COX-2– and IDO-associated immunomodulatory functions of stage II carcinoma-associated fibroblasts in breast cancer

Abstract

The immune response against cancer is influenced by the tumor stroma. Carcinoma-associated fibroblasts (CAFs), a significant part of the stroma, are important modulators of the tumour microenvironment (TME). The current study examined the immunomodulatory properties of CAFs isolated from a stage II mouse model of mammary invasive ductal cancer. Based on the expression of surface markers, we described Stage II CAFs and assessed their effects on cytokine release, nitric oxide (NO) generation, and splenocyte proliferation in direct co-culture systems. Class II MHC molecules (I-Ad/I-Ed), FAP-1, CD29, CD90, and CD105 were all expressed by isolated CAFs, indicating the possibility of direct immunological contact. In direct co-culture, CAFs demonstrated two distinct functions: their conditioned medium (CM) had immunosuppressive effects, whereas direct cell interactions promoted splenocyte growth. Additionally, CAFs changed the cytokine milieu, particularly by secreting high levels of TGF-β and PGE₂, and dramatically reduced splenocyte NO generation. COX-2, IDO, and MMP2 were significantly upregulated in CAFs, according to gene expression data. Our results emphasize the potential of Stage II CAFs as a therapeutic target in breast cancer by demonstrating their complex immunomodulatory character, which includes the ability to both stimulate and repress immune responses through cellular contact and soluble factors.

Introduction

Carcinoma-associated fibroblasts (CAFs) constitute a prominent cellular element within the tumour microenvironment (TME) and significantly influence tumour progression, angiogenesis, extracellular matrix remodelling, and immune responses. A growing body of research suggests that CAFs aid in immune evasion by secreting cytokines, chemokines, and lipid mediators such as prostaglandin E₂ (PGE₂), TNF-α, IL-6, and IL-1β, via NF-κB-dependent pathways (1). This inflammatory milieu not only suppresses anti-tumor immunity but can also facilitate metastasis (2).

CAF populations exhibit functional heterogeneity as a result of dynamic alterations in the stromal compartment that accompany the evolution of breast cancer (3–5). While epithelial tumour cells experience genetic changes, stromal cells predominantly undergo epigenetic and transcriptional reprogramming, which significantly affects immune cell recruitment, differentiation, and function (6). Different immune phenotypes, such as immunological-inflamed, immune-excluded, or immune-desert states, are formed within tumours as a result of these stromal-driven effects (7). As a result of a dynamically evolving TME, the immune response may change depending on the stage of breast cancer. In our previous work, we established murine models representing different stages of breast cancer progression and isolated CAFs from these stages (8).

Building on this paradigm, the current work examines the immunophenotype of CAFs isolated from stage II tumours and assesses their impact on cytokine production, splenocyte proliferation, and important immunoregulatory mechanisms. Since CAFs and mesenchymal stem cells (MSCs) share mesodermal lineage origins and phenotypic traits, we directly compared Stage II CAFs to adipose-derived MSCs (AD-MSCs). This approach allows us to identify functions that are specifically gained or amplified by CAFs in the tumour context, distinguishing them from the baseline immunomodulatory program of a normal mesenchymal cell. This is critical for defining tumor-specific therapeutic targets. In this study, we specifically investigate the functional consequences of COX-2 and IDO upregulation, namely enhanced PGE₂ secretion and the potential for tryptophan catabolism, and their association with the complex immunomodulatory profile of CAFs.

Materials and methods

Isolation and characterization of AD-MSCs and CAFs

In our previous work, we established a murine model in which CAFs were isolated from tumors at defined temporal stages of progression, herein referred to as stage II and stage IV, reflecting early versus advanced tumor development rather than clinical staging (8). Briefly, five 4–6 weeks old female Balb/C mice were inoculated subcutaneously into the left flank with 1 × 10⁶ MBL-6 murine breast cancer cell line. Stage II CAFs were isolated from tumors of Balb/C mice inoculated with the MBL-6 cell line. One gram tumor tissues were digested in an enzymatic cocktail containing collagenase IV, hyaluronidase and DNase I for 4 h in a shaking incubator at 37℃. CAFs were purified via Ficoll gradient separation. Adipose-derived mesenchymal stem cells (AD-MSCs) were isolated from Balb/C mice visceral fat. Both cell types were cultured in F12/DMEM supplemented with 10% FBS and penicillin/streptomycin (100 mg/mL). In all experiments cells in passage 2 or 3 were used. Isolated cells displayed a characteristic spindle-shaped, fibroblast-like morphology upon adherence to tissue culture plastic, as observed by phase-contrast microscopy.

Cell characterization was performed using flow cytometry using the following antibodies: anti-mouse PE-conjugated anti-CD29, anti-CD105, anti-I-A^d/I-E^d, anti-CD45, anti-CD90, and anti-FAP; anti-mouse FITC conjugated anti-CD104a (PDGFRa), anti-Sca-1 (eBioscience). BD FACS flow cytometry was used and results were analyzed using Flowjo software (version 7.6.1). Our previous study has shown that these cells express FAP, as indicator of CAFs, therefore in this study to compare the immunomodulatory properties we also used common MSC markers for comparison.

Evaluation of splenocyte proliferation

Immune activation is initially characterized by lymphocyte expansion against antigen. The ability of CAFs and AD-MSCs to influence immune cell proliferation was evaluated using the MTT assay. Briefly, Mitomycin-inactivated CAFs and AD-MSCs were co-cultured with splenocytes in cellular contact (CC) or Conditioned Media (CM) at four ratios of 1/10, 1/50, 1/100 and 1/250 in RPMI-1640 supplemented with 10% FBS, penicillin/streptomycin (100 mg/mL) and at 37 °C in a humid atmosphere containing 5% CO₂. For CM collection, CAFs and AD-MSCs were cultured in fresh media at predefined numbers for 24 h, after which the CM was collected and transferred to fresh wells. Splenocytes were added to all cultures. 5 µg/ mL PHA (Baharafshan, Tehran, Iran) was added to stimulate splenocytes. After 72 h, proliferation was assessed via MTT protocol. Co-culture ODs were corrected with mono-CAF culture. As a control, splenocytes were activated by 5 µg/ mL PHA alone and the negative control was untreated splenocytes. The optical density of splenocyte proliferation was calculated and expressed as stimulation index (SI).

Where OD_CC−P is the optical density of co-cultured CAFs and splenocytes stimulated with PHA (or their conditioned media), OD_SPL−P is the optical density of splenocytes stimulated with PHA alone, and OD_CAF and OD_SPL are the optical density of CAFs and splenocytes cultured without PHA stimulation.

Analysis of cytokine production

Splenocytes were co-cultured in direct contact or with conditioned medium of Mitomycin-inactivated CAFs/AD-MSCs in 24-well plate (TPP) at 1:10 ratio. PHA was added and after 72 h, the supernatant was collected and the levels of IL-10, IL-17, TGF-β1, and Prostaglandin E2 were assessed by ELISA (DuoSet ELISA Development kit, R&D Systems, Minneapolis, MN, USA). All procedures were followed according to the manufacturer’s protocol.

Evaluation of nitric oxide production

NO levels (nmol/ml) were measured in supernatants of co-cultures by the Griess reaction. Briefly, nitrite was measured by adding equal volumes of Griess reagent (1% sulphanilamide and 0.1% naphthylenediamine in 5% phosphoric acid). Optical density was measured at 550 nm. Fresh culture medium was used as blank. All chemicals were obtained from Merck (Darmstadt, Germany).

Gene expression analysis using qRT-PCR

Gene Expression was assessed using Real-Time PCR. Total cellular RNA was extracted using Trizol (Gibco-BRL, Life Technologies, MD). Random hexamer-primed reverse transcription (Metabion) was performed on aliquots of total RNA (1 µg) as a template. The resulting cDNA was used for Real-Time PCR amplification. Primers (Sinacolon, Tehran, Iran) for murine cycloxygenase 2 (COX-2), inducible Nitric Oxide Synthetase (iNOS), Indolamine Deoxygenase (IDO), Matrix Metalloproteinase 2 (MMP2) and 9 (MMP9) and beta-actin were synthesized based on the reported sequences and were as follows:

COX-2 (145 bp) forward: 5’-AGACAGATCATAAGCGAGGAC-3’, reverse: 5’-CCACCAATGACCTGATATTTC-3’; INOS (142 bp): forward: 5’-TGTGCGAAGTGTCAGTGG-3’, reverse: 5’-TCCTTTGAGCCCTTTGTG-3’; IDO (168 bp): forward: 5’-GGATGCGTGACTTTGTGG-3’, reverse: 5’-TGGAAGATGCTGCTCTGG-3’; MMP2 (150 bp): forward: 5’-AGACAAGTTCTGGAGATACAATG-3’, reverse: 5’-GCACCCTTGAAGAAGTAGC;-3’ MMP9 (136 bp): forward: 5’-GGCGTGTCTGGAGATTCG-3’, reverse: 5’-TGGCAGAAATAGGCTTTGTC-3’.

Real-time PCR reaction mixtures (final volume of 30 ul) contained 1 µl cDNA, 30 pmol of each primer, 3 µl of 200 µM dNTP, and 1U Taq-DNA polymerase (MBI Fermentas Inc., Burlington, ON). Amplification conditions were as follows: 40 cycles of 94 °C for 30 s; 55 °C for 60 s; and 72 °C for 1 min, followed by 72 °C incubation for 10 min (Corbett Life Sciences, Australia). The results of primer amplification were analyzed using Rotor-Gene 6000^® (Qiagen AG Hilden, Series Software 1.7, build 34) and REST software version 2009. Gene expression was assessed by the relative expression method compared to AD-MSC cells.

Statistical analysis

All data are presented as mean ± SD unless otherwise stated. After performing Shapiro–Wilk test for normality, data were analyzed by ANOVA-Tukey post hoc test or Kuskal-Wallis test using SPSS v.19 software. GraphPad Prism v.8 was used for visualization. The relative expressions were analyzed using REST software. A P-value less than 0.05 was considered statistically significant. This study was evaluated and approved by the ethical committee of the Tarbiat Modare University, Tehran, Iran, under license number IR.TMU.REC.1390.326. All methods in this research were performed following the relevant guidelines.

Results

CAFs and AD-MSCs exhibit similar phenotypic profiles

Table 1 provides a summarized overview comparing the immunomodulatory effects of Stage II CAFs and AD-MSCs. Isolated cells displayed a characteristic spindle-shaped, fibroblast-like morphology upon adherence to tissue culture plastic, as observed by phase-contrast microscopy. Flow cytometry analysis confirmed that Stage II CAFs were positive for CD29, CD90, CD105, FAP-1, and I-A^d/I-E^d (MHC class II), and negative for CD45, CD11b, CD140a, and Sca-1. A subset (~ 23%) of CAFs expressed MHC class II, indicating a potential for direct T-cell interaction. AD-MSCs shared mesenchymal markers (CD29, CD90, CD105) but lacked CD45, CD11b, CD140a, FAP, and MHC class II expression (Fig. 1). Quantification of the flow cytometry data is shown in the right panel of Fig. 1.

Table 1.

Comparative Summary of Immunomodulatory Effects of Stage II CAFs and AD-MSCs on Splenocytes

Immunomodulatory Function	Co-culture Condition	Effect of Stage II CAFs	Effect of AD-MSCs
Splenocyte Proliferation	Cellular Contact (CC)	Stimulatory (↑ SI)	No significant effect
	Conditioned Medium (CM)	Suppressive (↓ SI)	No significant effect
Cytokine Secretion (IL-10)	CC & CM	Increases production, primarily from splenocytes	CM induces high levels of IL-10 from splenocytes
Cytokine Secretion (TGF-β1)	CC & CM	High production, predominantly from CAFs	No significant production
Cytokine Secretion (IL-17)	CC	High production from splenocytes	No significant production
Cytokine Secretion (PGE₂)	CC & CM	High production, predominantly from CAFs	No significant production
Nitric Oxide (NO) Production	CC & CM	Strongly suppressive (↓ NO from splenocytes)	High basal production; no suppression of splenocyte NO
Key Gene Expression	Monoculture	Upregulation: COX-2 (+++), IDO (++), MMP2 (++)	Baseline expression (control)
Phenotype	Monoculture	CAF-like: FAP⁺, MHC-II⁺ (~ 23%)	MSC-like: FAP⁻, MHC-II⁻

SI, Stimulation Index; CC, Cellular Contact; CM, Conditioned Medium; +++ indicates very high upregulation; ++ indicates significant upregulation

Fig. 1.

surface marker analysis of STAGE II CAFs and AD-MSCs. flow cytometry analysis of surface marker expression of common mesenchymal markers revealed that CAFs express class II MHC molecules, a strong indicative of immunomodulatory function of CAF. Quantification of the flow cytometry data is shown in the right panel Results were analyzed using one-way ANOVA with Tukey post hoc test. Significant difference is indicated with Asterix. ns = not significant, *= P-value < 0.05, ** = P-value<0.01, ***= P-value < 0.001****= P-value<0.0001. Data are presented as mean ± SD (n = 3)

CAFs have a biphasic effect on splenocyte proliferation

As shown in Fig. 2, direct cellular contact (CC) with CAFs led to an increased stimulation index (SI) of splenocytes, indicative of an immune stimulatory function. This effect decreased with lower CAF-to-splenocyte ratios. Conversely, in conditioned medium (CM) cultures, the SI increased as the ratio of CAFs used to generate the CM decreased, indicating that the immunosuppressive properties of CAFs reside mainly in their soluble secretome. In contrast, neither AD-MSCs in direct co-culture nor their CM significantly altered PHA-induced splenocyte proliferation under these assay conditions.

Fig. 2.

evaluation of Stimulation index splenocyte co-cultures with Stage II CAFs or AD-MSCs under direct cellular contact (CC) or conditioned medium (CM) conditions. when activated using PHA, splenocytes showed higher SI in the presence of stage II CAFs and decreased SI in the presences of CM (CAF2c). also, when cocultured with AD-MSCs, no significant increase in splenocyte SI was observed. Results were analyzed using one-way ANOVA with Tukey post hoc test. Significant difference is indicated with Asterix. ns = not significant, *= P-value < 0.05, ** = P-value<0.01, ***= P-value < 0.001****= P-value<0.0001. Data are presented as mean ± SD (n = 5). Bars with no indication of significant difference are not significant and “ns” is included for emphasis. (SPL: splenocyte, SPL + PHA: splenocytes stimulated with PHA, CAF-II: stage II CAFs, AD-MSC: Adipose Derived-Mesenchymal stem cells, CC: cellular contact coculture, CM: conditioned media coculture)

CAFs alter the co-culture cytokine milieu

IL-10 is an adaptive immunity-produced cytokine. As shown in Fig. 3c, CAFs are unlikely to be the primary source of IL-10. Also, the serum used in the culture media contained very low amounts of IL-10. Therefore, the IL-10 detected is attributable to splenocytes. Additionally, results indicate that activated splenocytes decrease their production of IL-10. This production is highest when in cellular contact (CC) rather than in the presence of conditioned media (CM). Also, AD-MSCs’ scretome showed that these cells also produce high levels of IL-10 (Fig. 3a).

Fig. 3.

Elevated levels of suppressive cytokines were observed from splenocyte co-cultures with Stage II CAFs or AD-MSCs under direct cellular contact (CC) or conditioned medium (CM) conditions. Evaluating the production of suppressive cytokines in direct co-culture of splenocytes with stage II CAFs or AD-MSCs after 72 h using ELISA showed significant production of IL-10 and TGF-β1 compared to splenocytes group. Results were analyzed using ANOVA with tukey post hoc test. Significant difference is indicated with Asterix. ns = not significant, *= P-value < 0.05, ** = P-value<0.01, ***= P-value < 0.001****= P-value<0.0001. Bars with no indication of significant difference are not significant and “ns” is included for emphasis. (SPL: splenocyte, SPL + PHA: splenocytes stimulated with PHA, CAF-II: stage II CAFs, AD-MSC: Adipose Derived-Mesenchymal stem cells, CC: cellular contact coculture, CM: conditioned media coculture)

TGF-β is a ubiquitous cytokine produced from many cells in the TME. Results demonstrated that stage II CAFs produce high levels of TGF-β (Fig. 3d). Also, splenocytes without the influence of stage II CAFs do not produce high amount of TGF-β. Therefore, it is not clear to what extent splenocyte are responsible for TGF-β production. AD-MSCs also produce TGF-β however in lower levels compared to CAF-II (Fig. 3b).

PGE₂ is an inflammatory mediator produced by enzyme Cyclooxyegenase 2. Figure 4 demonstrates that splenocytes, activated or not, do not produce significant amount of PGE₂. However, high levels of PGE₂ are produced when splenocytes are under the influence of CAFs or their CM (Fig. 4c). This production was significantly higher in cellular contact with CAF-II. AD-MSCs showed low levels of PGE2 production. However, the production of PGE2 in co-cultures of AD-MSCs and splenocytes increased, but was not significant. This increase was prominent when stimulated with PHA (Fig. 4a).

Fig. 4.

levels of proinflammatory cytokines in cocultures from splenocyte co-cultures with Stage II CAFs or AD-MSCs under direct cellular contact (CC) or conditioned medium (CM) conditions. Production of inflammatory cytokines in direct co-culture of splenocytes with stage II CAFs or AD-MSCs after 72 h were assessed using ELISA. Significant production of IL-17 and PGE₂ was evident in cocultures of Stage II CAFs -SPL compared to AD-MSCs group. Results were analyzed using ANOVA with tukey post hoc test. Significant difference is indicated with Asterix. ns = not significant, *= P-value < 0.05, ** = P-value<0.01, ***= P-value < 0.001****= P-value<0.0001. Bars with no indication of significant difference are not significant and “ns” is included for emphasis. (SPL: splenocyte, SPL + PHA: splenocytes stimulated with PHA, CAF-II: stage II CAFs, AD-MSC: Adipose Derived-Mesenchymal stem cells, CC: cellular contact coculture, CM: conditioned media coculture)

IL-17 is a differentiation cytokine produced by TH₁₇ cells. As shown in Fig. 4d, the source of IL-17 in the culture media is not CAFs and is produced from splenocytes, only when activated in cellular contact. Also, AD-MSC co-cultures did not induce significant IL-17 production above baseline (Fig. 4b).

CAFs suppress nitric oxide production

Nitric oxide, a product of iNOS enzyme, was evaluated using the Griess reaction. Splenocytes alone produced high levels of NO upon PHA stimulation (3.27 ± 0.8 nmol/mL). This production was significantly reduced in co-cultures with CAFs (CC: 0.24 ± 0.3 nmol/mL and CM: 0.28 ± 0.3 nmol/mL; P-value < 0.01, Fig. 5b ). Without PHA stimulation, co-cultures of CAFs and splenocytes produced undetectable amounts of NO, however, when stimulated NO levels raised in both co-culture and CM-cultures. AD-MSCs alone produced high levels of NO, which decreased in direct co-culture with splenocytes but remained high when only AD-MSC CM was present., comparable to AD-MSC culture (Fig. 5a).

Fig. 5.

levels of NO production from splenocyte co-cultures with Stage II CAFs or AD-MSCs under direct cellular contact (CC) or conditioned medium (CM) conditions. Production of nitric oxide in direct co-culture of splenocytes with stage II CAFs or AD-MSCs after 72 h were assessed using Griess reaction. Significant production of NO was evident in cocultures of AD-MSCs-SPL compared to Stage II CAFs. Results were analyzed using ANOVA with tukey post hoc test. Significant difference is indicated with Asterix. ns = not significant, *= P-value < 0.05, ** = P-value<0.01, ***= P-value < 0.001****= P-value<0.0001. Bars with no indication of significant difference are not significant and “ns” is included for emphasis. (SPL: splenocyte, SPL + PHA: splenocytes stimulated with PHA, CAF-II: stage II CAFs, AD-MSC: Adipose Derived-Mesenchymal stem cells, CC: cellular contact coculture, CM: conditioned media coculture)

Stage II CAFs upregulate key inflammatory and remodeling genes

Expression of inflammatory genes in isolated CAFs were compared with AD-MSCs. Results, presented as relative expression compared to the AD-MSC control, showed significant increases in COX-2, IDO and MMP2 expression in stage II CAF (Fig. 6). COX-2 was the gene with the most significant increase (relative expressions: CAF-II: 580.0 ± 164.6 Vs AD-MSCs: 1.3294 ± 0.70, P-value < 0.001). IDO and MMP2 also showed a significant increase compared to AD-MSCs (relative expressions CAF-II: 111.98 ± 49.16 Vs AD-MSCs: 1.079 ± 0.38, P-value = 0.02; and CAF-II: 123.54 ± 67.02 Vs AD-MSCs: 1.54 ± 1.37, P-value = 0.01, respectively).

Fig. 6.

Expression of genes involved in inflammation and matrix remodeling. Real-time PCR analysis showed significant increase in the expression of COX-2, IDO and MMP2 in Stage II CAFs. Gene expressions were calculated using ΔΔ2Ct formula and expressed as Fold changes. Beta-actin was used as a reference gene and setting the mean expression of the AD-MSC control group to 1. Results were analyzed using ANOVA with Tukey post hoc test. Significant difference is indicated with Asterix. ns = not significant, *= P-value < 0.05, ** = P-value<0.01, ***= P-value < 0.001****= P-value<0.0001

Discussion

Studies have shown that carcinoma-associated fibroblasts (CAFs) display phenotypic similarities to adipose-derived mesenchymal stem cells (AD-MSCs) (9, 10), and play a crucial role in cancer development and immune regulation (11). The present research isolated cancer-associated fibroblasts (CAFs) from a mouse stage II breast cancer model, demonstrating their complex, context-dependent immunomodulatory effects on splenocytes. Our findings indicate that stage II CAFs possess both immune-stimulating and immune-suppressing activities, mediated through direct cell-to-cell interactions and soluble factors, respectively.

This study’s noteworthy finding is the detection of major histocompatibility complex class II (MHC II) molecules on a subset of CAFs, suggesting their potential for direct interaction with T cells. This trait is consistent with previous research showing that CAFs are multifunctionally diverse subsets rather than a single population(12, 13). Recent single-cell transcriptome investigations have identified three primary subtypes of CAFs: myofibroblastic CAFs (myCAFs), inflammatory CAFs (iCAFs), and antigen-presenting CAFs (apCAFs) (14). MyCAFs are characterized by increased expression of α-smooth muscle actin and genes related to the extracellular matrix, principally promoting tissue remodelling and augmenting stromal stiffness (15). Conversely, iCAFs exhibit a transcriptional profile that facilitates inflammation and secretes numerous cytokines and chemokines, including IL-6, IL-1β, and various members of the CXCL family (16). This modifies the processes of immune cell recruitment and the inflammatory miliue within the tumour microenvironment. A third variety, known as apCAFs, possesses MHC class II molecules and can interact with CD4⁺ T cells (17). Nevertheless, they lack the costimulatory molecules essential for the complete activation of T cells. It is suggested that these apCAFs assist in modulating the immune system rather than immune-activation(18).

This study’s noteworthy finding is the detection of major histocompatibility complex class II (MHC II) molecules on a subset of CAFs, suggesting their potential for direct interaction with T cells. The functional properties observed in our investigation suggest that stage II CAFs possess characteristics typical of both apCAF and iCAF populations. The presence of MHC class II molecules classifies them as apCAF, whereas the elevated production of inflammatory mediators, such as prostaglandin E₂ (PGE₂), along with the substantial overexpression of COX-2 and IDO, indicates an iCAF-like immunoregulatory profile. This overlap indicates the plasticity of CAF populations and exists along a functional continuum (19). This plasticity may enable CAFs to adjust to evolving tumor and immune system microenvironment as cancer advances.

Our direct co-culture experiments revealed that CAFs had dual effects on splenocyte responses. Direct contact between CAFs and splenocytes resulted in an enhanced lymphocyte population, indicating that cell–cell interactions could deliver activating or costimulatory signals via antigen presentation or adhesion-dependent mechanisms. Conversely, exposure to CAF-conditioned medium suppressed splenocyte growth, indicating that soluble substances produced by CAFs have significant paracrine immunosuppressive effects. This discrepancy illustrates the complexity of interactions between CAFs and the immune system, indicating that the tumour microenvironment significantly influences immune responses (20).

The analysis of cytokine production validates this dual regulatory role. CAFs were a significant source of prostaglandin E₂ (PGE2) and transforming growth factor-β (TGF-β). Both of these are recognized as the suppressors of immune system responses within the tumour microenvironment (21). Splenocytes were the possible primary source of IL-10 and IL-17, indicating that CAFs do not directly influence adaptive immune polarization. Rather, they constitute the primary stimulant of these cytokines. The elevated concentrations of PGE₂ in cultures with CAFs correlate with a marked upregulation of COX-2 at the transcriptional level, highlighting a vital mechanism via which CAFs may suppress effector immune responses(22, 23).

The synthesis of nitric oxide (NO) represents an alternate mechanism for the regulation of immune effector actions (24). This study found that splenocytes stimulated by phytohemagglutinin (PHA) produced significant amounts of NO, whereas co-culturing with CAFs substantially reduced NO levels. This finding suggests that CAFs interfere with iNOS-dependent pathways crucial for immune system activation. The significant overexpression of IDO in CAFs may contribute to this inhibitory effect, as IDO activity has been demonstrated to impede T-cell function and negatively influence nitric oxide production (25). It has been shown that NO preferentially induces TH1 differentiation through upregulation of IL-12 receptor β2 expression via cGMP (26). While direct mechanistic connections were not explored in this study, the results support a hypothesis suggesting that metabolic and enzymatic pathways derived from CAFs attenuate immune effector functions. A unique aspect of our study is the direct functional and genetic comparison between Stage II CAFs and AD-MSCs. Our goal was not to simply find a normal counterpart but to determine which immunomodulatory functions are uniquely acquired or amplified as a mesenchymal cell transition to a carcinoma-associated phenotype. For example, the suppression of NO production by splenocytes was a unique property of CAFs, not shared by AD-MSCs. This comparison pinpoints gain-of-function pathways in CAFs, such as the dramatic upregulation of COX-2 and IDO, as tumor-specific vulnerabilities. From a clinical perspective, targeting these pathways in stage II disease could reverse early immunosuppression without affecting the normal tissue-repair functions of resident MSCs.

Our findings collectively suggest a framework wherein stage II CAFs function as dynamic modulators of the immunological environment, facilitating lymphocyte activation through contact-dependent mechanisms while concurrently inhibiting immune effector functions via soluble mediators. This functional duality likely aids the immune system’s adaptation during the early stages of cancer development by permitting initial targeting of the tumour and subsequently enabling the tumour to avoid immune detection over time. The presence of both apCAF- and iCAF-like characteristics within CAF population underscores the adaptable and context-sensitive nature of stromal cells in breast cancer (27). A limitation of this work is that we have not yet directly compared these Stage II CAFs with Stage IV CAFs. Such a direct comparison is the logical next step to demonstrate how these immunomodulatory programs evolve with tumour progression and is currently being actively investigated in our laboratory.

From a translational viewpoint, these data underscore CAF-derived pathways, including COX-2/PGE₂, IDO, and TGF-β, as prospective therapeutic targets for modifying the cancer microenvironment (28, 29). Focusing on these systems, either individually or in conjunction with immunotherapies, may augment antitumor immunity by mitigating CAF-induced immunosuppression. Future investigations are required to verify these pathways in vivo, analyze CAF heterogeneity across various tumour stages, and investigate how targeted control of CAF subgroups may enhance immune-based therapeutic methods in breast cancer.

Authors’ contributions

Conceptualization; ZMH, LL. SMH, MSData curation; LL.Formal analysis; LL, Funding acquisition; ZMH, MS.Investigation; LL, SMH.Methodology; ZMH, LL. SMH, MS.Project administration; ZMH. SMH, MS, Resources; ZMH. SMH, MSSoftware; LL, SMH, Supervision; ZMH. SMH, MSValidation ZMHVisualization; LLRoles/Writing - original draft; LL.

Funding sources

This study was funded by the research department University of Medical Sciences with the grant reference number IR.TMU.REC.1390.326.

Data Availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

This study was approved by the ethical committee of the Kerman University of Medical Sciences with the reference number IR.TMU.REC.1390.326. Informed consent was obtained from all participants before the procedure.

Competing interests

The authors declare no competing interests.