A Structurally Simple Vaccine Candidate Reduces Progression and Dissemination of Triple-Negative Breast Cancer

Abstract

The Tn antigen is a well-known tumor-associated carbohydrate determinant, often incorporated in glycopeptides to develop cancer vaccines. Herein, four copies of a conformationally constrained mimetic of the antigen TnThr (GalNAc-Thr) were conjugated to the adjuvant CRM197, a protein licensed for human use. The resulting vaccine candidate, mime[4]CRM elicited a robust immune response in a triple-negative breast cancer mouse model, correlated with high frequency of CD4+ T cells and low frequency of M2-type macrophages, which reduces tumor progression and lung metastasis growth. Mime[4]CRM-mediated activation of human dendritic cells is reported, and the proliferation of mime[4]CRM-specific T cells, in cancer tissue and peripheral blood of patients with breast cancer, is demonstrated. The locked conformation of the TnThr mimetic and a proper presentation on the surface of CRM197 may explain the binding of the conjugate to the anti-Tn antibody Tn218 and its efficacy to fight cancer cells in mice.

Keywords: Cancer; Immunology; Medical Biochemistry.

Graphical abstract

Figure 1

Structure of Tn Antigen, TnSer, TnThr, and TnThr Mimetic, 1

Scheme 1

Synthesis of Activated Derivatives 2 and 3

Figure 2

CRM197 Decorated with 4, 19 and 34 Residues of Mimetic 1 or with Residues of Glucose CRM197 decorated with 4 residues (mime[4]CRM), 19 residues (mime[19]CRM), or 34 residues (mime[34]CRM) of TnThr mimetic 1, and CRM197 decorated with glucose residues (Glc-CRM, used as control, see Supplemental Information, Data S12).

Figure 3

Molecular Dinamic Simulation on Mimetic 1and mime[4]CRM (A) Representative conformer of TnThr (as a diamide derivative) in water solution, together with the structural ensemble derived from 0.5-μs MD simulation of mimetic 1. (B) Representative snapshot derived from 0.5-μs MD simulations of CRM197 (PDB CRM197: 4AE0) upon chemical modification with 4 molecules of mimetic 1. A root-mean-square deviation value of 4.7 ± 1.1 Å (protein backbone), relative to the starting structure, was derived from the MD simulations. CRM197 is shown as ribbons, and the unnatural residues are represented as sticks.

Figure 4

Affinity of Glycoconjugate mime[4]CRM to the Monoclonal anti-Tn Antibody Tn218 (A and B) SPR curves and fit obtained for mime[4]CRM (A) and isolated mimetic 1 (B) toward monoclonal antibody Tn218 (related to Figure S3).

Figure 5

Evaluation of Human DCs' Activation DCs isolated from three healthy donors have been stimulated with mime[4]CRM, Glc-CRM, or native Tn. Glc-CRM and native Tn showed some effect in DCs activation, whereas mime[4]CRM increased the expression of CD83 and CD86 markers. The data (median ±SD) of CD80, CD83, and CD86 expression on negative control (ctr−), positive control (ctr+, lipopolysaccharide), added with 0.2 mg/mL mime[4]CRM, with 0.2 mg/mL Glc-CRM, or with native Tn (0.05 or 0.2 mg/mL). The difference between mime[4]CRM treatment and the ctr− was assessed using paired t test; ∗p < 0.05 (see Supplemental Information for detail).

Figure 6

Description of the Vaccine Therapy Using Mime[4]CRM in Transplanted Mammary BC Cells and Their Rate of Proliferation In Vivo (A) Representative scheme for the in vivo preclinical trial showing 4T1-Luc cells' injection into the mammary fat pad of n = 19 syngeneic BALB/c mice (at T0), and the weekly subcutaneous administration of mime[4]CRM (17 mg/kg/weekly) started after 7 days from the time of cell implantation. CRM197 was administered to the control group. (B) Representative BLI images of mice orthotopically transplanted using 4T1-Luc cells implanted (n = 10 control mice, n = 9 treated mice). Mice were imaged every 7 days via in vivo BLI to monitor tumor growth from time of implantation (T0) to 28 days after tumor implantation. (C) Quantification of photon emission (p/s) from the region of interest (ROI) in mice treated with mime[4]CRM or with CRM197 as vehicle. The differences in total flux (photons/seconds, P/S) between the two groups of mice indicated a statistically significant reduction of tumor growth (14 days of treatment: 3.7-fold reduction; 28 days of treatment: 2-fold reduction) as measured by luminescence signal emission from tumorigenic 4T1-LUC cells after 1 week and 3 weeks from tumor injection (∗p < 0.01, ∗∗p < 0.05, see Supplemental Information for detail). The values are expressed as mean ± SD. (D) Evaluation of drug toxicity using mice body weight in mime[4]CRM-treated and control mice. The values are expressed as mean ± SD. No significant differences were observed between the two groups (related to Figure S5).

Figure 7

Representative Immunofluorescence (IF) Staining of Primary Mammary Tumor Sections from Transplanted Mice Treated with Mime[4]CRM or with CRM197 as a Vehicle (A–F) IF double staining with (A) T cell receptor (TCR, marker of T cells, red) and CD4 (expressed by both T helper and dendritic cells, green), (B) CD11c (marker of DCs, red) and CD4 (green), (C) programmed cell death protein 1 (PD1, marker of immunosuppressive immune cells, red) and CD4 (green), (D) FOXP3 (marker of regulatory T cells, Tregs, red) and CD4 (green), (E) CD163 (specific marker of M2-polarized macrophages, red) and CD68 (marker of macrophages, green), and (F) CD163 (red) and F4/80 (marker of macrophages, green). DAPI was used to stain the nuclei (blue) in yellow or in pink, the overlays indicating (A) CD4+TCR+ (T lymphocytes), (B) CD4+ CD11c+ (DCs), (C) CD4+ PD1+ (immunosuppressive lymphocytes), (D) CD4+ FOXP3+ (T regulatory cells), and (E and F) CD68+ CD163+ and F4/80+ CD163+ cells (M2 macrophages). Magnification 63×. (a–f) Graphs showing the percentage of positive cells counted in fluorescence staining by using ImageJ software (n = 3 different sections for each tumor were screened within the tumor mass and images and cell positivity were counted). Mice treated with mime[4]CRM, instead of CRM197, showed a statistically significant increase of T helper lymphocytes (CD4+ TCR+, ∗p < 0.03; a) and DCs (CD4+ CD11c+, ∗p < 0.04, b) and a reduction of immunosuppressive lymphocytes (CD4+ PD1+, ∗p < 0.04, c), T regulatory cells (CD4+FOXP3+, ∗p < 0.01, d), and M2 tumor-associated macrophages (TAMs; i.e., CD68+ CD163+, ∗p < 0.003 and F4/80+ CD163+ cells ∗∗p < 0.007, e and f). The values are expressed as mean ± SD. ∗p < 0.05; ∗∗p < 0.007. See Supplemental Information for detail.

Figure 8

In Vivo and Ex Vivo Analyses to Evaluate the Presence of Lung Metastases (A and B) In vivo BLI analyses showed only 50% of the mime[4]CRM-treated mice (four of eight animals) developed lung metastases (A), whereas the presence of metastatic foci was found in seven of nine control mice (i.e.,78%) at T42 (B). Ex vivo BLI analyses proved the presence of metastatic foci in eight of nine control mice (seven mice at T42 and one mouse at T54, B lower panel) and in four of eight mime[4]CRM-treated mice (one mouse at T42 and three mice at T54, A lower panel).