Placenta-derived trophoblast extracellular vesicles promote CD169⁺ macrophage migration and suppress ovarian tumour growth

Abstract

Background

A lower survival rate due to a later stage of diagnosis requires a more effective therapy for ovarian cancer. Extracellular vesicles (EVs) have shown therapeutic potential in various diseases, partly due to their ability to modulate immune responses. We previously showed that placental-derived trophoblast EVs suppressed ovarian tumour growth, accompanied by increased infiltration of CD169⁺ macrophages and NK cells in tumours. However, the underlying mechanism is unclear. Here, we investigated how immune cells migrate to tumors in response to placental-derived trophoblast EVs in vivo.

Method

SKOV-3 ovarian cancer cell xenografts were intraperitoneally injected with placental-derived trophoblast EVs. Tumour growth was monitored, and EV biodistribution was tracked at four time points. Additionally, the dynamics of CD169⁺ macrophages and NK cells were analysed in immune organs and tumour tissues.

Results

A reduction of tumour growth after EV treatment was observed. At early time points (within 24 hours), trophoblast EVs preferentially accumulated in immune organs, including the inguinal lymph nodes, but not in tumours. The significantly higher intensity of CD169⁺ macrophages was observed in the inguinal lymph nodes within 24 hours, compared to controls. However, by day 30, the infiltration of EVs and CD169⁺ macrophages was observed in tumours, with clear colocalization. Additionally, no infiltration of NK cells was observed at early time points, but was observed at day 30 in the inguinal lymph nodes, with the increase in IL-15, a known NK cell activator. Importantly, in non-tumour-bearing mice following EV treatment, the infiltration of CD169⁺ macrophages in the inguinal lymph nodes was not observed, compared to controls.

Conclusion

Placental-derived trophoblast EVs may act as an immune modulator, specifically in the presence of tumours, through promoting CD169⁺ macrophage recruitment to the tumour site, resulting in an anti-ovarian tumour response. These findings suggest the potential of trophoblast EVs as a novel immune-based therapeutic strategy.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-025-07320-6.

Background

Ovarian cancer is one of the most lethal gynaecological malignancies, with an increasing incidence, especially in young women [1], and only a 45% five-year survival rate [2]. The poor prognosis of ovarian cancer is primarily due to challenges in early diagnosis, limited effective therapeutic options, and a high recurrence rate [3]. Additionally, aberrant metabolism is a hallmark of cancer, including ovarian cancer. These factors underscore an urgent need to explore innovative therapeutic approaches, including strategies targeting cancer metabolism and hormone receptors [4, 5] as well as nanoparticles derived from natural sources such as plants [6, 7], to improve patient outcomes.

There are notable parallels between the human placenta and tumours [8]. During human pregnancy, trophoblasts, the epithelial cells unique to the placenta, invade deeply into the uterus and transform the uterine vasculature to ensure an adequate blood supply to the placenta. However, unlike tumour invasion, trophoblast invasion is tightly regulated during pregnancy [9]. Although the mechanisms underlying the controlled proliferation and invasion of trophoblasts during placental development remain largely unknown, the biological similarities between placental development and cancer have inspired interest in exploring placental-derived therapies for cancer treatment.

Several studies suggest that placental derivatives possess anti-tumour potential. For example, a placental lysate inhibited metastasis, reduced tumour growth in a Lewis lung carcinoma model [10], and suppressed melanogenesis in B16 melanoma cells [11]. Additionally, conditioned medium from human placenta-derived mesenchymal stem cells inhibited the growth of several cancer cell types [12, 13]. These findings highlight the therapeutic promise of placental-derived factors in cancer.

Extracellular vesicles (EVs), released from all cells, are lipid-enclosed packages of cellular contents that serve as complex interorgan communication systems. They are generally categorized by large EVs (100–1000 nm) and small EVs ( < 100 nm) [14]. These EVs are enriched with functional proteins, regulatory RNAs, DNA, and lipids. During pregnancy, a large number of EVs are released from the placenta to facilitate fetomaternal communication [15]. Trophoblast EVs carry anti-tumour cargos, particularly microRNAs (miRNAs) from the chromosome 19 microRNA cluster (C19MC) that are unique to the human placenta [16], which exhibit potent anti-tumour properties [17].

Our previous research demonstrated that trophoblast EVs inhibited ovarian cancer cell growth by delaying cell cycle progression in vitro [18], and a single intraperitoneal (i.p.) injection of trophoblast EVs significantly suppressed the growth of human SKOV-3 ovarian tumour xenografts [19]. The reduced tumour growth was associated with the infiltration of CD169⁺ macrophages and natural killer (NK) cells into the tumours. CD169 is predominantly expressed by metallophilic macrophages in the lymph nodes and the marginal zone of the spleen [20]. They enhance anti-tumour immunity by presenting tumour antigens to adaptive immune cells, and increased levels of CD169⁺ macrophages are associated with favourable outcomes in cancers [21–24].

The therapeutic potential of EVs has been given greater attention. A key therapeutic advantage of EVs compared to synthetic nanoparticles is their expression of “addressins” on the vesicle surface, such as integrins [25], which facilitates targeted delivery to specific cells and tissues [26]. EVs can travel through the cardiovascular and lymphatic systems, as well as extracorporeal and interstitial fluids, enabling the targeted delivery of their therapeutic cargo to specific organs or tumours [27]. Therefore, in this study, we sought to investigate the potential underlying mechanisms by which placental-derived trophoblast EVs contribute to the suppression of ovarian cancer xenograft growth. We hypothesized that these EVs facilitate the recruitment of CD169+ macrophages and NK cells into the tumor microenvironment. To address this, we conducted a biodistribution time-course study in which large SKOV-3 ovarian cancer cell xenografts, representing advanced disease, were grown in nude mice. Additionally, we analyzed the tumour-associated immune cell response. This study may provide the potential of placental-derived trophoblast EVs as a novel immune-based therapeutic strategy for ovarian cancer.

Materials and methods

Preparation of first-trimester trophoblast EVs

First-trimester placentae (n = 12) were collected from the Epsom Day Unit of Green Lane Hospital (Auckland, New Zealand), with informed consent and approval from the Regional Health and Disabilities Ethics Committee (Approval No. NTX/12/06/057/AM08). Trophoblast EVs were harvested as previously described [28]. Briefly, placental explants were cultured for 16 hours, and conditioned medium was collected and centrifuged sequentially for large and small EVs, and stored at 4 °C [29]. In some experiments, CellTracker^TM Red CMTPX dye (1 µg/mL, Thermo Fisher Scientific, New Zealand) was added to the cultures. EVs collected from culture media are referred to as vehicle controls.

Cell culture

The human epithelial ovarian adenocarcinoma cell line SKOV-3 was obtained by Dr. Wouter van Leeuwen (Auckland Cancer Society Research Centre, University of Auckland, New Zealand). Cells were cultured in DMEM high glucose medium supplemented with 10% FBS and 1% P/S. Cultures were maintained at 37 °C in a humidified incubator with 5% CO₂.

Mouse xenograft model and EV treatment

SKOV-3 ovarian cancer cell xenografts were employed, as it is the most common model for cancer investigation [30–33], and this was a follow-up of our previous study [19]. All animal breeding and experimental procedures were conducted at the Vernon-Jansen Unit of the University of Auckland, and were performed in accordance with the Animal Ethics Committee of the University of Auckland (Approval No. AEC2257). Female CD1 nude mice (6–8 weeks), obtained from the Vernon-Jansen Unit at the University of Auckland (New Zealand), were housed under controlled environmental conditions, including a constant temperature of 20 °C, regulated humidity, and a 12-hour light/dark cycle.

For subcutaneous (s.c.) tumour implantation, 1 × 10⁶ SKOV-3 cells were suspended in 100 μL of ice-cold DPBS and subcutaneously inoculated into the left flank of CD1 nude mice. Tumour-bearing mice were monitored daily, and tumour volumes were calculated using the formula: a²⋅b × 0.52, where ‘a’ and ‘b’ represent the minor and major tumour axes, respectively. When tumour volume reached ~65 mm³, animals were assigned to one of the three treatment or the control groups as follows: 1) mice with tumour xenografts treated with trophoblast EVs (n = 27), 2) control mice with tumour xenografts treated with vehicle control (n = 12), 3) mice without tumour xenografts treated with trophoblast EVs (n = 9), and 4) control mice without xenografts treated with vehicle control (n = 3) (Details in Supplementary table 1).

A total of 100 μg of trophoblast EVs (based on protein content) was diluted in 100 μL of DPBS and administered intraperitoneally (i.p.) to both SKOV-3 tumour-bearing mice and non-tumour-bearing mice. The trophoblast EV group consisted of three subgroups: trophoblast large EVs (n = 3), trophoblast small EVs (n = 3), and a mixture of trophoblast large and small EVs (n = 3). At predetermined time points (2 hours, 24 hours, 10 days, and 30 days) post-EV administration, animals were euthanized using CO₂ inhalation followed by cervical dislocation. Tissues were harvested, including tumours, spleen, liver, inguinal lymph node, hepatic lymph node, renal lymph node, kidney, and skeletal muscle (right forelimb). Part of the tissues were embedded in the optimal cutting temperature (OCT) compound, and the rest were preserved in RNA later and stored at −80 °C.

Distribution of trophoblast EVs

Tissue specimens were harvested, and 5-μm-thick cryosections were placed on microscope slides. Sections were washed in PBS for 10 minutes to remove residual OCT compound. After washing, sections were mounted using Fluoroshield Mounting Medium with DAPI (Abcam, New Zealand) to stain nuclei. The localisation of fluorescently labelled EVs at the cellular level was examined using an Eclipse Ni-E fluorescence microscope (Nikon, Japan). Images were acquired using a DS-Ri1 camera and processed using NIS-elements F software (Version 4.0) and ImageJ software.

miRNA levels in mouse tissues

Tissues (tumours, spleens, muscles) were powdered using liquid nitrogen to facilitate RNA extraction. The pestle was placed in the mortar, and liquid nitrogen was added to pre-cool both tools. Tissue samples were transferred into the pre-cooled mortar and crushed into a fine frozen powder. The powdered tissues were transferred to DNase/RNase-free Eppendorf tubes, and total RNA was extracted using the miRNeasy Mini Kit (Qiagen, New Zealand), following the manufacturer’s protocol. miRNA-specific primers for hsa-miR-512-3p and hsa-miR-519a-5p were purchased from Ribobio Co. Ltd (Guangzhou, China).

miRNAs were reverse transcribed using the RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, New Zealand) and quantified by SYBR^TM Select Master Mix (Thermo Fisher Scientific, New Zealand) according to the manufacturer’s instructions. Quantitative real-time PCR (qRT-PCR) was performed using the QuantStudio^TM 12K Flex Real-Time PCR System (Thermo Fisher Scientific, New Zealand). U6 small nuclear RNA (snRNA) was used as an internal normalization control. All experiments were conducted in triplicate on three independent occasions.

Haematoxylin-eosin (H&E) and immunofluorescence staining

Immunofluorescence staining of tumours and secondary lymphoid organs was performed on 5-μm- cryosections, according to standard protocols. Tissue sections were air-dried for 1 hour, and the embedding medium was removed by incubating slides in PBS for 10 minutes at room temperature. Sections were then fixed with 4% paraformaldehyde for 15 minutes, rinsed with PBS, and incubated in blocking reagent [10% normal horse serum (Gibco, New Zealand) in 0.5% PBST] at room temperature (RT) for 1 hour to block nonspecific binding. Primary antibodies were applied and incubated overnight at 4 °C, including monoclonal anti-CD169 (Bio-Rad, MCA947GA, 1:200), monoclonal anti-F4/80 (BioLegend, BIO123102, 1:200), monoclonal anti-IL-15 (PH Scientific, DF7654, 1:200), monoclonal anti-NKp46 (Abcam, ab89877, 1:200), and monoclonal anti-G-CSF (Abcam, ab181053, 1:200). Following primary antibody incubation, sections were washed three times with PBS and incubated with Alexa Fluor 488-conjugated anti-rat IgG (Jackson ImmunoResearch Laboratories, AB_2340683, 1:200) in blocking buffer for 2 hours at RT. After incubation, sections were washed three times with PBST for 5 minutes each, mounted using a DAPI-containing mounting medium (Abcam, New Zealand), and examined using an Eclipse Ni-E fluorescence compound microscope (Nikon, Japan). Images were acquired using a DS-Ri1 camera, processed with NIS-elements F software (Version 4.0), and analysed using ImageJ software. Additional tissue sections were stained with haematoxylin and eosin (Sigma-Aldrich, New Zealand) using standard protocols.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 9.3.1 (GraphPad Software, USA). The Shapiro-Wilk test was used to assess data normality. Unpaired t-tests were performed for normally distributed data. The Mann-Whitney test was applied for non-normally distributed data. Comparisons between multiple groups were analysed using two-way ANOVA, followed by Sidak’s post hoc multiple comparisons test. A p-value of < 0.05 was considered statistically significant. Data in bar graphs are presented as mean ± standard error of the mean (SEM).

Results

Trophoblast EVs slowed ovarian tumour growth

By day 30 post-injection, there was a significant reduction in mean tumour volume in the animals treated with trophoblast EVs compared to vehicle controls (Fig. 1, p = 0.0027). Tumour growth was significantly inhibited regardless of the subtype of trophoblast EVs (large EVs: p = 0.0342; small EVs: p = 0.0181; mixed EVs: p = 0.0374), with no significant differences observed between the EV subtypes (Supplementary Figure 1). H&E staining of tumours showed chronic inflammation and signs of necrosis in all groups, correlating with tumour size rather than EV treatment (Supplementary Figure 2).

Fig. 1.

Effects of a single intraperitoneal injection of trophoblast EVs on ovarian tumour growth in nude mice. Tumour volumes in mice treated with either trophoblast EVs (purple, n = 9) or vehicle controls (black, n = 3). Statistical analysis was performed using two-way ANOVA with post hoc Sidak’s test. Data are presented as mean ± SEM

Trophoblast EVs biodistribution in ovarian tumour-bearing mice

The distribution of trophoblast EVs revealed that trophoblast EVs were primarily localised in the spleen, liver, inguinal and hepatic lymph nodes at 2 hours and 24 hours (Fig. 2). No changes in the intensity of EVs in liver, inguinal, and hepatic lymph nodes by day 30. However, the intensity of trophoblast EVs was significantly reduced in the spleen by day 30 post-injection (p = 0.0008). Additionally, at 2 and 24 hours post-injection, no trophoblast EVs were observed in the tumours. However, by day 10, trophoblast EVs were observed at the tumour periphery, and by day 30, EVs had deeply infiltrated the tumours (Fig. 2b). In contrast, trophoblast EVs were not detected in the kidney or skeletal muscle (right forelimb), and red fluorescence in renal lymph nodes did not differ between vehicle control and trophoblast EV-treated groups at any time point (Supplementary Figure 3).

Fig. 2.

Biodistribution of trophoblast EVs in SKOV-3 tumour-bearing mice. (a) semi-quantitative fluorescence intensity analysis of CMTPX-labelled trophoblast EVs in the tumours, livers, hepatic lymph nodes, inguinal lymph nodes, and spleens. Fluorescence intensities were normalized to background fluorescence levels in organs from vehicle controls. Data are presented as mean ± SEM. (b) Representative fluorescence images showing the distribution of trophoblast EVs in various organs. Scale bar = 100 μm

The localisation of CD169⁺ macrophages and F4/80⁺ macrophages

Our previous study reported the infiltration of CD169⁺ macrophages into tumours by day 58 post-EV injection. We first investigated the trends of CD169⁺ and F4/80⁺ macrophage infiltration in different immune organs. In the spleen, liver, and hepatic lymph nodes, both CD169⁺ and F4/80⁺ macrophages were present throughout the time points, regardless of the treatment, but no changes in abundance were observed at any time point, compared to vehicle controls (data not shown).

In the inguinal lymph nodes, the intensity of CD169⁺ macrophages was significantly increased after 24 hours until day 30 post-injection of the EVs, compared to the vehicle controls (Fig. 3a–f). The intensity of F4/80⁺ macrophages in the inguinal lymph nodes was significantly increased on day 30 post-EV injection, compared to the vehicle controls (Fig. 3g and h).

Fig. 3.

CD169 and F4/80 macrophages in the inguinal lymph nodes of tumour-bearing mice following treatment with trophoblast EVs or vehicle control. (a,c,e) Representative images of immunofluorescent detection of CD169 in inguinal lymph nodes collected at 24 hours (a), day 10 (c), and day 30 (e) post-treatment with trophoblast EVs or vehicle controls. DAPI (blue), CD169-positive cells (green). Scale bar = 100 μm. (b,d,f) semi-quantitative analysis of CD169 levels in inguinal lymph nodes collected at 24 h (b), day 10 (d), and day 30 (f) post-treatment. (g) Representative images of F4/80 in inguinal lymph nodes at day 30 post-treatment. DAPI (blue), F4/80-positive cells (green). Scale bar = 100 μm. (h) semi-quantitative analysis of F4/80 levels in inguinal lymph nodes collected at day 30 post-treatment. Data are presented as mean ± SEM

In tumours, CD169⁺ macrophages were notably present, primarily at the edge of the tumours, at day 10 post EV injection, but absent from the tumours in the vehicle controls (Fig. 4a and b). By day 30 post-EV injection, CD169⁺ macrophages had deeply infiltrated the tumour tissues, with a significant increase in intensity compared to the vehicle controls (Fig. 4c and d). Small numbers of F4/80⁺ macrophages were present in the tumours at all time points (Supplementary Figure 4), with a significant increase in the intensity of F4/80⁺ macrophages in the tumours on day 30 post-EV injection (Fig. 4e and f).

Fig. 4.

CD169 and F4/80 levels in murine tumours following treatment with trophoblast EVs or vehicle control. (a,c) Representative images of CD169 in tumours at day 10 (a) and day 30 (c) post-treatment with trophoblast EVs or vehicle controls (scale bar = 200 μm). Regions in the white boxes are enlarged in the top left corner (scale bar = 100 μm). DAPI (blue), CD169⁺ cells (green), CMTPX (EVs, red). (b,d) semi-quantitative analysis of CD169 levels in tumours collected at day 10 (b) or day 30 (d) post-treatment. (e) Representative images of F4/80 in tumours at day 30 post-treatment (scale bar = 200 μm). Zoomed-in regions are shown in the corner (scale bar = 100 μm). DAPI (blue), F4/80⁺ cells (green), CMTPX (EVs, red). (f) semi-quantification of F4/80 levels in tumours at day 30 post-treatment. Data are presented as mean ± SEM

Colocalization of CD169⁺ and F4/80⁺ macrophages and trophoblast EVs

Double immunofluorescence staining was performed to understand the interaction between trophoblast EVs and CD169⁺ or F4/80⁺ macrophages. In the spleen, EVs were primarily colocalised with CD169⁺ macrophages in the splenic marginal zone at 2 hours and 24 hours post-EV injection. By days 10 and 30 post-injection, a greater number of trophoblast EVs were also observed in the splenic white pulp. However, there was no interaction between F4/80⁺ macrophages and trophoblast EVs in the spleen at any time point (Fig. 5).

Fig. 5.

Cellular localisation of trophoblast EVs in the spleen of tumour-bearing mice. Colocalisation of trophoblast EVs with CD169⁺, but not F4/80⁺ macrophages in the spleen at 2 hours, 24 hours, 10 days, and 30 days post-injection. Scale bar = 100 μm. Regions in the white boxes are enlarged in the bottom-left corner. DAPI (blue), CD169 or F4/80 positive cells (green), CMTPX-labelled trophoblast EVs (red)

In the inguinal lymph nodes, immunofluorescence analysis showed distinct macrophage interaction preferences, with trophoblast EVs primarily colocalizing with F4/80⁺ macrophages, with minimal association with CD169⁺ macrophages at all time points (Fig. 6). However, in the hepatic lymph nodes, trophoblast EVs did not show preferential association with any specific macrophage subtype.

Fig. 6.

Localisation of trophoblast EVs in the inguinal lymph nodes of ovarian tumour-bearing mice. Colocalisation of trophoblast EVs with F4/80⁺, but not CD169⁺ macrophages in the inguinal lymph nodes at 2 hours, 24 hours, 10 days, and 30 days post-injection. Scale bar = 100 μm. Regions in the white boxes are enlarged in the bottom-left corner. DAPI (blue), CD169 or F4/80 positive cells (green), CMTPX-labelled trophoblast EVs (red)

In the tumour tissues, by day 10 of post-injection, a noticeable and localized increase in CD169⁺ macrophage intensity, specifically at the periphery of the tumour, where they colocalized with trophoblast EVs, was observed (Fig. 4a). By day 30, CD169⁺ macrophages containing trophoblast EVs had infiltrated deeper and were located throughout tumours in the placental trophoblast EV-treated group (Fig. 4c). However, there was no interaction between F4/80⁺ macrophages and trophoblast EVs in the tumour tissues before day 30. On day 30, F4/80⁺ macrophages were found throughout tumours, with some colocalizing with trophoblast EVs (Fig. 4e).

IL-15 and NK cells in murine tissues after administration of trophoblast EVs

IL-15 is a key cytokine regulating NK cell activation and survival that enhances innate immune defense against tumours [34]. Immunofluorescent analysis indicated that the expression of IL-15 and NKp46 was not present in tumours from mice treated with trophoblast EVs or vehicle controls at any time point (Supplementary Fig. 5).

Additionally, no differences in splenic IL-15 levels were observed between placental EV and vehicle control-treated mice at any time point. However, a significant increase in IL-15 was observed in the inguinal lymph nodes of tumour-bearing mice at day 30 following trophoblast EV administration, but not at earlier time points (Fig. 7b, p = 0.0149). Furthermore, IL-15 selectively colocalized with trophoblast EVs in the inguinal lymph nodes (Fig. 7a). There were also significant increases in NKp46⁺ NK cells in the inguinal lymph nodes of placental EV-treated mice at day 30 post-EV administration (Fig. 7d, p = 0.0052), but not at earlier time points.

Fig. 7.

IL-15 and NKp46 levels in inguinal lymph nodes collected from tumour-bearing mice following treatment with trophoblast EVs or vehicle controls. (a) Representative images of IL-15 in the inguinal lymph nodes collected at day 30 post-treatment with trophoblast EVs or vehicle controls. Left scale bar = 200 μm. The image on the right is an enlarged view of the area in the red box on the left (scale bar = 100 μm). DAPI (blue), IL-15 (green), CMTPX (EVs, red). (b) semi-quantification of IL-15 levels in inguinal lymph nodes collected at day 30 post-treatment with trophoblast EVs or vehicle controls. Data are presented as mean ± SEM. (c) Representative images of NKp46⁺ natural killer cells in the inguinal lymph nodes collected at day 30 post-treatment with trophoblast EVs or vehicle controls. Left scale bar = 200 μm. The image on the right is an enlarged view of the area in the red box on the left (scale bar = 100 μm). DAPI (blue), NKp46⁺ NK cells (green). (b) semi-quantification of NKp46 levels in inguinal lymph nodes collected at day 30 post-treatment with trophoblast EVs or vehicle controls. Data are presented as mean ± SEM

NK cells, CD169⁺ and F4/80⁺ macrophages, as well as IL-15 in non-tumour-bearing mice treated with trophoblast EVs

As described above, CD169⁺ and F4/80⁺ macrophages, IL-15 and NKp46⁺ NK cells were significantly increased in the inguinal lymph nodes in tumour-bearing mice on day 30 following trophoblast EV injection. However, these increases were not observed in mice without SKOV-3 tumours that were injected with trophoblast EVs (Supplementary Figure 6).

Primate placenta-specific C19MC cluster miRnas were present in mouse tissues

To confirm EV cargo delivery, two placenta-specific miRNAs from the C19MC cluster (hsa-miR-519a-5p and hsa-miR-512-3p) were quantified in the spleens and tumours of mice following injection of trophoblast EVs. Hsa-miR-519a-5p was detected in the spleens 24 hours post-injection of placental EVs (Fig. 8b, p = 0.0013). In contrast, this miRNA was detected in tumours only at day 30 (Fig. 8h, p = 0.0489). Neither miRNA was detected in muscle tissues (negative control) at 2 hours post-EV injection.

Fig. 8.

Levels of hsa-miR-519a-5p and hsa-miR-512-3p in murine tissues at 2 h, 24 h, 10 days, and 30 days post EV injection. (a-d) hsa-miR-519a-5p levels in spleens collected from mice treated with trophoblast EVs or vehicle control at four time points. (e-h) hsa-miR-519a-5p levels in tumours collected from mice treated with trophoblast EVs or vehicle controls at four time points. (i-l) hsa-miR-512-3p levels in spleens collected from mice treated with trophoblast EVs or vehicle controls at four time points. (m-p) hsa-miR-512-3p in tumours collected from mice treated with trophoblast EVs or vehicle control at four time points. Data are presented as mean ± SEM. Experiments were performed in triplicate and repeated on three independent occasions

To further examine these findings, we analysed the large and small EV-injected subgroups separately. This analysis revealed that hsa-miR-512-3p was detectable in spleens as early as 2 hours post-injection of trophoblast large EVs (p = 0.0239, n = 3). In tumours, hsa-miR-512-3p was detected only at day 30 post-injection of trophoblast small EVs (p = 0.0217, n = 3) (Supplementary Figure 7).

Discusion

Placental-derived trophoblast EVs significantly slowed ovarian tumour growth, regardless of whether large, small, or a mixture of large and small EVs were administered. A selective biodistribution of trophoblast EVs, with EVs present in the immune organs starting from an early time point, and in the tumor at a later time point. Additionally, infiltration of CD169⁺ macrophages was observed in the immune organs at an early time point, but in the tumour at a later time point.

Consistent with our previous study [19], the slowed ovarian tumour growth seen in this study indicates that a component, or more likely components of the cargoes carried by trophoblast EVs that are shared between large and small EVs, are responsible for this inhibitive effect. Unlike our prior study, which was conducted with smaller tumours and treated for a longer treatment period, the signs of necrosis in tumours did not significantly increase following placental EV treatment. Instead, the signs of necrosis correlated with tumour size, with larger tumours showing more necrosis. This discrepancy may result from the shorter duration of this study (30 days) compared to 58 days in previous work, or more likely, the larger size of the tumours studied here.

Our findings revealed a selective biodistribution of trophoblast EVs in ovarian tumour-bearing mice after a single i.p. administration of the EVs. At early time points (2 hours and 24 hours post-injection), the EVs were localised in the spleen, liver, hepatic and inguinal lymph nodes, with similar biodistribution between placental large and small EVs. Trophoblast EVs were undetectable in the kidneys, renal lymph nodes, and negative control skeletal muscle (right forelimb). Surprisingly, the EVs were also absent from the tumours at early timepoints. A recent systematic review indicated that, in most published studies, following the administration of EVs into tumour-bearing mice, the EVs rapidly localised to the tumours (and other sites) [35]. However, in most of the studies in that review, the EVs were administered intravenously (i.v.) or directly into the tumours. One publication compared i.p. versus i.v. administration of EVs into tumour-bearing mice and showed that while i.v. administration resulted in rapid accumulation of EVs in the tumours, in agreement with our study, i.p. administration did not induce accumulation of EVs in tumours [36]. Another major difference between our study and several studies examining the biodistribution of EVs after administration to tumour-bearing mice was that many studies used EVs derived from the same cells as the tumours [37–39]. In this study, however, trophoblast EVs were unrelated to the SKOV-3 tumours. Unlike tumour-derived EVs, which may carry tumourigenic signals, trophoblast EVs come from a non-cancerous source, reducing the risk of supporting tumour growth or metastasis. Our finding that trophoblast EVs reduced ovarian tumour growth without being related to the tumours makes them an attractive option as a potential therapeutic. Additionally, the distribution of EVs depends on both their cargo content and size [14]. For example, a recent study indicated that large EVs exhibit a delayed clearance, shifting more slowly from lung reticuloendothelial system metabolism to hepatic metabolism. This delay may reflect uptake of large EVs via pathways other than clathrin-coated pits, consistent with size-dependent differences in EV internalization [40].

The liver is a major site of rapid EV localisation in EV biodistribution studies. Localisation to the liver most likely reflects the liver being the major organ that processes and clears circulating EVs, especially when given at supraphysiologic concentrations, as in biodistribution studies [35]. However, since the liver is an organ that generates immune tolerance, it is possible that EVs in the liver could modulate immune responses [41]. Since our previous study demonstrated that trophoblast EVs induce migration of CD169⁺ macrophages to the tumours, we have concentrated our analysis on the two major sites where CD169⁺ macrophages are typically located, the lymph nodes and the spleen. We found that trophoblast EVs were rapidly localised to the CD169⁺ macrophages in the marginal zone of the spleen. Saunderson et al. demonstrated that EV binding to CD169⁺ macrophages is mediated by the interaction between sialic acids on the EVs and macrophage-surface CD169 (also called sialoadhesin) [42]. CD169⁺ macrophages are not typical M1/M2 macrophages and are instead specialised for the uptake of particulate antigens and have recently been shown to be important in the infiltration of immune cells into tumours [42, 43]. CD169⁺ macrophages have also been shown to indicate better clinical outcomes when present in some tumours [44, 45]. In contrast to the spleen, in the inguinal lymph nodes, the EVs were primarily associated with F4/80+ macrophages, with only a few EVs colocalised with CD169⁺ macrophages at the 2 and 24 hour time points. Despite this lack of co-localisation of EVs with CD169⁺ macrophages, there was a very rapid and significant expansion of this cell population in the inguinal lymph node by 24 hours post-EV administration. This suggests that either the interaction of the EVs with F4/80⁺ macrophages induced secretion of a factor(s) that promoted expansion of the CD169⁺ population, or the small number of CD169⁺ macrophages that did interact with the EVs rapidly expanded. No EVs were present in the renal lymph nodes, suggesting that there was specific trafficking of the EVs to the inguinal lymph nodes. This is not surprising as the inguinal lymph nodes drain lymph from the hind limbs and lower abdomen, which was the site of i.p. injection of the trophoblast EVs in this current study.

By day 10 after administration, the EVs began appearing at the margins of the tumours, where they were colocalised with CD169⁺ macrophages. By day 30, the EVs were associated with CD169⁺ macrophages that had deeply infiltrated the tumours. The finding of labelled “EVs” in macrophages in these mice 30 days after administration raises the question, “are these intact EVs or is this simply the stain which was used to label the EVs remaining in the cells that had taken up and processed the EVs”. It seems highly unlikely that these are intact EVs, but it is more like that they are remnants of the stain used to label. In unpublished work, we found the CellTracker^TM stain used in this study to label the trophoblast EVs that can persist in organs 25 weeks after animal injection.

To address whether the CD169⁺ macrophages in the tumours had taken up and retained the cargo of the trophoblast EVs, we examined the tumours and spleens for the presence of two miRNAs from the C19MC cluster. We demonstrated that these miRNAs were detectable in the spleens at 2 and 24 hours post-EV administration and in the tumours, but not until day 30 post-EV administration when the “EVs” were colocalised with the infiltrating CD169⁺ macrophages. The C19MC miRNAs are uniquely expressed in primate placentae [16]. Thus, the presence of these miRNAs in the mouse tumours suggests that the CD169⁺ macrophages had taken up and processed the trophoblast EVs in the spleens, and that the miRNAs had remained in these macrophages, which had then migrated to the tumours. The miRNAs likely retained their function in these cells. The C19MC miRNAs have been shown to have multiple functions in regulating cancer cells [46] and have been shown to regulate the inflammatory STING pathway in macrophages [47]. The detection of C19MC miRNAs in these tissues confirmed that the red fluorescence originated from the injected trophoblast EVs. These results highlight the importance of the i.p. administration route in targeting immune organs and suggest that trophoblast EVs indirectly influence the tumour microenvironment through immune modulation.

Our previous study found increased numbers of both CD169⁺ macrophages and NK cells in tumours at the end point (day 58). A key function of NK cells is to attack tumours. In the current study, although we did not find NK cells in the tumours at day 30, which was 28 days earlier than the previous study, the population of NK cells in the inguinal lymph nodes increased on day 30 after treatment compared to the controls. Accompanying this increase in NK cells, we also found increased expression of IL-15 in the inguinal lymph nodes of tumour-bearing mice on day 30. IL-15 is a key cytokine that promotes NK cell proliferation, survival, and cytotoxic activity [48–53]. Combining the results of our two studies suggests that trophoblast EVs may influence macrophages in the inguinal lymph nodes, where they are associated with increased IL-15 expression and NK cell expansion. Once in the tumours, the macrophages may contribute to NK cell recruitment. This multiphase activation underscores the potential of trophoblast EVs to induce durable and targeted antitumor immunity. Additionally, we observed an indirect relationship between miRNA-519a-5p and miRNA-512-3p in tumours at day 10. The divergence may reflect that cytokine-driven regulation of distinct immune differential pathways, consistent with the parallels between pregnancy and cancer, described by Mor et al. [2].

Interestingly, none of the effects of trophoblast EVs on CD169⁺/F4/80 macrophages, IL-15, or NK cells were observed in non-tumour-bearing mice that had been administered trophoblast EVs, indicating a three-way interplay between trophoblast EVs, the immune system, and the tumours. The tumour microenvironment is often immunosuppressive, impairing NK cell function and reducing the effectiveness of immune responses [54, 55]. Trophoblast EVs appear to counteract this suppression by enhancing macrophage-mediated direct actions on the tumours and cytokine release, including IL-15, which stimulates NK cells in the lymph nodes [56, 57]. However, it is interesting that this study showed that the trophoblast EVs reduced ovarian tumour growth before NK cells were recruited to the tumours.

EVs have shown therapeutic potential in various diseases, in part due to their ability to modulate the immune system, including the activation of immune cells. In this study, we tracked the distribution of trophoblast EVs and investigated the association between trophoblast EVs and immune cells in ovarian cancer xenografts. The limitations of this study include that the xenografts used in this study lack T cells due to the human ovarian cancer cell line used. Further studies, including the use of patient-derived organoids, are required to confirm our findings. Additionally, our findings do not indicate a direct alternative therapy for ovarian cancer. Instead, our findings provide fundamental insights into the role of placental-derived EV in inhibiting ovarian tumour growth and modulating the immune response. These findings may inform future development of natural product-based strategies for ovarian cancer therapy.

Conclusion

In conclusion, following a single i.p. administration into ovarian tumour-bearing mice, trophoblast EVs exhibited selective localisation within the expected organs of EV clearance but also selective localisation to the inguinal lymph nodes and spleens. This resulted in the migration of CD169⁺ macrophages, specifically to the tumours and subsequent increases in CD169⁺ macrophages, IL-15 expression, and NK cell expansion in the inguinal lymph nodes. These changes did not occur in either tumour-bearing mice administered with vehicle control or non-tumour-bearing mice administered trophoblast EVs. Thus, there was a complex interaction between the trophoblast EVs, the immune system, and the tumours, resulting in reduced tumour growth. These findings provide valuable insights into the potential application of trophoblast EVs as therapeutic agents in cancer immunotherapy.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Qi Chen, Larry Chamley, and Lai-Ming Ching conceptualized and supervised the project. Xinyi Sun, Sofian Tijono, and Bridget Tsai executed the animal studies. Xinyi Sun conducted the biodistribution, immunofluorescence, and statistical analyses and drafted the manuscript. Terry Morgan evaluated the H&E-stained sections. Leana Terblanche secured funding for this study. All authors participated in redrafting and reviewing the manuscript and have agreed to the final content.

Funding

This study was supported by PBRF of the School of Medicine, Faculty of Medical Health and Sciences, The University of Auckland, and Nurture Foundation of the Department of Obstetrics, Gynaecology and Reproductive Sciences, The University of Auckland.

Data availability

The datasets used and analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethical approval and consent to participate

This study was approved by the Regional Health and Disabilities Ethics Committee (Approval No. NTX/12/06/057/AM08), and written informed consent was obtained from all participants. All animal experiments were performed in accordance with the Animal Ethics Committee of the University of Auckland (Approval No. AEC2257).

Consent for publication

All authors have reviewed the final version of the manuscript and consent to its publication.

Competing interests

The authors declare that there are no competing interests associated with the manuscript.