This is the first study to examine the emergence of binge eating proneness during puberty in animals. Results revealed dramatic increases in the binge prone phenotype across puberty, such that there was little evidence of individual differences in binge proneness during pre-early puberty, but significant differences during mid-late puberty and adulthood. Developmental effects were robust across two independent samples and categorical as well as continuous definitions of binge prone status. These findings are significant in suggesting that increases in binge eating and eating disorders characterized by binge eating during and after puberty may be at least partially due to biological factors. Indeed, the presence of these phenotypic effects in animals strongly suggests that factors other than psychological influences (e.g., increased body dissatisfaction) contribute to individual differences in binge eating risk in females during puberty.
The BER/BEP rats we identified in adulthood closely resembled those identified previously in terms of their patterns of PF intake, chow intake, and body weight. The BEP phenotype also resembles several aspects of binge eating observed in humans, including a preferential increases in PF (but not chow) consumption, particularly in response to stress (Boggiano et al., 2007
; Oswald et al., in press
). Nonetheless, the percentage (~30%) of adult rats identified as binge prone in our and previous research (Boggiano et al., 2007
; Oswald et al., in press
) is higher than estimates of binge eating in older adolescent and young adult women (~10–19%) (Gauvin, Steiger, & Brodeur, 2009
; Haines, Neumark-Sztainer, Eisenberg, & Hannan, 2006
; Hay, Mond, Buttner, & Darby; Jones, Bennett, Olmsted, Lawson, & Rodin, 2001
). In addition, the BEP rats do not experience the weight losses and gains commonly observed in women with binge eating. These differences highlight the continuing gaps in our knowledge regarding the face validity of the BER/BEP and other animal models of binge eating.
However, an alternative interpretation is that these differences may not be surprising when one considers that BEP rats do not experience the negative environmental (e.g., social disapproval) and psychological (e.g., guilt, self-blame, fears of weight gain) consequences (Fairburn, Marcus, & Wilson, 1993
) of binge eating that are common in women who binge eat. These negative consequences likely act as social and physical constraints on binge eating in women, such that a smaller proportion of women develop binge eating, and they engage in compensatory behaviors (e.g., dieting) that cause physical (i.e., weight fluctuations) and physical/psychological (e.g., weight suppression; Butryn, Lowe, Safer, & Agras, 2006
; Lowe, Thomas, Safer, & Butryn, 2007
) features that we do not observe in BEP rats. In essence, the BEP rats may represent a model of “pure” binge eating proneness as it exists in the population before moderation by environmental/psychological factors that decrease the likelihood of binge eating in all women. Clearly, this hypothesis requires empirical testing, but is an intriguing possibility to consider in future research. Indeed, although no animal model is an exact replica of all binge eating characteristics in humans, the BER/BEP model appears promising for understanding the biological basis of several aspects of human binge eating.
The emergence of the binge prone phenotype during puberty further contributes to the face validity of the model. All rats prefer PF, regardless of developmental stage; this is similar to what we see in humans where a preference for PF (e.g., candy) is present in most girls across development. Importantly, however, we begin to see a divergence in the preference for PF intake during puberty, with BEP rats consuming much more PF than BER rats. This increase in individual differences is similar to what is observed in humans, where binge eating symptoms increase during puberty in some, but not all, girls (Corcos et al., 2000
; Killen, Hayward, Hammer, Wilson, Miner, C.B. et al., 1992
). This emergence previously has been attributed to psychological (e.g., increased negative affect, body dissatisfaction), physiological (e.g., caloric deprivation from increases in dieting), and psychosocial (e.g., increased pressures for thinness) factors (Bulik, 2002
; Garber, Brooks-Gunn, Paikoff, & Warne, 1994
). The emergence of individual differences in binge eating proneness in female rats suggests that biological factors related to puberty are critical for this developmental pattern as well.
The biological factors contributing to puberty’s effects on individual differences in binge proneness are not yet known. Hormone-independent processes cannot be ruled out, but given the myriad hormonal changes associated with puberty in combination with robust hormonal influences on food intake (Asarian & Geary, 2006
), hormones are likely candidates. For example, ovarian hormones and leptin both become elevated during puberty and are linked to homeostatic regulation of food intake (Wilson et al., 1998
). Of these two classes of hormones, it seems more likely that ovarian hormones play a role in the pubertal manifestation of binge eating phenotypes. Ovarian hormones drive pubertal development in females in both rats and humans (Wilson et al., 1998
) and show phenotypic (Edler, Lipson, & Keel, 2007
; Klump, Culbert, Edler, & Keel, 2008
) as well as genetic associations (Klump, Keel, Sisk, & Burt, 2010
) with binge eating. For example, binge eating is negatively associated with estradiol levels, and positively associated with progesterone levels, in clinical (i.e., BN women) (Edler et al., 2007
) and non-clinical (Klump et al., 2008
) samples of women. Associations have been observed across the menstrual cycle where it can be confirmed that changes in ovarian hormones drive changes in binge eating rather than the reverse. These apparent causal relationships are not surprising given extant animal data showing that experimental manipulations of both hormones (via ovariectomy, hormone administration) cause predictable changes in food intake in a variety of species (Asarian & Geary, 2006
). Indeed, a recent study confirmed that estradiol reduces fat intake under binge-like conditions in ovariectomized, adult female rats (Yu, Geary, & Corwin, 2008
Ovarian hormones also exhibit genetic associations with binge eating. Previous research suggests that the heritability of disordered eating is activated at puberty, such that genes account for 0% of the heritability during pre-puberty and ~50% during and after puberty (Culbert, Burt, McGue, Iacono, & Klump, 2009
; Klump, McGue, & Iacono, 2003
; Klump, Perkins, Burt, McGue, & Iacono, 2007
). Follow-up research has confirmed a role for estradiol in these effects. After dividing twins by high versus low estradiol levels during puberty,Klump et al. (2010)
found no evidence for genetic effects on binge eating and disordered eating symptoms in twins with low estradiol levels, but significant genetic effects in twins with high estradiol levels. Findings remained unchanged when controlling for age, body mass index, and the physical changes of puberty (e.g., breast development), suggesting direct effects of estradiol on genetic risk for binge eating and disordered eating.
Taken together, previous data suggest that the emergence of individual differences in binge eating during puberty may be due to increases in ovarian hormones in females during this important developmental stage. These increases may “activate” genetic risk in vulnerable individuals and lead to increased expression of, and individual differences in, binge eating in both rats and humans. Unfortunately, our data are unable to directly examine this hypothesis, as we did not directly manipulate ovarian hormone exposure or examine gene expression. Future animal research should directly examine these possibilities by experimentally manipulating ovarian hormones (e.g., via ovariectomy) before, during, and after puberty to determine whether the emergence of individual differences in binge proneness is dependent upon the presence of these hormones. Ideally, these investigations would also investigate gene expression patterns within the central nervous system (CNS) in order to identify neural systems that contribute to individual differences in binge proneness.
Despite the strengths of our study (e.g., longitudinal data, replication across two samples), there are limitations that must be noted. First, although the BER/BEP model appears promising for understanding biological influences on binge eating, it cannot be determined for certain that the phenotype is the same as that observed in humans. Additional work examining the validity of the model is needed, including continued efforts to model cognitive (e.g., loss of control) and behavioral (e.g., weight suppression; Butryn et al., 2006
; Lowe et al., 2007
) symptoms of eating disorders in BEP versus BER rats.
Second, sample sizes in our BER/BEP groups were small in both experiments. Although our use of categorical and continuous measures of binge proneness partially addressed this concern, future research should examine larger samples of rats to replicate our results. Finally, because of limited resources for this internally funded project, we were unable to identify causal mechanisms underlying puberty’s effects. Additional research should investigate the role of ovarian hormones and other biological factors to understand the mechanisms underlying developmental changes in binge eating across puberty.