Keywords

1 Introduction

Fluorescence molecular imaging (FMI) utilizes molecular probes to label target organisms. Under certain external conditions, the molecular probe releases a fluorescent light in the visible or near-infrared spectrum using high-sensitivity detection equipment for fluorescence. The signal is collected, and the position and intensity of the fluorescent light source are displayed to obtain the physiological activity information of the organism (Fig. 1). FMI is the most intuitive imaging modality, and the fluorescent methods allow us to detect photons with familiar parallels to our eyesight, allowing spatial and temporal resolutions that are otherwise unachievable [1, 2]. Compared to the positron emission tomography (PET) imaging and magnetic resonance imaging (MRI), FMI is a noninvasive and nonionizing imaging modality with higher sensitivity and higher specificity, is safer, has a lower cost, and is easier to perform, and it offers anatomical, physiological, and even molecular information within the bodies of living subjects [3,4,5]. Therefore, FMI has been widely applied for tumor detection, drug development, image-guide surgery, and other biomedical fields.

Fig. 1
figure 1

The principle of fluorescence molecular imaging (FMI)

Full size image

From a medicinal chemistry perspective, FMI is a potent tool for probing biomolecules in their natural environment and for visualizing dynamic processes in complex biological samples, living cells, and organisms [6,7,8,9], which are well suited for highlighting molecular alterations associated with pathological disorders. Thereby, it offers means of implementing sensitive and alternative technologies for diagnostic purposes, which constitute as attractive tools for drug discovery programs from initial target validation and high-throughput screening identification campaigns to the final clinical translation phases. In this chapter, the FMI probe, imaging analysis methods, and medical application in drug discovery will be described.

2 Imaging Probe

Fluorescence probe is one of the basic elements of FMI. Fluorescence probes are usually made up of fluorophores and targeting ligands.

2.1 Fluorophores

The fluorophore can convert molecular recognition information into a fluorescence signal, which has the advantages of high sensitivity, rapid reaction time, and the ability to realize in situ detection. An ideal fluorophore has a large value of Stokes shift [difference between the absorption maximum (λmax) and the emission maximum (λem)] to minimize the reabsorption of emitted photons. According to the fluorophore with emission in different regions, it can be divided into three categories: visible light (λem < 700 nm), near-infrared I (NIR-I, 700 < λem < 1,000 nm), and near-infrared II (NIR-II, λem > 1,000 nm) fluorophores (Fig. 2) [10, 11].

Fig. 2
figure 2

Wavelengths for fluorescence molecular imaging. (a) Tissue penetration depth of light with different wavelengths. (b) Light when entering a tissue can be reflected or adsorbed by molecules within the tissue or excite fluorophores to emit light at a different wavelength. Reproduced from Ref. [11]

Full size image

2.1.1 Visible Light Imaging Fluorophores

Visible light fluorophores mainly include fluorescein isothiocyanate (FITC), cyanine (Cy), rhodamine, BODIPY, coumarin, quinoline, etc. (Fig. 3) [12,13,14,15,16]. These fluorophores are fluorescent indicators with rapid detection, good reproducibility, and low sample size, which can be used to detect the changes of cations (Na+, K+, Mg2+, Ca2+, etc.), anions (phosphate radical, etc.), free radicals (reactive oxygen species H2O2, superoxide ion), monolinear oxygen, hydroxyl radical, sugar (glucose, chitosan, etc.), nucleic acid (DNA, RNA), and enzymes (trypsin, viral protease, no synthase) in biological systems after drug therapy [17,18,19]. The visible light fluorophores (400–700 nm) are used quite often in biomedical studies, but the key issues in fluorescence imaging of visible regions include autofluorescence, quenching, photobleaching, and a low depth of tissue penetration. In comparison, fluorescence imaging in NIR offers considerable advantages. However, the use of NIR fluorophores requires a special camera, as the light is not visible to the naked eye or conventional video cameras.

Fig. 3
figure 3

Representative visible light fluorophores

Full size image

2.1.2 NIR-I Imaging Fluorophores

Fluorophores with emission in the NIR region possess less absorption and scatter from tissues much more efficiently than fluorophores based on visible light, which is favorable for in vivo imaging with a high signal-to-background ratio (SBR). This imaging modality also inherited quick feedback, high-resolution, and noninvasive properties of optical imaging and can be utilized to visualize the real-time dynamics in living organisms [20,21,22,23]. Over the past decade, fluorescence imaging in the first NIR window (NIR-I, 700–900 nm) has been widely studied in fundamental research and preclinical/clinical applications, which is partially because of the immediate availability of a wide range of fluorophores, such as NIR-760, IRDye800CW, indocyanine green (ICG), methylene blue, and their derivatives (Fig. 4) [24,25,26].

Fig. 4
figure 4

Representative NIR-I fluorophores

Full size image

2.1.3 NIR-II Imaging Fluorophores

The generation of new fluorophores and the development of fluorescent labeling technology provide specific and efficient contrast for FMI and greatly improve the detection sensitivity and specificity of in vivo imaging. Compared to NIR-I, fluorescence imaging at the second near-infrared region (NIR-II, 1,000–1,700 nm) can realize better fluorescence image quality (Fig. 5) [27]. The NIR-II biological window is broadly defined as wavelengths in range of 1,000–1,700 nm. Smaller optical sub-windows such as NIR-IIa (1,300–1,400 nm) and NIR-IIb (1,500–1,700 nm) have provided further improvements in fluorescence imaging metrics. The 1,400–1,500 nm window is typically avoided owing to the presence of an absorbance peak due to a water overtone. Significant improvements in imaging temporal and spatial resolution (~20 ms and ~25 μm) and penetration depth (up to ~3 cm), which are very difficult to achieve with NIR-I and also PET and SPECT imaging, have been fulfilled by this innovative NIR-II region on biomedical imaging, thanks to the reduced scattering, negligible tissue absorption, and minimal autofluorescence [28, 29].

Fig. 5
figure 5

Fluorescence imaging of the cerebrovasculature of mice without craniotomy in the (a) NIR-I, (b) NIR-II, and (c) NIR-IIb regions, with the corresponding SBR analysis shown in (d)–(f). Scale bars, 2 mm. Reproduced from Ref. [30]

Full size image

However, NIR-II fluorophores also suffer from poor water solubility, low photostability, low quantum yield, and the scarcity of molecules with suitable NIR-II band gaps have further limited the applications and development of NIR-II imaging techniques. So far, a series of fluorophores with emission wavelengths longer than 1,000 nm in the NIR-II region have been designed based on the donor-acceptor-donor scaffold (D-A-D). These NIR-II fluorophores are usually composed of various spacers (thiophene), electron donor (fluorene and triphenylamine), and the central electron-accepting aromatic backbone (benzobisthiadiazole, BBTD), which can greatly expand the library of small-molecule NIR-II fluorophores (Fig. 6) [31,32,33]. Fortunately, a series of organic small molecules and organic and inorganic nanomaterials with precisely controlled structures and intrinsic near-infrared emissions in the NIR-II window have been developed to enable the acquisition of high-definition NIR-II images at wavelengths well in excess of 1,000 nm. These NIR-I and NIR-II fluorophores can be used to develop diagnostics, biomedical imaging technologies, and drug discovery programs.

Fig. 6
figure 6

Representative NIR-II fluorophores

Full size image

2.2 Targeted Ligand of Fluorescent Probes

In the field of medicinal chemistry, the target is the key element for the drug design and development. FMI has the advantages of high sensitivity, convenience, reliability, and suitability for large-scale detection of drug targets. Using FMI, not only the receptors, proteins, and genes that interact with drugs can be imaged, but the drug targets can also be located and evaluated for their presence in an organism. The spatial and temporal distribution of the target is evaluated quantitatively. Verification of target expression is valuable for diagnosis, as well as the selection of treatment regimen and pre-evaluation. There are often overexpressed specific receptors on the surface of tumor cells. For example, abnormal overexpression of estrogen receptor (ER) [34,35,36] and fructose transporter (GLUT5) [37] has been found in breast cancer; prostate-specific membrane antigen (PSMA) is overexpressed in prostate cancer [38,39,40,41]. FMI can be used to determine the expression level of related hormone receptors in tumor biopsy specimens, which can help select the best treatment plan. Using these receptors as targets to develop new, potential drugs with high specificity and affinity is the main direction of development of antitumor drugs. Currently, many active targeted fluorescence probes have been developed.

2.2.1 Estrogen Receptor-Targeted Fluorescent Probes

Estrogen receptor (ER) is a ligand-regulated transcription factor that regulates many physiological and pathological processes and also plays a predominant role in breast cancer growth. Therefore, ER is regarded as an important pharmaceutical target for the treatment of breast cancer, and the development of ER-targeted fluorescence probes has emerged as an active research field for breast cancer detection, and many of these probes have been developed (compounds 1–3, Fig. 7) [34,35,36].

Fig. 7
figure 7

Representative estrogen receptor-targeted fluorescent probes 1–3

Full size image

2.2.2 GLUT5 Transporter-Targeted Fluorescent Probes

Facilitated hexose transporters (Gluts) are a group of transmembrane proteins responsible for transporting sugars such as glucose or galactose across the cell membrane. Tumor cells usually overexpress the Glut transporter to meet their high levels of energy consumption needs. For example, GLUT5 is overexpressed in breast cancer cells but not in normal breast cells. Therefore, the selection of a high GLUT5-binding affinity compound as the targeting ligand is an effective strategy for the development of imaging probes for breast cancer detection (compounds 4–5, Fig. 8) [37].

Fig. 8
figure 8

Representative GLUT5 ligand 4 and targeted fluorescent probe 5

Full size image

2.2.3 Prostate-Specific Membrane Antigen (PSMA)-Targeted Fluorescence Probes

Prostate cancer is the most commonly diagnosed malignancy in men, and the integral membrane protein PSMA is becoming increasingly recognized as a viable target for prostate cancer diagnosis and treatment. Therefore, PSMA-specific antibodies, peptides, peptide derivatives, or other small molecules have been developed as targeting ligands for the development of imaging probes for prostate cancer detection (compounds 6–9, Fig. 9) [38,39,40,41].

Fig. 9
figure 9

Representative PSMA-targeted fluorescent probes 6–9

Full size image

2.2.4 Folate Receptor-Targeted Fluorescent Probes

Folates are essential for the maintenance of all cells and tissue regeneration. Folates have a high affinity for their cell surface folate receptor (FR), which is primarily expressed on healthy cells where it does not readily encounter folate from the bloodstream. When the malignant transformation occurs, high levels of FR are expressed in a number of malignancies, including ovarian and endometrial cancers and myeloid leukemias. Thus, the high affinity of FR offers a potential means for tumor targeting, which has already become the main design strategy for the FR-targeted tumor imaging probes (compounds 10–11, Fig. 10) [42, 43].

Fig. 10
figure 10

Representative folate receptor-targeted fluorescent probes 10–11

Full size image

2.2.5 Cyclooxygenase-2 (COX-2)-Targeted Fluorescent Probes

COX-2 is a crucial biological mediator in the etiology of cancer. This enzyme is absent or present at low levels in normal cells but shows high expression levels in inflamed tissues as well as many premalignant and malignant tumors such as colorectal adenomas and adenocarcinomas. COX-2 has been used as an ideal imaging biomarker for cancer cells. Currently, many fluorescent probes have been engineered to target COX-2 for tumor detection (compounds 12–15, Fig. 11) [44, 45].

Fig. 11
figure 11

Representative COX-2-targeted fluorescent probes 12–15

Full size image

2.2.6 Other Targeted Fluorescent Probes

In addition to the targeted fluorescence probes mentioned above, various other targeted fluorescence probes have also been reported, such as carbonic anhydrases IX (CAIX) probes. CAIX has been associated with tumor progression and invasion, which is usually expressed in normal tissues at certain levels, but overexpressed in many solid tumors, such as colorectal tumors. Therefore, CAIX can be used as a potential tumor target for the development of imaging probes (compounds 16–17, Fig. 12) [46, 47].

Fig. 12
figure 12

Representative CAIX-targeted fluorescent probes 16–17

Full size image

In general, the targeted fluorescent probes were accomplished by conjugating fluorescence dye to the targeted ligand, which may have a great application prospect in the clinic for tumor diagnosis.

3 Imaging Analysis

Fluorescence molecular tomography (FMT) is a three-dimensional imaging method based on FMI, which is based on the distribution of fluorescence in biological tissues [48,49,50,51,52,53]. It was developed from two-dimensional (2D) qualitative imaging to three-dimensional (3D) quantitative research and further expanded the integration of stimulated fluorescence in the diagnosis and treatment of cancer and preclinical and clinical applications. The advent of FMT led to the three-dimensional reconstruction of FMI agent accumulation in living animals based on light recordings collected at the tissue boundary. FMT has been used to visualize and quantitate a variety of cellular and molecular events and, as opposed to planar fluorescence imaging, yields quantitative information and allows imaging at greater depth, up to several centimeters. It has developed rapidly in recent years and has become a research frontier and research hotspot for FMI technology [54,55,56].

When imaging spatial data needed for FMT reconstruction are obtained, then the reconstruction of the structural data and optical data based on the biological model can be carried out [57]. In general, the image reconstruction process includes two steps: solving the forward problem and solving the inverse problem. The solution of the forward problem is used to calculate the photon propagation model of the fluorescence transmitted in the imaging space to obtain the linear relationship between the fluorescence measurement data on the surface of the tissue and the fluorescence distribution inside the bio-tissue. After the linear relationship is obtained by solving the photon transfer model, various methods are used to solve the linear model, and the distribution of fluorescence inside the imaging space is obtained, which is called the inverse problem [52].

3.1 Photon Propagation Model

The process of transmitting fluorescence from a light source to a biological body through a specific biological tissue is extremely complicated and includes various physical processes such as scattering of light, inter-tissue reflection, refraction, diffusion, and absorption. For FMT imaging, imaging is usually performed in the visible and near-infrared optical bands, and the scattering and absorption effects of this band of light inside the biological tissues are the main forms of our study. Therefore, the FMT photon propagation model can be simplified to a photon stochastic propagation model that contains only the scattering and absorption effects without considering the reflection and refraction of different tissues. Current mainstream mathematical theory to solve these problems is mainly based on Boltzmann’s radiative transfer equation (RTE) [58], which is equivalent to photon propagation as transport of photon flux in a medium from particle fluctuation to energy transport and to study transport of light energy in biological tissue problems.

In three-dimensional biological tissue, the RTE solution is transformed into a six-dimensional space-time problem. There are few methods in solving mathematical and computer problems, and they are usually not able to directly close the analytical solution. Moreover, because of its unknowns, it can be solved precisely only in rare cases. Usually it cannot get a closed analytical solution. At the same time, it is extremely difficult to solve RTE directly, while the exact solution will only exist in rare cases. Therefore, it is common practice to replace itself with a simplified approximation of the radiation transfer equation [59], such as diffusion equation (DE), which is a widely used RTE-based simplified model [60,61,62,63,64,65]. It utilizes the first-order spherical harmonic function to expand important function items in the RTE equation and performs the approximate processing, which significantly reduces the computational complexity and is suitable for the visible and near-infrared bands of FMT imaging. In recent years, researchers have proposed such high-order approximations as RTE [66,67,68,69,70,71]. Compared with diffusion equations, higher-order approximation models can significantly improve FMT accuracy. The SN model [72], PN model [73], and SPN model [69] are three commonly used RTE high-order approximation models and usually give more accurate RTE solutions to the more diffusive equations. By these approximation methods, the traditional RTE equation can be transformed into several coupled higher-order partial differential equations for easy calculation and solution.

3.2 Forward Problem-Solving

The linear relationship between the measured data on the surface of the imaging area and the internal fluorescence distribution in the imaging area based on the photon propagation model is the core of the FMT forward problem. In recent years, researchers have proposed various mathematical solution methods including the analytic method, statistical method, and numerical analysis method to solve the forward problem of FMT [52]. Numerical analysis method is the main solving method currently used in optical molecular imaging reconstruction. Its computational efficiency is high and its applicability is wide. Numerical analysis methods include the finite difference method (FDM) [70], boundary element method (BEM), finite element method (FEM) [74], and meshless method (MM) [75]. FDM uses equidistant grid points and regular grids to solve the forward problem, which is more efficient than irregular grids. However, FDM has difficulty in dealing with geometrically complex imaging spaces and boundary conditions. In contrast, FEM is the mainstream solution to FMT forward problems in recent years. The main advantage of FEM is its effectiveness in dealing with complex geometric problems. In addition, the system matrices obtained by FEM are usually sparse and positive definite, which leads to a more stable solution and high computational efficiency, which is also beneficial to FMT reconstruction [76, 77]. However, the main drawback of FEM is that it is difficult to generate the FEM grid. In contrast, BEM only needs to discretize the imaging surface and the boundaries of the heterogeneous tissue within the space without the need to mesh the entire imaging space. Therefore, compared with FEM, BEM can effectively reduce the computational dimension and complexity to improve computational efficiency. However, fast and stable 3D mesh generation for complex geometry problems remains a challenging issue. In order to overcome the problem of 3D mesh generation, An et al. proposed a meshless method and applied it to solve the forward problem of FMT [78]. The method only needs to obtain nodes that are relatively independent from each other to discretize the imaging space and does not require a cumbersome gridding process.

3.3 Inverse Problem-Solving

In FMT preclinical and clinical trials, the fluorescence signal is usually only measured from the imaging surface. However, the dimension of the measurement data on the imaging space surface is usually much less than the number of internal nodes in the imaging space. Therefore, the inverse problem of FMT is ill-conditioned [55]. Moreover, because of the high scattering properties of photons in the imaging space, the inverse problem is also ill-posed, and it is difficult to find the exact solution [79,80,81]. At the same time, the noise generated during the experiment also affects the accuracy of the FMT reconstruction [82].

The ill-posedness of the FMT inverse problem is mainly due to the lack of information and uncertainty due to the high scattering of photons. In order to overcome the ill-posedness of reverse problems, researchers started from the light source prior information and combined it with a variety of a priori information related to the light source and photon transmission to reduce the uncertainty of the information so as to improve the accuracy of inverse problem-solving [83,84,85,86,87,88,89]. Researchers usually combined the prior information of the structure into the FMT reconstruction and proposed a nonhomogeneous imaging space model and a priori reconstruction method, which greatly improved the reconstruction accuracy. The structure of imaging space prior information can usually be obtained by high-resolution structural imaging modalities such as CT and MRI [90,91,92,93,94]. The optical parameters of various organs and tissues can be obtained by other imaging techniques such as DOT. The imaging technique that combines imaging modalities to increase imaging prior information is also known as multimodality imaging and is the focus of current medical imaging research [95].

Although researchers have put forward prior knowledge such as feasible regions, structural prior information to augment the information needed for reconstruction, the morbidity of the FMT reconstruction equation remains unresolved. Moreover, the actual FMT acquisition data usually contain a certain amount of noise, which has a great impact on the reconstruction of the pathological equation. A small signal disturbance may lead to a large reconstruction error. Therefore, researchers apply regularization techniques to FMT reconstruction to constrain the reconstruction process and reduce morbidity [96,97,98,99,100,101,103,104]. The classical regularization term is Lp-norm regularization. The Lp-norm regularization (p = 0.5–2) usually obtains a smoother reconstructed result of a large reconstructed area and has a good reconstruction effect for a large light source volume in an imaging space. Another available regularization method is total variation (TV) [105]. The main idea of TV norm regularization is to constrain the variation terms of the distribution of the fluorescent light sources while preserving the boundaries of the light source zones (Fig. 13).

Fig. 13
figure 13

(a) Coronal and (b) transverse sections of the CT image of the mouse-shaped phantom showing the two embedded fluorescent line sources. (c) Coronal and (d) transverse overlay of CT and FMT images. (e) Coronal and (f) transverse sections of the FMT imaging showing the two fluorescent line sources reconstructed using both L1 and TV penalties with regularization parameters of 10 and 1, respectively. The figure is reproduced from Ref. [105]

Full size image

4 Medical Application

Precision medicine has promoted the development of treatment modalities that are developed to specifically kill tumor cells but not normal cells. The traditional methods of drug discovery have many disadvantages, such as a long research period and the antitumor drug treatment effects in situ cannot be monitored in real time. Therefore, the use of new technologies such as FMI for drug discovery is urgently needed. It seems likely that FMI will meet this challenge for the evaluation of therapeutic effects. The results were more accurate and reliable than the traditional measurement of tumor size. In this chapter, the application of FMI will be described in drug discovery, including identification of therapeutic targets, candidate drug screening, pharmacokinetics of drugs, and prodrug development.

4.1 Identification of Therapeutic Targets

Specific therapeutic target is the key for therapy, but traditional drug chemistry methods find targets at slow speed. FMI can improve the process of target identification and identify suitable treatment regimens, hence improving patient treatment outcomes. For example, breast cancer is the most common cancer among women with different subtypes. Nearly 75% of patients demonstrate abnormally high expression of ERα. Therefore, ERα is regarded as an important pharmaceutical target for the treatment of breast cancer, and many ERα ligands have been developed into hormone agents. However, hormone therapy is ineffective for ERα(−) and triple-negative breast cancers (TNBCs). The ERα fluorescent probe P1 can be used to identify a suitable therapeutic regimen for breast cancer. As shown in Fig. 14, the fluorescence signals can be observed in the cell nucleus of the ERα-positive MCF-7 breast cancer cell, but not in MDA-MB-231 TNBC cells. Therefore, FMI is able to identify the target expression and determine treatment [45].

Fig. 14
figure 14

Fluorescence imaging of intracellular targets in triple-negative breast cancer cells MDA-MB-231 and ER(+) MCF-7 cells. Images of cells treated with compound P1. Reproduced from Ref. [44]

Full size image

4.2 Candidate Drug Screening

To evaluate the therapeutic effects of antitumor drugs in vivo, a traditional medical imaging method is used to measure the tumor volume at the late stage of tumor growth with treatment for a period of time. However, this method can only tell the changes in tumors when they show anatomical changes; in addition, it is difficult to evaluate the effect of in situ tumor therapy by the traditional method. FMI can completely overcome the shortcomings of the traditional method and can not only monitor the changes of the tumor biomarker but also evaluate the therapeutic effect in an early and dynamic manner. For example, histone deacetylases (HDACs) are overexpressed in TNBC. The FMI of the LBH589-Cy5.5 probe has been successfully applied not only for measuring the expression and functions of HDACs in tumors but also in evaluating the therapeutic response of HDAC inhibitor SAHA treatment, as evidenced by the significantly reduced HDAC signals in SAHA-treated breast tumors (Fig. 15) [106].

Fig. 15
figure 15

Detection of tumor and evaluation of antitumor activity of drugs by FMI. Reproduced from Ref. [106]

Full size image

4.3 Tracking the Drug Biodistribution and Metabolism

When the tumor cells were treated with drugs, it was difficult to observe the drug interaction with the corresponding targets by traditional pharmacochemical methods. The majority of drugs tested clinically exhibit off-target effects, which is easy to produce side effects. FMI is able to directly visualize the binding of the drug to the target, which can effectively improve the success rate of drug development. For example, 2-((3-(3-fluoro-4-hydroxyphenyl)-7-hydroxynaphthalen-1-yl) methylene) malononitrile (FPNM) can potently inhibit the growth of MDA-MB-231 tumors, and the relative binding affinity (RBA) value shows FPNM is an estrogen receptor β (ERβ) ligand. In order to confirm the interaction between FPNM and ERβ, FMI of FPNM is performed in MCF-7 cells. As shown in Fig. 16, ERβ is a nuclear hormone receptor, and the fluorescence derived from the complex between FPNM and ERβ was mainly detected in the cell nucleus. The data suggested that FPNM showed specifically selective affinity toward ERβ in the living MCF-7 cells. These results indicated that FPNM possesses the ability to selectively bind to the ERβ in living cells [107].

Fig. 16
figure 16

Co-localization of MCF-7 cells with FPNM (10.0 μM) and ERβ antibody. (a) Bright field of cells. (b) Nuclei were stained with DAPI. Cells were stained with FPNM (c) and ERβ antibody (d). Co-localization of FPNM and ERβ antibody. Scale bar = 10 μm. Reproduced from Ref. [107]

Full size image

4.4 Determination of Pharmacokinetics of Drugs

Targeting specific, small molecules as modulators of drug delivery may play a significant role in the development of therapeutics. Small molecules can alter the pharmacokinetics of therapeutic macromolecules leading to more efficient drug delivery with less systemic toxicity. McCann et al. used FMI to observe the biodistribution and excretion patterns associated with molecular probes composed of human serum albumin (HSA) conjugated to high and low numbers of various monosaccharides: Glc-α, Gal-β, Man-α, Fuc-α, and Fuc-β. First, the conjugation of IRDye800CW to HSA demonstrated nonspecific distribution throughout the body, suggesting the addition of IRDye800CW minimally changed the biodistribution of HSA. However, the conjugation with either low numbers or high numbers of sugar molecules resulted in rapid and specific changes in biodistribution. The conjugation of HSA to a low number of sugar molecules leads to slower clearance of the probe from the blood circulation compared to HSA conjugated to a high number of sugar molecules (Fig. 17) [108].

Fig. 17
figure 17

Chemical structure of the linkage between sugar and albumin. (a) the images of HSA glycosylated with a low (b) and high (c) sugar number at different time points as 10 min, 6 h, 24 h, and 48 h post-injection. Reproduced from Ref. [108]

Full size image

4.5 Fluorescence Prodrug Conjugates

Fluorescence prodrug conjugates are dual functional systems that offer both therapeutic promise and potential for concurrent diagnosis, which are able to target cancer cells selectively, provide cytotoxic chemotherapeutics, and allow facile monitoring of the location and efficacy of drugs [109, 110]. Fluorescence prodrug conjugates are of particular interest since they are stable in blood plasma, which can be activated efficiently by various cellular constituents, such as thiols, reactive oxygen species (ROS), and enzymes that are overexpressed in tumors [111, 112].

Fluorescence prodrug conjugates can realize both targeted therapeutic release and targeted FMI. Such prodrug conjugates usually contain fluorophores, cleavable linkers, and targeting ligands or chemotherapeutic agents. Fluorophores are usually naphthalimide, coumarin, BODIPY, rhodol, cyanine, etc. Cleavage linkers include hydrolysis of esters, amides, and hydrazine linkers, disulfide exchange-based scission, hypoxia-induced activation, enzymatic reactions, photolysis, and thermolysis [113]. Targeting ligands include folate, biotin, galactose, and RGD (Arg-Gly-Asp) peptide sequences. Doxorubicin, camptothecin, paclitaxel, gemcitabine, and cisplatin are commonly used chemotherapeutic agents. To date, much effort has been devoted to develop systems that undergo cleavage under physiological conditions. When the cleavable linkers serve to tether a fluorophore to a prodrug in such a way that the fluorescence signal can be controlled upon cleavage, it becomes a potential system that operates as both therapeutics and diagnostics (Fig. 18) [114].

Fig. 18
figure 18

Design principle for achieving fluorescent prodrug conjugates that are able to target cancer cells selectively, provide cytotoxic chemotherapeutics, and produce readily monitored imaging signals. Reproduced from Ref. [114]

Full size image

To date, many fluorescence prodrug conjugates have been reported, including cellular thiol-activatable fluorescent prodrugs (compound 18) [114], hydrogen peroxide-activated fluorogenic prodrugs (compounds 20–21) [115, 116], acidic pH-activated fluorogenic prodrugs (compounds 22–23) [117, 118], hypoxia-activated fluorogenic prodrugs (compound 24) [119], platinum reduction-based fluorogenic prodrugs (compound 28) [120], enzymatic cleavage-based fluorogenic prodrugs (compound 29) [121], light-activated fluorogenic prodrugs (compounds 32–33) [112, 122], etc. (Fig. 19).

Fig. 19
figure 19figure 19figure 19

Representative fluorescent prodrug conjugates

Full size image

5 Future Perspectives

Currently, FMI has become a powerful and effective tool for drug development. The study of specific probes and targets depends on the development of chemistry and biology, and the FMI tracers do not interfere with the biological process by themselves. The involvement of FMI in metabolic or specific biological processes can more realistically reflect the physiological and pathological processes occurring in organisms, such as gene expression, activation of biochemical pathways, protein interaction, and tracing of cell proliferation and death. Using this noninvasive method to detect cell function can help evaluate the role of new candidate drugs under the influence of complex biological responses in animals. The development of new probes, especially NIR probes, can facilitate the clinical application of this nonradioactive imaging technology. In the last 10 years, FMI has developed rapidly for its low cost, non-ionization, and high throughput, and it plays an important role in all stages of drug development. This technique also has some shortcomings:

  1. 1.

    The detection depth is limited: because of the different wavelengths of the fluorophores and the influence of light absorption and scattering, optical molecular imaging equipment cannot detect cell activity at deeper depths in vivo.

  2. 2.

    Target selection: in order to apply FMI to monitor the efficacy of drug treatment, it is necessary to find specific targets for the diseases. Currently, specific targets of diseases representing the occurrence and development of diseases are not fully discovered.

  3. 3.

    Probe development: the ideal probe should have a high sensitivity and specificity for detection and should not cause an immune response and can be easily cleared by the body. However, the existing FMI probes do not fully meet the above conditions, and the development of new probes is costly.

  4. 4.

    Clinical trials: safety and effectiveness of the application of fluorescence probes in the clinical trials or treatments. In summary, the development of new FMI probes and their application in various stages of drug development will ultimately improve the efficiency of developing new and effective drugs, reduce the cost of research and development, and provide a wider and clinical application prospect in the field of drug development.