A novel optical imaging probe for targeted visualization of NLRP3 inflammasomes in a mouse model of age-related macular degeneration

Abstract

Purpose: Wet form of age-related macular degeneration (wet AMD) is a progressive vascular disease that mainly affects older adults and causes severe and irreversible vision loss. A key complication of wet AMD is choroidal neovascularization (CNV), which may be driven in part by NLRP3 inflammasomes that are associated with macrophages migration to CNV lesions. Since activated NLRP3 is correlated with CNV, visualizing NLRP3 inflammasomes and their associated macrophages is of great interest to monitor wet AMD progression and develop effective therapies against it. However, to the best of our knowledge, current ophthalmic imaging systems do not permit such targeted imaging. Therefore, in this study, we developed InflammaProbe-1, an optical imaging probe for targeted visualization of NLRP3 inflammasomes in CNV lesions.

Methods: InflammaProbe-1 was synthesized by conjugating a clinically relevant fluorophore, Oregon Green^® 488, to the selective NLRP3 inhibitor, CY-09. The ability of InflammaProbe-1 to target NLRP3 was assessed with an enzyme-linked immunosorbent assay by comparing its ability to inhibit NLRP3-mediated secretion of IL-1β to that of CY-09 in LPS-primed and nigericin-stimulated BMDMs. In vitro confocal imaging of NLRP3 was performed on InflammaProbe-1-stained BMDMs that had been induced to express NLRP3 with LPS. In vivo imaging of NLRP3 was conducted on mouse laser induced choroidal neovascularization (LCNV), a model of AMD, 6 h after an intraperitoneal injection of InflammaProbe-1 at 10 mg/kg on day 4 post-LCNV.

Results: InflammaProbe-1 was just as effective as CY-09 at inhibiting IL-1β secretion (p < 0.01 at 10 μM for both the InflammaProbe-1 and CY-09 groups relative to the control). InflammaProbe-1-stained BMDMs that had been induced to express NLRP3 showed significantly brighter fluorescence than untreated cells (p < 0.0001 for LPS treatment group and p < 0.001 for LPS and nigericin treatment group). Furthermore, in vivo molecular imaging of NLRP3 was achieved in mouse LCNV.

Conclusion: We propose that InflammaProbe-1 may be a useful molecular imaging probe to monitor the onset, progression, and therapeutic response of AMD and other NLRP3-mediated diseases.

Keywords: CY-09; NLRP3 inflammasome; age-related macular degeneration; choroidal neovascularization; macrophages; optical imaging.

FIGURE 1

Synthesis and characterization of InflammaProbe-1. (A) InflammaProbe-1 was synthesized by conjugating a selective inhibitor of NLRP3 (CY09) to a commercially available fluorophore (Oregon Green^® 488). Conjugation was achieved with an EDCI-mediated coupling method. (B) Excitation and emission spectra of InflammaProbe-1 in ethanol containing 10% DMSO.

FIGURE 2

InflammaProbe-1 inhibits NLRP3-mediated secretion of IL-1β but not TNF-α. (A) InflammaProbe-1 and CY-09 dose-dependently inhibited NLRP3-mediated secretion of IL-1β in LPS-primed and nigericin-stimulated mouse bone marrow-derived macrophages. (B) InflammaProbe-1 and CY-09 had no significant effect on LPS-induced secretion of TNF-α in comparison to the LPS and nigericin control, except for CY-09 at 10 μM. These data suggest that InflammaProbe-1 retains the inhibitory ability of its parent compound, CY-09, that enables it to target the NLRP3 inflammasome. Levels of IL-1β and TNF-α were measured by performing ELISA. The data were expressed as the mean ± SD (n = 3). Statistical analysis by unpaired t-tests with Welch’s corrections; *p < 0.05 and **p < 0.01.

FIGURE 3

In vitro imaging of NLRP3 in bone marrow-derived macrophages (BMDMs) using InflammaProbe-1. (a–c) Untreated BMDMs, (d–f) LPS-primed BMDMs, and (g–i) LPS-primed and nigericin-stimulated BMDMs that were treated with 10 μM InflammaProbe-1, fixed on microscope slides, and imaged using confocal fluorescence microscopy. (j) Quantification of InflammaProbe-1 RFUs per cell of confocal microscopy images. Relative to the untreated control cells, those that were treated with LPS alone or with LPS and nigericin displayed significantly brighter InflammaProbe-1-dependent fluorescence (p < 0.0001 and p < 0.001, respectively). These results indicate that InflammaProbe-1 selectively stains macrophages that have been induced to express NLRP3. These data are representative of four replicates from each experimental group. Statistical analysis by unpaired t-tests with Welch’s corrections; ***p < 0.001 and ****p < 0.0001 relative to the untreated control.

FIGURE 4

In vivo retinal imaging of NLRP3 in laser-induced choroidal neovascularization (LCNV) using InflammaProbe-1. (a) Brightfield and (b) fluorescence fundus images of mouse LCNV taken 6 h after a 10 mg/kg intraperitoneal injection of InflammaProbe-1 on day 4 post-LCNV. The fluorescence fundus image shows InflammaProbe-1-dependent fluorescence that is localized to each of the LCNV lesions observed in the brightfield fundus image. These are representative images of 12 mouse LCNV eyes. (c,d) Higher magnification of panels (a,b), respectively. (e) Fluorescence intensity measurements expressed in relative fluorescence units (RFU) per pixel area were measured using ImageJ software. Statistically significant fluorescence enhancement was observed in LCNV lesions compared to those in the non-laser control groups. Statistical significance *p < 0.05.

FIGURE 5

Confocal imaging of NLRP3 in laser induced choroidal neovascularization (LCNV) tissues using InflammaProbe-1. Four days after LCNV, mice were intraperitoneally injected with InflammaProbe-1 at 10 mg/kg and enucleated after 6 h. Their choroids were dissected and co-stained with fluorescently tagged antibodies against IBA1, which targets macrophages. The stained choroidal lesions were then imaged with confocal fluorescence microscopy at 63x magnification (a–c). (d) Higher magnification of panel (b). The yellow arrows in panel (c) indicate InflammaProbe-1 localization in macrophages. These are representative confocal images taken from 12 choroids. See also Supplementary Figure 4 for co-localization of InflammaProbe-1 in 3D reconstructed LCNV lesions from confocal images.

FIGURE 6

Toxicity of InflammaProbe-1. (A) Cytotoxicity of InflammaProbe-1 was assessed in primary mouse retinal microvascular endothelial cells (MRMEC) using a fluorescence-based assay with Calcein Deep Red™ AM ester. The viability of MRMECs was not significantly reduced by a 20-h exposure to 1–20 μM InflammaProbe-1 in comparison to untreated cells. (B) Retinal toxicity was assessed in dark-adapted mice using ganzfeld electroretinography (ERG) 7 days after an intraperitoneal injection of InflammaProbe-1 at 10 mg/kg. Relative to the retinas of mice injected with saline or a vehicle control (10% DMSO in PBS), the retinas of mice that had been injected with InflammaProbe-1 did not show any significant reduction in the a-wave and b-wave amplitudes of their electrical response to a 1 Log cd⋅s/m² light flash. These results suggest that InflammaProbe-1 is not toxic to primary cells or retinal tissues. Cell viability data were expressed as the mean ± SD of six replicates from each group. Statistical analysis by unpaired t-tests with Welch’s corrections; **p < 0.01. ERG data were expressed as the mean ± SD of four retinas from each group.