Behavioural studies using temporal and spatial inactivation of the oxytocin receptor

Heon-Jin Lee1, Heather K. Caldwell2, Abbe H. Macbeth1 and W. Scott Young III1'*

1 Section on Neural Gene Expression, National Institute of Mental Health, National Institutes of Health, DHHS,

Bethesda, MD, USA 2Department of Biological Sciences, Kent State University, Kent, OH, USA

Abstract: Oxytocin (Oxt), synthesized in magnocellular neurons of the paraventricular (PVN) and supraoptic (SON) hypothalamic nuclei for transport to and release from the posterior pituitary, is released during parturition and is essential for lactation. Lesser amounts of Oxt are made by smaller cells of the PVN and a few other forebrain nuclei and released into the central nervous system (CNS) to influence various other behaviours. In both the periphery and CNS, Oxt actions are transduced by the oxytocin receptor (Oxtr). Previously, it has been reported that Oxt—/— (knockout, KO) mice show a failure of milk ejection and thus are incapable of rearing their offspring. Unexpectedly, these mice have largely normal reproductive and maternal behaviours, perhaps due to compensatory mechanisms through activation of the Oxtr by vasopressin or through development. To examine the specific roles of the Oxtr during development and in particular brain areas, we created conditional Oxtr—/— mice in which we could control the spatial and temporal inactivation of the Oxtr. We flanked the neomycin-resistance selectable marker in an Oxtr intron with FRT sites to enable its removal using FLP recombinase. Coding sequence within exons 2 and 3 was flanked by two loxP sites enabling subsequent inactivation of the gene by targeted expression of Cre recombinase. The first Oxtr KO lines we created have either total or relatively specific forebrain elimination. The latter was achieved by crossing the conditional Oxtr line with a transgenic line in which the Camk2a promoter drives expression of Cre recombinase to significant levels beginning 21-28 days after birth, thus eliminating potential compensation for a deleted Oxtr gene during early development. This Cre-expressing line also significantly spares the main olfactory bulb reducing the potential confound of an olfactory deficit. We have investigated various behaviours, most notably social recognition, in both Oxtr KO strains (Oxtr—/— and OxtrFB/FB).

Keywords: oxytocin; Oxtr; conditional; knockout; behaviour; transgenic; Camk2a; Cre

Introduction

The first gene knockout (KO) mouse created by homologous recombination was introduced in the

Corresponding author. Tel.: +301-496-8767; Fax: +301-402-6473; E-mail: [email protected]

late 1980s (Mansour et al., 1988; Thompson et al., 1989). This powerful technology has facilitated the study of various mammalian biological processes, from signal transduction to animal behaviour. The neurotransmitter oxytocin (Oxt) and its sole receptor, oxytocin receptor (Oxtr), have also been studied using the KO mouse model. Oxt is the most abundant hormone in the hypothalamus and is best known for its role in reproductive functions such as parturition and lactation (Gimpl and Fahrenholz, 2001). The majority of Oxt is synthesized in magnocellular neurons of the paraventri-cular (PVN) and supraoptic (SON) nuclei of the hypothalamus, and released from the posterior pituitary into the general blood circulation (Brownstein et al., 1980). Lesser amounts of Oxt are made by smaller cells of the PVN and a few other forebrain nuclei and released into the central nervous system (CNS) to influence behaviour. In both the periphery and CNS, Oxt actions are transduced by the Oxtr, which belongs to the G-protein-coupled receptor family and coupled with phospholipase C by interacting via Gaq11 (Young and Gainer, 2003). In the mouse brain, the Oxtr gene is expressed in various regions including the hypothalamus, olfactory system, amygdala and hippocampus (Insel et al., 1991; Gould and Zingg, 2003), implicating the Oxtr in various functions throughout the brain. In this chapter, the behavioural functions of Oxtr through use of the conditional KO technology are discussed.

Oxt and Oxtr knockout mice

A large number of behavioural studies have utilized KO mouse models, with many focusing on Oxt and Oxtr. The first reports on Oxt_/_ mice were made in 1996 by two independent groups, with the most notable deficits being an inability to lactate and support their own offspring (Nishimori et al., 1996; Young et al., 1996). In spite of the clear reproductive function of Oxt in milk production, Oxt_/_ mice display essentially normal parturition and maternal behaviour with some exceptions, such as pup retrieval (Nishimori et al., 1996; Young et al., 1996). To explain these results, it was hypothesized that a functional redundancy through the activation of the Oxtr by vasopressin or through developmental compensation occurs (Caldwell and Young, 2006). However, Oxtr_/_ mice generated by Takayanagi et al. (2005) show no obvious deficits in mating and parturition. Pups show increased exploration and decreased ultrasonic vocalizations consistent with less stress- or anxiety-like behaviour. These mice also display increased aggression attributed to lack of Oxtr activation by maternal Oxt prenatally (Takayanagi et al., 2005).

Additional studies in Oxt_/_ and Oxtr_/_ mice have found deficits in social memory (Ferguson et al., 2000; Takayanagi et al., 2005), increased anxiety (Mantella et al., 2003) and increased intake of sucrose (Amico et al., 2005).

Behavioural analysis using conditional knockout mice

Although the KO mice are great tools to study the function of the genes, the conventional KO technology has the following three limitations:

  • 1) genes that are essential for development could lead to a lethal embryonic phenotype in the KO;
  • 2) KO of a gene could show no phenotype due to functional redundancy; (3) KO of a widely expressed gene would not address its specific functions in different tissues (Gaveriaux-Ruff and Kieffer, 2007). In order to overcome these issues, Cre-loxP (Sternberg and Hamilton, 1981) and FLP-FRT (Broach and Hicks, 1980) systems have been applied to KO techniques. These site-specific recombinases (Cre and FLP) allow generation of temporally and/or spatially regulated KO mice (Fig. 1). Gene inactivation can be achieved through expression of Cre recombinase under the regulation of cell-type-specific promoters (Lewandoski, 2001). Early expression of Cre can cause functional inactivation in the same manner as conventional KO mice. Therefore, for behavioural studies, the use of late onset promoters is more useful to obtain the benefit of temporal regulation (Morozov et al., 2003; Gaveriaux-Ruff and Kieffer, 2007). In addition, brain and peripheral functions can interact and influence one another. For example, in the Oxt system, the inability of Oxt_/_ mice to lactate affects their ability to display maternal behaviours.

Phenotypical implication of Oxtr conditional knockout mice

We generated both forebrain-specific Oxtr conditional KO (OxtrFB/FB) and total Oxtr KO (Oxtr"/_) mouse lines. OxtrFB/FB mice were made by crossing floxed Oxtr mice with transgenic mice in which

Floxed mouse Cre mouse

Floxed Mice Cre

Fig. 1. Controlling the Oxtr gene expression by Cre-mediated DNA recombination. Coding sequence of the Oxtr gene allele that is flanked by loxP sites (floxed) is produced by homologous recombination. The Camk2a-Cre promoter drives Cre expression in transgenic mice relatively specifically in the forebrain. By crossing the floxed mice with the Cre transgenic mice, progeny are produced in which the conditional allele is inactivated only in the forebrain. Therefore, the Oxtr expression still remains in other tissues, such as the mammary glands. Cre recombinase recognizes two loxP sites (chevrons) and eliminates the floxed Oxtr exons.

Forebrain Oxtr KO mouse

Fig. 1. Controlling the Oxtr gene expression by Cre-mediated DNA recombination. Coding sequence of the Oxtr gene allele that is flanked by loxP sites (floxed) is produced by homologous recombination. The Camk2a-Cre promoter drives Cre expression in transgenic mice relatively specifically in the forebrain. By crossing the floxed mice with the Cre transgenic mice, progeny are produced in which the conditional allele is inactivated only in the forebrain. Therefore, the Oxtr expression still remains in other tissues, such as the mammary glands. Cre recombinase recognizes two loxP sites (chevrons) and eliminates the floxed Oxtr exons.

Ca2+/calmodulin-dependent protein kinase II alpha (Camk2a) promoter drives expression of Cre. This promoter allows the Cre gene to be expressed relatively specifically in the forebrain (Dragatsis and Zeitlin, 2000). Indeed, effective Cre levels appear between 21 and 28 days after birth, thus eliminating compensation for a deleted Oxtr gene during early development. This Cre-expressing line also significantly spares the main olfactory bulb reducing the potential confound of a deficit in olfaction, which is critical for rodent behaviours (Lee et al., 2008).

Both OxtrFB/FB and Oxtr—/— mice show similar performances in assessments of general health, including reflexes, sensory and motor tests. The OxtrFB/FB mice are able to lactate and display no deficits in maternal behaviour when left undisturbed. Since previously reported Oxtr—/— showed impaired maternal behaviours (Takayanagi et al., 2005), the different phenotypes between OxtrFB/FB and Oxtr—/— mice may be due to the incomplete elimination of Oxtr in the brain regions that underlie expression of maternal behaviour or differences in testing conditions. Interestingly, the OxtrFB/FB and Oxtr—/— show differences in social recognition that we are currently investigating. Besides the phenotypes mentioned above, OxtrFB/FB and Oxtr—/— mice show the same aspects of behaviour including general health and olfaction. Further in depth studies, including assessment of aggression, object memory and maternal behaviour in virgin females are progressing.

Future directions

Impairment of social recognition is strongly related to some neurodevelopmental disorders such as autism (Young et al., 2002). As mounting evidence exists suggesting an association between the Oxt system with autism in humans (Auranen et al., 2002; Shao et al., 2002; Wu et al., 2005; Jacob et al., 2007), the Oxtr KO mouse lines could provide useful model systems to study autism. Precise temporal and spatial regulation of Oxtr will allow finer investigations into the roles of Oxt and Oxtr in various behaviours. This precise temporal and/or spatial manipulation will be possible through the use of transgenic mice or lentiviruses expressing tetracycline-inducible Cre recombinase.

Abbreviations

Was this article helpful?

0 0

Post a comment