Conditioned taste aversion (CTA) is a form of associative learning in which an animal avoids and reacts aversively to a taste (conditioned stimulus, CS) that has been previously paired with a toxin or other malaise-inducing stimulus (unconditioned stimulus, US). CTA is unique among Pavlovian learning paradigms because conditioning is supported across long delays (minutes to hours) between taste and toxin [Garcia et al., 1966; Kalat & Rozin, 1973; Smith & Roll, 1967] and is robust in that an animal can form a strong aversion that can last for months [Houpt et al., 1996; Martin & Timmins, 1980; Steinert et al., 1980] after only a single trial of a taste-toxin pairing [Garcia & Koelling, 1967]. CTA learning is easily manipulated, as the strength or magnitude of the aversion is dependent on the concentration, saliency, and duration of the CS, as well as the amount or strength of the US [Barker, 1976; Dragoin, 1971; Nachman & Ashe, 1973].
The anatomical pathway involved in CTA is well characterized and includes the nucleus of the solitary tract, parabrachial nucleus, amygdala, and gustatory cortex [for a review, see Yamamoto, 2006]. The two forebrain regions, the amygdala and gustatory cortex, are particularly important as lesions of these areas cause deficits in CTA learning [Josselyn et al., 2004; Nerad et al., 1996], and both structures exhibit cellular changes during and after CTA acquisition, such as induction of long-term potentiation [Escobar & Bermudez-Ratoni, 2000], activation of immediate early genes such as c-fos [Lamprecht and Dudai, 1995], phosphorylation of markers such as MAPK [Berman et al., 1998], phosphorylation of NMDAR subunits such as NR2B [Rosenblum et al., 1997], and changes in the activation and expression of genes such as CREB, fra-2, and fen-1 [Desmedt et al., 2003; Kwon et al., 2008; Saavedra-Rodríguez et al., 2009].
As with other forms of associative learning, CTA is N-methyl-D-aspartate receptor (NMDAR)-dependent [Jimenez & Tapia, 2004]. Pharmacological inactivation of NMDARs attenuates or blocks CTA [e.g. Gutierrez et al., 1999] while activation by NMDAR agonists enhances CTA learning [Land & Riccio, 1997]. Data from our lab show that the NMDAR agonist, D-cycloserine (DCS), dose-dependently enhances taste learning, but only under certain parameters [Nunnink et al., 2007].
Although DCS potentiates NMDAR neurotransmission at the synaptic level, the site and functional mechanism by which DCS enhances CTA are unknown. I have found that DCS enhances short-, but not long-, delay CTA. I further found that prolonged taste exposure prevents DCS enhancement of short-delay CTA, suggesting an interaction between DCS and taste processing [Davenport & Houpt, 2009]. There are many ways in which DCS can act to enhance CTA learning. DCS may enhance the neural processing of the US (LiCl), the CS (taste), or the association of the two through alterations in intracellular events such as changes in protein activation levels or changes in gene expression. The goal of this dissertation is to explore the molecular events associated with DCS and taste learning, thereby elucidating the role of DCS in CTA enhancement as well as helping to characterize key neurochemical players involved in general taste learning.
There were four main studies: (1) To determine whether DCS affects the neural transmission of US (LiCl) processing, I administered DCS followed by LiCl, then measured changes in intracellular LiCl-induced c-Fos induction in central visceral relays using immunohistochemistry. (2) To determine whether DCS affects the neural transmission of CS (taste) processing, I administered DCS followed by a tastant, then measured changes in intracellular taste-induced phosphorylation of MAP kinase in the gustatory cortex and amygdala using immunohistochemistry. (3) NMDARs are disbursed throughout the entire central nervous system. To determine the site of action for DCS enhancement in the brain, I induced excitotoxic lesions in the gustatory cortex, a promising site given its role in processing both gustatory and visceral information, before administration of DCS and a CTA protocol. The effects of DCS on CTA acquisition in lesioned rats were determined daily by 24-hour two-bottle preference tests. (4) CTA learning is associated with changes in gene expression. To determine whether DCS, or the NMDAR antagonist MK-801, leads to alterations in gene activation or expression, I administered DCS or MK-801 then measure NMDAR and serine racemase mRNA using quantitative real time RT-PCR.