The transcription factor deltaFosB (ΔFosB), induced in nucleus accumbens (NAc) by chronic exposure to drugs of abuse, has been shown to mediate sensitized responses to these drugs. However, less is known about a role for ΔFosB in regulating responses to natural rewards. Here, we demonstrate that two powerful natural reward behaviors, sucrose drinking and sexual behavior, increase levels of ΔFosB in the NAc. We then use viral-mediated gene transfer to study how such ΔFosB induction influences behavioral responses to these natural rewards. We demonstrate that overexpression of ΔFosB in the NAc increases sucrose intake and promotes aspects of sexual behavior. In addition, we show that animals with previous sexual experience, which exhibit increased ΔFosB levels, also show an increase in sucrose consumption. This work suggests that ΔFosB is not only induced in the NAc by drugs of abuse, but also by natural rewarding stimuli. Additionally, our findings show that chronic exposure to stimuli that induce ΔFosB in the NAc can increase consumption of other natural rewards.
ΔFosB, a Fos family transcription factor, is a truncated product of the fosB gene (Nakabeppu and Nathans, 1991). It is expressed at relatively low levels compared with other Fos family proteins in response to acute stimuli, but accumulates to high levels in brain after chronic stimulation because of its unique stability (Nestler, 2008). This accumulation occurs in a region-specific manner in response to many types of chronic stimulation, including chronic administration of drugs of abuse, seizures, antidepressant drugs, antipsychotic drugs, neuronal lesions, and several types of stress [for review, see Cenci (2002) and Nestler (2008)].
The functional consequences of ΔFosB induction are best understood for drugs of abuse, which induce the protein most prominently in the nucleus accumbens (NAc), a response reported for virtually all types of drugs of abuse (Muller and Unterwald, 2005; McDaid et al., 2006; Nestler, 2008; Perrotti et al., 2008). The NAc is part of the ventral striatum and is an important neural substrate for the rewarding actions of abused drugs. Accordingly, increasing evidence suggests that ΔFosB induction in this region increases an animal's sensitivity to the rewarding effects of drugs of abuse and may also increase motivation to obtain them. Thus, overexpression of ΔFosB in the NAc causes animals to develop place preferences to cocaine or to morphine, or to self-administer cocaine, at lower drug doses, and enhances lever pressing for cocaine in a progressive ratio paradigm (Kelz et al., 1999; Colby et al., 2003; Zachariou et al., 2006).
In addition to its role in mediating drug reward, the NAc has been implicated in regulating responses to natural rewards, and recent work has suggested a relationship between natural rewards and ΔFosB as well. Voluntary wheel running has been shown to increase ΔFosB levels in the NAc, and overexpression of ΔFosB within this brain region causes a steady increase in running, which lasts for several weeks, compared with control animals, whose running plateaus over 2 weeks (Werme et al., 2002). Similarly, a high-fat diet induces ΔFosB in the NAc (Teegarden and Bale, 2007), whereas ΔFosB overexpression in this region increases instrumental responding for food reward (Olausson et al., 2006). Additionally, the fosB gene is implicated in maternal behavior (Brown et al., 1996). However, little information is available on the relationship between ΔFosB and sexual behavior, one of the strongest natural rewards. Moreover, less clear still is the possible involvement of ΔFosB in more compulsive, even “addictive,” models of natural reward behavior. For example, several reports have demonstrated an addiction-like aspect in sucrose intake paradigms (Avena et al., 2008).
To extend our knowledge of ΔFosB action in natural reward behaviors, we investigated the induction of ΔFosB in the NAc in sucrose drinking and sexual behavior models. We also determined how overexpression of ΔFosB in the NAc modifies behavioral responses to these natural rewards, and whether previous exposure to one natural reward can enhance other natural rewarding behaviors.
Materials and Methods
All animal procedures were approved by the Institutional Animal Care and Use Committee of University of Texas Southwestern Medical Center.
Sexually experienced male Sprague Dawley rats (Charles River) were generated by allowing them to mate with receptive females until ejaculation, ∼1–2 times per week for 8–10 weeks for a total of 14 sessions. Sexual behavior was assessed as described previously (Barrot et al., 2005). Control males were generated by exposure to the same arena and bedding, for the same amount of time, as experienced males. Females were never introduced to the arena with these control males. In a separate experiment, an additional experimental group was generated: males were introduced to a hormone-treated female that had not yet entered estrus. These males attempted mounts and intromissions; however, because females were nonreceptive, sexual behavior was not achieved in this group. Eighteen hours after the last session, animals were perfused or decapitated, and brains were taken for tissue processing. For another group of animals, ∼5 d after the 14th session, sucrose preference was tested as described below. For further details, see supplemental Methods (available at www.jneurosci.org as supplemental material).
In the first experiment (Fig. 1a), rats were given unlimited access to two bottles of water for 2 d, followed by one bottle each of water and sucrose for 2 d at increasing sucrose concentrations (0.125–50%). A 6 d period of two bottles of water only followed, then 2 d of one bottle of water and a bottle of 0.125% sucrose. In the second experiment (Figs. 1b,c, 2), rats were given unlimited access to one bottle each of water and 10% sucrose for 10 d. Control animals received two bottles of water only. Animals were perfused or rapidly decapitated, and brains were collected for tissue processing.
Two-bottle choice test.
A two-bottle choice paradigm was conducted as previously described (Barrot et al., 2002). Before surgery, to control for possible individual differences, animals were pretested during the first 30 min of the dark phase for a two-bottle choice procedure between water and 1% sucrose. Three weeks after viral-mediated gene transfer (see below) and before any additional behavioral testing, the animals given only water were then tested for a 30 min two-bottle choice procedure between water and the 1% sucrose solution.
Sexually experienced and control animals did not have a pretest procedure before sexual behavior. Five days after the 14th session of sexual (or control) behavior, animals were given a two-bottle choice test between water and a 1% sucrose solution during the first 30 min of their dark–light cycle. Separate groups of sexually experienced and control animals were used for measuring ΔFosB levels after sexual behavior and for studying the effect of sexual behavior on sucrose preference.
NAc dissections obtained by punch dissection were analyzed by Western blotting as described previously (Perrotti et al., 2004), using a rabbit polyclonal anti-FosB antibody [for antibody characterization, see Perrotti et al. (2004)] and a monoclonal antibody to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (RDI-TRK5G4-6C5; Research Diagnostics), which served as a control protein. ΔFosB protein levels were normalized to GAPDH, and experimental and control samples were compared. For further details, see supplemental Methods (available at www.jneurosci.org as supplemental material).
Animals were perfused, and brain tissues were treated using published immunohistochemistry methods (Perrotti et al., 2005). Because the last exposure to rewarding stimuli occurred 18–24 h before analysis, we considered all FosB-like immunoreactivity, detected with a pan-FosB antibody (SC-48; Santa Cruz Biotechnology), to reflect ΔFosB (Perrotti et al., 2004, 2005). For further details, see supplemental Methods (available at www.jneurosci.org as supplemental material).
Viral-mediated gene transfer.
Surgery was performed on male Sprague Dawley rats. Adeno-associated virus (AAV) vectors were injected bilaterally, 1.5 μl per side, into the NAc as described previously (Barrot et al., 2005). Correct placement was verified after experiments on 40 μm cresyl-violet-stained sections. Vectors included a control expressing only green fluorescent protein (GFP) (AAV-GFP) or an AAV expressing wild-type ΔFosB and GFP (AAV-ΔFosB) (Zachariou et al., 2006). Based on the time course of transgene expression within the NAc, animals were tested for behavior 3–4 weeks after injection of AAV vectors, when transgene expression is maximal (Zachariou et al., 2006). For further details, see supplemental Methods (available at www.jneurosci.org as supplemental material).
Significance was measured using two-factor repeated-measures ANOVAs as well as Student's t tests, which were corrected where noted for multiple comparisons. Data are expressed as means ± SEM. Statistical significance was defined as p < 0.05.
Chronic exposure to sucrose induces increased sucrose intake and sensitization-like behavior
We implemented a two-bottle choice paradigm in which the concentration of sucrose was approximately doubled every 2 d after 2 d of two bottles of water. The sucrose concentration started at 0.125% and increased to 50%. Animals did not show a sucrose preference until 0.25% sucrose, and then drank more sucrose than water at all higher concentrations. Starting at the 0.25% concentration, animals drank increasing volumes of sucrose until a maximum sucrose volume was reached at 5 and 10%. At 20% and higher, they began to decrease their sucrose volume to maintain steady levels of total sucrose consumption (Fig. 1a, inset). After this paradigm, animals spent 6 d with two bottles of water only, and were then presented with a choice of a 0.125% sucrose bottle or water for 2 d. Animals drank more sucrose than water at this concentration, and exhibited a significant sucrose preference compared with the lack of preference observed after initial exposure to this sucrose concentration on day 1.
Because maximum volume intake was reached at a 10% concentration, naive animals were given a choice between one bottle of water and one bottle of 10% sucrose for 10 d and compared with a control group given two bottles of water only. Sucrose animals built to higher levels of sucrose intake by day 10 (Fig. 1b). They also gained significantly more weight after continued sucrose exposure compared with control animals, with the difference in weight increasing over time (Fig. 1c).
Sucrose drinking increases ΔFosB levels in the NAc
We analyzed these animals on the 10% sucrose paradigm for levels of ΔFosB in the NAc by use of Western blotting (Fig. 2a) and immunohistochemistry (Fig. 2b). Both methods revealed induction of ΔFosB protein in this brain region in sucrose experienced compared with control animals. Because the entire ΔFosB protein sequence is contained within that of full-length FosB, antibodies used to detect FosB-like immunoreactivity recognize both proteins (Perrotti et al., 2004, 2005). However, Western blotting revealed that only ΔFosB was significantly induced by sucrose drinking. This indicates that the difference in signal observed by immunohistochemistry represents ΔFosB. The increase observed in Figure 2b was found in the NAc core and shell, but not dorsal striatum (data not shown).
Sexual behavior increases ΔFosB levels in the NAc
We next investigated the effects of chronic sexual behavior on induction of ΔFosB in the NAc. Sexually experienced male rats were allowed unlimited access with a receptive female until ejaculation for 14 sessions over an 8–10 week time period. Importantly, control animals were not home-cage controls, but instead were generated by similar handling on testing days and exposure to the open field arena and bedding in which copulation occurred for the same amount of time but without exposure to a receptive female, controlling for olfaction and handling effects. Using Western blotting, we found that sexual experience significantly increased levels of ΔFosB compared with the control group (Fig. 2a), with no detectable levels of full-length FosB observed. Consistent with these data, immunohistochemistry revealed an increase in ΔFosB staining in both core and shell of the NAc (Fig. 2c), but not dorsal striatum (data not shown).
To ensure that the increase in ΔFosB observed in the sexually experienced animals was not attributable to social interaction or some other non-mating-related stimulus, we generated nonmating males, which were exposed to hormone-treated females, but not allowed to copulate. These males showed no difference in ΔFosB levels compared with a separate set of olfaction-arena control animals (Fig. 2a), suggesting that ΔFosB induction occurs in response to sexual behavior and not social or nonmating cues.
Overexpression of ΔFosB in the NAc increases sucrose intake
Using a viral-mediated overexpression system, which enables the stable expression of ΔFosB over several weeks (Zachariou et al., 2006) (Fig. 3a), we investigated the influence of higher levels of ΔFosB, specifically targeting the NAc, on sucrose drinking behavior (Fig. 3b). We first insured that there were no differences in baseline sucrose behavior before surgery with a sucrose intake pretest (AAV-GFP, 6.49 ± 0.879 ml; AAV-ΔFosB, 6.22 ± 0.621 ml; n = 15/group; p > 0.80). Three weeks after surgery, when ΔFosB expression had been stable for ∼10 d, animals were given a postsurgery sucrose test. The AAV-ΔFosB group drank significantly more sucrose than the AAV-GFP control group (Fig. 3b). There was no difference in the amount of water intake between the two groups (AAV-GFP, 0.92 ± 0.019 ml; AAV-ΔFosB, 0.95 ± 0.007 ml; n = 15/group; p > 0.15), suggesting that the effect of ΔFosB is specific to sucrose.
Overexpression of ΔFosB in the NAc influences sexual behavior
Next, we examined whether ΔFosB overexpression in the NAc regulates sexual behavior of naive and experienced animals. Although we did not find any differences in parameters of sexual behavior between AAV-ΔFosB- and -GFP-treated experienced animals (see supplemental Table S1, available at www.jneurosci.org as supplemental material), overexpression of ΔFosB in naive animals significantly decreased the number of intromissions required to reach ejaculation for the first sexual behavior experience (Fig. 3c). There was also a trend for a decrease in the postejaculatory interval for the ΔFosB group after the first sexual experience (Fig. 3c). In contrast, no differences were observed in latencies for mounts, intromissions, or ejaculation in either the naive or experienced animals (see supplemental Table S1, available at www.jneurosci.org as supplemental material). Similarly, no difference was observed for the intromission ratio [number of intromissions/(number of intromissions + number of mounts)], although this may be because of the high variability in the number of mounts in each group.
Sexual experience increases sucrose intake
Because we found an increase in ΔFosB levels in the NAc after both sucrose drinking and sexual experience, and ΔFosB overexpression influences behavioral responses to both rewards, it was of interest to explore whether previous exposure to one of the rewards significantly affected behavioral responses to the other. Before sexual experience, naive animals were randomly assigned to control or sex conditions. Animals were then exposed to sexual experiences or control conditions, as described previously, over 8–10 weeks. Five days after the last sex session, animals were subjected to a 30 min two-bottle choice paradigm between one bottle of water and one of sucrose. We found that the sexually experienced animals drank significantly more sucrose than controls (Fig. 3b). No difference between sexually experienced and control animals was observed with water intake (control, 1.21 ± 0.142 ml; sex experienced, 1.16 ± 0.159 ml; n = 7–9; p = 0.79), suggesting that the effect is specific to sucrose.
This study bridges a previous gap in the literature in elucidating the role of ΔFosB in natural reward behaviors related to sex and sucrose. We first set out to determine whether ΔFosB accumulates in the NAc, a crucial brain reward region, after chronic exposure to natural rewards. An important feature of this work was giving the animals a choice in their behavior, by analogy to drug self-administration paradigms. This was to ensure that any effect on ΔFosB levels was related to voluntary consumption of the reward. The sucrose model (Fig. 1) demonstrates aspects of addiction-like behavior compared with other sucrose intake models: a choice between reward and control, an inverted U-shaped dose–response curve, a sensitized response after withdrawal, and excessive intake. This model also causes increased weight gain, not seen in other models such as the daily intermittent sugar model (Avena et al., 2008).
Our data establish, for the first time, that two key types of natural rewards, sucrose and sex, both increase ΔFosB levels in the NAc. These increases were observed by Western blotting and immunohistochemistry; using both methods insures that the observed protein product is indeed ΔFosB and not full-length FosB, another product of the fosB gene. The selective induction of ΔFosB by sucrose and sex is similar to the selective induction of ΔFosB in the NAc after chronic administration of virtually all types of drugs of abuse (see Introduction). Of note, however, is the observation that the degree of ΔFosB induction in the NAc observed here in response to natural rewards is smaller than that seen for drug rewards: sucrose drinking and sexual behavior produced a 40–60% increase in ΔFosB levels in contrast to the severalfold induction seen with many drugs of abuse (Perrotti et al., 2008).
The second objective of this study was to investigate the functional consequence of ΔFosB induction in the NAc on natural reward-related behaviors. Much of our previous work on the influence ΔFosB on drug reward has made use of inducible bitransgenic mice, in which ΔFosB expression is targeted to the NAc and dorsal striatum. These ΔFosB-overexpressing mice show enhanced behavioral responses to cocaine and opiates, as well as increased wheel running and instrumental responding for food (see Introduction). In this study, we used a more recently developed viral-mediated gene transfer system to stably overexpress ΔFosB in targeted brain regions of male rats (Zachariou et al., 2006). We found here that ΔFosB overexpression increased sucrose intake when compared with control animals, with no differences in water intake between the two groups.
We also investigated how ΔFosB influences sexual behavior. We demonstrated that ΔFosB overexpression in the NAc decreases the number of intromissions required for ejaculation in sexually naive animals. This did not correspond with other differences in naive sexual behavior, including alterations in mount, intromission, or ejaculation latencies. In addition, ΔFosB overexpression did not affect any aspect of sexual behavior in sexually experienced animals. The ability of a manipulation in the NAc to influence sexual behavior is not surprising given the growing evidence that this brain reward region regulates sexual behavior (Balfour et al., 2004; Hull and Dominguez, 2007). The ΔFosB-induced decrease in number of intromissions could reflect an enhancement of sexual behavior, in that naive animals with ΔFosB overexpression in the NAc behave more like experienced animals. For example, in tests of repeated sexual experience, animals require fewer intromissions to reach ejaculation (Lumley and Hull, 1999). In addition, the trend for a decrease in the postejaculatory interval with ΔFosB overexpression also reflects behaviors observed in more sexually motivated, experienced males (Kippin and van der Kooy, 2003). Together, these findings suggest that ΔFosB overexpression in naive animals may facilitate sexual behavior by making naive animals resemble more experienced or sexually motivated animals. On the other hand, we did not observe a significant effect of ΔFosB overexpression on experienced sexual behavior. More complex behavioral studies of sexual behavior (e.g., conditioned place preference) may better discriminate possible effects of ΔFosB.
Last, we investigated how previous exposure to one natural reward affects behavioral responses to another. Specifically, we determined the effect of previous sexual experience on sucrose intake. Although both control and sexually experienced animals showed a strong preference for sucrose, the sexually experienced animals drank much more sucrose, with no change in water consumption. This is an interesting finding, in that it suggests that previous exposure to one reward may heighten the rewarding value of another rewarding stimulus, as would be expected if there was a partly shared molecular basis (e.g., ΔFosB) of reward sensitivity. Similar to this study, female hamsters previously exposed to sexual behavior exhibited enhanced sensitivity to the behavioral effects of cocaine (Bradley and Meisel, 2001). These findings support the notion of plasticity within the brain's reward circuitry, in that the perceived value of present rewards is built on past reward exposures.
In summary, the work presented here provides evidence that, in addition to drugs of abuse, natural rewards induce ΔFosB levels in the NAc. Likewise, overexpression of ΔFosB in this brain region regulates an animal's behavioral responses to natural rewards, as has been observed previously for drug rewards. These findings suggest that ΔFosB plays a more general role in the regulation of reward mechanisms, and may help mediate the cross-sensitization observed across many types of drug and natural rewards. As well, our results raise the possibility that ΔFosB induction in the NAc may mediate not only key aspects of drug addiction, but also aspects of so-called natural addictions involving compulsive consumption of natural rewards.
This work was supported by grants from the National Institute of Mental Health and National Institute on Drug Abuse and from the National Alliance for Research in Schizophrenia and Depression.
- Correspondence should be addressed to Carlos A. Bolanos at the above address.