We describe the development of a lean PCOS-like mouse model with dihydrotestosterone pellet to study the pathophysiology of PCOS and the offspring from these PCOS-like dams.
Hyperandrogenemia plays a critical role in reproductive and metabolic function in females and is the hallmark of polycystic ovary syndrome. Developing a lean PCOS-like mouse model that mimics women with PCOS is clinically meaningful. In this protocol, we describe such a model. By inserting a 4 mm length of DHT (dihydrotestosterone) crystal powder pellet (total length of pellet is 8 mm), and replacing it monthly, we are able to produce a PCOS-like mouse model with serum DHT levels 2 fold higher than mice not implanted with DHT (no-DHT). We observed reproductive and metabolic dysfunction without changing body weight and body composition. While exhibiting a high degree of infertility, a small subset of these PCOS-like female mice can get pregnant and their offspring show delayed puberty and increased testosterone as adults. This PCOS-like lean mouse model is a useful tool to study the pathophysiology of PCOS and the offspring from these PCOS-like dams.
Hyperandrogenism is the hallmark of polycystic ovary syndrome (PCOS) according to NIH criteria and of the Androgen Excess and PCOS (AE-PCOS) Society. Women with PCOS have difficulty getting pregnant and have increased risk of pregnancy complications1. Even if they get pregnant, their female offspring have an adverse health outcomes2,3. Animal models have been developed using various strategies4,5,6,7,8,9,10,11,12 and exhibiting many features of PCOS (anovulation, and or impaired glucose and insulin tolerance) with increased body weight and obesity associated with enlarged adipocyte size and increased adipocyte weight. There are two major strategies to produce animal models that are used to study PCOS. One is treatment with high levels of androgens directly (exogenous androgen injection/insertion) or indirectly (such as blocking androgen conversion to estrogen with aromatase inhibitor) after birth13. Another is by fetal hyperexposure of androgens during gestation14,15 to study the offspring. For example, female offspring from rhesus monkey16,17, sheep18, and rodents exposed to male levels of androgen during the intrauterine period develop PCOS-like traits later in life. These models significantly enhanced our understanding of elevated androgen effects, and fetal programing, and uterine environmental effects. However, these models have their own limitations: 1) animals develop obesity and it is therefore difficult to separate the effects of hyperandrogenemia from obesity induced reproductive and metabolic dysfunction; 2) before pregnancy, women with PCOS already exhibit high levels of androgen, thus oocytes have been exposed to androgen excess before fertilization; 3) the pharmacological doses of testosterone (T) or dihydrotestosterone (DHT) used after birth or during gestation may not reflect the androgen environment of PCOS. Testosterone and DHT levels have been measured in ovarian follicular fluid and/or serum, and testosterone and DHT levels are 1.5 to 3.9 fold higher in women with PCOS5,19,20,21,22,23 compared to unaffected women. We created an adult mouse model23,24,25 that develops reproductive and metabolic dysfunction within two weeks of the initiation of chronic DHT exposure from insertion of a pellet with 4mm length of crystal DHT powder (total length of pellet is 8mm). This model produces serum DHT levels that are about 2-fold higher (referred to as 2xDHT) than that of control mice without DHT treatment. The 2xDHT mice do not exhibit alterations of basal serum estradiol, testosterone, LH and do not develop obesity, and show similar ovarian weight, serum levels of cholesterol, free fatty acids, leptin, TNFα and IL-623,24,25 relative to controls even up to 3.5 months after DHT insertion23,24,25. Additionally, by mating females that have already developed features of PCOS, we can study the impact of a hyperandrogenic maternal environment on the metabolic and reproductive health of the offspring15.
This new paradigm (relevant to NIH and AE-PCOS Society criteria) models the disease by producing relatively similar levels of androgens to those of women with PCOS 2- to 3-fold higher testosterone or DHT levels compared to unaffected women. However, this model is maintained by continual exogenous DHT and not from programmed endogenous hyperandrogenism once DHT is withdrawn. The overall goal of this article is to focus on 1) how to make the DHT pellet; 2) how to generate a lean-PCOS like mouse model; 3) strategies to evaluate female offspring from these dams. Other measurements and assessment of phenotypes are not addressed in this manuscript but can be found in5,15,23,24,25,26.
Here, we present detailed protocols for DHT pellet preparation and insertion, and for reproductive and metabolic testing. The mice used in this study were a mixed background (C57/B6, CD1, 129Sv) and were maintained with food and water ad libitum in a 14/10 h light/dark cycle at 24 °C in the Broadway Research Building animal facility at the Johns Hopkins University School of Medicine. All procedures were approved by the Johns Hopkins University Animal Care and Use Committee.
1. Create PCOS-like Mouse Model
2. Assess Reproductive Profiles of Female Offspring from Chronically DHT Inserted Dams
Serum DHT levels and Glucose tolerance test
DHT levels were measured from collected serum by both ELISA and by LC-MS according to protocol 1.24–1.25, and 2.9, 3.0. The DHT absolute values are different between mass spectrometry and ELISA, however, the relative fold (around 2-fold) of DHT vs no-DHT insertion is similar from both assays and across experiments15,23,24 (Figure 2A). DHT levels are significantly increased from preconception through pregnancy in both DHT and no-DHT mice, however, DHT levels are 2-fold higher in the DHT mice compared to the no-DHT mice at both pregestational (one day before mating) and gestational (around 14 days) time points (Figure 2B). Since absolute levels of DHT differ between ELISA and LC-MS, we calculate relative levels (fold change: DHT level with DHT insertion vs no-DHT insertion) within assays. Female mice with DHT showed impaired glucose tolerance relative to no-DHT (Figure 3) at 2 weeks following DHT insertion according to protocol 1.23.
Female offspring body weight and puberty
While generally infertile, some dams get pregnant and have pups (around 30% pregnancy rate, so testing fertility of 10 dams we will usually get 3 pregnant dams, refer to protocol 2.1). We therefore can evaluate chronic androgen effects in dams on the female offspring. Female offspring from DHT inserted dams are called DHT-daughters, and those from no-DHT inserted dams are called no-DHT daughters. We observed no difference between no-DHT and DHT-daughters for the age of vaginal opening. However, the first estrus of DHT-daughters is significantly delayed (day 35 for no-DHT daughter; day 42 for DHT-daughter). This is associated with reduced body weight at both 35 and 42 PND in DHT-daughters relative to no-DHT daughters (Figure 4; refer to protocol 2.4–2.7).
Hormonal levels and estrous cyclicity in female offspring
Testosterone levels are not altered at 21 PND, but are reduced at 26PND. However, testosterone is increased at 70 PND (Figure 5, refer to protocol 2.9–3.0). Adult DHT daughters showed disrupted estrous cyclicity compared to adult no-DHT daughters, experiencing significantly longer times in M/D and less time in P and E (Figure 6A, B; according to the method in protocol 1.17–1.22).
Figure 1: Mouse identification. Within one cage, mouse is ear punched at different position to represent the mouse number. Mouse #1 to 3 is punched on the right ear, and #4 to 6 is punched on the left ear viewed dorsally. If mouse is punched on ears at both #1 and #4 position, it is #7; at 2 and 5, is #8; at 3 and 6 is #9; at the middle of left ear is #10; at 1 and 10 position is #11; etc. Please click here to view a larger version of this figure.
Figure 2: Serum DHT levels. Serum DHT levels. (A) Serum DHT levels are measured by both ELISA and Mass-Spectrometry. Although the absolute values are different, the two assays showed similar fold differences between DHT and no-DHT treated mice. (B) Serum DHT fold change relative to non-DHT levels at preconceptional and gestational time points. No-DHT (open bars) and DHT-implanted (black bars) female mice before (one day before mating) and during pregnancy (around 14 days of gestation). Values are mean ±S.E.M. N = 5-8 per group. Please click here to view a larger version of this figure.
Figure 3: Glucose tolerance test (GTT). No-DHT and DHT implanted mice were fasted overnight and glucose (2 g/kg BW) was injected intraperitoneally, and tail blood glucose was measured at different time points. DHT treated mice showed significantly increased glucose levels between 30 to 120 min compared to no-DHT treated mice. Values are mean ±S.E.M. N = 4-12 per group. * P <0.05 Please click here to view a larger version of this figure.
Figure 4: Pre-pregnancy maternal DHT treatment resulted in reduced body weight in female offspring at 35 and 42 PND. Body weight (Y-axis) was measured on post-natal days as shown (X-axis). DHT-exposed offspring: black bar; no –DHT offspring: open bar. Values are mean ±S.E.M. N = 9-14 per group. Please click here to view a larger version of this figure.
Figure 5: Serum DHT levels. Blood was collected in the morning on PND 21, 26, 70 in the morning. Serum DHT levels (Y-axis) from female offspring of no-DHT daughters (open bars) and DHT-daughters (black bars). Values are mean ±S.E.M. N = 5-11 per group. This figure has been modified from reference2 Please click here to view a larger version of this figure.
Figure 6: Chronic maternal androgen excess leads to disturbed cyclicity in adult daughters. (A) Representation of estrous cyclicity of female offspring. (B) The percentage of time spent at each estrous stage (Y-axis) during 15 days (X-axis) measured by cytological examination of vaginal cells. Values are mean ±S.E.M. N = 5-9 per group. The estrous cycle stage (Y-axis). M/D: met/diestrus; P: proestrus; E: estrus. Please click here to view a larger version of this figure.
Table 1: Ketamine/xylazine cocktail.
Hyperandrogenism is a key feature of PCOS. The serum DHT levels (two fold higher in DHT mice than in no-DHT mice) used in this protocol are lower than those reported by other investigators in previous studies and are calibrated to proportionally mimic women with PCOS5,19,20,21. Unlike other models, this 2-fold DHT model does not alter the body weight and whole body composition compared with no-DHT mice for up to 3.5 months after DHT insertion23,24. These adult DHT implanted mice are maintained by continual exogenous DHT. Although whole body composition is not altered, alteration of the structure and function of the adipocytes has been observed in lean PCOS women20. A careful examination of different depots is warranted. We observed impaired estrous cyclicity within one week after DHT insertion15 and reduced fertility during a 3 month fertility evaluation23.
The small subset of DHT treated female mice that were able to get pregnant and have offspring provided an opportunity to assess the impact of pre-gestational, gestational and nursing hyperandrogenemia on fetal development. In order to get a properly controlled experiment, multiple breeding females are required which necessitates that at least 10 mating pairs be established. DHT-daughters are only exposed to DHT during gestation and before 21 PNA. DHT-daughters showed reduced body weight compared to no-DHT daughters, indicating that obesity or overweight is not a confounding variable in the observed physiological effects15.
In initial studies, we validated the serum DHT levels over time. We found no significant differences in serum DHT at 1, 2, 3, and 4 weeks23,24. DHT levels declined after 4 weeks, therefore, we replace DHT pellet every 4 weeks. We do not observe reduced effects of DHT even they have been stored at room temperature for 3 months. It is important to incubate pellets in saline for 24 h just before insertion. This step is critical for even release of DHT from the pellet (DHT release slowly through the silastic tubing). Estrous cyclicity can be examined after 3 days of DHT insertion, and vaginal smear can be also examined directly in the saline without dry and stain under a light microscope after you familiar with the cell types27.
Metabolic phenotype can be assessed at, or after, 2 weeks (14 days) of DHT insertion. However, the effects of DHT on metabolism at the end of the fourth week (28 days) are slightly attenuated, though no attenuation is observed for reproductive dysfunction. Therefore, we assess metabolism function normally at the end of 2, 3, 5 , 6, 7 weeks (14, 21, 35, 42, and 49 days, 2 times total insertion). The production of slow release steroid pellets in the lab is a mature technique that has been widely adopted by different laboratories33,34. It represents a cost-effective alternative to commercially available products, which can exceed $50/pellet (90 day DHT pellet, 5mg/pellet, IRA, NA-161). The commercial product does have the advantage of not needing to be replaced for up to 90 days, while the pellet described in this report needs to be replaced every 28 days. For large-scale studies, investigators may find it more economical to produce their own pellets even with the added work required to replace pellets every 4 weeks.
As observed by others, models with chronic DHT implantation after birth do not show increased LH pulsatile frequency and this may point out different pathological mechanisms of elevated androgen induced reproductive function between developmental and late-onset acquired hyperandrogenism. With this protocol, we are able to investigate female offspring from chronically hyperandrogenic dams. We can leverage this model to try to fully understand the reproductive (e.g. follicle development, ovulation (corpora lutea), and fertility) and metabolic consequences (e.g. glucose or lipid metabolism, body composition, adipose depot distribution). Our protocol adds new tool to investigate elevated androgen induced reproductive and metabolic dysfunction; and dissociates the pathophysiology caused by elevated androgen from those that could be due to obesity. This model can be used to explore the tissue specific effects of elevated androgens in females. In addition, the methodology reported here for DHT pellet production can be easily applied to the study of other steroid hormones.
The authors have nothing to disclose.
This work was supported by the National Institutes of Health (Grants R00-HD068130 to S.W.) and the Baltimore Diabetes Research Center: Pilots and Feasibility Grant (to S.W.).
Crystalline 5α-DHT powder | Sigma-Aldrich | A8380-1G | ||
Dow Corning Silastic tubing | Fisher Scientific | 11-189-15D | 0.04in/1mm inner diameter x0.085in/2.15mm outer diameter | |
Medical adhesive silicone | Factor II, InC. | A-100 | ||
Goggles, lab coats, gloves and masks. | ||||
10 µL pipette tips without filter | USA Scientific | 11113700 | ||
Microscope slide for smear | Fisher Scientific | 12-550-003 | ||
Diff Quik for staining cells | Fisher Scientific | NC9979740 | ||
Lancet | Fisher Scientific | NC9416572 | ||
3 mL Syring | Becton, Dickinson and Company (BD), | 30985 | ||
attached needle: 20G | BD | 305176 | ||
Ruler: any length than 10cm with milimeter scale. | ||||
Xylazine | Vet one AnnSeA LA, MWI, Boise | NDC13985-704-10 | 100mg/ml | |
Ketamine Hydrochloride | Hospira, Inc | NDC 0409-2051-05 | 100mg/ml | |
Surgical staple | AutoClip® System, Fine Science Tool | 12020-00 | ||
Insulin syringe | BD | 329461 | 1/2 CC, low dose U-100 insulin syringe | |
Trochar | Innovative Research of America | MP-182 | ||
Microscope | Carl Zeiss Primo Star | 415500-0010-001 | Germany | |
Ear punch | Fisher Scientific | 13-812-201 | ||
Testosterone rat/mouse ELISA kit | IBL | B79174 | ||
DHT ELISA kit | Alpha Diagnostic International | 1940 | ||
One touch ultra glucometer | Life Scan, Inc. | |||
One touch ultra test stripes | Life Scan, Inc. | |||
Eppendorf tube | Fisher Scientific | 05-402-18 | ||
Razor blade | Fisher Scientific | 12-640 | ||
Clidox | Fisher Scientific | NC0089321 | ||
surgical underpad | Fisher Scientific | 50587953 |
|
|
Betadine Antiseptic Solution | Walgreens | |||
3M Vetbond (n-butyl cyanoacrylate) | 3M Science. Applied to Life | |||
Animal tattoo ink paste | Ketchum manufacturing Inc. | Brockville, Ontario, Canada | ||
Scale | Ohaus Corporation | HH120D | Pine Brook, NJ | |
Electronic digital caliper | NEIKO Tools USA | 01407A | available from Amazon |