Acute hypothalamic suppression significantly affects trabecular bone but not cortical bone following recovery and ovariectomy surgery in a rat model

Background. Osteoporosis is “a pediatric disease with geriatric consequences.” Bone morphology and tissue quality co-adapt during ontogeny for sufficient bone stiffness. Altered bone morphology from hypothalamic amenorrhea, a risk factor for low bone mass in women, may affect bone strength later in life. Our purpose was to determine if altered morphology following hypothalamic suppression during development affects cortical bone strength and trabecular bone volume (BV/TV) at maturity. Methods. Female rats (25 days old) were assigned to a control (C) group (n = 45) that received saline injections (.2 cc) or an experimental group (GnRH-a) (n = 45) that received gonadotropin releasing hormone antagonist injections (.24 mg per dose) for 25 days. Fifteen animals from each group were sacrificed immediately after the injection protocol at Day 50 (C, GnRH-a). The remaining animals recovered for 135 days and a subset of each group was sacrificed at Day 185 ((C-R) (n = 15) and (G-R) (n = 15)). The remaining animals had an ovariectomy surgery (OVX) at 185 days of age and were sacrificed 40 days later (C-OVX) (n = 15) and (G-OVX) (n = 15). After sacrifice femurs were mechanically tested and scanned using micro CT. Serum C-terminal telopeptides (CTX) and insulin-like growth factor 1 (IGF-1) were measured. Two-way ANOVA (2 groups (GnRH-a and Control) X 3 time points (Injection Protocol, Recovery, post-OVX)) was computed. Results. GnRH-a injections suppressed uterine weights (72%) and increased CTX levels by 59%. Bone stiffness was greater in the GnRH-a groups compared to C. Ash content and cortical bone area were similar between groups at all time points. Polar moment of inertia, a measure of bone architecture, was 15% larger in the GnRH-a group and remained larger than C (19%) following recovery. Both the polar moment of inertia and cortical area increased linearly with the increases in body weight. Following the injection protocol, trabecular BV/TV was 31% lower in the GnRH-a group compared to C, a similar deficit in BV/TV was also measured following recovery and post-OVX. The trabecular number and thickness were lower in the GnRH-a group compared to control. Conclusion. These data suggest that following a transient delay in pubertal onset, trabecular bone volume was significantly lower and no restoration of bone volume occurred following recovery or post-OVX surgery. However, cortical bone strength was maintained through architectural adaptations in the cortical bone envelope. An increase in the polar moment of inertia offset increased bone resorption. The current data are the first to suppress trabecular bone during growth, and then add an OVX protocol at maturity. Trabecular bone and cortical bone differed in their response to hypothalamic suppression during development; trabecular bone was more sensitive to the negative effects of hypothalamic suppression.


159
160 Bone Mineral Content 161 After mechanical testing, the right femora were flushed with phosphate-buffered saline 162 to discard the marrow. Dry weight of the bones was determined after drying in an oven 163 at 100°C for 24 hr. Ash weight was determined after ashing the bone in a muffle furnace 164 (Fischer Scientific) at 700°C for 24 hr. Ash fraction was calculated as ash weight / dry 165 weight.
166 167 Blood Chemistry 168 C-terminal telopeptides or carboxy-terminal collagen crosslinks (CTX) were measured in 169 serum using an immunoenzymometric assay (Rat-Laps EIA, Immunodiagnostic 170 Systems Inc., Fountain Hills, AZ, USA). The detection limit of the assay was 2.0 ng/mL. 171 Serum insulin-like growth factor 1 (IGF-1) was measured using an 172 immunoenzymometric assay (Rat/Mouse IGF-1, Immunodiagnostic Systems Inc., 173 Fountain Hills, AZ, USA). The sensitivity of the assay was 63 ng/mL. There was a significant interaction between group and time point for peak 230 moment of the femur. The G-R group was 28% stronger following recovery compared to 231 C-R (Table 1). However, no differences in peak moment between GnRH-a and C 232 groups were found following the injection protocol or post-OVX (Table 1). Stiffness in 233 the femur was significantly greater in the GnRH-a groups compared to control. No group 234 differences in mechanical strength were found in the tibia (Table 1).

237
The polar moment of inertia (J) of the femur was significantly larger in the GnRH-238 a groups compared to control. In addition, cortical area, the amount of bone, trended 239 higher (p=0.0619) in the GnRH-a groups. The total area, marrow area and cortical 240 thickness were not significantly different between groups. Polar moment of inertia 241 increased with later time points compared to the first time point following the injection 242 protocol (Table 1). The cortical area did significantly increase with each time point with 293 The relationship between cortical area and polar moment of inertia were not different 294 between groups ( Figure 6C). ). In the current study, bone structure and strength were assessed at maturity 301 following ovariectomy (OVX) surgery while still maintaining the acute hypothalamic 302 suppression via gonadotropin releasing hormone antagonist injections during growth; 303 thus investigating the response of both cortical and trabecular bone growth on bone 304 strength, geometry and mineral at maturity, particularly after the menopause. The 305 hypothesis that environment during bone development affects the structure at maturity 306 was partially supported by our data. Functional bone strength was maintained through 307 architectural adaptations in the cortical bone envelope throughout the lifespan however, 308 a lower trabecular bone mass during growth was retained through recovery and post 309 ovariectomy surgery. Our data support the hypothesis that trabecular bone is more 310 vulnerable to factors negatively affecting growth and are less likely to recover from 311 these deficits. Manuscript to be reviewed 315 recovery the femoral peak moment was greater in the GnRH-a group (G-R) compared 316 to control (C-R). The cortical bone architecture specifically the polar moment of inertia 317 was greater in the GnRH-a groups which rescued bone strength following the delayed 318 puberty. Cortical bone stiffness increased with aging (recovery and post-OVX).
319 However, trabecular bone volume (BV/TV) was significantly lower in the GnRH-a groups 320 compared to control immediately following the injection protocol, after recovery and post 321 OVX surgery. Both trabecular number and thickness were lower for all time points as 322 well.  Schally, 1970). Lower uterine weights are indicative of suppressed estrogen release 332 from the ovaries. Although estrogen levels were not measured in the current study, 333 previous data corroborates suppressed serum estradiol levels using the GnRH-a 334 injection protocol in pre-pubescent female rats (Yingling & Khaneja, 2006;Yingling et 335 al., 2007;Yingling & Taylor, 2008;Saine et al., 2011). In addition, serum CTX levels 336 were significantly increased by 59% in the GnRH-a group an indicator of increased 337 bone resorption also associated with suppressed estrogen secretion. Reduced trabecular bone volume was primarily due to fewer trabeculae following 362 the injection protocol (29% less); the thickness of the trabeculae was only 3-5% thinner 363 in the GnRH-a groups. In osteoporotic women, trabecular bone is typically reduced due 364 to trabecular thinning and loss of trabecular connectivity (Aaron, Makins & Sagreiya, 365 1987;Borer, 2005). A previous study using the GnRH-a protocol at a 100 μg/day 366 dosage measured a loss in percent trabecular bone volume (BV/TV) due to a larger     Manuscript to be reviewed Linear regression of of body weight and bone structural variables for the GnRH-a and control groups.
A) There was a significant relationship between body weight and polar moment of inertia but no difference between groups (C: R 2 =.9188; GnRH-a: R 2 =.7846). B) There was a significant relationship between body weight and cortical area but no difference between groups (C: R 2 =.8909; GnRH-a: R 2 =.9079). C) There was a significant relationship between polar moment of inertia and cortical area but no difference between groups (C: R 2 =.7893; GnRH-a: R 2 =.9043).