Objective
To evaluate the effect of selective estrogen receptor modulators and ethinyl estradiol on the biomechanical and biochemical properties of the uterosacral and round ligaments in the monkey model of menopause.
Study Design
A randomized, double-blind, placebo-controlled study on 11 female macaque monkeys. Ovariectomized monkeys received 12 weeks of placebo, raloxifene, tamoxifen, or ethinyl estradiol. Biomechanical step-strain testing and real-time polymerase chain reaction was performed on the uterosacral and round ligaments.
Results
Tamoxifen and raloxifene uterosacrals expressed differing collagen I /III receptor density ratios, but both selective estrogen receptor modulators showed decreased tensile stiffness compared to ethinyl estradiol and controls.
Conclusion
These findings support a possible effect of selective estrogen receptor modulators on biomechanical and biochemical properties of uterosacrals. This may play a role in pelvic organ prolapse.
Weakness in the pelvic support structures including the uterosacral ligament (USL), cardinal ligament, and paravaginal connective tissues is thought to result in pelvic organ prolapse (POP). Estrogen receptors, ERα and Erβ, are present in the major pelvic support structures in women, including the vaginal wall and uterosacral ligaments, and their expression declines after menopause. Changes in pelvic support ligaments tissue quality is thought to be multifactorial. One factor having an impact on tissue quality may be hormonal, thus menopause is long suspected to contribute to POP; however, little is understood about the role of hormonal status on pelvic support pathology.
Biomechanical studies with animal models in the orthopedic literature have demonstrated that estrogen can change joint ligament stiffness. Both conjugated equine estrogens with medroxyprogesterone acetate and ethinyl estradiol (EE) plus norethindrone acetate have been shown to increase the tensile modulus of the USL and decrease the tensile modulus of the round ligament (RL) in the Macaque monkey model. The tensile modulus is defined as stiffness of the material, or the resistance of the material to uniaxial elongation when a stress is applied to the material in the direction in which it elongates. These changes may allow adaptation to the increase in load imposed by pregnancy without concomitant excessive motion (laxity) of the uterus at the uterosacral junction. This suggests that supportive ligaments may be end-organs for hormonal effect. It may therefore be possible that selective estrogen receptor modulators (SERMs) likewise have an effect on tissue quality.
Currently, SERMs such as raloxifene and tamoxifen are widely used in clinical practice for the treatment of and the prevention of osteoporosis and breast cancer. The use of SERMs has increased after 2002 when the Women’s Health Initiative Study reported a greater risk for cardiovascular adverse effects and breast cancer in women taking estrogen/progesterone supplementation. SERMs include a relatively large number of compounds, each with different profiles of estrogenic/antiestrogenic actions on the genital tract. Two classes of SERMs approved for use by the Food and Drug Administration (FDA) are as follows: (1) raloxifene, a benzothiophene that has estrogen-antagonistic effects in both the breast and uterus and estrogen-like agonistic effects in bone and on serum lipid levels; and (2) tamoxifen, a triphenylethylene that has estrogen antagonistic effects in breast tissue, and stimulatory effects in endometrial tissue. Little is understood about their potential clinical effects on the urogenital system with the exception of the endometrium. Specifically, little is known about the potential effect of SERMs on uterovaginal prolapse and urinary incontinence.
In 2002, our colleagues published a randomized, placebo controlled, double-blind study of 57 postmenopausal women suggesting worsening prolapse with raloxifene and tamoxifen compared with conjugated equine estrogen or placebo, when evaluated by standardized POP quantification methodology. The findings suggested that both tamoxifen and raloxifene increase the incidence of pelvic floor prolapse, although this was not apparent from the licensing studies data for either of the drugs. Increase uterine prolapse and incontinence were also adverse events reported to the FDA in prematurely terminated clinical trials of 2 SERMs that are tamoxifen derivatives, levoremeloxifene and idoxifene. It may therefore be possible that SERMs may likewise have an effect on tissue quality. These studies suggest a relationship between SERMs and POP; however. There are no preclinical studies showing a mechanistic basis for SERMs induction of POP. The effect of SERMs on the tissue properties of pelvic supportive ligaments has been suggested but never evaluated through formal biomechanical and biochemical testing.
The cynomolgus monkey has served as important nonhuman primate models in several studies of aging, including menopause. The pelvic anatomy of the macaque species is almost identical to that of the human, providing opportunities for studying an analogous support system. The animal model allows hormonal manipulation, isolation, and testing of supportive tissues in a way not feasible in human subjects. We used this model to evaluate quantitatively the effects of estrogen deprivation, and the effects of SERMs on ligament composition and tensile properties of the pelvic support ligaments. We compared ovariectomized controls to animals fed raloxifene, tamoxifen, and EE. Supportive ligament receptor density, load deformation and elastic modulus were measured.
We hypothesized that SERMs such as tamoxifen and raloxifene, commonly used in postmenopausal and breast cancer prevention therapy, enhance the hypoestrogenic effects of menopause and may reduce tissue stiffness. The impact of SERMs on biomechanical properties may be correlated to changes in estrogen and progesterone receptor density and collagen type ratio. These translational studies may help elucidate the hormonal factors that impact POP in the postmenopausal female.
Materials and Methods
Experimental design
Randomized, double-blind, placebo-controlled pilot study with outcome measurements performed by persons blinded to animal status ( Figure 1 ).
Animals and treatment protocol
Eleven Indonesian-origin (Bogor, Indonesia) adult female cynomolgus macaques ( Macaca fascicularis ) were obtained from a licensed vendor (Primate Products, Miami, FL). Age as estimated by dentition was 7-13 years with a mean of 9.85 years. The typical life span for this species is 30 years and the average age at menopause is approximately 20 years. Ovariectomies were performed under aseptic conditions using isoflurane anesthesia and appropriate postoperative care and monitoring. Ovariectomized animals were randomly assigned into 4 groups: a control placebo group that received only the vehicle (n = 3); raloxifene 3 mg/kg/d (n= 3); tamoxifen 1 mg/kg/d (n = 2); and EE 3 μg/kg/d (n = 3). Duration of administration was 12 weeks. The doses were scaled from those doses used in women. Drug doses were calculated based on the assumption that an average woman consumes approximately 1800 calories per day. The animal (subjects) were fed a 120 calories/kilogram of body weight atherogenic diet and kept in individual cages. All drugs were administered by training the animals for cooperative oral dosing with the drugs in a flavored aqueous vehicle (Crystal Lite; Kraft Foods Inc, Winnetka, IL; Splenda; Johnson and Johnson/McNeil, New Brunswick, NJ). Serum samples were drawn at 1.5 hours and 24 hours postdosing. After 12 weeks of treatment, animals were sedated with ketamine (10 mg/kg) and humanely euthanized by intravenous injection of pentobarbital (100 mg/kg), followed by assessment of multiple organ systems. The Institutional Animal Care and Use Committee of Wake Forest University approved all procedures involving animals, under accreditation by the Association for the Assessment and Accreditation of Laboratory Animal Care.
Ligament preparation
After necropsy the USL and RLs were harvested. All USLs were ligated with 0 silk in situ at the sacral end to within 0.3 cm to the sacrum, and dissected around the sutures ( Figure 2 ). All tissues for biomechanical testing were wrapped in saline-soaked gauze, sealed in airtight bag, and stored at −80°C. The ligaments were then shipped to Columbia University Center for Orthopaedic Research Biomechanics Laboratory and were stored in −20°C freezers. Before testing, each ligament was thawed for 30 minutes in normal saline solution. During storage, dissection, preparation, and geometric measurement, all specimens were moistened with 0.15 M physiologic saline solution to maintain hydration. Ligaments were tagged with emery cloth at each end with cyanoacrylate and mounted in grips on the Instron 5848 Micro Tester material testing apparatus. All testing was performed with approval from the Institutional Animal Care and Use Committee at Columbia University.
Biomechanical testing
A 10-N load cell with an accuracy of ± 0.005 N was set to zero with the aligned specimen in the lax state. A 0.05-N tare load was applied to determine the initial test length, width, and thickness of the specimen. A grid of known dimensions was placed in alignment with the ligament and 2 digital photographs, orthogonal to each other and the specimen, were taken. The initial length, width, and thickness of the specimen was obtained by analysis of the digital photo using the Image J image analysis software program ( http://rsbweb.nih.gov.easyaccess1.lib.cuhk.edu.hk/ij ; National Institutes of Health, Bethesda, MD).
Mechanical testing was performed using the technique of Akizuki et al, by the same blinded tester, using a computer-controlled Instron 5848 MicroTester (Instron, Norwood, MA). The ligaments were tested at constant step load ramp rate of 1.0 mm/s. A series of step strains, in 5% increments, ranging from 5–30%, were applied to the ligaments with an elongation rate of 1.0 mm/s. The resulting force in the ligament, which was continuously bathed in physiologic (0.15 M) saline solution, was recorded as the ligament underwent stress relaxation at each strain level. Relaxation time was 40 minutes at each strain level. Peak and equilibrium forces were recorded at the progressively increasing strain levels. After the last step strain of 30%, the ligament was unloaded and allowed to recover for 20 minutes. A failure load ramp rate of 1.0 mm/s was then applied to the ligament until the ligament failed. Digital video was obtained to evaluate the mode of ligament failure and to ensure no slippage occurred at the clamp sites.
RNA extraction and amplification
A section of ligament, approximately 3 mm in length, was excised from the ligament before biomechanical testing and homogenized in 1.0 mL Tri-reagent (Molecular Research Center, Inc, Cincinnati, OH) using a Brinkman Polytron (Metrohm USA, Riverview FL) for 30 seconds. RNA was extracted using the Tri-reagent protocol. Isopropanol (500 μL) was added to 1.0 mL aqueous phase and added to a Qiagen RNeasy Mini Kit Column (Qiagen Sciences, Germantown, MD) and the RNA was purified according to Qiagen’s instructions. RNA was eluted from the column in 50 μL DEPC-treated water. RNA quality for samples was determined by Agilent 2100 Bioanalyzer chip analysis. Total RNA (50 ng) from each sample was amplified with a WT-Ovation Pico Kit (Nugen Technologies, San Carlos CA), according to the manufacturer’s instructions. This produced 7.4 ± 1.3 μg amplified DNA. The amplified DNA was diluted 10-fold in nuclease-free water.
Real-time quantitative polymerase chain reaction assay development
Specific quantitative assays for collagen IA1, collagen 2A1, collagen IIIA1, collagen IVA1, estrogen receptor alpha (ERα), estrogen receptor beta (ERβ), elastin precursor (EL), progesterone receptor (PR), and mammalian 18SrRNA pan-specific assay were developed using Beacon Designer software (Premier Biosoft International, Palo Alto, CA) based on the reference sequences from the National Center for Biotechnology Information (NCBI), all primer sequences used for real-time quantitative polymerase chain reaction (qPCR) are shown in Table 1 . All probes were purchased from Biosearch Technologies (Novato, CA), all primers were from IDT (Coralville, IA), and all long oligo PAGE purified DNA standards from Sigma-Genosys (The Woodlands, TX). Technical assay and dual-labeled probe information are provided in Table 2 . Assay design in Beacon Designer used m-fold analysis. The m-fold web server for nucleic acid folding and hybridization prediction was used to avoid secondary structure in the templates. Exon coordinates were based on human sequences. Alignments with the macaque sequences were used as splice junctions because these are greatly conserved among mammalian species.
| Assay | NCBI accession number | Primer sequence a |
|---|---|---|
| Collagen IA1 | XM_001096194 | Forward 2929(+) GATCCCCTGGTGCTGATG Reverse 2997(−) GCTGTCCAGCAATACCTTG |
| Collagen IIA1 | XM_001100278 | Forward 4487(+) CCAATCCCAGTCACTCTAGG Reverse 4572(−) AAAGTCCGAACTGTGAGAGG |
| Collagen IIIA1 | XM_001105968 | Forward 3469(+) GGAGCAAAACAGTCTTTGAATATC Reverse 3556(−) TCAGGACCACCAATGTCATAG |
| Collagen IVA1 | XM_001088660 | Forward 5232(+) TTGTTAACAGCAACGAACCC Reverse 5317(−) GCAAATTCCTATAAGGCACTTTAC |
| ERα | XM_001097228 | Forward 1073(+) GGAATGATGAAAGGTGGGATAC Reverse 1144(−) ATCTCTCTGGCGCTTGTG |
| ERβ | XM_001101433 | Forward 1436(+) TCCTCCCAGCAGCAATCCAT Reverse 1511(−) CCTTGTTACTCGCATGCCTGA |
| EL | XM_001083671 | Forward 798(+) TGGAATCCACATCAGTATCCG Reverse 872(−) AGAGAGACTGCACATCTTGG |
| PR | XM_001095317 | Forward 1694(+) TACCTCAGAAGATTTGTTTAATCTG Reverse 1772(−) TTACAGCTCCCACAGGTAAG |
| Mammalian 18SrRNA | M10098 | Forward 1335(+) CGGCTTAATTTGACTCAACAC Reverse 1401(−) ATCAATCTGTCAATCCTGTCC |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
