Development of molecularly imprinted polymers for the detection of human chorionic gonadotropin

Download PDF Article Open access Published: 26 March 2025 Development of molecularly imprinted polymers for the detection of human chorionic gonadotropin Radvilė Zubrytė1,2, Liliia

Mavliutova2, Yadiris García3,4, Mark V. Sullivan5, Nicholas W. Turner5, Francesco Patitucci3,6, Laura C. Polania4, Verónica A. Jiménez4, Robert Porter1, Alice Mattsson1 & …Börje

Sellergren2,3 Show authors Scientific Reports volume 15, Article number: 10436 (2025) Cite this article

2184 Accesses

Metrics details

Diagnostic pregnancy tests are the most widely used immunoassays for home-based use. These tests employ the well-established lateral flow assay (LFA) technique, reminiscent of affinity

chromatography relying on the dual action of two orthogonal anti-hCG antibodies. Immunoassays suffer from several drawbacks, including challenges in antibody manufacturing, suboptimal

accuracy, and sensitivity to adverse storing conditions. Additionally, LFAs are typically designed for single use, as the LFA technique is non-reusable. An alternative to overcome these

drawbacks is to leverage molecularly imprinted polymer (MIP) technology to generate polymer-based hCG-receptors and, subsequently, non-bioreceptor-based tests. Here, we report the

development of MIP nanogels for hCG detection, exploiting epitopes and magnetic templates for high-yielding dispersed phase imprinting. The resulting nanogels were designed for orthogonal

targeting of two immunogenic epitopes (SV and PQ) and were thoroughly characterized with respect to physical properties, binding affinity, specificity, and sensitivity. Molecular dynamics

simulations indicated a pronounced conformational overlap between the templates and the epitopes in the native protein, supporting their suitability for templating cavities for hCG

recognition. Quartz crystal microbalance (QCM)-based binding tests and kinetic interaction analysis by surface plasmon resonance (SPR) revealed nanomolar dissociation constants for the MIP

nanogels and their corresponding template peptides and low uptake of lutenizing hormone (LH), structurally resembling to hCG. Receptor reusability was demonstrated in the multicycle SPR

sensing mode using a low pH regeneration buffer. The results suggest the feasibility of using imprinted nanogels as a class of cost-effective, stable alternatives to natural antibodies for

hCG detection. We foresee applications of these binders with respect to reusable pregnancy tests and other hCG-related disease diagnostics.

Molecular fingerprinting of biological nanoparticles with a label-free optofluidic platform Article Open access 15 May 2024 Soft molecularly imprinted nanoparticles with simultaneous lossy

mode and surface plasmon multi-resonances for femtomolar sensing of serum transferrin protein Article Open access 11 July 2023 DNA tetrahedron-mediated immune-sandwich assay for rapid and

sensitive detection of PSA through a microfluidic electrochemical detection system Article Open access 25 April 2021 Introduction

Human chorionic gonadotropin (hCG) is a 36 kDa glycoprotein hormone produced by the placenta during early pregnancy. This protein is composed of an α and β chain forming a heterodimer, with

variable glycosylation levels comprising up to ca 30% of its total molecular weight. hCG plays a pivotal role during early pregnancy by promoting the thickening of the uterine lining to

support the nourishment of a growing embryo. Consequently, hCG becomes detectable shortly after fertilization1,2. Most pregnancy tests available on the market today demonstrate a detection

accuracy of 99% when used on the day of a missed period or 14 days post-ovulation correlating with hCG levels of about 50 mIU/mL3. However, these levels can vary significantly among

individuals. Given the frequency with which women use pregnancy tests throughout their reproductive years, they represent today one of the most commonly utilized immunoassays4. In other

medical contexts, hCG plays a vital therapeutic role for in vitro fertilization and is a significant tumor marker across various cancers, including testicular, ovarian, and trophoblastic

tumors5 Quantitative and specific hCG tests are thus indispensable for both pregnancy and disease management purposes. The development of corresponding tests is associated with different

requirements in terms of sensitivity ranges and the need for detecting various modified forms of hCG6. All hCG tests are antibody-based immunoassays designed to distinguish hCG from closely

related hormones such as luteinizing hormone (LH), follicle-stimulating hormone (FSH), and thyroid-stimulating hormone (TSH). While these hormones share identical α-chains with hCG, their

distinct β-chains confer unique biological activities despite significant sequence homology with hCG (e.g., LH shares ~ 80% sequence identity). Accurate hCG tests depend on access to

specific antibodies displaying no or minimal cross-reactivity with the other hormones7. Developing such antibodies, in turn, depends on a comprehensive structural mapping of all isoforms to

identify proteotypic and solvent-accessible epitopes. This is further complicated by the wide range of hCG variants differing in glycosylation state and fragment size8. In addition,

antibody-based assays suffer from inherent challenges such as complicated manufacturing, suboptimal accuracy, sensitivity to adverse storing conditions, and restriction to single use. This

has led to an emerging need for robust and readily available recognition elements as sustainable alternative antibodies9,10,11.

In recent years, molecularly imprinted polymers (MIPs) have emerged as promising robust synthetic antibodies for detecting fluid biomarkers12,13,14,15. These are prepared in the presence of

a template that, post-removal, leaves behind binding sites with template-complementary shapes and functionalities. MIPs are robust and stable, negating the need for costly,

temperature-controlled supply chains. Meanwhile, they exhibit molecular recognition properties comparable with antibodies, are inexpensive to produce, and are made using animal-free methods.

Moreover, unaffected by harsh cleaning conditions, MIPs are reusable, potentially paving the way for more sustainable diagnostic tools16,17. MIPs adapted for diagnostic applications are

preferably nanoparticles or nanogels produced by precipitation, emulsion, or graft polymerization,13,14,18 and thin-film materials19 that exhibit homogenous binding sites due to the spatial

restrictions imposed by the limited film thickness or nanoparticle radius. Following a quasi-generic protocol, polyacrylamide-gel-based nanogels can now be manufactured to target

well-established antigens20,21,22 Like antibodies, these binders can bind biological targets ranging from small to large molecules and have been implemented in sensor and assay formats. This

notable progress stems from recent advances in polymer, colloid, and host–guest chemistry, particularly through the application of epitope imprinting22,23,24,25,26 combined with

solid-phase27,28 or dispersed phase synthesis using magnetic template carriers29,30,31. In the latter methods, the template is strategically immobilized in a site-directed orientation on

non-porous glass beads or magnetic nanoparticles, respectively. Subsequently, high dilution polymerization of appropriate water-soluble monomers occurs, resulting in polymers partially

adhering to the surface of the template carriers. The magnetic templating method (Fig. 1) is preferred for high-yielding synthesis of epitope imprinted nanoparticles targeting hCG.

Monomers used to synthesize nanogels for hCG recognition and principle of hCG-epitope imprinting exploiting dispersions of magnetized templates to produce imprinted nanogels.

image

Similar to antibody-antigen interactions, the epitope refers to a short, solvent-exposed peptide sequence (8–20 amino acids), acting as an antigenic determinant, thus constituting the

specific site on the protein surface interacting with the MIP. Preferred epitopes are linear solvent-exposed C- or N-terminal sequences or internal conformationally defined loop structures.

Moreover, the epitopes should be free from interfering post-translational modifications and generate adequate affinity and specificity for the target hormone, in this case, allowing

detection of hCG levels in the range 10 pM – 1.0 µM and absence of cross-reactivity with luteinizing hormone (LH). Most hCG-reactive antibodies bind to assembled discontinuous epitopes and

rarely to linear continuous sequences8,32 Two exceptions are the C-terminal peptide (βCTP) of the β-chain, aa 135–145 (PGPSDTPILPQ = PQ), and the β-chain loop structure sequence aa 66–80

(SIRLPGCPRGVNPVV = SV) (Fig. 2A). βCTP is a flexible terminal sequence lacking a defined secondary structure. Hence, its immunogenicity is limited, constraining the production of

high-affinity antibodies. Given that βCTP is lacking in LH, monoclonal antibodies against this sequence show high selectivity to hCG versus LH and have found use in various commercial

sandwich immunoassays33. Interestingly, glycosylation of this sequence at Ser138 did not seem to influence affinity for this epitope34. On the other hand, the β-chain loop structure sequence

aa 66–80 is a highly conserved motif present in all hCG variants that has been proposed as an ideal epitope for developing a universal, single epitope hCG assay. Indeed, monoclonal

antibodies recognizing this sequence show an exceptionally high target selectivity with only minor cross-reactivity with the homologous LH fragment (aa 86–100) that features an identical

sequence except for Asn77, which in LH corresponds to Asp. This reflects the high performance of the corresponding monoclonal antibodies35.

(A) Structure of the full hCG hormone in which the epitopes SV and PQ are highlighted. SV corresponds to the β-chain loop residues 66–80 (SIRLPGCPRGVNPVV). PQ is the C-terminal peptide of

the β-chain, corresponding to residues 135–145 (PGPSDTPILPQ). (B) Structures of the SV and PQ peptides conjugated to the N-acetylated-6-aminohexanoyl (Ahx) moiety at the N-termini. (C) MD

simulation box used to evaluate the conformational properties of the peptide templates (Ahx-SV or Ahx-PQ) within a pre-polymerization mixture environment. The simulated systems were built by

placing the templates Ahx-SV or Ahx-PQ at the center of the box surrounded by 200 monomers in proportions consistent with the experimental setup and explicit solvent molecules.

image

In this report, we used these two immunogenic sequences (SV and PQ) as templates for preparing high-affinity polymer-based receptors for hCG. Molecular dynamics simulations confirmed that

the templates retain their mimotopic conformations within a pre-polymerization mixture environment, thus being suitable for imprinting cavities capable of hCG recognition. The obtained MIPs

recognize their templates with high affinities when used as receptors in a QCM-based or a regenerable SPR-based biosensor.

Ferric chloride hexahydrate (FeCl3·6H2O), sodium acetate, polyethylene glycol 6000 (PEG), sulfuric acid, glycine and sodium dodecyl sulfate (SDS) were obtained from Merck. Ethylene glycol,

ammonium hydroxide 25%, triethylamine (TEA), (3-aminopropyl)trimethoxysilane (APTMS), succinic anhydride, 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDC),

N-hydroxysuccinimide (NHS), ammonium persulfate (APS), N,N,N’,N’-tetramethylethylenediamine (TEMED), N-isopropylacrylamide (NIPAm), N-tert-butylacrylamide (TBAm), acrylic acid (AA),

tetraethyl orthosilicate (TEOS), phosphate-buffered saline (PBS), ethanolamine hydrochloride, 16-mercaptohexadecanoic acid (MHA), 11-mercapto-1-undecanol (MU), acetic acid, sodium chloride

(HCl), triethylamine, 4-(2-hydroxyethyl)piperazine-1-ethane-sulfonic acid (HEPES), 2-[morpholino]ethanesulfonic acid (MES), albumin from Bovine Serum (BSA), human chorionic gonadotropin

(hCG) and luteinizing hormone (LH) were obtained from Sigma-Aldrich. N,N-Dimethylformamide (DMF), acetone, hydrochloric acid (HCl) and tetrahydrofuran (THF) were purchased from VWR.

N,N′-methylene-bis-acrylamide (MBAm) was obtained from Alfa Aesar, N-(3-aminopropyl)methacrylamide hydrochloride (APM) was obtained from Polysciences Inc. Ethanol was obtained from Solveco.

6-Aminohexanoyl-hCGβ60-80 (Ahx-SIRLPGCPRGVNPVV) and 6-Aminohexanoyl-hCGβ135-145 (Ahx-PGPSDTPILPQ) (> 95%) were purchased from Lifetein LLC (New Jersey, US). N-fluoresceinacrylamide was

synthesized as reported previously36 Fluoresceine was purchased from TCI, and Tween 20 was obtained from AppliChem.

MD-simulations were carried out in a computer cluster equipped with (a) 2 Supermicro Quantum TXR413-1500R servers with 2 Intel Xeon® processors E52620, and 16 NVIDIA GeForce GTX1080 GPUs and

(b) 2 TensorEX TS4-1,598,415-AMB servers with 2 Intel Xeon Silver® processors and 16 NVIDIA RTC2080 GPUs.

Effective hydrodynamic diameters (dh) of the particles were determined by dynamic light scattering (DLS) with a Zetasizer Ultra (Malvern Panalytical) equipped with a He–Ne laser (688 nm) and

set to backscatter mode, with measurements of samples performed in triplicate at 25 °C. Data was analyzed using the ZS Xplorer software.

Spectra were collected using a Nicolet 6400, equipped with a DTGS detector. The smartiTR accessory was used to characterize dried modified magnetic nanoparticles. 500 spectra were collected

at resolution 4. Compressed air was continuously run through the instrument during and before the measurements. Baseline correction and data management were performed with the OMNIC 6

software.

The affinity of the imprinted nanoparticles for the epitope target was investigated using a Reichert 2 SPR system (Reichert Technologies, Buffalo, USA) with an attached autosampler and

degasser. The specificity of the imprinted nanomaterial was investigated by binding a non-target peptide of similar shape and size.

The interaction between the peptide/protein and nanogels was analyzed with Q-Sense QCM-D E4 unit equipped with a standard flow module (Biolin Scientific AB, Sweden). The sensors used for the

experiments were the QCM5140TiAu120-050-Q (5 MHz) from Quartz Pro AB (Jarfalla, Sweden). All sensors were cleaned in line with the manufacturer’s recommendations.

of magnetic nanoparticles

Magnetic nanoparticles (MNP) were obtained by solvothermal synthesis, as reported by Mahajan et al.37 Briefly, FeCl3·6H2O (16.6 mmol), sodium acetate (95.0 mmol), SDS (21.8 mmol) and PEG

6000 (2.7 g) were dissolved in 150 mL of ethylene glycol. The mixture was magnetically stirred at 100 °C for 30 min and then transferred to a 150 mL Teflon-lined stainless-steel autoclave.

The autoclave was sealed and heated at 180 °C for 24 h. Then, the container was cooled to room temperature, and the MNPs were separated by a magnet. The solid was washed five times with

deionized water, three times with 100 mL of ethanol, three times with 100 mL of acetone, and then dried in a desiccator under vacuum at 22 °C for 24 h. Subsequently, 1000 mg of MNPs were

dispersed in 1 L of 80% ethanol containing 0.25 mM ammonium hydroxide, and ultrasonicated for 2 min at 50% intensity using a sonifier (BRANSON). A 0.5 M TEOS solution was then added, and the

reaction mixture was shaken in an orbital shaker for 6 h. The obtained nanoparticles were washed with deionized water until neutral pH, followed by three washes with 100 mL of ethanol, and

dried in a desiccator over activated silica at 22 °C for 24 h. The produced silica-coated nanoparticles (MNP@Si,1000 mg) were surface-functionalized by sonication in a bath sonicator for 2 h

with APTMS (57 mM) in 75% ethanol. The resulting primary amine-modified MNPs (MNP-NH2) were washed three times with ethanol and dried in a 37 °C oven overnight. Particles were further

functionalized with succinic anhydride (2.1 M) in DMF by sonication in a bath for 3 h. The final product (MNPs-COOH) was washed ten times with deionized water, three times with ethanol, and

dried at 37 °C oven.

MNPs-COOH (100 mg) was placed into a 20 mL vial along with 6 mL of deionized water at room temperature, followed by sonication for 30 s twice at 35% amplitude. The carboxylic groups on the

surface of the magnetic nanoparticles were activated by adding 2 mL of freshly prepared EDC solution (100 mg·mL-1) and 2 mL of NHS solution (62.5 mg·mL-1) in deionized water. The mixture was

then shaken on an orbital shaker for 1 h at room temperature. Afterward, the activated magnetic nanoparticles were washed twice with 10 mL of deionized water and once with 5 mL of 10 mM PBS

(pH 7.4). The immobilization of Ahx-PQ and Ahx-SV was carried out by adding 5 mL of epitope solution (0.50 mg·mL-1 in 10 mM PBS, pH 7.4) to the vial, which was then shaken at room

temperature overnight. Subsequently, the peptide conjugation was assessed via the fluorescamine test.38,39 A portion of the initial supernatant (100 μL) from the reaction was mixed with 25

μL of Fluorescamine solution (0.50 mg·mL-1 in acetone). Fluorescence was then measured (λex = 400, λem = 510 nm), and the specific amount of conjugated peptide was calculated from the

reduced signal upon solution depletion. To remove any excess epitope, the MNPs were collected using an external magnet, rinsed with PBS and deionized water, then suspended in 5 mL of

deionized water and stored at 4 °C.

Monomer feed ratios used to synthesize the epitope-imprinted MIPs are reported in Table S2. Synthesis of MIP-PQ was conducted as follows. A pre-polymerization mixture with a total monomer

concentration of 10 mM was prepared by dissolving NIPAm (5.43 mg, 48 μmol), BIS (0.31 mg, 2 μmol), TBAm (5.09 mg, 40 μmol dissolved in 1 mL of ethanol), APMA (1.79 mg, 10 μmol), and 0.4 mg

of N-fluoresceinylacrylamide (dissolved in ethanol) in deionized water (9 mL) in a 20 mL vial. The pre-polymerization mixture was homogenized for 30 min, followed by the addition of 1 mL of

25 mg·mL-1 MNP-SV or MNP-PQ dispersion. The mixture was stirred at room temperature under N2 atmosphere for 15 min. Then, 0.26 mmol of ammonium persulfate (APS) and 0.40 mmol of

N,N,N’,N’-tetramethylethylenediamine (TEMED) were added and the dispersion was left overnight at 40 °C under constant stirring (480 rpm). Afterward, the particles were collected using an

external magnet and washed with deionized water (5 × 10 mL) at 40 °C to remove unreacted functional monomers and low-affinity nanoparticles. Finally, the high-affinity imprinted nanogels

were obtained by adding 4 mL of 1 mM PBS to the magnetic particles and leaving at 4 °C for 4 h. The elution step was performed by collecting the first 4 mL, and this process was repeated

once on the magnetic nanoparticles by adding an additional 1 mL of PBS and incubating overnight at 4 °C. Thus, the elution process resulted in a total volume of 5 mL of imprinted nanogel

solution. Concentrations of the nanogel solutions were calculated by taking 400 µL of the solution (in triplicate) and evaporating to dryness. The mass of the dried particles was then

measured, and the amount was multiplied by 2.5 to give the concentration in μg mL−1.

To evaluate the hydrodynamic size of nanogels, a 100 μg mL−1 dispersion of the MIP nanogels in water was prepared. The sample was sonicated in a bath sonicator for 30 min and left at 4 °C

for 2 h. To evaluate MNP synthesis and surface modifications, a 50 µg mL-1 suspension of the MNPs was prepared in water. Before the measurement samples were sonicated 3 times for 10 s at 50%

amplitude. Size and zeta potential measurements were done in triplicates.

A carboxymethyl dextran hydrogel-coated Au chip (Reichert, USA) was preconditioned within the SPR using a running buffer consisting of PBS (0.010 M) and 0.010% Tween 20 at pH 7.4, at a flow

rate of 10 µL min-1. The carboxylic acid groups on the dextran chip were activated with an injection of 1 mL of aqueous solution containing 40 mg EDC and 10 mg NHS passed over the chip (6 

min at 10 µL min-1). The MIP nanogels (approximately 300 µg), were activated by dissolving in 1 mL of 10 mM sodium acetate in PBST solution. This was injected over the left channel (working

channel) of the chip for 1 min. The amine groups of the MIP nanogels react with the functionalized surface of the chip, leading to the covalent immobilization of the nanoparticles. A

quenching solution of ethanolamine (1 M at pH 8.5) was injected over both channels (working and reference) for 8 min, followed by a continuous flow of PBST at 10 µL min-1. All injections

were taken from a stable baseline.

The kinetic analysis for the affinity of the target peptide to the MIP nanogels was performed in a set pattern of a 2 min association (PBST with the analyte in a concentration range of 4–64 

nM), 5-min dissociation (PBST only) and a regeneration cycle (regeneration buffer 10 mM Glycine–HCl, pH 2 for 1 min) followed by a final stabilization cycle (PBST for 1 min). An initial

injection of blank PBST was used as the first run, with increasing analyte concentration for subsequent runs. After the analyses were completed, signals from the reference channel were

subtracted from signals from the working channel. Selectivity of the MIP nanogels was investigated by repeating the kinetic analysis, but with a non-target analyte with the same

concentration range (4–64 nM). All experiments were performed in triplicate (n = 3). The SPR responses were fitted to a 1:1 Langmuir fit bio-interaction (BI) model using the Reichert

TraceDrawer software. The association rate constants (ka), dissociation rate constants (kd), and maximum binding (Bmax) were fitted globally, whereas the BI signal was fitted locally.

Equilibrium dissociation constants (KD) were calculated by kd/ka. For each MIP nanogel/analyte epitope combination, a calibration curve was generated across the concentration range 4–64 nM,

using the SPR fitted curve maxima. From this information, a theoretical limit of detection (LOD) was calculated.

After washing with piranha solution (H2SO4:H2O2, 4:1), a gold-coated QCM sensor chip was dried with N2. The chip, cleaned via plasma cleaner for 5 min, was then placed in a petri dish and

immersed in 4 mL of a thiol solution composed of 1 mM 16-Mercaptohexadecanoic acid (MHA) and 1 mM 11-Mercapto-1-undecanol (MU) in absolute ethanol:acetic acid (ratio 9:1) for 18 h at room

temperature in the dark. Afterward, the chip was washed three times with absolute ethanol and 10 mM MES buffer (pH 6) and dried with N2. To activate the carboxylic group of MHA, 100 μL of a

sulfo-NHS:EDC mixture (1:1, 100 mg·mL-1 in MES buffer) was introduced for 30 min, followed by washing with 10 mM PBS and drying with N2. The immobilization of the epitope was conducted using

a standard solution of Ahx-PQ or Ahx-SV prepared at a concentration of 1 mg·mL-1 in 10 mM PBS (pH 7,4). Immobilization of hCG was carried out using a protein solution at 10 μg·mL-1 in PBS.

The protein was injected into a QCM-D instrument to verify the bound amount. Additionally, to confirm the specificity of imprinted nanogels, the same procedure was conducted for LH.

measurements

The experiments aimed at examining the binding of imprinted nanogels with PQ, SV and hCG were carried out utilizing various concentrations of MIP-PQ or MIP-SV (12.5, 25, 50, 100 μg·mL-1)

prepared through the same procedure employed for DLS measurements. The fundamental frequency was approximately 5 MHz and the baseline frequency signal was calibrated by injecting a running

buffer solution composed of 1 mM PBS and Tween 20 (0.005%, v/v) for 1 h at a flow rate of 10 μL·min-1 until the frequency signal response stabilized within ± 0.2 Hz/5 min. The nanogel

solutions were introduced using a flow rate of 100 µL/min for 3min followed by 10 µL/min for 20 min. The frequency variations attributed to binding with analytes were recorded in accordance

with the frequency signal response. To evaluate the specificity of MIP-PQ and MIP-SV in detecting hCG and LH, the proteins were immobilized on QCM sensor chips, and the coupling was analyzed

using the same dilutions as before. The apparent molarities of nanogels were determined using Eq. 1.

where, d is the hydrodynamic diameter of particles, NA is Avogadro’s constant (6.023·1023 mol-1), X is the nanogel concentration in g·mL-1 (1·10–4 g·mL-1) and ρ is the polymer density. The

results revealed an apparent molarity of ~ 40 nM.

The coordinates of the full hCG protein were retrieved from the Protein Data Bank server (PDB code 1HRP)40 The missing protein segments were built using the Chimera X interface to Modeller

tools. The coordinates of the epitopes SV and PQ were obtained from residues 60–80 and 135–145 of the β-chain, respectively. Both epitopes were modified at the N-terminal end by attaching an

N-acetylated-6-aminohexanoyl (Ahx) moiety. The modified epitopes were placed at the center of a cubic box of 80 Å side and surrounded by 200 randomly located monomers in proportions

consistent with the formulations reported in this work using the PACKMOL software41 The simulated mixtures were neutralized with Na+ or Cl- ions and solvated in a cubic box of explicit OPC

waters considering an outer layer of 5 Å measured from the outermost atoms. The ff19SB force field was used to model the peptidic segments, whereas the Ahx moieties were modeled using

parameters consistent with the gaff2 force field with AM1-BCC charges. MD simulations were carried out using the following protocol: (a) 1500 steepest descent minimization steps followed by

3500 conjugate gradient minimization steps for water molecules relaxation, (b) 1500 steepest descent minimization steps followed by 6500 conjugate gradient minimization steps for the entire

system, (c) 500 ps of progressive heating from 0 to 300 K in a canonical ensemble (NVT) with a constant number of particles (N), volume (V), and temperature (T), (d) 10 ns of equilibrium at

300 K to ensure density equilibration in a isothermal–isobaric ensemble (NPT) with a constant number of particles (N), pressure (P), and temperature (T), (e) 20 ns of NPT equilibrium at 300 

K, and finally (f) 150 ns of NPT production dynamics at 300 K and 1 bar from which production data were collected. During MD simulations the cutoff for non-bonded terms was 10 Å, long-range

electrostatics were treated using the Particle-Mesh Ewald approach, and the SHAKE algorithm was employed to constrain all bonds involving hydrogen. Positional restraints were applied to the

terminal acetyl group of the Ahx moiety throughout the simulation protocol to mimic the immobilization of the epitopes to the solid support through Ahx. All Calculations were carried out

using the pmemd.CUDA software implemented in AMBER20. Trajectory analysis was carried out with the CPPTRAJ42 and VMD43 software.

This study reports the development of molecularly imprinted polymer nanogels (MIPs) exploiting an epitope-based approach for hCG detection. To this aim, we focused on the 15aa sequence

corresponding to the β3-loop of the β-subunit from residues 60 to 80 (SIRLPGCPRGVNPVV = SV) and the C-terminal undecapeptide of the β-chain from residues 135 to 145 (PGPSDTPILPQ = PQ) as

templates. To confirm that these sequences properly mimic the three-dimensional conformation of the corresponding protein epitopes, we performed molecular dynamics (MD)

simulations.

MD simulations were used to investigate the conformational properties of the SV and PQ segments in the full hCG protein and in the context of pre-polymerization formulations. This is

relevant to confirm whether the selected templates preserve the structural features required for imprinting molecular cavities capable of efficiently recognizing the full protein. To that

end, we conducted MD simulations on the structures of the SV and PQ peptides conjugated with the Ahx moiety (Ahx-SV and Ahx-PQ, respectively) (Fig. 2B) and surrounded by a mixture of

functional monomers and explicit solvent molecules in proportions consistent with the experimental setting reported in this work (Fig. 2C). Parallel simulations were carried out on the

structure of the full hCG protein in water to evaluate the intrinsic dynamic properties of the SV and PQ epitopes.

The global conformational properties of the template peptides Ahx-SV and Ahx-PQ and the corresponding epitopes in the hCG protein were evaluated from root-mean-square deviation (RMSD)

calculations on the peptide backbone atoms throughout each simulation run, using the epitope coordinates in the 1HRP crystallographic model as the reference structure (Fig. 3A). RMSD data

indicate that the template peptides Ahx-SV and Ahx-PQ have similar mobilities as the corresponding epitopes in the full protein, with PQ being more flexible than SV. The size and compactness

of each peptide were evaluated from radius of gyration (Rg) calculations along each MD trajectory (Fig. 3B). SV has very similar Rg distributions in the Ahx-SV template and in the full hCG

protein, whereas PQ has slightly larger Rg values in the Ahx-PQ system, suggesting more extended conformations than in the hCG epitope. The structural consistency between the SV peptide in

Ahx-SV and the full protein was confirmed by the visual inspection of the conformations achieved by the sequence in their MD trajectories (Fig. 3B,C). SV retained the loop structure required

to imprint cavities suitable for hCG recognition, which is a relevant outcome to support the choice of this segment as a template for the synthesis of molecularly imprinted nanogels. In the

case of PQ, the peptide adopts more extended conformations in the Ahx-PQ than in the hCG protein, as inferred from Rg data. Regarding the intrinsic arrangement of the peptide sequences,

DSSP secondary structure analysis confirmed that SV and PQ are highly disordered peptides that exist mostly in turn and bend states both in the Ahx-conjugated templates and the full hCG

protein (Fig. 3D). This conformational plasticity is relevant for the epitopes to adapt to imprinted cavities of variable size and shape, which supports the choice of SV and PQ as template

peptides in our experimental design.

(A) Root-mean-square deviation RMSD (Å) distributions calculated for the peptide backbone in Ahx-SV and Ahx-PQ and the corresponding epitopes in the hCG protein. Data was collected from the

analysis of 150 ns MD simulations in pre-polymerization mixtures (Ahx-SV and Ahx-PQ) or water (full hCG), using the coordinates of the SV and PQ epitopes in the 1HRP crystallographic

structure as reference models. (B) Peptide radius of gyration Rg (Å) for the SV and PQ sequences in Ahx-SV, Ahx-PQ, and the full hCG protein calculated from 150 ns MD trajectories. (C)

Structural representations of the backbone conformations achieved by the SV and PQ sequences in Ahx-SV, Ahx-PQ, and the full hCG protein throughout 150 MD trajectories. Structures are

displayed from MD trajectories aligned to the first and last amino acids of each sequence. (D) Heatmaps for the frequency of secondary structure assignations according to DSSP analysis on SV

and PQ sequences in Ahx-SV, Ahx-PQ, and the full hCG protein throughout 150 MD trajectories.

Besides conformational features, we used MD simulations to estimate the strength of peptide-monomer interactions within the pre-polymerization mixtures under a LIE approach. This method

decomposes the total interaction energy between a molecule and its surrounding environment (Etotal) as the sum of electrostatic (Elec) and van der Waals (EvdW) energy terms. Energy

components were calculated between all atoms in the peptides with all atoms in the surrounding monomer mixtures with a distance cutoff of 12 Å along the entire MD trajectories (Fig. 4A). For

Ahx-SV, the interaction with the monomer mixture is almost equally driven by the contributions from Elec and EvdW terms. On the other hand, the interaction of the monomer mixture with

Ahx-PQ shows a stronger Elec component over EvdW. A possible explanation for this effect arises from the presence of the highly polar SDT triad in PQ, which can engage in strong

electrostatic or hydrogen bonding interactions with opposite-charged moieties of the surrounding monomers. Added to the energetics of peptide-monomer interactions, we examined the

organization of the pre-polymerization mixtures by counting the number of monomers interacting with the peptides at distances ≤ 5 Å throughout the MD trajectories (Fig. 4B). Distribution

data reveals that the monomer organization mostly relies on TBAm and NIPAm, which are the monomers with higher proportions in both monomer mixtures.

(A) Peptide-monomer interaction energies calculated under the LIE approach for the Axh-SV and Ahx-PQ peptides with their surrounding monomer mixtures. The total interaction energy (Etotal)

corresponds to the sum of electrostatic (Elec) and van der Waals (EvdW) energy contributions. Energy terms are reported as mean values (kcal/mol) with their corresponding standard errors.

(B) Distribution data for the number of monomers that interact with the Axh-SV and Ahx-PQ peptides at distances ≤ 5 Å throughout the MD trajectories.

Magnetic nanoparticles were synthesized via solvothermal method and subsequently surface-functionalized to yield carboxyl-terminated magnetic nanoparticles (MNP-COOH). DLS analysis (Table 1)

and FTIR (Fig. S3) confirmed the successful surface modifications. The MNP-COOH particles were conjugated with the peptides SV and PQ via their Ahx linkers, resulting in the

template-conjugated magnetic nanoparticles referred to as MNPs-SV and MNPs-PQ. The surface coverages of the peptides were estimated to be 8.3 and 11.7 µg·g-1 for Ahx-SV and Ahx-PQ,

respectively, which are comparable to coverages reported elsewhere.22,30,31 DLS measurements (Table 1) confirmed that the template-conjugated MNPs maintained nanoscale hydrodynamic radii (ca

280 nm) and exhibited Zeta potential changes agreeing with the net charges of the epitopes (Table S1).

reported in this work.Full size tableMIP synthesis

MIPs were synthesized by free radical polymerization of acrylamide monomers in the presence of the template-conjugated MNPs. The monomers were rationally selected based on previous designs,

to enable diverse intermolecular interactions with the immobilized peptides. The template peptides SV (SIRLPGCPRGVNPVV) and PQ (PGPSDTPILPQ) comprise 15 and 11 amino acids, respectively. SV

contains two positively charged and eight hydrophobic amino acids reflected in both a high pI and Gravy index (10.9 and 0.30). On the other hand, PQ features one negatively charged and seven

hydrophobic residues resulting in a low pI and Gravy index (3.1 and -0.64) (Table S1). Building upon previously reported examples, poly-NIPAm-based nanogels were synthesized using a

template matching ratio of the charged monomers N-(3-aminopropyl)methacrylamide hydrochloride (APM) and acrylic acid (AA) in addition to the hydrophobic functional monomer

N-tert-butylacrylamide (TBAm) (Table S2). Additionally, N,N’-methylenebisacrylamide (BIS) (2%) was used as a cross-linker, and the fluorescent functional monomer N-fluoresceinylacrylamide

(FITC-AAm) was doped to allow fluorescent tracking of the nanogels30,44 The polymeric nanogels were purified by magnetic decantation with pure water, resulting in a mass yields of 633 mg

SV-MIP and 153 mg PQ-MIP per gram of MNP carrier. This confirms that mass yields can be dramatically enhanced (> > 100x) using high surface area MNPs as template carriers30,31 The average

size, polydispersity index, and Z-potential of the MIP nanogels assessed through DLS (Fig. S1, Table 1) confirmed their nanoscale dimensions although with a minor aggregation tendency,

notably in the case of MIP-SV (Fig. S1). The latter display a second, albeit much smaller, distribution peak, which we attribute to slight aggregation of the nanogels31 Nevertheless, the DLS

size distribution supports that this synthetic method for the production of imprinted nanogels generally produces regular homogenous nanoparticles. Any potential aggregation issues was

taken into account, and to prevent any potential issues during immobilisation onto the SPR chip surface, the dried nanogels were redissolved in the running buffer of PBST (PBS containing

0.01% Tween20). The MIPs were also characterized by FTIR spectroscopy as shown in Fig. S2. FTIR analysis revealed the characteristic amide bands for both MIP-SV and MIP-PQ at 1640 cm⁻1

(amide 1), 1536 cm⁻1 (amide 2), and 1223 cm⁻1 (amide 3). Additionally, the symmetric and asymmetric CH stretch bands are present at 2933 cm⁻1 and 2875 cm⁻1, respectively.

interaction analysis by surface plasmon resonance (SPR)

For a preliminary assessment of the MIP recognition properties, we performed a kinetic interaction analysis using surface plasmon resonance (SPR). Deposition of the nanogels onto the surface

of the SPR chip was achieved through amide coupling on pre-functionalized gold surfaces with a carboxymethyl dextran layer17,31. The carboxyl groups on the chip surface were activated by

EDC/NHS and conjugated with excess nanogels through their amine functionalities provided by the side chains of the APM moieties. Ethanolamine was then used to deactivate any unreacted

carboxyl groups left on the SPR chip surface after nanogel immobilization while washing away the fraction of unbound nanogel. This deposition method is expected to leave a single nanogel

layer on the chip surface with maximum coverage of available binding sites. The nanogel-modified SPR chips were then tested in the multicycle mode with solutions of increasing concentrations

of the target peptides SV and PQ with intermediate regeneration of the sensors using a pH 2 glycine buffer. The SPR plots showing the RU changes occurring upon injections of five different

concentrations of the target peptides are displayed in Fig. S3. The overall equilibrium dissociation constant (KD) values for the target interacting with their nanogels were calculated from

the curves using a 1:1 kinetic model, leading to KD:s of 85 nM and 92 nM for PQ and SV, respectively (Table 2). These results are consistent with previously published values for nanogels

imprinted for peptides and demonstrate the capacity of MIP-SV and MIP-PQ to rebind their template epitopes with high affinity31. Cross-reactivity SPR experiments with swapped templates (SV

for MIP-PQ-conjugated sensors and PQ for MIP-SV-conjugated sensors) resulted in nearly fourfold increases in KD values, which support the selectivity of the nanogels for their corresponding

target peptides. Nevertheless, both SV and PQ exhibit non-specific nanomolar dissociation constants to the conjugated sensors, which might arise from intermolecular interactions involving

the nanogel surface or the highly polar groups available from the ethanolamine or uncovered carboxymethyl dextran layers on the SPR chip.

constant (KD) of the nanogels obtained from SPR data presented in Fig. S3. All experiments were performed under ambient conditions and with three replicates.Full size tableNanogels binding

affinity towards target peptides and proteins assessed by quartz crystal microbalance (QCM)

QCM was used for quantifying the interaction between the nanogels and their epitope templates, the full hCG protein, and the decoy protein LH20,21,45 For the measurements, we adopted the

most commonly reported technique based on template-modified sensor chips and the MIP injected in the running buffer. First, the sensor surface was modified with a mixed Self-Assembled

Monolayer (SAM) of mercaptohexadecanoic acid (MHA) and mercaptoundecanol (MU), the latter serving as a filler for improved ligand accessibility (Fig. S4). Subsequently, the ligands (SV, PQ,

hCG or LH) were immobilized via EDC/NHS catalyzed coupling through their amine functionalities. The progression of this step was monitored through changes in resonant frequency (ΔF), with

estimated surface coverages of 4.9·10–11 mol·cm-2 and 4.1·10–11 mol·cm-2 for PQ and SV, respectively. For hCG and LH, the estimated immobilization ratios were 1.2·10–12 mol·cm-2 and

1.3·10–12 mol·cm-2, respectively (Fig. S5).

The epitope-functionalized QCM chips were first used to assess the binding of MIP-PQ and MIP-SV to their corresponding templates (Fig. 5A,B). Increasing concentrations of colloidal MIP

solutions were introduced at predetermined time intervals, and the frequency change (ΔF), which is proportional to mass uptake on the sensor surface, was monitored in real time. Introducing

the MIPs elicited pronounced frequency decreases relative to the injection of dispersions of non-imprinted nanoparticles (NIP) of identical chemical composition. This highlights the

relevance of the SV- and PQ-imprinted cavities to trigger the template recognition over non-specific intermolecular interactions on the nanogel surface. Parallel experiments with swapped

nanogels-chips pairs (MIP-SV against PQ-functionalized chips and MIP-PQ against SV-functionalized chips) also resulted in significant frequency changes compared to NIPs, which suggests the

presence of non-specific binding cavities capable of hosting flexible peptide moieties. Apparent dissociation constants (KD) and maximum frequency shifts (∆Fmax) were estimated for each

system by fitting non-equilibrium frequency changes against injected concentrations to a Langmuir adsorption isotherm model (Fig. 6). This analysis yielded a KD of ca 8.9 and 30 nM

respectively for MIP-PQ and MIP-SV for their corresponding peptide templates (Table 3). These values are in the same order of magnitude as those determined by SPR, with expected variations

arising from the QCM-associated errors due to a lack of equilibrium during data acquisition, making SPR more robust for kinetic and affinity characterization. To demonstrate the binding

specificity, we performed a competitive inhibition test by pre-incubating the nanogels with soluble PQ or SV peptides (Fig. 5A,B, labels MIP-PQ + PQ and MIP-SV + SV). Pre-incubation

significantly reduced the subsequent binding of MIP-SV and MIP-PQ to the template-functionalized sensors, implying that epitope occupancy of imprinted sites inhibits further nanogel

interactions. These results provide compelling evidence that the MIPs achieve the design goal of selective, epitope-specific binding to the templates.

Real-time resonant frequency changes ΔF (Hz) after repeated injections of MIP-PQ, MIP-SV, and NIP dispersions of increasing concentration on functionalized QCM-D sensor chips modified with

(A) PQ, (B) SV, (C) hCG, and (D) LH. Figures (A) and (B) also show the frequency changes upon injecting nanogels dispersions pre-incubated with 0.1 mg·mL-1 peptide solutions (MIP-PQ + PQ and

MIP-SV + SV, respectively).

(A) Apparent non-equilibrium binding isotherms of MIP-PQ on a PQ (blue circles) and hCG (red squares) functionalized sensor chip and MIP-SV on an SV (green diamonds) and hCG (red circles)

functionalized sensor chip, obtained from the QCM experiments in Fig. 5. The data were fitted to a 1:1 Langmuir adsorption model to determine the apparent dissociation constants given in

Table 3. (B) Maximum frequency changes registered for the different sensor chips at a nanogel concentration of 40 nM.

(KD) of the nanogels interacting with their corresponding epitopes and hCG from the QCM data presented in Fig. 6A. All experiments were performed under ambient conditions and with three

replicates.Full size table

Additional QCM experiments were conducted to assess the nanogel affinity for hCG and LH proteins. To that aim, the two proteins were immobilized on the sensor chips. Following

concentration-dependent exposures of MIP-SV and MIP-PQ, the hCG-functionalized sensor led to pronounced frequency drops, significantly exceeding those of NIPs (Fig. 5C). Apparent KD values

for hCG were estimated to 4.4 nM for MIP-PQ and 49 nM for MIP-SV (Table 3) (Fig. 6A). Parallel experiments using the LH-modified sensor chip produced significantly lower resonance frequency

drops (Fig. 5D,6B), highlighting the selectivity of the MIPs towards the target protein hCG. The combined outcomes of QCM experiments with hCG- and LH-modified sensors suggest a higher

protein discrimination capacity for MIP-PQ, which can be attributed to either an unfavorable protein immobilization masking the SV epitope or an intrinsically lower affinity of MIP-SV for

the epitope.

A solid-phase epitope-based synthetic approach was applied to obtain water-soluble molecularly imprinted nanoparticles for hCG detection. MD simulations were used to investigate the

structure of the immobilized peptide templates within the context of pre-polymerization mixtures. Our findings strongly indicate that the chosen peptides have conformational properties that

mimic the dynamic behavior of the epitopes in the full protein, thereby supporting their potential to imprint molecular cavities suitable for hCG recognition. Template-monomer interactions

were found to be driven by TBAm and NIPAm, with a minor contribution of BIS and negligible interactions with AA and APM monomers. These findings offer an opportunity for optimizing the

pre-polymerization mixture by varying the monomer types or proportions to maximize the interactions with the template.

The synthesized MIPs have homogeneous size distributions and exhibit high binding affinity, and specificity toward the SV and PQ epitopes and hCG as shown through SPR and QCM analysis. The

selectivity towards hCG over other proteins and potential interferants, including LH, was confirmed. Overall, the study suggests the feasibility of using imprinted nanogels as a class of

cost-effective, stable alternatives to natural antibodies for hCG detection. In view of the receptor´s robustness and the demonstrated SPR sensors compatibility with multiple regeneration

cycles, we foresee applications of the MIPs in reusable pregnancy tests and other hCG-related disease diagnostics.

All data generated or analysed during this study are included in this published article [and its supplementary information files].

References Cole, L. A. New discoveries on the biology and detection of human chorionic gonadotropin. Reprod. Biol. Endocrinol. https://doi.org/10.1186/1477-7827-7-8 (2009).

Article  PubMed  PubMed Central  MATH  Google Scholar 

Cole, L. A. Biological functions of hCG and hCG-related molecules. Reprod. Biol. Endocrinol. https://doi.org/10.1186/1477-7827-8-102 (2010).

Article  PubMed  PubMed Central  MATH  Google Scholar 

Gnoth, C. & Johnson, S. Strips of hope: Accuracy of home pregnancy tests and new developments. Geburtshilfe und Frauenheilkunde 74, 661–669. https://doi.org/10.1055/s-0034-1368589 (2014).

Article  CAS  PubMed  PubMed Central  MATH  Google Scholar 

Kennedy, C. E., Yeh, P. T., Gholbzouri, K. & Narasimhan, M. Self-testing for pregnancy: a systematic review and meta-analysis. BMJ Open 12, e054120–e054120.

https://doi.org/10.1136/BMJOPEN-2021-054120 (2022).

Article  PubMed  PubMed Central  Google Scholar 

Stenman, U.-H., Alfthan, H. & Hotakainen, K. Human chorionic gonadotropin in cancer. Clin. Biochem. 37, 549–561 (2004).

CAS  PubMed  MATH  Google Scholar 

Sisinni, L. & Landriscina, M. Advances in Cancer Biomarkers: From biochemistry to clinic for a critical revision (Springer, 2015).

MATH  Google Scholar 

Stenman, U. H., Tiitinen, A., Alfthan, H. & Valmu, L. In Human Reproduction Update (2006).

MATH  Google Scholar 

Berger, P. et al. Candidate epitopes for measurement of hCG and related molecules: The second ISOBM TD-7 workshop. Tumor Biol. 34, 4033–4057. https://doi.org/10.1007/s13277-013-0994-6

(2013).

Article  CAS  MATH  Google Scholar 

Boehringer, H. R. & O’Farrell, B. J. Lateral flow assays in infectious disease diagnosis. Clin. Chem. 68, 52–58. https://doi.org/10.1093/CLINCHEM/HVAB194 (2021).

Article  PubMed  PubMed Central  MATH  Google Scholar 

Fridley, G. E., Le, H. Q., Fu, E. & Yager, P. Controlled release of dry reagents in porous media for tunable temporal and spatial distribution upon rehydration. Lab Chip 12, 4321–4321.

https://doi.org/10.1039/C2LC40785J (2012).

Article  CAS  PubMed  PubMed Central  MATH  Google Scholar 

Dineva, M. A., Candotti, D., Fletcher-Brown, F., Allain, J. P. & Lee, H. Simultaneous visual detection of multiple viral amplicons by dipstick assay. J. Clin. Microbiol. 43, 4015–4021.

https://doi.org/10.1128/JCM.43.8.4015-4021.2005 (2005).

Article  CAS  PubMed  PubMed Central  Google Scholar 

Hoshino, Y. et al. Recognition, neutralization, and clearance of target peptides in the bloodstream of living mice by molecularly imprinted polymer nanoparticles: a plastic antibody. J. Am.

Chem. Soc. 132, 6644–6645. https://doi.org/10.1021/ja102148f (2010).

Article  CAS  PubMed  PubMed Central  MATH  Google Scholar 

Shinde, S. et al. Sialic acid-imprinted fluorescent core-shell particles for selective labeling of cell surface glycans. J. Am. Chem. Soc. 137, 13908–13912.

https://doi.org/10.1021/jacs.5b08482 (2015).

Article  CAS  PubMed  MATH  Google Scholar 

Bui, T. S. et al. Molecularly Imprinted Polymers as Synthetic Antibodies for Protein Recognition: The Next Generation (John Wiley and Sons Inc, 2023).

MATH  Google Scholar 

Zhang, H. Molecularly imprinted nanoparticles for biomedical applications. Adv. Materials 32, 1806328. https://doi.org/10.1002/adma.201806328 (2020).

Article  CAS  Google Scholar 

Koeber, R. et al. Evaluation of a multidimensional solid-phase extraction platform for highly selective on-line cleanup and high-throughput LC-MS analysis of triazines in river water samples

using molecularly imprinted polymers. Anal. Chem. 73, 2437–2444. https://doi.org/10.1021/ac001483s (2001).

Article  CAS  PubMed  MATH  Google Scholar 

Blackburn, C., Sullivan, M. V. & Wild, M. I. Utilisation of molecularly imprinting technology for the detection of glucocorticoids for a point of care surface plasmon resonance (SPR) device.

Analytica Chimica Acta. 1285, 342004. https://doi.org/10.1016/j.aca.2023.342004 (2024).

Article  CAS  PubMed  Google Scholar 

Li, Q. et al. Selective detection of phospholipids using molecularly imprinted fluorescent sensory core-shell particles. Sci. Rep. 10, 9924. https://doi.org/10.1038/s41598-020-66802-3

(2020).

Article  ADS  CAS  PubMed  PubMed Central  MATH  Google Scholar 

Halhalli, M. R., Schillinger, E., Aureliano, C. S. A. & Sellergren, B. Thin walled imprinted polymer beads featuring both uniform and accessible binding sites. Chem. Mater. 24, 2909–2919.

https://doi.org/10.1021/cm300965t (2012).

Article  CAS  Google Scholar 

Hoshino, Y., Kodama, T., Okahata, Y. & Shea, K. J. Peptide imprinted polymer nanoparticles: a plastic antibody. J. Am. Chem. Soc. 130, 15242–15243 (2008).

CAS  PubMed  Google Scholar 

Zeng, Z., Hoshino, Y., Rodriguez, A., Yoo, H. & Shea, K. J. Synthetic polymer nanoparticles with antibody-like affinity for a hydrophilic peptide. ACS Nano 4, 199–204 (2010).

CAS  PubMed  PubMed Central  Google Scholar 

Mier, A. et al. Molecularly imprinted polymer nanogels for protein recognition: direct proof of specific binding sites by solution STD and WaterLOGSY NMR spectroscopies. Angewandte Chem.

Int. Ed. 60, 20849–20857. https://doi.org/10.1002/anie.202106507 (2021).

Article  CAS  Google Scholar 

Rachkov, A. & Minoura, N. Recognition of oxytocin and oxytocin-related peptides in aqueous using a molecularly imprinted polymer the epitope approach. J. Chromatogr. A. 889(111), 118 (2000).

Google Scholar 

Titirici, M. M., Hall, A. J. & Sellergren, B. Hierarchical imprinting using crude solid phase peptide synthesis products as templates. Chem. Materials 15, 822–824 (2003).

CAS  Google Scholar 

Nishino, H., Huang, C.-S. & Shea, K. J. Selective protein capture by epitope imprinting. Angewandte Chemie, International Edition 45, 2392–2396 (2006).

CAS  PubMed  MATH  Google Scholar 

Pasquardini, L. & Bossi, A. M. Molecularly imprinted polymers by epitope imprinting: a journey from molecular interactions to the available bioinformatics resources to scout for epitope

templates. Analytical Bioanalytical Chem. 413, 6101–6115. https://doi.org/10.1007/s00216-021-03409-1 (2021).

Article  CAS  MATH  Google Scholar 

Poma, A. et al. Solid-phase synthesis of molecularly imprinted polymer nanoparticles with a reusable template–“plastic antibodies”. Adv. Functional Mater. 23, 2821–2827.

https://doi.org/10.1002/adfm.201202397 (2013).

Article  CAS  MATH  Google Scholar 

Ambrosini, S., Beyazit, S., Haupt, K., Bui, T. S. & B.,. Solid-phase synthesis of molecularly imprinted nanoparticles for protein recognition. Chem. Commun. 49, 6746–6748.

https://doi.org/10.1039/C3CC41701H (2013).

Article  CAS  MATH  Google Scholar 

Berghaus, M., Mohammadi, R. & Sellergren, B. Productive encounter: molecularly imprinted nanoparticles prepared using magnetic templates. Chem. Commun. 50, 8993–8996.

https://doi.org/10.1039/C4CC01346H (2014).

Article  CAS  Google Scholar 

Mahajan, R. et al. Highly efficient synthesis and assay of protein-imprinted nanogels by using magnetic templates. Angew. Chem. Int. Ed. Engl. 58(727), 730.

https://doi.org/10.1002/anie.201805772 (2019).

Article  CAS  MATH  Google Scholar 

Hix-Janssens, T., Davies, J. R., Turner, N. W., Sellergren, B. & Sullivan, M. V. Molecularly imprinted nanogels as synthetic recognition materials for the ultrasensitive detection of

periodontal disease biomarkers. Analytical Bioanalytical Chem. https://doi.org/10.1007/s00216-024-05395-6 (2024).

Article  Google Scholar 

Dirnhofer, S. et al. The molecular basis for epitopes on the free β-subunit of human chorionic gonadotrophin (hCG), its carboxyl-terminal peptide and the hCGβ-core fragment. J. Endocrinol.

141, 153–162. https://doi.org/10.1677/JOE.0.1410153 (1994).

Article  CAS  PubMed  Google Scholar 

Klonisch, T. et al. Relative location of epitopes involved in synergistic antibody binding using human chorionic gonadotropin as a model. Eur. J. Immunol. 26, 1897–1905.

https://doi.org/10.1002/EJI.1830260834 (1996).

Article  CAS  PubMed  MATH  Google Scholar 

Valmu, L., Alfthan, H., Hotakainen, K., Birken, S. & Stenman, U. H. Site-specific glycan analysis of human chorionic gonadotropin β-subunit from malignancies and pregnancy by liquid

chromatography - Electrospray mass spectrometry. Glycobiology 16, 1207–1218. https://doi.org/10.1093/glycob/cwl034 (2006).

Article  CAS  PubMed  Google Scholar 

Lund, H. et al. Epitope analysis and detection of human chorionic gonadotropin (hCG) variants by monoclonal antibodies and mass spectrometry. Tumor Biol. 35, 1013–1022.

https://doi.org/10.1007/S13277-013-1135-Y/FIGURES/4 (2014).

Article  CAS  Google Scholar 

Canfarotta, F., Poma, A., Guerreiro, A. & Piletsky, S. Solid-phase synthesis of molecularly imprinted nanoparticles. Nature Protocols 11, 443. https://doi.org/10.1038/nprot.2016.030 (2016).

Article  CAS  PubMed  Google Scholar 

Mahajan, R., Suriyanarayanan, S. & Nicholls, I. A. Improved Solvothermal Synthesis of γ-Fe(2)O(3) Magnetic Nanoparticles for SiO(2) Coating. Nanomaterials (Basel)

https://doi.org/10.3390/nano11081889 (2021).

Article  PubMed  MATH  Google Scholar 

Udenfriend, S. et al. Fluorescamine: a reagent for assay of amino acids, peptides, proteins, and primary amines in the picomole range. Science 178, 871–872 (1972).

ADS  CAS  PubMed  Google Scholar 

Castell, J. V., Cervera, M. & Marco, R. A convenient micromethod for the assay of primary amines and proteins with fluorescamine. A reexamination of the conditions of reaction. Analytical

Biochem. 99, 379–391 (1979).

CAS  Google Scholar 

Lapthorn, A. J. et al. Crystal structure of human chorionic gonadotropin. Nature 369, 455–461. https://doi.org/10.1038/369455a0 (1994).

Article  ADS  CAS  PubMed  Google Scholar 

Martínez, L., Andrade, R., Birgin, E. G. & Martínez, J. M. PACKMOL: A package for building initial configurations for molecular dynamics simulations. J. Comput. Chem. 30, 2157–2164 (2009).

PubMed  MATH  Google Scholar 

Roe, D. R. & Cheatham Iii, T. E. PTRAJ and CPPTRAJ: Software for processing and analysis of molecular dynamics trajectory data. J. Chem. Theory Comput. 9, 3084–3095 (2013).

CAS  PubMed  MATH  Google Scholar 

Humphrey, W., Dalke, A. & Schulten, K. VMD: visual molecular dynamics. J. Mol. Graph. 14, 27–38. https://doi.org/10.1016/0263-7855(96)00018-5 (1996).

Article  Google Scholar 

Bedwell, T. S. et al. New protocol for optimisation of polymer composition for imprinting of peptides and proteins. RSC Adv. 9, 27849–27855. https://doi.org/10.1039/c9ra05009d (2019).

Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

Koide, H. et al. A polymer nanoparticle with engineered affinity for a vascular endothelial growth factor (VEGF(165)). Nat. Chem. 9, 715–722. https://doi.org/10.1038/nchem.2749 (2017).

Article  CAS  PubMed  MATH  Google Scholar 

Chianella, I. et al. Direct replacement of antibodies with molecularly imprinted polymer nanoparticles in ELISA—development of a novel assay for vancomycin. Analytical Chem. 85, 8462–8468.

https://doi.org/10.1021/ac402102j (2013).

Article  CAS  MATH  Google Scholar 

Download references

We acknowledge financial support from EU through the WomenTechEU program (Pharmista) and stipends from the Sten K. Johnsson Foundation (Surecapture Technologies, Pharmista). Yadiris García

thanks to the financial support of Postdoctoral Project ANID/FONDECYT 3200508. Verónica Jiménez thanks Agencia Nacional de Investigación y Desarrollo through Grant EXPLORACION 13220020 for

financial support. Francesco Patitucci was funded by PON “Ricerca e Innovazione” 2014–2020, Asse IV “Istruzione e ricerca per il recupero”, Azione IV.4—“Dottorati e contratti di ricerca su

tematiche dell’innovazione”.

Open access funding provided by Malmö University.

European Commission WomenTechEU, Sten K. Johnsson foundation grants to, Pharmista and Surecapture Technologies, ANID/FONDECYT,3200508,Ricerca e

Innovazione,2014–2020,EXPLORACION,13220020,13220020.

Author informationAuthors and Affiliations Pharmista Technologies AB, Scheelevägen 3, 223 63, Lund, Sweden

Radvilė Zubrytė, Robert Porter & Alice Mattsson

Surecapture Technologies AB, Per Albin Hanssons Väg 35, 214 32, Malmö, Sweden

Radvilė Zubrytė, Liliia Mavliutova & Börje Sellergren

Biofilms Research Center for Biointerfaces, Malmö University, Per Albin Hanssons Väg 35, 214 32, Malmö, Sweden

Yadiris García, Francesco Patitucci & Börje Sellergren

Departamento de Ciencias Químicas, Facultad de Ciencias Exactas, Universidad Andres Bello, Autopista Concepción-Talcahuano 7100, Talcahuano, Chile

Yadiris García, Laura C. Polania & Verónica A. Jiménez

University of Sheffield, Dainton Building, Brook Hill, Sheffield, S3 7HF, Great Britain

Mark V. Sullivan & Nicholas W. Turner

Department of Pharmacy, Health and Nutritional Sciences, University of Calabria, Arcavacata, 87036, Rende, (CS), Italy

Francesco Patitucci

AuthorsRadvilė ZubrytėView author publications You can also search for this author inPubMed Google Scholar

Liliia MavliutovaView author publications You can also search for this author inPubMed Google Scholar

Yadiris GarcíaView author publications You can also search for this author inPubMed Google Scholar

Mark V. SullivanView author publications You can also search for this author inPubMed Google Scholar

Nicholas W. TurnerView author publications You can also search for this author inPubMed Google Scholar

Francesco PatitucciView author publications You can also search for this author inPubMed Google Scholar

Laura C. PolaniaView author publications You can also search for this author inPubMed Google Scholar

Verónica A. JiménezView author publications You can also search for this author inPubMed Google Scholar

Robert PorterView author publications You can also search for this author inPubMed Google Scholar

Alice MattssonView author publications You can also search for this author inPubMed Google Scholar

Börje SellergrenView author publications You can also search for this author inPubMed Google Scholar

A.M. and R.P. conceived the idea. B.S. designed the initial strategy, wrote the final version of the manuscript, directed the project and designed the part of the experiments. L.M. screened

and evaluated the hCG epitopes and designed and tested the first generation MIPs. R.Z. carried out most experiments relating to MIP synthesis and characterisation as well as protein binding

tests. Y.G. carried out MIP synthesis and characterisation. M.S. performed the SPR investigation and evaluated the results. N.T. supervised the SPR experiments and advised on the general

strategy. F.P. performed the QCM investigation and evaluated the results. L.P. assisted the dynamic simulation experiments. V.J. performed the dynamic simulations and contributed to writing

of the paper. R.P. advised on the general strategy of the work.

Corresponding author Correspondence to Börje Sellergren.

Pharmista AB is a holder of a patent protecting the disclosed technology. The other authors have no competing interests.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or

format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or

other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in

the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the

copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this articleCite this article Zubrytė, R., Mavliutova, L., García, Y. et al. Development of molecularly imprinted polymers for the detection of human chorionic gonadotropin. Sci Rep

15, 10436 (2025). https://doi.org/10.1038/s41598-025-94289-3

Download citation

Received: 08 October 2024

Accepted: 12 March 2025

Published: 26 March 2025

DOI: https://doi.org/10.1038/s41598-025-94289-3

Share this article Anyone you share the following link with will be able to read this content:

Get shareable link Sorry, a shareable link is not currently available for this article.

Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative