Urocortin 1 and 3 Impair Maternal Defense Behavior in Mice Kimberly L. D’Anna, Sharon A. Stevenson, and Stephen C. Gammie University of Wisconsin—Madison Lactating female mice fiercely defend offspring while exhibiting decreased fear and anxiety. Recent work (J. S. Lonstein & S. C. Gammie, 2002) found that intracerebroventricular (icv) injections of corticotropin releasing factor (CRF), a putative anxiogenic peptide, inhibit maternal defense behavior. This study examines effects of CRF-related peptides, urocortin (Ucn) 1 and Ucn 3, on maternal aggression in mice. Intracerebroventricular injections of both Ucn 1 (0.2 g) and Ucn 3 (0.5 g) reduced aggression but not pup retrieval. c-Fos levels were elevated by intracerebroventricular injections of Ucn 1 (0.2 g) and Ucn 3 (0.5 g) in 2 and 6 brain regions, respectively; however, both triggered increases in bed nucleus of the stria terminalis dorsal (BNSTd) and lateral septum (LS). These findings suggest that CRF-related peptides similarly modulate maternal aggression and that BNSTd/LS may be critical sites for negative regulation of maternal aggression. Keywords: corticotropin-releasing factor, maternal aggression, corticotropin-releasing factor receptor 2, lateral septum, bed nucleus of the stria terminalis Maternal aggression is a fierce behavior displayed by lactating females against intruders that plays an important role in defense and protection of offspring. In rodents, this heightened aggression has been associated with reduced indices of fear and anxiety (relative to virgin females) in several experimental paradigms, including the elevated plus-maze (Ferreira, Pereira, Agrati, Uri- arte, & Fernandez-Guasti, 2002; Kellogg & Barrett, 1999), acous- tic startle stimulus (Hard & Hansen, 1985; Toufexis, Rochford, & Walker, 1999), open field test (Fleming & Luebke, 1981), defen- sive burying (Picazo & Fernandez-Guasti, 1993), freezing behav- ior (Ferreira et al., 2002), and light– dark choice tests (Lonstein & Gammie, 2002; Maestripieri, Badiani, & Puglisi-Allegra, 1991). Recent work (Gammie, Negron, Newman, & Rhodes, 2004) has supported the idea that decreased fear and anxiety during lactation play a critical role in maternal defense behavior (possibly by making the dam less hesitant to attack a normally fear-evoking stimulus) by showing that central elevations of the putative anx- iogenic peptide, corticotropin releasing factor (CRF), decreases maternal aggression. Prior studies have implicated CRF in regu- lating maternal aggression by indicating that the central nervous system (CNS) of the lactating rat is less sensitive (compared with the virgin rat) to CRF (da Costa, Kampa, Windle, Ingram, & Lightman, 1997) and that CRF synthesis is decreased at certain stages of lactation (Walker, Toufexis, & Burlet, 2001). Thus, for normal maternal aggression to occur, decreased neurotransmission of CRF may be necessary. CRF belongs to a peptide family composed of three other peptides: urocortin (Ucn) 1; Ucn 2 or stresscopin-related peptide; and Ucn 3, or stresscopin (Hsu & Hsueh, 2001; Lewis et al., 2001). This family of peptides can act on two G-protein coupled recep- tors, CRF receptor 1 (CRF1) and CRF receptor 2 (CRF2; Hauger et al., 2003). CRF1 has a widespread distribution in the brain, whereas mRNA for CRF2 is found in a more limited distribution that is concentrated mostly in subcortical regions and with highest expression in the lateral septal nuclei (Chalmers, Lovenberg, & De Souza, 1995). CRF and Ucn 1 have a high affinity for both receptors, but Ucn 1 binds more readily to CRF2 than CRF (Vaughan et al., 1995). Ucn 2 and Ucn 3 bind only to CRF2 and are considered to be the endogenous ligands for this receptor (Hauger et al., 2003; Lewis et al., 2001; Venihaki et al., 2004). Similar to CRF, Ucn 1 is a putative anxiogenic peptide (Koob & Heinrichs, 1999; Spina et al., 2002). In support of this fact, central administration of Ucn 1 causes an increase in anxiety-like behav- iors during a light– dark choice test, open field test, and elevated plus-maze test (Moreau, Kilpatrick, & Jenck, 1997). Ucn 3 has been shown to be produced in stress-related areas of the brain (Li, Vaughan, Sawchenko, & Vale, 2002; Venihaki et al., 2004), but few studies have examined the effects of Ucn 3 on measures of anxiety and the results are less clear. Central admin- istration of Ucn 3 did not alter elevated plus-maze performance in two studies (Pelleymounter, Joppa, Ling, & Foster, 2004; Venihaki et al., 2004), but in a third study, an increase in percentage of time on the open bars was reported (Valdez, Zorrilla, Rivier, Vale, & Koob, 2003). The CRF peptide family and its receptors have also been exam- ined in the behavioral anxiety-like responses to uncontrollable stress, known as learned helplessness. A principal role for CRF, Ucn 2, and CRF2 within the dorsal raphe (DRD) has been sug- gested in mediating the behavior (Hammack et al., 2001; Ham- mack, Richey, et al., 2003; Hammack, Schmid, et al., 2003). These findings also suggest that the family of CRF-related peptides and Kimberly L. D’Anna and Sharon A. Stevenson, Department of Zoology, University of Wisconsin—Madison; Stephen C. Gammie, Department of Zoology and Neuroscience Training Program, University of Wisconsin— Madison. This work was supported by National Institutes of Health Grant R01 MH066086 to Stephen C. Gammie. We thank Justin Rhodes for statistical consultation. We also thank Emily Bethea, Kelly Clinkenbeard, Heidi Gierahn, Justin Friske, Kyle Blake, Amber Frank, and Nina Hasen for technical assistance, and Kate Skogen and Jeff Alexander for animal care. Correspondence concerning this article should be addressed to Kimberly L. D’Anna, Department of Zoology, University of Wisconsin, 1117 West Johnson Street, Madison, WI 53706. E-mail: kldanna@wisc.edu Behavioral Neuroscience Copyright 2005 by the American Psychological Association 2005, Vol. 119, No. 4, 1061–1071 0735-7044/05/$12.00 DOI: 10.1037/0735-7044.119.4.1061 1061