In vivo monitoring of tissue regeneration using a ratiometric lysosomal AIE probe

Tissue regeneration is a crucial self-renewal capability involving many complex biological processes. Although transgenic techniques and fluorescence immunohistochemical staining have promoted our understanding of tissue regeneration, simultaneous quantification and visualization of tissue regeneration processes is not easy to achieve. Herein, we developed a simple and quantitative method for the real-time and non-invasive observation of the process of tissue regeneration. The synthesized ratiometric aggregation-induced-emission (AIE) probe exhibits high selectivity and reversibility for pH responses, good ability to map lysosomal pH both in vitro and in vivo, good biocompatibility and excellent photostability. The caudal fin regeneration of a fish model (medaka larvae) was monitored by tracking the lysosomal pH change. It was found that the mean lysosomal pH is reduced during 24–48 hpa to promote the autophagic activity for cell debris degradation. Our research can quantify the changes in mean lysosomal pH and also exhibit its distribution during the caudal fin regeneration. We believe that the AIE-active lysosomal pH probe can also be potentially used for long-term tracking of various lysosome-involved biological processes, such as tracking the stress responses of tissue, tracking the inflammatory responses, and so on.

The in vivo co-localization was also studied. Medaka larvae were firstly exposed to 5 µM CSMPP for 2 h at 28 °C, and then were fed with 200 nM LysoTracker deep red (LTDR, Invitrogen Co., Carlsbad, CA) together with CSMPP for another 2 h at 28 °C. After being washed with ERM, medaka larvae were imaged under two different channels. Green channel was excited at 405 nm and detected from 468 to 630 nm; red channel was excited at 633 nm and detected from 647 to 713 nm. These images were obtained by using a Zeiss LSM 710 confocal microscope equipped with the accessory of a Bioptechs Focht Chamber System 2 (FCS 2) and analyzed by using ZEN 2009 software (Carl Zeiss).

Cytotoxicity study
2-(4,5-Dimethyl-2-thiazolyl)-2,5-diphenyltetrazolium bromide (MTT) assay was used to evaluate the cytotoxicity of CSMPP to HeLa cells and ARPE-19 cells. HeLa cells or ARPE-19 cells were seeded in a 96-well plate at a density of 5,000 cells per well. After 24 h incubation, the cells were exposed to a series of doses of CSMPP (0-10 μM) in culture medium at 37 °C for 24 h. Six replicates were performed for each concentration. Next, the dye solution in wells were removed and 0.5 mg/mL freshly prepared MTT solution was added into each well. After incubation for 4 h, 100 μL of solubilization solution (10% SDS in 0.01 M HCl) was added to dissolve the formed purple crystals. After another 4 hours' incubation, the absorbance at 570 nm was recorded using a Perkin-Elmer Victor plate reader.

Intracellular pH calibration and measurement
The pH calibration buffer contains 20 mM HEPES, 20 mM MES, 20 mM acetate, 100 mM KCl, 20 mM NaCl, 1 mM CaCl2, 0.5 mM MgCl2, 5 mM glucose, 12 µM nigericin and 5 µM monensin. The pH buffers were adjusted to different pH in the range of 2.50-7.00 by using 2 N NaOH solution and 2 N hydrochloric acid.
HeLa cells or zebrafish cells were cultured in 35 mm confocal dishes (VWR International) and used when they were grown up to about 80% confluence. Cells were first stained by 3 µM CSMPP for 1 h at 37 º C. Then, cells were equilibrated in pH calibration buffers for 8 min at 25 º C. The fluorescence images of green channel and red channel at four fields of vision for each pH were acquired in less than 30 min. The fluorescence images for pH calibration in HeLa cells were captured by CLSM with a 63x oil objective lens. The fluorescence images for pH calibration in zebrafish cells were captured by CLSM with a 40x oil objective lens. Confocal images of cells without incubation by dyes were also captured and analyzed as the background signals, which will be used as the minimal threshold of the fluorescence images for pH calibration.
The ratiometric analysis was carried out by using Image J software. The background of every fluorescence image was subtracted by using the minimal threshold after transforming the image into 32-bit image. For HeLa cells captured by 63x oil objective lens, the minimal threshold is 5. For zebrafish cells captured by 40x oil objective lens, the minimal threshold is 2. The ratiometric image was acquired by dividing the red image by the green image with Image J software. From the ratiometric image, the mean lysosomal pH was calculated. The ratiometric image was shown in 16color type under the lookup tables. The same settings (e.g. objective lens, channel range, pinhole, laser gain, resolution) as those used in performing pH calibration experiments were employed for the imaging of corresponding cells during lysosomal pH measurement.
The H + calibration equation was acquired based on Grynkiewicz's formula for calibration of calcium ion, 1 assuming that the fluorescence contribution from any given molecular species is proportional to the concentration of that species.
Therefore, the relationship between [H + ] and ratio is shown as below: The fluorescence ratio R is the ratio of the dye's fluorescence intensities F1 and F2 at two emission wavelengths λ1 and λ2 (R = F1/F2). Ka is the acid dissociation constant (Ka = ). S factor is the proportionality coefficient in relation with dye concentration, and Sf2 is symbolized for free dye measured at wavelength λ2, Sb2 for H + -bound dye at λ2. It is well known that, in principle, each S factor is determined by the excitation intensity, extinction coefficient, path length, quantum efficiency and the instrumental efficiency of collecting emitted photons. Taking the equation pH =log[H + ] and pKa = -logKa into equation (1), the linear relationship between pH and log[

Rmax-R
] was acquired as below: After performing the pH calibration experiments, the linear fitting relationship between pH and log[

Rmax-R
] under the specific imaging conditions can be obtained based on the equation (2). Then, the measured mean ratio R of samples was taken into the fitted linear calibration curve to calculate the corresponding pH.

In vivo toxicity to medaka
The biocompatibility of CSMPP was assessed by exposing medaka larvae to different concentrations of CSMPP (0−10 μM) in a glass beaker containing 1 L of ERM for 96 h at 28 °C. In each beaker, 10 individuals of 1-day old medaka larvae were added. Three replicates were performed for each treatment. All the beakers were stored in an incubator at 28 °C. Mortality of medaka larvae was then monitored for 96 h. In addition, the biocompatibility of CSMPP was also evaluated by recording the heartbeat of larvae. Fifteen medaka larvae were first incubated with 5 μM CSMPP for 4 h, then transferred into new ERM, later the heartbeat of the larvae was recorded at different time (0, 6, 24, 48, 72 and 96 h). Heart rate was measured as mean heartbeat per minute. Larvae were allowed to "rest" for roughly one minute before being measured. A hand-held counter was used to record heartbeats for 15 seconds.

Photostability study
HeLa cells were stained by 2 μM CSMPP for 12 min, 500 nM LTR for 6 min and 500 nM LTG for 5 min at 37 °C, respectively. Then, the confocal fluorescence images of the CSMPP-stained HeLa cells were continuously scanned for 100 times using its normal imaging condition. The confocal fluorescence images of the LTR-stained or LTG-stained HeLa cells were continuously scanned for 50 times using their normal imaging conditions. And the changes of fluorescence signals were compared. For CSMPP: λex = 405 nm; λem = 470-650 nm. For LTR: λex = 561 nm, λem = 565-650 nm. For LTG: λex = 488 nm, λem = 495-580 nm.

Results
Scheme S1. The scheme of the synthetic route of CSMPP.            S12. The biocompatibility of CSMPP to medaka larvae was assessed by (a) measuring the survival rate after exposing larvae to different concentrations of CSMPP for 96 h and by (b) recording the mean heartbeat of larvae at different times after feeding them with 5 μM CSMPP for 4 h. Three replicates were included in each concentration treatment, and each replicate contained 10 individual larvae. The heartbeat of fifteen larvae was recorded each time. Data are the mean ± SD. Figure S13. (a) The pH calibration of CSMPP probe in HeLa cells captured by using CLSM with a 63x oil objective lens. Cells were first stained by 3 µM CSMPP for 1 h at 37 º C. Then, cells were equilibrated in pH calibration buffers containing 12 µM nigericin and 5 µM monensin for 8 min at 25 ºC. For green channel: λem = 470-560 nm; For red channel: λem = 560-700 nm. λex = 405 nm. The ratiometric image was acquired by dividing the red channel by the green channel with Image J software. Scale bar is 20 μm. (b) The mean ratio of Em-red/Em-green as a function of pH. Mean ± SD between three images were presented. The calibration curve was fitted based on the relationship between the mean ratio, Rmin, Rmax and the pH, whose calculation equation is presented in equation 2. Figure S14. The CSLM images after HeLa cells were stimulated by chemical stimulants, including brightfield, green channels, red channels, merged two-channel images, and ratiometric images showing the lysosomal pH distribution. Chemical stimulants: normal lysosomal pH without stimulants; 50 nM bafilomycin A1 (BFA) incubated for 30 min, 0.1 mM H2O2 incubated for 30 min; 40 μM tamoxifen (TMX) incubated for 15 min; 1.75 mM and 10 mM acetic acid (HAc) incubated for 8 min at 37 o C. The captured confocal fluorescence images of the red and the green channels were analyzed by using Image J software to acquire the ratiometric images. Excitation: 405 nm. The emission of the green channel at 470−560 nm and the red channel at 560−700 nm were collected. Scale bar is 20 μm. Figure S15. The lysosomal pH of HeLa cells after being incubated by different concentrations of CSMPP for 5 min. Excitation wavelength is 405 nm. The emission of the green channel at 470−560 nm and the red channel at 560−700 nm were collected. The ratio image was acquired by dividing the red channel with the green channel using Image J software. Scale bar is 20 μm. Figure S16. The CSLM images of medaka larva's caudal fin without amputation at different times after being fed with 5 μM CSMPP for 4 h, including brightfield, green channels, red channels, merged two-channel images, and ratiometric images showing the lysosomal pH distribution. For the green channel: λem = 416-555 nm; for the red channel: λem = 557-704 nm. λex = 405 nm. The ratiometric images of Em-red/Em-green were analyzed by using Image J software based on the fluorescence images. As a comparison with the amputated medaka larvae, the medaka larvae without amputation after being fed with CSMPP were photographed at the same time as that of photographing the amputated ones. The hours shown in the figure were named based on the hours after amputation. And "0 h" means the time after being fed with CSMPP and before amputation. Scale bar is 50 μm. Figure S17. The confocal images of fish cells and the caudal fin without being incubated by dyes. The green channel (416−555 nm) and the red channel (557−704 nm) images were captured with the same imaging condition as when tracking lysosomal pH. Scale bar is 50 μm.