Digital Immunoassay for Biomarker Detection Based on Single-Particle Laser Ablation ICP MS

This study presents a digital dot-blot particle-linked immunosorbent assay (PLISA) that uses laser ablation single-particle ICP-MS (LA SP ICP MS) to count upconversion nanoparticle (UCNP)-labeled immunocomplexes. The method offers precise nanoparticle desorption and counting, minimizing noise and aggregates through digital signal processing. It was optimized for human serum albumin and prostate-specific antigen (PSA), key biomarkers for kidney and prostate conditions.

The digital LA SP ICP MS readout achieved sub-pg/mL detection limits, outperforming both upconversion luminescence (UCL) and analog LA ICP MS. Applied to clinical serum samples for PSA, the results strongly correlated with electrochemiluminescence reference values. This approach highlights the power of digital particle counting for ultrasensitive biomarker detection in clinical diagnostics.

The original article

Digital Immunoassay for Biomarker Detection Based on Single-Particle Laser Ablation ICP MS

Vilém Svojanovský, Jakub Máčala, Antonín Hlaváček, Aleš Čermák, Jaromír Stráník, Pavel Bouchal, Ivana Mašlaňová, Petr Skládal, Zdeněk Farka, and Jan Preisler

Anal. Chem. 2025, 97, 26, 13832–13839

https://doi.org/10.1021/acs.analchem.5c00641

licensed under CC-BY 4.0

Selected sections from the article follow. Formats and hyperlinks were adapted from the original.

Immunoassays are routinely used bioanalytical techniques, combining antibodies as biorecognition elements capable of specific binding to target antigens with various labels for signal generation. (1) Heterogeneous immunoassays, representing the most common format, are performed on the surface of a solid phase, which allows for washing out of the unbound reagents, leading to a decrease of the background signals caused by nonspecific binding. Consequently, background reduction allows for achieving lower LODs. (2)

One of the heterogeneous immunoassay types is dot-blot, representing a simple assay format performed on the membrane surface, where the capture immunoreagents are adsorbed by noncovalent interactions. (3) Dot-blot immunoassays were successfully used to detect numerous analytes, including biomarkers, (4) infectious agents, (5) and organic pollutants; (6) however, the detection is mostly only semiquantitative. (7)

Several types of labels have been frequently used in dot-blot immunoassays. Enzymes (8) and fluorophores (9) are commonly employed for signal generation due to their effectiveness; however, they come with certain drawbacks, such as limited stability of enzymes and photobleaching of molecular fluorophores. To address these drawbacks, various nanoparticle (NP)-based labels have emerged in recent decades, offering enhanced stability and signal intensity, with some even enabling the quantitative analyte detection in dot-blots. (4,10) For example, gold NPs (AuNPs) were used as colorimetric labels, (11) and quantum dots as a replacement for organic fluorophores. (12) Photon upconversion nanoparticles (UCNP) represent another class of nanomaterials recently used as a label in dot-blot immunoassays. (4) UCNPs are nanocrystals composed of inorganic matrix (e.g., NaYF₄) doped with lanthanide ions (e.g., Yb³⁺, Er³⁺, Tm³⁺). Their main advantage is connected with the anti-Stokes emission of light of shorter wavelength under the excitation with light of longer wavelength, usually infrared. (13) Therefore, using UCNP labels reduces background signal, as fluorophores naturally present in biological samples are not excited by infrared radiation. (14) Thus, immunoassays utilizing UCNP labels can achieve lower LODs, making them suitable for ultrasensitive quantitative analysis of various biomarkers. (15) Using optical microscopy, even individual UCNPs can be detected, enabling the development of single-molecule immunoassays. (14)

Detection of various types of NP labels can be combined with different readout methods, e.g., absorbance and fluorescence or surface-enhanced Raman spectroscopy. (16) Another option for the sensitive detection of various NP-based labels containing suitable metals, is laser ablation inductively coupled plasma mass spectrometry (LA ICP MS). Simple chelate labels, which carry only a single metal atom, are commonly used but provide limited sensitivity. (17) To amplify the signal, polymer chelates like Maxpar, binding around 30 metal atoms, have been developed. (18) However, the most effective signal amplification comes from metal NPs, which offer the highest number of metal atoms per label size. (19) Although the incorporation of metal NP resulted in enhanced signals, conventional LA lacked the capability of individual label detection.

Nanomaterials of various compositions and shapes can be analyzed by a single-particle ICP MS (SP ICP MS). This extension of the established ICP MS technique enables the detection of individual NPs, determining their elemental composition, concentration, and size if the shape is known. (20) After introducing an NP into the plasma, it is atomized and ionized to form an ion cloud, which is recorded as a distinct peak upon reaching the detector. (21) In the overwhelming majority of cases, NPs are introduced into the plasma by nebulization from a dispersion. (22,23) The use of laser ablation (LA) for introducing intact NPs was reported in 2016; AuNPs (56 and 86 nm in diameter) were introduced directly into the ICP MS from the polyethylene terephthalate glycol that mediated the desorption and prevented NP disintegration, achieving a detection efficiency of 61%. (24) Maintaining the UV laser fluence below 0.15 J/cm² was crucial to ensure the desorption of intact AuNPs, as commonly used fluences exceeding 1 J·cm^–2 would lead to NP disintegration during ablation. In 2022, utilizing an IR laser (2940 nm) allowed the efficient release of intact NPs from biological tissue due to the large difference in absorption of the tissue and NPs. More than 80% of 20 nm AuNPs were detected from labeled cancerous biological tissue, enabling digital mapping of biomarkers in colorectal carcinoma cells. (25)

In this work, we introduce a novel dot-blot, particle-linked immunosorbent assay (PLISA) method utilizing LA SP ICP MS. The single-particle readout allows the detection of single immunocomplexes labeled with UCNP-streptavidin conjugates (UCNP-SA). Individual UCNPs are introduced into the plasma torch and counted (Figure 1). The potential of PLISA was tested by the detection of human serum albumin (HSA), a biomarker of kidney damage when found in urine, and prostate-specific antigen (PSA), a key serum biomarker of prostate cancer. The performance of UCNP counting was compared to conventional analog signal recording of ICP MS and upconversion luminescence (UCL) scanning as a reference method, demonstrating the potential of LA SP ICP MS for ultrasensitive analysis. Finally, the clinical applicability of PLISA was confirmed by the detection of PSA in serum samples of patients tested for prostate cancer.

Experimental Section

LA SP ICP MS

The 2940 nm laser beam was guided from the source into a 4× Galilean beam expander composed of an anti-reflective (AR)-coated CaF2 plano-convex lens with a 12.7 mm diameter and −25.0 mm focal length and an AR-coated CaF2 plano-convex lens with 25.4 mm diameter and 100.0 mm focal length. Afterward, the beam was reflected by 90° using a silver mirror in an optical cube, followed by reduction by an iris to a top-hat near profile (all from Thorlabs, USA). The beam was focused on the sample using an AR-coated aspheric plano-convex lens with 8.00 mm diameter and 5.95 mm focal length (Edmund Optics, USA) to a diameter of 22 μm and a fluence of 11.4 J/cm2.

The ablation cell was operated in two configurations: standard and fast washout. In the standard configuration, the cell dimensions above the slide were 76.1 mm × 26.1 mm × 3.0 mm. Helium served as the carrier gas at a flow rate of 1.0 L/min, with an additional 1.0 L/min of argon introduced after the cell exit. The fast washout configuration featured a rectangular main channel above the slide, measuring 76.1 mm × 4.7 mm × 3.0 mm. In this setup, helium flowed at 1.6 L/min and 0.4 L/min of argon. For maximum transport efficiency, all laser ablations were performed on the half of the glass slide positioned closer to the outlet of the ablation cell.8 To ablate the samples from the second half, the slide was rotated 180° to maintain LA conditions as consistent as possible.

The laser spot position was set by the XY motion of the cell with 8MT-175-150 and 8MT-175-100 stages (both from Standa, Lithuania). The stages were controlled using a program developed in LabVIEW (National Instruments, USA). The laboratory-built shutter was employed to block the laser beam during the returning phase of the flyback raster mode. In this mode, the beam scans from right to left and from top to bottom; it returns to the start of a new line without collecting data. A stainless steel capillary tube with a 1.34 mm internal diameter (Swagelok, USA) was used for aerosol transport. Steel capillary was connected to a laser ablation adapter (31-808-4034, Glass Expansion, Australia) and flowed into a plasma torch with a 2.5 mm injector (G3280-80053; AHF, Germany). The sample aerosol was introduced into a quadrupole mass spectrometer (ICP MS 7900, Agilent, USA) through nickel sampler and skimmer cones (G3280-67040, G8400-67200; AHF, Germany). The ICP operated at 1550 W was sustained by 15 L/min of Ar plasma gas and 0.9 L/min of Ar auxiliary gas. The plasma torch position and ion optics voltage settings were optimized using a MicroMist nebulizer equipped with a Scott spray chamber. A 1 ppb yttrium solution in 2% HNO₃ served as the tuning standard (Analytika, Czech Republic).

Results and Discussion

Optimization of PLISA for HSA Determination

The LA SP ICP MS results of dried UCNP droplet samples demonstrated the feasibility of SP detection from the nitrocellulose pads, suggesting the potential of LA SP ICP MS for single-label detection in immunoassays. To test the performance of the method for the detection of individual biomarkers, the UCNP-based dot-blot immunoassay for HSA was performed on nitrocellulose pads. As shown in our previous reports, UCNP-SA conjugates represent suitable labels for the sensitive dot-blot immunoassays with UCL readout. (4,14,31) However, compared to our previously published dot-blot immunoassay, (4) additional optimizations were necessary because of the different nitrocellulose membrane type used in this work. The most important was the optimization of the coating of the nitrocellulose pads with the capture antibody, as it is crucial for the successful capture of the target antigen and included volume, concentration and composition of coating buffer as well as concentration of UCNP-SA conjugate and scanning method of LA SP ICP MS (Figures S3–S7). The optimal procedure utilized a piezo-driven dispenser (Figure S8) (details on optimization experiments are provided in SI). This approach resulted in greater consistency and homogeneity of the detection area (Figure 2), leading to an LOD of 0.18 ng/mL for UCL readout and an improved LOD of 0.12 ng/mL utilizing the LA SP ICP MS. The working ranges of this immunoassay were 30–750 ng/mL and 50–3400 ng/mL for UCL and LA SP ICP MS, respectively, indicating the higher accuracy of LA SP ICP MS for higher analyte concentrations. Although the antibody deposition with a pipette on a larger area resulted in a lower LOD for the UCL readout compared to the piezo-driven spotting, incorporating the automatic deposition system significantly improved the consistency of results obtained from both detection methods; therefore, it was used for further experiments.

Anal. Chem. 2025, 97, 26, 13832–13839: Figure 2. Calibration curves for the determination of HSA utilizing PLISA based on piezo-driven spotting of the capture antibody. Pad images in pseudocolor scale obtained for (A) UCL intensity readout and (B) UCNP counting by LA SP ICP MS. (C) Corresponding UCL and LA SP ICP MS calibration curves. Empty triangles represent LODs; error bars represent standard deviations.

Determination of PSA in Clinical Samples

The concentration of PSA in the serum of healthy men typically remains below 4 ng/mL, with PSA levels exceeding this threshold, indicating an elevated risk of prostate cancer. (33) In addition, when prostate cancer treatment involves radical prostatectomy, the surgical removal of the prostate, PSA levels typically drop to units of fg/mL. (34) Persistent PSA levels rising above 0.1 ng/mL then indicate significant concerns, such as the presence of residual tumor tissue or cancer recurrence. (35) Both the UCL and the LA SP ICP MS methods are highly relevant for clinical PSA analysis, as their LODs are in the pg/mL range. Furthermore, these methods enable the precise monitoring of subtle changes in PSA concentrations, which is essential for the early detection of postsurgical complications.

Human serum samples obtained from 15 patients tested for prostate cancer were analyzed to assess the clinical performance of PLISA with UCL and LA SP ICP MS readouts. Serum contains many components that can affect immunoassay performance, such as other proteins, various small organic molecules, and possibly even trace amounts of medications. (36) All of them can negatively influence assay sensitivity by nonspecific interactions with the nitrocellulose surface and immunoassay reagents. (37) The achieved sensitivity enabled a reduction of matrix effects through simple dilution; serum samples were diluted 100-fold prior to analysis, which also ensured alignment with the assay working range. The PSA levels obtained by immunoassay with UCL and LA SP ICP MS readout were compared to the reference concentrations measured by the standard electrochemiluminescence immunoassay method. The results of PLISA aligned well with the reference method (Figure 4, Table S1); the slopes of 0.974 and 1.018 and coefficients of determination of 0.75 and 0.69 for UCL and LA SP ICP MS, respectively, demonstrated a strong correlation between the methods. Recovery rates ranged from 78 to 127% for UCL and 76 to 137% for LA SP ICP MS detection. These variations in recovery highlight some limitations, such as the anonymization of clinical samples, which prevented access to patient information, including medication for secondary diseases and other factors that could influence the immunoassay performance. Nevertheless, the results highlight LA SP ICP MS as a highly sensitive readout method, particularly suited for immunoassays for the analysis of clinical samples with low sample volumes and concentrations. Moreover, the strong correlation with the reference method confirmed the applicability of the developed immunoassay for the analysis of complex samples, as other proteins contained in the human serum did not cause significant interference with PSA detection.

Anal. Chem. 2025, 97, 26, 13832–13839: Figure 4. Correlation of PSA concentrations in the clinical samples found by UCL and LA SP ICP MS with the reference electrochemiluminescence immunoassay. Error bars represent standard deviations.

Conclusions

This work introduces a novel type of immunoassay, PLISA, utilizing the LA SP ICP MS readout method. It allows counting individual NP-labeled immunocomplexes from nitrocellulose surface by IR LA of intact UCNPs with SP ICP MS detection. The immunoassay procedure and parameters of LA SP ICP MS readout were first optimized using HSA as a model analyte, reaching LODs of 0.18 ng/mL and 0.12 ng/mL for UCL and LA SP ICP MS readouts, respectively. Under optimal conditions, the detection of PSA in serum was carried out, achieving LODs of 2.4 pg/mL, 1.4 pg/mL, and 0.3 pg/mL for the UCL, analog LA ICP MS, and digital LA SP ICP MS readouts, respectively. The 8-fold and 5-fold improvement of digital readout compared to UCL and LA ICP MS, respectively, clearly demonstrates the superior sensitivity of the approach based on counting individual NP-labeled immunocomplexes. The total analysis time per sample was approximately 3 min for UCL and 15 min for LA SP ICP MS. Digital signal processing allows precise filtering of aggregates of NP labels that could lead to a positive readout error while also reducing noise. Compared to similar methods based on single NP detection or LA ICP MS, the LOD of PLISA is approximately 1 to 6 orders of magnitude lower. Finally, to demonstrate the robustness of our approach, clinical samples from patients screened for prostate cancer were analyzed, showing a strong correlation with the reference electrochemiluminescence immunoassay. This work represents a promising option for ultrasensitive digital readout of immunoassays or microarrays. Moreover, due to the inherent specificity of MS, possible simultaneous detection of labels containing different elements opens doors to extensive multiplex analysis, which will further contribute to the development of more effective biomarker analysis and diagnostic technologies.