Introduction

Exosomes (Exos) are double phospholipid membrane extracellular vesicles (EVs) with an average size of 50 to 150 nm [1, 2]. Compared to other EV types such as apoptotic bodies and microvesicles, Exos are produced by the activity of the endosomal system using different molecular types of machinery such as ESCRT, tetraspanins, and SNARE factors [3]. Several investigations have revealed that EVs are rich in protein, miRNA, and DNA and play an essential role in cell-to-cell communication. These nanosized particles can reflect real-time changes in the metabolic status of parent origin cells, indicating EVs as a valid diagnostic tool [1, 4]. Besides their diagnostic roles, EVs can be involved in the promotion of cancer cell proliferation, angiogenesis, and metastasis [5,6,7]. EVs released by cancer cells can easily distribute in several biofluids such as blood and thus could be used for monitoring and diagnosis the several pathologies especially anaplastic changes [8]. According to published data, identifying prostate cancer antigens in the early stages of development can increase the probability of a treatment intervention and increase survival rates from 10 to 90% [9].

Recently, some methods such as nanoparticle tracking analysis (NTA) [10], flow cytometry [11], western blotting [12], and ELISA [13] have been used for the detection of EVs or related proteins. NTA can be used when the concentration of EVs is between 107 and 109 particles mL−1, but when the concentration is lower than that, quantification is laborious. NTA is mainly used for particle size and potential distribution detection. Flow cytometry can be applied for high-throughput detection, but light scattering from EVs with particle sizes < 100 nm is weak and cannot be accurately quantified. ELISA and western blot analyses require a large sample volume and the sensitivity of both of them is limited.

Since EV particles are limited in peripheral blood in the early stages of cancers, early and precise detection is difficult due to the low sensitivity of traditional detection methods [14]. Therefore, there is an urgent need for accurate diagnostic approaches with high sensitivity and selectivity for the early detection of EVs [15]. Recently, biosensors, including recognition elements and energy transducers, have received great interest in the detection of biomolecules, especially EVs [16]. After the detection of target molecules, biosensors can generate signals that can be read by alternative signal transduction elements. Signal transduction elements mainly include colorimetry [17], surface plasmon resonance (SPR) [18], surface-enhanced Raman scattering (SERS) [19], fluorescence [20], electrochemiluminescence (ECL) [21], field-effect transistors (FETs) [14, 22] as gate-controlled structures with visible light [23], and electrochemistry [24, 25]. The electrochemical biosensor enables the conversion of identification information of biological molecules into electrochemical signals of current, potential, and impedance [26]. Furthermore, electrochemical biosensors have been recognized as a vital tool due to their inherent advantages such as rapid analysis, cost/time effective, easy operation, portability, miniaturization, selectivity, and high sensitivity [27,28,29].

Nanomaterials have specific features like reactive surface topography, better mechanical features, quantum effects, superior sensitivity, high surface area, and catalytic activities [30, 31]. These properties make them a more desirable recourse for biosensing approaches compared to the other available materials [32]. Over the past few decades, perovskite oxides, characterized by their cubic-symmetry structure represented by the chemical formula ABO3 (where A: rare-earth or alkaline earth cations, B: a transition cation), have gained considerable significance as a crucial group of catalysts [33,34,35]. A key feature of perovskites is their capacity for partial or/and total substitution of different cations at the A and B sites, resulting in a versatile and adaptable structure (denoted as A1−xA’xB1−yB’yO3), which exhibits tunable physical–chemical properties [36,37,38,39]. These unique characteristics have led to a significant focus on perovskite oxides in various applications, including their utilization as catalysts and electrocatalysts in processes such as De-NOx and dry reforming [40], supercapacitors [41], biosensors [42], water splitting [43,44,45], fuel cells [46], and batteries [47].

However, one notable drawback of perovskite oxides is their inherent low electron conductivity, primarily attributed to the high calcination temperatures employed during their synthesis [48]. On the other hand, noble metals such as palladium and platinum have garnered significant interest due to their exceptional conductivity, facilitating efficient electron transport [49]. To overcome this limitation, partial substitution of noble metals in small concentrations at the B site of perovskites has known a promising solution.

Herein, we established a label-free electrochemical Exosensor using Pt-perovskite for the quantification of EVs derived from prostate cancer cells. Also, this proposed method efficiently discriminated between normal and cancer EV prostate cancer patient sera compared to control samples. Overall, the developed Exosensor shows a potential value for the detection of EVs in the screening and early diagnosis of cancers.

Experimental section

Chemicals and materials

La(NO3)3·6H2O, Fe(NO3)3·9H2O, Mn(NO3)2·4H2O, Pt(NH3)4·2H2O, glycine, NH3·H2O, and 6-mercaptohexanol (MCH) were purchased from Sigma-Aldrich. Biotinylated human monoclonal CD63 antibody (Clone: H5C6, Cat: 353,018) and purified streptavidin protein were purchased from BioLegend Company. Anti-PSA was purchased from Abcam (ab76113). H2SO4, KH2PO4, Na2HPO4, KCl, K4[Fe(CN)6], and K₃[Fe(CN)₆] were obtained from Merck company.

Apparatus

All electrochemical measurements were carried out with a conventional three-electrode system on RadStat-10 potentiostat/galvanostat electrochemical analyzer (Kianshardanesh, Iran) including a reference electrode (Ag/AgCl), a counter electrode (platinum wire), and a working electrode (glassy carbon with 2-mm diameter). An ultrasonic device (Strasonic 35) and a magnetic stirrer (Heidolph) were applied for the homogenization of the prepared solutions. A centrifuge (Beckman) was used to separate EVs from the serum samples.

Composite characterization

The scanning electron microscopy (SEM) images and semi-quantitative amounts of the elements were acquired by a Quanta FEG 450 instrument. Transmission electron microscopy (TEM) was carried out on the Hitachi HighTech HT7700 to complete the structural analysis. The X-ray diffraction (XRD) pattern was obtained using a Tongda TD-3700 X-ray diffractometer from China. Cu-Kα radiation with a wavelength (λ) of 1.5406 Å was used, and the scanning range spanned from 10 to 90°. Also, X-ray photoelectron spectroscopy (XPS) measurements were carried out using a Thermo Scientific K-Alpha instrument, with Al Kα emission as the excitation source for elemental analysis and characterization.

Synthesis of La(Fe0.7Mn0.3)0.98Pt0.02O3

To synthesize La(Fe0.7Mn0.3)0.98Pt0.02O3 perovskite oxide, the combustion sol–gel method was employed, as outlined in reference [50]. First, the calculated amounts of nitrates were dissolved in deionized water and heated to 80 °C. Subsequently, glycine was added to the solution, maintaining the temperature at 80 °C. Upon gel formation, the mixture underwent combustion, resulting in the formation of a black powder. The obtained powder was calcinated at 700 °C for 5 h.

Fabrication of Pt-perovskite-GCE Exosensor

To prepare a suspension of perovskite, 0.03 g of Pt-perovskite powder was ground and dispersed in 10 mL of distilled water. Then the obtained suspension was placed in an ultrasonic device for 2 h to disperse the nanocomposite well. Next, 2 mL of dispersed suspension and 8 mL of 0.1 M KCl were mixed and used for the electrodeposition of platinum perovskite through the chronoamperometric technique. In this regard, 10 mL of the obtained solution was transferred to the electrochemical cell. The deposition process was performed using the chronoamperometric (CHA) technique (E = − 2.4 V, t = 45 s). Finally, the glassy carbon electrode modified with Pt-perovskite was applied for the next steps of analysis (Pt-per-GCE).

Surface functionalization

Biotinylated monoclonal CD63 antibody and streptavidin were incubated under optimal conditions according to previously published data [51]. In brief, to modify the biotinylated monoclonal CD63 antibody on the surface of Pt-per-GCE, the streptavidin (STRP) protein (1 µg mL−1) was first incubated on the Pt-per-GCE at 4 °C for 90 min. STRP was used to increase the loading capacity of anti-CD63 immobilization. The STRP, due to its unique molecular structure, could provide more binding active sites, where the sensitivity of the developed Exosensor could increase four times [52]. Next, anti-CD63 (1 µg mL−1) was introduced on the STRP-Pt-per-GCE for incubation at 4 °C for 90 min. Finally, 1 mg mL−1 MCH was applied to prevent nonspecific adsorption.

Cell culture and EV isolation

To calibrate the designed Exosensor, prostate cancer cell EVs were isolated, purified, and introduced to the Exosensor platform. Human prostate cancer DU145 cells were purchased from Iranian Cell Bank Pasture and cultured in DMEM/HG medium supplemented with 10% fetal bovine serum (FBS; Gibco) and 1% penicillin–streptomycin. Upon 70–80% confluence, cells were washed with pre-warmed PBS and incubated with EV-free FBS (Gibco). After 48-h incubation time, supernatants were collected and centrifuged at 400, 2000, and 10,000 g for 5, 10, and 20 min respectively to eliminate cells, debris, and apoptotic bodies. To get the EV pellet, samples were centrifuged at 100,000 g for 1 h. Finally, the morphology and zeta potential, and the number of EVs were approximately determined using SEM images and DLS analysis.

Result and discussion

Structure and morphology

Upon comparing the XRD pattern of the synthesized La(Fe0.7Mn0.3)0.98Pt0.02O3 catalyst (Figure S1) to the standard pattern of LaFeO3 (01–075-0439), we observed that the synthesized nanocomposite exhibits an orthorhombic structure, similar to that of LaFeO3. However, a slight shift towards higher angles (to the right) is observed in La(Fe0.7Mn0.3)0.98Pt0.02O3 structure. This shift may be attributed to the incorporation of Mn, which has a different ionic radius and consequently alters the unit cell size. Remarkably, the addition of Pt did not result in the emergence of distinct peaks, indicating that the noble metals do not significantly modify the perovskite host lattice [52,53,54]. The morphology of the prepared perovskite was examined using SEM and TEM (Fig. 1). SEM analysis revealed the agglomerated and non-uniform spherical nanoparticles. Furthermore, TEM images provided additional insights, showing that the nanoparticles have an average crystallite diameter of 40 nm. The surface elemental composition and oxidation states of La(Fe0.7Mn0.3)0.98Pt0.02O3 nanoparticles were analyzed using XPS (Fig. 2).

Fig. 1
figure 1

SEM (a-1, a-2) and TEM (b-1, b-2) images of La(Fe0.7Mn0.3)0.98Pt0.02O3

Full size image
Fig. 2
figure 2

XPS survey and elemental spectra of La(Fe0.7Mn0.3)0.98Pt0.02O3

Full size image

Figure 2A displays the XPS spectra of La(Fe0.7Mn0.3)0.98Pt0.02O3, indicating the presence of all elements within the perovskite structure. Figure 2B shows the XPS spectra of La 3d, with two doublet peaks corresponding to La 3d3/2 and 3d5/2. The peaks at 834.48 and 837.98 eV peaks correspond to La 3d5/2, while peaks around 851.28 and 854.68 eV are attributed to La 3d3/2. The presence of peaks at 854.78 and 834.48 eV confirms the La3+ oxidation state. The other peaks may be due to charge transfer from neighboring La atoms to the vacant 4f subshell during ionization [54]. The Fe 2p spectra show two distinct peaks at binding energies of 712.413 eV and 726.18 eV, corresponding to the 2p3/2 and 2p1/2 levels, respectively. These peaks exhibited typical shake-up structures at 717.98 eV, indicative of ferric ions in oxide forms [55]. The Mn 2p XPS spectra (Fig. 2D) exhibit single peaks at 642.15 eV (Mn 2p3/2) and 652.66 eV (Mn 2p1/2), indicating the coexistence of Mn4+ and Mn3+ in the perovskite structure (Fig. 2C) [56]. In the Pt 4f spectra (Fig. 2E), the binding energies around 71.98 eV and 74 eV correspond to Pt 4f7/2 and Pt 4f5/2, indicating the presence of adsorbed oxygen (PtOads) and oxidized Pt2+ [57]. The XPS spectra of O1s (Fig. 2F) display two asymmetric peaks at 528.88 eV and 531.18 eV, suggesting the presence of lattice oxygen and chemisorbed oxygen species. The binding energy of chemisorbed oxygen species (O−/OH−) is generally higher by 2.1–2.5 eV than that of lattice oxygen [58].

Optimization of electrodeposition potential

To improve the performance and sensitivity of the Exosensor and the effective surface of the electrode for loading the antibody, the perovskite density was optimized on the surface of the electrode using the chronoamperometric technique. For this purpose, the potential was optimized and the results were obtained using differential pulse voltammetry (DPV) in 5 mM [Fe(CN)6]3−/4/0.1 M KCl solution (pH 7.4) as an electrochemical redox system. In this regard, the range of potentials was selected from − 2.7 to − 2.2 V to reduce and deposit Pt-perovskite on the electrode surface [59]. Figure S2 shows the current generated from each of the potentials as well as the related histogram. Based on the results, the potential of − 2.4 V was chosen as the optimal potential for Pt-per electrodeposition on the electrode surface. Data confirmed that platinum, the active site of the composite, is lost when sweeping the potential towards positive potentials. These features cause a decrease in the current and consequently the performance in this analytical approach [60].

Optimization of electrodeposition time

For enhancing the efficacy of the designed Exosensor, the time of the electrodeposition of Pt-based perovskite nanocomposite was optimized using the chronoamperometric technique. In this regard, time intervals of 15 to 360 s were selected for optimization. Based on the obtained results, the maximum electric current was recorded in 45 s, which shows that there is enough time for nanocomposite deposition on the electrode surface. According to the data (Figure S3), with increasing time of chronoamperometry, the corresponding current gradually decreases because of excessive accumulation and saturation of the electrode surface with nanocomposite. Therefore, 45 s was selected as the optimal time for Pt-perovskite electrodeposition.

EV detection (electrode preparation steps)

The GCE was modified as mentioned. EVs at an appropriate concentration (V = 15 µL ~ 6000 EVs/µL) were incubated on STRP-Pt-per-GCE functionalized biotinylated monoclonal CD63 antibody at 4 °C for 60 min. Then the prepared Exosensor was placed in the electrochemical cell containing 5 mM [Fe(CN)6]3−/4−/0.1 M KCl solution (pH 7.4) to monitor the EV number (Fig. 3).

Fig. 3
figure 3

Electrode preparation steps: DPV signals (A) and its related histogram (B). SWV signals (C) and related histogram (D). All experiments were performed in 5 mM [Fe(CN)6]3−/4−/0.1 M KCl (pH 7.4) solution

Full size image

Electrochemical detection of EVs (calibration curve)

The developed Exosensor was applied for the quantification of different concentrations of EVs under the optimized conditions. For this purpose, semi-quantitative analysis was carried out on the EVs that were previously isolated with approximate numbers in certain volumes. Thus, different volumes were properly immobilized on the designed Exosensor and their related electrical signals were recorded by DPV technique in 5 mM [Fe(CN)6]3−/4−/0.1 M KCl solution (pH 7.4). As shown in Fig. 4, a linear relationship between the recorded electrical signal and the number of EVs with an equation of Y = − 1.7668x + 32.843 (R2 = 0.9789) was obtained, and the lower limit of quantitation (LLOQ) is 2000 EVs/μL. Based on the results of a calibration curve, the linear dynamic range (LDR) was reported to be 2000 to 14,000 EVs/μL. In Fig. 4, panels A and B show the related voltammograms and the average electric current versus the number of EVs.

Fig. 4
figure 4

A DPV signal readouts and (B) calibration curve for different numbers of the EVs. Error bars indicate the standard deviation of ten independent experiments

Full size image

Extracellular vesicle detection in clinical samples

Here, an Exosensor was designed to monitor prostate cancer EV PSA and data were compared to conventional ELISA assay (Fig. 5). ELISA indicated that prostate cancer marker PSA levels were high in prostate cancer patients compared to the healthy control individual. The difference in PSA levels in cancer patients can correlate with the tumor size and progression of anaplastic changes [9]. Our laboratory data indicated that systemic levels of PSA increase in patients with the relevant prostate cancer manifestation. To examine whether prostate cancer EVs harbor specific PSA and investigate the close relationship between the blood content and exosomal levels, the protein contents of PSA were measured using the Exosensor. Based on the data, the electrical current of PSA-modified electrodes was recorded as the same as the calibration curve. We noted statistically significant differences regarding the electrical current between EV PSA in prostate cancer patients and healthy control samples (p < 0.0001; Fig. 5). It is postulated that attachment of anti-PSA antibody to EVs yielded lower electrical current (µA) values in prostate cancer patients. Of note, the reduction of µA was prominent in samples with higher levels of PSA measured by ELISA. To be specific, in prostate cancer patient 3, the levels of systemic PSA were 35.9 U/mL with 3.9 µA indicated by Exosensor. Data indicated that by increasing the levels of Exosomal PSA, the intensity of the electrical current is reduced in which EVs of prostate cancer patient 3 had the lowest electrical current (µA) value compared to EVs of prostate cancer patients 1 and 2. These data show that the levels of exosomal PSA are increased with the progression of anaplastic changes in prostate cancer. Like blood samples, EVs can appropriately reflect real-time changes in prostate cancer patients. The main features of the most recent electrochemical biosensors, which diagnose various illnesses by analyzing linked EVs, are outlined in Table 1.

Fig. 5
figure 5

Measuring the levels of PSA using ELISA (A) and Exosensor (B). ELISA indicates the increase of PSA levels in prostate cancer patients compared to the healthy control. Designed Exosensor is eligible to detect different contents of exosomal PSA in both control and cancer patient samples. Data indicate the fluctuation in terms of PSA in patients with different prostate cancer stages. By the progression of prostate cancer, the recorded electrical signals are correspondingly reduced, showing the sensitivity of designed Exosensor in the validation of PSA in biological samples. One-way ANOVA and Tukey post hoc analysis. ****p < 0.0001. Each sample was read in 10 technical replicates

Full size image
Table 1 Comparison of different electrochemical biosensors for evaluation of EVs
Full size table

By generating geno/apta/immuno/sensors for evaluation of EVs, various nanomaterials were used to study different cancers. The performance of noted electrochemical biosensors is comparable in a number of ways: (I) the nanomaterial used in electrode modification should be biocompatible and cost-effective [44, 45, 64, 65]; (II) analytical performances such as linear dynamic range, sensitivity, and selectivity should have the highest levels; and (III) the working electrodes and techniques used must be suitable for theragnostic purposes. Table 1 shows that our developed biosensor is the most sensitive approach among previously reported works. Our research work has desirable lower limit of quantification (LLOQ) among others, which appears to be due to the presence of perovskite nanostructure and platinum nanoparticles. This nanocomposite as modifier increases the effective surface area and catalyzes the electrochemical reaction on the electrode, which in consequently enhances the sensitivity in detecting target EVs. The results of our work show that the modified electrodes have desirable stability because the modifiers are properly attached on the electrode surface. The techniques that are most common are DPV and SWV, which are used in the fabrication of electrochemical biosensors due to their low detection limits and high sensitivities.

Real sample collection

EVs were isolated from blood sample remnants of prostate cancer patients who referred to Shahid Ghazi Hospital affiliated to Tabriz University of Medical Sciences. EVs were isolated using similar protocols described for the isolation of EVs from cell culture samples. All phases of this study were approved by local ethics committee of Tabriz University of Medical Sciences and steps were in accordance with previously published guidelines of the Declaration of Helsinki.

Statistical analysis

In this study, data are expressed as mean ± SD. To compare statistical differences between the groups, one-way ANOVA with post hoc analysis was used. p < 0.05 was considered statistically significant.

Conclusion

In this research, an electrochemical Exosensor was successfully designed for prostate cancer derived-extracellular vesicles analysis. For this purpose, platinum-based perovskite nanostructures were synthesized as modifying nanomaterials. The advantages of these nanostructures can be mentioned as high conductivity and surface area, biocompatibility, and cost-effectiveness. The existence of the noble platinum metal as an active site provides the possibility of better binding of streptavidin and then antibody on the surface of the electrode. Based on this, simultaneously the stability and the sensitivity of the prepared Exosensor improved. The detection method is based on the specific binding of biotinylated monoclonal anti-CD63 and anti-PSA to EVs through the sandwiching method to determine prostate cancer EVs. Among the applications of this designed Exosensor, we can point out the differentiation between normal and tumorous EVs, which leads to the evaluation of the early stages of PSA. Also, this developed method can be used as a POC test in clinical laboratories.