Application of Azobenzene Derivative Probes in Hypoxia Cell Imaging

Yunxue Wu; Hengyi Zhang; Yu Liu

doi:10.7536/PC200934

Progress in Chemistry ›› 2021, Vol. 33 ›› Issue (3) : 331-340 DOI: 10.7536/PC200934

review-article

Scan to read & Share

Application of Azobenzene Derivative Probes in Hypoxia Cell Imaging

Author information +

History +

PDF

Abstract

Tumor tissues have lower oxygen concentration compared to normal tissues, due to the inadequate oxygen supply caused by uncontrolled cell growth and proliferation, in addition to an abnormal vasculature. As a common feature of solid tumors, hypoxia can be an indicator of malignant tissues or cancer progression. Accurate hypoxia detection and imaging are essential for the diagnosis and clinical treatment of cancer patients. Fluorescence imaging has been used in cancer detection because of its high sensitivity, non-invasive and real-time characteristics. During recent years, azo groups have been widely used to construct fluorescent probes for hypoxia cell imaging, owing to the fluorescence quenching effect on fluorophores and their reductive cleavage resulting in fluorescence recovery. This review summarizes various azobenzene derivative probes according to different construction strategies, and explores their mechanism and application in imaging. The limitations and future development of these probes are also discussed.

Contents

1 Introduction

2 Azobenzene derivative probes based on covalent strategies

2.1 Azobenzene derivative probes linked to dyes

2.2 Azobenzene derivative probes linked to AIEgens

3 Azobenzene derivative probes based on noncovalent strategies

4 Conclusion and outlook

Graphical abstract

Keywords

hypoxia / azobenzene derivative / fluorescent probes / cell imaging / cancer

Cite this article

Download citation ▾

Yunxue Wu, Hengyi Zhang, Yu Liu. Application of Azobenzene Derivative Probes in Hypoxia Cell Imaging. Progress in Chemistry, 2021, 33(3): 331-340 DOI:10.7536/PC200934

登录浏览全文

4963

注册一个新账户忘记密码

1 Introduction

Cancer is a complex disease. Uncontrolled cell growth and abnormal cell proliferation at the tumor site lead to cancer with aggressive phenotypes such as infiltration and metastasis, and with high morbidity and mortality^[1,2]^[3]. According to the statistical report of 2018, about 18.1 million new cancer patients and more than 9 million deaths were diagnosed worldwide in the survey of 36 cancers^[4]. Horribly, the incidence of cancer continues to increase^[5]. Although today's medical technology has greatly improved in molecular imaging, clinical diagnosis and chemotherapy, cancer is still a world medical problem that can not be ignored. For cancer, early detection and diagnosis can improve survival and even cure rates^[6,7]. Therefore, it is very important to explore new methods for early detection of cancer.

Compared with normal tissue, the life activity of tumor cells increases, the oxygen demand increases sharply, and the mutated tumor blood vessels can not provide enough oxygen^[8]. This imbalance between oxygen supply and demand causes the oxygen concentration in the tumor site to be lower than that in normal tissue, which can drop to less than 4%, or even to 0% locally^[9⇓~11]. Pathological hypoxia, which is present in most solid tumors, is the most important feature of malignancy and can be used as an indicator of cancer progression^[12,13]^[13,14]. Early detection of hypoxia can not only support the establishment of clinical treatment strategies, but also help to develop individual treatment plans for patients^[15⇓~17].

Fluorescence imaging has become the preferred method for cancer detection because of its high sensitivity, low toxicity, good spatial and temporal resolution, simple operation and non-invasive real-time^[18⇓~20]. Hypoxic fluorescent probes based on reductases, such as NTR (nitroreductase), CyP450 (cytochrome P450), quinone reductase and AzoR (azoreductase), have been developed as a result of the increased reductive stress caused by the low oxygen content at the tumor site^[21]. Nitro or p-nitrobenzyl probes sensitive to NTR appeared earlier and have been widely used. In addition, quinone probes can be reduced by quinone reductase, and azobenzene derivative probes have been widely studied in recent years because of their excellent performance^{[22⇓⇓⇓~26]}^[27,28].

The mode of action of the azobenzene derivative probe is to regenerate the original fluorophore by reductive cleavage of the azo bond, and then reduce the fluorescence. AzoR is overexpressed in hypoxic cells, reducing azobenzenes to anilines by a stepwise transfer of electrons, as shown in Figure 1^[29]. After the azo-group quenching group and the fluorophore are combined in a certain way, in a normal oxygen environment, the fluorophore is excited by light, and the fluorescence is quenched due to the unique absorption of the quenching group on the fluorescence or the rapid conformation inversion around the azo bond; Under hypoxic conditions, the azo bond breaks, the quenching process disappears, and the fluorescence recovers. More and more literatures show that it is a good strategy to construct fluorescent probes by using azo bonds as response elements. The azobenzene derivative probe constructs a controllable off-on fluorescent switch, expands the potential range of the hypoxia sensor, and can dynamically monitor the change of oxygen concentration. In this paper, the azobenzene derivative probes for hypoxic cell imaging are summarized and discussed, and the limitations and development trends of these probes are summarized and prospected.

2 Azobenzene Derivative Probe Constructed by Covalent Bond

2.1 Dye molecule-azobenzene derivative probe

2.1.1 Probe based on FRET mechanism

In 2010, Nagano et al. First used azo groups as hypoxia response factors to construct fluorescent probes^[30]. In order to satisfy both the quenching agent's full effect on the fluorophore and the fluorophore's non-interference with the reductase, they turned their attention to the FRET (fluorescence resonance energy transfer) mechanism and covalently linked the black hole quenching agent (BHQ) with azo bond to the cyanine dye to obtain QCys series probes (Fig. 2b). The probe alters FRET and quenching efficiency by breaking the azo bond, resulting in a change in fluorescence intensity (fig. 2a). In the normal environment, the FRET process between the cyanine dye fluorophore and the BHQ occurs, the fluorescence of the cyanine dye is quenched, and the probe has almost no fluorescence emission; Under hypoxic conditions, AzoR caused the reductive cleavage of the azo double bond, the FRET and fluorescence quenching disappeared, and the fluorescence was restored. The QCys probe has a switching rate of 50 to 100 times, emits near-infrared (NIR) fluorescence with the strongest tissue penetration, has rapid fluorescence reaction, and shows excellent imaging performance in vitro and in vivo.

In 2015, Zhu et al. Constructed a fluorescent probe HP for hypoxia imaging based on rhodamine B and naphthalimide fluorophore^[31]. Naphthalimide is often used to design fluorescent probes because of its high chemical stability, high fluorescence quantum yield and multiple chemical modification sites. 1,8-naphthalimide was directly covalently linked with an azo bond to obtain HPN with both hypoxia response and fluorescence quenching, which was then linked with rhodamine B via an ether bond to construct probe HP (fig. 3). The fluorescence of rhodamine B is very weak due to the FRET effect from rhodamine B to HPN. Under hypoxic conditions, the breakage of the azo bond leads to the decomposition of the HPN structure, the FRET process and quenching disappear, and the fluorescence emission of Rhodamine B is restored (Figure 3), with a significant increase in fluorescence at 581 nm. The reduction potential of cyclic voltammetry showed that the probe HP could be rapidly reduced chemically and enzymatically by CyP450. When co-cultured with HeLa cells (human cervical cancer cells), the fluorescence on-off ratio of HP under hypoxic and normoxic conditions reached 9 times, and the fluorescence intensity was significantly different under various oxygen concentrations.

In addition to responding to hypoxia stimuli, probes sensitive to biomacromolecules have also attracted much attention in recent years. In 2018, Zhu et al. Reported a COX-2 (cyclooxygenase 2) specific hypoxia-responsive NIR fluorescent probe YZP1^[32]. The concentration of COX-2 in tumor is higher than that in normal tissue, and the hypoxic environment will lead to higher concentration. The probe takes a near-infrared fluorescent dye Cy5.5 as a fluorophore, takes indomethacin (IMC) as a targeting group, and takes an azo group as a quencher. After the FRET process and the quenching effect of azo, YZP1 has almost no fluorescence emission. After reduction with CyP450 under hypoxic conditions, the azo bond of YZP1 was broken, Cy5.5 was released and fluoresced, and the fluorescence quantum yield increased from 0.0028 to 0.125. Red fluorescence was observed when YZP1 was co-cultured with HeLa cells (COX-2 overexpression) under hypoxic conditions, and the ratio of fluorescence intensity between hypoxic and normoxic cells could reach 9 times. After treatment with different concentrations of COX-2 inhibitors, the fluorescence gradually decreased with the increase of concentration, which confirmed the binding of YZP1 and COX-2. YZP1 performed well in in vivo imaging experiments using mouse inflammatory models and tumor xenograft models.

Since then, Zheng et al. Developed a probe that simultaneously responds to hypoxia and Cyt C (cytochrome C)^[33]. Cytochrome C is one of the important factors of apoptosis, which is released under the induction of hypoxia. The probe comprises a gold nanoparticle as a core, AZB-DNA-TAMRA (AZB = azobenzene, TAMRA = tetramethylrhodamine) and a Cyt C targeting aptamer embedded with BHQ. Only under the co-stimulation of hypoxia and Cyt C, the azo bond is broken and the aptamer is released, and TAMRA emits fluorescence. In 2019, Zheng et al. Also constructed an AMOF (azoreductase-responsive metal-organic framework) -based nano-fluorescent probe responsive to hypoxia and VEGF (vascular endothelial growth factor) mRNA^[34]. The lack of oxygen causes the azo bond to break, the AMOF structure to be destroyed, and the internal MB (molecular beacon) to be released. Then the pairing of VEGF mRNA with MB destroys the stem-loop structure of MB, the chromophore donor TAMRA and the receptor Cy5 on MB are separated, the original FRET is inhibited, and the fluorescence emission is restored. In addition, Tan et al. synthesized an aptamer linked to a PEG₅₀₀₀-Azo-NHS(NHS=N- hydroxysuccinimide ester) and a dye molecule Cy5/FAM^[35]. The hypoxic environment causes the aptamer to break away from the outer PEGs, resulting in receptor recognition and in situ fluorescence emission. The above four works have constructed hypoxia and biomacromolecule bispecific probes, realized the recognition and imaging of biomacromolecules under hypoxia conditions, solved the problem of detection failure caused by hypoxia, and are expected to be applied to cancer diagnosis and targeted therapy.

2.1.2 Probe based on non-FRET mechanism

In 2013, Hanaoka et al. explored a new way of azo-based probes in view of the defects of QCys series probes (only oxygen concentration below 1% can respond, quencher BHQ-3 has no chemical modification space, etc.)^[36]. Instead of using BHQ and FRET mechanisms, they directly linked the azo bond to the fluorophore covalently, and used the photochemical properties of the azo bond-the rapid conformational change around the NN bond led to the rapid dissipation of excited state energy to achieve fluorescence quenching. In the absence of BHQ, the fluorophores 2Me RG (2-methylrhodamine green) and 2Me SiR600 (2-methylrhodamine 600) were covalently attached to an azo bond, respectively, resulting in MAR (monoazorhodamine) and MASR (monoazosilicorhodamine) probes (fig. 4A). Ultrafast cis-trans isomerization of the azobenzene moiety after light excitation causes the probe to show little fluorescence, which has been verified by quantum chemical calculations. Under hypoxic conditions, the participation of AzoR leads to the breakage of the azo bond and the release of the fluorophore to restore fluorescence^[37]. Hypoxic reactivity was examined in rat liver microsomes containing multiple reductases, and MAR and MASR showed clear fluorescence recovery with on-off ratios of 630 and 20-fold, respectively. Subsequent in vitro experiments further validated the reactivity of the probes to hypoxia (fig. 4B). The MAR probe increases the hypoxia threshold to 5%, and its simple chemical structure is easier to modify, which can be applied to a series of hypoxia detection and imaging with different oxygen concentrations.

In 2015, Renard et al. Developed a far-infrared emission fluorescence probe SR101-NaphtNH₂-Hyp-diMe based on sulforhodamine^[38]. The water-soluble fluorescent probe is prepared by a simple azo-coupling reaction between a fluorescent primary arylamine derived from a sulforhodamine 101 scaffold (SR101-NaphtNH₂) and a tertiary aniline. The azo group is directly attached to the conjugated system of the rhodamine fluorophore, blocking the fluorescence emission of the fluorophore. Its fluorescence quenching is mainly based on two processes: the xanthene group is locked in the unconjugated spironolactone form; Rapid reversal of the E/Z isomerism of the azo bond after photoexcitation. Under hypoxia conditions, the probe regenerates the original sulforhodamine SR101-NaphtNH₂ through the biological reductive cleavage of the azo bond, thereby recovering the strong far-red fluorescence (peak at 625 nm), and the fluorescence on-off ratio reaches 94. A549 cells were used for in vitro imaging test, and the fluorescence intensity of the probe changed significantly. Compared with the MASR probe which also emits far-red fluorescence, the synthesis method of the SR101-NaphtNH₂-Hyp-diMe is simpler, and the structure of the SR101-NaphtNH₂-Hyp-diMe is easier to modify, thereby having larger transformation and application space.

In 2017, Cui et al. Synthesized an NIR fluorescent probe Azo-DCM (AZO = azo group) with high selectivity and long wavelength, which can be used for biological reductase (CyP450 overexpressed in tumor cells) and hypoxia detection^[39]. The azo bond is used as a quencher and a hypoxia response element and is conjugated with a stable fluorophore dicyanomethylene-4H-pyran dye (DCM) to form the probe. The probe showed strong anti-interference ability, and could not be interfered by many substances such as biological thiols, inorganic salts (including sulfites) and amino acids. After treatment with CyP450 and NADH (reduced nicotinamide adenine dinucleotide), the azo bond was broken, DCM was released and emitted fluorescence, and the intensity increased by more than 150 times, showing good selectivity and sensitivity to CyP450 and hypoxia environment. Microscopic images of A549 cells co-cultured with AZO-DCM showed that the fluorescence intensity of AZO-DCM increased with the decrease of oxygen concentration, and the cell imaging ability was good.

Azobenzene is not only a fluorescence quencher, but also an inactivator of nitrogen mustards, which can be used to design hypoxia responsive probes for diagnosis and treatment. In 2017, Kim et al. Covalently linked rhodamine dyes to targeting groups, and then linked them to nitrogen mustard drugs through azo bonds to obtain a mitochondrial targeting fluorescent probe^[40]. In hypoxic cells, the azo bond is broken, fluorescence is restored, and the drug acts on the mitochondria to cause apoptosis. The experimental results at the cell, tissue and animal levels showed that the probe had good selectivity, targeting and activation ability. In the future, azobenzene derivative probes may be used in the construction of drug delivery systems for therapeutic diagnostics.

In 2018, Li et al. Designed a combination probe using dye IR-780, which is an azobenzene derivative probe AXNN and a nitro probe AXNO₂^[41]. The probe showed high specificity for NTR + CyP450 combination enzyme, and could distinguish HepG2 (human hepatoma cells) and 4T1 (breast cancer cells) by fluorescence signal, with good safety, distribution in vivo, accumulation and metabolism. The combination probe is superior to the single probe in the performance of multiple indicators, and may play a huge potential in cancer diagnosis in the future.

2.2 AIE molecular-azobenzene derivative probe

Aggregation-induced luminescence (AIE) is a photophysical phenomenon, which was first proposed by Tang Benzhong et al in 2001^[42]. In contrast to the aggregate fluorescence quenching (ACQ) effect of conventional fluorophores, AIE molecules exhibit weak fluorescence in the dispersed state and strong fluorescence in the aggregated state^[42,43]. This unique phenomenon has attracted the attention of many scientific researchers. In the current exploration, the fluorescence of AIE molecules covers the whole visible light region, which has a wide range of applications in biosensing and optical devices^[44,45]. AIE molecules are important for fluorescence imaging due to their high fluorescence brightness, resistance to photobleaching, and ability to track cells for long-term imaging^[46]. In recent years, a variety of AIE molecules based on azo, nitro and N⁺-O^- have been used to design hypoxia-responsive fluorescent probes^[47,48].

The classical AIE unit TPE (tetraphenylethylene) is easy to synthesize and modify, and is often used to design and synthesize AIE derivatives with relatively complex structures and specific functions^[49]. In 2018, He et al. Successfully prepared a hypoxia responsive polymer containing TPE and azobenzene (Fig. 5)^[50]. Uniform nanoparticles with TPE-Azo as the core and biocompatible PEG (polyethylene glycol) as the shell were prepared by Azo-coupling reaction in aqueous solution. The polymer exhibits an extremely weak fluorescence emission due to the FRET process and the typical absorption of the azo chromophore. After treatment with the enzyme NQO 1 (human quinone oxidoreductase) and the coenzyme NADPH (reduced nicotinamide adenine dinucleotide phosphate), the fluorescence recovery of TPE was evident due to the reduction of azo bonds and aggregation of hydrophobic TPE (fig. 5). This work not only provides a new method for the preparation of azobenzene-containing polymers, but also provides inspiration for researchers to construct hypoxia responsive nanoprobes based on TPE and azobenzene.

In 2019, Zhou et al. reported a novel hypoxia-responsive AIE probe, PCL-TPE-Azo-PEG (PCL = polycaprolactone)^[51]. TPE-Azo, as a middle part, is connected with a hydrophilic PEG chain and a hydrophobic PCL chain respectively to form an amphiphilic block copolymer. Due to the quenching effect of azo groups on the fluorescence of TPE, the polymer self-assembled into rod-like micelles without fluorescence in phosphate buffer solution. After treatment with sodium dithionite or AzoR, the azo bond was reductively broken and the assembly was fragmented into PEG and PCL segments. The fluorescence of hydrophobic segment PCL-TPE recovered and showed AIE effect, which caused a significant increase in the fluorescence intensity near the wavelength of 500 nm. This means that the probe has potential applications in biosensing and imaging of tumor hypoxia.

It has been proved that the introduction of hydrophilic PEG chain and hydrophobic azobenzene moiety into a small molecule through azo-coupling reaction is an effective method for the synthesis of amphiphilic diblock copolymers. Based on the above strategy, He et al. Successfully constructed an amphiphilic PEG-Azo-TPE hypoxia-responsive AIE probe for cell imaging^[52]. Due to its amphiphilic structure, the probe can form micelles in aqueous solution, resulting in stable activatable AIE sites, and the fluorescence of the AIE fluorophore in the azobenzene cage is quenched by the presence of FRET and azo. After excitation with 365 nm light and treatment with AzoR, the azo bond was reduced and the fluorescence of TPE was restored, and the fluorescence on-off ratio was 25. Microscopic images after co-culture with A549 cells under hypoxic and normoxic conditions (fig. 6) indicate that the probe responds well to the tumor hypoxic microenvironment. In addition, due to the two-photon absorption cross section of the AIE polymer, the probe can be excited by 760 nm light in addition to ultraviolet light and can be used for two-photon imaging. Furthermore, the excitation wavelength is in the NIR region, providing the possibility of deep tissue imaging.

3 Azobenzene Derivative Probes Based on Non-covalent Bond

Supramolecular chemistry is a new branch of chemistry that mainly studies non-covalent interactions and supramolecular assemblies with specific structures and functions formed by non-covalent bonds. The concept was first proposed by French chemist Jean-Marie Lehn in 1978^[53]. Common noncovalent interactions include hydrogen bonds, electrostatic interactions, π-π stacking, hydrophobic effects, and Van der Waals forces^[54⇓~56]. Supramolecular assembly is not a simple molecular superposition, it has functions beyond the host and guest molecules, and can be used in drug loading, dye adsorption, phase transfer catalysis and other fields^[57,58]. Supramolecular chemistry provides a new way to construct non-covalent systems for many fields, such as materials, biology, medicine and so on^[59,60]. In recent years, azobenzene derivatives have been widely used to construct supramolecular assemblies and applied in many fields due to their photoisomerism and the ability to form stable inclusion complexes with macrocyclic hosts^[61].

At present, most of the hypoxia-responsive fluorescent probes reported in the literature are covalently linked to dye molecules by azo groups, and then fluorescence recovery is achieved through covalent bond breakage (Fig. 7A). This construction method has some disadvantages that can not be ignored, such as complex molecular design, time-consuming and expensive synthesis process, activity change caused by covalent linkage, and potential toxicity caused by the introduction of exogenous groups. Supramolecular chemistry can provide a non-covalent construction method that is easier to implement: the host-guest interaction between the azo macrocyclic host molecule and the dye molecule is used to establish the non-covalent connection between the azo group and the dye molecule to construct the supramolecular assembly probe. In order to realize the above non-covalent strategy, it is necessary to find suitable host-guest molecules. The macrocyclic host of azobenzene derivatives should have the following conditions: (1) it can establish a tight connection with the dye guest; (2) to achieve maximum fluorescence quenching in a normal environment to achieve a smaller imaging background; (3) It has good hypoxia response, and can release dye molecules and restore fluorescence under hypoxia conditions^[62].

Guo et al. Reported a supramolecular assembly probe CAC4A-Rho123 for hypoxic cell imaging^[62]. In order to meet the above host conditions, they synthesized a carboxyl-modified azocalixarene CAC4A, which included the dye molecule Rho123 (Rhodamine 123) in the cavity of CAC4A (Fig. 7B). The tight binding between CAC4A and Rho123 is due to the synergy of several noncovalent interactions, namely, the N — H … O and C — H … O hydrogen bonds, the electrostatic force between electron-rich CAC4A and electron-poor Rho123, and the π-π stacking between the two molecules. Unlike most of the covalently linked azo derivative probes, the fluorescence quenching of CAC4A-Rho123 is due to the electron transfer from the donor CAC4A to the acceptor Rho123 after photoexcitation, resulting in a significant decrease in the fluorescence intensity of Rho123. After treatment with the reducing agent sodium dithionite, all four azo bonds were reduced and broken, and the dye molecules were released and the fluorescence was restored, and the fluorescence was enhanced by 8 times. When the probe was exposed to various biological species (adenosine triphosphate, metal ions, amino acids, glucose, urea, creatinine and bovine serum albumin), the fluorescence intensity did not change significantly.

It is proved that the supramolecular assembly can exist stably in complex biological environment. A549 cells were used for fluorescence detection under normoxic (20%O₂) and hypoxic (below 0.1%O₂) conditions, and the fluorescence of the probe was almost invisible under normoxic conditions and significantly enhanced under hypoxic conditions (fig. 7c). CAC4A-Rho123 probe is the first azo-derivative probe constructed by non-covalent strategy, which not only opens up a new way for the construction of hypoxia fluorescence imaging probes, but also provides a new idea for drug loading and cancer therapy due to its non-covalent linkage and easy-to-modify macrocyclic host.

Tang et al. reported a hypoxia-responsive nanogel for anticancer drug loading^[63]. The nanogels were prepared by host-guest interaction between PLG-g-mPEG/Azo and PLG-g-mPEG/β-CD (PLG-G-mPEG = (L-glutamic acid) -graft-poly (ethylene glycol) methyl ether, β-CD = β-cyclodextrin), and loaded with anti-cancer drug RNase. Azo was included in β-CD to form a cross-linked site, which was destroyed by the cleavage of the Azo bond and the release of RNase after treatment with NTR and NADPH. In vitro release experiment, 75.0% of RNase was released in 72 hours under hypoxia condition, while only 19.7% was released under normoxia condition, which proved that the nanogel could release drugs under hypoxia stimulation. This shows that the hypoxia-responsive azo supramolecular assembly probe constructed by Guo et al is not an example, and the construction of such supramolecular assembly by host-guest interaction is a feasible method. Because there is no complex synthesis, it can achieve high-fidelity release and build a universal platform for easy modification, which has potential applications in hypoxia imaging and anti-cancer drug loading.

The molecule to be loaded with that drug or dye may be encapsulate in vesicles or micelles in addition to being encapsulate in a macrocyclic host. In 2016, Mallik et al. Synthesized a diblock copolymer PLA₈₀-(AZB)-PEG₄₇(PLA= polylactic acid), which can self-assemble into vesicles in aqueous medium and encapsulate carboxyfluorescein^[64]. Under hypoxia, the vesicles were disintegrated due to the breakage of the azo bond, and 90% of the dye molecules were released in 50 min, and the fluorescence intensity was greatly increased (Fig. 8). The fluorescence quenching of vesicles comes from the ACQ (aggregate fluorescence quenching) phenomenon of dye molecules, and its background fluorescence intensity is higher than that of traditional probes, but it provides an example for hypoxia-responsive vesicles.

In 2019, Zheng et al developed a hypoxia imaging fluorescent probe MCM @ TATp for mitochondrial autophagy^[65]. Amphiphilic polymers containing azobenzene groups self-assemble to form micelles encapsulating Mito-rHP, a rhodamine derivative with mitochondrial targeting and pH sensitivity. Under hypoxia, the azo bond is broken and Mito-rHP is released into the mitochondria, and the acidic environment caused by mitochondrial autophagy leads to fluorescence emission. The successful construction of MCM @ TATp shows that it is a feasible method for hypoxia imaging by loading dye in micelles.

In 2020, He et al. Developed an in-situ self-assembled AIE probe^[66]. The probe comprises an Azo copolymer WS-AC connected by a carbamate and an AIE molecule AWS-TPE modified by a sulfonic acid group with negative charge. The two components have high solubility in aqueous solution and can penetrate into deep tissues, and the fluorescence emission of AWS-TPE in a dispersed state is weak. In the absence of oxygen, the azo bond of WS-AC is cleaved and the carbamate undergoes a 1,6 self-elimination cascade to give a positively charged protonated primary amine. Under the electrostatic interaction, WS-AC and AWS-TPE self-assembled into nanoparticles, and AWS-TPE was encapsulated in the nanoparticles. The fluorescence intensity was significantly enhanced due to the molecular aggregation (Fig. 9). In vitro experiments using multicellular spheroids of cervical cancer demonstrated the feasibility of in situ self-assembled probes. This in situ self-assembly method is expected to achieve both high permeability and long residence time of the probe in the tumor site.

4 Conclusion and prospect

Azobenzene derivative probes have been widely developed in recent years due to their good sensitivity, specificity and other advantages, and have been used in hypoxic cell imaging, which is helpful for the early diagnosis of cancer. Azo bonds can not only be broken and reduced in response to hypoxia, but also affect the fluorescence intensity of adjacent fluorophores, which is the basis and inspiration for the development of azobenzene derivative probes. In this review, we focus on the azobenzene derivative probes that use the azo bond as a hypoxia responsive switch, classify and introduce the probes in detail according to different construction strategies, and explore their mechanism of action and application ability in hypoxic cell imaging.

The construction strategies of azobenzene derivative probes are mainly divided into two kinds: the first is that the azobenzene group is covalently connected with a strong fluorescent chromophore, the azo bond is broken under hypoxia conditions, and the fluorophore restores fluorescence emission to realize imaging, which is the construction method of most hypoxic cell imaging probes at present; The second is to construct supramolecular assemblies by non-covalent means. Under hypoxia conditions, the azo bond breaks, the assembly disintegrates, the dye molecules are released, and the fluorescence is restored. The construction of hypoxic cell imaging probes through covalent bonds has made great progress, and they are more mature and sensitive in hypoxia imaging. The non-covalent construction strategy has the advantages of simple and safe synthesis, and can use different dye molecules, so it has great potential for development.

However, the reported azobenzene derivative probes are still in the stage of in vitro and in vivo imaging tests in the laboratory, and no probe has reached the clinical standard. On the one hand, in order to enter the clinical stage, various properties of the probe, such as toxicity, photostability, biocompatibility, biodegradability, tissue distribution, pharmacokinetic behavior and so on, need to be tested rigorously and thoroughly. On the other hand, the probe still has defects in imaging performance and needs further modification. The fluorescence emission wavelength of most probes is in the short wave region, which is not only easy to cause damage to cells and tissues, but also can not be used for deep tissue imaging, so it is necessary to develop NIR fluorescence imaging probes. In addition, the fluorescence emission intensity of the probe could not establish a linear relationship with the oxygen concentration of the tissue, which could only qualitatively judge the tissue hypoxia and roughly show the degree of hypoxia, but could not quantitatively describe the hypoxia situation. Therefore, optimizing the probe performance and exploring the diversity of probes are very important for the clinical application and future development of azobenzene derivative probes.

As a new field just over ten years old, azobenzene derivative probes for hypoxic cell imaging have great development space. The emergence of new AIE luminescence and non-covalent construction strategies not only optimizes the performance and simplifies the synthesis, but also broadens the development direction of this field. Molecular recognition of hypoxia response, phototherapy, chemotherapy and drug delivery are also developing, and the combination of hypoxia imaging and complex probes provides inspiration for the creation of new cancer diagnosis and treatment methods and the development of synergistic therapies^{[67⇓⇓⇓⇓⇓⇓⇓⇓~76]}. The development of azobenzene derivative hypoxia imaging probes is of great significance for cancer monitoring and diagnosis, and is worthy of further study.

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Gilkes D M, Semenza G L, Wirtz D. Nat. Rev. Cancer, 2014, 14(6):430.

[2]	Vaupel P. Oncol., 2008, 13(S3):21.

[3]	Valastyan S, Weinberg R A. Cell, 2011, 147(2):275.

[4]	Bray F, Ferlay J, Soerjomataram I, Siegel R L, Torre L A, Jemal A. CA: A Cancer J. Clin., 2018, 68(6):394.

[5]	Torre L A, Bray F, Siegel R L, Ferlay J, Lortet-Tieulent J, Jemal A. CA: A Cancer J. Clin., 2015, 65(2):87.

[6]	Bach P B, Cramer L D, Warren J L, Begg C B. N Engl J. Med., 1999, 341(16):1198.

[7]	Zhang Z, Bast R C, Yu Y H, Li J N, Sokoll L J, Rai A J, Rosenzweig J M, Cameron B, Wang Y Y, Meng X Y, Berchuck A, van Haaften-Day C, Hacker N F, de Bruijn H W A, van der Zee A G J, Jacobs I J, Fung E T, Chan D W. Cancer Res., 2004, 64(16):5882.

[8]	Brown J M, Wilson W R. Nat. Rev. Cancer, 2004, 4(6):437.

[9]	Muller M, Padberg W, Schindler E, Sticher J, Osmer C, Friemann S, Hempelmann G. Anesth. Analg., 1998, 87(2):474.

[10]	CervÓs-Navarro J, Diemer N H. Crit. Rev. Neurobiol., 1991, 6:149.

[11]	Vaupel P, Schlenger K, Knoop C, Hçckel M. Cancer Res., 1991, 51:3316.

[12]	Parks S K, Cormerais Y, PouyssÉgur J. J. Physiol., 2017, 595(8):2439.

[13]	Schito L, Semenza G L. Trends Cancer, 2016, 2(12):758.

[14]	Eltzschig H K, Carmeliet P. N Engl J. Med., 2011, 364(7):656.

[15]	Minn H, Gronroos T, Komar G, Eskola O, Lehtio K, Tuomela J, Seppanen M, Solin O. Curr. Pharm. Des., 2008, 14(28):2932.

[16]	Jubb A M, Buffa F M, Harris A L. J. Cell. Mol. Med., 2010, 14(1/2):18.

[17]	Rajendran J G, Hendrickson K R G, Spence A M, Muzi M, Krohn K A, Mankoff D A. Eur. J. Nucl. Med. Mol. Imaging, 2006, 33(S1):44.

[18]	Chan J, Dodani S C, Chang C J. Nat. Chem., 2012, 4(12):973.

[19]	Zipfel W R, Williams R M, Webb W W. Nat. Biotechnol., 2003, 21(11):1369.

[20]	Stephens D J. Science, 2003, 300(5616):82.

[21]	Liu J N, Bu W B, Shi J L. Chem. Rev., 2017, 117(9):6160.

[22]	Cui L, Zhong Y, Zhu W P, Xu Y F, Du Q S, Wang X, Qian X H, Xiao Y. Org. Lett., 2011, 13(5):928.

[23]	Guo T, Cui L, Shen J N, Zhu W P, Xu Y F, Qian X H. Chem. Commun., 2013, 49(92):10820.

[24]	Li Z, He X, Wang Z, Yang R, Shi W, Ma H. Biosens. Bioelectron., 2015, 63:112.

[25]	Zhang J, Liu H W, Hu X X, Li J, Liang L H, Zhang X B, Tan W H. Anal. Chem., 2015, 87(23):11832.

[26]	Zhou J, Shi W, Li L H, Gong Q Y, Wu X F, Li X H, Ma H M. Chem. Asian J., 2016, 11(19):2719.

[27]	Zhang P Y, Huang H Y, Chen Y, Wang J Q, Ji L N, Chao H. Biomaterials, 2015, 53:522.

[28]	Komatsu H, Shindo Y, Oka K, Hill J P, Ariga K. Angew. Chem. Int. Ed., 2014, 53(15):3993.

[29]	Zbaida S, Levine W G. Chem. Res. Toxicol., 1991, 4(1):82.

[30]	Kiyose K, Hanaoka K, Oushiki D, Nakamura T, Kajimura M, Suematsu M, Nishimatsu H, Yamane T, Terai T, Hirata Y, Nagano T. J. Am. Chem. Soc., 2010, 132(45):15846.

[31]	Cai Q, Yu T, Zhu W P, Xu Y F, Qian X H. Chem. Commun., 2015, 51(79):14739.

[32]	Ti Y Z, Yu L, Tang Y, Jin T X, Yang M, Wang R, Xu Y F, Zhu W P. Sensor Actuat. B: Chem., 2018, 265:582.

[33]	Tang J R, Huang C X, Shu J Y, Zheng J, Ma D D, Li J S, Yang R H. Anal. Chem., 2018, 90(9):5865.

[34]	Liu N, Zou Z, Liu J, Zhu C, Zheng J, Yang R H. Anal., 2019, 144(21):6254.

[35]	Zhou F, Fu T, Huang Q, Kuai H L, Mo L T, Liu H L, Wang Q Q, Peng Y B, Han D M, Zhao Z L, Fang X H, Tan W H. J. Am. Chem. Soc., 2019, 141(46):18421.

[36]	Piao W, Nagano T, Hanaoka K. Angew. Chem. Int. Ed., 2013, 52:13028.

[37]	Shin N, Hanaoka K, Piao W, Miyakawa T, Fujisawa T, Takeuchi S, Takahashi S, Komatsu T, Ueno T, Terai T, Tahara T, Tanokura M, Nagano T, Urano Y. ACS Chem. Biol., 2017, 12(2):558.

[38]	Chevalier A, Piao W, Hanaoka K, Nagano T, Renard P Y, Romieu A. Methods Appl. Fluoresc., 2015, 3(4):044004.

[39]	Cui L, Shi Y P, Zhang S P, Yan L L, Zhang H, Tian Z R, Gu Y Y, Guo T, Huang J H. Dye. Pigment., 2017, 139:587.

[40]	Verwilst P, Han J Y, Lee J, Mun S, Kang H G, Kim J S. Biomaterials, 2017, 115:104.

[41]	Tian X W, Li Z, Sun Y, Wang P, Ma H M. Anal. Chem., 2018, 90(22):13759.

[42]	Luo J D, Xie Z L, Lam J W Y, Cheng L, Tang B Z, Chen H Y, Qiu C F, Kwok H S, Zhan X W, Liu Y Q, Zhu D B. Chem. Commun., 2001,(18):1740.

[43]	Hong Y N, Lam J W Y, Tang B Z. Chem. Commun., 2009,(29):4332.

[44]	Li K T, Lin Y J, Lu C. Chem. Asian J., 2019, 14(6):715.

[45]	Zhan R Y, Pan Y T, Manghnani P N, Liu B. Macromol. Biosci., 2017, 17(5):1600433.

[46]	Feng G X, Liu B. Acc. Chem. Res., 2018, 51(6):1404.

[47]	Xu G P, Tang Y H, Ma Y Y, Xu A, Lin W Y. Spectrochimica Acta Part A: Mol. Biomol. Spectrosc., 2018, 188:197.

[48]	Xu C H, Zou H, Zhao Z, Zhang P F, Kwok R T K, Lam J W Y, Sung H H Y, Williams I D, Tang B Z. Adv. Funct. Mater., 2019, 29(34):1903278.

[49]	Yang B, Zhang X Y, Zhang X Q, Huang Z F, Wei Y, Tao L. Mater. Today, 2016, 19(5):284.

[50]	Li S, Wang J L, Shen J J, Wu B, He Y N. ACS Macro Lett., 2018, 7(4):437.

[51]	Yuan X J, Wang Z, Li L S, Yu J W, Wang Y Q, Li H K, Zhang J D, Zhang Z B, Zhou N C, Zhu X L. Mater. Chem. Front., 2019, 3(6):1097.

[52]	Xue T H, Jia X Q, Wang J L, Xiang J Y, Wang W, Du J J, He Y N. Chem. Eur. J., 2019, 25(41):9634.

[53]	Lehn J M. Pure Appl. Chem., 1978, 50(9/10):871.

[54]	Ma J C, Dougherty D A. Chem. Rev., 1997, 97(5):1303.

[55]	Smithrud D B, Sanford E M, Chao I, Ferguson S B, Carcanague D R, Evanseck J D, Houk K N, Diederich F. Pure Appl. Chem., 1990, 62(12):2227.

[56]	Hancock R D. J. Chem. Educ., 1992, 69(8):615.

[57]	Tian R, Wang H M, Niu R F, Ding D. J. Colloid Interface Sci., 2015, 453:15.

[58]	Chen X M, Zhang Y M, Liu Y. Supramol. Chem., 2016, 28(9/10):817.

[59]	Vögtle F. Supramoleculare Chemie. Teubner, Stuttgart, 1991.

[60]	Lehn J M. Angew. Chem. Int. Ed. Engl., 1988, 27(1):89.

[61]	Liu Y H, Liu Y. Progress in Chemistry, 2019, 31(11): 1528.

[62]	(刘耀华, 刘育. 化学进展, 2019, 31(11): 1528.)

[63]	Geng W C, Jia S R, Zheng Z, Li Z H, Ding D, Guo D S. Angew. Chem. Int. Ed., 2019, 58(8):2377.

[64]	Si X H, Ma S, Xu Y D, Zhang D W, Shen N, Yu H Y, Zhang Y, Song W T, Tang Z H, Chen X S. J. Control. Release, 2020, 320:83.

[65]	Kulkarni P, Haldar M K, You S, Choi Y, Mallik S. Biomacromolecules, 2016, 17(8):2507.

[66]	Ma D D, Huang C X, Zheng J, Zhou W, Tang J R, Chen W J, Li J S, Yang R H. Anal. Chem., 2019, 91(2):1360.

[67]	Xue T H, Shao K C, Xiang J Y, Pan X Y, Zhu Z X, He Y N. Nanoscale, 2020, 12(14):7509.

[68]	Wang W L, Lin L, Ma X J, Wang B, Liu S R, Yan X X, Li S R, Tian H Y, Yu X F. ACS Appl. Mater. Interfaces, 2018, 10(23):19398.

[69]	Zhang X L, Wu M, Li J, Lan S Y, Zeng Y Y, Liu X L, Liu J F. ACS Appl. Mater. Interfaces, 2018, 10:21909.

[70]	Yan Q, Guo X L, Huang X L, Meng X, Liu F, Dai P P, Wang Z, Zhao Y J. ACS Appl. Mater. Interfaces, 2019, 11(27):24377.

[71]	Yang G B, Phua S Z F, Lim W Q, Zhang R, Feng L Z, Liu G F, Wu H W, Bindra A K, Jana D, Liu Z, Zhao Y L. Adv. Mater., 2019, 31(25):1901513.

[72]	Xu Z T, Pan C, Yuan W Z. Biomater. Sci., 2020, 8:3348.

[73]	Ihsanullah K M, Kumar B N, Zhao Y Y, Muhammad H, Liu Y, Wang L, Liu H, Jiang W. Biomaterials, 2020, 245:119982.

[74]	Lee J, Oh E T, Yoon H, Woo Kim C, Han Y J, Song J, Jang H, Joo Park H, Kim C. Nanoscale, 2017, 9(20):6901.

[75]	Li J J, Meng X, Deng J, Lu D, Zhang X, Chen Y R, Zhu J D, Fan A P, Ding D, Kong D L, Wang Z, Zhao Y J. ACS Appl. Mater. Interfaces, 2018, 10(20):17117.