o c a t e / b r a i n r e s

Research Report

Identification of a novel non-AT1, non-AT2 angiotensin binding site in the rat brain Vardan T. Karamyan, Robert C. Speth ⁎ Department of Pharmacology and Research Institute of Pharmaceutical Sciences, School of Pharmacy, University of Mississippi, University, MS 38677, USA

A R T I C LE I N FO

AB S T R A C T

Article history:

Efforts to protect radiolabeled angiotensins from metabolism during receptor binding

Accepted 18 January 2007

assays date back more than 30 years. However, this continues to be a problem. This study

Available online 24 January 2007

focused on the effects of a protease inhibitor, p-chloromercuribenzoate (PCMB), on the binding of 125I-Ang II to rat brain membranes. Addition of PCMB to the incubation medium

Keywords:

revealed a high affinity binding site for 125I-Ang II in brain membranes (Kd = 1–4 nM) with a

Angiotensinases

greater amount of binding than revealed in previous studies of brain Ang II receptors.

Proteases

Further characterization of this binding, revealed it to be insensitive to inhibition by losartan

p-chloromercuribenzoate (PCMB)

(an AT1 receptor antagonist) and PD123319 (an AT2 receptor antagonist). This non-AT1, non-

Sulfhydryl

AT2 binding site was not present in liver or adrenal membranes. It was activated by a limited

125

I-angiotensin II

range of concentrations of PCMB, with maximal activation at 0.3–1 mM. This binding site

Angiotensin receptor

was equally abundant in cerebral cortex (a brain region with few Ang II receptors) and the hypothalamus (a brain region with abundant Ang II receptors). The binding site was also present in mouse brain, but not mouse liver. The binding site shows high affinity for Ang I, Ang II and Ang III (Ki ∼ 40–100 nM), but lesser affinity for smaller angiotensin fragments and other neuropeptides. This binding site shares some characteristics with the liver cytosolic Ang II binding proteins, later identified as endopeptidases EC 3.4.24.15 and/or EC 3.4.24.16. However, some unique characteristics of this non-AT1, non-AT2 binding site suggest that it may be a novel angiotensin binding substance. © 2007 Elsevier B.V. All rights reserved.

1.

Introduction

The brain renin–angiotensin system (RAS) is recognized as one of several tissue RASs described in a number of organs, having an important role in the control of cardiovascular functions, body fluid and mineral balance and overall homeostasis (Fitzsimons, 1998; Saavedra, 1992, 2005; Severs and DanielsSevers, 1973; Speth et al., 2003; Unger et al., 1988). The major effector peptide of the RAS is angiotensin II (Ang II) which acts through two receptor subtypes-the

angiotensin type 1 (AT1) and type 2 (AT2) receptors (de Gasparo et al., 2000; Kaschina and Unger, 2003). The AT1 receptor mediates classic physiological as well as pathological effects of Ang II, including regulation of blood pressure and hydromineral balance, thirst and sodium intake associated behaviors, cyclicity of reproductive hormones, cell proliferation and angiogenesis. The effects of the AT2 receptor are less clear. Many of its effects are opposite to those mediated by the AT1 receptor, e.g., it causes vasodilation. It is also involved in development, cell differentiation,

⁎ Corresponding author. Fax: +1 662 915 5148. E-mail address: [email protected] (R.C. Speth). URL: http://www.pharmacy.olemiss.edu/pharmacology/Speth.html (R.C. Speth). 0006-8993/$ − see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2007.01.051

84

BR A IN RE S EA RCH 1 1 43 ( 20 0 7 ) 8 3 –91

tissue regeneration and apoptosis (de Gasparo et al., 2000; Kaschina and Unger, 2003). The brain RAS has a pivotal role in normal function of the brain and cardiovascular system, and is important in the pathogenesis of many disease conditions (Fitzsimons, 1998; Saavedra, 1992, 2005; Speth et al., 2003; Unger et al., 1988). Our knowledge of the function of the brain RAS under normal and disease conditions has improved considerably since its discovery over 35 years ago. However, many questions remain as to how the brain RAS controls brain and cardiovascular functions, e.g., the identity of the effector peptide(s), specificity/similarity of the receptors compared with those in the periphery, the role of AT2 receptors, new functions of the receptors, their downstream signaling pathways, existence of the receptors in specific brain nuclei, proteases involved in metabolic processing and degradation of angiotensin peptides, alternative pathways of generation and degradation of angiotensin peptides and the possible existence of functional, uncharacterized angiotensin binding sites. The initial goal of this study was to establish experimental conditions using peptidase inhibitors that would insure stability of Ang II and other angiotensin peptides during receptor binding assays of brain AT1 receptors without impairment of angiotensin receptor binding. However, in the course of these experiments, it was discovered that the organomercurial sulfhydryl agent p-chloromercuribenzoic acid (PCMB) which is reported to be an “angiotensinase inhibitor” (Fujioka et al., 1995; Kohara et al., 1991), unmasked a non-AT1, non-AT2 angiotensin binding site in the rat brain. In this paper we, report for the first time, a preliminary pharmacological characterization of this non-AT1, non-AT2 angiotensin binding site.

2.

Results

There was very little specific 125I-Ang II binding (not significantly different from zero) in the rat hypothalamus in the presence of 10 μM PD123319 (to restrict binding to only the AT1 receptor subtype), incubated at room temperature for 1 h in assay buffer without any additional protease inhibitor (Fig. 1a). In contrast, 125I-SI Ang II showed a moderate amount of specific, saturable binding (Fig. 1b) under the same assay conditions. Addition of 0.3 mM of PCMB to the assay buffer containing 10 μM PD123319 with 1 h incubation gave substantial specific and saturable 125I-Ang II binding in the hypothalamus Bmax = 1.73± 0.1 fmol/mg wet weight and Kd = 2.53 ± 0.29 nM (n = 4). However, when losartan (10 μM) was added, the binding of 125I-Ang II was largely unaffected (Bmax = 2.07 ± 0.21 fmol/mg wet weight and Kd = 4.36 ± 0.78 nM, n = 9). When 125I-Ang II binding was assessed over varying incubation times, maximum specific 125I-Ang II binding in the presence of 0.3 mM PCMB, and 10 μM losartan and PD123319 occurred at 45–60 min, after which it declined (Fig. 2). As the binding of 125I-Ang II did not reach equilibrium conditions (mainly because of its severe metabolic degradation), all incubations of 125I-Ang II binding in the presence of 0.3 mM PCMB, and 10 μM losartan and PD123319 were carried out for 1 h to have maximum specific binding of 125I-Ang II.

Fig. 1 – Representative saturation curves of 125I-Ang II (a) and I-SI-Ang II (b) in the rat hypothalamus. The specific binding of 125I-Ang II is not significantly different from zero. The specific binding of 125I-SI-Ang II is Bmax = 0.6 ± 0.04 fmol/mg wet wt., and Kd = 0.18 ± 0.05 nM. Rat hypothalamic membrane preparations contained 10 μM PD123319. Tot—total binding, NSP—non-specific binding, SP—specific binding.

125

Since 10 μM losartan (along with PD123319, but in the absence of PCMB) completely eliminated specific 125I-Ang II binding, as well as 125I-SI Ang II binding to hypothalamic, adrenal and liver tissue, this indicates that the 125I-Ang II binding found in the presence of 0.3 mM PCMB, 10 μM losartan and 10 μM PD123319 is to a non-AT1, non-AT2 binding site (Fig. 3). Also shown in Fig. 3, 0.3 mM PCMB revealed non-AT1, nonAT2 binding of 125I-Ang II not only in the hypothalamus, but also in the cerebral cortex as well (Bmax = 2.73 ± 0.29 fmol/mg wet weight and Kd = 4.24 ± 0.89 nM, n = 11). The cerebral cortex shows extremely low AT1 receptor binding; however, the nonAT1, non-AT2 binding of 125I-Ang II was similar in both regions. The AT1 antagonist 125I-SI-Ang II also showed high affinity for the non-AT1, non-AT2 binding site in the brain (Bmax = 2.92 ± 0.21 fmol/mg wet weight and Kd = 1.0 ± 0.16 nM in hypothalamus; and Bmax = 2.48 ± 0.13 fmol/mg wet weight and Kd = 0.5 ± 0.11 nM in cerebral cortex; n = 2) in the presence of 0.3 mM PCMB, and 10 μM losartan and PD123319. Addition of 0.3 mM PCMB to suspensions of liver and adrenal membranes containing 10 μM losartan and PD123319 did not cause expression of an 125I-Ang II binding site (Fig. 3), suggesting that the non-AT1, non-AT2 binding site may be specific to the brain. Moreover, it is important to note, that the binding of 125I-Ang II to AT1 receptors in the liver was substantially inhibited by 0.3 mM PCMB (Fig. 4).

BR A IN RE S E A RCH 1 1 43 ( 20 0 7 ) 8 3 –9 1

Fig. 2 – Representative time course of specific binding of I-Ang II (1nM) in the rat hypothalamus (HT) and cerebral cortex (Ctx). The membrane preparations contained 10 μM PD123319 and losartan, to exclude AT1 and AT2 receptor binding and 0.3 mM PCMB (15, 30, 45, 60, 90 and 120 min incubations at 24 °C. Values shown are mean ± SEM, n = 2).

125

Addition of varying concentrations of PCMB, ranging from 0.05 to 10 mM, to membrane suspensions of cerebral cortex containing 10 μM losartan and PD123319, indicated that the optimal concentration of PCMB for expression of the non-AT1, non-AT2 binding site in the rat brain is 0.3 to 1 mM (Fig. 5 and Table 1). Qualitatively similar, but less potent effects were produced by 3–10 mM p-chloromercuriphenylsulphonic acid (PCMPS) in the cerebral cortex (Bmax = 0.59 ± 0.08 fmol/mg wet weight, Kd = 3.29 ± 0.96 nM at 3 mM PCMPS; Bmax = 0.9 ± 0.08 fmol/mg wet weight, Kd = 3.69 ± 0.66 nM at 5 mM PCMPS; Bmax = 1.66 ± 0.280 fmol/mg wet weight, Kd = 4.38 ± 1.41 nM at 10 mM PCMPS; n = 2). In contrast, the disulfide reducing agents dithiothreitol (DTT) (0.3, 1, 3 and 5 mM) and beta-mercaptoethanol (β-MET)

Fig. 3 – Representative saturation binding analyses of I-Ang II binding in the rat hypothalamus (HT), cerebral cortex (Ctx), liver (Liv.) and adrenals (Adr.) in the presence of 10 μM PD123319 and losartan in the absence or presence (indicated with “+”) of 0.3 mM PCMB. In “HT+” Bmax = 1.97 ± 0.12 fmol/mg wet wt., and Kd = 3.27 ± 0.46 nM; in “Ctx+” Bmax = 1.6 ± 0.17 fmol/mg wet wt., and Kd = 3.6 ± 0.78 nM; in the rest of the cases Bmax and Kd were not significantly different from zero.

85

Fig. 4 – Representative saturation curves of 125I-Ang II binding in the rat liver in the absence (“Liver”) or presence (“Liver + PCMB”) of 0.3 mM PCMB. In “Liver” Bmax = 14.5 ± 0.52 fmol/mg wet wt. and Kd = 1.5 ± 0.1 nM; while in the presence of 0.3 mM PCMB the amount of AT1 binding (Bmax = 3.8 ± 1.6 fmol mg wet wt.) and the binding affinity (Kd = 5.8 ± 3.2 nM) are considerably decreased.

(5, 10, 20 and 30 mM), organomercurial sulfhydryl reagent mersalyl acid (0.3, 1, 3 and 5 mM), and sulfhydryl reagents Nethylmaleimide (NEM) and 5′,5′-dithiobis (2-nitrobenzoic acid) (DTNB) (0.3, 1 and 3 mM) did not activate the non-AT1, nonAT2 binding site in the rat brain. Moreover, 5 mM DTT and 30 mM β-MET inhibited the ability of PCMB to activate the nonAT1, non-AT2 binding site in the rat hypothalamus (Fig. 6). As shown in Fig. 7 and Table 2, the pharmacological specificity of the non-AT1, non-AT2 binding site in the hypothalamus and cerebral cortex demonstrated considerable specificity for angiotensin peptides. However, the competition profile was slightly different than that of the AT1 and AT2 receptors. Ang I was nearly as potent a competitor as Ang II, while Ang III was slightly more potent than Ang II. Ang IV (the 3–8 hexapeptide) showed dramatically lower affinity for the non-AT-1, non-AT-2 binding site, while the 4–8 pentapeptide of Ang II showed an even lower affinity for the binding site. Five non-angiotensin peptides, bradykinin, neurotensin,

125

Fig. 5 – Representative saturation curves of 125I-Ang II binding in the rat cerebral cortex in the presence of 10 μM PD123319 and losartan, and 0.05–10 mM PCMB. Average Bmax and Kd values are given in Table 1.

86

BR A IN RE S EA RCH 1 1 43 ( 20 0 7 ) 8 3 –91

Table 1 – Effect of varying concentrations of PCMB on 125I-Ang II binding in cerebral cortex PCMB

Kd (nM) Bmax

0.05 mM

0.1 mM

0.3 mM

1 mM

3 mM

5 mM

10 mM

1.54 ± 1.09 0.19 ± 0.05

6.38 ± 1.87 1.36 ± 0.25

4.26 ± 0.87 2.73±0.29

6.67 ± 2.24 3.92 ± 0.78

5.3 ± 1.28 2.5 ± 0.34

5.0 ± 1.29 1.13 ± 0.16

8.45 ± 9.9 1.1 ± 0.88

Summary of kinetic parameters of 125I-Ang II binding in the rat cerebral cortex in the presence of 10 μM PD123319 and losartan, and 0.05–10 mM PCMB (1 h incubation at 24 °C; n = 11 for 0.3 mM PCMB, and n = 3 for the other concentrations).

vasoactive intestinal polypeptide (VIP), gonadotropin releasing hormone (GnRH or LHRH) and substance P, showed low, micromolar affinities for the non-AT1, non-AT2 binding site. Ang (5–8) and Ang (1–7) at 10 μM had no effect on 125I-Ang II binding to the non-AT1, non-AT2 binding site, whereas Ang (1–4) significantly increased specific binding of 125I-Ang II by 25–30% both in hypothalamus and cerebral cortex (Fig. 8). High affinity, saturable binding of 125I-Ang II was observed in the mouse hypothalamus and cerebral cortex as well, but was not present in the liver using the same experimental conditions (Bmax = 1.43 ± 0.13 fmol/mg initial wet wt. and Kd = 2.79 ± 0.58 nM in hypothalamus; and Bmax = 1.94 ± 0.16 fmol/mg initial wet wt. and Kd = 3.45 ± 0.57 nM in cerebral cortex; n = 2).

3.

Discussion

of 125I-Ang II at 24 °C with a brain membrane preparation containing 10 μM PD123319 to block AT2 receptor binding, only 3.1 ± 0.8% (n = 3) remained intact. But, in the presence of 0.3 mM PCMB, the amount of intact 125I-Ang II was 38.2 ± 4.9% (n = 2). However, the initial supposition, that PCMB increased measurable binding of 125I-Ang II to AT1 receptors in the hypothalamus by protecting it from degradation, was quickly disproved when it was demonstrated that the enhanced specific binding of 125I-Ang II was retained in the presence of a saturating concentration of the AT1 receptor antagonist losartan (Fig. 3). Moreover, it was further shown that PCMB impairs the binding of 125I-Ang II to AT1 receptors in rat liver membranes (Fig. 4). PCMB also unmasked abundant high-affinity binding of 125 I-Ang II in rat cerebral cortical membranes in the presence of PD123319 and losartan indicating that its distribution does not follow that of brain Ang II receptors (de Gasparo et al.,

As noted in the Introduction, the initial goal of this study was to establish conditions to protect Ang II from metabolic degradation during receptor binding assays with brain membrane preparations using peptidase inhibitors. Interest in PCMB arose from its ability to inhibit “angiotensinase” activity, particularly aminopeptidase and prolyl endopeptidase activity (Fujioka et al., 1995; Kohara et al., 1991). The ability of PCMB to inhibit Ang II metabolism was confirmed. After 1 h incubation

Fig. 6 – Representative saturation curves of 125I-Ang II binding in the rat hypothalamus in the presence of 10 μM PD123319 and losartan, 0.3 mM PCMB with and without sulfhydryl reducing agents. In the absence of sulfhydryl reducing agent, Bmax = 1.57 ± 0.16 and Kd = 4.1 ± 0.82. In the presence of 30 mM ß-MET, Bmax = 0.11 ± 0.02 and Kd = 1.6 ± 0.72. In the presence of DTT, Bmax = 0.02 ± 0.008 and Kd = 0.44 ± 0.57.

Fig. 7 – Representative competition curves of for 125I-Ang II binding by angiotensin (a) and non-angiotensin (b) peptides in rat hypothalamus in the presence of 1 nM 125I-Ang II, 10 μM PD123319 and losartan, and 0.3 mM PCMB. Average values are presented in Table 2.

87

BR A IN RE S E A RCH 1 1 43 ( 20 0 7 ) 8 3 –9 1

Table 2 – Competition for 125I-Ang II binding by angiotensin and non-angiotensin peptides in the rat brain in the presence of PCMB Angiotensin peptides

Ki values in HT, n = 3 Ki values in Ctx, n = 3

Ang I

Ang II

Ang III

Ang IV

Ang (4–8)

1.0 × 10− 7 8.2 × 10− 8

7.5 × 10− 8 6.0 × 10− 8

4.2 × 10− 8 3.9 × 10− 8

3.3 × 10− 6 2.3 × 10− 6

1.67 × 10− 5 1.5 × 10− 5

LHRH

Substance P

Non-angiotensin peptides Bradykinin Ki values in HT, n = 3 Ki values in Ctx, n = 3

−5

2.84 × 10 3.45 × 10− 5

Neurotensin −5

1.19 × 10 2.24 × 10− 5

VIP −5

1.7 × 10 1.7 × 10− 5

−5

3.79 × 10 3.5×10− 5

6.2 × 10− 6 4.92 × 10− 6

Summary of competition binding studies (n = 3) of angiotensin and non-angiotensin peptides in rat hypothalamus (HT) and cerebral cortex (Ctx) in the presence of 1 nM 125I-Ang II, 10 μM PD123319 and losartan, and 0.3 mM PCMB (1 h incubation at 24 °C).

2000; Saavedra, 1992; Speth et al., 1991). The absence of such a binding site in liver and adrenal membranes, two tissues richly endowed in Ang II receptors, suggests that this nonAT1, non-AT2 binding site is not associated with Ang II receptors and may be brain specific (Fig. 3). It should also be mentioned, that the affinities of 125I-Ang II and 125I-SI-Ang II for the non-AT1, non-AT2 binding site are similar to their affinities for AT1 receptors in liver 1.8 ± 0.19 nM and Kd = 0.15 ± 0.01, respectively, n = 5). It was also established that disulfide-reducing agents DTT (0.3–5 mM) and β-MET (5–30 mM), the organomercurial sulfhydryl reagent mersalyl (0.3–5 mM) and sulfhydryl reagents DTNB and NEM (0.3–3 mM) had no ability to activate the non-AT1, non-AT2 binding site in the rat brain. Moreover, DTT (5 mM) and β-MET (30 mM) which can reverse the binding of organomercurials to the cysteine residues of proteins (Capobianco et al., 2004; Karlin and Bartels, 1966; Karniski, 1989) inhibited the binding site-revealing effect of PCMB in the rat brain (Fig. 6). The sulfonic acid derivative of PCMB, PCMPS, also showed an ability to unmask the non-AT1, non-AT2 binding site in the rat brain. However its potency was much less than PCMB. PCMB and PCMPS are organomercurial sulfhydryl reagents, that specifically react with free − SH groups of cysteine

Fig. 8 – Single concentration (10 μM) competition for 125I-Ang II binding by angiotensin fragments Ang (5–8), Ang (1–7) and Ang (1–4) in rat hypothalamus (HT) and cerebral cortex (Ctx) in the presence of 1 nM 125I-Ang II, 10 μM PD123319 and losartan, and 0.3 mM PCMB. *p = 0.01 (n = 2).

residues in proteins and peptides while remaining inert towards disulfide bonds in general (Capobianco et al., 2004; Han et al., 1987; Karlin and Bartels, 1966; Karniski, 1989; Olami et al., 1997). PCMB, however, differs in one important aspect: PCMB is a lipid-soluble weak acid (pK = 4), which can reach cysteine residues in proteins both within and along the margins of the lipid bilayer. PCMPS, because of the strongly acidic sulfonic acid group (pK = 1.5) is much less lipid-soluble and is largely membrane-impermeant. It mainly attacks the cysteines in proteins located on the external membrane surface (Karniski, 1989; Olami et al., 1997). This suggests that the critical cysteines needed for unmasking of this binding site are located within the lipid bilayer rather than on the domains located at the inner or outer surface faces of the protein. These results indicate that unmasking of the non-AT1, nonAT2 binding site requires a portion of the free −SH group(s) of cysteine residues to be blocked/conjugated by the organomercurial agent. However, if too many −SH groups of cysteine residues are blocked, e.g., at concentrations of PCMB in excess of 1 mM, there is a diminution in the number of binding sites. The most probable explanation for the unmasking of the non-AT1, non-AT2 binding site is that the blockage/conjugation of critical free − SH group(s) in the membrane spanning domain(s) of the protein causes a conformational change of the non-AT1, non-AT2 binding protein, facilitating its ability to bind angiotensin peptides. Conversely, this binding site may be an enzyme and these −SH groups may be critical for the catalytic activity of the enzyme to metabolize Ang II. Thus the unmasking of this binding site by PCMB may simply reflect the ability to block the metabolism, but not the binding of Ang II to this protein. With increased blockage/ conjugation of the cysteine residues of this protein at higher concentrations of PCMB, there may be a conformational change or steric hindrance that interferes with Ang II binding to the protein. The pharmacological specificity of the non-AT1, non-AT2 binding site, Ang III ≅ Ang II ≥ Ang I >> Ang IV >> Ang (4–8) (Fig. 7a and Table 2) reveals a strong preference for angiotensin peptides similar to, but not identical to that of the AT1 and AT2 subtypes. Ang (5–8) and Ang (1–7) at 10 μM had no effect on 125I-Ang II binding to the non-AT1, non-AT2 binding site, whereas Ang (1–4) increased specific binding of 125I-Ang II by 25–30% both in hypothalamus and cerebral cortex (Fig. 8). The

88

BR A IN RE S EA RCH 1 1 43 ( 20 0 7 ) 8 3 –91

ability of Ang (1–4) to increase the binding of 125I-Ang II to the non-AT1, non-AT2 binding sites was not investigated. Among non-angiotensin peptides tested (Fig. 7b and Table 2), substance P with a Ki = 6.2 μM in hypothalamus, and 4.92 μM in cerebral cortex (n = 3) showed the highest affinity for the binding site. Bradykinin, neurotensin, luteinizing hormonereleasing hormone (LHRH or GnRH) and vasoactive intestinal peptide (VIP) all had Ki values in excess of 10 μM, suggesting again that this non-AT1, non-AT2 binding site is specific for angiotensin peptides. However, it remains to be determined if there are other neuropeptides that might display affinities in the range of the angiotensins. A similar phenomenon of unmasking of an Ang II binding site by the organomercurial sulfhydryl reagents, PCMB and PCMPS was described by two independent groups who characterized a soluble angiotensin-binding protein in rabbit and porcine liver (Hagiwara et al., 1989; Kiron and Soffer, 1989). Later it was established that this soluble Ang II binding protein is endopeptidase 24.15 (EC 3.4.24.15, EP24.15, thimet oligopeptidase, TOP, Pz-peptidase, endo-oligopeptidase A) (McKie et al., 1993). Subsequently, a closely related peptidase, endopeptidase 24.16 (EC 3.4.24.16, neurolysin) (Kato et al., 1994, 1997), was also reported to have the characteristics of the soluble Ang II binding protein. Both proteins are members of the M3 family of Zn-dependent metalloendopeptidases ubiquitously distributed in the central nervous system and in peripheral organs of mammals (Ferro et al., 2004; Shrimpton et al., 2002; Tisljar, 1993). The highest EP 24.15 activity is in brain, testis and in the anterior and posterior lobes of the pituitary, while the activities in spleen, liver, kidney, lung, adrenals and thyroid are only 10–20% of that found in brain. EP 24.15 is present in both glial and neuronal cells, having primarily nuclear localization. It degrades several endogenous peptides, including Ang I and Ang II (cleavage of Pro7-Phe8 and Tyr4-Ile5 bonds, respectively) (Chu and Orlowski, 1985; Healy and Orlowski, 1992). In the rat brain 20–30% of EP 24.15 activity is associated with membranes, including plasma membranes, endosomes and synaptic vesicles (Acker et al., 1987; Dahms and Mentlein, 1992; Fontenele-Neto et al., 2001; Molineaux et al., 1991). In rat liver, 90% of EP 24.15 activity is in the soluble fraction of the liver homogenate, derived from the cytoplasm of hepatocytes (Gioli-Pereira et al., 2003). The highest EP 24.16 activity in the rat is in the liver and kidney, whereas brain contains less than 30% of the activity of the liver (Checler et al., 1989; Dauch et al., 1992). EP 24.16 occurs in both membrane-associated and cytosolic forms in all of these tissues (Checler et al., 1983, 1986; Kato et al., 1997; Rioli et al., 1998; Serizawa et al., 1995). In the rat brain, EP 24.16 occurs in selective subpopulations of glia and neurons, with a mainly extranuclear localization. In the brain, EP 24.15 is exclusively bound to the cytoplasmic side of membranes, whereas EP 24.16 is found on both sides of the membranes (Fontenele-Neto et al., 2001; Massarelli et al., 1999; Vincent et al., 1996; Woulfe et al., 1992). Both Kiron and Soffer (1989) and Hagiwara et al. (1989) reported that the presence of PCMB or PCMPS, along with EDTA, was an obligatory condition for binding of 125I-Ang II to the soluble protein, which could be impaired by DTT or β-MET. The soluble angiotensin-binding protein isolated from porcine

liver cytosolic fraction showed maximum Ang II binding activity in the presence of 0.1–1 mM PCMPS, and at higher concentrations the degree of activation declined (Hagiwara et al., 1989). In the rat brain membranes, unmasking of the nonAT1, non-AT2 binding site was also observed in the presence of small range of the organomercurial agent with inhibition at higher concentrations. Kiron and Soffer (1989) and Hagiwara et al. (1989) showed that among angiotensin peptides Ang III has the highest affinity followed by Ang II and Ang I. In our case, the affinity of angiotensin peptides for the brain non-AT1, non-AT2 binding site was also Ang III ≅ Ang II > Ang I >> Ang IV >> Ang (4–8) (Table 2). These observations are consistent with, the non-AT1, non-AT2 binding site in the rat brain being like EP 24.15 and/or EP 24.16, the soluble angiotensin binding protein isolated from liver, but in a membrane-bound state. Despite the similarities between the liver soluble angiotensin binding protein identified as EC24.15 and/or 24.16, and the membrane-bound non-AT1, non-AT2 binding site in brain, there are significant differences. Hagiwara et al. (1989) reported that the period of exposure to PCMPS was a critical factor for binding 125I-Ang II; PCMPS had to be added simultaneously with 125I-Ang II to have binding. If the addition of 125I-Ang II was delayed, then the binding of 125I-Ang II was decreased. In the rat brain membranes, binding of 125I-Ang II to the non-AT1, non-AT2 binding site did not depend on time of exposure to the organomercurials. The non-AT1, non-AT2 binding site in brain is not likely to be EP 24.16 because EP 24.16 occurs as a membrane bound protein in liver (Checler et al., 1983, 1986; Kato et al., 1997; Rioli et al., 1998; Serizawa et al., 1995) yet no non-AT1, non-AT2 binding was observed in liver membranes in this study. The binding affinities of non-angiotensin peptides for brain non-AT1, non-AT2 angiotensin binding site differ considerably from the peptidases EP 24.15 and EP 24.16. Ang I, Ang II, neurotensin, bradykinin, substance P and LHRH are cleaved by rat and recombinant EP 24.15 and EP 24.16 at similar rates (Dahms and Mentlein, 1992; Dando et al., 1993; Rioli et al., 1998). The angiotensin and non-angiotensin peptides show similar micromolar Ki/Km values for the soluble fraction of EP24.15 in rat brain (Chu and Orlowski, 1985). However, the affinity of angiotensin III, II and I for the non-AT1, non-AT2 binding site is ∼ 100 times higher than that of neurotensin, bradykinin, substance P and LHRH (Table 2). In summary, this report describes a high-affinity, nonAT1, non-AT2 angiotensin binding site in rat brain membranes that is unmasked by specific organomercurial sulfhydryl reagents; PCMB and PCMPS, but not by other sulfhydryl reagents mersalyl, NEM, DTNB or the disulfide-reducing agents DTT and β-MET. The requirement for organomercurial sulfhydryl reagents to unmask the brain non-AT1, nonAT2 angiotensin binding site raises the question of whether artificial conditions might be creating an artifactual binding site. However, the possibility of activation of the brain nonAT1, non-AT2 angiotensin binding site by endogenous factor (s) or conditions which mimic the environment created by the presence of PCMB and PCMPS, cannot be ruled out. This concept is supported by the report of Shrimpton et al. (2003) identifying two components of cerebrospinal fluid that activate recombinant EP 24.15 by affecting its multimerization state.

BR A IN RE S E A RCH 1 1 43 ( 20 0 7 ) 8 3 –9 1

Regarding the possible functionality of the non-AT1, nonAT2 binding site: its role may be different from that of the other Ang II receptors. The binding site could be a clearance receptor akin to the “silent receptors of atrial natriuretic factor” (Anand-Srivastava and Trachte, 1993; Maack et al., 1987) which may not be so silent (Trachte, 2005). In view of the uncertainties regarding the sites of synthesis of Ang II in the brain and questions as to its ability to act as a neurotransmitter, such a clearance receptor might play a critical role either to remove Ang II from the extracellular milieu of the brain, or to promote its recycling as a neurotransmitter by neurons capable of taking up Ang II via such a clearance receptor. It may also mediate responses to angiotensins that until now have not been recognized, e.g., subtle alterations in mood, motor activity, pain transmission or arousal states. Preliminary studies of the distribution of the non-AT1, nonAT2 binding site in the brainstem suggest that it binds to the ventral and lateral tegmentum, with little resemblance to the distribution of brainstem AT1 receptors. It is also possible that PCMB, which is a known inhibitor of EP 24.15 (McKie et al., 1993; Ray et al., 2004) along with EDTA could be conformationally altering a membrane-bound variant of EP 24.15 in a state capable of binding angiotensin peptides with high affinity, but unable to enzymatically cleave the peptides. As such the non-AT1, non-AT2 binding site may be a key enzyme responsible for metabolic degradation of angiotensin peptides in brain, perhaps even more important than aminopeptidases, and thus be a primary regulator of their physiological effects. The fact that the distribution of the non-AT1, non-AT2 binding site does not parallel the distribution of Ang II receptors in the brain, suggests that if it is an enzyme, it may have other peptide transmitter or hormone substrates. While this non-AT1, non-AT2 binding site was not observed in the liver and adrenal, a more detailed study of distribution of the binding site in other organs and in specific brain nuclei, as well as studies towards identification of the protein representing the non-AT1, non-AT2 binding site should be carried out.

4.

Experimental procedure

Adult Sprague–Dawley rats (Harlan) 300–350 g maintained in 12 h light/dark cycle, fed ad libitum were used for the study. The protocol for these studies was approved by the University of Mississippi IACUC. Angiotensin and non-angiotensin peptides were obtained from Bachem, Phoenix Pharmaceuticals or American Peptides. Losartan was a gift of Dr. Ron Smith of Dupont Merck. pChloromercuribenzoic acid (PCMB) sodium salt was purchased from MP Biomedicals. p-Chloromercuriphenylsulphonic acid (PCMPS) was purchased from Toronto Research Chemicals. PD123319, β-mercaptoethanol (β-MET), N-ethylmaleimide (NEM), 5,5′-dithiobis (2-nitrobenzoic acid) (DTNB) and mersalyl acid were purchased from Sigma Chemical Co. Dithiothreitol (DTT) was purchased from Invitrogen. Unless otherwise stated, losartan, PD123319, PCMB or other chemicals were added into the tissue membrane homogenate 10–15 min before incubation.

89

The measurement of 125I-Ang II binding in rat brain membranes was carried out using established procedures (Speth, 2003; Speth et al., 2005). Briefly, rat liver, adrenals and brain were excised upon decapitation. Hypothalamic tissue and cerebral cortex were quickly dissected from the brain on ice. Tissues were weighed and immediately homogenized in ice-cold hypotonic, 20 mM NaPO4, 5 mM EDTA, pH 7.2 solution by a mechanical homogenizer (Tekmar Tissuemizer). The homogenates were centrifuged (48,000×g for 20 min at 4 °C) and the supernatants decanted. The membrane pellets were resuspended by homogenization in assay buffer (150 mM NaCl, 5 mM EDTA, 0.1 mM bacitracin, 50 mM NaPO4, pH 7.1– 7.2). The homogenates were recentrifuged as before and the pellets resuspended by homogenization in the assay buffer (50 mg/ml initial wet tissue weight for hypothalamus and cortex, and 20 mg/ml for liver and adrenals). When present in the homogenates, PCMB was derived from a 100 or 200 mM stock solution in 50 mM NaOH. PD123319 and losartan were added as 1/1000 dilutions from 10 mM stocks in water. To measure Ang II receptor and non-receptor binding, membrane preparations were incubated with 125I-Ang II or 125 I-Sar1, Ile8 Ang II (125I-SI Ang II) prepared by the chloramine T procedure (Hunter and Greenwood, 1962). The mono-radioiodinated angiotensin was purified by HPLC using a C18 reverse phase column (Varian, Microsorb-MV 5–100 C18, 250 × 4.6 mm), eluted isocratically with triethylamine phosphate (pH 3.0)/ acetonitrile mobile phase (Speth and Harding, 2001). Saturation binding studies were carried out by incubation of 50 μl membrane preparations with six concentrations of 125 I-Ang II (0.3–6 nM) or 125I-SI-Ang II (0.25–3 nM) in 100 μl total assay volume for 1 or 2 h at room temperature, respectively. The 1 h incubation for 125I-Ang II is based on a time course analysis indicating that maximal specific binding occurred at 45–60 min, after which it decreased. Non-specific binding was estimated in the presence of 3 μM SI-Ang II or Ang II for 125I-Ang II and 125I-SI-Ang II, respectively. Values shown represent specific (total minus non-specific) binding. Competition binding studies were conducted by incubation of 50 μl of membrane preparation with 1 nM 125I-Ang II in the presence of four different concentrations of angiotensin and non-angiotensin peptides (1 nM to 10 μM) in 100 μl total assay volume for 1 h at room temperature. Non-specific binding was determined in the presence of 3 μM SI-Ang II or 3 μM Ang II. Specific binding is presented as the difference between total and non-specific binding. Free and bound radioligands were separated on glass fiber filters (Whatman, Schleicher and Schuell, #32 glass), prewetted with 1 mg/ml bovine serum albumin, using a cell harvester (Model M24R, Brandel, Gaithersberg, MD). The bound radioligand retained on the filter disks was assayed with a Beckman Gamma 5500 gamma counter at a counting efficiency of 65%. Determination of Bmax (fmol of radioligand bound per mg initial wet weight), Kd and IC50 values were carried out using one-site saturation and competition binding models of Prism software (Graphpad Software, San Diego, CA). Ki values were determined using the Cheng-Prusoff equation: Ki = IC50 / (1 + H/Kd) where H is the radioligand concentration and Kd is the Kd for the radioligand. Values reported are mean ± SEM.

90

BR A IN RE S EA RCH 1 1 43 ( 20 0 7 ) 8 3 –91

Acknowledgments This work was supported by an American Heart Association Grant in Aid: 0350481Z and the Peptide Radioiodination Service Center of the University of Mississippi. Losartan was a gift of Dr. Ron Smith, Dupont-Merck, Whitehouse, NJ.

REFERENCES

Acker, G.R., Molineaux, C., Orlowski, M., 1987. Synaptosomal membrane-bound form of endopeptidase-24.15 generates Leu-enkephalin from dynorphin1–8, alpha- and beta-neoendorphin, and Met-enkephalin from Met-enkephalin-Arg6-Gly7-Leu8. J. Neurochem. 48, 284–292. Anand-Srivastava, M.B., Trachte, G.J., 1993. Atrial natriuretic factor receptors and signal transduction mechanisms. Pharmacol. Rev. 45, 455–497. Capobianco, L., Impagnatiello, T., Ferramosca, A., Zara, V., 2004. The mitochondrial tricarboxylate carrier of silver eel: chemical modification by sulfhydryl reagents. J. Biochem. Mol. Biol. 37, 515–521. Checler, F., Vincent, J.P., Kitabgi, P., 1983. Degradation of neurotensin by rat brain synaptic membranes: involvement of a thermolysin-like metalloendopeptidase (enkephalinase), angiotensin-converting enzyme, and other unidentified peptidases. J. Neurochem. 41, 375–384. Checler, F., Vincent, J.P., Kitabgi, P., 1986. Purification and characterization of a novel neurotensin-degrading peptidase from rat brain synaptic membranes. J. Biol. Chem. 261, 11274–11281. Checler, F., Barelli, H., Vincent, J.P., 1989. Tissue distribution of a novel neurotensin-degrading metallopeptidase. An immunological approach using monospecific polyclonal antibodies. Biochem. J. 257, 549–554. Chu, T.G., Orlowski, M., 1985. Soluble metalloendopeptidase from rat brain: action on enkephalin-containing peptides and other bioactive peptides. Endocrinology 116, 1418–1425. Dahms, P., Mentlein, R., 1992. Purification of the main somatostatin-degrading proteases from rat and pig brains, their action on other neuropeptides, and their identification as endopeptidases 24.15 and 24.16. Eur. J. Biochem. 208, 145–154. Dando, P.M., Brown, M.A., Barrett, A.J., 1993. Human thimet oligopeptidase. Biochem. J. 294 (Pt2), 451–457. Dauch, P., Masuo, Y., Vincent, J.P., Checler, F., 1992. Endopeptidase 24-16 in murines: tissue distribution, cerebral regionalization, and ontogeny. J. Neurochem. 59, 1862–1867. de Gasparo, M., Catt, K.J., Inagami, T., Wright, J.W., Unger, T., 2000. International union of pharmacology: XXIII. The angiotensin II receptors. Pharmacol. Rev. 52, 415–472. Ferro, E.S., Carreno, F.R., Goni, C., Garrido, P.A., Guimaraes, A.O., Castro, L.M., Oliveira, V., Araujo, M.C., Rioli, V., Gomes, M.D., Fontenele-Neto, J.D., Hyslop, S., 2004. The intracellular distribution and secretion of endopeptidases 24.15 (EC 3.4.24.15) and 24.16 (EC 3.4.24.16). Protein Pept. Lett. 11, 415–421. Fitzsimons, J.T., 1998. Angiotensin, thirst, and sodium appetite. Physiol. Rev. 78, 583–686. Fontenele-Neto, J.D., Massarelli, E.E., Gurgel Garrido, P.A., Beaudet, A., Ferro, E.S., 2001. Comparative fine structural distribution of endopeptidase 24.15 (EC3.4.24.15) and 24.16 (EC3.4.24.16) in rat brain. J. Comp. Neurol. 438, 399–410. Fujioka, H., Okabe, T., Yamaguchi, H., 1995. Purification and characterization of angiotensin II degradation factor from porcine endothelial cells. Tohoku J. Exp. Med. 177, 183–192.

Gioli-Pereira, L., Fontenele-Neto, J.D., Borges, D.R., Kouyoumdjian, M., 2003. Localization of EP24.15, a major liver kininase. J. Histochem. Cytochem. 51, 125–127. Hagiwara, H., Sugiura, N., Wakita, K., Hirose, S., 1989. Purification and characterization of angiotensin-binding protein from porcine liver cytosolic fraction. Eur. J. Biochem. 185, 405–410. Han, K.K., Delacourte, A., Hemon, B., 1987. Chemical modification of thiol group(s) in protein: application to the study of anti-microtubular drugs binding. Comp. Biochem. Physiol., B 88, 1057–1065. Healy, D.P., Orlowski, M., 1992. Immunocytochemical localization of endopeptidase 24.15 in rat brain. Brain Res. 571, 121–128. Hunter, W.M., Greenwood, F.C., 1962. Preparation of iodine-131 labelled human growth hormone of high specific activity. Nature 194, 495–496. Karlin, A., Bartels, E., 1966. Effects of blocking sulfhydryl groups and of reducing disulfide bonds on the acetylcholine-activated permeability system of the electroplax. Biochim. Biophys. Acta 126, 525–535. Karniski, L.P., 1989. Activation of Cl−/OH− exchange by parachloromercuribenzoic acid in rabbit renal brush-border membranes. J. Membr. Biol. 112, 59–66. Kaschina, E., Unger, T., 2003. Angiotensin AT1/AT2 receptors: regulation, signalling and function. Blood Pressure 12, 70–88. Kato, A., Sugiura, N., Hagiwara, H., Hirose, S., 1994. Cloning, amino acid sequence and tissue distribution of porcine thimet oligopeptidase. A comparison with soluble angiotensin-binding protein. Eur. J. Biochem. 221, 159–165. Kato, A., Sugiura, N., Saruta, Y., Hosoiri, T., Yasue, H., Hirose, S., 1997. Targeting of endopeptidase 24.16 to different subcellular compartments by alternative promoter usage. J. Biol. Chem. 272, 15313–15322. Kiron, M.A., Soffer, R.L., 1989. Purification and properties of a soluble angiotensin II-binding protein from rabbit liver. J. Biol. Chem. 264, 4138–4142. Kohara, K., Tabuchi, Y., Senanayake, P., Brosnihan, K.B., Ferrario, C.M., 1991. Reassessment of plasma angiotensins measurement: effects of protease inhibitors and sample handling procedures. Peptides 12, 1135–1141. Maack, T., Suzuki, M., Almeida, F.A., Nussenzveig, D., Scarborough, R.M., McEnroe, G.A., Lewicki, J.A., 1987. Physiological role of silent receptors of atrial natriuretic factor. Science 238, 675–678. Massarelli, E.E., Casatti, C.A., Kato, A., Camargo, A.C., Bauer, J.A., Glucksman, M.J., Roberts, J.L., Hirose, S., Ferro, E.S., 1999. Differential subcellular distribution of neurolysin (EC 3.4.24.16) and thimet oligopeptidase (EC 3.4.24.15) in the rat brain. Brain Res. 851, 261–265. McKie, N., Dando, P.M., Rawlings, N.D., Barrett, A.J., 1993. Thimet oligopeptidase: similarity to ‘soluble angiotensin II-binding protein’ and some corrections to the published amino acid sequence of the rat testis enzyme. Biochem. J. 295 (Pt1), 57–60. Molineaux, C.J., Yu, B., Ayala, J.M., 1991. Distribution of endopeptidase-24.15 in rat brain nuclei using a novel fluorogenic substrate: comparison with endopeptidase-24.11. Neuropeptides 18, 49–54. Olami, Y., Rimon, A., Gerchman, Y., Rothman, A., Padan, E., 1997. Histidine 225, a residue of the NhaA-Na+/H+ antiporter of Escherichia coli is exposed and faces the cell exterior. J. Biol. Chem. 272, 1761–1768. Ray, K., Hines, C.S., Coll-Rodriguez, J., Rodgers, D.W., 2004. Crystal structure of human thimet oligopeptidase provides insight into substrate recognition, regulation, and localization. J. Biol. Chem. 279, 20480–20489. Rioli, V., Kato, A., Portaro, F.C., Cury, G.K., te, K.K., Vincent, B., Checler, F., Camargo, A.C., Glucksman, M.J., Roberts, J.L., Hirose, S., Ferro, E.S., 1998. Neuropeptide specificity and inhibition of recombinant isoforms of the endopeptidase

BR A IN RE S E A RCH 1 1 43 ( 20 0 7 ) 8 3 –9 1

3.4.24.16 family: comparison with the related recombinant endopeptidase 3.4.24.15. Biochem. Biophys. Res. Commun. 250, 5–11. Saavedra, J.M., 1992. Brain and pituitary angiotensin. Endocr. Rev. 13, 329–380. Saavedra, J.M., 2005. Brain angiotensin II: new developments, unanswered questions and therapeutic opportunities. Cell. Mol. Neurobiol. 25, 485–512. Serizawa, A., Dando, P.M., Barrett, A.J., 1995. Characterization of a mitochondrial metallopeptidase reveals neurolysin as a homologue of thimet oligopeptidase. J. Biol. Chem. 270, 2092–2098. Severs, W.B., Daniels-Severs, A.E., 1973. Effects of angiotensin on the central nervous system. Pharmacol. Rev. 25, 415–449. Shrimpton, C.N., Smith, A.I., Lew, R.A., 2002. Soluble metalloendopeptidases and neuroendocrine signaling. Endocr. Rev. 23, 647–664. Shrimpton, C.N., Wolfson, A.J., Smith, A.I., Lew, R.A., 2003. Regulators of the neuropeptide-degrading enzyme, EC 3.4.24.15 (thimet oligopeptidase), in cerebrospinal fluid. J. Neurosci. Res. 74, 474–478. Speth, R.C., 2003. Sarcosine1,glycine8 angiotensin II is an AT1 angiotensin II receptor subtype selective antagonist. Regul. Pept. 115, 203–209. Speth, R.C., Harding, J.W., 2001. Radiolabeling of angiotensin peptides. In: Wang, D.H. (Ed.), Angiotensin Protocols. Humana Press, Totowa, NJ, pp. 275–295.

91

Speth, R.C., Rowe, B.P., Grove, K.L., Carter, M.R., Saylor, D., 1991. Sulfhydryl reducing agents distinguish two subtypes of angiotensin II receptors in the rat brain. Brain Res. 548, 1–8. Speth, R.C., Brown, T.E., Barnes, R.D., Wright, J.W., 2003. Brain angiotensinergic activity: the state of our current knowledge. Proc. West. Pharmacol. Soc. 46, 11–15. Speth, R.C., Kim, K.H., Elton, T.S., Simasko, S., 2005. Sarcosine1, glycine8 angiotensin II is a functional AT1 angiotensin receptor antagonist. Endocrine 26, 83–87. Tisljar, U., 1993. Thimet oligopeptidase-a review of a thiol dependent metallo-endopeptidase also known as Pz-peptidase endopeptidase 24.15 and endo-oligopeptidase. Biol. Chem. Hoppe-Seyler 374, 91–100. Trachte, G., 2005. Neuronal regulation and function of natriuretic peptide receptor C. Peptides 26, 1060–1067. Unger, T., Badoer, E., Ganten, D., Lang, R.E., Rettig, R., 1988. Brain angiotensin: pathways and pharmacology. Circulation 77, I40–I54. Vincent, B., Beaudet, A., Dauch, P., Vincent, J.P., Checler, F., 1996. Distinct properties of neuronal and astrocytic endopeptidase 3.4.24.16: a study on differentiation, subcellular distribution, and secretion processes. J. Neurosci. 16, 5049–5059. Woulfe, J., Checler, F., Beaudet, A., 1992. Light and electron microscopic localization of the neutral metalloendopeptidase EC 3.4.24.16 in the mesencephalon of the rat. Eur. J. Neurosci. 4, 1309–1319.