Two-Photon Enzymatic Probes Visualizing Sub-cellular/Deep-brain Caspase Activities in Neurodegenerative Models

Caspases work as a double-edged sword in maintaining cell homeostasis. Highly regulated caspase activities are essential during animal development, but dysregulation might lead to different diseases, e.g. extreme caspase activation is known to promote neurodegeneration. At present, visualization of caspase activation has mostly remained at the cellular level, in part due to a lack of cell-permeable imaging probes capable of direct, real-time investigations of endogenous caspase activities in deep tissues. Herein, we report a suite of two-photon, small molecule/peptide probes which enable sensitive and dynamic imaging of individual caspase activities in neurodegenerative models under physiological conditions. With no apparent toxicity and the ability of imaging endogenous caspases both in different subcellular organelles of mammalian cells and in brain tissues, these probes serve as complementary tools to conventional histological analysis. They should facilitate future explorations of caspases at molecular, cellular and organism levels and inspire development of novel two-photon probes against other enzymes.


Supplementary Methods
All chemicals were purchased from commercial vendors and used without further purification, unless otherwise noted. All non-aqueous reactions were carried out under a nitrogen/argon atmosphere in oven-dried glassware. Reaction progress was monitored by TLC on pre-coated silica plates (Merck 60 F254 nm, 250 μm thickness) and spots were visualized by UV light or appropriate staining (e.g., ceric ammonium molybdate (CAM), basic KMnO4). Flash column chromatography was carried out using silica gel (Merck 60 F254 nm 0.040-0.063μm). All 1 H NMR and 13

S60
The flow rate was 0.6 mL/min for analytical HPLC and 10 mL/min for preparative HPLC. UV-vis absorption and fluorescence spectra were measured by using a Shimadzu UV 2550 spectrometer and a Perkin Elmer LS50 spectrofluorometer, respectively. The two-photon excited fluorescence measurements were performed using a Spectra Physics femtosecond Ti:sapphire oscillator (Tsunami) as the excitation source. The output laser pulses have a tunable center wavelength from 760 nm to 850 nm with pulse duration of 40 fs and a repetition rate of 76 MHz. The laser beam was focused onto the sample that was contained in a cuvette with path length of 1 cm. The emission from the sample was collected at 90° angle by a pair of lenses and an optical fiber that was connected to a monochromator (Acton, Spectra Pro 2300i) coupled with CCD system (Princeton Instruments, Pixis 100B). A short pass filter with cut-off wavelength at 700 nm was placed before the spectrometer to minimize the scattering from the pump beam. All optical measurements were performed at room temperature. All  Signaling. Anti-caspase-7 antibody (sc-6138) was bought from Santa Cruz. Anti-caspase-3 (both full   length and p17 of caspase-3, CST #9662, human and mouse) and anti-caspase-8 (ALX-804-447, mouse) antibodies were gifts from Prof. Han-Ming Shen (NUS).

(E3)
To a stirred solution of methyl 5-chlorosalicylate (5 g, 26.8 mmol) in DMF (40 mL) was added K2CO3 (11.1 g, 80.0 mmol), followed by iodomethane (4.2 g, 29.5 mmol). The mixture was stirred for 4 h at rt, TLC showed (EA:Hexane = 1:4) that reaction was complete. 200 mL of water was added to the mixture and extracted with EA (2×100 mL). The combined organic layers were washed with brine, S66 dried over Na2SO4, and evaporated to afford the desired product E1 as yellow oil, which was contaminated with DMF and used for next step without further purification.
The above residue was dissolved in ethanol (30 mL) and to which an aqueous solution of NaOH (3.2 g, 80.4 mmol) in water (30 mL) was added. After being stirred for 1-3 min, white solid precipitates, and the mixture was continued to stir for 1 h. TLC showed (EA:Hexane = 1:1) starting material had disappeared. The mixture was concentrated to remove most part of ethanol and acidified with 1 M HCl to pH = 2-3 and extracted with EA (2×100 mL). The combined organic layers were washed with brine, dried over Na2SO4, and evaporated to afford the desired product E2 as yellowish solid, which was used for next step without further purification.
To the crude E2 obtained above was added concentrated sulfuric acid 50 mL to dissolve the solid and the solution was cooled to 0°C. 1.9 mL of concentrated nitric acid was added dropwise in 2 min.
After the addition, the solution was stirred for 15 min, and TLC showed (EA:Hexane = 1:1) the reaction finished. The solution was poured into ice in a 250 mL of conical flask, and the product was collected by filtration, and the filter cake was washed with a large amount of water to completely remove the acid. The solid was dried to afford the desired product E3 (5.4 g, 87% yield for three steps) as offwhite solid.

(E4)
To a stirred solution of E3 (5 g, 21.6 mmol) in THF cooled to 0°C was added Et3N (4.4 g, 43.2 mmol), followed by tert-butyl chloroformate (4.3 g, 32.4 mmol) at 0°C. The solution was stirred for 1 h, and TLC showed (EA:Hexane = 1:1) the reaction finished. Volatiles were removed under vacuum and DMF (50 mL) was added, followed by a solution of 4-(2-aminoethyl)benzenesulfonamide (4.7 g, 23.7 mmol) in DMF (20 mL) at rt. The mixture was stirred for 2 h, TLC showed (EA:Hexane = 1:1) the reaction completely finished. 200 mL of water was poured into the reaction mixture and the product was collected by filtration, and the filter cake was washed with water to completely remove DMF. The solid was dried to afford the desired product E4 (7.6 g, 85% yield) as off-white solid.

(E8)
In an ice bath, N-hydroxysuccinimide and N,N'-dicyclohexylcarbodiimide (DCC) were added to a solution of E7 in anhydrous THF. The suspension was filtered off and the filtrate was used for the subsequent coupling reaction directly.

Determination of Photophysical and Biochemical Properties of Probes in vitro
Hydrolytic stability assay with C1RS/C1FS. The hydrolytic stability assay was carried out as described previously. 8  As expected, the proteolytic stability test on both C1RS and C1FS showed they did not display any noticeable degradation in BV2 lysates over 24 h (Supplementary Fig.3). All issues considered, C1RB and C1FS were therefore chosen as the designated caspase-1 detecting probes in subsequent bioimaging studies.
In vitro selectivity. For each probe, different caspases (-1, -3, -6, -8 and -9) were applied under identical conditions. Normalized, relative fluorescence values (RFU) were obtained after 2-h incubation at room temperature with λex = 360 ± 40 nm; λem = 528 ± 20 nm for AAN/λem = 460 ± 40 nm for AMC and DAN. All probes showed virtually identical selectivity profiles towards their intended caspases as those obtained with the AMC substrates. We did not include recombinant caspase-7 in the screening, as its reactivity toward our probes would be redundantly similar to that of caspase-3, based on published literature. 9, 10 Finally, both C6RA and C9RA, similar to their AMC counterparts, exhibited comparatively poor selectivity profiles towards caspase-6 and -9, respectively, and were unlikely to be useful in real bioimaging applications. Taken together, in addition to C1RB and C1FS (for caspase-1 bioimaging), we chose C3RA/C3RM/C3RE and C8RA for further two-photon, livecell bioimaging of caspase-3/-7 and caspase-8 activities, respectively, as these probes possess both S79 reasonable in vitro reactivity and selectivity profiles towards their intended caspase targets. and induced channel (set to 1) of the C3RA readouts.

Permeability of free dye and probes in live cells.
To check the permeability and distribution of the free dye (i.e., AAN) and our probes in live cells, HeLa cells were seeded in the glass-bottom dishes and grown to be 70% confluency. Then each compound at a suitable concentration was added into S80 HeLa cells and incubated for 1 h at 37 o C before images were taken. LysoTracker (100 nM, LysoTracker ® Red DND-99, L-7528) was added together to check the cellular localization of the compound. The free dye AAN was microscopically observed to diffuse freely throughout the entire intracellular environment (em = 470-500 nm; Supplementary Fig. 5a). By taking advantage of the weakly blue-shifted fluorescence of AAN-modified probes (C1RB/C3RA/C8RA; ex = 405 nm, em = 420-480 nm with Carl Zeiss system), we were able to determine they were cell-permeable ( Supplementary Fig. 5a). Similar AMC-containing probes are known to be cell-permeable as well.
Furthermore, incubation at low temperature ( Supplementary Fig. 5c, down), and no obvious endocytosis and endo/lysosomal trapping was observed. To be noted, for C1FS, we were not able to obtain direct imaging-based evidence of its intracellular localization profiles due to its nearly nonfluorescent state, but given its small uniform molecule-like feature, and that its analog C1RS was cellpermeable, we expected this probe to be taken up readily by cells as well.  for probes and λex = 650 nm, λem = 655-740 nm for the membrane tracker ( Supplementary Fig. 5b).

Staurosporine (STS)-induced caspase activation in mammalian cells. Induction of apoptosis by
STS in mammalian cells was performed as previously described. 5,9,11 Cells were grown to be ~70%    Fig. 6). To better visualize the cellular signals, 3D images were acquired from PerkinElmer Ultraview Vox Spinning Disc confocal microscope and processed with Volocity 3D image analysis software (Fig. 3k and Supplementary Videos). 12,13 S83 TPFM imaging of caspase activities in primary cortical neurons and fresh tissues. All twophoton images were taken on a Leica TCS SP5X Confocal Microscope System. Procedures involving animals were approved by and conformed to the guidelines of Institutional Animal Care and Use Committee at the National Neuroscience Institute (Singapore).
Dissociated neuron-enriched cell cultures of cerebral cortex were established from day 16 C57BL/6 mouse embryos, as described. 14 Primary neurons were cultured in the neurobasal medium (Gibco) with 2% B27 and 0.5 mM GlutaMAX. Experiments were performed in 7-9 day-old cultures.
For glucose-deprivation (GD) studies, glucose-free neurobasal medium was used (with other components fixed). The cultured neurons were incubated in glucose-deprived medium for 8 h, while control cells were incubated in normal neurobasal medium. Afterwards, C1RB (24 μM)/C1FS (12 μM) was introduced with further incubation for 1 h at 37 o C. To check whether the signal observed was related to caspase-1 activity, the caspase-1 specific inhibitor, Pralnacasan (50 μM), was added to the cultures 1 h prior to probe introduction (t = 7 h during GD) ( Supplementary Fig. 7a). To verify the presence of cleaved caspase-1, WB of total lysates from GD treated primary neurons for different time periods were performed as well ( Supplementary Fig. 7b).
Tissues used in the imaging experiments were fresh brains of 7-day-old C57BL/6 mice subcutaneously injected with either ethanol (20% solution in normal saline with 2.5 g/kg at 0 h and again at 2 h) or the same amount of normal saline. 15,16 At 8 h following the first ethanol dose, the brains were surgically removed from the mouse head and immediately transferred into an ice-artificial cerebrospinal fluid (ACSF; 138.6 mM NaCl, 3.5 mM KCl, 21 mM NaHCO3, 0.6 mM NaH2PO4, 10 mM D-glucose, 1 mM CaCl2 and 3 mM MgCl2). The brain was cut into 200 μm-thick sections using a vibrating blade microtome in ACSF. Slices were incubated with C3RA (120 µM) in ACSF at 37 o C for 3 h before image acquisition. For inhibition experiments, the slices were treated with caspase-3/7 inhibitor I (200 μM) 3 h prior to addition of C3RA. Treated brains were then transferred to poly-Llysine-coated cover slips and images were acquired at different depths by changing the Z-axis thickness on the microscope. To prepare lysates, the brain was first homogenized in lysis buffer (HEPES containing 1% triton X-100 and 1 mM PMSF) with plastic homogenizer and then sonicated. The supernatant was obtained by centrifugation at 13,000 rpm for 20 min at 4 o C and protein concentration was determined. 30 μg lysates per lane were loaded for WB (Fig. 4c). At the same time, 300 μg lysates in 60 μL was taken for enzymatic assay (diluted in HEPES containing 0.02% triton X-100) with 6 μM