Biosensors; a Novel Concept in Real-Time Detection of Autophagy

Abstract

Autophagy is an early-stage response with self-degradation properties against several insulting conditions. To date, the critical role of autophagy has been well-documented in physiological and pathological conditions. This process involves various signaling and functional biomolecules, which are involved in different steps of the autophagic response. During recent decades, a range of biochemical analyses, chemical assays, and varied imaging techniques have been used for monitoring this pathway. Due to the complexity and dynamic aspects of autophagy, the application of the conventional methodology for following autophagic progression is frequently associated with a mistake in discrimination between a complete and incomplete autophagic response. Biosensors provide a de novo platform for precise and accurate analysis of target molecules in different biological settings. It has been suggested that these devices are applicable for real-time monitoring and highly sensitive detection of autophagy effectors. In this review article, we focus on cutting-edge biosensing technologies associated with autophagy detection.

1. Introduction to autophagy

The term autophagy refers to several mechanisms of cytoplasmic degradation via the lysosome or the analogous organelle in plants and yeast, the vacuole. These include chaperone-mediated autophagy (CMA), microautophagy, and macroautophagy, the latter of which has been best characterized. Macroautophagy is a housekeeping and early-stage cellular mechanism involved in the elimination of excessive and dysfunctional proteins, protein aggregates, and organelles via the formation of phagophores; these sequestering compartments mature into double-membrane autophagosomes that fuse with lysosomes leading to cargo degradation and recycling (Chen et al. 2021). This process of organelle and protein degradation is integral to cellular homeostasis; accumulation of this material—especially in nondividing cells—can lead to the progression of abnormal growth and various pathological conditions (Rezabakhsh et al. 2017a). In eukaryotes, the process of degradation and elimination is orchestrated primarily using two distinct types of machinery, namely the ubiquitin-proteasome system and macroautophagy. Of note, the ubiquitin-proteasome system is associated with the removal of short-lived protein substrates, whereas macroautophagy participates in the degradation of long-lived proteins (Zhou et al. 2022). There are several types of macroautophagy that function in response to different forms of stress. In this regard, the precise cellular mechanisms support the selective degradation of, for example, damaged mitochondria, aggregated proteins, endoplasmic reticulum, and microorganisms using mitophagy, aggrephagy, reticulophagy, and xenophagy, respectively (Vainshtein and Grumati 2020). The physiological significance of macroautophagy is not solely limited to the maintenance of cell survival and emerging data have indicated the possible role of macroautophagy in the stimulation of specific effectors which are associated with cell death mechanisms (Lee et al. 2021; Zada et al. 2021). In general, macroautophagy acts as a cytoprotective mechanism in different cellular settings. However, under specific conditions, the promotion of a macroautophagic response can lead to cell death and cytostatic outcomes (Rezabakhsh et al. 2017c; Yu and Klionsky 2022). Therefore, the dual activity of macroautophagy under different situations correlates with the host cell metabolic activity, synthetic function, and cell-death stimuli (Noguchi et al. 2020).

According to emerging data, there is an intricate cross-talk of macroautophagy machinery with regulated cell death mechanisms such as inflammation, apoptosis, necroptosis, ferroptosis, etc. in both normal and cancer cells (Gao et al. 2022; Liu et al. 2022). Concurrently, numerous in vitro and in vivo studies have shown that the activation of the TLR (toll-like receptor) signaling pathway, especially the TLR4 axis, in the presence of danger/damage-associated molecular patterns/DAMPs and pathogen-associated molecular patterns/PAMPs leads to the activation of macroautophagy signaling effectors. In particular, the stimulation of specific TLR factors such as MYD88, and TICAM1/TRIF triggers the separation of BECN1 from BCL2 and promotes the formation of autophagosomes (Hayat 2017). It was suggested that the depletion of Casp1 in mouse hepatocytes leads to the suppression of macroautophagy-related factors and the accumulation of injured organelles within the host cells (Sun et al. 2013). Due to the complexity of the macroautophagy machinery, it has been suggested that macroautophagy can efficiently support the process of stem cell development, maturation, and differentiation into target lineages; thus, macroautophagy is extremely important in the area of regenerative medicine (Hassanpour et al. 2018).

2. Autophagy machinery and molecular bases

As noted above, there are three major types of autophagy. In microautophagy, the autophagic cargo is directly engulfed by lysosomes through a protrusion and septation or invagination process, followed by enzymatic digestion (Schuck 2020). The second type of autophagy that transfers protein substrates directly into lysosomes is CMA (Hubert et al. 2022). Unlike micro- and macroautophagy, CMA is only used for the degradation of individual, unfolded proteins. The recognition of KFERQ motifs in target substrates by cytosolic HSPA8/Hsc70 is critical to the promotion of CMA (Schnebert et al. 2022). This process is followed by unfolding of the substrate and translocation directly across the lysosome membrane via an oligomeric channel and the action of lumenal HSPA8 (Su et al. 2020). The morphological hallmark of macroautophagy is the formation of double-membrane vesicles, autophagosomes, at phagophore assembly sites. This phenomenon is orchestrated with the active participation of ATG (autophagy related) genes, leading to the fusion of autophagosomes; upon fusion with endosomes and/or lysosomes these compartments mature into autolysosomes (Figure 1) (Wang et al. 2022). In the remainder of this article, the term autophagy refers to macroautophagy. The identification and characterization of many of the components of the molecular machinery initially in yeast, revealed multiple complexes that control the progression of autophagy. The Atg1-Atg13 kinase complex supports the formation of phagophores from clusters of Atg9-containing vesicles (Stjepanovic et al. 2014). Atg9 is a transmembrane protein that functions as a lipid scramblase, working in conjunction with the Atg2 lipid transfer protein (Noda 2021). Atg1 is a protein kinase, and Atg13 is a regulatory component. These proteins are part of a pentameric complex consisting in addition of Atg17, Atg29, and Atg31. The Atg17-Atg31-Atg29 trinary complex acts as a scaffold to recruit Atg1 and Atg13 to the phagophore assembly site; furthermore, the Atg1 kinase complex also acts in later stages of autophagy including fusion. In human cells, the autophagic initiating complex consists of ULK1 (unc51-like autophagy activating kinase 1) and ULK2, ATG13, RB1CC1/FIP200, and ATG101.

Figure 1.

Schematic diagram of autophagic response. This phenomenon consists of different steps as follows: (A) Initiation; (B) nucleation-expansion-maturation; (C) fusion; and (D) degradation and recycling. ATG proteins are the main molecular counterparts in the promotion and acceleration of the autophagic response (Li et al. 2020c). This figure was modified from Figure 1, originally published in (Li et al. 2020c) with the permission of the publisher. (Copyright 2020, Molecular Cancer).

Downstream of the Atg1/ULK1 kinase is the class III phosphatidylinositol 3-kinase (PtdIns3K) complex, which includes Vps34/PIK3C3/VPS34, Vps15/PIK3R4/VPS15, Vps30/Atg6/BECN1 (beclin 1) and Atg38/NRBF2. The existence of Atg14/ATG14 in the PtdIns3K complex facilitates the transmission of the complex to phagophore formation sites and the activity of the PtdIns3K complex leads to phagophore expansion and maturation (Hurley and Young 2017).

Formation and maturation of the phagophore also involve two unique conjugation reactions that involve the ubiquitin-like proteins Atg8 and Atg12. The conjugation systems include Atg3, Atg4, Atg5, Atg7, Atg10 and Atg16, and the process has been described in detail (Rubinsztein et al. 2012). In brief, the Atg12–Atg5-Atg16 complex acts as an E3 ligase to facilitate the conjugation of Atg8 to phosphatidylethanolamine (PE). This process, termed lipidation, anchors Atg8 to the phagophore, and subsequently the autophagosome, membrane. On the concave surface of the phagophore, Atg8–PE plays an important role in cargo recognition. As with much of the autophagy machinery, the conjugation proteins are conserved in more complex eukaryotes. In mammals there are two subfamilies of Atg8 proteins, terms MAP1LC3/LC3 and GABARAP. These proteins bind various receptors including SQSTM1/p62, CALCOCO2/NDP52, OPTN and NBR1, which link different cargoes to the autophagy machinery. The progression of autophagy involves completion of the phagophore to generate the autophagosome, fusion with the lysosome and formation of autolysosomes. These steps lead to the degradation of the autophagosome cargo via the enzymatic activity of a multitude of hydrolases and subsequent release of the breakdown products back into the cytosol via lysosomal permeases, allowing their reuse for cellular metabolism.

3. Autophagy detection

To date, there is no single applicable gold standard protocol for the determination and characterization of autophagic activity in all experimental platforms. The previously established guidelines lack enough criteria to precisely indicate an autophagic response applicable to the wide range of conditions that can induce this process (Klionsky et al. 2021). The conventional analyses applied for this purpose are not applicable to all cell types and experimental systems. There are also issues with regard to accuracy, sensitivity, and specificity. In addition, the complexity and dynamic nature of the autophagy molecular machinery often cause researchers to incorrectly discriminate between autophagy induction and subsequent blockage versus a complete autophagic response (Ylä-Anttila et al. 2009). For example, an increase in autophagosome numbers in the ultrastructural analysis of target cells using electron microscopy can be associated with enhanced autophagosome formation or inhibition of autophagosome turnover. Although it is possible to calculate autophagosome and cytosol ratios based on an analysis of early and late autophagosomes, this methodology requires considerable skill and experience, which is not practical in many circumstances (Ylä-Anttila et al. 2009). Furthermore, in situations associated with prolonged autophagic processes, it is difficult to differentiate autophagosomes from other endocytic compartments such as heterophagic vacuoles.

LC3 is initially proteolytically processed by ATG4, to generate LC3-I. This form is then conjugated to PE to generate LC3–PE, commonly referred to as LC3-II. The latter is involved in the maturation of autophagosomes; accordingly, the LC3-II:LC3-I ratio is commonly calculated using a western blot for measuring the extent of autophagic status in several experimental settings (Rezabakhsh et al. 2017c; Rezabakhsh et al. 2019). LC3 has four isoforms LC3A, LC3B, LC3B2 and LC3C, and the LC3B isoform in particular is integral to autophagosome formation and maturation (Song et al. 2019b; Tang et al. 2022). Most conventional antibodies can detect this isoform; however, antibodies have a differential affinity to LC3-I and LC3-II while the varied cellular distribution of LC3 in several tissues and cell types can lead to misinterpretations (Kimura et al. 2009). In fluorescence microscopy, the term “punctate LC3” or green fluorescent protein (GFP)-LC3 refers to direct or indirect immunofluorescence detection of this protein, which has been used to determine an approximate number of autophagosomes in each cell type. Unfortunately, most cells exhibit small numbers of punctate LC3 under normal physiological conditions (Kimura et al. 2009). Thus, it is mandatory to define stimulated and basal (background) autophagy states in different tissues. Despite their applicability in clinical areas, these methods are time consuming, expensive, laborious, and need skillful staff. These bottlenecks force scientists to investigate new recognition strategies with better detection sensitivity and specificity. Due to the simple preparation steps, low costs, easy operation, and high sensitivity, biosensors have been broadly exploited for analyzing autophagy (Kim and Seong 2021). In this review, we discuss the current advances in biosensors for autophagy monitoring and address the challenges and gaps for future projects.

4. Sensing and Biosensing types

Diagnostics is a vast scientific bridge that connects chemistry engineering to clinical analytics. Generally, diagnostic systems are divided and classified into two main categories: sensors and biosensors. This classification is based on different strategies and modalities in using or neglecting biorecognition elements used in the construction of these sensing systems. Indeed, biorecognition molecules refer to some antigen-specific biological molecules with the potential to recognize a specific antigen. Considering the difference between the sensors and biosensors, this section gives a clearer view of the applied methods and constructions applied to fabricate the correlated structures. The presence or absence of certain bioreceptors in the detection framework has a sensible prominent effect on selectivity, sensitivity, response time, and other analytical factors (Mansouri et al. 2020; Nakhjavani et al. 2023).

4.1. Chemical Sensors

The detection of chemical molecules is of great significance for analyzing various matrices, including environmental, biological, and food samples. In this scenario, this category consists of, varied types of organic and inorganic molecules such as toxins, pesticides, and signaling molecules that can be monitored in gas, liquid, or solid phases. In the construction of sensors, no bioreceptor-antigen coupling mechanism has been employed to analyze target molecules. Based on the fact that there is a wider range of chemical molecules compared to biological molecules, this category sensor type is more diverse in design and analysis. The target antigens are commonly extracted from biofluids; these molecules can be extracted from biological matrices (such as blood, serum, and urine), environmental samples (such as soil, water, and wastewater), food products, etc. (Meyyappan 2016; Schroeder et al. 2018).

In electrochemical sensing, a chemical sensor is used and there are two methods of evaluation used for analysis. Some analyte (target recognition molecule) types can be monitored based on a redox reaction, oxidation, or reduction activity. In this class, an analyte has an oxidation or reduction activity that can be employed as an indicator for recognition. This strategy has better selectivity and accuracy because two signal indicators including peak position and peak height are used in the analysis process. As a case in point, the detection of H₂O₂ is of great use among the chemical sensing approaches for cancer cell screening; cancer cells can produce large amounts of H₂O₂ resulting in a high concentration around the cancerous cells compared to normal cells (Lin et al. 2019). In many different cancer screening studies using H₂O₂ sensing, supernatant H₂O₂ levels are analyzed to detect malignancy and the anaplastic rate, which can be used as an indicator for cancer incidence and severity (Asif et al. 2022; Dou et al. 2018; Li et al. 2019c). Due to their redox activity and absence of luminescence properties, electrochemical sensors are the most prominent strategies for the direct detection of H₂O₂ (Kaur et al. 2018a; Sumalekshmy et al. 2007).

The analysis of metal ions is another important application of chemical sensors, which is crucial in the evaluation of water quality and contamination of biological systems. Especially of note, heavy metal ions such as Hg²⁺, Cd^2+, and Pb²⁺ are hazardous to biological organisms, and their analysis is of utmost importance. However, many chemical analytes cannot be recognized using a straightforward approach of screening them via the production of signaling activity. To circumvent this pitfall, the second format of electrochemical sensing, in which the analysis is based on the indirect detection of an analyte, has been introduced. In such strategies, the analytes have no apparent redox activity, and quantitative and qualitative monitoring is achieved by reducing or enhancing the electrochemical signals of known redox agents such as potassium ferrocyanide (K₄[Fe[CN]₆] (Doiron et al.)). In contrast to the former strategy, this approach can be used for all molecules but with lower selectivity because there is no analyte-associated peak position (Lake et al. 2019; Paderni et al. 2021). In luminescence sensors, the existence of analytes can affect the luminescence emission in terms of both intensity (increase or decrease) and light shift in the emission peak position (blue shift or red shift) (Berhanu et al. 2019; Gruber et al. 2017). Along with the application of different detection methods, enzymatic approaches are one of the highly selective methods for designing chemical sensors. For example, in the case of H₂O₂, the enzyme horseradish peroxidase/HRP has often been used for the direct detection of H₂O₂ or indirect analysis of other molecules based on their effects on the activity of H₂O₂ (Lian et al. 2016; Sun et al. 2018). Another strategy for enzymatic approaches is based on the suppression of catalytic activity in the presence of specific analytes. Pesticide detection, for example, falls within this category (Arduini et al. 2019).

In addition to enzymatic approaches, other protocols have been designed for chemical sensors. For example, a molecularly imprinted polymer (MIP)-based framework has been used for the fabrication of chemical sensors. These constructs are based on the host-guest mechanism in which the MIP is generated onto the electrode in the presence of the analyte. The procedure is followed by the extraction of the analyte from the holes (hosts) on the MIP-modified electrode (Ahmad et al. 2019; BelBruno 2018; Haupt et al. 2020). The selectivity of these sensors is based on the specificity of the hosts to target molecules via physical interaction compared to the other interferences (Chen et al. 2017; Lian et al. 2015). It is thought that the selectivity of these strategies is lower than that of the other enzymatic approaches, which results in the lower selectivity of the MIPs compared to those of the enzymatic reactions.

One of the attractive applications of chemical sensors is the detection of gasses and volatile compounds which is crucial for human health care systems and environmental protection. Although various traditional gas sensing systems have been broadly employed, their inherent restrictions have barricaded their further development. Therefore, numerous attempts have been made to fabricate improved sensing materials and frameworks with excellent performance for gasses and volatile material compounds. To this end, recent decades have witnessed the advent and development of different sensor types, such as chemiresistive, luminescence, and electrochemical approaches, which are frequently studied in the literature (Dai et al. 2020; Hein et al. 2020).

The basic principle of chemiresistive sensing strategies is an alteration in the electronic features or resistance of the sensing substrates upon the interaction and adsorption mechanisms with the analyzed compounds. One important tip is that the applied transducing materials like metallic nanomaterials should possess a desirable charge-transferring ability. Also, the used sensing substrate should be able to effectively capture the analyte molecules. Boosting both of these features could improve the analysis efficiency performance. In the case of luminescence sensors, the major sensing mechanisms include photoelectron transfer between the luminophore and the guest analytes, intermolecular charge transferring, Förster resonance energy transferring from the luminophore to target molecules, and competitive adsorption mechanisms (Burratti et al. 2018; Li et al. 2020a; Wu et al. 2020a). In all of these mentioned mechanisms, the energy is transmitted from emission molecules to the analytes, in which the signals decline in the presence of an analyte.

Monitoring of oxygen is one of the most common uses of gas sensors (Ehgartner et al. 2016; Kelly et al. 2020). Oxygen is the most important and substantial element in human and other organisms’ lives as it participates in metabolic mechanisms to produce energy for cells through cellular respiration. The oxygen concentration must be adjusted within a defined range in normal cells, tissues, and organs. Decreasing or increasing reduction or elevation of oxygen content level in cells causes hypoxia and hyperoxia, respectively. Severe hypoxia may lead to some disorders such as cardiac inflammatory diseases, various cancers, ischemic diseases, and lung diseases, whereas hyperoxia leads to oxygen toxicity syndrome (Terraneo et al. 2017). Oxygen also has extensive applications in industrial and medical domains. Thus, the recognition of oxygen is of great significance in biological and chemical monitoring and industry (Pereira et al. 2017; Stoeckel et al. 2017; Zhu et al. 2017).

In total, chemical sensors are an important part of diagnostics which are employed in different analytical approaches such as luminescence and electrochemical methods, and for a wide range of molecules from clinically relevant molecules (such as glucose and H₂O₂) to gas detection (such as greenhouse gases). Overall, this field needs continued progress toward increasing the sensitivity and especially the specificity of these applications.

4.2. Biosensors

The concept of biosensors refers to methods in which biorecognition elements such as antibodies, DNA strands, proteins, and enzymes are applied to target specific molecules (Nasrollahpour and Khalilzadeh 2023). The detection process in biosensors is based on an indirect effect of the analyte on the signaling system (Chen and Wang 2020; Scheller and Schubert 1991). The first and most popular effect is the steric hindrance of the analyte, which takes signaling molecules away from the transducer surface (Liu et al. 2019b). This strategy is very useful in designing electrode-based biosensors (electrochemical, electrochemiluminescence (Chenaghlou et al. 2021; Nasrollahpour et al. 2021b), and photoelectrochemical). In these methods, the signal readouts decrease with an increase in the concentration of an analyte (Wei et al. 2021). These methods are known as direct strategies and are very simple, time effective, and easy to prepare. Of note, some of these methods are more difficult to implement but exhibit more selectivity compared to direct methods (Quintela and Wu 2020). Using a layer-by-layer preparation strategy, the target analyte is sandwiched between the probe and reported biorecognition elements (Seo and Gu 2017a). In these architectures, biorecognition element 2 carries signaling or amplifier molecules. In the presence of an analyte, the signals are increased due to the label attached to this element (Dong et al. 2019a; Seo and Gu 2017b). The other format of biosensing configurations is ratiometric strategies, which are favorites for fluorescence biosensors (Huo et al. 2019). Two (or even more) signals are employed in this class, and their ratio is defined by analyte concentration. The ratiometric and sandwich-type methods are more complicated but have more selectivity and accuracy than direct methods (Farka et al. 2017; Youssef et al. 2019b).

Generally, biosensors can be categorized into four different configurations based on the applied bioreceptor: Immunosensors, genosensors, aptasensors, and cytosensors (Babamiri et al. 2019). These terms are attributed to the biorecognition molecules and target analytes. In immunosensors, the biorecognition elements and analytes are antibodies and proteins, respectively (Aizawa 2019). Antibody molecules consist of several functional groups, including amine, carboxylic, and hydroxyl groups, which are useful in designing immunosensors (Campbell 1991). Amine and carboxylic moieties are very popular in attaching antibodies on different substrates via amide bounds or coordination reactions (Sharafeldin et al. 2020). Without further treatment, antibodies could be used as biorecognition elements due to their intrinsic structure with rich functional groups. Modification of the Fc moiety with biotin is another method to attach antibodies to the transducing surfaces. According to data obtained from several studies, this strategy is based on the high-affinity interaction of biotin and streptavidin, making antibodies flexible to adapt to different attachment strategies (Sangili et al. 2020; Tang et al. 2021).

Genosensing is another format of biosensors in which a nucleotide sequence is used to detect a specific single strand of DNA (Crevillen et al. 2022). The detection mechanism is based on the coupling of purine and pyrimidine bases between capture and target DNA. The most common protocols for the recognition and monitoring of DNA strands are based on molecular biology strategies, including polymerase chain reaction/PCR (Zou et al. 2018), northern blot procedures (Schneider et al. 2018), and DNA microarray (Marzancola et al. 2016). These techniques have high performance, but often are tedious, complicated, expensive, time consuming, and also need highly skilled operators (Eivazzadeh-Keihan et al. 2019). In addition to its invasiveness, samples obtained from cancer tissues need several stages of extraction and enhancement of DNA sequences to reach a high level of sensitivity. Unlike solid cancer types, recognizing circulating DNA strands in biofluids does not require any challenging steps (Abu-Salah et al. 2010). microRNAs (miRNAs) with 18–25 nucleotides are the most important DNA analytes which have been used in bioanalysis (Bastami et al. 2019; Saliminejad et al. 2019). MiRNAs can control the expression of mRNAs involved in varied cellular functions (Beyrampour-Basmenj et al. 2022; Yao et al. 2019). Biofluid-circulating miRNAs are promising biomarkers for cancer detection and monitoring (Kilic et al. 2018). Hence, the rapid and cost-effective protocols for miRNA analysis have attracted growing attention in biological analysis and clinical aspects (Alzate et al. 2020; Mohammadi et al. 2019).

Aptasensing is another biosensing strategy in which aptamers, synthesized DNA strands, are used as bio-recognition elements to detect proteins, cells, or other molecules (Kilic et al. 2018). Indeed, aptamers are single-strand DNA or RNA with 15 to 40 nucleotides. These bio-recognition elements can efficiently interact with specific target biochemical molecules, such as cancer biomarkers, metal ions, antibiotics (Dezhakam et al. 2023b), environmental pollutants (Hasanzadeh et al. 2009; Hasanzadeh et al. 2010), nucleotides, proteins (Karimzadeh et al. 2020; Nasrollahpour et al. 2021a), and drugs (Balal et al. 2009; Karim-Nezhad et al. 2009; Khalilzadeh et al. 2011; Saghatforoush et al. 2009; Shafaei et al. 2022), making them high-performance recognition elements (Kim et al. 2021). Aptamers are fabricated through an in vitro evolution of ligands, pointing to an exponential enrichment/SELEX process. These recognition elements play a crucial role in treatment and diagnostics (Kaur et al. 2018b; Zhou and Rossi 2017). Due to appropriate stability, it is suggested that aptamers can be used under different conditions in which antibody-based assays are not possible (Arshavsky-Graham et al. 2022). Unlike antibodies, aptamers maintain their physical stability after being exposed to high temperatures and different pH values. These features make aptamers promising tools and suitable substitutes for antibodies in designing sensing and biosensing frameworks for various analytes (Eivazzadeh-Keihan et al. 2018; Mahmoudpour et al. 2021; Mao et al. 2020). Because of a particularly short length of aptamers, these elements are more eligible for nucleases in comparison with antibodies. Along these lines, it is noteworthy to mention that the applicability of aptamers is diminished in samples containing nucleases such as blood (Bauer et al. 2019; Crivianu-Gaita et al. 2016).

Monitoring of circulating tumor cells (CTCs) in the metastatic cancer state is crucial for the early-stage detection and therapy of cancer (He et al. 2019). Therefore, the screening of CTCs holds great promise to minimize the application of other high-risk procedures for cancer analysis (Vajhadin et al. 2020b). Generally, the number of CTCs is a valuable criterion for evaluating cancer medication because increasing systemic cancer cells indicates the therapeutic regimens are ineffective (Williams 2013). Cytosensors are biosensing platforms for the identification and quantification of specific cell types (Vajhadin et al. 2020a). These frameworks can predict the possibility of malignancy and survival expectancy via the rapid and effective recognition of CTCs. Various cytosensing approaches have been designed to recognize CTCs, including electrochemical (Sun et al. 2019), electrochemiluminescence (Li et al. 2019b), fluorescence (Guerreiro et al. 2019), photoelectrochemical (Xu et al. 2021), etc. Depending on the different analytical purposes, cytosensing efforts can be classified into two main types including cell detection and cell imaging (Duan et al. 2020). Most studies have determined the type and number of CTC with a focus on the evaluation of physiological characteristics, pharmaceutical assessment, and detection and measuring of the quality of specific molecules located on the membrane or inside the cytosol of target cells.

5. Applications

5.1. Fluorescence Assays

Fluorescence is the phenomenon of light emitting by a subject (an atom or molecule) from an electronic singlet excited state level to its ground electronic level after absorption of a photon (Jameson 2014). In recent decades, fluorescence diagnostics have been utilized as screening and analysis strategies in developing new and highly efficient assays (Valeur and Berberan-Santos 2011). There is a broad range of compounds and nanomaterials that represent intrinsic fluorescence properties, making them suitable for the preparation of several frameworks (Basabe-Desmonts et al. 2007; Nasrollahpour et al. 2023c). Over the last decades, numerous fluorescent-based assays have been developed for the determination of various types of cations (Na²⁺, Mg²⁺, Zn²⁺, Cu²⁺, Al³⁺, etc.), anions (Cl⁻, Br⁻, carboxylates, PO₄³⁻, ATP, etc.) and neutral compounds (glucose, O₂) (Carter et al. 2014). Employing fluorescence phenomena for evaluations of large biomolecules such as enzymes, proteins, DNA strands, and cells is one of the biggest achievements in diagnostics, leading to the revolution of the clinical-analytical field over the last 30 years (Wu et al. 2020b). The diagnostic efficiency of fluorescence-based biosensors results from their high sensitivity rate, fast response time, selectivity, simplicity, and flexibility. The most important drawbacks of fluorescent methods are the intrinsic fluorescence features of nontarget chemical-biological materials which cause background noise via non-specific false-positive responses (Liu et al. 2019a). Thus, a clear sample solution is needed to prohibit possible interference emissions. In addition to increasing the cost of tools, this property also enhances the size of the instrumentation to limit unwanted noise, making it difficult to miniaturize, use on-site analysis, and apply in resource-poor settings (Romei and Boxer 2019). It has been documented that utilizing a ratiometric design with multiple fluorescence centers could answer these challenges to some extent (Lu et al. 2020; Youssef et al. 2019a).

In this configuration type, dual emission centers are irradiated from one or two materials with the potential to minimize possible interference (Laliwala et al. 2022; Raja et al. 2021; Zhan et al. 2021). Among the many methodologies developed for ratiometric methods, applications based on fluorescence resonance energy transfer/FRET can achieve superior efficiency because of its benefits of high efficiency, nice stability, good reproducibility, and large stokes shifts (Vangara et al. 2018; Zehra et al. 2018). This strategy has become the most popular application of ratiometric fluorescence protocols and was widely employed for the recognition of various analytes (Gong et al. 2020; Wang et al. 2019c). The development of fluorescence sensors and biosensors aims to explore ways to increase the quantum yield of current emitters or produce new fluorophores with improved properties, using some methods and tools to decrease the background noise and size of instrumentation (Feng et al. 2022; Wang et al. 2021a). Fluorescence diagnostics can be categorized into two different classes: i) methods in which the emitting moieties interact directly with the target molecules. In these methods, the analyte may enhance or quench the emission intensity of the applied fluorophore via complexation, dissolution, or decomposition mechanisms (Abu-Ali et al. 2022; Chu et al. 2022). This strategy is very nice for the detection of enzymes, small molecules, anions, and specialty metal cations which can easily make a coordination bond with conventional fluorophores; ii) the second strategy is based on the indirect monitoring of target analytes in which the fluorophore has no interaction with the analyte. This strategy is the most common for large biomolecules such as proteins, nucleic acids, and cells which need a more specific detection mechanism than a simple interaction with a fluorophore (Choi et al. 2020; Nasu et al. 2021). In these protocols, the interaction between an analyte and bio-recognition elements (e.g., antibody, aptamer, peptide, etc.) makes a change in the emission conditions of the fluorophore resulting in emission with different intensities and/or different wavelengths.

Fluorescence biosensing is the most common technique in monitoring autophagy in recent decades (Li et al. 2021b). Among many strategies, photoluminescence (PL) based on fluorescent probes is one the pioneer methods used in analyzing autophagy (Li et al. 2021b). This is because of the capability of various fluorescent emitters to enter the cells (Li et al. 2021b). In addition, these probes are easily prepared and non-destructive, which are critical aspects in routine large-scale point of care (Wang et al. 2021b). The fluorescent probes for autophagy target lysosomes or mitochondria and attach to them, making them visualizable with a fluorescence instrument (Hou et al. 2018; Park et al. 2020). As mentioned above, the autophagy process involves lysosomes or mitochondria in living cells changing their environment and morphologies. Thus, monitoring these organelles is crucial in studying certain aspects of autophagy. In one experiment, Liu et al. developed a new fluorescent probe ([E]-4-(2-[7-{diethylamino}-2-oxo-2H-chromen-3-yl]vinyl)-1-dodecylpyridin-1-ium iodide [CVP]) for long-term monitoring of mitophagy in living cells and tissues. The phospholipid-biomimetic construction of CVP makes it highly selective to mitochondria without affecting the membrane potential. Because of the long alkyl chain, CVP can efficiently enter almost all tissue in which mitochondria are stained with this probe during the first 3 minutes. In this study, HeLa cells were co-stained with green CVP, and then LysoTracker Deep Red and exposed to mitophagy inducers pepstatin A and carbonyl cyanide m-chlorophenyl hydrazone (CCCP). Immunofluorescence imaging indicated that the number of yellow cells with overlapped CCCP and CVP increases over time after induction of mitophagy. Figure 2 represents the interface-targeting strategy of CVP. CVP can accumulate in mitochondria via electrostatic and hydrophobic interactions. Using an excitation wavelength (λ_ex) of 473 nm a maxima wavelength of emission (λ_em) was observed at 540–640 nm. This strategy visualizes four types of mitochondria in tissues (Liu et al. 2020).

Figure 2.

Long-term tracking of mitochondria and monitoring mitophagy using CVP. Reproduced with permission (Liu et al. 2020). Copyright 2020, Elsevier.

In a study conducted by Zhu and coworkers, a cell-permeable pyridium salt derivative (TpsPym) was used for monitoring autophagic death in HepG2 cells (Zhu et al. 2020). This fluorochrome can specifically label mitochondria and endoplasmic reticulum because of its cationic nature and affinity for lipid-rich compounds after 30 minutes. Co-staining of cells with MitoTracker Far Red, the endoplasmic reticulum marker ER-Tracker Blue-White DPX, and TpsPym reveals the selectivity of TpsPym to target these organelles (Zhu et al. 2020). Using 4D-STED evaluation and TpsPym staining, it was suggested that the cytoplasmic area covered by the endoplasmic reticulum is diminished in autophagic cells after starvation. These features coincide with the swelling and increase of cristae distance within the mitochondrial mass (Zhu et al. 2020).

With the progression of autophagic response, the fusion of lysosomes with autophagosomes leads to the reduction of pH values in autophagosomes (Yim and Mizushima 2020). The same story is also true with mitochondria during mitophagy. Hereupon, designing and developing fluorescent probes with proper sensitivity to pH fluctuation is an efficient strategy during the autophagic response (Xu et al. 2020). Xu and colleagues fabricated pH-responsive chondroitin sulfate-based polymersomes (Tuni/HCQ@CS-PAE) for co-delivery of autophagy blocker Hydroxychloroquine and tunicamycin to exert anti-tumor activity via inhibiting autophagy response and reticulum endoplasmic injury (Xu et al. 2020). After exposure to a pH value of 5, both Hydroxychloroquine and tunicamycin are released. Due to protonation properties, Tuni/HCQ@CS-PAE can deacidify the lysosomes and prevent the formation of autophagolysosomes. Under these conditions, the release of drug and Hydroxychloroquine from polymersomes leads to the inhibition of cancer cell growth and autophagolysosome formation. Acridine orange staining indicated that treatment of mouse breast cancer cell line 4T1 with Tuni/HCQ@CS-PAE increased the number of red spots inside the cells and the red/green signal ratio was evaluated by flow cytometry analysis. Inside the cell, acidic compartments like lysosomes and autophagolysosomes become red in the presence of Acridine orange while the cytoplasm and nucleic acids appear green (Xu et al. 2020). Besides, they monitored the status of autophagy after exposure to Tuni/HCQ@CS-PAE in cancer cells transfected with a green fluorescent protein (GFP)––mCherry––LC3 expressing adenoviral vector. In non-treated control cells, the expression of mCherry-LC3, and GFP-LC3 promotes yellow puncta in autophagosomes. Further fusion with lysosomes, leads to the quenching of GFP-LC3, and cells exhibit only a red appearance due to the existence of mCherry-LC3, indicating the formation of autophagolysosomes. In the presence of Tuni/HCQ@CS-PAE, cells exhibit yellow puncta because of lysosome deacidification and inhibition of the autophagic flux (Figure 3). A pH-responsive polysiloxane-based ratiometric fluorescence probe, namely PA1 designed by Zhang et al., can follow H⁺ activity around the mitochondrial surface in liver HepG2 cells (Zhang et al. 2022). PA1 can efficiently track lysosomes inside cells. Immunofluorescence imagining reveals colocalization of green PA1 and LysoBlue in HepG2 cells under normal conditions. Upon the activation of the autophagic response and fusion of lysosomes with autophagosomes, the pH values are reduced in autolysosomes of starved HepG2 cells. Along with these changes, PA1-stained lysosomes appear as red fluorescence and these values are induced by the progression of autolysosome formation. Treatment of these cells with the autophagy inhibitor chloroquine reduces the fusion of lysosomes with autophagosomes and results in a moderate decrease in red fluorescence intensity (Zhang et al. 2022). The current system can indicate the subcellular localization of lysosomes and subsequent fusion with autophagosomes for monitoring the autophagic response.

Figure 3.

Evaluation of autophagic response in 4T1 cells after exposure to Tuni/HCQ@CS-PAE. Immunofluorescence imaging (A) and flow cytometry analysis (B) of cells stained with Acridine orange. In the group treated with Tuni/HCQ@CS-PAE, numerous intracellular red puncta are apparent in the cytosol indicating the accumulation of lysosomes and autophagosomes compared to the other group. Besides, flow cytometry analysis revealed an increased red/green ratio in Tuni/HCQ@CS-PAE cells related to other groups, indicating the inhibition of autophagic flux. Immunofluorescence imaging of mCherry-GFP-LC3 expressing cells 48 hours after treatment with Tuni/HCQ@CS-PAE (C). The number of LC3 puncta in treated cells (D). TEM imaging (E). Data showed an increase in autophagosome number (green arrows) and a reduction of autophagolysosomes (red arrows). Reproduced with permission (Xu et al. 2020). Copyright 2020, Science Advances.

All physiological processes originate from certain biochemical reactions inside the host cells by production-specific factors (Wang et al. 2019a) and numerous lines of data have indicated crosstalk between varied molecular signaling pathways (Wang et al. 2019a). Of note, the precise discrimination between autophagy and programmed cell death, namely apoptosis, is at the center of debate under several pathological conditions (Rezabakhsh et al. 2017b). To monitor superoxide anion (O^2•−) levels and pH values in HeLa cells with apoptosis and autophagy, Yand and colleagues used a dual-ratiometric fluorescent nanoprobe (Yang et al. 2017). For this purpose, MSN@RhB@SiO2@CS nanocomposites were prepared to hold nano-emitters DBZTC and Tpy-Cy, sensitive to O^2•− and pH, respectively. The obtained MSN@RhB@SiO2 @DBZTC@Tpy-Cy@CS assembly was then modified with triphenylphosphonium/TPP for mitochondria targeting in HeLa cells (Figure 4). In this study, rhodamine B (RhB) was used as a control fluorophore to enhance the accuracy of the results. RhB and Tpy-Cy represent high PL emissions at purple and red wavelengths after exciting at 543 and 633 nm, respectively. The calibration curves for both analytes are obtained by plotting the PL of each emitter divided by the PL of RhB versus concentration. The data indicate that with the initiation of apoptosis and autophagy, mitochondrial pH values are decreased at the early stages. Along with these changes, cellular O^2•− contents are increased in apoptotic cells and remain unchanged in cells with autophagy. The progression of autophagy and apoptosis elevate O^2•− content at late stages. Monitoring the pH index shows that in cells with a progressive autophagy response, pH values are lower whereas these features remain unchanged in apoptotic cells (Yang et al. 2017).

Figure 4.

Preparation of MSN@RhB@SiO2 @DBZTC@Tpy-Cy@CS nano-emitter with three emitters in the nanocomposite structure. Reproduced with permission (Yang et al. 2017). Copyright 2017, American Chemical Society.

Mitochondrial function and morphology are integral to cell viability (Walter et al. 2019), and the level of ion homeostasis has a vital role in determining mitochondrial morphology (Li et al. 2020b). Studies have shown that the process of ion homeostasis is regulated via several genes and their associated proteins (López-Hernández et al. 2020; Pirouz et al. 2020). For instance, Mdm38/Mkh1 is part of a K⁺/H⁺ exchange system in the mitochondrial inner membrane. Mutations in the encoding gene can dysregulate the function of ion channels and lead to mitochondrial swelling and reduced cell growth rate (Tolkovsky 2009). Nowikovsky et al. used a PL fluorescent biosensor for monitoring mitophagy after changes in K⁺/H⁺ exchange function and downregulation of the MDM38 gene after morphological changes (Nowikovsky et al. 2007). In Mdm38-depleted cells, numerous spherical mitochondria with reduced cristae, fragmentation, swelling, and vacuolar structure appear in the cytosol, indicating the promotion of mitophagy (Nowikovsky et al. 2007). The biosensor has two components including a pH-sensitive GFP and a pH-insensitive red fluorescent protein (Nowikovsky et al. 2007). They found that with the progression of mitochondrial dysfunction and mitophagy, GFP⁺ mitochondrial fragments are reduced with retained RFP appearance, indicating the lowering of pH values and mitochondrial turnover. In addition, this biosensor can be used for monitoring mitophagy induced via osmotic stress in the target cells and tissue.

One of the drawbacks associated with the application of pH-responsive probes in autophagy monitoring is the difficulty of spanning the entire autophagy process (Ding et al. 2022a). This restriction is imposed by the pKa of the probes which cannot be sensitive in the entire pH range (Ding et al. 2022a). It was suggested that pH values differ in each step of autophagy (Ying et al. 2022). As a correlate, more applicable parameters should be monitored in the entire process. Viscosity-responsive probes can address this problem (Shi et al. 2022), and Zhai and coworkers monitored autophagy using a ratiometric viscosity-sensitive probe, Lyso-Vis (Zhai et al. 2022). They demonstrate that this probe is viscosity responsive when tagged with lysosomal targeting moieties and provides a dual fluorescence response—two photons in red and green wavelengths. Using an 810-nm two-photon excitation, they monitored the relationship between autophagy induction and viscosity changes in PC12 and BV-2 cells. According to the data, the exposure of cells to low temperatures and incubation with dexamethasone (a viscosity inducer), causes a reduction in the green fluorescence intensity coincident with the promotion of red fluorescence (Zhai et al. 2022). The study showed that the incubation of cells with an autophagy inducer, rapamycin, leads to a red appearance in cells, whereas co-treatment of these cells with rapamycin and an autophagy inhibitor, 3-methyladenine, leads to the reduction of red fluorescence and retention of green fluorescence (Zhai et al. 2022). These features showed that changes in viscosity rate could be an alternative approach for monitoring autophagy under different pathological conditions.

Long noncoding RNAs (lncRNAs) are RNAs with sequences longer than 200 nucleotides without protein-coding properties (Statello et al. 2021). It was suggested that lncRNAs are key regulators in anaplastic changes and can be regarded as valuable biomarkers in theranostics (Dezhakam et al. 2023a; Li et al. 2021a). The function of lncRNAs correlates with the development of carcinomas by affecting several cellular functions, especially autophagy (Luo et al. 2021; Shafabakhsh et al. 2021). Therefore, monitoring lncRNAs can give us many unknown facts about metabolism and autophagy status, and the progression of pathological changes. In this regard, Cai et al. previously fabricated a PL assay for monitoring autophagy flux using a fluorescent gold-short hairpin RNA nano-complex (Au–shRNA NCs). This system was utilized for monitoring autophagy in hepatocellular carcinoma cell lines HepG2, SMMC-7721 cells, and an orthotopic tumor model in mice (Cai et al. 2021). Inside the cancer cells, Au nanoparticles undergo reduction. This feature can increase the possibility of interaction between Au³⁺ with negatively charged shRNA and self-assembling of Au–shRNA NCs compared to the normal control cell lines (L02). After excitation of Au–shRNA NCs at 488 nm, immunofluorescence imaging indicates that the formation of this complex and attachment to MALAT1 reduces autophagic flux in these cells. These features coincide with the increase of RFP-SQSTM1/p62 puncta (corresponding to protein aggregates and/or autophagosomes) and reduction of LAMP2 (a lysosome marker)-positive spots, indicating the inhibition of autophagy machinery at the latter steps (Figure 5). The conventional technique for detecting RNA strands is real-time polymerase chain reaction (PCR), which is a relatively expensive method requiring skilled personnel. However, the fluorescence biosensor designed in this study offers improved applicability without the need for highly skilled individuals (Mahdipour et al. 2010). Table 1 illustrates the advances in autophagy analysis using PL assays.

Figure 5.

The regulation of MALAT1 in hepatocellular carcinoma using bio-self-assembled fluorescent Au–shRNA NCs (a-e). TEM imaging indicates intracellular accumulation of numerous autophagosomes in HepG2 cells after treatment with Au–shRNA NCs. Au particles can be detected inside the autophagosomes (red arrows). Western blotting indicates the reduction of the LC3-II/-I ratio and increase of P62, indicating the inhibition of autophagic flux (c and d). Cells were exposed to RFP-LC3 for 24 hours before the treatment with Au–shRNA NCs. Immunofluorescence imaging indicates that the incubation of HepG2 with Au–shRNA NCs reduced the conversion of RFP-LC3 into LC3-II, resulting in the accumulation of RFP-LC3 in the cytosol. Along with these changes, autophagosome and lysosome fusion protein LAMP2 is also diminished (d: Scale bar: 5 μm). These data indicate that Au–shRNA NCs can reduce autophagic response in cancer cells via the control of autophagosome formation and fusion with the lysosome. Au–shRNA NCs caused the accumulation of P62 in cancer cells, indicating the reduction of autophagic flux (e: Scale bar: 5 μm). (n= 3-5); *p<0.05, **p<0.01. Reproduced with permission (Cai et al. 2021). Copyright 2021, MDPI.

Table 1.

Developed photoluminescence assays for autophagy monitoring.

Method	Analytes	Organelles	Nano-emitters	Ref
Ratiometric PL	(O2•−) and H+	Mitochondria	MSN@RhB@SiO2 @DBZTC@Tpy-Cy@CS (MSN@RhB@SiO2@CS was employed to hold the emitters from leaking)	(Yang et al. 2017)
PL	Mitochondria and endoplasmic reticulum	Mitochondria and endoplasmic reticulum	TpsPym	(Zhu et al. 2020)
PL	Mitochondria	Mitochondria	CVP	(Liu et al. 2020)
PL	Mdm38	Mitochondria	JC-1	(Nowikovsky et al. 2007)
Ratiometric PL	H+	Lysosomes	PA-1	(Zhang et al. 2022)
Ratiometric PL	Viscosity	Lysosomes	Lyso-Vis	(Zhai et al. 2022)
PL	H+	Mitochondria and Lysosomes	Rosella	(Sargsyan et al. 2015)
PL imaging+ prognosis	MALAT1 mRNA	Autophagic flux	Au–shRNA NCs	(Cai et al. 2021)
Ratiometric PL	RALB	Endomembrane in HEK-HT-H-RasV12 cells	fluorescent proteins (CFP, YFP)	(Singh et al. 2019)
Ratiometric PL	H+	Lysosomes	SN-Lyso	(Zhang et al. 2021)
Dual PL	Viscosity	Lysosomes	IQ-LVs	(Park et al. 2020)
Ratiometric PL	H+	Lysosomes	Lyso-SP and Lyso-SQ	(Ding et al. 2022b)
Ratiometric PL	Viscosity	Lysosomes	Lyso-NP	(Hou et al. 2018)
LAMP2A	LAMP2A protein	Lysosomes	KFERQ	(Issa et al. 2018)
Ratiometric PL	H+	Lysosomes	RpH-LAMP1-3xFLAG	(Ponsford et al. 2021)

O2•−: superoxide anion; MSN: mesoporous silica nanoparticle; RhB: rhodamine B; DBZTC: 2-chloro-1,3-dibenzothiazo-linecyclohexene; CS: chitosan; Tpy-Cy: 2-{4-[4′-aminomethylphenyl)-2,2′:6′,2″-terpyridinyl]-7-(1-ethyl-3,3-dimethyl- (indolin-2-ylidene)}-3,5-(propane-1,3-diyl)-1,3,5-heptatrien-1- yl)-1-ethyl-3,3-dimethyl-3H-indolium; PL: photoluminescence; TpsPym: pyridium salt derivative; CVP: JC-1: (E)-4-(2-(7-(diethylamino)-2-oxo-2H-chromen-3-yl)vinyl)-1-dodecylpyridin-1-ium iodide; 5,50,6,60-tetrachloro-1,10,3,30-tetraethyl benzimidazole carbocyanine iodide; PA-1: polysiloxane-based probe.

5.2. Chemiluminescence assays

The term chemiluminescence (CL) is a phenomenon in which luminescence is produced as a result of a chemical reaction (Hananya and Shabat 2019; Yang et al. 2020b). Certain compounds, including intermediates and fluorophores, are generally excited by an oxidation step forming an excited state and high-energy intermediates. Decomposition and return of excited molecules to the ground state lead to the production of luminescence at different wavelengths (Yang et al. 2020a). Considering the diverse chemical energy exchange mechanisms, CL can be classified into two different formats: i) direct CL and ii) indirect CL. In the former route, a chemiluminescent indicator is oxidized to generate a highly energy-state intermediate which decomposes in the following step to produce emitting species (Liu et al. 2021). For example, luminol, which is a popular CL substrate, can be directly oxidized by H₂O₂ to induce CL (Song et al. 2019a). In the indirect CL form, an energy-transfer step often contributes to the CL mechanism (Gnaim et al. 2018). The excited-state intermediates transfer their energy to the nearby fluorophores in a non-irradiation manner via resonance energy transfer. This brings luminophore molecules to their excited state energies, making energy in the form of luminescence and decaying to their ground energy levels (Yan et al. 2020).

In situ generated photons can replace conventional excitation light to formulate chemi-excited photodynamic therapeutic protocols or drug delivery systems with smart pharmaceutical releases (Chen et al. 2021; Tang et al. 2019). CL methods are very beneficial for the screening and remediation of deeply seated diseases or cancers. Meanwhile, CL experiments require simpler and cost-effective instrumentation and operation, have wide dynamic ranges compared to optically exciting techniques and are more appropriate for miniaturization. CL research is highly motivated by the recognition of different chemical and biological analytes (Lin et al. 2020; Wang et al. 2019b).

Lin and coworkers studied the effect of quercetin-rich guava juice and trehalose on the protection of the kidney and pancreas against type 2 diabetes in rats (Lin et al. 2016). In this research, kidney H₂O₂ and hypochlorite (HOCl) were monitored by a luminol-amplified CL assay. Guava juice can eliminate both H₂O₂ and HOCl while trehalose eliminates only H₂O₂. In this protocol, CL is generated from the interaction of H₂O₂ and HOCl in the presence of luminol. Following the consumption of guava juice and trehalose, the total concentrations of H₂O₂ and HOCl are decreased indicating their positive effect on preventing harmful diabetic effects (Lin et al. 2016). Along with these changes, the protein levels of autophagy-related proteins such as LC3, and BECN1 (beclin 1) are reduced in diabetic rats that receive quercetin-rich guava juice. Western blotting indicates that protein levels of factors related to apoptosis (CASP3 [caspase 3]) and pyroptosis (IL1B/IL-1β) are also reduced. Tsai et al. used another CL method for measuring reactive oxygen species (ROS) in the presence of L-theanine in rats with hyperactive bladder (Tsai et al. 2017). In vitro analysis indicated that L-theanine can inhibit the generation of HOCl, and H₂O₂ from H₂O₂-luminol-phosphate-buffered saline (PBS) and HOCl-luminol-PBS mixtures evaluated using an ultrasensitive chemiluminescent analyzer. The production of ROS in the rat bladder exposed to substance P is also decreased after the administration of L-theanine. It was suggested that the scavenging properties of L-theanine can inhibit the pro-inflammatory response via the MPO (myeloperoxidase) activity of neutrophils recruited into the injured sites (Tsai et al. 2017). Protein levels of cleaved CASP3 and LC3 are increased in substance P-exposed urinary bladder tissue. By reducing the local ROS contents, cleaved CASP3 and LC3 levels are diminished after the administration of L-theanine. These data show that monitoring the ROS using CL approaches is an effective method to reveal the progression of cellular injuries during varied pathological conditions. Compared to conventional methodologies such as Western blotting or biochemical assays, CL offers a more convenient detection method with lower costs and higher sensitivity. The CL sensing methods for autophagy are indicated in Table 2.

Table 2.

Chemiluminescence assays for autophagy analysis.

Analytes	Organelles	CL system	Reference-year
Reactive oxygen species (ROS)	Kidney and Pancreas	Luminol/ROS	(Lin et al. 2016)
Reactive oxygen species (ROS)	Hyperactive Bladder	Luminol/ROS	(Tsai et al. 2017)

5.3. Electrochemical assays

Electrochemical recognition is being broadly investigated for the analysis of cancers and pathological conditions (Nasrollahpour et al. 2023a; Nasrollahpour et al. 2022; Nasrollahpour et al. 2021c; Singh et al. 2021). This method is one of the most prevalent modalities being studied and advanced due to its potential for cost-effectiveness, miniaturization, and portability (Yan et al. 2017). The most applicable example of electrochemical sensors is the glucometer, which is frequently used worldwide by diabetic patients (Baek et al. 2020). Three major layers including transducing, biorecognition elements, and analyte are the main parts of electrochemical biosensors (Singh et al. 2021). The design of the transducing surface and the format of employing biorecognition molecules define the performance of the fabricated biosensor. In this regard, many attempts have been made to optimize the previous and current methodologies (Hai et al. 2020; Menon et al. 2020). Advantages such as simpler instrumentation and ease of use make this biosensor superior to other types. Simple instrumentation such as an electrode (e.g., carbon electrode) and a potential source along with electrochemical analysis procedures such as amperometric and voltammetric techniques (e.g., pulse and cyclic voltammetry techniques) are used in electrochemical biosensors (Pashchenko et al. 2018). To date, different electrochemical modalities have been developed. Features such as transducing frameworks, molecular analysis, and configurations have been improved (Anderson et al. 2020; Rahn and Anand 2020). The integration of electrochemical sensing and nanotechnology is one of the greatest advances, leading to the extension of the domain to many applications and analytes (Chen et al. 2021; Wongkaew et al. 2018). The design of new frameworks using different nanomaterials provides new and valuable opportunities to analyze chemical and biological markers because of certain properties like increased specific surface area, functionality, and conductivity associated with the application of nano-scale structures (Nasrollahpour et al. 2023a). Wang et al. developed a multiplex biosensor for detecting BECN1 and LC3B-II using a sandwich-type electrochemical immunosensor (Figure 6) (Wang et al. 2017). To this end, AuNPs-rGO composites are deposited on the electrode surface to increase the conductivity and raise the signals. After attaching anti-BECN1 and anti-LC3B-II antibodies, the modified electrode is incubated with the relevant antigens. In this study, two redox signal probes (thionine/Thi, and 2, 3-diaminophenazine/DPA) loaded inside Au nanocages (AuNCs) are used as indicators for the BECN1 and LC3B-II proteins. Using differential pulse voltammetry/DPV, a high limit of detection of 0.02 and 0.03 ng/mL was obtained for BECN1 and LC3B-II, individually (Wang et al. 2017).

Figure 6.

Schematic illustration of the electrochemical immunosensor for autophagy monitoring. This figure was modified from Figure 1 originally published in (Wang et al. 2017) with permission of the publisher (Copyright 2017, Nature)

Aptasensors are another biosensing type that uses aptamers as bioreceptors instead of antibodies in immunosensors. for this purpose, synthesized DNA sequences are employed to monitor proteins or other small molecules that contribute to the autophagy mechanism (Yoon and Rossi 2018; Yoshikawa et al. 2021). Lima and co-workers used an electrochemical aptasensor to assess the interaction between APOE4 (apolipoprotein E variant 4) and CLEAR and SIRT1 (related to the autophagy response) during Alzheimer disease. In this study, a gold electrode was used as a working electrode to monitor APOE4 using a double-strand DNA sequence with SIRT1 and APOE4 attached to the gold electrode. By the promotion of an autophagy response, two nucleotide strands are attached to the APOE4 protein, making qualitative and quantitative calculations. This aptasensor also distinguishes the interaction between target nucleotide sequences with a specific APOE type (e.g., APOE4) from other isoforms (Lima et al. 2020). In addition to antibodies, which are the common bioreceptors in conventional methods like ELISA strategies, biosensors can recruit aptamers which have several advantages such as higher stability and enabling reusability of the developed frameworks (Jalili et al. 2022; Mirzaie et al. 2023). Table 3 presents a brief illustration of the sensing advances for electrochemical diagnosis of autophagy.

Table 3.

The currently developed electrochemical biosensors for autophagy monitoring.

Analytes	Sensing type	Designed framework	LOD (ng/mL)	Linear range (ng/mL)	Ref
apolipoprotein E isoforms	Aptasensor	AuE/p-DNA	-	-	(Lima et al. 2020)
BECN1 and LC3B-II	Immunosensor	AuE/AuNPs/rGO AuNCs/Thi or DAP	0.02 (for BECN1) 0.03 (for LC3B-II)	0.1–100	(Wang et al. 2017)

AuNCs: Au nanocages; AuE: Gold electrode

5.4. Plasmonic biosensors

Plasmonic sensing is well-known for excellent sensitivity, multiplexing capability, and successful commercialization. These features make plasmonic assays one of the most exciting methods in centralized and decentralized areas (Yang et al. 2021). Plasmonic biosensors are optical diagnostics based on the plasmon mechanism in which the free electrons in the conduction band of elements absorb the electromagnetic wavelengths and begin to oscillate (Balbinot et al. 2021; Mejía-Salazar and Oliveira Jr 2018; Szunerits et al. 2022). Free electrons absorb the wavelengths with the same frequency as their oscillation leading to a phenomenon termed surface plasmon resonance (SPR) (Doiron et al. 2019). Often gold and silver are used as metal elements with SPR properties (Cady et al. 2021). In plasmonic sensors, a metallic substrate is exposed to light that is refracted with a defined angle. Any changes on the metal surface and refraction angle are integral to the sensing indicator.

Nanomaterial-based plasmonic sensing is an extremely powerful strategy in designing plasmonic and plasmon-enhanced nano-biosensors (Cady et al. 2021). With the advent of nanomaterials, specific plasmonic sensing like localized surface plasmon resonance (LSPR) has been introduced as a nano-scale phenomenon (Ai et al. 2021). The LSPR produced on the surface of nanostructures generates a strong electric field in the vicinity of the nanomaterials (Agrawal et al. 2018). This phenomenon exhibits higher field amplification presenting a great application in signal enhancement methods. The wavelength used in the LSPR system is associated with the size and surface chemistry of nanomaterials. LSPR-based biosensors have also attracted tremendous attention in designing biosensors with extraordinary applicability (Lu et al. 2022; Zhan et al. 2018). Plasmon-enhanced fluorescence (Dong et al. 2019b; Huang et al. 2022), electrochemiluminescence (Akbari Nakhjavani et al. 2023; Kitte et al. 2021; Nasrollahpour et al. 2023b), electrochemical (Liang et al. 2022; Ribeiro et al. 2022), and scattering (Li 2018; Wang et al. 2020) biosensors are frequently utilized in the literature and represent great achievement in the detection of chemical and biological molecules. Due to the applicability and ability to miniaturize plasmonic biosensors, these tools can discover many aspects of autophagy. Previously, Choi et al. used an LSPR-based label-free nanobiosensor for autophagy monitoring using nanoplasmonic gold nanoparticles (AuNPs) (Choi et al. 2022b) (Figure 7). AuNPs are deposited on the transducing surface followed by the immobilization of anti-LC3 antibodies on the electrode surface. This platform was employed to monitor the conversion of LC3-I to LC3-II, which involves PEBP1 (phosphatidylethanolamine binding protein 1), a protein naturally produced in the body. Using this strategy, the total concentration of LC3 and also the conversion ratio was detected in the wide linear range of 10² to 10⁶ fM (Choi et al. 2022b). This example highlights one of the advantages of utilizing biosensors over traditional methods. Biosensing enables multiplex analysis, which addresses a limitation of conventional approaches.

Figure 7.

Schematic illustration of the LSPR biosensor for LC3 analysis. (A) Step-by-step preparation and detection. (B) The obtained signals represent a red shift for each two sensing steps. This figure was modified from Scheme 1 and Figure 1, originally published in (Choi et al. 2022b) with permission of the publisher. Copyright 2022, Elsevier.

In a study reported by Li et al., gold nanostars (AuNSTs) with an average diameter of 60 nm are used as plasmonic nanomaterials in a plasmon-enhanced Raman scattering system for monitoring pH changes in lysosomes during autophagy in HeLa cells (Li et al. 2019a). In this study, AuNSTs are modified with 4-mercaptopyridine/4-MPy (as a Raman probe) and bovine serum albumin to monitor pH changes in the range of 3 to 9 (Li et al. 2019a). Based on the obtained data, an 8-h starvation time leads to the formation of apparent red puncta associated with acidic vesicular organelles (autolysosomes) within the cytosol. In line with these changes, the mean lysosomal pH value reaches 4.45 ± 0.10, indicating the reduction of pH in lysosomes after fusion with autophagosomes whereas these values are 5.67 ± 0.14 in apoptotic HeLa cells. Considering the differences in lysosomal pH values, it is possible to type cell death in terms of cellular acidity. Plasmonic biosensing methods for autophagy are briefly presented in Table 4.

Table 4.

Plasmonic assays for autophagy analysis.

Method	Analytes	Framework	LOD	Linear range	Reference-year
LSPR	LC3 (I, II)	AuNRs	102 to 106 fM	64.61 fM	(Choi et al. 2022b)-2022
LSPR	LC3 (I, II)	AuNRs	102 to 107 fM	64.1 fM	(Choi et al. 2022a)-2022
SERS	pH	AuNSTs	-	-	(Li et al. 2019a)-2019

AuNSTs: gold nanostars; SERS: surface-enhanced Raman scattering; AuNRs: gold nanorods

6. Conclusions

Due to recent progress in the fabrication and development of varied biosensing techniques, real-time monitoring and sensitive detection of autophagy signaling pathways have become possible under physiological and pathological conditions. Regarding the complexity and sensitivity of biosensors, it is possible to discriminate the autophagic response from other cell death mechanisms. The integration of other scientific disciplines such as nanotechnology contributed to the extension of the domain to many applications and analytes. It seems that the application of biosensors is one of the interesting fields awaiting expanded application in the analysis of autophagy-related changes during pathological and physiological conditions.

Abbreviations:

ATG

autophagy related

AuNPs

gold nanoparticles

AuNSTs

gold nanostars

CL

chemiluminescence

CMA

chaperone-mediated autophagy

CTCs

circulating tumor cells

GFP

green fluorescent protein

HOCl

hypochlorite

LncRNA

long noncoding RNAs

LSPR

localized surface plasmon resonance

MIP

molecularly imprinted polymer

NC

nano-complex

PL

photoluminescence

PtdIns3K

phosphatidylinositol 3-kinase

RFP

red fluorescent protein

RhB

rhodamine B

ROS

reactive oxygen species

SPR

surface plasmon resonance