Abstract

Background

   As a widespread epidemic, obesity poses a significant risk to health
   and leads to physiological abnormalities, including diabetes mellitus
   and inflammation. Obesity-induced inflammation can accelerate the
   development of various cancers; however, the role of obesity in the
   migration of ovarian carcinoma is still unclear.

Results

   Twenty-four commonly upregulated genes were identified from single-cell
   RNA sequencing datasets of both ovarian carcinoma and adipose tissue of
   obese humans, with the chemokine CXCL10 showing a significant increase
   in adipose tissues associated with obesity. And CXCL10 treated primed
   macrophages exhibit both direct and indirect effects on the
   proliferation, apoptosis, migration, and invasion of ovarian
   adenocarcinoma cells. The treatment of CXCL10 on the SKOV3 cells
   enhances FAK expression and phosphorylation, thereby accelerating the
   migration and invasion of ovarian cancer cells. Conditioned
   medium-derived from CXCL10-treated THP-1 cells significantly promoted
   ovarian cancer cell migration and invasion, which may be attributed to
   the increased expression of C1QA, C1QC, CCL24, and IL4R in macrophages.

Conclusions

   Obesity exacerbates the production of CXCL10 from adipose tissues in
   obese women. CXCL10 is a key hub factor between developments of ovarian
   cancer and adipose tissues in obese. Targeting adipose-derived CXCL10
   or its downstream macrophages may be a potential strategy to alleviate
   ovarian cancer accompanied by obesity.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s13048-024-01568-0.

   Keywords: CXCL10, Ovarian carcinoma, Obesity, Adipose tissue,
   Macrophage

Introduction

   Obesity with excessive fat accumulation presents a risk to health and
   increases incidence of obesity-related complications. Excess nutrient
   intake can result in the expansion of adipose tissue to store energy
   and augmented adipose tissue induced inflammation, which is one of the
   features of obesity [[38]1]. Adipose tissue in obesity can enhance the
   production of proinflammatory cytokines (such as IL-1, IL-6, and TNF)
   and chemokines from adipocytes and immune cells [[39]2]. The
   chemokines, including CXCL1, CXCL5, CXCL10, and CXCL11, are upregulated
   in the adipose tissue of obesity [[40]3, [41]4], and the expression of
   chemokine receptors, including CCR1, CCR2, CCR3, and CCR5, is also
   increased in the omental and subcutaneous adipose tissues of obesity
   [[42]5, [43]6]. Moreover, inflammation is thought to be linked to
   obesity, insulin resistance, metabolic syndrome, and even cancer
   [[44]7, [45]8].

   Several studies have demonstrated a positive correlation between
   obesity, as measured by BMI, and cancer incidence [[46]9–[47]12]. As
   the most fatal gynecologic cancer, ovarian cancer is an important
   source of cancer-related mortality worldwide. Most patients with
   ovarian cancer ultimately succumb to abdominal metastatic tumors rather
   than the primary carcinoma in situ [[48]13]. The spread of ovarian
   cancer cells to abdominal tissues commonly occurs via peritoneal fluid
   [[49]14]. Thus, the attachment between ovarian cancer cells and the
   omentum is the initial step for the further formation of peritoneal
   disease [[50]15, [51]16]. Although the relationships between omentum
   metastasis and ovarian cancer were widely acknowledged, the role of
   obesity in the metastasis of ovarian carcinoma remains unclear.

   Chemokines, classified into four subfamilies—CXC, CC, (X)C, and CX3C
   based on their structure—play a pivotal role in immune regulation and
   tumor development [[52]17]. The chemokine itself can regulate tumor
   cell growth, invasiveness, metastasis, and angiogenesis
   [[53]18–[54]20]. Moreover, the chemokine can also recruit anti- or
   pro-tumor immune cells to the tumor microenvironment and affect tumor
   progression indirectly [[55]21, [56]22]. For instance, CCL5 has been
   shown to recruit Treg cells to the tumor, facilitating tumor metastasis
   in ovarian cancer [[57]20, [58]23]. CXCL10 and the receptor CXCR3 were
   overexpressed and associated with poor survival outcomes in pancreatic
   cancer [[59]24, [60]25]. Obesity is characterized by the enormous
   production of proinflammatory cytokines and chemokines from adipocytes
   [[61]26]. While epidemiologic studies revealed that the prognosis of
   ovarian cancer is significantly worse in obese women compared to those
   with a normal BMI [[62]27–[63]29]. Thus, it is of significant interest
   to investigate the roles and mechanisms of adipocyte-derived chemokines
   in the process of ovarian cancer, under the condition of obesity.

   In this study, we integrated datasets of bulk RNA-sequencing or single
   cell RNA-sequencing in adipose tissues and ovarian cancer cells to
   discover common DEGs (differentially expressed genes) between obesity
   and ovarian cancer, and dissect the effects and underlying mechanisms
   of the chemokines produced by the adipose tissues on the biological
   characteristics of ovarian cancer cells. Our study may help identify a
   specific target for treating ovarian cancer occurred under the setting
   of obesity.

Materials and methods

Differentially expressed gene analysis in human and mice adipose tissues

   The datasets of [64]GSE152991 [[65]30], [66]GSE110729, and
   [67]GSE132885 [[68]31] were used to analyze the DEGs in white adipose
   tissue between obese and normal mice based on the R package DESeq2
   [[69]32]. The differentially expressed genes (OB_DEG) were identified
   as a P-value < 0.05 and log2FC > 1 between obese and normal mice. For
   [70]GSE152991, all samples in the groups of metabolically-healthy lean
   (MHL), metabolically-healthy obese (MHO), and metabolically-unhealthy
   obese (MUO) were included in the analyses, among which MHL was
   considered a control group [[71]30]. For [72]GSE110729, samples with a
   BMI of 20–23 were categorized as the control group, while samples with
   a BMI of 47–68 were classified as the obese group. For [73]GSE132885,
   subcutaneous (or visceral) adipose tissues from C57BL/6 mice fed a
   standard chow diet (SCD) for 8 or 20 weeks were classified as the
   control groups, while tissues from mice fed a high-fat diet (HFD) were
   classified as the obese group.

Analysis of single-cell RNA sequencing data in white adipose between SCD and
HFD mice

   To illustrate the potential relationship between obesity and ovarian
   carcinoma at the single-cell level, we analyzed the datasets of
   [74]GSE176171 [mouse (C57/B6J) inguinal white adipose from
   [75]GSM5820692 (HFD), [76]GSM5820693 (SCD), [77]GSM5820694 (HFD), and
   [78]GSM5820695 (SCD); mouse peri-ovarian white adipose from
   [79]GSM5820698 (HFD), [80]GSM5820699 (SCD), [81]GSM5820700 (HFD),
   [82]GSM5820701 (SCD)] following the typical pipeline of Seurat 4.4.0
   [[83]33]. Briefly, the matrix was normalized by the NormalizeData
   function, and the top 2000 highly variable genes were identified with
   the FindVariableGenes function. The tSNE and UMAP dimensionality
   reductions were used to project these populations in two dimensions,
   respectively. The cell types were annotated based on the published
   paper [[84]34] and known markers [mesothelial cell: Upk3b; adipose stem
   and progenitor cell (ASPC): Pdgfra; adipocytes (APC): Adipoq;
   endometrial cell: Prlr; epithelial cell: Erbb4; endothelial cell:
   Pecam1; myeloid cell: Mafb; B cell: Ms4a1; T cells: Il7r; smooth muscle
   cell: Acta2]. The FeaturePlot function in R package Seurat (v4.4.0) and
   the function DotPlot in R package ggplot2 (v3.4.4) were used to
   visualize the gene expression.

Differentially expressed gene analyses between ovarian carcinoma and
para-carcinoma tissues

   The GEPIA website ([85]http://gepia.cancer-pku.cn/) was utilized to
   analyze the transcriptome data of ovarian carcinoma (n = 426) and the
   paired normal tissues (n = 88) in humans, and the DEGs were identified
   as a P-value < 0.05 and log2FC > 1 between ovarian carcinoma and
   para-carcinoma tissues.

Cell culture

   All the ovarian cancer cell lines (SKOV3 and HO-8910) were cultured
   with RPMI 1640 medium (R5886, Sigma-Aldrich, UK) supplemented with 10%
   fetal calf serum (F0193, Sigma-Aldrich, UK) and incubated at 37 °C and
   5% CO[2]. The THP-1 monocytes (human leukemia monocytic cell line) were
   treated with PMA (100 ng/ml; P8139, Sigma-Aldrich) for 48 h to
   differentiate into macrophages.

   For the preparation of the conditional medium from the macrophages,
   macrophages derived from THP-1 cells were treated with recombinant
   human CXCL10 (200 ng/mL; HY-P7226, MedChemExpress, USA) dissolved in
   ddH2O at a stock concentration of 200 μg/mL, or with ddH2O (control)
   for 24 h. Following this treatment, the CXCL10-containing medium was
   discarded, and the macrophages were incubated with fresh culture medium
   for an additional 48 h. The medium collected after this 48-h period was
   used as the conditioned medium for treating ovarian cancer cells.

Cell counting Kit-8 (CCK-8)

   For the cellular proliferation assays, ovarian cancer cells were plated
   in a 96-well plate (2 × 10^3 cells/well) and treated with or without
   the addition of CXCL10 (200 ng/mL) or the CM from the CXCL10-treated
   macrophages, as described previously. After 24 h and 48 h, the cells
   were incubated with the CCK-8 solution (C0005, TargetMol, USA) for 2 h.
   The absorbance values were measured by a multifunctional microplate
   reader (Cytation 5, BioTek, USA) at a wavelength of 450 nm. All of the
   experiments were repeated at least 3 times.

Transwell invasion assay

   The cell invasion assay was performed with a transwell chamber (3422,
   Corning, USA). Matrigel (C0372, Beyotime, China) was diluted in
   serum-free medium at a 1:6 ratio, then applied to the upper surface of
   the transwell and incubated at 37 °C for 2 h, following the
   manufacturer’s instructions. After the matrigel had set, the transwells
   were placed into the 24-well plate with 200 μL of the cell suspension
   (5 × 10^4 cells) on top of the matrigel layer. We added CXCL10
   (200 ng/mL) or CM (from CXCL10-treated macrophages) to the 24-well
   plate, which in the underside of the transwell. After 12 h of
   incubation at 37 °C with 5% CO[2], the cells in transwells were fixed
   with 4% paraformaldehyde fix solution (P0099, Beyotime), stained with
   crystal violet staining solution (C0121, Beyotime) and removed the
   matrigel in the interior part of the transwell before visualization. At
   least three replications were conducted.

Wound-healing assay

   The ovarian cancer cells were seeded in a 24-well plate (5 × 10^4 cells
   per well) for 12 h and mechanically wound. The serum-free medium
   containing CXCL10 (200 ng/mL), or CM from macrophages treated with or
   without CXCL10 was added into the wells. The images of the wounded
   areas were captured under the light microscope at 0 h and 10 h after
   the treatment. The area of the wound was quantified with Image J
   software ([86]http://rsb.info.nih.gov). At least three replications
   were conducted.

Apoptosis assay

   The cells were treated with CXCL10 (direct) or CM from THP-1 (indirect)
   for 24 h. The apoptotic cells were detected with the Annexin V-FITC
   apoptosis detection kit (C1062L, Beyotime), following the
   manufacturer’s instructions. A CytoFLEX Flow Cytometer (Beckman
   Coulter, USA) was used to examine the apoptotic rate of cells in each
   group.

Colony-formation assay

   The SKOV3 cells were seeded in a 6-well plate (1 × 10^3 cells per
   well), and were treated with CXCL10 (200 ng/mL) for 48 h. The culture
   medium containing 200 ng/mL of CXCL10 was replaced every two days.
   After 12 days, the cells were washed with PBS and stained with the
   crystal violet staining solution (C0121, Beyotime) for visualization
   under the light microscope.

Western blot

   The SKOV3 cells treated with CXCL10 (200 ng/mL) were harvested and
   lysed in the RIPA lysis buffer (P1045, Beyotime) containing protease
   inhibitors and phosphatase inhibitors (C0001 and C0004, TargetMol,
   USA). The processes of protein extraction and western blot were
   consistent with our previous protocol [[87]35]. The following primary
   antibodies and dilutions were used: FAK (1:1000; 1700–1, Epitomics,
   USA), phospho-tyrosine (Y) 397-FAK (1:1000; 3283S, Cell Signaling
   Technology, USA), phospho-Y576/577-FAK (1:1000; ab76244, Abcam, UK),
   and GAPDH (1:1000; ab9485, Abcam). The secondary antibodies were
   horseradish peroxidase–conjugated anti-rabbit or anti-mouse IgG
   (1:5000; Jackson, USA). The chemiluminescent signal for each target
   protein was quantified by Image J software
   ([88]http://rsb.info.nih.gov). Phosphorylated protein levels were
   normalized to their respective total protein levels, and total protein
   intensities were normalized to GAPDH. All the raw gel figures are
   supply as the supplementary materials.

RNA sequencing and pathway enrichment analysis

   The macrophages derived from THP-1 cells were treated with CXCL10
   (200 ng/mL) or control solution (ddH[2]O) for 24 h, as mentioned
   previously. Then the total RNA was isolated with RNAiso Plus reagent
   (9108; Takara, Japan) according to the manufacturer’s instructions.
   After RNA purity and integrity were assessed, 2 μg of total RNA per
   sample were reverse transcribed for cDNA library construction. The cDNA
   fragments were sequenced on an Illumina HiSeq platform (Illumina). TPM
   (transcripts per kilobase million) was used to compare the expression
   of genes between CXCL10-treated macrophages and the control
   macrophages. A gene was considered a DEG when it met the following
   criteria: a twofold cut-off change (log2-fold change > 1) in transcript
   levels and q (adjusted p) value < 0.05. GO analysis and KEGG datasets
   were used to calculate the significance of the enrichment of DEGs in
   each pathway. The RNA sequencing data were deposited in the NCBI’s Gene
   Expression Omnibus database (GSE253899).

Statistical analysis

   All data were analyzed using the R software (version 4.2.1). The
   unpaired Student’s t-test (T.test in R) was used compare the
   differences between two groups. All data are shown as the
   mean ± standard deviation (SD). Statistically significant differences
   were considered at a P value < 0.05.

Results

Dysregulated immune chemokines are present in both obesity and ovarian
carcinoma

   To investigate the potential relationship between obesity and ovarian
   carcinoma, we analyzed the differentially expressed genes (DEGs)
   (P < 0.05 and log2FC > 1) in adipose tissues associated with obesity
   and in ovarian carcinoma, respectively. First, RNA-seq datasets of
   subcutaneous adipose tissue from metabolically-healthy lean (MHL)
   individuals were compared with those of metabolically-unhealthy obese
   (MUO) and metabolically-healthy obese (MHO) individuals, respectively,
   to obtain the DEGs of adipose tissues. Additionally, DEGs of the long
   non-coding RNAs (lncRNAs) profiling of the adipose tissues were
   identified from the dataset [89]GSE110729. Finally, we intersected the
   DEGs associated with obesity (from [90]GSE152991 and [91]GSE110729)
   with those associated with ovarian carcinoma (from TCGA database),
   identifying 127 positively correlated genes (Fig. [92]1A). And GO
   enrichment shows these 127 genes are related to immune cell activation
   and immune responses (Fig. [93]1B), suggesting that immunity and immune
   activation play roles in the interaction between adipose tissue and
   ovarian carcinoma.

Fig. 1.

   [94]Fig. 1
   [95]Open in a new tab

   The dysregulated genes in both obese adipose tissues from humans and
   mice and ovarian carcinoma. A The volcano plot showed differentially
   expressed genes (P-value < 0.05 and logFC > 1) in obese or ovarian
   carcinoma compared to their controls. GSE152991_MUO_MHL showed the
   differentially expressed genes in white adipose tissue between
   metabolically unhealthy obese (MUO) and metabolically healthy lean
   (MHL). GSE152991_MHO_MHL showed the differentially expressed genes in
   white adipose tissue between metabolically healthy obese (MHO) and
   metabolically healthy lean (MHL). GSE110729_OB indicated differentially
   expressed genes in samples with a BMI of 47–68 compared to those with a
   BMI of 20–23. TCGA_OV indicated differentially expressed genes in
   ovarian carcinoma compared to normal samples. The Venn plots indicated
   the shared up-regulated genes in both obese and ovarian carcinomas in
   humans. B The GO plot shows the GO enrichment analysis for the 127
   shared genes. C, D The expression of 24 shared up-regulated genes in
   both obese and ovarian carcinoma from both humans and mice

   Given the complex factors associated with progressive obesity, we
   further analyzed the datasets of the adipose tissues from mice fed with
   a high-fat diet (HFD) or standard-chow diet (SCD) for 8 weeks
   ([96]GSE132885). By intersecting these datasets, we identified 24
   overlapping DEGs (Fig. [97]1C). Among these, the chemokines CXCL10 and
   CCL3, which are involved in inflammation, stand out as potential key
   linkages between obesity and ovarian carcinoma (Fig. [98]1D). These
   findings suggest that inflammation induced by adipose tissue in obesity
   may play a crucial role in the development of ovarian carcinoma.

The enhanced CXCL10 in white adipocyte of mouse HFD model

   To determine the origin of chemokines in adipose tissue, we further
   analyzed the expression of CXCL10 and CCL3 in white adipose tissues
   from the peri-ovarian fat pad and inguinal fat pad of HFD and SCD mice,
   based on single-cell RNA sequencing datasets. For the peri-ovarian fat
   pad, we clustered all cells into eight cell types: smooth muscle cells
   (SMC), NK and T cells (T_NK), B cells (B), myeloid cells (Mye),
   endothelial cells (Endo), adipocytes (APC), adipose stem and progenitor
   cells (ASPC), and mesothelial cells (Mesothelial; Fig. [99]2A-B).
   CXCL10 was specifically expressed in APCs, while CCL3 showed minimal
   expression across all cell types (Fig. S1A). Additionally, the
   expression of CXCL10 was significantly higher in the adipocytes of HFD
   mice compared to those of SCD mice (P < 0.05; Fig. [100]2C), which is
   consistent with the enhanced expression of CXCL10 in the adipose tissue
   of obese mice.

Fig. 2.

   [101]Fig. 2
   [102]Open in a new tab

   The expressions of Cxcl10 in white adipose from high-fat diet or
   standard-chow diet mice. A Cell types of mouse peri-ovarian white
   adipose from single-cell datasets of high-fat diet (HFD) and
   standard-chow diet (SCD). B The specific markers for each cell type of
   the mouse peri-ovarian white adipose. C The expression and the
   expressed ratio of CXCL10 in peri-ovarian white adipose from HFD and
   SCD mice. D Cell types of mice inguinal white adipose from single-cell
   datasets of HFD and SCD. E The specific markers for each cell type. F
   CXCL10 expression and ratio in inguinal white adipose from HFD and SCD
   mice

   We also analyzed the datasets from the inguinal fat pads of HFD and SCD
   mice ([103]GSE176171). 37,994 single-cell transcriptome data in total
   were obtained and classified into 10 cell clusters [mesothelial cell
   (Mesothelial), adipose stem and progenitor cell (ASPC), adipocytes
   (APC), endometrial cell (Endometrial), epithelial cell (Epi),
   endothelial cell (Endo), myeloid cell (Mye), B cell, T cells, and
   smooth muscle cell (SMC)] with the specific cell markers
   (Fig. [104]2D-E). Consistent with the results from the peri-ovarian fat
   pad, CCL3 showed very low expression, but CXCL10 showed specifically
   abundant expression in APCs (Fig. S1B). In addition, the expression
   levels of CXCL10 were also increased in APCs from HFD mice compared to
   those from SCD mice (P < 0.05; Fig. [105]2F). These results indicate
   that CXCL10 expression in APCs is increased under the status of
   obesity.

CXCL10 accelerates the migration in ovarian cancer cells

   The expression of CXCL10 and its receptor, CXCR3, was significantly
   elevated in ovarian cancer, as shown by data from the GEPIA database,
   indicating a correlation between CXCL10 and ovarian cancer
   (Fig. [106]3A). To confirm the direct effects of CXCL10 on ovarian
   cancer cells, we conducted cellular assays to evaluate its impact on
   the proliferation, migration, and invasion of SKOV3 and HO-8910 cells.
   As shown in Fig. [107]3B and Fig. S2A, CXCL10 significantly increased
   the migration rate of SKOV3 and HO-8910 cells in the wound healing
   assay. Similarly, CXCL10 treatment markedly enhanced the invasion
   ability of SKOV3 and HO-8910 cells (Fig. [108]3C, Fig. S2B). However,
   there was no significant effect on proliferation, apoptosis, or colony
   formation of SKOV3 cells between the groups with or without CXCL10
   (Fig. [109]3D-3E). For HO8910 cells, CXCL10 partially increased cell
   viability but had no effect on cell apoptosis (Fig. S2C-2D). These
   results suggest that CXCL10 may directly promote the migration of
   ovarian cancer cells without significantly affecting cell proliferation
   or apoptosis.

Fig. 3.

   [110]Fig. 3
   [111]Open in a new tab

   The effect of CXCL10 on migration, invasion, colony-formation,
   apoptosis, and proliferation of SKOV3 cells. A The expression of CXCR3
   and CXCL10 in the ovarian carcinoma from the GEPIA database. B The
   significantly enhanced of migration in the SKOV3 cell line with the
   addition of CXCL10 (P < 0.001). C The significantly enhanced invasion
   in the SKOV3 cell line with the addition of CXCL10 (P < 0.001). D The
   apoptosis of SKOV3 cells treated with or without CXCL10. E, F The
   proliferation includes CCK8 assay (E) and colony-formation assay (F) of
   SKOV3 cells treated with or without CXCL10 (P > 0.05). Bar plots show
   the mean ± SEM. *** indicates a P < 0.001

CXCL10 increases the phosphorylation of FAK in SKOV3 cells

   Given the important roles of FAK signaling in tumor cell adhesion and
   migration, we further elucidate the possible mechanisms underlying the
   effect of CXCL10 on the migration of SKOV3 cells. CXCL10 significantly
   increased protein levels of FAK and phospho-FAK (Tyr576/577) in SKOV3
   cells (Fig. [112]4), but it did not significantly affect the
   autophosphorylation of FAK at the tyrosine 397 site. These findings
   suggest that CXCL10 may enhance the migration of ovarian cancer cells
   by increasing FAK expression and activating the phosphorylation of FAK.

Fig. 4.

   [113]Fig. 4
   [114]Open in a new tab

   CXCL10 enhanced the expression and activation of FAK in SKOV3 cell
   lines. A The expression and phosphorylation of FAK in the SKOV3 cells
   treated with or without CXCL10. B, C The quantification of the relative
   expression of FAK in the western blot. Bar plots show the mean ± SEM. *
   indicates a P < 0.05

CXCL10 accelerates migration and invasion of ovarian cancer cells through
immune microenvironment

   Since CXCL10 is a key chemokine that recruits immune cells to the tumor
   microenvironment and influences tumor progression, we further
   investigated its effects by treating SKOV3 and HO-8910 cells with
   conditioned medium generated from CXCL10-treated macrophages derived
   from the THP-1 cell. Interestingly, the conditional medium from
   CXCL10-treated macrophages significantly increased the migration rate
   and invasive ability of these two ovarian cells compared to the medium
   from NC-treated macrophages (Fig. [115]5A, 5C; Fig. S2E-2F). However,
   there were no significant differences in proliferation or apoptosis in
   ovarian cancer cells upon the addition of the conditioned medium
   (Fig. [116]5B, 5D-E; Fig. S2G-2H). Beyond its direct effects, CXCL10
   may also indirectly influence the invasion and migration ability of
   ovarian cancer cells by priming and activating macrophages within the
   tumor microenvironment.

Fig. 5.

   [117]Fig. 5
   [118]Open in a new tab

   The effects of conditional medium from CXCL10-treated macrophages on
   migration, invasion, proliferation, and apoptosis of SKOV3 cells. A The
   conditional medium from CXCL10-treated macrophages (THP-1 cell)
   enhanced the invasion of SKOV3 cells. B The apoptosis of SKOV3 cells
   treated with the conditional medium from CXCL10-treated macrophages. C,
   D The migration of SKOV3 cells treated with the conditional medium from
   CXCL10-treated macrophages. E The proliferation of SKOV3 cells treated
   with conditional medium from CXCL10-treated macrophages. *** indicates
   a P < 0.001

The increased C1Q + macrophages induced by CXCL10 may accelerate migration of
ovarian cancer cells

   To explore the mechanisms by which CXCL10 primes macrophages, we
   performed RNA sequencing analysis between the macrophages treated with
   CXCL10 and control. As shown in Fig. [119]6A, CXCL10 treatment did not
   significantly differentiate the macrophages to M1 or M2 polarization.
   However, markers for tumor-associated macrophages (TAM), i.e., C1QA,
   C1QC, IL4R, and CCL24, were markedly upregulated in CXCL10-treated
   macrophages compared to controls.

Fig. 6.

   [120]Fig. 6
   [121]Open in a new tab

   The increased expression of C1Qs in CXCL10-treated macrophages and
   macrophages from adipose tissue. A The addition of CXCL10 increased the
   expression of C1QA and C1QC in macrophages. + indicates the
   up-regulation in CXCL10-added macrophages with the fold
   change > 1.2;—indicates the down-regulation in CXCL10-added macrophages
   with the fold change > 1.2. B The sub-clusters in macrophages. C-H The
   increased expression ratios of C1qa (P value < 0.05), C1qb (P
   value < 0.01), C1qc (P value < 0.05), Il4ra (P value < 0.05), Il4 (P
   value < 0.05), and Ccl24 (P value = 0.102) in macrophages from HFD mice
   compared to those from SCD mice

   To further confirm the characteristics of these macrophages and their
   interactions with adipose tissue, we analyzed gene expression patterns
   in macrophages from the white adipose tissues of HFD and SCD mice.
   Interestingly, the expression levels of C1qa (P < 0.05), C1qb
   (P < 0.01), C1qc (P < 0.05), Il4ra (P < 0.05), Il4 (P < 0.05), and
   Ccl24 (P = 0.102) were all significantly elevated in macrophages from
   the white adipose tissues of HFD mice (Fig. [122]6B-H). Combined with
   our previous finding of abundant CXCL10 expression in APCs
   (Fig. [123]2B), these results suggest that elevated CXCL10 levels in
   APCs may increase the proportion of C1Q + macrophages and enhance CCL24
   production, playing a critical role in regulating the tumor
   microenvironment and promoting ovarian cancer metastasis.

Discussion

   Due to drastic lifestyle changes, particularly the increased
   consumption of high-fat diets and decreased physical activity,
   overweight or obesity has become a worldwide epidemic problem for human
   beings and poses a significant risk to human health [[124]36],
   including diabetes mellitus, cardiovascular disease, infertility, and
   even the development of some cancers [[125]37, [126]38].
   Obesity-induced inflammation may increase cancer risk through multiple
   mechanisms [[127]7, [128]39], and validating the shared genes that are
   simultaneously dysregulated in both obesity and tumors is a promising
   way to reveal the interaction between obesity and cancer. In this
   study, we first verified 24 common genes that were synchronously
   up-regulated in both ovarian cancer cells from obese women and adipose
   tissues from HFD-induced obese mice, based on the public RNA sequencing
   datasets. Given the potential functions of chemokines in immune and
   inflammatory response, which play an important role in both obesity and
   cancer progression, CXCL10 and CCL3 drew our attention. Moreover, we
   further verified that CXCL10, but not CCL3 was abundantly expressed in
   the adipocyte at the single-cell level in both the peri-ovarian fat pad
   and the inguinal fat pad. CXCL10 manifested a significant increase in
   adipocytes from HFD mice compared to those from SCD mice, leading us to
   focus on CXCL10 in the following study.

   CXCL10 showed opposite effects on the development of tumors mediated by
   different mechanisms [[129]40–[130]48]. For example, the CXCL10
   produced by tumors can recruit CXCR3 + T cells and NK cells to suppress
   tumors [[131]41, [132]45, [133]47, [134]48]. However, CXCL10 can also
   increase the metastasis of CXCR3 + tumor cells [[135]43, [136]44,
   [137]49–[138]51]. In this study, we found that CXCL10 can increase the
   expression and phosphorylation of FAK (Y576/577) and improve the
   migration and invasion of ovarian cancer cells in vitro. As an
   important regulator of cell migration and angiogenesis [[139]52], FAK
   plays an important role in the formation and maturation of focal
   adhesions that are linked to the intracellular actin cytoskeleton
   [[140]53]. FAK(Y397) is an autophosphorylation site initializing the
   activation of FAK signaling and facilitates FAK combination with the
   actin-linking protein vinculin, which is required for highly dynamic
   tissue remodeling during development [[141]35, [142]54, [143]55]. After
   the autophosphorylation of FAK (Y397), the FAK phosphorylation at
   Y576/577 site cause maximal FAK catalytic activation [[144]35, [145]56,
   [146]57]. Our findings reveal that CXCL10 significantly increases
   phosphorylation at the FAK Y576/577 sites, while exerting no
   significant effect on phosphorylation at the Y397 site. This suggests
   that adipose derived CXCL10-driven catalytic activation of FAK is a key
   process that contributing to the migration of ovarian cancer.
   Additionally, other intricate signaling pathways and regulatory
   mechanisms associated with CXCL10-regulated ovarian cancer cell
   migration, such as AKT and MAPK, also merit further investigation.

   Of interest, we found that the conditional medium from the macrophages
   treated with CXCL10 also enhanced the migration and invasion of ovarian
   cancer cell lines. This result suggested that the excess CXCL10
   secreted by adipose tissues may transform the state of macrophages,
   subsequently influencing the metastatic behavior of ovarian cancer
   cells. Previous studies have shown that macrophages are key immune
   cells in the tumor microenvironment [[147]58–[148]60], and the direct
   communication between macrophages and tumor cells can promote the tumor
   development [[149]59, [150]61, [151]62].

   The classical inflammatory markers for M1/M2 macrophage polarization
   manifested no significant changes, while the markers for
   tumor-associated macrophages, namely C1QA, C1QC, CCL24, and IL4R, were
   all significantly increased in macrophages treated with CXCL10. C1Q has
   been shown to promote the migration and adhesion of various human
   malignant tumors, thereby facilitating tumor growth and migration
   through non-classical pathways. Notably, C1Q is present in the ascitic
   fluid associated with ovarian cancer, rather than within the ovarian
   cancer cells themselves [[152]63, [153]64]. And the C1Q + macrophage is
   the driver of cancer progression, as reported in previous studies
   [[154]65]. CCL24 and the IL4/IL4R pathway can facilitate migration and
   invasion of tumor cells [[155]66, [156]67]. Similarly, the expression
   levels of C1qa, C1qb, C1qc, Ccl23, Il4, and Il4ra were significantly
   elevated in macrophages from the subcutaneous adipose tissue of HFD
   mice. Thus, the excess CXCL10 secreted by adipose tissues potentially
   accelerates the migration of ovarian cancer in obese patients by
   altering the tumor microenvironment, at least partly mediated by
   modification of tissue resident macrophages. Since the DEGs in
   C1Q + macrophages include both transcriptional factors and secreted
   proteins, further investigation is warranted to determine whether the
   interactions between C1Q + macrophages and ovarian cancer cells are
   mediated exclusively by secreted factors or also involve direct
   cell–cell contact.

   In conclusion, we propose that high CXCL10 levels secreted by
   adipocytes of obese individuals could exacerbate migration of ovarian
   cancer through direct and indirect mechanisms. Directly, CXCL10
   enhances the migration and invasion of ovarian cancer cells through
   activation of FAK signaling. Indirectly, CXCL10 increases the
   expression of C1QA, C1QC, CCL24, and IL4R in macrophages, leading to
   their transformation into tumor-associated macrophages (TAMs), which
   further enhance the migration of ovarian cancer cell. We highlight
   CXCL10 as a hub between obesity and ovarian cancer, particularly its
   role in promoting the migration of ovarian cancer cells. However, due
   to the absence of in vivo verification, the significance of CXCL10 in
   the immune microenvironment and its impact on ovarian cancer metastasis
   in vivo warrant further research.

Supplementary Information

   [157]13048_2024_1568_MOESM1_ESM.jpg^ (1.6MB, jpg)

   Supplementary Material 1: Figure S1. The expression of Cxcl10 and Ccl3
   in white adipose from both HFD and SCD mice.
   [158]13048_2024_1568_MOESM2_ESM.jpg^ (6.2MB, jpg)

   Supplementary Material 2: Figure S2. The effects of CXCL10 and
   conditional medium from CXCL10-treated macrophages on migration,
   invasion, proliferation, and apoptosis of HO-8910 cells. (A-D) The
   effect of CXCL10 on the migration (A), invasion (B), apoptosis (C), and
   proliferation (D) of HO-8910 cells. (E–H) The effect of the conditional
   medium from CXCL10-treated macrophages on the migration (E), invasion
   (F), apoptosis (G), and proliferation (H) of HO-8910 cells. *:
   P < 0.05; **: P < 0.01; ***: P < 0.001.
   [159]Supplementary Material 3.^ (130.7KB, docx)

Acknowledgements