Nano-Theranostic Modality for Visualization of the Placenta and Photo-Hyperthermia for Potential Management of Ectopic Pregnancy.

Nanocarriers have been extensively employed for selective targeting of various tissues and their associated vasculature. [ 33 ] Nanomaterials provide a number of advantages over free small molecule contrast and therapeutic agents, including the potential to transport a variety of cargos, extended blood circulation time, targeted delivery, and reduced systemic toxicity. [ 34 ] Because of these characteristics, several fundamental principles gleaned from the use of nanosystems to detect and treat other diseases and disorders could be applied to reproductive medicine, including complications such as ectopic pregnancy. We have previously shown that such a nanoplatform enables visualization of angiogenesis-dependent disease tissues such as cancer and endometriotic lesions with real-time NIR fluorescence in mice, and effectively generates heat within tissue under NIR light exposure. [ 30 , 32 ] In this study, we use a photostable, hydrophobic NIR dye (SiNc) encapsulated in a biodegradable PEG-PCL polymeric nanoparticle (NIR-NP), that demonstrates prominent NIR fluorescence, photoacoustic, and PHT properties. [ 30 – 32 ] The constructed NIR-NP have a hydrodynamic size of ~ 40 nm ( Figure 1B ), spherical shape ( Figure 1C ), and a nearly neutral surface charge (−1.88 ± 0.90 mV, n=3). NIR-NP generate a strong NIR fluorescence signal in aqueous solution at ~ 780 nm ( Figure 1D ). In addition to the pronounced fluorescence properties, these nanoparticles are characterized by a strong and narrow PA peak around 780 nm ( Figure 1E ), where endogenous contrast agents, such as oxygenated and deoxygenated hemoglobin (Oxy-Hb and Deoxy-Hb), demonstrate the lowest PA signal. Of note, SiNc exhibits many characteristics of an ideal PA contrast agent, including peak absorption in the NIR 650–950 nm window to maximize penetration depth by avoiding the strong absorption by blood and tissues, a high molar extinction coefficient (4.7 × 10 5 M −1 cm −1 in THF) to maximize the amount of absorbed light, high photostability to assure imaging reliability, and a narrow photoacoustic peak for definitive identification via spectral unmixing ( Figure 1E ). [ 35 ] Therefore, NIR-NP have the potential for producing a distinguished PA signature in placenta that is distinct from background ( Figure S1 ). The ability of NIR-NP to generate heat under NIR light was also confirmed in aqueous solution where, after 10 min of 780 nm laser light exposure, the temperature reached 68 °C, as monitored by FLIR NIR camera, whereas no heat was generated by illumination of control saline solution ( Figure 1F , S2 ). In addition, the imaging (fluorescence and PA) and photo-thermal capabilities of NIR-NP are maintained in biological fluids such as blood ( Figure 3 S), even at 10x lower concentration of NIR-NP (0.03 mg mL −1 ). To avoid serious health complications, EP must be identified as soon as possible in the first trimester. Effective detection of the growing placenta would drastically improve the accurate and timely identification of EP. The placenta is an organ that is composed of several cell types and facilitates oxygen delivery, nutrient transport, and waste removal; features which are crucial for fetal development and which occur in the chorionic villi - the basic structural unit of the placenta. [ 36 ] During the initial 12 weeks of pregnancy (first trimester) in humans, the villous cytotrophoblasts differentiate into syncytiotrophoblasts that eventually develop into villous trees ( Figure 2A ). Syncytiotrophoblasts are a suitable tracking target within the growing placenta, as they begin to form immediately before embryo implantation and surround the outside of the blastocyst as it adheres to the decidua (the maternal-fetal interface). Extravillous trophoblasts erupt from the tips of chorionic villi to invade the decidua and anchor the conceptus to the uterus, and may serve as potential targets for visualization of the developing placenta ( Figure 2A ). To determine whether NIR-NP can be effectively internalized by placental cells, we exploited the intrinsic NIR fluorescence and photoacoustic properties of SiNc to visualize the nanoparticles within human placenta-derived cells. Cultured primary human syncytiotrophoblast cells were incubated with NIR-NP (25 µg mL −1 ), and an obvious SiNc-derived fluorescence signal was observed only in the nanoparticle treated cells ( Figure 2B , top) as compared to untreated controls ( Figure 2B , bottom). By overlaying DAPI (blue) and SiNc (red) fluorescence images of the treated cells, it appears that nanoparticles are found only within the cytoplasm and do not penetrate nuclei, thus diminishing the threat of genotoxicity. As shown in Figure 2C , fluorescence imaging of human extravillous trophoblast cells (HTR-8/SVneo) incubated with NIR-NP for 24 h also confirmed the location of nanoparticles within the cytoplasm. Furthermore, PA imaging of HTR-8/SVneo cells treated with NIR-NP demonstrated a high signal-to-noise ratio (SNR) at 780 nm ( Figure 2D , red-colored image) – a region of the NIR spectrum where SiNc exhibits a narrow PA peak ( Figure 2D , 780 nm spectrum inset). For the same sample, no distinguishable SiNc-associated PA signal was detected at other wavelengths outside of the SiNc PA peak, as confirmed for images at 680 and 970 nm ( Figure 2D ), ensuring unambiguous identification at specific wavelengths. In the same experiment, untreated control trophoblast cells did not produce any distinguishable PA signal ( Figure S4 ). Based on these findings in vitro, NIR-NPs demonstrate successful internalization within both human syncytiotrophoblast and extra-villous trophoblast cells after 24 hours as confirmed by fluorescence and PA imaging. This validates that the imaging properties of SiNc are preserved within human placental cells and suggests that NIR-NP can be an effective NIR fluorescence and PA imaging agent for visualization of placental tissue. To determine if NIR-NP-mediated PHT can produce intracellular heat sufficient to induce placental cell death, human extravillous trophoblast cells (HTR-8/SVneo) were incubated with NIR-NP (25 μg mL −1 ) for 24 h, loosely pelleted, and then illuminated with NIR light (780 nm, 15 min). During NIR light exposure, the internal temperature within treated cell pellets increased during 1–2 min from ~37 to ~43–45 °C, which was maintained for 15 min, resulting in approximately 90% cell death ( Figure 2E , PHT). In contrast, cells treated with either NIR light, or NIR-NP, alone, led to only a negligible increase in temperature (<0.5°C) without compromising cell viability ( Figure 2E ). These results confirmed that NIR-NP, internalized within placental cells and subsequently exposed to NIR light, induce cell death when the resulting increase in temperature exceeds 42–43 °C. Previous research revealed that heating tissues above 42 °C can result in cell death via apoptosis (42–46 °C) or necrosis (> 46 °C, thermoablation). [ 37 , 38 ] Of note, the individual components of this photo-hyperthermic modality (NIR-NP and NIR light) are non-toxic when administered separately, and only produce the desired hyperthermic effect when combined. While EP is a common disorder in human pregnancies, it is rarely detected in animals; hence there is no animal model of EP. [ 39 ] The pregnant murine model has been used to study the delivery of imaging and therapeutic agents to the placenta because human and mouse placentas share many anatomical and functional similarities. [ 40 ] In addition, even though normal and ectopic pregnancies differ in the location of their implantation, the processes of implantation and placentation are similar. [ 41 , 42 ] Furthermore, the murine placental labyrinth is functionally comparable to human chorionic villi in its facilitation of nutrient exchange between maternal and fetal circulation ( Figure 3A ). Given these similarities, we used a pregnant mouse model to evaluate the imaging and photothermal capabilities of NIR-NP. While both humans and mice display chorio-allantoic placentation and share commonalities in the early placentation process, the gestation length and the time of formation of main placental structures are very different. In humans, both formation of the main placental structure and establishment of maternal blood flow is completed in the first trimester of pregnancy; placental structures are formed by day 21 (week 3 of gestation), and maternal blood flow is established by day 80 (between weeks 11–12 of gestation of its ~40-week gestational period). [ 41 , 43 , 44 ] However, in mice, these processes are completed between gd10.5 to 12.5, mid-way through its gestation. Hence, we focused our mouse studies on gd13.5 and 17.5 as these gestational points in the mouse placenta are more reflective of the first-trimester placenta in humans. NIR-NP-mediated visualization of the developing placenta was evaluated in the early-mid gestational stages both before (gd8.5–9.5) and after (gd12.5–13.5) vascular blood flow into the placental tissue had been established, and at the late gestational stage (gd16.5–17.5), to determine how remodeling of the placental vasculature and changes in blood flow affect the distribution of NIR-NP in the placenta ( Figure 3 ). When recorded at the same intensity settings, fluorescence images indicated no background fluorescence prior to intravenous ( i.v. ) injection of NIR-NP at each tested gestational day (8.5, 12.5, and 16.5). At 24 h post-injection (gd 9.5, 13.5, and 17.5), following laparotomy and externalization of the pregnant uterine horns, a pronounced SiNc fluorescence signal was detected at each implantation site in the uterus at gd9.5, 13.5 and gd17.5 ( Figure 3B and Figure S5 ). Following separation of fetoplacental units from the uterine wall, SiNc fluorescence signal was detected only in the placenta, whereas signal in the fetuses (gd13.5 and 17.5) remained at the background level ( Figure 3C , D , Figures S5 , separated fetuses and placentas, and Figure S6 , control). These results suggest that NIR-NP effectively accumulates in the mouse placenta after systemic administration within 24 h and does not cross the placenta barrier at 12.5/13.5 and 16.5/17.5 gestational ages, and thus does not pose direct harm to the developing fetus. Further evaluation of NIR-NP biodistribution at 6, 24, 48, and 72 h post-injection revealed peak fluorescence intensity in the liver at 24 h. Fluorescence levels significantly decreased at 48 h and continued to approach background levels at 72 h ( Figure S7 ). In mice administered NIR-NP at gd8.5, prior to the establishment of blood flow into the labyrinth, a strong SiNc fluorescence signal was observed at each implantation site along the entire resected uterus ( Figure 3B , uterus). At this stage in mouse pregnancy, the yolk sac absorbs nutrients from the maternal circulation via capillary channels in the decidual tissue. These channels eventually form sinuses between the primary trophoblastic giant cells of the developing placenta and the Reichert membrane, a specialized basement membrane that wraps around the implanted mouse embryo. [ 43 , 44 ]. We believe that at gd8.5 to 9.5, the injected nanoparticles enter the decidual region via these sinuses and are seen on the surface of the Reichert’s membrane ( Figure S11 ). Lacunae formed in the decidua during human placentation, which later develop into inter-villous spaces, are similar to the sinuses in mouse placentation. Additionally, both in humans and mice, the extravillous trophoblast cells of the placenta are actively replacing the smooth muscle cells of maternal arteries and are thus in contact with maternal blood even before blood supply to the labyrinth (in mice) and inter-villous space (in humans) is established. This may explain the intense fluorescence signal of NIR-NP observed in the decidua at this time, following systemic administration of NIR-NP in the pregnant dam. In addition, we demonstrate that these nanoparticles accumulate within implantation sites of pregnant mice as early as gd7.5 ( Figure S8 ), 24 h post IV injection of NIR-NP, demonstrating visualization of the developing placenta with these nanoparticles before establishment of blood flow. To further examine the distribution of NIR-NP within the mouse fetoplacental unit at different gestational stages, we performed a detailed examination of conceptuses by relating fluorescence imaging to the corresponding H&E stained tissues. Localization of NIR-NP in tissue sections across a complete fetoplacental unit is shown 24 h after a single NIR-NP injection on gd 8.5, 12.5 and 16.5 of mouse pregnancy ( Figure 4 ). In a pregnant mouse, injected on day 8.5, prior to establishment of blood flow into the labyrinth, the nanoparticles are localized in the decidua and the membrane surrounding the amniotic cavity when imaged on gd9.5 ( Figure 4A , D ). By gd12.5, the pregnancy has developed sufficiently to enable distinction of the different structures within the placenta. At gd13.5 ( Figure 4B , E ), the nanoparticles are clearly present in the placental labyrinth. Enlargement of the fluorescence image ( Figure 4G ) shows NIR-NP distributed primarily in the labyrinth and junctional zone ( Figure 5E , G ). Interestingly, following systemic administration of NIR-NP at gd16.5, there appears to be little evidence of the nanoparticle in the placental labyrinth ( Figure 4C , F ) when imaged on gd17.5, and it is instead restricted to the decidua. This discrepancy in localization of NIR-NP at different gestational stages is likely due to extensive remodeling of the placental vasculature between gd11.5 and gd17.5, when several-fold changes in pressure drop and blood flow are observed. [ 45 ] Of note, there is no indication that the nanoparticle manages to reach the fetus at any gestational age examined, suggesting lack of NIR-NP transport across the placenta. As controls, tissue fluorescence images collected from gd9.5, gd13.5 and gd17.5 mice injected with saline did not display any background fluorescence signal associated with SiNc ( Figure S9 ). Together, these data suggest that NIR-NP targets the placenta more efficiently during early mouse pregnancy (gd12.5), when placental blood flow is established (comparable to 10 weeks of human pregnancy), as opposed to later pregnancy (gd16.5). We hypothesized that, following systemic administration, NIR-NP would localize to highly vascular tissues and rapidly dividing cells of the developing placenta. Such a pattern of biodistribution would be similar to that observed in other highly vascularized tissues such as solid tumors and endometriotic lesions. [ 30 , 31 ] The placental labyrinth facilitates the exchange of oxygen and nutrients between the maternal and fetal circulation and, as a result, contains a complex network of blood flow, implying that NIR-NP would localize to the labyrinth. To confirm this, we stained fetoplacental thin sections with antibodies specific to Meca-32 (which is expressed on fetal endothelial cells found in the placental labyrinth, Figure S10 , -Meca-32 vs +Meca-32) and compared the images to fluorescence micrographs of fetoplacental units at the same gestational ages 24 h following systemic administration of NIR-NP. Light micrographs of immunohistochemical staining ( Figure 5A , B , C ) and corresponding fluorescence micrographs ( Figure 5D , E , F ) of fetoplacental units at gd 8.5, 12.5 and 16.5, and imaged 24 h later (at gd 9.5, 13.5 and 17.5), following intravenous injection of NIR-NP, reveal that, prior to the establishment of blood flow in to the placenta (gd10.5 ), NIR-NP localizes to the decidua and the amnion layer surrounding the amniotic cavity. At gd13.5, following the establishment of blood flow, extensive accumulation of NIR-NP is observed in the placental labyrinth, confirming our expectations and observations made in the H&E stains ( Figure 5B , E ). At gd17.5, NIR-NPs are highly accumulated in the decidua. The efficacy of NIR-NP as a PA contrast agent for imaging of placenta and associated implantation sites was then assessed. Using PA imaging, we aimed to confirm the presence of nanoparticles within the placenta, and their absence from the fetus, based on our findings provided by fluorescence imaging in vivo and histological analysis. Conveniently, ultrasound and photoacoustic imaging can be performed with the same transducer co-registering the two corresponding images at significant tissue depth, and can correlate anatomical features on ultrasound with the photoacoustic signal ( Figure 6A ). Thus unmixed PA images, obtained prior to (control) and following i.v. administration of NIR-NP in mice at gd12.5 ( Figure 6 ), show the distinct unmixed SiNc-associated signal within the placenta ( Figure 4C , green) at 24 h post-injection, whereas no noticeable SiNc-signal was detected within the fetus. Of note, control images (prior to NIR-NP injection) do not show SiNc-associated (green) signal ( Figure 6B , D ). When images were unmixed for both Oxy-Hb and SiNc ( Figure 6D , E ), the blood signal is detected as expected in placenta prior to and after NIR-NP administration, whereas SiNc is only observed in the placenta 24 h post-treatment. The PA signal associated with Oxy-Hb (red) is especially strong in placenta, presumably due to the large volume of blood. When performing PAI on the gd13.5 conceptus from a different angle and with the total PA signal at 780 nm (Spectro mode, Figure 6F ), NIR-NP (green ROI) is observed within the placenta, as confrmed wth corresponding spectrum (echoing with Figure 6C , E ), where the labyrinth is located, and thus confirms the findings from fluorescence imaging. Notably, no SiNc-associated signal was detected within the fetus ( Figure 6F and Figure S12 ). So far, PA imaging confirmed accumulation of NIR-NP within placenta with no transplacental migration to the fetus. To further confirm these findings, an ex vivo 3D scan of fetoplacental units (imaged at gd13.5) was performed using unmixed (for Oxy-Hb and SiNc) nanostepper imaging settings with a 0.102nm 3D step size. Recorded either top-to-bottom (Z-plane, Figure 4G ) or left-to-right (x- or y-plane, Figure 6H ), the green SiNc signal is only present in the placental tissue and not in the fetus (refer to Supplementary videos of corresponding 3D scans: PAI 3D top-down and PAI 3D left-right). Thus, this first example of NIR-NP-mediated PA imaging, performed from various angles, provided non-invasive visualization of the placenta. To address concerns associated with the chemotherapeutic (methotrexate) and surgical methods for management of EP following diagnosis, our nanomedicine approach to image-guided therapy utilizes a dual-function agent that enables visualization of the developing placenta and produces localized heat sufficient for termination of EP upon illumination with NIR light. Our visualization data shows that, in mice, NIR-NP localizes to placenta tissue and does not cross the placental barrier to the fetus. This would suggest that, in the case of a non-visualized but suspected EP, the location of the developing fetoplacental unit can be confirmed as either eutopic or ectopic following administration of NIR-NP and subsequent imaging. If the pregnancy was later determined to be viable, then the agent should pose no risk to the developing fetus, as previously demonstrated by the inability of NIR-NP to cross the placental barrier ( Figure 4 ). If, however, fluorescence and/or PA imaging reveals the pregnancy to in fact be ectopic, the combination of photoactivatable nanoparticle and NIR light can be used to eliminate the products of conception without damaging surrounding healthy tissue. As a proof-of-concept, we evaluated the therapeutic potential of localized hyperthermia induced by NIR light within a setting where heat is only produced when both NIR light and photo-responsive nanoparticles are delivered to the same area. Our experimental PHT data shows that when NIR-NP accumulates in the mouse placenta, following i.v. administration, and is subsequently activated using NIR light ( Figure 7A ), temperatures in the range of 42–50 °C can be generated within the developing placenta by varying the power of light used (0.7 – 0.9 W cm −2 ), compared to a negligible increase in temperature when using NIR light alone ( Figure 7A , black). After confirming the generation of heat sufficient for apoptotic or necrotic cellular damage (≥43 °C), [ 37 ] pregnant mice underwent localized PHT 24 h following intravenous administration (at gd11.5) of NIR-NP guided by SiNc fluorescence ( Figure 7B ). A group of fluorescently identified implantation sites was illuminated for 10 min with focused 780 nm light delivered via fiber optic cable ( Figure 7B ). Mice were then allowed to continue gestation for 6 days, and were euthanized on gd17.5 to evaluate the development of PHT-treated vs untreated conceptuses. After opening the peritoneal cavity and exposing the uterus, several underdeveloped conceptuses were identified in the treated region ( Figure 7C , D ). On close inspection of the excised uterus, these conceptuses appeared to correlate with the location of the treated implantation sites ( Figure 7E ). This observation was confirmed by histological analysis of H&E stained cryosections of these PHT-treated fetuses and implantation sites (sites 3–5, Figure 7E ) within a single uterine horn, demonstrating no indication of a normal fetus but rather resorbed fetal tissue ( Figure 7G ). H&E staining revealed that PHT-treated conceptuses contain a defined placenta surrounded by viable maternal uterine tissue but with only fetal remnants, whereas untreated adjacent fetoplacental units (e.g., implantation site #2) from the same uterine horn displayed no structural or developmental abnormalities at gd17.5 ( Figure 7F ). Notably, when the same NIR light (0.9 W cm −2 , 10 min) was applied to the mouse abdominal injected with saline as a control, no fetal resorption was visible at 6-days post-PHT (gd17.5), indicating that the applied NIR light is safe and does not cause hyperthermia in tissues lacking a photoabsorbent molecule in the placenta ( Figure S13 ). The localized effects of PHT treatment were further validated by staining cryosections of PHT-treated conceptuses using a TUNEL assay ( Figure 7H ) which revealed that parts of the placenta and the entirety of partially resorbed fetal remnants contained apoptotic cells, in contrast to the viable perimetrium, myometrium, endometrial glands, and placental spongiotrophoblasts highlighted by hematoxylin counterstain. Infarcted junctional zones of the placenta are also visible with the TUNEL stain, and these features are consistent with those observed in H&E stained cryosections of the same fetoplacental units. Following PHT, the maternal myometrium and endometrial glands and areas of the decidual metrial triangle remain viable, but the embryo and the interface between the decidua and the labyrinth (junctional zone) are infarcted and nonviable, as seen in higher magnification images of H&E stained cryosections of PHT-treated conceptuses ( Figure 8 ). Increased magnification of the junctional zone highlights the contrast between viable maternal uterine tissue (arrows) and placental trophoblasts (arrowheads), and the infarcted junctional zone (*) of the fetal placenta. This is reproducible in multiple separate PHT-treated fetoplacental units within a single pregnant mouse. To assess the effect of nanoparticles and nanoparticle-mediated PHT on dams and pups, we performed safety/toxicity studies by evaluating neonate development, and blood plasma toxicity markers in dams, following photo-hyperthermia treatment (along with saline, NIR-NP, and NIR light control groups) ( Figure 9 ). Dams were allowed to continue gestation following treatment and successfully delivered pups in all 4 studied groups. At post-natal day (pnd) 1, neonates appeared healthy and active, and no pup deaths were recorded during the study. At the end of the study (pnd14), pups in all four groups were euhydrated, active, vocal, and were well-groomed with smooth coats, indicating that there were no obvious signs of fetal toxicity following administration of NIR-NP, NIR light, or PHT. Mouse pup weights fit within the baseline range determined from data provided by Charles River Laboratories for CD1 mice. [ 46 ] Additionally, to evaluate the safety of NIR-NP and PHT treatment, we measured concentrations of surrogate markers of cardiac (creatine kinase (CK)), hepatic (alanine transaminase (ALT) and alkaline phosphatase (ALP)), and renal (blood urea nitrogen (BUN) and creatinine) function in dams, 26 days post-treatment. The obtained values in control and treatment groups were similar, with no statistically significant differences between groups, suggesting that NIR-NP, NIR light, and PHT are safe and well-tolerated ( Figures 9B , C ).

Ectopic pregnancy (EP) is a significant cause of morbidity, mortality, and subsequent reduced fertility. Current limitations in diagnostic imaging can result in false diagnoses, and elimination of extrauterine implantations often employs surgical procedures that can significantly impact subsequent fertility. We theorize that the fluorescence and PA imaging capabilities of biocompatible nanoparticles will enable more accurate and timely diagnosis of EP by facilitating improved visualization of the developing placenta using existing diagnostic technology. Moreover, exploitation of the photo-hyperthermic (PHT) properties of NIR-NP could more effectively and safely eliminate EP, while avoiding the side effects associated with methotrexate treatment and eliminating the need for mechanically destructive surgical procedures. The present work demonstrates the imaging utility and PHT capabilities of NIR-NP in vitro and in a pregnant mouse model. We validate that NIR-NP efficiently accumulates in both primary human syncytiotrophoblasts and an immortalized extravillous trophoblast cell line in culture and is readily detectable using both fluorescence and PA imaging. In addition, illumination of NIR-NP-treated human placenta cell pellets with NIR light elevates their temperature from 43–45°C, resulting in ~90% cell death. Notably, in the pregnant mouse NIR-NP efficiently accumulates in the developing placenta following intravenous delivery at gd8.5, gd12.5, and gd16.5, without traversing the placental barrier to the fetus at either gestational day, as evidenced by fluorescence and PA imaging, and histological analysis. Placental function appeared unaffected by NIR-NP alone, as fetuses continued to develop normally following systemic administration of nanoparticles. Our study also confirmed that heat (43–45 °C) produced under locally applied NIR light, following systemic administration of NIR-NP, is sufficient to selectively impair placental function and induce termination and resorption of pregnancies, as evidenced by the infarcted placental junctional zones and necrotic fetuses observed in PHT-treated implantation sites. Furthermore, pups adjacent to PHT-treated implantation sites continued to develop normally, indicating a significant degree of spatial control over photo-induced hyperthermia. The data presented herein demonstrate the feasibility of alternative management strategies for EP using nanomaterial-based delivery platforms, as well as the application of nanomaterial-based modalities for diagnosis and treatment of other placenta-related complications that pose significant health risks to expectant patients and fetuses. The main purpose of this work was to evaluate our nanoparticle’s ability to identify and visualize the developing placenta, and demonstrate its photothermal capabilities in a pregnant mouse model for potential application of EP management. Although our experimental results are promising, the establishment of a reliable animal model of ectopic pregnancy is needed to further advance the application of this technology.

PEG–PCL (SUNBRIGHT ME-050CL100, poly(ethylene glycol)-block-poly( ε -caprolactone), MW: 5k-10k) and SiNc (silicon 2,3-naphthalocyanine bis (trihexylsilyloxide)) were purchased from NOF American Corporation (White Plains, NY) and Alfa Chemistry (Ronkonkoma NY), respectively. NIR-NPs employed in this study, constructed from PEG-PCL and SiNc, were prepared via a solvent evaporation method ( SI ). [ 31 , 32 ] Characterization of NIR-NP’s properties in solution is provided in Supporting Information ( SI ). Other reagents and supplies were obtained from VWR International, LLC (Radnor, PA), MilliporeSigma (Milwaukee, WI), and Fisher Scientific Inc. (Hampton, NH). Human extravillous trophoblast cells HTR-8/SVneo (a kind gift from Dr. K. Swan, Tulane University Medical School, New Orleans, LA) were seeded in a 6-well plate (50,000 cells per well) and grown in RPMI 1640 (10% fetal bovine serum) for 24 h. To obtain primary human syncytiotrophoblast cells, human placental tissue was collected at term cesarean section in the absence of labor with informed consent under a protocol approved by the OHSU Institutional Review Board. Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board (or Ethics Committee) of Oregon Health and Science University (00016328, approved July, 2021). Cytotrophoblast cells were isolated as we have previously described, [ 47 ] plated, and allowed to aggregate and fuse over 72 h to form syncytiotrophoblast. Both cell types were incubated with NIR-NP (25 µg mL −1 SiNc) for 24 h. Next, cells were washed twice with Dulbecco’s Modified PBS (DPBS), and brightfield (BF) images and fluorescence images were acquired on a fluorescence microscope (Keyence, Osaka, Japan) with DAPI (ex./em. 360/460nm) and Cy7 (ex./em. 710/810 nm) fluorescence filters. Cell nuclei were stained with NucBlue (DAPI) Fixed Cell ReadyProbes Reagent (ThermoFisher Scientific, Waltham, Massachusetts). Human extravillous trophoblast HTR-8/SVneo cells were seeded in T-25 flasks at a density of 0.7×10 6 cells per well. At ~75% confluency, cells were then incubated with NIR-NP at a concentration of 25 µg mL -1 . After 24 h, cells were trypsinised, washed 3X with DPBS, pelleted, and re-suspended in DPBS (50 µL). The cell suspension was then mixed with 50 µL of 2% activated agarose and pipetted into a recessed cavity within a cast 2% agarose gel and allowed to set. Ultrasound and PA images were acquired on Vevo LAZR Photoacoustic Imaging System (LZ550 transducer, FUJIFILM VisualSonics, Inc., Toronto, Canada). A spectral PA scan was performed by acquiring one image every 5 nm within the 680 to 970 nm spectral range. Then, regions of interest (ROIs) placed around imaged cell pellets were evaluated at 680, 780, and 970 nm, and the photoacoustic spectra were generated as average photoacoustic signal intensity against wavelength within the 680–970 nm NIR region. The non-treated cell pellets within a cast 2% agarose gel were used as a control. The HTR-8/SVneo trophoblast cells were cultured and treated in T-75 flasks (3.2×10 6 cells) with NIR-NP (25 µg mL −1 SiNc) for 24 h. Cells were then washed three times with DPBS (to remove nanoparticles that were not internalized), trypsinized, and loosely pelleted. Next, the loose cell pellets were illuminated with 780 nm laser (Wavespectrum Laser Group Limited, China) light (0.9 W cm −2 ) for 15 min. Untreated cells, cells illuminated only with 780 nm light for 15 min, and cells treated only with NIR-NP (25 µg·mL −1 SiNc) were used as controls. [ 31 ] Temperature was recorded using a fiber optic thermal probe (Neoptix, Québec, Canada) as previously reported. [ 48 ] After treatment, cells were then re-suspended in complete growth media and seeded in a 96-well plate (1.0×10 4 cells per well). After 24 h, cell viability was determined using Calcein AM assay (Corning Inc., NY) as previously reported. [ 49 , 50 ] All animal experiments were approved by Institutional Animal Care and Use Committee (IACUC) of Oregon Health & Science University and were carried out in accordance with national and local guidelines and regulations (IP00002578 and IP00003848). Pregnant CD1 wild-type mice at gestational days 8.5, 12.5 and 16.5 were obtained from Charles River Laboratories (Wilmington, MA) and caged in a pathogen-free animal room. Pregnant mice were injected intravenously in the tail vein with NIR-NP (100 μL at 0.3 mg mL −1 SiNc in saline). Live, whole-body images were obtained using a LI-COR Pearl Impulse Imaging System with an 800 nm channel prior to and 24 h after injection. Pregnant mice were euthanized using an institutionally approved protocol. Following euthanasia, an abdominal incision was made to exteriorize the entire uterus for fluorescence imaging. Major organs and uterus were then collected. Fluorescence images were recorded for the exteriorized uterus, the major organs, and the separated fetuses and placentae using a LI-COR Pearl Impulse Imaging System with an 800 nm channel. The uterine wall was then removed from each implantation site, as previously described. [ 51 ] Conceptuses at gd9.5 were left undisturbed in their amniotic sacs and immediately frozen in cryomolds using liquid nitrogen and optimal cutting temperature medium (OCT) for subsequent sectioning, histology, and fluorescence imaging. Other tissues of interest (placentas, fetuses, and residual tissues from resorbed pregnancies following photothermal therapy) were either left undisturbed in their amniotic sacs and immediately frozen in cryomolds, or first separated by gently peeling away the amniotic sac, and then frozen in cryomolds. Where applied, regions of interest (ROI) of the same size were drawn over tissue (for instance, placenta and fetuses), and the average fluorescence signal for each area was determined on the LI-COR Pearl Impulse Imaging System software. Statistical analysis was performed using a one-way analysis of variance (ANOVA) for multiple comparison with GraphPad Prism 8.0 (GraphPad Software Inc. USA). **** p value < 0.0001. Mouse placentas, conceptuses/fetuses, and PHT-treated fetoplacental units were frozen in cryomolds using OCT and liquid nitrogen immediately following necropsy and stored at −80 °C. Thin sections (10 µm) were obtained using a Leica CM 1860 cryostat (Leica Biosystems, Buffalo Grove, IL). Fluorescence images were captured using an EVOS FL Cell Imaging System (Life Technologies, Grand Island, NY) equipped with a Cy7 filter cube (Ex:710/40, Em: 809/81) prior to staining of thin sections with hematoxylin and eosin (H&E). Following H&E staining, brightfield images were captured using the EVOS FL Cell Imaging System. Individual fluorescence and brightfield images were stitched using EVOS FL Cell Imaging System software to provide macroscopic views of tissue thin sections. Cryopreserved tissues were sectioned at 10µm using a Leica CM1860 cryostat (Leica, Wetzlar, Germany) and adhered to Fisherbrand Superfrost Plus microscope slides (Thermo Scientific, Waltham, Massachusetts). For staining, sections were fixed in pre-cooled acetone for 10 min, followed by two washes in 1× PBS. Endogenous peroxidase activity was quenched using 0.3% H 2 O 2 solution in 1×PBS at room temperature for 10 min. The slides were gently washed twice and blocked with blocking buffer (10% fetal bovine serum in 1×PBS) at room temperature (RT) for an hour. After 1 h, the sections were incubated with anti-Meca32 antibody (1:50, BD Biosciences, New Jersey, USA) at 4°C overnight. The following day, the slides were washed twice and incubated with biotinylated Goat Anti-Rat Biotin IgG (1:100, BD Biosciences, New Jersey, USA) for 1 h at RT the slides were gently washed twice, and incubated with pre-diluted HRP-conjugated Streptavidin (BD Biosciences, New Jersey, USA) at RT in the dark for 30 min. The slides were then washed and 200 μL of DAB substrate (BD Biosciences, New Jersey, USA) was applied for 30 seconds (or until the development of color was observed). The slides were then washed gently again and counterstained with Hematoxylin (Gill’s Hematoxylin, Sigma Aldrich) for 3 min. The slides were then gently rinsed under tap water for 5 min, followed by dehydration through 6 changes of alcohol (70%, 80%, 95%, 95%, 100% and 100%), cleared in 2 changes of xylene and cover-slipped using Permount mounting medium (Fischer Scientific, New Hampshire, USA). Images were acquired using a Zeiss Axio Scan.Z1 microscope slide scanner with a Plan-Apochromat 10x/0.45 M27 objective and Hitachi HV-F202SCL camera. The 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, providing real-time acquisition and simultaneous overlay of ultrasound and photoacoustic images on 2D and 3D planes. The LZ550 (center operating frequency of 40 MHz, axial resolution 40 µm) probe was used to acquire images. Animals were anesthetized (inhaled isoflurane at 4% for induction and 1.5–2% during the procedure administered in 100% oxygen, 1–3 L min −1 ), and secured to a heated animal handling platform (37 °C) which allowed for monitoring of heart rate, respiration, and body temperature. Eye lubrication was applied. Fur from the abdomen region of mice was removed with a commercial hair remover, and pre-warmed ultrasonic gel was applied to the skin to guarantee optimal image quality and to provide a coupling interface between the ultrasound probe and the animal. The Spectro sub-mode was used to record PA images across 680–970 nm region. Nanostepper PA sub-mode was then employed to obtain 2D information and generate Unmixed images. The system automatically selects the specific wavelengths, associated with the stored spectra for the selected endogenous or exogenous agent (e.g., Deoxy-Hb, Oxy-Hb, NIR-NP, etc.) to produce spectral unmixed images. Next, 3D scan was performed for NP-treated mice as well as exteriorized uteri ex vivo producing the video or individual frames of unmixed 2D images. The corresponding images were recorded prior to and 24 h post-injection of NIR-NP (100 μL at 0.3 mg mL −1 SiNc in saline). Pregnant mice (gd11.5) were administered NIR-NP (100 μL at 0.3 mg mL −1 SiNc in saline) or saline (control) via tail vein injection. Mice were euthanized 24 h later (gd12.5), and laparotomy was immediately performed to expose the pregnant uteri. Mice were positioned supine on a heated stage to maintain body temperature, and a Neoptix thermal probe (Neoptix, Québec, CA) was inserted in the placenta and held in place during illumination of the placenta with a 780 nm laser, mounted such that the aperture was ~0.5 cm above the placenta. Laser light was shone directly on the placenta at two different powers (0.7 and 0.9 W cm −2 ). Thermal readings were recorded at various time points and graphed using Origin software (OriginLab, USA). Pregnant mice (gd10.5) were administered with NIR-NP (100 μL at 0.3 mg mL −1 SiNc in saline) via tail vein injection. PTT was performed 24 h following administration of NIR-NP. On gd 11.5, pregnant mice were anesthetized using 1.5% isoflurane and positioned on a heated platform to maintain body temperature under anesthesia. A 780 nm laser, collimated to a 0.6 cm beam, was mounted such that the aperture was ~0.5 cm above the exposed, depilated abdomen, and positioned to target implantation sites identified via whole-body fluorescence imaging. Laser light was shone through the intact dermis and abdominal wall at 0.9 W cm −2 for 10 min. The following control groups were used: saline, NIR-NP only and NIR light only. At gd17.5, mice were imaged, euthanized according to the approved institutional protocol, and organs and uteri were collected for ex vivo imaging and subsequent histological analysis. All steps were performed at room temperature using TUNEL Assay Kit - HRP-DAB (Abcam, Cambridge, UK). 10 µm cryosections of mouse feto-placental units were fixed with 4% formaldehyde and permeabilized using Proteinase K, followed by quenching of endogenous peroxidase activity in 30% H 2 O 2 and subsequent equilibration. Sections were then incubated in TdT labeling solution for 1.5 h in a humidified chamber, then rinsed prior to termination of the labeling reaction with stop buffer. Sections were then incubated with blocking buffer prior to the addition of conjugate. DAB solution was applied, and slides were then rinsed in dH 2 O. Sections were immediately counterstained using Gill’s hematoxylin, and rinsed of excess counterstain in two changes of 100% EtOH. Slides were then cleared in 100% xylene, dried, and cover-slipped using Richard-Allen Scientific organic mounting medium (Thermo Fisher Scientific, Waltham, MA). Images were acquired using a Zeiss Axio Scan.Z1 microscope slide scanner with a Plan-Apochromat 10x/0.45 M27 objective and Hitachi HV-F202SCL camera. In a separate study, pregnant mice were allowed to continue gestation following IV administration of saline or NIR-NP (at gd10.5), and NIR light (at gd11.5), and combination of NIR-NP + NIR light (PHT, at gd11.5). Following parturition, pups were observed to evaluate overall appearance, locomotor activity, feeding, and vocalization. Pup weights were collected on post-natal day (pnd) 1, 7, and 14. Datapoints for body weights of Baseline:Females and Baseline:Males growth curves (body weight as a function of time) were bootstrapped for pnd1,7, and 14 for a total of 350 samples with respect to n (sample size), body weight, SD, and growth rate slopes, derived from data published by Charles River Laboratories for CD1 mice. [ 46 ] Error bars for Baseline:Females and Baseline:Males growth curves indicate bootstrapped estimates for SD. Bootstrapping was performed using a linear model with GraphPad Prism v9. Upon completion of the study, pups were euthanized following an institutionally approved protocol. Whole blood was collected from dams prior to euthanasia (26 days following treatment) and submitted to IDEXX Laboratories for Total Health Profile screen to determine plasma levels of cardiac, renal, and hepatic function indicators, including blood urea nitrogen (BUN), creatinine, creatine kinase (CK), alkaline phosphatase (ALP), and alanine transaminase (ALT). Data were analyzed using GraphPad Prism v9 (GraphPad Software, CA, USA) using descriptive statistics, two-tailed unpaired t-test, and presented as mean values ± standard deviation from three to eight independent measurements. To test statistical significance for more than two groups, one-way analysis of variance (ANOVA) was used. Statistical significance is denoted as ****p < 0.0001.

When a pregnancy test is positive, but an intrauterine or ectopic gestation cannot be detected via transvaginal ultrasonography, a condition called ‘pregnancy of unknown location’ (PUL) is identified. [ 1 ] Definitive diagnoses may include ectopic pregnancy, miscarriage, or viable or nonviable intrauterine pregnancy. An ectopic pregnancy (EP) is suspected when serum human chorionic gonadotropin (hCG) levels exceed 3,000 mIU mL −1 with no intrauterine gestational sac detectable by transvaginal ultrasound. [ 2 ] One in fifty pregnancies (~ 120,000 pregnancies) in the USA each year are ectopic with a significant societal and economic cost, while ectopic pregnancies account for 1–2% of all pregnancies worldwide. Bleeding from ectopic pregnancy is still the greatest cause of first-trimester maternal deaths, causing 10% of the 700 pregnancy-related deaths per year and is responsible for 16% of emergency room visits in the first trimester. [ 3 , 4 ] Current management strategies for EP include diagnosis using transvaginal ultrasound, expectant management, medical treatment with methotrexate, and surgical intervention including salpingostomy/salpingectomy. However, these strategies can result in misdiagnosis, and are associated with the risk of tubal rupture, reduced fertility, and increased risk of subsequent EP (patients with EP are 10% more likely to have a second EP). [ 5 ] Using the current clinical approach, the presumed diagnosis of ectopic pregnancy could be erroneous in ~40% of cases, [ 6 ] and subsequent methotrexate treatment can lead to failed intrauterine pregnancy or live-born infants with developmental abnormalities. [ 7 ] Even in the case of accurate diagnosis, methotrexate treatment has a failure rate exceeding 10%, is associated with systemic side effects when successful, and rupture of EP during this medical management ranges from 7–14%. [ 8 ] Methotrexate treatment is also associated with reduced quality of life due to various side effects ranging from mild (nausea, vomiting, diarrhea, elevated liver enzymes, etc.) to severe (nephrotoxicity, interstitial pneumonitis). [ 8 ] As EP can be fatal, prompt and accurate diagnosis, followed by successful treatment, is crucial to identifying and managing EP when ultrasound findings are inconclusive. Advanced strategies ensuring faster and more accurate diagnosis could lead to more effective management of ectopic pregnancies and/or reduce the number of unnecessary treatments. Although experimental use of nanomedicines during pregnancy is still a nascent field of investigation, it has been demonstrated that nanoparticles can accumulate in the placenta or fetus following systemic administration. [ 9 ] Several teams recently reported nanocarriers functionalized with placenta-specific targeting moieties for delivery of therapeutic payloads to the placenta in pregnant mice. [ 10 – 12 ] Using lipid–polymer nanoparticles coated with a peptide that binds Chondroitin sulfate A (CSA) expressed on the surface of trophoblasts within the placenta, Zhang et al. demonstrated delivery of methotrexate specifically to placental trophoblasts in pregnant mice. [ 12 ] In another study, trophoblast-specific delivery of doxorubicin encapsulated within nanoparticles targeted to epidermal growth factor receptor (EGFR) was reported to address abnormal trophoblast growth in a mouse model of choriocarcinoma. [ 10 ] Liposomes modified with the tumor-homing peptide sequences CGKRK and iRGD were also shown to successfully deliver therapeutic cargo (insulin-like growth factor 2) to the murine placenta. [ 11 ] While nanoparticles of various compositions, arrangement, size, charge, and surface modification can be employed for delivery to specific tissues of interest, there are no clear correlations between these characteristics and the ability of nanoparticles to cross the placental barrier to the developing fetus following intravenous administration in pregnant rodents. [ 9 , 13 – 15 ] Our research team aims at developing a novel strategy for detection and management of ectopic pregnancy based on accurately designed nanoparticles that preferentially accumulate in the placenta without the risk of harming the fetus, as they do not readily penetrate the placental barrier. A number of research groups are seeking to develop blood-based biomarkers that could accurately identify EP using a minimally invasive blood test. [ 16 ] However, improvements in real-time imaging technology may be more valuable for EP identification, as the most reliable sign of ectopic pregnancy is the visualization of extrauterine gestation. Photoacoustic imaging (PAI) is an emerging real-time non-invasive imaging technology that combines the spatial resolution of ultrasound imaging with the high contrast of optical imaging, offering improved resolution at greater tissue depth. [ 17 , 18 ] PAI works by detecting sound waves produced by thermal expansion in tissue when photo-absorbers are activated by pulsed laser light. [ 19 ] PAI operates without the use of harmful ionizing radiation and exploits the natural photoacoustic signals of tissue components such as melanin, blood, or collagen (endogenous contrasts). [ 20 , 21 ] Employment of additional (exogenous) photoacoustic absorbers in the near-infrared (NIR) optical imaging window (~ 650–950 nm), where the predominant endogenous signals from water and blood are at their minima, offers to produce a unique PA signal at the site of interest. Exogenous contrast agents with a narrow absorbance peak have the ability to produce PA signals clearly distinguishable from the background (blood, water, etc.) and thus allow users to generate unmixed images using pre-recorded PAI spectra. By synergistically combining light and sound, PAI provides deeper tissue penetration than other optical modalities such as fluorescence imaging. [ 22 , 23 ] While in vivo fluorescence imaging offers favorable contrast and resolution at superficial tissue depths (e.g., subcutaneous, intradermal, etc.), the effect of optical scattering within biological tissues results in rapidly degraded spatial resolution as tissue depth increases (>1–2mm). [ 23 ] Because PAI relies on the propagation and detection of acoustic waves, this imaging modality can significantly improve spatial resolution in comparison to pure optical imaging as acoustic wave scattering in biological tissues is considerably less than that of optical scattering at increased depth. [ 22 , 24 , 25 ] However, effective exogenous contrast agents for accurate PAI diagnosis are required. Fluorescence imaging can be complementary to PAI, when employing fluorophores in the NIR region (~700–1800 nm) [ 26 ] with minimal autofluorescence and absorption/scattering of NIR light, providing improved tissue penetration depth and reduced autofluorescence background. [ 26 ] The photo-responsive agents used for PA and fluorescence imaging can also generate heat upon exposure to NIR light, allowing for localized hyperthermia. To date, hyperthermia, as an alternative or adjunct to conventional chemotherapy, has been employed as a clinical procedure to destroy cancer tissues, and can enable the circumvention of chemo-resistance. [ 27 ] Nanomaterials afford the opportunity of delivering hyperthermia agents to a tissue of interest in the body and subsequent activation of the cargo with external stimuli such as a magnetic field, ultrasound, light, etc. [ 28 ] Photo-hyperthermia (PHT) uses non-harmful light to convert photo-active small molecules into localized heaters when the nanoparticles carrying them reach specific targeted tissue. [ 29 ] The above-outlined imaging and therapeutic approaches have not been applied to diagnose and treat EP. Herein, we report the first theranostic approach for potential diagnosis and treatment of EP in a pregnant mouse model. It is based on a nano-agent composed of a single NIR photo-responsive molecule, silicon naphthalocyanine (SiNc), encapsulated within a biocompatible polymeric PEG-PCL (poly(ethylene glycol)-block-poly(ε-caprolactone) methyl ether) nanoparticle ( Figure 1A ) that selectively accumulates in highly vascularized tissues, including the placenta. [ 30 – 32 ] Following systemic administration, the constructed nanoparticles specifically deliver SiNc to the highly vascularized placental tissue, but not to the fetus. As a result, the developing placenta can be visualized using fluorescence and photoacoustic imaging ( Figure 1 ). Following accumulation in the placenta, this photoresponsive nanoparticle is also capable of generating heat under exposure to NIR light. This enables the application of selective, localized photo-hyperthermia (PHT), causing disruption of placental function and enabling pregnancy demise. To our knowledge, this is the first report of an exogenous nanoparticle-based contrast agent for photoacoustic imaging of the placenta and the first application of nanoparticle-mediated PHT for the potential management of ectopic pregnancy. Because the agent does not cross the placental barrier, and is non-toxic according to plasma markers evaluated in this work, the PHT phase may be omitted, and gestation may proceed normally if the decision is made in a clinical setting to prolong a pregnancy that is later determined to be a viable intrauterine implantation.