Phototheranostics Using Erythrocyte-Based Particles

Abstract

There has been a recent increase in the development of delivery systems based on red blood cells (RBCs) for light-mediated imaging and therapeutic applications. These constructs are able to take advantage of the immune evasion properties of the RBC, while the addition of an optical cargo allows the particles to be activated by light for a number of promising applications. Here, we review some of the common fabrication methods to engineer these constructs. We also present some of the current light-based applications with potential for clinical translation, and offer some insight into future directions in this exciting field.

Keywords: cancer; delivery systems; imaging; photodynamic therapy; photothermal therapy; red blood cells.

Figure 1

Schematic illustration of RBC membrane coating. RBCs are depleted of their hemoglobin using a hypotonic treatment to form erythrocyte ghosts (EGs). For illustration purposes, CD47 is shown as one of the RBC transmembrane proteins retained on EGs. The inset illustrates the sonication and extrusion size reduction methods. During sonication, sound waves are applied to EGs, while during extrusion, the EGs are pushed through filters with nano-sized pores, both of which result in formation of nEGs.

Figure 2

Volcano plot associated with proteomics analysis of human RBCs, and ICG-EGs. Each dot represents a particular protein. There are no statistically significant differences in the ratios of ICG-EGs proteins to RBCs proteins located below the red dashed line.

Figure 3

Schematic illustration of RBCs, EGs, and nEGs loaded with an optical cargo to serve as RBC-derived carriers. For illustration purposes CD47 is shown as one of the RBC transmembrane proteins retained on EGs. A representative spherical nanoparticle, to be coated with RBC membrane, is shown as the optical cargo for constructs formed using sonication and/or extrusion. Indocyanine green (ICG) is shown in its dimeric and monomeric forms as a representative optical cargo.

Figure 4

Z-stacks of SKBR3 breast cancer cells images obtained by laser scanning confocal fluorescence microscopy. Images were obtained three hours after incubating the cells with ICG-nEGs at 37 °C, and are falsely colored. Blue channel: DAPI staining of the nucleus; red channel: NIR emission from ICG-nEGs internalized by the cells. Values of normalized fluorescence emission intensity associated with the circled region on each stack are plotted as a function of relative depth of the corresponding stack. Scale bar = 10 µm in all panels.

Figure 5

Fluorescence images of SKBR3 breast cancer cells following 808 nm laser irradiation at irradiance of 680 mW/cm² for 15 min. Cells were incubated with PBS (negative control), free ICG (positive control), or ICG-nEGs for three hours at 37 °C, and then washed three times before laser irradiation. Live cells are identified by Calcein acetoxymethyl staining, and falsely-colored in green. Dead cells are identified by ethidium homodimer-1 (EthD-1), and falsely-colored in red.

Figure 6

(A) Absorption and (B) fluorescence emission spectra of nEGs loaded with ICG and Fe₃O₄ NPs (ICG + Fe₃O₄ NPs-nEGs). For comparison, spectra for Fe₃O₄ NPs, ICG-nEGs, and nEGs loaded with Fe₃O₄ NPs (Fe₃O₄ NPs-nEGs) are also shown. Diameters of Fe₃O₄ NPs were in the range of ~ 5–10 nm. nEGs were formed by extrusion of EGs, and then loaded with ICG and Fe₃O₄ NPs. Concentrations of ICG and Fe₃O₄ NPs in the hypotonic loading buffer were 20 µM and 1 mg/mL, respectively. The inset on panel (A) shows, from left to right, photographs of ICG-nEGs, Fe₃O₄ NPs-nEGs, and ICG + Fe₃O₄ NPs-nEGs suspended in isotonic PBS. Absorption > 600 nm is due to ICG. Fluorescence emission was recorded in response to photoexcitation at 720 nm. Presence of Fe₃O₄ NPs in ICG + Fe₃O₄ NPs-nEGs had minimal effect on absorption and fluorescence emission characteristics of the particles as compared to those for ICG-nEGs.

Figure 7

(A) Open-abdomen whole-body NIR fluorescence images of immunodeficient mice inoculated with intraperitoneal SKOV ovarian cancer cells at 24 h post-intravenous administration of ICG-nEGs, or folate-functionalized ICG-nEGs (folate-ICG-nEGs). Images are falsely colors with the emission intensity scale shown on the right side. White arrows point to the tumors. (B) Corresponding images of the extracted organs and tumors from each mouse. (C) Average fluorescence emission intensity normalized to weight of each tumor, and subsequently averaged among the number of homogenized tumors (n = 3 tumors from three different animals). Asterisk indicates a statistically significant difference (p < 0.05). For imaging of the whole body, and the extracted organs and tumors, photoexcitation wavelength was 710 ± 25 nm, and emission > 785 nm was recorded. Homogenized tumors were photoexcited at 720 nm, and emission in 735–900 nm band was recorded.

Figure 8

Quantification of PS on the outer leaflet of (A) ICG-EGs (negative control), (B) cholesterol-depleted ICG-EGs (c^--ICG-EGs) (positive control), and (C) cholesterol-enriched ICG-EGs (c⁺-ICG-EGs). Particles were fabricated using human RBCs, and loaded with ICG in hypotonic buffer. We show representative dot plots of side scattering vs. fluorescence emission of AlexaFluor 488-labeled annexin V, which targets surface-exposed PS. Boxed regions correspond to PS-positive particles. The indicated percentages refer to the fraction of the particles in the sample population that are PS-positive, and represent the average of four measurements on each sample.