Standardized SDS-PAGE Workflow for Personalized Protein Corona Profiling in Early Cancer Detection

Summary

This protocol standardizes SDS-PAGE analysis for personalized protein corona profiling on nanoparticles, enabling reproducible, scalable, and low-cost detection of cancer-specific signatures. Designed for early diagnosis of pancreatic ductal adenocarcinoma, it offers a practical, REASSURED-aligned alternative to complex proteomic methods in both research and clinical settings.

Abstract

The formation of a personalized protein corona (PC) on nanoparticles (NPs) upon exposure to biological fluids has emerged as a promising, minimally invasive strategy for early cancer detection. This study introduces a standardized protocol for profiling the protein corona using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), optimized to ensure robustness, reproducibility, and automation. The method was rigorously validated across a variety of nanoparticle (NP) platforms, plasma sources, and experimental conditions, demonstrating consistently reproducible profiles with minimal operator-dependent variability. In comparison to conventional proteomics techniques, this approach offers a faster, more cost-effective solution that meets the World Health Organization's REASSURED criteria for diagnostics. When applied to pancreatic ductal adenocarcinoma (PDAC), a highly aggressive malignancy with limited early biomarkers, this workflow achieved classification accuracies of up to 90%, even in early-stage patients. This standardized platform marks a significant advancement towards scalable, clinically translatable, and affordable diagnostic tools for cancer screening.

Introduction

Nanoparticle protein corona (NP-PC)

The characterization of the PC that forms on the surface of NPs following their exposure to biological fluids has emerged as a powerful strategy for disease diagnostics and personalized medicine¹. This layer of adsorbed proteins, dynamic and disease-specific in its composition, transforms the synthetic identity of NPs and drives their biological interactions². The composition of the PC is not random but highly dependent on both the physicochemical properties of the NPs and the biological environment in which they are introduced³. Factors such as NP size, shape, surface charge, and composition can significantly influence the affinity, kinetics, and structural arrangement of adsorbed proteins. Likewise, the biological fluid, its source, complexity, and disease state, play a key role in determining the corona's final makeup^4,5,6.

The dynamic nature of the corona further complicates its analysis, as its structure evolves over time from a soft, loosely bound outer layer to a more stable and tightly adsorbed inner core^7,8. This evolving architecture, resulting from competitive protein adsorption events, requires time-resolved and standardized approaches to be meaningfully characterized⁹. Despite this complexity, it is now widely recognized that the PC is not a mere artifact, but a biologically informative layer capable of capturing systemic disease alterations in a non-invasive manner. The goal of the method described in this paper is to standardize the use of SDS-PAGE for the robust, reproducible, and accessible profiling of the personalized PC, particularly for early cancer detection through liquid biopsy approaches¹⁰.

The personalized PC and its relevance in cancer detection

The scientific basis for this method lies in the concept of the personalized PC. Upon incubation with patient plasma, NPs adsorb a unique fingerprint of proteins that reflects the individual's physiological or pathological status^11,12. This concept has been substantiated by studies showing significant alterations in PC composition in the presence of disease, including pancreatic ductal adenocarcinoma (PDAC), lung cancer¹³, meningioma¹⁴, and glioblastoma multiforme¹⁵. Specifically, in PDAC, PCs derived from patient plasma differ substantially from those of healthy donors. These differences have been exploited in classifier models that achieve up to 90% diagnostic accuracy. The PC thus serves as a nanoscale concentrator of disease-specific signals¹⁶.

Recent work has demonstrated that PCs from cancer patients are enriched in proteins involved in inflammation, coagulation, complement activation, and extracellular vesicle transport, biological pathways known to be dysregulated in malignancy¹⁷. These protein patterns are highly reproducible within individual patient classes, yet clearly distinct across different disease states, suggesting that the corona can function not only as a general disease sensor but also as a disease-specific classifier. In the context of PDAC, one of the deadliest and most insidious malignancies, our group has pioneered NP-enabled blood (NEB) tests based on PC characterization^18,19,20. This minimally invasive approach leverages the capacity of NPs to selectively enrich low-abundance and disease-altered plasma proteins, making them detectable via simple, benchtop electrophoretic separation.

Advantages over conventional proteomic techniques

The rationale for developing this approach stems from the limitations of conventional proteomics techniques such as mass spectrometry (MS), which, while offering deep proteome coverage, are expensive, time-consuming, and ill-suited for high-throughput screening or point-of-care diagnostics²¹. Another key challenge lies in the intrinsic complexity of MS, where variations in sample preparation, instrument calibration, acquisition parameters, and data analysis workflows contribute to significant inter-laboratory differences in the reported PC composition²². In contrast, SDS-PAGE analysis of the PC offers a rapid and cost-effective means to capture disease-related proteomic patterns. Crucially, the SDS-PAGE readout is not reliant on the absolute identification of individual biomarkers but on the detection of systematic changes in the global protein profile, which can be analyzed through densitometry and multivariate statistics.

Unlike antibody-based assays, which are limited to known targets and prone to cross-reactivity, SDS-PAGE offers an unbiased view of the protein landscape, preserving the integrity of the original sample and allowing downstream integration with complementary platforms such as LC-MS/MS for protein identification. Moreover, since SDS-PAGE does not require sample fractionation or depletion steps, it is especially suitable for analyzing low-volume biological samples, an essential feature for applications involving early diagnosis or screening of fragile populations²³. For all these reasons, this protocol is consistent with several of the WHO's REASSURED criteria (Real-time connectivity, Ease of specimen collection, Affordable, Sensitive, Specific, User-friendly, Robust, Equipment-free, and Deliverable), making it a practical tool for scalable diagnostics²⁴. Based on this evidence, it was hypothesized that NPs incubated with patient plasma adsorb a reproducible, disease-specific PC that can be systematically profiled using SDS-PAGE. We further propose that standardized electrophoretic analysis of these personalized coronas can distinguish cancer patients from healthy individuals, thereby providing a robust and clinically translatable diagnostic tool.

Workflow and clinical applicability

The SDS-PAGE-based corona profiling protocol systematizes each step: from NPs preparation and incubation with diluted plasma, to corona isolation, gel electrophoresis, and quantitative densitometric analysis (Figure 1). Key parameters such as incubation time, plasma dilution, and NP type²⁵ are finely tuned to ensure optimal protein capture and gel image resolution. The protocol also addresses practical concerns such as sample throughput, storage stability, and batch-to-batch consistency²⁶. Moreover, since it is a non-destructive technique, SDS-PAGE can be seamlessly combined with complementary proteomic approaches such as MS to identify the molecular composition of individual protein bands¹⁵. The workflow has been optimized for high inter-operator reproducibility, including compatibility with manual and automated pipetting platforms, and it is robust across different NP chemistries and plasma concentrations. These features ensure that the method can be applied across diverse laboratory settings, making it amenable to both centralized diagnostic hubs and decentralized research laboratories. Thus, this technique is particularly well-suited for researchers and clinicians involved in early diagnostics, biomarker discovery, and translational nanomedicine. It is applicable to a wide range of pathologies and adaptable to diverse biological fluids beyond plasma, including saliva or urine. The flexibility and robustness of this protocol render it a strong candidate for integration into clinical validation pipelines, large-scale screening studies, and point-of-care platforms.

Protocol

This protocol provides a detailed and standardized workflow for the preparation, incubation with human plasma (HP), isolation, and analysis of the PC formed on various types of NPs. Figure 2 is a graphical representation of the entire protocol, outlining the sequential experimental steps from NP synthesis to multivariate SDS-PAGE analysis of protein profiles. Importantly, the methodology has been standardized by our group, as demonstrated in previous works^18,26, ensuring reproducibility and reliability of the results. The reagents and the equipment used are listed in the Table of Materials.

1. Preparation and characterization of NPs

1. GO nanosheets
   1. Dilute a 4 mg/mL stock solution in ultrapure water to the desired concentration. Sonicate for 2 min at 28% amplitude (pulse: 0.8/0.6 s) to ensure proper dispersion.
   2. Quantify concentration using UV-Vis spectroscopy
   3. Measure hydrodynamic size and zeta potential by dynamic light scattering (DLS) using a 633 nm He-Ne laser. Prepare samples by dilution in ultrapure water based on cuvette detection limits (e.g., 1:100 dilution for 1 mL cuvettes). Record mean ± standard deviation (SD) from three replicates.
2. Gold NPs
   1. Use citrate-stabilized 100 nm gold NPs (concentration from stock: 3.8E+9 particles/mL) without further modification.
   2. Measure hydrodynamic size and zeta potential by dynamic light scattering (DLS) using a 633 nm He-Ne laser. Prepare samples by dilution in ultrapure water based on cuvette detection limits (e.g., 1:100 dilution for 1 mL cuvettes). Record mean ± SD from three replicates.
3. DOTAP liposomes
   1. Dissolve 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP) powder in chloroform inside a certified chemical fume hood while wearing appropriate PPE (lab coat, gloves, goggles).Evaporate under vacuum for 2 h, resuspend the lipid film with ultrapure water to reach a final concentration of 1 mg/mL, and hydrate overnight at 4 °C. Finally, extrude the liposome solution 20 times through a 0.1 µm polycarbonate filter.
   2. Measure hydrodynamic size and zeta potential by dynamic light scattering (DLS) using a 633 nm He-Ne laser. Prepare samples by dilution in ultrapure water based on cuvette detection limits (e.g., 1:100 dilution for 1 mL cuvettes). Record mean ± SD from three replicates.
      CAUTION: Handle chloroform in a fume hood with full PPE. Dispose of organic solvent waste in accordance with institutional chemical safety procedures.
      NOTE: For GO with size distributions centered approximately at 300 nm, sonicate for 10 min at 125 W, centrifuge at 7,700 x g for 30 min at 4 °C, collect supernatant, centrifuge again at 18,620 x g for 30 min, and resuspend the pellet. For GO with size distributions centered approximately at 100 nm, sonicate 3 × 180 min at 4 °C, centrifuging at 10,000 x g for 30 min, at 4 °C after each round, and collecting the final supernatant.

2. Preparation of human plasma

1. Commercial human plasma (HP)
   1. Dilute commercial lyophilized plasma with ultrapure water as needed.
   2. Centrifuge the reconstituted plasma at 18620 x g for 15 min at 4 °C, collect the supernatant and store at -80 °C until use.
2. Patient-derived HP
   1. Centrifuge blood samples at 1216 x g for 10 min at 4 °C and collect the supernatant for plasma separation.
   2. Aliquot 450 µL plasma samples in 0.5 mL protein low-bind tubes and store at -80 °C until use.
      NOTE: All human plasma samples are collected in accordance with institutional ethical guidelines, and informed consent was obtained from all participants.
      CAUTION: Human plasma is biohazardous. Handle inside a biosafety cabinet (BSC) using gloves, a lab coat, and protective eyewear. Dispose of all plasma-containing materials according to institutional biosafety and waste-disposal regulations.

3. Formation of PC

1. Manual liquid handling
   1. Determine optimal NP: plasma ratio by quantifying protein concentration via Bicinchoninic Acid (BCA) assay to adjust plasma dilution.
   2. Add 100 µL of NP suspension to 100 µL of plasma to reach desired plasma concentrations.
   3. Incubate at 37 °C for 1 h.
      ​NOTE: For BCA assay, mix Reagents A and B (50:1), prepare BSA standard dilutions, add 10-25 µL sample/standard + 200 µL reagent, incubate 30 min at 37 °C, and measure absorbance at 562 nm. Plasma concentrations are typically adjusted between 0.5% and 50% v/v relative to NP volume, depending on NP type. For GO, use lower concentrations (from 0.5% to 5% recommended); for DOTAP/gold NPs, use higher concentrations (up to 50%). Use protein 1.5 mL low-bind tubes.
2. Automated incubation protocol
   1. Load the liquid handling software protocol onto the automated pipetting platform (20 µL + 200 µL heads)
   2. Dispense 100 µL of each NP type into low-bind 1.5 mL tubes.
   3. Add reconstituted plasma prepared as in 2.1 and 2.2, according to the following formula:
      where VNP is the NP suspension volume. For 100 µL of NP:
      
      5%: 5.3 µL HP
      20%: 25 µL HP
      50%: 100 µL HP
   4. Add ultrapure water to reach 200 µL:
      5%: 94.7 µL
      20%: 75 µL
      50%: none
   5. Mix samples:
      5%: 15 µL cycles
      20%: 50 µL cycles
      50%: 100 µL cycles3.2.6. Incubate at 37 °C for 1 h
      NOTE: Supplementary Figure 1 illustrates the graphical interface of the automated setup used to prepare NP-HP mixtures at three final plasma concentrations (5%, 20%, and 50% v/v), each in triplicate. The deck layout includes tip racks, a waste reservoir, and a 3 × 4 low-bind tube rack containing NP suspensions (GO, gold nanospheres, and DOTAP liposome) (NP), HP, and ultrapure water (H₂O). Use protein low-bind tubes.

4. PC analysis by SDS-PAGE

1. PC isolation
   1. Centrifuge the NP-HP samples at 18,620 x g, 4 °C, for 15 min. A visible pellet should be present at the bottom of the tube. Carefully discard the supernatant, which should appear clear before proceeding (see Supplementary Figure 2).
   2. Wash pellet with 200 µL ultrapure water, resuspend it, and centrifuge it again. Repeat this step twice. After the final wash, a compact pellet should still be visible, confirming successful isolation of NP-protein complexes (see Supplementary Figure 2).
   3. Prepare loading buffer (LB):
   4. Calculate total LB volume based on plasma condition and mix reducing agent (1:10), SDS sample buffer denature proteins (1:2), and water.
   5. Resuspend pellet in appropriate volume of 1x LB and boil at 100 °C for 10 min.
   6. Centrifuge at 18,620 x g, 4 °C, for 15 min. Collect the supernatant.
2. SDS-PAGE sample loading
   1. Prepare the ladder by diluting the molar weight (MW) standard with the loading buffer.
   2. Dilute 10x Tris/Glycine/SDS running buffer to 1x.
   3. Assemble the electrophoresis chamber with 4%-20% stain-free gradient gel and add running buffer until it reaches the recommended level indicated by the manufacturer.
   4. Load 10 µL samples and 7 µL ladder in each well. Run gel at 150 V for ~90 min at room temperature.
      NOTE: The volume of LB should be calculated based on the plasma concentration selected in 3.1.1 to ensure optimal protein visualization in the gel image (typically 10 µg proteins in 20 µL LB).
      CAUTION: For optimal reproducibility, perform SDS-PAGE analysis using freshly prepared NP-plasma samples. Storage of washed pellets at -80 °C for 24 h resulted in detectable alterations of both physicochemical properties (after resuspension and DLS/zeta potential measurements) and protein corona fingerprints (when pellets were processed starting from Step 4.3.1 of the protocol; see Supplementary Figure 3).

5. Protein patterns analysis

1. Image processing
   1. Image gels using a chemiluminescence imager.
   2. Process the gel image as follows:
      1. Import the gel image and save it as a TIFF file, correct it for undesired effects, e.g., oblique lane distortions that particularly affect peripheral lanes. Detect all gel lanes. Subtract the background signal from the image.
      2. Calibrate the vertical displacements by converting pixel units into molecular weight units, based on a double-term exponential fit of the marker lane. Generate graphical outputs to visualize intensity profiles and lane-specific signal distributions.
   3. Export the processed data, i.e., the computed one-dimensional profiles.
   4. Compute the molecular weight distributions by normalizing the profiles, and calculate integral areas across the desired MW ranges for each lane.
   5. Export the output data, i.e., the integral areas within the specified molecular weight ranges.
2. Multivariate analysis for diagnostic
   1. Apply linear discriminant analysis (LDA) or other multivariate methods for classification analysis.
   2. Calculate accuracy, sensitivity, specificity, and perform ROC analysis.
      NOTE: The employed scripts used for image processing, profile normalization, fitting, and classification are available from²⁷, where the corresponding user manual is also provided.
      CAUTION: Follow biosafety protocols when handling human plasma.

Results

One of the major challenges in standardizing PC experiments lies in minimizing technical and operator-dependent variability, as previously highlighted by this study²⁶. Open questions still persist regarding the influence of experimental variables such as operator bias and the impact of automated tools on reproducibility^18,28.

Firstly, this study p...

Discussion

Method versatility and reproducibility across NP types

The protocol presented here provides a standardized and reproducible workflow for isolating and characterizing the PC formed on NP upon incubation with HP. As demonstrated in previous and current studies³³, the method enables the generation of SDS-PAGE-based molecular fingerprints that capture biologically and diagnostically relevant features of the personalized PC. The observed consistency across m...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
1.5ML Protein LoBind TUBE	eppendorf	22431081
10x Tris/Glycine/SDS	Biorad	1610732
Centrifuge	Labortechnik	Z 216 MK Hermle	Used at 18,620 RCF for 20 min at 4 °C
ChemiDoc imaging system	Bio-Rad	12003153	Used for gel image acquisition
Chloroform	Sigma Aldrich	C2432-1L	Solvent for DOTAP
Corning 1-200 µL Universal Fit Bulk Packed Pipet Tips	corning	4845
Criterion TGX Stain-Free Precast Gels	Biorad	5678095
Distilled Water 6 x 1L	thermo fisher	15230001
DOTAP (1,2-dioleoyl-3-trimethylammonium-propane)	Avanti Polar Lipids	890890P-200mg	Used to prepare cationic liposomes; dissolved in chloroform, evaporated, hydrated, and extruded
Enduro 300V Power Supply	Labnet	E0303-230V
GloMax Discover Microplate Reader	Promega	GM3000
Gold nanoparticles (citrate-stabilized)	Sigma-Aldrich	742031-25ML	Used as nanomaterial; used without further purification
Gradient polyacrylamide gel (4–20% TGX)	Bio-Rad	5678095	Used for SDS–PAGE
Grant Instrument Thermoshaker HC24N (24 x 1.5mL microtubes)	Fisher scientific	PHMT-PSC24N
Graphene Oxide (GO) solution	Graphenea	GO-4-1000	Used as nanomaterial; diluted to 0.25 mg/mL and sonicated
Human Plasma (HP)	Sigma-Aldrich	P9523-5ML	Used for nanoparticle incubation
ImageLab Software	Bio-Rad. Version 6.1.0.07	/	Used for exporting gel image as tiff files
Laemmli loading buffer 2x	Biorad	1610737	Used for resuspending pellets
MATLAB	MathWorks. Version R2022a	/	Used for image processing, data analysis and profile computation
Mini-Extruder	Avanti Polar Lipids	610020	Used for liposome preparation
NuPAGE Sample Reducing Agent (10X)	Thermo Scientific	NP004
Phosphate Saline Buffer	Corning	21-031-cm	Used for DOTAP-HP samples
Pierce BCA Protein Assay Kit, 1 Kit (1 L)	ThermoFisher	23225
Pipet tips 0.5-10ul clear	Axygen	T-300-STK
PIPETMAN L P10L, 0.5-10 µL, Metal Ejector	Gilson	FA10002M
PIPETMAN P20, 2-20 µL, Metal Ejector	Gilson	F144056M
PIPETMAN P200, 20-200 µL, Metal Ejector	Gilson	F144058M
PIPETMAX automated pipetting system	Gilson	GSYS0021	Used for automated sample handling
Polycarbonate filter (0.1 µm)	Avanti Polar Lipids	610005-1EA	Used for liposome extrusion
Precision Plus Protein All Blue Prestained Protein Standards	Biorad	1610373
Precision Plus Protein Unstained Standards	Biorad	1610363EDU
Vibra cell sonicator VC505	Sonics and Materials inc.	VC505
ZetaSizer	Malvern	ZS90	Used for dynamic light scattering (DLS) and microelectrophoresis (ME) measurements