IL-17A promotes fatty acid uptake through the IL-17A/IL-17RA/p-STAT3/FABP4 axis to fuel ovarian cancer growth in an adipocyte-rich microenvironment

Abstract

Pro-inflammatory cytokines are crucial mediators of cancer development, representing potential targets for cancer therapy. The molecular mechanism of a vital pro-inflammatory cytokine, IL-17A, in cancer progression and its potential use in therapy through influencing fatty acid (FA) metabolism, especially FA uptake of cancer cells, remains unknown. In the present study, we used IL-17A and ovarian cancer (OvCa), a representative of both obesity-related and inflammation-related cancers, to explore the interactions among IL-17A, cancer cells and adipocytes (which can provide FAs). We found that in the presence of palmitic acid (PA), IL-17A could directly increase the cellular uptake of PA, leading to the proliferation of OvCa cells via the IL-17A/IL-17RA/p-STAT3/FABP4 axis rather than via CD36. Moreover, in vivo experiments using an orthotopic implantation model in IL-17A-deficient mice demonstrated that endogenous IL-17A could fuel OvCa growth and metastasis with increased expression of FABP4 and p-STAT3. Furthermore, analysis of clinical specimens supported the above findings. Our data not only provide useful insights into the clinical intervention of the growth and metastasis of the tumors (such as OvCa) that are prone to growth and metastasis in an adipocyte-rich microenvironment (ARM) but also provides new insights into the roles of IL-17A in tumor progression and immunomodulatory therapy of OvCa.

Electronic supplementary material

The online version of this article (10.1007/s00262-019-02445-2) contains supplementary material, which is available to authorized users.

Introduction

Recently, the lipid metabolism of cancer cells has been reported to be a promising therapeutic target for cancer [1]. The adipocyte-rich microenvironment (ARM) plays an essential role in the growth and metastasis of various tumors, such as gastric, breast, ovarian, prostate and colon cancers [2–6]. In the ARM, in addition to glucose, cancer cells can uptake and utilize free fatty acids (FAs) provided by adipocytes through certain and/or uncertain mechanisms, such as by upregulating the expression level of fatty acid-binding protein 4 (FABP4, also known as aP2) or CD36 molecules [2–6]. Moreover, studies in the last decade have revealed that various types of immune cells are recruited to white adipose tissue (WAT) beds, where they can secrete cytokines to modulate glucose metabolism within WAT [7]. Currently whether and how these immune cells and the secreted cytokines within the ARM influence the FA metabolism (particularly FA uptake) of cancer cells and impact the growth and/or metastasis of cancer remains unclear.

Recently, the interleukin IL-17 (IL-17, also called IL-17A) family, IL-17-producing cells, and the IL-17/IL-17R axis have garnered much attention as vital components of the tumor microenvironment (TME) [8–15]. IL-17 cytokines are primarily produced by CD4⁺ T helper 17 (Th17) cells, natural killer (NK) cells, γδ T cells, and innate lymphoid cells (ILC). IL-17RA is ubiquitous and is a common co-receptor subunit for the IL-17 family cytokines [9, 13]. IL-17A and IL-17RA have been established as therapeutic targets for colorectal cancer, and immune neutralization of IL-17A can inhibit tumor progression [16]. Therefore, neutralization of inflammation by targeting IL-17 may be considered as an immune strategy in addition to the currently available targeted cancer therapies. However, there are very few reports about the involvement of IL-17A in cancer progression and/or therapy as it relates to influencing FA metabolism, especially the FA uptake of cancer cells.

FABP4 is an important lipid chaperone protein that governs diverse biology depending upon the tissue source and disease context [17]. Nieman et al. reported that cancer-associated adipocytes (CAAs) release FAs through lipolysis, and the FAs are transferred to OvCa cells with the help of FABP4; then they are utilized for energy production (ATP) through β-oxidation to support tumor progression and uncontrolled growth [2, 3]. FABP4 has been reported to be a key mediator between adipocytes and cancer progression in prostate and ovarian cancer [3, 18]. The mechanisms by which FABP4 is regulated are complicated. Researchers have identified that adipocyte-derived IL-8, PPARγ and endothelial DLL4–NOTCH can upregulate FABP4 expression [3, 19, 20] whereas miR-409-3p downregulates FABP4 expression [21]. Very few studies have examined the impacts of pro-inflammatory cytokines in the ARM (such as IL-17A) on FABP4 expression.

Ovarian cancer (OvCa) is a representative of both obesity-related and inflammation-related cancers, and it remains the most lethal of all gynecological malignancies, carrying a 5-year survival rate of 45% [22–24]. Since it lacks specific clinical symptoms at early stages, most cases of OvCa are diagnosed at advanced stages (FIGO stage III–IV), after disseminated intra-abdominal metastasis has already occurred [25, 26]. OvCa typically metastasizes to the omentum, which is rich in adipocytes and OvCa rarely metastasizes hematogenously [27].

Based on the above data, we speculated that inflammatory IL-17A might directly increase the FA uptake of OvCa cells via the IL-17A/IL-17RA/FABP4 axis to fuel OvCa growth and/or metastasis. We have now endeavored to test this hypothesis with OvCa cell lines, IL-17A-deficient (IL-17A^−/−) mice and clinical specimens. This study provides new insights into the roles of IL-17A in tumor progression and immunomodulatory therapy in OvCa.

Materials and methods

Quantitative real-time polymerase chain reaction (qRT-PCR)

To examine the effect of recombinant human IL-17A (rhIL-17A, PeproTech Cat# 200-17) on the mRNA level of FABP4 in OvCa cells, OVCAR3 and A2780 cells were seeded in six-well plates at 5 × 10⁵ cells per well, and then they were treated with rhIL-17A (0, 1, 10 or 50 ng/ml) for 24 h. Then the cells were harvested and homogenized for RNA extraction using TRIzol Reagent (TIANGEN BIOTECH Cat# DP430). cDNA was synthesized from the total RNA (2 µg) in a 20 µl reaction using a reverse transcription system (TIANGEN BIOTECH Cat# KR103-04). The primers for FABP4 and GAPDH are as follows: FABP4 (forward sequence, 5′-TACTGGGCCAGGAATTTGAC-3′; reverse sequence, 5′-GTGGAAGTGACGCCTTTCAT-3′); GAPDH (forward sequence, 5′-GAAGGTGAAGGTCGGAGTC-3′; reverse sequence, 5′-GAAGATGGTGATGGGATTTC-3′). The reactions were conducted according to the manufacturer’s protocol. For data analysis, the expression level of the FABP4 gene was normalized to the GAPDH gene. Relative gene expression was calculated with the $2^{-\mathrm{\Delta }{C}_t}$ formula.

Western blotting assay

Total protein was extracted from OvCa cells and murine OvCa tissues. The protein concentrations were determined with protein assay reagents (KeyGEN BioTECH Cat# KGP250). Equal amounts of proteins were separated by 10–13% SDS-PAGE, transferred onto PVDF membrane and incubated at 4 °C overnight with specific primary antibodies against CD36 (Cat# 18836-1-AP) from Proteintech and FABP4 (Cat# DF6035), STAT3 (Cat# AF6294), p-STAT3 (Tyr705,Cat# AF3293) and GAPDH (Cat# AF7021) from Affinity. The membranes were probed with corresponding secondary antibodies (Affinity Cat# S0001) and scanned by Odyssey^®CLx equipment (LI-COR Biosciences) to detect the bands. The density of the bands was detected by Odyssey software 3.0 (LI-COR Biosciences).

Cell proliferation assay

The OvCa cells were cultured in DMEM/Low, (Neuronbc Cat# 07-01F) containing 10% (v/v) FBS (Gibco, USA). To examine the effects of rhIL-17A and palmitic acid (PA, Sigma-Aldrich Cat# P9767) on cell proliferation, OVCAR3 and A2780 cells were seeded in 96-well plates at 3 × 10³ cells per well and were then treated with (1) 0, 0.1, 1, 10, 100 ng/ml rhIL-17A for 12, 24, 48 or 72 h; (2) 0, 12.5, 25, 50, 100 μM PA for 0, 24 or 48 h; (3) 10 ng/ml rhIL-17A and PA (0, 12.5, 25, 50, or 100 μM) for 48 h. To determine the signals involved in the effects of rhIL-17A and PA on cell proliferation, OVCAR3 and A2780 cells were pretreated with the FABP4 inhibitor BMS309403 (25 μM; Merck Cat# 341310) or the STAT3 inhibitor STATTIC (A2780, 0.3125 μM; OVCAR3, 1.25 μM; Selleck Cat# S7024) followed by 48 h of treatment with 10 ng/ml rhIL-17A and 25 μM PA. After treatments, cells were incubated with MTS (KeyGEN BioTECH Cat# KGA327) for an additional 2 h at 37 °C in the dark. The absorbance was determined at 490 nm (OD490) using an ELISA microplate reader (Thermo Fisher, Multiskan GO).

FA transfer assay and qualitative and quantitative oil red-staining assay

A2780 or OVCAR3 cells were seeded in 24-well plates and then treated with 10 ng/ml IL-17A, 25 μM PA, or 10 ng/ml IL-17A and 25 μM PA for 24 h. To determine the signals involved in the FA uptake of OvCa cells in the presence of rhIL-17A and PA, A2780 or OVCAR3 cells were pretreated with BMS309403 or STATTIC for 2 h. To determine whether the FA uptake by OvCa cells in the presence of rhIL-17A and PA was mediated by IL-17RA, IL-17RA-siRNA and control-siRNA were transfected into A2780 and OVCAR3 cells as previously described in detail [9]. Then the cells were incubated with 10% formaldehyde for 30 min. Subsequently, an oil-red O (KeyGEN BioTECH Cat# KGA329) working solution was added into each well to stain the cells for 30 min. After washing twice with washing solution (KeyGEN BioTECH Cat# KGA329), images of the stained cells were captured under a microscope (Olympus, DP80). Then absorbance at OD490 was measured using an ELISA microplate reader after dissolving oil-red O dye particles with 100% isopropanol.

Bodipy-FL-C16 capture assay

A2780, OVCAR3 and SKOV3 cells were seeded in six-well plates at 3 × 10⁵ cells per well and were pretreated with or without 25 μM BMS309403 for 2 h; then they were treated with or without 10 ng/ml rhIL-17A for 6 h. After treatment, cells were resuspended in 100 µl of PBS with or without 1 µM fatty acid Bodipy-FL-C16 (D-3821, Thermo Fisher), and then they were incubated at room temperature for 30 min. Unstained cells were used as control. Cells were washed twice in ice-cold PBS containing 2% FBS, which was followed each time by centrifugation at 450 g for 5 min. Fluorescence was measured by flow cytometry and analyzed using FlowJo software [7].

Establishment of orthotopic implantation models

To evaluate the in vivo effects of IL-17A on the growth and peritoneal metastasis of OvCa, the orthotopic implantation models were established. A cell suspension of 5 × 10⁶ ID8 cells (mouse ovarian epithelial papillary serous adenocarcinoma cell line) in 20 μl of PBS was prepared for the unilateral orthotopic/intrabursal injection into C57BL/6 wild-type (WT) mice and IL-17A^−/− mice (n = 6), as described by Denise C. Connolly’s report [28]. The mice were euthanized and killed for analysis 8 weeks after tumor cell implantation.

Immunohistochemistry (IHC) staining

IHC staining of tumors from human epithelial OvCa clinical specimens or murine OvCa samples was performed as previously described in detail [12]. Primary antibodies against IL-17A (Affinity Cat# DF6127) and FABP4 (Affinity Cat# DF6035) were used in staining, which was followed by microscopic analysis. The intensity of the IHC staining of clinical specimens was assessed as previously described in detail [12], using a four-step grading system, i.e., −, +, ++, +++, for negative, low, high, and very high, respectively (−: ≤ 5% positive staining; +: 5–20% positive staining; ++: 20–50% positive staining; and +++: ≥ 50% positive staining) [29]. For statistical analysis of IL-17A staining, we divided the cases into two groups: negative/low (−/+) expression and high/very high (++/+++) expression.

Statistical analysis

Statistical analysis was performed as previously described in detail [12]. A p value of less than 0.05 was considered statistically significant.

Results

rhIL-17A increased FABP4 expression in OvCa cells via STAT3 signaling

First, we evaluated the effect of rhIL-17A on the expression of FABP4 at the mRNA and protein levels. Our previous study [12] demonstrated that five commonly used human OvCa cell lines (A2780, OVCAR3, SKOV3, HO8910, and HO8910/PM) all expressed IL-17RA. Therefore, we performed dose–effect and time–effect experiments analyzing the effect of rhIL-17A on FABP4 expression in A2780 and OVCAR3 cell lines. As shown in Fig. 1a, b, treatment with rhIL-17A (0–50 ng/ml) for 6 h increased the mRNA level and protein expression of FABP4 in a dose-dependent manner, and the maximum effect of rhIL-17A occurred with 10 ng/ml and 50 ng/ml treatments. Consequently, a working concentration of 10 ng/ml for rhIL-17A treatment was applied in the following experiments. Meanwhile, a time–effect study indicated that the protein level of FABP4 was significantly increased by treatment with rhIL-17A for 3, 6, 12, or 24 h (A2780) and for 6, 12 or 24 h (OVCAR3) (Fig. 1c). Therefore, the rhIL-17A treatment time applied in the following experiments was 6 h, 12 h or 24 h.

Fig. 1.

rhIL-17A increased FABP4 expression in OvCa cells via STAT3 signaling. Dose–effect (a, b) and time–effect (c) experiments were performed in A2780 and OVCAR3 cells. a mRNA level of FABP4 after rhIL-17A treatment. b, c Protein expression of FABP4 after rhIL-17A treatment. d-(a) Protein expression of FABP4, p-STAT3 and STAT3 after rhIL-17A and/or STATTIC treatment (A2780: 0.3125 μM; OVCAR3: 1.25 μM). d-(b) The relative expression of proteins in d-(a). Three independent experiments were performed and a representative image is shown. Data represent the mean ± SD from three independent experiments. *p < 0.05, **p < 0.01

However, there was no significant change in protein expression of another FA transporter [4], CD36, after the treatment with rhIL-17A (sFig), indicating that IL-17A might enhance the FA uptake by OvCa cells through FABP4 rather than through CD36.

Signal transducer and activator of transcription 3 (STAT3) has been reported to be essential for tumor progression to advanced malignancy [30] and for tumor cell proliferation [31]. We investigated the mechanisms of rhIL-17A activity and focused our attention on STAT3 signaling. The results showed that rhIL-17A treatment could enhance the protein level of both p-STAT3 and FABP4, whereas pretreatment with STATTIC (a STAT3 inhibitor) could significantly block the upregulation of FABP4 expression caused by rhIL-17A (Fig. 1d). The above data suggest that the increased level of p-STAT3 was at least partially responsible for the impact of IL-17A on increased FABP4 expression.

rhIL-17A enhanced the proliferation of OvCa cells in the presence of PA via the IL-17A/IL-17RA/STAT3/FABP4 axis

First, palmitic acid (PA), a commonly used long-chain FA, was used as the FA source within the ARM to detect the effect of rhIL-17A on OvCa cell proliferation. As shown in Fig. 2a, individual treatment with rhIL-17A or PA had no significant effects on the proliferation of either OvCa cell line. When the PA concentration ranged from 25 μM to 50 μM (for A2780) or from 25 μM to 100 μM (for OVCAR3), the treatment with rhIL-17A and PA markedly increased cell proliferation compared to what was observed in the PA treatment alone. However, when compared with other control groups, the degree of change was quite small in OvCAR3 cells and not as obvious as it was in A2780 cells, even though the increases were statistically significant. When the PA concentration reached 100 μM for A2780 cells, rhIL-17A and PA treatment resulted in notably decreased cell proliferation compared to what was observed in the cells treated with only PA, indicating that rhIL-17A could enhance the proliferation of OvCa cells in the presence of PA within an appropriate dose range. Accordingly, 10 ng/ml of rhIL-17A and 25 μM of PA were applied in the following experiments.

Fig. 2.

rhIL-17A enhanced the proliferation of OvCa cells in the presence of PA via the IL-17A/IL-17RA/p-STAT3/FABP4 axis. a Proliferation assays of 2780 and OVCAR3 cells after treatment with rhIL-17A, PA, or rhIL-17A and PA for designated time periods. Cell proliferation was detected by MTS assay. b Proliferation assay for 2780 and OVCAR3 cells after pretreatment with BMS309403 or STATTIC for 2 h and then treatment with rhIL-17A, PA, or rhIL-17A and PA for 48 h. Data represent the mean ± SD from three independent experiments. *p < 0.05, **p < 0.01

Second, proliferation experiments were conducted to determine whether the effect of rhIL-17A was mediated by FABP4 via STAT3 signaling. As shown in Fig. 2b, the OD490 values of both A2780 and OVCAR3 cells that were treated with rhIL-17A and PA were obviously higher than they were in the cells treated with rhIL-17A or PA. Moreover, pretreatment with BMS309403 (a FABP4 inhibitor) or STATTIC could remarkably block the above effects. These results indicate that rhIL-17A might promote OvCa cell proliferation in the presence of PA to affect FABP4 expression, and this effect may be achieved at least partially through STAT3 pathway.

rhIL-17A increased the uptake of PA by OvCa cells via the IL-17A/IL-17RA/p-STAT3/FABP4 axis

FA transfer assays and qualitative and quantitative oil red-staining assays were performed to determine whether rhIL-17A could promote the uptake of FA by OvCa cells via the IL-17A/IL-17RA/p-STAT3/FABP4 axis. A2780 and OVCAR3 cells were transfected with control-siRNA or IL-17RA-siRNA as previously described in detail [12]. Oil-red quantitative results showed that after treatment with rhIL-17A and PA, the OD490 values (indicating the PA level inside the cells) of parental A2780 or OVCAR3 cells were notably higher than those from cells treated individually with rhIL-17A or PA (Fig. 3a). Moreover, pretreatment with BMS309403 or STATTIC remarkably blocked the effects. Furthermore, when rhIL-17A and PA were applied, the OD490 values of IL-17RA-siRNA A2780/OVCAR3 cells were significantly lower than those of control-siRNA cell lines (Fig. 3a). The morphological results of oil-red staining (Fig. 3b) were in line with the quantitative results. These data suggested that rhIL-17A might promote the uptake of PA in OvCa cells via the IL-17A/IL-17RA/p-STAT3/FABP4 axis.

Fig. 3.

rhIL-17A increased the uptake of PA in OvCa cells via the IL-17A/IL-17RA/p-STAT3/FABP4 axis. A2780 and OVCAR3 cells were treated with rhIL-17A, PA, rhIL-17A and PA, BMS309403 or STATTIC in parental cells or control/IL-17RA-siRNA cells. After treatments, oil-red O staining was performed. a Bar graph for OD490 values (indicating the PA level inside the cells) from each group. Data represent the mean ± SD from three independent experiments. *p < 0.05, **p < 0.01. b Representative photos of oil-red O staining for each group. Magnification, ×100. 1: local enlarged image of A2780 cells after treatment with rhIL-17A and PA (shown in inset); 2: local enlarged image of OVCAR3 cells after treatment with rhIL-17A and PA (shown in inset). c Bodipy-FL-C16 capture assay after pretreatment with BMS309403 for 2 h and then rhIL-17A for 6 h in A2780, OVCAR3 and SKOV3 cells. Three independent experiments were performed and a representative image is shown

The Bodipy-labeled fatty acid analogs (such as Bodipy-FL-C16) have been employed to study the cellular uptake and metabolism of lipids [32]. Therefore, to evaluate the effect of rhIL-17A on the ex vivo FA uptake by OvCa cells, a Bodipy-FL-C16 capture assay was performed [32]. To investigate the generality of the effect in OvCa cell lines, we added a third commonly used OvCa cell line: SKOV3 cells. The results showed that, compared with the control group, the fluorescence intensity in all three cell lines treated with rhIL-17A was increased (Fig. 3c). After treatment with BMS309403 and rhIL-17A, the fluorescence intensity was lower than the group treated with rhIL-17A, but it was still higher than the value from those cells treated individually with BMS309403. The results suggested that IL-17A treatment could promote the uptake of Bodipy-labeled fatty acid (Bodipy-FL-C16) by OvCa cells via the upregulation of FABP4.

Endogenous IL-17A increased the growth and metastasis of OvCa in the peritoneal cavity of a murine model

Orthotopic implantation may provide an ideal model that closely mimics the growth and metastasis of OvCa in the human body [28]. To examine the impact of IL-17A on the development of OvCa in vivo, an ID8 cell suspension was prepared for intrabursal injection into C57BL/6 WT mice and IL-17A^−/− mice. As shown in Fig. 4a, c, much larger tumor nodules (indicated by black arrows) developed in the ovarian tissue of WT mice injected with ID8 cells than what was observed in the IL-17A^−/− mice. Statistical analysis demonstrated a significant difference in the mean diameter of the tumor nodules in WT mice and IL-17A^−/− mice (WT mice: 0.62 ± 0.07 cm, IL-17A^−/− mice: 0.29 ± 0.03 cm; p = 0.022). Necropsy results revealed the widespread abdominal dissemination of tumor cells to the peritoneum in WT mice (Fig. 4b), including multiple large tumor nodules throughout the omentum (Fig. 4b-(a)), bowel (Fig. 4b-(b)), mesentery (Fig. 4b-(c)) and abdominal wall (Fig. 4b-(d)). Conversely, in IL-17A^−/− mice (Fig. 4d), fewer and smaller tumor masses developed in the peritoneal cavity (Fig. 4d-(a)–d-(d)). Figure 4e displays the significant difference in the number of tumor nodules between the two groups. These data indicated that IL-17A might fuel tumor growth and metastasis in the ARM. Western blotting analysis of protein extraction from tumor samples showed that the levels of FABP4 and p-STAT3 in the WT mouse group were strikingly higher than they were in the IL-17A^−/− mouse group (Fig. 4f). IHC staining for FABP4 expression revealed that the level of FABP4 in tumor samples from the WT mouse group was markedly higher than it was in the IL-17A^−/− mouse group (Fig. 4g). In brief, western blotting and IHC staining results suggested that the IL-17A level is positively related to FABP4 expression, and both contribute to the progression and metastasis of OvCa.

Fig. 4.

IL-17A increased the OvCa growth and metastasis in the peritoneal cavity of a murine model. ID8 cells (murine OvCa) were prepared for orthotopic/intrabursal injection into C57BL/6 WT and IL-17A^−/− mice. Eight weeks after injection, the mice were killed. a, c Representative photos of ovarian tissues in the ID8-injected side (indicated by black arrow). b, d Representative photos of tumor nodules distributed in the abdominal cavity. a Omentum. b Bowel. c Mesentery. d Abdominal wall. e Number of tumor nodules in WT and IL-17A^−/− mice. **p < 0.01. f Protein lysates were prepared from the tumor tissues of WT and IL-17A^−/− mice, and the protein levels of FABP4, p-STAT3 and STAT3 were analyzed by Western blotting. Three independent experiments were performed and a representative image is shown. Data represent mean ± SD from three independent experiments. **p < 0.01. g The sections were prepared from the tumor tissues of WT and IL-17A^−/− mice, and IHC staining was performed to determine FABP4 expression was performed. Three independent experiments were performed, and the representative image is shown

The level of IL-17A was positively correlated with FIGO staging and FABP4 expression in clinical OvCa settings

To verify whether there were correlations among IL-17A, FABP4 and OvCa progression in OvCa patients, sixty clinical OvCa specimens and ten normal ovary specimens were analyzed. First, we detected the status of IL-17A expression using a four-step grading system. The IHC staining results revealed that compared with normal ovary samples, the number of cells staining positively for IL-17A (IL-17A^+ve) slightly increased in FIGO stage I and II samples, and there was a marked increase in stage III and IV samples, with the percentages of highly positive IL-17A-stained specimens (++/+++) being 26.7% and 77.0%, respectively (Table 1). The data suggested that the IL-17A level remarkably increased as OvCa progressed based on FIGO staging (p = 0.004). Meanwhile, the percentages of highly IL-17A-stained specimens (++/+++) in OvCa samples with lymphatic or peritoneum metastasis were notably higher than they were in nonmetastatic samples (80.9% vs. 35.9% for lymphatic metastasis, and 73.7% vs. 41.5% for peritoneum metastasis, respectively). The IL-17A^+ve cells were generally located in the stroma of the tumor (Fig. 5a), indicating that in the OvCa microenvironment, IL-17A-producing cells were mainly macrophages (large irregularly shaped cells) and lymphocytes (small round cells). We also analyzed the expression of FABP4 in normal ovary tissues and OvCa specimens. In normal ovary tissues, FABP4 was either absent or expressed at a low level. As the disease progressed, the proportion of FABP4-positive staining (FABP4^+ve) OvCa cells increased (Fig. 5a), which is similar to the results in Fig. 4, suggesting that FABP4 is related to OvCa progression and metastasis. Figure 5b shows that the percentage of FABP4-positive cells in the IL-17A-negative/low (−/+) group was significantly lower than it was in the IL-17A-high/very high (++/+++) group (p < 0.05). Based on these results, we proved the positive correlation among the expression of IL-17A and FABP4 and OvCa progression in clinical specimens.

Table 1.

Relationship between IL-17A protein level and clinical outcomes in clinical ovarian cancer settings

	Number of cases	IL-17A (%)a		p value
		−/+	++/+++
Normal ovary	10	8 (80.0)	2 (20.0)
Epithelial ovarian cancer	60	29 (48.3)	31 (51.7)	0.004b
Age
≤ 50	24	14 (58.3)	10 (41.7)
> 50	36	15 (41.6)	21 (58.4)	0.296c
FIGO stage
I&II	30	22 (73.3)	8 (26.7)
III&IV	30	7 (23.0)	23 (77.0)	0.004c
Lymphatic metastasis
–	39	25 (64.1)	14 (35.9)
+	21	4 (19.1)	17 (80.9)	0.003c
Peritoneum metastasis
–	41	24 (58.5)	17 (41.5)
+	19	5 (26.3)	14 (73.7)	0.005

^aNumbers before parentheses are the number of cases

^bCompared with normal ovary group; Chi-square test

^cCompared between the sub-categories; Chi-square test

Fig. 5.

Relationship between the levels of IL-17A and FABP4 in clinical OvCa settings. a Expression of IL-17A and FABP4 in normal ovary tissues and OvCa specimens, as determined by IHC staining and FIGO staging. Magnification, ×400; scale bar, 25 μm. Blue box: Typical characteristics of IL-17A^+ve cells in an OvCa environment. 1. An IL-17A^+ve small round lymphocyte characterized by a small round nucleus (enlarged in inset); 2. An IL-17A^+ve large irregularly shaped macrophage with a kidney-shaped nucleus (shown in inset). b Percentage of FABP4 positivity in the IL-17A-negative/low (−/+) group and the IL-17A-high/very high (++/+++) group. The black bar represents the median of each group. A minimum of 10 fields per section was counted and analyzed. *p < 0.05

Discussion

Many studies have shown that FABP4, a major molecule responsible for FA uptake, plays critical roles in the progression of various cancers including prostate and ovarian cancer [3, 18], and these cancer cells incline to growth and metastasis in adipocyte-rich organs such as bone marrow and omentum. However, the molecular mechanisms performed by FABP4 in the ARM need to be further investigated. In the present study, we explored whether and how IL-17A, a vital cytokine in the ARM, influences the expression of FABP4 and impacts OvCa growth and/or metastasis.

In proliferation assays, although the blocking impact of the STAT3 inhibitor on the effect of IL-17A was statistically significant, it does not seem to completely block IL-17A, indicating that STAT3 signaling might be an important but not exclusive pathway that regulates IL-17A effects on FABP4 expression. There may exist other mechanisms that need to be further explored, e.g., IL-17 can induce the release of IL-8 by human renal cancer cells [33], whereas IL-8 upregulates FABP4 expression [3]; hence, the IL-17/IL-8/FABP4 axis might be another interesting possible pathway.

We utilized IL-17A^−/− mice as animal models to observe the effects of IL-17A on the tumor growth and metastasis of OvCa in vivo. It should be noted that the impacts on the tumor may also be partially caused by the immune responses that were mediated by the immune cells due to IL-17 deficiency in the mice, e.g., IL-17 could block the entry of cytotoxic CD8⁺ T cells into the tumor microenvironment and subsequently reduce the local immune responses towards the local tumor cells [34]. A seemingly preferable alternative method would be to inject IL-17RA^−/− tumor cells into the mice to eliminate the above interfering influences, yet Wang K et al. reported that in the study of colitis-associated cancer, ablation of IL-17RA decreases the expression of IL-6, a cytokine that contributes to the development of the tumor [16], suggesting that models with injected IL-17RA^−/− tumor cells also have shortcomings. In our in vivo studies, the protein expression levels (Fig. 4f) were strongly in line with the in vitro results, which further confirmed the existence of the IL-17A/IL-17RA/p-STAT3/FABP4 axis in the animal models. Therefore, even though the impacts caused by the immune responses due to IL-17 deficiency may exist in the mice, our in vitro and in vivo data still demonstrated the effects of IL-17A on FABP4 in OvCa, which may be at least an important partial mechanism influencing tumor growth.

In the present study, both experimental data using animals and clinical specimen analysis indicate that FABP4 correlates with OvCa progression and metastasis. These results coincide with other studies [3, 21, 35], indicating the crucial role of FABP4 in OvCa progression and metastasis, which may be linked to (1) FABP4 expression in adipocytes and adjacent tumor cells being closely related to “scavenging” of adipocyte-derived FA in tumor cells; (2) FABP4 requirement for angiogenesis especially in tumors that are embedded in FA-rich tissues [20]; (3) high expression of FABP4 being involved in the regulation of multiple metabolites and pathways that contribute to the metastasis of cancer. Hence, studies on the regulation of FABP4 have become extremely important in cancer research. In the present study, we identified that IL-17A increased FABP4 expression in OvCa cells partially via STAT3 signaling. Thus, we provide a new understanding of the FABP4 regulation mechanism, which proposes a potential synergistic therapeutic target to interfere with OvCa progression and metastasis.

It is worth mentioning that in the OvCa TME, FABP4^+ve cells mainly included adipocytes and cancer cells, and the levels of FABP4 in these cells from the WT mice group were significantly higher than they were in the cells from the IL-17A^−/− mice (Fig. 4g), indicating that IL-17A could upregulate the expression of FABP4 not only in OvCa cells but also in adipocytes. The Mita group reported that FABP4 is expressed in adipocytes, and elevated plasma FABP4 levels are associated with an obesity-mediated metabolic phenotype [36]. Meanwhile, obesity has been reported to be associated with increased IL-17A production in humans [37]. Therefore, we speculated that IL-17A might also increase the FABP4 level in adipocytes within the ARM, which could facilitate FA exportation from the adipocytes into the ARM. The effects and the corresponding mechanisms need to be further investigated.

In conclusion, we established a model (Fig. 6) to illustrate a novel mode of the interactions among adipocytes, OvCa cells, and a certain mediator (such as pro-inflammatory IL-17A) in the ARM. Our findings may provide potential therapeutic targets for clinical intervention of the growth and metastasis of the tumors (such as OvCa) which are prone to metastasizing to adipocyte-rich sites (such as omentum). Our study may also expand the knowledge of immunomodulatory strategies in improving the efficacy of tumor therapies.

Fig. 6.

Proposed model for the mechanism by which IL-17A links adipocytes with OvCa cells in the ARM. In the ARM, IL-17A-producing cells secrete IL-17A, which upregulates FABP4 expression via p-STAT3 signaling. Meanwhile, adipocytes provide FAs, which are transported by FABP4 and are then utilized for ATP production by β-oxidization; subsequently, OvCa cell proliferation will be increased

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

ARM

Adipocyte-rich microenvironment

CAA

Cancer-associated adipocyte

DLL4-NOTCH

Delta-like 4-NOTCH

FA

Fatty acid

FABP4

Fatty acid-binding protein 4

FABP4^+ve

FABP4-positive staining

FIGO

International Federation of Gynecology and Obstetrics

GAPDH

Glyceraldehyde-3-phosphate dehydrogenase

IL-17A^−/−

IL-17A deficient

IL-17A^+ve

IL-17A-positive staining

IL-17R

Interleukin 17 receptor

ILC

Innate lymphoid cell

OvCa

Ovarian cancer

PA

Palmitic acid

PPARγ

Peroxisome proliferator-activated receptor gamma

p-STAT3

Phosphorylated STAT3

PVDF

Polyvinylidene fluoride

qRT-PCR

Quantitative RT-PCR

rhIL-17A

Recombinant human IL-17A

TME

Tumor microenvironment

WAT

White adipose tissue

WT

Wild type

Author contributions

CYY, XLN, ZY and WMD contributed to the conception and design of the study. XLN, YRD, YC, LLX, XXM, and YL contributed to the development of methodologies and acquisition of data. XLN, YRD, YC, XML, LLX, XXM and YL contributed to the analysis and interpretation of data. YI contributed to the establishment of the mouse model. CYY, XLN, and WMD wrote the manuscript. All authors revised the manuscript critically and approved the final manuscript version.

Funding

This project has been funded in part by grants from the Natural Science Foundation of China (30670801 to Weimin Deng.), the Tianjin Natural Science Foundation (18YFZCSY00040 and 15JCYBJC26000 to Weimin Deng.) and the Key projects of Tianjin Municipal Education Commission (2016YD01 to Weimin Deng).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflicts of interest.

Ethical approval and ethical standards

All procedures performed in studies involving human participants were in accordance with the ethical standards of Tianjin Central Hospital of Gynecology Obstetrics and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards (Ethical approval number 2017KY009). All experiments with C57BL/6 mice were performed in compliance with the principles and procedures approved by the Animal Ethical and Welfare Committee (AEWC) of Tianjin Medical University (Protocol number SYXK 2016-0012).

Informed consent

Informed oral consent was obtained from all individual participants included in the study. Patients consented to the use of their tumor material and clinical data for research and publication. Healthy donors also consented to the use of their tissue for research and publication.

Animal source

C57BL/6 wild-type (WT) mice were purchased from the Laboratory Animal Center of the Academy of Military Medical Sciences (Beijing, China). IL-17A^−/− C57BL/6 mice were kindly provided by Hong Zhou (Nanjing Medical University, China) with permission from Yoichiro Iwakura (Tokyo University of Science, Chiba, Japan).

Cell line authentication

Authentication is not applicable since the human OvCa cell line A2780 was purchased from KeyGEN BioTECH (Cat# KG039, Nanjing, China), OVCAR3 and SKOV3 cells were gifts from Zhi Yao (Tianjin Medical University, Tianjin, China), and the mouse ovarian epithelial papillary serous adenocarcinoma cell line ID8 was a gift from Luyuan Li (Nankai University, Tianjin, China). Hence, no authentication information was obtained.