点击显示 收起
1 From the Department of Pharmacology and Physiology, New Jersey Medical School, 185 South Orange Avenue, Newark, NJ 07103–2714.
2 Supported by a supplement from the Office of Research for Women’s Health to NIH grant no. AG11403 and grant no. IBN0235011 from the National Science Foundation. 3 Address reprint requests to RP Ferraris, Department of Pharmacology and Physiology, New Jersey Medical School, 185 South Orange Avenue, Newark, NJ 07103–2714. E-mail: ferraris{at}umdnj.edu.
ABSTRACT
Background: Approximately 25% of women retain 5 kg of the weight gained during pregnancy, but the physiologic factors underlying excessive postpartum weight gain are not known.
Objective: The objective of the study was to determine whether pregnancy-related adaptive increases in intestinal nutrient transport are retained after parturition and therefore contribute to postpartum weight gain.
Design: We measured body weight and intestinal nutrient transport in virgin (V, control), primiparous (P, one pregnancy), and multiparous (M, 3 pregnancies) mice at parturition (day 1), during lactation (days 14 and 21), at weaning (day 28), after weaning (day 40), and during aging (days 70, 120, 200, and 300).
Results: In M and P mice, body weight and the weight and length of the small intestine were greatest during lactation; they then decreased but did not return to prepregnancy values until 300 d after parturition. Intestinal villus heights were maximal at lactation and remained high 200 d after parturition. Total intestinal transport capacity for D-glucose, D-fructose, and L-proline was also greatest during lactation, and the lactation-enhanced transport capacity was retained 70 d after parturition. M mice retained more body weight and intestinal transport capacity postpartum than did P mice. Aging per se had little or no effect on body weight or intestinal weight, length, and nutrient transport. The dramatic, lactation-related increases in intestinal nutrient transport capacity were due mainly to increases in intestinal mass.
Conclusions: Postpartum retention of pregnancy- and lactation-related increases in intestinal nutrient uptake capacity may play a significant role in postpartum body weight retention. These adaptive increases may be cumulative and may result in greater weight retention in mice with multiple pregnancies.
Key Words: Intestinal transport • intestinal adaptation • pregnancy • lactation • weight retention • D-glucose • D-fructose • L-proline
INTRODUCTION
Pregnancy and lactation are associated with complex physiologic adaptations that allow the mother to sustain the offspring. Hyperphagia, increased intestinal nutrient absorption, and a reduction in renal or fecal excretion (1) are some adaptations that lead to an increase in maternal fat stores and body weight. Postpartum weight gain in humans averages 1–4 kg per pregnancy (2). After correction for an age-related weight gain of 0.45 kg/y (3), the Institute of Medicine (4) concluded that childbearing is associated with an average weight gain of 1 kg per childbirth. However, 14–25% of women, especially those who have had several children, retained > 5 kg per pregnancy (5). Moreover, 73% of obese patients in the Karolinska Institute retained > 10 kg of weight gained during pregnancy (6). When augmented by the age-related increase in body weight, the postpartum retention of pregnancy-related weight gain will certainly lead to a significant incidence of excessive weight in women. In fact, in over half of overweight women who had normal weight before pregnancy, excessive weight is pregnancy- and lactation-related (7).
The current approach to preventing pregnancy-related obesity is based on weight control during pregnancy, but the main causes of this obesity problem have not been investigated. The weight and surface area of the small intestine increase markedly during pregnancy and lactation in mice, thereby increasing nutrient uptake (8, 9). Unfortunately, there have been no studies of lactation-related changes in human intestinal anatomy and nutrient uptake. However, if the intestine does not adapt during pregnancy and lactation, women may not have the capacity to absorb the nutrients needed to maintain their pregnancy (10). Little is known about the process of down-regulating pregnancy and lactation-related intestinal adaptations. If the up-regulation of intestinal absorption due to offspring sustenance were only partially reversible after completion of lactation and were retained through anestrus or menopause, an abnormally high intestinal uptake capacity and rate of nutrient transport may chronically overload regulatory mechanisms that control the delivery of nutrients to various metabolic pathways, and this process may lead to weight gain. In humans, a more efficient and rapid absorption of a test meal in the upper part of the intestine is associated with obesity (11).
We studied changes in intestinal nutrient transport in mice at different times after parturition. We chose mice because they have been used in a large number of studies of the adaptation of intestinal nutrient transport in pregnancy and lactation (12), and the results were later confirmed in humans (13). Moreover, the mouse has been used extensively in studies of intestinaladaptation to calorie restriction during lactation (13, 14). Finally, the chronic effects of pregnancy and lactation on intestinal function can be practically studied only in animals with a short life span, such as mice [24 mo (15)]. The use of mice allowed our study to span the whole period from parturition through senescence and still to be completed within an acceptable time frame.
We measured intestinal nutrient transport in dams from the day of parturition up to 300 d postpartum in virgin (V, control subjects), primiparous (P, first pregnancy), and multiparous (M, 3 pregnancies) mice.
MATERIALS AND METHODS
Experimental animals
V, P, and M mice were obtained from the breeding colonies of Taconic Farms (Germantown, NY). P and M mice were timed-pregnant. Pregnant mice arrived at our facility =" BORDER="0"> 1 week before parturition, to allow for a reasonable period of adaptation. All mice were kept in microisolators on a schedule of 12 h dark and 12 h light at 25 °C, and they had free access to a complete, sterile standard rodent diet and water. Their food consumption was monitored twice a week, but only the feeding rates of the week before the experiment are presented (eg, the feeding rates of mice on day 200 are an average of the feeding rates calculated between days 193 and 200). Because mice were in regular cages (the enormous number of mice—180—precluded the use of metabolic cages), weanling pups had access to the adult rodent diet, and we cannot exclude the contribution of the pups to the rates of feed consumption attributed to the dam for the span from day 14 (when pups develop teeth) to day 21, when all pups were removed from the cages and killed. Hence, food consumption for that interval could be slightly overestimated, and the feeding rate calculated and presented on day 21 could also be overestimated; the food consumption of the dams between day 7 and day 14, as presented on day 14, should be correct, as we did not observe consumption of the diet by any mouse < 14 d old. By comparing the mean weight of their litters at birth, we ascertained that possible weight differences between P and M mice were due to the number of pregnancies and not to differences in litter size (see Results). There was no significant difference in litter weight at birth or in the number of pups per litter between P and M mice.
P and V mice were exactly the same age (8 wk); M mice are < 30-60 d older. It was not possible, without altering production schedules and procedures (and hence dramatically increasing costs), for 60 timed-pregnant M mice of the exact same age as 60 timed-pregnant P mice and 60 V mice to be delivered in batches of 20 per group. As will be shown in the Results section, the effect of this 45-d (average) difference in age is virtually insignificant as far as major conclusions are concerned.
To minimize the effect of experimental variation, one mouse from each of the P, M, and V groups was killed at each interval, when intestinal nutrient uptakes were determined, so that we could have paired comparisons. Thus, for every experiment, 3 mice of the same age and time after parturition (not fasted overnight) were anesthetized and killed by an overdose of pentobarbital sodium (3.5 mg · kg-1 · body wt-1). The small intestine was gently flushed with cold saline, excised, and everted on a glass rod. Tissues from the proximal (12 cm distal to the pylorus), distal (12 cm proximal to the cecum), and middle (50% of total intestinal length) small intestine were used for the preparation of everted sleeves. Experimental procedures were approved by the Institutional Animal Care and Use Committee of the University of Medicine and Dentistry of New Jersey.
Nutrient transport measurements
Everted sleeves, 1 cm long, were prepared by the method of Karasov and Diamond (16), mounted on a glass rod, and incubated as described by Casirola et al (17). We measured the uptake of 2 sugars (D[14C]glucose and D[14C]fructose) and one test amino acid (L[3H]proline) in the intestinal mucosa. Incubation times were 1 min for D-glucose and 2 min for D-fructose and L-proline, according to the criteria of Karasov and Diamond (16). L[3H]glucose was used to simultaneously correct for adherent fluid and for the diffusive component of total D-glucose and D-fructose uptake (16). We used [14C]polyethylene glycol (molecular weight: 4000) to correct for L[3H]proline in the adherent fluid, so we actually measured mediated plus diffusive L-proline transport. Radioisotopes were from DuPont New England Nuclear (Boston).
Transport results were expressed per milligram of small-intestine wet weight to detect specific changes in rate of transport and per total small-intestine weight to detect changes in total absorptive capacity for each nutrient tested. Because of differences in tissue mass between V, P, and M mice, uptake results were also expressed per centimeter of small intestine. Total absorptive capacity was determined by integrating transport per centimeter along the length of the small intestine by following the method of Casirola et al (17). Test nutrient concentrations (each, 50 mmol/L) were chosen to yield max, a condition in which unstirred layer effects are minimal (16).
Histologic measurements
To evaluate possible changes in mucosal morphology, we measured the weight of and villus height in 1-cm tissue samples from the proximal, middle, and distal regions of the small intestine. These were fixed, paraffinized, and sectioned at 6-µm thickness. Villus heights were measured according to Ferraris and Diamond (18).
Statistical analysis
Results are expressed as means ± SEs. Intestinal weight, intestinal length, villus heights, and total intestinal capacity were analyzed for the simultaneous effects of pregnancy and of time after last parturition by two-factor analysis of variance (ANOVA) and then by Bonferroni post hoc tests. Transport results at each interval after parturition were analyzed by two-factor ANOVA for the simultaneous effects of pregnancy and intestinal position (STATVIEW, version 5; SAS Institute Inc, Cary, NC). By subsequent one-way ANOVA, we analyzed for the effect of pregnancy at each intestinal position.
Because P, M, and V mice were represented at each interval, we were able to normalize the transport rate for P and M mice relative to that for V mice. By normalizing to V mice, we highlighted the potential differences among V, P, and M mice and minimized the effect of individual and batch variation. Transport rates and total intestinal transport capacity were analyzed in V mice for the effects of age by one-way ANOVA. We then used the data from the V mice to help us distinguish the effects of pregnancy from those of age on uptake capacity in P and M mice.
RESULTS
Effects of parturition, lactation, and aging on clinical indexes
Body weight and litter size
Body weight (Figure 1A) was highest around day 14 in both M and P mice. There was a significant effect on body weight of pregnancy, time after parturition, and the interaction between pregnancy and time (P < 0.001 for all). The effect of pregnancy on body weight was significant at each time point from day 1 through day 200 (P < 0.05), but not on day 300 (P = 0.87). M mice were heavier than were V mice from day 1 through day 200 (P < 0.005). P mice were heavier than were V mice on days 7, 14, and 21 (ie, during lactation) (P < 0.0002 for all). Finally, M mice were heavier than were P mice on days 7 and 21 (P < 0.001). These differences in body weight on days 1–21 when M mice > P mice > V mice indicated that previous pregnancies contributed incrementally to increases in body weight during lactation and that the greater the number of previous pregnancies, the greater the body weight at parturition.
FIGURE 1.. Mean (± SE) changes during parturition, lactation, and aging in body weight (A), feeding rate (B), and small-intestine weight (C) and length (D) in virgin (V), primiparous (P), and multiparous (M) mice (n = 6 per group). Time points were parturition (day 1), lactation (days 14 and 21), weaning (day 28), postweaning (day 40), and aging (days 70, 120, 200, and 300). Data for body weight and feeding rate 7 d after parturition were also included in panels A and B. Values for each time point within a panel with different superscript letters are significantly different, P < 0.05 (one-way ANOVA followed by Bonferroni test). By two-factor ANOVA, there was a highly significant effect of time (P < 0.0001) and pregnancy (P < 0.0001). There was also a significant interactive effect between time and pregnancy (A: P = 0.0010; B–D: P < 0.0001). One-way ANOVA in V mice indicated a significant effect of time (aging) only on body weight, which increased after day 70 (P = 0.0014), and on small-intestine length, which increased after day 40 (P = 0.006). A: Pregnancy- and lactation-related increases in body weight were clearly retained for 200 d postpartum, a time that represents one-third of the mouse life span. B: Dramatic increases in feeding rate due to lactation ceased immediately after the pups were removed on day 21. C: Pregnancy- and lactation-related increases in intestinal weight in P and M mice were retained for 200 d postpartum, and they paralleled postpartum retention of body weight. D: Intestinal length also varied with time in P and M mice (P < 0.01 for all). Pregnancy- and lactation-related increases in intestinal length explained some but not all of the increases in intestinal weight in P and M mice; the increases in length were also retained for 200 d postpartum.
At peak lactation (day 14; 9), there was no difference between P and M mice, which indicated that the maximum body weight increase may be independent of the number of pregnancies. However, from day 28 to day 200, M mice retained more weight than did P mice. These differences in body weight between P and M mice did not arise from differences in litter size or litter weight: for P mice and M mice, respectively, the average litter weight at birth was 23.4 ± 0.7 and 25.5 ± 0.9 g (P = 0.07), and the average number of pups per litter was 11 ± 0.4 and 12 ± 0.6 (P = 0.07).
We found a gradual increase in the body weight of V mice with time (P = 0.0014). This increase, which occurred mainly at the end of the study (mostly between days 70 and 300), represented the effect of aging per se. It may confound the marked effect of pregnancy on body weight. However, the body weight of V mice did not change within any 45-d period in the study, which indicated that the small age difference between M and P or V mice did not result in a change of body weight significant enough to confound pregnancy- and lactation-related changes in body weight.
Feeding rate
There was a highly significant effect of pregnancy, time after parturition, and the interaction between pregnancy and time (P < 0.0001 for all) on feeding rate (Figure 1B). The mean feeding rate across all time points was 5.00 ± 0.11, 8.31 ± 0.73, and 8.45 ± 0.79 g/d for V, P, and M mice, respectively (P mice = M mice > V mice). In both P and M mice, the feeding rate was maximal around days 14 and 21, when it was 3 times that in V mice (see Materials and Methods regarding the possible contribution of weanling mice to the estimated feeding rate on day 21). After the pups were weaned and removed from the cages on day 21, the feeding rates reverted to prelactation values. In V mice, the feeding rate did not vary significantly with time (and, thus, not with age).
Small-intestine weight and length
There was a significant effect of pregnancy, time after parturition, and the interaction between pregnancy and time (P < 0.0001 for all) on small-intestine weight (Figure 1C), which was greater in M and P mice than in V mice from day 1 through day 200 (P < 0.001 to P < 0.05). Weight increased dramatically during lactation (days 14 and 21), when small-intestine weight in P and M mice was about twice that in V mice. The effect of time (aging) on the intestinal weight of V mice was not significant (P = 0.103). This suggests that differences in intestinal weight between M or P mice and V mice were not due to age.
Small-intestine length (Figure 1D) was also greater in M and P mice than in V mice. There was a significant effect of pregnancy, time, and the interaction between pregnancy and time (P < 0.0001 for all) on intestinal length. The intestines of P and M mice were longer than those of V mice from day 1through day 200 (P < 0.001 to P < 0.05). It is interesting to note that, in V mice, small-intestine length increased gradually with time (aging) (P = 0.005). However, intestinal length in V mice did not change within any 45-d period in the study, which indicated that differences in intestinal length between V and M mice were not due to the 45-d age difference between the 2 groups.
Sleeve weight
There was a significant effect of pregnancy, time, and the interaction between pregnancy and time (P < 0.004 for all) on intestinal sleeve weights in the proximal, middle, and distal small intestine (Figure 2). On day 14, sleeve weights of M and P mice, which represent the amount of tissue per centimeter of intestine, were significantly greater than those of V mice in the proximal region (Figure 2A); on day 21, sleeve weights of both P and M mice were significantly greater than those of V mice in all regions [Figure 2 (A, B, and C)]. Sleeve weights of M and P mice reached a maximum on day 21 and then reverted to prelactation values but tended to remain greater than those of V mice. Compared with sleeve weights on days 40–120, sleeve weights were also slightly higher on day 300 in all intestinal regions, but there was no pregnancy-related difference, which indicated that aging resulted in a modest increase in mucosal mass in all groups of mice. Measured sleeve weights are in agreement with the calculated ratio of intestinal weight to intestinal length (data not shown).
FIGURE 2.. Mean (± SE) changes in sleeve weights during parturition, lactation, and aging in virgin (V), primiparous (P), and multiparous (M) mice (n = 6 per group) in the proximal (A), middle (B), and distal (C) small intestine. At least 3 sleeves per intestinal region per mouse were examined, and the average was used to represent the intestinal region of that mouse. By 3-way ANOVA, there was a highly significant effect of time, pregnancy, and intestinal region and a significant interactive effect (P < 0.001 for all). Sleeve weight per centimeter decreased along a proximal to distal gradient. By two-factor ANOVA at each intestinal region, there was a significant effect of both pregnancy and time after parturition (P < 0.005 for all). In all 3 positions, the sleeve weight in M mice increased on day 300; however, by one-way ANOVA, the effect of time was significant only in the proximal region (P < 0.0001). In all 3 intestinal regions, weight per centimeter tended to be greater in P and M mice than in V mice at each time point. These modest increases, though mostly significant, accumulated along the proximal to distal regions and eventually contributed to significant increases in total intestinal weight. Values for each time point within a panel with different superscript letters are significantly different, P < 0.05.
Villus height
There was a significant effect of pregnancy and of time after parturition (P < 0.001 for all) on villus height in the proximal and middle small intestine. The average (from all sampling intervals) villus height in the proximal small intestine (Figure 3A) followed the pattern of M mice (646 ± 17 µm) > P mice (625 ± 17 µm) > V mice (564 ± 6 µm) (P = 0.0001). Similar results were obtained in the middle small intestine (Figure 3B), where M mice (450 ± 9 µm) > P mice (367 ± 7 µm) > V mice (342 ± 4 µm) (P < 0.0001). In the distal small intestine (Figure 3C), average villus height varied with time after parturition (P < 0.001 by two-factor ANOVA) but not with pregnancy. There was no significant effect of time (P < 0.001 by two-factor ANOVA) on villus height in the proximal, middle, and distal small intestine of V mice.
FIGURE 3.. Mean (± SE) changes in villus height during parturition, lactation, and aging in virgin (V), primiparous (P), and multiparous (M) mice (n = 10–16 per group) in the proximal (A), middle (B), and distal (C) small intestine. Mice on days 120 and 300 were not sampled because no changes were observed for day 40 or later. Villus heights decreased along a proximal to distal gradient. By two-factor ANOVA, there was a significant effect of both pregnancy and time after parturition in the proximal and middle small intestine (P < 0.001 for all). In the distal small intestine, there was a significant effect of time (P < 0.001) only; pregnancy had no effect.
In both P and M mice, heights of villi in the proximal small intestine initially increased from day 1 after parturition to day 21 and gradually decreased thereafter to day 200. In fact, villus height on days 14 and 21 was significantly greater than that on other days (P < 0.001), which indicated that villus height increased during lactation. In the middle intestine, there was a similar pattern for both P and M mice: mean villus heights were greater on days 1–28 than on days 40–200.
Total intestinal transport capacity for sugars
There was a significant effect of pregnancy, time, and the interaction between pregnancy and time (P < 0.0001 for all) on total intestinal transport capacity for D-glucose and D-fructose. Intestinal capacity for both sugars throughout time after parturition followed the order of M mice > P mice > V mice. Intestinal transport capacity was analyzed for pregnancy effect at each time after parturition and for time effect within each pregnancy group.
D-Glucose
D-Glucose transport capacity increased in P and M mice from day 1 to day 40 (P < 0.01 for all) and reverted to the values in V mice from day 70 to day 300 (Figure 4 A). Capacity was greater in P mice than in V mice from day 14 through day 40. In turn, capacity was greater in M mice than in V mice from day 1 through day 40.
FIGURE 4.. Mean (± SE) total small-intestine transport capacity for D-glucose (A), D-fructose (B), and L-proline (C) during parturition, lactation, and aging in virgin (V), primiparous (P), and multiparous (M) mice (n = 6 per group). Time points were parturition (day 1), lactation (days 14 and 21), weaning (day 28), postweaning (day 40), and aging (days 70, 120, 200, and 300). By two-factor ANOVA, there was a highly significant effect of time and pregnancy and a significant interactive effect between time and pregnancy (P < 0.001 for all) on intestinal D-glucose, D-fructose, and L-proline transport capacity. One-way ANOVA showed that total transport capacity for all 3 nutrients at lactation was significantly higher than that at all other times after parturition. It also showed that lactation-related increases in uptake capacity for all nutrients were retained well past the end of weaning and 70 d after parturition. Values for each time point within a panel with different superscript letters are significantly different, P < 0.05.
In P and M mice, capacity at day 14 was significantly greater than that at all other time points (P < 0.0001 to P < 0.01). Total intestinal transport capacity for D-glucose increased during the period of litter sustenance until after the completion of weaning (day 14). Compared with D-glucose capacity on day 1, that on day 14 was 210% greater in P mice and 153% greater in M mice. Although uptake capacity in M mice tended to be greater than that in P mice, the difference between the uptake capacities of M and P mice for D-glucose at peak lactation was not significant, and so the adaptive response was maximal at maximal energy demand and did not depend on number of pregnancies. In V mice, D-glucose transport capacity did not change with age (P = 0.469 by one-way ANOVA).
D-Fructose
Pregnancy increased D-fructose transport capacity from day 1 through day 70 (P < 0.03 for all), but not from day 120 through day 300 (Figure 4B). Compared with that for D-glucose, the uptake capacity for D-fructose decreased more rapidly, which indicated a short-term adaptation to an increased energy requirement. Capacity for D-fructose in M mice at parturition (day 1) was significantly greater than that in P and V mice, and there was no difference between P and V mice, which again indicated an effect of previous pregnancies. Capacity in P mice was greater than that in V mice on days 14 and 21. Capacity in M mice was greater than that in V mice from day 1 through day 70. The D-fructose transport capacity in V mice did not change with age (P = 0.233), whereas that in both P and M mice peaked on day 14 (P < 0.001 for all).
Total intestinal transport capacity for L-proline
Intestinal L-proline transport capacity varied significantly with pregnancy (M mice > P mice > V mice) and with time (P < 0.0001 for all); there was a significant interaction between these 2 factors (P = 0.0013) (Figure 4C). The L-proline transport capacity was greater in M and P mice than in V mice from day 14 through day 40 (P < 0.05 for all). As was seen earlier for D-glucose and D-fructose transport capacity, L-proline transport capacity was greatest in M mice on day 1 (P < 0.05).
The L-proline transport capacity in V mice did not change with age (P = 0.18). In P mice, L-proline transport capacity peaked on day 21, when it was 150% higher than that on day 1, and it was significantly greater than that on day 28 and thereafter (P < 0.05 for all), which indicated that capacity increased during peak lactation. In M mice, L-proline transport capacity was highest (P < 0.007) on days 14 through 28. Capacity on day 1 was lower than that on days 21 and 28 (P < 0.03 for all), but it did not differ significantly from that on days 40 through 300, which indicated that the capacity for L-proline transport reverted to prelactation values by day 40.
Transport by weight or length of intestine
Nutrient transport by weight (in mg) and length (in cm) in the proximal, middle, and distal intestines of V mice did not change with age, and hence the uptake capacity for D-glucose, D-fructose, and L-proline did not change significantly with age (Figure 4). To determine whether nutrient transport per milligram and per centimeter of intestine in M and P mice changed relative to that in V mice at each time after parturition and to minimize the confounding effect of experimental variation, we normalized transport results in P and M mice to those in V mice at each time and position. This was possible because uptakes in V (control) mice were always determined at the same time as were those in P and M mice.
D-Glucose
When transport was expressed per milligram of intestine and then normalized to V mice, there were no significant differences among M, P, and V mice in the transport rates in any of the 3 intestinal regions at any time after parturition (Figure 5). When transport was expressed per centimeter of intestine, however, significant differences among V, P, and M mice (P < 0.001 to P < 0.05) were found from parturition through postweaning (days 1–28), mainly in the proximal and middle small intestine.
FIGURE 5.. Mean (± SE) ratios of D-glucose uptake in virgin (V), primiparous (P), and multiparous (M) mice to the uptake in V mice (n = 6 per group). The effect of parturition, lactation, and aging on relative intestinal D-glucose transport is shown per milligram and per centimeter of small intestine. Transport results were normalized to V mice transport at each time and in each intestinal region to emphasize the relative magnitude of pregnancy- and lactation-related changes and to minimize the confounding effect of experimental variation. The SEs of V mouse bars were obtained by dividing each V mouse transport by the average V mouse transport value at each time and in each position. Pro, proximal small intestine; Mid, middle small intestine; Dis, distal small intestine. Uptakes were also measured for mice on days 40, 120, and 200, but they are not shown because the results were similar to those in mice on day 300. Glucose transport per centimeter, but not per milligram, in the proximal and middle small intestine increased in P and M mice during lactation, weaning, and postweaning. Bars in the same intestinal region with different superscript letters are significantly different, P < 0.05.
D-Fructose
As were D-glucose transport rates per milligram of intestine, D-fructose transport rates per milligram of intestine were the same in M, P, and V mice, regardless of intestinal region (Figure 6). Significant differences were found between M and V mice when transport was expressed per centimeter of intestine and then normalized to that in V mice. Values for P mice did not differ significantly from those for V or M mice except on day 28, when P mice = M mice > V mice (P < 0.01). On days 1, 14, and 28, uptake in M mice was greater than that in V mice only in the proximal small intestine; on day 21, it was greater in the proximal and middle small intestine. On day 1, uptake in M mice was significantly higher than that in V mice in the proximal small intestine; however, uptake in P mice did not differ significantly from that in either M or V mice.
FIGURE 6.. Mean (± SE) ratios of D-fructose uptake in virgin (V), primiparous (P), and multiparous (M) mice to uptake in V mice (n = 6 per group). The effects of parturition, lactation, and aging on relative intestinal D-fructose transport are shown per milligram and per centimeter of small intestine. Transport results were normalized to V mouse transport at each time and in each intestinal region. Pro, proximal small intestine; Mid, middle small intestine; Dis, distal small intestine. Uptakes were also measured for days 40, 120, and 200, but they are not shown because the results were similar to those in mice on day 300. Fructose transport per centimeter, but not per milligram, in the proximal and middle small intestine increased in P and M mice during parturition, lactation, weaning, and postweaning. Bars in the same intestinal region with different superscript letters are significantly different, P < 0.05.
L-Proline
There were no parturition-related differences in transport per milligram of intestine. When transport was expressed per centimeter of intestine, however, significant differences among V, P, and M mice were found in the proximal or middle small intestine at parturition (P < 0.01), during lactation (P < 0.05), and at weaning (P < 0.05; Figure 7). On day 1, L-proline uptake per centimeter of intestine in the proximal small intestine of M mice was greater than that in P and V mice. On days 14 and 21, L-proline uptake per centimeter of intestine was greater in P and M mice than in V mice.
FIGURE 7.. Mean (± SE) ratios of L-proline uptake in virgin (V), primiparous (P), and multiparous (M) mice to uptake in V mice (n = 6 per group). The effect of parturition, lactation, and aging on relative intestinal L-proline transport is shown per milligram and per centimeter of small intestine. Transport results were normalized to V mouse transport at each time and in each intestinal region. Pro, proximal small intestine; Mid, middle small intestine; Dis, distal small intestine. Uptakes were also measured for days 40, 120, and 200, but they are not shown because the results were similar to those in mice on day 300. Proline transport per centimeter, but not per milligram, in the proximal and middle small intestine increased in P and M mice during lactation, weaning, and postweaning. Bars in the same intestinal region with different superscript letters are significantly different, P < 0.05.
DISCUSSION
The novel findings of this study are that 1) pregnancy- and lactation-related increases in body weight and intestinal weight are retained for a significant portion of the female mouse life span, 2) postpartum adaptations in intestinal nutrient absorptive capacity last well beyond parturition, and 3) the postpartum increase in nutrient uptake capacity per gram of metabolic body weight is due mainly to increases in intestinal mass.
Postpartum retention of pregnancy- and lactation-related increases in body weight and intestinal weight
Mean postpartum body weight typically reaches a maximum at peak lactation, and this increase in body weight is proportional to the summed body weight of the pups, 8 pups (9). After peak lactation, some of the pregnancy- and lactation-related increase in body weight was retained by P and M mice 200 d postpartum, a time that represents 30% of the median life span of a typical mouse (15). Body weights generally followed the pattern of M mice > P mice > V mice, which indicated that postpartum body weight gain is cumulative and proportional to the number of previous pregnancies. By this reasoning, P mice, if impregnated shortly after the weaning of their first litter, would have a greater body weight on day 1 after giving birth to their second litter than they would have on day 1 after giving birth to their first litter. Hence, M mice that gave birth to their third litter weighed almost 40% more than did V mice and 20% more than did P mice.
The lactation-related increases in intestinal weight are also retained long after litter weaning, so that the long-term postpartum retention of an enlarged small intestine is closely correlated with long-term postpartum body weight retention. The retained increase in intestinal weight is due in part to an increase in length, as M and P mice had 20% longer small intestines than did V mice. Most of the increase in intestinal weight, however, may be accounted for by increases in mucosal thickness, as indicated by longer villi and greater tissue weight per centimeter of intestine in P and M mice. Mucosal hyperplasia is known to take place in rats during pregnancy and lactation (19–21), but no studies have been done in humans. Two likely proximate signals for this lactation-related mucosal adaptation could be lactation-associated hyperphagia (larger amount of food in the gut) and secretion of prolactin (22-24). However, the hyperphagia associated with lactation is brief (14 d) and cannot explain the long-term postpartum retention of hypertrophic intestinal mucosa and enlarged intestines in mice. Intestinal mucosal cells are proliferative, so that cell populations are replaced every 3–5 d. There is a need to identify the signal or signals that regulate cell proliferation and, eventually, mucosal hypertrophy during pregnancy and lactation, because reduced effectiveness of such signal or signals may be responsible for postpartum retention of mucosal hypertrophy.
Postpartum adaptation in intestinal absorptive capacity lasts well beyond parturition
Pregnancy and lactation lead to well-documented increases in intestinal nutrient absorptive capacity in rodents (25) and cows (26). This pregnancy- and lactation-related increase in intestinal absorptive capacity for various nutrients not only is retained up to 70 d after parturition but also is cumulative. Had M or P mice been impregnated before 70 d, adaptive increases in intestinal absorptive capacity during the new pregnancy would have supplemented previous adaptive increases for earlier pregnancies. M mice differed from P mice in that the former already had a much higher D-glucose, D-fructose, and L-proline absorptive capacity on day 1, which indicated that that capacity was increased at parturition, because the M mice were likely impregnated before their intestinal absorptive capacity (from a previous pregnancy) was completely down-regulated.
What are the consequences of increased intestinal uptake capacity for D-glucose (and other nutrients)? In this study, there is a strong correlation between body weight and the uptake capacity for D-glucose (Figure 8), D-fructose (not shown), and L-proline (not shown). It has been shown that obesity in humans is strongly associated with a chronically elevated rate of uptake and a more efficient absorption of nutrients in the small intestine (11, 27). In obese humans, enhanced intestinal uptake may be the primary mechanism responsible for an abnormally rapid time course of and a greater amplitude in the increase in plasma nutrient concentrations. The abnormally rapid transfer of energy to the circulation of calories from ingested food may, over time, chronically overload organs, notably the endocrine pancreas and liver, with metabolic and regulatory functions. In fact, altered rates of intestinal nutrient absorption in obese humans cause abnormal plasma concentration patterns of gastrointestinal hormones such as cholecystokinin (28). Altered nutrient transport and hormonal patterns, in turn, may potentially lead to long-term retention of pregnancy- and lactation-related weight gain (6) and eventually to deleterious changes in body fat distribution that are related to menopause (29).
FIGURE 8.. Correlation between intestinal D-glucose uptake capacity and body weight of virgin, primiparous, and multiparous mice (r2 = 0.43, P < 0.0001; n = 150). D-glucose, D-fructose (data not shown; r2 = 0.32, P < 0.0001), and L-proline (data not shown; r2 = 0.26, P < 0.0001) intestinal uptake capacity increased with body weight.
The link between the enhanced rate of intestinal nutrient absorption and postpartum weight gain is supported by findings that chronic consumption of rapidly absorbed refined carbohydrates (eg, simple sugars) increases blood pressure as well as the risk of obesity and cardiovascular diseases, whereas the consumption of complex carbohydrates, which are less rapidly absorbed, lowers blood pressure and reduces the risk of obesity and cardiovascular diseases (30). Two factors then emerge as important regulators of postprandial increases in concentrations of plasma nutrients: the type of diet consumed and the rate of intestinal absorption. The role of diet in postprandial nutrient concentrations and obesity has received much attention and has been the focus of much research effort; unfortunately, the role of the small intestine has been virtually neglected.
Postpartum increases in transport capacity are due mainly to increases in intestinal mass
Metabolism changes as a function of metabolic mass (ie, body weight0.75; 31, 32). The total intestinal uptake capacity normalized to metabolic mass (J/BW0.75) indicates the maximum amount of a nutrient that can be absorbed per unit of metabolic weight, so that
ACKNOWLEDGMENTS
We are deeply indebted to Yvette Suarez and Sandra Basantes for help in the care of the dams and pups and for assistance in the uptake experiments.
DMC was involved in data collection, data analysis, and writing of the manuscript, and RPF was involved in study design, data collection, and writing of the manuscript. Neither author had any financial or personal interest in any company mentioned in this manuscript or in organizations sponsoring this research.
REFERENCES