A description of the surgical induction of endometriosis in mice and rats by auto-transplantation of uterine tissue to the arterial cascade of the intestinal mesentery.
Endometriosis is a chronic, painful disease whose etiology remains unknown. Furthermore, treatment of endometriosis can require laparoscopic removal of lesions, and/or chronic pharmaceutical management of pain and infertility symptoms. The cost associated with endometriosis has been estimated at 22 billion dollars per year in the United States1. To further our understanding of mechanisms underlying this enigmatic disease, animal models have been employed. Primates spontaneously develop endometriosis and therefore primate models most closely resemble the disease in women. Rodent models, however, are more cost effective and readily available2. The model that we describe here involves an autologous transfer of uterine tissue to the intestinal mesentery (Figure 1) and was first developed in the rat3 and later transferred to the mouse4. The goal of the autologous rodent model of surgically-induced endometriosis is to mimic the disease in women. We and others have previously shown that the altered gene expression pattern observed in endometriotic lesions from mice or rats mirrors that observed in women with the disease5,6. One advantage of performing the surgery in the mouse is that the abundance of transgenic mouse strains available can aid researchers in determining the role of specific components important in the establishment and growth of endometriosis. An alternative model in which excised human endometrial fragments are introduced to the peritoneum of immunocompromised mice is also widely used but is limited by the lack of a normal immune system which is thought to be important in endometriosis2,7. Importantly, the mouse model of surgically induced endometriosis is a versatile model that has been used to study how the immune system8, hormones9,10 and environmental factors11,12 affect endometriosis as well as the effects of endometriosis on fertility13 and pain14.
1. Planning for live-animal surgery
2. Prepare the surgical area for live animal surgery
3. Anesthetize and prepare the mouse for surgery
4. Uterine ligation
5. Prepare endometriotic implants from excised uterus
6. Suturing endometriotic implants in peritoneal cavity
7. Sham surgeries
8. Closing the surgical wound
9. Recover animal
10. Post-operative care
11. Necropsy and tissue excision
Representative Results
Endometriotic lesions in the mouse model of surgically induced endometriosis morphologically and histologically resemble those observed in women. Histological analysis of endometriosis in both women and the mouse model indicates that endometriotic lesions contain endometrial glands and stroma (Figure 2A). Endometriotic lesions in mice also contain hemosiderin-laden macrophages, which are a common hallmark of endometriosis in women (Figure 2B)19.
Endometriotic lesions removed from mice three days post-induction appear inflamed and hemorrhagic (Figure 3A). After two to four weeks of growth endometriotic lesions in the mouse model are cyst-like, fluid filled and surrounded by peritoneal adhesions (Figures 3B and 3C). Compared to lesion weight at induction, fluid filled lesions were 306% and 862% larger at one and two months post-induction and lanced lesions were 51% and 172% larger, respectively (Figures 4A and 4B). We have obtained consistent fluid filled and lanced endometriotic lesion weights at one-month post-induction over five different experiments (Figure 5). At one month post-induction fluid filled (7.44±3.75 mg) and lanced (2.92±1.23 mg) endometriotic lesion weight were significantly correlated (Pearson’s correlation coefficient = 0.669, p < 0.001).
Age of the mouse did not affect lesion size for mice between three and ten months of age. Neither the fluid filled or lanced endometriotic lesion weight at one month post-induction was significantly correlated with the age of the animal (r = -0.136, p = 0.380 and r = -0.063, p = 0.698, respectively).
The mouse uterus undergoes changes in size, fluid retention, cell proliferation and appearance due to the influence of steroid hormones during the estrus cycle. We compared the endometriotic lesion weight to the weight of the remaining intact uterine horn from animals in different estrus stages. We did not find a significant correlation between uterine weight and fluid filled or lanced endometriotic lesion weight at one-month post induction (r = -0.046, p = 0.765 and r = 0.232, p = 0.155, respectively).
The gene expression pattern observed in the endometriotic lesions of mice closely mirrors that reported in women with the disease5. By three days post-induction genes regulating extracellular matrix remodeling, cell adhesion, and angiogenesis are highly upregulated and many of these genes remain upregulated through one month of growth.
Figures and Tables
Figure 1. Surgical induction of endometriosis by autologus uterine tissue transfer in the mouse. The left uterine horn is ligated, excised, and opened longitudinally to expose the endometrium. Three 2 mm2 biopsies are prepared and each is sutured to an artery in the arterial cascade of the intestinal mesentery. By one month post-induction the endometriotic lesions are fluid filled and surrounded by adhesions.
Figure 2. Hematoxylin and eosin stained section of an endometrial lesion from the mouse model of endometriosis at one month post-induction demonstrating (A) the presence of endometrial glands and stroma; scale bar = 50 μm and (B) hemosiderin-laden macrophages, some of which are indicated by arrows; scale bar = 20 μm.
Figure 3. Endometriotic lesions in the mouse model following euthanasia, either three days post-induction (A) or one month post-induction (B and C).
Figure 4. Endometriotic lesions from mice surgically induced to have endometriosis were excised and weighed at one or two months post-induction. Data are average±SEM. Data were log transformed and different letters indicate significance within each panel by one-way ANOVA followed by one-sided Fisher’s Least Significant Difference Mulitple Comparisons. (A) Cyst like, fluid filled endometriotic lesions (N = 10, 7 or 5 for induction, one month, or two month post-induction, respectively). (B) Lanced endometriotic lesions (N = 10, 8 or 7 for induction, one month or two month post-induction respectively).
Figure 5. Endometriotic lesion wet weight with fluid and lanced at one month post-induction from five separate experiments. Data are average±SEM. Mice N=10, 6, 8, 7 and 7 for fluid filled lesions and 0, 7, 10, 8, and 8 for lanced lesions in experiment 1, 2, 3, 4, and 5, respectively.
Table 1. Observation of Estrus Stage by Vaginal Cytology and Visual Appearance of Ovaries and Uterus and Induction.
Appearance of ovary and uterus will be time dependent. The following are based on sacrifice around 8:00 am the morning of each cycle day. Further, the observations are subjective and comparing the ovary and uterine horns will be a better estimate than uterine horns only. These observations are meant to supplement the information obtained from daily vaginal cytology readings.
Table 2. Comparison of surgery in mouse and rat.
There are several critical parameters that should be noted while performing the surgical induction of endometriosis in mice. First, endometriosis is an estrogen dependent disease and as such this surgery should be performed in intact animals or alternatively in ovariectomized animals supplemented with estrogens20. Second, suturing the endometrial biopsies to the arterial cascade must be performed with extreme care. We have found that using only two relatively loose knots with one throw each keeps the biopsy in place while preventing ligation of the blood supply to the bowel and subsequent tissue necrosis and animal death. We strongly recommend practicing on several mice prior to the actual experiment to ensure that sutures are placed securely enough that the tissue is not lost, but not too tight as to cause bowel necrosis. Third, it is very important to keep the bowel tissue hydrated with sterile PBS containing penicillin and streptomycin during the surgery. Fourth, efforts should be taken to ensure that endometriotic implant size is consistent. To this effect we use a 2 mm biopsy punch to create consistently sized endometriotic implants from excised uterine tissue.
There are several modifications that can be made to this protocol to suit the individual needs and scientific questions of the researcher. One potential modification pertains to the hormonal profile of the animal. Gene and protein expression in the endometriotic lesions, remaining intact uterine horn and immune system may be directly influenced by the timing of collection in relation to the estrus cycle stage. There are several approaches to control for the estrus stage at the time of collection, each with its own advantages and disadvantages. For example, intact, cycling animals can be synchronized by transferring urine-soaked male bedding into the female cage three days prior to induction or collection such that animals are surgically induced and collected during the same estrus stage15. If intact cycling animals are used, estrus cyclicity should be monitored daily by examination of vaginal cytology17. This has the advantage of being more physiologically relevant, but achieving cycle synchronicity is not 100% successful. Alternatively, intact animals could be synchronized by exogenous gonadotropin administration21,22. While somewhat more accurate, this method creates higher estrogen levels than normal. Another approach has been to ovariectomize animals and give back a constant level of exogenous hormones either through a silastic capsule, mini-osmotic pump, or daily injections4,8. This method has the advantage of obtaining a uniform hormone profile across many animals, but is disadvantageous because the animals are not cycling.
Sham surgeries can be performed to assess which effects are the result of undergoing the surgery versus effects attributable to endometriosis. In sham surgeries the left uterine horn is removed and sutures are placed around the arteries in the arterial cascade of the intestinal mesentery but no endometriotic implants are sutured5,8,23. Alternatively, fat removed from the excised uterine horn can be sutured to the intestinal mesentery3,14. We recently reported that there are relatively few differences in gene expression as measured by cDNA microarray between sham uterus and the intact remaining uterine horn from animals surgically induced to have endometriosis5. Furthermore, using quantitative real-time PCR for seven endometriosis-related genes, we found that only haptoglobin expression was significantly different in the uteri of sham and endometriosis mice. Lee et al. however, reported differential expression of five genes in the uterus from mice induced to have endometriosis compared to sham controls23. This suggests that the presence of the endometriotic lesions can cause altered gene expression in the eutopic uterus.
The timing of collection of mice surgically induced to have endometriosis should be determined by the particular research question. Collection of mice three days post-induction allows one to assess the critical, early events in the establishment of endometriosis including the initial neutrophil and machrophage infiltration and cytokine production as has been reported previously8. We have shown that endometriotic lesions collected three days post-induction are small, hemorraghic, and have significantly altered gene expression relative to the remaining uterine horn4. By two weeks post-induction endometriotic lesions are well established and have often formed cyst-like structures. In the mouse, endometriotic lesions continue to increase in size, as assessed by weight of the fluid filled and lanced lesions (figure 4) as well as by lesion volume determined by length, width and height measurements through two months post-induction9.
As mentioned above, the surgical model of endometriosis was first developed in the rat and is still extensively used3. When performing this surgery in the rat we make the following modifications to the protocol, as summarized in Table 2. In the rat, all of the sutures used are 4-0 size and can and should be tied more tightly. Additionally, we suture six 5 mm2 endometriotic implants into the intestinal mesentery of the peritoneal cavity.
In this protocol we describe the use of an inhaled anesthetic, isoflurane. Alternatively, combined injectable anesthetic composed of ketamine and Domitor (medetomidine hydrochloride) or another alpha-2 agonist can be used, although the recovery time is somewhat longer. Ketamine, administered at 75 mg/kg by intraperiotneal injection, is a dissociative anesthetic agent with minimal acute pain relieving properties and is a Schedule III drug requiring a DEA license and detailed drug inventory. Domitor, administered at 1 mg/kg, is an alpha-2 adrenergic agonist and is readily reversed with Antisedan (atipamezole hydrochloride) administration. Domitor is a sedative that provides muscle relaxation and pain relief. Domitor and ketamine can be prepared ahead of time in combination in PBS using sterile technique. Antisedan, the reversing agent for Domitor, can be prepared in PBS in combination with buprenorphine and should be administered at 1 mg/kg post-operatively. Buprenorphine is also a Schedule III drug and alternatives include the non-steroidal anti-inflammatories banamine (flunixin meglumine) and ketoprofen.
As with all models there are certain limitations associated with the surgical induction of endometriosis in rodents. Foremost is that rodents do not have a menstrual cycle and therefore do not spontaneously develop endometriosis. In an effort to make the rodent model more physiologically similar to the condition in humans some researchers have opted to inject autologous or homologous uterine tissue fragments from syngenic animals into the peritoneal cavity directly without suturing24,25. In mice, the injected tissue forms cyst-like endometriotic lesions, however, the injection method of induction does not seem to work in rats as the tissue fails to attach and invade in the peritoneal cavity3.
The rodent model has been extensively used to study the etiology, pathology, and risk factors of endometriosis2,8,26-29 as well as to explore novel therapeutics14,30-34. The rodent model of surgically induced endometriosis demonstrates many similarities to the disease in humans, including reduced fertility and fecundity and altered gene and protein expression5,6,35,36. Taken together, with their relatively low costs and ready availability, the rodent model of surgically induced endometriosis is an excellent model for endometriosis in women.
The authors have nothing to disclose.
Special thanks to Chris Kassotis and Audrey Bailey for critical review of this manuscript and to Dr. Scott Korte, Joseph Beeman, Alison Curfman, Paul Kimball, Bridget Neibreggue, Jacob Redel, Amy Schroder, Maija Steinberg, and Stacey Winkeler for their assistance in optimization of this model in our laboratory. Funding was provided by the Clinical Biodetectives Training Grant (NIH T90) (KEP), University of Missouri Life Sciences Undergraduate Research Opportunities Program, MU Research Council, MU Research Board grants and NIH R21HD056441 (SCN).
Name of the reagent/equipment | Company | Catalogue number |
Wax pencil | Fisher | NC9954135 |
Glass slide | Fisher | 12-550-433 |
Eyedropper | Fisher | S79383 |
Standard light microscope for evaluating vaginal cytology smears | ||
Buprenorphine HCL c3 (CARJET) 10X1ml | Butler Animal Health Supply | 022891 |
Sterile phosphate buffered saline (PBS) | Gibco | 14040-117 |
10,000U/ml Penicillin, 10,000μg/ml Streptomycin in 0.85% NaCl | Hyclone | SV30010 |
Isoflurane | Abbott Animal Health | 05260-05 |
Isoflurane non-rebreathing anesthetic system | ||
Recirculating hot water heating pad | ||
30 ml syringe sheath | Fisher | 14-823-16G |
Powder free sterile gloves | Fisherbrand | 19020558 |
Ophthalmic ointment | Major Pharmaceuticals | 10033691 |
Small electrical clippers | Wahl | 9861-600 |
Chlorhexidine scrub | Fisher | NC9863042 |
70% Ethanol | ||
Polylined sterile field | Busse Hospital Disposables | 696 |
Size 3 scalpel | Fisher | 22-079-657 |
Number 10 scalpel blades | Fisher | 22-079-681 |
Small surgical scissors | Roboz | RS-5850 |
Small serrated semi-curved forceps | Roboz | RS-5135 |
5-0 black braided silk suture | Ethicon | K870H |
Sterilized pyrex glass Petri dishes | Corning | 70160-101 |
2 mm biopsy punch | Miltex | 33-31 |
Sterile gauze | Kendall | 1806 |
6-0 black monofilament ethilon nylon suture | Ethicon | 697G |
Needle drivers (optional) | World Precision Instruments | 500023 |
5-0 undyed braided coated vicryl suture | Ethicon | J490G |
9mm Autoclip wound clips | Becton Dickinson | 427631 |
Autoclip applier & remover | Becton Dickinson | 427630 |
23G needle | Becton Dickinson | 305193 |
1cc syringe | Becton Dickson | 301025 |
5X magnifying glass stand (optional) | Fisher | 14-648-23 |
10% Buffered formalin | Fisher | SF100-4 |
Calipers | Roboz | RS-6466 |
Processing/embedding cassettes | Fisher | 15-197-700A |
Biopsy foam pads | Fisher | 22-038-222 |
RNAqueous RNA isolation kit | Ambion | AM1912 |
Liquid nitrogen | ||
Snap cap microcentrifuge flat top tube | Fisher | 02-681-240 |
Ketamine (optional) | Simga | K4138 |
Domitor (medetomidine hydrochloride) (optional) | Tocris | 2023 |
Antisedan (atipamezole) (optional) | Sigma | A9611 |