3.1.Molecular design and insighted fluorescence mechanism
It is due to the cellular localization of the active BACE enzyme that inhibitors fail in their attempts to inhibit cellular activity. When exposed to an extracellular environment, BACE is inactive at pH 7.4, so may not cause the cleavage of APP. A pH value of four to five is required for BACE1 to acquire activity once it is endocytosed to an endosome[11]. BACE1 cannot be inhibited by inhibitors that cannot access these endosomal compartments. Due to limited accessibility to intracellular vesicles, conventional FRET probes are ineffective as well for monitoring BACE1 activity in vitro.
Fluorophore’s molecular design and on achieving excitation coefficients and optimal excitation/emission spectra have attracted an increasing experimental interest, but few efforts are devoted to fundamental mechanistic studies. In this paper, we in-sighted the fluorescence mechanism in the ESIPT systems of HBAE probe with the ideal quantum chemical tools. Using time-dependent density functional theory (TD-DFT), we explore the potential energy surfaces (PESs) of the lowest-lying excited states rather than analysing frontier orbital energy diagrams and performing ESIPT thermodynamics calculations and the radiative and nonradiative decay rates from the involved excited states are computed from first-principles using a thermal vibration correlation function formalism [25, 26]. With such strategies, our results reveal the real origins of the fluorescence intramolecular proton transfer.
The emission mechanism of these fluorescence plays an important role in the molecular design and construction of a functional system. We simulated three possible nonradiative decay channels from the S1-state Franck − Condon structure of Z-enol, then consuming its excited-state energy and the ESIPT process along the O − H distances from Z-keto*. The PBE0/Def2-TZVP optimized S1-MEPs, are shown in Fig. 1. The computational results indicate that the probe being excited to the S1 state, a nonradiative decay from the excited to the ground state was observed and barrierless ESIPT from enol to keto tautomer in the waterable solution.
According to the reported structure, Our calculations confirm Z-enol as the most stable configuration in the S0 state at the PBE0 level. we calculated configurations of HBT the vertical absorption energy at Z-enol including Z-keto, E-enol, and E-keto to predict its absorption spectra. The S0 → S1 excitation leading to a spectroscopically bright state with π − π* character, mainly results from the HOMO to LUMO excitation. The S0 → S1 vertical excitation energy calculated is 540 nm (3.85 eV, f = 0.46) by TD-PBE0 with def2-TZVP approach. Both of those results are in agreement to 540 nm the experimentally measured absorption maximum (Figure S1).
The tautomerized product, Z-keto*, shows vertical emission energy of 650 nm (2.53 eV) at the TDPBE0 level with the def2-TZVP approach (650 nm, 2.50 eV with SS-PCM calculation), which is in excellent agreement with the experimentally fluorescence maximum of 650 nm. Therefore, Z-keto* is assigned as the most likely emissive structure on the S1 state. In short, using the electronic-structure calculations in combination with the TDPBE0 level with the def2-TZVP approach, we have comprehensively investigated the possible emission channels of HBAE, thus confirming that fluorescence emission is owing to the effect of the ESIPT and subsequent bond-rotation relaxation, both of which are essential.
3.2. Mechanism studies
Inspired by the prominent fluorescent performances of HBAE towards neuro cells, we were then unveiled the underlying mechanism of the probe specificity and activity. According to previous reports, the effects of many brain-targeting compounds are attributed to the strong and selective inhibition of BACE1.[3, 5, 19] Thus, we hypothesized that the reason of the strong affinity of HBAE on neuro cells probably was the strong targeting binding of it towards BACE1, by which the probe ultimately achieved specific target of neuro cells. In other words, as well demonstrated above that the probe could perform neuro cells specific bio-imaging, we conclused that the intracellular fluorescence intensity should be highly associated with the content of BACE1, and it in turn could signal the location of the targeting BACE1.
To prove our thoughts, the corresponding docking between HBAE and BACE1 (PDB: 5I3Y) was carried out initially. It was performed through the ducking method by swiss predict. As depicted in Fig. 2A, the benzene ring on benzothiazole had Van der Waals force with Thr72 and Gln73, while the benzene-bridge had the same interaction with Val309. It was shown that HBAE was inserted into a nonpolar binding cavity of BACE1. It was further reflected that the fragments of the hydroxyl group and the cyano group of the probe were docked into the catalytic site of BACE1. These groups were interacted with Ser325 and Lys321 through hydrogen bonds, respectively. Besides, the N atom on benzothiazole also had a strong hydrogen bond with the amino acid residues of BACE1, including Gln73 and Thr 232 in Fig. 2B. These results suggested that the targeting ability of HBAE towards neuro cells could be based on the specific binding between the probe and BACE1.
3.3. Specific imaging of BACE1 in human and mouse cells
Since BACE1 in different human and mouse cell lines (HEK293 is a human embryonic kidney cell, U87-MG is a human glioma cell, N2a and bEnd. 3 are mouse derived cells) was analyzed using western blot images. As shown in Figure. S5., BACE1 contents obviously increased in N2a, HEK293, U87-MG and bEnd.3, which is the component part of the blood-brain barrier. These results demonstrated that the levels of BACE1 were highly expressed in the four cell lines. Binding affinity was another important factor for probes to imaging the BACE1. After the addition of HBAE to the BACE1, it was self-assembled to form nanomaterial and the corresponding NIR fluorescence enhancement was immediately found at 665 nm, demonstrating that HBAE showed specific binding affinity with BACE1. So, the highly sensitive NIR fluorescence response of HBAE with BACE1 made it promising nanomaterials for mapping BACE1 in AD mouse brain.
To detect and image BACE1 in the four cell lines, the four cells were stained with DAPI, HBAE and BACE1 specific antibody followed by the treatment with fluorescently (Alex fluo 488) labeled secondary antibody, and the images were captured under the fluorescence microscope (Fig. 3). Staining with Alex 488-BACE1, a specific antibody of BACE1, the green fluorescence is observed in the four cells. BACE1 is incubated with primary antibody binding and then decolored with green secondary antibody(Alex fluo 488), so green fluorescence represents BACE1, which is highly expressed in four cell lines (Fig. 3). These results revealed that the four cell lines all have the high levels of BACE1, and HBAE can also be well imaged on cells and overlapped with BACE-1 specific antibody fluorescence. it provides a basis for the HBAE probe qualified endogenous BACE1 detection in human and mouse brain cells.
Next, for imaging and sensing of endogenous BACE1 in live human and mouse cells, 5.0 µM HBAE probe was incubated with cells for 20 min. From Fig. 4, we can see that the probe successfully entered into the human and mouse cells as shown in the overlay channel, the probe HBAE can be obvious observed from the merge image that it is well colocated with Alex fluo 488, indicating it can target BACE1 protein. The images obtained from the four cells showed green fluorescence signal which confirmed the presence of BACE1 mainly in the cell membrance and cytoplasm (Fig. 4), and demonstrating that the developed HBAE probe was quite qualified for endogenous BACE1 detection as expected.
3.4.Evaluation of the blood brain barrier permeability
Having confirmed BACE1 specific NIR light-up responses in BACE1 highly expressed cells. we further comfirmed of the probe is proved by simulating the blood-brain barrier in vitro with Transwell plate. The bEnd. 3 cells planted in the upper chamber to simulate the blood-brain barrier in vitro, and then U87-MG cells are planted in the lower chamber. It can be found that within four hours, strong fluorescence could be captured in U87-MG cells indicating that HBAE can cross smoothly and has good permeability, the fluorescence staining diagram that the staining effect of U87-MG cells in the lower layer is obvious, indicating that the probe can penetrate the in vitro BBB model. The Transwell experiment analysis further confirmed that HBAE showed ideal permeability (Fig. 5). and realize the staining of lower cells, indicative of its potential for matching BBB penetrability. It provides ideal results for HBAE further imaging in living animals. In fact, via intravenous injection of HBAE (perfused with PBS), respectively, high-resolution imagine from the brain homogenate extraction of wild-type mice could be obtained to verify the BBB penetrability (Fig. 6).
3.5. In vivo imaging in the AD model
As shown in Fig. 6A, 22-month-old male AD-model (5XFAD) mice and age-matched wild-type mice were employed to observe the brain imaging by intravenous injection, to further confirm the practicability of HBAE for in vivo imaging BACE1. The strong fluorescence signals were observed in the brain position. In particular, the fluorescence intensity of HBAE in the brain regions of the AD-model mice was much higher than that in the control of wild-type mice at 120 min after postinjection, there is strong fluorescence at 2 h, and the fluorescence will be stronger with time lapse, and there is always a strong fluorescence at 24 h. In contrast, in wild-type mice, the fluorescence became very weak after 24 hrs indicative of specifically trapping BACE1 in vivo with probe HBAE. In addition, the cell viability of the probe HBAE by the CCK8 assays demonstrated its lower toxicity (Figure S6 in the Supporting Information). As shown in Fig. 6B, the fluorescent image of ex vivo brain of AD-model mice was obviously higher than that in the wild-type mice at corresponding time. As shown in Fig. 6C and D, the results indicated that HBAE could cross the blood–brain barrier and image BACE1 in vivo.
Ex vivo histology of HBAE binding to BACE1 in AD mice was carried out to further confirmed the in vivo performance. After 120 min of intravenous injection of HBAE, a higher number of fluorescence was observed in the mouse brain,liver and kidney,indicating the probe mainly distribution in these organs. It was further confirmed that the in vivo fluorescence signal was resulting from HBAE specifically binding to BACE1.
3.6. In vitro imagine of the hippocampus region in AD
It is well-known that the areas of the brain such as the hippocampus and cortex, that are primarily involved in memory processing, are likely to be first affected by AD memory loss. Immunohistochemistry was performed on the AD brain section to see which part had high expression of BACE1, from the experimental results in Figure.8A it can be seen that the levels of BACE1 in hippocampus and cortex areas were higher than those in other regions in AD mice. The images under red fluorescence channel revealed that the probe HBAE can well target proteins in AD mouse brain slices.
Next, the microscope images of the hippocampus region in AD mouse brain tissue labelled with the synthesized probe were obtained. After 120 min of intravenous injection of HBAE, the brain slice was obtained and incubated with BACE1 antibody, staining with secondary antibody, and finally stain the nucleus, a higher number of BACE1 were observed in the brain slices from 5XFAD mice (Fig. 8). As shown in Fig. 8B the probe HBAE showed the excellent colocalization of BACE1 during staining the same section with BACE1 antibody (5.0 µM) was subsequently used for evaluating the levels of BACE1 in AD (5XFAD) mouse brain tissues. Figure 8B illustrates different regions of the mouse brain slice, such as field of hippocampus. From Fig. 8B, the pseudocolor of the Fgreen/Fred channel changed from green to red in the regions hippocampus and their arounds, demonstrating that BACE1 in these regions of AD mouse brain was higher than other regious. As shown in Fig. 8A, BACE1 obviously stained in hippocampus areas by BACE1-antibody, compared with those in other areas in AD mouse brain. As shown in Fig. 8B, HBAE could well staining the highly expressed BACE1 in the hippocampus area after entering the mouse. These results demonstrated that the levels of BACE1 were nonuniform in different regions of AD mouse brain. From the experimental results in Fig. 8, the mapping of fluorescent imaging of BACE1 in hippocampus and cortex areas was obviously beyond regions in AD mice, which can be regarded as that BACE1 levels were closely related to the pathogenesis of AD.