|Home | About | Journals | Submit | Contact Us | Français|
Obese female OLETF rats display increased nursing time and frequency compared to lean LETO controls, suggesting a maternal contribution to pup pre-obesity. In previous studies, OLETF pups presented high adiposity, showed greater suckling efficiency, initiative and weight gain from nursing than controls throughout lactation. To further elucidate maternal-infant interactions contributing to pup pre-obesity, we cross-fostered pups a day after birth and examined maternal behavior. Nursing frequency decreased in OLETF dams raising LETO pups (OdLp) in the third postnatal week, while LETO dams raising OLETF pups showed no significant changes. Fat % was greater in the milk of OLETF vs. LETO dams. OdLp pups showed long-term body weight (BW) increase, suggesting that maternal environment can induce BW increases even in the absence of a genetic tendency. Additionally, interaction between OLETF dams and pups produces high nursing frequency, exposing the pups to abundant high-fat milk, thus strengthening their pre-obese phenotype.
Variations in maternal care have frequently been considered to have a critical influence on the offspring’s future development (e.g., Champagne, Francis, Mar, & Meaney, 2003). There are several examples of the importance and influence of the early postnatal environment on the development of obesity. Obesity-prone mice fostered to obesity-resistant dams displayed attenuated obesity, whereas fostering obesity-resistant mice to obesity-prone dams has been shown to induce the development of obesity and insulin resistance (Reifsnyder, Churchill, & Leiter, 2000). Similarly, (diet-induced-) obesity-prone rat pups cross-fostered to obesity resistant dams remained obese but had improved insulin sensitivity as adults; in contrast, obesity resistant rat pups cross-fostered to genetically obese dams showed a diet-induced increase in adiposity, reduced insulin sensitivity and changes in hypothalamic neuropeptide expression (Gorski, Dunn-Meynell, Hartman, & Levin, 2006).
Another factor of importance influencing the emergence of obesity in the offspring is the obesity level of the dam. Maternal obesity during pregnancy and lactation have been found to influence the offspring in the long term, predisposing them to develop larger fat pads and higher leptin and glucose levels later in life, especially in the presence of a genetic tendency (Levin & Covek, 1998; Gorski, Dunn-Meynell, & Levin, 2007). Moreover, the milk of (diet-induced) obese rats has been reported to contain more energy, with more fat but less protein than that of lean rats, and this is also influenced by pre-existing maternal obesity and the diet during pregnancy (Rolls, Gurr, Van Duijvenvoorde, Rolls, & Rowe, 1986). High-fat milk has been found in other animal models of obesity, adding a further factor affecting the early adiposity profile of the offspring (Rolls & Rowe, 1982; Rolls, Van Duijvenvoorde, & Rowe, 1984; Gorski et al., 2006; Javonovic-Peterson, et al., 1989; and Rolls et al, 1986).
The Otsuka Long Evans Tokushima Fatty (OLETF) rat lacks CCK1 receptors due to a spontaneous genetic mutation, and represents a broadly established model of non-insulin dependent diabetes mellitus (NIDDM) (Kawano, 1992; Kawano, Hirashima, Mori, & Natori, 1994) and early-onset hyperphagia-induced obesity (Schroeder, Zagoory-Sharon, Lavi-Avnon, Moran, & Weller, 2006; Schroeder, Shbiro, Zagoory-Sharon, Moran, & Weller, 2009; Moran, 2000). Cholecystokinin (CCK) is one of the most abundant brain-gut peptides, it is produced and secreted in endocrine cells of the intestine and is widely distributed in peripheral nerves of the gastrointestinal tract (Little, Horowitz, & Feinle-Bisset, 2005). The effects of CCK on food intake, digestion and satiety occur through its binding to CCK1 receptors (Moran & Kinzig, 2004; Little et al., 2005). Through this pathway, CCK also reduces feeding in newborn and infant rats in their first independent meal away from the dam (Robinson, Moran, & McHugh, 1988; Smith, Tyrka, & Gibbs, 1991; Weller, 2006; Weller, Smith, & Gibbs, 1990).
OLETF males and females are heavier from birth (Schroeder et al., 2006) and show hyperphagic characteristics in independent ingestion tests as early as PND 2 (Blumberg, Haba, Schroeder, Smith, & Weller, 2006). OLETF pups also consume more milk during nursing bouts whether they are suckling from an OLETF or a LETO dam (Schroeder, Lavi-Avnon, Zagoory-Sharon, Moran, & Weller, 2007b) and also initiate more nursing bouts and this becomes a significant difference during the 3rd postnatal week (Schroeder, Lavi-Avnon, Dagan, Zagoory-Sharon, Moran, & Weller, 2007a). OLETF pups present increased fat percentages, larger adipocytes and waist circumference as early as PND7 compared to LETO controls (Schroeder et al., 2009). In addition to these differences between the pups, there is disparity in the behavior of OLETF and LETO dams that may contribute to the difference in the rate of weight gain of the offspring. OLETF dams spend more time nursing (Schroeder et al., 2007a) and behave differently than LETO dams when interacting with one pup (Lavi-Avnon, Malkesman, Hurwitz, & Weller, 2004).
Previous studies on OLETF females showed that they are obese and hyperphagic during pregnancy (Schroeder et al., 2009) and lactation (Zagoory-Sharon et al., 2008; Schroeder et al., 2009). Although obesity in this strain develops as a consequence of hyperphagia of standard chow (instead of cafeteria/high fat diet in other models), maternal OLETF milk is still likely to contain large amounts of fat. We examined this issue in the current study. In addition, there is evidence for behavioral, non-genomic transmission of postpartum behavior patterns from mothers to female offspring (Champagne, 2008), which in the case of an obese strain might imply excessive nursing, perpetuating the disorder.
It is well known that rat pups stimulate their own nurturance by providing their mother with attractive and appropriate stimuli (Stern, 1989, 1996). In the OLETF strain, our previous findings suggest that the excessive nursing time provided by the OLETF dams to their pups might not only be the result of their own experience, but also of the pups’ demands (Schroeder et al., 2007a, 2007b). Therefore, in the present study we used the cross-fostering strategy in order to further investigate the interaction between dams and pups and how this affects maternal behavior. This manipulation allows at least partial identification of the factors affecting maternal and nursing behavior in the dams rearing pups of both strains, permitting a more clear recognition of the side responsible for the pre-obese features of the pups during the suckling period.
Sixty-nine nulliparous and primiparous OLETF and LETO rats were mated in the Developmental Psychobiology lab’s colony, at Bar-Ilan University, Ramat-Gan, Israel. The original dams were received as a generous gift from the Tokushima Research Institute, Japan. OLETF and LETO pregnant females were housed individually in clean polycarbonate cages (18.5 cm height × 26.5 cm width × 43 cm length), with stainless steel wire lids and wood shavings as bedding material. Food (Koffolk, Tel-Aviv, #19510; 4% fat) and water were freely available. The animals were on a 14:10 hr light: dark cycle, with lights on at 05:00. Room temperature was maintained at 22+/−2 °C. The females were checked daily for parturition. Newborn litters found until 12:00 hr each day were designated as born on that day (PND 0). On PND 1, litters were culled to 10 pups (minimum 8), with sex distribution kept as equal as possible in each litter. At this point, entire (culled) litters were either fostered to another dam, or returned to their natal dam (controls). Follow-up of pup weight and N per group are described below.
Data from dams that failed to deliver adequate (≤8) numbers of pups were not used. The adoption of LETO pups by inexperienced OLETF dams was often unsuccessful; therefore the number of primiparous dams in the OdLp group was rather small. The final distribution was as follows: LdOp: 9 primiparous dams and 4 multiparous dams; OdLp: 3 primiparous dams and 4 multiparous dams; OLETF: 4 primiparous dams and 6 multiparous dams; LETO: 6 primiparous dams and 4 multiparous dams. Despite the small N that resulted from the split of the data by maternal experience, we compared the maternal profile of primiparous and multiparous dams and found that multiparous dams of both strains nursed less frequently, but this did not achieve statistical significance. The primiparous OLETF dams that failed to adopt LETO pups were excluded from the experiment.
Behaviors were rated by two independent observers who were “blind” to the cross-fostering status of the litters. Correlations between the two observers were checked for inter-rater reliability. The research protocol was approved by the Institutional Animal Care and Use Committee, and it adhered to the guidelines of the American Psychological Association and the Society for Neuroscience.
On PND 1, pups were cross-fostered to a dam of the opposite strain. OLETF and LETO dams were observed with their litters in their home cages (N=13 for the LdOp (LETO dams rearing OLETF litters), N=7 for the OdLp (OLETF dams rearing LETO litters), N=10 OLETF controls and N=10 LETO controls), 3 times every week for 3 weeks. Days of observation ranged between 3–7 days postpartum in the first week, 9–13 days postpartum in the second week and 16–20 days postpartum in the third week (observation days were counterbalanced between the groups). The observations took place between 10:00–14:00 hr. During each 240 min. session, each mother was observed every 15 minutes, for 1–3 seconds. This allowed identification of the ongoing maternal behavior at the observation time. Various maternal and non-maternal behaviors were recorded in every observation (see below). The score was “1” if the behavior occurred and “0” if it did not occur.
Maternal Behaviors: measures are based on existing literature and on our studies (e.g., Lavi-Avnon, Yadid, Overstreet, and Weller, 2005; Schroeder et al 2006, 2007a). Maternal measures were divided into pup-directed and self-directed behaviors and were expressed as frequencies.
All littermates (or all but one) are relatively inactive, in contact with each other (or close enough to be covered by the dam’s ventrum).
The dam is crouching over her pups, with at least four of them under her ventrum and attached to her nipples.
All littermates (or all but one) are found suckling during the nursing episode observed.
The dam engages in active behaviors directed towards the pups. She might be hovering over the pups or in near contact with them, licking them or huddling with at least four of them.
Self grooming-the dam is observed cleaning or scratching her body and face using her tongue or paws.
Activity- the dam is standing on her two hind legs/is moving from one side of the cage to the other or is manipulating the wood shavings.
Eating- the dam is holding the food in her front paws and clearly chewing it.
Resting- the dam is located in a distance from all pups, while lying inactive.
Behaviors were analyzed as means per strain/week.
In a separate set of animals, analysis was performed on milk collected the 1st and 3rd PN weeks. On PND6–7 and 18–20 (N=6–10 females per strain/per week), dams were anesthetized with Ketamine- Xylazine (40–80mg/ml Ketamine, 5–10 mg/ml Xylazine), and injected subcutaneously with 4IU Oxytocin (Sigma) to stimulate milk flow. Milking was initiated about 5 min after Oxytocin injection. Milk drops were expressed manually by gentle stroking of the nipple and collected directly into vials. Samples were frozen at −20°C until analyzed.
All milk samples were pooled within each strain and PN week. Lipids content in milk was analyzed by extraction using mixture of 2:1 v/v chloroform: methanol, as suggested by Folch, Lees, & Stanley (1957). Briefly, the samples were dried in a lyophilizer for three days until powder milk was obtained. The dry milk was homogenized with chloroform/methanol (2:1) to a final volume 20 times the volume of the dry milk sample, stirred at room temperature for 30 min and vacuumed filtered. The obtained solution containing the lipids was roto-evaporated to dryness and weighed. Percentage of lipids in milk was calculated by dividing the weight of the fats remaining after the procedure by the weight of the original sample. The entire analysis procedure was made by an investigator “blind” to the group of origin of the samples (week and strain).
Pups remained with the dams until weaning at PND 22 and were housed with their same-sex littermates from then on [N=7 for LETO litters reared by OLETF dams (OdLp), N=14 for OLETF litters reared by LETO dams (LdOp)]. Pup body weights were examined at noon every fifth day from PND 1 to 65. Body weights were compared to those of regular litters (N=15 OLETF and n=18 LETO).
In order to have a control for the fostering procedure itself, 4 LETO litters and 4 OLETF litters were In-fostered on PND 1 to another dam of the same strain. No overall significant differences were found between In-fostered OLETF and LETO males and females and regular OLETF and LETO litters in body weight (data not shown) and maternal behavior (summarized in table 1). For data analysis, the control litters used were the regular, non-fostering groups.
For the undisturbed observations of the litters, a multivariate analysis of variance (MANOVA) was performed for each week’s observations and for each dams’ strain separately, in order to find differences in the patterns of the behaviors examined. In these analyses, the pup’s strain was the independent factor and the behaviors were the dependent measures.
Weights of cross-fostered OLETF and LETO pups were compared to regular, non-fostering litters by repeated measures analysis of variance (ANOVA). In this analysis, the mean weight of the male and of the female pups of each dam was the measure of interest, assessed by separate ANOVAs. The pups’ and dam’s strain (OLETF/LETO) were independent factors, and weight in grams over the 14 time points studied was the dependent, repeated measure. Significant day × strain interactions were followed by post-hoc t-tests comparing the weight of the cross-fostered and regular LETO and cross-fostered and regular OLETF pups at each age.
When examining LETO dams’ maternal behavior, MANOVA revealed an overall effect of strain of the pups only in the second postpartum week (F (8,14) = 2.83, p< 0.05). When examining specific behaviors in all the PN weeks, an increased eating frequency was observed in LETO dams raising OLETF pups compared to controls (F (1,21) = 6.02, p< 0.05) in the first PN week and decreased non-nutritive contact was observed in cross-fostering dams compared to regular LETO dams in the second PN week (F (1,21) = 11.33, p< 0.01). Moreover, in this week OLETF pups raised by LETO dams (LdOp) were more frequently observed in their nest compared to LETO pups raised by their dam (F (1,21) = 12.85, p< 0.01). During PN week 3, no differences in maternal behaviors were found between LETO dams raising LETO or OLETF pups. Self directed and pups directed behaviors are presented in Figure 1.
In OLETF dams, no overall effects of pup strain were observed across the weeks, although during PNW 3, a tendency to significance was observed (F (8,8) = 31.16, p= 0.062). Examination of specific behaviors showed no differences in maternal behaviors of the OLETF dam when raising pups from the different strains on PNW 1 (Fig. 2).
During PNW 2 the only significant difference observed was increased non-nutritive contact towards LETO pups when compared with regular OLETF litters (F (1,15) = 5.00, p< 0.05) (Fig. 2).
During the 3rd PNW, there were significant differences in pups directed behaviors, with OLETF dams showing decreased nursing frequency when raising LETO pups compared to OLETF pups (F (1,15) = 12.78, p< 0.01) (Fig. 2). In the self-directed behaviors, increased self-grooming frequency (F (1,15) = 7.62, p< 0.05) and sleeping (F (1,15) = 5.61, p< 0.05) were observed when OLETF dams raised LETO pups compared to controls (Fig. 2).
The analysis of maternal milk revealed higher amounts of fat in the OLETF strain compared to LETO controls (25.7% vs. 22.2% for PNW1 and 17.6% vs. 12.8% on PNW3 respectively). In addition, LETO milk contained more water, and presented a higher density than that of the OLETF females, suggesting larger amounts of protein and carbohydrates. As expected, fat concentrations decreased from PNW1 to 3 in both strains, to be replaced by larger amounts of water, protein and carbohydrates. The results in the LETO strain are consistent with previous findings characterizing rat milk throughout lactation (Keen, Lonnerdal, Clegg and Hurley, 1981) and the high fat levels found in the OLETF females are consistent with other rat models of obesity (Rolls, Gurr, Van Duijvenvoorde, Rolls, & Rowe, 1986, Gorski et al., 2006; Javonovic-Peterson et al., 1989).
As shown in Figure 3, the growth curve of male pups was affected by maternal strain (F(13,39)=3.84, p≤0.001) and pup strain (F(13,39)=17.27, p<0.001). There was also a significant interaction between age (growth), maternal and pup strains (F(13,39)=3.04, p<0.01). Post hoc t-tests revealed significant differences between the regular and cross-fostered LETO males on PND 5 and 25, when cross-fostered LETO weighed significantly more than regular controls. Even though cross-fostered LETO males tended to weigh more than controls during the whole period, this did not reach statistical significance at the other time points. Cross-fostered OLETF males were not significantly different from the regular OLETF litters at any point during lactation, but weighed significantly less from PND 55 and on.
For the females, the growth curves of the pups were affected by maternal strain (F(13,37)=2.07, p<0.05), pup strain (F(13,37)=12.69, p<0.001), and there was a significant interaction between age (growth), maternal and pup strains (F(13,37)=2.90, p<0.01). Cross-fostered LETO females were significantly heavier than LETO controls from PND 5 to 30 and again from PND55 on. Cross-fostered OLETF females showed significant differences from the regular OLETF litters during the PND 30–45 period, weighing even more than regular OLETF controls. Before and after that period, their body weights did not differ from those of the OLETF control group (Fig. 3).
The OLETF rat develops obesity early in life as a direct consequence of hyperphagia caused by a genetic deficiency. Still, studies in other strains suggest that the maternal environment itself may also have long-term effects on the offspring and can even predispose to diet-induced obesity (Caluwaerts, Lambin, van Bree, Peeters, Vergote, & Verhaeghe, 2007; Samuelsson, Matthews, Argenton, Christie, McConnell, Hansen, Piersma, Ozanne, Twinn, Remacle, Rowlerson, Poston, & Taylor, 2008), even in the absence of a genetic tendency. In previous studies, we found significant differences between the nursing behavior of OLETF and control LETO dams, especially in the first and third postnatal weeks (Schroeder et al., 2006, 2007a, 2007b). In the first PN week, it appeared that OLETF females spent more time nursing both at day and night, by their own initiative. On the other hand, OLETF pups at this age also presented increased suckling efficiency. This enthusiastic suckling may have increased the frequency of milk letdowns. On the third PN week, OLETF dams still presented extended nursing time, but the initiative switched to the pups, which started around 70–80% of the nursing episodes (Schroeder et al., 2007a). Taken together, it seems that beyond the genetic mutation, the interaction between the OLETF dams and pups may exacerbate the offspring’s pre-obese phenotype.
The milk of (diet-induced) obese rats has been reported to contain more energy, with more fat but less protein than that of lean rats; this is also influenced by pre-existing maternal obesity and the diet during pregnancy (Rolls et al., 1986). High-fat milk has been found in other animal models of obesity, adding a further factor affecting the early adiposity profile of the offspring (Rolls & Rowe, 1982; Rolls et al., 1984; Gorski et al., 2006; Javonovic-Peterson et al., 1989; Rolls et al., 1986). While obesity in the OLETF strain develops as a consequence of hyperphagia of standard chow (instead of cafeteria/high fat diet like in other models), we now report for the first time that OLETF milk contains large amounts of fat, both in the first and third PN weeks. Thus, the efficient nursing of OLETF pups from their natal OLETF dam’s high-fat milk can potentially contribute to their pre-obese profile.
The first two weeks of lactation represent a period when pups are completely dependent on their mother and are relatively passive. In the third week, when pups are active and more independent, nursing behaviors demonstrate a more explicit interaction between the pups’ demands and the dam’s innate maternal tendency; and it is here where we saw the most significant effects.
LETO females did not significantly change their behavior when rearing obese instead of lean pups in the first PN week. However, there was an interesting increase in eating frequency when rearing OLETF pups, a finding that might reflect higher energetic demands by the pups. In the second PN week, LETO females provided more non-nutritive contact to LETO compared to OLETF pups, a finding that is consistent with previous findings where LETO pups seem to receive increased non-nutritive contact from the dams, compared to OLETF pups (Schroeder et al., 2006). No further differences were observed in the LETO dams’ behavior. It is interesting that OLETF pups retained a relatively high body weight under these conditions, and while their suckling efficacy may have helped them obtain more milk during the limited nursing time (Schroeder et al., 2007b); their body composition was probably affected. After weaning, OLETF males presented decreased BW and females presented a transitory increase in BW apparently compensating for that “low fat” time.
OLETF dams also behaved almost equally when raising OLETF versus LETO pups during the first two PN weeks. The only significant difference, again, consisted in increased non-nutritive contact toward the LETO pups compared to OLETF pups. Still, the high fat milk provided by the dam during this period permanently affected their body weight. In contrast, a different profile occurred during the third week. OLETF females nursed with less frequency when rearing LETO versus OLETF pups, an outcome that is strongly supported by our previous findings where OLETF pups were found to be responsible for the increased nursing time and frequency they receive at this specific PN week. When rearing “lean” pups that do not initiate so many episodes by themselves, OLETF dams spent less time nursing and more time in self directed behaviors such as self-grooming and sleeping. Still, the decrease in nursing frequency observed was not enough to compensate for the high fat milk obtained and did not prevent these pups from becoming overweight during lactation (females more than males).
In summary, the results suggest that although OLETF rats develop obesity as a consequence of lifelong hyperphagia, the early maternal environment further contributes to their disorder. By providing their pups with high fat maternal milk and frequent nursing throughout lactation (even if this comes as a response to the pups’ demands and not from her natural tendency), OLETF dams worsen the pups’ phenotype. LETO pups reared by OLETF females presented increased body weight throughout lactation (an effect that persisted until adulthood), further supporting the role of maternal environment on the pups’ phenotype, even in the absence of a genetic tendency. On the other hand, LETO dams firmly adhered to their typical behavioral phenotype, by not increasing nursing time according to the nutritional needs of the OLETF pups.
While long-term body weight changes within the strains were not very significant, deeper physiological changes may have taken place after the manipulation that escaped the present measurements. The importance of the present study relies on its contribution to our understanding of the early stages of obesity development in this strain, which serves as a productive model of obesity (Moran, 2008). In light of the current findings, a follow up study will address the changes in the adiposity profile and intake of pups of both sexes from weaning until adulthood, and will potentially elucidate the short- and long-term influences of the postnatal environment on obesity development and resistance.
The authors thank Dr. Kawano of the Otsuka Tokushima Research Institute for the generous gift of the OLETF and LETO rats. This work was supported by grant NIH/NIDDK – RO1 DK57609 (PI: THM, sub-contract: AW).
A portion of this research was presented at the 13th Annual Meeting of the Society for the Study of Ingestive Behavior. Pittsburgh, PA, USA (2005).
The research reported in this paper was completed as part of the first author’s Ph.D. dissertation, in the Faculty of Life Sciences, Bar-Ilan University, Ramat-Gan, Israel