Podocalyxin may be down-regulated by progesterone in high grade serous ovarian cancer spheroids to increase responsiveness to carboplatin and NK cells

Abstract

Background

Ovarian cancer is the most lethal gynaecological malignancy with limited therapeutic options. We have previously reported that the transmembrane protein podocalyxin (PODXL) promotes survival of high grade serous carcinoma (HGSC) spheroids against chemotherapy agent carboplatin and NK cell cytotoxicity, and that silencing PODXL by gene editing significantly sensitizes HGSC cells to these treatments. Progesterone, a widely used hormone, has been reported to down-regulate PODXL in endometrial epithelial cells.

Methods

In this current study, we investigated whether progesterone could also lower PODXL in HGSC spheroids to sensitize them to carboplatin and NK cells. Kuramochi cells, a high PODXL expressing HGSC model, was first used to examine the effect of progesterone on PODXL expression. Cells were treated with progesterone or a vehicle control and PODXL levels were assessed. Kuramochi spheroids were then generated from pre-treated cells and exposed to carboplatin to evaluate changes in chemotherapy sensitivity. To assess immune susceptibility, spheroids were co-cultured with primary human NK cells isolated from peripheral blood mononuclear cells, and the effects were examined following 24, 48, and 72 h of co-culture. Additionally, primary HGSC cells were treated with progesterone to assess its ability to modulate PODXL expression in patient-derived samples.

Results

Progesterone pre-treatment significantly reduced PODXL expression in Kuramochi spheroids resulting in increased sensitivity to carboplatin and increased NK cell infiltration and cytotoxicity. Furthermore, progesterone showed potential to reduce PODXL expression in ascites-derived primary HGSC cells.

Conclusions

These findings suggest that further studies should explore the utility of progesterone-mediated down-regulation of PODXL in HGSC cancer to increase responsiveness to chemotherapy and NK cells.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12885-026-16390-8.

Introduction

Ovarian cancer (OC) is the third most common female reproductive cancer, affecting over 300,000 women worldwide and accounting for approximately 107,000 deaths each year [1]. OC remains the most lethal gynaecological malignancy with the lowest 5-year survival rate [1]. High grade serous carcinoma (HGSC) accounts for majority of OC cases and is reported to be responsible for approximately 90% of OC deaths [2]. This is largely attributed by the late subtle presentation of symptoms and lack of sensitive early detection methods [3]. The standard management of OC includes initial cytoreductive surgery followed by a combination of platinum-taxane based chemotherapy consisting mainly of carboplatin and paclitaxel [4]. While most patients respond to initial treatment, up to 90% of patients with advanced stage disease will develop resistance to treatment, resulting in limited therapeutic options and inevitable fatality [5]. In contrast to most cancers, OC primarily metastasises through exfoliation of tumour cells either as single cells, small clusters or tight multicellular aggregates, also known as spheroids, which disseminate throughout the abdominal cavity via ascites fluid [6]. In particular, spheroids are highly tumorigenic and are inherently chemoresistant, however, the mechanisms remain unclear [7–9].

In recent years, natural killer (NK) cells, a cytotoxic subset of innate immune cells in the body, have shown great value in OC treatment which is recognised as an immunogenic disease [10, 11]. Studies have shown that infiltration of NK cells within OC tissues is associated with improved patient outcomes [12, 13]. Therefore, NK cell-based immunotherapy has potential as an innovative therapeutic for targeting HGSC. However, the efficacy of NK cells is often hampered by inadequate NK cell infiltration into the tumor which is closely linked to the tumor microenvironment [14–16].

Reproductive states such as pregnancy have been associated with overall reduced OC risk, with multiparous individuals who carry full term pregnancies exhibiting the greatest risk reduction [17]. It has been widely accepted that this is due to the shift in hormone balance during pregnancy with increased progesterone [18]. Furthermore, OC typically presents as a postmenopausal disease with pronounced incidence in women over the age of 65 with an age range of 50–79 [19–21]. During menopause, due to cessation of ovarian function, serum progesterone concentrations remain very low [22], estrogen levels also decline significantly but other tissues such as fat still produce this hormone [23]. While the use of hormonal replacement therapy remains a controversial matter, oral contraceptive use prior to menopause is associated with a 30% decreased risk of developing OC [24, 25], and progesterone is widely recognised to have an overall protective effect against OC [26, 27]; in particular it is suggested to have a protective role through reduced ovulatory cycles and regulation of epithelial cell proliferation [28]. In HGSC, progesterone receptor expression has been associated with improved clinical outcomes, and progesterone has been reported to exert pro-apoptotic and anti-proliferative effects [29, 30]. In contrast, studies in mice have demonstrated that ovarian progesterone signalling can also be pro-tumorigenic by promoting primary tumor development and metastasis [31]. Therefore, the role of progesterone in HGSC can be complex and multifaceted with its effects depending on the context of model system being used.

Podocalyxin (PODXL) is an extensively glycosylated sialomucin normally expressed in kidney podocytes, endothelia, hematopoietic progenitor cells and epithelial cells [32, 33]. However, PODXL is aberrantly increased in several cancers including breast [34] and ovarian cancer [35, 36]. In HGSC, increased PODXL expression has been associated with poor prognosis and significant decrease in disease free survival [35, 36]. We have recently reported that PODXL promotes the formation of chemoresistant HGSC cancer spheroids in both HGSC-derived cell lines and ascites-derived primary HGSC cells [36]. We have also reported that PODXL protects HGSC spheroids from NK cell infiltration and spheroid destruction [37], which was again observed in both cell lines and primary HGSC cells. Furthermore, silencing PODXL expression by CRISPR/Cas9 gene editing has significantly sensitized HGSC cancer spheroids to both carboplatin treatment and NK cell infiltration [36, 37]. These studies indicate that, lowering PODXL in HGSC therapeutically, if possible, may sensitize HGSC spheroids to chemotherapeutics and NK cell infiltration and toxicity.

To date, it is unclear how PODXL expression is regulated in HGSC. However, recent studies have reported that progesterone can down-regulate PODXL in endometrial epithelial cells for embryo implantation [38, 39]. It was further suggested that this PODXL down-regulation is partly due to the actions of certain microRNAs which are up-regulated by progesterone [39]. Nevertheless, these studies indicate that progesterone may down-regulate PODXL in reproductive tissues and cells.

This study thus investigated whether progesterone can down-regulate PODXL in HGSC spheroids to render them more vulnerable to the chemo drug carboplatin and NK cell infiltration. We first employed a cell line model of HGSC spheroids using Kuramochi cells, which express the highest level of PODXL among known HGSC cell lines [36]. These cells were treated with progesterone and PODXL protein was analysed by Western blot and immunofluorescence. The progesterone-treated cancer spheroids were then exposed to carboplatin and spheroid characteristics were examined. The progesterone-treated cancer spheroids were also co-cultured with primary human NK cells isolated from peripheral blood mononuclear cells (PBMCs), and NK cell infiltration and related cytotoxicity were investigated. We next explored whether progesterone could also reduce PODXL in primary ascites-derived HGSC cells.

Methods

Cell culture

Kuramochi cell line was purchased from CellBank Australia (NSW, Australia) and cultured in RPMI 1640 Medium + GlutaMAX supplement (Thermo Fisher Scientific, MA, USA, #61870036). Ascites-derived primary HGSC cells were isolated as previously described [36], with ethics approval obtained from Monash University Human Research Ethics Committee (#06032 C). They were maintained in a 1:1 ratio of Medium 199 (Thermo Fisher Scientific, #11150-059) and MCDB131 (Thermo Fisher Scientific, #10372-019). All media were supplemented with 10% (for Kuramochi cells) or 15% (for primary cells) fetal bovine serum (FBS, Thermo Fisher Scientific) and 1% antibiotic–antimycotic (Thermo Fisher Scientific, #15240062); all cells were cultured at 37 °C under 5% CO₂.

For hormonal treatment, the previously published method for endometrial cells was utilized [39, 40]. Briefly, 20,000 cells were seeded into 6-well plates and incubated for 5 h at 37 °C under 5% CO₂ conditions. To upregulate progesterone receptors, cells were then primed for 48 h with 10 nM of 17β-estradiol (Sigma Aldrich, MO, USA) in RPMI 1640 Medium + GlutaMAX supplement (Thermo Fisher Scientific, MA, USA, #61870036), supplemented with 2% charcoal stripped FBS (CS-FBS). Cells were then changed to fresh medium containing 5 µM progesterone (4-pregnene-3,20-dione, P) (Sigma Aldrich) or an equivalent amount of 100% ethanol as the vehicle control (VC) and cultured for 24 h. The cells were then washed, trypsinized, and seeded at 2,500 cells per well in 96-well round bottomed ultra-low attachment plate (Merck, Darmstadt, Germany, #CLS7007-24EA) and cultured for 72 h in their respective treatments (P or VC) to form spheroids. Cells were snap frozen on dry ice and stored at -80 °C for subsequent protein extraction.

Western blot

This was carried out as previously described [36]. In brief, cells were thawed from − 80 °C and lysed in buffer containing 50 mM Tris, 150 mM NaCl, 2 mM EDTA, 25 mM NaF (serine/threonine protein phosphatase inhibitor), 0.2% Triton X-100, 0.3% Nonidet P-40 (Millipore, Sigma-Aldrich), 25 mM glycerolphosphate (a phosphatase inhibitor) with the addition of a complete protease inhibitor cocktail (Roche Molecular, Mannheim, Germany). Proteins were separated on a 10% polyacrylamide SDS-PAGE gel, transferred to PVDF membranes (GE Healthcare, NSW, Australia). The membranes were first blocked for 5 h at room temperature with 5% BSA (Bovogen, VIC, Australia) in Tris-buffered saline [10 mmol/L Tris (pH 7.5) 232 and 0.14 mol/L NaCl] containing 0.2% Tween20 (TBST-T), they were next incubated with polyclonal goat antibodies against human PODXL (2.5 µg/ml in TBS-T, R&D Systems, MN, USA #AF1568) overnight at 4 °C, then with a rabbit anti-goat IgG-HRP (diluted 1:2000 in TBS-T; Dako, USA #P0449) for 1 h. Bands were visualised using Lumi-Light Western Blotting Substrate (Roche Molecular, #12015200001) and ChemiDoc MP Imaging System (Bio-Rad, CA, USA). Membranes were subsequently probed for β-actin as a loading control using an HRP-conjugated β-actin antibody (2 µg/ml, Cell Signalling Technology, MA, USA). Densitometry of the bands were performed using the ImageJ software (National Institutes of Health, Bethesda, MD, USA) and normalised to β-actin. Data are presented as fold change relative to the VC. Western blot analyses were repeated using lysates from 3 independent experiments.

Treatment of cancer spheroids with carboplatin

Spheroids pre-treated with progesterone or VC were exposed for 24 h to a dose of 348 µg/ml (937µM) carboplatin (Hospira, IL, USA) [31], then maintained in complete media containing 10% CS-FBS and analysed at day 2 and 4 as previously reported [36]. Briefly, culture media were refreshed every 2 days by replacing half the media with fresh media. To determine the number and viability of cells within spheroids, individual spheroids were trypsinised and dissociated into single cells and analysed using an automated cell counter Countess 3 (Thermo Fisher Scientific). Images of the treated spheroids were taken at different time points using the Nikon eclipse TS100 microscope equipped with a Nikon DS-Fi1 camera (Tokyo, Japan). All experiments were repeated 3 times independently.

Immunofluorescence

All procedures were performed as previously described [36], at room temperature unless stated otherwise. In brief, spheroids were fixed with 4% paraformaldehyde for 30 min, permeabilised with 0.1% Triton X-100 for 10 min, and blocked with 15% normal donkey serum for 2 h. Spheroids were next incubated overnight at 4 °C with goat anti-human PODXL antibodies (2.5 µg/ml diluted in 1% BSA/PBS, R&D Systems, #AF1568) or goat IgG (2.5 µg/ml diluted in 1% BSA/PBS, R&D Systems, #AB-108-C), then 2 h with Alexa Fluor 488 donkey anti-goat antibodies (1 µg/ml diluted in 1% BSA/PBS, Invitrogen, #A11055), and nuclei were counterstained with DAPI (0.5 µg/ml diluted in PBS, R&D Systems, #AB-108-C). For imaging, 10 mm glass coverslips (Marienfeld, Laidao-Königshofen, Germany, #0111500) were mounted onto the corners of the microscope slide (Marienfeld, #0705002) to function as spacers, and spheroids were then pipetted to the middle of the slide with a drop of fluorescent mounting agent (Dako, #S3023), mounted with a coverslip, and analysed using the ECLIPSE Ti A1R confocal microscope (Nikon, Japan). Images were taken at 20X magnification.

To assess Ki-67 positive cells, spheroids were resuspended into single cells in trypsin and pipetted onto a droplet of Histogel (Epredia, MI, USA, #HG-4000-012), then smeared onto a microscope slide. After the gel was air dried, cells were fixed and immunostained as described above but with anti-Ki-67 rabbit antibody (1:250 dilution in 1% BSA/PBS, Abcam, Cambridge, UK, #ab16667) or rabbit IgG (4 ug/ml diluted in 1% BSA/PBS, Dako, #X0936), and a rabbit-anti mouse Alexa Fluor 568 antibody (1:200 dilution in 1% BSA/PBS, Thermo Fisher Scientific, #a10042). Coverslips were mounted with a fluorescent mounting agent (Dako) and imaged using the Olympus BX60 (Nikon, Japan) fluorescent microscope. The percentage of Ki67 positive cells over the total number of live cells were calculated.

Isolation of primary human NK cells

Primary human NK cells were isolated from frozen PBMCs of 4 healthy female volunteer donors as previously described [37], which was provided by the Australian Red Cross with ethics approval by RMIT College of Human Ethics Advisory Network (#28056). All work was conducted according to the Declaration of Helsinki Principles and the Australian National Health and Medical Research Council (NHMRC) Code of Practice. Signed informed consent was obtained from all donors before the study. NK cells were isolated using the NK Cell Isolation Kit (Miltenyibiotec, Bergisch Gladbach, Germany, #130092657), LS Column (Miltenyibiotec, #130122729,) and QuadropMACS™ Separator (Miltenyibiotec, #130090976) according to the manufacturers’ protocol. The resulting NK cells were confirmed by analysis on the LSR Fortessa X-20 flow cytometer (BD Biosciences, USA) on the basis of CD3- and CD56 + expression as previously reported [37]. On average, 93% of the enriched cells were confirmed to be NK cells. NK cells were activated and maintained in complete RPMI 1640 Medium supplemented with 10ng/ml IL-15 (STEMCELL technologies, BC, Canada, #78031.1) for 72 h at 37 °C under 5% CO₂; they were then stained with carboxyfluoroscein succinimidyl ester (CFSE, Thermo Fisher Scientific, #C34570) for 10 min (diluted 1:1000 in PBS) as per manufacturer’s protocol before being co-cultured with cancer spheroids.

Co-culture of cancer spheroids with NK cells

Kuramochi spheroids pre-treated with progesterone or VC were co-cultured for 24, 48 and 72 h respectively with NK cells that were pre-labelled with fluorescent dye CFSE. The number of NK cells added was estimated at an effector-to-target ratio of 1:3 (NK: cancer cells) as previously described [37], and NK cell infiltration and cytotoxicity of cancer cells were assessed.

Assessment of NK cell infiltration and the impact on spheroid size and cell number

At each time point of assessment, brightfield images of the co-cultured spheroids were taken using the Nikon eclipse TS100 microscope equipped with a Nikon DS-Fi1 camera (Tokyo, Japan); subsequently confocal images of co-cultured spheroids were taken with an A1R confocal microscope (Nikon, Japan) to identify the fluorescently labelled NK cells. The co-cultured spheroids were then washed with PBS using a 10 µl pipette tip until all attached NK cells on the outside of the spheroids were completely removed, after which brightfield images of the washed spheroids were taken.

To analyse NK cell infiltration into the spheroid, total NK cell fluorescence intensity within the co-cultured and unwashed spheroids was quantified (as washing may lead to loss of NK cells) as previously described [37]. Briefly, the outline of each spheroid was determined on the washed spheroid then overlayed onto the confocal image of the unwashed spheroid, the total NK cell fluorescence intensity within the spheroid were then determined using the ImageJ software version 1.53c (NIH, USA). The final data were expressed as total fluorescence after subtracting the background. The experiment was repeated 4 times using NK cells isolated from 4 different donors (n = 4). For each donor, 3 individual spheroids were analysed for each group and time point, and their average was used for data representation. This was applied consistently to all subsequent quantifications following NK cell co-culture experiments.

The volume of spheroids following the PBS wash was also determined at each time point. The diameter of each spheroid was measured using the average length value of 4 different angles using the “straight line” function of the ImageJ software, and the spheroid volume was calculated using the formula of 4/3πr³. The number and viability of cells contained within the washed spheroids were also analysed. To do this, the washed spheroids were trypsinized and dissociated into single cells, they were then mixed with trypan blue (Thermo Fischer Scientific) and live cells were analysed with the automated cell counter countess 3 (Thermo Fisher Scientific).

Assessment of NK cell induced cytotoxicity

To assess NK cell induced apoptosis, CellEvent Caspase-3/7 red detection reagent (Thermo Fisher Scientific, #C10430) was added to complete RPMI 1640 media in 1:100 dilution in both co-cultured and untreated spheroids for 1 h. Spheroids were then live imaged using the confocal microscope. Subsequently, the VC and progesterone treated spheroids were dissociated through gentle pipetting before fluorescence at ~ 502/530 nm (excitation/emission) was immediately measured on the CLARIOstar^® Plus plate reader (BMG LABTECH, Ortenberg, Baden-Württemberg, Germany).

Statistical analyses

GraphPad Prism version 10 (GraphPad Software, San Diego, CA) was used for statistical analysis. Two-way ANOVA or Paired T-test was applied wherever appropriate, and data were expressed as mean ± standard deviation (SD); P ≤ 0.05 was considered significant.

Results

Progesterone down-regulates PODXL in HGSC cell line of Kuramochi

We used Kuramochi cell line as a model for this study, since these cells naturally express the highest levels of PODXL among known HGSC cell lines [36]. To determine whether progesterone can down-regulate PODXL in these cells, Kuramochi cells were treated with progesterone (5 µM) or the vehicle control (VC) for 72 h and PODXL protein was examined by western blot. As shown in Fig. 1A, PODXL protein levels were significantly lower in cells treated with progesterone than VC, and the mean reduction was quantified to be 44% (Fig. 1A). Immunofluorescent staining of spheroids formed with these cells further demonstrated that PODXL was much lower in those treated with progesterone (Fig. 1B). These data thus showed that progesterone can reduce PODXL in HGSC Kuramochi cell line, consistent with previous findings made in human endometrial epithelial cells [39, 40].

Fig. 1.

Down-regulation of PODXL in Kuramochi cells by progesterone. Cells treated with vehicle control (VC) or progesterone (P) were compared. A Western blot analysis of PODXL protein, β-actin was used as the loading control (full-length blots/gels are presented in Supplementary Fig. 1) graph: densitometric analysis of PODXL levels, data normalized to β-actin and expressed as fold changes relative to VC. B Immunostaining of PODXL in spheroids assembled from VC or P treated cells. Representative brightfield and confocal images are shown, PODXL in green and nuclei in blue. Scale bar: 100 μm. Mean ± SD, n = 3. *P < 0.05

Kuramochi spheroids pre-treated with progesterone are more sensitive to carboplatin

We next examined the chemosensitivity of Kuramochi spheroids with lower levels of PODXL following treatment with progesterone. Spheroids were pre-treated with progesterone or VC, then exposed for 24 h to a single dose of 348 µg/ml (937µM) carboplatin, a mean concentration found in the perfusate of Epithelial OC patients undergoing hyperthermic intraperitoneal chemotherapy [41], spheroids were then monitored in complete media for up to 4 days to examine post-treatment recovery (Fig. 2). Bright field images showed that spheroids pre-treated with progesterone were visibly smaller at all time points following the exposure to carboplatin (Fig. 2A). Cell viability was comparable between the two groups immediately after the carboplatin treatment (Fig. 2B), however, during post-treatment recovery, cell viability decreased more sharply and was significantly lower at both day 2 and 4 in spheroids pre-treated with progesterone than VC (Fig. 2B). This was also reflected in the live cell numbers detected inside spheroids, which were similar between the two groups at day 0, but reduced more quickly over time in spheroids that were pre-treated with progesterone than VC (Fig. 2B). To examine whether these cells were capable of proliferating, day 0 spheroids were dissociated into single cells and immunostained for cell proliferation marker Ki67 (Fig. 2C, images). Percentage of cells positively stained for Ki67 was significantly lower in spheroids that were pre-treated with progesterone than VC (20% vs. 40%, Fig. 2C, bar graph), indicating fewer cells were more actively proliferating in the former than latter spheroids. These data suggest that spheroids pre-treated with progesterone were more sensitive to carboplatin.

Fig. 2.

Sensitisation of Kuramochi spheroids to carboplatin by pre-treatment with progesterone. Spheroids pre-treated with vehicle control (VC) or progesterone (P) were exposed to carboplatin for 24 h, then monitored for recovery in complete media for up to 4 days, time immediately after the carboplatin treatment was designated as day 0. A Representative brightfield images of spheroids. B Cell viability (assessed by trypan blue) and total number of live cells within spheroids. C Immunofluorescent staining of Ki67 in spheroids at day 0. Red, Ki67; blue, DAPI; scale bar, 50 μm. Bar graph: quantification of Ki67 positive cells over total cells inside spheroids. Mean ± SD, n = 3. *P < 0.05, **P < 0.005, **** P < 0.0001

Kuramochi spheroids pre-treated with progesterone are more vulnerable to NK cell infiltration

We next examined whether suppression of PODXL by progesterone could influence spheroid susceptibility to NK cells. Spheroids pre-treated with progesterone or VC were co-cultured for 24, 48 and 72 h respectively with human NK cells isolated from PBMCs of healthy female donors (n = 4), and these NK cells were pre-labelled with a green fluorescent CFSE dye. Figure 3A shows representative brightfield and confocal images of spheroids immediately after the co-culture, and after being washed with PBS to remove un-infiltrated NK cells that surrounded the spheroids. For both types of spheroids (treated with VC or progesterone), a thick “ring” of NK cells was present on the periphery of each spheroid, which was more obvious on the confocal images (Fig. 3A-a and b, both panels). The two groups showed subtle differences especially at later time points of co-culture with NK cells, with the spheroids pre-treated with progesterone displaying more NK cells towards the spheroid centre than those pre-treated with VC.

Fig. 3.

Sensitisation of Kuramochi spheroids to NK cell infiltration by pre-treatment with progesterone. Spheroids pre-treated with vehicle control (VC) or progesterone (P) were co-cultured with NK cells for 24, 48 and 72 h respectively. A Representative images of spheroids co-cultured with NK cells. Top panel: spheroids pre-treated with VC. Bottom panel: spheroids pre-treated with P. For each panel: a and b, immediately after the co-culture; c, after PBS wash to remove NK cells still present outside the spheroids; a and c, brightfield images; b, confocal images of NK cells (fluorescently stained in green). B Analysis of NK cell infiltration into spheroids. Images of b in A) (NK cells in green) overlaid with the outlines of spheroids as yellow circles derived from images of c in A). Bar graph, total NK cell fluorescence present inside spheroids. C Analysis of spheroid size following PBS wash. Images of c in A) presented together with yellow circle outlines shown in B). Bar graph, spheroid volume. D Analysis of live cells present within the spheroid after the wash. Data presented as percentage of live cells of untreated counterpart spheroids at each time point. Data as mean ± SD, n = 4. *P < 0.05, ** P < 0.01

We next further examined NK cell infiltration into spheroids and the impact on spheroid size and cell number. To analyse NK cell infiltration, we determined the total fluorescence of NK cells present within the co-cultured spheroids before PBS wash, as washing may cause leakage of NK cells out of the spheroids. To do this, the outline of each spheroid was determined on washed spheroids then overlaid onto the confocal image of the un-washed spheroid (Fig. 3B, top panel), and the total NK cell fluorescence within this outline was quantified (Fig. 3B, bar graph). While large degrees of NK cell infiltration were observed in both groups at 24 h without significant differences, by 48 and 72 h, significantly more NK cells were present inside the spheroids that were pre-treated with progesterone than VC, indicating that the former had higher levels of sustained NK cell infiltration (Fig. 3B), and that higher levels of PODXL in VC group had a protective role against NK cell infiltration over time.

To examine the impact of NK cell infiltration on spheroid size, the images of the washed spheroids (Fig. 3C, images also show spheroid outlines used in Fig. 3B) were used to measure the diameters of spheroid outlines to calculate their volumes (Fig. 3C, bar graph). With longer time of co-culture with NK cells, both groups of spheroids increased in size over time, however, spheroids pre-treated progesterone were smaller than those pre-treated with VC (Fig. 3C). In the absence of NK cells, spheroids of both groups exhibited similar sizes (Supplementary Fig. 2A), indicating that the differences observed were largely associated with NK cell infiltration.

We further assessed the total live cell numbers remaining inside the spheroids (Fig. 3D). Within the first 24 h of co-culture, both groups of spheroids showed a steep 50% decline in cell counts. However, with increasing co-culture time, live cell numbers stared to increase in spheroids pre-treated with VC but not in those pre-treated with progesterone, therefore the latter spheroids had significantly fewer live cells than the former spheroids at both 48 and 72 h (Fig. 3D). These data suggest that spheroids pre-treated with progesterone were more vulnerable to NK cell infiltration.

Kuramochi spheroids pre-treated with progesterone show higher NK cell-induced cytotoxicity

Next, we assessed if pre-treatment with progesterone would affect NK cell-induced cytotoxicity in these cancer spheroids. Spheroids pre-treated with progesterone or VC were co-cultured with NK cells for 24, 48 and 72 h respectively as described above, then caspase-3/7 activity within spheroids was measured as a fluorescence readout at each time point (Fig. 4). Visually, caspase-3/7 activity was apparent in both groups of spheroids, but the signals were stronger in spheroids pre-treated with progesterone than VC (Fig. 4A-b and c, both panels). Furthermore, NK cells overlapped more with caspase-3/7 activities inside spheroids that were pre-treated with progesterone than VC (Fig. 4A-b, both panels).

Fig. 4.

Sensitisation of Kuramochi spheroids to NK cell-induced caspase 3/7 activity by pre-treatment with progesterone. Spheroids pre-treated with vehicle control (VC) or progesterone (P) were co-cultured with NK cells for 24, 48 and 72 h respectively then analysed for caspase-3/7 activity. A Representative images of spheroids co-cultured with NK cells. Top panel: spheroids pre-treated with VC. Bottom panel: spheroids pre-treated with P. For each panel: a, brightfield images of spheroids co-cultured with NK cells; b and c, confocal imaging of caspase-3/7 activity (red) overlaid with (b) or without (c) NK cells (green). B and C Analysis of caspase-3/7 activity. B) Representative images of c in A) overlaid with spheroid outlines as yellow circles. C) Quantification of caspase-3/7 activity. Data presented as total fluorescence reading. D Representative images of Ki67 immunostaining in spheroids. Data after 48 h co-culture with NK cells are presented. Red, Ki67; blue, DAPI; scale bar, 50 μm. Bar graph, quantification of Ki67-positive cells over all live cells. Mean ± SD, n = 4. *P < 0.05, ** P < 0.01

To further analyse the caspase-3/7 activity inside the spheroids, confocal images of caspase-3/7 activity shown in Fig. 4A was overlayed with spheroid outlines as done for Fig. 3C, which indicated that the red signals were contained inside spheroids but more intensely in those pre-treated with progesterone than VC (Fig. 4B). We then measured the total fluorescence readings inside these spheroids (Fig. 4C). While no apparent difference was seen at 24 h, with increasing time of co-culture with NK cells, caspase-3/7 activity remained higher in spheroids pre-treated with progesterone than VC, and these differences were associated with co-culture with NK cells; as caspase-3/7 was negligible in spheroids of both groups in the absence of NK cells (Supplementary Fig. 2B). Thus, these results indicate that overtime spheroids pre-treated with progesterone endured higher rates of NK cell-induced cytotoxicity than VC-treated spheroids.

Next, we assessed Ki67 to examine whether the proliferative capacity of the surviving cells within the spheroids were also affected (Fig. 4D). As our data consistently showed no differences in cell numbers at 24 h, here we only examined the Ki67 staining at 48 h where spheroids started exhibiting difference in cell numbers. Fewer Ki67-positive cells were displayed in spheroids pre-treated with progesterone than VC (Fig. 4D, images). Quantification showed that 45% and 64% of cells in spheroids pre-treated with progesterone and VC respectively were positive for Ki67 (Fig. 4D, bar graph). Thus, Ki67 positivity was inversely correlated to caspase-3/7 activities in these spheroids (Fig. 4). Again, in the absence of NK cells, spheroids of both groups consistently showed over 50% Ki67 positivity (Supplementary Fig. 2C).

PODXL may also be down-regulated by progesterone in ascites-derived primary HGSC cells

We next endeavoured to investigate ascites-derived primary cells obtained from three high PODXL-expressing HGSC patients as reported in our previous studies [36, 37]. However, cells from only one patient could be cultured successfully for this trial, and these cells were treated with progesterone or the vehicle control (VC) for 72 h and PODXL protein was examined by western blot. As shown in Fig. 5, PODXL protein levels were significantly lower in cells treated with progesterone than VC, with a mean reduction of 50%. We tried to further study how this reduction of PODXL would influence the responsiveness of these cells to carboplatin and NK cells, unfortunately, further experimentation could not be conducted due to the very low proliferative capacity of these primary cells. Nevertheless, these initial findings provide important support for the observations made with Kuramochi cell line and the rationale for further exploring the potential utility of progesterone to down-regulate PODXL in HGSC.

Fig. 5.

Down-regulation of PODXL by progesterone in ascites-derived primary HGSC cells. Cells treated with vehicle control (VC) or progesterone (P) were compared. A Western blot analysis of PODXL protein, β-actin was used as the loading control (full-length blots/gels are presented in Supplementary Fig. 3) graph: densitometric analysis of PODXL levels, data normalized to β-actin and expressed as fold changes relative to VC. Mean ± SD, n = 3. ** P < 0.01

Discussion

We have previously reported that PODXL promotes the formation of chemoresistant HGSC cancer spheroids, and that PODXL protects these spheroids from NK cell infiltration [36, 37]. Furthermore, silencing PODXL in these spheroids by CRISPR/Cas9 gene editing significantly sensitize them to both carboplatin treatment and NK cell infiltration [36, 37]. Extending these observations, the current study investigated whether progesterone, which was reported to down-regulate PODXL in endometrial epithelial cells [39], could also downregulate PODXL in HGSC cancer spheroids and sensitize them to carboplatin and NK cells. Using Kuramochi cell lines as a model, we demonstrated that progesterone could indeed significantly reduce PODXL levels in these HGSC cells and spheroids. Importantly, pre-treatment with progesterone significantly increased sensitivity of Kuramochi spheroids to carboplatin, resulting in significantly higher cell death and less proliferation. Furthermore, spheroids pre-treated with progesterone also exhibited significantly higher NK cell infiltration and higher caspase-3/7 activities inside the spheroids. We also showed the potential of progesterone to reduce PODXL in some ascites-derived primary HGSC cells. These findings suggest that further studies are warranted to investigate the value of down-regulating PODXL in HGSC by progesterone to possibly sensitise them to chemotherapy and NK cells.

In oral tongue squamous cell carcinoma and osteosarcoma, PODXL has been reported to significantly mediate cisplatin resistance [42, 43]. We have previously reported that, in both Kuramochi cell line as well as ascites-derived primary HGSC cells, PODXL promotes cell proliferation and fosters the formation of dense cancer spheroids which are resilient to physical fragmentation and chemotherapy drugs [36]. When PODXL was knocked out in Kuramochi cell line by gene editing, spheroids of the resulting cells were less resistant to carboplatin (a second-generation analogue of cisplatin), leading to poorer cell viability and higher cell death [36]. Furthermore, spheroids formed with primary HGSC cells also showed differential sensitivity to carboplatin depending on the level of PODXL expression, those with higher levels were more resilient than those with lower PODXL [36]. The current study further explored therapeutic reduction of PODXL and showed that progesterone could suppress PODXL in Kuramochi cells, and that this suppression was sufficient to sensitise spheroids to carboplatin. Spheroids pre-treated with progesterone exhibited significantly lower cell viability and faster decline in cell numbers than spheroids pre-treated with VC in response to carboplatin treatment. This was also reflected by significantly lower numbers of Ki67-positive cells remaining within the spheroids pre-treated with progesterone, which consequently also appeared to be smaller in size, compared to those pre-treated with VC. These data also suggest that a complete ablation of PODXL may not be necessary to elicit beneficial effects, as PODXL was reduced but not totally removed by progesterone.

Several studies have reported that OC cells are susceptible to NK cell-mediated cytotoxicity in both in vitro and in vivo [14, 44–46]. Tumor spheroid models, which mimic the 3D structure of the tumor and its microenvironment, have been increasingly used in co-culture studies over monolayer models. However, to date only a few studies have used spheroids to assess NK cell infiltration and intratumoral cytotoxicity [47–50]. When OC cell lines SKOV3, IGROV1 and OVCAR3 were co-cultured with NK cells derived from haematopoietic stem and progenitor cells, NK cells showed high capability to migrate and infiltrate spheroids to mediate intratumoral killing of cancer cells [51]. Our previous studies showed that NK cells isolated from PBMCs of healthy women were also able to infiltrate into spheroids formed with HGSC cell line Kuramochi as well as primary HGSC cells [37]; however, PODXL played a protective role, and knockout of PODXL significantly enhanced NK cell infiltration and apoptosis of cancer cells [37]. In the current study, we showed that lowering PODXL by progesterone also sensitised Kuramochi spheroids to NK cells. Spheroids pre-treated with progesterone had significantly higher infiltration of NK cells and higher apoptosis of cancer cells over time, resulting in smaller spheroids with fewer live cells remaining within. We noted that at 24 h, while both groups showed similar number of infiltrated NK cells and viable cancer cells, spheroids pre-treated with progesterone were smaller those pre-treated with VC, which may reflect subtle differences between the two groups in shedding of dead or loosely attached cells after the initial attack of NK cells. They may also reflect subtle differences in spheroid compactness.

Interestingly, by 72 h of co-culture with NK cells, spheroid volume started to increase in all cases, suggesting that all spheroids were recovering from NK cell attack. However, number of live cells as well as percentage of proliferating cells inside spheroids remained significantly lower in those pre-treated with progesterone than VC, suggesting that the former spheroids were recovering at a much slower rate. However, spheroid sizes at this time point did not differ significantly between the two groups, which may be due to dead cells still remaining within the spheroids.

Since progesterone is a widely used medicine in women’s reproductive health and has an overall protective effect against OC [26, 27], the potential benefit of progesterone in pre-treating HGSC patients warrants further investigation. This is the first study showing that progesterone can sensitise HGSC spheroids expressing high levels of PODXL to chemotherapeutics such as carboplatin and NK cell infiltration in a cell line model. We also obtained limited but encouraging data showing that progesterone could also reduce PODXL in some ascites-derived primary HGSC cells. However, with the scarce availability of such primary cells, and they were hard to maintain in culture and could not be expanded indefinitely, we encountered difficulties to do more extensive validations, which is a limitation of this study. Given the preliminary nature of these findings, no definitive conclusions can yet be drawn from data derived from primary HGSC spheroids. Further studies are required to optimise culturing conditions of primary HGSC cells and to determine whether progesterone-induced reduction of PODXL may enhance responsiveness of these cells to chemotherapy and NK cell-mediated cytotoxicity. In addition, a larger cohort of HGSC patients will also help address the scarcity of primary HGSC cells and account for possible donor to donor variability.

In summary, this study suggests that even lowering PODXL levels by 44% may profoundly reduce HGSC cell survival against carboplatin and NK cell activity.

Abbreviations

OC

Ovarian cancer

HGSC

High grade serous carcinoma

NK

Natural killer

PODXL

Podocalyxin

PBMCs

Peripheral blood mononuclear cells

CFSE

Carboxyfluoroscein succinimidyl ester

VC

Vehicle control

P

Progesterone

Authors’ contributions

G.N conceived and oversaw the project, G.N and Y.W and N.L.T designed the study. N.L.T conducted the experiments, analysed the data, and wrote the manuscript under the guidance of Y.W and G.N. M.B and A.S provided carboplatin. K.Q. provided the blood samples and guidance on PBMC isolation. All authors have read and approved of the manuscript.

Funding

This study was supported by the National Health and Medical Research Council (NHMRC) of Australia (#2012523 to G.N) and Contributing to Australian Scholarship and Science (CASS) foundation (#10453 to Y.W).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Primary HGSC cells were acquired from ascites fluid of patients admitted to Eastern Health hospitals in Melbourne, Australia; ethics approval was obtained from Monash University Human Research Ethics Committee (#06032 C). Human NK cells were isolated from PBMC’s derived from whole blood of healthy patients provided by the Australian Red Cross with ethics approval by RMIT College of Human Ethics Advisory Network (#28056). All work was conducted according to the Declaration of Helsinki Principles and the Australian National Health and Medical Research Council (NHMRC) Code of Practice. Signed informed consent was obtained from all donors before the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.