Targeting of M2-like tumor-associated macrophages with a melittin-based pro-apoptotic peptide

Abstract

Background: Tumor-associated macrophages (TAMs) are the major component of tumor-infiltrating immune cells. Macrophages are broadly categorized as M1 or M2 types, and TAMs have been shown to express an M2-like phenotype. TAMs promote tumor progression and contribute to resistance to chemotherapies. Therefore, M2-like TAMs are potential targets for the cancer immunotherapy. In this study, we targeted M2-like TAMs using a hybrid peptide, MEL-dKLA, composed of melittin (MEL), which binds preferentially to M2-like TAMs, and the pro-apoptotic peptide d (KLAKLAK)₂ (dKLA), which induces mitochondrial death after cell membrane penetration.

Methods: The M1 or M2-differentiated RAW264.7 cells were used for mitochondrial colocalization and apoptosis test in vitro. For in vivo study, the murine Lewis lung carcinoma cells were inoculated subcutaneously in the right flank of mouse. The dKLA, MEL and MEL-dKLA peptides were intraperitoneally injected at 175 nmol/kg every 3 days. Flow cytometry analysis of tumor-associated macrophages and immunofluorescence staining were performed to investigate the immunotherapeutic effects of MEL-dKLA.

Results: We showed that MEL-dKLA induced selective cell death of M2 macrophages in vitro, whereas MEL did not disrupt the mitochondrial membrane. We also showed that MEL-dKLA selectively targeted M2-like TAMs without affecting other leukocytes, such as T cells and dendritic cells, in vivo. These features resulted in lower tumor growth rates, tumor weights, and angiogenesis in vivo. Importantly, although both MEL and MEL-dKLA reduced numbers of CD206⁺ M2-like TAMs in tumors, only MEL-dKLA induced apoptosis in CD206⁺ M2-like TAMs, and MEL did not induce cell death.

Conclusion: Taken together, our study demonstrated that MEL-dKLA could be used to target M2-like TAMs as a promising cancer therapeutic agent.

Keywords: Cancer immunotherapy; Melittin; Pro-apoptotic peptide; Therapeutic agent; Tumor-associated macrophages.

Fig. 1

Selective cytotoxicity of MEL-dKLA in M2-differentiated RAW264.7 macrophages. (a, b) Effects of dKLA, MEL, or MEL-dKLA on cell viability after incubation for 24 h. After incubation, cell viability was measured with MTS assays. The percent cell viability was standardized based on the absorbance of the PBS-treated control. *P < 0.05, **P < 0.01, ***P < 0.001 versus the dKLA group; #P < 0.05, ##P < 0.01, ###P < 0.001 versus the MEL group. (c) Cell cycle analysis revealed by propidium iodide (PI) staining. All results are expressed as means ± SEMs. (d, e) Annexin V-FITC and MitoTracker-Red CMXRos staining were measured by flow cytometry in (d) M1- and (e) M2-differentiated RAW264.7 macrophages. The cells were treated with PBS or 0.8 μM peptides. Each population of MitoTracker^low Annexin V⁺ cells was normalized to the population treated with PBS as a control. *P < 0.05, **P < 0.01 versus the PBS group; ##P < 0.01 versus the dKLA group; $P < 0.05, $$P < 0.01 versus the MEL group. (n = 3)

Fig. 2

Induction of cell death due to mitochondrial membrane disruption. (a, b) Cellular respiration (a) and glycolysis (b) of peptide-treated cells were analyzed in real time with an XF24 flux analyzer. Oligomycin, FCCP, and Rot/AA were injected into XF24 plates sequentially after the baseline measurement. The y-axis shows the OCRs or ECARs, and the x-axis indicates the measurement number. (c–e) Basal respiration (c), ATP production (d), and maximal respiration (e) were compared between the experimental and PBS-treated control groups. (f-h) The cell energy phenotype, OCRs, and ECARs under basal and stressed conditions. All values are means ± SEMs (n = 3). *P < 0.05, **P < 0.01 versus the PBS group; #P < 0.05, versus the dKLA group; $P < 0.05 versus the MEL group (n = 3)

Fig. 3

Colocalization of MEL-dKLA with mitochondria in vitro. (a) Representative images of mitochondria (green), peptides (red), and DAPI (blue) are shown. (b) Correlation coefficients were measured by Pearson’s statistics. Values ranged from − 1 to + 1, representing full negative or positive correlations. Zero indicates no correlation between the two colors. Results are expressed as means ± SEMs; ***P < 0.001. Scale bar, 20 μm (n = 3)

Fig. 4

Antitumor activities of MEL and MEL-dKLA in vivo. (a) Tumor size, (b) tumor weight, (c) tumor size fold change, (d) and body weight were measured. Tumor-bearing mice were treated with each peptide every 3 days (for a total of three times) starting on day 5 after tumor inoculation. Tumor fold change was calculated based on the tumor size of the last day normalized to the initiation of tumor growth on day 5 (n = 5). (e) Immunofluorescence staining of endothelial cells in subcutaneously implanted LLC tumor paraffin sections. (f) Vessel density was quantified as integrated density per area (μm²) by Image J. Total magnification, 630×. Scale bar, 50 μm. (n = 6). All data are presented as means ± SEMs; *P < 0.05, **P < 0.01, ***P < 0.0001 versus the PBS group; #P < 0.05, ##P < 0.01, ###P < 0.0001 versus the dKLA group

Fig. 5

Selective reduction of the M2-like TAM population by MEL and MEL-dKLA. (a) M1-like macrophages that infiltrated into the tumor stroma were stained as CD45⁺F4/80⁺CD86⁺ (upper panel) and M2-like TAMs were marked as CD45⁺F4/80⁺CD206⁺ (bottom panel). The cells are shown as dot plots within the F4/80 and CD86 or CD206 axis gated on CD45⁺ cells from total live-gated cells. (b, c) Data for each cell phenotype are displayed as the percentages of M1- and M2-like TAMs in CD45⁺ cells (left panel) and the exact number of M1- and M2-like TAMs counted per 10⁴ total single cells (right panel). (d) The M1/M2 ratio was calculated based on the percentage of F4/80⁺CD86⁺ cells and F4/80⁺CD206⁺ cells in CD45⁺ cells (left panel) or using the counted population in total sing cells (right panel). (e, f) The number of tissue-resident M1 or M2 macrophages were counted per 10⁴ total single cells. (g) CD4 T cells (CD45⁺CD4⁺CD8⁻), (h) regulatory T cells (CD4⁺CD25⁺Foxp3⁺), (i) CD8 T cells (CD45⁺CD4⁻CD8⁺), and (j) dendritic cells (CD45⁺CD11b⁺CD11c⁺) were detected within total live-gated cells. All values are presented as means ± SEMs (n = 5); *P < 0.05, **P < 0.01, ***P < 0.0001 versus the PBS group; #P < 0.05, ##P < 0.01 versus the dKLA group; and $P < 0.05 versus the MEL group (n = 5)

Fig. 6

Selective cell death of M2-like TAMs by MEL-dKLA in the tumor stroma. (a) The representative histograms of Annexin⁺ cells in F4/80⁺CD86⁺ M1 macrophages or (b) F4/80⁺CD206⁺ M2 macrophages were normalized to maximum peak and displayed as half offset histograms. (c) The percentages of Annexin V⁺ cells in M1 or M2 macrophages are shown in a bar graph. **P < 0.01 versus the PBS group, ##P < 0.01 versus the dKLA group, $$P < 0.01 versus the MEL group, &&&P < 0.0001 versus the M1 group (n = 6)

Fig. 7

Antitumor effect of MEL-dKLA in an orthotopic model of lung cancer. (a) Representative images showing the lungs of wild mice (WT) and LLC tumor-inoculated mice treated with 175 nmol/kg of each peptide every 3 days. Drugs were administered 5 days after tumor injection into the left lung. Tumors are marked with black arrows. Size bar, 1 cm. (b) Representative H&E-stained histologic sections of the paraffin-embedded left lung. Magnification, × 1.5; size bar, 1 mm. Percentage of tumor area was calculated as tumor area per total area by Image J (left bottom panel). (c) Tumor size and (d) weight of left lung were measured at the end of the experiment on day 16. *P < 0.05, **P < 0.01 versus the PBS group; #P < 0.05, ##P < 0.01 versus the dKLA group; $$P < 0.01 versus the MEL group (n = 7 per group; WT n = 4)