Colon cancer exosome-associated HSP90B1 initiates pre-metastatic niche formation in the liver by polarizing M1 macrophage into M2 phenotype

Abstract

Background

Colorectal cancer (CRC) frequently metastasizes to the liver, worsening patient outcomes. The formation of a pre-metastatic niche (PMN) is essential for this process, but how the primary colon tumor orchestrates the PMN formation remains unclear.

Methods

This study investigated the role of CRC-derived exosomes using CT-26 murine colon carcinoma cells. The effects of these exosomes on immune cells, specifically M1 macrophage polarization and CD8 + T cell viability, were assessed. HSP90B1 expression in CT-26-derived exosomes was analyzed to understand its contribution to PMN formation. HSP90B1 silencing experiments were conducted to evaluate its impact on immunosuppressive PMN creation and liver metastasis. Patient blood samples were also examined to correlate exosomal HSP90B1 levels with CRC progression.

Results

Exosomes from CT-26 cells were found to polarize M1 macrophages into an M2 phenotype and decrease CD8 + T cell viability, promoting liver metastasis. High expression of HSP90B1 in CT-26 cell-derived exosomes was identified as a key factor in inducing M2 macrophage polarization and creating an immunosuppressive PMN. Silencing HSP90B1 significantly inhibited the exosome-mediated formation of the immunosuppressive PMN and reduced liver metastasis. Furthermore, elevated levels of HSP90B1 in patient-derived exosomes were associated with advanced CRC and poorer prognosis.

Conclusions

CRC-derived exosomes promote liver metastasis by forming an immunosuppressive PMN through HSP90B1. Targeting HSP90B1 in CRC exosomes may offer a new therapeutic strategy to prevent liver metastasis and improve patient outcomes.

Significance

This study elucidates the role of colorectal cancer (CRC) cell-derived exosomes in promoting liver metastasis by creating an immunosuppressive pre-metastatic niche (PMN). We demonstrate that high HSP90B1 expression in these exosomes polarizes M1 macrophages into the M2 phenotype, reducing CD8 + T cell viability. Importantly, HSP90B1 expression correlates with advanced CRC and poor patient outcomes. These findings suggest that targeting HSP90B1 in CRC-derived exosomes could offer a novel therapeutic strategy to inhibit PMN formation and prevent liver metastasis, improving clinical outcomes for CRC patients.

Introduction

Colorectal cancer (CRC) remains one of the most prevalent malignancies worldwide, ranking third in incidence and second in mortality [1, 2]. The metastatic spread of CRC, particularly to the liver, significantly worsens the patient’s prognosis and complicates treatment [3]. Approximately 50% of CRC patients develop liver metastases [4], making it imperative to understand the underlying mechanisms that facilitate this process.

The concept of the pre-metastatic niche (PMN), first introduced by Lyden and colleagues, has revolutionized our understanding of metastasis [5]. The PMN refers to a microenvironment in distant organs that is pre-conditioned by the primary tumor to become conducive to cancer cell colonization [6]. This process is orchestrated by the primary tumor through the release of various factors, including exosomes, cytokines, and growth factors, which collectively remodel the target organ’s microenvironment [6,7,8].

Exosomes, small extracellular vesicles ranging from 30 to 150 nanometers, are critical mediators in the formation of the PMN [7]. These vesicles are secreted by cancer cells and carry a diverse array of bioactive molecules, such as proteins, lipids, RNA, and DNA, reflecting the molecular composition of their cells of origin [9]. In CRC, tumor-derived exosomes travel through the bloodstream to the liver, where they facilitate the establishment of a supportive microenvironment for incoming metastatic cells [10]. Exosomes contribute to PMN formation through several mechanisms. They promote extracellular matrix (ECM) remodeling by transferring matrix metalloproteinases (MMPs) to the liver, facilitating the degradation of ECM components and creating a scaffold that supports metastatic cell growth [11]. Additionally, exosomes carry immunosuppressive molecules such as TGF-β and PD-L1, which modulate the immune environment by suppressing the activity of cytotoxic T lymphocytes (CTLs) and natural killer (NK) cells, thus enabling metastatic cells to evade immune detection and destruction [5, 12, 13]. Furthermore, exosomes facilitate the recruitment of bone marrow-derived cells (BMDCs) like myeloid-derived suppressor cells (MDSCs) and tumor-associated macrophages (TAMs), which contribute to the immunosuppressive milieu and promote angiogenesis [14].

Macrophages play a crucial role in PMN formation, particularly through their polarization into M1 and M2 phenotypes [15]. M1 macrophages, typically induced by interferon-gamma (IFN-γ) and microbial products, are pro-inflammatory and have tumoricidal properties [16]. In contrast, M2 macrophages, which are induced by interleukins such as IL-4 and IL-13, exhibit anti-inflammatory and tissue-remodeling functions, promoting tumor growth and metastasis [17]. Tumor-associated macrophages (TAMs) predominantly exhibit an M2-like phenotype in the tumor microenvironment, which secrete cytokines and growth factors such as VEGF, TGF-β, and IL-10, which promote immunosuppression, ECM remodeling, and angiogenesis [18]. By shaping the local immune landscape and supporting structural changes in the liver, macrophages are pivotal in establishing a niche that favors metastasis [19].

In this study, we demonstrate that colon cancer-associated exosomes enriched with heat shock protein 90B1 (HSP90B1), a molecular chaperone that facilitates the maturation of substrates [20], lead to the polarization of M1 macrophages into the M2 phenotype. This polarization promotes immunosuppression in the liver, consequently resulting in the formation of a pre-metastatic niche and facilitating liver metastasis of colon cancer.

Materials and methods

Cell culture

CT-26 (CL-0071, RRID: CVCL_7256) was obtained from Procell (Wuhan, China). Mouse colonic epithelial cells (MCEC, BFN608006441) and HEK 293T (BFN60700191, RRID: CVCL_0063) were obtained from Bluefcell (Shanghai, China). CT-26 cells were cultured in RPMI-1640 medium (GIBCO, Rockville, MD, USA) supplemented with 10% fetal bovine serum (FBS; Hyclone, Logan, UT, USA). Mouse colonic epithelial cells (MCEC, Bluefcell, Shanghai, China) and HEK 293T (Bluefcell, Shanghai, China) cells were cultured in DMEM medium (GIBCO) supplemented with 10% FBS (Hyclone). All cells were grown in a medium supplemented with 100 µg/mL streptomycin (GIBCO) and 100 units/mL penicillin (GIBCO). Cells were maintained in a humidified 37 °C incubator under a 5% CO2 atmosphere. All cell lines used in this study were routinely tested to be negative for mycoplasma contamination and were kept at low passages to maintain their identity and were authenticated by morphology check and growth curve analysis.

Plasmid transfection and lentiviral infection

Recombinant lentiviruses were amplified by transfecting HEK 293T cells at 75% confluence with pMD2.G and psPAX2 packaging plasmids and lentiviral-based shRNAs specific for green fluorescent protein (GFP, CAAATCACAGAATCGTCGTAT) or HSP90B1 (#1, TAATAGTGAGGAACGTAGGCT; #2, AAATAAAAATCTTGAAAAGAA) using Lipofectamine 2000 (11668, Invitrogen, Carlsbad, CA, USA). Viruses were collected 72 h after transfection. CT26 cells at 40% confluence were infected with recombinant lentivirus in the presence of 10 µg/mL polybrene, followed by 12 h of incubation at 37 °C with 5% CO2.

Western blot analyses

Cells were collected, washed with cold PBS, and resuspended in EBC250 lysis buffer (250 mM NaCl, 50 mM Tris pH 8.0, 0.5% Nonidet P-40, 50 mM NaF, 1 mM phenylmethylsulfonyl fluoride, 2 µg/mL aprotinin, and 2 µg/mL leupeptin). Equal amounts of total protein were loaded, separated by SDS-PAGE, transferred to PVDF membranes (Millipore), and hybridized to an appropriate primary antibody and HRP-conjugated secondary antibody for subsequent detection by enhanced chemiluminescence. Antibodies for CD11b (ab184308), β-actin (ab8226), CD63 (ab217345), CD206 (ab64693), and TSG101 (ab125011) were purchased from Abcam (Cambridge, UK).

Quantitative PCR (qPCR)

Total RNA was extracted from cells using the RNeasy Plus Mini Kit (Qiagen) according to the manufacturer’s protocol. RNA was reverse-transcribed into cDNA using the M-MLV First Strand Kit (Invitrogen). qPCR analyses of CD206 (F: GTTCACCTGGAGTGATGGTTCTC; R: AGGACATGCCAGGGTCACCTTT), CD11b (F: TACTTCGGGCAGTCTCTGAGTG; R: ATGGTTGCCTCCAGTCTCAGCA), and HSP90B1 (F: GTTTCCCGTGAGACTCTTCAGC; R: ATTCGTGCCGAACTCCTTCCAG) were performed in a CFX96 Real-Time PCR System (Bio-Rad) using SoFast EvaGreen Supermix (Bio-Rad) according to the manufacturer’s instructions. The reactions were carried out in a 96-well plate at 95 °C for 5 min, followed by 40 cycles of 95 °C for 15 s and 58 °C for 30 s. GAPDH expression was used as an internal control to normalize gene expression by the ΔΔCt method.

Isolation of cell-derived exosomes

Exosomes were isolated from the culture medium of MCEC or CT-26 cells as previously described [21]. Briefly, the culture medium was centrifuged at 3,000 g for 15 min, and the supernatant was filtered through a 0.22-µm PVDF filter (Millipore). Exoquick-TC exosome precipitation solution (System Biosciences) was added to the filtered culture medium at a ratio of 1:5. After mixing and refrigeration for at least 12 h, the mixture was centrifuged at 1,500 g for 30 min. The exosome pellet was then resuspended in PBS for further analysis.

Collection of patient samples and isolation of colon cancer-derived exosomes

Peripheral blood samples from colorectal cancer patients (n = 27) were collected at Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, China., following the institutional regulations of the Ethics Committee of Tongji Hospital. Ten milliliters of blood were drawn from each patient, serum was separated, and exosomes were isolated using the Umibio^® exosome isolation kits (Umibio, Cat. No: UR52136, China) according to the manufacturer’s instructions. In brief, an initial spin was performed at 3000×g at 4 °C for 10 min followed by 10,000×g at 4 °C for 20 min for each sample to remove cells and debris. The corresponding amounts of reagents were then added proportionally to the starting sample volume as per the manufacturer’s instructions. Mixtures were vortexed and incubated at 4 °C for up to 2 h, and then centrifuged at 10,000×g at 4 °C for 60 min to precipitate exosome pellets. Pellets were resuspended in 1× PBS and purified using an Exosome Purification Filter at 3000×g at 4 °C for 10 min. The resuspension volume for exosome pellets was 200 µL for 20 mL starting volumes according to the manufacturer’s instructions. All exosomes were stored at − 80 °C immediately after isolation until further analysis.

Induction of mouse M1 macrophages

To induce M1-type macrophages in mice, BALB/c mice are euthanized using cervical dislocation and then immersed in 75% ethanol for 10 min. The femur and tibia are cut intact, and sterile PBS solution is used to flush the bone marrow cavity until it turns white, resulting in a single-cell suspension of bone marrow cells. This suspension is centrifuged at 350 g for 4 min to obtain the cell pellet. The pellet is then treated with 2 mL of red blood cell lysis buffer for 5 min, followed by the addition of 8 mL of sterile PBS to stop the lysis, and centrifuged again at 350 g for 4 min. The resulting cell pellet is resuspended in 10 mL of complete culture medium and transferred to a culture dish, where M-CSF is added at a concentration of 25 µg/L. After 3 days, the medium is changed and M-CSF is replenished. Mature bone marrow-derived macrophages (BMDMs) are obtained after 6 days. To induce M1 macrophages, 50 µg/L LPS and 20 µg/L IFN-γ are added to the culture medium.

Flow cytometry

Mouse M1 macrophages (1 × 10⁶) were treated with 20 µg of exosomes derived from MCEC or CT-26 cells for 48 h, followed by co-culture with CD8 + T cells (1 × 10⁶) derived from mouse spleen for an additional 48 h. The cells were then resuspended in 100 µL of cold FACS buffer (PBS + 2% FBS) and incubated with 1 µL of FITC-conjugated Anti-Mouse CD8 for 10 min at 4 °C in the dark. After surface staining, the cells were washed with FACS buffer and fixed with 4% formaldehyde solution for 30 min at 4 °C in the dark. The stained cells were washed twice and resuspended in 300 µL of staining buffer. The percentages of lymphocytes were analyzed using a flow cytometry instrument (FACSCanto, BD, USA) and BD FACSCanto software (version 3.0).

Mass spectrometry

The protein solution derived from exosomes was reduced with 5 mM dithiothreitol for 30 min at 56 °C and alkylated with 11 mM iodoacetamide for 15 min at room temperature in darkness. The protein sample was then diluted by adding 200 mM TEAB to urea concentration less than 2 M. Finally, trypsin was added at 1:50 trypsin-to-protein mass ratio for the first digestion overnight and 1:100 trypsin-to-protein mass ratio for a second 4 h-digestion. The tryptic peptides were dissolved in solvent A, and directly loaded onto a home-made reversed-phase analytical column (25-cm length, 100 μm i.d.). The mobile phase consisted of solvent A (0.1% formic acid, 2% acetonitrile/in water) and solvent B (0.1% formic acid, 90% acetonitrile/in water). Peptides were separated with the following gradient: 0–14.5 min, 6-22%B;14.5–17.5 min, 22-34%B; 17.5–19 min, 34-80%B; 19–20 min, 80%B, and all at a constant flow rate of 700 nl/min on an EASY-nLC 1200 UPLC system (ThermoFisher Scientific). The separated peptides were analyzed in Orbitrap Exploris 480 with a nano-electrospray ion source. The electrospray voltage applied was 2300 V. FAIMS compensate voltage (CV) was set as-45 V. Precursors and fragments were analyzed at the Orbitrap detector. The full MS scan resolution was set to 60,000 for a scan range of 350–1400 m/z. The MS/MS scan was fixed first mass as 120.0 m/z at a resolution of 15,000. The HCD fragmentation was performed at a normalized collision energy (NCE) of 27%. The automatic gain control (AGC) target was set at 1E6, with a maximum injection time of 22 ms.

ELISA assay

ELISA tests were conducted to measure the concentrations of IL-6, TGF-β, TNF-α, IL-12, CCL-1, and IFN-γ in culture supernatants using ELISA kits (BioLegend San Diego, CA, USA). The assay was performed under the manufacturer’s recommended protocols. The colorimetric absorbance was assessed in 450 nm by SpectraMax^® Gemini™ EM Microplate and the cytokine concentrations were analyzed using SoftMax Pro 7 software.

Animal models

All animal experiments were conducted on 5-week-old female BALB/c mice (SLAC Laboratory Animal Co., Ltd, Shanghai, China). Mice were maintained in individual cages at a room temperature of 22 ± 2 °C and humidity of 50–60%, on a 12:12 light–dark cycle (lights on at 09:00 h).

To evaluate the roles of exosomes in the formation of the liver pre-metastatic niche of colon cancer, the mice were injected with exosomes (20 µg in 100 µL PBS) derived from MCEC or CT-26 cells via the tail vein every 3 days for 3 weeks. Three weeks after the initial exosome injections, all mice were euthanized, and liver tissues were collected for the analysis of M1 macrophages, M2 macrophages, and CD8 + T cells. For analyzing the roles of exosomes in liver metastasis of colon cancer, mice were treated with exosomes (20 µg in 100 µL PBS) derived from MCEC or CT-26 cells via tail vein injection every 3 days for 3 weeks. Three weeks after the initial exosome injections, luciferase-labeled CT-26 cells were injected into the spleens of the mice to induce a liver metastasis model, which was evaluated using the IVIS Lumina LT series III (PerkinElmer, MA, USA).

Statistical analyses

GraphPad Prism 8.0 (GraphPad Software Inc.) was used for data recording and calculation. Data were presented as means ± s.d. Quantitative data were analyzed statistically using Student’s t-test for two groups or one‑way ANOVA for multiple groups to assess significance.

Results

Colon cancer cell-derived exosomes create an immunosuppressive microenvironment in the mouse liver to promote liver metastasis

Colorectal cancer (CRC) is highly prevalent, ranking third in incidence and second in mortality worldwide [1]. Liver metastasis, occurring in about 50% of CRC patients, greatly worsens prognosis and complicates treatment [3, 4]). It has been documented that the formation of the pre-metastatic niche (PMN) plays a crucial role in cancer cell metastasis [5]. The PMN is orchestrated by the primary tumor through the release of various factors, including exosomes [7]. However, whether colon cancer cell-associated exosomes promote PMN formation in the liver to facilitate metastasis remains unclear. To address this issue, we first isolated exosomes from CT-26, a murine colon carcinoma cell line, and MCEC, a mouse colon epithelial cell line. As shown in Figure S1A-C, CT-26 cell-derived exosomes exhibited similar morphology, size, and expression levels of Tsg101 and CD63, two well-established exosome markers [22, 23], compared to MCEC cell-derived exosomes. We then examined whether colon cancer cell-derived exosomes affect the immune environment in mouse liver. As shown in Fig. 1A-D, compared to MCEC cell-derived exosomes, CT-26 cell-derived exosomes significantly downregulated the number of M1 macrophages (CD86+/CD11b+, a common marker combination used to identify M1 macrophages) and upregulated the number of M2 macrophages (CD206+/CD11b+, a marker combination typically used to identify M2 macrophages) in the mouse liver. Moreover, CT-26 cell-derived exosomes also markedly decreased the number of CD8 + T cells (Fig. 1E-F). These results suggest that colon cancer cell-derived exosomes create an immunosuppressive microenvironment in the mouse liver. Next, we examined whether colon cancer cell-derived exosomes promote liver metastasis of colon cancer. As shown in Fig. 1G-O, compared to MCEC cell-derived exosomes, CT-26 cell-derived exosomes significantly facilitated liver metastasis of colon cancer cells, concomitant with decreased the number of M1 macrophages and CD8 + T cells and increased the number of M2 macrophages.

Fig. 1

Colon cancer cell-derived exosomes inhibit the immune microenvironment of the mouse liver. (A-F) Five-week-old female BALB/c mice (n = 8 per group) were injected with 20 µg of CT-26 cell-derived exosomes (CT-26/exo), MCEC cell-derived exosomes (MCEC/exo), or PBS via the tail vein every 3 days. After 3 weeks, the mice were sacrificed, and liver tissues were collected. Immunofluorescence staining was performed to detect the number of M1 macrophages (CD86+/CD11b+) (A, B), M2 macrophages (CD206+/CD11b+) (C, D), and CD8 + T cells (E, F) in the liver tissues. Scale bar = 50 μm. (G-Q) Five-week-old female BALB/c mice (n = 8 per group) were injected with 20 µg of CT-26/exosome, MCEC/exosome, or PBS via the tail vein every 3 days. After 3 weeks of exosome treatment, luciferase-labeled CT-26 cells (1 × 10^6) were injected into the tail vein of each group. (G) Experimental protocol. On day 21 after tumor cell injection, metastasis in each group was monitored using a small animal in vivo imaging system (H, I), followed by dissection of the livers. Hematoxylin and eosin (HE) staining was used to quantify the number of metastatic nodules in the liver per mouse (J, K). Immunofluorescence staining was then performed to detect the number of M1 macrophages (CD86+/CD11b+) (L, M), M2 macrophages (CD206+/CD11b+) (N, O), and CD8 + T cells (P, Q) in the liver tissues. Scale bar = 50 μm. Data were presented as mean ± SEM (B, D, F, I, K, M, O, Q). Statistical analyses were performed with one-way ANOVA with Tukey’s test (B, D, F, I, K, M, O, Q)

Together, these results indicate that colon cancer cell-derived exosomes create an immunosuppressive PMN in the mouse liver to facilitate liver metastasis.

Colon cancer cell-derived exosomes polarize M1 macrophage into M2 phenotype in vitro

Next, we explored the mechanisms by which colon cancer cell-derived exosomes lead to an immunosuppressive pre-metastatic niche (PMN). Our previous data indicate that colon cancer cell-derived exosomes significantly decreased the number of M1 macrophages and increased the number of M2 macrophages. Therefore, we speculated that colon cancer cell-derived exosomes could polarize M1 macrophages into the M2 phenotype to create an immunosuppressive PMN. To test this hypothesis, we first established mouse M1 macrophages. As shown in Figure S2A, mouse peritoneal macrophages were transformed into M1 macrophages after LPS treatment, consistent with previous reports [24]. Moreover, M1 macrophages exhibited similar uptake efficiency for both CT-26 cell-derived exosomes and MCEC cell-derived exosomes (Figure S2B-C).

Importantly, compared to MCEC cell-derived exosomes, CT-26 cell-derived exosomes significantly increased CD206 mRNA and protein expression, a marker of M2 macrophages (Fig. 2A-C). FACS analyses further confirmed that CT-26 cell-derived exosomes could significantly convert M1 macrophages to M2 macrophages (CD206+/CD11b+) in a dose-dependent manner (Fig. 2D-E).

Fig. 2

Colon cancer cell-derived exosomes polarize M1 macrophage into M2 phenotype in vitro. (A-C) Mouse M1 macrophages were treated with 20 µg of CT-26 cell-derived exosomes (CT-26/exo) or MCEC cell-derived exosomes (MCEC/exo) for 48 h. Cells were subjected to qPCR (A) or western blot (B-C) analyses (n = 3 independent experiments). (D-E) Mouse M1 macrophages were treated with 20 µg of CT-26/exo, MCEC/exo, or PBS or were treated with 40 µg of CT-26/exo (CT-26/exo High) for 48 h. Cells were subjected to FACS analyses (n = 3 independent experiments). (F) Mouse M1 macrophages were treated with 20 µg of CT-26/exo, MCEC/exo for 48 h followed by examination of as indicated cytokines levels in the medium (n = 3 independent experiments). (G-H) Mouse M1 macrophages (1 × 10⁶) were treated with 20 µg of CT-26/exo, MCEC/exo for 48 h, followed by co-culture with CD8 + T cells (1 × 10⁶) derived from mouse spleen for 48 h. Cells were subjected to FACS analyses (n = 3 independent experiments). Data were presented as mean ± SD (A, C, E, F, H). Statistical analyses were performed with one-way ANOVA with Tukey’s test (A, C, E, F, H)

Consistently, exosomes derived from CT-26 cells markedly suppressed the secretion of TNF-α, IL-6, and IL-12 by M1 macrophages (Fig. 2F). In contrast, CT-26 cell-derived exosomes dramatically decreased the levels of IL-10, TGF-β, and CCL-1, which are secreted by M2 macrophages (Fig. 2F). Importantly, compared to MCEC/exo-treated M1 macrophages had little effect on CD8 + T cell viability, CT-26/exo-treated M1 macrophages could significantly inhibit CD8 + T cell viability (Fig. 2G-H).

Together, these results indicate that colon cancer cell-derived exosomes polarize M1 macrophages into the M2 phenotype, thereby suppressing CD8 + T cell viability.

Colon cancer exosome-associated HSP90B1 promotes the polarization of M1 macrophage into M2 phenotype to facilitate pre-metastatic niche formation and liver metastasis

We then investigated the molecular basis by which colon cancer cell-derived exosomes polarize M1 macrophages into the M2 phenotype. To address this issue, we performed mass spectrometry analyses. As shown in Fig. 3A, CT-26 cell-derived exosomes (CT-26/exo) exhibited higher HSP90B1 expression, a molecular chaperone that facilitates the maturation of substrates [20, 25], compared to MCEC cell-derived exosomes, which was further confirmed by western blot analyses (Fig. 3B-C and S3A-C). Importantly, silencing HSP90B1 in CT-26 cells significantly suppressed the CT-26/exo-mediated conversion of M1 macrophages to M2 macrophages (CD206+/CD11b+), as evidenced by FACS analyses (Fig. 3D-H).

Fig. 3

Colon cancer exosome-associated HSP90B1 promotes the polarization of M1 macrophage into M2 phenotype in vitro and in vivo. (A-C) 20 µg of CT-26 cell-derived exosomes (CT-26/exo) or MCEC cell-derived exosomes (MCEC/exo) were subjected to mass spectrometry analyses (A) or western blot analyses (B-C) (n = 3 independent experiments). (D-F) CT-26 cells stably expressing shGFP, shHSP90B1-1, or shHSP90B1-2 were subjected to qPCR (D) or western blot (E-F) analyses (n = 3 independent experiments). (G-H) Mouse M1 macrophages were treated with 20 µg of CT-26-shGFP/exo, CT-26-shHSP90B1-1/exo, or CT-26-shHSP90B1-2/exo for 48 h. Cells were subjected to FACS analyses (n = 3 independent experiments). (I) Mouse M1 macrophages were treated with 20 µg of CT-26-shGFP/exo, CT-26-shHSP90B1-1/exo, or CT-26-shHSP90B1-2/exo for 48 h followed by an examination of as indicated cytokines levels in the medium (n = 3 independent experiments). (J-K) Mouse M1 macrophages (1 × 10⁶) were treated with 20 µg of CT-26-shGFP/exo, CT-26-shHSP90B1-1/exo, or CT-26-shHSP90B1-2/exo for 48 h, followed by co-culture with CD8 + T cells (1 × 10⁶) derived from mouse spleen for 48 h. Cells were subjected to FACS analyses (n = 3 independent experiments). (L-Q) Five-week-old female BALB/c mice (n = 8 per group) were injected with 20 µg of CT-26-shGFP/exo, CT-26-shHSP90B1-1/exo, or CT-26-shHSP90B1-2/exo, or PBS via the tail vein every 3 days. After 3 weeks, the mice were sacrificed, and liver tissues were collected. Immunofluorescence staining was performed to detect the number of M1 macrophages (CD86+/CD11b+) (L-M), M2 macrophages (CD206+/CD11b+) (N-O), and CD8 + T cells (P-Q) in the liver tissues. Scale bar = 50 μm. Data were presented as mean ± SD (C, D, F, H, I, K) or SEM (M, O, Q). Statistical analyses were performed with one-way ANOVA with Tukey’s test (D, F, H, I, K, M, O, Q) and unpaired two-tailed Student’s t-test (C)

Furthermore, the downregulation of TNF-α, IL-6, and IL-12 levels and the upregulation of IL-10, TGF-β, and CCL-1 levels mediated by CT-26/exo could be significantly reversed by silencing HSP90B1 in CT-26 cells (Fig. 3I). Additionally, the inhibition of CD8 + T cell viability by CT-26/exo-treated M1 macrophages was also reversed by silencing HSP90B1 (Fig. 3J-K). In vivo assay showed that CT-26/exo-mediated the downregulation of M1 macrophages and CD8 + T cells and the upregulation of M2 macrophages in the mouse liver, all of which could be reversed by silencing HSP90B1 (Fig. 3L-Q). These results suggest that colon cancer exosome-associated HSP90B1 promotes the polarization of M1 macrophages into the M2 phenotype, creating an immunosuppressive microenvironment in the mouse liver.

Next, we explored the role of HSP90B1 in colon cancer exosome-mediated liver metastasis. As shown in Fig. 4A-K, CT-26/exo significantly facilitated liver metastasis of colon cancer cells, concomitant with a decreased number of M1 macrophages and CD8 + T cells and an increased number of M2 macrophages, all of which could be largely reversed by silencing HSP90B1.

Fig. 4

Colon cancer exosome-associated HSP90B1 promotes the formation of a pre-metastatic niche and facilitates liver metastasis. Five-week-old female BALB/c mice (n = 8 per group) were injected with 20 µg of CT-26-shGFP/exo, CT-26-shHSP90B1-1/exo, or CT-26-shHSP90B1-2/exo, or PBS via the tail vein every 3 days. After 3 weeks of exosome treatment, luciferase-labeled CT-26 cells (1 × 10⁶) were injected into the tail vein of each group. On day 21 after tumor cell injection, metastasis in each group was monitored using a small animal in vivo imaging system (A-B), followed by dissection of the livers. Hematoxylin and eosin (HE) staining was used to quantify the number of metastatic nodules in the liver per mouse (C-D). Immunofluorescence staining was then performed to detect the number of M1 macrophages (CD86+/CD11b+) (E-F), M2 macrophages (CD206+/CD11b+) (G-H), and CD8 + T cells (I-J) in the liver tissues. Data were presented as mean ± SEM (B, D, F, H, J). Statistical analyses were performed with one-way ANOVA with Tukey’s test (B, D, F, H, J). Scale bar = 50 μm

Together, these results indicate that colon cancer exosome-associated HSP90B1 promotes the formation of a pre-metastatic niche to facilitate liver metastasis of colon cancer.

High HSP90B1 expression in exosomes derived from peripheral blood is associated with advanced colon cancers and poor clinical outcomes

We then investigated the clinical correlation between HSP90B1 expression in exosomes and colon cancer prognosis. As shown in Fig. 5A, mass spectrometry analyses revealed that exosomes derived from the peripheral blood of colon cancer patients had higher HSP90B1 expression compared to those from healthy individuals. This finding was further confirmed by western blot analyses (Fig. 5B-C). Moreover, exosomes from the peripheral blood of colon cancer patients with lymph node-positive or hepatic metastasis exhibited significantly higher HSP90B1 expression than those from patients with lymph node-negative status (Fig. 5D-E). Importantly, high HSP90B1 expression in exosomes derived from peripheral blood is associated with poor prognosis in colon cancer patients (Fig. 5F). These results suggest that HSP90B1 expression in circulating exosomes could serve as a biomarker for advanced colon cancer and may predict clinical outcomes.

Fig. 5

High HSP90B1 expression in exosomes derived from peripheral blood is associated with advanced colon cancers and poor clinical outcomes. (A-C) 20 µg of exosomes derived from the peripheral blood of healthy individuals (n = 10) or colon cancer patients (n = 10) were subjected to mass spectrometry (A) or western blot (B-C) analyses. (D-E) 20 µg of exosomes derived from the peripheral blood of colon cancer patients with lymph node-negative (n = 5), lymph node-positive (n = 5), or hepatic metastasis (n = 5) were subjected to western blot analyses. The classification of node-positive, node-negative, or distant metastasis was determined by experienced clinicians based on imaging assessments, followed by confirmation through subsequent pathological verification. (F) The correlation between the HSP90B1 levels in exosomes derived from the peripheral blood of colon cancer patients and overall survival (OS) was analyzed. (G) A model depicts that colon cancer exosome-associated HSP90B1 initiates pre-metastatic niche formation in the liver by polarizing M1 macrophage into M2 phenotype to facilitate liver metastasis. Data were presented as mean ± SEM (C and E). Statistical analyses were performed with one-way ANOVA with Tukey’s test (E), unpaired two-tailed Student’s t-test (C), and Log-rank (Mantel-Cox) test (F)

Discussion

The formation of a pre-metastatic niche (PMN) is a crucial step in the process of CRC liver metastasis. The PMN prepares the distant organ, in this case, the liver, to receive and support incoming cancer cells. This process is facilitated by various factors secreted by the primary tumor, including exosomes, cytokines, and growth factors [7]. The establishment of a PMN in the liver involves the recruitment of bone marrow-derived cells, remodeling of the extracellular matrix (ECM), and suppression of local immune responses, creating a favorable microenvironment for tumor cells [5]. In this study, we demonstrate that exosomes secreted by colorectal cancer (CRC) cells contain the protein HSP90B1, which plays a pivotal role in liver metastasis. HSP90B1 in these exosomes can be taken up by liver-resident M1 macrophages, leading to their polarization into M2 macrophages. This polarization process significantly suppresses CD8 + T cell activity, thereby establishing a pre-metastatic niche conducive to cancer cell colonization and subsequent liver metastasis (Fig. 5G).

Despite these advances, the precise molecular mechanisms by which HSP90B1 mediates the polarization of macrophages remain unclear. It is known that HSP90B1 is involved in protein folding and stabilization, but its role in immune modulation and macrophage polarization is less well understood [26]. Understanding these pathways could reveal novel targets for therapeutic intervention. Future research should focus on elucidating the signaling cascades activated by HSP90B1 within macrophages, potentially involving pathways such as NF-κB, STAT3, or other transcription factors involved in macrophage differentiation [27,28,29].

Our results corroborate previous studies that emphasize the role of M2 macrophages in tumor progression. M2 macrophages are known for their tissue remodeling and immunosuppressive properties, which facilitate tumor growth and metastasis [30]. The uptake of HSP90B1 by M1 macrophages and their subsequent polarization to an M2 phenotype underscores the complexity of the tumor microenvironment. This shift not only creates an immunosuppressive environment but also enhances the tumor’s ability to evade immune surveillance. Similar mechanisms have been observed in other cancer types, where exosome-mediated delivery of specific proteins and RNAs modulates immune cell function and supports metastatic spread [9].

The implications of these findings are significant for therapeutic strategies targeting the metastatic process. By inhibiting the secretion of exosomes or blocking the expression of HSP90B1, it might be possible to prevent the formation of a pre-metastatic niche. This approach could enhance the efficacy of existing treatments and improve the prognosis for patients with advanced CRC. Targeting the exosomal pathway and its components offers a novel strategy to disrupt the communication between tumor cells and the immune system, thereby hindering metastatic progression.

Conclusions

In conclusions, our results indicated that colon cancer cell derived exosomes create an immunosuppressive microenvironment in the liver, facilitating metastasis. Notably, exosome-associated HSP90B1 was identified as a promoter of M1 to M2 macrophage polarization, aiding PMN formation and liver metastasis. Furthermore, high HSP90B1 expression in exosomes from peripheral blood was correlated with advanced colon cancer and poorer clinical outcomes. Overall, these findings suggest that targeting HSP90B1 in CRC-derived exosomes could provide a novel therapeutic strategy to inhibit PMN formation and prevent liver metastasis, ultimately improving clinical outcomes for CRC patients.