3.1. Stem Cells
The Wnt signaling pathway is known to be involved in regulating the self-renewal capacity of non-malignant stem cells. For example, an association between Wnt and adult stem cells was established with the gene disruption of mouse
TCF7L2. This gene encodes for T cell factor-4 (Tcf-4), which forms the β-catenin/Tcf-4 transcription complex. A study by Korinek et al. demonstrated that cessation of Wnt signaling resulted in a loss of intestinal stem cells, leading to a breakdown of the intestinal epithelium [
30]. Wnt signaling has also been found to help maintain the pluripotency of embryonic stem cells [
31]. In contrast, DKK1 overexpression, which results in Wnt pathway inhibition, has been shown to eliminate hair follicles and other skin appendages, suggesting a possible blockade of stem cell initiation [
32]. Intriguingly, many studies are now analyzing tumor stem cells as a source of immune evasion, as they have been shown to selectively acquire expression of CD80, a surface ligand that dampens immune recognition by binding Cytotoxic T lymphocyte Antigen-4 (CTLA-4) present on activated T cells [
33]. Overexpression of WNT5A, a common Wnt ligand, via epigenetic activation in glioblastoma is thought to lead to stem cell differentiation and invasive growth [
34]. Additionally, epithelial-to-mesenchymal transition (EMT) and metastasis are supported by WNT5A [
35]. In addition, mammary tumor stem cells were found to rely on Wnt proteins as rate-limiting self-renewal signals [
36]. There is evidence to support Wnt pathway inhibition with the downregulation of PD-L1 expression, associated with a decreased stemness score signature, in triple negative breast cancer [
37]. Potential relationships between cancer stem cells and CD8
+ tumor infiltration, as related to tumor PD-L1 expression and the effects on cancer progression, are also being investigated [
38]. Due to these documented effects of Wnt signaling on promoting stem cell viability and function, it is possible that aberrant Wnt signaling pathways may promote the stem-cell-like qualities of tumor stem cells, thereby facilitating intratumoral immune evasion.
With its role in cellular regulation, there is ample evidence that Wnt signaling affects hematopoietic stem cells. Wnt stimulation of hematopoietic stem cells may increase their self-renewal capacity [
39]. For example, limiting β-catenin activation via noncanonical Wnt signaling stimulation was shown to inhibit the differentiation of hematopoietic stem cells [
40,
41]. However, other animal studies where β-catenin was mutated did not show a significant change in hematopoiesis [
42,
43]. Induction of the Wnt/β-catenin pathway through inhibition of GSK3β or the use of WNT3A, a stimulating Wnt ligand, was found to arrest CD8
+ T cell differentiation into effector cells, promoting self-renewing multipotent CD8
+ memory stem cells and maintaining the “stemness” of mature memory CD8
+ T cells [
44]. This correlation at the cellular developmental level may provide one of the key links between Wnt signaling and alterations in immune functionality in the tumor.
3.2. Wnt Signaling and Cancer
The associations between Wnt signaling and cancer progression are the topic of intense investigation. This correlation was first established when a WNT factor gene was identified as an oncogene in the mouse mammary cell line RAC311c [
45]. Our understanding of these associations has now progressed to include the well-established relationship between APC gene mutations and colorectal cancers. APC gene mutations can be found in most sporadic colorectal cancers, and familial adenomatous polyposis, or the hereditary colon cancer syndrome [
46,
47]. Alterations in core Wnt regulators were found in a sequencing project of 1134 colorectal cancer samples, noting the incidence of oncogenic Wnt activation in 96% of human colorectal cancers [
48]. Axin2 gene mutations have been found in colorectal cancers as well [
49]. Hepatocellular carcinomas were found to have Axin1 mutations [
50]. Deletions of
GSK3B may lead hematopoietic stem cells to progress to acute myeloid leukemia [
51]. Mutations in these proteins cause a stabilization of cytoplasmic β-catenin due to an inappropriately functioning degradation complex, thus mimicking an upregulation of Wnt signaling. Furthermore, direct mutations in β-catenin have been found in colon cancer and melanoma [
52,
53]. It is known that β-catenin is also used in cellular adhesion junctions. Through immunohistochemistry evaluation of epithelial ovarian cancer samples, a build-up of β-catenin in the cellular membrane was associated with a decrease in progression-free survival, and resistance to platinum-based chemotherapy [
54]. Inactivating gene mutations in the pathway, such as Rnf43 in pancreatic cancer and Znrf3 in adrenocortical cancer or additional cancers, have implicated new links between Wnt signaling and cancer transformation [
55,
56]. While these direct changes of the Wnt pathway have been established in cancer progression, evolving investigations suggest more depth to Wnt’s role, specifically in regulating anti-tumor immunity.
3.3. Immunity in Cancer
Despite tremendous advancements in traditional chemotherapies, many oncology patients continue to experience rapid progression of disease characterized by chemotherapy resistance, which limits therapeutic options. This has led to the pressing need to investigate alternative methods of treatment, such as immune-directed therapies. Many key aspects of the relationship between the immune system and solid tumors have been elucidated over the last 20 years. There is an appreciable complexity of this system that results in either tumor suppression or tumor progression. It is now established that in order for immune cells to recognize cancer cells and control cancer progression and metastasis, they must first infiltrate the tumor, then remain activated in the TME [
57]. The presence and activation of these tumor infiltrating lymphocytes (TILs) typically correlates with tumors that are more sensitive to chemotherapeutic treatment [
58,
59]. However, if the tumor has been invaded by tumor-promoting T cells, such as Tregs, there may be decreased treatment sensitivity [
60]. In particular, higher frequencies of CD8
+ T cells, and high CD8
+/CD4
+ ratios have been correlated with improved overall survival in ovarian cancer [
61]. The main goal of immunotherapy is to convert the tumor milieu from an immunologically suppressed state to an inflamed state, for tumor recognition, cell destruction, and improved treatment sensitivity.
The Cancer Genome Atlas (TCGA) has provided much insight into the role of the immune system in various types of cancers through combined analysis of genomic and patient outcome data. With this resource, TME may be examined from a transcriptional viewpoint, permitting more nuanced cancer categorizations to be made. For example, Thorsson et al. recently performed an immunogenomic analysis of 10,000 tumors from 33 cancer types to identify underlying immune subtypes that are common to all cancers examined [
62]. The six subtypes, namely wound healing, interferon-γ dominant, inflammatory, lymphocyte depleted, immunologically quiet and transforming growth factor beta (TGF-β) dominant, help identify the immunological differences present in TME signatures. From this and similar studies, future therapies may be more directed toward the appropriate targets.
3.4. Wnt Signaling and Leukocyte Differentiation
The differentiation of multiple leukocyte populations is regulated by Wnt signaling pathways. Alterations in canonical Wnt signaling may have a genomic influence on T cell development. TCF1 and LEF1 genes have been linked to epigenetic changes that may promote CD8
+ T cell differentiation by repressing CD4 genetic networks [
63,
64]. Furthermore, deletions of
Ctnnb1, the gene encoding for β-catenin, or deletions in
Tcf7, the gene encoding TCF1, were shown to block thymocyte development [
65,
66]. When
Tcf7 is deleted in CD8
+ T cells, functional T cell memory is impaired [
67]. However, when p45, a TCF1 variant, is combined with stabilized β-catenin, there is an enhancement of central memory T cell production [
68]. Thus, there is support that genomic Wnt pathway alterations correlate with changes in T cell development and differentiation.
Other immune cell lineages are also influenced by canonical Wnt signaling. Innate lymphoid cells, including natural killer (NK) cells, require TCF1 for development, as shown by the earliest linage-specific precursor expressing high levels of TCF1, and defective NK cell survival in
Tcf7-/- mice [
69,
70]. In
Lef1-/- and
Fzd9-/- mice, B cell precursors were diminished in the bone marrow [
71]. However, B cell development was not impaired despite a
Ctnnb1 deletion in B cell precursors [
72]. Furthermore, dendritic cell (DC) differentiation is promoted when there is an upregulation of Fz receptors in DC precursors [
73]. The observed results of Wnt signaling alterations throughout multiple cell types remains vast, reflecting the importance of this pathway in immune cellular regulation.
3.5. Wnt Signaling and Immunomodulation
Multiple leukocyte populations within the TME are influenced by both Wnt stimulators and inhibitors. Key Wnt-related immunological alterations are illustrated in . For example, the Wnt/β-catenin inhibitor DKK1 has been found to be overexpressed in several different TMEs. High levels of DKK1 were found in serum samples of patients with pancreas, stomach, liver, bile duct, breast, and cervical carcinoma [
74]. One might speculate that this could be a negative feedback result of deregulated Wnt signaling from the tumor. However, DKK1 binding to its receptor cytoskeleton associated protein 4 (CKAP4) promoted tumor progression [
75]. Additional studies have shown tumor stroma-derived DKK1 targeted β-catenin downregulation in myeloid-derived suppressor cells (MDSCs), leading to an accumulation of these cells, a suppressed T cell response, and tumor proliferation [
76]. When a DKK1 vaccination was given in a murine model of myeloma, it was shown to elicit CD4
+ and CD8
+ T cell protective immunity [
77]. This insinuates a potential for a DKK1 vaccination as an immunotherapeutic adjunct. While the relationship between malignancy and DKK1 remains unclear, there appears to be a strong immunologic influence.
Figure 2. Wnt/β-catenin signaling and tumor immunomodulation. A magnified tumor cell (top left), illustrates Wnt signaling leading to elevated cytosolic and nuclear β-catenin. Increased Wnt signaling is associated with heightened survival of Tregs, skewed differentiation of CD4+ T cells to a pro-tumorigenic Th17 subtype, conversion of dendritic cells to a regulatory state with enhanced IL-10 and IL-12 secretion, and decreased effector differentiation and function in CD8+ T cells. DKK1 inhibits Wnt ligand/receptor interactions. Elevated DKK1 leads to an accumulation of MDSC in the TME and subsequent inhibition of effector CD8+ T cell function. Created with BioRender.com. Not drawn to scale.
Noncanonical pathway stimulation may support a tumor proliferative environment. WNT5A stimulation has been shown to increase IL-12 production from DCs, causing an increase in T
H1 responses [
78]. Alternatively, a deficiency of WNT5A in another model showed low levels of interferon gamma producing T
H1 cells [
79]. These Wnt effects on peripheral T cells may alter their functions in tumor recognition. For example, RAR-related orphan receptor C (RARC) was upregulated when CD4
+ T cells had sustained β-catenin activation, resulting in T
H17 polarization and production of proinflammatory cytokines that favor tumorigenesis [
80]. It has also been shown that Treg survival is increased with increased β-catenin expression [
81]. These recognized changes in tumor-infiltrating leukocytes provide insight into the influence of Wnt pathways on tumor immunity and provide a platform for new intervention concepts.
Immune tolerance is also impacted by Wnt signaling. Irregularities in antitumor cytotoxic T lymphocyte (CTL) priming are associated with Wnt signaling in DCs. The high levels of Wnt ligands found in the TME condition DCs to a regulatory state [
82] (). This suppressed antitumor immunity was explored via DC-specific LRP5/6 deletions in a murine tumor model. Results showed delayed tumor growth with enhanced effector T cell differentiation, and decreased Treg differentiation [
83]. This idea was mimicked pharmacologically via use of the PORCN inhibitor IWP-L6 [
83]. In one study, denilukin diftitox (ONTAK; a diphtheria toxin fragment/ IL-2 fusion protein) was given prior to DC vaccinations in patients with melanoma. This resulted in increased β-catenin in the skin and immune tolerance with an increased survival of resting Tregs [
84]. In another study, forced expression of non-degradable β-catenin in melanoma cells or DCs led to secretion of the anti-inflammatory cytokine IL-10, which impaired the ability of DCs to cross-prime CD8
+ CTLs for tumor recognition [
85,
86]. Through these influences on DCs, related to tumor-induced increases in β-catenin signaling, tumors may acquire tolerogenic characteristics, allowing immune evasion. However, in DC-β-catenin
-/- mice, with a CD11c-specific deletion of β-catenin, vaccination with tumor antigen failed to provide tumor protection. Interestingly, these mice were found to be deficient in CD8
+ T cell immunity [
86]. In a mouse model of human melanoma, the β-catenin-dependent transcription blocker, PKF115-584, stimulates DCs to cross-prime tumor-specific CTLs, altering Wnt-induced immunosuppression and improving therapeutic response [
85]. These findings suggest complex roles for β-catenin in DCs, where an aberrant amount, through either depletion or overabundance, may lead to immune tolerance.
Evidence suggests a strong correlation between Wnt pathway changes and immune exclusion. This was recently evaluated with TCGA data. Multiple cancers were analyzed for gene expression of TILs and categorized into a high, intermediate, or low T cell-inflamed tumor environment, based on their expression for genes associated with T cell infiltration. These tumors were then profiled for Wnt/β-catenin-related gene expression profiles. Up to 90% of tumor types showed an inverse correlation between Wnt/β-catenin pathway activation and a T cell-inflamed gene expression signature [
87], suggesting that Wnt signaling in the TME suppresses T cell infiltration and/or function. Additional support of immune exclusion was found through a mouse melanoma model that was engineered to express β-catenin. These cells were unable to express C-C motif chemokine ligand 4 (CCL4), leading to decreased CTL infiltration into the TME due to defective recruitment of DCs [
88]. Furthermore, in patients with primary and metastatic melanomas treated with BRAF inhibitors, tumor immune infiltration and survival were inversely correlated with β-catenin signaling [
89,
90]. Collectively, these findings imply a role for altered β-catenin levels in the exclusion of TILs in the TME.
Immunoevasion mechanisms may also be represented through alterations in immune checkpoint molecules due to Wnt signaling component changes.
GSK3 inactivation in mouse melanoma resulted in a repression of the PD-1 gene, allowing an improved CD8
+ response [
91]. However, in mouse mammary carcinoma, GSK3β was shown to interact with PD-L1, inducing its degradation, which led to increased CTL infiltration [
92]. As immune therapies become increasingly important in cancer therapeutic options, these interactions need to be further investigated.
3.6. Immunotherapy
Some immunotherapies are directed at the proteins that regulate T cell function and cytolytic activity. In particular, immune checkpoint inhibitors (ICIs) are monoclonal antibodies against receptors such as CLTA-4 and Programmed Death-1 (PD-1) that act to down-modulate T cell effector function. ICI are now approved for the treatment of malignant melanoma, non-small-cell lung cancer, classical Hodgkin lymphoma, head and neck squamous cell carcinoma, urothelial carcinoma, and renal cell carcinoma [
93]. In addition, anti-programmed death-1 ligand (PD-L1) has also shown many promising clinical results [
94]. Atezolizumab is an FDA-approved PD-L1 inhibitor that has shown improved progression-free survival and overall survival when used in combination with nabpaclitaxel for metastatic triple negative breast cancer [
95]. These results were in patients with known positive PD-L1 tumors. Other anti-PD-L1 therapies include durvalumab and avelumab. Additionally, the monoclonal antibody pembrolizumab (anti-PD-1) was combined with platinum-based chemotherapy in metastatic non-small-cell lung cancer treatment in those who lacked targetable gene mutations. The combination of therapies increased progression-free and overall survival [
96].
In a similar fashion, ipilimumab and tremelimumab are monoclonal antibodies to CTLA4, which normally functions to restrain effector T cell activation. Ipilimumab is known for its significant progression-free and overall survival advancement in melanoma [
97]. Further improvements in metastatic melanoma were seen following a combination of ipilimumab and anti-PD-L1 therapy [
98].
With the rising use of mono- and combinatorial immunotherapy in the clinical setting, there is a need to further understand the TME in patients who do not respond. It has been reported that decreased TILs result in resistance to ICIs [
99]. Wnt pathway regulation may play a role in this lack of response, as tumor-intrinsic, active β-catenin signaling in human melanoma has been associated with resistance to anti-CTLA4 and anti-PD-L1 antibodies due to T cell exclusion [
88]. It has also been speculated that tumor cells may lack the antigens needed for recognition by TILs, perhaps leading to immunotherapy resistance in these cancers. This was investigated with an analysis of 266 melanoma tumor samples from TCGA. Tissues were divided into categories based on a high or low expression of genes that were associated with T cell infiltration. This was correlated with nonsynonymous somatic mutations as a representative of mutational neoantigens that the tumor may possess. It was concluded that the change in tumor gene expression of infiltrating T cells did not correlate with increased mutational neoantigens [
100]. However, gene signatures of these melanoma tissues did support prior evidence of a correlation between Wnt/β-catenin pathway activation and a reduction in T cell infiltration in the tumor [
100]. This evidence, in addition to prior stated support, shows a strong correlation between increased Wnt signaling and decreased T cell infiltration in tumors.