A high‐performance cell‐labeling NIR‐II dye for in vivo cell tracking

Fluorescent dyes that emit in the second near‐infrared (NIR‐II, 1000–3000 nm) region have provided significant advances toward real‐time and high‐resolution imaging of vessel and lymphatic system. However, in vivo NIR‐II tracking of the fate of labeled cells still remains challenging. Here, we develop a shielding unit–donor–acceptor–donor–shielding unit (S‐D‐A‐D‐S) NIR‐II fluorophore (FE‐4ZW) with zwitterionic terminal groups for high‐efficiency cell labeling without using cell‐penetrating peptides, which provides for enhanced non‐invasive in vivo determination of the location of cell migration. The tethering terminal sulfoammonium inner salts are featured with its high affinity for cell membranes, thereby enabling the stable labeling even for fixed cells. The fate of transplanted stem cell and the tumor cell migration along lymphatic system in brain or periphery tissues are clearly monitored by the cell‐internalized FE‐4ZW. We also confirmed that a clinically used surfactant, D‐α‐tocopheryl polyethylene glycol‐1000 succinate, can reduce the liver and spleen uptake of FE‐4ZW. The fluorophore design strategy and cell‐labeling technology reported here open a new realm in the visualization of cell migration and insight into the relocation process, thereby ultimately providing an opportunity to investigate in greater detail of the underlying mechanisms of stem cell therapy and tumor metastasis.


INTRODUCTION
Fluorescence imaging in the second near-infrared (NIR-II) region allows for its application in disease diagnosis, imaging-guided surgery, and the recovery process by affording the monitoring of post-surgery therapy, especially in the fields of colon/brain tumor imaging/resection, tumor metastasis, sentinel lymph node mapping, and the determination of inflammation, stroke, and acute kidney injury. 1 Despite the significant advancements that have been achieved using NIR-II bioimaging in vivo recently, challenges still remain, especially in the non-invasive tracking of the fate of transplanted cells and the ability to discern the underlying migration/location mechanism through NIR-II bioimaging at the cellular level. 2 Especially, the important development of glymphatic system and central nervous system lymphatic vessels (meningeal lymphatic vessels, MLVs), which provide pathway for T cells or brain tumor cells migration, urge us to develop an imaging modality with non-invasive and high-resolution advantages for monitoring the cellular behavior. 3Traditional cell imaging generally relies on the use of fluorophores providing fluorescence emission in the visible and NIR-I spectral regions (400-700 nm and 700-900 nm, respectively). 4These fluorophores are primarily oriented toward in vitro cell imaging and are limited in use for in vivo cell tracking, given the reasons of substantial tissue autofluorescence and light scattering at such short wavelength regions.4b,f,5 Cell imaging at an optical window that offers less interference, such as the NIR-II region, could significantly contribute to this important field. 6ecent reports had made great efforts on stem cell tracking with NIR-II imaging or based on photoacoustic imaging in the NIR-II window.6c,7 The high temporal and spatial resolution of NIR-II imaging enabled the visualizing of the fate of stem cells after transplantation and tracking of the responsive behavior of transplanted stem cells in diseased environments.7a However, the applied NIR-II fluorophores were required to be conjugated with cell-penetrating peptides (CPPs) in order to effectively label the stem cells prior to transplantation.6c, 7a-c The use of a small organic NIR-II fluorophore was recommended as the first choice for in vitro and in vivo cell imaging due to low toxicity level and showing tremendous promise for potential clinical translation.Cell imaging and in vivo cell tracking using natively permeable organic small-molecule NIR-II fluorophores, especially without conjugated CPPs moiety, have not been achieved to date.Thus, developing such forms of cell imaging contrast agents is essential toward discerning the cell migration and cell relocation processes. 8he parameter of long-term photostability is an essential feature of an NIR-II fluorophore to meet the requirement for in vivo cell tracking. 9Among the various types of NIR-II fluorophores, recent literature had highlighted that shielding unit-donor-acceptor-donorshielding unit (S-D-A-D-S) dyes could demonstrate exceptional photostability. 10Herein, we capitalized on a well-established small-molecule core structure that has an S-D-A-D-S backbone to develop an NIR-II fluorophore that could readily interact with the lipid mono/bilayer of cellular walls and compartments.In the meantime, we systematically screened different terminal groups (zwitterionic and non-zwitterionic), which were tethered to the peripheral shielding units of the NIR-II fluorescent core (Scheme 1A).Ultimately, this method of screening allowed us to develop a permeable NIR-II fluorophore (FE-4ZW) with zwitterionic groups that consequently provided high-efficiency in vivo cell tracking (Scheme 1B).More specifically, the reasonable molecular structure design of its zwitterionic sulfoammonium groups afforded us with an ideal cell tracking probe for cellular dynamics research.The migration and biodistribution of the labeled cells (neural stem cells [NSCs] and tumor cells) were non-invasively visualized in the NIR-II imaging window, which conferred high temporal and spatial resolution.Importantly, the tracker assisted us to achieve the realtime imaging of brain tumor cells metastasis into cervical lymph node through glymphatic system and MLVs.Both active and passive cellular transport pathways for internalization of FE-4ZW were observed.The displayed FE-4ZW NIR-II dye affords a convenient synthesis direction to obtain natively permeable cell contrast agent based on small organic NIR-II fluorophores and offers researchers potential opportunities for non-invasively monitoring the fate of labeled cells and better understanding disease progression at the cellular level.

Characterization and imaging of the zwitterionic NIR-II fluorophores
To obtain a bright NIR-II fluorophore for in vitro and in vivo cell imaging, we followed a robust synthetic protocol to manufacture a series of three S-D-A-D-S-based dyes.We chose to use the S-D-A-D-S framework for our dyes also because they are chemically inert and photophysically stable such that degradation and photobleaching would not be of concern. 10We maintained the same backbone for the S-D-A-D-S dyes with benzobisthiadiazole moiety as the acceptor, 3,4-ethylenedioxy thiophene as the donor moiety, and dialkyl fluorene as the shielding unit, but adjusting the tethering terminal groups of the shielding units by appending to them a small panel of different moieties ranging from the commonly used uncharged polyethylene glycol (PEG, M w = 600) to zwitterionic sulfoammonium groups.By doing so, we were able to prepare and characterize  S1).Compared to the commercially available and clinically used fluorophore, indocyanine green (ICG), all synthesized FE dyes exhibited stable peak wavelengths of fluorescence emission within the NIR-II optical imaging window, thereby allowing a higher resolution imaging with enhanced visual contrast of biological features (Figure S1).For our experiments, a stock solution of FE dyes in phosphate buffer saline (PBS) media was prepared in advance.The stock solution was diluted accordingly for in vitro and in vivo experiments.
Meanwhile, the dynamic light scattering (DLS) and zeta potential measurements were performed to test the status of FE-4ZW, FE-2PEG, and FE-2PEG2ZW in culture buffer (Figure S2).The hydrodynamic diameter of FE-4ZW was determined to be approximately 70 nm and the zeta potential measurement was zero.FE-2PEG and FE-2PEG2ZW showed smaller hydrodynamic diameter compared with FE-4ZW, which was determined to be approximately 15 and 7 nm, respectively.The finding of a larger hydrodynamic diameter for FE-4ZW was considered possibly attributed to intermolecular aggregation induced by the tethering zwitterionic sulfoammonium groups.The mean zeta potentials of FE-2PEG and FE-2PEG2ZW were determined to be −1.77and 0.16 mV, respectively.
We further measured the photostability of FE-4ZW fluorophore (with ICG as a reference) in microcentrifuge tubes and mice lymph nodes under continuous 808 nm excitation.The results revealed that the FE-4ZW demonstrated superior photostability with only marginal photobleaching than the clinically used ICG probe (Figure S3).This property provides opportunities for long-term and repeatable monitoring of regions of interest.
Next, we investigated the in vivo pharmacokinetics of FE-4ZW and FE-2PEG after separate i.v.administration.Compared to FE-2PEG, which has a long circulation time and features partial renal excretion, FE-4ZW showed faster accumulation in the liver and spleen within a short blood circulation time (∼50 min).This was followed by an increase in fluorescence signal intensity in liver and spleen over time, which gradual decreased (Figures 1A-C  and S13).Based on the faster liver-uptake feature of FE-4ZW versus FE-2PEG, we hypothesized that when the FE-4ZW was injected into the blood circulation system of mouse, they were also endocytosed by liver cells and/or macrophages (Kupffer cells).This difference of finding tentatively points toward the FE-4ZW having a much faster accumulation rate compared with FE-2PEG, thereby indicating that FE-4ZW may have stronger interaction with cell membranes, and thus could potentially serve as the better cell contrast agent.Thus, we designed extensive in vitro and/or in vivo experiments to measure the capacity of FE-4ZW to function as a cell contrast agent.

Cell imaging in the NIR-II window
Methods to observe cellular fate with high resolution are necessary to investigate a host of important physiological and pathological processes.However, those methods are limited to implanted imaging chamber with visible fluorophores that inevitably provide lower imaging resolution and penetration depth.Thus, in vitro highefficiency cell labeling, following by in vivo cell tracking with high-definition imaging are essential for evaluating the cellular dynamics in clinical treatment, such as burgeoning use of stem cell therapies. 11For this purpose, we systematically investigated the cell labeling and in vivo cell tracking properties of FE-4ZW against a series of cell types using NIR-II fluorescence imaging.
Biosafety is a major concern in the development and application of ideal cell tracker. 12We first used cell counting kits (CCK8) assay to evaluate the cell toxicity of the FE dyes series (FE-4ZW, FE-2PEG, and FE-2PEG2ZW) in the concentration range from 0 to 20 µM by the way of separately incubating selected cell lines (4T1 and L-O2) with the respective FE fluorophores for 12 h.As expected, the FE dyes (FE-4ZW, FE-2PEG, and FE-2PEG2ZW) exhibited reasonable cell compatibility at the tested concentration up to 20 µM, thereby eliminating any toxicity concerns when they were later used (within the evaluated concentration range) as NIR-II fluorescence imaging cell tracker (Figure S4).In our cell-labeling protocol, a concentration of 10 µM of the FE dyes was separately used to label the selected cell lines.Our protocol, which does not call for the use of CPPs is more streamlined and straightforward for obtaining fluorescent cells due to the potential interaction and/or affinity between FE-4ZW and the cell membranes (Figure 1D-F).Several types of cell lines, including liver, macrophage, stem, and tumor cells were utilized to evaluate the series of FE fluorophores.Throughout the duration of co-incubation of FE-4ZW with different cell lines, the NIR-II fluorescence intensity of FE-4ZW-labeled cells increased over time, wherein sufficient labeling density at the 12 h timepoint allowed for bright NIR-II imaging (Figures 1F and  S5).For all tested cell lines, the bright NIR-II fluorescence signal by FE-4ZW allowed unambiguous outlining of the cell shapes upon internalization (Figures 1E,F  and S5).
In contrast, in the L-O2 cells group, we obtained very low fluorescence intensities when they were incubated with the FE-2PEG or FE-2PEG2ZW fluorophores.These results indicated that the molecular structure played an important role in functioning as an efficient cell-labeling biotechnology (Figure 2A,B).The intracellular photostability of FE-4ZW was also determined by placing the labeled L-O2 cells under continuous laser irradiation (Figure 2C).The fluorescence signal only decayed at an extremely slow rate when monitored over a time course of 2 h.Conversely, as expected, intracellular ICG photostability exhibited instantaneous photobleaching in a matter of seconds (Figure S6).Collectively, our FE-4ZW was found to be an ideal cell imaging agent with bright intracellular NIR-II fluorescence emission and excellent photostability.Such superior properties of FE-4ZW set the stage for non-invasively tracking cellular fate in vivo.
Except photostability, intracellular stability is another essential factor of a cell tracking agent for having the ability of long-term tracking and repeated imaging.To measure the intracellular stability of FE-4ZW as a long-term cell tracking agent, a leakage experiment was conducted according to a well-established protocol.7c After 12 h incubation using FE-4ZW in cells, the cell media, which contained non-internalized FE-4ZW, was replaced with fresh media.The labeled cells were sequentially cultured up to 7 days and the supernatants were collected each day and imaged by an NIR-II detector.No detectable NIR-II signal was observed for all collected supernatants, while the cells themselves exhibited bright NIR-II fluorescence signal, which indicated stable internalization of the intracellular FE-4ZW dyes in the labeled cells (Figure 2E,F).Another complementary need for in vivo cell tracking to adequately detect the proliferated cells even after the primary labeling signal shows depletion during cell proliferation.Stable NIR-II fluorescence signals were detectable for up to 6 days even as the cell count increased exponentially (confirmed by the CCK8 assay) (Figure 2G-I).The above results revealed that the labeling and/or internalization of FE-4ZW in cells could maintain for a long time, thereby providing an opportunity for monitoring complicated physiological process on the cellular level by using high-contrast NIR-II fluorescence bioimaging.

Cellular internalization mechanism of FE-4ZW
After establishing an in vitro cell labeling protocol, we remained interest in the process and mechanism of FE-4ZW internalizing into the several types of cell lines.At the beginning of this study, we hypothesized that energy-dependent endocytosis exclusively contributed to the FE-4ZW internalization. 13Thus, the co-incubation of FE-4ZW with L-O2 cells was performed at 37 • C, 27 • C, and 4 • C, respectively, for 3 h.After detecting and quantifying the NIR-II fluorescence signal at wavelength over 1200 nm for each group of cells, we verified that lower incubation temperature resulted in lower FE-4ZW labeling efficacy (Figure 3A,B).Incubation of FE-4ZW with the cells at 37 • C revealed a "saturated" level of uptake, but incubation at 27 • C and 4 • C lowered the FE-4ZW uptake, instead of completely inhibiting the uptake (Figure 3A,B).In addition to modulating the co-incubation temperature, typical uptake inhibitors were utilized to further confirm the internalization pathway of FE-4ZW.Well-established inhibitors that included chlorpromazine, methyl-β-cyclodextrin, and amiloride were selected for the inhibition of  13 L-O2 cells were separately incubated with the selected inhibitors for 45 min, followed by incubating with FE-4ZW for 3 h.Results verified that endocytosis is an important, and not the only, pathway for cell internalization of FE-4ZW, as no inhibitors could completely inhibit the uptake process (Figure 3C,D).Again, consistent with the above results, it was revealed that endocytotic processes, as active transport pathways, were not the exclusive pathway for FE-4ZW internalizing into cell.Next, we aimed to provide corroborative evidence that would help verify the occurrence of passive transport of FE-4ZW using fixed L-O2 cells.No fluorescence signal was detected for either FE-2PEG or FE-2PEG2ZW (at the same dosage as FE-4ZW of 10 µM) when incubated with fixed cells.Conversely, FE-4ZW-treated fixed cells emitted an intense NIR-II fluorescence signal above 1200 nm, thereby indicating that FE-4ZW could automatically internalize into cell and further unveiling FE-4ZW has potential as a permeable cell imaging agent (Figure S7).
To further unveil the reason for high intracellular delivery of FE-4ZW, another FE fluorophore derivative (FE-4SO 3 Na) that lacks zwitterionic terminal groups (due to lack of an ammonium group) but maintains the negatively-charged sulfonate terminal groups was compared to FE-4ZW following its synthesis (Figures 3E  and S17).The measured zeta potential of the FE-4SO 3 Na was determined to be sharply negative in value (Figure S8).First, the fluorescence signal intensity of FE-4ZW and FE-4SO 3 Na at the same concentrations (2.5, 5, and 10 µM) was tested.Results showed that FE-4ZW and FE-4SO 3 Na exhibited similar fluorescence signal intensity over 1200 nm (Figure 3F,G).Subsequently, the cell-labeling protocol was performed, which involved administration of the FE-4ZW or FE-4SO 3 Na fluorophores.The FE-4ZWtreated cells provided higher fluorescence signal intensity than FE-4SO 3 Na-treated cells after separately co-cultured for 3 h.Extending the co-culturing time to 12 h, the fluorescence signal intensity of FE-4ZW-treated cells showed notable enhancement, as expected.Meanwhile, negligible fluorescence signal enhancement resulted from the FE-4SO 3 Na-treated cells (Figure 3H,I).To further prove the rationality of introducing zwitterionic sulfoammo-nium groups responsible for the enhanced internalization ability, a high concentration (50 µM) of FE-2PEG and FE-2PEG2ZW with lower-staining efficiency was utilized to co-culture with live or fixed cell (Figure S9).The results showed that FE-2PEG failed to stain both live and fixed cell at the high concentration.For FE-2PEG2ZW, luminous cells represented weak affinity between FE-2PEG2ZW and the cells.The distinction of molecular structure and staining ability between FE-2PEG and FE-2PEG2ZW further highlighted the important role of the zwitterionic sulfoammonium groups.These results verified the zwitterionic sulfoammonium group is an important factor for cell internalization of fluorophores, which possibility contributed to enhance the interaction with cell membranes or permeable ability.

In vivo fluorescence cell tracking in the NIR-II optical window
We demonstrated intravital NSC tracking by pre-culturing NSCs with FE-4ZW, followed by i.v.administration of the labeled NSCs.According to an established protocol, the luminescent NSCs that were labeled by FE-4ZW were carefully collected by centrifugation, and subsequently were intravenously transplanted into the blood circulation system of a Balb/C murine model (Figure 4A).The NIR-II fluorescence signal afforded clear outlining of the lung upon immediate observation, thereby revealing that the labeled NSCs first accumulate in the lung (Movie S1).This result was consistent with previous report in term of mesenchymal stem cells monitoring.7c The NIR-II fluorescence signal from the FE-4ZW-labeled NSCs within the lung gradually decreased over time and nearly disappeared after 72 h timepoint.Throughout the duration of this experiment, an NIR-II fluorescence signal in the liver progressively increased.These results verified that the transplanted NSCs after i.v.administration were gradually transported to the liver following initial accumulation in the lung (Figure 4B-E).Longitudinal non-invasive in vivo imaging in real time, via tracking NSCs, could provide a convenient cell tracking technology for investigating the physiological mechanism and discerning the obstacles for NSC therapy in clinic. 11,14tracellular FE-4ZW.(E and F) After being labeled with FE-4ZW, the combination of lower signal of cultured supernatant up to 7 days along with the higher NIR-II fluorescence signal of the trypsinized cells verified the intracellular stability of FE-4ZW.(G-I) The remaining NIR-II fluorescence emission intensity along with the cell proliferation further highlights the stability of FE-4ZW as a dye that affords efficient cell tracking performance.White scale bar: 50 µm.The imaging condition for (E) is 808 nm laser, 1000 nm long-pass emission filter, and 65 mW/cm 2 .The imaging condition for (A), (C), and (G) is 808 nm laser, 1200 nm long-pass emission filter, and 4.25 W/cm 2 .****p < .0001;two-way analysis of variance (ANOVA).Monitoring the fate of tumor cells in vivo is critical for investigating potential tumor metastasis, yet current non-invasive strategies have failed to achieve this goal.FE-4ZW-labeled 4T1 tumor cells were intradermally injected and the tumor cells (or cell aggregation clusters) were observed to migrate along the lymphatic vessels over time (Figure 4F), whereby they eventually seeded in the adjacent lymph nodes (Figure 4G,H).The in vitro leakage experiment and cell proliferation experiment were con-ducted to test the intracellular stability of FE-4ZW.The results showed that after cellular internalization, FE-4ZW could steadily exist in cell compartments (Figure 2D-I).Furthermore, in vivo NSC cell tracking results showed that the NSCs labeled with FE-4ZW were retained in the lung for a long time and gradually circulated into the liver.This suggested that the internalized FE-4ZW was not detached from the NSCs.If the FE-4ZW was detached from cells, the outline of lung would rapidly disappear (Figure 4B,D).The presence of free-moving fluorescence dots along lymphatic vessels in Figure 4F indicated that cell aggregation clusters were being transported along lymphatic vessels, rather than the FE-4ZW dyes were being free moving.The method reported here is potentially applicable to the monitoring of the fate and migration pathway of tumor cells, such that key parameters in tumor metastasis could be identified.

FE-4ZW enabled imaging glymphatic system and tracking brain tumor cells migration
The cerebrospinal fluid (CSF) circulation is significant physiological process in brain, contributing to clearance of potentially harmful metabolic wastes. 15The glymphatic system and meningeal lymphatic vessels (MLVs) have been considered as important pathways for assisting CSF circulation.Some interesting studies have demonstrated that the MLVs have function for draining immune cells, brain tumor cells or neurotropic viral to peripheral lymph nodes (cervical deep lymph nodes [dCLNs] and cervical superficial lymph nodes [sCLNs]).3b,c,8a,16 However, the imaging modalities for studying the role of MLVs and cervical lymph nodes in uptaking cells mainly depend on twophoton microscopy and/or harvest these tissues for in vitro microscope imaging.Hence, we try to demonstrate the potential of FE-4ZW as contrast agent for in vivo tracking CSF circulation and tracking cell migration in MLVs and cervical lymph nodes.Figure 5A illustrates the experimental process for demonstrating the potential of FE-4ZW as CSF tracker.The processes of exposure of Cisterna Magna (CM), CM injection of FE-4ZW, and acquirement NIR-II images were all performed under ISO anesthesia.The supplement drug dexmedetomidine was delivered by intraperitoneal injection at a dose of 0.2 mg/kg for facilitating the CSF flow along perivascular space (PVS).Figure 5B shows representative time-course images of FE-4ZW distribution at the dorsal brain surface that indicates that the FE-4ZW could image the CSF circulation pathway by CM delivery through the intact scalp.Afterwards, we collected the brain tissues and imaged the distribution of FE-4ZW on brain tissue surfaces at different poses.Figure 5C shows that the CM-injected FE-4ZW distribution along PVS and the outline of PVS surrounding the middle cerebral artery were clearly lighted (Figure 5D).After demonstrating FE-4ZW as a desired CSF tracer, we performed the in vivo monitoring of brain tumor cells migration by CM injection of FE-4ZW-labeled GL261 cells (Figure 5E).After injecting FE-4ZW-labeled GL261 cells, we monitored the traveling pathway of GL261 brain tumor cells.The unambiguous MLV structures were imaged, demonstrating that the CM-injected cells traveled along MLVs following with CSF circulation (Figure 5F).The disappeared MLV structure implied that the injected cells had time-dependent migration process (Figure 5F,G).We collected the MLVs after allowing the cell circulation for 2 h and imaged the distribution of cells within MLVs by in-house NIR-II fluorescence microscope.The representative images showed that the injected cells mainly distributed within the MLVs covering the transverse sinus and olfactory bulb (Figures 5H,I and S10).We then removed the cervical skin and exposed the CLNs under the imaging setup after allowing the cells circulation for 2 h, the surrounding fluorescence signal demonstrated that the CLNs (sCLNs and dCLNs) may have important function for central nervous system diseases, such as brain tumor (Figure 5J).We used the NIR-II fluorescence microscope to visualize the detailed distribution of injected cells in sCLNs and dCLNs.The sCLNs and dCLNs covered with bright fluorescence dots imply that GL261 cell clusters have migrated into sCLNs and dCLNs (Figure 5K).The larger fluorescence area in FE-4ZW-labeled group than PBS group further confirmed that the sCLNs and dCLNs have filled with labeled GL261 cells (Figure 5L).After circulation for 12 h, the bright CLNs further demonstrated that the CLNs were important draining lymph nodes for brain tumor (Figure S11).The results from in vitro cell imaging and in vivo cell tracking studies confirmed the FE-4ZW could be competent as a cell contrast agent.The advantages of our FE-4ZW include: (1) sufficient maintenance in photostability that enables repeated imaging requirements; (2) high-efficiency celllabeling ability without using CPPs such as TAT, a type of CPPs.So, it is important to pay more effort to expand the family of cell contrast agents based on organic NIR-II small-molecule fluorophore.

2.6
Reduction of liver and spleen uptake of FE-4ZW when using D-α-tocopheryl polyethylene glycol-1000 succinate/FE-4ZW cocktail We have estimated that the FE-4ZW possessed high affinity with cells and had potential as a cell tracker.The high affinity with cells may be an underlying reason for faster accumulated in liver and spleen after i.v.injection of FE-4ZW.Next, we tried to regulate the in vivo pharmacokinetics of FE-4ZW by supplement with a surfactant D-α-tocopheryl polyethylene glycol-1000 succinate (TPGS).The surfactant TPGS could form a shell with a hydrophilic outer surface that could chaperone FE-4ZW in aqueous solution, and thus alter its affinity for and interaction with cell membranes (Figure S12A).The remarkable decline of the size from FE-4ZW to FE-4ZW/TPGS cocktail verified the co-assembly process.The size change from FE-4ZW to FE-4ZW/TPGS cocktail was also verified by transmission electron microscopy and DLS measurement (Figure S12B-D).We also measured the fluorescence signal intensity of FE-4ZW using fixed cells in the presence of 2 mg/mL TPGS, and the lower fluorescence signal intensity (∼75% reduction) verified that the TPGS could inhibit entrance of FE-4ZW into cells (Figure S12E,F) over long-term incubation (12 h).Thus, we hypothesized that additional TPGS could also regulate the bio-distribution of FE-4ZW in vivo by helping the injected FE-4ZW escape the cytophagy.At the initial stage (within 30 min) after separate i.v.administration of FE-4ZW and the FE-4ZW/TPGS cocktail, there were no obvious differences in the liver/spleen NIR-II fluorescence signal intensities (Figure S13A-D), which were due to the inherent accumulation of S-D-A-D-S dyes in the liver and spleen. 17However, allowing circulation up to 3 h, the lower fluorescence signal of liver and spleen in FE-4ZW/TPGS group verified that the TPGS@FE-4ZW cocktail could limit the interaction between the FE-4ZW and immune cells due to the exposed outer surface PEG chains (Figure S13A-D).Taken together, we revealed that the commercially/clinically available surfactant TPGS could reduce liver and spleen accumulation of FE-4ZW in the small animal model.

CONCLUSION
NIR-II fluorescence-emitting S-D-A-D-S fluorophores with tethering zwitterionic sulfoammonium terminal groups provided a means to afford high-efficiency labeling of living cells without the need of CPPs.The four introduced zwitterionic sulfoammonium terminal groups not only conferred to FE-4ZW excellent aqueous solubility, but also improved the affinity with cell membranes when co-incubation with various types of cells.Notably, FE-4ZW was used for in vivo monitoring of the fate of the transplanted cells and brain tumor cells migration into cervical lymph nodes.We demonstrated that the FE-4ZW dye internalizes in the cell and is retained in the cells without obvious cellular cytotoxicity based on morphology.Moreover, the NIR-II fluorescent dye, FE-4ZW, afforded long-term monitor and visualization of labeled cells.In addition, although the NIR-II fluorescence signal intensity decreased with the proliferation of primary FE-4ZW-labeled cells, the NIR-II fluorescence signal intensity after 6 days of proliferation was still detectable.Stem cell therapy can repair damaged cells at the targeted location by reducing inflammation and modulating the immune system, and thus it is critical to evaluate the accumulation efficiency of administered NSCs. 18Benefiting from the high NIR-II fluorescence signal intensity and excellent intracellular chemical/photostability, the transfer and bio-distribution of NSCs were non-invasively traced using FE-4ZW.This is the first time to utilize an organic small-molecule NIR-II fluorescence-emitting fluorophore that does not have CPPs, to allow for monitoring of the migration process of stem cells after i.v.administration in a small animal model.In tumor metastasis, cancer cells can escape from the primary tumor, migrate along with the blood and/or lymphatic vessels, and seed a new tumor in distant organs/tissues.8a, 19 As a natural consequence borne out of our cell-labeling protocol, our tailored fluorophores also provide a potential means to visualize particular aspects of such metastatic processes in vivo using a non-invasive pathway.
Previous reports indicated that the modification of zwitterionic groups on a cyanine NIR-I fluorescenceemitting fluorophore can lower the interaction between the fluorophore and proteins, and thus impart the cyanine dye with faster renal excretion than if otherwise. 20owever, the zwitterionic S-D-A-D-S structure of our NIR-II-emitting fluorophore (FE-4ZW) promoted high affinity for and interaction with cell membranes.The designed zwitterionic sulfoammonium group was the identified factor contributing for the high internalization ability of FE-4ZW.This valuable conclusion was supported by the highest labeled efficiency of FE-4ZW than of that for other FE dyes (FE-2PEG, FE-2PEG2ZW, and FE-4SO 3 Na).The cell uptake mechanism of FE-4ZW included both energy-dependent endocytosis and energy-independent passive transport.Apparent fluorescence signal intensity depletion that was observed by the decrease in the culture temperature and additional endocytosis inhibitors revealed that endocytosis was one of a few potential uptake pathways for FE-4ZW.Another uptake mechanism could be explained in the collective following points: (1) the FE-2PEG failed to internalize into fixed cell, and even enhance the work concentration up to 50 µM; (2) performing the cell uptake under the low-temperature condition or in the presence of inhibitors could not completely inhibit the uptake of FE-4ZW, respectively.Furthermore, the reduced liver and spleen accumulation of FE-4ZW was realized upon using the TPGS/FE-4ZW cocktail.FE-4ZW offers exciting possibilities for enhancing the imaging quality for in vivo cell tracking.To address some unmet practical needs, the following improvements could be considered: (1) extending the emission wavelength of FE-4ZW to the NIR-IIb could further improve the imaging quality due to the wavelength-dependent penetration depth; (2) simplifying the synthesis process of FE-4ZW would make it more suitable for clinical applications.Overall, our NIR-II fluorescence-emitting fluorophore, FE-4ZW, provides a convenient strategy for non-invasively tracking the transplanted cells with high resolution and offers a rational design strategy to produce high-performance cell imaging agents emitting fluorescence signal in the NIR-II window.

C O N F L I C T O F I N T E R E S T S TAT E M E N T
The authors declare no conflict of interest.

S C H E M E 1
Illustration of the fabrication of FE-4ZW as cell tracer and in vivo tracking the cell migration in blood circulation and glymphatic system.(A) Screening the functional units yielded the superior cell imaging probe.(B) Organic small-molecule fluorophore FE-4ZW served as a high-performance contrast agent for in vivo cell tracking.

F I G U R E 1
The abilities of FE fluorophores to afford effective second near-infrared (NIR-II) fluorescence imaging contrast levels.(A) Delivery of the FE-2PEG and FE-4ZW into blood circulation system by tail-vein injection.(B and C) Whole-body imaging of FE-4ZW and FE-2PEG (300 µM, 200 µL) in mice within a 50-min period (808 nm laser, 1100 nm long-pass emission filter, and 65 mW/cm 2 ).White scale bar: 1 cm.(D) Schematic showing the protocol for FE-4ZW labeling of select cell lines and fluorescence microscopy imaging in the NIR-II window.(E and F) NIR-II fluorescence images of FE-4ZW (10 µM) cultured in select cell lines for 12 h using an in-house setup for NIR-II microscopy (808 nm laser, 1200 nm long-pass emission filter, and 4.25 W/cm 2 ).White scale bar: 50 µm.

F
I G U R E 2 FE-4ZW as a high-performance cell-labeling dye.(A and B) Uptake comparison of a series of synthesized second near-infrared (NIR-II) fluorescence-emitting FE fluorophores.High NIR-II fluorescence emission intensity from labeling cells (L-O2) with FE-4ZW is indicative of efficient cell uptake.Both FE-2PEG and FE-2PEG2ZW failed to label L-O2 cells, and thus exhibited negligible fluorescence emission intensity upon irradiation.Scale bar: 20 µm.(C) Photostability of internalized FE-4ZW under continuous irradiation by 808 nm laser at a power density of 4.25 W/cm 2 .White scale bar: 50 µm.(D) Schematic showing the protocol for leakage assay of the clathrin-mediated endocytosis, caveolae-mediated endocytosis, and macropinocytosis, respectively.

F I G U R E 3
Second near-infrared (NIR-II) imaging-guide studying the internalization mechanism of FE-4ZW.(A and B) The L-O2 cell line was used as model cell line to evaluate FE-4ZW uptake via temperature-dependent endocytosis.(C and D) The influence of select uptake inhibitors on the cellular internalization of FE-4ZW.(E) The chemical structures of FE-4ZW and FE-4SO 3 Na.(F and G) The fluorescence signal intensity of FE-4ZW and FE-4SO 3 Na at the concentrations of 2.5, 5, and 10 µM in microcentrifuge tubes, respectively.The NIR-II fluorescence image was acquired at the exposure time 100 ms with such emission collected over 1200 nm.(H and I) Cell imaging in the NIR-II window following co-incubation with FE-4ZW and FE-4SO 3 Na (10 µM) at different timepoints 3 and 12 h.FE-4ZW has higher cell-labeling efficacy compared with FE-4SO 3 Na.The imaging condition for cell images is 808 nm laser, 1200 nm long-pass emission filter, and 4.25 W/cm 2 .White scale bar: 50 µm.**p < .01,***p < .001,****p < .0001;two-way analysis of variance (ANOVA).

F I G U R E 4
Tracking the location of transplanted neural stem cells (NSCs) via performing second near-infrared (NIR-II) fluorescence imaging using FE-4ZW.(A) Schematic illustration of the protocol for intravital NSCs tracking by cellular internalization of FE-4ZW.(B-E) In vivo distribution of i.v.administered FE-4ZW-labeled NSCs evaluated via using NIR-II fluorescence imaging.The imaging condition is 808 nm laser, 1100 nm long-pass emission filter, and 65 mW/cm 2 .(F) Real-time monitoring of tumor cells (or cell aggregation clusters) migration using 4T1 cells labeled with internalized FE-4ZW.The images were acquired from supine position.The yellow dashed line is a baseline for indicating the movement of injected cell clusters.The yellow arrow indicates the location of injected cell clusters.(G and H) Anatomical structure revealed the migration route of FE-4ZW-labeled 4T1 cells after intradermal injection.The imaging condition is 808 nm laser, 1000 nm long-pass emission filter, and 65 mW/cm 2 .White scale bar: 1 cm.

F I G U R E 5
FE-4ZW allowed for in vivo imaging of glymphatic pathway and tracking brain tumor cells circulation.(A) Schematic representation of the FE-4ZW as cerebrospinal fluid (CSF) tracer for imaging the glymphatic pathway.(B) Monitoring of FE-4ZW-filling CSF distribution after Cisterna Magna (CM) injection by local brain imaging with intact scalp.(C) Imaging the distribution of the second near-infrared (NIR-II) tracer at the brain surfaces after allowing circulation for 70 min.The imaging condition is 808 nm laser, 1100 nm long-pass emission filter, and 65 mW/cm 2 .(D) Cross-sectional fluorescence signal along the white dashed line inserted in (C) was collected to show the FE-4ZW movement along perivascular space (PVS).(E) CM injection of exogenous FE-4ZW-labeled GL261 brain tumor cells for in vivo imaging the interrelation between brain tumorous metastasis and glymphatic system.(F and G) In vivo imaging of exogenous GL261 cells migrating into meningeal lymphatic vessels (MLVs) with the removal of scalp.The imaging condition is 808 nm laser, 1100 nm long-pass emission filter, and 65 mW/cm 2 .White scale bar: 2 mm.(H and I) Representative NIR-II fluorescence images of FE-4ZW-labeled GL261 cells along MLVs after injection for 2 h.White scale bar: (H) 5 mm and (I) 3 mm.(J) In vivo imaging of exogenous GL261 cells migrating into cervical deep lymph nodes (dCLNs) and cervical superficial lymph nodes (sCLNs).White scale bar: 2 mm.(K) NIR-II macroscopy imaging of FE-4ZW-labeled GL261 cells migrating into cervical lymph nodes after injection for 2 h.(L) NIR-II macroscopy imaging of cervical lymph nodes after receiving injection of PBS solution for 2 h.The imaging condition is 808 nm laser, 1000 nm long-pass emission filter, and 65 mW/cm 2 .White scale bar: 3 mm.
This research was supported by the National Key Research and Development Program of China (No. 2022YFC2408100), the Guangdong Provincial Natural Science Foundation-Yueshen Joint Fund (Youth Project) (No. 2019A1515110464), the Shenzhen Science and Technology Innovation Commission-Free Exploration/General Project (No. JCYJ20190812151209348), and the Shenzhen Science and Technology Innovation Commission-Technology Breakthrough Project (No. JSGG20191231141403880).