Chapter 1: Fluorescent Probes for NIR-I Bioimaging

Fluorescence imaging in the near-infrared (NIR) region provides precise spatiotemporal information on biomolecules, particular for in vivo imaging, since the imaging depth is greatly improved relative to fluorescence imaging in the visible region. Small molecule fluorescent probes are the main tools for realizing fluorescence imaging. In principle, a good fluorescent probe is composed of a recognition unit and a fluorophore separated by a short linker spacer. So, the design of fluorescent probes for NIR bioimaging should be first based on the preparation of NIR fluorophores. The NIR window (650–1700 nm wavelength) is divided into two regions: the first near-infrared window (NIR-I, 650–900 nm) and the second near-infrared window (NIR-II, 1000–1700 nm).¹  A variety of NIR-I fluorophores have been developed and most of them can be classified as types of cyanine dyes, xanthene dyes, BODIPY dyes, aggregate-induced emission luminogens (AIEgens) and “crossbreeding” dyad-based dyes.²  In this chapter, we will discuss small molecule fluorescent probes by highlighting the latest publications for NIR-I bioimaging, focusing on the fundamental principles of the design, synthetic protocols and photophysical properties of the NIR-I fluorophores and fluorescent probes. The future direction of bioimaging with these NIR-I fluorescent probes will also be discussed.

Generic cyanine dyes consist of two nitrogen centers, one of which is positively charged and linked by a conjugated chain of an odd number of carbon atoms to the other nitrogen (Scheme 1.1a). This prominent feature has been studied for “push–pull” alkenes and also forms the basis of the polymethine dyes, which contain the streptopolymethine unit as the chromophore.³ 

These derivatives have an analytical wavelength (600–900 nm) in the near-infrared region, and the properties of cyanine dyes can be improved by modifying their structure, such as increasing water solubility and stability and red-shifting the emission wavelength.

Cyanine dyes can decompose or degrade into hemicyanine products. Structurally, hemicyanines may have a quaternary ammonium heterocycle (serving as an electron acceptor) at one end and a phenolic hydroxyl or amino group (acting as an electron donor) at the other end, connected by a conjugated system (Scheme 1.1b).⁴  This creates a typical push–pull alkene structure, characterized by an electron donor, a conjugated bridge and an electron acceptor. The innovative hemicyanine dyes stand out for their adjustable optical characteristics, including NIR absorption and emission capabilities. Significantly, these dyes demonstrate excellent photostability and a high fluorescence quantum yield. Due to their advantageous NIR photochemical properties, hemicyanine dyes and their derivatives have been widely employed as robust foundations for developing NIR probes for in vivo imaging applications. This section focuses on the applications of cyanine dyes and hemicyanine dyes in NIR-I bioimaging.

Cyanine dyes can be classified into monocyanines, trimethylcyanines (Cy3), pentamethine cyanines (Cy5), heptamethine cyanines (Cy7) and so on (Scheme 1.2), according to the length of their methine chain. However, it is worth noting that the absorption and emission spectrum of polymethylcyanines is related to the degree of conjugation of the methyl chains. As displayed in Table 1.1, the absorption and emission wavelengths of trimethylcyanine are mostly in the visible light range (500–600 nm) and the absorption and emission wavelengths of pentamethine cyanines are shifted to the near-infrared region (>650 nm), while the absorption wavelength of heptamethine cyanines is generally greater than 780 nm and the emission wavelength is greater than 800 nm. Therefore, heptamethine cyanines and their derivatives are widely used in NIR-I bioimaging.

One typical dye is indocyanine green (ICG), which was approved by the US Food and Drug Administration (US FDA) in 1959, and its excitation and emission wavelengths are about 785 nm and 810 nm, respectively (Scheme 1.3).⁷  Due to its NIR-I absorption and emission, the fluorescence signal of ICG can penetrate deeper into living tissue, which is particularly useful for in vivo imaging. Until now, ICG has been widely used for fluorescence-guided surgery (FGS), including ICG angiography,⁸  non-invasive imaging of lymphatic system,⁹  laparoscopic cholecystectomy,¹⁰  etc.¹¹ 

Since the success of ICG in clinical applications, a variety of different heptamethine cyanine dyes have been developed. With modification at the R1 position of heptamethine cyanine, the water solubility, light stability and singlet oxygen production efficiency of Cy7 are greatly improved, which is particularly important in the biological field (Scheme 1.4). The introduction of various functional groups as R2 substituents facilitates tumor targeting, enabling the design of targeted fluorescent probes for NIR-I bioimaging. Modification at the X and Y sites mainly affects the photostability and fluorescence quantum yield of the cyanine dyes. Therefore, the design and preparation of cyanine dye-based fluorescent probes for NIR-I imaging is prevalent and has been widely explored.³ 

For instance, Patonay’s group increased the water solubility of dyes by introducing hydrophilic groups, such as –SO₃H, –NH₂ and –NCS (compounds 4a, 4c and 4f in Scheme 1.5).¹²  And Ryu’s group introduced pyridine salt ions into the alkyl chain to prepare IR-Pyr¹³  (Scheme 1.5), which not only increased the water solubility of the dye, but also greatly enhanced the dark stability and photostability of the dye. So, IR-Pyr is better for use for biological imaging.

Different target fluorescent probes were developed by conjugating the target groups to the heptamethyocyanine dyes for NIR-I bioimaging.¹⁴  In 2017, Krämer’s group successfully prepared a series of tumor-targeting fluorescent probes by coupling the folic acid group to heptamethyocyanine dyes (compounds 20, 21, 22 and 23 in Scheme 1.6). These probes exhibited excellent performance in NIR-I fluorescence imaging of folate receptors that were overexpressed on cancer cells¹⁵  due to the high selective interaction between the folic acid-conjugated fluorescent probes and the folate receptors on the surface of the tumor cells.

Hemicyanine is another important fluorophore that has been used recently for the development of fluorescent probes for NIR-I fluorescence imaging.^16,17  One of the milestones for NIR-I hemicyanines was the development of hemicyanine dyes (HD dyes) from Lin’s group¹⁸  (Scheme 1.7). The serendipitous discovery of NIR-I HD dyes arose from the formation of a chloro-substituted “half” cyanine enol from ICG-Cl via a retro-Knoevenagel reaction in the presence of base (Scheme 1.7a). The tailorable structures of these hemicyanines enabled the design of versatile activatable fluorescent probes for NIR-I fluorescence imaging, and so there has been great demand for the development of synthetic methods for such HD dyes. In 2015, Richard provided a new synthetic approach for preparation of NIR-I HD dyes. As shown in Scheme 1.7b, an aldehyde was first obtained as a key intermediate via a ‘‘one-pot’’ reaction of two commercially available raw materials, which could be used for a condensation reaction with Fischer’s base to open the door for the high-throughput preparation of NIR-I HD fluorophores.¹⁹ 

In 2020, Yin’s group used a tetrahydropyridine ring to replace the cyclohexene ring at the center of IR-780, forming a cyanine dye, Cy-NH, with NIR-I emission²⁰  (Figure 1.1a). Cy-NH contains two functional reaction sites (e.g. Cl and NH sites on the tetrahydropyridine ring), which are easily modified for the synthesis of Cy-NMe and Cy-MP (Figure 1.1a). Cy-NH, Cy-NMe and Cy-MP revealed similar absorption bands and their maximum absorptions were at 756 nm, 764 nm and 766 nm, respectively. Cy-NH showed a stronger near-infrared emission at 790 nm.

The confocal experiment showed a bright red fluorescence signal after the cells were treated with Cy-NH, Cy-NMe and Cy-MP, respectively (Figure 1.1b). The cells are completely stained red, emphasizing the excellent cellular staining ability. Additionally, the NIR-I fluorescence signal in zebrafish was also observed after treatment with Cy-NH, Cy-NMe and Cy-MP, respectively (Figure 1.1c). These results suggested that Cy-NH, Cy-NMe and Cy-MP could be used for NIR-I fluorescence bioimaging in living cells and in zebrafish.

In 2022, Chen's group constructed a new cyanine dye-based fluorescent probe, Cy-Mp (Figure 1.2a).²¹  Cy-Mp demonstrates high hepatocyte-targeting ability and responds effectively to polarity changes, making it valuable for diagnosing non-alcoholic fatty liver (NAFL). Compared to the control group, the signals in both channels were stronger in the NAFL model mice (Figure 1.2b), which demonstrated that the probe was capable of monitoring polarity in NAFL model mice through ratiometric imaging. The research not only introduced an effective probe for the early detection of NAFL disease, but also unveiled an innovative approach for creating polarity-sensitive probes.

In 2021, Zhu’s group developed a novel probe, Hcy-Br (Figure 1.3a), for monitoring human CYP3A5.²²  The Hcy-Br probe was constructed by using hemicyanine as the fluorophore and bromoethoxy as the recognition group. In the presence of CYP3A5, the fluorescence emission intensity at 670 nm gradually decreased, accompanied by great enhancement of the fluorescence emission at 713 nm. This phenomenon could be ascribed to the fact that CYP3A5 catalyzes the decomposition of the bromoethyl ether units, triggering a 1,6-rearrangement-elimination reaction, and resulting in the formation of a cyano platform with long-wavelength emission (Figure 1.3a). Live-cell imaging suggested that Hcy-Br could be used for tracking the endogenous CYP3A5 enzyme in both HepG2 and A549 cells (Figure 1.3b). In addition, fluorescence imaging of Hcy-Br in tumor-bearing mice was also conducted. At 10 minutes after the injection of Hcy-Br into the mouse tumor area, near-infrared fluorescence signals became visible throughout the tumor region (Figure 1.3c), indicating that Hcy-Br can be used for NIR-I fluorescence imaging of CYP3A5 in tumor-bearing nude mice. These findings suggest that Hcy-Br may be a promising candidate for studying the role of CYP3A5 in evaluating drug metabolism and potential drug interactions.

In 2021, based on hemicyanine dyes, Su’s group constructed a leucine aminopeptidase (LAP)-activated NIR-I fluorogenic probe, hCy-CA-LAP (Figure 1.4a) with high hepatocyte-targeting ability, for the accurate and sensitive imaging of LAP in drug-induced liver injuries (DILI).²³  The probe greatly improves the targeting ability of hepatocytes by introducing a cholic acid group. At the same time, in the presence of LAP, hCy-CA-LAP was selectively cleaved to generate a strong NIR-I fluorescence emission, showing high sensitivity and low detection limit for LAP (Figure 1.4a). In vivo imaging in a mouse model of acute liver injury induced by acetaminophen (APAP) showed that hCy-CA-LAP rapidly accumulates in the liver and responds to changes in LAP activity, indicating its potential for early-stage DILI diagnosis (Figure 1.4c). Therefore, Su’s work may provide an effective strategy for facilitating DILI diagnosis, drug evaluation and further research on the clinical and biological characteristics of biomarkers.

Xanthene dyes are another typical type of dye, which can be classified into three groups according to their main chemical structures, namely fluoresceins, rhodamines and rhodols²⁴  (Scheme 1.8a). By extending the π-conjugated system of the xanthene core, xanthene dyes can achieve NIR-I fluorescence (or absorption).²⁴  Another design strategy to create NIR-I xanthene dyes is replacing the moiety at the 10 position (“X” position in Scheme 1.8) with silicon (Si), borinate, phosphinate, etc.²⁵ 

Tsubaki and co-workers developed a new type of NIR-I fluorescein by extending the π-conjugated system, for instance, the dye SNAFL (Scheme 1.9).²⁶  Additionally, they prepared a new fluorescein-based NIR-I fluorescent probe for fluorescence imaging of HEK 293 cells. Similarly, Shi and co-workers extended the π-conjugation of the xanthene core through the Suzuki coupling reaction, to provide the fluorescent dyes VIX-1 and VIX-2 (Scheme 1.9) that exhibited NIR-I emission.²⁷  Xian and co-workers prepared a new rhodol (WR9 in Scheme 1.9) that displayed NIR-I fluorescence emission. Then, two new NIR-I fluorescent probes based on similar structures to WR9 were prepared for sensing hydrogen sulfide (H₂S) in live HeLa cells.²⁸  Subsequently, a variety of NIR-I rhodols were developed for bioimaging.²⁹  Scott and co-workers combined the electron-deficient xanthene core with electron-rich pyrrole and indole to prepare the NIR-I xanthene-based dyes 3B and 3D (Scheme 1.9). This design strategy leads to a decrease in the π–π* energy gap due to the extended π-conjugated system. Thus, a red-shift in the absorption and emission maxima was acheived.³⁰ 

Another important design strategy for xanthene-based NIR-I fluorescent dyes is replacing the xanthene oxygen (10-position) with different elements. One of the typical structures is the silicon-replacing rhodamine dyes (Si-rhodamine). The pioneering work is the replacement of the 10-position oxygen of rhodamine with silicon to produce red-shifted rhodamine dyes.²⁵  Nagano and co-workers developed a series of NIR-I Si-rhodamine dyes, SiR700, SiP720 and SiR720 (Figure 1.5a).³¹  Then, they prepared amine-reactive succinimidyl ester-bearing SiR700, 2-Me-4-COOSu SiR700 (Figure 1.5b), which was further coupled with anti-tenascin-C antibody to provide a targeting fluorescent probe (SiR700-RCB1). The xenograft tumor exhibited a strong fluorescence signal after injection of SiR700 at 24 h and the fluorescence signal remained for more than 10 days (Figure 1.5c). Additionally, fluorescence images of frozen sections of tumors suggested the improved photostability of SiR700-RCB1 compared to Cy5.5-RCB1, demonstrating the enhanced suitability of SiR700 over Cy5.5 for longer term observation under high-power laser irradiation (Figure 1.5c). Finally, SiR700 also demonstrated its potential use in multicolor imaging of tumors with IRDye800 (Figure 1.5d).

In 2015, Wang and co-workers developed NIR-I phosphorus-substituted rhodamines (PR, Me-PR and tMe-PR in Figure 1.6a) by replacing the bridging oxygen atom with phosphine oxide groups.³²  These exhibited great tolerance to photobleaching and tMe-PR worked well as a NIR-I fluorescent probe for imaging of HepG2 cells (Figure 1.6c). Stains and co-workers prepared similar phosphinate-based NIR-I rhodamines (NR₇₀₀, NR₆₉₈, NR₇₄₄ in Figure 1.6b). The phosphinate ethyl ester not only allows red-shift of these rhodamine dyes, but also makes the NR₇₀₀ cell-permeable for live-cell imaging.³³ 

In 2016, Guo and co-workers replaced the rhodamine 10-position O atom with a sulfone group to provide a series of sulfone-rhodamines (SO₂R1, SO₂R2, SO₂R3, SO₂R4 in Figure 1.7a).³⁴  These dyes exhibited absorption and emission maxima up to 700–710 nm and 728–752 nm, respectively. This photophysical property can be attributed to the fact that the sulfone group serves as a bridge to not only inhibit the vibration of the part of malachite green in sulfone-rhodamines, but also to act as the electron acceptor to red-shift the maximum emission of the dye. Additionally, the strong electron-withdrawing sulfone group decreases the π-electron density, thus enabling the anti-photobleaching of these sulfone-rhodamines, which is particularly important in time-lapse, long-term bioimaging applications. Interestingly, SO₂R4 and SO₂R5, with disubstituted meso-phenyl groups, could be used for specific labeling of lysosomes, and are comparable to commercially available LysoTrackers, including LysoTracker Red DND-99 and LysoTracker Green DND-26 (Figure 1.7b).

Detty and co-workers replaced the rhodamine 10-position O atom by the S, Se and Te atoms to prepare a series of new NIR-I fluorescent dyes (4-S, 5-S, 4-Se, 5-Se, 29 and 30 in Figure 1.8a).^35,36  They initially prepared the Te analogues 27 and 28 (Figure 1.8a), which exhibited no fluorescence in the NIR region, and which were easily oxidized to tellurorhodamine telluroxides 29 and 30, which exhibited absorption at wavelengths ≥690 nm and emission at wavelengths >720 nm. So, 27 was used for fluorescence imaging of singlet oxygen (¹O₂) in live cells. The results suggested that 27 localized in the mitochondria of Colo-26 cells (Figure 1.8b).³⁶ 

The full name of BODIPY is boron dipyrromethene difluoride, or difluoroboron dipyrromethene, with the IUPAC name of 4,4-difluoro-4-bora-3α,4α-diaza-s-indacene. The BODIPY dye core is composed of fipronil rings on both sides and a six-membered boron–nitrogen heterocycle in the middle, and has a good rigid planar hole structure with fluorine atoms on both sides of the planar structure (Scheme 1.10).^37,38 

BODIPY dyes can be synthesized in two steps: the first step is the condensation of pyrrole and its derivatives with electrophilic carbon compounds (e.g. acid anhydrides, chlorides or aldehydes) under an acid catalyst to produce dipyrromethene 1 (Scheme 1.11). Step 2, oxidation of dipyrromethene 1 to dipyrromethene 2 and complexation with BF₃·OEt₂ in the presence of an alkali provides the BODIPY (3 in Scheme 1.11).³⁹ 

BODIPY dyes have good photophysical and chemical properties,⁴⁰  such as high molar absorption coefficient and quantum yield, narrow fluorescence half-peak width, strong absorption of light radiation, insensitivity to polarity and pH, as well as good chemical and photostability, which makes them suitable for use as fluorescent probes. In principle, the absorption and emission wavelengths of the BODIPY core are not in the near-infrared region, with absorption wavelengths being around 505 nm and emission wavelengths around 516 nm. In order to prepare NIR-I BODIPY, two main strategies have been developed:^41,42  (1) functionalization of the core BODIPY structure at the α-, β- and meso-positions to increase the π-conjugated system, which results in a red-shift of the absorption and emission; and (2) replacing the carbon atoms at the position of the intermediates in the basic structure of the dye with nitrogen atoms, to form Aza-BODIPY.^43,44 

For example, Liao et al. reported the synthesis of BDP 4 in 2016,⁴⁵  in which a carbazole ring was added to the β-position of BODIPY. The additional conjugation produces a pronounced red-shift in the maximum absorption and emission values (Scheme 1.12a and Table 1.2).

In 2015, Zhang et al. reported a new BODIPY molecule substituted with tetra-styryl carbazole⁴⁶  for extended π-conjugation (TAC in Scheme 1.12c). The corresponding emission maximum showed a pronounced red-shift and a strong absorption maximum around 728 nm (Table 1.2).

In 2014, Gupta et al. reported the synthesis and photophysical properties of an alternative meso-substituted BODIPY 2 (Scheme 1.12b).⁴⁷  The conjugation system of the molecule was amplified to increase the maximum red-shift, with an emission peak near 631 nm (Table 1.2).

In 2002, O’Shea et al. firstly synthesized an Aza-BODIPY by replacing the carbon atom on the BODIPY ring with a nitrogen atom (Scheme 1.12d).⁴³  Compared to conventional BODIPY analogues, Aza-BODIPY not only maintains good optical and thermal stability, but also has remarkable red-shift absorption and maximum emission values which could be observed from the maximum UV absorption and fluorescence emission of phenyl-substituted Aza-BODIPY (Scheme 1.12e and Table 1.2).

Fluorescent probes based on BODIPY dyes have great advantages in NIR-I fluorescence imaging due to their high photostability and bright fluorescence signal.^48–53  Here, we select some representative fluorescent probes for NIR-I bioimaging. In 2021, Xiong et al. reported a fluorescent probe for the simultaneous detection of Fe²⁺ and H⁺ based on the N-oxide strategy and photoinduced electron transfer (PeT) mechanism (Figure 1.9a).⁵⁴  BODIPY-Fe exhibited near-infrared fluorescence at 671 nm and a fast response to Fe²⁺. Initially, BODIPY-Fe exhibited strong fluorescence due to the inhibition of the PeT process. After the N-oxide partially reacts with Fe²⁺, the fluorescence is quenched due to the PeT effect from the nitrogen of the tertiary amine group. Finally, the tertiary amine group was protonated at acidic pH, and BODIPY-H recovered the fluorescence, since the PeT process was again inhibited (Figure 1.9a). In addition, BODIPY-Fe was used for sensitive and selective detection of Fe²⁺ and H⁺ in live mice (Figure 1.9b and c).

In 2021, Banala et al. designed and synthesized a new conjugated BODIPY probe (ROS-BODIPY) (Figure 1.10a) for the reversible detection of reactive oxygen species (ROS),⁵⁵  which can be restored to their initial state using bioreducing agents after reacting with physiological levels of ROS, but persist in higher (pathological) concentrations of ROS, thus enabling reversible detection. One of the probes, 3d (Figure 1.10a), was used for reversible fluorescence imaging of ROS in live mice (Figure 1.10b). Additionally, 3d was also used for multispectral optoacoustic tomography (MSOT) imaging of ROS due to an absorption value of λ_max > 680 nm (Figure 1.10c).

In 2023, Jiang et al. designed a fluorescent probe based on Aza-BODIPY,⁵⁶  which can be used for NIR-I imaging of CN⁻. The introduction of an electron-withdrawing group (–CF₃) into Aza-BODIPY (CF₃-BDP) resulted in an electron-deficient nature for the C═N double bond (Figure 1.11a), thus enhancing the nucleophilic addition of CN⁻ to CF₃-BDP. By destroying the conjugated structure through the addition reaction, the fluorescence of CF₃-BDP is greatly decreased, which enables highly sensitive detection of CN⁻. CF₃-BDP can rapidly enter cells and bind to cyanide, and has the ability to show NIR-I fluorescence imaging of CN⁻ in live cells (Figure 1.11b).

Typically, organic molecules that emit NIR-I light tend to have a large conjugated structure, resulting in a tendency for low brightness, photobleaching and aggregation-caused quenching (ACQ). Aggregate-induced emission luminogens (AIEgens) in the NIR-I are distinguished by their high brightness and photostability,⁵⁷  which facilitates their application in NIR-I fluorescence bioimaging.^58–60  A variety of NIR-I AIEgens have been developed for bioimaging. Herein, we present typical tetraphenylethylene (TPE), triphenylamine (TPA) and quinoline malononitrile (QM)-based fluorescent probes for NIR-I imaging.

TPE contains four phenyl groups that can rotate freely, and the traditional TPE core design exhibits a notable Stoke’s shift and strong blue fluorescence.^61–63  With its ease of modification and impressive luminescent efficiency in the aggregated state, this unit stands out as a key player for AIE systems. Herein, we discuss the design of some typical TPE-based AIEgens with NIR-I emission properties (Table 1.3).

In 2017, Tang et al. developed a durable TPE-based NIR-I fluorescent dye (t-BPITBT-TPE in Scheme 1.13), which exhibits powerful luminescence in the aggregated form, and its use in bioimaging in live cells and zebrafish was also investigated.⁶⁴  In 2018, Tang et al. introduced TPE into the periphery of electron-deficient spiro-bisindane-2,1-biphenyl (BI), synthesizing another NIR-I AIEgen (2TPE-2T-BI in Scheme 1.13).⁶³  The introduction of the thiophene group enlarges the conjugated system of 2TPE-2T-BI, thus realizing NIR-I emission (Table 1.3). 2TPE-2T-BI was further developed into AIE dots that can be used for in vivo metabolic labeling. In the same year, Wong et al. reported a new NIR-I platinum complex (TPE/TPY-Pt-PA/PEG in Scheme 1.13) with AIE characteristics, by introducing TPE into a terpyridyl Pt(ii) complex.⁶⁵  The use of TPE/TPY-Pt-PA/PEG in creating nanoparticles resulted in excellent biocompatibility and a strong affinity for lysosomes in HeLa cells. Thus, TPE/TPY-Pt-PA/PEG could be used for NIR-I fluorescence imaging of lysosomes in live cells.

In 2019, Zhang et al. synthesized a new NIR-I emitting AIEgen TPE-Ph-DCM (Scheme 1.13).⁶⁶  TPE-Ph-DCM was constructed by incorporating the TPE group into the dicyanomethylene-4H-pyran (DCM), which is a typical donor–π–acceptor (D–π–A) scaffold. The dye was further elaborated to afford NIR-I nanoparticles. Due to their improved fluorescence brightness, the nanoparticles performed exceptionally well for in vivo imaging, and were particularly useful for tumor targeting due to the enhanced permeability and retention (EPR) effect. Similarly, through manipulation of the D–π–A structure, in 2022 He et al. successfully synthesized a series of photosensitizers (PSs) with AIE and NIR-I properties.⁶⁷  The top-performing AIE-PS (MTPE-DT-Py in Scheme 1.13) exhibits a strong emission at 740 nm (Table 1.3). This study presented a practical approach for creating highly effective AIE-PSs and provided superior options for fluorescence imaging-guided photodynamic therapy.

In recent years, the fields of fluorescence sensing^68,69  and biomedicine⁷⁰  have seen significant growth in the use of TPA and its derivatives, which have a high level of electron content. Through linkage of diverse electron-withdrawing groups, the emission wavelength of TPA derivatives can be altered from the ultraviolet-visible regions to the NIR-I window. This characteristic is highly beneficial for creating structures that can effectively absorb long-wavelength light, penetrate deep into tissues and enhance biological imaging.

In 2016, Xie and colleagues discovered a TPA-based NIR-I fluorescent probe (5 in Scheme 1.14) that is sensitive to cysteine (Cys) and homocysteine (Hcy). This probe was constructed by incorporating a triphenylamine analogue as its electron donor and a dicyanovinyl component as its electron acceptor and reactive site.⁷¹  Upon interaction of Cys or Hcy with dicyanoethylene, the intramolecular charge transfer capacity was diminished, causing the resulting products to aggregate in an aqueous solution and to emit strong NIR-I fluorescence (Table 1.4). Probe 5 exhibited outstanding characteristics, including high selectivity, rapid response and low toxicity, enabling successful imaging of intracellular biothiols with a robust signal-to-noise (S/N) ratio.

In 2018, Qian and co-workers reported a red/NIR-emitting AIEgen (TPA-T-CyP) for mitochondria-specific imaging (Scheme 1.14 and Table 1.4).⁷²  TPA-T-CyP has the ability to aggregate on mitochondria without the need for encapsulation or surface modification. The use of TPA-T-CyP allowed for the dynamic visualization of mitochondria, providing clear observation of the movement, fusion and fission of mitochondria at a super-resolution level. The use of NIR-I AIEgens for super-resolution organelle visualization shows promise in various areas of fundamental biomedical research.

In 2021, Li and Yu et al. reported another viscosity-dependent AIE probe TPA-Py (Scheme 1.14) for monitoring mitophagy.⁷³  TPA-Py was synthesized through a Knoevenagel condensation reaction between the AIE unit and a pyridine salt, resulting in a highly efficient near-infrared emission at 680 nm (Table 1.4) with strong water solubility and specific targeting for mitochondria.

In 2022, Song and Liu et al. reported a new D–A–D AIEgen (TPA-BTZ, Scheme 1.14) decorated with long and branched alkyl chains.⁷⁴  With the TPA-BTZ structure, the TPA served as the donor and the BTZ group as the acceptor. This strong D–A interaction enables the TPA-BTZ to exhibit NIR-I emission (Table 1.4). Additionally, PEG2000 was used to construct TPA-BTZ@PEG2000 nanoparticles, which were successfully used for NIR-I imaging-guided synergistic photodynamic and photothermal therapy.

Quinoline malononitrile is a new type of AIE structure unit that has significant features, such as good photostability and high brightness.⁷⁵  Generally, QM derivatives are constructed using the dicyanomethylene-4H-pyran (DCM) group, which possess exceptional versatility and simple functionalization, making them excellent candidates as dye matrices for creating fluorescent probes for bioimaging.

In 2015, Guo and Zhu et al. developed a series of QM derivatives that could be used for targeted imaging of tumors (QM-1, QM-2, QM-3, QM-4, QM-5 and QM-6 in Scheme 1.15).⁷⁶  QM-1 exhibited an emission wavelength of 612 nm (Table 1.5). QM-3 was synthesized by attaching a potent electron-donating alkoxytriphenylamine group to QM-1, demonstrating a NIR-I emission at 705 nm. Additionally, incorporation of the 3,4-ethylene-dioxythiophene group to provide QM-6 extended the emission wavelength to 719 nm (Table 1.5). With great potential for optimization of the organic quinoline and malononitrile, these QM derivatives can further form nanoprobes that are a safe and efficient option for tracking cells, as well as for tumor-targeting NIR-I fluorescence imaging.

In 2019, the same research group further designed an ultrasensitive off–on NIR-I AIE-activated probe for mapping amyloid-β (Aβ) plaques in situ (QM-FN-SO₃ in Scheme 1.15).⁷⁷  QM-FN-SO ₃ is composed of a lipophilic thiophene bridge and a hydrophilic sulfonate group, making it ideal for long-wavelength emission and enhanced blood–brain barrier (BBB) penetration. Through its strong hydrophilicity, the probe remains in a fluorescence-off state until it attaches to Aβ plaques. QM-FN-SO₃ has exceptional qualities enabling the in situ mapping of Aβ plaques, including an ultra-high S/N ratio, strong BBB penetration capabilities and high-performance, near-infrared emission properties.

A study by Dong et al. in 2021 reported a new NIR-I fluorescent probe (CS-Py-BC in Figure 1.12a) that was used for tracking mitochondrial viscosity changes during mitophagy.⁷⁸  CS-Py-BC was constructed using a cyanostilbene skeleton, a pyridinium cation and a benzyl chloride group. It exhibits an off–on fluorescence response to changes in viscosity. The probe localizes in mitochondria and enables the real-time tracking of viscosity changes during starvation or rapamycin-induced mitophagy (Figure 1.12b).

The AIE mechanism in this probe is based on a cyanostilbene framework, combined with a viscosity-sensitive element and a pyridinium cation to target mitochondria. Additionally, a benzyl chloride unit is incorporated to immobilize at the mitochondria effectively. Due to the inherent differences in the microenvironment between autophagosomes and mitochondria, such as pH value, viscosity, polarity, etc., detecting alterations in the mitochondrial microenvironment offers a practical approach to real-time monitoring of mitochondrial autophagy. Disease states within the body can be determined by monitoring the viscosity of mitochondria, which is highly important within the microenvironment. CS-Py-BC can effectively attach to mitochondria, regardless of any variations in mitochondrial membrane potential. CS-Py-BC exhibits great potential for analyzing the pathological and physiological aspects of mitophagy.

In 2022, Zhou et al. prepared a series of D–π–A systems (TPA-OS, TPA-T-OS and TPA-2T-OS) using TPA and ionic salt (OS) as the electron donor and acceptor, respectively (Figure 1.13). TPA-2T-OS was developed for enhancing multimodal imaging-guided therapy.⁷⁹ 

The probe TPA-2T-OS was enhanced with thienyl as a π-bridge in order to improve both the planarization and donor rotation. At the same time, planarization boosted the intramolecular charge transfer (ICT) effect, promoting the production of type I ROS and NIR fluorescence emission at 833 nm. As a result, this one-step regulation approach provides a hassle-free and efficient way to develop theranostic agents that enhance both imaging and treatment outcomes simultaneously.

In 2022, Ding et al. developed two probes, TCP-PF₆ and TTCP-PF₆, which possess comparable D–A structures. Integration of a thiophene ring into the TTCP-PF₆ framework facilitated the solid-state packing modes (Figure 1.14).⁸⁰ 

Integrating a thiophene segment with high electron density increases the ICT effect, resulting in a shift towards longer emission wavelengths in the NIR region, rendering the probe suitable for biological applications. The tightly packed arrangement and strong affinity for lipids caused by the thiophene component allow TTCP-PF₆ to achieve superior imaging capabilities and enhanced production of ROS by specifically targeting mitochondria, surpassing the performance of TCP-PF₆ (Figure 1.14). With the specialized ability to focus on mitochondria, TTCP-PF₆ is an excellent choice for efficient photodynamic therapy.

The widespread occurrence of stroke is a significant cause of both permanent disability and death globally. Subarachnoid hemorrhage (SAH) is a form of stroke caused by a burst blood vessel on the outer layer of the brain, leading to bleeding in the subarachnoid space.^81,82  In 2023, Situ et al. developed an ultrasensitive fluorescence probe (TTVP in Figure 1.15a) for detecting and evaluating SAH. TTVP is a water-soluble, small molecule probe that specifically interacts with blood (Figure 1.15).⁸³ 

Through the utilization of its AIE features, TTVP is able to give out a strong fluorescence signal in bleeding areas after it specifically targets albumin, providing a strong S/N ratio for precise blood detection. Even trace amounts of blood can be visualized by TTVP. Additionally, a system has been developed that uses fluorescence to evaluate the degree of the SAH bleeding in vivo (Figure 1.15b and c). This probe could be used as an efficient fluorescence imaging tool for the sensitive detection of hemorrhagic-related diseases.⁸³ 

Compared to visible light, NIR-I light with a longer wavelength is less likely to be absorbed and scattered by biological tissues, enabling NIR-I fluorescent probes to exhibit deeper penetration and more effective imaging of biological tissues, such as skin and blood.^84,85  This chapter presents the design strategies and photophysical properties of typical NIR-I fluorophores and some recently developed small organic fluorescent probes for NIR-I bioimaging. Despite the great advancements, there are still some limitations for fluorescent probes in NIR-I bioimaging. For instance, except for the limited NIR-I cyanine dyes, including ICG and IR820, etc., that have been approved by the US FDA, the biocompatibility of most NIR-I fluorophores has rarely been evaluated in large animals or humans, which prevents the further clinical application of NIR-I fluorescent probes. Another concern is the low photostability of various NIR-I fluorescent probes, since the NIR-I emissions were realized by enlarging the π-conjugated systems of the fluorophores, which are then prone to be broken by continuous and strong irradiation. In addition, the S/N ratio for in vivo NIR-I fluorescence imaging still needs to be improved, which is particularly significant in tumor imaging in the field of FGS. Thus, one direction for further exploration of fluorescent probes for NIR-I bioimaging should be towards improvement of the biocompatibility of NIR-I fluorescent probes, particularly for large animals. For practical clinical applications, the photostability during continuous irradiation of the fluorescent probes needs to be improved. Finally, fluorescent probes with high S/N ratio are urgently required, since they exhibit great potential in FGS.

ACQ

Aggregation-caused quenching

AIE

Aggregate-induced emission

BBB

Blood–brain barrier

BODIPY

Difluoroboron dipyrromethene

DILI

Drug-induced liver injury

FGS

Fluorescence-guided surgery

FDA

Food and Drug Administration

FRs

Folate receptors

ICG

Indocyanine green

LAP

Leucine aminopeptidase

NAFL

Non-alcoholic fatty liver

QM

Quinoline malononitrile

ROS

Reactive oxygen species

SAH

Subarachnoid hemorrhage

TPE

Tetraphenylethylene

TPA

Triphenylamine

S/N

Signal-to-noise

The authors gratefully acknowledge the financial support from the National Natural Science Foundation of China (Grants 21672150), Shanghai Rising-Star Program (19QA1406400), Shanghai Municipal Education Commission and Shanghai Engineering Research Center of Green Energy Chemical Engineering.