European Journal of Pharmaceutics and Biopharmaceutics 130 (2018) 272–280 Contents lists available at ScienceDirect European Journal of Pharmaceutics and Biopharmaceutics journal homepage: www.elsevier.com/locate/ejpb Research paper Inﬂuence of the estrus cycle of the mouse on the disposition of SHetA2 after vaginal administration T Sanjida Mahjabeena, Manolya Kukut Hatipoglua, Doris M. Benbrooka,b, Stanley D. Kosankec, ⁎ David Garcia-Contrerasd, Lucila Garcia-Contrerasa, a Department of Pharmaceutical Sciences, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA Department of Obstetrics and Gynecology, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA c College of Dentistry, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA d Medicina Veterinaria, Facultad de Estudios Superiores Cuautitlán, Mexico b A R T I C LE I N FO A B S T R A C T Keywords: Cervical dysplasia SHetA2 Vaginal drug delivery Tissue absorption Estrous cycle Preclinical PK/PD SHetA2 is a novel compound with the potential to treat cervical dysplasia, but has poor water solubility. A vaginal suppository formulation was able to achieve therapeutic concentrations in the cervix of mice, but these concentrations were variable. Histological analysis indicated that mice in the same group were in diﬀerent stages of their estrous cycle, which is known to induce anatomical changes in their gynecological tissues. We investigated the eﬀects of these changes on the pharmacokinetics and pharmacodynamics of SHetA2 when administered vaginally. Mice were synchronized to be either in estrous or diestrus stage for administration of the SHetA2 suppository. Pharmacokinetic parameters were calculated from the SHetA2 concentrations vs. time data. The reduction in the expression of cyclin D1 protein in the cervix was used as pharmacodynamic endpoint. Mice dosed during diestrus had a larger AUCcervix (335 μg mL h−1), higher Cmax (121.8 ± 38.7 µg/g) and longer t1/2−1 , 44.6 ± 29.5 µg/g and 3.6 h respectively). cervix (30.3 h) compared to mice dosed during estrus (120 μg mL h Therapeutic concentrations of SHetA2 were maintained for 48 h in the cervix of mice dosed during diestrus and for only 12 h in the estrus group. The treatment reduced the expression of cyclin D1 protein in the cervix of mice in the estrus to a larger extent. These results indicate that the estrous cycle of mice inﬂuences signiﬁcantly the disposition of SHetA2 after vaginal administration and may also inﬂuence its eﬃcacy. 1. Introduction Cervical dysplasia is a precancerous condition of the uterine cervix, mainly initiated by high risk papilloma virus (HPV) infection . The prevalence of HPV infection in developed countries is reported to range between 40% and 80% in young adult females [2,3]. This infection can be asymptomatic and resolved by the immune system of the individual, but if the infection persists it can progress towards invasive cervical cancer, which is second most common form of gynecologic cancer worldwide . Because of this possibility, physicians may over treat the patient. Current treatments for cervical dysplasia involve invasive procedures such as ablation therapy and cold knife-conization therapy, which are expensive, cause discomfort and can lead to infertility [5,6]. Presently, there are no drug-based treatments in clinical practice for this disease. SHetA2 (Fig. 1) is a novel, non-toxic chemotherapeutic agent , with strong chemopreventive activity in human cell-culture  and ⁎ murine tumors . However, it has low oral bioavailability (∼10%) due to its poor water solubility . To overcome these limitations, we employed Quality by Design (QbD) approaches to develop a vaginal suppository formulation for direct delivery to the cervix, the site of drug action . The formulation was optimized using two consecutive designs of experiments (DoE). The composition of a hydrophilic suppository base (PEG 400: PEG3350) and the percentage of solubilizing agent (Kolliphor) were optimized in DoE1 (16 experiments). Subsequently, the optimized hydrophilic base (35% PEG 400, 60% PEG 3350 and 5% Kolliphor) was entered in a second DoE (DoE2), which compared the eﬀect of the type of base (hydrophilic versus lipophilic base (cocoa butter)) and the % drug content on the integrity and disintegration/ melting time of the suppository. DoE2 determined that a cocoa butter base and 40% drug content produced suppositories that remained solid at room temperature and melted in 6 min. We performed proof-of concept pharmacokinetic and pharmacodynamics studies in mice to evaluate the potential of the SHetA2 vaginal Corresponding author at: The University of Oklahoma Health Sciences Center 1110 N. Stonewall Ave., Oklahoma City, OK 73126-0901, USA. E-mail address: firstname.lastname@example.org (L. Garcia-Contreras). https://doi.org/10.1016/j.ejpb.2018.07.004 Received 22 May 2018; Received in revised form 29 June 2018; Accepted 3 July 2018 Available online 04 July 2018 0939-6411/ © 2018 Elsevier B.V. All rights reserved. European Journal of Pharmaceutics and Biopharmaceutics 130 (2018) 272–280 S. Mahjabeen et al. 2.2. Methods 2.2.1. Suppository manufacturing and quality control SHetA2 suppositories containing 15 mg/kg body weight dose were manufactured by the fusion-molding method . Suppositories were evaluated for content uniformity (85% to 115% of intended content), weight variation (no more than two units having a relative standard deviation (RDS) > 7.8%) and softening time (less than 30 min) as outlined by the USP . Fig. 1. Chemical structure of SHetA2. 2.2.2. Animals Friend Leukemia Virus B (FVB) female mice of 7 weeks of age (National Cancer Institute Charles River Frederick Research Facility) were used in this study because they are the wild type species for the K14-HPV16 mouse model of cervical neoplasia  that will be used in eﬃcacy studies. Animals were housed in a facility in a constant temperature room at 22 ± 1 °C with a 12 h light/12 h dark cycle and provided access to food and water ad libitum. All animal experiments were approved by University of Oklahoma Health Sciences Center Institutional Animal Care and Use Committee (IACUC). suppositories to treat cervical dysplasia . SHetA2 suppositories achieved cervix concentrations signiﬁcantly higher than its predicted therapeutic concentration (4 µM) and induced a 9-fold decrease in the levels of cyclin D1 protein, a marker of eﬃcacy associated with the prognosis in cervical cancer patients . However, a large variability was observed in the SHetA2 concentrations among mice receiving the same dose, at the same time point, even though the quality control indicated that drug variation in the batch of suppositories was within United States Pharmacopoeia (USP) limits . Thus, we investigated the inﬂuence of the anatomical and physiological features of the reproductive system of the mouse on SHetA2 absorption. Unlike humans that have a menstrual cycle, the reproductive cycle of the mouse, known as estrous cycle, has four diﬀerent stages: proestrus (P), estrus (E), metestrus (M) and diestrus (D) . Due to differential hormonal regulation, the composition, thickness and structure of the stratiﬁed squamous and mucosal epithelium in the gynecologic tissues of mice are variable across the estrous cycle . Histological examination of the uterine horns of the mice treated with the same SHetA2 dose in preliminary studies revealed that mice within the same time point group were in diﬀerent stages of their estrus cycle. Currently, there are no published reports describing the eﬀects of anatomical and physiological changes due to estrus cycle on drug disposition after vaginal administration of compounds to mice. A single report from Hsu et al.  using an excised vaginal membrane in a diﬀusion chamber indicated that the permeability coeﬃcients for vidarabine, an antiviral drug with activity against herpes, were 10–100 fold higher during the diestrus stage compared to those in the estrus stage. Therefore, the objective of the present study was to evaluate the inﬂuence of the anatomical and physiological changes due to the estrus cycle of the mouse on SHetA2 disposition after vaginal administration, and SHetA2 pharmacodynamic endpoint, the reduction in the levels of cyclin D1 protein. 2.2.3. Monitoring estrus cycle by the visual method The vaginal openings of mice were monitored every 24 h to determine their stage in the estrous cycle as described by Byers et al . According to this method, during proestrus the vaginal opening remains swollen, moist and pink; during estrus, the opening becomes less moist and less swollen; during metestrus, a distinctive white cellular debris is observed, but the tissue is not swollen; whereas during diestrus, the vaginal opening is very narrow and not swollen. 2.2.4. Synchronization of estrous cycle The estrous cycle in mice was synchronized by the Whitten eﬀect, where female mice are exposed to male pheromones from their urine, which induce them to enter the estrus stage on the third day of exposure [18,19]. Since urine from male mice was not readily available, we used the soiled bedding from male mice cage for a modiﬁed Whitten eﬀect . To verify the stage of the estrus cycle in these mice, vaginal lavage was performed every 24 h on the third, fourth and ﬁfth day of exposure as described below. 2.2.5. Monitoring estrus cycle by observation of vaginal cytology The vaginal lavage was performed in each mouse under light sedation with isoﬂurane as described by McLean et al . A sterile pipette tip was ﬁlled with approximately 10 μL of sterile saline and inserted gently into the vaginal cavity of the mouse, followed by gentle aspiration, and the procedure repeated three to ﬁve times. The collected ﬂuid was then smeared onto a microscope glass slide and air-dried. The slides were stained with crystal violet after drying. Slides were examined with a microscope to determine the type of cells that were present in the smear. The stages of the estrous cycle were determined according to the percentage of anucleated corniﬁed cells, nucleated epithelial cells and leukocytes presented in the smear as follows . A mouse was determined to be in the: (1) pro-estrus stage when nucleated epithelial cells were predominant; (2) estrus stage when mainly anucleated corniﬁed cells were present; (3) metestrus stage when three types of cells, leukocytes, corniﬁed, and nucleated epithelial cells, were present; and (4) diestrus stage when the majority of cells present were leukocytes . 2. Materials and methods 2.1. Materials SHetA2 was synthesized by Cayman Chemical company, Inc. under a contract from the Rapid Access to Preventive Intervention Development (RAPID) National Cancer Institute (NCI) program. Cocoa butter was purchased from Nature’s Oils (Streetsboro, OH). Kolliphor HS-15 was obtained from BASF (Germany). Sterile saline and isoﬂurane were obtained from Henry Schein Animal Health Inc. Acetonitrile (HPLC grade ≥ 99.5%), methanol (HPLC grade ≥ 99.5%), phosphoric acid, hydrochloric acid, crystal violet stain and sodium acetate trihydrate were purchased from Sigma Aldrich (St Louis, MO). Captiva® ﬁltration equipment was purchased from Agilent Technologies Inc. for extraction of drug from tissues. Mouse cyclin D1 enzyme linked immunosorbent assay kit (ELISA) was purchased from Cedarlane lab (NC). T-PER (tissue protein extraction reagent) was purchased from ThermoFisher Scientiﬁc (Waltham, MA). Protease inhibitor cocktail tablets were purchased from Sigma Aldrich (St Louis, MO). 2.2.6. Histology of the uterine horns A piece of a uterine horn from each mouse was collected, ﬁxed in 10% neutralized buﬀered formalin and embedded in a paraﬃn block. Afterwards, sections were cut perpendicularly to the transverse axis from the paraﬃn blocks, so that the lumen and stromal areas were included in each section. Each tissue section was ﬁxed onto a microscope slide and stained with hematoxylin and eosin (H&E) for 273 European Journal of Pharmaceutics and Biopharmaceutics 130 (2018) 272–280 S. Mahjabeen et al. MRTcervix = AUMCcervix/AUCcervix were determined by non-compartmental analysis using Phoenix WinNonlin® software. The average SHetA2 cervix concentration of the 5 mice in each time point was used in the calculation. Statistical analysis of cyclin D1 expression levels in the cervix of mice was performed by Graphpad Prism software using two way ANOVA and Tukey’s multiple comparison test. A p value of less than 0.01 was considered to be statistically signiﬁcant. histological analysis. 2.2.7. Pharmacokinetic study SHetA2 suppositories (15 mg/kg) were administered vaginally to female mice when they were either in diestrus or estrus stage, in order to determine the inﬂuence of the resulting anatomical and physiological changes on the extent of drug absorption. We selected these two stages based on a preliminary study, in which mice in diestrus stage exhibited the highest drug concentrations in the cervix and those in estrus stage exhibited the lowest. Also, it is at these two stages that the hormonal levels are the most diﬀerent, with diestrus and estrus stages corresponding to the secretory and proliferative phases of the human uterine cycle, respectively . Mice were lightly sedated with isoﬂurane to facilitate suppository insertion and maintained under sedation for 3 min in horizontal position to improve the retention of the suppository. Mice were then transferred to a clean cage lined with paper and observed for 10 additional minutes to account for any possible leakage of the yellow colored suppository. Mice dosed during diestrus were euthanized at 0.5, 1, 4, 8, 12, 24, 36 and 48 h after dosing (n = 5 mice, per time point), whereas mice dosed during estrus stage were euthanized at 0.5, 1, 4, 8 and 12 h after dosing (n = 5 mice, per time point). Blood was collected by cardiac puncture into heparinized tubes and plasma was separated after centrifugation for 10 min at 12,500g. The cervix tissues were collected and thoroughly washed by saline to remove unabsorbed drug. Plasma and gynecologic tissues were kept frozen at −80 °C until analysis, where each cervix was divided into two pieces, to be used in pharmacokinetic and pharmacodynamic determinations. In addition, mice (n = 3 mice) were dosed with placebo suppositories (cocoa butter and Kolliphor only) and euthanized at the times of highest and lowest drug concentrations in cervix to be used as controls for the pharmacodynamic endpoint. Cervix tissues were collected, washed with saline and kept frozen at −80 °C. SHetA2 was extracted from one-half of each cervix tissue (n = 5 per time point) using a Captiva® ﬁltration system and drug levels were analyzed by HPLC using a method validated in our laboratory . A Waters Alliance HPLC System equipped with Waters Xbrigde C18 3.5 μm, 2.1 × 150 mm column and Waters Xbridge BEH C18, 3.5 μm, 2.1 × 5 mm guard column were used to determine drug concentrations. The mobile phase consisted of acetonitrile and water (80:20 v/v) at ﬂow rate of 0.3 mL/min. 3. Results 3.1. Quality control for suppositories All manufactured SHetA2 suppositories met USP speciﬁcations : content uniformity of 101.3 ± 9.9% and average weight of 54.4 ± 3.8 mg (relative standard deviation 7.1%). The average softening time for the suppositories was 5.6 ± 0.1 min. 3.2. Estrus cycle monitoring by visual method The stages of the estrus cycle for mice in the same cage, as monitored by the physical appearance of their vaginal opening, are presented in Table 1. Despite being housed in same cage, the mice exhibited diﬀerent stages of the estrus cycle. For example, on day 1, two mice were in metestrus, one was in diestrus and one in proestrus. On day 2, one of the mice that was in metestrus remained in that stage whereas the other had already progressed past diestrus into proestrus. Due to the unpredictable variability in the cycle of the mice in the study, it was determined that synchronization of the estrous cycle was needed to reduce variability in the study. 3.3. Monitoring estrus cycle by observation of vaginal cytology The Whitten eﬀect method was eﬀective to synchronize the estrus cycle in the mice in the study as shown in Table 2. On the third day of exposure to male urine, 77 ± 6% of the mice were in estrus stage, and on the fourth day 64 ± 13% of mice were in diestrus stage. Vaginal cytology was also a more unequivocal way to determine the stage of estrus cycle in these mice as the diﬀerent types of cells are readily recognizable in the smears of vaginal lavages. The smear of mice in estrus stage consisted mostly of corniﬁed epithelial cells (Fig. 2A), whereas most cells in the smear of mice in proestrus stage were nucleated epithelial cells (Fig. 2B). In contrast, leukocytes and corniﬁed epithelial cells were abundantly present in the smear of mice in metestrus stage (Fig. 2C), whereas the smear of mice in diestrus stage consisted mainly of leukocytes (Fig. 2D). 2.2.8. Pharmacodynamic study The expression of cyclin D1 protein in the cervix of mice treated with SHetA2 suppositories, placebo suppositories and untreated mice was determined by Enzyme Linked Immunosorbant Assay (ELISA). The one-half of the cervix of each mouse (n = 5, per each time point) was homogenized in T-PER reagent (10 μL per each mg of tissue) containing protease inhibitor cocktail. Homogenization was performed in an ice bath using an OMNI-GLH general laboratory homogenizer. Homogenates were centrifuged at 4 °C for 5 min at 10,000g, the supernatants were collected and the assay was performed according to manufacturer’s instruction. Each standard blank and test specimen was evaluated in duplicate. The replicates with > 15% coeﬃcients of variance was eliminated from the analysis. The average of the test samples was compared to the standard curve to derive the cyclin D1 concentration. 3.4. Histology of uterine horns Fig. 3 shows sections of the uterine horns of mice that conﬁrmed the diﬀerent stages of the estrous cycle: the lumen of the uterine horn of mice in proestrus stage was large with a few mitotic cells in the epithelium (Fig. 3A), whereas the lumen of the uterine horns of mice in estrus stage is the largest in the cycle and has many mitotic cells in the Table 1 Visual monitoring of estrous cycle in female FVB mice of the same cage, without synchronization (P = proestrus, E = estrus, M = metestrus, D = diestrus): 2.2.9. Data analysis Pharmacokinetic parameters to characterize the vaginal disposition of SHetA2 were determined as follows: The Cmax-cervix = maximum concentration in cervix and Tmax-cervix = time to achieve maximum concentration in cervix, were determined from the cervix concentration versus time plot. The AUCcervix = area under the cervix concentration versus time curve, t1/2-cervix = half-life in cervix, Vz/F = apparent volume of distribution/F, Clcervix/F = Clearance/F, and Mouse ID Black Red Blue Green 274 Observation Day 1 2 3 4 5 6 7 8 9 D M P M D M D P P P E E P P E P M M M P P E D P P E D E E M D E M M P D European Journal of Pharmaceutics and Biopharmaceutics 130 (2018) 272–280 S. Mahjabeen et al. diestrus stages, but this concentration was maintained to a diﬀerent extent in these groups. While therapeutic concentrations of SHetA2 in the cervix of mice treated with the suppository during diestrus were maintained for 48 h, the SHetA2 cervix concentrations in mice treated during estrus fell below therapeutic level after 12 h. Even though the time of maximum SHetA2 concentration was the same (Tmaxcervix = 0.5 h) in both treatment groups, the maximum concentration (Cmax) was more than threefold higher when mice were dosed during diestrus (121.8 ± 38.7 µg/ml) compared to when mice were dosed during estrus (44.6 ± 29.5 µg/ml). Interestingly, the cervix concentration versus time proﬁle for both groups exhibited an extra peak that occurred at 4 h for the estrus group and at 12 h for the diestrus group. The pharmacokinetic parameters characterizing the disposition of SHetA2 after vaginal administration are presented in Table 3. The area under the curve (AUCcervix) was almost three-fold higher when mice were dosed during diestrus (335 μg mL h−1) than when dosed treated during estrus (120 μg mL h−1). These AUC correlated with a faster drug clearance in mice treated during estrus (CL/FCervix = 2.6 mL/h) compared to that of mice dosed during diestrus (CL/Fcervix = 0.9 mL/h). Consequently, the elimination half-life (t1/2-cervix) of SHetA2 from the cervix of mice dosed during estrus (3.6 h) was almost 10-fold shorter than in mice dosed during diestrus (30.3 h) and the mean residence time (MRTcervix) was 8-fold shorter. Table 2 Monitoring of estrus cycle by microscopic examination of vaginal smears from female FVB mice of the same cage, after synchronization by the Whitten eﬀect. % of mice in each stage Third day of exposure Fourth day of exposure Estrus/Proestrus Diestrus 77% ± 6% 23% ± 6% 35% ± 15% 64% ± 13% epithelium (Fig. 3B). In contrast, the lumen of the uterine horns of mice in metestrus stage becomes smaller and the number of mitotic cells decreases (Fig. 3C), whereas the lumen of the uterine horns of mice in diestrus stage is the smallest in the cycle, there are no mitotic cells present and the epithelial layer is thin (Fig. 3D). 3.5. Pharmacokinetic study SHetA2 suppositories melted inside the vaginal cavity of mice within the 3 min that they were kept sedated after insertion and the formulation absorbed in the majority of cases as evidenced by very little to no leakage of the formulation noted within the 10 min that the mice were observed after recovering from anesthesia. The SHetA2 cervix tissue concentration versus time proﬁle obtained from mice receiving the vaginal suppositories as a function of the estrus stage is shown in Fig. 4. Administration of the optimized SHetA2 vaginal suppositories achieved concentrations above the therapeutic level (4.0 μM or 1.6 µg/mL)  in the cervix of mice in both estrus and Fig. 2. Cell composition of the vaginal lavages from mice synchronized by the modiﬁed Whitten eﬀect showing the four diﬀerent estrus stages: (A) Proestrus (mostly nucleated epithelial cells, marked with red arrow), (B) Estrus (mostly corniﬁed epithelial cells, marked with black arrow), (C) Metestrus (mostly nucleated epithelial cells, few leukocyte and corniﬁed epithelial cell); (D) Diestrus (mostly leukocytes, marked with blue arrow). 275 European Journal of Pharmaceutics and Biopharmaceutics 130 (2018) 272–280 S. Mahjabeen et al. Fig. 3. H&E stained slides from transversal sections of the uterine horns of mice receiving a 15 mg/kg SHetA2 vaginal suppository during: (A) Proestrus, (B) Estrus, (C) Metestrus, and D) Diestrus. Diestrus group SHetA2 concentration in cervix ( g/g) 1000 Table 3 Pharmacokinetic parameters characterizing the disposition of SHetA2 in the cervix tissue after vaginal administration to FVB mice (n = 5–7 mice per time point in each group). Estrus group 100 10 Therapeutic conc 0 1 PK parameter 0.1 −1 AUCcervix (μg mL h Cmax (μg/g) t1/2-cervix (h) Cl/Fcervix (mL/h) Vz/Fcervix (mL) MRTcervix (h) 0.01 0.001 0.0001 0.00001 0 10 20 30 40 ) Estrus group Diestrus group 120 44.6 ± 29.5 3.6 2.6 13.5 5.2 335 121.8 ± 38.7 30.3 0.9 41.3 41.0 50 Time (h) cyclin D1 expression in mice receiving placebo suppositories was also 15% lower compared to untreated controls, but this diﬀerence was not statistically signiﬁcant. However, while the cyclin D1 levels in mice receiving the placebo suppositories returned to the same level as untreated controls after 24 h of treatment, the levels of cyclin D1 remained signiﬁcantly reduced in mice treated with SHetA2 suppositories (Fig. 5A). Notably, the diﬀerences in cyclin D1 levels between groups dosed during estrus was more signiﬁcant, as the cyclin D1 reduction in mice treated with SHetA2 suppositories was statistically lower than that in untreated mice and those receiving placebo suppositories at both 0.5 h and 24 h (Fig. 5B). Fig. 4. SHetA2 cervix tissue concentration versus time proﬁle obtained from FVB female mice treated with 15 mg/kg vaginal suppositories as a function of diestrus the estrus stage (n = 5–7 mice, per time point in each treatment) ( estrus group). group, 3.6. Pharmacodynamic study The levels of cyclin D1 protein employed as a pharmacodynamic endpoint to evaluate the eﬀect of SHetA2 levels in cervix tissue are depicted in Fig. 5. Mice dosed during diestrus with SHetA2 suppositories exhibited signiﬁcantly lower levels of cyclin D1 (approximately 50% lower) at Tmax = 0.5 h compared to untreated mice (Fig. 5A). The 276 European Journal of Pharmaceutics and Biopharmaceutics 130 (2018) 272–280 S. Mahjabeen et al. Diestrus group A) cervical dysplasia, based on their capability to achieve therapeutic concentrations at the site of action and the reduction of the cyclin D1 protein expression . However, it was necessary to address the cause of the large variations in drug concentrations observed in the cervix from mice within the same experimental group because they could possibly compromise the eﬀectiveness of the proposed treatment. The present study revealed that the disposition of SHetA2 after vaginal administration of the same dose was signiﬁcantly inﬂuenced by the stage of the estrus cycle in mice. The diﬀerence between the SHetA2 cervix concentrations observed in mice dosed during diestrus and estrus (Fig. 4) was attributed to diﬀerences in the cell composition of the vaginal epithelium during these stages, which in turn inﬂuenced drug absorption. During estrus, the epithelium in the vaginal cavity of mice is thick and the superﬁcial layer is formed by large, anuclear corniﬁed epithelial cells [29,30] that would limit drug absorption. In contrast, during diestrus, the epithelium is thin and is formed by polygonal, plump epithelial cells due to early muciﬁcation, which are more likely to absorb drug as indicated by a higher Cmax-cervix in the diestrus group. Another signiﬁcant diﬀerence was the mean residence time of SHetA2 in the cervix of mice, which was 8-fold shorter in mice dosed during estrus (5.2 h), compared to that of mice dosed during diestrus (41.0 h). As a result, SHetA2 concentration in the cervix of mice dosed during estrus fell below therapeutic levels 12 h after the suppository was administrated, whereas, SHetA2 concentration in the cervix of mice dosed during diestrus was maintained at a therapeutic level until 48 h. The possible explanations for these diﬀerences are oﬀered in Fig. 6. The duration of the estrus stage in mice is reported to be 12 h [30,31], at the end of estrus, the corniﬁed superﬁcial layer of the vaginal epithelium (Fig. 6A) is gradually lost (Fig. 6B) and becomes completely delaminated at the beginning of metestrus . Consequently, when SHetA2 is administered during the estrus stages, it is likely that most of the drug is lost together with the anucleated corniﬁed cells at the end of the stage, which correlates with cervix concentrations falling below therapeutic levels (12 h, Fig. 6C). In contrast, the duration of the diestrus stage is 65 h [30,31], with the epithelial plump cells remaining intact for this period of time (Fig. 6B) . Thus, when SHetA2 is administered during the diestrus stages, it is likely that the drug remains in the epithelial cell layers with a larger probability to be absorbed, which correlates with cervix concentrations remaining above therapeutic levels for a longer period of time (> 48 h, Fig. 6C). These diﬀerences observed in the disposition of SHetA2 were also reﬂected in their respective pharmacokinetic parameters (Table 3), with mice dosed during diestrus group exhibiting larger AUCcervix, longer t1/2-cervix and MRTcervix time compared to those observed in mice dosed during estrus. An unexpected secondary peak was observed in the SHetA2 cervix tissue concentration versus time proﬁle for both treatment groups. It is unlikely that these peaks are due to enterohepatic recirculation as the drug was under the levels of detection in plasma at all time points and the concentrations are from the cervix tissue. A more plausible explanation is that SHetA2 was ﬁrst dissolved and then absorbed in tissues, but as the concentration increased inside the tissue, the drug may have precipitated and then dissolved and reabsorbed again. The appearance of double peaks in the plasma concentration versus time proﬁles following extravascular (mainly oral) administration of drugs has also been reported for other drugs [32–36]. Among these drugs, danazol is poorly soluble in water (0.017 mg/mL) like SHetA2 (0.4 µg/ mL) and has a log P value (3.48) similar to that of SHetA2 (3.84) [4,10]. It has been hypothesized that a portion of the danazol administered orally is rapidly solubilized in the stomach but precipitates in the intestine and might again solubilize aided by bile salts . Jackson et al  performed a dissolution study to investigate the precipitation and solubilization kinetics of danazol formulations using nanoparticle tracking analysis and microscopy. They reported that as the drug concentration increased and the solution became saturated, nucleation increased and droplet like precipitates were formed. The dissolution proﬁle of SHetA2 in simulated vaginal ﬂuid reported in our previous Untreated Control Placebo SHetA2 (15mg/kg) suppository ns ns Fold change in cyclin D1 expression 1.5 ** ** ** ns 1.0 0.5 0. 24 5 ho ho ur ur 0.0 Treatment time (h) Estrus group B) Control Placebo 1.5 SHetA2 suppository Fold change in cyclin D1 expression ** ** ** ** 1.0 0.5 ho 12 0. 5 ho ur ur 0.0 Treatment time (h) Fig. 5. Cyclin D1 expression level in the cervix of mice treated with 15 mg/kg SHetA2 vaginal suppository: (A) diestrus group (n = 3–5 mice, per time point in each treatment), (B) estrus group (n = 3–5 mice, per time point in each treatment). Cyclin D1 expression in placebo and treatment group was normalized with the expression level in untreated control (** indicates p value less than 0.005, ns = not signiﬁcant p values were derived using two-way ANOVA Tukey’s post test, n = 3). 4. Discussion Cervical cancer aﬀects about a half million women worldwide, but the number of women aﬀected by cervical dysplasia is almost 30-fold larger . Thus, treating cervical dysplasia would have a larger impact on the patient population and potentially reduce the number of patients aﬀected by cervical cancer. To date, only a handful of animal models for cervical dysplasia and cervical cancer have been established, and the majority of these models are based in mice . In particular, the K14-HPV16 transgenic mouse model develops multiple hyper-proliferative and dysplastic lesions that are similar to those observed in human [26,27]. FVB female mice were employed in the present study, as it is the wild type strain to generate the K14-HPV16 mouse model [27,28]. Proof-of-concept studies support the potential of vaginal suppositories to deliver SHetA2 as a promising, non-invasive therapy for 277 European Journal of Pharmaceutics and Biopharmaceutics 130 (2018) 272–280 S. Mahjabeen et al. Fig. 6. Schematic diagram of the vaginal epithelium during, (A) estrus and (B) diestrus: (A) The superﬁcial layer of the vaginal epithelium is corniﬁed and this layer is gradually lost at the end of this stage (12 h) [30,31]. The corniﬁed layer can aﬀect drug absorption and when is lost at the end of the stage, drug may be lost with it. The superﬁcial layer consists of nucleated epithelial cell and the epithelium is intact for the duration of the stage (65 h), therefore a prolonged residence time is expected. (C) Time course of SHetA2 cervix concentration according to length of stages: diestrus (in black line) and estrus (in red line). protein, which plays essential role in G1 to S phase entry in cell cycle . In G1 phase, cyclin D1 promotes cell cycle progression to S phase upon binding to cyclin dependent kinase 4 and 6. SHetA2 induces G1 cell cycle arrest by degradation of cyclin D1 . In Fig. 3, a large number of mitotic bodies are observed around the lumen of the uterine horns from mice in estrus stage indicating active cell division and thus more opportunity for SHetA2 activity. In contrast, the mitotic activity around the lumen of the uterine horns from mice in diestrus stage is minimal and thus the reduction in cyclin D1 levels is minimal. These observations are in agreement with other publications which indicate that the highest mitotic activity in the inner layer of the vaginal epithelium and uterine horns occurs during estrus, decreased during metestrus and its minimal during diestrus [12,39]. Given the diﬀerences in the menstrual cycle of humans and the estrus cycle of mice, it would be interesting to speculate about the implication of the ﬁndings of the present study. In contrast to the four stages of the estrus cycle of mice, a regular menstrual cycle has three phases, according to the events in both ovaries and uterus . The study exhibited a similar pattern . Therefore, it is plausible that SHetA2 may have precipitated at the site of absorption and then dissolved and re-absorbed again, but more studies are needed to conﬁrm this assumption. The eﬀect of the estrus cycle was also observed on the pharmacodynamic endpoint marker cyclin D1. Although the levels of cyclin D1 in the cervix of mice treated with SHetA2 during diestrus were signiﬁcantly reduced at 0.5 h compared to those in untreated controls (Fig. 5A), this reduction was not statistically diﬀerent to the reduction of cyclin D1 levels in the cervix of mice treated with placebo suppositories. In contrast, the levels of cyclin D1 in the cervix of mice treated with SHetA2 during estrus were signiﬁcantly reduced compared to both, placebo treated and untreated mice (Fig. 5B). The cyclin D1 reduction was also larger in mice dosed during estrus compared to that observed during estrus (Fig. 5A and 5B). The diﬀerences in the pharmacodynamic eﬀect of SHetA2 during estrus and diestrus stages may be explained by the diﬀerence in the mitotic activity of the cells of the epithelium during these stages. Cyclin D1 is a cell cycle regulatory 278 European Journal of Pharmaceutics and Biopharmaceutics 130 (2018) 272–280 S. Mahjabeen et al. for future eﬃcacy studies in the K14-HPV16 mouse model of cervical dysplasia. 6. Acknowledgements This work was funded by the MD Anderson Cancer Center Moon Shot grant with the help of Dr. Michael Frumovitz, the Stephenson Cancer Center (SCC) Gynecologic Cancers Program at the OUHSC and R01 CA196200126-01A1 (DMB and LGC, Co-PIs). The authors are grateful to SCC tissue pathology core for preparation of uterine horns histology slides and H&E staining. References  F. Remoue, N. Jacobs, V. Miot, J. Boniver, P. Delvenne, High intraepithelial expression of estrogen and progesterone receptors in the transformation zone of the uterine cervix, Am. J. Obstet. Gynecol. 189 (2003) 1660–1665.  F.X. Bosch, T.R. Broker, D. Forman, A.B. Moscicki, M.L. Gillison, J. Doorbar, P.L. Stern, M. Stanley, M. Arbyn, M. Poljak, J. Cuzick, P.E. Castle, J.T. Schiller, L.E. Markowitz, W.A. Fisher, K. Canfell, L.A. Denny, E.L. Franco, M. Steben, M.A. Kane, M. Schiﬀman, C.J. Meijer, R. Sankaranarayanan, X. Castellsague, J.J. Kim, M. Brotons, L. Alemany, G. Albero, M. Diaz, S. de Sanjose, Comprehensive control of human papillomavirus infections and related diseases, Vaccine 31 (Suppl. 8) (2013) I1–I31.  M. Saraiya, E.R. Unger, T.D. Thompson, C.F. Lynch, B.Y. Hernandez, C.W. Lyu, M. Steinau, M. Watson, E.J. Wilkinson, C. Hopenhayn, G. Copeland, W. Cozen, E.S. Peters, Y. Huang, M.S. Saber, S. Altekruse, M.T. Goodman, US assessment of HPV types in cancers: implications for current and 9-valent HPV vaccines, J. Natl. Cancer Inst. 107 (2015) djv086.  M. Scott, M. Nakagawa, A.-B. Moscicki, Cell-mediated immune response to human papillomavirus infection, Clin. Diagn. Lab. Immunol. 8 (2001) 209–220.  M. Herfs, C.P. Crum, Cervical cancer: Squamocolumnar junction ablation[mdash] tying up loose ends? Nat. Rev. Clin. Oncol. 12 (2015) 378–380.  IQWiG, Institute for Quality and Eﬃciency in Health Care. Cervical cancer: Cell changes in the cervix (dysplasia), in: I.f.Q.a.E.i.H. Care (Ed.), Informed Health Online [Internet]. Cologne, Germany: IQWiG, Cologne, Germany, 2013.  K.K. Kabirov, I.M. Kapetanovic, D.M. Benbrook, N. Dinger, I. Mankovskaya, A. Zakharov, C. Detrisac, M. Pereira, T. Martin-Jimenez, E. Onua, A. Banerjee, R.B. van Breemen, D. Nikolic, L. Chen, A.V. Lyubimov, Oral toxicity and pharmacokinetic studies of SHetA2, a new chemopreventive agent, in rats and dogs, Drug Chem. Toxicol. 36 (2013) 284–295.  D.M. Benbrook, S. Lightfoot, J. Ranger-Moore, T. Liu, S. Chengedza, W.L. Berry, I. Dozmorov, Gene expression analysis of biological systems driving an organotypic model of endometrial carcinogenesis and chemoprevention, GeneRegul Syst. Bio. 2 (2008) 21–42.  D.M. Benbrook, S. Guruswamy, Y. Wang, Z. Sun, A. Mohammed, Y. Zhang, Q. Li, C.V. Rao, Chemoprevention of colon and small intestinal tumorigenesis in APC(min/+) mice by SHetA2 (NSC721689) without toxicity, cancer prevention research, (Philadelphia, Pa.) 6 (2013) 908–916.  S. Mahjabeen, M.K. Hatipoglu, V. Chandra, D.M. Benbrook, L. Garcia-Contreras, Optimization of a vaginal suppository formulation to deliver SHetA2 as a novel treatment for cervical dysplasia, J. Pharm. Sci. 107 (2018) 638–646.  D.S. Bae, S.B. Cho, Y.J. Kim, J.D. Whang, S.Y. Song, C.S. Park, D.S. Kim, J.H. Lee, Aberrant expression of cyclin D1 is associated with poor prognosis in early stage cervical cancer of the uterus, Gynecol. Oncol. 81 (2001) 341–347.  The Staﬀ of the Jackson Laboratory, Biology of the Laboratory Mouse. http://www. informatics.jax.org/greenbook/tables/table11-1.shtml.  K. Bertolin, B.D. Murph, https://www.alnmag.com/article/2014/07/reproductivetract-changes-during-mouse-estrous-cycle.  C.C. Hsu, J.Y. Park, N.F. Ho, W.I. Higuchi, J.L. Fox, Topical vaginal drug delivery I: eﬀect of the estrous cycle on vaginal membrane permeability and diﬀusivity of vidarabine in mice, J. Pharmaceut. Sci. 72 (1983) 674–680.  C.E. Wainwright, A.L. Quittner, D.E. Geller, C. Nakamura, J.L. Wooldridge, R.L. Gibson, S. Lewis, A.B. Montgomery, Aztreonam for inhalation solution (AZLI) in patients with cystic ﬁbrosis, mild lung impairment, and P. aeruginosa, J. Cystic Fibrosis: Oﬀ. J. Eur. Cystic Fibrosis Soc. 10 (2011) 234–242.  K. Smith-McCune, Y.H. Zhu, D. Hanahan, J. Arbeit, Cross-species comparison of angiogenesis during the premalignant stages of squamous carcinogenesis in the human cervix and K14-HPV16 transgenic mice, Cancer Res. 57 (1997) 1294–1300.  S.L. Byers, M.V. Wiles, S.L. Dunn, R.A. Taft, Mouse estrous cycle identiﬁcation tool and images, PLoS One 7 (2012) e35538.  S.J. Dalal, J.S. Estep, I.E. Valentin-Bon, A.E. Jerse, Standardization of the Whitten Eﬀect to induce susceptibility to Neisseria gonorrhoeae in female mice, Contemp. Top. Lab. Anim. Sci. 40 (2001) 13–17.  B.K. Gangrade, C.J. Dominic, Studies of the male-originating pheromones involved in the Whitten eﬀect and Bruce eﬀect in mice, Biol. Reprod. 31 (1984) 89–96.  A.C. McLean, N. Valenzuela, S. Fai, S.A.L. Bennett, Performing vaginal lavage, crystal violet staining, and vaginal cytological evaluation for mouse estrous cycle staging identiﬁcation, J. Visual. Exp.: JoVE (2012) 4389.  C. Caligioni, Assessing reproductive status/stages in mice, current protocols in neuroscience/editorial board, Jacqueline N. Crawley, et al., APPENDIX (2009) Fig. 7. Comparison between human uterine cycle and mouse estrus cycle according to estradiol and progesterone levels: (A) hormonal level in human uterine cycle, (B) hormonal level in mouse estrous cycle. Blue arrow indicates the ideal time of vaginal drug administration to maximize the drug concentration in gynecological tissues. thickness and viscosity of the human cervical mucus is altered during these diﬀerent phases , but unlike the mouse, the thickness of the human vaginal epithelium does not drastically change across the cycle. Although a few studies indicate minor changes , the clinical relevance of these changes has not been demonstrated. In order to apply the ﬁndings of the present study, the menstrual cycle of humans as inﬂuenced by its main hormones has been schematically depicted (Fig. 7A) next to the estrus cycle of the mouse (Fig. 7B). According to the hormonal levels, the diestrus stage in the mouse corresponds to the human secretory phase, whereas the proestrus and estrus stages correspond to the human proliferative phase of the menstrual cycle [39,41]. Even though the epithelium of the human vagina and uterus is not signiﬁcantly aﬀected by the stages, the changes in mucus volume secretion during the menstrual cycle may inﬂuence the extent of drug absorption . Consequently, drug absorption may be aﬀected during secretory phase when mucus production is high. Therefore, the ideal time for vaginal administration of drugs to humans would be at the beginning of the secretory phase or right after menses, as shown by the blue arrows in Fig. 7A and B. 5. Conclusion The extent of drug absorption in the cervix of FVB mice appears to be inﬂuenced by the stage of the estrus cycle. SHetA2 absorption and residence time in the cervix is maximized during diestrus stage, whereas the pharmacodynamics eﬀects appears to be favored during estrus stage. These eﬀects need to be balanced and taken into account for the design of the dosing regimen with SHetA2 vaginal suppositories 279 European Journal of Pharmaceutics and Biopharmaceutics 130 (2018) 272–280 S. Mahjabeen et al. Appendix-4I.  H. Wang, S.K. Dey, Roadmap to embryo implantation: clues from mouse models, Nat. Rev. Genet. 7 (2006) 185–199.  J.N.G.P. Loyd, V. Allen, Howard C. Ansel, Ansel's Pharmaceutical Dosage Forms and Drug Delivery Systems, 9th ed.,, Wolters Kluwer/ Lippincott Williams & Wilkins, Philadelphia, PA, 2011.  C.P. Masamha, D.M. Benbrook, Cyclin D1 degradation is suﬃcient to induce G1 cell cycle arrest despite constitutive expression of cyclin E2 in ovarian cancer cells, Cancer Res. 69 (2009) 6565–6572.  R. Siegel, J. Ma, Z. Zou, A. Jemal, Cancer statistics, 2014, CA: Cancer J. Clin. 64 (2014) 9–29.  L.I. Larmour, T.W. Jobling, C.E. Gargett, A review of current animal models for the study of cervical dysplasia and cervical carcinoma, Int. J. Gynecol. Cancer 25 (2015) 1345–1352.  R.R. Riley, S. Duensing, T. Brake, K. Munger, P.F. Lambert, J.M. Arbeit, Dissection of human papillomavirus E6 and E7 function in transgenic mouse models of cervical carcinogenesis, Cancer Res. 63 (2003) 4862–4871.  J.M. Arbeit, K. Münger, P.M. Howley, D. Hanahan, Progressive squamous epithelial neoplasia in K14-human papillomavirus type 16 transgenic mice, J. Virol. 68 (1994) 4358–4368.  E. Hernández-Pichardo, F. Fernández-Reyes, S.S. Cortés, U.U. Xochimilco, Fundamentos teórico prácticos de la citología exfoliativa en medicina veterinaria, Universidad Autónoma Metropolitana, Xochimilco Mexico, 1999.  P.F. Galassi-Geréz, F.A. Gullace, Reproducción en animales de laboratorio, Universidad de Buenos Aires, Buenos Aires, Argentina, Facultad de Ciencias Veterinarias, 2004.  F.D.M.V.Y. Zootecnia, Manual de prácticas de ﬁsiología animal, Universidad Nacional Autónoma de México, Mexico, D.F., 1980.  W.N. Charman, M.C. Rogge, A.W. Boddy, W.H. Barr, B.M. Berger, Absorption of danazol after administration to diﬀerent sites of the gastrointestinal tract and the          280 relationship to single- and double-peak phenomena in the plasma proﬁles, J. Clin. Pharmacol. 33 (1993) 1207–1213. Y. Plusquellec, G. Campistron, S. Staveris, J. Barre, L. Jung, J.P. Tillement, G. Houin, A double-peak phenomenon in the pharmacokinetics of veralipride after oral administration: a double-site model for drug absorption, J. Pharmacokinet Biopharm. 15 (1987) 225–239. S. Staveris, G. Houin, J.P. Tillement, G. Jamet, M. Schneider, L. Jung, J.C. Koﬀel, Primary dose-dependent pharmacokinetic study of veralipride, J. Pharm. Sci. 74 (1985) 94–96. Y. Wang, A. Roy, L. Sun, C.E. Lau, A double-peak phenomenon in the pharmacokinetics of alprazolam after oral administration, Drug Metab. Dispos. 27 (1999) 855–859. R. Zhang, P. Mao, T. Zhang, C. Ma, B. Jin, T. Li, Pharmacokinetic analysis and tissue distribution of Vam3 in the rat by a validated LC-MS/MS method, Acta Pharmaceutica Sinica B 5 (2015) 254–263. M.J. Jackson, U.S. Kestur, M.A. Hussain, L.S. Taylor, Characterization of supersaturated danazol solutions – impact of polymers on solution properties and phase transitions, Pharm. Res. 33 (2016) 1276–1288. J.P. Alao, The regulation of cyclin D1 degradation: roles in cancer development and the potential for therapeutic invention, Mol. Cancer 6 (2007) 24. D.L. Patton, S.S. Thwin, A. Meier, T.M. Hooton, A.E. Stapleton, D.A. Eschenbach, Epithelial cell layer thickness and immune cell populations in the normal human vagina at diﬀerent stages of the menstrual cycle, Am. J. Obstet. Gynecol. 183 (2000) 967–973. S.R. Davis, I. Lambrinoudaki, M. Lumsden, G.D. Mishra, L. Pal, M. Rees, N. Santoro, T. Simoncini, Menopause, Nat. Rev. Dis. Primers 1 (2015) 15004. B.G. Reed, B.R. Carr, Endotext [Internet], The Normal Menstrual Cycle and the Control of Ovulation. In: L.J. De Groot, C. G., K. Dungan (Eds.). MDText.com, Inc.: South Dartmouth (MA), 2015 May 22.