Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID- 19. The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information website. Elsevier hereby grants permission to make all its COVID-19-related research that is available on the COVID-19 resource centre - including this research content - immediately available in PubMed Central and other publicly funded repositories, such as the WHO COVID database with rights for unrestricted research re-use and analyses in any form or by any means with acknowledgement of the original source. These permissions are granted for free by Elsevier for as long as the COVID-19 resource centre remains active. Colloids and Surfaces B: Biointerfaces 193 (2020) 111126 Contents lists available at ScienceDirect Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb Simultaneous optical and magnetophoretic monitoring of DNA T hybridization using superparamagnetic and plasmonic colloids Maria Benelmekkia,b,*, Sergi Gassob, Lluis M. Martinezc a College of Engineering, Swansea University, Bay Campus, Fabian Way, Swansea, UK b Pragmatic Diagnostics, Parc de Recerca, Campus UAB, E-08193 Bellaterra, Spain c Sepmag, ParcTecnologic del Valles, E-08290 Barcelona, Spain A R T I C LE I N FO A B S T R A C T Keywords: The detection and separation of small biomolecules from complex mixtures and the possibility of their re- Magnetophoresis covering for further analyses have great benefits for the early diagnosis and prognosis of diseases. Developing Optical-monitoring simple, sensitive, and cost-effective tools that allow the rapid and accurate assembly and isolation of molecular Oligonucleotide-hybridization biomarkers has the potential to improve both patient care and hospital logistic efficiency towards personalized Surface-functionalization and affordable treatments of diseases.In this work, we presenta method consisting ofUV–vis-spectroscopy as- Colloidal-stability Magnetic-separation sisted-magnetophoresis for the monitoring of DNA hybridization. For this purpose, a magnetic device generating Magnetic-isolation 7.5 T/m uniform magnetic field gradient was designed and incorporated to a commercial spectrophotometer. Molecular-diagnosis Different batches of colloidal superparamagnetic particles (SMPs), with different elemental compositions, were functionalized with twenty-mer complementary oligonucleotides, TB1 and TB2. When the functionalized SMPs- TB1 and SMPs-TB2 are mixed and incubated, the hybridization process of TB1 and TB2 occurs resulting in the formation of colloidal aggregates. When brought under the magnetic field, depending on the magnetic strength (Γ) of the formed aggregates, they separate either faster or slower than the non-functionalized SMPs. The dif- ference in magnetic separation time (Δt) is optically monitored by measuring the real time transparency of the suspension at specific wavelengths. The detection of aggregates at concentrations of 0.001% w/v was achieved, showing |Δt| ranging from 113−228 s. Based on the changes of Δt, the study addresses how electrosteric, magnetic, and hydrogen bonding interactions affect the hybridization process and suggests optimum experi- mental conditions for accurate monitoring of TB1-TB2 hybridization. 1. Introduction expected to provide solutions for early diagnosis and targeted therapy. Such expectations are projected in different paths and include in- Within the field of biomedical applications, concepts and tools de- expensive rapid tests for genetic predisposition, viral infection and the rived from nanotechnology and nanomaterials have been applied to first signs of diseases long before symptoms manifest themselves [6]. overcome the problems of conventional techniques for advanced diag- Biomarkers are very important as they allow early diagnosis and in- nosis and prognosis of diseases. In particular, merging nanomaterials tervention prior to the occurrence of potentially serious clinical events. with different properties have opened a new horizon for multi- They can be used to establish the prognosis of patient and help to plan functional nanostructures within a single platform. For example, novel the treatment. In addition they can provide early information about the nanostructures that combine, plasmonic and magnetic properties can efficacy of treatment and indicate whether to continue, switch or stop lead to high sensitive and cost-effective biosensors, providing important the therapy [7,8]. For example, in the field of oncology, when tumor improvements in patient care and at the same time reduce costs, con- cells die, they release circulating tumor DNA (ctDNA) into the blood. tribute to the efficiency of the hospital logistics, and enhance safety, Scientists recourse to liquid biopsy to detect and purify ctDNA, thus, the allowing the early detection of specific biomarkers at a single molecule purified biomaterial is analyzed using advanced and complex technique level [1–5]. in the field of genomics such as digital PCR, next generation sequencing With the increasing interest in the development and implementation (NGS), fluorescence in situ hybridization (FISH) and BEAMing [9,10]. of accessible and affordable “personalized medicine”, nanomedicine is However, these techniques require an operation by specialized ⁎ Corresponding author at: College of Engineering, Swansea University, Bay campus, Fabian Way, Swansea, UK. E-mail address: maria.benelmekki@swansea.ac.uk (M. Benelmekki). https://doi.org/10.1016/j.colsurfb.2020.111126 Received 27 March 2020; Received in revised form 3 May 2020; Accepted 9 May 2020 Available online 15 May 2020 0927-7765/ © 2020 Elsevier B.V. All rights reserved. M. Benelmekki, et al. Colloids and Surfaces B: Biointerfaces 193 (2020) 111126 personnel in large infrastructures such as hospitals or research centres. were purchased from Spherotech (Sphero™ CM-025-10H), and used as Currently, most of the reported results are performed in optimal la- magnetic carriers. The initial concentration of the as received MPs is 1% boratory settings as a proof-of-concept for specific biomarkers, instead (w/v). Magneto-plasmonic nanoparticles (NPs) of Fe core coated with of using complex environment (e.g., whole blood, urine, cells), where Au shell (hereafter FeAu-NPs) of ∼50 nm were purchased from the mixture of biomolecules and ions often provoke false signals and Nanoimmunotech (NITmagold Cit 50 nm). The Fe core is ∼13 nm of reduced the sensitivity [11]. Extensive studies have proven that plas- diameter, and the thickness of the Au shell is ∼38 nm. The surface of monic nanoparticles (especially gold) exhibit properties that facilitate the Au-shell is modified with citrate anions. The initial concentration of their implementation in molecular assays for the detection of genetic the as received FeAu-NPs is 0.02% (w/v). All chemicals were of ana- mutations in biological samples. However the response of those sys- lytical grade and used without further purification. Sodium phosphate, tems, relies exclusively on the color change of the sample [12,13]. EDAC (N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydro- The term “Magnetophoresis” is used to describe the motion of chloride)), MES (2-(N-morpholino)ethanesulfonic acid),and Imidazole magnetic particles when brought under a magnetic field gradient. where purchased from Sigma-Aldrich. The twenty-mer complementary Nowadays, magnetophoresis is being used in a wide range of research oligonucleotides were purchased from Sigma Life Science. Their se- and technological areas ranging from environment remediation to quences are: Sequence (5′to 3′) Amino-CCCCCATTCGTACCATTTTT biomedical applications [14,15]. Numerous applications, encompassing (probe TB1), and Sequence (5′to 3′) Amino-AAAAATGGTACGAATGG wastewater treatments [16–18], pollutant removal [19], biomolecules GGG (probe TB2). isolation, drug delivery, and magnetic-particle imaging [20–22], have Zeta potential measurements were performed with a Zetasizer Nano been developed. Within biomedical applications, magnetic separation ZS (Malvern instruments), provided by a He/Ne laser of 633 nm wa- enables the separation and isolation of targeted biological materials velength. Hysteresis loops were measured with a superconducting such as proteins, cells, small biomolecules from complex mixtures, by quantum interference device (SQUID) magnetometer (Quantum Design the use of superparamagnetic colloids. Generally, these super- MPMS5XL). The experimental results were corrected for the holder paramagnetic colloids present a spherical shape with sizes ranging from contribution. UV–vis spectra were performed using Evolution 201 a few tens of nanometers to microns. These nanostructures, are com- Thermo Scientific spectrophotometer. monly made by embedding superparamagnetic nanocrystals in a non- magnetic matrix such as polystyrene or nanoporous silica [23–26]. 2.2. Conjugation of the particles to TB1 and TB2 probes Other nanostructures consist of magnetic nanoparticles coated with inorganic materials such as Au, Ag or silica, resulting in core/shell Generally speaking, the conjugation process consists of five prin- nanostructures [27,28]. Prior to their use as magnetic carriers, the cipal steps [32]: 1) washing of the as received particles, 2) surface ac- surfaces of these nanostructures are modified, so that they bind to the tivation of the particles, 3) coupling (or coating) of the particles with targeted biological material in a solution to produce magnetic biolo- the probes, 4) blocking process to avoid any excess reactive sites, 5) gical composites. In the particular case of genetic and pathogenic dis- resuspension of the functionalized particles in a suitable storage buffer. ease detection, the surface of the particles can be modified by grafting All the conjugation protocols used in this work are performed according complementary oligonucleotides. Thus, a magnetic field is applied to the technical recommendations in ref [32]. the sample (commonly a complex colloidal mixture) to drive the magnetic biological complex towards selected regions of the containing 2.2.1. Polystyrene (PS) magnetic particle (MPs) conjugation vessel, enabling the separation of the solid content from the liquid The as received MPs were washed with a 50 mM sodium phosphate mixture, and thus the removal of the liquid phase [29–31]. buffer (pH 7.2), and magnetically separated. 100 μM (100 μmole/L) The emphasis of the present work is on the magnetophoretic se- suspensions of TB1 and TB2 probes were prepared in sterilized H2O paration, and isolation of small biomolecules such as single-stranded- (Milli-Q). For TB1 coupling to MPs, 200 μg of MPs were suspended in DNA. Particularly, this report addresses a new and sensitive UV–vis 180 μL of the coupling suspension (0.1 M imidazole buffer pH = 7.2, spectroscopy assisted-magnetophoresis method for the monitoring of 0.1 M EDAC (N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydro- the hybridization process of twenty-mer complementary oligonucleo- chloride)), then 20 μL of TB1 suspension (100 μmole/L) was added. The tides (hereafter TB1 and TB2). For this purpose, two batches of mag- sample was incubated for 16 h at room temperature with gentle netic particles are separately functionalized with TB1 and TB2. When shaking. For TB2 coupling to MPs, the same procedure was followed: the magnetic particles from two different batches are mixed and in- 200 μg of MPs were suspended in 180 μL of the coupling suspension cubated, colloidal aggregates composed of the magnetic particles, TB1, (0.1 M imidazole buffer pH = 7.2, 0.1 M EDAC (N-(3-dimethylamino- and TB2 are formed as a result of the hybridization process. The ag- propyl)-N'-ethylcarbodiimide hydrochloride)), then 20 μL of TB2 sus- gregates, when brought under a uniform magnetic field gradient, move pension (100 μmole/L) was added, and the sample was subject to 16 h at a different velocity than the initial non-functionalized magnetic incubation at room temperature under gentle shaking. The MPs func- particles. The velocity of the formed agglomerates depends on their tionalized with TB1 and TB2 were washed separately and resuspended resulting magnetic strength, which is quantified with the parameter Γ. in 100 mM Tris (tris(hidroximetil)aminometano)buffer to block excess This behavior induces an important difference in the magnetic se- reactive sites. (Reacted at room temperature for 2 h with gentle paration time of the formed aggregates, which is optically monitored by shaking). Finally the particles were washed and resuspended in 50 mM measuring the real time transparency of the colloidal suspension at sodium phosphate buffer (separately). specific wavelengths. The method is highly sensitive as it allows the detection of aggregates at concentrations of 0.001% w/v. Moreover, the 2.2.2. Magneto-plasmonic FeAu-NPs conjugation aggregates concentrate in the region where the magnetic field is max- The as received FeAu-NPs were washed with a 50 mM sodium imum (on the walls of the recipient containing the suspension), al- phosphate buffer (pH 7.5). For TB1 coupling, 5.024 μg of FeAu-NPs lowing their collection for further analyses. were suspended in 200 μL of the coupling suspension (0.1 M imidazole buffer pH = 7.2, 0.1 M EDAC (N-(3-dimethylaminopropyl)-N'-ethyl- 2. Experimental details carbodiimide hydrochloride), then 0.31 μL of TB1 suspension (100 μM) was added. The sample was incubated for 16 h at room temperature 2.1. Materials and methods with gentle shaking. The same procedure was used for the conjugation of FeAu-NPs to TB2. The functionalized FeAu-NPs were washed sepa- Carboxyl polystyrene (PS) magnetic particles (hereafter MPs) of rately and resuspended in 100 mM Tris (tris(hidroximetil)aminome- ∼140 nm of diameter, ∼20% of ferrite content, density ∼1.24 g/cm3, tano)buffer for 2 h with gentle shaking to block excess reactive sites. 2 M. Benelmekki, et al. Colloids and Surfaces B: Biointerfaces 193 (2020) 111126 Fig. 1. Setup used for magnetophoresis mea- surements. (1) Shows the magnetic accessory designed to place inside the commercial spec- trophotometer used in this study. (2) The dis- tribution of the magnetic field gradient in the half-height plane crossing the set of magnets inside the device in (1), indicating the position of the sample (cuvette), and the light trajectory trough the sample. (3) Illustrates an overview on the magnetic separation steps using the device in (1); (a) the cuvette containing the magnetic particle suspension (brown balls) is introduced into the system illustrated in (1); (b) top-view showing the initial distribution of the particles, the incident light is absorbed (blocked) by the particles; (c) the particles are driven by the magnetic force and move to the wall of the cuvette, then the light absorbance is partially reduced; (d) when all the particles reach the wall, the suspension reaches a max- imum transparency (minimum absorbance), indicating the end of the separation process. Finally the NPs were washed and resuspended in 50 mM sodium measurement. Prior to magnetophoresis measurements, a standard phosphate buffer (separately). transmittance spectrum of the suspension (in the range 250–900 nm) was collected (Fig. 2-b). The wavelengths 350 nm, 580 nm, and 750 nm 2.3. Opto-magnetophoresis experimental setup were selected to collect the magnetophoresis data. Fig. 2-c shows the obtained magnetophoresis curves of the suspension. The curves show a The magnetophoresis setup employed in the present study is the similar behavior consisting of a progressive increase of the suspension SEPMAG QUV7.5 device (7.5 T/m) [33]. The system consists of a cu- transmittance. The transmittance increases by 50% respect to the initial bical cavity containing a permanent magnetic field with a uniform ra- state at a time t50 of 255.04 ± 2.67 s at 350 nm, 239.44 ± 0.86 s at dial gradient pointing toward the walls of the cubical non-magnetic- 580 nm, and 236.86 ± 0.98 s at 750 nm. The time t50 is slightly larger support, as shown in Fig. 1. The magnetic force within the magnetic when the data are collected at 350 nm wavelength, which is likely due device is generated by a set of permanent magnets placed in a ring-like to the strong absorbance of iron at this wavelength range [35,36]. In distribution following the Halbach progression for a quadrupole [34]. order to obtain the maximum contrast in separation times, all the The magnetic device is placed inside the spectrophotometer, and sub- magnetophoresis measurements presented in this study were collected sequently, a standard polystyrene cuvette containing the MPs suspen- at 350 nm wavelength. In addition, both MPs and FeAu-NPs are likely to sion (the sample) is introduced into the device in the position as illu- show a strong absorbance around 350 nm wavelength as they both strated in Fig. 1–1. An empty cuvette is used as a reference. Pictures contain iron component. showing the experimental setup are presented in figure S1 of Sup- Previous works reported on two different models for magneto- porting Information. The assembly generates a constant gradient in the phoretic separation kinetics [37–39]. A fast separation kinetics driven sample volume (Fig. 1–2). The magnetic device presents symmetric by a reversible aggregation of the MPs, and a slow separation kinetics apertures for the entrance and exit of the beam light. The light is dominated by a single particle magnetophoretic motion. Within this transmitted in a direction perpendicular to the magnetic force direction. context, a new theoretical frame work was developed by Andreu et al. The design of the device allows the insertion of the sample (cuvette) in [38] for the prediction of the aggregation behavior of MPs, when they the working area where the magnetic field is higher than the saturation are brought under a uniform magnetic field gradient. Briefly, the de- magnetic field, and thus the MPs become saturated. veloped analytical model suggests that the aggregation behavior of The transmittance of the initial suspension increases progressively, colloidal magnetic particles is controlled by a dimensionless parameter reaching a transparent final state with all particles close to the wall of N* defined as: the cuvette (Fig. 1-3d). The transmittance measurements are performed using the UV–vis lamp of the spectrophotometer as a source of light. N* = ∅0 e Γ − 1 (1) The light beam passes through the suspension, in a perpendicular di- where Φ0 is the volume fraction of particles uniformly dispersed in the rection of the generated magnetic force, and reaches the spectro- solution, and Γ is the magnetic strength constant defined as: photometer detector (Fig. 1), while the data of the transmittance of the suspension versus time is collected [33]. μ0 ms2 Γ= 2πD3KB T (2) 3. Results and discussions where μ0 is the free space permeability, ms is the magnetic moment of a 3.1. Opto-magnetophoretic behavior of the pristine MPs particle at saturation magnetization, D is the diameter of the particle, KB is the Boltzmann constant, and T the absolute temperature. The as received MPs were washed several times, magnetically se- Physically, Γ is the ratio between the magnetic energy associated to parated, and resuspended in H2O Milli-Q to form a suspension of the dipole-dipole attraction, and the thermal energy. The aggregation of 0.001% (w/v). Their magnetic hysteresis loop shows a super- magnetic colloids depends on Γ and Φ0 through N* [40]. In order to paramagnetic behavior at room temperature (Fig. 1-a). A volume of observe long chain formation (aggregation) of magnetic colloids, the 100 μL of the suspension was introduced into the cuvette to perform the condition N* > 10 should be verified. In the case where 1 < N* < 10, 3 M. Benelmekki, et al. Colloids and Surfaces B: Biointerfaces 193 (2020) 111126 Fig. 2. (a) Magnetization curve showing a superparamagnetic behavior of the pristine MPs at 300 K. (b) UV–vis transmittance spectrum of the pristine MPs suspension. (c) Magnetophoresis behavior under 7.5 T/m magnetic field gradient of 0.001% (w/v) MPs suspen- sion at 350 nm, 580 nm, and 750 nm. The curves show the increase in the suspension transmittance versus the separation time. The horizontal black arrow in (c) indicates when the sus- pension reaches 50% of its transmit- tance. The vertical arrows (dashed line) indicate the corresponding separation times (t50). small chains are expected to form. If N* < 1, aggregation is not likely MPs-TB2. In fact the oligonucleotides are negatively charged due to to occur, and the separation process is induced by the individual motion phosphate (PO2−) groups along their backbones, inducing a stability of the magnetic colloids, resulting in a slow separation process. based on electrostatic repulsions particle-particle as illustrated in Fig. 3- In the case of the MPs used for this study, considering the saturation d, which is confirmed from zeta-potential measurements as shown in magnetization (∼ 17 emu/g) Fig. 2-a, the density of the MPs (∼1.24 g/ Table 1. On the other hand, steric stabilization induced by the anchored cm3), and the concentration of the suspension (0.001% w/v), the cal- TB1 and TB2 on the surface of the MPs is likely to contribute to the culated values of the magnetic strength (Γ0) and the dimensionless observed stability of MPs-TB1 and MPs-TB2 suspensions, resulting in an parameter (N0*) are ∼14.79 and ∼2.82, respectively. Therefore, the electrosteric colloidal stabilization. magnetophoresis process of the pristine MPs is based on a cooperative In the case of pristine MPs (Fig. 3-d (1)), the scheme illustrates the motion driven by the formation of relatively small chains, inducing mechanism of the formation of particle chains when the suspension is separation time t50 of ∼255 s at 350 nm for concentrations of 0.001% brought under a magnetic force. MPs align to form short chains that (w/v). move faster in response the magnetic force. when MPs are functiona- lized with TB1 or TB2 (Fig. 3d-(2) and d-(3)), the induced repulsive ⎯→ ⎯ ⎯→ ⎯ forces between the particles (annotated as F rep1 and F rep2, respec- 3.2. Opto-magnetophoretic monitoring of TB1-TB2 hybridization tively) disturb the formation of chains of particles, which lead to in- dividual magnetophoretic motion of MPs-TB1 and MPs-TB2, and thus, 3.2.1. Hybridization of MPs-TB1 and MPs-TB2 in an important increment in the magnetic separation times. As described in Section 2.2, the MPs were functionalized with For the monitoring of TB1 and TB2 hybridization using the opto- twenty-mer complementary oligonucleotides TB1 and TB2, and two magnetophoresis method, 50 μL of MPs-TB1 and 50 μL of MPs-TB2 separated suspensions of MPs-TB1 and MPs-TB2 were prepared. For the suspensions were mixed and incubated at room temperature for dif- opto-magnetophoresis measurements, different concentration of MPs- ferent times. In all cases, the working concentrations and volumes were TB1 and MPs-TB2 were prepared in H2O Milli-Q, and volumes of 100 μL 0.001% (w/v) and 100 μL, respectively. The suspensions were in- were used for all the measurements. troduced in the magnetic generator of 7.5 T/m (described in Fig. 1), Figs. 3a and b show the obtained separation curve of the suspen- which was previously placed inside the spectrophotometer, after the sions at different concentrations for MPs-TB1 and MPs-TB2, respec- incubation process. The values of the separation times t50 are shown in tively. The separation times t50 decrease exponentially when the con- table S2 of the supporting information. When MPs-TB1 and MPs-TB2 centrations of the suspensions are increased, as illustrated in Fig. 3-c. suspensions (0.001%) are mixed in a ratio 1:1, and as shown in Fig. 4-a, More details are given in the caption of Fig. 3. The values of t50 in each 60 min of incubation time is the threshold for the hydrogen bonding to case are presented in Table S1 of the Supporting Information. occur between the bases A–T (Adenine-Thymine) and C–G (Cytosine- Of note, separation times t50 for both MPs-TB1 and MPs-TB2 sus- Guanine), overcoming the electrosteric repulsion. This hybridization pensions (0.001% w/v) are ∼30% longer in comparison to the sus- results in the formation of agglomerates, as schematically illustrated in pension of pristine MPs. This behavior might be attributed to the kinetic Figs. 4-c and 4-d. When the suspension is brought under the magnetic and thermodynamic colloidal stabilities acquired by the MPs-TB1 and 4 M. Benelmekki, et al. Colloids and Surfaces B: Biointerfaces 193 (2020) 111126 Fig. 3. (a) and (b) Magnetophoresis curves of MPs-TB1 and MPs-TB2 suspensions at different concentrations, respectively. (c) Shows the exponential decrease of separation time when increasing the suspension concentrations. (d) Scheme illustrating (1) the formation of a small chain of pristine MPs when brought under ⎯→ ⎯ magnetic force ( F mag), resulting in a faster separation. (d-2 andD-3) illustrate the repulsive ⎯→ ⎯ ⎯→ ⎯ forces F rep1 and F rep2 between tow particles of MPs-TB1 and MPs-TB2, respectively. Table 1 concentration of pristine FeAu-NPs in buffer suspension, which suggests Zeta potential values of the different suspensions measured at 0.001% con- the formation of chemical bonds between the FeAu-NPs and the TB1 centrations. oligonucleotide. On the other hand, the additional decrease of SPR peak Sample MPs MPs-TB1 MPs-TB2 MPs-TB / MPs-1TB2 (1H intensity after reacting the functionalized FeAu-TB1 NPs with TB2 (as incubation) illustrated in figure S3) further confirms the successful functionaliza- tion of the NPs, in agreement with previous reports [41,42]. Zeta potential (mV) −23.1 −29.4 −31.2 −34.2 Fig. 5-a shows the normalized UV–vis spectra (absorbance) of the Standard Dev. (mV) 3.9 4.5 4.9 5.7 MPs-TB2 and FeAu-TB1 suspensions of concentrations of 0.01% (w/v) and 0.02% (w/v), respectively. Fig. 5-b illustrates the normalized force, the agglomerates move faster toward the walls of the container, UV–vis spectra of the mixture of both suspensions for different times of following a cooperative magnetophoretic mode [37–39]. The velocity incubation. For the preparation of the mixture, 20 μL of MPs-TB2 sus- of separation of MPs-TB1/MPs-TB2 suspension is higher than the ve- pension was mixed with 80 μL the suspension of FeAu-TB1. This pro- locity of MPs-TB1 and MPs-TB2, showing differences Δt1 ranging from portions result in a total volume of 100 μL containing 0.002% w/v of ∼37 s to ∼113 s, depending on the incubation time. MPs-TB2 particles and 0.016% w/v of FeAu-TB1 NPs (ratio ∼ 1:8). However, comparing the separation velocities of pristine MPs and Three samples of the mixture were prepared and incubated for different the MPs-TB1/MPs-TB2, the separation times show smaller differences times. The resulting UV–vis spectra show a red shift of the SPR peaks of (Δt2), observed for incubation times superior to 10 h, as shown in Fig. 4- ∼15 nm and ∼32 nm when incubation times are increased to 3 h and a. this behavior is expected as both samples follow a cooperative 20 h, respectively (Fig. 5-b). A plausible explanation of the observed red magnetophoretic process, where the magnetic particles form small shifts, is the spatial proximity between FeAu-TB1 NPs at the occurrence chain under the magnetic field gradient, and therefore move faster than of the hybridization process on the surface of MPs-TB2 particles, as il- the individual NPs (See Fig. 4) [37–39]. lustrated in the inset in Fig. 5b, providing an efficient optical platform for a strong near-field plasmonic coupling between the FeAu-NPs [3,42]. The position of the SPR peak corresponding to the sample after 3.2.2. Hybridization of FeAu-TB1 and MPs-TB2 1 h of incubation time doesn’t show any shift with respect to the UV–vis FeAu-NPs were functionalized with TB1 and TB2 under the condi- spectrum of FeAu-TB1 (Fig. 5-a), which might be a result of the short tions described in the experimental section, and two separate batches of incubation time. This is in full agreement with our previous discussion FeAu-NPs functionalized with TB1 (FeAu-TB1) and TB2 (FeAu-TB2) in section 3.2.1., where we pointed-out the nature’s role of the different were prepared. UV–vis absorbance spectra corresponding to FeAu-TB1, interactions between the functionalized MPs. At the beginning, a bal- FeAu-TB2 are shown in Figure S2 of the supporting information. ance between the repulsive interactions between FeAu-TB1 NPs and the However, the emphasis of our discussion will be on FeAu-TB1 compo- attractive interactions between FeAu-TB1 and MPs-TB2 would take sites (the results are the same for FeAu-TB2). To confirm the successful place, resulting in a small number of FeAu-TB1 NPs attached to the functionalization of the FeAu-NPs, the changes in the SPR (Surface surface of each MP-TB2. After one hour of incubation time, a new Plasmon Resonance) peak were used. For this purpose, UV–vis spectra balance between the different interactions would be established; the of FeAu-TB1 suspension were collected, and the kinetics of the intensity hydrogen bonding between the bases A–T and C–G is likely to become of the SPR peak were monitored during time. The obtained spectra of dominant, resulting in an increased number of FeAu-TB1 NPs attached FeAu-TB1 suspensions are illustrated in figure S3 of Supporting to the surface of each MP-TB2. The spatial proximity of the FeAu-NPs Information. The spectra show a progressive decay of the intensity of resulting from the increasing number of FeAu-TB1 anchored to the the SPR peak around 530 nm indicating the decrease in the surface of a MP-TB2 leads to a strong near-field plasmonic coupling 5 M. Benelmekki, et al. Colloids and Surfaces B: Biointerfaces 193 (2020) 111126 Fig. 4. (a) Evolution of separation time (t50) versus incubation time of 0.001% w/v con- centrations of MPs-TB1 suspension (red line), MPs-TB2 suspension (blue line) and MPs-TB1/ MPs-TB2 suspension (black line). Δt1 and Δt2 indicate the separation time differences be- tween the mixture of MPs-TB1/MPs-TB2, MPs- TB1 (or MPs-TB2), and pristine MPs, respec- tively. (b) shows a small chain of pristine MPs formed when a magnetic field is applied. (c) illustrates the repulsive forces resulting from the electrosteric stabilization of MPs-TB1 and MPs-TB2 after mixing both suspensions. (d) Shows an agglomerate formed after hy- bridization process. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) between the FeAu-NPs, which justifies the SPR peak shift to higher • Magnetic strength of FeAu-NPs: wavelengths, as can be observed in Fig. 5-b [3,42]. μ0 m12 For magnetophoresis measurements, 100 μL of the mixture suspen- Γ1 = ≈ 0.69 sion were incubated for 3 h. Fig. 6-a shows the separation kinetics of the 2πD13 KB T (4) mixture in comparison to the 0.01% suspension of MPs-TB2. The se- where m1 is the magnetic moment at saturation. D1 (∼ 50 nm) the paration time t50 of MPs-TB2 (∼ 228 s) is shorter than the t50 corre- diameter of the FeAu-NPs, D1 ≈ D0 /3. Thus: sponding to FeAu-TB1/MPs-TB2 mixture, indicating that the formed aggregates FeAu-TB1/MPs-TB2 move slower than the MPs-TB2. This Γ0 m2 D 3 14.79 = so3 12 ≈ behavior will be approached based on the magnetic strength of the Γ1 D0 m1 0.69 (5) formed agglomerates. As discussed in the previous section, during magnetophoresis process, the aggregation of the particles depends • Magnetic strength of the mixture MPs + FeAu-NPs: strongly on the magnetic strength, Γ. In the case of FeAu-NPs, based on μ0 (ms0 + Pm1)2 their physical and chemical properties (saturation magnetization of Γ2 = , 2π (D0 + 2D1)3KB T (6) ∼5 emu/g, diameter of ∼50 nm, calculated density of ∼13.28 g/cm3, and the working concentration of ∼0.016% w/v), the calculated value Where P is the number of FeAu-NPs bound to each MP. of ΓFeAu-NPs is ∼0.69 (and N* ∼ 0.01), indicating a non-cooperative By substituting all the parameters from Eq.s 3,4 and 5 in Eq.6: magnetophoretic process based on the individual motion of FeAu-NPs 27 P 2⎤ [37–39], in agreement with the experimental magnetophoresis kinetics Γ2 ≈ Γ0 ⎡ ⎛1 + ⎞ ⎢ ⎣ 125 ⎝ 24 ⎠⎥⎦ (7) obtained for FeAu-NPs, showing separations times > 30 h when the suspension is brought under 7.5 T/m (not shown). An approximation of the number P within the experimental condi- Considering the magnetic strength Γ for the different systems as tions of the presented study, was calculated considering a spherical follows: particle of a diameter D0 = 140 nm, surrounded with a continuous film formed of spherical NPs of a diameter D1 = 50 nm nm, as illustrated in • Magnetic strength of the MPs: D Fig. 6-b. The volume of surface-NPs can be estimated as [43]4π ( 20 )2D1, μ0 ms20 and thus the maximum number of the spherical NPs fitting in this vo- Γ0 = ≈ 14.79 lume of surface is P ∼ 54, which indicates that a maximum of 54 FeAu- 2πD03 KB T (3) NPs can bind to the surface of a MP, resulting in a magnetic strength of Where ms0 is the magnetic moment at saturation and D0 (∼ 140 nm) is the composite, Γ2 ∼ 2Γ0. the diameter of the MPs Fig. 6-c illustrates the behavior of the parameter Γ2 versus the number P. From the obtained curve, one can deduce that the formed 6 M. Benelmekki, et al. Colloids and Surfaces B: Biointerfaces 193 (2020) 111126 suspensions containing the biomolecule to be detected and isolated. For this purpose: i) A magnetic device generating a uniform magnetic field gradient of 7.5 T/m was designed and incorporated into a commercial spec- trophotometer. ii) MPs composed of Fe NPs embedded in PS matrix, and magneto- plasmonic NPs composed of Fe core and Au shell, were successfully functionalized with twenty-mer complementary oligonucleotides (named TB1 and TB2). iii) The study demonstrates that incubation time plays an important role in TB1-TB2 hybridization process, and shows that, for con- centrations of 0.001% (w/v), 60 min of incubation is the threshold for the hybridization process to occur. iv) Hybridization process results in the formation of agglomerates composed of the magnetic particles and the biomolecules. When brought under the 7.5 T/m magnetic field gradient, those agglom- erates follow a cooperative magnetophoretic separation, and move faster than non-hybridized composites. v) The agglomerates composed of hybridized MPs-TB2/FeAu-TB1 se- parate either faster or slower than the non-hybridized colloids, depending the number of FeAu-NPs bound to each MP. vi) By the use of magnetoplasmonic NPs, the intensity and the shift of the SPR peak are used to track the surface functionalization of the FeAu-NPs, and the formation of the aggregates in the suspension, respectively. Our findings suggest exciting possibilities towards a simple and sensitive method for the detection and isolation of target small bio- molecules in complex mixtures. This method is promising for the im- provement of conventional latex agglutination tests, used to detect a wide range of analytes in clinical laboratories [33,44]. In addition, the possibility of recovering the agglomerates at the walls of the container allows their further analysis. In the specific case of agglomerates con- taining plasmonic nanoparticles, further optical analysis of the sample, such as Surface enhanced Raman scattering (SERS), can be conducted allowing the detection of biomolecules, such as proteins and antigens, at a very low concentrations [3,45]. This method represents a big step towards a powerful quantitative tool for rapid molecular diagnosis of diseases, and thus contributes to the prevention of societies and eco- nomic-systems collapsing, as witnessed with the current coronavirus pandemic (COVID-19). Fig. 5. Normalized UV–vis absorbance spectra of (a) MPs-TB1 and FeAu-TB2 Author’s contributions suspension, (b) mixture of MPs-TB2 and FeAu-TB1 suspensions after different incubation times, showing red-shifts of the SPR peak when increasing incuba- MB, SG, and Ll.M conceived the idea. MB planned the experiments, tion time. performed the opto-magnetophoretic measurements, analyzed the data, and wrote the manuscript. SG designed, supervised and validated the agglomerates will move slower than MPs-TB2 while P < 28, and faster magnetic particle functionalization. LlM designed, fabricated and vali- for cases where P > 28.These results can be further developed to dated the magnetic device. All the authors discussed the results and quantify the exact number of the single hybridizations occurring on the edited the manuscript. surface of each functionalized MPs. For example, assuming that each FeAu-TB1 NPs is anchored to the surface of the MP-TB2 via a specific number of hybridizations (or points), the number of the anchored FeAu- Declaration of Competing Interest NPs can be directly associated to the number of the oligonucleotides. Which is highly promising for a quantitative and rapid molecular di- The authors declare that they have no known competing financial agnostic test. Thereby, more quantitative studies combined with theo- interests or personal relationships that could have appeared to influ- retical simulations are needed to further advance these findings. ence the work reported in this paper. 4. Conclusions Acknowledgments We present a simple method for a rapid and accurate detection and separation of twenty-mer oligonucleotides using superparamagnetic The authors thank Ms. Almudena Perez and Dr. Xavier Serra for and magnetoplasmonic colloidal particles. The method consists of a their technical support during the functionalization process of the simultaneous optical and magnetophoretic monitoring of the colloidal magnetic particles. 7 M. Benelmekki, et al. Colloids and Surfaces B: Biointerfaces 193 (2020) 111126 Fig. 6. (a) Magnetic separation curves of MPs- TB2 suspension (black) and MPs-TB2/FeAu- TB1 suspension (red), after 3 h incubation. (b) Illustrates the formed agglomerates after hy- bridization process. (c) Shows the magnetic strength behavior of the formed agglomerates in (b) versus P, the number of FeAu-NPs bound to each MPs. The formed agglomerates will move slower than the MPs while P < 28 (Γ2 < Γ0). (For interpretation of the references to color in this figure legend, the reader is re- ferred to the web version of this article.) Appendix A. Supplementary data [21] J.L. Corchero, A. Villaverde, Trends Biotechnol. 27 (2009) 468. [22] K.M. Krishnan, IEEE Trans. Magn. 46 (2010) 2523. [23] D. Leun, A.K. Sengupta, Environ. Sci. Technol. 34 (2000) 3276–3282. Supplementary material related to this article can be found, in the [24] T.K. Jain, J. Richey, M. Strand, D.L. Leslie-Pelecky, C.A. Flask, V. Labhasetwar, online version, at doi:https://doi.org/10.1016/j.colsurfb.2020.111126. Biomaterials 29 (2008) 4012–4021. [25] S. Behrens, Nanoscale 3 (2011) 877–892. [26] M. Benelmekki, C. Caparros, E. Xuriguera, S. Lanceros-Mendez, E. Rodriguez- References Carmona, R. Mendoza, J.L. Corchero, Ll.M. Martinez, Colloids Surf. B Biointerfaces 101 (2013) 370–375. [1] M.C. Leake, Philos. Trans. R. Soc. Lond. B 368 (2013) 20120248. [27] M. Benelmekki, Designing hybrid nanoparticles, IOP concise physics, Morgan & [2] Y. Zhu, Q. Zhan, J.-C. Yang, Y. Bitla, P. Liu, C.-I. Li, J.H. Liu, V.S. Kumar, Claypool Publishing, 2015. E. Arenholz, Q. He, Y.-H. Chu, ACS Appl. Mater. Interfaces 8 (2016) 1104–1111. [28] M. Benelmekki, A. Montras, A.J. Martins, P.J.G. Coutinho, Ll.M. Martinez, J. Magn. [3] J.-H. Kim, M. Benelmekki, ACS Appl. Mater. Interfaces 8 (2016) 33121–33130. Magn. Mater. 323 (2011) 1945. [4] B. Gao, M.J. Rozin, A.R. Tao, Nanoscale 5 (2013) 5677–5691. [29] G. Friedman, B. Yellen, Curr. Opin. Colloid Interface Sci. 10 (2005) 158. [5] P. Sharma, A. Singh, S.C. Brown, N. Bengtsson, G.A. Walter, S.R. Grobmyer, [30] R. Gerber, M. Takayasu, F.J. Friedlaender, IEEE Trans. Magn. 19 (1983) 2115. N. Iwakuma, S. Santra, E.W. Scott, B.M. Moudgil, Methods Mol. Biol. 624 (2010) [31] G. de las Cuevas, J. Faraudo, J. Camacho, J. Phys. Chem. C 112 (2008) 945. 67–81. [32] Greg T. Hermanson, Bioconjugate Techniques, 3rd ed, Academic Press, 2013 ISBN [6] M.G. Krukemeyer, V. Krenn, F. Huebner, W. Wagner, R. Resch, J. Nanomed, 978-0-12-382239-0. Nanotechnol. 6 (2015) 336. [33] M. Benelmekki, S. Gasso, Ll. M. Martinez (2019) WO/2019/229276. [7] S.M. Mahendran, V. Chandran, Proteomes 6 (2018) 5. [34] Ll. M. Martinez, G. Montero, S. Garcia, EP1904237 B.1 (2013). [8] A. Chamorro-Garcia, A. Merkoc, Nanobiomedicine 3 (2016) 1–26. [35] J. Tang, M. Myers, K.A. Bosnick, L.E. Brus, J. Phys. Chem. B 107 (2003) 7501–7506. [9] Y. van der Pol, F. Mouliere, Cancer Cell 36 (4) (2019) 350–368. [36] P.J. Vikesland, R.L. Rebodos, J.Y. Bottero, J. Rose, A. Masion, Environ. Sci. Nano 3 [10] S.M. Ranuncolo, J. Cancer Therapy 8 (2017) 3. (2016) 567. [11] https://www.cancer.gov/news-events/cancer-currents-blog/2017/liquid-biopsy- [37] J.S. Andreu, J. Faraudo, J. Camacho, M. Benelmekki, C. Rebollo, Ll.M. Martinez, detects-treats-cancer Noational Institute of Cancer. Phys. Rev. E 84 (2011) 021402. [12] W. Li, H. Wang, Z. Zhao, H. Gao, C. Liu, L. Zhu, C. Wang, Y. Yang, Adv. Mater. [38] J.S. Andreu, J. Camacho, J. Faraudo, Soft Matter 7 (2011) 2336–2339. (Weinheim, Ger.) 31 (2019) 1805344. [39] M. Benelmekki, C. capparos, A. Montras, R. Goncalves, S. Lanceros-Mendez, Ll. M. [13] M.S. Iglesias, M. Grzelczak, Beilstein J. Nanotechnol. 11 (2020) 263–284. Martinez J. Nanopart. Res. 13 (2011) 3199–3206. [14] C.T. Yavuz, A. Prakash, J.T. Mayo, V.L. Colvin, Chem. Eng. Sci. 64 (2009) 2510. [40] J. Faraudo, J.S. Andreu, C. Calero, J. Camacho, Adv. Funct. Mater. 26 (2016) [15] J. Lim, S.P. Yep, S.C. Low, Sep. Purif. Technol. 123 (2014) 171–174. 3837–3858. [16] C. De Latour, IEEE Trans. Magn. 9 (1973) 314. [41] D. Roy, J. Fendler, Adv. Mater. 16 (2004) 497–508. [17] G. Mariani, M. Fabbri, F. Negrini, P.L. Ribani, Sep. Purif. Technol. 72 (2010) 147. [42] M.G. Manera, J. Spadavecchia, A. Taurino, R. Rella, J. Opt. 12 (2010) 035003. [18] C. Caparros, M. Benlemekki, P. Martins, E. Xuriguera, C.J. Ribeiro, Ll.M. Martinez, [43] M. Benelmekki, Nanomaterials: the original product of nanotechnology, IOP S. Lanceros-Mendez, "Mater. Chem. Phys. 135 (2012) 510–517. Concise Physics, Morgan & Claypool Publishing, 2019. [19] C.T. Yavuz, Science 314 (2006) 964. [44] F.J. Gella, J. Serra, J. Gener, Pure&App. Chem. 63 (1991) 1131–1134. [20] M. Benelmekki, E. Xuriguera, C. Caparros, E. Rodriguez, R. Mendoza, J.L. Corchero, [45] J. Baniukevic, I.H. Boyaci, A.G. bozkurt, U. Tamer, A. Ramanavicius, S. Lanceros-Mendez, Ll.M. Martinez, J. Colloid Interface Sci. 365 (2012) 156–162. A. Ramanaviciene, Biosens. Bioelectron. 43 (2013) 281–288. 8