Peptide-Drug Conjugate for Therapeutic Reprogramming of Tumor-Associated Macrophages in Breast Cancer

Abstract

In triple-negative breast cancer (TNBC), pro-tumoral macrophages promote metastasis and suppress the immune response. To target these cells, a previously identified CD206 (mannose receptor)-binding peptide, mUNO was engineered to enhance its affinity and proteolytic stability. The new rationally designed peptide, MACTIDE, includes a trypsin inhibitor loop, from the Sunflower Trypsin Inhibitor-I. Binding studies to recombinant CD206 revealed a 15-fold lower K_D for MACTIDE compared to parental mUNO. Mass spectrometry further demonstrated a 5-fold increase in MACTIDE's half-life in tumor lysates compared to mUNO. Homing studies in TNBC-bearing mice shows that fluorescein (FAM)-MACTIDE precisely targeted CD206⁺ tumor-associated macrophages (TAM) upon intravenous, intraperitoneal, and even oral administration, with minimal liver accumulation. MACTIDE was conjugated to Verteporfin, an FDA-approved photosensitizer and YAP/TAZ pathway inhibitor to create the conjugate MACTIDE-V. In the orthotopic 4T1 TNBC mouse model, non-irradiated MACTIDE-V-treated mice exhibited anti-tumoral effects comparable to those treated with irradiated MACTIDE-V, with fewer signs of toxicity, prompting further investigation into the laser-independent activity of the conjugate. In vitro studies using bone marrow-derived mouse macrophages showed that MACTIDE-V excluded YAP from the nucleus, increased phagocytic activity, and upregulated several genes associated with cytotoxic anti-tumoral macrophages. In mouse models of TNBC, MACTIDE-V slowed primary tumor growth, suppressed lung metastases, and increased markers of phagocytosis and antigen presentation in TAM and monocytes, increasing the tumor infiltration of several lymphocyte subsets. MACTIDE-V is proposed as a promising peptide-drug conjugate for modulating macrophage function in breast cancer immunotherapy.

Keywords: CD206; peptide‐drug conjugate; targeting peptides; triple negative breast cancer; tumor‐associated macrophages.

Figure 1

Design and molecular dynamics of MACTIDE. A) MACTIDE structure with the CD206‐binding motif mUNO in red. B) RMSD to the average structure for heavy atoms of mUNO (red lines) and MACTIDE (green line), in two 400 ns molecular dynamics for mUNO and MACTIDE in solution. C) The two most populated conformations of MACTIDE, C0, and C1, with the mUNO motif in red. D) RMSF of each residue for MACTIDE in solution. E) Intra hydrogen bonds in MACTIDE (dashed blue lines), between Pro7‐Cys10 with an occupancy of 75% and between Ile6‐Thr3 with an occupancy of 10% (present in the C1 conformation); the yellow line indicates the disulfide bond.

Figure 2

Docking shows high binding energy of MACTIDE to CD206 and ligand‐induced conformational change. A) CTLD2 domain for cluster node 2 (green) compared with cluster node 0 (purple). MACTIDE is represented as VDW golden spheres. A large alpha helix displacement of 3.8 Å was found in node 2 making space for the peptide to bind. B) Node 2 compared to the crystal structure 5XTS and colored by RMSD. Significant differences (in red) were found in the CysR domain. C) Hydrogen bonds (dashed blue and red lines) formed at the docking pose, where MACTIDE is shown in yellow and CD206 in cyan.

Figure 3

MACTIDE has higher affinity and proteolytic stability than mUNO. A) QCM experiment of MACTIDE, mUNO, and control peptides at a final concentration of 10 µm in PBS on multilayers of PAH/CD206 or control multilayers of PAH/BSA. The black arrow indicates when the peptides were added, blue and red arrows indicate when the washing step with PBS started. B) Integrity of FAM‐MACTIDE and FAM‐mUNO measured by LC‐MS at different time points after incubation of peptides with lysate derived from a 4T1 tumor.

Figure 4

FAM‐MACTIDE targets CD206⁺ TAM using different administration routes. Thirty nmol of FAM‐MACTIDE or FAM‐mUNO were administered i.v. (A), i.p. (B), or orally (C) and left to circulate for 24 h. After 24 h, mice were sacrificed, and the organs were collected, fixed, cryoprotected, sectioned, and immunostained for FAM (shown in green) and CD206 (shown in red). The CD206/FAM colocalization indices were calculated from representative images from n = 3 tumors, using Fiji (Mandler's tM2 index) and the signal intensity per CD206⁺ TAM (i.p. administration) was quantified using ImageJ (B). Scale bars = 100 µm. ^* p ≤ 0.05, ^** p ≤ 0.01 (Anova one‐way fisher LSD).

Figure 5

In vitro and in vivo photodynamic therapy with MACTIDE‐V. A) Structure of MACTIDE‐V. B) MACTIDE‐V, CtrlPep‐V, and DOX were incubated with primary human macrophages (obtained from monocytes derived from human blood buffy coat) at 30 µm for 1 h at 37 °C, followed by 2 washes with media. Then, the groups shown in pink were irradiated with 10 J cm⁻² (irradiance: 170 mW cm⁻², spot diameter: 0.5 cm). Then, the cells were incubated for 48 h and the cell viability was determined using the MTT((3‐(4,5‐Dimethylthiazol‐2‐yl)‐2,5‐Diphenyltetrazolium Bromide) assay. Here, the graph shows the average of n = 6 experiments. C) Treatment with MACTIDE‐V and CtrlPep‐V with and without laser (denoted by + and −, respectively), in mice bearing orthotopic 4T1 tumors (n = 6/group) in Balb/C. Peptide‐V conjugates were administered i.p. at a dose of 30 nmol (1 mg Kg⁻¹ in Verteporfin). For the irradiated groups, 4 h after administration of conjugates or PBS, the mice were irradiated with 100 J cm⁻²; shown is primary tumor volume progression during treatment. Gray arrows indicate injection days. D) Bodyweight during the treatment as % of the initial one. E) Representative mIHC images of PBS, CtrlPep‐V, and MACTIDE‐V‐treated tumors, 20X magnification, scale bar = 100 µm, white triangles point to Foxp3⁺ cells. F) CD8⁺ T cells, CD4⁺ T cells, and Treg density, n = 6 mice/group, with every dot representing the mean cell density of each tumor obtained from 20 images. G) CD8⁺ T cells density in CK8⁺‐dense tumor regions. n = 5 mice/group, with every dot representing the mean cell density of each tumor obtained from 20 images. Median ± interquartile range. One‐way ANOVA with multiple comparisons, ^* p ≤ 0.05, ^** p ≤ 0.01.

Figure 6

MACTIDE‐V promotes a YAP‐ mediated anti‐ tumoral switch in bone marrow‐derived macrophages. Day 5 BMDM were incubated for 4 h with 10 µm conjugates, washed, and followed‐up in media. A) YAP Immunofluorescence (shown in red) was analyzed 3 h after treatment and B) phagocytosis of fluorescent E. coli particles at 48 h follow‐up. C) Flow cytometry on BMDM 48 h after treatment. Left panel, GeoMean of MHC II, CD206, CD80, and PD‐L1 in Balb/c BMDM. Median ± interquartile range. Repeated measures one‐way ANOVA with multiple comparisons, for n = 4 independent experiments. ^* p ≤ 0.05, ^** p ≤ 0.01. D) Heatmap of the mRNA expression of genes involved in the functional activation of BMDM, YAP signaling, and adhesion, measured by real‐time PCR 48 h after treatment in BALB/c BMDM. n = 3 independent experiments.

Figure 7

MACTIDE‐V slows tumor growth and suppresses lung metastasis in orthotopic 4T1.2. 4T1.2 bearing mice were treated with 9 doses of CtrlPep‐V, MACTIDE‐V (1 mg Kg⁻¹ in Verteporfin per dose), or PBS i.p. every other day, while monitoring the primary tumor volume (A). Mice were sacrificed on day 25, tumor weights were measured (B) and pulmonary metastasis areas were quantified from H&E sections (C). D–G) Tumor cell suspensions were analyzed by flow cytometry (n = 4 per group). H) Representative mIHC images of PBS, CtrlPep‐V, and MACTIDE‐V treated tumors, scale bar = 100 µm. I) CD8⁺ T cells, CD4⁺ T cells, and Treg density obtained from mIHC images (n = 6 mice/group), with every dot representing the mean cell density of each tumor obtained from 20 images. Median ± interquartile range. One‐way ANOVA with multiple comparisons, ^* p ≤ 0.05, ^** p ≤ 0.01.

Figure 8

MACTIDE‐V‐treated tumors do not benefit from concomitant anti‐PD‐1 blockade in orthotopic 4T1.2. 4T1.2 bearing mice were treated with 9 doses of CtrlPep‐V, MACTIDE‐V (1 mg Kg⁻¹ in Verteporfin per dose), or PBS i.p. every other day according to the scheme in (A). Anti‐PD‐1 injections started 16 days post‐tumor induction (three injections of 200 µg each). Mice were sacrificed on day 29, their tumor weights were analyzed (B), pulmonary metastases area quantified from H&E‐stained sections (C), and their tumors were analyzed with flow cytometry (D–H).