A Two-Photon Probe Based on Naphthalimide-Styrene Fluorophore for the In Vivo Tracking of Cellular Senescence

Abstract

Cellular senescence is a state of stable cell cycle arrest that can negatively affect the regenerative capacities of tissues and can contribute to inflammation and the progression of various aging-related diseases. Advances in the in vivo detection of cellular senescence are still crucial to monitor the action of senolytic drugs and to assess the early onset or accumulation of senescent cells. Here, we describe a naphthalimide-styrene-based probe (HeckGal) for the detection of cellular senescence both in vitro and in vivo. HeckGal is hydrolyzed by the increased lysosomal β-galactosidase activity of senescent cells, resulting in fluorescence emission. The probe was validated in vitro using normal human fibroblasts and various cancer cell lines undergoing senescence induced by different stress stimuli. Remarkably, HeckGal was also validated in vivo in an orthotopic breast cancer mouse model treated with senescence-inducing chemotherapy and in a renal fibrosis mouse model. In all cases, HeckGal allowed the unambiguous detection of senescence in vitro as well as in tissues and tumors in vivo. This work is expected to provide a potential technology for senescence detection in aged or damaged tissues.

Figure 1

Synthesis of the probe and the mechanism of action in mice. (A) Synthetic route used for the preparation of the probe: (a) CH₃ONH₂·3HCl, Et₃N, and dioxane; (b) TBDPSCl, imidazole, and DMF; (c) n-BuLi, Ph₃PCH₃I, and THF; (d) Pd(OAc)₂, (O-tolyl)₃P, Et₃N, and DMF; and (e) NaOH/MeOH and acetobromo-α-D-galactose. (B) Schematic representation of the application of the probe in two in vivo models of senescence: (i) kidney fibrotic C57BL/6 J male mice induced by treatment with folic acid and (ii) BALB/cByJ female mice bearing 4 T1 breast cancer tumors treated with senescence-inducing chemotherapy. (iii) Fluorescence emission spectra (λ_ex = 488 nm) of HeckGal (yellow) and Heck fluorophore (orange) in aqueous solutions (pH 7)–DMSO (0.01%).

Figure 2

Probe enables the detection of senescence in various cell lines regardless of the induction method. (i) Senescence induction was assessed by SA-β-Gal staining in (A) nontreated and (H) palbociclib-treated cells. Note that senescent SK-Mel-103 cells present the typical blue staining. Confocal images of SK-Mel-103 and SK-Mel-103 treated with palbociclib. (B–E and I–L) One-photon confocal images of (B–E) control SK-Mel-103 in the (B) absence or (C–E) presence of 10, 15, and 20 μM of a HeckGal probe, respectively, and (I–L) SK-Mel-103 treated with palbociclib in the (I) absence or (J–L) presence of 10, 15, and 20 μM of the HeckGal probe, respectively. (F, G, M, N) Two-photon microscopy images of (F, G) nontreated and (M, N) palbociclib-treated (senescent) SK-Mel-103 cells in (F, M) the absence and (G, N) presence of 10 μM of the HeckGal probe. Cells were incubated with HeckGal in a DMEM (10% FBS, 0.1% DMSO) in 20% O₂ and 5% CO₂ at 37 °C for 2 h, and then one-photon images were acquired by using a confocal microscope (Leica TCS SP8 AOBS), and two-photon images were acquired by using a multiphoton confocal microscope (Olympus FV1000MPE). (ii) SA-β-Gal staining of (A) nontreated and (H) palbociclib-treated 4 T1 cells. Note that senescent 4 T1 cells present the typical blue staining. (B, C, I, J) Confocal images of (B, C) control 4 T1 cells in the presence of (B) 15 μM of the HeckGal probe or (C) 15 μM of Heck and 4 T1 cells treated with (I, J) palbociclib in the presence of (I) 15 μM of the HeckGal probe or (J) 15 μM of Heck. SA-β-Gal staining of (D) nontreated and (K) cisplatin-treated A549 cells. Note that senescent A549 cells present the typical blue staining. Confocal microscopy images of (E−G) nontreated and (L–N) cisplatin-treated A549 cells, exposed to the HeckGal probe. Cells were incubated with HeckGal (15 μM) in a DMEM + 10% FBS in 20% O₂ and 5% CO₂ at 37 °C for 2 h, and images were acquired by using a confocal microscope (excitation at 488 nm). (iii) Quantification of the fluorescence emission intensity relative to the cell surface of control and palbociclib-treated SK-Mel-103 cells incubated with HeckGal visualized with (A) one-photon confocal imaging and (B) two-photon confocal imaging. Quantification of the fluorescence emission intensity relative to the cell surface of control and palbociclib-treated 4 T1 cells incubated with HeckGal or Heck visualized with (C) one-photon confocal imaging. Quantification of the fluorescence emission intensity relative to the cell surface of control and cisplatin-treated A549 cells incubated with HeckGal visualized with (D) one-photon confocal imaging. Error bars represent SEM (n = 3). (E) Fluorescence-activated cell sorting (FACS) analysis for control SK-Mel-103 (gray) human melanoma cells and doxorubicin-treated SK-Mel-103 (red) cells after treatment with HeckGal. (F) FACS analysis for control BJ (gray) human fibroblast cells and doxorubicin-treated BJ (red) cells after treatment with HeckGal. Both cell lines were treated with 250 nM doxorubicin for 24 h in order to induce cellular senescence, or with DMSO as the vehicle. After 14 days, upon complete development of the senescent phenotype, cells were incubated with 7 μM HeckGal for 2 h, detached from the plates, and washed twice with PBS. HeckGal fluorescence was subsequently evaluated by a Sony SA3800 spectral analyzer.

Figure 3

HeckGal probe enables the detection of senescence in different disease models of senescence. (A) Representative images of tumors stained for the SA-β-Gal assay: tumors from vehicle (left) and palbociclib-treated mice (right). (B) Immunohistochemical detection of the proliferation marker Ki67 in paraffin sections of tumors from vehicle (top) and palbociclib-treated mice (bottom). (C–F) IVIS images of organs and tumors from BALB/cByJ female mice bearing 4 T1 breast cancer cells: From left to right and from top to bottom: lungs, liver, tumor, kidney, and spleen; (C) Vehicle mice, (D) vehicle mice treated with (13.33 mg/mL, 100 μL), (E) mice treated with palbociclib for 1 week, (F) palbociclib-treated mice injected with HeckGal (13.33 mg/mL, 100 μL). Mice were sacrificed 2 h post-HeckGal treatment. (G) Quantification of the Ki67 signal in paraffin sections of tumors from vehicle (top) and palbociclib-treated mice (bottom). Error bars represent s.d. (H) Quantification of average radiance intensity from organs and tumors showed in images (C), (D), (E), and (F). Error bars represent SEM (n = 3 for each condition). (I) Two-photon fluorescence depth images of HeckGal in tumor tissue slices from vehicle (up) and palbociclib-treated mice (down). The slices were incubated with HeckGal (10 mM) for 2 h at 37 °C in a dry incubator. The images were acquired at different penetration depths (λ_ex = 820 nm). (J) 3D representation of images shown in Figure 3I demonstrating the greater penetrability of HeckGal in tumor tissue slices from palbociclib-treated mice (down) compared to tumor tissue slices from vehicle mice (up).

Figure 4

(A) Immunostaining for p21 in kidney slides. (B) IVIS images of kidneys from mice with renal fibrosis induced by FA treatment. From left to right: Vehicle mice + HeckGal (6 mg/mL, 200 μL), FA-treated mice (with renal fibrosis) + HeckGal (6.6 mg/mL, 200 μL), and FA-treated mice (with renal fibrosis). Mice were sacrificed 5 h post-HeckGal injection. (C) Quantification of the p21 signal in paraffin sections of kidney from vehicle and FA-treated mice. Error bars represent SEM. (D) Quantification of average radiance intensity from kidneys showed in the 4B image. Error bars represent SEM (n = 3 for control mice treated with the probe and FA-treated mice, and n = 4 for FA-treated mice + HeckGal).