{"paper_id":"a3650886-b315-438a-b5d8-c4728d8b141e","body_text":"Endometrial cancer is the fourth most common cancer in women in the United States, and the most common gynecologic cancer. Based on estimates from the American Cancer Society, 67,880 women were diagnosed with endometrial cancer in 2024 in the U.S., and an estimated 13,250 died from the disease [ 1 ]. While hysterectomy is the leading treatment option, better therapies and earlier detection methods are needed to increase survival rates, particularly in younger women hoping to have children. In endometrioid endometrial carcinoma (EEC), 17-β-estradiol (E2) drives epithelial cell proliferation while progesterone (P4) inhibits E2-induced epithelial cell proliferation. Thus, the unopposed actions of E2 can lead to endometrial carcinoma, and P4 analogs are often used as a treatment option [ 2 ]. E2 signaling indirectly activates the AKT pathway of cell survival and proliferation. A negative regulator of AKT signaling is the phosphatase and tensin homologue (PTEN). The loss of PTEN activity via gene mutation leads to elevated AKT activity and increased endometrial epithelial cell proliferation [ 3 ]. Inflammation likely plays a prominent role in the development and progression of endometrial cancer [ 4 ]. Notably, PTEN is mutated in 83% of type I endometrial carcinomas [ 5 ]. In mice, the deletion of both alleles of  Pten  in the uterus leads to endometrial cancer in 100% by one month of age, and  Pten  heterozygosity results in 100% hyperplasia and 22% endometrial cancer [ 6 , 7 , 8 ].  Pten  heterozygosity in mice is often used as a model to study endometrial oncogenesis in an effort to assess the contributions of other genes thought to be involved in endometrial neoplasm formation or those that prevent the development and progression of the disease (e.g., [ 9 , 10 , 11 ]).\nProgesterone receptor membrane component (PGRMC) family members PGRMC1 and PGRMC2 were first cloned from porcine vascular smooth muscle cells [ 12 ] and human cells [ 13 ]. Whereas PGRMC1 expression is highest in the liver and kidney, PGRMC2 expression is most abundant in placenta [ 13 ]. Both proteins are expressed in most tissues. Initial characterization studies using porcine liver fractions demonstrated that PGRMC1 binds steroid hormones with affinities in the low- to mid-nanomolar range [ 14 ]. PGRMC1 was found to have the highest affinity for P4 (i.e., 11 nM), but it also bound corticosterone, cortisol, testosterone, and cholesterol. PGRMC1 has a poor affinity for 17-β estradiol and aldosterone [ 14 ]. Despite being labelled as a P4 receptor, PGRMC2 has not yet been shown to bind and become activated by P4. PGRMC proteins have pleiotropic functions in female reproduction and P4 signaling [ 15 , 16 , 17 ], mitosis and meiosis [ 18 ], energy metabolism and Warburg glycolysis [ 19 ], angiogenesis [ 20 ], receptor trafficking to the plasma membrane [ 21 ], drug metabolism [ 22 ], heme transport [ 23 ], and neural development [ 24 ]. Gene mutations and aberrant expression profiles in PGRMC1 and PGRMC2 are associated with a host of reproductive diseases [ 15 , 25 , 26 ]. Through association-based studies in humans, as well as cell culture and xenograft studies in immunocompromised mice, PGRMC1 is thought to contribute to the development and progression of solid tumors in many different organs [ 26 , 27 , 28 , 29 , 30 , 31 , 32 ]. Similarly, much of what is known about PGRMC2 stems from descriptive studies in cell lines and reproductive tissues [ 15 , 25 , 26 , 27 ]. PGRMC2 expression correlates with the development of endometriosis in a non-human primate model of the disease (reviewed in [ 26 ]). The conditional ablation of  Pgrmc2  in mice results in subfertility that progresses to premature reproductive senescence [ 16 , 26 , 27 ]. PGRMC2 was shown to function in adipocytes as an essential intracellular heme chaperone required for normal function [ 23 ], as well as a pressure–volume regulator for accommodating stress responses in cardiomyocytes [ 33 ]. PGRMC2 plays an important role in human placental extravillous trophoblast invasion during early pregnancy where it also helps regulate immune homeostasis at the maternal–fetal interface [ 34 , 35 ]. Considerably less is known about a role for PGRMC2 in cancer. As with PGRMC1, PGRMC2 is elevated in ovarian cancer and a large number of ovarian cancer cell lines [ 27 ]. However, to date, we are unaware of any studies that assess the function of PGRMC2 in the context of the development or progression of cancer in vivo.\nGiven that PGRMC proteins are generally overexpressed in many tumor types, we hypothesize that selectively deleting a PGRMC family member from the endometrium with the heterozygous or complete loss of  Pten  would prevent the development of or reduce the severity of  Pten  loss-of-function-induced hyperplasia and cancer. Because of the general lack of information about PGRMC2 in the context of cancer,  Pgrmc2  was specifically evaluated in this context.\n\nAll animal protocols were approved by the Institutional Animal Care and Use Committees at Washington State University or the University of Wyoming. The floxed  Pgrmc2  mice ( Pgrmc2 fl ) and their use with  Pgr cre/+  mice were described previously [ 16 ]. Floxed  Pten  ( Pten fl/fl ) mice were obtained from Jackson laboratories. Mice of 10 different genotypes were generated for this study by crossing  Pgr cre/+  mice with  Pten fl/fl  and/or  Pgrmc2 fl/fl  mice. Tissues were collected from mice at 6–12 weeks of age or at 9 months of age at the completion of the study. In some cases, such as with  Pgr cre/+ ;  Pten fl/fl  ( Pten d/d ) mice, animals were euthanized at 4–9 months of age due to rapid development of cancer. Animals aged 4–9 months were first ovariectomized and tissues were collected 1 week later for fixation in 4% paraformaldehyde (PFA) and paraffin embedded or snap frozen in liquid nitrogen and stored at −80 °C until use. For assessing the endometrial epithelial mitotic response to 17β-estradiol (E2), young (6–12 weeks)  Pten +/fl ;  Pgrmc2 fl/fl ,  Pten +/d , and  Pten +/d ; Pgrmc2 d/d  mice were ovariectomized, allowed to rest for 1–2 weeks, and then subjected to sesame oil vehicle treatment that mirrored treatment with E2 (100 ng E2 for 2 days, 5 days without treatment, 50 ng E2, and tissue collection 18 h later). Tissues were again collected for paraffin embedding and some were snap frozen.\nAll human and mouse tissues were fixed in 10% buffered formalin or PFA, then stored in 70% ethanol until paraffin embedding. Tissues were processed through a graded series of ethanol (70–100%) and xylenes, embedded in paraffin, and sectioned at 5 µm by microtome. Human tissues were collected deidentified, and contributed to a Northwestern University tissue repository as part of NCATS/UH3TR00120. Proliferative and secretory specimens were obtained from women with benign gynecological conditions such as fibroids. For general histological analyses, tissue sections were deparaffinized with xylenes, rehydrated, and stained with hematoxylin and eosin. Immunohistochemistry (IHC) was completed as previously described [ 16 , 17 , 31 , 32 ]. Antibodies used for IHC included anti-phosphohistone H3 (phH3) primary antibody (Ser10, Millipore 06-570, 1:750, Burlington, MA, USA), anti-PGRMC2 primary antibody (Sigma, 1:100, St. Louis, MO, USA), and biotinylated anti-rabbit IgG secondary antibody (Vector BA-1000, 1:750, Burlingame, CA, USA). From young ovariectomized  Pten +/fl ;  Pgrmc2 fl/fl ,  Pten +/d , and  Pten +/d ; Pgrmc2 d/d  mice treated with vehicle or E2 and aged ovariectomized  Pten +/fl ;  Pgrmc2 fl/fl ,  Pgrmc2 d/d ,  Pten +/d , and  Pten +/d ; Pgrmc2 d/d  mice, mitotic (phH3+ cells) counts were made from three different tissue sections for each biological replicate. From 9-month-old  Pten +/fl ; Pgrmc2 fl/fl ,  Pten +/d , and  Pten +/d ; Pgrmc2 d/d  mice ovariectomized one week prior to euthanasia, the number of granulated and degranulated mast cells was counted on a per section basis. Following deparaffinization and rehydration, 5 µm sections were stained with toluidine blue working solution (1 mg/mL) and counter-stained with hematoxylin. Mast cells were quantified by counting dark blue cells and establishing a mean value from three non-adjacent sections for each biological replicate.\nFor assessing mRNA levels of the classical estrogen receptor ( Esr1 ) and progesterone receptor ( Pgr ) from uterine tissues collected from ovarictomized and sesame oil vehicle treated  Pten +/fl ;  Pgrmc2 fl/fl ,  Pten +/d , and  Pten +/d ; Pgrmc2 d/d  mice, total cellular RNA was isolated using TRI-Reagent (Sigma, St. Louis, MO, USA). RNA samples were subjected to DNAse I digestion (Promega, Madison, WI, USA), and complementary DNA was generated using iScript Reverse-Transcription Supermix (BioRad, Hong Kong, China). qPCR was performed to compare expression of  Esr1  and  Pgr  in complementary DNA samples.  Rpl13a  was included for normalization. A negative control (no reverse transcription) was included to confirm the absence of genomic DNA.  Table 1  provides primer information for qPCR, as well as for primers used for genotyping animals.\nAnimals with the same genotype were randomly assigned to the various treatment groups. All data are presented as the mean ± SEM for n = 3–28 samples. Individual animals represented a single biological replicate. Differences between genotypes were assessed by Student’s  t -test, where the mean values of two groups were compared. Differences between treatment groups or genotypes with more than two groups were analyzed by one-way analysis of variance followed by Tukey’s post hoc test. A chi-square analysis was used for incidence studies and a Kaplan–Meier survival curve was generated for lifespan analyses. For semi-quantitatively comparing PGRMC2 protein levels in human endometrial tissues, a composite h-score was established by multiplying the staining intensity [scored as 1 (low) or 2 (high)] and percentage of positive cells [scored as 0 (no staining), 1 (1–10% staining), 2 (11–50% staining), or 3 (>50% staining)] within each tissue specimen (n=10 each). Contingency graphs were then generated to show the relationship between staining intensity and the percentage of positive cells by showing the frequencies of different combinations of these variables. All data were analyzed using GraphPad 5.0 software (San Diego, CA, USA) where  p  ≤ 0.05 was considered statistically significant and  p  > 0.05 ≤ 0.10 was considered a trend.\n\nThe level of PGRMC2 protein was evaluated by immunohistochemistry in human endometrial samples during the proliferative and secretory phases of the menstrual cycle, as well as in endometrial cancer specimens ( Figure 1 ). Semi-quantitative h-scoring was used to show that PGRMC2 protein levels did not change in epithelial or stromal tissues across the different groups. During the proliferative phase, PGRMC2 was abundant in both the glandular and luminal epithelia and, to a lesser extent, in the stroma. PGRMC2 remained abundant in epithelia during the secretory phase, and appeared to be elevated in the stromal compartment compared with stromal tissue from the proliferative phase. However, a difference in PGRMC2 between epithelial and stromal tissues was not found to be statistically different during the menstrual cycle or endometrial cancer. PGRMC2 seemed more abundant in the neoplastic epithelium of endometrial cancer tissues when compared to adjacent stromal tissue, but this may likely be due to the presence of stratified epithelium in this neoplastic tissue.\nThe most commonly mutated gene in human endometrioid endometrial carcinoma is  PTEN  [ 36 ]. The deletion of a single  Pten  allele in mice results in a condition similar to Cowden’s syndrome in women. This phenotype is accompanied by endometrial hyperplasia in 100% of mice and endometrial carcinoma in 22% of mice by six months of age [ 37 ]. The conditional ablation of both alleles causes endometrial carcinoma in 100% of mice by around one month of age [ 6 ]. Conditional mutagenesis in mice in which one or both  Pten  alleles are ablated is commonly used as a model to study  Pten -based endometrial carcinoma to understand how other gene mutations contribute to endometrial neoplasms [ 6 , 7 , 8 , 10 , 11 , 38 ]. Given the lack of in vivo information about a role for PGRMC2 in oncogenesis, we sought to determine if endometrial  Pgrmc2  ablation would impact the development and progression of endometrial neoplasia in a well-established model of endometrial cancer. In this study, the  Pgr Cre/+  mouse line was used in which Cre recombinase is expressed in cells of the female reproductive tract under the control of the  Pgr  promoter [ 39 ].  Pgr Cre/+  mice were crossed with  Pten fl/fl  and/or  Pgrmc2 fl/fl  mice to conditionally ablate one or both copies of these genes in  Pgr -expressing cells.  Figure 2 A shows the gross morphology of female reproductive tracts isolated from control ( Pten +/fl ;  Pgrmc2 fl/fl ),  Pten +/d ,  Pten +/d ;  Pgrmc2 d/d , Pten d/d , and  Pten d/d ;  Pgrmc2 d/d  mice one week after ovariectomy at nine months of age. The exterior surface of tracts from  Pten +/d  and  Pten d/d  were rough compared to the smooth surface of tracts isolated from  Pten +/d ; Pgrmc2 d/d  and  Pten d/d ;  Pgrmc2 d/d  mice. Overt signs of carcinoma were clearly present in  Pten d/d  mice, particularly at the distal end of the uterus near the oviduct and the cervix. Images of the H&E-stained uterine cross-sections are shown in  Figure 2 . No signs of hyperplasia or carcinoma were found in control mice ( Figure 2 B). As previously described [ 16 ], non-hyperplastic cystic glands formed in uteri from  Pgrmc2 d/d  mice, consistent with a premature aging phenotype ( Figure 2 C). Also shown are representative images of atypical endometrial hyperplasia ( Figure 2 D) and carcinoma ( Figure 2 E) commonly observed in uteri isolated from  Pten d/d  and  Pten d/d ;  Pgrmc2 d/d  mice.\nThe incidence of endometrial hyperplasia and cancer was determined in mice with 10 different  Pten  and  Pgrmc2  genotypes as outlined in  Table 2 . Reproductive tracts were collected for analyses at nine months of age, except for  Pten d/d  mice which were collected at 4–9 months of age due to the lethality of this phenotype. While the tract wet weight was not different between  Pten +/d  and  Pten +/d ; Pgrmc2 d/d  mice, hyperplasia was reduced from 88.9% in  Pten +/d  mice to 46.4% in  Pten +/d ; Pgrmc2 d/d  mice ( Figure 3 A,B). The ablation of  Pgrmc2  on a  Pten +/d  background also decreased the incidence of endometrial cancer to 7.1% compared to the 22.2% in  Pten +/d  mice ( Figure 3 C). Similarly, the tract weight did not differ between  Pten d/d ,  and  Pten d/d ; Pgrmc2 d/d  mice ( Figure 4 ). The ablation of  Pgrmc2  (i.e.,  Pten d/d ; Pgrmc2 d/d ) did not reduce the 100% incidence of hyperplasia or cancer observed in  Pten d/d  mice ( Table 2 ). However, whereas all  Pten d/d  mice had to be euthanized prior to the end of the study at nine months, 100% of  Pten d/d ; Pgrmc2 d/d  mice survived to nine months of age, suggesting that the ablation of  Pgrmc2  attenuated the progression of endometrial cancer ( Figure 4 B).\nThe mRNA levels of  Pgr  and  Esr1  were evaluated in uterine tissues from  Pten +/fl ; Pgrmc2 fl/fl  (control),  Pten +/d ,  and  Pten +/d ; Pgrmc2 d/d  mice and found not to be different between the groups ( Figure 5 ). Given that PGRMC proteins regulate proliferation in granulosa cells and xenograft tumors [ 15 , 18 , 25 , 26 , 27 , 30 , 32 , 40 ], we next evaluated endometrial luminal and glandular epithelial cell mitosis by phospho-histone H3 immunohistochemsitry in young sexually mature ovariectomized  Pten +/fl ; Pgrmc2 fl/fl ,  Pten +/d ,  and  Pten +/d ; Pgrmc2 d/d  mice treated with either vehicle or E2. While there was no difference in basal luminal epithelial cell mitosis across genotypes, basal glandular epithelial mitosis was approximately six times higher in uterine tissue from  Pten +/d  and  Pten +/d ; Pgrmc2 d/d  mice than from  Pten +/fl ; Pgrmc2 fl/fl  mice. The ablation of  Pgrmc2  had no effect on the elevated basal mitosis observed in young  Pten +/d  mice ( Figure 6 A). The treatment of  Pten +/fl ; Pgrmc2 fl/fl  (control),  Pten +/d ,  and  Pten +/d ; Pgrmc2 d/d  mice with E2 equitably elevated luminal epithelial cell proliferation to approximately 10%. The ablation of a single copy of the  Pten  gene tended ( p  = 0.07) to increase E2-induced glandular epithelial mitosis by about 40% over glandular epithelium from  Pten +/fl ; Pgrmc2 fl/fl  mice. Interestingly, the elevated mitosis observed in  Pten +/d  mice was lost upon the ablation of  Pgrmc2  in  Pten +/d ; Pgrmc2 d/d  mice ( Figure 6 B). In aged ovariectomized mice, basal endometrial luminal epithelial cell mitosis was not different in  Pten +/fl ; Pgrmc2 fl/fl ,  Pgrmc2 d/d ,  Pten +/d ,  and  Pten +/d ; Pgrmc2 d/d  mice. In contrast, glandular epithelial cell mitosis was significantly higher in  Pten +/d  mice than in all other groups, including  Pten +/d ; Pgrmc2 d/d  mice ( Figure 6 C).\nConditional  Pten  heterozygosity or deficiency in the uterus causes inflammation [ 6 ]. In the final study, we quantified the uterine infiltration of mast cells and their degranulation in  Pten +/fl ; Pgrmc2 fl/fl ,  Pten +/d ,  and  Pten +/d ; Pgrmc2 d/d  mice. As shown in  Figure 7 , an approximate 2.5-fold increase in mast cells infiltrated occurred in uterine tissue from  Pten +/d  mice compared with  Pten +/fl ; Pgrmc2 flfl  control mice. The conditional ablation of  Pgrmc2  in  Pten +/d  mice failed to reduce mast cell infiltration. Approximately 10% of the total mast cells were degranulated in all tissues examined. As with total mast cells, the number of degranulated mast cells was about 2.5-fold higher in uterine tissues isolated from  Pten +/d  and  Pten +/d ; Pgrmc2 d/d  mice compared with uterine tissue from  Pten +/fl ; Pgrmc2 fl/fl  mice ( Figure 7 B).\n\nCoupled with supporting mutations in the cell cycle, adhesion, and/or apoptosis genes, mutations of tumor suppressor genes are causally responsible for neoplastic transformation in various organs. Endometrioid endometrial carcinoma, which generally exists as type 1 endometrial cancer, is the most common gynecological cancer. Mutations in the tumor suppressor gene  PTEN  occur in over 80% of type 1 endometrial cancer cases [ 5 ]. PTEN is a phosphatase that attenuates the activity of the AKT, a kinase activated by pro-survival and pro-growth ligand:receptor-initiated phosphorylation cascades. The loss of even one  PTEN  allele reduces the overall cellular PTEN phosphatase activity. This haploinsufficiency results in accelerated proliferation and neoplastic transformation in epithelial tissues of several organs due, in part, to its role in maintaining the chromatin structure and the integrity of the genome [ 3 , 41 ]. Because many tumor suppressor genes play vital roles in diverse normal cellular functions, the global ablation or mutation of these genes generally causes embryonic lethality. To circumvent this limitation, the  loxP-Cre  system allows for the ablation, mutation, or overexpression of genes in a cell-specific fashion [ 42 ]. The objective of this study was to conditionally ablate the  Pten  gene from endometrial cells and evaluate the consequences of concomitant  Pgrmc2  ablation. This was accomplished by using the  Pgr-Cre  mouse, which expresses Cre recombinase in cells that normally express the classical  Pgr . Of note, an endometrial-specific Cre mouse line does not exist, so the  Pgr-Cre  mouse is routinely used to ablate, mutate, or overexpress genes in the endometrium despite having Cre recombinase expression in other organs of the female reproductive tract, peri-ovulatory follicles, mammary gland, and gonadotropes [ 39 ]. This approach has been used extensively to determine causal relationships between  PTEN  and other genes. For instance, deleting the  glucose-regulated protein-78  ( Grp78 ) gene prevented the development of endometrial carcinoma in  Pten -ablated mice [ 43 ]. Similarly, because  Pten  deletion deregulated lipid metabolism [ 7 ], the overexpression of an  mfat-1  transgene in mice reinstated lipid homeostasis and prevented  Pten -heterozygous mice from developing hyperplasia and cancer [ 44 ]. The overexpression of the tumor suppressor  mitogen-inducible gene-6  ( Mig-6 ) abrogated the development of  Pten -loss-of function-induced endometrial carcinoma [ 45 ], whereas the ablation of  Mig-6  accelerated the timeline of developing endometrial carcinoma and its severity [ 46 ]. The deletion of  epithelial cadherin  ( Cdh1 ) or the tumor suppressor  liver kinase B1  (Lkb1) along with  Pten  increases the severity of endometrial cancer invasiveness in mice [ 9 , 47 ]. The use of this genetic approach should facilitate the identification of gene targets for the treatment of  Pten -related endometrial cancer. By example, the pharmacological inhibition of mammalian target of rapamycin (mTOR) signaling was shown to substantially inhibit the growth and progression of  Pten / Lkb1 -deficient endometrial cancer. Given that E2 enhances and P4 attenuates type 1 endometrial cancer, the  Pten  loss-of-function model of endometrial cancer has utility in understanding the endocrine actions of these female sex steroids [ 2 , 48 , 49 , 50 ].\nPGRMC2 expression was not different between endometrial cancer tissue compared with endometrial tissue obtained during the proliferative and secretory phase of the menstrual cycle ( Figure 1 ). The ablation of  Pgrmc2  reduced the percentage of  Pten +/d  mice developing hyperplasia by about 50% and those developing endometrial cancer by 66% ( Table 2  and  Figure 3 ). More striking was the survival data when both  Pten  alleles were ablated. Here, 100% of  Pten d/d  mice developed endometrial cancer and were euthanized prior to the end of the nine-month trial. In contrast, 100% of  Pten d/d ; Pgrmc2 d/d  mice survived to nine months despite all developing endometrial cancer suggesting the Pgrmc2 not only attenuates tumor growth, but also the likely metastatic spread.\nThe actions of PGRMC2 are consistent with prior studies in which  PGRMC1  knockdown in endometrial cancer cells dramatically reduced xenograft tumor growth in immunocompromised mice while also elevating chemosensitivity [ 31 ]. Similar findings were observed in PGRMC1-deplete ovarian and breast cancer xenograft tumors [ 30 , 32 ]. Our prior evaluation of  Pgrmc1 d/d  and/or  Pgrmc2 d/d  endometrial tissue demonstrated that the ESR1 and PGR levels do not differ from the corresponding control tissues in which  Pgrmc  genes remain intact [ 16 , 17 ]. Consistent with these findings,  Esr1  and  Pgr  mRNA levels did not change following ablation with  Pten  and/or  Pgrmc2  in the present study ( Figure 5 ). Additional studies are needed to determine if  Pgrmc2  fits the generally accepted definition of a proto-oncogene, a normal gene that produces proteins which helps cells grow and proliferate, or to simply help cells survive. The ablation of  Pgrmc2  clearly attenuated the incidence/development, progression, and aggressiveness of  Pten  loss-of-function-induced endometrial cancer by inhibiting endometrial glandular epithelial cell proliferation ( Figure 6 ). A great deal of excitement surrounds the notion that inflammation initiates neoplastic transformation and/or promotes the tumor progression of solid tumors [ 51 , 52 ]. The conditional ablation of  Pten  using  Pgr-Cre  mice elevates endometrial inflammation [ 6 ]. We assessed the recruitment of mast cells in  Pten +/fl ; Pgrmc2 fl/fl ,  Pten +/d , and  Pten +/d ; Pgrmc2 d/d  mice and observed an increase in endometrial mast cells in  Pten +/d  mice compared with  Pten +/fl ; Pgrmc2 fl/fl  control mice ( Figure 7 ). The ablation of  Pgrmc2  had no effect on mast cell recruitment, indicating that the ability of a  Pgrmc2  deficiency to slow the progression of endometrial cancer on a  Pten  haploinsufficient background does not stem from changes in mast cell recruitment or degranulation.\n\nThrough the use of a well-established mouse model of  Pten  loss-of-function-induced endometrioid endometrial carcinoma, this study demonstrates that the loss of  Pgrmc2  attenuates endometrial hyperplasia and cancer incidence and severity, in part, by inhibiting endometrial glandular epithelial cell proliferation. These findings on a member of the PGRMC family conducted using a mouse model of endometrial cancer advance our prior studies demonstrating that the knockdown of  PGRMC1  in a human endometrial xenograft tumor attenuated tumor development, growth, and progression in vivo [ 31 ], a finding also supported by ovarian and breast cancer xenograft studies [ 30 , 32 ].","source_license":"CC-BY-4.0","license_restricted":false}