Targeting stromal cell sialylation reverses T cell-mediated immunosuppression in the tumor microenvironment

Highlights

• •Tumor-conditioned stromal cells and CRC patient-derived CAFs are hypersialylated
• •Hypersialylated stromal cells and CAFs suppress T cell activation
• •Hypersialylated stromal cells induce Siglec receptor expression on T cells
• •Targeting sialylation activates T cells and reduces PD-1/Siglec-expressing CD8 T cells

Summary

Immunosuppressive tumor microenvironments (TMEs) reduce the effectiveness of immune responses in cancer. Mesenchymal stromal cells (MSCs), precursors to cancer-associated fibroblasts (CAFs), promote tumor progression by enhancing immune cell suppression in colorectal cancer (CRC). Hyper-sialylation of glycans promotes immune evasion in cancer through binding of sialic acids to their receptors, Siglecs, expressed on immune cells, which results in inhibition of effector functions. The role of sialylation in shaping MSC/CAF immunosuppression in the TME is not well characterized. In this study, we show that tumor-conditioned stromal cells have increased sialyltransferase expression, α2,3/6-linked sialic acid, and Siglec ligands. Tumor-conditioned stromal cells and CAFs induce exhausted immunomodulatory CD8⁺ PD1⁺ and CD8⁺ Siglec-7⁺/Siglec-9⁺ T cell phenotypes. In vivo, targeting stromal cell sialylation reverses stromal cell-mediated immunosuppression, as shown by infiltration of CD25 and granzyme B-expressing CD8⁺ T cells in the tumor and draining lymph node. Targeting stromal cell sialylation may overcome immunosuppression in the CRC TME.

Introduction

Immunosuppressive tumor microenvironments (TMEs) reduce the effectiveness of immune-based therapies for the treatment of cancer.¹ In colorectal cancer (CRC), the heterogeneous environment is composed of numerous cellular components that can facilitate an immunosuppressive microenvironment, including immune, stromal, and endothelial cells.² CRC can be classified into subtypes based on distinct molecular and clinical features, termed consensus molecular subtypes (CMSs).³ Four CMS groups are recognized (CMS 1–4), with differential prognosis.⁴ CMS4, characterized by a mesenchymal signature and reflecting an increased stromal cell content compared with other CRC subgroups, has been associated with poor prognosis.^{5, 6} Additionally, the type, density, and location of immune cells within tumors can predict clinical outcome and tumor recurrence.⁷ Lower densities of CD3 and CD8 immune cells identify patients with tumor recurrence compared with patients whose tumors did not recur.^4,7 The strongest predictors for poor prognosis for patients with CRC are associated with elevated levels of mesenchymal stromal cell (MSC) signatures,^6,8,9 lack of activated innate and adaptive immune cells,^{10,11,12,13,14} inflammation,¹⁵ and lack of neo-antigens in the TME.¹² Similarly, studies of multiple myeloma (MM), a cancer that arises in the bone marrow in a stromal cell-dense environment, indicates the complex role of stromal cells in regulating tumor progression and immune evasion.¹⁶ These observations indicate the importance of understanding the key cellular and molecular events that dictate immunosuppressive TMEs and tumor extrinsic mechanisms that influence tumor progression.

Non-hematopoietic intestinal MSCs, precursors to cancer-associated fibroblasts (CAFs), are major components of the CRC and MM TME.¹⁷ They are a heterogeneous population of stromal cells of mesenchymal lineage, characterized by a combination of morphological features, tissue origin, and lack of defined lineage markers.^18,19 Stromal cells in the TME arise from local tissue MSCs, fibroblasts, trans-differentiation events, and recruited bone marrow MSCs.²⁰

 The definition of cell types and subtypes within the classification of stromal cells is complicated by the lack of distinct markers.¹⁸ The high proportion, localization, and function of MSCs in CRC and other stromal-rich tumors suggest that these cells are crucial to tumor development.¹⁹ MSCs are positioned between the epithelial cells and the underlying vasculature and can passively or actively impair immune cell trafficking and activation.^17,21

 The immunological hallmarks of stromal cells in the TME include regulation of immune cell infiltration, regulation of anti-tumor immune responses, and responsiveness to immunotherapy.¹⁸ Stromal cell signatures in colon tumors and MM are associated with tumor progression, a poorer prognosis, immune evasion, and therapy resistance.^6,22,23

 It is unclear whether this association is due to inherent tumor-promoting or immunosuppressive functions in the TME.¹⁷ This knowledge highlights the need to investigate immunosuppressive mechanisms to improve discovery of effective stroma-targeting therapeutic strategies.

Chronic inflammation can alter glycosylation, and emerging knowledge on the role of glycosylation in tumor progression indicates its association with poorer prognosis.^24,25 One of the more common changes in cancer glycosylation is an upregulation of sialylated glycans (termed hypersialylation²⁶). Sialic acid is a common component of glycan molecules, and its presence can result in altered protein function and immune recognition.²⁶ A common development in tumor progression is the upregulation of sialic acid expression on cancer cells, which can be exploited to evade immune clearance.

 Sialoglycans are recognized by Siglec receptors (sialic acid-binding immunoglobulin-type lectins), which are expressed on the surface of both innate and adaptive immune cells.³¹

 Siglecs are a class of self-pattern recognition receptors that co-regulate the function of immune cells. Inhibitory Siglecs contain tyrosine-based inhibitory signaling motifs (ITIMs) that mediate inhibitory signals upon binding sialoglycans.³² Hypersialylation of glycans is linked to increased immune evasion, drug resistance, tumor invasiveness, and metastasis. The inhibitory Siglecs most strongly implicated in immune evasion in cancer are Siglec-7, Siglec-9, and Siglec-10, which are expressed on natural killer (NK) cells, macrophages, and T cells.

 In the TME, tumor cell sialic acid can drive immune cell differentiation by Siglec receptor engagement, indicating that cell-cell sialic acid interactions can dictate immune cell function.

Recent conflicting reports regarding the pro- and anti-tumor effects of targeting Siglec ligands, specifically on tumor cells, have indicated the complexity of sialic acid-dependent signaling in the regulation of tumor growth.

 These data reflect the heterogeneity of cell type-specific sialyation in the TME. Overcoming immunosuppression and enhancing immunotherapy responses is a key challenge in cancer treatment, and the role of stromal cell sialylation in this regard is unknown.

We propose that stromal cell sialylation in the TME is an unexplored immunological target to reverse TME immunosuppression. We hypothesized that the TME induces stromal cell sialylation, which in turn regulates stromal cell-mediated immunosuppression. Here, we show that tumor secretome can enhance stromal cell sialylation and Siglec-E (mouse), Siglec-7, and Siglec-9 (human) ligand expression, which is associated with CD8⁺ T cell suppression and exhaustion. Targeting sialyltransferase activity in stromal cells inhibits Siglec ligand expression and reverses T cell suppression and exhaustion. In a preclinical tumor model, we show that stromal cell sialylation can suppress T cell activation, function, and phenotype through cell-cell contact-dependent mechanisms, which is reversed by targeting sialyltransferase activity in stromal cells. Targeting stromal cell sialylation and/or Siglec-Siglec ligand interactions changes the immune contexture in stromal-dense tumors and may represent an innovative strategy to enhance anti-tumor immunity in immunosuppressive TMEs.

Results

Tumor-conditioned stromal cells have increased levels of cell surface α2,3- and α2,6-linked sialic acid

MSCs can sense and switch immune responses through secretion of soluble immunosuppressive molecules, as well as cell-cell contact-mediated immunomodulatory ligand expression.

 However, the impact of MSC glycosylation on their immunomodulatory potential is not well characterized. We therefore sought to assess the role of glycosylation on MSC-mediated suppression of T cell proliferation and activation. We used biotinylated lectins Concanavalin A (Con A), Galanthus nivalis (GNA), wheat germ agglutinin (WGA), and Sambucus nigra (SNA-I) to assess the expression levels of the respective glycan structures by flow cytometry.. A schematic depicting the preferential binding of each lectin is shown in Figure 1A. The sialic acid-binding lectins SNA-I and WGA showed the highest level of expression on MSCs compared with control unstained MSCs (Figure 1B). The two most common glycosidic linkages of sialic acid are α2,3 and α2,6. As shown in the schematic in Figure 1C, we conditioned murine MSCs with CT26 tumor cell secretome (CT26 MSCs^TCS) and analyzed sialic acid expression by incubating cells with the α2,3- and α2,6-binding lectins Maackia Amurenesis (MAL-II) and SNA-I, respectively. Flow cytometric analysis confirmed significantly increased expression of α2,6, but not α2,3, sialic acid on MSCs^TCS (Figure 1D). Using RNA sequencing, we analyzed mRNA expression levels of enzymes that control sialylation, namely, α2,3- and α2,6-specific sialyltransferases. Figures 1E and 1F show differential expression of both α2,3 and α2,6-specific sialyltransferases, respectively, between TCS-conditioned and unconditioned MSCs. This highlights the complexity of regulation of sialic acid synthesis, as, apart from increased expression of St3gal4 by MSCs^TCS, there was no clear trend in sialyltransferase mRNA expression. Finally, MSCs^TCS inhibit both CD4+ and CD8+ T cell proliferation in co-culture assays compared with unconditioned MSCs (Figure 1G)…