Aptasensor for the Detection of Mycobacterium tuberculosis in Sputum Utilising CFP10-ESAT6 Protein as a Selective Biomarker

A portable electrochemical aptamer-antibody based sandwich biosensor has been designed and successfully developed using an aptamer bioreceptor immobilized onto a screen-printed electrode surface for Mycobacterium tuberculosis (M. tuberculosis) detection in clinical sputum samples. In the sensing strategy, a CFP10-ESAT6 binding aptamer was immobilized onto a graphene/polyaniline (GP/PANI)-modified gold working electrode by covalent binding via glutaraldehyde linkage. Upon interaction with the CFP10-ESAT6 antigen target, the aptamer will capture the target where the nano-labelled Fe3O4/Au MNPs conjugated antibody is used to complete the sandwich format and enhance the signal produced from the aptamer–antigen interaction. Using this strategy, the detection of CFP10-ESAT6 antigen was conducted in the concentration range of 5 to 500 ng/mL. From the analysis, the detection limit was found to be 1.5 ng/mL, thereby demonstrating the efficiency of the aptamer as a bioreceptor. The specificity study was carried out using bovine serum albumin (BSA), MPT64, and human serum, and the result demonstrated good specificity that is 7% higher than the antibody–antigen interaction reported in a previous study. The fabricated aptasensor for M. tuberculosis analysis shows good reproducibility with an relative standard deviation (RSD) of 2.5%. Further analysis of M. tuberculosis in sputum samples have shown good correlation with the culture method with 100% specificity and sensitivity, thus making the aptasensor a promising candidate for M. tuberculosis detection considering its high specificity and sensitivity with clinical samples.


Introduction
Biosensors play a vital role in a plethora of applications, especially in the medical and food industrires, which comprise of the detection and quantification of proteins in clinical and food samples. Biosensors are described as the analytical devices used to detect the presence or quantify the concentration of a biological analyte by transducing the biochemical reactions of a bioreceptor with a specific target analyte into an optical, thermal or electrical signal [1]. Among different types of biosensors, electrochemical biosensors are of particular interest because of the lower limit of detection, simplicity, and easily miniaturized characteristics [2][3][4].
Bioreceptors are one of the key factors for explicit biosensor performance. Aptamers are single-stranded DNA or RNA sequences (most recently, peptides) that are generally less than hundred bases long [5]. Since their discovery in the 1990s, aptamers have received extensive interest for their application in biosensor development as an alternative to antibodies, which act as bioreceptors [6]. In general, the aptamers have to be isolated from the pools of random nucleic acid sequences by systematic evolution of ligands by an exponential enrichment (SELEX) process [7,8]. To specifically bind to their target, aptamers must fold into particular three-dimensional structures. Aptamers have been synthetically designed against a wide variety of targets, from small human molecules and viral proteins to whole microorganisms [9]. As compared to antibodies, aptamers are relatively easy to produce at a low cost, have high affinity towards their target and are generally stable [6,10]. On the other hand, aptamers can be synthesized in a large quantity with high reproducibility and purity [11]. A number of aptamer-based sensors have been developed using different transducer techniques and have demonstrated their excellent performance, which validates the promising potential of aptasensors [12][13][14].
Tuberculosis (TB) is one of the most common causes of death for those with human immunodeficiency virus (HIV), and this is due to the Mycobacterium tuberculosis (M. tuberculosis) pathogen [15]. This contagious disease is classified as airborne because it can be transmitted to other people when the infected person spews the bacteria into the air by coughing, sneezing, or even talking [16]. This pathogen's main route of infection is through the respiratory system and will further travel into the body towards the lungs. Other than the lungs, this bacterium can become active and spread to other parts of the body when the immune system becomes weak. Globally, the number of reported TB cases increased from 6.4 million in 2017 to 7.1 million in 2019 [17]. In Malaysia, the number of TB cases are high, with a current estimated incidence of 92 cases per 100,000 population during the year 2019 [17]. This is due to the high influx of illegal immigration, HIV, drug resistance, delayed diagnosis, high smoking rates, and diabetes [18][19][20][21]. Therefore, early and accurate detection of TB will be of great help to isolate patients and curb the spread of the disease.
The aptamer used in this study is specifically designed with a high affinity towards M. tuberculosis through the SELEX process. It is used as the bioreceptor immobilized on the working electrode surface to specifically detect various biomarkers of M. tuberculosis such as CFP10 [22], ESAT6 [23], CFP10-ESAT6 complex [9,13], Ag85A [24] and MPT64 [5,25]. In this study, CFP10-ESAT6 complex, an antigen secreted by M. tuberculosis, was chosen as a biomarker because of its better sensitivity compared to either CFP10 or ESAT6 antigen alone. Besides that, it is capable of avoiding false-positive results (better specificity) that usually happens due to a Bacillus Calmette-Guérin (BCG) vaccination [13]. Aptamer is widely known to be more selective than other commonly used antibody bioreceptors and this will increase the stability and robustness of the developed sensor.
Graphene (GP), in the form of a two-dimensional honeycomb, has a lot of useful properties such as a large surface area (2630 m 2 /g), good electrical conductivity (200,000 cm 2 /Vs) and good thermal conductivity of~5000 Wm/K [26]. It has been commonly used in biosensors due to its biocompatibility and non-toxic nature at optimal concentrations [27,28]. Meanwhile, polyaniline (PANI) is a conductive polymer that has been widely used because of its easy preparation process and low cost. In electrochemical sensors, PANI is reported to have fast electron transfer and excellent electrochemical activity [29]. Therefore, the combination of GP and PANI could enhance the conductivity on the surface of the working electrode in our proposed electrochemical biosensors, thus enhancing the signal transduction during biomolecule immobilization and detection [30]. This GP/PANI combination forms nanoparticle clusters on the working electrode's surface and hence offers a high surface area to volume ratio for biorecognition of molecules immobilized as compared to direct immobilization on the working electrode [31]. The optimum surface immobilization could also contribute to the high-performance detection of the disease.
Signal amplification via sandwich-type electrochemical antibodies or aptamer-based assays can be achieved by using labelling tracers such as enzymes and nanoparticles (NPs). This method is commonly used to obtain high detection sensitivity along with the improvements in the selectivity. This limits the crosstalk and interfering signal [32]. The various types of nanomaterials have been investigated as labels, particularly metal NPs due to their high stability compared to enzymatic signal tags [33,34]. Besides, they also have a high surface area to volume ratio that could anchor a large number of biomolecules, which then enables a much higher signal response [35].
In recent years, gold NPs (AuNPs) have been widely used as a label in biosensing technologies due to their advantages. Different detection schemes implementing AuNPs as labels have been reported, such as the direct electrooxidation of Au(III) and a two-step detection scheme. A two-step detection scheme involves the pre-oxidation of Au 0 to Au(III) by dissolution in acidic medium or HBr/Br 2 mixture followed by stripping voltammetry detection or the direct reduction of the released Au(III) to Au 0 . The determination of AuNPs labels using electrochemical oxidation in hydrochloric acid (HCl) is the most common method in electrochemical biosensors [34,36]. In this detection approach, the AuNPs undergo the electrochemical oxidation process in order to form AuCl 4 − , which is called the pre-oxidation step. Then, the AuCl 4 − formed is immediately reduced to metallic gold (AuCl 4 − + 3e − → Au 0 + 4Cl − ) and measured by voltammetric techniques, usually by differential pulse voltammetry (DPV) or square wave voltammetry (SWV). However, the acidity of HCl used needs to be evaluated since the use of highly acidic solutions contributes to the denaturation of biological material when used in decentralized settings [37]. Meanwhile, the performance of AuNPs could be enhanced by combination with magnetic NPs (MNPs). A work by Freitas et al. has shown that MNPs help to easily separate the target analyte from large and complex samples using an external magnetic field as well as enhance the reaction signal produced due to the high catalytic activity of MNP [38,39].
In this study, we proposed a sandwich-type electrochemical aptasensor for the detection of M. tuberculosis with modifications on the screen-printed gold electrode (SPGE) using iron/gold MNPs (Fe 3 O 4 /Au MNPs) as a label, while the GP/PANI nanocomposite serves as a signal amplification layer. The DPV technique was used to detect the CFP10-ESAT6 antigen. It was measured by the direct electrooxidation of gold in a neutral medium to form Au(III) ions. As compared with other sensors used for M. tuberculosis detection, this aptasensor is more specific in its choice of target.

Instrumentation
An Eco Chemie Autolab PGSTAT302 benchtop potentiostat with a NOVA 1.5 module (Metrohm, Utrecht, the Netherlands) was used for the electrochemical immunoassay analysis while a portable reader (DRP-DROPCAST, DropSens, Oviedo, Spain) was used for the on-site clinical sample analysis. A screen-printed gold electrode (SPGE) was purchased from DropSens, Spain. It consists of a working electrode (WE) and counter electrode (CE) that are made of gold ink, while the reference electrode is made of silver/silver chloride (Ag/AgCl). All of these electrodes were printed on a ceramic support. Other equipment used included circular dichroism (CD) (Jasco, Portland, OR, USA), X-ray diffraction (XRD) (X'Pert-MPD, PHILIPS, Amsterdam, Netherlands), high-resolution transmission electron microscopy (HR-TEM) (JEOL, Akishima, Japan), and field emission scanning electron microscopy (FESEM) (NOVA NANOSEM 230, FEI, Hillsboro, OR, USA).

Preparation of Fe 3 O 4 /Au MNPs-Labelled Antibodies
Iron oxide/gold magnetic nanoparticles (Fe 3 O 4 /Au MNPs) were firstly prepared according to the previous protocol [31]. About 10 mg of Fe 3 O 4 /Au MNPs was mixed with 2-mercaptoethanol and 12-mercaptododecanoic acid (ME-MDDA) (1 mM, 4:1) prior to conjugation with Ab and incubated in dark conditions for 24 h. The functionalized Fe 3 O 4 /Au MNPs was washed several times with ethanol and water before being dispersed in 0.1 M PBS at pH 7.4. Then, the functionalized Fe 3 O 4 /Au MNPs were incubated for 2 h with 25 µg/mL Ab followed by washing using PBS solution. Lastly, the bio-conjugated MNPs (Fe 3 O 4 /Au-Ab) were blocked from non-specific binding by incubation with 1% BSA (2 h) and washed to remove excess BSA several times using PBS solution.

Electrochemical Aptamer-Based Assay Detection Scheme
The development of aptasensor for CFP10-ESAT6 detection was conducted according to the previous study [9]; but with a different type of transducer. The WE surface of SPGE was electrochemically treated with 0.5 M H 2 SO 4 using a CV technique (potential range: 0.0-1.6 V, scan rate 100 mV/s) followed by deposition of 4 µL of 1 mg/mL GP/PANI onto the WE surface. The GP/PANI-modified SPGE was dried under room temperature overnight. Then, it was washed with ethanol to remove the unbound GP/PANI, followed by drying at 70 • C for 30 min. The capture aptamer (CapApt) immobilization process on the electrode surface was conducted through a cross-linking reaction using glutaraldehyde. The modified SPGE was immersed in 1% glutaraldehyde for 30 min at 4 • C. After the reaction, the modified electrode was rinsed with deionized water and dried at 37 • C. Then, 6 µL CapApt (20 µg/mL) was dropped onto the electrode surface, left to react for an hour and washed using deionized water. The surface of electrode was blocked with 0.25% BSA for 1 h at 37 • C to avoid non-specific binding. After washing with PBS washing buffer, the prepared electrode was incubated with 4 µL of CFP10-ESAT6 antigen solution with different concentrations at 37 • C for 1 h. Finally, the prepared Fe 3 O 4 /Au-Ab buffer solution (4 µL) was dropped onto the electrode surface. After incubation for 40 min followed by washing, the electrode was ready for the practical measurements.

Optimization of Experimental Conditions for the Bioreceptor
The concentration and incubation time of CapApt immobilized on the GP/PANImodified SPGE were optimized to obtain the maximum performance of the aptasensor.
The concentration of CapApt was tested in the range of 5 to 100 µg/mL with a constant concentration of CFP10-ESAT6 (20 ng/mL) immobilized on the electrode. The binding of CapApt and CFP10-ESAT6 was evaluated at various times ranging from 30 to 150 min. The binding was monitored by measuring the change in the oxidation current using DPV in 0.1 M PBS (pH 7.4) solution as described above.

Detection Study
The CapApt/GP/PANI-modified SPGE surface was incubated in different concentrations (5 to 500 ng/mL) of CFP10-ESAT6 for 1 h, rinsed twice with PBS to remove unspecific adsorbed CFP10-ESAT6 molecules and the response was recorded with DPV taken in 0.1 M PBS (pH 7.4) at the potential range of 0.0 to 1.0 V.

Specificity and Reproducibility Studies
For specificity studies, the sensor was incubated with 20 ng/mL CFP10-ESAT6, 2.5 mg/mL BSA (widely distinct protein), 20 ng/mL MPT64 (secreted by M. tuberculosis as well) and 10× dilution human serum for 1 h. After washing using PBS, Fe 3 O 4 /Au-Ab was dropped on the surface of the electrode and incubated for 40 min to complete the assay. The current response of different analytes was analysed using DPV by immersing the prepared electrodes in 0.1 M PBS (pH 7.4) at a potential ranging from 0.0 to 1.0 V. Meanwhile, for the reproducibility study, five individual electrodes were prepared for 20 ng/mL CFP10-ESAT6 antigen detection. After incubating the antigen for 1 h, the electrode was further incubated with Fe 3 O 4 /Au-Ab to complete the assay. The DPV was measured to evaluate the reproducibility performance of the developed sensor.

Real Sample (Sputum) Analysis
Hospital Universiti Sains Malaysia (HUSM), Kelantan, Malaysia supplied 10 samples collected from patients. The patients were diagnosed with TB using the standard method of smear microscopy and a culture test to validate our developed electrochemical aptasensor. A culture test is a confirmation test for TB detection in hospitals in Malaysia. The results from the culture method will be used as a reference, while the performance of smear microscopy will be compared with the developed electrochemical aptasensor. The analysis was carried out at the laboratory of Medical Microbiology and Parasitology, HUSM. The clinical samples (sputum) were diluted in 4% NaOH prior to detection [40]. Decontaminated samples were dropped on the modified SPGE and incubated for 1 h; for the testing, at least 4 µL of decontaminated sample is required. After that, 4 µL of Fe 3 O 4 /Au-Ab was dropped onto the electrode and incubated for another 40 min. Finally, the electrode was immersed in 0.1 M PBS (pH 7.4) and ready for the measurements using a portable reader. The data displayed a value in the unit of µA. This whole process is shown in Figure 1.

Aptamer-Antibody Sandwich Interaction
As described in the experimental section, the aptamer was covalently immobilized on the working electrode surface to act as a capture probe for CFP10-ESAT6 protein. The selection of aptamer is based on a previous study [9], which developed the sequence of an aptamer for a CFP10-ESAT6 antigen. The structure of the aptamer is shown in Figure 2a. Beforehand, CD characterization was performed to confirm the binding and determine the conformational changes of CFP10-ESAT6 antigen and its aptamer. The CD is usually used for molecules such as nucleic acids and proteins containing chiral atoms. It measures differential absorption of left and right polarized light by the analytes [41,42]. As shown in Figure 2b, CFP10-ESAT6 antigen alone has a negative peak at 218 nm. Significant changes in the antigen structure were observed upon addition of CapApt to CFP10-ESAT6 antigen, indicating that aptamer folding occurred due to binding interactions.  Figure 3a illustrates the different steps involved in the fabrication of a biosensor for the detection of CFP10-ESAT6 based on its specific bioreceptor by an aptamer-antibody assay. The covalent immobilization was conducted through a glutaraldehyde crosslinking reaction on the GP/PANI-modified SPGE surface. Figure 3b shows the field emission scanning electron microscope (FESEM) images of GP/PANI nanocomposite. The image of the GP/PANI nanocomposite showed that the GP sheets were mostly covered by PANI nanotubes and formed porous structures. This indicates that the presence of GP promoted the formation of agglomerated PANI nanofibers. The effect of the GP/PANI modification electrode surface has been elaborated in the Supplementary Materials (Section S2). Then, this step was followed by a blocking step with BSA to avoid non-specific adsorption on the transducer surface. Then, the Ab that was conjugated with Fe 3 O 4 /Au MNPs was deposited on the sensing platform to enhance the detection signal through sandwich type detection. Figure 3c shows the TEM images of the Fe 3 O 4 /Au MNPs that were used as nanoparticle labelling to increase the signal produced from the aptamer-antigen interaction. The average particles of Fe 3 O 4 /Au MNPs are around 35 nm, the molecular d-spacing was calculated from Image J software, which resulted in around 0.25 nm for the lighter part (iron core) and around 0.23 nm for the darker part (gold shell). The lattice distances measured for the shell correspond to the known Au lattice parameters for the (1 1 1) plane and those measured for the core match well the Fe 3 O 4 lattice parameters for the (3 1 1) plane [43]. Preparation and crystalline structure characterization of this material have been further evaluated in the Supplementary Materials (Section S1). Finally, the biosensor was immersed in a PBS buffer solution of pH 7.4 and CFP10-ESAT6 was determined by differential pulse voltammetry of the generated AuCl 4 − as the reaction product.

Optimization Studies of Aptasensor
Optimization steps are very important and necessary to construct a good, stable, and reproducible biosensor. For this purpose, capture probe deposition parameters such as concentration of CapApt as well as incubation time of CapApt were examined. These conditions need to be optimized, as they influence the binding of the capturing probe with CFP10-ESAT6 antigen and affect the peak current of DPV. Different concentrations of CapApt (5,20,40,60,80, and 100 µg/mL) were studied by drop casting them onto GP/PANI-modified SPGE for 2 h so that the optimum concentration of CapApt bonding on the sensor could be determined. As seen in Figure 4a, the peak current increased with an increase in CapApt concentration (5 to 20 µg/mL). Nevertheless, the peak current started to decrease as the concentration increased from 40 to 100 µg/mL. This might be due to the over saturation of CapApt on the surface of GP/PANI-modified SPGE, which made CapApt less effective at binding with target antigens. Hence, the concentration of CapApt that was chosen to be used in further study is 20 µg/mL instead of 5 µg/mL. Besides, 20 µg/mL CapApt concentration provides better reproducibility as compared to 5 µg/mL. Figure 4b shows the effect of CapApt incubation time on peak current. From the results, the peak current reached the maximum peak at 60 min. Upon extending the incubation time, the current slowly dropped, which might be due to the instability of CapApt left at 37 • C for a longer time on the electrode surface. In general, the incubation time of 60 min is sufficient to record the detection of CFP10-ESAT6 antigen.

Limit of Detection (LOD), Specificity, and Reproducibility
Different concentrations of CFP10-ESAT6 antigen were detected by utilizing a specific constructed aptasensor. The analytical performance of our modified electrode was studied by using one of the most sensitive electrochemical analysis techniques, known as DPV. Figure 5a shows the current response of aptasensor upon interaction with the Fe 3 O 4 /Au MNPs conjugated Ab used for the detection of CFP10-ESAT6. It was shown that, with an increase in CFP10-ESAT6 concentration, the current response of Fe 3 O 4 /Au MNPs increased. This indicates an elevated interaction between Ab and antigen as the amount of Fe 3 O 4 /Au MNPs conjugated Ab increases.
A linear relationship between the analytical responses, ∆I/I 0 % signal and the logarithmic values of CFP10-ESAT6 concentration was found within the concentration of CFP10-ESAT6 ranging from 5-500 ng/mL as shown in Figure 5b. ∆I/I 0 % indicating; I 0 = Peak current for blank (0 ng/mL CFP10-ESAT6), I = current response for certain concentration and ∆I = I -I 0 . The graph indicates an increase in sensor response signal with that of increased CFP10-ESAT6 concentration, which generated a linear positive slope. The calibration plot was fitted to a linear equation: where y is the ∆I/I 0 % signal, m is the gradient slope and c is the y-intercept. The aptasensor exhibited a linear regression, expressed as y = 10.54x + 0.53 with a correlation coefficient of R 2 = 0.9948. The LOD was obtained at 1.52 ng/mL of CFP10-ESAT6 concentration by using the 3σ/s calculation formula, where σ is the standard deviation of the blank and s is the slope of the calibration curve (Figure 5b). The specificity study was conducted to prove the biosensor response is specific between the CFP10-ESAT6 antigen and the capturing probes immobilized on the electrodes. The study was conducted using 20 ng/mL CFP10-ESAT6, 2.5 mg/mL BSA, 20 ng/mL MPT64 and 10x dilution of human serum. As portrayed in Figure 5c, the highest peak current for the sensor was shown by CFP10-ESAT6 antigen, while the other pathogens were detected at a low peak current. This is because the pathogens other than CFP10-ESAT6 were not captured by CapApt. The aptasensor showed that the current difference between CFP10-ESAT6 with other pathogens is about 15% or more. The reproducibility of the developed aptasensor was evaluated by means of relative standard deviation (RSD). Five independently modified electrodes were prepared for the detection of 20 ng/mL CFP10-ESAT6 antigen. From the bar chart in Figure 5d, the peak current obtained from 20 ng/mL CFP10-ESAT6 antigen detection on the five different modified aptasensor electrodes resulted in a current value in the range of 30-31 µA with an RSD of 2.5%. The consistency of signals recorded from the modified electrodes proved the reliability of the modification and stability of the detection process, thus showing its potential for mass production.
The analytical performance of the developed sensor for M. tuberculosis detection with other detection methods reported in the literature, together with the results, is shown in Table 1. Most of the detection was based on optical transduction techniques such as enzyme-linked immunosorbent assay (ELISA) [44,45], optical waveguide [46], fluorescence resonance energy transfer (FRET) [47], calorimetric magnetophoretic assay [48], and electrochemical sensor [49]. All of them except ELISA recorded a lower LOD when compared with our developed sensor, while our previous work recorded almost the same LOD. However, when compared with this work, the other methods still require multi-step amplification, are labour-intensive and time-consuming, and use heavy equipment, which obviously limit their use for point-of-care testing (POCT) for an extremely resource-limited environment. To the best of our knowledge, the electrochemical aptasensor that we developed in this work is capable of the direct detection of M. tuberculosis in sputum samples and can be implemented on POCT. Even though the LOD of our developed sensors was higher than several studies reported in Table 1, the LOD still falls in the physiological range of interest for M. tuberculosis, which is around 100 ng/mL in TB patients [20]. Besides that, our developed sensor can be easily miniaturized and transported for on-site detection during clinical application.

Clinical Sputum Samples Detection
A total of 10 sputum samples were analysed for clinical sample analysis. As the gold standard, results from the smear microscopy and culture method were used for comparison with our aptasensor reader. Each sample was identified as positive (P) or negative (N) according to the culture method (reference method). The culture method takes around 2-3 weeks to obtain the results, while smear microscopy takes around 15 min. The detection time using our portable reader is 2 h. From the reference method, it shows that there are six patients (out of 10) who were diagnosed positive and four were negative. From our portable reader, the results obtained were simplified in bar chart form as shown in Figure 6, where all of the positive samples show positive ∆I/I 0 % currents while the negative samples show negative ∆I/I 0 % currents.
Despite qualitative detection, this sandwich-type aptasensor is also able to quantify the concentration of CFP10-ESAT6 antigen in sputum samples, as shown in Table 2. By inserting the value of ∆I/I 0 % in a linear equation: y = 10.54 log x + 0.53, the concentration of CFP10-ESAT6 antigen were able to be determined.
The sensitivity and specificity of the smear microscopy method and our portable reader were compared by using the culture method as a reference and is shown in Table 3. The results obtained for the sensitivity and specificity of the smear microscopy and aptasensor reader was identical, which are 100%. From these findings, it was supported that our developed aptasensor is suitable for the detection of CFP10-ESAT6 antigen in sputum samples.

Conclusions
In this study, a sandwich type electrochemical aptasensor was designed for the detection of the M. tuberculosis antigen biomarker, CFP10-ESAT6. The SPGE was modified with GP/PANI for the immobilization of the capturing probe (CapApt). Meanwhile, Fe 3 O 4 /Au MNPs were used as a label to amplify the signal generation. The Fe 3 O 4 /Au MNPs were conjugated with primary antibodies to complete the sandwich format. The analytical performances of both sensors were successfully conducted by direct electrooxidation of AuNPs in PBS solution to form AuCl 4 − due to the presence of chloride ions measured by DPV. It was found that the signal generated increased as the concentration of CFP10-ESAT6 antigen increased. The developed aptasensor was evaluated through the detection of tuberculosis in sputum samples and demonstrated 100% sensitivity and specificity. This suggests promising performance for the aptasensor and the commercialization of the aptasensor for TB and other diseases' detection should be considered for early and effective illness prevention.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.