Dominant Negative AT2 Receptor Oligomers Induce G-protein Arrest and Symptoms of Neurodegeneration*

Neurodegeneration in Alzheimer's disease (AD) correlates with dysfunction of signaling mediated by Gαq/11. Nondissociable angiotensin II AT2 receptor oligomers are linked to the impaired Gαq/11-stimulated signaling of AD patients and transgenic mice with AD-like symptoms. To further analyze the role of AT2 receptor oligomers, we induced the formation of AT2 oligomers in an in vitro cell system. Similarly as in vivo, sequential oxidative and transglutaminase-dependent cross-linking steps triggered the formation of AT2 oligomers in vitro. Elevated reactive oxygen species mediated oxidative cross-linking of AT2 monomers to dimers involving tyrosine residues located at putative interreceptor contact sites of the cytoplasmic loop connecting transmembrane helices III/IV. Cross-linked AT2 dimers were subsequently a substrate of activated transglutaminase-2, which targeted the carboxyl terminus of AT2 dimers, as assessed by truncated and chimeric AT2 receptors, respectively. AT2 oligomers acted as dominant negative receptors in vitro by mediating Gαq/11 protein sequestration and Gαq/11 protein arrest. The formation of AT2 oligomers and G-protein dysfunction could be suppressed in vitro and in vivo by an AT2 receptor mutant. Inhibition of AT2 oligomerization upon stereotactic expression of the AT2 receptor mutant revealed that Gαq/11-sequestering AT2 oligomers enhanced the development of neurodegenerative symptoms in the hippocampus of transgenic mice with AD-like pathology. Thus, AT2 oligomers inducing Gαq/11 arrest are causally involved in inducing symptoms of neurodegeneration.

Dysfunction of G␣ q/11 is a characteristic feature of AD 2 patients and transgenic mice with AD-like symptoms (see the accompanying article (26) and Refs. [1][2][3][4][5]. The G␣ q/11 -protein defect is a well established hallmark of clinical AD (5). However, the pathophysiological role of the G-protein defect is barely understood. A direct relationship between G␣ q/11 dysfunction and the pathogenesis of AD leading to neurodegeneration and dementia is suggested by the important role of G␣ q/11 proteins in neuronal survival and memory (6,7). In addition, many of the cognition-enhancing effects of acetylcholine are mediated by the G␣ q/11 -coupled muscarinic receptors (8). The M 1 receptor is a major target of G␣ q/11 dysfunction of AD patients and mice, and impaired G␣ q/11 -coupling of M 1 receptors correlates with disease severity (5).
To better understand the pathophysiological function of the G-protein defect in AD, we analyzed the mechanism accounting for the G-protein defect in AD. In the first article (26), we identified cross-linked AT 2 receptor oligomers in the brains of AD patients and transgenic mice with AD-like symptoms. Coenrichment studies showed that AT 2 receptor oligomers resembled dominant negative receptors by sequestering G␣ q/11 in the absence of agonist. The cross-linked AT 2 receptors were directly linked to the progression of AD, because (aggregated) amyloid ␤ (A␤) triggered AT 2 oligomerization in vivo in a dosedependent manner by a two-step process of oxidative and transglutaminase-dependent cross-linking. Down-regulation of AT 2 receptors by RNA interference revealed that AT 2 receptor (oligomers) contributed to G␣ q/11 dysfunction of transgenic mice with AD-like symptoms. Specifically, AT 2 receptor oligomers impaired signaling of G␣ q/11 -coupled muscarinic M 1 receptors in the hippocampus.
Because activation of hippocampal G␣ q/11 -stimulated signaling retards the development of AD-like symptoms in transgenic mice (9), we thought that AT 2 receptor oligomers were directly involved in symptoms of neurodegeneration. To address this question, we used an AT 2 receptor mutant to suppress AT 2 oligomerization in vitro and in vivo. We show here that inhibition of AT 2 oligomerization by stereotactic expression of a mutated AT 2 receptor retarded the development of neurodegenerative symptoms in the hippocampus of transgenic mice with AD-like pathology. These observations provide strong evidence for an involvement of AT 2 oligomers in mediating symptoms of neurodegeneration.

EXPERIMENTAL PROCEDURES
Cell Culture, Protein Expression, and Functional Assays-Cultivation of human embryonic kidney (HEK) cells and transfection with plasmids encoding AT 2 , AT 2 ⌬CTer , M 1 , G␣ 11 Q209L , or transglutaminase-2 under control of the cytomegalovirus promoter was performed as described (10,11). Reactive oxygen * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1  species (ROS) was induced by expression of NOX-3. Changes in total inositol phosphates and measurement of [Ca 2ϩ ] i were determined with intact cells (11,12). Basal and stimulated binding of [ 35 S]GTP␥S (specific activity 1250 Ci/mmol; final concentration 0.5 nM) to G␣ q/11 was determined in the presence of 0.1 M GDP in a volume of 200 l in triplicates with membranes (25 g of protein/point) prepared from cells and mouse hippocampal tissue followed by immunoaffinity enrichment of G␣ q/11 . The method measures specifically the activation of G␣ q/11 , because the applied antibodies cross-react specifically with G␣ q/11 as determined in immunoblot. Transglutaminase activity was determined by a [ 3 H]putrescine incorporation assay (11). Immunoaffinity enrichment of G␣ q/11 followed by immunoblot detection of co-enriched AT 2 was performed by a method that was described (12). Quantification of the carbachol-stimulated M 1 -G␣ q interaction in the presence of increasing levels of oligomeric AT 2 (0 -300 fmol/mg protein) was performed after immunoaffinity enrichment of M 1 receptors from M 1 receptor-expressing HEK cells (ϳ900 fmol/mg protein) co-expressing oligomeric AT 2 as indicated. Briefly, HEK cell membranes (50 g/point) in GTPfree buffer supplemented with 2 mM MgCl 2 were incubated in the absence or presence of 1 mM carbachol for 60 min at 12°C followed by 20 min at 24°C and the addition of 1 mM DSP. Membranes were washed three times with 20 mM Tris (pH 7.5) supplemented with 1 mM EDTA and protease inhibitors, collected by centrifugation, and solubilized in 200 l of buffer (1% deoxycholate, 0.1% Nonidet P-40, 0.1% SDS in 20 mM Tris, pH 7.4, supplemented with protease inhibitors). The solubilisate was incubated for 2 h at 4°C with an immunoaffinity matrix of M 1 -specific antibodies coupled to Affi-Gel 10. After washing, G␣ q co-enriched with the M 1 receptor was quantified by the addition of 125 I-labeled F(ab) 2 fragments of immunoselected anti-G␣ q antibodies (1 ϫ 10 Ϫ8 M, specific activity ϳ0.02 Ci/mg) to the affinity matrix. After 2 h of incubation at 4°C, the affinity matrix was washed three times with ice cold buffer, and radioactivity was determined in a ␥-counter. To determine the carbachol-stimulated M 1 -G␣ q interaction, bound radioactivity in the absence of carbachol (Ͻ10%) was subtracted. In a parallel approach, G␣ q co-enriched with the M 1 receptor was quantified in immunoblot with G␣ q -specific antibodies after elution from the affinity matrix and SDS-PAGE under reducing conditions. Activities of ␣-secretase present in hippocampal tissue of 13-month-old stressed APP Sw mice were determined using a commercially available kit (R&D Systems).
Antibodies for Immunoblotting and Immunohistochemistry-The following antibodies were used for immunoblotting, affinity purification, and immunohistochemistry (10 -12): affinitypurified rabbit/rat polyclonal anti-AT 2 antibodies (raised against an antigen encompassing amino acids 320 -349 of the human AT 2 receptor); affinity-purified rabbit polyclonal anti-AT 2 antibodies (raised against an antigen encompassing amino acids 326 -349 of the human AT 2 receptor); affinity-purified rabbit polyclonal anti-AT 2 antibodies (raised against an antigen encompassing amino acids 16 -35 of the human or mouse AT 2 receptor); affinity-purified rabbit/rat polyclonal anti-M 1 antibodies (raised against an antigen encompassing amino acids 231-350 of the human M 1 receptor); affinity-purified rabbit polyclonal anti-transglutaminase antibodies (raised against an antigen encompassing amino acids 1-20 of mouse transglutaminase-2); affinity-purified rabbit polyclonal anti-G␣ q/11 antibodies (raised against the COOH terminus of G␣ q/11 ). Immunoblotting and immunohistochemistry were routinely used to determine and confirm cross-reactivity of the antibodies with the respective proteins (10 -12).
Protein Detection in Immunoblot-Membranes were prepared by sucrose density gradient centrifugation at 4°C followed by partial enrichment (see accompanying manuscript (26)). For immunoblotting analysis, protein samples were separated by SDS-PAGE under reducing conditions and supplemented with urea followed by transfer to polyvinylidene difluoride membranes, as described (11). Affinity-purified antibodies or F(ab) 2 fragments of the respective antibodies preabsorbed to human and mouse proteins, respectively, were used for detection of AT 2 , AT 1 receptors, and G␣ q/11 . Applied antibodies were characterized in previous studies (accompanying manuscript (26) and Refs. 10 -13). Bound antibody was visualized by preabsorbed F(ab) 2 fragments of enzyme-coupled secondary antibodies or by enzyme-coupled Protein A followed by enhanced chemiluminescence detection (ECL plus).
Fluorescence-activated Cell Sorting Analysis-Cell surface M 1 receptors of transfected HEK cells were determined by fluorescence-activated cell sorting analysis using affinity-purified M 1 -specific antibodies raised against an epitope corresponding to the receptor's amino-terminal region according to standard methods (11). Transgenic Animals-Transgenic mice used in this study express human APP695 with the double mutation (K670N/ M671L; APP Sw ) that was identified in a Swedish family with early onset Alzheimer disease (14).
Lentiviral Vector Production-Vector plasmids were constructed for the production of third generation lentiviruses expressing wild-type AT 2 and AT 2 ⌬CTer under control of the cytomegalovirus promoter. The generation of viral particles pseudotyped with the vesicular stomatitis virus G glycoprotein was described in the accompanying article (26). For stereotactic injection in mice, high titer lentiviral stocks were used (Ͼ1 ϫ 10 7 transduction units/l).
Immunohistochemistry-For immunohistochemistry, paraffin-embedded sections (8 m, taken at 50-m intervals for analyses, 10 -15 sections/set) were deparaffinized, followed by antigen retrieval (11). Immunohistochemical staining of AT 2 receptor was performed with F(ab) 2 fragments of preabsorbed affinity-purified polyclonal antibodies (10,11). Immunohistochemistry with antibodies to MAP2 (microtubule-associated protein 2) was used as a marker of neuronal cell bodies and dendrites (anti-MAP2 antibodies; Sigma). DNA strand breaks were determined in situ, applying TUNEL technology (Roche Applied Science). All sections were imaged with a Leica DMI6000 microscope equipped with a DFC420 camera.
Cerebral Injection of Lentiviruses-To inhibit the formation of cross-linked AT 2 oligomers, lentivirus preparations encoding AT 2 ⌬CTer were injected bilaterally into the CA1 area of 12-month-old APP Sw mice similarly as described in the accompanying article (26) (anteroposterior Ϫ1.6 to Ϫ2.3; lateral 1; dorsoventral Ϫ1.8). The control group received an injection of a control lentivirus encoding ␤-galactosidase or wild-type AT 2 receptor. After 1 week of recovery, mice were subjected to the stress paradigm for 4 weeks. Five weeks after the lentiviral injection, behavioral analysis, biochemical analysis, and immunohistochemistry were performed.
Protein expression levels of lentivirus-driven expression were routinely monitored in immunoblot with anti-AT 2 receptor antibodies (ϳ1.8 -2.3-fold hippocampal overexpression compared with mice injected with a control lentivirus encoding ␤-galactosidase).
Induction of Stress-The generation of A␤ was enhanced in 12-month-old APP Sw mice by 4 weeks of stress as described in the accompanying article (26). Stressed mice with a significant decrease in sucrose preference (Յ50% of sucrose consumption compared with nonstressed controls) were included in the study. After 28 days of stress, 4 h before the beginning of the dark phase, brains from stressed and nonstressed APP Sw mice were removed and processed for immunoblotting, functional studies, and immunohistochemistry as described above.
Behavioral Studies-We used the standard water maze task (hidden platform) to test for spatial memory (15). Testing involved four trials per day over 10 days starting after the stress period. On the day following the 10 days of acquisition testing, memory retention was determined in a single 60-s probe trial for which the submerged platform was removed (16).
Animal experiments were reviewed and approved by the committees on animal research at the Universities of Hamburg and Cairo and were conducted in accordance with National Institutes of Health guidelines.
Statistics-Unless otherwise stated, data are expressed as mean Ϯ S.E. To determine significance between two groups, we made comparisons using the unpaired two-tailed Student's t test. p values of Ͻ0.05 were considered significant. (26) we have shown that (aggregated) A␤ induces the formation of AT 2 receptor oligomers in vivo in a dose-dependent manner. The formation of AT 2 oligomers in vivo is due to a two-step process of oxidative and transglutaminase-mediated cross-linking. To analyze whether ROS and transglutaminase were indeed sufficient to support the formation of AT 2 oligomers, we reconstituted the cross-linking in a nonneuronal in vitro system. For detection of AT 2 receptors in immunoblot, we applied AT 2specific antibodies cross-reacting specifically with the AT 2 receptor of ϳ65 Ϯ 5 kDa of AT 2 receptor-transfected HEK cells (cf. accompanying article (26)) ( Fig. 1). Oxidization of AT 2 receptor monomers of ϳ65 kDa could be induced in HEK cells expressing NOX-3 as a ROS-generating system, leading to the appearance of an SDS-stable dimeric AT 2 form of ϳ130 kDa (Fig. 1A, lane 1). NOX-3 was chosen as a constitutively active ROS-generating NADPH oxidase that also shows increased expression in AD patients (17). Cross-linking of AT 2 receptors in NOX-3-expressing HEK cells was suppressed by pretreatment with the antioxidant diferuloylmethane, confirming involvement of ROS (Fig. 1A, lane 2).

Oxidative and Transglutaminase-mediated Cross-linking of AT 2 Receptors in Vitro-In the accompanying article
ROS targets tyrosine residues of proteins, leading to dityrosine formation. Tyrosine residues of AT 2 are located at putative interreceptor contact sites (i.e. the cytoplasmic loop connecting transmembrane helices III/IV) (18). Those interreceptor contact sites appear crucial for ROS-mediated crosslinking of AT 2 , because oxidative cross-linking was strongly diminished in a mutated AT 2 receptor lacking three tyrosines of connecting loop III-IV, AT 2 Tyr3 Ala143,148,162 (Fig. 1, B

versus A).
ROS-dependent cross-linking assembled AT 2 for transglutaminase-mediated cross-linking, because the concomitant action of ROS and transglutaminase-2 led to AT 2 receptor oligomers, whereas inhibition of transglutaminase-2 by monodansyl cadaverine prevented only the appearance of AT 2 oligomers and left the oxidized AT 2 dimers intact (Fig. 1C). Thus, transglutaminase targeted AT 2 receptors preassembled by oxidative cross-linking. Transglutaminase-dependent cross-linking of oxidized AT 2 receptors seemed to require the receptor's carboxyl terminus, because a mutated AT 2 receptor, AT 2 ⌬CTer , lacking the carboxyl terminus was not significantly cross-linked by transglutaminase-2, although oxidative dimerization was evident (Fig.  1D). This finding is in agreement with a model of receptor oligomers identifying COOH-terminal residues as contact sites between dimers (18).
ROS/transglutaminase targeted specifically the AT 2 receptor, because the related AT 1 receptor appeared as a pure monomer under similar cross-linking conditions (Fig. 1E). The specificity of transglutaminase for oxidized AT 2 receptor dimers was further analyzed by a chimeric AT 2 receptor with the carboxyl terminus of AT 1 . The chimeric AT 2 receptor, AT 2 /AT 1 COO-, was not significantly cross-linked to oli-gomers by concomitant ROS/ transglutaminase action, although (oxidized) AT 2 /AT 1 COOdimers were detected in the presence of the transglutaminase inhibitor (Fig. 1F). This observation further confirms that the carboxyl terminus of the AT 2 receptor is required for transglutaminase-dependent cross-linking of (oxidized) AT 2 receptor dimers.
Altogether, a two-step process of oxidative and transglutaminase-dependent cross-linking leads to AT 2 oligomers in vitro (this work) and in vivo (accompanying article (26)).
AT 2 Receptor Oligomers Sequester G␣ q/11 in Vitro-The biochemical similarity between in vitro formed AT 2 oligomers and in vivo detected AT 2 oligomers was further analyzed. Similarly to AD patients and mice, AT 2 oligomers of HEK cells sequestered G␣ q/11 , as determined by co-enrichment with G␣ q/11 proteins, whereas AT 2 monomers/ dimers did not significantly interact with G␣ q/11 (Fig. 2, A (lanes 2 and 4  versus lanes 1 and 3) and B).
The interaction with an activated G-protein-coupled receptor is an essential part of the G-protein activation step. In agreement with this paradigm, stimulation of muscarinic M 1 receptors by carbachol strongly enhanced the interaction of M 1 receptors with G␣ q , as determined by co-enrichment (Fig. 2C). The presence of AT 2 oligomers prevented the carbachol-stimulated M 1 /G␣ q interaction almost entirely (Fig. 2C). Thus, oligomeric AT 2 is capable of suppressing the interaction of G␣ q with an activated heterologous receptor, as would be expected from its G␣ q/11 -sequestering capacity.
Next, we quantified the carbachol-stimulated interaction of G␣ q with the M 1 receptor (ϳ900 fmol/mg) in the presence of increasing levels of AT 2 oligomers. We chose the G␣ q/11 -coupled muscarinic M 1 receptor, because impaired G-protein coupling of M 1 correlates with disease severity in AD (5). Crosslinked AT 2 oligomers potently inhibited the stimulated M 1 -G␣ q interaction in a concentration-dependent manner with an EC 50 of ϳ30 fmol/mg protein (Fig. 2D). As a control, comparable levels of the AT 2 ⌬CTer mutant that was not crosslinked to oligomers under similar conditions (cf. Fig. 1D) had no significant effect on the carbachol-stimulated M 1 -G␣ q interaction (Fig. 2E). These observations strongly indicate that physiological AT 2 receptor expression levels of AD brain are suffi- (lanes 1 and 2) shows an immunoblot of AT 2 receptors with cell membranes (IB: anti-AT2). The right panel shows an immunoblot of AT 2 receptors co-enriched with G␣ q/11 from solubilized membranes subjected to immunoaffinity purification of G␣ q/11 (AP: anti-q/11; IB: anti-AT2) (upper blot). The lower panel shows a control immunoblot demonstrating equal enrichment of G␣ q/11 (AP/IB: anti-q/11). B, the reciprocal experiment confirmed that G␣ q/11 was not significantly co-enriched with AT 2 receptors under basal conditions (lane 1), whereas upon induction of AT 2 oligomers by NOX-3/Tg, AT 2 interacted with G␣ q/11 (lane 2), as determined by immunoaffinity enrichment of AT 2 and immunoblot detection of co-enriched G␣ q/11 (AP: anti-AT2; IB: anti-q/11). The right panel (lanes 3 and 4) shows equal enrichment of AT 2 receptors under both conditions (AP/IB: anti-AT2). C, AT 2 oligomers (ϮAT2 olig ; ϳ150 fmol/mg) suppress the carbachol-stimulated (ϮCch) interaction of M 1 receptors with G␣ q , as determined by immunoaffinity enrichment of M 1 receptors from M 1 -transfected HEK cells and detection of co-enriched G␣ q (AP: anti-M1; IB: anti-G␣ q ). The lower panel shows equal enrichment of M 1 receptors (AP/IB: anti-M1). D and E, carbachol-stimulated M 1 -G␣ q interaction in the presence of increasing levels of oligomeric AT 2 (AT2 olig ; ϩNOX-3/Tg; D) or mutated AT 2 ⌬CTer receptors (AT 2 ⌬CTer ; ϩNOX-3/Tg; E). The G␣ q bound to the affinity-purified M 1 receptor from solubilized HEK cell membranes was quantified by 125 I-labeled F(ab) 2 fragments of anti-G␣ q antibodies added to the affinity matrix. Data represent mean Ϯ S.E., n ϭ 3. *, p Ͻ 0.001 versus column 0 (i.e. the stimulated G␣ q /M 1 interaction of HEK cells expressing only M 1 ); analysis of variance followed by Dunnett's multiple comparison test. The insets show representative immunoblots of the eluates detecting (i) G␣ q co-enriched with stimulated M 1 (upper panels; AP: anti-M1 and IB: anti-G␣ q ) and (ii) enriched M 1 (lower panels; AP/IB: anti-M1). MARCH 6, 2009 • VOLUME 284 • NUMBER 10 JOURNAL OF BIOLOGICAL CHEMISTRY 6569 cient to exert major effects on the interaction of G(␣) q/11 with activated receptors (43.4 Ϯ 2.7 fmol of AT 2 /mg of protein Ϯ S.E., n ϭ 10, as determined with 125 I-labeled AT 2 receptorspecific antibodies on prefrontal cortex membranes isolated from AD patients).

Dominant Negative AT 2 Receptors
G␣ q/11 -sequestering AT 2 Receptor Oligomers Inhibit Basal and Carbachol-stimulated Activation of G␣ q/11 -The AT 2 oligomer-mediated sequestration of G␣ q/11 proteins was accompanied by reduced G␣ q/11 activation either under basal conditions or upon carbachol stimulation (Fig. 3A), whereas the AT 2 ⌬CTer mutant that is not cross-linked to oligomers did not significantly affect the M 1 receptor-stimulated signal under similar conditions (Fig. 3B).
AT 2 Receptor Oligomers Suppress G␣ q/11 -stimulated Signaling-We analyzed the effect of AT 2 oligomers on G␣ q/11stimulated signaling in intact cells by measuring the G␣ q/11stimulated activation of phospholipases C. AT 2 oligomers led to a strongly decreased inositol phosphate signal generation either under basal conditions or upon carbachol stimulation of transfected muscarinic M 1 receptors (Fig. 4A). Inhibition of M 1 receptor-stimulated signaling by AT 2 oligomers was confirmed by significantly reduced potency and efficacy of carbachol (Fig. 4B). As a control, ROS and transglutaminase did not significantly affect M 1 -stimulated signaling in the absence of AT 2 under similar conditions (data not shown). Cell surface M 1 receptor levels as determined by fluorescence-activated cell sorting analysis with M 1 -specific antibodies were similar to those of HEK cells expressing dissociable AT 2 receptors or cross-linked AT 2 oligomers (Fig. 4D, panel 1

versus panel 2).
Signaling stimulated by direct activation of G-proteins with GTP␥S or AlF 4 Ϫ was also strongly decreased in the presence of crosslinked AT 2 oligomers, whereas G-protein-independent inositol phosphate generation triggered by direct activation of phospholipases C with calcium was not changed (Fig. 4A). These findings show that AT 2 oligomers inhibit G␣ q/11 -stimulated signaling.
AT 2 Oligomers Arrest Constitutively Activated G␣ 11 Q209L -AT 2 oligomers also interacted with a constitutively active mutant of G␣ 11 , G␣ 11 Q209L , and suppressed the inositol phosphate signal stimulated by this mutant (Fig. 4A). This observation reveals that AT 2 oligomers act in a dominant negative manner and induce the virtual  Ϫ (300 M), or Ca 2ϩ (10 M in permeabilization buffer) or upon expression of the constitutively activated G␣ 11 Q209L mutant. Data are the mean Ϯ S.E., n ϭ 6. *, p Ͻ 0.0002. B, concentration-response relationship of the carbachol-stimulated inositol phosphate signal of M 1 receptor-expressing HEK cells co-expressing dissociable AT 2 receptors or AT 2 receptor oligomers. Inositol phosphates are expressed as percentage of maximum stimulation (i.e. the signal obtained with 100 M carbachol on cells with dissociable AT 2 receptors) (12,710 Ϯ 420 dpm). Data represent mean Ϯ S.D., n ϭ 3. C, carbachol-stimulated (100 M) rise in the free intracellular calcium concentration, [Ca 2ϩ ] i , of M 1 receptor-expressing HEK cells co-expressing dissociable AT 2 (panels 1-3) or AT 2 oligomers (panels 4 -6). As indicated, cells were co-stimulated with 100 nM angiotensin II (Ang; panels 2 and 5). In the experiments of panels 3 and 6, cells were desensitized by prior stimulation with carbachol for 30 min at 37°C, followed by washing steps (Cch desens.). Single experiments are representative of three independent experiments, each with similar results. D, fluorescence-activated cell sorting analysis (FACS) of cell surface M 1 receptors of HEK cells co-expressing dissociable AT 2 receptors (panels 1 and 3) or AT 2 oligomers (panels 2 and 4). In the experiments of panels 3 and 4, M 1 receptor redistribution (Redistrib.) was induced by stimulation with 1 mM carbachol for 60 min at 37°C. Cell surface M 1 receptors were determined by FACS using affinity-purified F(ab) 2 fragments of M 1 -specific antibodies.
arrest of an activated G␣ q/11 protein. The dominant negative activity of the nondissociable oligomeric AT 2 receptor platform could account for the potent inhibition of G␣ q/11 -stimulated signaling at the cell membrane, which is dependent on the release of a limited number of activated G␣ q/11 proteins triggered by stimulation of specific receptors.
Inhibition of Calcium Signaling and Desensitization by AT 2 Oligomers-Impaired M 1 receptor-stimulated G␣ q/11 -protein activation in the presence of AT 2 oligomers was also reflected by a decreased rise in the intracellular free calcium concentration, [Ca 2ϩ ] i (Fig. 4C, panels 1 and 2 versus panels 4 and 5). The G␣ q/11 inhibitory effect of the cross-linked AT 2 oligomers was independent of the activation of AT 2 by angiotensin II (Fig. 4C,  panels 4 and 5).
Activated G-proteins trigger signal generation as well as signal desensitization. To determine whether AT 2 oligomers inhibited the desensitization of the M 1 receptor, we pretreated the cells with carbachol. Carbachol pretreatment led to a significant decrease of the M 1 -stimulated rise in [Ca 2ϩ ] i on cells expressing dissociable AT 2 receptors indicative of M 1 receptor desensitization (Fig. 4C, panel 3 versus panels 1 and 2). In contrast, the pretreatment with carbachol did not markedly decrease the M 1 signal on cells expressing oligomeric AT 2 receptors (Fig. 4C, panel 6 versus panels 4 and 5). Thus, AT 2 oligomers inhibit signal generation and desensitization of the G␣ q/11 -coupled M 1 receptor.
Receptor desensitization is followed by receptor redistribution. Stimulation with carbachol induced prominent M 1 receptor redistribution/internalization on cells expressing dissociable AT 2 receptors (Fig. 4D, panel 3 versus panel 1). In contrast, inhibition of G␣ q/11 proteins by AT 2 oligomers led to a significantly reduced carbachol-stimulated M 1 receptor redistribution (Fig. 4D, panel 4 versus panel 2). Together, these findings show that cross-linked AT 2 oligomers arrest G␣ q/11stimulated signaling, as reflected by inhibition of different signal transduction events triggered by a prototypic G␣ q/11 -coupled receptor (i.e. the muscarinic M 1 receptor).
The Truncated AT 2 ⌬CTer Mutant Suppresses AT 2 Oligomerization in Vitro and in Vivo-We further assessed the impact of AT 2 oligomers on G␣ q/11 activity and applied the truncated AT 2 ⌬CTer mutant to interfere with AT 2 oligomerization. The AT 2 ⌬CTer receptor when co-expressed with wild-type AT 2 prevented the transglutaminase-dependent cross-linking step accounting for AT 2 dimer/oligomer transition, as determined in immunoblot with AT 2 -specific antibodies detecting specifically wild-type AT 2 receptors (Fig. 5A). Upon inhibition of AT 2 oligomerization, the basal and M 1 receptor-stimulated G␣ q/11 activation were not significantly impaired by AT 2 even in the presence of ROS and activated transglutaminase ( Fig. 5B; cf. Fig.  3A). These findings confirm that AT 2 oligomers prevent G␣ q/11 activation.
To suppress A␤-induced AT 2 oligomerization in vivo, we expressed the AT 2 ⌬CTer mutant (that prevented AT 2 oligomerization in vitro) in the hippocampal CA1 area of stressed APP Sw mice by stereotactic injection of a lentivirus. Similarly as in HEK cells, expression of AT 2 ⌬CTer led to a marked decrease of AT 2 oligomerization in the hippocampus of treated "AD mice," as determined by immunoblotting (Fig. 6A). Thus, AT 2 ⌬CTer interfered with transglutaminase-mediated cross-linking of AT 2 receptors in vitro and in vivo.
Inhibition of AT 2 Oligomerization Slows Progression of Neurodegeneration in AD Mice-Detection of the endogenous AT 2 receptor and the transduced AT 2 ⌬CTer by immunohistochemistry revealed prominent neuritic/dendritic AT 2 staining in a representative treated AD mouse expressing AT 2 ⌬CTer that was not visible in the control (i.e. a representative APP Sw mouse with stress-enhanced A␤ generation and injection of a control lentivirus) (Fig. 6B). Notably, there was prominent neuritic/dendritic AT 2 staining in treated AD mice expressing AT 2 ⌬CTer that was not visible in the controls (Fig. 6B). Immunohistochemistry applying MAP2-specific antibodies confirmed this observation and revealed largely intact neurites/dendrites in the injected CA1 area of mice expressing AT 2 ⌬CTer , whereas there was extensive neuritic/dendritic degeneration in the control mice ( Fig. 6C; cf. Fig.  7A). These findings strongly suggest that the AT 2 ⌬CTer mutant retarded the development of neurodegenerative symptoms in stressed APP Sw mice (Fig. 6C).
In accordance with these observations, TUNEL-positive cells were barely detectable in the injected CA1 area of treated AD mice expressing AT 2 ⌬CTer , whereas CA1 neurons of control mice showed prominent TUNEL labeling (Fig. 6D, right versus  left panel). Decreased symptoms of neuronal degeneration correlated with significantly reduced hippocampal transglutaminase activity (Fig. 6E).
Treated AD mice expressing AT 2 ⌬CTer also displayed a marginally better performance in the CA1-dependent learning task of spatial memory acquisition and retention compared with control AD mice with a high load of cross-linked AT 2 receptor oligomers (Fig. 6, F and G). Together these findings strongly suggest that inhibition of AT 2 oligomerization by expression of a protein inhibitor (i.e. the AT 2 ⌬CTer mutant) retarded the CTer ; ϳ150 fmol/mg) expressed in HEK cells upon ROS generation by NOX-3 and transglutaminase-2 activation (NOX-3/Tg) in the absence (Ϫ) or presence (ϩ) of AT 2 ⌬CTer suppressing the formation of cross-linked AT 2 oligomers (lane 2 versus lane 1). The applied antibodies were raised against a carboxyl-terminal epitope of AT 2 that is lacking in AT 2 ⌬CTer and cross-react specifically with wild-type AT 2 . B, effective G␣ q/11 activation (Basal, columns 1 and 2; ϩCch, columns 3 and 4) upon inhibition of AT 2 oligomerization by AT 2 ⌬CTer . Activation of G␣ q/11 was determined in the presence of NOX-3 and activated transglutaminase-2 (NOX-3/Tg) with membranes of M 1 receptor-expressing HEK cells co-expressing ϳ150 fmol/mg protein of AT 2 in the absence (Ϫ) or presence (ϩ) of AT 2 ⌬CTer (ϳ150 fmol/mg protein). Data represent mean Ϯ S.E., n ϭ 6. *, p Ͻ 0.001. development of neurodegenerative symptoms in stressed APP Sw mice.
Inhibition of AT 2 Oligomerization Prevents G-protein Dysfunction in AD Mice-Overexpression of wild-type AT 2 receptors in AD mice further confirmed the involvement of AT 2 receptor(s) (oligomers) in the neurodegenerative process. Although immunohistochemistry revealed intact neurons in the CA1 area of mice with AT 2 ⌬CTer -dependent inhibition of oligomerization, there was profound dendritic pathology in CA1 neurons of stressed APP Sw mice overexpressing comparable levels of oligomerization-prone wild-type AT 2 receptors (Fig. 7A). Concomitantly, overexpression of AT 2 (oligomers) led to strongly impaired M 1 receptor signaling compared with the prominent activation of G␣ q/11 in mice expressing similar levels of AT 2 ⌬CTer (Fig. 7B). These findings further confirm that AT 2 (oligomers) induce G␣ q/11 impairment, whereas inhibition of AT 2 oligomerization by AT 2 ⌬CTer prevents the development of G-protein dysfunction in transgenic mice with AD-like symptoms.
In agreement with a role of M 1 in nonamyloidogenic processing of APP (19, 20), dysfunctional M 1 receptor-stimulated G-protein activation of AT 2 receptor-expressing mice was accompanied by a marginally decreased hippocampal ␣-secretase activity compared with mice with expression of AT 2 ⌬CTer and intact M 1 -stimulated signaling (Fig. 7C). Thus, the ␣-secretase activity of AD mice was preserved upon inhibition of AT 2 oligomerization by expression of an AT 2 mutant (this work) and also upon downregulation of AT 2 (oligomerization) by RNA interference (cf. accompanying article (26)) ( Fig. 6). Altogether, the experiments reveal that targeting of AT 2 receptor oligomers by expression of a protein inhibitor is feasible and slows the development of neurodegenerative symptoms in transgenic mice with ADlike pathology.

DISCUSSION
G-protein dysfunction in the brains of AD patients is a characteristic hallmark of clinical AD with neurodegeneration (5). In the accompanying article (26), G-protein dysfunction was linked to AT 2 oligomers that form in the brains of AD patients and mice with AD-like symptoms. AT 2 receptor oligomers assemble in vivo by a two-step process of oxidative and transglutaminase-dependent cross-linking, which is triggered consecutively by (aggregated) A␤ in a dose-dependent manner (see the accompanying article (26)). To further characterize the function of AT 2 oligomers, the current study reconstituted the two-step formation of AT 2 receptor oligomers in vitro in a transfected cell system. The in vitro experiments revealed that the initial oxidative cross-linking step accounting for the formation of AT 2 receptor dimers targets specifically AT 2 with its tyrosine residues at putative interreceptor contact sites (18), because a mutated AT 2 receptor lacking the tyrosines FIGURE 6. Inhibition of AT 2 oligomerization slows the progression of neurodegeneration in AD mice. A, immunoblot of hippocampal AT 2 receptors of four stressed APP Sw mice upon stereotactic injection of a lentivirus encoding AT 2 ⌬CTer (Treated) revealed decreased AT 2 receptor oligomerization compared with mice injected with a control lentivirus (Control). Applied antibodies cross-react with a COOH-terminal epitope of AT 2 that is lacking in AT 2 ⌬CTer (IB: anti-AT 2 CTer ). B, immunohistochemistry applying antibodies cross-reacting with an NH 2 -terminal epitope of AT 2 (anti-AT 2 NTer ) to visualize AT 2 and AT 2 ⌬CTer revealed AT 2 receptor expression in CA1 neuronal cell bodies and largely intact dendrites of stressed APP Sw transgenic mice co-expressing AT 2 ⌬CTer (Treated) compared with mice receiving injection of a control lentivirus (Control) encoding ␤-galactosidase (original magnification, ϫ300). C, inhibition of AT 2 oligomerization by AT 2 ⌬Cter retarded the development of symptoms of dendritic degeneration in the injected CA1 area of stressed APP Sw transgenic mice (Treated) compared with mice injected with a control lentivirus (Control), as determined by immunohistochemistry with anti-MAP2 antibodies (original magnification, ϫ300). D, DNA strand breaks were barely detectable in the injected CA1 area of stressed APP Sw mice upon inhibition of AT 2 oligomerization by AT 2 ⌬CTer compared with stressed APP Sw mice injected with a control lentivirus, as determined by in situ TUNEL labeling (original magnification, ϫ300). E, inhibition of AT 2 oligomerization in the CA1 area of stressed APP Sw transgenic mice by AT 2 ⌬CTer (Treated) led to a significantly decreased hippocampal transglutaminase activity compared with stressed APP Sw transgenic mice injected with a control lentivirus (Control). Data represent mean Ϯ S.E., n ϭ 8. *, p Ͻ 0.01. F and G, inhibition of AT 2 oligomerization by AT 2 ⌬CTer led to a significantly better performance of stressed APP Sw transgenic mice of the water maze acquisition (F) and retention (G) test compared with mice injected with a control lentivirus. Data represent mean Ϯ S.E., n ϭ 12 mice per subgroup. F, *, p Ͻ 0.05; **, p Ͻ 0.008. G, *, p Ͻ 0.003; analysis of variance with Dunn's multiple comparison test.