Circulating Disease Biomarker Detection in Complex Matrices: Real

Jul 12, 2016 - Phone number: +39 040-375-8755., *E-mail: [email protected]. Phone number: +39 040-375-8291. ... Label-Free Electrochemical S...
0 downloads 11 Views 12MB Size
Download PDF
Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Circulating disease biomarkers detection in complex matrices: real-time, in-situ measurements of DNA/miRNA hybridization via Electrochemical Impedance Spectroscopy Pietro Capaldo, Serena Rosa Alfarano, Luca Ianeselli, Simone Dal Zilio, Alessandro Bosco, Pietro Parisse, and Loredana Casalis ACS Sens., Just Accepted Manuscript • DOI: 10.1021/acssensors.6b00262 • Publication Date (Web): 12 Jul 2016 Downloaded from http://pubs.acs.org on July 18, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Sensors is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Circulating disease biomarkers detection in complex matrices: realtime, in-situ measurements of DNA/miRNA hybridization via Electrochemical Impedance Spectroscopy Pietro Capaldo a,b, Serena Rosa Alfarano c, Luca Ianeselli a, Simone Dal Ziliod, Alessandro Boscoa,†*, Pietro Parissea,e,*, Loredana Casalisa,e,* a

Elettra-Sincrotrone Trieste S.C.p.A., S.S. 14 km 163.5, in Area Science Park, 34149 Basovizza, Trieste, Italy University of Trieste, PhD Course in Nanotechnology, Trieste, Italy c University of Trieste, Physics Department, Trieste, Italy d CNR-IOM, Laboratorio TASC, Area Science Park, Strada Statale 14 km 163.5, 3419 Basovizza, Trieste, Italy e INSTM-ST Unit, Trieste, Italy b

KEYWORDS Capacitive biosensors, Nucleic Acid Hybridization, Electrochemical Impedance Spectroscopy, Langmuir adsorption model ABSTRACT: The quantification of significant amounts of disease biomarkers circulating in the bloodstream represents one of the challenging frontiers in biomedicine. The complexity of blood composition has opened the quest for novel detection technologies, capable of discerning small amount of specific biomarkers from other blood proteins/oligonucleotides and of reliably measuring them. In this context, we have developed a label-free device based on differential double-layer capacitance readout at microfabricated gold electrodes and demonstrated its detection performance in real bio-sample volumes. By means of electrochemical impedance spectroscopy (EIS) measurements in a three electrodes setup, we first calibrated the system following the in situ hybridization of a self-assembled monolayer of single-stranded, short oligonucleotides on the gold microelectrode through the measurement of differential capacitance changes as a function of time, for different concentrations of complementary DNA in a saline buffer. Based on this calibration we used the device to quantify the presence of microRNAs (miRNAs) in human plasma. We demonstrated that our device is fast, sensitive, reusable, reproducible and perfectly suited to detect biomarkers in complex matrices, as cell lysate, serum and blood. We put forward the possibility to apply this platform to the bioaffinity detection of protein biomarkers as well as circulating drugs in blood, for therapeutic drug monitoring applications.

In the last decades the growing interest towards personalized therapies has motivated the development of an increasing number of miniaturized, label-free devices to be used as fast diagnostic tools for medical treatment 1-5, with the final goal of detecting biological biomarkers circulating in the bloodstream. Label-free biosensors require only a single recognition element, leading to simplified assay design, decreased assay time and reduction in reagent costs. Another advantage of label-free method is the ability to perform quantitative measurement of molecular interaction in real-time, allowing continuous data recording. In this class of biosensors, the best performances are expected by detectors based on electrical readout, both in terms of cost reduction and of multiplexing analysis of different biomarkers. These sensors integrated with microfluidic networks in a Labon-a-Chip platform, can develop into easy-to-use, rapid and reliable diagnostic kits to be operated as medical practitioner’s bench tool 6. Electrical readout assays monitor the response of the system to the application of a small amplitude (well below 1 V) AC voltage to functionalized gold electrodes, and can measure, as a function of the AC frequency, the capacitance variation at

the electrode/electrolyte interface upon the occurrence of biorecognition events. These types of measurements are known as electrochemical impedance spectroscopy (EIS) measurements. When working at low frequencies the so called double layer capacitance dominates, as demonstrated by Zou et al 7. In this regime, it is crucial to control the voltage applied to the electrodes with high accuracy, to avoid voltage-induced damaging of the functional molecules adsorbed on the gold surface. To this aim, a third reference electrode is often used. We have recently demonstrated that a three-electrode, miniaturized capacitive device allows for high time stability, enables rapid, realtime response and improves resolution 8, hence being an ideal candidate for fast personalized diagnostics, and/or therapeutic drug monitoring 9-10. However, device performance in terms of quantification of biomolecule in complex matrices, as blood or serum, still needs to be established. In the present work we exploit an improved version of the device as conceived in the work of L. Ianeselli et al.8, to perform the detection of microRNA (miRNA) sequences in human plasma or serum.     miRNAs are ∼22 nucleotide-long non-coding RNA molecules that regulate gene expression of target genes at the posttranscriptional level by translational repression or degradation of messenger RNAs (mRNAs) 11, 12. They contribute to di-

ACS Paragon Plus Environment

ACS Sensors

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

verse physiological processes in mammals, including epithelial regeneration, cardiac function and the progression of cancer 13. The  mechanism  of  translational  silencing  or  deg-­‐ radation  of  mRNA  is  implemented  through  the  ribonucleo-­‐ protein   complex   called   miRNA-­‐induced   silencing   complex   (miRISC).     In  this  process,  the  single  strand  “guide”  miRNA   is  bound  to  the  Argonaute  Ago2  protein,  the  catalytic  core   of  miRISC,  which  uses  the  miRNA  strand  to  recognize  and   bind  complementary  messenger  RNA  transcripts  to  inhibit   the  relative  protein  translation  or  promote  degradation  of   the  mRNA  tracts.  Perfect  matching  is  not  necessary  for  the   silencing   process,   and   one   single   miRNA   can   promote   the   repression   of   several   mRNA   tracts. Some miRNAs are down- or up-regulated in different types of diseases, as cancer or cardiovascular diseases 14-23, affecting the expression of genes causing the disorder. Therefore, miRNAs constitute a novel class of robust and reliable biomarkers for the prognosis and even treatment of diseases. However, despite the relevance of miRNAs for the clinical practice, their diagnostic/therapeutic use is still hindered by the lack of sensitive and reproducible devices, ultimately capable of real time measurements to monitor the response to therapy. We demonstrate here that our device, which detects biorecognition events through the capacitance changes at a DNA probe coated gold electrode as a function of time, is instead sensitive to miRNA quantification, reproducible, reusable, and not affected by the unspecific adsorption of proteins or other substances present in serum. In fact, it relies not only on steady-state values associated to miRNAs biorecognition, but mostly on kinetics parameters. By exploiting the DNA directed immobilization of protein binder-DNA chimeras, we put forward the possibility to implement the device to detect also circulating protein biomarkers directly from the blood.

EXPERIMENTAL SECTION Device fabrication Devices were fabricated using proximity UV-optical lithography, improving the protocols described in 8. Working and counter electrodes were produced on clean microscope slides by a classical lift-off process: positive resist MEGAPOSITTM SPRTM 220 1.2 (Series Photo-Resist) was patterned on the glass slide to a thickness of 1.4 µm. A 20-nm Ti adhesion promoter was deposited first, followed by an 80-nm Au layer. An overnight acetone bath was then performed for a complete lift-off. Electrodes were then coated again with an insulating layer (about 2.5 µm) of NANOTM SU8-2002, shaped to expose the circular part of the working electrode (WE) and the counter electrode (CE) only to the solution. The portion of the cell exposed to the solution, as well as the profile of the protecting SU8 layer, are visible in Figure 1b.

Page 2 of 9

Figure 1. a) Picture of our device and its holder equipped with an electronic card to bring the signal to the HEKA-bipotentiostat. b) SEM image (EHT=2 kV and Mag 82x) of the microfabricated working and counter gold electrodes. The gold electrodes appear lighter than the background of the picture. The patterned insulation layer is shown and it is higher with respect to the plane of the electrodes. c) Model of the electrode-electrolyte interface for a ssDNA SAM on the Au(111) surface. The total capacitance Cd is the sum of capacitances linked in parallel and associated to the DNA-covered area or to the ions (see text).

Functionalization of the gold electrodes DNA functionalization of the gold electrodes was carried out using the well-established procedures for single stranded (ss) DNA self-assembled monolayers (SAMs) on gold 24-25. The electrodes were wetted for 12 min, with a drop of 1 µM thiolated (C6H13-SH, namely C6) ssDNA (probe1 in Table S-1) in TE NaCl 1M. In this way a low density ssDNA SAM (2.1 x 1012 molecules/cm2) was obtained. We experimentally confirmed that the probe density can be used as a controlling factor for the target hybridization efficiency. In fact, in this low probe density regime, almost the totality of probes can be hybridized and the kinetics of binding is well approximated by the Langmuir adsorption model. After DNA-SAM formation the electrodes were rinsed with the same salt solution used for the measurements, KCl 100 mM. Then the differential capacitance at the electrode-electrolyte interface was measured. Hybridization was performed by wetting the electrodes with a drop of the same salt solution containing the complementary DNA strand (target1 in Table S-1) in different concentrations. In the DNA/miRNA experiments the DNA sequences (probe2 and probe3 in Table S-1) were chosen to be complementary to human miRNAs (miR-451a and miR-154-5p) relevant to cardiovascular disease 18, 20. In these experiments the presence of miRNA in buffer solution and in complex matrix (human plasma), respectively, was detected. All the used oligonucleotide-molecules were purchased from Biomers.net. A complete list of oligonucleotide sequences can be found in Table S-1 in SI.

ACS Paragon Plus Environment

Page 3 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Differential capacitance measurements As already explained in the work of Ianeselli et al.8, our device measures the differential capacitance Cd at the electrode electrolyte interface, defined as: Eq.1                                                                              𝐶! =

!!! !"

a two-step cleaning protocol was used: the electrode was first soaked in a PBS plus tween 1.25% solution for a time dependent on the miRNA buffer dilution used, before proceeding with the previously described thermal treatment.

.

Cd reflects the charge density (σM) change at the metal surface for a small variation of the applied potential  𝜑 between electrode and solution. In the case of a bio-functionalized metal electrode immersed in a saline solution, Cd can be modeled as a series connection of two capacitances 26: the one due to the absorbed layer of molecules (Cmol in Figure S-1) and the one related to the ions in solution, the so-called diffusive layer capacitance (Cdl in Figure S-1). The smaller of the two, Cmol, dominates the series. Using the planar capacitor representation, we obtain: Eq.2

𝐶! ≈ 𝐶!"# = 𝜀𝜀!

! !

where A (cm2) is the electrode area in contact with the solution, d the molecular layer height expressed in nm, ε0 and ε the vacuum and molecular layer dielectric constants, respectively. The total capacitance at the interface is, anyway, the sum of capacitances linked in parallel and associated to the area composed of DNA strands or water ions, as shown in Figure 1c. Cd variations at the functionalized WE were measured by means of a Heka PG340 usb potentiostat. As in a classical electrochemical cell, the current generated between the microfabricated WE and CE was monitored, upon applying a voltage measured between WE and a reference electrode (RE), a classical 4 mm-sized Ag/AgCl pellet electrode immersed directly in the liquid cell (a silicone pool with 6 mm diameter and 4 mm height, holding a 125 µL volume) above the microelectrodes, whose stability assures a fine control of the absolute value of the applied potential. Cd was measured in solution, fitting the current response upon the application of a 10 mV amplitude AC voltage at 4 frequencies: 100 Hz, 200 Hz, 250 Hz and 400 Hz. Measurements were restricted to this frequency range since at higher frequencies (above 1.2 kHz) the solution resistance contribution becomes dominant 27. At each frequency 200 complete periods were collected, from which the root-mean-square value of the measured current, Irms, and the relative uncertainties using error propagation analysis were determined. Irms vs. frequency data were then fitted by using a linear regression procedure. To follow Cd variation upon hybridization, first Cd at the ssDNA SAM functionalized WE was measured, testing signal stability over a period of about one hour. Then a solution containing complementary DNA was added to the pool, and the Cd variation measured as a function of time. To better follow the kinetics of the hybridization, the measurements were performed at the rate of four samplings per minute for the first 15 minutes, and one per minute afterwards, for a total of at least an hour.

De-hybridization protocols for device regeneration Thermal de-hybridization cycles were introduced to test the reusability of the device. The thermal treatment in the case of DNA/DNA hybridized SAM on the WE, in buffer solution, consisted of sample incubation in a basic solution (pH=9) of TE buffer for 1 hour in oven at a temperature 10° C higher than the melting temperature of the specific DNA sequence (55°C for the sequence probe1 of Table S-1 in SI). In the case of DNA/miRNA SAMs hybridized in human plasma samples

Figure 2. Kinetics of DNA-hybridization via differential capacitance measurements. The green signal represents the differential capacitance of a low density ssDNA (probe1) SAM functionalized WE measured in KCl 100 mM. The hybridization (red points) was performed by adding in the experimental pool 10 pM complementary DNA (target1). Cartoon-insets show the variation of thickness and dielectric constant between a layer of ss and dsDNA, respectively. The blue and black-dotted lines represent the fits based on first-order Langmuir adsorption kinetics and double exponential kinetics, respectively.

RESULTS AND DISCUSSION In-situ study of DNA-hybridization As a first step, the device was calibrated for DNA/DNA hybridization in saline buffer. The WE was functionalized with a low-density ssDNA SAM (using the 22-mer DNA sequence shown as probe1 in Table S-1 in SI), following the procedure described in Experimental section. The Cd values measured over different samples in KCl 100 mM saline buffer (see Figure 2) at 10 mV applied potential were in the range C! ≈ 1 ÷ 1.3  nF, corresponding to a capacitance density of ≈ 10 µF/cm2, in agreement with the literature 8, 28, 29. This small variability is supporting the high reproducibility of the fabrication steps. Moreover, the Cd absolute values found supports the model of planar capacitor. Assuming a SAM-height of 2.38 ± 0.07 nm, as determined via atomic force microscopy (AFM) measurements 30, and using the ssDNA dielectric constant values found in literature 31-36, a Cd variation between 0.6 nF (εssDNA≈ 20) and 2.3 nF (εssDNA≈ 78) can in fact be derived from Equation 2. The ssDNA SAM hybridization kinetics was followed in-situ, similarly to what described in Ianeselli and co-workers 8, the only difference being in the lower concentration (pM range) now detectable and in the use of real-time detection conditions. The Cd variation due to strand pairing is represented by the red curve of Figure 2, for the case of 10 pM complementary DNA (target1) in KCl 100 mM salt solution. The lowering of the capacitance upon hybridization is justified by the DNA SAM height increase due to the different persistence length of ss- and ds- (double stranded) DNA (1 nm and 50 nm, respectively) 8, 24, 37, and by the replacement of water mole-

ACS Paragon Plus Environment

ACS Sensors

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

cules (with a dielectric constant of  𝜀 ≃ 76.7 ± 0.17, for 100mM KCl-doped aqueous solution 38) with DNA molecules (lower ε) upon DNA pairing. Two different fitting curves were used here: a first order Langmuir adsorption model (blue line in Figure 2) and a double exponential model (dashed black curve), as described in SI. Although the first model (Equation S-2) is the most accepted for ssDNA-SAMs hybridization, the quality of the fit improves (from 𝜒!! ≈ 0.73 to 0.9) if considering the double exponential function (Equation S-3). In this model it is considered that, beside direct pairing by DNA target free in solution, DNA hybridization might also occur through non-specific adsorption onto regions not covered with immobilized DNA, followed by surface diffusion of the target to the probe 39-41. The fitting parameters have been summarized in Table S-3 in SI. The time constant values obtained at different target concentrations with the two fitting curves, reported in Figure S-3, are following the same trend, giving hybridization times 1.5-5 fold shorter than what reported in the literature for DNA-SAM hybridization in similar experimental conditions (e.g.: DNA length, ionic strength, etc.) 25, 42. The fast hybridization observed in our experiments can be justified with a combination of the low probe density used ((2.1 ± 0.1)×1012 mol/cm2, as determined by XPS measurements, Figures S-4 and S-5, corresponding to 3.4 nm average chain-chain distance) and the presence of an applied potential. In this surface density conditions in fact, electrostatic and steric hindrance effects are minimized, while the presence of the potential might help the highly charged target DNA to get in proximity of the surface to bind the complementary, surface immobilized probes and complete hybridization with relatively fast kinetics can be achieved 43-48. To prove this last point, Cd-ssDNA has been measured as a function of the applied potential (Figure S-6): at higher potentials, the dielectric constant of the layer (and so the capacitance) increases due to the progressive orientation of the dipole moment of the ions in solution in the direction of the electrical stimulus (Supporting Information Figures S-7 and S-8). The reusability of our three-electrode sensor obtained by using the thermal regeneration procedure described in the Experimental section, is shown in Figure S-2 of SI. At each cycle, the Cd value after thermal regeneration recovers the initial CdssDNA value with an error ranging from 1 % to 3% (Table S-2 in SI), attesting that the used protocol does not damage the ssDNA probe layer on the electrode. Hybridization was then performed after each regeneration cycle, using different concentrations of complementary-strand DNA (from 1 pM to 100 nM, see Figure S-2), monitoring the differential capacitance at the WE versus the incubation time. At each hybridization cycle, the measurements were run until the Cd differential variations between successive points were lower than 6%. Such “steady-state” Cd values, named Cd-dsDNA, are shown in Table S-2 of SI. Cd-dsDNA values are monotonically inversely proportional to the used concentration of DNA target. This can be rationalized considering that the capacitance decrement is proportional to the number of hybridized sites at the electrode.

Page 4 of 9

Figure 3. Percentage change of Cd upon hybridization of the ssDNA SAM, as a function of [cDNA] and fitted with a Hill equation (dotted black curve). In our operating conditions, DNA binding can be well described using a concentration-limited model (solid green line). The solid red line represents the average variation of Cd when a non-complementary DNA (no_target1 in Table S-1) is present in solution at high concentration (1 µM), whereas the dashed red lines represent 3 standard deviations around this value.

The maximum variation of Cd upon hybridization, Cd% (calculated from Cd-ssDNA to Cd-dsDNA) obtained from real-time measurements, plotted versus complementary DNA concentration [cDNA] in solution, results in the DNA binding affinity curve shown in Figure 3. Assuming a first order Langmuir adsorption kinetics 25, 38, 49, data were fit with a Hill Equation: Eq.3

𝐶!% =

!! ∙[!"#$] !!  !! ∙[!"#$]

An affinity constant for DNA hybridization KA = kon /koff of (0.35 ± 0.09) x 109 M-1, corresponding to an equilibrium dissociation constant Kd = KA -1≈ (2.8 ± 0.7) nM, was derived, about 6 times higher than reported in our previous work8, and closer to the value reported in literature 40, 50, 51 for thiolated DNA strands of similar length. This is justified with the increased sensitivity of the device. Experimental data were also analysed using a concentration-limited regime deriving a Kdvalue, Kd ≈ (2.6 ± 0.7) nM, in agreement with the previous one. More details can be found in SI. From data of Figure 3, a new limit of detection (LOD) of the device, defined as the smallest measured capacitance variation that can be unambiguously assigned to the hybridization process, of 1 pM, is determined. Such value is well above the control measurements, obtained for non-complementary strands (no_target1 shown in Table S-1in SI) in solution at high (1 µM) concentration (region comprised between the two dotted red lines). We believe that this LOD value may be further decreased by lowering the dimension of the gold electrode active area.

Detection of DNA/miRNA Hybridization in Human Plasma After device calibration, we proceeded with the detection of miRNAs of clinical interest in human plasma. As their dysregulation leads to disease development, miRNA profiles are potentially useful as early prognostic and predictive biomarkers. Moreover, miRNAs can be detected in blood

ACS Paragon Plus Environment

Page 5 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

and other body fluids, allowing for non-invasive liquid biopsy 52-55 . Several techniques (RT-qPCR, Next Generation Sequencing, microarray-based assays, etc.) have been employed so far for miRNA analysis 56. However, these methodologies still lack consistency and standardization and often the sensitivity and specificity depend on type and volumes of samples. Here, we focused on a specific miRNA family (miR-451a and miR-154-5p defined in Table S-1) related, among others, to heart failure disease 18, 20.

Figure 4 DNA/miRNA (miR-451a) hybridization detection in KCl 100 mM on DNA-probe film (probe2 in SI). The black line is the fit based on first-order Langmuir adsorption kinetics. The relative fitting parameters are listed in Table S-4 in SI.

In Figure 4, we show the differential capacitance changes relative to the hybridization with 100 pM miR-451a (blue curve) of a low density, complementary DNA SAM (probe2 in Table S-1 in SI) deposited on the WE (red curve) in KCl 100 mM. Fitting data with a first order Langmuir kinetics, we found a hybridization time of τ = 12.35 minutes (Table S-4). Interpolating this value with the DNA/miR-451a hybridization data (blue points of Figure S-10 obtained by applying the same potential of 10 mV), a target concentration of (96.6 ± 8.2) pM is derived (Figure S-10) in close agreement with the expected 100 pM. To move towards the detection of blood circulating miRNAs, we first tested the specificity of the device. As a negative control, we functionalized the electrode with a thiolated ssDNA complementary (shown as probe4 in Table S-1 in SI) to the murine mmu-miR-351-5p miRNA, not traceable in human blood. Then, we incubated the system in a human plasma sample, observing no hybridization kinetics, as reported in SI. At this point, we proceeded with the quantification of miR154-5p in a human plasma sample. PBS 1x buffer solution diluted plasma samples with different ratios (1:20, 1:4 and 1:3), were injected successively into a cell with cDNA SAM functionalized WE. Between each injection, the SAM was annealed (following the two-step procedure described in the Experimental section), and the ssDNA capacitance measured, as a reference, in pure PBS 1x buffer. In order to increase the device sensitivity, we set an AC voltage of 150 mV amplitude. Results are shown in Figure 5a. Remarkably, hybridization kinetics curves are well defined also in a complex matrix. The main difference with respect to saline buffer is the presence of an initial stepwise decrease of Cd. We associate this gap to the blocking effect of the protein contaminants (mostly HSA) present in human plasma.

Figure 5. (a) ssDNA (probe3 in Table S-1 in SI) functionalized WE (red points) in PBS 1x followed by cDNA/miRNA detection in Human Plasma diluted in PBS 1x in ratio 1:3 (green points);1:4 (blue points) and 1:20 (pink points). Purple and black points were measured in PBS 1x after a two step regenerating protocol. (b) Hybridization time (red points) and Cd gap (blue points) due to unspecific proteins binding, as a funcion of the number of serial dilution of the same sample.

To prove it, we monitored the effect of different concentrations of BSA (Bovine Serum Albumin) on ssDNA SAM Cd (probe1 used in Figure 6). At each concentration, we registered a stepwise signal drop, without any further decrease of the capacitance vs. time. The absence of any typical absorption kinetics (yellow curve in the inset of Figure 6) confirms the occurrence of aspecific binding only. The sharp drop of Cd is rationalized assuming a displacement of adsorbed ions by the proteins, reducing the electric susceptibility of the medium. The Cd reduction amount is BSA-concentration dependent, as seen in Figure 6.

Figure 6 Percentage change of Cd obtained dissolving increasing concentrations of BSA on the same probe film. Inset: Effects, in time, on the measured Cd at the ssDNA functionalized WE due to the presence in solution of 10% of BSA (Bovine serum Albumin).

The time constants corresponding to data of Figure 5a are plotted in Figure 5b, together with the Cd gap values. As we can see, there is an internal consistency in this set of data.

ACS Paragon Plus Environment

ACS Sensors

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

To calibrate the system properly, the effect of the Argonauts, which are protecting miRNA from enzymatic degradation in serum, on hybridization kinetics needs to be estimated. In particular, since Argonauts are proteins of about ≈ 110 kDa weight 13 we expect a slower kinetics in real conditions. To this aim, we calibrated the system using a target strand conjugated with streptavidin, to simulate the Argonaut. Since DNA/cDNA and DNA/miRNA calibration curves obtained in similar experimental conditions (Figure S-10, SI, data taken at 100 mM KCl and 10 mV applied potential) were comparable, within the experimental errors, we decided to use DNAstreptavidin chimera instead of more complex miRNAstreptavidin ones. We measured DNA/cDNA hybridization in PBS 1x (Figure 7, red points, 150 mV applied potential): reporting the τHyb value corresponding to the highest dilution (1:20) on this curve, an estimated miR-154-5p concentration of (2.0 ± 0.3) nM is obtained.

 

Figure 7 Hybridization times as a function of concentration of complementary solution. In blue we plot complementary DNA conjugated with Streptavidin, in red DNA/cDNA detection.

Then, in another experiment, the same ssDNA SAM was hybridized with DNA-streptavidin chimeras obtained from the conjugation of a complementary, biotinilated DNA strand with streptavidin (≈ 60 kDa). Data are reported in Figure 7, blue curve. As expected, we found a slower kinetics in the case of streptavidin-DNA conjugates, and an estimated miR-154-5p concentration of approximately (3.1 ± 0.3) nM.

CONCLUSIONS High sensitivity detection of circulating biomarkers as miRNAs and proteins relevant to specific disease is still challenging when it comes to real, complex matrices 57. Biosamples as human plasma, serum, saliva, urine, etc. contain an elevated numbers of proteins that can participate to signal generation (e.g.: fluorescence, charge distribution) hiding the signal coming from (few) specific biorecognition events. Here we proposed a miniaturized, label-free, electrochemical impedance spectroscopy based device with a careful optimization of the electrode surface functionalization, for the realtime, quantitative detection of double layer capacitance changes derived from the adsorption of molecules at the functionalized interface. Our device measures not only the steady-state,

Page 6 of 9

but also the biorecognition kinetics, easily discriminating the blocking effect of other biomolecules present in the matrix. In particular, we developed a three-electrodes sensor for the analysis of miRNAs relevant to specific diseases through the determination of the hybridization kinetics onto a cDNA probe monolayer self-assembled onto the working electrode. We demonstrated that: i) the device is reusable up to several times; ii) data are reproducible on different devices (standard deviation of only few percent); iii) the device has a low detection limit of 1 pM (improvable) in buffer solution; iv) it can be used to quantify miRNAs in complex matrices, as human plasma, by using DNA/cDNA calibration curves. In the human sample tested so far, a miRNA concentration in the low nM (few nM) range was detected. Other healthy/diseased patient samples are at present under measurement. Very remarkably, the high sensitivity of our kinetics studies allows not only to quantify unknown concentrations of DNA/miRNA-target in solution, but in principle also to discriminate different miRNAs families 13,17,55 through the detection of single-base mismatches along the target sequence, as recently demonstrated for the case of DNA/cDNA hybridization in buffer solution 58. Moreover, we can easily extend the device design to protein biomarker detection 59 or therapeutic drug monitoring 9: by exploiting DNA directed immobilization (DDI) 60 of DNAprotein conjugates, we can in fact immobilize protein or drug binders on the surface, and then perform a biomarker/drug screening in complex matrices, similarly to what already described for miRNA analysis. The only precaution in this case will be to use small binders, as in-silico designed short peptides, aptamers or single domain antibodies (e.g. VHH fragments), in order to match the dynamic polarization length of the device 61, which is around 8 nm at 150 mV of applied potential. We are confident that our device, when opportunely combined with the available paper-based microfluidics and integrated to perform multiplexing analysis of 5-10 different biomarkers in a biosample droplet, will constitute a sensitive, fast and cheap point-of-care device for medical diagnostics/prognosis and for the monitoring of the effect of a therapy from liquid biopsies, at the single patient level.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website Additional details and figures include: the list of oligonucleotides sequences used in this work (Table S-1); in situ DNA hybridization measurements (Table S-2, Figures S-2); results of the kinetics analysis of DNA Hybridization (Table S-3); calibration curves obtained fitting experimental data with 1st order Langmuir adsorption model and double exponential model (Figure S-3); X-Ray Photoemission measurements for the evaluation of DNA SAM density (Figures S-4 and S-5); electrical characterization of the electrodes in presence and absence of ssDNA SAM (Figures S-6 and S-8); calibration of DNA/miRNA hybridization in complex matrix (Figure S-9,

ACS Paragon Plus Environment

Page 7 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Table S-4), device specificity in Human Plasma for miRNA detection in Figure S-10.

AUTHOR INFORMATION Corresponding Author * : [email protected] Phone number: +39 040-375-8291 : [email protected] Phone number: +39 040-375-8755 : [email protected]    

Present Addresses †Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden.

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

ACKNOWLEDGMENTS We would like to thank Laura Locchi for the help in setting up the measurements in complex matrices, and Denis Scaini for stimulating discussions. We are also grateful to Daniela Cesselli and Antonio Paolo Beltrami from University of Udine for providing the human plasma samples, under the approval of the Ethical Committee of Azienda Ospedaliero Universitaria di Udine. This work was supported by a FIRB 2011 grant RBAP11ETKA “Nanotechnological approaches toward tumor theragnostic” (P.P. and L.C), a grant from Associazione Italiana per la Ricerca sul Cancro (AIRC) to L.C., A.B. and P.C. (AIRC 5 per mille 2011, no. 12214).

REFERENCES   (1) (2)

(3) (4)

(5)

(6) (7)

(8)

(9)

Tsouti, V.; Boutopoulos, C.; Zergioti, I.; Chatzandroulis, S. Capacitive microsystems for biological sensing. Biosens. Bioelectron. 2011, 27(1), 1–11. Beyette, F. R.; Kost, G. J.; Gaydos, C. A.; Weigl, B. H. “Point-of-care technologies for health care,” IEEE Trans. Biomed. Eng. 2011, 58(3), 732-735. Neužil, P.; Giselbrecht, S.; Länge, K.; Huang, T. J.; Manz, A. Revisiting lab-on-a-chip technology for drug discovery. Nat. Rev. Drug. Discov. 2012, 11, 620–632. Chin CD, Linder V, Sia SK. Commercialization of microfluidic point-of-care diagnostic devices. Lab Chip 2012, 12, 2118–2134. Kumar, S.; Kumar, S.; Ali, M. A.; Anand, P.; Agrawal, V. V.; John, R.; Maji, S.; Malhotra, B. D. Microfluidicintegrated biosensors: prospects for point-of-care diagnostics. Biotechnol. J. 2013, 8(11), 1267-1279. Mao, X.; Huang, T. J. Microfluidic diagnostics for the developing world. Lab Chip 2012, 12(8), 1412-1416. Zou, Z.; Kai, J.; Rust, M. J.; Han, J.; Ahn, C. H. Functionalized nano interdigitated electrodes arrays on polymer with integrated microfluidics for direct bio-affinity sensing using impedimetric measurements. Sensor. Actuat. A-Phys. 2007, 136, 518–526. Ianeselli, L.; Grenci, G.; Callegari, C.; Tormen, M.; Casalis, L. Development of stable and reproducible biosensors based on electrochemical impedance spectroscopy: Three-electrode versus two-electrode setup. Biosens. Bioelectron. 2014, 55, 1–6. Aarnoutse, R. E.; Schapiro, J. M.; Boucher, C. A. B.; Hekster, Y. A.; Burger, D. M. Therapeutic drug monitoring: an aid to optimising response to antiretroviral drugs?. Drugs 2003, 63, 741–753.

(10) Kuo, Y. C.; Chen, C. S.; Chang, K. N.; Lin, C. T.; Lee, C. K. Sensitivity improvement of a miniaturized label-free electrochemical impedance biosensor by electrode edge effect. J. Micro/Nanolith. MEMS MOEMS. 2014, 13(3), 033019. (11) Ambros, V. The functions of animal microRNAs. Nature 2004, 16 (431), 350-355. (12) Xuhang, Liu; Fortin, K.; Mourelatos, Z. MicroRNAs: biogenesis and molecular functions. Brain Pathol. 2008, 18(1), 113-121.   (13) Schirle, N.T.; Sheu-Gruttadauria, J.; MacRae, I. J. Structural basis for microRNA targeting. Science 2014, 346(6209), 608-613. (14) Fabian, M. R.; Sonenberg, N. The mechanism of miRNAmediated gene silencing: a look under the hood of miRISC. Nat. Struct. Mol. Biol. 2012, 19(6), 586-593. (15) Gilad, S.; Meiri, E.; Yogev, Y.; Benjamin, S.; Lebanony, D.; Yerushalmi, N.; Benjamin, H.; Kushnir, M.; Cholakh, H.; Melamed, N.; Bentwich, Z.; Hod, M.; Goren, Y.; Chajut, A. Serum microRNAs are promising novel biomarkers. PLoS One 2008, 3(9):e3148. (16) Andorfer, C. A.; Necela, B. M.; Thompson, E. A.; Perez, E. A. MicroRNA signatures: clinical biomarkers for the diagnosis and treatment of breast cancer. Trends Mol. Med. 2011, 17, 313-319. (17) Abell, J. L.; Garren, J. M.; Driskell, J. D.; Tripp, R. A.; Zhao, Y. Label-Free Detection of Micro-RNA Hybridization Using Surface-Enhanced Raman Spectroscopy and LeastSquares Analysis. JACS 2012, 134(31), 12889-12892. (18) Iorio, M. V.; Croce, C. M. MicroRNA dysregulation in cancer: diagnostics, monitoring and therapeutics. A comprehensive review. EMBO Mol. Med. 2012, 4, 143–159. (19) Goren, Y.; Kushnir, M.; Zafrir, B.; Tabak, S.; Lewis, B. S.; Amir, O. Serum levels of microRNAs in patients with heart failure. Eur. J. Heart Fail. 2012, 14, 147–154. (20) Sempere, L. F.; Korc, M. A method for conducting highly sensitive microRNA in situ hybridization and immunohistochemical analysis in pancreatic cancer. Methods Mol. Biol. 2013, 980:43-59. (21) Bauters, C.; Kumarswamy, R.; Holzmann, A.; Bretthauer, J.; Anker, S. D.; Pinet, F.; Thum, T. Circulating miR-133a and miR-423-5p fail as biomarkers for left ventricular remodeling after myocardial infarction. Int. J. Cardiol. 2013, 168(3), 1837-1840. (22) Ogata-Kawata, H.; Izumiya, M.; Kurioka, D.; Honma, Y.; Yamada, Y.; Furuta, K.; Gunji, T.; Ohta, H.; Okamoto, H.; Sonoda, H.; Watanabe, M.; Nakagama, H.; Yokota, J.; Kohno, T.; Tsuchiya, N. Circulating Exosomal microRNAs as Biomarkers of Colon Cancer. PLoS One 2014, 9(4), e92921. (23) Markou, A.; Zavridou, M.; Lianidou, E. S. MicroRNA signatures as clinical biomarkers in lung cancer. Curr. Biomark. Find. 2015, 5, 35-45. (24) Mirmomtaz, E.; Castronovo, M.; Grunwald, C.; Bano, F.; Scaini, D.; Ensafi, A.; Scoles, G.; Casalis, L. Quantitative study of the effect of coverage on the hybridization efficiency of surface-bound DNA nanostructures. Nano Lett. 2008, 8, 4134–4139. (25) Peterson, A. W.; Heaton, R. J.; Georgiadis, R. M. The effect of surface probe density on DNA hybridization. Nucleic Acids Res. 2001, 29, 5163–5168. (26) Guiducci, C.; Stagni, C.; Zuccheri, G.; Bogliolo, A.; Benini, L.; Samorì, B.; Riccò, B. A Biosensor for Direct Detection of DNA Sequences Based on Capacitance Measurements. 32nd ESSDERC 2002, 479–482. (27) Hong, J.; Yoon, D. S.; Kim, S. K.; Kim, T. S.; Kim, S.; Pak, E. Y.; No, K. AC frequency characteristics of coplanar impedance sensors as design parameters. Lab Chip 2005 ,5, 270–279.

ACS Paragon Plus Environment

ACS Sensors

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(28) Carrara, S.; Bhalla,V.; Stagni, C.; Benini, L.; Ferretti, A.; Valle, F.; Gallotta, A.; Riccò, B.; Samorì, B. Label-free cancer markers detection by capacitance biochip. Sensor. Actuat. B-Chem. 2009, 136(1), 163–172. (29) De Venuto, D.; Carrara, S.; Riccò, B. High Resolution Electronics for Label Free Capacitive DNA Detection. In Sequence and Genome Analysis: Methods and Application II; iConcenpt Press Pty Ltd., Annerley, PRC, 2011, pp 157176. (30) Ianeselli., L. PhD Thesis: “Development of a Miniaturized Electro-fluidic Detector for Medical Diagnostics”, 2013. (31) Mandel. L. M. Dielectric properties of cherged linear macromolecules with particular reference to DNA. In Annals of the New York Academy of Sciences, 1978, vol. 303, pp 7489. (32) Baker-Jarvis, J.; Jones, C.A.; Riddle, B. Electrical Properties and Dielectric Relaxation of DNA in Solution. NIST Technical Note 1509 USA, 1998. (33) Fumagalli, L.; Ferrari, G.; Sampietro, M.; Gomila, G. Dielectric-constant measurement of thin insulating films at low-frequency by nanoscale capacitance microscopy. Appl. Phys. Lett., 2007, 91, 243110. (34) Fumagalli, L.; Esteban-Ferrer, D.; Cuervo, A.; Carrascosa, J. L.; Gomila, G. Label-free identification of single dielectric nanoparticles and virus with ultraweak polarization forces. Nat. Mater. 2007, 11, 808-816. (35) Tomiz, S.; Dolanski Babic, S.; Vuletic, T.; Krca, S.; Ivankovic, D.; Griparic, L.; Podgornik, R. Dielectric relaxation of DNA aqueous solutions. Phys. Rev. E 2007, 75, 021905. (36) Liu, Y. S.; Banada, P. P.; Bhattacharya, S.; Bhunia, A. K.; Bashir, R. Electrical charactherization of DNA molecules in solution using impedance measurements. Appl. Phys. Lett. 2008, 92,143902, 1-3. (37) Rao, A. N.; Rodesch, C. K.; Grainger, D. W. Real-time fluorescent image analysis of DNA spot hybridization kinetics to assess microarray spot heterogeneity. Anal. Chem. 2012, 84, 9379-9387. (38) Hubbard, J. B.; Onsager, L.; van Beek, W. M.; Mandel, M. Kinetic polarization deficiency in electrolyte solutions. Proc. Natl. Acad. Sci. USA, 1977, 74(2), 401-404. (39) Holford T.R.J., Davis F., and Higson S.P.J. Recent trends in antibody based sensors. Biosens. Bioelectron., 2012, 34(1), 12-24. (40) Chan, V.; Graves, D. J.; McKenzie, S. E. The Biophysics of DNA hybridization with immobilized oligonucleotide probes. Biophys. J. 1995, 69, 2243-2255. (41) Sorgenfrei, S.; Chiu, C.; Gonzalez Jr, R. L.; Yu, Y-J.; Kim, P.; Nuckolls, C.; Shepard, K. L. Label-free single-molecule detection of DNA-hybridization kinetics with a carbon nanotube field-effect transistor. Nat. Nanotech. 2011, 6(2), 126-132. (42) Auer, S.; Nirschl, M.; Schreiter, M.; Vikholm-Lundin, I. Detection of DNA hybridization in a diluited serum matrix by surface plasmon resonance and film bulk acoustic resonators. Ana. Bioanal. Chem. 2011, 400(5), 1387-96. (43) Erickson, D.; Li, D.; Krull, U. J. Modeling of DNA Hybridization Kinetics for Spatially Resolved Biochips. Anal. Biochem. 2003, 317, 186–200. (44) Monserud, J. H.; Schwartz, D. K. Mechanisms of surfacemediated DNA hybidization. ACS Nano 2014, 8(5), 44884499. (45) Wong, I. Y.; Melosh, N. A. Directed hybridization and melting of DNA linkers using counterion-screened electric fields. Nano Lett., 2009, 9, 3521-3526. (46) Fixe, F.; Cabeça, R.; Chu, V.; Prazeres, D.; Ferreira, G.; Conde, J. Electric – field – pulse - assisted covalent immobilization of DNA in the nanosecond time scale. Appl. Phys. Lett., 2003, 83, 1465ectric fields. Nano Lett., 9, 35213526.

Page 8 of 9

(47) Irving, D.; Gong, P.; Levicky, R. DNA Surface Hybridization: Comparison of Theory and Experiment. J. Phys. Chem. B 2010, 114, 7631–7640. (48) Wong, I.Y.; Melosh, N.A. An electrostatic model for DNA surface hybridization. Biophys. J. 2010, 98, 2954-2963. (49) Peterson, A.W.; Wolf, L. K.; Georgiadis, R. M. Hybridization of mismatched or partially matched DNA at surfaces. J. Am. Chem. Soc. 2002, 682, 14601-14607. (50) Li, D.; Zou, X.; Shen, Q.; Dong, S. Kinetic study of DNA/DNA hybridization with electrochemical impedance spectroscopy. Electrochem. Commun. 2007, 9, 191-196. (51) Kambhampati, D.; Nielsen, P.E.; Knoll, W. Investigating the kinteics of DNA-DNA and PNA-DNA interactions using surface plasmon resonance-enhanced fluorescence spectroscopy. Biosens. Bioelectron. 2001, 16, 1109-1118. (52) de Plannell-Saguer, M.; Rodicio, M. C. Analytical aspects of microRNA in diagnostics: A review. Anal. Chim. Acta 2011, 699(2), 134-152. (53) Heneghan, H. M.; Miller, N.; Lowery, A. J.; Sweeney, K. J.; Newell, J.; Kerin, M. J. Circulating microRNAs as novel minimally invasive biomarkers for breast cancer. Ann. Surg. 2010, 251(3), 499-505. (54) Mitchell, P. S.; Parkin, R. K.; Kroh, E. M.; Fritz, B. R.; Wyman, S. K.; Pogosova-Agadjanyan, E. L.; Peterson, A.; Noteboom, J.; O'Briant, K. C.; Allen, A.; Lin, D. W.; Urban, N.; Drescher, C. W.; Knudsen, B. S., Stirewalt, D. L.; Gentleman, R.; Vessella, R. L.; Nelson, P. S.; Martin, D. B.; Tewari, M. Circulating microRNAs as stable blood-based markers for cancer detection. Proc. Natl. Acad. Sci. USA 2008, 105(30), 10513-10518. (55) Mo, M-H.; Chen, L.; Fu, Y.; Wang, W.; Fu, S. W. Cell-free Circulating miRNA Biomarkers in Cancer. J. Cancer 2012, 3, 432-448. (56) Ono, S.; Lam, S.; Nagahare, M.; Hoon, D. S. B. Circulating microRNA biomarkers as lquid biopsy for cancer patients: pros and cons of current assays. J. Clin. Med. 2015, 4, 18901907. (57) Gaster, R. S.; Hall, D. A.; Neilson, C. H.; Osterfeld, S. J.; Yu, H.; Mach, K. E.; Wilson, R. J.; Murmann, B.; Liao, J. C.; Gambhir, S. S.; Wang, S. X. Matrix-insensitive protein assays push the limits of biosensors in medicine. Nat. Med. 2009, 15, 1327-1332. (58) Capaldo, P.; Nkoua Ngavouka, M. D.; Ambrosetti, E.; Scoles, G.; Casalis, L.; Parisse, P. Mismatch detection in DNA monolayers by Atomic Force Microscopy and Electrochemical Impedance Spectroscopy. Beilstein J. Nanotechnol. 2016, 7, 220-227. (59) Boozer, C.; Ladd, J.; Chen, S.; Yu, Q.; Homola, J.; Jiang, S. DNA directed protein immobilization on mixed ssDNA/oligo (ethylene glycol) self-assembled monolayers for sensistive biosensors. Anal. Chem. 2004, 76(23), 69676972. (60) Bano, F.; Fruk, L.; Sanavio, B.; Glettenberg, M. Toward Multiprotein Nanoarrays Using Nanografting and DNA Directed Immobilization of Proteins, Nano Lett. 2009, 9(7), 2614-2618. (61) Singh, M. B.; Kant, R. Debye-Falkenhagen dynamics of electric double layer in presence of electrode heterogeneities. J. Electroanal. Chem., 2013, 704, 197-207.  

ACS Paragon Plus Environment

Page 9 of 9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

For  TOC  only  

   

ACS Paragon Plus Environment