Pulsed administration for physiological estrogen replacement in mice

Animals

C57BL/6J mice (Taconic, Denmark) were kept in the animal facility at the University of Gothenburg (Sweden) under regular lighting conditions (12 h light/dark cycles), fed with soya-free laboratory chow and tap water ad libitum. Mice were acclimatized for 7 days before initiating the surgical procedures. The experiments were carried out as described in Figure 1. All the experimental procedures were performed in accordance with the ethical permit approved by the Regional Ethical Review Board in Gothenburg, Sweden (Dnr: 2814/2020), which included also the criteria for the earlier termination of the experiment (weight loss and signs of pain and distress), and according to the Institutional Animal Care and Usage Committee (ARRIVE guidelines²¹). The sample size (n=3group) was decided to minimize the number of animals and allow the statistical analysis. Animals were randomly allocated in the different experimental groups and randomly sacrificed to minimize potential confounders.

Figure 1. Schematic description of the experimental plan.

This figure describes the timeline of the experiments and the experimental groups included in the study.

Ovariectomy

In this study, 96 female mice, 8 weeks old, were employed and underwent OVX or sham surgery. Mice were anesthetized by a mix of isoflurane and oxygen (1.5-2.5 %). A 10 mm incision was made through the shaved and disinfected skin at the lower back area. Then, a 5 mm incision was done in the peritoneum on one side to carefully pull out the ovarian fat pad and excise the ovary by cauterization. The incision in the peritoneum was closed using absorbable sutures. The procedure was repeated on the other side with the second ovary. The external wound on the skin was closed by surgical clips. In mice undergoing sham surgery, the same procedure was performed, but without excising the ovaries. Meloxicam (5mg/kg) was used as a postoperative analgesic.

Preparation of solutions and estradiol treatments

The oil-based stock solution of E2 (17β-estradiol-3-benzoate, SIGMA Aldrich) (1mg/ml) was prepared by mixing E2 with inert miglyol oil (Miglyol812 OmyaPeralta GmbH, Hamburg, Germany). E2 was dissolved by stirring the solution for 3 hours at 150°C and then further diluted with miglyol to 0.5 µg/ml and 1.5 µg/ml concentrations.

For the PBS-based formulation, a stock solution (1mg/ml) of E2 dissolved in absolute ethanol was prepared and stored at -20°C. At the time of injections, the stock solution was diluted in PBS to 0.5 µg/ml and 1.5 µg/ml final concentrations.

Mice were allowed to recover from the OVX procedure for 10 days before the initiation of treatments. Sham and OVX mice were divided into different treatment groups receiving subcutaneous (s.c.) injections (100 µl) every four days of E2 (0.05 μg or 0.15 μg/mouse/injection) dissolved in miglyol or PBS, or vehicle (miglyol or PBS). Mice were sacrificed at 2, 4, 6 or 8 weeks after the start of treatments.

Termination of experiments and tissue collection

Throughout the experiment, all the efforts were made to minimize the pain and distress of the experimental animals. For this purpose, mice were anesthetized with a mixture of ketamine / dexmedetomidine hydrochloride before the procedure. Their body composition was determined using dual-energy x-ray absorptiometry (DXA) scan (UltraFocus^DXA, Faxitron Bioptics, Tuscon, AZ). Anesthetized mice were euthanized by exsanguination followed by cervical dislocation. Blood samples were collected for serum isolation. Uterus, thymus, spleen, and liver were dissected, and weights were noted. Femurs were collected for micro-computed tomography (µCT) analysis.

Micro-computed tomography

After sacrifice, the left hindleg was excised and the femur dissected. The soft tissue was carefully removed from the bone. Samples were fixed in PFA 4% for 3 days and then stored in 70% ethanol. The area for the trabecular bone evaluation was selected starting at 500 µm from the growth plate and extending a further longitudinal distance of 192 µm in the proximal direction. Cortical measurements were performed in the diaphyseal region of the femur starting at 5195 µm from the growth plate and extending a further longitudinal distance of 192 μm in the proximal direction. The selected area was evaluated in a scanning tube providing a voxel size of 4.49 µm isotropically and scanned at 50 kV, 200 µA (Skyscan 1172 scanner; Bruker MicroCT, Aartselaar, Belgium). Samples were kept in paper soaked in PBS to avoid dehydration. Analysis of the morphology and measurement of bone features by μCT was performed using the software CtAN (1.13.2.1, Bruker microCT, RRID: SCR_021338).

Motor tests

To assess the effect of E2 replacement on locomotor ability, two motor tests were performed 2 days before termination of the 8 weeks of treatment experiment. Mice were acclimatized in the procedure room for 1 hour before the start of the tests.

Spontaneous locomotor activity was analyzed using the open-field test. Each mouse was placed into the center of a 60 × 60 × 60 cm chamber to allow free exploration. The experiments were performed for 15 min. The motor parameters were measured by computerized analysis at 3 min intervals. Mice behavior was recorded, and videos were analyzed by using Viewer software (RRID:SCR_014337; Biobserve, Germany).

Forced locomotor activity was tested by using the rotarod test. Mice were placed on a rotarod apparatus (Panlab, Harvard Apparatus, Spain) and tested for 5 min with constantly increasing acceleration from 4 to 40 r.p.m. The latency to fall was registered for each animal. To exclude differences in learning skills between the groups of mice, each group was assessed over three trials per day for 2 consecutive days. Mice were given a 30 min inter-trial rest interval.

Sex steroid measurements in serum

Peripheral blood from all mice was collected at termination in 500 µl tubes containing serum gel with clotting activator (Microvette 500 Z-Gel, Sarstedt). The serum was extracted and stored at -80° C until use. Steroids were extracted from serum (200 µl) and concentrations of estradiol, estrone, progesterone, 17β-hydroxiprogesterone, testosterone, dihydrotestosterone (DHT) and androstenedione were analyzed by Liquid Chromatography-Mass Spectrometry (See underlying data,²²).

Data analysis

Data on levels of sex steroids in serum are expressed as median (range) with non-detectable values represented as half of the lowest level of quantification (LLOQ). Statistical analysis was not performed on serum steroid levels due to the presence of values under the detection limit for some hormones, in combination with the low number of samples per experimental group. For all other data Gaussian distribution was assumed and results are expressed as mean ± SEMGraphPad Prism software (RRID:SCR_002798; GraphPad, San Diego, CA). No animals were excluded from the data analysis. Among the authors of the manuscript, CC was aware of the group allocation at the different stages of the experiment.

Effects of OVX and pulsed E2 treatment on serum levels of sex steroids

Mice were subjected to sham or OVX surgery and treated with s.c. injections every 4 days with vehicle (miglyol or PBS) or two different doses of E2 in the corresponding vehicle (Figure 1). The concentrations of estradiol, estrone, progesterone, 17α-hydroxiprogesterone, testosterone, dihydrotestosterone (DHT), and androstenedione were measured in serum from mice terminated at 8 weeks after start of the treatments.

Estradiol and estrone. After 8 weeks of treatment, a reduced concentration of estradiol was shown in the miglyol OVX+veh group compared with miglyol sham+veh mice, and both doses of E2 dissolved in miglyol increased the serum level of estradiol compared to OVX+veh. However, in mice receiving the PBS formulation no differences between OVX+veh and any of the other groups could be detected. A comparison between the miglyol sham+veh and PBS sham+veh groups showed that the PBS treated sham group had a tendency towards lower serum concentration of estradiol (Figure 2A).

Figure 2. Levels of estradiol and estrone in serum after 8 weeks of treatments.

The amount of estradiol (A) and estrone (B) was measured in Sham and OVX mice after 8 weeks of s.c. injections every 4 days with vehicle (miglyol or PBS) or 17 β-estradiol (E2; 0.05 µg/mouse or 0.15 µg/mouse). Data are expressed as median (range); n=3 for each experimental group. (nd = not detectable; LLOQ=lower level of quantification).

The fluctuations of estrone levels in the miglyol preparation groups, showed a similar pattern to that described for estradiol. Estrone levels were generally low, and under the detection limit in the miglyol OVX+veh group. Treatment with both doses of E2 dissolved in miglyol increased the amount of serum estrone compared to OVX mice. However, animals injected with the PBS preparation of E2 displayed estrone values below the detection limit (Figure 2B).

Progesterone and 17β-hydroxiprogesterone. As expected, OVX induced a reduction in serum progesterone levels. Treatments with E2, dissolved in miglyol or PBS, did not affect the serum concentrations of progesterone (Figure 3A).

Figure 3. Levels of progesterone and 17α-hydroxyprogesterone in serum after 8 weeks of treatments.

The amount of progesterone (A) and 17α-hydroxyprogesterone (B) was measured in Sham and OVX mice after 8 weeks of s.c. injections every 4 days with vehicle (miglyol or PBS) or 17 β-estradiol (E2; 0.05 µg/mouse or 0.15μg/mouse). Data are expressed as median (range); n=3 for each experimental group. (LLOQ = lower level of quantification).

Serum levels of 17β-hydroxiprogesterone were unaltered between the groups after 8 weeks of treatment (Figure 3B).

Testosterone, dihydrotestosterone (DHT) and androstenedione. The levels of testosterone tended to decrease in OVX+veh groups for both formulations compared to the sham+veh groups in mice terminated after 8 weeks. Testosterone concentrations were not altered by E2 replacement, in neither the miglyol nor the PBS preparation groups (Figure 4A).

Figure 4. Levels of testosterone, dihydrotestosterone (DHT) and androstenedione in serum after 8 weeks of treatments.

The amount of testosterone (A), dihydrotestosterone (DHT; B) and androstenedione (C) was measured in Sham and OVX mice after 8 weeks of s.c. injections every 4 days with vehicle (miglyol or PBS) or 17 β-estradiol (E2; 0.05 µg/mouse or 0.15 µg/mouse). Data are expressed as median (range); n=3 for each experimental group. (nd = not detectable; LLOQ=lower level of quantification).

Serum levels of DHT were reduced after OVX in both miglyol and PBS preparations. Treatment with E2 dissolved in miglyol resulted in DHT levels under the detection limit after 8 weeks, while E2 dissolved in PBS partially restored the serum levels of DHT (Figure 4B).

The serum levels of androstenedione in the different groups resembled those for DHT (Figure 4C). Androstenedione was reduced in the OVX groups after 8 weeks of treatment compared to the sham mice, with values under the detection limit in the miglyol OVX+veh and OVX+E2 groups.

Effects of OVX and pulsed E2 treatment over time in soft tissues

The body weight and the weight of perigonadal fat, uterus, thymus, liver, and spleen were measured at the time of each termination (Figure 5; extended data figure. 1²¹). The miglyol sham+veh mice had constant body weight for the whole duration of the experiment. However, the miglyol OVX+veh group showed a significant increase in body weight compared to the sham+veh group after 8 weeks of treatment, which was efficiently prevented by treatment with both doses of E2 dissolved in miglyol (Figure 5A). On the contrary, the PBS OVX+veh group did not differ in body weight from PBS sham+veh mice at any time-point. Treatment with E2 in PBS resulted in a gradual increase of the body weight compared to body weights of the PBS OVX+E2 groups measured after 2 weeks of treatment and reached a significant increase at 6 and 8 weeks of treatment for the low and high dose of E2, respectively (Figure 5A).

Figure 5. Body weights and weights of soft tissues.

Mice were subjected to Sham or OVX surgery and treated with s.c. injections of vehicle (miglyol or PBS) or 17β-estradiol (E2; 0.05µg/mouse or 0.15µg/mouse) every 4 days. Experiments were terminated at 2, 4, 6 and 8 weeks after the start of treatments. Body weight (A), perigonadal fat (B) and uterus (C) weights were determined at sacrifice. Data are expressed as mean±SEM; n=3 for each experimental group. * statistical differences between sham+veh vs OVX+veh; or OVX+veh vs the E2 treatment groups; # statistical difference vs week 2 of the same treatment group. *or #, p<0.05; **or ##, p<0.01; ***or ###, p<0.001; ***** or ####, p<0.0001; Two-way ANOVA followed by Dunnet’s post-hoc test).

Consistent with the results of body weights, the perigonadal fat in the miglyol OVX+veh group increased significantly at 6 and 8 weeks of treatment compared to the miglyol sham+veh group (Figure 5B). Furthermore, miglyol E2 treatment of OVX mice prevented perigonadal fat accumulation. At 6 weeks of treatment, the PBS OVX+veh group had significantly increased perigonadal fat weight compared to the PBS sham+veh group. Consistent with the results of body weights there was a gradual increase in perigonodal fat in mice treated with both doses of PBS E2. After 8 weeks of treatment, the perigonadal fat of the PBS OVX+E2 0.05µg group was also significantly higher compared to the PBS OVX+veh mice (Figure 5B).

As expected, the uterus weight was substantially lower in the OVX+veh mice compared to sham+veh mice in both miglyol and PBS groups (Figure 5C). An increase of the uterus weight was induced in the miglyol OVX+E2 0.15 µg group, starting at 2 weeks of the treatment. At the termination after 8 weeks of treatment, no differences between the miglyol OVX+E2 groups compared to the OVX+veh group were detected due to the unexpectedly increased uterus weight in miglyol OVX+veh mice (Figure 5C, left). On the contrary, injections of E2 dissolved in PBS did not increase the uterus weight when compared to the PBS OVX+veh mice (Figure 5C, right).

After 2 weeks of treatment, the PBS OVX+veh group had increased thymus weight compared to the PBS sham+veh mice, however the differences were equalized during the later time points. No major differences between the groups were detected for liver and spleen weights in either miglyol or PBS preparations (Extended data, figure. 1A–C,²¹).

Analysis of the body composition after OVX and pulsed E2 treatment

Before each sacrifice, the whole-body bone, fat and lean mass composition were measured using DXA scan (Figure 6A) (Underlying data²²). The percentage of fat in miglyol OVX+veh mice increased over time and reached statistically significant differences at weeks 6 and 8 compared to 2 weeks of treatment. After 8 weeks of treatment, both groups of miglyol OVX+E2 displayed decreased levels of body fat compared to the miglyol OVX+veh group (Figure 6B, left). On the contrary, both the low and the high dose of E2 dissolved in PBS induced an increase in whole body fat over time, compared to the same treatment at week 2 (Figure 6B, right). No differences between treatments were detected in lean mass (Extended data, figure. 2A,²¹).

Figure 6. Analysis of body composition by DXA.

Mice were subjected to Sham or OVX surgery and treated with s.c. injections of vehicle (miglyol or PBS) or 17β-estradiol (E2; 0.05 µg/mouse or 0.15 µg/mouse) every 4 days. The analyses were performed using a dual energy x-ray absorptiometry (DXA) scanner before sacrifice after 2, 4, 6 and 8 weeks of treatment. Representative DXA scan images for the experimental groups after 8 weeks of treatment (A). Percentage of fat (B), bone mineral content (BMC; C) and bone mineral density (BMD; D) were determined for the whole body. Data are expressed as mean±SEM; n=3 for each experimental group. * statistical differences between sham+veh vs OVX+veh; or OVX+veh vs the E2 treatment groups; # statistical difference vs week 2 of the same treatment group. *or #, p<0.05; **or ##, p<0.01; ***or ###, p<0.001; ***** or ####, p7lt;0.0001; Two-way ANOVA followed by Dunnet’s post-hoc test).

In the miglyol groups, the OVX+veh mice showed an increase of bone mineral content (BMC) and bone area after 8 weeks of treatment compared to week 2 (Figure 6C, left; extended data, figure. 2B, left,²¹). At week 4, the BMC of PBS OVX+E2 0.15µg mice were higher compared to week 2 of treatment. The other PBS experimental groups also showed an increase of the BMC and bone area at weeks 6 and 8 compared to the same treatment at week 2 (Figure 6C, right, extended data 2B, right,²¹).

The bone mineral density (BMD) was significantly lower in miglyol OVX+veh mice compared to the sham+veh group at 8 weeks of treatment. Neither the high nor the low dose of miglyol E2 treatments affected the BMD compared to OVX+veh (Figure 6D, left). Similarly, a decrease in BMD was detected in the PBS OVX+veh group at weeks 6 and 8 compared to the PBS Sham+veh groups (Figure 6D, right).

Effects of OVX and pulsed E2 replacement on the features of long bones

Micro-computed tomography was performed on dissected femurs of all mice for each time point (Figure 7A) (Underlying data²²). After 8 weeks of treatment a decrease in bone volume over total volume (BV/TV) was detected in both miglyol and PBS OVX+veh mice compared to the sham counterparts (Figure 7B). The loss of bone was faster (2 weeks) in OVX mice receiving PBS-vehicle. No significantly protective effects on BV/TV were detected by E2 replacement in the miglyol groups (Figure 7B, left). An early protective effect (2 weeks) was measured in the PBS OVX+E2 mice (Figure 7B, right).

Figure 7. Features of femur analysed by µCT analysis.

Mice were subjected to Sham or OVX surgery and treated with s.c. injections of vehicle (miglyol or PBS) or 17β-estradiol (E2; 0.05µg/mouse or 0.15µg/mouse) every 4 days. Experiments were terminated at 2, 4, 6 and 8 weeks after the start of treatments and femurs were dissected. Trabecular and cortical parameters of the femoral bone were determined by micro-computed tomography (µCT) analysis. Representative µCT scan pictures of the trabecular and cortical femur for the experimental groups after eight weeks of treatment (A). µCT analyses of bone volume/total volume (BV/TV); B), trabecular number (C), trabecular thickness (D), trabecular bone mineral density (BMD; E), and cortical thickness (F) were performed on femur. Data are expressed as mean±SEM; n=3 for each experimental group. * statistical differences between sham+veh vs OVX+veh; or OVX+veh vs the E2 treatment groups; # statistical difference vs week 2 of the same treatment group. *or #, p<0.05; ##, p<0.01; **** or ####, p<0.0001; Two-way ANOVA followed by Dunnet’s post-hoc test).

After 8 weeks of treatment with miglyol OVX+veh, mice displayed reduced trabecular number compared to sham+veh. Treatment with both doses miglyol E2 resulted in a trend towards increased trabecular number (Figure 7C, left). In the PBS OVX+veh group, the trabecular number was significantly reduced at all time points compared to PBS sham+veh (Figure 7C, right).

For trabecular thickness similar patterns were detected between the miglyol and PBS OVX+veh groups at week 8, showing increased trabecular thickness compared to the corresponding OVX+veh groups at week 2 (Figure 7D).

The trabecular BMD was reduced in the PBS OVX+veh mice compared to the PBS sham+veh group, constantly over time (Figure 7E, right). No significant differences were detected for any of the miglyol experimental groups (Figure 7E, left).

Loss of cortical thickness was detected in the miglyol OVX+veh group after 4 weeks of treatment, and the high dose of E2 (0.15µg) in miglyol protected from the decrease in cortical thickness (Figure 7F, left). In addition, both sham+veh, OVX+veh, and treatment of OVX mice with both doses of miglyol E2, resulted in an increase in cortical thickness over time (Figure 7F, left). For the PBS group the cortical thickness was significantly decreased in OVX+veh compared to sham+veh starting at week 6 and the high dose E2 treatment increased the cortical thickness from 4 weeks of treatment (Figure 7F, right).

The cortical area increased over time compared to week 2 in all the experimental groups of mice receiving miglyol, as well as in the PBS sham+veh group from 6 weeks of treatment (Extended data, figure. 3A,²¹).

The porosity of the bone was decreased in the PBS OVX+veh mice compared to the PBS sham+veh group at 4 weeks after the start of the treatments. Also, an overall reduction over time among the PBS OVX treatment groups were detected starting at week 6 (Extended data, figure. 3B,²¹).

Effects of OVX and pulsed E2 replacement on motor ability

The capacity to run on the rotarod (set to a constant acceleration) was reduced in OVX+veh mice compared to the sham+veh in both PBS and miglyol groups (Figure 8A). An improvement in the latency to fall was registered in OVX mice treated with miglyol E2 0.15µg compared to the OVX+veh mice (Figure 8A, left).

Figure 8. Motor ability determined by rotarod and open field tests after 8 weeks of treatments.

Mice were subjected to Sham or OVX surgery and treated with s.c. injections of vehicle (miglyol or PBS) or 17β-estradiol (E2; 0.05 µg/mouse or 0.15 µg/mouse) every 4 days. After 8 weeks of treatments, motor behavior was tested using the rotarod test for measurement of latency to fall (A) and the open field test to measure activity (B), track length (C) and average velocity (D). Data are expressed as mean±SEM; n=3 for each experimental group. One-way ANOVA followed by Dunnet’s post-hoc test was used to analyze differences between sham+veh vs OVX+veh; or OVX+veh vs the E2 treatment groups. *, p<0.05; **, p<0.01.

In the open field test, miglyol OVX+veh mice tended to move less and slower compared to the sham+veh operated mice (Figure 8B–D, left). A similar and more pronounced pattern was registered in the mice from the PBS groups. In this group, E2 replacement with the lower dose was able to restore the activity parameters (Figure 8B–D, right).

Underlying data

Zenodo: Pulsed administration for physiological estrogen replacement in mice.

https://doi.org/10.5281/zenodo.5036241²²

The project contains the following underlying data:

Extended data

Zenodo: Pulsed administration for physiological estrogen replacement in mice.

https://10.5281/zenodo.5140017²¹.

The project contains the following extended data:

Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication).

Reporting guidelines

Zenodo: ARRIVE checklist for “Pulsed administration for physiological estrogen replacement in mice”. https://10.5281/zenodo.5140017²¹.

The file is available under the terms of the Creative Commons Attribution 4.0 International.