Colorectal cancer (CRC) remains one of the most commonly diagnosed and deadliest types of cancer worldwide. CRC displays a desmoplastic reaction (DR) that has been inversely associated with poor prognosis; less DR is associated with a better prognosis. This reaction generates excessive connective tissue, in which cancer-associated fibroblasts (CAFs) are critical cells that form a part of the tumor microenvironment. CAFs are directly involved in tumorigenesis through different mechanisms. However, their role in immunosuppression in CRC is not well understood, and the precise role of signal transducers and activators of transcription (STATs) in mediating CAF activity in CRC remains unclear. Among the myriad chemical and biological factors that affect CAFs, different cytokines mediate their function by activating STAT signaling pathways. Thus, the harmful effects of CAFs in favoring tumor growth and invasion may be modulated using STAT inhibitors. Here, we analyze the impact of different STATs on CAF activity and their immunoregulatory role.

Colorectal cancer (CRC) is the third most common tumor and the second leading cause of cancer-related deaths worldwide[1]. In recent decades, CRC mortality and incidence have decreased continuously in the adult population ≥ 65 years of age. However, this pathology has drastically increased in both men and women worldwide, including young people (< 50 years old)[2]. Despite the increase in the survival rate of this type of cancer, patients diagnosed at advanced clinical stages have unfavorable prognoses, with a survival rate of 14% at five years[1]. Once the disease spreads to distant organs, conventional chemotherapy or other targeted therapies do not significantly improve patients. At least one-third of patients develop metastases to the liver, and deaths are mostly attributed to distant metastasis[3]. Some key poor prognostic factors related to metastasis in CRC include lymphovascular invasion, degree of differentiation, and positive lymph nodes[4,5]. Approximately 70% of CRCs are sporadic, and some cases (10%-15%) are related to a personal history of adenoma/sessile serrated polyps and inflammatory bowel diseases that significantly increase the risk of developing CRC[6]. Less frequently, CRC may develop because of genetic syndromes (lynch syndrome, familial adenomatous polyposis, and hyperplastic polyposis, among others); these have a higher risk of tumor development and occur most often at an earlier age[2,6].

Currently, CRC management is based on the TNM staging system[7]. Surgical resection is the mainstay treatment. According to the TNM system, systemic chemotherapy and local pelvic radiotherapy are necessary adjuvant treatment modalities[8]. Correct histological stratification of the tumor is the primary tool used for selecting the best patient treatment. Most CRCs (90%) are adenocarcinomas that originate from epithelial cells of the colorectal mucosa[9]. The rare types include neuroendocrine, squamous cell, adenosquamous, spindle cell, and undifferentiated carcinomas[9].

Macroscopically, CRC usually has a polypoid appearance with a depressed central ulcer. The intraluminal region may cause obstruction. When the tumor invades the wall of the intestine and the surrounding tissue, it induces varying degrees of desmoplasia[4]. The presence of desmoplasia or a desmoplastic reaction (DR) is an important feature[10].

The DR is characterized by excessive accumulation of connective tissue that encapsulates the tumor, composed of recruited immune cells, activated fibroblasts, capillary formation, and the secreted extracellular matrix proteins such as collagen, fibronectin, tenascin C, periostin, laminin, growth factors, cytokines, and chemokines[11]. The DR has been associated with poor prognosis in patients[12,13] and has been reported in various solid tumors, including ductal adenocarcinoma of the pancreas (PDAC), breast cancer, lung cancer, ovarian cancer, head and neck cancer, and CRC[14]. The cells responsible for generating DR are cancer-associated fibroblasts (CAFs), which promote tumor development by increasing the proliferation, invasion, and migration of tumor cells as well as resistance to chemotherapy and radiotherapy[15]. The expression of stromal genes associated with DR in CRC increases the frequency of tumor-initiating cells through transforming growth factor β (TGF-β). Current CRC staging based on histopathology and imaging has limited scope for predicting prognosis, leading to the need for molecular classifications[13]. One is based on the classification of the Subtype Consortium for CRC, which has identified four consensus molecular subtypes (CMS): CMS1, CMS2, CMS3, and CMS4. The infiltration of immune cells characterizes CMS1. CMS2 has marked activation of the WNT and MYC signaling pathways, and CMS3 has marked alterations in metabolic signaling pathways. CMS4 is linked to genetic signatures associated with the epithelial-mesenchymal transition (EMT), as well as with the activation of TGF-β, and the activation of vascular endothelial growth factor (VEGF) and VEGFR receptor[16]. In addition, CMS4 is characterized by the high expression of the stroma-derived genes POSTN and CALD1 in CAFs, which is directly associated with the development of tumor biology and poor prognosis[13]. It should be noted that CMS4 has the worst prognosis in terms of overall survival for all clinical stages, and the worst relapse-free survival among cases classified as clinical stages I-III compared with other subtypes of this classification[16].

In recent years, it has been reported that CAFs not only have pro-tumoral functions but may also have the ability to generate an immunosuppressive microenvironment in the tumor, recruiting and differentiating immunosuppressive cells, creating a barrier for cytotoxic T cells that keeps them away from the tumor[17]. One of the mechanisms that has been observed to be involved in the immunosuppressive function of CAFs in solid tumors is the signal transduction pathway regulated by the signal transducer and activator of transcription (STAT) signaling[18]. It is linked to various cellular processes, such as proliferation, apoptosis, immune system regulation, and hematopoiesis[19]. In cancer, STAT pathway dysregulation not only contributes to the malignant transformation of the cell but also regulates the tumor microenvironment (TME)[20]. Therefore, the STAT pathway is of therapeutic interest in cancer treatment. However, little information exists on the interaction between this signaling pathway and its association with the DR and CAFs. In this paper, we review the role of STAT in the immunosuppressive and pro-tumoral potential of CAFs in CRC.

The TME plays a crucial role in fundamental biological processes involved in the development of solid tumors, such as tumor growth, immunosuppression, lack of response to treatment, and distant metastasis. The TME is a complex biological system that is composed of tumor cells, infiltrates of immune cells (macrophages, dendritic cells, and leukocytes), cancer-associated stromal cells (CAFs, endothelial cells, and adipocytes), along with the extracellular matrix (ECM) and multiple signaling molecules[17]. Interestingly, CAF cell populations are critical players in the development of the DR and are considered fibroblasts that secrete ECM proteins that surround and encapsulate solid tumors.

Microarray expression studies and immunohistochemical analyses have revealed that the DR gradually increases during CRC carcinogenesis from the normal mucosa and low- and high-grade adenomas to adenocarcinomas[21], suggesting a close relationship between CRC progression and DR.

Fibroblasts have been studied under normal conditions using in vitro multi-tissue models; they have antitumorigenic properties and may suppress tumor cell proliferation[22]. However, under conditions of damage, inflammation, and neoplastic transformation, intestinal mesenchymal cells (IMCs) in the colon respond to stimuli from the microenvironment (mainly transformed epithelial cells and inflammatory cells)[23]. These inflammatory and carcinogenic processes activate stromal cells (fibroblasts and IMCs) to acquire a myofibroblastoid (CAF) phenotype[24]. Activated CAFs primarily express alpha-smooth muscle actin (α-SMA), fibronectin, vimentin, fibroblast-specific protein 1, and fibroblast activation protein (FAP)[17,24]. In addition to these markers, CD90, actin, desmin, platelet-derived growth factor receptor alpha (PDGFRα), and glial neuron antigen 2 (NG2) have been used to identify the CAFs presence[25]. But they are not always expressed together[24]. This has been proven by activated cell sorting fluorescence experiments where CAFs have been isolated by the absence of the epithelial cell adhesion molecule marker (EpCAM), the hematopoietic marker (CD45), and the endothelial marker (CD31), as well as by the positive expression of the FAP and PDGFRα markers[13,26]. It should be noted that although there are markers that identify CAFs, they are not specific because they are shared with other cell types and healthy tissues[27]. Different subpopulations of CAFs have been identified, which, depending on the expression of markers, act in favor of or against tumor development in other neoplasms (Table 1)[14,15,17]. Studies have used sequence analysis of single-cell RNA from human CRC samples to identify two main subtypes of CAFs, based on the TGF-β gene expression pathway. The CAF-A subtype is characterized by the high expression of matrix metalloproteinase-2, decorin (DCN), collagen type Iα2 (COL1A2), and FAP. The CAF-B subtype has high expression of myofibroblastic markers such as α-SMA, transgelin (TAGLN), and PDGFα[28]. In murine models of PDAC and colon cancer (CAC), CAFs lacking FAP have been reported to affect tumor development, suggesting that populations positive for this marker have a promoting function[13,29].

CAFs can originate from cell populations other than quiescent fibroblasts (the primary source)[30]. This process is mediated by multiple signaling pathways and activation modulators that allow the transdifferentiation of epithelial/mesothelial cells, endothelial cells, monocytes, smooth muscle cells, pericytes, adipocytes, mesenchymal stem cells from bone marrow cells (MSCs), and IMCs from colon tissue[14,17,25] (Figure 1). This may explain why CAFs are a heterogeneous cell population in terms of their phenotypic markers and functions within the TME[11,31]. Understanding the mechanisms that give rise to CAFs and the particular functions of each subtype derived from the various cell populations provides an opportunity to direct targeted treatments against the DR[14]. However, most studies on the origin of CAFs are based on in vitro experiments and bone marrow transplant studies evaluated in one or a few tumor models[14]. Further information is needed to understand the importance of CAF subtypes in tumor development and how to inhibit them.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 The cellular origins of cancer-associated fibroblasts in solid tumors and colorectal cancer. (1) Colon resident fibroblasts, intestinal mesenchymal cells and mesenchymal stem cells are converted to cancer-associated fibroblasts (CAFs) by stimulation of different activators including stroma-derived factor 1, transforming growth factor beta (TGF-β), TGF-β1, C-X-C chemokine ligand 12, reactive oxygen species, and the Notch and Akt signaling pathways; (2) Monocytes are transformed to CAF through monocyte-myofibroblast transdifferentiation; (3) Pericytes, adipocytes and smooth muscle cells are transformed to CAF by stimulation of the activators TGF-β1 and Wnt-3a; and (4) Epithelial cells, mesothelial cells, tumor cells are transformed into CAF through the process of epithelial mesenchymal transition by stimulation of the activator TGF-β1, endothelial cells are transformed into CAF by the process of endothelial-mesenchymal transition by stimulation of the TGF-β activator. In addition to all these factors involved in myofibroblast activation, interleukin 1 β (IL-1β), IL-6 through nuclear factor kappa B, and Janus kinase have been reported to function as transducers and activators of the signal transducer and activator of transcription 3 to promote CAF activation. Created with BioRender.com. TGF-β: Transforming growth factor beta; IMC: Intestinal mesenchymal cells; MSCs: Mesenchymal stem cells; CAFs: Cancer-associated fibroblasts; CXCL12: C-X-C chemokine ligand 12; IL: Interleukin; NF-κB: Nuclear factor kappa B; SDF-1: Stroma-derived factor 1; ROS: Reactive oxygen species.

Fibroblasts are generally quiescent in colon tissue, have negligible transcriptomic and metabolic activity, and are located mainly in the lamina propria (the connective tissue adjacent to the epithelium). The primary function of colonic fibroblasts is to maintain the ECM through the secretion of ECM proteins and modulating enzymes, as well as the regulation of epithelial proliferation and differentiation. Fibroblasts are the most active cell type in the stroma and healthy connective tissue. When activated (myofibroblasts), they change their morphology to a stellar form; they are metabolically more active and have a higher rate of proliferation, and they regulate inflammation and tissue remodeling under certain physiological conditions (wound healing and fibrosis). Once they fulfill their function, the number of myofibroblasts is reduced through apoptosis or reversion to their quiescent state; failure of this process allows fibrosis in the tissue and chronic inflammation, which can lead to the development of the DR and the progression of CRC[25].

Mesenchymal cells also transform into myofibroblasts and are the main cellular constituents of the normal gut and solid intestinal tumors, contributing to carcinogenesis through their interaction with tumor cells and other cell types in the TME. Recently, the role of these cells in intestinal tumorigenesis has been studied, particularly in carcinogenesis associated with colitis[25]. IMCs sustain their structure and maintain intestinal homeostasis, and comprise a spectrum of cell types that are similar in origin, function, and molecular markers (intestinal subepithelial myofibroblasts, lamina propria fibroblasts, pericytes, MSCs, smooth muscle cells of the muscularis mucosae, and external muscularis)[25,32] (Table 2).

Normal IMCs have been proposed as regulators of the tumor immune microenvironment (TIME) and inflammation. In vitro experiments with IMCs were used to determine the expression of Toll-like receptors (TLRs) and nucleotide-binding oligomerization domains 1 and 2 (NOD1 and NOD2). IMCs respond to cytokines secreted in vitro by immune cells, such as macrophages, T and B cells, mast cells, epithelial cells, and viral and bacterial products, generating inflammatory and anti-inflammatory mediators, cytokines, and chemokines that affect the infiltration of inflammatory cells in the intestine[25,33].

In CRC, CAFs are capable of secreting different inflammatory mediators, including chemokines, cytokines, transcription factors, proteases, and growth factors, which promote an immunosuppressive TME and affect the infiltration of inflammatory cells[25,33]. CAFs in murine models of cancer-associated colitis secrete high levels of the cytokine interleukin 6 (IL-6), osteopontin, chemokine ligands with CC motifs (CCL2, CCL8, CCL11), CXCL2, CXCL5, and amyloid serum A3; CCL2 and CCL8 were reported to promote the migration, invasion, and adhesion of a CRC tumor line[34].

CCL2, IL-6, and macrophage colony-stimulating factor 1 (M-CSF1) secreted by CAFs recruit and increase the macrophage population in pancreatic cancer[35]. The macrophages that infiltrate tumors are known as tumor-associated macrophages (TAMs), and these are divided into two groups that are activated by different cytokines. Lipopolysaccharides with or without Th1 cytokines generate M1-type macrophages and Th2 cytokines generate M2-type macrophages. It has been reported that TAMs are the closest immune cells to CAFs, which suggests close communication between both cell populations. The co-expression of CAF markers and M2 macrophages in tumors has been associated with malignant progression and poor prognosis in CRC[24]. CAFs, through various regulatory molecules (IL-8, IL-10, TGF-β, and CCL2), promote the recruitment of monocytes and their differentiation to type M2 macrophages, decreasing the immune response of effector T cells in solid tumors[17]. M2 macrophages that are polarized by their interaction with CAFs display elevated expression of programmed cell death protein (PD-1) on the cell surface, which confers suppression of innate and adaptive immunity in the tumor, with a reduced capacity for phagocytic potential, while inhibiting T cell infiltration and proliferation[36]. TAMs with the M2 phenotype reciprocally regulate CAF activation by secreting stroma-derived factor 1 (SDF-1) and IL-6[37].

Experimental data in mice suggest that tumor associated neutrophils (TANs) in the TIME may acquire either the N1 antitumor phenotype or the N2 protumor phenotype depending on TGF-β activation[38]. In TAMs, their polarization is molecule-dependent, whereas in TANs it is based on the degree of activation[39]. In addition, the partnership between CAFs and TANs acts bidirectionally. TANs differentiate to an N2 phenotype in the presence of CAF-derived cardiotrophin-like cytokine factor 1 secretion by upregulating CXCL6 and TGFβ in tumor cells from hepatocellular carcinomas[40]. It has also been discovered that gastric cancer MSCs (G-MSCs) can be activated to a CAF phenotype in the presence of N2-type TANs[41]. This close relationship between CAFs and TANs could be mediated by the chemokine receptor C-X-C 2 and SDF-1α expressed by CAFs, which enable neutrophil recruitment into solid tumors, and by IL-6, which stimulates the STAT3 signaling pathway in TANs, inhibiting T cell activation and inducing immune tolerance through the expression of PD-1 and programmed death ligand (PD-L1)[17,42]. In CRC TANs can activate TLR9 and promote tumor growth, migration, and invasion by activity of mitogen-activated protein kinase (MAPK) signaling. Therefore, an increase in neutrophils in the blood has been associated with poor prognosis in patients with CRC[3].

Natural killer (NK) cells are part of the innate immune system, and can generate a response against tumor cells through ligand-receptor recognition mechanisms, cytokines, and cytotoxicity (perforins and granzymes)[43]. CAFs can inhibit the antitumoral activity of NK cells through processes that include receptor activation recognition, toxicity, and cytokine production. Prostaglandin E 2 (PGE-2) and indoleamine 2,3-dioxygenase (IDO) secreted by CAFs have been described as one of the mechanisms that inhibits activation receptors in solid tumors (melanoma and hepatocellular carcinoma)[44,45]. They can exert an immune-regulating effect on NK cells through interactions with other cells such as macrophages, favoring the inhibition of their activation and cytotoxicity[46]. Furthermore, in in vitro co-cultures, it has been observed that NK cells favor the elevated secretion of PGE-2 in CAFs more than in normal fibroblasts (dual relationship) and that the secretion of TGF-β by CAFs inhibits the activation and cytotoxic activity of NK cells by reducing interferon-γ (IFN-γ) expression which downregulates cell surface receptor activation, such as NKG2D[47]. On the other hand TGF-β1 secreted by CAFs downregulates NKp30, NKp46, and DNAM-1 by activating the SMAD 2/3 signaling pathway[48].

Wei et al[49] reported a crucial role for CAFs in the recruitment of immunosuppressive cell populations mediated by CXCL12/CXCR4/CXCR7, favoring an immunosuppressed TIME in CRC through the reduction of T and NK cells. They observed that treatment with the ketogenic enzyme 3-hydroxy-3-methylglutaryl-CoA synthase 2 inhibited CAF proliferation and CXCL12 expression, thereby promoting a non-immunosuppressed TIME[49]. This further reinforces the idea that tumors displaying desmoplastic activity have a complex TIME that, in addition to helping cancer develop, protects them against therapy and helps evade the innate and adaptive immune responses.

In 2022, for the first time, it was reported in a model of murine CRC-associated colitis that CAFs can also come from a population of intestinal cells with the pericryptal leptin receptor (cells of mesenchymal origin), which develop an α-SMA positive phenotype, expressing the melanoma cell adhesion molecule (MCAM), a specific marker of CRC stroma. High expression of MCAM induced by TGF-β has been associated with poor prognosis in patients with CRC. In murine models in which stromal MCAM was knocked out, a decrease in the growth of orthotopically injected colorectal tumor cells was observed by reducing the recruitment of tumor-associated macrophages[21]. MSCs have been previously reported in other models of DR, particularly in PDAC, in which the secretion of GM-CSF regulated tumor proliferation and metastasis. In that study, GM-CSF knockdown was used in CAFs of tumor origin co-transplanted with pancreatic cancer tumor cells, and metastatic activity was completely blocked in an orthotopic murine model of PDAC[50]. In CAC, it has been reported that MSCs modulate tumorigenic activities, such as tumor growth, migration, and metastasis, through the proinflammatory cytokine IL-6[51]. In breast and ovarian tumors TGF-α, TGF-β, and CXCL12 secreted by CAFs have been reported as promoters of the development of metastasis[25,52].

In gastrointestinal tumors induced by inflammation, it has been described that 20% of CAFs come from an α-SMA-positive population of MCSs that migrated from the bone marrow and were recruited mainly by mechanisms involving TGF-β and CXCL12. These CAFs in turn, promote tumorigenesis through the secretion of IL-6, Wnt family member 5A, and bone morphogenetic protein 4 (BMP4)[25,53]. CAFs are a highly active population within the TME that, based on marker expression and protein secretion, exhibit tumor-inhibiting or tumor-promoting properties. This indicates that inflammatory cytokines play an important role in CAF activity.

In a murine model of skin carcinogenesis, it was observed that activation of nuclear factor-κB (NF-κB) due to IκB kinase-β (IKKβ) in CAFs plays a vital role in inflammation that promotes tumor development[54]. This has been studied in models of CAC associated with colitis, where IKKβ-deficient CAFs, created by deletion of IKKβ in type VI collagen-positive fibroblasts, had reduced production of IL-6 and reduced tumor size along with infiltration of immune cells[55]. However, the results are controversial given that in another study, deleting IKKβ in a CAF COL1A2-positive population in a similar CAC model accelerated tumor growth through hepatocyte growth factor (HGF) secretion[56]; this could be because the CAFs have different markers and could correspond to different subpopulations; while some promote tumor development, others restrict tumor activity.

CAFs which activate the receptor for HGF, MET, promote stemness and chemoresistance in murine models of CAC and hepatocellular carcinoma (HCC), and induce the upregulation of keratin 19 in HCC by the activation of MET, which has been observed to be associated with poor patient prognosis[15,57]. Both types of myofibroblasts in HCC and CAC can secrete proteins of the EGF family and promote tumor progression through ERBB receptors, including the EGF receptor (EGFR)[58]. Insulin-like growth factor-2 (IGF-2) secreted by CAFs has also been reported to be a key mediator of stemness in tumor cells[59], and IL-6 and IL-11 increased tumor cell proliferation and liver metastasis in murine models of CAC by increasing STAT3 signaling[25,55,60]. In addition, the WNT signaling pathway and BMP signaling are closely related to the fate of intestinal stem cells normally and in cancer, and HGF expressed by CAFs plays a vital role in the maintenance of tumor stem cells by mediating an increase in WNT[61].

In these studies, it was evident that the inflammatory TME and related cytokines are important factors in the activity of CAF in CRC and in types of gastrointestinal cancers. However, subpopulations of CAFs define the growth or restraint of tumors, and more studies are necessary to define and characterize them because understanding their functions could be useful for targeted therapies.

An increase in CAFs in the TME is correlated with poor prognosis and recurrence of CRC[33]. Several studies have investigated the role of CAFs in gastrointestinal tumor models. Findings in the KPC transgenic murine model of PDAC (KrasLSL−G12D/+;Trp53LSL−R172H/+; Pdx1-Cre) where the DR is active[62] and murine models of CAC associated with colitis due to the use of azoxymethane and dextran sulfate sodium, have shown that there are subpopulations of CAFs that can restrain tumor development. Simultaneously, other factors can favor tumorigenesis[63,64]. When the α-SMA+ population is depleted or the Sonic hedgehog signaling pathway (critical signaling pathway necessary for CAF and DR activation) is blocked in these murine models of PDAC and CAC, tumor progression is accelerated. When CAFs are eliminated from these models, both the CAF subpopulation that restrains the development of the neoplasm and the CAFs that promote its development are depleted and the tumor cells themselves are primed to spread, resulting in poor prognosis in the murine models[63].

Based on the pattern of marker expression in CAFs, the presence of these cells has diagnostic and clinical prognostic value for patients with cancer[65]. It has been reported that it is not only tumor cells that can enter the circulation. Circulating stroma is characterized by the presence of FAP-positive CAFs in the peripheral blood of patients with CRC metastasis and other types of solid tumors[66]. For this reason, several strategies have been used to eliminate this cell population; one of them is the depletion of CAFs using as a target the expression of surface markers, mainly FAP, α-SMA, and PDGFR[67,68]. Therefore, it is important to investigate the use of inhibitors of these markers.

α-SMA is another of the most commonly used markers to identify CAFs; however, due to the controversy of dual effects on disease progression, research into this marker is in progress. Depleting α-SMA-positive populations suppresses angiogenesis and metastasis in breast cancer and PDAC, however, there are opposite results when blocking populations of α-SMA-positive CAFs because by depleting the signaling pathway that generates the DR or by inhibiting α-SMA targeting, the tumor becomes more aggressive and progresses faster, increasing the infiltration of CD3+Foxp3+ Tregs in the TME[69]. Elevated stromal expression of TGF-β-related genes is associated with poor prognosis in patients diagnosed with CRC, and histological studies have reported that high expression of α-SMA or stromal cells predicts poor clinical outcome in patients with CRC, PDAC, and HCC[70-73]. Clinical trials of PDGFR inhibitors are still underway[17,74].

Notably, none of these markers are specific to CAFs. Rather than eliminating the stromal compounds associated with cancer, the activation of the stroma and, therefore, its function within the TME, could be suppressed through the blockade of the effector molecules that activate CAFs, for example, TGF-β, one of the main activators of fibroblasts. It has been suggested that inhibition could restore the damaged immune response in the TME[11]. Some studies have shown that the use of alisertib, a TGF-β inhibitor, has a better effect against solid tumors when combined with chemotherapy or immunotherapy such as the anti-PDL1 checkpoint inhibitor[75-77], the latter of which could be explained by the expression of PDL1 in CAFs and colon myofibroblasts[78,79].

Cellular reprogramming, via vitamin D/A, of tumor-promoting fibroblasts (pCAFs) to quiescent phenotypes or tumor-restraining fibroblasts (rCAFs) is another of the most studied therapeutic models in PDAC and CRC[80,81]. Target-directed therapy using these cells yielded good results when combined with chemotherapy (gemcitabine) in KPC mice. This has been done in activated myofibroblasts, where different methods have been used, including administering a vitamin D analog or all-trans retinoic acid, and produced antitumor effects, such as increasing CD8+ T cells in the compartments near the tumor and limiting the invasion of tumor cells[82,83].

Evasion of the immune system, tumor proliferation, apoptosis evasion, and drug resistance through inflammatory mediators secreted by CAF correlate with other characteristics involved in the poor prognosis of patients with CRC, including the EMT that ultimately participates in cell stemness and tumor migration[33]. One of the mechanisms that has been proposed as part of the development of solid tumors is the STAT pathway, which has also been reported to be activated by cytokines secreted by CAFs; however, few studies have documented evidence of its role in tumor progression and immunosuppression in CRC[84]. Therefore, in this work, we discuss the relationship between STAT and CAFs in the progression of CRC.

Cancer progression requires modifications and adaptations in cancer cells and the cooperation of cells from the microenvironment, including stromal and immune cells[18]. The STAT pathway is involved in diverse cellular processes as previously mentioned. Additionally, STAT pathway dysregulation is involved in the progression of pathological conditions such as rheumatoid arthritis, atopic dermatitis, and cancer[85-87]. In cancer, dysregulation of the STAT pathway contributes to the malignant transformation of cells and regulates the TME[88,89]. Therefore, the STAT pathway is of therapeutic interest in cancer treatment.

The STAT family comprises seven members. STAT3 is an important regulator of cancer progression[90]. STAT3 activation is involved in response to several cytokine and growth factors, including IL-6, IL-10, EGF, fibroblast growth factor and IGF[48,91,92]. Among the stimuli that lead to STAT3-pathway activation, IL-6 is the most well described[93]. Once IL-6 is detected at the cell membrane, Janus kinase (JAK) induces tyrosine residue phosphorylation (p-Tyr) at the receptor elements. STAT3 then docks to p-Tyr by employing its Src homology 2 (SH2) domain, and then STAT3 becomes a Tyr substrate. Thus, STAT3 p-Tyr phosphorylation induces homodimerization via reciprocal phosphotyrosine-SH2 interactions between STAT3 monomers. Then, STAT3 homodimer translocates to the nucleus and acts as a transcription factor[93-95]. Once in the nucleus, STAT3 homodimers regulate the expression of different genes such as c-MYC, cyclin D1, CCND1, BCL2L1, VEGF, IL-8, MMP-2, MMP-9, Bcl-xL, Bcl-2, and survivin. STAT3 regulates the expression of genes related to apoptosis, cell growth, survival, the cell cycle, cell differentiation, proliferation, metastasis, invasion, migration, angiogenesis, immune activation, and inflammation[96]. Under normal conditions, p-STAT3 expression is transient; however, in multiple cancer cell lines, p-STAT3 remains phosphorylated and activated. STAT3 activation presents an opportunity in numerous cancer research fields[97,98].

STAT signaling plays a vital role in the development and progression of CRC[99]. Different reports have pointed out that STAT signaling activation increases the regulation of a particular gene expression signature in CRC compared to normal tissues, including IFNL3, IFNE, CSF2, IFNL2, IL23A, AGT, IL20, OSM, leukemia inhibitory factor (LIF), IL13, PRL, EPO, HAMP, CENPJ, MGAT5, and PIGU[100]. These up-regulated genes may also be involved in CRC survival. However, to determine the role of each upregulated gene in CRC progression, further studies are required. Additionally, it is not clear which STAT isoform regulates the upregulation of this battery of genes; therefore, it is important to continue focusing efforts to address this issue. One possible mechanism is the constitutive activation of STAT3 in cancer[97]. A relationship has been reported between high p-STAT3 levels and poor prognosis in CRC patients, including poor survival and metastasis[101]. Moreover, an increase in p-STAT3 levels is related to radiotherapy resistance. It has been reported that in the CRC cell lines HCT-116 and LoVo, and CRC patient-derived cells, p-STAT3 is increased with radiotherapy, which confers resistance to apoptosis; this resistance to the therapy may induce CRC progression[102]. It has also been reported that LYN mediates p-STAT3, which regulates the EMT, a fundamental process in CRC metastasis[103]. In addition, it has been observed that STAT3 regulates the overexpression of Snail in CRC by inducing EMT processing[104].

STAT3 is also critical in regulating different signaling pathways, for example, the JAK/STAT3, EGF/STAT3, and Abl/Src pathways[105]. In CRC, it has been reported that PDGF induces the activation of the Abl/Src pathway, which, in turn, mediates the EMT and, therefore, invasion and metastasis[106]. PDGF also participates in the mechanism of resistance to treatments, as observed in RKO cells, in which the transfer of p-STAT3 through exosomes, which reverses STAT3 activity inhibition, induces chemoresistance to 5-fluorouracil. Therefore, these data suggest a possible relationship between p-STAT3 activity and resistance to 5-fluorouracil treatment in CRC cells.

Radiotherapy is a treatment option for patients with advanced CRC. However, similar to other types of cancer, a small population of cells can become resistant to radiotherapy and develop a more aggressive phenotype[107-110]. Recent reports have shown that STAT3 directly interacts with the CCND2 promoter to induce its expression. CCND2 belongs to the cyclin family and regulates the transition from the G1 phase to the S phase. Additionally, it has been observed that radiotherapy can induce STAT3 activation in CRC cells, thereby promoting cell cycle progression. These data suggest that STAT3 activation is an important mechanism of resistance to radiotherapy in patients with CRC[102,111].

In normal tissues, fibroblasts are responsible for remodeling the ECM by secreting proteases, collagen, tenascins, and periostins[112]. However, in the TME, fibroblasts transform into CAFs and play a fundamental role in tumor establishment and progression[113]. Excessive activation of STAT3 in CAFs, mediated by IL-6 and IL-11, is associated with poor prognosis in patients with CRC. Additionally, co-culture of CAFs with oral cancer cells has been shown to induce STAT3 activation through the production of both CCL2 and reactive oxygen species[114]. However, this mechanism has not been explored in CRC and could involve interactions between CAFs and CRC cells to promote tumor progression.

STAT3 could regulate CRC interaction with the microenvironment and then cancer progression. CRC interacts with ECM components through different integrins such as αVβ3 and αVβ5; these integrins then activate focal adhesion kinases, leading to the activation of Src family kinases, which regulate the activation of YAP/TAZ signaling and induce the production and secretion of IL-6. Consequently, IL-6 induces collagen expression in CAFs, promoting the interaction of the tumor with the ECM[115,116]. Cell-to-cell interaction between CAFs and tumor cells is a fundamental mechanism in tumor progression, and it has been reported that STAT3 may be involved in the regulation of this interaction. Immunohistochemistry shows that α-SMA is expressed in the cytoplasm of stromal cells while the transcription factor STAT3 is expressed in the nucleus of both glandular tumor cells and stromal cells (Figure 2).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Immunohistochemistry detection of alpha-smooth muscle actin and the signal transducer and activator of transcription 3 in representative images of colorectal cancer. A: Cytoplasmic expression of alpha-smooth muscle actin (α-SMA) in stromal cells; B: Diffuse expression of activated smooth muscle actin and the transcription signaling pathway 3 (STAT3) in the nuclei of both glandular and stromal neoplastic cells. Black arrows (α-SMA-positive stroma), orange arrows (STAT3-positive tumor and stroma). Magnification 20 ×. Scale bars: 200 μm.

CAFs comprise diverse heterogeneous cell subpopulations and show plasticity. Studies have described that inflammatory tumor-associated fibroblasts (iCAFs) from PDAC can be converted to tumor-promoting myofibrolasts (myCAFs) through the activation of TGF-β that inhibits IL 1 receptor type 1 (IL-1R1). This induces the JAK/STAT signaling pathway that activates the iCAFs through the stimulation of a cytokine cascade after the activation of NF-κB that acts predominantly by LIF. The JAK/STAT signaling pathway maintains the inflammatory phenotype of CAFs through a positive feedback loop involving STAT3 via the upregulation of IL-1R1. MyCAFs have been located adjacent to tumor cells. In contrast, iCAFs were located away from the tumor within the DR zone, suggesting that different phenotypes were associated with their spatial distribution (Figure 3). Interestingly, an intermediate population between iCAF and myCAF expressing α-SMA+/p-STAT3+ has been investigated as a potential target in immune therapy in solid tumors[117]. Even in CRC, the CAF-A subpopulation may be converted to CAF-B when healthy fibroblasts are transformed into CAFs[28].

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Plasticity between cancer-associated fibroblasts[117]. A: This model demonstrates the antagonistic signaling pathway that determines the ability of cancer-associated fibroblasts (CAFs) to transform into tumor or inflammatory promoter phenotypes. Two subtypes of CAF have been reported in the different zones within the tumor microenvironment in pancreatic ductal adenocarcinoma; inflammatory tumor-associated fibroblasts located in the tumor periphery, and tumor-promoting myofibrolasts (myCAFs) close to the tumor; B: Tumor cells secrete transforming growth factor β ligand (TGF-βL), which allows activation of TGF-β receptor (TGF-βR) in adjacent myCAFs, inhibiting interleukin 1 receptor type 1 (IL-1R1) expression; C: Tumor cells secrete IL-1 alpha (IL-1α) which activates the IL-1 signaling pathway in CAFs. IL-1 signaling induces a cytokine cascade that includes IL-6 and C-X-C chemokine ligand 1, allowing activation of Janus kinase/transcription signaling pathway (JAK/STAT) signaling through nuclear factor kappa B and autocrine leukemia inhibitory factor signaling. Activated JAK/STAT signaling establishes a positive feedback loop involving transcription signaling pathway 3 (STAT3) by upregulation of IL-1R1 expression. Created with BioRender.com. iCAF: Inflammatory tumor-associated fibroblasts; IL: Interleukin; JAK/STAT: Janus kinase/transcription signaling pathway; myCAFs: Tumor-promoting myofibrolasts; LIF: Leukemia inhibitory factor; CXCL: C-X-C chemokine ligand; TGF-β: Transforming growth factor β; NF-κB: Nuclear factor kappa B. Citation: Biffi G, Oni TE, Spielman B, Hao Y, Elyada E, Park Y, Preall J, Tuveson DA. IL1-Induced JAK/STAT Signaling Is Antagonized by TGFβ to Shape CAF Heterogeneity in Pancreatic Ductal Adenocarcinoma. Cancer Discovery 2019; 9 (2): 282-301, with permission from American Association for Cancer Research (Supplementary material).

In pancreatic ductal cancer, CAFs form fibrotic tissue around the tumor, which acts as a barrier against the immune system and provides cancer resistance to treatment. However, this hinders the entry of nutrients and oxygen into the TME[118]. In addition, both IL-6 and IL-17 secretion induced activation of STAT3 in CAFs, and activation of STAT3 regulated the expression of hypoxia-inducible factor 1 (HIF-1), which is required in response to the low-oxygen TME, suggesting that STAT3 and HIF-1 could activate the mechanism of autophagy in CAFs[119-121]. In the TME, autophagic CAFs are an important source of metabolites such as lactate, ketones, and alanine[122-124]. In addition, it has been described that in both ovarian and head and neck cancers, in response to the aforementioned metabolites, cancer cells secrete IL-6, which promotes a feedback loop in the TME[125,126].

It has been documented that STAT3 is of paramount importance in the mechanism of evasion of the immune system in cancer, a crucial mechanism for tumor progression. Secretion of cytokines and growth factors by both tumor cells and CAFs induces the recruitment of immune cells, inducing the activation of STAT3. In addition, the activation of STAT3 in immune cells induces the production of more proinflammatory cytokines, which perpetuate inflammation in the TME[127,128]. In patients with CRC, it has been observed that the presence of M2 macrophages is associated with low disease survival[129]; the secretion of both IL-6 and macrophage colony-stimulating factor (CSF-1) by CAFs induces polarization to M2 via STAT3 activation[37,130]. In addition, it has been observed that M2 macrophages also promote the transformation of fibroblasts into CAFs mediated by the secretion of IL-6, which promotes STAT3 activation in fibroblasts[131]. Therefore, communication between CAFs and macrophages appears to be crucial for regulating immune system evasion during tumor development and progression. However, the role played by other immune cells and whether STAT3 is involved in the recruitment, differentiation, and activation mechanisms have yet to be fully elucidated.

As described above, STAT3 is involved in multiple processes that are required for CRC progression. However, many of these processes also occur regularly; for example, during the inflammatory process, which is necessary to stop infection. STAT3 plays an essential role[129,132]. STAT3 has also been shown to be significantly involved in tissue repair[133]. In addition, STAT3 regulates the expression of genes related to apoptosis, cell growth, survival, cell cycle, cell differentiation, and angiogenesis[134]. Therefore, it is of utmost importance to develop strategies focused on regulating the functions mediated by STAT3. However, this goal is not simple. STAT3 represents a valuable opportunity for the treatment of not only CRC but also multiple tumors and other diseases related to its dysregulation.

One of the concerns in CRC and other tumors is the acquisition of malignancy by tumor cells, which makes treatment challenging and frequently leads to relapse in patients[135]. Metastasis and invasion are events in tumor malignancy that require diverse processes, changes in tumor cell biology, and various histological reactions[136]. The DR, mainly promoted by CAFs, appears to be a central component of the TME involved in cancer progression, and the acquired capacity for invasion and metastasis through extensive interactions with cancer cells and other stromal cells is increasingly recognized[137]. As previously mentioned, CAFs are highly heterogeneous cells, and their crosstalk with cancer cells is mediated by complex signaling pathways such as TGF-β, PI3K/AKT JAK/STAT, and NF-kB, among others[138]. Therefore, therapies that aim to control these cell populations by regulating key signaling pathways involved in CAF development could be an excellent strategy to mitigate the growth and metastasis of CRC.

Previous reports have indicated that the JAK/STAT signaling pathway is constitutively activated in CAFs in a positive feedback loop, promoting the activation of CAF-derived cytokines, such as IL-6, IL-10, IL-11, and IL-22, which act as ligands for the JAK/STAT signaling pathway[138,139]. The relationship between STAT3 and CAF activity has been widely studied[140,141]. Studies on primary cell cultures from breast cancer tissues indicated that the inhibition of STAT3 activity through 5-azacytidine and ruxolitinib treatment led to the inhibition of the tumor-promoting invasive phenotypes of CAFs[142]. The use of a selective inhibitor of IGF-1R and STAT3 (NT157) in a mouse model of sporadic colorectal tumorigenesis significantly decreased the tumor burden by affecting cancer cells, myeloid cells, and CAFs[143]. Additionally, a decrease in cell proliferation and increase in apoptosis were related to the inhibition of CAF activation by a STAT3 inhibitor. Hence, inhibition of the activity of STAT3 in tumors impacts the regulation of processes such as proliferation and metastasis and influences the regulation of cells in the TME.

Patients with desmoplastic tumors have poor survival rates. These tumors are characterized by the presence of the most abundant cell type in all TMEs, CAFs[144]. Some reports have indicated that natural and synthetic STAT3 inhibitors could be promising candidates for modulating DR s in CRC. A previous report using a non-conventional treatment with helminth-derived molecules in a mouse model of colitis-associated CRC demonstrated the capacity to inhibit the activity of the STAT3/NF-kB signaling pathway, decreasing colon tumorigenesis[145]. Even though the use of these molecules in CAF regulation has not been explored, the evident inhibitory effect on STAT3 activity suggests the potential use of these molecules in the regulation of CAFs, which needs to be analyzed further. Curcumin is a natural product that inhibits CAF activity, blocks JAK/STAT3 signaling, and promotes the recovery of 5-fluoruracil sensitivity in chemoresistant gastric cancer cells[146]. Moreover, studies found that an anti-IL-17a neutralizing antibody or AG490, an inhibitor of JAK2, attenuated CAF activity over STAT3, leading to a significant decrease in the migration and invasion of gastric cancer cells as a consequence of the reduced activity of this molecule[147]. In pancreatic cancer, quercetin acts as an inhibitor of the STAT3 signaling pathway, triggering the inhibition of the EMT, invasion, and metastasis[148]. These data are in accordance with a report on desmoplastic tumors, in which this natural compound downregulated CAF-induced cancer drug resistance[144]. In lung cancer, an aptamer-based conjugate (Gint4.T-STAT3) containing a STAT3 siRNA that inhibited PDGFRβ resulted in both STAT3-specific silencing and interfered with CAF-pro-tumorigenic functions[149].

Inhibition of STAT3 activity is a favorable strategy for the management of different types of cancer, such as non-small cell lung cancer, gastric cancer, and CRC, for two main reasons: (1) The continuous activation of STAT3 regulates the expression of downstream proteins associated with carcinogenesis, progression, metastasis, and cancer survival[150,151]; and (2) the master regulator STAT3 mediates crosstalk between cancer cells and their microenvironment, driving tumor immune escape to regulate the function of stromal cells, particularly immune cells and CAFs. Additionally, it has been demonstrated that STAT3 hyperactivity induces TME interstitial remodeling, reducing the immune response by immune escape through increased expression of immune checkpoint molecules (such as CTLA-4, PD-1, PD-L1, and B7-H4)[152].

Therefore, it is possible that the currently used conventional chemotherapies and immunotherapies could be complemented with STAT3 inhibitors, and their effects could be improved.

Despite extensive evidence for using STAT3 inhibitors as a therapy, few molecules have reached the clinical use phase, mainly due to their limited stability, low affinity, cell permeability, and bioavailability, among other events[153]. However, recently, small-molecule inhibitors targeting STAT3, TTI-101, OPB- 31121, and OPB-111077, have advanced to clinical trials. TTI-101 is a selective small molecule that binds to STAT3 and prevents its phosphorylation, homodimerization, nuclear translocation, and STAT3-mediated transcriptional activity[154]. TTI-101 showed no toxicity and evidence of clinical benefit in a phase I study in patients with solid tumors who were refractory to prior therapies, similar to OPB-31121 in patients with CRC and other solid tumors[154,155].

Napabucasin (BBI608) is a novel oral small-molecule STAT3 inhibitor evaluated in a broad spectrum of solid tumors, including CRC[156-159]. Multicenter phase I/II trials of napabucasin and pembrolizumab, a blocker of PD1, in patients with metastatic CRC found that the combination of the two molecules showed antitumor activity with acceptable toxicities for patients with metastatic CRC, including those with high microsatellite instability (MSI-H); however, further investigations are required[159].

In summary, therapeutic strategies aimed at inhibiting STAT3 activity have demonstrated effectiveness in increasing the chemosensitivity of refractory tumor cells and modulating the TME, reducing factors that promote tumor metastasis via reduction of the EMT and immune modulators such as CAFs, and reducing the expression of anti-apoptotic molecules such as Bcl-2 and overexpression of caspase 3[160,161]. However, further clinical trials of pharmacologic STAT3 inhibitors and new alternatives are necessary. Natural compounds with inhibitory properties against STAT3 could act as template compounds for the synthesis of more efficient molecules[153] for the treatment of patients with CRC and other tumors characterized by significant expression of STAT3.

Other members of the STAT family have recently been analyzed for their possible roles in CRC progression, primarily at the experimental level. For example, STAT6, a critical molecule for IL-4 and IL-13 signaling, has been demonstrated by different authors to be a crucial signaling pathway that favors CRC progression. Animals deficient in STAT6 develop fewer colon tumors and mount a better immune response[162]. Moreover, using a chemical inhibitor of STAT6 together with 5-fluorouracil in wild-type animals with CRC at advanced stages reduced the tumor load by almost 75% compared to that in control mice[163]; however, whether these treatments affect the composition of CAFs or the DR requires further research.

The DR, in addition to promoting tumor progression and resistance to chemotherapy and radiotherapy, is fundamental for the development of local immunosuppression, which further encourages the growth of tumor cells that can enter the circulation and invade other organs, resulting in poor prognosis. CAFs are a heterogeneous population with different subtypes with pro-tumor, immunosuppressive, antigen-presenting, and tumor cell growth restriction functions. It is difficult to treat the DR directly; it is counterproductive to eliminate it as doing so results in greater aggressiveness and the ability of tumor cells to generate metastatic lesions in less time. This could be because some CAFs restrict tumor growth. In contrast, others are promoters of tumor development, and by eliminating CAFs in a general way, there are already tumor cells ready to enter the circulation and colonize a secondary organ. The study of the mechanisms involved in activating CAFs has opened an opportunity to consider strategies that target therapies to mitigate their promoter and immunosuppressive activities without affecting their tumor cell growth-restriction activity. However, these treatments are still in the testing phase owing to their toxicity to healthy cells. One of the mechanisms involved in the activation of CAFs is the JAK/STAT signaling pathway, and STAT3 has been reported in several studies to be closely related to the development of DR and tumor malignancy. Therefore, the use of synthetic or natural-origin inhibitors of STAT3 to modulate the activity of CAFs has been suggested; however, the use of these molecules to inhibit STAT3 is still being studied. This could significantly affect the regulation of CAFs by these molecules and should be studied in in vivo models, where the TME in which these cells interact can be analyzed in greater depth. Over time, this could be one of the alternatives used in therapy to reinforce conventional treatment; however, even though some molecules are currently used in the clinic and have positive results, there is still much to learn about these STAT3 inhibitors, how they work at the clinical level, and the potential side effects. More studies are needed on solid tumors and CRC to identify which CAF subtypes are acting and how this, in turn, modulates the activity of other subtypes in the complexity of the DR.

We thank the excellent technical assistance of Alam Palma Guzmán, member of National Laboratory of Advanced Microscopy-IMSS, National Medical Center, Siglo XXI, Mexico City. Dra. María De Los Ángeles Hernández Cueto, Central Epidemiology Laboratory, National Medical Center la Raza, IMSS. Both for their support in the immunohistochemistry assays and photomicrography.