Glutathione-Responsive Methotrexate Polymersomes for Potential Management of Ectopic Pregnancy.

Because MTX remains the primary treatment for EP, we hypothesized that MTX would function more effectively and safely when encapsulated within a specifically constructed biocompatible nanocarrier capable of effective drug delivery and release at the implantation site. Polymersomes, self-assembled bilayer polymeric vesicles that are composed of amphiphilic di-block copolymers, [ 24 ] were selected for this purpose due to their capability of encapsulating water-soluble molecules such as MTX (sodium salt) within the hydrophilic core and hydrophobic molecules (e.g., NIR dye) within the hydrophobic bilayer ( Figure 1A – C ). Using a microfluidic mixing approach, we prepared MTX-loaded polymersomes composed of an amphiphilic di-block copolymer ( Figure S1 ) with a hydrophobic polycaprolactone (PCL) block to generate the stable polymersome bilayer and a polyethylene glycol (PEG) block to create a hydrophilic core as well as a water-soluble shell. In addition, the polymersomes utilized in this study were constructed from an amphiphilic PEG-PCL copolymer containing a disulfide bond between the PEG and PCL blocks (PEG-SS-PCL, Figure S1 ) for glutathione (GSH)-triggered intracellular release of the drug cargo. The high concentrations of intracellular GSH are expected to rapidly reduce an incorporated disulfide bond within the polymersome upon internalization into placental cells, resulting in the disintegration of nanocarriers and efficient drug release. [ 25 , 26 ] The developed disulfide bond-containing polymersomes loaded with MTX (MTX-SS-Ps, Figure 1A ) have a spherical shape ( Figure 1G ), nearly neutral surface charge (−2.62 ± 0.45 mV) ( Table S1 ), uniform distribution (PDI (polydispersity index) of 0.11 ± 0.02) and a hydrodynamic size of 38.2 ± 0.4 nm ( Figure 1D ), that is comparable in size and charge to the same polymersome lacking a disulfide bond (MTX-Ps, Figure 1E , H ), or when loaded with NIR dye ( Figure 1F and Table S1 ). Storage of MTX-loaded polymersomes for 8 weeks at 4°C resulted in only a minimal change in size ( Figure 1I ), suggesting favorable stability of this formulation and extended shelf life. The release of MTX from polymersomes (MTX-Ps and MTX-SS-Ps) was studied in the presence and absence of GSH at physiological pH (7.4) ( Figures 1J – L ). Cytosolic concentrations of GSH are between 1 to 10 mM, [ 27 ] while plasma concentrations of GSH are within 1 to 5 μM. [ 28 , 29 ] As shown in Figure 1J , MTX-SS-Ps efficiently release the drug at both 1 and 10 mM GSH concentrations, exceeding 90% release at 10 mM GSH and 80% release at 1mM GSH, reaching a plateau at ~20 hours. However, at 5 μM GSH, only a negligible release of drug from MTX-SS-Ps was observed after 7 hours, and less than 15% release after 20 hours. MTX release from both MTX-Ps and MTX-SS-Ps in PBS at pH 7.4 follows a similar pattern in the absence of GSH with a minimal release ( Figure 1K ). When comparing the MTX release from MTX-Ps to that of MTX-SS-Ps at 10 mM GSH ( Figure 1L ), drug release from MTX-Ps was minimal and resembled the release pattern observed for MTX-Ps in the absence of GSH. These release studies suggest that MTX-SS-Ps could ensure the GSH-triggered intracellular release of MTX for effective therapeutic outcomes. Recently, we showed that naphthalocyanine derivatives, when encapsulated within polymeric nanoparticles, are capable of producing a strong photoacoustic (PA) signature that is distinct from the background. [ 30 ] In the present study, we constructed a polymersome-based companion imaging agent by encapsulating silicon 2,3-naphthalocyanine bis(trihexylsilyloxide) (SiNc), a hydrophobic, photostable, near infra-red dye ( Figure 1C and S1 ). Polymersomes loaded with SiNc (NIR-SS-Ps) demonstrate a strong fluorescence signal in aqueous solution at ~ 780 nm ( Figure S2A ), as well as a strong and narrow PA peak at ~ 780 nm ( Figure S2B ), a region of the electromagnetic spectrum where endogenous contrast agents, such as oxygenated and deoxygenated hemoglobin, demonstrate their lowest PA signal. [ 22 , 31 ] The aforementioned PA properties, confirmed in solution studies, make these promising candidates for further in vivo evaluation. In the absence of an animal model for EP, [ 32 – 35 ] the delivery of imaging and therapeutic agents to the placenta has been studied using the pregnant murine model since the human and mouse placenta are anatomically and functionally comparable. [ 20 , 23 , 36 ] Additionally, while the location of implantation varies between normal and ectopic pregnancies, the mechanisms of implantation and placentation remain the same. [ 37 ] Thus, with the goal of timely management of EP, we evaluated the distribution of NIR dye-loaded polymersomes with and without disulfide bonds (NIR-SS-Ps and NIR-Ps) in a pregnant mouse model at gestational day (Gd)6.5, which corresponds to the early phase of the first trimester of a human pregnancy ( Figure 2 ). [ 38 , 39 ] Since the process of implantation and placentation is the same regardless of location, our goal was to demonstrate that MTX-loaded polymersome with S-S-bond can accumulate in the developing vascularized placenta, efficiently release MTX in placental cells with high levels of GSH, and as a result cause the demise of the associated fetus. This would imply that EP, presumably with a similarly vascularized placenta, could act as a sink for the developed polymersomes following systemic administration, and the released MTX would provide the required therapeutic effect. At 24 hours post-intravenous (i.v.) injection, both NIR-Ps and NIR-SS-Ps produce a prominent fluorescence signal in each implantation site of the gravid uterus ( Figure 2B ). Fluorescence imaging of resected organs revealed similar accumulation of both NIR-Ps and NIR-SS-Ps in uteri and major organs of pregnant mice at Gd7.5 - mostly in kidney, liver, and spleen - 24 hours following i.v. administration at Gd6.5 ( Figure 2A , C ). While some accumulation of nanoparticles was observed in the lungs and heart, the signal was similar to background levels, and there was no statistically significant difference in fluorescence intensity between these organs and those of control mice injected with saline. Importantly, polymersomes preferentially accumulate in the uterus as compared to other organs. For instance, the fluorescence signal of NIR polymersomes in kidneys and liver was ~50% and 80% lower than that observed in the uterus, respectively ( Figure 2C ). It is important to emphasize that the mean fluorescence signal generated by NIR-SS-Ps (0.18±0.06) in the liver is only 10% lower than the signal (0.20±0.05) of the non-responsive polymersome (NIR-Ps), and this difference is not statistically significant ( Figure 2C ). Similarly, the mean fluorescence signal generated by NIR-SS-Ps (0.45±0.06) in the kidneys is only 8% lower than the signal (0.49±0.05) of the non-responsive polymersome (NIR-Ps), and this difference is also not statistically significant. Therefore, it would be reasonable to conclude that NIR-SS-Ps and NIR-Ps demonstrate similar accumulation in various tissues following systemic administration. The similar biodistribution of both formulations is expected because NIR-SS-Ps and NIR-Ps have very similar hydrodynamic size (39.3 nm vs 37.4 nm, Table S1 ), charge (−0.58 mV vs −0.29 mV) and surface modification (2kDa PEG vs 2kDa PEG). Following ex vivo assessment of polymersome distribution in organs and implantation sites 24 hours after administration with a small animal imaging system, fluorescence microscopy was employed to further characterize the distribution of polymersomes in fetoplacental tissues of mice administered NIR-Ps and NIR-SS-Ps ( Figure 2D , E ). The intense fluorescence signal was observed in the antimesometrial decidua, the ectoplacental cone, and the region surrounding the implantation crypt, as confirmed by subsequent H&E staining of the same tissue sections ( Figure 2D , E ). No fluorescence signal was observed within the primitive streak-stage embryo. In addition to the biodistribution profile of NIR-SS-Ps at 24 hours after injection, we also examined their distribution in the uterus and other organs at earlier time points. Our results revealed that the developed polymersomes efficiently accumulate in the implantation sites as early as 1 hour after injection ( Figure S3 ). At this time point, NIR fluorescence generated by NIR-SS-Ps was also detected in the liver, kidneys, lungs, heart, and spleen, albeit at a lower intensity than at the implantation sites. Our findings further revealed that polymersomes were significantly eliminated from the aforementioned organs within 12 hours while remaining at the implantation sites ( Figure S3 ). The application of nanoparticles for the delivery of imaging and therapeutic agents to the implantation sites is still a nascent field of research; therefore, there is insufficient knowledge about the physicochemical parameters of nanoparticles that can provide efficient and preferential accumulation of nanomedicines in the placenta after systemic administration. Given the fact that nanoparticles can efficiently accumulate in angiogenesis-dependent cancer and endometriotic lesions via passive targeting [ 40 ] and that the placenta is a highly vascularized tissue, [ 41 ] we hypothesized that some fundamental principles of cancer and endometriosis nanomedicine can be used to develop nanoparticles for preferential delivery of MTX to implantation sites. Previous research has shown that polymeric nanoparticles with sizes ranging from 30 to 100 nm, a near neutral charge, and a PEGylated surface can efficiently deliver cargo to cancer and endometriosis lesions. [ 42 , 43 ] Literature also suggests that PEG with a molecular weight of 2kDa and higher reduces the recognition of nanoparticles by the mononuclear phagocyte system, increases their circulation time, and decreases accumulation in various organs. [ 44 – 46 ] In addition, it was demonstrated that a suitable PEG coating can reduce retention of polymeric nanoparticles in the liver, spleen, and lungs. [ 47 ] Therefore, we developed disulfide bond containing polymersomes with a hydrodynamic size of ~39 nm, nearly neutral surface charge and a 2kDa PEG coating. Our biodistribution studies validated our hypothesis by demonstrating that polymersomes with the above-mentioned physicochemical parameters can efficiently accumulate and retain in the placenta while being substantially eliminated from other organs. The mechanism of nanoparticle retention in the placenta when compared to other organs remains unclear; thus, additional research is required. As the detection of extrauterine gestation is the most accurate indicator of EP, advancements in real-time imaging technologies are greatly valuable for EP diagnosis. Photoacoustic imaging (PAI) offers deeper tissue penetration than other optical modalities, such as fluorescence imaging, due to its generation of acoustic waves, which encounter significantly less scattering and subsequent signal attenuation than optical waves. [ 48 ] PAI operates without the use of harmful ionizing radiation and exploits the natural endogenous photoacoustic contrast of tissue components such as melanin, blood, etc. [ 49 ] Exogenous contrast agents with pronounced PA signals further enhance clear distinction of targeted tissue from the background (tissue, blood, or water). Therefore, we evaluated the PAI capabilities of NIR polymersomes as well as their distribution in the murine placenta following the establishment of blood flow. In humans, the main placental structure forms by day 21 and maternal blood flow is established by day 80, and both of these processes are completed in the first trimester of human pregnancy, which corresponds to early gestation in mice of Gd6.5–13.5. [ 50 – 53 ] Thus, the effectiveness of NIR polymersome as a PA contrast agent was evaluated for visualizing the fetoplacental unit at Gd12.5. We presume that NIR-SS-Ps would concentrate in the rapidly dividing cells and highly vascular tissues of the growing placenta after systemic administration. The placental labyrinth contains an extensive network of blood flow and enables the transfer of oxygen and nutrients between the maternal and fetal circulations, suggesting that NIR polymersomes would localize preferentially within the labyrinth. Ultrasound (US) and PA images were acquired with the same transducer enabling overlay of anatomical characteristics of US images with the photoacoustic signal. A clear, intense PA signal (green) was observed in the amnion and placenta 24 hours post-injection, but not in the fetus, of mice administered NIR-SS-Ps at Gd12.5 ( Figure 3B ), when compared to the same tissues of control mice administered saline, which only displayed PA signal for blood (red) ( Figure 3A ). Recorded PA spectra confirm these findings, as evidenced by a spectral peak for NIR dye (780 nm) in the placenta of mice administered NIR-SS-Ps and the absence of this signal in the fetus ( Figure 3D ), compared to control mice administered saline, where only spectral background signal is observed in fetus and placenta ( Figure 3C ). The histological analysis of the implantation sites further suggests that the developed nano-formulations are predominantly localized in the placenta ( Figure 3F ). Fluorescence microscopy was used to confirm NIR-SS-Ps distribution in fetoplacental units of mice injected at Gd12.5. The intense fluorescence signal was observed in the placental labyrinth and decidua but absent from the junctional zone that separates these regions ( Figure 3E , F ). Fluorescence was also observed in the avascular amnion surrounding the fetus (embryo) but completely absent from the fetus itself. There is no indication that the nanoparticle reaches the fetus at mid-gestation, suggesting that NIR polymersomes are not transported across the placental barrier. Hence, these disulfide-containing polymersome delivery vehicles may be employed to transport other imaging and therapeutic cargos, for the treatment of pregnancy complications or other diseases during pregnancy, without endangering the fetus. Together, our data suggest that NIR polymersomes efficiently accumulate in implantations during early mouse pregnancy (Gd6.5–13.5), before and after placental blood flow is established (comparable to 10 weeks of human pregnancy), and facilitate PA visualization of the placenta. Our nanomedicine approach has the potential to address issues with small molecule methotrexate-chemotherapy for EP, by increasing the efficacy of delivery and release of MTX at the implantation site thanks to specifically designed nanocarriers. Embryonic development in CD-1 pregnant dams was monitored using a US imaging system (Vevo 2100, VisualSonics, Canada) every other day through Gd13.5 after i.v. administration of a two-dose regimen ( Figure 4A ) of either MTX-SS-Ps (1 mg kg −1 of MTX at Gd6.5 and 8.5) or various controls (saline, free MTX (1 mg kg −1 at Gd6.5 and 8.5), empty polymersome (Ps), empty disulfide containing polymersome (SS-Ps), or MTX-Ps (1 mg kg −1 of MTX at Gd6.5 and 8.5)). Although free MTX was administered at the clinically recommended dose of 1 mg kg −1 for EP treatment, [ 54 , 55 ] it did not affect fetal development in the pregnant murine model, similar to saline and empty polymersomes, according to US imaging and measurements of the gestational sac length ( Figure 4B , C ). The administration of MTX-Ps at the same dosing regimen (1 mg kg −1 at Gd6.5 and 8.5) demonstrated inhibition of fetal development after the 2nd dose as seen by US, suggesting increased MTX delivery to the implantation site as compared to free MTX at the same dosing regimen. However, while fetal development was severely inhibited by treatment with MTX-Ps, fetal and placental tissues had not been completely resorbed by Gd13.5. In contrast, US imaging revealed that MTX-SS-Ps with a disulfide bond effectively inhibited fetal development after gestational day 8.5, and pregnancies were completely resorbed by Gd13.5 5 in all mice (n=3) treated with MTX-SS-Ps. ( Figure 4C ). This pronounced therapeutic effect can be attributed to improved MTX delivery to the placenta using the constructed nanocarrier ( Figure 2 ) and efficient triggered MTX release from the disulfide polymersomes in the presence of high cytosolic concentrations of GSH within placental cells. Our drug release studies demonstrates that ~80% of MTX is released from the S-S-bond-containing nanocarrier within 20 hours in the presence of GSH at the concentration of 1 mM and higher ( Figure 1J ). GSH is present in most mammalian cells at concentrations ranging from 1 to 10 mM, with the highest levels of GSH found in the liver at a concentration of 10 mM. [ 27 , 56 ] Previous reports also demonstrate that the level of GSH in the placenta of pregnant mice and rats is only ~1.8 times lower than in the liver. [ 57 , 58 ] Therefore, MTX can be efficiently released from MTX-SS-Ps after accumulation in the placenta containing GSH at concentrations greater than 1 mM. Our results demonstrated the impact that the SS bond, incorporated within the polymersome-based carrier, exerts on drug release and, consequently, therapeutic efficacy. To evaluate the safety of MTX polymersomes, we collected blood samples following euthanization of treated mice at Gd13.5 and measured serum concentrations of proteins and surrogate markers indicating hepatic (alkaline phosphatase (ALP), alanine transaminase (ALT), aspartate aminotransferase (AST)), cardiac (creatine kinase (CK), and renal (blood urea nitrogen (BUN) and creatinine) function. The obtained values in the treatment (MTX-SS-Ps) and control groups were similar, with no statistically significant differences between groups, suggesting that MTX-SS-Ps are safe and well-tolerated ( Figures S4 ). After euthanization at Gd13.5, murine uteri were exteriorized for a more detailed evaluation of fetal development ( Figure 5 ). Uteri from the saline, Ps, SS-Ps, and free MTX groups were virtually indistinguishable from one another, each containing a large litter of apparently healthy, well-developed fetuses and associated placentae ( Figure 5A , B ). Uteri of mice administered MTX-Ps contained severely underdeveloped embryos and no discernable placentae, and the length of the extended uterus was similar to that of a non-pregnant mouse. Distinctively, administration of MTX-SS-Ps, at the same dosing regimen of MTX given in the free MTX and MTX-Ps groups (1 mg kg −1 2x), had a profound effect on implantations, resulting in complete fetal resorption and the uteri resembling those of non-pregnant mice ( Figure 5A , B ). The aforementioned findings demonstrated that polymersomes lacking a stimuli-responsive linkage between their hydrophobic and hydrophilic components (MTX-Ps) can successfully deliver drugs to the site of implantation. Adding such a linkage, on the other hand, significantly improves drug release and, as a result, therapeutic effect. Consequently, MTX can be efficiently released and, through its antimetabolite function, impair embryonic development by inhibiting normal cell growth and division. [ 6 , 59 , 60 ] A detailed histopathological assessment of fetoplacental tissues was performed by a board-certified pathologist for the following treatment groups: saline, free MTX, MTX-Ps, or MTX-SS-Ps ( Figure 6 ). Uteri of mice administered free MTX (1 mg kg −1 MTX, 2x) contained apparently healthy, well-developed fetuses and placentae, with no indications of infarcted tissues or growth retardation, and were histologically similar to those of the saline control group ( Figure 6A , B ). Free MTX at this dosing regimen appeared to exert no effect on fetal, placental, or maternal uterine tissues. Conversely, uteri of mice administered MTX-Ps at the same dose (1 mg kg −1 MTX, 2x) contained much smaller, undeveloped, necrotic embryos that lacked nuclear staining, and placentae with severely infarcted regions, primarily in the decidua ( Figure 6C ). The maternal uterine tissues of these mice appeared unaffected, with abundant, viable endometrial glands. Notably, administration of MTX-SS-Ps at the same dose (1 mg kg −1 MTX, 2x, at Gd6.5 and Gd8.5) resulted in the demise and complete resorption of all pregnancies by Gd13.5, and uteri of these mice resembled virgin mouse uterus with healthy endometrial glands and stroma( Figure 6D ), as described by a board-certified pathologist. We assessed the safety of MTX-SS-Ps for the mother and fetus. Three groups of pregnant mice at Gd6.5 were i.v. injected with saline, free MTX (1 mg kg −1 MTX at Gd6.5 and 8.5), and MTX-SS-Ps (1 mg kg −1 MTX at Gd6.5 and 8.5). Dams in all groups were allowed to continue gestation following treatment, give birth, nurse pups, and mate again, aiming to assess the success of mating and subsequent pregnancy. Following delivery, individual pup weights from each litter of each group were recorded and compared to evaluate neonatal development. Pups in the saline and free MTX groups were delivered successfully and developed properly, confirming once more that free MTX (at 1 mg kg −1 MTX, 2x) had no effect on the development of the pups. Pup weights in free MTX and saline treatment groups were not statistically different ( Figure 7A ). Because administration of MTX-SS-Ps resulted in complete fetal resorption ( Figure 5 and 6 ), no pup weights were reported for this group, and weights for this group are given as 0 g on each post-natal day (pnd) for the duration of the study ( Figure 7A ). These results indicate that a two-dose regimen of free MTX at 1 mg kg −1 had no effect on fetal development, while the same dose administered in the disulfide polymersome formulation was successful at inducing the demise of all pregnancies. Pups from the free MTX group continued to gain weight through pnd 21 ( Figure 7A ), and resembled the neonate growth rate in the control group. Pups appeared healthy, ambulatory, actively feeding, well hydrated, and had smooth coats by pnd 21, at which time pups were weaned. To assess the effects of MTX-SS-Ps on subsequent fertility, dams from the previous experiments (all three groups) were paired with male mice until copulation plugs appeared, and mice were then separated. All dams from the three groups successfully mated and demonstrated similar gestation lengths. All mice subsequently birthed large litters (average 14 pups/litter) of healthy pups, implying that the previous treatment with MTX-SS-Ps had no noticeable effect on the dams’ subsequent fertility. Individual pup weights were recorded weekly from pnd 1 to pnd 21 ( Figure 7B ), and pups from all three groups appeared healthy and continued to gain weight, with no obvious indications of toxicity resulting from the dams’ previous treatment with either free MTX or MTX-SS-Ps, and together with no indicated changes in blood markers in dams ( Figure S4 ), MTX polymersomes appear to be a safe treatment approach. Because there was no observable effect of free MTX on fetal development when administered in a two-dose regimen of 1 mg kg −1 , and in order to compare the efficacy of MTX-SS-Ps to that of free drug, we proceeded to identify a dose of free MTX that would exert an inhibitory effect on fetal development. Pregnant mice were administered a two-dose regimen of free MTX at 2, 4, or 6 mg kg −1 , with a single i.v. injection on Gd6.5 and Gd8.5. Pregnant mice from the 2 and 4 mg kg −1 groups continued to increase in size and ultimately birthed litters (average 11 pups), while dams administered 6 mg kg −1 free MTX appeared non-pregnant throughout the study and did not birth any pups ( Figure 7C ). These results indicate that the MTX-SS-Ps formulation was capable of terminating pregnancy at a dose 6-fold less than that of free MTX.

EP, the abnormal implantation of an embryo in the uterus or sites outside of the uterine cavity, can lead to rupture and hemorrhage if not identified and treated in a timely manner, and is the primary contributor to maternal morbidity and mortality in the first trimester. In this study, we tested our hypothesis that GSH-responsive polymersomes can efficiently encapsulate and preferentially deliver MTX to implantation sites while ensuring nearly complete triggered drug release. The MTX-loaded polymersomes were produced with a size of less than 40 nm and a nearly neutral surface charge, and released 95% of MTX during 20 hours. These polymersomes successfully accumulate in mouse implantation sites within 24 hours of administration at both Gd6.5 and Gd12.5, as shown by fluorescence and PA imaging in addition to histological analysis. The developed MTX nanomedicine effectively suppressed fetal development and completely eliminated conceptuses in a pregnant mouse model at a six-fold lower dose than free MTX. Because the developed formulation is biocompatible, and dams successfully conceived and birthed healthy pups following a prior complete pregnancy demise induced by MTX-SS-Ps, MTX nanomedicine appears to be a safe therapeutic method. These results suggest that the developed MTX nanomedicine can address several challenges associated with the currently used free MTX regimen for ectopic pregnancy management, including insufficient accumulation of the drug at the ectopic implantation site, a high failure rate of MTX treatment and various side effects. It has the potential to improve the clinical management of ectopic pregnancy, minimize side effects associated with free MTX, and consequently reduce associated mortality rates and costs. In addition, these disulfide-containing polymersomes could also be used for the efficient delivery of various therapeutic cargos for the treatment of placental disorders or other pregnancy complications without harming the fetus.

PEG(2k)-PCL(5k) (methoxy poly (ethylene glycol)- b -poly(ε-caprolactone) copolymer was purchased from Akina Inc (West Lafayette, IN). PEG(2k)-SS-PCL(5k) (methoxy poly (ethylene glycol)- b -disulfide-poly(ε-caprolactone) copolymer was synthesized by Ruixibiotech (Xian, China). NIR dye (silicon 2,3-naphthalocyanine bis (trihexylsilyloxide)) was purchased from Alfa Chemistry (Ronkonkoma, NY, USA). USP-grade methotrexate sodium salt (MTX) was obtained from OHSU Pharmacy (Hikma Pharmaceuticals USA Inc., Berkeley Heights, NJ). MilliporeSigma (Milwaukee, WI), Fisher Scientific Inc. (Hampton, NH) and VWR International, LLC (Radnor, PA) provided other common chemicals and supplies. MTX-loaded polymersomes were prepared using PEG-PCL and PEG-SS-PCL copolymers [ 61 , 62 ] via a microfluidic mixing method. Copolymer (10.0 mg) and MTX (15.0 mg) were dissolved in 1 mL of acetone and 1 mL of saline, respectively. Then, obtained solutions were loaded into two Hamilton glass syringes, and a Harvard DDS dual independent channel syringe pump (Holliston, MA) was used for mixing aqueous and organic phases at a flow rate of 2.2 mL min −1 through a microfluidic mixer chip (Precigenome, San Jose, CA). Next, the organic solvent was evaporated, and an Amicon Ultra-4 centrifugal filter unit (Merck Millipore, MWCO: 30KDa, 10 min, 5500 rpm) was used to remove the unencapsulated drug. Finally, the MTX-loaded nanoparticle solution was passed through a 0.22 μm filter for the final purification to produce MTX-Ps or MTX-SS-Ps. NIR dye (naphthalocyanine derivative) was loaded into polymersomes using a similar method; copolymer (10.0 mg, PEG-PCL or PEG-SS-PCL) and NIR dye (0.3 mg) were dissolved in tetrahydrofuran (THF, 1 mL total), and NIR-Ps or NIR-SS-Ps were generated using a glass microfluidic mixer chip (Precigenome, San Jose, CA) compatible with THF. The size, surface charge, and polydispersity of the polymersomes were determined using a Malvern ZetaSizer (Worcestershire, UK). Cryogenic Transmission Electron Microscopy (Cryo-TEM) (Thermo Fisher Scientific, Waltham MA) was used for assessing the morphology of nanoparticles. [ 63 – 66 ] Loading of MTX in the polymersome nanoparticles was evaluated at 302 nm by preparing a calibration curve of serial MTX solutions [ 3 – 5 ] using high-performance liquid chromatography (HPLC, Shimadzu, Japan) with an Agilent ZORBAX C-18 column (3.5 μm, 4.6 × 75 mm) at a flow rate of 1 mL min −1 and a mobile phase comprised of acetonitrile/water (35:65 v/v%) stabilized with 0.1% trifluoroacetic acid. The stability of MTX-encapsulated polymersomes in saline, stored at 4°C, was evaluated for 8 weeks by examining the size and polydispersity index of the nanoparticles. [ 4 ] Drug (MTX) release from the polymersomes was monitored in the presence of different concentrations of glutathione (GSH) in phosphate-buffered saline (PBS) as a function of time, described in more detail in Supporting Information (SI) . [ 62 ] Fluorescence and photoacoustic images (and spectra) were recorded using the Pearl Impulse Small Animal Imaging System (LI-COR, USA) and Vevo LAZR imaging system (FUJIFILM VisualSonics, Toronto, Canada) with the LZ550 probe (operating frequency of 40 MHz), respectively. [ 22 , 40 ] All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Oregon Health and Science University and were carried out in accordance with national and local guidelines and regulations (IP00003848). Pregnant CD1 wild-type mice at selected gestational days (Gd) were obtained from Charles River Laboratories (Wilmington, MA). Pregnant mice at selected Gd6.5 and GD12.5 (3 animals per group, n =3) were injected intravenously (i.v.) via tail vein with NIR-Ps or NIR-SS-Ps (100 μL at 0.3 mg mL −1 SiNc in saline) along with saline as the control group, and biodistribution of NIR dye loaded polymersomes was evaluated in major organs and uterus 24 hours after i.v. injection, using a LI-COR Pearl Impulse Imaging System (LI-COR, Lincoln, NE) with an 800 nm channel, as previously reported. [ 22 ] The mean fluorescence intensity of the region of interest was quantified using Pearl Impulse Software. For fluorescence tissue imaging, conceptuses at Gd7.5 and Gd13.5 (left undisturbed in their amniotic sacs) were immediately frozen in cryomolds using liquid nitrogen and optimal cutting temperature (OCT) medium for subsequent sectioning, histology, and fluorescence imaging as described in Supporting Information . The ultrasound/photoacoustic imaging instrumentation (Vevo LAZR, FUJIFILM VisualSonics, Inc., Toronto, Canada) setting used in these studies employs an ultrasound transducer integrated with a tunable pulsed laser operating within 680–970 nm (LZ550), providing real-time acquisition and simultaneous overlay of ultrasound and photoacoustic images on 2D and 3D planes. [ 22 ] The recommended clinical dose for MTX is ~1 mg kg −1 (or 50 mg m −2 ). In the first step, to convert human effective dose (HED) in mg kg −1 to animal (mouse) effective dose (AED) in mg kg −1 , the following formula was used: AED = HED × 12.3. [ 67 ] Thus, AED = 1 mg kg −1 dose of MTX × 12.3 = 12.3 mg kg −1 for mouse. In the next step, to determine experimentally the lowest dose of free MTX and dosage regimen required for complete pregnancy demise, we conducted a dose-escalation study by administering the drug to mice at a range of doses less than 12.3 mg kg −1 . Briefly, pregnant mice at Gd6.5 (3 animals per group, n =3) were administered a single dose of free MTX (1, 2, 4, or 6 mg kg −1 ) on Gd6.5, and the same dose again on Gd8.5. Mice were allowed to continue gestation after i.v. administration free MTX. Following parturition, pup weights were recorded and analyzed on post-natal day (pnd) 1, 7, 14 and 21. Upon completion of the study, pups were euthanized following an institutionally approved protocol. To compare the therapeutic efficacy of different formulations, pregnant mice at Gd6.5 were divided into 6 different groups ( n =3): control, free MTX, MTX-Ps, MTX-SS-Ps, empty Ps, and empty SS-Ps. The control groups were injected with saline and empty polymersomes (Ps and SS-Ps). The MTX-treatment groups (free MTX, MTX-Ps, and MTX-SS-Ps) were administered a single dose of 1 mg kg −1 free MTX or MTX-loaded polymersome formulations (MTX-Ps and MTX-SS-Ps,1 mg kg −1 MTX) on Gd6.5, and the same dose again on Gd8.5. Embryonic development in all groups was monitored using a Vevo 2100 high-frequency ultrasound imaging system (Visual Sonics, Toronto, Canada) with an MS-550 high-resolution transducer to obtain images. US images were obtained prior to treatment on Gd6.5, and on Gd8.5, 11.5, and 13.5. The images were analyzed using Vevo LAB software version 5.7.0 and gestational sac lengths were measured. Mice were euthanized on Gd13.5, and uteri were collected for ex vivo imaging and subsequent histological analysis. A board-certified pathologist performed a detailed histopathological assessment of H&E-stained thin sections of fetoplacental units from different treatment groups. When euthanized at Gd13.5, blood samples were also collected and analyzed by IDEXX Laboratories (Portland, OR, USA) for the total health profile screen to determine plasma levels of cardiac, renal, and hepatic function indicators. In a separate study, pregnant mice (3 animals per group, n =3) were allowed to continue gestation following i.v. administration of saline, free MTX (1 mg kg −1 MTX, 2x) or MTX-SS-Ps (1 mg kg −1 MTX, 2x) on Gd6.5 and Gd8.5. Following parturition, pups were observed to evaluate overall appearance, locomotor activity, feeding, and vocalization. Pup weights were recorded adn analyzed on post-natal day (pnd) 1, 7, 14 and 21. Upon completion of the study, pups were euthanized following an institutionally approved protocol. After 2 weeks interval, each female mouse from the treated and control groups was paired with a male for the second breeding. When the pregnant mice gave birth, the pups were monitored and weighted on post-natal day (pnd) 1, 7, 14 and 21 (the same way as for the first pregnancy) to evaluate the safety or toxicity effect of the developed formulations on the subsequent pregnancy. The pups’ weights in treatment groups were analyzed and compared to the control groups. In these studies, no data pre-processing was performed. The data was presented using a mean and standard deviation format (mean+/− SD), with the sample size ( n ) for each study specified in the figure legends. For comparisons between two groups, a two-tailed unpaired t-test was employed. For more than two groups, one-way analysis of variance (ANOVA) was used to examine the statistical significance. GraphPad Prism v9 (GraphPad Software, CA, USA) was used to perform all statistical analyses.

Ectopic pregnancy (EP) is defined as the abnormal implantation of an embryo, most often outside of the uterus, and accounts for ~1–2% of pregnancies in the USA. [ 1 , 2 ] Hemorrhage caused by ruptured EP continues to be the primary cause of first-trimester maternal death, accounting for 16% of first-trimester emergency room visits annually. [ 3 , 4 ] EP is presumed when serum human chorionic gonadotropin (hCG) levels surpass 3,000 mIU mL −1 and there is no intrauterine gestational sac visible by ultrasound, and if hCG levels do not decline after uterine evacuation. [ 5 ] The most common treatment of confirmed or presumed early unruptured EP is systemic administration of the chemotherapy agent methotrexate (MTX), an inhibitor of folate-dependent steps in nucleic acid synthesis, which effectively destroys the rapidly dividing ectopic trophoblast and thus prevents the placenta from developing and invading adjacent tissues. [ 6 ] The recommended clinical dosing regimen for MTX is a single intramuscular injection of 1 mg kg −1 or 50 mg m −2 . [ 6 – 9 ] Unfortunately, the failure rate of MTX treatment can exceed 10%; [ 7 – 9 ] the risk factors for which are poorly understood but may include high body mass index, rapid clearance, or inaccurate diagnosis. Rapid clearance of MTX most likely leads to insufficient accumulation of the drug at the ectopic implantation site, which ultimately results in therapeutic failure. In this scenario, repeated or higher doses are required, which can result in various side effects ranging from nausea and vomiting to interstitial pneumonitis and bone marrow suppression. [ 10 – 12 ] As MTX remains the first-line treatment for EP, there is a pressing need to improve MTX efficacy. Because systemic MTX treatment is also associated with substantial side effects, [ 7 , 8 ] enhancing its precise delivery at an effective dose to the site of ectopic implantation and, consequently, increasing its safety and efficacy, can greatly improve therapeutic results. A nanomedicine approach can improve the delivery of imaging or therapeutic payloads to targeted tissue, by employing specifically constructed nano-sized carriers, due to their multidrug loading capacity, stability, extended blood circulation, and stealth properties. [ 13 – 18 ] Although the application of nanomedicine to treat pregnancy complications, including ectopic pregnancy, is still in its infancy, [ 19 , 20 ] our and other research teams have demonstrated that specifically designed nano-agents can improve the delivery and retention of the intended cargo in the developing placenta. [ 21 – 23 ] Our lab seeks to develop nanomedicines for EP management based on specifically designed nanoparticles with effective drug delivery and release at the implantation site, aiming at decreasing the necessary dose and adverse effects of MTX while enhancing its therapeutic effect. Herein, we report a nanomedicine strategy to improve EP management, demonstrating proof-of-concept in a pregnant mouse model. We developed glutathione-responsive MTX polymersomes comprised of an amphiphilic polyethylene glycol-disulfide-polycaprolactone (PEG-SS-PCL) copolymer, containing a disulfide bond between PEG and PCL blocks for efficient delivery and triggered intracellular release of the drug cargo in the placenta ( Figure 1A ). To our knowledge, this is the first report employing polymersomes for drug delivery to the placenta/implantation site.