¹²⁵I-Antisauvagine-30: a novel and specific high-affinity radioligand for the characterization of corticotropin-releasing factor type 2 receptors

Abstract

Corticotropin-releasing factor (CRF) receptors type 1 (CRF₁) and type 2 (CRF₂) differ from each other in their pharmacological properties. The human and ovine CRF versions bind to CRF₁ receptors with significantly higher affinity than to CRF₂ receptors. Recently antisauvagine-30, an N-terminally truncated version of the CRF analog sauvagine, was characterized as a specific antagonist to mouse CRF_2B. We have synthesized the radiolabeled version ¹²⁵I-antisauvagine-30 and tested it for its affinity at human CRF₁ (hCRF₁), hCRF_2A, Xenopus CRF₁ (xCRF₁) and xCRF₂ receptors. In control binding studies ¹²⁵I-labeled hCRF, sauvagine and astressin were also bound to these receptors. ¹²⁵I-antisauvagine-30 exclusively bound to hCRF_2A and xCRF₂ but not to hCRF₁ and xCRF₁ receptors. ¹²⁵I-antisauvagine-30 binding to hCRF_2A and xCRF₂ receptors was saturable and of high affinity (hCRF_2A: K_d=125 pM; xCRF₂: K_d=1.1 nM). In displacement binding experiments using ¹²⁵I-antisauvagine-30 as radioligand several CRF analogs bound to hCRF_2A and xCRF₂ receptors with similar rank orders as reported with other CRF radioligands. Finally, preliminary studies using ¹²⁵I-antisauvagine-30 binding to membrane homogenates prepared from different rat brain structures showed that the peptide bound specifically to brain areas expressing CRF₂ receptors. These data demonstrate that ¹²⁵I-antisauvagine-30 is the first high-affinity ligand to specifically label CRF₂ receptors.

Introduction

The 41-amino acid peptide corticotropin-releasing factor (CRF), a main regulator of the body's stress response (for a review see Vale et al., 1997), is assumed to play a major role in several psychiatric disorders (Dunn and Berridge, 1990, Arborelius et al., 1999). CRF and its structurally related analogs mammalian urocortin (UCN), amphibian sauvagine (SVG) and fish urotensin I (URO) mediate their effects through two receptors, both of which belong to the class B subfamily of G protein-coupled receptors (GPCR) (reviewed in Kilpatrick et al., 1999). Two principal subtypes of the CRF receptor, type 1 (CRF₁) and type 2 (CRF₂), which are 70% identical on amino acid level, have been cloned from a variety of species (reviewed in Holsboer, 1999).

Despite their high degree of sequence homology, CRF₁ and CRF₂ receptors differ markedly from each other in their pharmacological properties and tissue distribution. The mammalian CRF₁ receptor is nonselective for CRF, UCN, URO and SVG. These peptides are bound with similar affinity and are equipotent in their ability to stimulate cAMP accumulation in cells expressing recombinant mammalian CRF₁ receptors (Vaughan et al., 1995, Donaldson et al., 1996, Dautzenberg et al., 1997, Dautzenberg et al., 1999a, Palchaudhuri et al., 1998). CRF₂ receptors, however, are ligand-selective (Donaldson et al., 1996, Dautzenberg et al., 1997, Dautzenberg et al., 1999a, Ardati et al., 1999, Palchaudhuri et al., 1999). While UCN, URO and SVG are bound by CRF₂ receptors with high affinity, CRF isolated from different species are bound with significantly lower affinity and thereby are CRF₁ receptor-specific ligands.

The tissue distribution of CRF receptors has mainly been studied by detection of their mRNAs using in situ hybridization or RNAse protection analyses. Thereby, CRF₁ receptor mRNA was found to be abundantly expressed in the brain, especially in cortex and cerebellum, and in the pituitary (Chalmers et al., 1995, Palchaudhuri et al., 1998). In the periphery, on the other hand, CRF₁ receptor mRNA is restricted to low level expression in a few organs (Palchaudhuri et al., 1998). CRF₂ receptor mRNA expression in the brain, in contrast to the CRF₁ receptor, is very discrete, occurring only in the hypothalamus, lateral septum, hippocampus and dorsal raphe (Chalmers et al., 1995, Palchaudhuri et al., 1999, Sanchez et al., 1999), loci of the neuroendocrine stress response, behavioral components of the startle response, contextual fear conditioning, and explicit memory acquisition (reviewed in Chalmers et al., 1996, Hauger and Dautzenberg, 1999). In addition to its expression in the brain, CRF₂ receptor mRNA is abundantly expressed in the skeletal muscle and the heart (Chalmers et al., 1995, Palchaudhuri et al., 1999).

Another method to study the distribution of CRF receptors in the brain is receptor autoradiography. For CRF₁ receptors, this has been accomplished using ¹²⁵I-labeled oCRF (Wynn et al., 1984), a radioligand displaying subnanomolar affinity to CRF₁ receptors but >100 nM affinity to CRF₂ receptors (Perrin et al., 1999). Although high-affinity ¹²⁵I-labeled ligands displaying equal affinity to CRF₁ and CRF₂ receptors have been developed (Grigoriadis et al., 1996, Perrin et al., 1999), there is still the need for a high-affinity CRF₂ receptor selective radioligand. Recently, [D-Phe¹¹, His¹²]SVG_(11–40), named antisauvagine-30 (aSVG-30), has been reported as the first CRF antagonist with ∼100-fold higher affinity to CRF₂ than to CRF₁ receptors (Rühmann et al., 1998).

In this study, we report the characterization of the labeled compound, (¹²⁵I-aSVG-30) that has high affinity to human CRF_2A (hCRF_2A) and Xenopus CRF₂ (xCRF₂) receptors but is devoid of any activity at hCRF₁ and xCRF₁ receptors, allowing for the first time the exclusive labeling of the CRF₂ receptor subtypes in vitro and providing a potential high-quality pharmacological tool for the study of this receptor in mixed populations of cells or tissues expressing both CRF receptor subtypes.