| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Division of Radiobiology, Department of Radiology, School of Medicine, University of Utah,Salt Lake City, Utah 84108
ABSTRACT
There are substantial changes in maternal skeletal dynamics during pregnancy, lactation, and after lactation. The purpose of this study was to correlate changes in cortical and cancellous bone mass, structure, and dynamics with mechanical properties during and after the first reproductive cycle in rats. Rats were mated and groups were taken at parturition, end of lactation and 8 wk after weaning, and were compared with age-matched, nulliparous controls. Measurements were taken on femoral cortical bone and lumbar vertebral body cancellous bone. At the end of pregnancy, there was an increase in cortical periosteal bone formation and an increase in cortical volume, but a suppression of turnover in cancellous bone with no change in cancellous or cortical mechanical properties. Lactation was associated with a decrease in cortical and cancellous bone strength with a decrease in bone volume, but an increase in turnover on cancellous and endocortical surfaces. After lactation, there was a partial or full restoration of mechanical properties. This study demonstrates substantial changes in bone mechanics that correlate with changes in bone structure and dynamics during the first reproductive cycle in rats. The greatest changes were observed during the lactation period with partial or full recovery in the postlactational period.
calcium, lactation, parturition, pregnancy
INTRODUCTION
The female reproductive cycle substantially alters maternal mineral metabolism to meet the demands of the mineralizing fetus and to supply calcium for milk production. Although calcium homeostasis is altered during pregnancy, substantial bone loss does not typically occur, except in rare cases of pregnancy-induced osteoporosis [1]. During early pregnancy the intake and intestinal absorption of calcium is increased in many species, resulting in a positive calcium balance [2–4]. In rats, calcium retention during early pregnancy is associated with an increase in bone mass [5]. Changes in bone mass during early pregnancy are typically not significant in humans [6, 7], but a positive calcium balance exists [4]. It has been suggested that the early gain in bone mass may be a mechanism for storing calcium for the impending fetal skeletal mineralization process and for milk production.
In contrast to pregnancy, lactation results in substantial bone loss, particularly at cancellous bone sites. Humans [6, 8–10], rats [5, 11, 12], dogs [13, 14], sheep [15], pigs [16, 17], and monkeys [18] demonstrate varying degrees of bone loss during lactation. The decrease in bone mass is transient, however, and bone mass appears to be at least partially restored after weaning in both humans [6, 19] and rats [12]. Studies in humans suggest that lactation does not increase later fracture risk and may even have a protective effect [20]. Fox et al. [21], for example, reported that bone mass increased by 1.4% at the distal forearm for each pregnancy. There is, however, recent concern about the effects of pregnancy on peak bone mass in adolescents [22].
Histomorphometric studies in rats [5, 23–25] and dogs [13, 14, 26] indicate that bone turnover is significantly elevated during lactation with a greater increase in bone resorption than bone formation. Rats, in particular, have very pronounced bone loss during the first lactation, with cancellous bone volume decreased by as much as 79% at the proximal tibia [5–24]. These decreases are greater than the bone loss observed following estrogen deficiency or immobilization in the rat [24]. Despite the dramatic bone loss, bone mass is rapidly restored after weaning in rats, particularly after the first reproductive cycle [12, 25].
There are few data explaining how the rapid loss and subsequent gain of bone mass affects bone mechanics. Most previous investigations have reported the effects of pregnancy and lactation on bone mechanics at cortical bone sites [27–33], but do not report on the postlactational period. A few studies suggest that cortical bone strength is only partially recovered following lactation [29, 31], which is in contrast to fracture risk in humans. There are limited data regarding changes in cancellous bone mechanics throughout the reproductive cycle. Although cancellous bone mass is at least partially restored after lactation, bone architecture may be altered as a result of the dramatic bone loss [5, 12]. For example, perforations in bone trabeculae may result in an irreversible loss of connectivity leading to decreased bone strength and stiffness. The objective of this study was to determine the effects of pregnancy, lactation, and recovery on bone mechanics and turnover at both cortical and cancellous bone sites in rats.
MATERIALS AND METHODS
Animals
Female Sprague-Dawley rats (Simonsen, Gilroy, CA) were allowed to acclimate to housing conditions and a daily photoperiod of 14L:10D for 5 days. Rats were fed standard laboratory rodent chow (#8640, Harland Teklad, Madison, WI) and water ad libitum. The diet contained 1.17% calcium, 1.0% phosphorus, and 6 IU/g of vitamin D. Rats were randomly divided into six groups; each experimental group contained six to eight rats that were approximately 90 days old at the start of the experiment. Three of the groups were mated and allowed to carry their pups to parturition. Rats from the first group were killed at parturition (about 120 days of age). Rats from the second group were allowed to complete a 21-day lactation and were killed at weaning (about 142 days of age). The third group was allowed to recover from lactation and the animals were killed 8 wk after weaning (about 200 days of age). At each of these times, a group of age-matched, nulliparous animals was killed and served as controls.
All animals received i.p. injections of the fluorochrome bone markers, calcein and tetracycline. Animals were given calcein (fluorescein-methylene iminodiacetic acid, Sigma Chemical Co., St. Louis, MO) at a dose of 10 mg/kg of body weight 8 days prior to necropsy. Tetracycline-HCl (Sigma) was administered 1 day prior to necropsy at a dose of 25 mg/kg of body weight.
Mechanical Testing of Cancellous and Cortical Bone
The right femur and sixth lumbar vertebra were removed at necropsy, wrapped in saline-soaked gauze, frozen, and stored. Specimens were thawed at room temperature in a saline bath prior to mechanical testing. One endplate from each of the sixth lumbar vertebra was mounted on an aluminum stub using a cyanoacrylate glue. The stub was placed in a low-speed bone saw (Isomet; Buehler, Lake Bluff, IL), and two plane parallel cuts were made, removing the endplates from the vertebral body [34]. Some drying of the specimen is likely to have occurred during the gluing process, but specimens were fully rehydrated during sectioning and subsequent mechanical testing. Drying and rewetting bone tissue is reported to have minimal influence on bone mechanical properties [35]. A compressive load was applied along the cranio-caudal axis at a rate of 2 mm/min until failure. Maximum load and flexural rigidity were measured from the load-deformation curve for each specimen.
The femora were placed in a materials testing system (Instron, Canton, MA) and loaded to failure in three-point bending. The central load support was applied to the anterior surface of the midfemoral diaphysis at a loading rate of 10 mm/min. Maximum load, flexural rigidity, and work-to-fracture were measured from the load-deformation curve. Elastic modulus and maximum stress were calculated based on standard engineering equations for three-point bending [36].
Tissue Preparation for Morphometry
The destructive nature of mechanical testing precludes any further analysis of these tissues. To investigate changes in skeletal morphometry and dynamics the contralateral (left) femur and lumbar vertebrae 1 through 3 were collected at necropsy, cleaned of soft tissue, dehydrated in ascending concentrations of ethanol, and embedded undecalcified in methyl methacrylate. Comparisons between contralateral limbs and among vertebral bodies is generally believed to be valid as these intraanimal variations are typically very small. Mid-diaphyseal transverse sections of the femur and longitudinal sections of the lumbar vertebrae were cut on a low-speed bone saw mounted on plastic slides, and ground to approximately 30 µm in thickness. Some sections of the lumbar vertebral bodies were stained by the von Kossa reaction for image analyses.
Histomorphometry of Cancellous Bone
Structural analyses were done using digitized (binary) images captured from von Kossa-stained sections of the vertebral bodies. This was performed using a semiautomated television microscope system [13] and structural analyses were performed using a bone structure program (KSS Computer Consultants, Magna, UT) as used by us in other bone structure studies in the rat [37]. The indices collected were the percentage of total tissue area occupied by bone (% bone) and the perimeter-to area-ratio (Pm/Ar) of the trabecular elements. The Pm/Ar ratio is a measure of trabecular thickness.
For measurements of bone formation indices, a 3.6 mm2 area of vertebral body cancellous bone was measured using a fluorescence microscope (Nikon, Tokyo, Japan) with a camera lucida attachment and a digitizing tablet interfaced with a microcomputer (Apple SE, Cupertino, CA) running histomorphometry software (KSS Scientific Consultants, Magna, UT). The indices measured included the percentage of single fluorochrome-labeled surface, percentage of double fluorochrome-labeled surface, percentage of mineralizing surface, mineral apposition rate (MAR), and the surface- and volume-referent bone formation rates (BFRs and BFRv, respectively). The mineralizing surface was calculated as the doubled-labeled surface plus one-half of the single-labeled surface. The MAR was corrected for section obliquity. The histomorphometric methods and indices are in accordance with standardized procedures and nomenclature [38].
Histomorphometry of Cortical Bone
Structural and dynamic histomorphometric measurements were made on the cross-sections of the mid-diaphyseal femoral shaft and included measurements on both the endocortical and periosteal surfaces. The structural measurements included the total cortical cross-sectional area, the average cortical thickness, the area of the marrow or medullary cavity, and the perimeters of the periosteal and endocortical surfaces.
On the periosteal and endocortical surfaces, histomorphometric indices included the percentage of single-, double-, and mineralizing-surface, the MAR, and the BFRs and BFRv. In many animals from some groups, the endocortical surfaces contained no visible double labels, therefore MAR and BFR are reported only for periosteal surfaces.
Statistical Analysis
Comparisons between experimental groups and nulliparous controls were performed with a Student t-test at each time point. A P value of < 0.05 was considered statistically significant.
RESULTS
Mechanical Testing of Cancellous and Cortical Bone
Maximum compressive load of the lumbar vertebrae showed a transient decrease associated with the reproductive cycle (Fig. 1a). Maximum load was not significantly changed at parturition, but decreased substantially during lactation (64% decrease; P < 0.0001) relative to nulliparous controls. However, by 8 wk after lactation, the bone strength had recovered to near control values. A similar pattern was observed with vertebral flexural rigidity (Fig. 1b): no change at parturition, a significant decline at weaning (56% decrease; P < 0.01), and full recovery 8 wk after weaning. Work-to-fracture remained unchanged at parturition and weaning, but was slightly decreased in mated rats after recovery (Table 1).
|
|
Maximum bending load of the diaphyseal cortical bone of the femur followed a temporal pattern similar to that observed at the lumbar vertebrae (Fig. 1c). Maximum load of the femur was not significantly different between parturition and nulliparous control rats. Maximum load was, however, significantly reduced (26% lower; P < 0.001) in lactating rats at weaning compared with control rats. By the end of an 8-wk recovery period, maximum load had increased, but remained significantly lower than control values (16% lower; P < 0.001). Flexural rigidity was unchanged at parturition, significantly declined at weaning (27% decrease; P < 0.01), and returned to control values by the end of recovery (Fig. 1d). Work-to-fracture remained unchanged at all three time points (Table 1).
Structure and Dynamics of Cancellous Bone
Some of the structural features of the lumbar cancellous bone are presented in Table 2 and illustrated in Figure 2. There was no difference in the percentage of bone at the end of parturition, but as expected, the percentage of bone was significantly less at the end of weaning compared with the nulliparous controls. At 8 wk after weaning (recovery), the amount of cancellous bone had increased and there was no significant difference from controls. The perimeter/area ratios were unchanged at parturition but were significantly elevated at the end of weaning compared with the controls (Table 2). This indicates that the trabecular elements are thinner. At 8 wk after lactation, this had reversed such that the perimeter/area ratios were significantly less compared with controls, which is indicative of thicker trabeculae.
|
|
The reproductive cycle significantly altered normal bone formation rates and turnover in the rat lumbar vertebral body. At the end of pregnancy, all of the indices of bone formation were significantly less than the age-matched, nulliparous controls (Table 3). This would suggest a reduction of bone turnover at the end of pregnancy. After lactation, however, all indices of bone formation rates were significantly greater than those observed in the controls, which is indicative of higher bone turnover rates. At 8 wk after weaning (recovery), none of the indices of bone formation and turnover were significantly different from the controls.
|
Structure and Dynamics of Cortical Bone
Some of the structural features of mid-diaphyseal femoral cortical bone are presented in Table 4. The average cortical cross-sectional area was significantly greater at the end of pregnancy than in the nulliparous controls. None of the other structural indices were significantly different at this time. At weaning, the average cortical cross-sectional area and the average cortical cross-sectional width were less than controls and remained less than controls at 8 wk after weaning. The periosteal circumference perimeter was slightly, but significantly less at this same time, compared with controls.
|
Bone formation indices taken from the periosteal and endocortical surfaces are presented in Table 5. At the end of pregnancy, the percentages of double-labeled surface, mineralizing surface, and surface- and volume-referent bone formation rates were greater than in the nulliparous controls. The opposite was observed on the endocortical surfaces with most indices of bone formation less than those observed in the controls. Double fluorochrome markers were observed in only two of the seven animals at this time, whereas double markers were observed in all of the control animals. Mineral appositional and bone formation rates could be calculated only from these two animals in the parturition group. The mineral appositional rates were about the same as those observed in the controls (data not shown), whereas the bone formation rates in the two animals were less than the mean from the controls (data not shown).
|
At weaning, bone formation activity on the periosteal surface had reversed from that seen at the end of pregnancy (Table 5). The percentage of double-labeled surface, percentage of mineralizing surface, and surface- and volume-referent bone formation rates on the periosteal surface were less than in the nulliparous controls. On the other hand, the percentage of single-labeled surface on the endocortical surface was greater than in the controls. At 8 wk after weaning, there were no significant differences in any of the bone formation and turnover indices at the periosteal or the endocortical surfaces.
DISCUSSION
There are significant changes in bone turnover, structure, and mechanical properties during the first reproductive cycle in the rat. Consistent with previous reports, periosteal bone formation rates were increased at the end of pregnancy [5] and this may have led to the greater average cortical cross-sectional area observed at this time. It is also known that endochondral bone elongation increases during pregnancy, particularly early to midpregnancy, and the increase in periosteal bone formation may be coupled with this increase in bone elongation to maintain the normal dimensionality of the bone [39, 40]. It has been suggested that the increase in endochondral and periosteal growth during pregnancy may be adaptive mechanisms to protect the maternal skeleton for the mineral requirements of the developing fetus and later for milk production during lactation. Contrarily, some indices of bone formation were decreased on the endocortical surface at the end of pregnancy. There were, however, no significant differences in the mechanical properties of the cortical bone at the end of pregnancy, indicating that the changes that occurred on the cortical periosteal and endocortical surfaces during pregnancy had no net effect on bone mechanics.
Similar to the changes on the endocortical surface at the end of pregnancy, some indices of bone formation were less on the endosteal cancellous bone surfaces of the lumbar vertebral bodies at this time. There were, however, no changes in bone volume, average trabecular thickness, or mechanical properties. Cancellous bone formation rates are elevated in early to midpregnancy in the rat [5] and perhaps some bone changes occur in early to midpregnancy that are reversed in later pregnancy as the fetal skeleton draws mineral from the maternal skeleton. Tojo et al. [41] also reported suppressed bone formation rates in the lumbar vertebral cancellous bone at the end of pregnancy in rats. However, they also observed a modest decrease in the amount of trabecular bone that correlated with an increase in bone resorption, which is likely associated with fetal skeletal mineralization in late pregnancy. In humans, there are some reports of an increase in bone turnover in late pregnancy that may be associated with some decreases in bone mineral density [42].
There were significant changes in skeletal dynamics, structure, and strength at the time of weaning. In cancellous bone, all indices of bone formation and turnover were significantly greater at the end of lactation. These included indices of bone formation and turnover and the mineral appositional rate. The high turnover of cancellous bone during lactation has been observed in other studies in rats, dogs [5, 12–14], and primates [43]. The higher bone turnover mechanism during lactation may ensure a readily available supply of mineral for milk production [26]. The loss of cancellous bone during lactation in rats is well documented, and a loss of bone mineral density during lactation has been observed in humans [6–10]. The structural changes that occurred during lactation in cancellous bone likely were the cause of the changes in mechanical properties. Some reductions in cancellous bone mechanical properties during lactation have been reported in the pig [17] but we are unaware of any other data from other species.
Although lactation is known to result in decreases in bone mass and, as demonstrated in the present study, some decreases in biomechanical properties, it is rarely associated with an increase in skeletal fractures or pathological osteopenia in humans. Some studies in humans suggest that the loss of bone mineral density and associated changes in mineral metabolism are "obligatory" in that they are independent of dietary calcium intake [44, 45].
There are rapid gains in bone mineral density after lactation in humans that at least partially restore the skeletal mass loss during lactation [6, 19]. This has also been observed in primates [43] and in rats [12]. We have recently reported that within the first several weeks after weaning in rats, there are substantial increases in bone formation rates that contribute to the rebuilding and restructuring of skeletal mass [25]. The data from the present study show that this postlactational "anabolic" period also restores bone mechanical properties. In the cortical bone of the femur, the mechanical properties were substantially improved over those observed at the end of lactation, but did remain less than those observed in nulliparous controls at 8 wk after weaning. This may not be the case in subsequent reproductive cycles in the rat, because there is evidence that the nulliparous rat has a skeletal mass in excess of that needed for its normal mechanical usage [12]. The first reproductive cycle in the rat is considered to be metabolically "inefficient" [46] and this excess in skeletal mass may be useful in protecting the maternal skeleton during the first reproductive cycle. Thus, in subsequent reproductive cycles, there may be more a complete restoration of skeletal mechanics in the postlactational period. The first reproductive cycle was examined in the present study to allow for comparison with previous work, most of which analyzed the first reproductive cycle. Future investigations into later reproductive cycles in the rat may provide additional insights into the relationship between bone mechanics, turnover, and lactation.
There was a substantial decrease in strength of the lumbar vertebral cancellous bone at the end of lactation, but by 8 wk after weaning it had been restored to levels comparable to controls. Cancellous bone formation rates are rapidly and substantially elevated soon after weaning [25], and this likely contributed to the improvements in cancellous bone mass and structure that were observed after the recovery period. For example, at 8 wk after weaning there was a substantial increase in bone mass (% bone) and a decrease in the Pm/Ar ratios indicative of a thickening of the individual trabeculae. The increased cancellous bone mass and thicker trabeculae would help account for the improvement in mechanical properties after lactation.
The endocrine mechanisms involved in altered bone mechanics during the reproductive cycle are not clear. Dramatic changes in pituitary and ovarian hormones occur during pregnancy and the postpartum period, but none of these have been conclusively linked with changes in bone dynamics. Late pregnancy is associated with a reduction in blood ionized calcium levels in rats and a corresponding increase in the calciotrophic hormones 1,25 dihydroxyvitamin-D (1,25-D), parathyroid hormone (PTH), and calcitonin [47]. The combined influence of these endocrine fluctuations are likely to be small, however, as pregnancy had minimal influence on bone mechanics in the present study.
In contrast, lactation resulted in substantial increases in bone turnover and decreases in bone strength. Numerous endocrine factors could be responsible for these alterations. PTH and vitamin D levels are elevated during lactation in the rat [48], but studies suggest that lactation-induced bone loss is independent of PTH and vitamin D levels [49]. Lactation is also a hypoestrogenic state, and the increased bone turnover and bone loss in lactation is consistent with observations of cancellous bone loss and turnover in ovariectomized rats [24] and dogs [50]. Ovariectomized rats, however, have marked differences in cortical bone dynamics and growth plate kinetics relative to lactating rats [24]. PTH-related peptide (PTHrP) has also been considered as a possible endocrine mediator involved in increased bone turnover during lactation [47]. Mammary secretion of PTHrP is increased during lactation leading to elevated PTHrP levels in maternal serum. Sera levels of PTHrP are negatively correlated with changes in bone mineral density in the spine and femoral neck in humans [51], but no direct link has been established.
The endocrine factors involved in the postlactational recovery period are poorly understood. Relatively few studies have examined this phase of the reproductive cycle, and data have not been correlated with skeletal dynamics. This is of particular interest as this represents one of the most potent "anabolic" phases reported in the adult skeleton.
In summary, this study demonstrates substantial changes in bone mechanical properties during and after the first reproductive cycle in rats. There were substantial decreases in cortical and cancellous bone strength after lactation that correlated with changes in bone structure and dynamics. After lactation, however, there was a rebuilding of both cortical and cancellous bone that resulted in greatly improved mechanical properties. These data demonstrate that the postlactational bone-forming period also greatly improves the mechanical properties of both cortical and cancellous bone.
FOOTNOTES
First decision: 25 January 2001.
1 This work was supported by grant AR-44806 from the National Institutes of Health. 
2 Correspondence: Scott C. Miller, Center for Advanced Medical Technologies,
729 Arapeen Dr., Suite 2334, Salt Lake City, UT 84108-1218.
FAX: 801 581 7008; scott.miller{at}hsc.utah.edu
Accepted: April 10, 2001.
Received: December 4, 2000.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  J. R Speakman The physiological costs of reproduction in small mammals Phil Trans R Soc B, January 27, 2008; 363(1490): 375 - 398. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Wysolmerski Conversations Between Breast and Bone: Physiological Bone Loss During Lactation as Evolutionary Template for Osteolysis in Breast Cancer and Pathological Bone Loss After Menopause IBMS BoneKEy, August 1, 2007; 4(8): 209 - 225. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Ardeshirpour, P. Dann, D. J. Adams, T. Nelson, J. VanHouten, M. C. Horowitz, and J. J. Wysolmerski Weaning Triggers a Decrease in Receptor Activator of Nuclear Factor-{kappa}B Ligand Expression, Widespread Osteoclast Apoptosis, and Rapid Recovery of Bone Mass after Lactation in Mice Endocrinology, August 1, 2007; 148(8): 3875 - 3886. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Miller, B. L. Anderson, and B. M. Bowman Weaning Initiates a Rapid and Powerful Anabolic Phase in the Rat Maternal Skeleton Biol Reprod, July 1, 2005; 73(1): 156 - 162. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. N. VanHouten and J. J. Wysolmerski Low Estrogen and High Parathyroid Hormone-Related Peptide Levels Contribute to Accelerated Bone Resorption and Bone Loss in Lactating Mice Endocrinology, December 1, 2003; 144(12): 5521 - 5529. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
