Cell-based approach for the detection of functional activating autoantibodies against the ß₁-adrenergic receptor (anti-ß₁-Abs). (A) Receptor activation induced by binding of antibodies to its accessible extracellular loops leads to an increase in cyclic adenosine monophosphate (cAMP) through sequential activation of G_s proteins and adenylyl cyclase (AC). The cAMP is detected by fluorescence resonance energy transfer (FRET) (grey arrow) between cyan fluorescent proteins (CFP) and yellow fluorescent proteins (YFP) fused to the cAMP-binding domain of Epac1. The sensor (Epac1-based fluorescent cAMP sensor [Epac1-camps]) changes its conformation upon cAMP binding (black arrow), resulting in a decrease in FRET. (B) Measuring cAMP levels in isoproterenol (Iso)-stimulated human embryonic kidney HEK293 cells stably expressing human ß₁-AR by conventional radioimmunoassay (RIA) (Amersham, Freiburg, Germany) or Epac-FRET. One of 3 independent RIA concentration-response curves (EC₅₀ = 0.53 ± 0.19 nmol/l) is shown. The cAMP range that can be monitored by Epac-FRET until it gets saturated is presented by 2 horizontal dotted lines. At 0.05 nmol/l isoproterenol, Epac1-camps becomes saturated, indicating an intracellular cAMP concentration of approximately 20 µmol/l (according to in vitro measurements in cell lysates with pure cAMP [19]). This extremely sensitive sensor is characterized by a high dynamic range at physiologically relevant cAMP concentrations 0.1 to 20 µmol/l (which are covered by only 15% of the maximal cAMP-RIA signal), making minor cAMP changes previously undetected by conventional assays well detectable by Epac-FRET. (C) The immunoglobulin (Ig)G prepared from rats immunized with glutathione-S-transferase (GST) fusion proteins containing the first (EC_I-immunized, center) or second extracellular loop of the human ß₁-AR (EC_II-immunized, right) were assayed for activity using human embryonic kidney HEK293 cells stably expressing human ß₁-AR transfected with Epac1-camps, and compared with nonimmunized animals (control animals, left) to assess the reliability of the method. Representative FRET ratio traces of 1 of 6 independent experiments are presented (% corresponds to the relative change in YFP/CFP intensity ratio). The decrease in FRET reflects an increase in intracellular cAMP.

Measuring cAMP by Epac-FRET detects functional anti-ß₁-Abs in dilated cardiomyopathy (DCM) patients. (A) None of the IgG prepared from healthy control patients (n = 50) induced a significant cAMP response in living cells (left). The IgG from DCM patients previously judged anti-ß₁-Abs-positive (15) (Abs+) elicited marked cAMP responses (49.5 ± 3.8% of maximal Iso signal, right). The IgG from 23 of 38 (60%) previously anti-ß₁-Abs-negative judged patients (6) (Abs–) showed a robust but significantly smaller increase in cAMP (31.3 ± 6.8%, p < 0.01 according to the Student t test, center). Representative experiments of at least 3 different cells for each IgG preparation are shown. (B) Histogram with Gaussian distribution curves based on the strength of the FRET signal in healthy control patients versus DCM patients (R² = 0.98 for a 3-peak Gaussian distribution). The antibody induced FRET signals allow discrimination of 3 groups of activity: no activity (n = 15), low activity (n = 22), high activity (n = 18); grouping was performed as described in the text. (C) Frequency and distribution of FRET signals obtained with IgG preparations from n = 22 previously antibody-typed ischemic cardiomyopathy (ICM) patients. Five of them showed high FRET activity, and the remainder was classified antibody-negative. (D) Concentration-response relationship between high-activator and low-activator IgG showing significantly different activation capacities over a wide range of antibody concentrations (p < 0.05; Student t test). The percent ± SEM of maximal Iso-induced cAMP responses is presented (normalized to maximal changes in FRET induced by high-activator anti-ß₁-Abs; representative experiments, n = 4). Abbreviations as in Figure 1.

Blocking experiments for different classes of functional anti-ß₁-Abs. (A) The cAMP generation induced by high-activator IgG is attenuated only by specific peptides derived from the second extracellular ß₁-AR loop (ß₁-EC_II). (B) Low-activator signals could be blocked in all instances by peptides corresponding to the first extracellular ß₁-AR loop (ß₁-EC_I) but not by ß₁-EC_II peptides (representative experiments from at least 3 cells per condition are shown). Analysis of the kinetics of the different FRET responses is presented in the lower panels of (A) and (B). Ratio traces were fitted with a mono-exponential function, and the time constant of the decay of the respective FRET signals was determined. In both cases, the effects of EC_I and EC_II peptides differ significantly (*p < 0.05, **p < 0.01; one-way analysis of variance [ANOVA]). (C) Columns represent mean ± SEM (error bars) of cAMP responses induced by (n = 7) representative high-activator IgG, or (D) (n = 10) representative low-activator IgG prepared from DCM patients in the absence (n.b. = not blocked) or presence of peptides corresponding to the first (EC_I) and/or second extracellular ß₁-AR loop (EC_II), or to EC_II of the ß₂-AR. Statistical significance was tested by one-way ANOVA (**p < 0.01 compared with the n.b. group). (E) Dose-response curve showing the effect of increasing concentrations of the ß₁-EC_II peptide on the inhibition of high-activator-induced cAMP responses (representative experiments, n = 5). The curve was fitted to the data using Origin 6.1 (OriginLab Corp., Northampton, Massachusetts) assuming a single binding site. Abbreviations as in Figures 1 and 2.

Beta-blockers do not fully abolish the Epac-FRET signals in cells stimulated with high-activator anti-ß₁-Abs. Neither the selective ß₁-AR anatagonists bisoprolol (Biso) (A) or metoprolol (Meto), nor nonselective drugs such as alprenolol (C) (Alpren) or carvedilol (Carved) blocked more than 70% of the FRET response induced by high-activator IgG under saturating conditions (3 µmol/l). Bisoprolol combined with EC_II peptides, however, could fully block high-activator-induced FRET signals (Biso + ECII) (D). (B) Concentration-response curve for the inhibitory effect of bisoprolol alone on high-activator-induced FRET signals. Data, means, and SEMs in (D) are calculated from at least 4 independent experiments. (E) The stimulatory effect of low-activator IgG could be fully abolished by bisoprolol. Representative experiments, n = 6. Quantification is shown in (F) together with the effect of a combination of bisoprolol with EC_I peptides (n = 5). Abbreviations as in Figures 1 and 2.