This is the accepted version of the article: Mauriz E., Dey P., Lechuga L.M.. Advances in nanoplasmonic biosensors for clinical applications. Analyst, (2019). 144. : 7105 - . 10.1039/c9an00701f. Available at: https://dx.doi.org/10.1039/c9an00701f View Article Online View Journal Analyst Accepted Manuscript This article can be cited before page numbers have been issued, to do this please use: E. Mauriz, P. Dey and L. M. Lechuga, Analyst, 2019, DOI: 10.1039/C9AN00701F. Volume 143 Number 10 21 May 2018 This is an Accepted Manuscript, which has been through the Pages 2185-2436 Royal Society of Chemistry peer review process and has been accepted for publication. Analyst Accepted Manuscripts are published online shortly after acceptance, rsc.li/analyst before technical editing, formatting and proof reading. Using this free service, authors can make their results available to the community, in citable form, before we publish the edited article. We will replace this Accepted Manuscript with the edited and formatted Advance Article as soon as it is available. You can find more information about Accepted Manuscripts in the Information for Authors. Please note that technical editing may introduce minor changes to the text and/or graphics, which may alter content. The journal’s standard ISSN 0003-2654 Terms & Conditions and the Ethical guidelines still apply. In no event PAPER Huinan Yang, Chengxing Shen, Xiaoshu Cai et al. Noninvasive and prospective diagnosis of coronary heart disease shall the Royal Society of Chemistry be held responsible for any errors with urine using surface-enhanced Raman spectroscopy or omissions in this Accepted Manuscript or any consequences arising from the use of any information it contains. rsc.li/analyst Page 1 of 47 Analyst 1 2 3 4 Advances in Nanoplasmonic biosensors for clinical applications View Article Online DOI: 10.1039/C9AN00701F 5 Elba Mauriz1*, Priyanka Dey2 and Laura M. Lechuga2 6 7 1 Department of Nursing and Physiotherapy, Universidad de León, Campus de Vegazana, s/n, 8 24071 León, Spain 9 10 2 Nanobiosensorsand Bioanalytical Applications Group, Catalan Institute of Nanoscience and Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 Nanotechnology (ICN2), CSIC, BIST, and CIBER-BBN, Campus UAB, 08193 Barcelona, 12 Spain. 13 14 15 16 Abstract Biomarkers are unquestionable biological indicators for diagnosis and therapeutic 17 18 interventions providing appropriate classification of a wide range of health disorders and risk Analyst Accepted Manuscript 19 20 factors. Nonetheless, detection and quantification of biomarkers need to be tested with sufficient 21 reliability by robust analytical methods in order to assure clinical performance in health care 22 23 settings. Since the analytical performance is determined by the sensitivity and specificity of the 24 method employed, techniques have been intensively refined in order to avoid misinterpretation of 25 26 results and undesirable bias. Although biomarkers can be detected with the existing analytical 27 techniques, to reproducibly quantify them in decentralized settings or remote locations with the 28 29 required accuracy is still a challenge. Currently, only a few point-of-care devices for biomarkers 30 31 evaluation are commercially available. Thus, more focused research efforts are needed to 32 overcome those limitations in order to provide universal patient-centered care platforms. 33 34 To this end, plasmonic biosensors can be conveniently used as portable diagnostic devices for 35 36 attaining timely and cost-effective clinical outcomes. The development of enhanced performances 37 38 based on nanoplasmonics technology opens the way for sensor miniaturization, multiplexing and 39 point of care testing. This review covers recent advances and applications of plasmonic and 40 41 nanoplasmonic biosensors intended for biomarker diagnosis in the clinical practice, including 42 cancer, cardiovascular and neurodegenerative diseases. The review specially focuses on: (i) recent 43 44 progress in plasmonics developments including the design of singular nanostructured surfaces, 45 (ii) novel chemical functionalization strategies for the appropriate incorporation of the 46 47 bioreceptors and (iii) plasmonic applications as real operative devices in the clinical field. Future 48 49 prospects in the use of nanoplasmonic sensor platforms for personalised quantification and 50 management of biomarkers directly in body fluids will also be discussed. 51 52 1. Introduction 53 54 55 Biomarkers in body fluids are commonly utilized in clinical practice as diagnostic or prognostic 56 indicators for the onset, progression and remittance of diseases.1, 2 Biomarkers provide objective 57 58 quantification of normal or pathological diagnostic outcomes and aid to predict the biological 59 response after exposure to therapeutic drugs or environmental agents. The variety of biomarkers 60 Page 1 of 47 Analyst Page 2 of 47 1 2 3 that can be tested ranges from protein, gene expression profiles to blood and urine concentrationsView Article Online DOI: 10.1039/C9AN00701F 4 5 of drugs.1, 3-6 In general, the classification of biomarkers relies on the sequence of events from 6 exposure to disease onset, categorizing them as either diagnostic or monitoring markers. 7 8 Diagnostic biomarkers allow the identification of a medical condition via genetic and 9 pathophysiologic parameters while monitoring biomarkers assess the extent and progression of a 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 disease by testing drug concentration or toxicity effects (FDA-NIH Biomarker Working Group). 12 13 14 Diagnostic or prognostic biomarkers allow for early detection of diseases and have been at the 15 16 spotlight for researchers, especially for cancer biomarkers2 as it promises significant 17 improvement in chances of survival. Biomarker-based disease diagnostics relies on detecting the 18 Analyst Accepted Manuscript 19 specific biomarkers associated to a disease and their presence in body fluids like blood, urine, 20 21 saliva, sputum, cerebrospinal fluid (CSF), among others.1, 3, 4 These biomarkers can exhibit a 22 significant and specific concentration deviation (up/down regulation) in disease conditions 23 24 compared to healthy conditions, followed by accurate and specific determination of concentration 25 of the specific biomarker. This requires collection of body fluids, preferentially in a minimally 26 27 invasive manner. 28 29 The phases of clinical diagnosis can be summarized as shown in Fig. 1. The usual laboratory 30 31 procedures involve sample collection, sample pre-processing (for example, pre-concentration, 32 DNA/RNA replication, growth incubation for pathogens, etc.), laboratory analytical test by expert 33 34 technicians (to evaluate the biomarker concentration) and analysis of the data (comparison of the 35 outcome to healthy condition). A summary of the complete procedure is then delivered to the 36 37 medical team who decides the appropriate treatment. Though this analytical procedure provides 38 39 reliable outcomes, an inherent limitation is its’ inaccessibility at decentralized clinics, 40 ambulatories, ambulances or hospitals itself, requiring in most cases an outsourced laboratory 41 42 and, thereby, delaying critical care, particularly for cases like sepsis7, 8 and other infections. Thus, 43 medical diagnostics need a transformation towards portable diagnostic devices with quick 44 45 turnaround time such that regular health biomarkers can be tested as point-of-care, at the doctor´s 46 47 office, emergency rooms, ambulances, hospitals and even at home. This can be achieved by 48 combining the sample pre-processing and the analytical test and read-out all into one simple- 49 50 functioning, easy to-use, direct read-out point-of-care (POC) device platform. 9-11 This concept 51 has been summarized in Fig. 1, where the research focus lies in POC device development. 52 53 54 55 56 57 58 59 60 Page 2 of 47 Page 3 of 47 Analyst 1 2 3 View Article Online DOI: 10.1039/C9AN00701F 4 5 6 7 8 9 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 13 14 15 16 17 18 Analyst Accepted Manuscript 19 20 21 22 23 24 25 26 Fig. 1. Overview of the flow in clinical diagnostics in present and in the near future. 27 28 29 30 To accomplish this transformation, analytical methods can take advantage of biosensing-based 31 32 platforms. Biosensors enable simultaneous and real-time detection of multiple biomarkers. They 33 provide a measurable signal that can be quantitatively related to the amount of biomarker in the 34 35 body fluids. An ideal biosensor for clinical diagnostics12, 13 should fulfill the following 36 characteristics: (a) be able to detect a specific biomarker in its clinically relevant range with a low 37 38 limit of detection (LOD) and a minimum signal-to-noise (S/N) ratio >2 for high accuracy, (b) 39 40 have a detection with high specificity (no cross-reactivity) and provide sensor-to-sensor 41 reproducibility, (c) able to provide multiplexed analysis of several biomarkers related to the same 42 43 disease, which would not only reduce the cost per test, but also will assure diagnostics accuracy, 44 (d) minimal patient body-fluid (sample) required, (e) minimal operational complications and 45 46 minimal additional reagents required, (f) fast testing turnaround times (potentially in minutes), 47 (g) sensing instrument fairly compact and portable, as well as user-friendly and cost-effective, (h) 48 49 sensor chips or substrates with a long shelf-life and stability in a wide working range of 50 51 temperature and humidity. 52 53 54 An IDTechEx research report titled “Biosensors for Point-of-Care Testing: Technologies, 55 56 Applications, Forecasts 2017-2027”, predicts biosensors for point-of-care testing market will 57 58 grow to $33 billion by 2027, with molecular diagnostic devices as the main driver for this growth. 59 The role of biosensors as clinical analytical tools has already been proven in the health care sector. 60 Page 3 of 47 Analyst Page 4 of 47 1 2 3 One of the most relevant examples that have popularized POC biosensor devices is the View Article Online DOI: 10.1039/C9AN00701F 4 5 electrochemical based glucometers14 that measure glucose levels from a drop of blood with digital 6 read-out which is a must-have instrument for diabetic patients at home now-a-days. The other is 7 8 pregnancy strips providing a visual color-based positive or negative read-out,based on gold 9 nanoparticles changing colour due to aggregation upon interaction with pregnancy hormones. 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 Products that are more recent utilize a smartphone as the read-out device, non-invasive wearable 12 13 sensors15, while others employ an injectable sensor material making the POC devices smarter and 14 more lucrative. 15 16 17 Enzyme-linked immunosorbent assay (ELISA)16 has been the gold standard in biomarker sensing 18 Analyst Accepted Manuscript 19 with dominating market presence for laboratory-based technologies. Recently, with the drive 20 21 towards POC devices, ELISA has been upgraded to lab-on-chip system by utilizing microfluidic 22 technologies. Although ELISA has advantages of being a robust platform, it still requires long 23 24 incubation and washing times along with the necessity of a detection label. Although plasmonic 25 biosensing technology utilizes similar sensing principles as ELISA where the bioreceptor is 26 27 immobilized onto a surface and the captured biomarker is detected, it is label-free and does not 28 require extensive incubation/washing steps nor is labor-intensive. This in addition to the 29 30 plausibility of plasmonic biosensor miniaturization, makes it an ideal technology for developing 31 32 POC devices for clinical settings. Optical biosensors have thus gained immense limelight due to 33 their capability of biomarker detection in real-time and significantly faster turnaround times with 34 35 lower manual handling time. Recent reviews3, 5, 12, 13, 17 justify the importance of optical 36 nanobiosensors for biomarker detection. 37 38 39 40 A quick evaluation of the scientific community´s effort in the field of nanoplasmonic biosensors 41 is portrayed in Fig. 2, which shows the publication contribution, in the past decade, of advances 42 43 in nanoplasmonics covering all aspects of nanoplasmonic sensor substrate and/or associated 44 optical technologies including Surface Plasmon Resonance (SPR), Localized Surface Plasmon 45 46 Resonance (LSPR) and Surface enhanced Raman scattering (SERS). In order to gain a better 47 understanding of specific technologies that have moved towards clinical biomarker detection, we 48 49 analyze the above publications in two categories: “general” (all and any advances) and “biomarker 50 51 sensing” (publications which demonstrate some clinically relevant studies). Fig. 2 (right-most bar 52 for % translation) clearly depicts that SPR and LSPR technologies have moved into clinical 53 54 applications for biomarker detection while overall nanoplasmonic and SERS domains are 55 predominantly in the research phase. 56 57 58 59 60 Page 4 of 47 Page 5 of 47 Analyst 1 2 3 View Article Online DOI: 10.1039/C9AN00701F 4 5 6 7 8 9 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 13 14 15 Fig. 2. Statistics of published papers in the last decade (2007-onwards) demonstrating the higher 16 translation of SPR/LSPR technologies employed for biomarker sensing applications. Here, 17 “nanoplasmonics” refers to the work done in the overall area of plasmonic sensors and their 18 Analyst Accepted Manuscript applications in various optical technologies. Source: Web of Science. 19 20 21 22 This review contains a critical discussion of the recent advances in clinical biomarker evaluation 23 24 employing nanoplasmonic biosensors. Special attention has been focused on SPR and LSPR 25 26 biosensor devices employed as diagnostic, predictive, prognostic or monitoring tools for assessing 27 cancer, cardiovascular, autoimmune and neurodegenerative diseases. Current progress in the 28 29 analytical performance of biomarker evaluation in biological media has also been discussed. 30 Finally, we consider different approaches in multiplexing and miniaturization of plasmonic 31 32 biosensors as steps towards designing fully-automated POC instruments. 33 34 35 36 2. SPR and LSPR biosensors: working principles 37 38 39 SPR. Both SPR and LSPR are examples of surface plasmon (SP) based sensors. When an incident 40 41 light interacts with a noble metal, photons induce a collective oscillation of the free electrons in 42 the conduction band of the metal generating surface plasmon polaritons (SPP), a form of 43 44 electromagnetic (EM) waves, propagating at the sensor metal-dielectric interface. It is important 45 to note that only a p-polarized electromagnetic or transverse magnetic wave is able to sustain 46 47 SPPs. The simplest geometry sustaining SPPs is that of a flat interface between a metal and a 48 49 dielectric medium with dielectric constants of opposite signs. Gold, silver, copper, and aluminium 50 are some of the metals that feature a negative real and a small positive imaginary dielectric 51 52 constant and hence are able to support SPPs. In terms of choice of the sensor, silver (Ag) has the 53 largest negative real dielectric constant and hence is more sensitive to refractive index (RI) 54 55 changes, but has poor chemical stability as it is readily oxidized in air. On the other hand, gold 56 57 (Au) has lower RI sensitivity than Ag but has higher stability, as well as higher chemical affinity 58 which can be employed to easily functionalize the sensor surface for target biomolecule capture. 59 60 Thus, Au has a higher employability in plasmonic biosensors and SPR sensor chips are often Page 5 of 47 Analyst Page 6 of 47 1 2 3 made of a thin gold film of approximately 50 nm coated onto a dielectric substrate like glass View or Article Online DOI: 10.1039/C9AN00701F 4 5 silica. 6 7 The bulk RI sensitivity defined as the ability of the plasmonic sensor to detect changes in the RI 8 9 is widely used to understand the effect of different parameters on the sensor performance. The 10 bulk RI sensitivity (SB) is defined as (1) Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 𝑑𝜆𝑟 13 𝑆𝐵 = …………………….. (1) 𝑑𝑛𝐵 14 15 16 where λr is the SP excitation wavelength (resonant wavelength) and n is the RI of the medium in 17 contact with the sensor surface. The review by Špačková and co-workers18 discuss in detail the 18 Analyst Accepted Manuscript 19 effects of various parameters on SB including the type of supporting EM mode (decay length), 20 21 resonant wavelength, excitation geometry, and properties of the sensor. Among them, one of the 22 main factors that affect SB is the localization of the EM mode. The generated electromagnetic 23 24 plasmonic waves extend to around 100400 nm (SPR decay length) from the metal-dielectric 25 interface making it capable of detecting any biomolecular binding event via dielectric RI change 26 27 occurring within that range in the dielectric medium. 28 29 Additionally, the ability of a plasmonic sensor to measure minute changes in the RI is directly 30 31 proportional to SB and, furthermore, inversely proportional to the width  of the resonant spectral 32 33 feature being monitored. The figure of merit (FOM) (equation 2) is a combination of these 34 parameters and serves as a quantitative comparison of sensing efficiency among different 35 36 nanostructures. The importance and comparisons of FOM has been discussed in dedicated 37 reviews.13, 18 38 39 𝑆𝐵 40 41 𝐹𝑂𝑀𝐵 = …………………….. (2)  42 43 A combination of the above factors and the ability to tune the optical response specifically and 44 45 sensitively culminates into a limit of detection (LOD) with a sensitivity of 10-610-7 Refractive 46 47 Index Units (RIU), making SPR a desirable choice for optical biosensor technology. Fig. 3(a) and 48 (b) shows a standard SPR biosensor configuration along with data read-out of change in reflection 49 50 angle () as a snap shot measurement, as well as a real-time measurement. 51 52 On the other hand, several optical device configurations have been explored depending on the in 53 54 coupling of the light to excite the resonance, SPR devices can employed either a prism, grating 55 waveguide or fiber optic to achieve the excitation of surface plasmons.19, 20 The incident light 56 57 coupling in fibre optic sensors are guided by total internal reflection inside the fibre core 58 59 waveguide. At the fibre core waveguide-metal interface a evanescent field is generated. This 60 configuration is increasingly becoming popular due to numerous advantages such as smart size, Page 6 of 47 Page 7 of 47 Analyst 1 2 3 high resolution, flexibility, and miniaturization, which allow sensing in harsh conditionsDOI: and over View Article Online 10.1039/C9AN00701F 4 5 long distances. A recent review by Tokel and co-workers6 discusses the role of the prism dielectric 6 constant, energy-momentum conservation and the plasmon resonance angle which is highly 7 8 sensitive to minute changes in the refractive index (RI) at the metal-dielectric interface. A SPR 9 biosensor setup in the frequently used prism-coupled Kretschmann configuration is shown in Fig. 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 3(a). The main advantage of such a configuration is that it allows monitoring of the incidence 12 13 angle continuously and hence real-time detection is possible. 14 15 16 17 18 Analyst Accepted Manuscript 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 Fig. 3. Schematic overview of (a) the biosensing detection (read-outs) for SPR on a thin gold film 42 and (b) SPR read-out change in refection angle . (c) Miniaturization of the SPR instrument: from 43 large commercial SPR (Biacore, GE Healthcare) (left), to hand-held devices21 (middle) and plan- 44 held22 (right) Adapted with permission from Scientific reports, 2014, 4, 6789 and Sensors and 45 Actuators B: Chemical, 2010, 150, 1-6. Copyright (2010) Elsevier. 46 47 48 49 Furthermore, the light coupling strategies need to be efficient and for commercial high-throughput 50 51 devices can cost a fair share. Miniaturization of SPR devices (as shown in Fig. 3c) into portable 52 53 bench-top21 and palm-top22 set-ups has thus been investigated for boosting usability in clinical 54 settings. Microfluidics, as well as smart optical components have revolutionized this 55 56 transformation.4, 17, 19, 23-27 Interesting reviews of lab-on-chip integrated sensor devices justify the 57 strategies and advantages of such approaches for miniaturization.28 Researchers have also 58 59 included LED light source as opposed to laser, CMOS (Complementary Metal Oxide 60 Page 7 of 47 Analyst Page 8 of 47 1 2 3 Semiconductor) detector instead of CCD (Charge Coupled Device) camera, as well asDOI: a batteryView Article Online 10.1039/C9AN00701F 4 5 power source in order to minimize the size and the cost of the SPR instrument, making it attractive 6 for POC market.8, 21, 22 Some of the above features, which have mainly been implemented in the 7 8 research phase, have been depicted in Fig. 3c. Commercial SPR devices which are still 9 comparatively bulky, are available from approximately 20 companies and costs approximately 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 50-300k euros. Thus, the size and cost are major hurdles for its market penetration to every 12 13 decentralized clinic in the world. The other major drawback of SPR instruments is its limited 14 multiplexing capability; they often require multiple sensors chips (similar to Fig. 3c-middle panel- 15 16 d) in addition to multi-flow channels requiring complex microfluidics. Various reports and 17 reviews suggest improved nanoscale manufacturing, microarray assay development and 18 Analyst Accepted Manuscript 19 microfluidics channel design as some of the choices for achieving miniaturization and improving 20 21 cost-effectiveness.4, 19, 29-32 A review by Masson provides an outlook of SPR biosensors, its 22 limitations and benefits.33 23 24 25 26 SPRi. Following the initial discovery of SPR microscopy in 1998 where SPP field was used to 27 28 image microscopic interfacial structures, researchers explored the use of surface plasmon 29 resonance imaging (SPRi) for biosensing applications. SPRi allows detection and imaging in the 30 31 entire sensor area and provides an added benefit of multiplexed detection, as opposed to SPR. 32 33 SPRi sensors have been based on intensity, angular wavelength, phase, and polarization 34 interrogations. A more extended explanation of SPRi fundamentals can be found in reviews by 35 36 Wong and Olivio.34 One review compares SPR and SPR imaging35, while others discuss the 37 importance of sensor interface design, biofunctionalization36, 37 and multiplexing in SPRi25. 38 39 Though SPRi is a high throughput multiplexed technique, it suffers from reduced detection 40 sensitivity of about one order of magnitude compared with SPR sensing. This reduction is due to 41 42 the adaptation of a less compact optical configuration with a less sensitive mode of intensity- 43 44 based measurement and less sensitive detectors. The review by Liu and coworkers38 summarizes 45 the recent developments in SPRi sensitivity discussing signal enhancement and/or amplification, 46 47 and noise suppression. A more recent and detailed review39 of SPRi encompasses the significance 48 of optical platforms, functional coatings, biomolecular interactions (drug–receptor, protein– 49 50 protein, protein–DNA, protein–cell) and their applications of a broad variety of analytes (nucleic 51 acids, proteins, bacteria). Though there has been some progress in the field of SPRi, it is not yet 52 53 the most suitable technique for clinical applications. 54 55 56 57 58 LSPR. When metallic nanostructures interact with light, localized field oscillations are observed 59 inspiring the name “localized” SPR i.e., LSPR. In this situation, part of the incident light is 60 Page 8 of 47 Page 9 of 47 Analyst 1 2 3 absorbed and part is scattered. Both absorption and scattering are greatly enhanced DOI: when the View Article Online 10.1039/C9AN00701F 4 5 LSPR resonance mode is excited. The large amplification observed in extinction (absorption and 6 scattering) can be explained by an analytical solution of Maxwell´s equation with spherical 7 8 boundary conditions, referred to as Mie theory which illustrates the extinction spectra of a 9 nanoparticle. To extend this theory to more complex nanoparticles shapes and calculate dielectric 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 constants at different wavelengths, the Modified Long Wavelength Approximation (MLWA) of 12 13 Mie theory is used as stated below.40 14 3 15 24 𝜋2𝑅2𝜀2𝑚𝑁 𝜀𝑖 16 𝐶𝑒𝑥𝑡 = …………………….. (3) 17 λln (10) (𝜀𝑟 + χ 𝜀𝑚 )2 + 𝜀2𝑖 18 Analyst Accepted Manuscript 19 where R is the radius of the particle, λ is the wavelength of the incident light, εm is the dielectric 20 21 constant of the surrounding medium, ε = εr + iεi is the complex dielectric constant of the bulk 22 metal, N is the electron density, and χ accounts for the shape of the particle (equals to 2 for a 23 24 sphere and up to 20 for high-aspect ratio particles like nanorods). This proves the importance of 25 the plasmonic nanoparticle physical properties in LSPR sensing, as opposed to that in SPR. It is 26 27 also worth mentioning that with the increase in nanoparticle (NP) size, the relative contribution 28 29 of scattering to extinction increases, with scattering observed for NPs ≥30 nm. The above equation 30 (3) defines the importance of the different parameters including the shape of the NP, wavelength 31 32 of incident light, type of material, and last but most importantly, the surrounding media and its 33 dielectric constant. For a given nanoplasmonic system, a change in Cext and max (which lies in 34 35 the visible region for Au and Ag, making them highly important) will be observed if the media 36 37 surrounding of the nanoplasmonic system changes i.e., changes in its dielectric constant (ε) and 38 refractive index (n) related by ε = n2. This has been utilized in determining analytes or biomarkers 39 40 that can bind to a functionalized nanoplasmonic surface, where the refractive index change 41 translates into a LSPR max shift. Thus, the LSPR Δ read-out plot would be similar to that of 42 43 SPR  plot shown in Fig. 3b. Furthermore, the shift in LSPR frequency upon biomolecular binding 44 45 is described by the following relation: 40 46 47 ―2𝑑 48 ∆ = 𝑚(∆𝑛)[1 ― exp ( 𝑙𝑑 )] …………………….. (4) 49 50 Where, m is the refractive index sensitivity, Δn is the change in refractive index induced by the 51 biomolecule, d is the effective biomolecular layer thickness and ld is the electromagnetic field 52 53 decay length (approximated as an exponential decay length). 54 55 56 Therefore, the continuous thin gold film used for SPR would not be sufficient for LSPR. LSPR 57 58 sensor substrates demands nanostructured features onto a substrate, this is an added complexity 59 and cost factor as opposed to SPR technologies. The electromagnetic field decay length for metal 60 Page 9 of 47 Analyst Page 10 of 47 1 2 3 nano-surfaces is of the order of 1040 nm for NPs from the sensor surface as compared to View Article Online DOI: 10.1039/C9AN00701F 4 5 100400 nm for SPR continuous metal films. This thus improves surface sensitivity of the 6 7 nanoplasmonic substrates for biomolecular sensing which happens closer to the sensor surface. 8 As compared to SPR, LSPR has negligible contribution from the bulk sensitivity and bulk 9 10 temperature fluctuations and thus feature high nano-surface sensitivity. SPR requires an external Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 light coupling method, whereas, light is directly coupled to the sensor for LSPR. This thereby 12 13 reduces the complexity of the instrumentation and with the market availability of portable 14 sensitive spectrometers, LSPR demonstrates strong potential for clinical diagnostic applications. 15 16 A more complete description of SPR, LSPR and their comparison can be found in our previous 17 18 reviews41 and other literature reports42. Analyst Accepted Manuscript 19 20 21 22 23 24 3. Nanoplasmonic biosensors 25 26 27 Research in engineering and optimization of plasmonic nanostructures supporting SPP has been 28 29 immensely boosted due to its application as LSPR biosensors.43-46 The simplest form of substrate 30 31 fabrication methodology is the bottom-up approach where colloidal plasmonic nanoparticles 32 (NPs) are firstly synthesized and further employed for biosensing applications. In this domain, 33 34 extensive effort has been directed towards plasmonic NP synthesis (shape, size, composition 35 variants), linker directed self-assembly (morphology variants) for improving the stability, optical 36 37 response and reproducibility of such colloidal nanostructures.44, 47-50 Although, simple and cheap 38 39 to synthesize and covers a wide range of LSPR resonant wavelength of 400-1000 nm, its low 40 shelf-life restricts the applicability of colloidal NPs in clinical applications, especially where 41 42 laboratory-like controlled storage facilities are not available. Other factors include batch-to-batch 43 variation in production, presence of stabilizing agents that interfere with biosensing, as well as 44 45 the large handling or shipping volume (in colloidal form) limiting transportability. 46 47 48 In contrast, top-down nanostructures formed via lithography51-58 benefit from the long shelf-life, 49 50 easy transportability and ease of usage as sensor chips for detection. This has instigated a research 51 interest in fabrication techniques to obtain reproducible high performance nanoplasmonics. 52 53 Various fabrication techniques have been employed including e-beam lithography, mask 54 lithography, etc.57, 58 An interesting bottom-up lithographic technique referred to as the colloidal 55 56 lithography has gained limelight as it offers an easy control and fabrication of various shapes and 57 58 sizes onto the sensor substrate while still being a low-cost technology.51, 56 Nanoimprint 59 lithography is another low-cost and easy to implement technology which is gaining momentum 60 Page 10 of 47 Page 11 of 47 Analyst 1 2 3 for nanoplasmonic structure fabrication.57 Furthermore, nanostructures featuring increase of Article Online View DOI: 10.1039/C9AN00701F 4 5 polarizability via multipolar resonance, plasmon coupling for boosting LSPR, as well as 6 employing optical antenna structures (like arrays of nanocrescents, nanostars, nanocross etc. with 7 8 varying pitch) utilizing the lightening rod effect has been employed, few of which has been shown 9 in Fig. 4. 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 13 For example, plasmonic nanodiscs have been widely investigated where studies have focused on 14 fabricating them with control over the height, nanogap, diameter, as well as employing a trunk 15 16 for elevated nanodiscs.55, 59, 60 Another novel methodology uses Blu-ray discs as the base substrate 17 which inherently offers nano-gratings and which require only a one-step coating with 18 Analyst Accepted Manuscript 19 nanoplasmonic metals as Au or Ag (Fig. 4).61 Other structures like nano split rings, nanocrescents, 20 21 nanopillars have also gained interest. Nanoholes, have both LSPR properties, as well as 22 extraordinary transmission (EOT) properties.8 Interesting research is being conducted in forming 23 24 various array types with the nanoholes and features that resemble nanocups.62 In summary, taking 25 into account the increase in sensitivity, the plasmonic nanostructures can be categorized into 26 27 colloidal NPs with lowest sensitivity, followed by nanoholes and nanodiscs and, finally, by other 28 varied structures as nanopillars, nanocrescents, nanorings etc., offering highest sensitivity. 29 30 Nanoholes and nanodisc plasmonic structures operate in the wavelength range of 450-1000 nm 31 32 and are typically fabricated using colloidal or nanoimprint lithography.8 They offer higher shelf- 33 life and better scalability and hence are favored for mass production than other nanostructures 34 35 like nanopillars etc. which otherwise feature higher sensitivity. Nanopillars, nanocrescents55 36 require complicated fabrication technologies that are difficult to scale up and are less amenable 37 38 for biofunctionalization. Until now, nanodiscs and nanoholes have been dominating the field. 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 11 of 47 Analyst Page 12 of 47 1 2 3 View Article Online DOI: 10.1039/C9AN00701F 4 5 6 7 8 9 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 13 14 15 16 17 18 Analyst Accepted Manuscript 19 20 21 22 23 24 25 26 Fig. 4 (a) Nanodiscs with varying height and nanogaps55, (b) Blu-ray disk coated with 27 28 nanoplasmonic metals61, (c) nanodiscs both flat on the surface and elevated59, (d) nanohole 29 30 arrays8, (e) nanocup arrays62 and (f) nanocrescents55. Adapted with permission from Advanced 31 materials, 2016, 28, 4658-4664 and Journal of biophotonics, 2018, 11 (Copyright Wiley); Light, 32 33 science & applications, 2017, 6; and Biosensors & bioelectronics, 2015, 67, 237-242). Copyright 34 (2015) Elsevier. 35 36 37 38 39 4. Nanoplasmonic biosensors for clinical biomarkers analysis 40 41 4.1. Neurodegenerative diseases 42 43 The prognostic use of plasmonic biosensors for the diagnostics of neurodegenerative diseases 44 demands the characterization of aggregation prone proteins related to the etiology of Alzheimer’s, 45 46 Parkinson and Prion diseases. Although there is an increasing interest in the development of 47 neurodegenerative diagnostic tools, particularly for Alzheimer’s disease (AD), plasmonic 48 49 biosensors are still under-utilized in this domain. This gap is especially pronounced for the 50 51 diagnosis of Parkinson disease biomarkers, wherein the scarce use of optical biosensors mainly 52 relies on quantum dots-related approaches. 53 54 55 Amyloid beta peptides and tau proteins are the most frequent biomarkers employed as diagnostic 56 57 targets for Alzheimer disease. Among them, amyloid beta levels have been detected and evaluated 58 59 in serum and cerebrospinal fluid (CSF) by SPR, LSPR and SPRi biosensors. The most common 60 amyloid beta oligomers detected are 1-40 and 1-42, since they are frequently overexpressed in Page 12 of 47 Page 13 of 47 Analyst 1 2 3 amyloid plaques as a result of the degradation of amyloid precursor protein (APP). ForDOI: instance,View Article Online 10.1039/C9AN00701F 4 5 a shape-code LSPR nanoplasmonic biosensor for multiplex detection of amyloid beta (A beta) 1- 6 40, A beta 1-42 and tau protein at 34.9 fM, 26 fM and 23.6 fM levels, respectively, was developed 7 8 in mimicked blood by functionalizing heterobifunctional polyethylene glycol coated gold 9 nanoparticles with specific antibodies and detecting directly the target proteins.63 The shape-code 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 plasmonic biosensor was based on the optical properties of gold nanoparticles of three different 12 13 shapes and sizes that allow the detection of distinct conjugated biomolecules from separate 14 colored spots and plasmonic spectra. Determination of Alzheimer protein biomarkers onto the 15 16 antibody functionalized gold nanoparticles was measured by quantifying Rayleigh light scattering 17 shifts after analyte binding through dark-field microscope, a spectrograph, and a CCD camera. 18 Analyst Accepted Manuscript 19 When using SPR biosensors, A beta 1-42 have been detected in the 22 to 440 pM range with 20 21 waveguide-coupled bimetallic chips.64 The plasmonic biosensor makes use of Ag/Au films (Au 22 as an outer layer) in order to improve the sensitivity by increasing the reflectance change obtained 23 24 in an intensity interrogation detection mode. The assay proved to detect selectively A beta 1-42 25 from A beta 1-4a oligomer although results in human biofluids such as blood or cerebrospinal 26 27 fluid were not shown. 28 Other prominent biomarker of Alzheimer’s and Parkinson disease, detected either in plasma or 29 30 CSF samples, is tau Protein which is associated with defective aggregation of microtubules and 31 32 abnormal expression in neurodegenerative diseases. 33 Recently, Truong et al. described an immunoassay capable of detecting tau and phosphorylated 34 35 tau proteins, characteristic of tau pathologies, at pM in human sera of AD patients employing a 36 SPR fiber sensor.65 The detection method consisted of the conventional Tau antibody 37 38 immobilization onto a thiol Self-Assembled Monolayer (SAM) followed by direct detection of 39 40 total and phosphorylated tau proteins. The analytical performance was tested using real blood 41 samples of 40 participants over 65 years, divided equally into experimental and control groups. 42 43 The results obtained contribute to define the rule of unphosphorylated tau proteins in the progress 44 of AD disease since concentrations of phosphorylated tau proteins (3-fold higher) were less 45 46 increased than tau proteins (6-fold higher) in AD disease patients when compared with the control 47 group. This suggests that unphosphorylated tau proteins are more likely to be produced in blood 48 49 of AD patients than phosphorylated tau proteins. In addition, the compact format of the SPR 50 51 device envisages its potential application for POC testing of tau proteins present in human blood 52 in the POCT mode. 53 54 Another SPR approach has demonstrated the determination of human tau 381 protein at 10 fM 55 levels in undiluted plasma66 by a multichannel platform. The detection format consisted of a DNA 56 57 aptamer/antibody sandwich assay over a tailored mixed monolayer on the sensor surface. Results 58 were validated by ELISA showing a 1000-fold sensitivity improvement for SPR, allowing Tau 59 60 determination in both spiked and native samples (see Fig. 5). Page 13 of 47 Analyst Page 14 of 47 1 2 3 View Article Online DOI: 10.1039/C9AN00701F 4 5 6 7 8 9 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 13 14 15 16 17 18 Analyst Accepted Manuscript 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Fig 5. SPR Analysis in plasma for three different chip monolayer compositions: 60% 11- 41 42 mercaptoundecanoic acid (MUA):40% 11-mercaptoundecanol (MUD), 40% MUA:60% MUD, 43 60% MUA:40% PEG monolayers. (a), (d), (g) Real-time SPR evaluation of anti-Tau adsorption 44 45 following exposure to different Tau concentrations spiked directly into plasma, P. (b), (e), (h) Δ 46 47 R.U plots obtained by calculating difference in R.U. signals in buffer before and after plasma 48 exposure. (c), (f), (i) is where the Δ R.U signal has been normalized with respect to the average 49 50 Δ R.U. signal of all non-specific controls (NC). The filled data point marker in the Δ R.U and 51 normalized Δ R.U. plots represents the plasma signal with no spiking. Representative NC3(anti- 52 53 recombinant human complement) and NC4 (aptamer control sequence) plots are also shown 54 55 in (a), (d) and (g). 56 Reprinted with permission from (Analytical Chemistry, 2016, 88, 7793-7799). Copyright (2016) 57 58 American Chemical Society. 59 60 Page 14 of 47 Page 15 of 47 Analyst 1 2 3 Finally, an original approach involving SPR detection by employing multiwalled carbonView Article Online DOI: 10.1039/C9AN00701F 4 5 nanotubes (MWCNTs) was reported by Lisi et al.67 The functionalization of the nanotubes with 6 a secondary Tau antibody afforded tau determination using a sandwich detection format over a 7 8 previously immobilized primary antibody layer. A signal amplification of 102-fold was obtained 9 in comparison with the direct and conventional unconjugated format, reaching a detection limit 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 of 125 pM. The major limitation of the study is referred to the lack of analysis in real biological 12 13 fluid since all the measurements were performed in buffer or in artificial cerebrospinal fluid. 14 Nevertheless, further improvement in avoiding matrix interference would permit its potential 15 16 transfer as a disposable device for POC testing. 17 18 Analyst Accepted Manuscript 19 Additionally, prion protein detection has been explored by SPR as a promising method for 20 21 screening of other neurodegenerative diseases such as the Creutzfeldt Jakob syndrome and its 22 variant Bovine spongiform encephalopathy (BSE).68 Lou et al. described an aptamer-based assay 23 24 that exploits the magnetic properties of microspheres to selectively capture prion proteins from 25 complex environments, while preventing interference effects of other compounds present in the 26 27 same sample. The assay comprised of the fabrication of magnetic nanoparticles and the 28 subsequent immobilization of aptamers by glutaraldehyde cross-linking. At this stage, Prion 29 30 proteins could be captured by the functionalized magnetic nanoparticles, although the amphiphilic 31 32 copolymer was required to embed and isolate prion proteins from the environment. Finally, the 33 generation of an alternating magnetic field was needed to release the labeled prion proteins and 34 35 allow SPR detection. In spite of the number of multiple steps and the complicated assay sequence, 36 prion proteins were evaluated in the 0.01-1000 ng mL−1 range and demonstrated good correlation 37 38 with determination in serum samples. 39 40 41 The identification of amyloid fibrils as final products of α-synuclein aggregation pathways may 42 43 be associated with the diagnosis and treatment of several neurodegenerative diseases. A 44 nanoplasmonic biosensor was employed to investigate the formation of amyloid fibrils as a result 45 46 of the aggregation of α-Synuclein proteins by using the chiroptical properties of helical fibril 47 proteins to induce chiral nanoparticle assembly of gold nanorods.69 Furthermore, the detection of 48 49 infectious prion proteins is reported by monitoring the circular dichroism (CD) signal at the 50 51 longitudinal plasmon wavelength after the arrangement of prion fibrils to gold nanorods as a 52 consequence of the helical assembly of gold nanoparticles to the fibrilar surface. The formation 53 54 of helical nanorods was associated with the identification of α-Synuclein protein fibrils in samples 55 of human brain homogenates of patients with Parkinson’s disease in contrast with the absence of 56 57 optical activity observed in human healthy brain samples. The plasmonic technique therefore 58 opens a route not only to the early detection of the Parkinson disease and novel therapeutic 59 60 Page 15 of 47 Analyst Page 16 of 47 1 2 3 strategies but also to a methodology that provides better understanding of the characterization of Article Online View DOI: 10.1039/C9AN00701F 4 5 aggregated species and template-driven plasmon chirality. 6 7 8 4.2. Cancer biomarkers 9 The detection of cancer biomarkers as predictive, diagnostic and prognosis tools by plasmonic 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 biosensors takes advantage of their multiplexing performance for the rapid and simultaneous 12 13 screening of a variety of biomarkers. The selection of the most appropriate cancer biomarkers is 14 crucial since a wide range of biomolecules could be included in this category according to the 15 16 clinical findings. Thereby, a universal classification relying on the biological nature of the 17 biomarker should be considered in order to include the main cancer biomarkers categories, 18 Analyst Accepted Manuscript 19 namely: proteins, exosomes, nucleic acids and circulating tumor cells. 20 21 22 4.2.1. Proteins 23 24 25 The interest in the determination of protein cancer biomarkers relies on their relevance to 26 27 demonstrate translational alterations in genetic related oncology disorders while permitting the 28 quantification of deviations in body fluid concentrations between cancer patients and healthy 29 30 subjects. Among proteins, the most significant and commonly reported biomarkers for cancer 31 32 screening have been PSA, VEGF and CEA. 33 PSA 34 35 “The prostate-specific antigen (PSA) has been extensively used as the most accepted biomarker 36 for the diagnosis of prostate cancer, in spite of its low accuracy and limited specificity.”. PSA is 37 38 a glycoprotein produced by the prostate gland commonly measured in serum as the sum of free 39 40 (fPSA) and complexed PSA. Although total PSA (tPSA) concentrations above 10.0 ng mL−1 may 41 indicate high probability of prostate cancer and tPSA concentrations below 4.0 ng mL−1 are 42 43 normally found in healthy subjects, patients with tPSA levels between 4.0 ng mL−1 and 10.0 ng 44 mL−1 are included in a diagnosis grey zone. Therefore, the assessment of novel biomarkers in 45 46 serum is essential for the diagnosis of prostate cancer malignancy. Since sensitive determinations 47 of fPSA are difficult to obtain by conventional methods, as concentration in serum is below 1 ng 48 49 mL−1, the fPSA/tPSA ratio has recently become a relevant diagnosis value for differentiating 50 51 between benign and malignant states. In this sense, several SPR and combined nanoplasmonic 52 approaches have reported PSA detection of either total or free/total serum fractions, involving 53 54 from immunoassay to microcontact patterning and aptamer-based assay deposition techniques. 55 56 57 Ertuk et al. described a bottom-up functionalization strategy based on the fabrication of SPR 58 sensor chips with PSA imprints.70 The method makes use of methacrylate derivatives and UV 59 60 polymerization to imprint the PSA protein via microcontact patterning. The design of the SPR Page 16 of 47 Page 17 of 47 Analyst 1 2 3 biosensor chip allowed PSA determination at highly sensitive levels (91 pg mL-1), improving the View Article Online DOI: 10.1039/C9AN00701F 4 5 limit of detection of conventional analytical methods by an order of magnitude. Clinical analysis 6 was performed in spiked human serum samples (1/4 diluted) of prostate cancer patients. 7 8 Additionally, relevant analytical parameters such as specificity against another proteins (HSA and 9 lysozyme) and association kinetic analysis were also reported, while good results were observed 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 for recovery, reusability and correlation with ELISA immunoassays. This method shows an 12 13 alternative patterning strategy to conventional molecular imprinting techniques enabling a high 14 specificity and binding capacity of the template molecule, while maintaining long-term 15 16 conformational stability of the recognition layer. These features are inherent advantages that 17 enable its application as a POC detection platform. 18 Analyst Accepted Manuscript 19 Another approach involving a micropatterned plasmonic strategy is described by Breault.71 The 20 21 method employs a dual detection system that combines the wavelength interrogation of a four- 22 channel SPR instrument and a fluorescence microscope mounted on the same SPR platform. SPR 23 24 biosensor chips were fabricated by photolithography to generate microhole arrays thus permitting 25 fluorescence detection from the solution side of the micropatterned gold film and SPR 26 27 interrogation using the interface between the glass prism and the gold film. The fabrication of a 28 micropatterned surface by incorporating microhole arrays allowed the improvement of SPR 29 30 response in 2-3 orders of magnitude. Likewise, the use of a sandwich-immunoassay format 31 32 comprising a capture anti-PSA antibody immobilized at the sensor chip and a labeled 33 horseperoxidase detection antibody incubated with Ampliflu™ Red solution increases the 34 35 sensitivity when using fluorescence detection in comparison with the SPR format. The singular 36 combination of SPR and fluorescence assays in a single chip using the same microfluidics 37 38 provided highly sensitive PSA determinations at clinical concentrations between 142 pg mL-1 and 39 40 1.42 g mL-1. 41 Discrimination between PSA free and total serum fractions from benign and malignant disease 42 43 state has been reported by Jiang and co-workers using a dual-channel SPR immunosensor72. Total 44 PSA fraction was measured directly with a linear range from 1.0 to 20.0 ng mL-1, whereas the 45 46 free fraction was detected via an asynchronous competitive inhibition immunoassay enhanced by 47 48 gold nanoparticles, reaching a linear range between 0.010 to 0.40 ng mL-1. By measuring 49 simultaneously total and free PSA serum fractions in each channel the f/t-PSA ratio was 50 51 calculated showing significant differences between benign prostatic hyperplasia and prostatic 52 cancer patients. The application of this methodology provides meaningful clinical information 53 54 about prostate cancer diagnosis in addition to the analytical results. 55 Single detection of PSA free fraction has also been reported using a LSPR-coupled fiber-optic 56 57 biosensor.73 A direct immunoassay was employed, based on the anti-PSA antibody 58 59 immobilization onto a SAM modified sensor chip. The major contribution is the fabrication of a 60 gold nanodisc at the fiber end facet allowing the reusability of the dielectric-metallic hybrid Page 17 of 47 Analyst Page 18 of 47 1 2 3 interface and a detection limit of 0.85 ng mL-1. The development of a a miniaturized multiplexedView Article Online DOI: 10.1039/C9AN00701F 4 5 POC device is suggested although. clinical applications were not demonstrated since PSA 6 evaluation was only done in buffer. 7 8 9 Another LSPR system with functionalized gold nanodiscs array is presented by Khan et al.74, 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 where PSA detection was performed by using DNA aptamers as specific receptors. The technique 12 13 takes advantage of the symmetry of gold nanodiscs thus providing detectable responses in a short 14 range near the vicinity of the sensor surface. Another strategy presented was the functionalization 15 16 of gold nanostructures with PSA specific aptamers providing a limit of detection of 1.49 ng mL-1 17 within a linear range of 1.7 to 20.4 ng mL-1 in buffer. The interference in human serum was also 18 Analyst Accepted Manuscript 19 tested showing good recovery values. Likewise, the reusability of the gold sensor was also 20 21 demonstrated by removing the applied aptamers with NaBH4 solution. 22 23 24 VEGF 25 Vascular endothelial growth factor (VEGF) is a hypoxia-inducible protein that promotes 26 27 vasculogenesis and angiogenesis. VEGF appears frequently upregulated in most human tumors 28 and it is a valuable cancer-specific protein biomarker for assessing the disease status during cancer 29 30 progression and metastasis. SPR biosensing of VEGF is primarily based on aptamer or antibody 31 32 assay formats. 33 One of the most singular approaches involve the application of rolling circle amplification (RCA) 34 35 to enhance the SPR signal by targeting different VEGF domains with two DNA aptamers, 36 employed as capture and detection probes, respectively. VEap121 is named as Aptamer 1 and 37 38 provides high affinity to the receptor binding domain of VEGF165 while VEa5 or Aptamer 2, 39 40 shows high binding affinity to the VEGF165 heparin binding domains. The aptasensor is based on 41 the immobilization of Aptamer 1 for monitoring VEGF in real-time. The RCA process makes use 42 43 of hybridization with the primers of the aptamer 2- polystyrene microspheres complex75 to 44 increase the VEGF sensitivity. A limit of detection of 100 pg mL-1 with a linear range from 1x10- 45 46 12 to 1x10-5 g mL-1 was achieved. The method shows the potential applicability of RCA to assist 47 48 SPR signal enhancement although lacks results with real samples (see Fig. 6). 49 50 51 52 53 54 55 56 57 58 59 60 Page 18 of 47 Page 19 of 47 Analyst 1 2 3 View Article Online DOI: 10.1039/C9AN00701F 4 5 6 7 8 9 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 13 14 15 16 17 18 Analyst Accepted Manuscript 19 20 Fig 6. (a) Concentration dependence of VEGF detection from 1x10-12 to 1x10-5 g mL-1, (b) the 21 contrast of with (red points) or without (black points) the SPR signal amplification by RCA 22 23 process. Reprinted with permission from (Biosensors and Bioelectronics 2014, 61, 83–87). 24 Copyright (2014) Elsevier. 25 26 27 28 Another VEGF biosensor application based on DNA aptamer is presented by Cennamo et al.76 29 The main characteristic of this system is the integration of a plastic optical fiber into the SPR 30 31 platform avoiding the prism-based Kretschman configuration. Following the aptamer layer 32 characterization, VEGF was detected directly at 81 ng mL-1 levels. Further experiments related to 33 34 other analytical parameters such as selectivity, reusability or interference of complex matrix are 35 not reported. 36 37 A different approach employs the immobilization of the fragment crystallizable (Fc) region of an 38 39 anti-VEGF antibody, reached a linear detection range of 0.01-100 ng mL-1 for VEGF165 detection 40 in human serum. The biosensor comprised of nanogold dot array that exploits the localized surface 41 42 plasmon properties at narrow bandwidths. The specificity of the plasmonic biochip was tested by 43 analyzing the interference of glucose and ascorbic acid and was also correlated with ELISA 44 45 results.77. 46 47 48 49 50 CEA 51 Another specific protein biomarker for cancer diagnosis and prognosis is the Carcinoembryonic 52 53 antigen (CEA) glycoprotein. CEA is a well-established tumor biomarker with increased reported 54 levels in several types of cancer at several locations (tumor size, staging) from the gastrointestinal 55 56 tract to body organs such as liver, lung, ovary, prostate or pancreas. 57 58 Clinical determination of CEA in serum and blood plasma has been reported using SPR 59 biosensors employing different immobilization formats. An immobilization method based on 60 Page 19 of 47 Analyst Page 20 of 47 1 2 3 biofunctionalized gold nanoparticles (bio-AuNPs) with either streptavidin or bovine serumView Article Online DOI: 10.1039/C9AN00701F 4 5 albumin was described to determine CEA in blood plasma at 0.1 ng mL-1 levels. The assay made 6 use of both high affinity of bio-AuNPs and lack of nonspecific binding towards the surface of the 7 8 SPR sensor to improve the LOD by a factor of more than 1,000 in comparison with previous CEA 9 determinations in undiluted plasma.78 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 CEA glycoprotein has also been detected via a two-channel SPR platform in spiked human serum 12 13 by exploiting streptavidin affinity in different detection formats, including direct, sandwich and 14 gold nanoparticles enhanced sandwich assays. The most sensitive detection format was obtained 15 16 for the streptavidin-gold nanoparticles enhanced sandwich format allowing CEA determination 17 at 1.0 ng mL-1 levels while achieving an improvement of 13.2-fold in sensitivity in comparison 18 Analyst Accepted Manuscript 19 with the direct format. The specificity of the assay was proved against other protein cancer 20 21 biomarkers, such as alpha fetal protein and prostatic specific antigen.79 22 23 Another recent approach80 described the fabrication of a liquid core coupling unit as an 24 25 amplification method to increase the sensitivity of a SPR biosensor by changing the liquid core 26 27 refractive index. The so-called liquid core SPR coupling unit consisted of a semi-cylindrical 28 container with an inlet and an outlet and a micro flow cell with a volume of 250 μL assembled to 29 30 the SPR chip via a poly(dimethylsiloxane) (PDMS) layer. The applicability of the method was 31 tested for CEA determination in real serum samples from early stage cancer patients and healthy 32 33 subjects, showing an 18-fold increase of concentration in cancer samples in comparison with 34 35 controls. Similarly, another novel strategy exploited81 the surface plasmon field enhanced 36 fluorescence (SPFS) from plasmonic nanogratings to create a protein array with a LOD of 0.36 37 38 ng mL-1, showing a 10-fold signal enhancement due to the fluorescence amplification in 39 comparison to the bare glass sensor surface. 40 41 42 43 44 ALCAM 45 Activated leukocyte cell adhesion molecule (ALCAM) is a cell surface glycoprotein of the 46 47 immunoglobulin superfamily that mediates adhesion and interactions with the CD6 immune 48 antigen. ALCAM is overexpressed in many types of cancers such as colorectal, gynecological, 49 50 and pancreatic while it is associated with poor prognosis for esophageal, breast and ovarian 51 cancers. 52 53 The analysis of ALCAM has been explored by plasmonic biosensors using both SPR imaging and 54 55 LSPR detection formats. Homola’s group82 described an SPR imaging sensor that was able to 56 detect human chorionic gonadotropin (hCG) and ALCAM with a limit of detection of 45 ng mL- 57 58 1 in 10% diluted blood plasma. The SPRi platform comprised of a high-density protein array with 59 120 sensing areas. DNA directed immobilization with antibody-DNA conjugates were 60 Page 20 of 47 Page 21 of 47 Analyst 1 2 3 microspotted on the sensor surface in combination with covalently immobilized bovine serumView Article Online DOI: 10.1039/C9AN00701F 4 5 albumin to reduce non-specific binding. 6 Another approach to detect ALCAM antigen was reported by Pai et al.83 using an immunoassay- 7 8 based LSPR sensor. The main feature of this assay was the incorporation of polyethylene glycol 9 adlayers in order to optimize the immobilization of gold nanorods bioconjugated to anti-ALCAM 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 antibodies. The model was able to detect ALCAM at 1.253 ng mL-1 in buffer although no other 12 13 analytical parameters were reported. The ALCAM gene expression has also been monitored by a 14 LSPR design involving the use of nanoparticles as the target-specific probes capable to detect 15 16 specific sequences of DNA or RNA. 17 18 Analyst Accepted Manuscript 19 4.2.2. Nucleic acids 20 21 22 Telomerase activity 23 24 Telomerase is a ribonucleoprotein reverse transcriptase that adds tandem telomeric repeats units 25 (TTAGGG)n to the 3' end of telomeres in order to prevent chromosomal DNA damage. The 26 27 expression of telomerase differs between normal and cancer cells. In normal cells, telomerase 28 activity is inhibited and the telomeres length diminishes after each replication cycle. In contrast, 29 30 the reactivation of telomerase is observed in over 85% of human cancer types thus inducing the 31 32 division of cancer cells indefinitely by counteracting telomere shortening. Monitoring of 33 telomerase over-expression during tumor diagnostic and therapy emerges as a promising tool 34 35 while conferring to telomerase activity a valuable role as tumor biomarker. 36 Conventional detection methods take advantage of the formation of G-quadruplex structures by 37 38 stranded repeats of TTAGGG for the development of telomere repeat amplification protocols 39 40 based on PCR and electrophoresis procedures. Although extensively used, the sensitivity and 41 reproducibility of the conventional method may be affected by the contamination of the sample 42 43 and can produce false results. To overcome these limitations, several approaches have measured 44 telomerase activity employing plasmonic biosensors. 45 46 Commonly, dual-detection mode strategies based on either fluorescence or colorimetric 47 functionalities in combination with plasmonic properties such as SERS, Rayleigh scattering and 48 49 LSPR have been described. 50 51 52 For instance, Wang et al.84 described a LSPR biosensor combining plasmonic resonance Rayleigh 53 54 scattering spectroscopy with dark-field microscopy (DFM) for in-situ detection and intracellular 55 monitoring of telomerase activity in response to telomerase-based drug treatments. The 56 57 biosensing method comprised core-satellites assembled nanostructures of Au50@Au13 58 functionalized with single chain DNA and complemented with the nicked hairpin DNA of O1. 59 60 The increase of telomerase activity induced the hybridization of the telomeric repeated sequence Page 21 of 47 Analyst Page 22 of 47 1 2 3 extended at the 3’-end of O1 to its complementary sequences at 5’-ends. Due to the elongation of Article Online View DOI: 10.1039/C9AN00701F 4 5 O1, a rigid hairpin structure was formed and the core-satellites nanostructure of Au50@Au13 6 might disassemble with the release of O1. The increase of telomerase activity caused by the 7 8 disassembly of Au50@Au13 was measured in HeLa cancer cell lines by the LSPR spectral shift 9 accompanied with color changes from orange to green with a detection limit of 1.3×10-13 RIU. 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 13 MicroRNA 14 15 16 MicroRNAs (miRNAs) are short (18–24 nucleotides) non-coding RNA molecules that regulate 17 gene expression by inhibiting the translation of mRNA. Their use as biomarkers is prominent 18 Analyst Accepted Manuscript 19 since deregulation of miRNA signaling pathways may affect biological functions from gene 20 21 expression to cellular processes, while can be also associated with cardiovascular or neurological 22 disorders and the development of a wide number of cancer types. Due to their valuable role as 23 24 protein translation regulators and the difficulties for an accurate and sensitive detection, the 25 number of miRNAs plasmonic applications has increased remarkably in the last five years. 26 27 28 For instance, noninvasive monitoring of exosomes is particularly relevant in clinical analysis as 29 30 they are extracellular microvesicles secreted by cells and neoplastic cell lines present in blood, 31 32 urine, saliva and breast milk which contain cytosolic and plasma proteins, lipids, DNAs and 33 RNAs. The major RNA component of exosomes is micro-RNA (miRNAs) which are of special 34 35 interest for cancer monitoring as they can be associated with epigenetic changes in the recipient 36 cells. Exosomes promote intercellular communication without direct cell-cell contact and are 37 38 involved in various physiological and pathological processes such as cancer development, 39 40 migration, metastasis, treatment resistance and the regulation of immune responses. 41 The use of exosomes has appeared to be a promising tool for the study of cancer development and 42 43 metastasis. Among exosomal proteins, CD63, CD9, CD24, CD44, EpCAM, HER2, EGFR, LRP1, 44 and LG3BP have been reported as potential biomarkers and have been already evaluated by 45 46 plasmonic biosensors. For instance, two studies report the determination of exosomal expression 47 derived from mast cell lines85 and breast cancer cell lines86 respectively by using a SPR 48 49 commercial biosensor. Rupert et al. quantified the CD63 expression of the total mass of exosomes 50 51 in µg L-1 levels while the latter one was able to distinguish between the exosomal expressions of 52 EpCAM in different types of cells in real plasma samples from healthy subjects. 53 54 Another approach involving a SPR biosensor was developed by Sina et al. in order to isolate bulk 55 exosome populations and detect exosome subpopulations using the tumor-specific marker human 56 57 epidermal growth factor receptor 2 (HER2). The assay was capable of identifying that 14-35% of 58 the bulk population from a small cohort of breast cancer patients expressed HER2. Patients with 59 60 Page 22 of 47 Page 23 of 47 Analyst 1 2 3 HER2+ exosomes exhibited 10-fold higher HER2 exosome concentration than HER2 and healthyView Article Online DOI: 10.1039/C9AN00701F 4 5 patients.87 6 Alternative immobilization methods take advantage of nanoholes87 or colloidal gold enhancement 7 8 strategies to exploit nanoplasmonics.88 For example, Park et al. designed a nanoplasmonic system 9 that uses nanohole arrays as a base sensing element. Multiplexed detection of both transmembrane 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 (EpCAM and CD63) and intravesicular (AKT1) proteins in exosome vesicles derived from 12 13 ovarian cancer cells was carried out by a plasmonic sensor with 100 sensing sites, showing good 14 correlation in the expression profiles between unperturbed cells and their exosome vesicles (see 15 16 Figure 7). The detection format enabled the identification of drug-specific protein signature from 17 drug-treated cancer cells, although the interference generated by the binding of AuNP aggregates 18 Analyst Accepted Manuscript 19 and the use of a pair of antibodies for target capture and detection are some of its limitations.88 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 Fig.7 New intravesicular nanoplasmonic system (iNPS) assay. (a) EVs are lysed to release 49 50 molecular cargos. Each target is captured on the iNPS chip via affinity ligands, and further labeled 51 52 with Au nanoparticles (AuNPs). Note that a single assay format is used both for transmembrane 53 and intravesicular proteins. (b) Scanning electron micrograph showing AuNPs after iNPS assay 54 55 steps. (c) FDTD electromagnetic simulation. AuNP on the iNPS surface concentrates electrical 56 fields. Compared to a protein binding (left) or a whole Extracellular Vesicles (EV) binding 57 58 (middle), the field intensity enhanced up to 70-fold with AuNP (right). (d) Measured signal 59 60 enhancement. Compared to EV binding, the spectral shift was about 9-fold higher when the same Page 23 of 47 Analyst Page 24 of 47 1 2 3 concentration of AuNPs (100 nm) bound to the iNPS chip. (e) Validation of iNPS assay. AuNPs View Article Online DOI: 10.1039/C9AN00701F 4 5 enables both membrane protein (EpCAM, CD63) and intravesicular protein (AKT1) detection 6 with enhanced spectral shifts. The error bars represent the standard deviation of signals. Reprinted 7 8 with permission from (ACS Photonics, 2018, 5 (2), pp 487–494). Copyright (2018) American 9 Chemical Society. 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 13 As a last example, Di Noto et al.89 exploited the colloidal properties of myeloma multiple (MM) 14 derived exosomes by a colorimetric nanoplasmonic assay. The method allowed the determination 15 16 at nanomolar concentration and cell protein binding. They demonstrate that exosomes are present 17 in four-fold higher concentrations in myeloma multiple patients in comparison with healthy 18 Analyst Accepted Manuscript 19 subjects, while only MM-derived exosomes have a significant binding affinity for heparin, 20 21 indicating that exosomes show different binding affinities for cell-associated heparan sulfate 22 proteoglycans. 23 24 25 Regarding to microRNA specific detection, LSPR and SPR biosensors have mainly exploited 26 27 nanoparticle and amplification approaches. For example, Ding et al.90 described the detection of 28 real miRNA from total RNA extracted from human breast adenocarcinoma MCF-7 cells at 9 pM 29 30 levels by a SPR biosensor. The detection method is based on the formation of DNA super- 31 32 sandwich assemblies between target miRNA and surface bound DNA-probes after hybridization. 33 Signal amplification was observed via streptavidin-biotin binding to the DNA sandwich structure. 34 35 The specificity of the assay was tested against five different miRNA sequences, while the 36 interference of complex components was evaluated by measuring spiked samples containing 37 38 miRNA and total RNA, showing good recovery values (108.3%). Although the assay lacks 39 40 evaluation in real biological samples, these results demonstrate its potential for discriminating a 41 single base mismatched miRNA sequence in complex matrices. 42 43 44 Another laboratory custom designed SPR platform has been described for the detection of 45 46 miRNA-145, a miRNA sequence involved in tumor suppression of human colon and gastric 47 cancer cell lines. The SPR based-assay takes advantage of modified parallel tail-clamps carrying 48 49 8-amino-2′-deoxyguanosine derivatives, which are used as bioreceptors, to form stable triplex 50 51 structures with their target miRNA-145. The results demonstrated that the formation of triplex 52 structures stabilizes the hybridization and is more stable than the conventional duplex format, 53 54 improving the detection limit 1.5 times (Fig. 8). This work contributes to strengthen the potential 55 of label-free methods to profile miRNA expression by evaluating alternative approaches, such as 56 57 the introduction of parallel tail-clamps bioreceptors with a short polypyrimidine track to form the 58 most stable triplex structures.91 59 60 Page 24 of 47 Page 25 of 47 Analyst 1 2 3 View Article Online DOI: 10.1039/C9AN00701F 4 5 6 7 8 9 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 13 14 15 16 17 18 Analyst Accepted Manuscript 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Fig. 8 Scheme of the gold sensor chip functionalized with the 3AG PT clamp bioreceptor (a). 33 34 Schemes of the different bioreceptors employed for the SPR biosensor study. LODs obtained for 35 36 each bioreceptor are indicated (b). SPR calibration curves of the detection of miR-145 with 37 different bioreceptors. (c). Real-time sensorgrams of 30 nM miRNA-145 using the different 38 39 bioreceptor variants. Dashed arrow indicates the equilibrium signal of the SPR sensorgrams 40 (d)Reprinted with permission from (Anal Bioanal Chem 2016, 408, 885–893). Copyright (2016) 41 42 Elsevier. 43 44 45 SPR analysis involving three-one dimensional spot arrays parallel designed for miRNA-93 46 47 capture by locked nucleic acid (LNA) immobilized probes is presented by Shmieder et al.92 Two 48 distinct amplification strategies were used to increase the assay sensitivity. First, an RNA-DNA- 49 50 hybrid antibody binding to the LNA-RNA double strand after successful hybridization of 51 miRNA-93, enabled miRNA-93 detection in RNA lysates from HEK-293 cell cultures with a 52 53 limit of detection of 10 pM. Alternatively, signal amplification by using the Poly(A) polymerase 54 55 enzyme after binding to the LNA probe reached a limit of detection of around 1 fM. 56 57 58 Several approaches involving LSPR biosensor have been reported by exploiting gold 59 nanoparticles. Hu et al.93 described trace oligonucleotide detection by measuring changes in the 60 Page 25 of 47 Analyst Page 26 of 47 1 2 3 dielectric constant on the surface of DNA modified gold nanoparticles after hybridization between View Article Online DOI: 10.1039/C9AN00701F 4 5 probe single-strand DNA and miRNA at 3 nM level. The system made use of a single plasmon 6 nanoparticle sensor based on Surface Plasmon Resonance Scattering (SPRS) to monitor a red shift 7 8 during the process of hybridization due to the change of the refractive index of the surface of 9 individual gold nanoparticles. 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 13 Similarly, a singular LSPR approach reports MicroRNA205 (miR-205) detection by monitoring 14 the variation of the dielectric constant which causes a red shift in the LSPR scattering spectra 15 16 during the hybridization of the target miRNA with a single DNA modified on the surface of a 17 gold nanocube94. Measurements were carried out in serum samples with a limit of detection of 5 18 Analyst Accepted Manuscript 19 pM showing an improvement of three orders of magnitude in comparison with the detection 20 21 efficiency of gold nanospheres (Fig.9). This achievement agrees with the theoretical analysis 22 about the effect of the nanocrystal configuration, since the presence of more vertexes and fewer 23 24 faces of polyhedral nanoparticles induce more surface plasmons in a wider energy range. 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 Fig 9. (a) LSPR scattering spectra of one AuNC-ssDNA probe treated with 1 μM miRNA sample 54 55 at different times. (b) LSPR spectra peak redshifts with the time on various AuNC-ssDNA probes 56 with miR-205 in different concentrations (10 pM, 100 pM, 1 nM, 10 nM, 100 nM, 1 μM). (c) 57 58 Calibration curve of the LSPR peak shift versus different concentrations of miR-205 (10 pM, 100 59 pM, 1 nM, 10 nM, 100 nM, 1 μM). (d) LSPR Peak shift after treatment of 100 nM target miR- 60 Page 26 of 47 Page 27 of 47 Analyst 1 2 3 205 with different modification times of 1 μM probe ssDNA. Reprinted with permission from View Article Online DOI: 10.1039/C9AN00701F 4 5 (ACS Sens. 2017, 2, 1435−1440). Copyright (2017) American Chemical Society. 6 7 8 Sensitivities in the nanomolar range (LOD of 10 nM) were obtained for miRNa-155 detection by 9 10 combining LSPR biosensing and colorimetric changes in the UV-visible spectrum95. The main Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 singularity of the method consisted on the aggregation of the gold nanoparticles without ligating 12 13 a linker to increase miRNA length in order to shorten the process to a single step. 14 Significant higher sensitivity, at attomolar levels, was achieved by Na et al.96 by using a LSPR 15 16 biosensor based on a 3D plasmonic nanostructure comprising gold strips on a 4-in.-scale substrate. 17 Determination of miRNA in total RNA from extracts of cancer lines allowed for discrimination 18 Analyst Accepted Manuscript 19 of single base mismatches (Fig. 10). Although the method has several limitations in comparison 20 21 with qRT-PCR, such as labeling or gene amplification, other analytical performance parameters 22 like response time or real-time analysis were reported to permit the integration of the 3D gold 23 24 strips nanostrutrctures in point of care diagnostic devices. 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Fig. 10. Sequence specificity of the LSPR sensor for miRNA. (a) LSPR spectra of the pattern 46 47 after incubation in the absence (green) or presence of miR-let-7c (blue), miR-let-7f (gray), or 48 miR-let-7a (red, 25 nM each). (b) Probe sequence for miR-let-7a detection and sequences of miR- 49 50 let-7a, miR-let-7c, and miR-let-7f. Mismatch positions in miRNA sequences are in red. LNA 51 positions are underlined and in bold. (c) miR-200a-3p detection in total RNA extracts from human 52 53 stomach adenocarcinoma cells. (d) LSPR peak shift in total RNA from human primary gastric 54 55 cancer cells expressing (MKN45) and not expressing (SNU1) miR-200a-3p and quantification of 56 miR-200a-3p in each cell line by qRT-PCR. Reprinted with permission from (Biosensors and 57 58 Bioelectronics, 2018, 113, 39–45). Copyright (2018) Elsevier. 59 60 Page 27 of 47 Analyst Page 28 of 47 1 2 3 DNA methylation View Article Online DOI: 10.1039/C9AN00701F 4 5 Among the epigenetic alterations associated with the origin of cancer, DNA methylation is one 6 of the most important mechanisms that regulate the genome imprinting while maintaining the 7 8 stability to prevent mutations. DNA methylation comprises the addition of a methyl group at the 9 fifth carbon of the cytosine ring (5-methylcytosine, 5mC) mediated by DNA methyltransferase 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 enzymes. Since the methylation pattern is closely related to cell differentiation and normal gene 12 13 expression, DNA methylation plays an important role as a promising biomarker for cancer 14 prevention and prognosis. 15 16 17 In recent years, several plasmonic-based methodologies have been successfully applied to the 18 Analyst Accepted Manuscript 19 sensitive and reliable quantification of DNA methylation. For instance, Huertas et al.97 describe 20 21 a SPR-based assay consisting of the monitoring of straightforward interactions between poly- 22 purine reverse-Hoogsten hairpin probes (PPRH) and ds-DNA fragments by triple helix formation. 23 24 The method proved to be effective for the specific methyl-cytosine quantification and the specific 25 capture of ds-DNA fragments in comparison with the standard duplex-forming probes. The label- 26 27 free method skips previous DNA strand de-hybridization or PCR amplification. In particular, the 28 analysis of DNA-methylation levels of the promoter region of PAX-5 gene showed a good 29 30 correlation between the sequences with a low signal variation (≤8% CV) for 35 31 32 hybridization/regeneration cycles. Additionally, no cross-reactivity results were reported by using 33 non-specific antibody indicating its accuracy and precision for screening of other DNA 34 35 modifications such as 5-hydroxymethylcytosines, 5-formylcytosine and 5-carboxylcytosine. 36 37 38 The analysis of methyltransferase activity plays an important role in cellular processes since it is 39 40 involved not only in the status of DNA methylation but also in genome regulation and stability, 41 thus making it a valuable biomarker with regard to cancer origin and progression. 42 43 A singular approach involving both the quantitative analysis of DNA methylation and the 44 detection of adenine methylation (Dam) methyltransferase (MTase) activity is presented by Li et 45 46 al.98 A SPR biosensor makes use of gold nanorods end-to-end assemblies, which functioned as 47 signal-enhancing scaffolds to amplify the SPR response. The assay comprises the monitoring of 48 49 the methylation process by injecting the methylation reaction solution in the first place into the 50 51 fluidic channels and subsequently the end-to-end AuNR assemblies. The shift in the SPR angle 52 after the hybridization process is enhanced by the introduction of the end-to-end AuNR 53 54 assemblies showing a good correlation between the SPR angle shift and the Dam MTase 55 concentration, with a detection limit of 0.2 U/mL (U: units of activity typically used to describe 56 57 enzyme catalytic activity). The main advantage of the method comprises the capacity to screen 58 inhibitors for Dam MTase with broad antimicrobial action, thus becoming a promising target for 59 60 antimicrobial drug development. Page 28 of 47 Page 29 of 47 Analyst 1 2 3 View Article Online DOI: 10.1039/C9AN00701F 4 5 Another SPR-based assay for DNA methyltransferase activity and the screening of its inhibitors 6 is described by Xia et al.99 The method was capable of monitoring the interaction between p53 7 8 protein and the methylation of double-stranded (ds)-DNA consensus sites consisting of a specific 9 sequence of 5′-CCGG-3′ in which the second C base can be methylated by methyltransferase. A 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 limit of detection of 0.09 U/mL for the methyltransferase activity was obtained due to the strong 12 13 binding between the methylated ds-DNA and p53 protein. The screening of the methyltransferase 14 inhibition demonstrated that nucleoside inhibitors are more effective than non-nucleoside 15 16 inhibitors. The potential of the method for exploring DNA methylation related diseases was 17 proved in either normal or cancer cell lysates. 18 Analyst Accepted Manuscript 19 20 21 4.2.3 Cells 22 Circulating tumor cells (CTCs) are cells within the blood stream that are released from the tumors 23 24 into the blood. CTCs are scarcely present in blood since only one CTC per billion normal blood 25 cells is circulating in patients with advanced cancer. Their presence in blood is associated with 26 27 epithelial cancers and implies poor prognosis for breast, colorectal, prostate and lung cancers. 28 Thus, characterization of CTCs by measuring either morphological properties or surface antigen 29 30 expression is essential for monitoring the progression of both localized and metastatic cancer 31 32 lesions. Until now, the use of plasmonic biosensors mainly rely on either the identification of 33 cancer cell lines from the specific patterns generated by SPR signals100 or the monitoring of cell 34 35 viability through morphological changes101. An interesting approach is described by Hong et al. 36 for the recognition of the proteolytic activity of membrane type 1 matrix metalloproteinase (MT1- 37 38 MMP), a zinc proteinase extracted from invasive cancer cells. They make use of a LSPR biosensor 39 40 with PEG-coated gold nanorods immobilized onto the gold sensor surface to detect the proteolytic 41 activity of MT1-MMP. Additionally, the biosensor was capable of discriminating the proteolytic 42 43 activity from whole cell lysate of two different live cancer cell lines (HT1080 and MCF7 cells). 44 The assay sensitivity for the proteolytic activity in both MT1-MMP enzymes and cell lysates were 45 46 measured by determining the maximum wavelength blue-shift changes of the LSPR spectra with 47 regard to the response time and the MT1-MMP concentration. The kinetics proteolytic response 48 49 for HT1080 cells showed higher proteolytic kinetic constant values (11.58 min-1) than for MCF7 50 51 (6.10 min-1) thus indicating higher proteolytic activity and overexpression of MT1-MMP 52 proteinases. 53 54 Other interesting approach reports the monitoring of intracellular signaling events as variations 55 of the angle of resonance of a SPR biosensor after activation of the epidermal growth factor 56 57 (EGFR) from Chinese hamster ovary (CHO) cells in response to EGF induction100. Results 58 showed different patterns in SPR signals when comparing carcinoma cell lines to non-tumorigenic 59 60 normal cell lines. A triphasic change of the angle of resonance was observed as induction of EGF Page 29 of 47 Analyst Page 30 of 47 1 2 3 in cancer lines whereas the activation of other receptors (as IgE, adenosine A3) produced eitherView a Article Online DOI: 10.1039/C9AN00701F 4 5 monophasic change or a minimal variation of the SPR signal. These findings provide a 6 complementary method to conventional histological techniques for the diagnosis of cancer. 7 8 Similarly, the phosphorylation of the cell surface signaling receptor VEGFR2 in intact 9 hepatocellular carcinoma HepG2 or Huh7 cells after VEGF stimulation was reported by another 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 SPR biosensor approach103. Two distinct immobilization strategies consisting of either living cells 12 13 directly cultured on the sensor chip or VEGF molecules covalently attached on the sensor surface 14 were examined. The latter strategy proved to be the most reliable method for monitoring VEGFR– 15 16 VEGF interactions in terms of stability and reproducibility while distinguishable SPR 17 sensorgrams were obtained for each cell line depending on the time required for VEGFR2 18 Analyst Accepted Manuscript 19 activation. The reusability of the sensor surface over more than 60 complete regeneration cycles 20 21 demonstrated significant benefits over conventional analytical methods thus indicating the 22 capacity of the sensor surface to recover the initial assay conditions without alteration of VEGF 23 24 active binding sites. 25 26 27 Liu et al. reported the effect of a targeted drug (bevacizumab, anti-VEGF monoclonal antibody) 28 on the binding between VEGFR and VEGF in live SKOV-3 carcinoma cells104. A customized 29 30 polymer flow gasket was fabricated to attach living cells on the ceiling of the flow chamber, 31 32 whereas VEGFR was immobilized on the SPR sensor chip. After stimulating the induction of 33 VEGF from cells, bevacizumab was added to the flow chamber and the regulation effect against 34 35 the VEGF-VEGFR angiogenic switch was monitored. Results showed good sensitivity and linear 36 range for VEGF and bevacizumab interactions and the blockage of VEGF-VEGFR binding after 37 38 bevacizumab addition. 39 40 41 4.3 Inflammation biomarkers 42 43 Inflammatory disorders comprehend a heterogeneous group of diseases that involve acute or 44 45 chronic inflammation such as autoimmune diseases, allergies, hepatitis, rheumatoid arthritis, 46 47 coeliac disease or transplant rejection. Monitoring of the inflammation process requires the 48 analysis of biomarkers to allow early diagnosis and control of the illness progression while 49 50 avoiding the relapse of the disease. Among inflammation biomarkers, including from cellular to 51 molecular factors such as serum amyloid A and procalcitonin, plasmonic biosensors have mainly 52 53 focused on cytokines and C reactive protein (CRP) detection. 54 55 4.3.1 Cytokines 56 57 Cytokines are soluble proteins of low molecular weight involved in cell communication as 58 59 immunomodulatory agents of autocrine, paracrine and endocrine cell secretions. Additionally, 60 Page 30 of 47 Page 31 of 47 Analyst 1 2 3 cytokines play a remarkable role in inflammatory reactions, host defense tissue damage repair, View Article Online DOI: 10.1039/C9AN00701F 4 5 apoptosis and cancer progression. They are commonly classified as pro-inflammatory (IL-&, 6 TNF-) responsible of the inflammatory response and anti-inflammatory cytokines (TGF-β, and 7 8 IL-4) which act as mediators of the immunological equilibrium. Monitoring the cytokine levels 9 10 may provide relevant information to improve our understanding of the immune response and Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 disease progression. Cytokines quantification primarily rely on immunoassay detection. 12 13 Multiplex cytokine analysis based on either LSPR or SPR imaging platforms appear to be the 14 15 most singular approaches for cytokine detection by plasmonic biosensors. Hendriks et al. 16 17 described an SPRi immunoassay relying on a reaction cascade which involved an antibody 18 sandwich assay with subsequent addition of neutravidin and gold biotinylated gold nanoparticle Analyst Accepted Manuscript 19 20 signal enhancement105-. Measurements were attained for four cytokines, IL-1β, IL-6, IFN-γ, and 21 TNF-α, in spiked synovial fluid and compared with buffer. The assay reported significantly high 22 23 sensitivity (fg mL-1) with an improvement of sensitivity over 40000 times and low variability 24 when using the signal enhancement cascade. Differences in sensitivity and cross-reactivity values 25 26 between cytokines were related to the affinity of the antibody pairs. The lack of validation in real 27 28 human samples and signal fluctuations are limitations to be addressed in a potential POC platform. 29 Chen et al106 presented another multiplexing approach based on a LSPR biosensor microarray 30 31 including eight parallel microfluidic channels fabricated by one-step microfluidic patterning and 32 480 nanoplasmonic sensing spots . The biosensor took advantage of antibody- conjugated gold 33 34 nanorods (AuNRs) while scanning the scattering light intensity across the microarrays of AuNR 35 36 conjugates with dark-field imaging optics. The assay achieved a sensitivity in the pg mL-1 range 37 for six cytokines: interleukin-2 (IL-2), interleukin-4 (IL-4), interleukin- 6 (IL-6), interleukin-10 38 39 (IL-10), interferon-gamma (IFN-γ), and tumor-necrosis-factor alpha (TNF-) in serum samples. 40 An assay cycle was completed in 40 minutes by utilizing the entire 480 LSPR biosensor array for 41 42 multianalyte detection along with ten repeated cycles including loading, incubation and washing 43 44 of eight different samples. As a proof of concept, the LSPR biosensor was used to monitor the 45 cytokine levels in serum during the immediate post-operative period of two neonates undergoing 46 47 congenital heart surgery. They obtained similar results to previous studies, indicating that 48 increased cytokine levels return to pre-operative concentrations within 48 hours of surgery. A 49 50 particular strategy combining a LSPR biosensor integrated with microelectrodes pursued the 51 enhancement of biosensing performance by applying an AC voltage to microelectrodes while 52 53 scanning the scattering light intensity variation of the gold nanorods. The quantification of the IL- 54 55 1β cytokine involved the functionalization of the gold nanorods followed by the monitoring of 56 the scattering intensity change every five seconds after applying an AC bias of 180 phase 57 58 difference for around 15 min107. The limit of detection of the assay was 158.5 fg mL-1 (9.1 fM) 59 for spiked samples in PBS and 1 pg mL-1 (58 fM) for diluted human serum (see Figure 11). 60 Page 31 of 47 Analyst Page 32 of 47 1 2 3 View Article Online DOI: 10.1039/C9AN00701F 4 5 6 7 8 9 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 13 14 15 16 17 18 Analyst Accepted Manuscript 19 20 21 22 23 24 Fig. 11 Schematic and principle of AC electroosmosis-enhanced localized surface plasmon 25 26 resonance (ACE-LSPR) biofunctional nanoparticle imaging. Pt microelectrodes were first 27 patterned on a glass substrate by photolithography and metal lift-off. AuNRs were then deposited 28 29 to form line-shaped sensor pattern between the microelectrodes using a microfluidic patterning 30 31 technique. Reprinted with permission from (Nano Letters, 2017, 17, 2374-2380.). Copyright 32 (2017) American Chemical Society. 33 34 35 36 37 38 4.3.2 CRP 39 40 41 42 C-reactive protein (CRP) is an annular blood plasma protein of liver origin with pentameric form 43 due to five identical non-glycosylated polypeptide sub-units. CRP is associated with infectious, 44 45 inflammatory conditions and other disorders. Its level increases substantially from pM to mM 46 during an acute-phase response of inflammation. The need of continuous monitoring of CRP in 47 48 clinical settings mostly rely on immunonephelometric, immunoturbidimetric and enzyme-linked 49 immunosorbent assays (ELISA). Plasmonic biosensing of CRP has been demonstrated for over a 50 51 decade. The most common approaches have focused on immunoassay applications based on 52 53 either non-oriented or site-directed immobilization formats with limit of detection in the µg mL-1 54 level. 55 56 An SPR immunoassay comprised of protein A/G oriented immobilization of an anti-CRP 57 antibody followed by direct CRP detection was described by Vashist et al108. The method 58 59 achieved good sensitivity with a limit of detection of 1.2 ngmL-1 in whole blood and serum. 60 Page 32 of 47 Page 33 of 47 Analyst 1 2 3 Determination of CRP was also performed in EDTA plasma samples of patients showing good View Article Online DOI: 10.1039/C9AN00701F 4 5 correlation with the clinically-accredited Roche COBAS® 8000 modular analyzer-based 6 immunoassay and conventional sandwich ELISA-based IA (human CRP Duoset ELISA kit). 7 8 Additionally, CRP detection through aptamer-based strategies has been developed in recent years. 9 Vance et al. reported an SPR imaging-oriented immobilization assay relying on the use of biotin- 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 labeled aptamers109. The detection method took advantage of signal amplification by coating the 12 13 aptamers with quantum dots emitting in near infrared. A limit of detection of 5 fg mL-1 was 14 obtained in diluted (spiked) human serum. 15 16 Another biotinylated-aptamer SPR-based approach was employed by Wu et al. for CRP detection 17 by using anti-CRP gold nanoparticles conjugates enhancement. Determination of CRP was 18 Analyst Accepted Manuscript 19 attained by the aptamer-antibody sandwich assay in the pM or nM range for 110 times diluted 20 21 human serum, while the specificity of the assay was tested against human immunoglobulin G 22 (IgG), human serum albumin (HSA), hemoglobin (Hb), transferrin (TRF) and myoglobin 23 24 (Myo)110. 25 Alternatively, an immobilization method based on plasma parylene N film was used for the 26 27 functionalization of the sensor biochip of a SPR biosensor. The method consisted of anti-CRP 28 immobilization onto the parylene film with a previous introduction of peroxide, hydroxyl and 29 30 carboxylic acids. Results showed that sensitivity levels needed to be enhanced since a linear 31 32 detection range between 1 ng/mL and 1 g mL-1 was obtained.111 33 Finally, a bioactive surface consisting of a calixarene-based antibody immobilization strategy 34 35 (ProLinker™ B) was evaluated by Soler et al. to directly detect CRP from urine and undiluted 36 human serum samples.112 Results were obtained for SPR and LSPR (based on nanodiscs 37 38 structures) biosensors -schemes by using a laboratory nanoplasmonic biosensor device capable of 39 40 working in both SPR and LSPR configurations. The comparison between both biosensing 41 approaches showed a two-fold improvement in sensitivity for LSPR detection on gold nanodiscs 42 43 (LOD(SPR) = 30.8 ng mL-1 and LOD(LSPR) = 16.2 ng mL-1). Additional advantages of this 44 biofunctionalization methodology on nanoplasmonic surfaces were the reduction of nonspecific 45 46 adsorptions and the achievement of higher sensitivity while assuring the reproducibility and 47 stability of the bioactive surface. 48 49 50 51 52 4.4 Cardiac biomarkers 53 54 55 Troponin forms a complex of three regulatory proteins (C, I and T) associated with muscle 56 57 contraction in skeletal and cardiac muscle. Among them, troponins I and T are considered the 58 target biomarkers for the diagnosis of acute myocardial infarction (AMI) since they are rapidly 59 60 released into the bloodstream after any event associated with myocardial damage. In addition, Page 33 of 47 Analyst Page 34 of 47 1 2 3 troponins are more sensitive and specific that other cardiac biomarkers,as creatinine DOI: (CK-MB) View Article Online 10.1039/C9AN00701F 4 5 and or myoglobin (Mb). While the prominence of the latter ones is not supported by the number 6 of plasmonic-based applications the presence of the former one in recent literature is increasing 7 8 every year. The monitoring of increased troponin levels in blood is essential for improving the 9 prognosis of patients while reducing deaths from heart attacks. Therefore, the rapid and real-time 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 detection conferred by plasmonic biosensors has triggered the number of troponin applications in 12 13 recent years. Most of the strategies rely on immunoassays using either SPR or LSPR biosensor 14 platforms. 15 16 SPR detection methods commonly involve gold nanoparticles secondary signal enhancement. 17 Pawula et al. describe Troponin T(tnT) detection in serum samples by using a SPR biosensor.105 18 Analyst Accepted Manuscript 19 The immobilization procedure included conventional carbodiimide coupling over a thiol 20 21 monolayer followed by the antibody immobilization113. Determination of Troponin T was attained 22 by using either a direct or a sandwich assay with gold nanoparticles anti-tnT antibody conjugates. 23 24 The optimized assay obtained a limit of detection of 0.5 ng mL-1 for 50 % diluted serum samples 25 (Fig. 12). The assay could be further improved by chemical modifications to achieve enhanced 26 27 sensitivity in the pM levels as required for clinical monitoring. 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 Page 34 of 47 Page 35 of 47 Analyst 1 2 3 View Article Online DOI: 10.1039/C9AN00701F 4 5 6 7 8 9 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 12 13 14 15 16 17 18 Analyst Accepted Manuscript 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 Fig. 12. Sensorgram of binding assays using AuNP modified detector antibodies on active (a) and 42 control antibody surfaces (b). Calibration plot for the measurement of TnT in human serum using 43 44 detector antibody conjugated AuNPs for amplification of the final binding response(c) (0.5 ng 45 mL-1 of antigen was injected on the active sensor surface). Reprinted with permission from 46 47 (Talanta, 2016, 146, 823–830). Copyright (2016) Elsevier 48 49 50 Another SPR biosensor reported significant advancements by developing a novel troponin I 51 52 detection strategy that involved both signal amplification by gold nanoparticles surface 53 modification and the generation of a magnetic field via multi-walled carbon nanotubes114. The 54 55 assay made use of a sandwich immunoassay format comprising the oriented immobilization of a 56 capture antibody into a gold nanoparticle-polydopamine (PDA) modified sensor surface. The 57 58 subsequent detection of the target analyte was carried out by conjugating the detection antibodies 59 60 to nanocomposites of polydopamine and multiwalled nanotubes. The formation of the antibody Page 35 of 47 Analyst Page 36 of 47 1 2 3 nanocomposites allowed the separation and enrichment of the target molecule while minimizing View Article Online DOI: 10.1039/C9AN00701F 4 5 diffusion limited mass transfer and matrix interference effects. The nanocomposite assay achieved 6 a 1000-fold improvement of the detection limit in comparison with barely PDA modified sensor 7 8 surfaces. The assay was not employed for testing clinical samples. 9 A SPR commercial biosensor demonstrated Troponin T detection by taking advantage of PDA 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 universal adhesive properties for immobilization of a biocompatible molecular imprinted polymer 12 13 (MIP)-based receptor. The characterization of the MIP optimal properties is reported by 14 evaluating different MIP receptors. After selecting the optimum MIP, determination of Troponin 15 16 was performed in human serum at nanomolar levels115.The improvement of sensitivity was 17 attributed to the bottom-up functionalization approach involving the assembly of molecular 18 Analyst Accepted Manuscript 19 nanostructures. 20 21 22 LSPR biosensor advances in troponin detection have allowed the determination of Troponin at 23 24 attomolar level in human biological samples (plasma, serum and urine). Liyanage et al. described 25 an immunoassay comprising the immobilization of the anti-TnT antibody on gold nanoprisms and 26 27 the subsequent determination of troponin by measuring the LSPR wavelength shift after 28 binding116. Low limit of detection at approximately 15 atM level was explained by the larger 29 30 sensing surface functionalized with gold nanoprism-antibody conjugates, the appropriate SAMs 31 32 length, and the number of receptor molecules that contributed to reduce the steric interference 33 between analytes. In spite of the excellent detection limits and reproducible results reported for 34 35 human body fluids, the biosensor was not capable of detecting TnT in cell extracts. 36 Another LSPR biosensor for TnT detection has been reported by Ashaduzzaman et al. by using 37 38 gold nanorods (GNR) and thermosensitive poly(N-isopropylacrylamide) (PNIPAAM) to form 39 40 stables anti-TnT conjugates117. The method attained an on/off-switching textural ability that 41 generated a highly sensitive reversible immune interaction between 25 and 37 °C. The reversible 42 43 behavior of the immunoassay was assigned to the induction of temperature structural changes by 44 PNIPAAM that generated differential LSPR responses in the gold nanorods while monitoring the 45 46 interaction between the anti-TnT and Tn-T. Following this singular approach, a limit of detection 47 of only 8.4 fg mL-1 with a fast response time of only 10 seconds at 25ºC was achieved. 48 49 50 51 52 53 54 55 4.5 Singular clinical applications 56 57 Insulin 58 Monitoring of insulin and glycosylated hemoglobin (HbA1c) by a SPRi biosensor microarray is 59 60 proposed as an innovative method for the control of diabetes instead of the successful Page 36 of 47 Page 37 of 47 Analyst 1 2 3 electrochemical glucose biosensor118. The dual biomarker detection format is based on theView Article Online DOI: 10.1039/C9AN00701F 4 5 functionalization of the SPR microarray spots with insulin antibody or HbA1c-antibody (Fig. 13). 6 To enhance the assay sensitivity, magnetic nanoparticles amine-functionalized with quantum dots 7 8 and subsequently the carbodiimide-activated COOH groups were covalently attached to specific 9 insulin and HbA1c aptamers. The conjugates formed were capable of capturing insulin and 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 HbA1c in whole blood diluted 20-times in buffer before being magnetically separated and 12 13 introduced to the SPR microarray spots immobilized with insulin antibody or HbA1c-antibody. 14 The limit of detection using this combined SPRi fluorescence imaging assay was 4 pM for insulin 15 16 and 1% for HbA1c, respectively, which are both within the clinical relevant range. 17 18 Analyst Accepted Manuscript 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Fig. 13 (a) SPRi microarray immunosensor developed for insulin detection in human serum, 33 and (b) SPRi responses inferring the serum matrix effect and the performance of direct vs. 34 35 sandwiched 50% serum insulin immunoassay. Reprinted with permission from (Analyst, 2018, 36 143, 1544-1555.). Copyright (2018) Royal Society of Chemistry 37 38 39 40 HIV-related DNA 41 Human immune deficiency virus (HIV)-related DNA has been detected by SPR biosensors using 42 43 entropy-driven strand displacement reactions (ESDRs) and double-layer DNA tetrahedrons 44 (DDTs)119. By using the ESDRs, target DNA could be utilized by cyclic formation of a great 45 46 number of double stranded DNA products, that after binding to the immobilized hairpin, could 47 48 combine with DDTs nanostructures. The detection limit of HIV-related DNA was 48 fM. The 49 major advancement over conventional methods relied on the reduction of the detection time to 6 50 51 minutes while avoiding enzyme labeling and complex chemical modifications. 52 53 54 Sintrom 55 A compact LSPR device based on gold nanodiscs has been used for the therapeutic drug 56 57 monitoring of Acenocoumarol (Sintrom®), an oral anticoagulant prescribed for the treatment of 58 59 a variety of thromboembolic disorders. By using a specific and reproducible label-free indirect 60 competitive assay, acenocumarol was monitored in 1:1 plasma diluted samples120. The interaction Page 37 of 47 Analyst Page 38 of 47 1 2 3 of the antibody with the custom-biofunctionalized antigen-coated surface resulted inDOI: detectionView Article Online 10.1039/C9AN00701F 4 5 levels of 0.77 nM, which are within the required clinical values for acenocumarol testing. The 6 main contribution of the method is its potential use as complementary tool in decentralized 7 8 settings as primary healthcare units. 9 Key analytical features and biomarker classification of the assays have been summarized in Table 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 1. 12 13 14 Table 1. Key analytical features and biomarker classification for detection in biological matrix, 15 16 sensitivity and linear range 17 18 Analyst Accepted Manuscript 19 20 Type of biomarker Detection format Matrix Detection Reference 21 (instrument range/LOD 22 configuration, 23 recognition element) 24 NEURODEGENERATIVE 25 DISEASES 26 Amyloid beta oligomers LSPR (PEG- gold Mimicked blood 34.9 and 26 fM 63 27 nanoparticles different (A 40 y A) 28 sizes) 29 SPR ( waveguide-coupled Assay buffer 22 to 440 pM 64 30 bimetallic) (A 42) SPR (fiber intrinsic Real blood 0.3 pM (total and 65 31 coupling) samples phosphorylated) 32 SPR (multichannel Undiluted 10 fM 66 33 platform DNA aptamers) plasma samples 34 SPR (Immmunoassay Assay buffer and 125 pM 67 multiwalled carbon artificial 35 nanotubes) cerebrospinal 36 fluid 37 Prion proteins SPR (magnetic Serum samples 0.01-1000 ng mL- 68 nanoparticles, amphiphilic 1 38 copolymer) 39 Amyloid fibrils (α-Synuclein LSPR (gold nanorods) Human brain Plasmonic 69 40 proteins) samples chirality down 41 nanomolar Prion proteins concentrations 42 CANCER BIOMARKERS 43 PROTEINS 44 SPR (PSA imprinted Spiked human 91 pg mL-1 (18 x 70 PSA 45 proteins) samples 10-14M) 46 SPR (photolithography, Whole blood 142 pg mL-1-1.42 71 47 microhole array without g mL-1 fluorescence SPR treatment 48 interrogation) 49 SPR (gold nanoparticles) Serum 0.010 to 0.40 ng 72 50 mL-1 (total PSA/ PSA free 51 fractions) 52 LSPR (gold nanodiscs) Assay buffer 0.85 ng mL-1 73 53 LSPR (gold nanodiscs UV Assay buffer 1,7 to 20.4 ng 74 54 spectra detection) Human serum mL-1 LOD 1.49 VEGF SPR (rolling circle Assay buffer 1x10-12 to 1x10-5 75 55 amplification DNA g mL-1 56 aptamers polystyrene LOD 100 pg mL-1 57 microsphere) SPR (plastic optical fiber, Assay buffer 81 ng mL-1 76 58 DNA aptamers) 59 LSPR (nanogold dot array) Human serum 0.01-100 ng mL-1 77 60 SPR (gold nanoparticles) Blood plasma 0.1 ng mL-1 78 Page 38 of 47 Page 39 of 47 Analyst 1 2 3 CEA SPR (gold nanoparticles) Spiked human 1.0 ng/mL-1 79 View Article Online 4 serum DOI: 10.1039/C9AN00701F SPR (liquid core coupling) Serum 0.5 ng mL-1 80 5 SPR field enhanced Assay buffer 0.36 ng mL-1 81 6 fluorescence (nanograting PBS 1ng mL-1 7 microarray) 8 ALCAM SPR imaging (DNA- Diluted blood 45 ng mL-1 82 directed array of plasma 9 antibodies) 10 LSPR (gold nanorods) Assay buffer 1.253 ng mL-1 83 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 Telomerase activity LSPR (Rayligh Culture medium 1.3×10-13 IU 84 spectroscopy+dark field 12 microscope) Gold 13 nanostructures 14 EXOSOMES SPR Buffer CD 63 expression 85 15 6.3 g mL-1 SPR Plasma Exosomal 86 16 expression 17 differentiation 18 between cell lines Analyst Accepted Manuscript 19 SPR custom built Identification of 87 exosomal 20 subpopulations 21 2070 22 exosomes/μL SPR (nanohole multiplex Culture medium Transmembrane 88 23 detection) PBS and intravesicular 24 proteins 10E4 25 exosome vesicles 26 SPR (Colloidal gold Serum Myeloma 89 nanoplasmonics) vesicles, 27 production from 28 patient and 29 healthy subjects nanomolar- 30 picomolar 31 32 MicroRNA SPR (coupled with DNA miRNA in total 9 pM 1x10-11M 90 33 super-sandwich assemblies RNA samples to 1x10-6 M and biotin–streptavidin 34 based amplification.) 35 SPR laboratory custom human miRNAs 2 to 30 nM 91 36 designed 37 SPR commercial design miRNA-93 in 10pM -1fM 92 total RNA lysate 38 from HEK-293 39 cells 40 LSPR (gold nanoparticles DNA probes 3 nM 93 41 + SPR scattering) LSPR (red shift gold DNA probes 5 pM (miRNA 94 42 nanocubes) 205) 43 LSPR (colorimetric UV DNA probes 1 nM 95 44 spectrum) (miRNA155) LSPR (3D nanostructure) Total RNA 13 fM: 2.6 96 45 Extract cell lines attomole in 200 46 µL 47 Single base 48 mismatch DNA methylation SPR laboratory custom DNA probes 115 pM 97 49 design 50 SPR (gold nanorods DNA probes 0.2 U mL-1 98 51 methyl transferase activity) 52 SPR (methyl transferase Cell lysates 3.04 nM 99 53 activity) 54 Cells LSPR (PEG-coated Whole Cell Metalloproteinase 102 55 nanorods) lysates proteolytic activity of cancer 56 cell lines 57 SPR Culture medium Signalling event 100 58 assay buffer Change in the angle of 59 resonance 60 Page 39 of 47 Analyst Page 40 of 47 1 2 3 SPR Culture medium Change in the 103 View Article Online 4 Assay buffer angle of DOI: 10.1039/C9AN00701F resonance 5 depending on the 6 time required for 7 activation. SPR Culture medium VEGF binding 104 8 (polydimethylsiloxane assay buffer and drug 9 flow chamber gasket) blocking-live cell 10 lines interaction, Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 variation of baseline 12 13 INFLAMMATION 14 BIOMARKERS 15 Cytokines 16 IL-1β, IL-6, IFN-γ, and SPRimaging Spiked synovial 50 fg mL-1 105 17 TNF- (immunoassay) fluid 18 interleukin-2 (IL-2), LSPR (multiplex Serum samples 11.43 (TNF-R), 106 Analyst Accepted Manuscript 19 immunoassay arrays 480 6.46 (IFN-γ), interleukin-4 (IL-4), nanoplasmonic spots 20.56 (IL-2), 20 interleukin- 6 (IL-6), antibody conjugated gold 4.60 (IL-4), 11.29 21 interleukin-10 (IL-10), nanorods) (IL-6), and 10.97 22 interferon-gamma (IFN-γ), (IL-10) pg mL-1 23 and tumor-necrosis-factor 24 alpha (TNF-R) 25 IL-1β cytokine LSPR (AC voltage Assay buffer 158.5 fg mL-1 107 microelectrodes) Diluted human PBS; 1pg/mL 26 serum human serum 27 CRP SPR (immunoassay Plasma samples 1.2 ng mL-1 108 28 protein A/G oriented) 29 SPR imaging (oriented Spiked human 5 fg mL-1 109 immobilization: biotin serum 30 labelled aptamer coated NI 31 quantum dots) 32 LSPR (biotin labelled Diluted human 10 pM-100 nM 110 antibodies, anti-CRP gold serum range 33 nanoparticles) 34 SPR (plasma parylene Assay buffer 1 ngmL-1 and 1 111 35 film) gmL-1 36 SPR/LSPR (gold Pure urine LOD(SPR) = 112 nanodiscs, calixarene Diluted plasma 30.8 ng mL-1 37 immobilization) LOD(LSPR) = 38 16.2 ng mL-1 39 CARDIAC 40 BIOMARKERS 41 Troponin SPR (immunoassay) 50% Diluted 0,5 ng mL-1 113 serum 42 SPR (gold nanoparticles Spiked human 1.25 ng mL-1 114 43 magnetic field multiwalled serum 44 carbon nanotubes nanocomposite) 45 SPR (molecular imprinted Human serum 15.4 nM 115 46 polymers) 47 LSPR (gold nanoprisms Plasma, serum 15 attomolar 116 48 antibody conjugates) and urine LSPR (gold nanorods Assay buffer 8.4 fg mL-1 117 49 reversible (PbS,MES) 50 immunoreaction) 51 OTHER CLINICAL 52 APPLICATIONS 53 Insulin SPRimaging (microarray Diluted whole 4 pM 118 54 magnetic nanoparticles, blood quantum dots aptamers) 55 HIV-related DNA SPR (entropy-driven Assay buffer 48 fM 119 56 strand displacement 57 reactionsand double-layer DNA tetrahedrons). 58 LSPR (goldnanodiscs) 1:1 diluted 0.77 nM 120 Sintrom 59 plasma 60 Page 40 of 47 Page 41 of 47 Analyst 1 2 3 View Article Online DOI: 10.1039/C9AN00701F 4 5 6 7 8 Conclusions and future perspectives 9 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 Although plasmonic biosensors have exploited clinical applications for more than a decade, recent 12 13 progress in nanotechnology has enabled their consolidation in a new and more advanced stage. 14 The arrival of the new era of nanoplasmonics relies on several paramount aspects: (i) the 15 16 utilization of complementary analytical performances using detection formats unable to be 17 combined in the past (i.e. colorimetric, electrochemiluminescence, SERS or fluorescence 18 Analyst Accepted Manuscript 19 imaging); (ii) the development of novel chemical functionalization strategies by employing top- 20 21 down (lithographic, patterning, etc.) and bottom-up (nanoparticles and their self-assemblies, etc.) 22 approaches and (iii) the selection and appropriate integration of the bioreceptors improving sensor 23 24 specificity while reducing cross-reactivity and thereby eliminating false outcome. Such 25 advancements have permitted to overcome several hurdles including poor sensitivity, limited 26 27 multiplexed analysis and interference from complex biological matrices for clinical applications. 28 29 30 This review covers mainly analytical improvements for the determination of several types of 31 32 biomarkers concerning neurodegenerative, inflammation, cancer and cardiac-related disorders by 33 describing different plasmonics biosensor methodologies. Firstly, approaches involving 34 35 modifications of conventional Surface Plasmon based-arrangements such as well-known LSPR, 36 SPR imaging and SPR fiber optic have also contributed to achieve ultrasensitive plasmonic 37 38 applications. Additionally, the use of distinct biosensor approaches with novel configuration 39 40 strategies involving the combination of plasmonics with electrochemical, fluorescence or 41 colorimetric detection has enabled the enhancement of the biosensor response while improving 42 43 the assay sensitivity. The number of combined applications is represented in almost every type of 44 biomarkers comprising from PSA and CEA cancer proteins detection to microRNA analysis. 45 46 47 Secondly, the design of the plasmonic sensor chip using top–down (metal nanoparticles) and 48 49 bottom up (lithographic) procedures determines the stability of the biosensor device for long-term 50 51 use. The fabrication of nanopatterned structures by microcontact printing, nanoimprinting and 52 polymer or electron beam lithography provides consistent nanoarchitectures that improve the 53 54 capacity of the immobilized component for interacting with the target analyte. Imprinting 55 technology has been successfully applied to the detection of PSA, troponin or cytokine 56 57 biomarkers. Similarly, the functionalization of nanomaterials by assembling nanocomposites and 58 tagging nanoparticles for enhancing the signal response has been extensively considered in 59 60 plasmonic sensing for clinical diagnosis. In particular, LSPR arrangement makes intended use of Page 41 of 47 Analyst Page 42 of 47 1 2 3 the chemical activation of different types of nanoparticles from nanorods to nanoprisms for the View Article Online DOI: 10.1039/C9AN00701F 4 5 detection of a broad range of biomarkers from CRP protein to living circulating tumor cells or 6 telomerase activity. 7 8 9 Finally, the feasibility of any biosensor depends on the appropriate immobilization of the 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 biorecognition element. The bioreceptors that recognize biomarkers include antibodies and 12 13 aptamers to more specialized biomolecules such as cell membrane receptors, DNA probes or 14 molecular imprinted polymers. The biocompatibility is a key aspect to consider during 15 16 immobilization of the biorecognition element since the functionalization may affect its activity. 17 The biological nature should also be preserved along measurements by generating the appropriate 18 Analyst Accepted Manuscript 19 environment to facilitate the accessibility while maintaining the binding activity. For this reason, 20 21 recent biomarker detection strategies take advantage of bottom-up surface modifications to 22 effectively functionalize the sensor chips. The use of anti-fouling materials as PEG/OEG on 23 24 nanosurfaces intends both to conserve the bioactive area from non-specific binding and to avoid 25 the interference of complex biological media. By achieving these outcomes, the majority of 26 27 plasmonic biosensors could be employed for analyzing body fluids (blood, urine or cerebrospinal 28 fluid) with better detection limits than conventional techniques. Likewise, improved selectivity 29 30 has also been demonstrated by attaining, for example, the mismatch differentiation of a single 31 32 nucleic acid base pair. 33 34 35 In spite of the apparent progress, several aspects still need to be considered in order to reach 36 reliable clinical applications in the future. Miniaturization, automation and integration of 37 38 microfluidics are essential blocks to construct POC biosensors. Primarily, miniaturization seems 39 40 an achievable goal to build efficient POC platforms owing to both the availability of lithographic 41 techniques and the development of novel flexible materials, whilst software can be easily 42 43 incorporated for the design of automatable POC devices. 44 Therefore, the main focus for the fabrication of POC platforms is to set the integration of 45 46 microfluidics without the minimum support of additional equipment. Although several 47 approaches are under development, particularly involving microchannels with micropneumatic or 48 49 electrical pumping, more efforts are needed to develop digital microfluidic devices that enable 50 51 the delivery, transport and separation of a tiny droplet of sample in a single step. 52 53 54 Likewise, it is worth pointing that the majority of works have concentrated on improving the 55 analytical performance of novel and existing detection designs, whereas less attention has been 56 57 paid to the integration of truly POC devices. Thus, the development of plasmonic devices has 58 mainly focused in the improvement of sensing schemes for attaining ultrasensitive detection 59 60 limits, while the incorporation of advanced software and microfluidic designs have been barely Page 42 of 47 Page 43 of 47 Analyst 1 2 3 considered. In this sense, many plasmonics works address novel functionalization strategies either View Article Online DOI: 10.1039/C9AN00701F 4 5 to enhance the active sensing area or to prevent non-specific adhesion of proteins. This matter has 6 been of paramount importance in order to provide antinonfouling layers that succeed in the sensor 7 8 detection of biomarkers in undiluted body fluids. However, the transfer of fully-integrated devices 9 to the clinical settings has not yet been entirely overcome and the gap between the laboratory and 10 Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 the market still remains. 12 13 14 On the other hand, validation of multiplexed detection formats, specifically for cancer, 15 16 inflammation, cardiac and neurodegenerative biomarkers, is seriously needed as the first step for 17 the clinical use of plasmonic biosensors. The second step involves both commercialization and 18 Analyst Accepted Manuscript 19 integration of plasmonic biosensors in clinical settings for routine analysis. Since manufacturing 20 21 and scale-up for mass production are the major bottleneck of clinical biosensors, extensive 22 research on the assembly and component integration process will aid to overcome these 23 24 challenges in relatively near future. Improvements in micro- and nanofabrication technologies 25 will help to introduce plasmonic biosensors as biodiagnostic tools for the massive screening of 26 27 the most prominent disorders. Furthermore, the incorporation of soft materials and wireless 28 telemetry systems might enable wearable or implantable biosensors in the human body for non- 29 30 invasive health monitoring. 31 32 33 Conflict of interest 34 35 There are no conflicts of interest to declare. 36 37 Acknowledgments. 38 39 40 ICN2 is supported by the Severo Ochoa program from Spanish MINECO (Grant No. SEV-2013- 41 0295) and EPISENS project (TEC2012-34280). 42 43 Author information: 44 ORCID id: 45 EM: 0000-0002-6459-8107 46 47 PD: 0000-0003-2896-7611 48 LL: 0000-0001-5187-5358 49 50 51 52 53 References 54 55 56 1. B. Bohunicky and S. A. Mousa, Nanotechnology, science and applications, 2010, 4, 1-10. 57 2. L. Wu and X. Qu, Chemical Society reviews, 2015, 44, 2963-2997. 58 59 3. A. Qureshi, Y. Gurbuz and J. H. Niazi, Sensors and Actuators B: Chemical, 2012, 171-172, 60 62-76. Page 43 of 47 Analyst Page 44 of 47 1 2 3 4. S. T. Sanjay, G. Fu, M. Dou, F. Xu, R. Liu, H. Qi and X. Li, The Analyst, 2015, 140, 7062- View Article Online DOI: 10.1039/C9AN00701F 4 7081. 5 5. M. Swierczewska, G. Liu, S. Lee and X. Chen, Chemical Society reviews, 2012, 41, 2641- 6 7 2655. 8 6. O. Tokel, F. Inci and U. Demirci, Chemical reviews, 2014, 114, 5728-5752. 9 7. J. Min, M. Nothing, B. Coble, H. Zheng, J. Park, H. Im, G. F. Weber, C. M. Castro, F. K. 10 Swirski, R. Weissleder and H. Lee, ACS nano, 2018, 12, 3378-3384. Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 8. P. Dey, N. Fabri-Faja, O. Calvo-Lozano, R. A. Terborg, A. Belushkin, F. Yesilkoy, A. Fabrega, 12 J. C. Ruiz-Rodriguez, R. Ferrer, J. J. Gonzalez-Lopez, M. C. Estevez, H. Altug, V. Pruneri 13 and L. M. Lechuga, ACS sensors, 2019, 4, 52-60. 14 9. J. H. Guo, X. W. Huang and X. Ma, Sensors and Actuators B-Chemical, 2018, 275, 446- 15 450. 16 17 10. J. Guo and X. Ma, Biosensors & Bioelectronics, 2017, 94, 415-419. 18 11. J. Guo, Analytical Chemistry, 2017, 89, 8609-8613. Analyst Accepted Manuscript 19 20 12. C. I. L. Justino, A. C. Duarte and T. A. P. Rocha-Santos, TrAC Trends in Analytical 21 Chemistry, 2016, 85, 36-60. 22 13. C. I. L. Justino, T. A. Rocha-Santos, A. C. Duarte and T. A. Rocha-Santos, TrAC Trends in 23 Analytical Chemistry, 2010, 29, 1172-1183. 24 14. C. Chen, Q. Xie, D. Yang, H. Xiao, Y. Fu, Y. Tan and S. Yao, RSC Advances, 2013, 3, 4473. 25 26 15. A. J. Bandodkar and J. Wang, Trends in biotechnology, 2014, 32, 363-371. 27 16. S. Zhang, A. Garcia-D'Angeli, J. P. Brennan and Q. Huo, The Analyst, 2014, 139, 439-445. 28 17. J. Sun, Y. Xianyu and X. Jiang, Chemical Society reviews, 2014, 43, 6239-6253. 29 18. B. Špacˇková, P. Wrobel, M. Bocková and J. í. Homola, Proceedings of the IEEE, 2016, 30 104, 2830-2408. 31 19. A. Abbas, M. J. Linman and Q. Cheng, Biosensors & bioelectronics, 2011, 26, 1815-1824. 32 20. A. Zybin, V. Shpacovitch, J. Skolnik and R. Hergenröder, Sensors and Actuators B: 33 Chemical, 2017, 239, 338-342. 34 35 21. A. F. Coskun, A. E. Cetin, B. C. Galarreta, D. A. Alvarez, H. Altug and A. Ozcan, Scientific 36 reports, 2014, 4, 6789. 37 22. Y.-B. Shin, H. M. Kim, Y. Jung and B. H. Chung, Sensors and Actuators B: Chemical, 2010, 38 150, 1-6. 39 23. J. C. Contreras-Naranjo, H. J. Wu and V. M. Ugaz, Lab on a chip, 2017, 17, 3558-3577. 40 24. Y. T. Kim, Y. Chen, J. Y. Choi, W. J. Kim, H. M. Dae, J. Jung and T. S. Seo, Biosensors & 41 bioelectronics, 2012, 33, 88-94. 42 25. M. J. Linman, A. Abbas and Q. Cheng, The Analyst, 2010, 135, 2759-2767. 43 26. Y. K. Oh, H. A. Joung, S. Kim and M. G. Kim, Lab on a chip, 2013, 13, 768-772. 44 45 27. D. Zhang, L. Men and Q. Chen, Sensors, 2011, 11, 5360-5382. 46 28. M. C. Estevez, M. Alvarez and L. M. Lechuga, Laser & Photonics Reviews, 2012, 6, 463- 47 487. 48 29. K. Cooper, Micromachines, 2017, 8, 20. 49 30. H. J. Lee, A. W. Wark and R. M. Corn, The Analyst, 2008, 133, 975-983. 50 31. M. I. Mohammed and M. P. Desmulliez, Lab on a chip, 2011, 11, 569-595. 51 32. C. Parolo and A. Merkoci, Chemical Society reviews, 2013, 42, 450-457. 52 33. J.-F. Masson, ACS sensors, 2017, 2, 16-30. 53 54 34. C. L. Wong and M. Olivo, Plasmonics, 2014, 9, 809-824. 55 35. M. Puiu and C. Bala, Sensors, 2016, 16. 56 36. J. B. Fasoli and R. M. Corn, Langmuir : the ACS journal of surfaces and colloids, 2015, 31, 57 9527-9536. 58 37. L. Simon, G. Lautner and R. E. Gyurcsányi, Analytical Methods, 2015, 7, 6077-6082. 59 38. C. Liu, F. Hu, W. Yang, J. Xu and Y. Chen, TrAC Trends in Analytical Chemistry, 2017, 97, 60 354-362. Page 44 of 47 Page 45 of 47 Analyst 1 2 3 39. Markéta Bocková, J. Slabý, T. Špringer and J. Homola, Annual Review of DOI: Analytical View Article Online 10.1039/C9AN00701F 4 Chemistry, 2019, 12. 5 40. J. A. Jackman, A. Rahim Ferhan and N. J. Cho, Chemical Society reviews, 2017, 46, 3615- 6 7 3660. 8 41. M. C. Estevez, M. A. Otte, B. Sepulveda and L. M. Lechuga, Analytica chimica acta, 2014, 9 806, 55-73. 10 42. M. Svedendahl, S. Chen, A. Dmitriev and M. Ka¨ll, Nano letters, 2009, 9, 4428-4433. Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 43. J. N. Anker, W. P. Hall, O. Lyandres, N. C. Shah, J. Zhao and R. P. V. Duyne, Nature 12 materials, 2008, 7. 13 44. M. Li, S. K. Cushing and N. Wu, The Analyst, 2015, 140, 386-406. 14 45. D. K. Gramotnev and S. I. Bozhevolnyi, Nature Photonics, 2010, 4, 83-91. 15 46. N. J. Halas, S. Lal, W. S. Chang, S. Link and P. Nordlander, Chemical reviews, 2011, 111, 16 17 3913-3961. 18 47. H. Chen, L. Shao, Q. Li and J. Wang, Chemical Society reviews, 2013, 42, 2679-2724. Analyst Accepted Manuscript 19 48. S. Neretina, R. A. Hughes, K. D. Gilroy and M. Hajfathalian, Accounts of chemical 20 research, 2016, 49, 2243-2250. 21 49. K. C. Ng, F.-C. Lin, P.-W. Yang, Y.-C. Chuang, C.-K. Chang, A.-H. Yeh, C.-S. Kuo, C.-R. Kao, 22 C.-C. Liu, U. S. Jeng, J.-S. Huang and C.-H. Kuo, Chemistry of Materials, 2017, 30, 204- 23 213. 24 50. J. Reguera, J. Langer, D. Jimenez de Aberasturi and L. M. Liz-Marzan, Chemical Society 25 26 reviews, 2017, 46, 3866-3885. 27 51. B. Ai, H. Möhwald, D. Wang and G. Zhang, Advanced Materials Interfaces, 2017, 4, 28 1600271. 29 52. K. Chen, B. B. Rajeeva, Z. Wu, M. Rukavina, T. D. Dao, S. Ishii, M. Aono, T. Nagao and Y. 30 Zheng, ACS nano, 2015, 9, 6031–6040. 31 53. Y. Chen, Microelectronic Engineering, 2015, 135, 57-72. 32 54. A. M. Lopatynskyi, V. K. Lytvyn, V. I. Nazarenko, L. J. Guo, B. D. Lucas and V. I. Chegel, 33 Nanoscale research letters, 2015, 10, 99. 34 35 55. R. Ogier, L. Shao, M. Svedendahl and M. Kall, Advanced materials, 2016, 28, 4658-4664. 36 56. V. Peksa, P. Lebruskova, H. Sipova, J. Stepanek, J. Bok, J. Homola and M. Prochazka, 37 Physical chemistry chemical physics : PCCP, 2016, 18, 19613-19620. 38 57. M. C. Traub, W. Longsine and V. N. Truskett, Annual review of chemical and biomolecular 39 engineering, 2016, 7, 583-604. 40 58. O. Vazquez-Mena, L. Gross, S. Xie, L. G. Villanueva and J. Brugger, Microelectronic 41 Engineering, 2015, 132, 236-254. 42 59. S. S. Acimovic, H. Sipova, G. Emilsson, A. B. Dahlin, T. J. Antosiewicz and M. Kall, Light, 43 science & applications, 2017, 6, e17042. 44 45 60. M. Soler, M. C. Estevez, R. Villar-Vazquez, J. I. Casal and L. M. Lechuga, Analytica chimica 46 acta, 2016, 930, 31-38. 47 61. G. A. Lopez-Munoz, M. C. Estevez, M. Vazquez-Garcia, M. Berenguel-Alonso, J. Alonso- 48 Chamarro, A. Homs-Corbera and L. M. Lechuga, Journal of biophotonics, 2018, 11, 49 e201800043. 50 62. D. Zhang, Y. Lu, J. Jiang, Q. Zhang, Y. Yao, P. Wang, B. Chen, Q. Cheng, G. L. Liu and Q. 51 Liu, Biosensors & bioelectronics, 2015, 67, 237-242. 52 63. H. Kim, J. U. Lee, S. Song, S. Kim and S. J. Sim, Biosensors & Bioelectronics, 2018, 101, 96- 53 54 102. 55 64. Y. K. Lee, K. S. Lee, W. M. Kim and Y. S. Sohn, PLoS One, 2014, 9, e98992. 56 65. N. Truong Thi Vu, T. Nhu Hoa Thi, E. Nam, N. Tan Tai, W. J. Yoon, S. Cho, J. Kim, K.-A. 57 Chang and H. Ju, Rsc Advances, 2018, 8, 7855-7862. 58 66. S. Kim, A. W. Wark and H. J. Lee, Analytical Chemistry, 2016, 88, 7793-7799. 59 67. S. Lisi, S. Scarano, S. Fedeli, E. Pascale, S. Cicchi, C. Ravelet, E. Peyrin and M. Minunni, 60 Biosensors & Bioelectronics, 2017, 93, 289-292. Page 45 of 47 Analyst Page 46 of 47 1 2 3 68. Z. Lou, H. Han, D. Mao, Y. Jiang and J. Song, Nanomaterials, 2018, 8. View Article Online DOI: 10.1039/C9AN00701F 4 69. J. Kumar, H. Erana, E. Lopez-Martinez, N. Claes, V. F. Martin, D. M. Solis, S. Bals, A. L. 5 Cortajarena, J. Castilla and L. M. Liz-Marzan, Proceedings of the National Academy of 6 7 Sciences of the United States of America, 2018, 115, 3225-3230. 8 70. G. Erturk, H. Ozen, M. A. Tumer, B. Mattiasson and A. Denizli, Sensors and Actuators B- 9 Chemical, 2016, 224, 823-832. 10 71. J. Breault-Turcot, H.-P. Poirier-Richard, M. Couture, D. Pelechacz and J.-F. Masson, Lab Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 Chip, 2015, 15, 4433-4440. 12 72. Z. X. Jiang, Y. Qin, Z. Peng, S. H. Chen, S. Chen, C. Y. Deng and J. Xiang, Biosensors & 13 Bioelectronics, 2014, 62, 268-273. 14 73. M. Sanders, Y. B. Lin, J. J. Wei, T. Bono and R. G. Lindquist, Biosensors & Bioelectronics, 15 2014, 61, 95-101. 16 17 74. Y. Khan, A. R. Li, L. Chang, L. D. Li and L. Guo, Sensors and Actuators B-Chemical, 2018, 18 255, 1298-1307. Analyst Accepted Manuscript 19 75. H. X. Chen, Y. F. Hou, F. J. Qi, J. J. Zhang, K. Koh, Z. M. Shen and G. X. Li, Biosensors & 20 Bioelectronics, 2014, 61, 83-87. 21 76. N. Cennamo, M. Pesavento, L. Lunelli, L. Vanzetti, C. Pederzolli, L. Zeni and L. 22 Pasquardini, Talanta, 2015, 140, 88-95. 23 24 77. H. W. Yang, R. Y. Tsai, J. P. Chen, S. P. Ju, J. F. Liao, K. C. Wei, W. L. Zou and M. Y. Hua, 25 26 ACS Appl. Mater. Interfaces, 2016, 8, 30845–30852. 27 78. T. Springer and J. Homola, Analytical and Bioanalytical Chemistry, 2012, 404, 2869-2875. 28 79. R. Li, F. Feng, Z. Z. Chen, Y. F. Bai, F. F. Guo, F. Y. Wu and G. Zhou, Talanta, 2015, 140, 29 143-149. 30 80. J. Xu and Y. Chen, Talanta, 2018, 184, 468-474. 31 81. C. Liu, F. Meng, B. Wang, L. Zhang and X. Cui, Analytical Methods, 2018, 10, 145-150. 32 82. M. Piliarik, L. Parova, H. Vaisocherova and J. Homola, in Optical Sensors 2009, eds. F. 33 Baldini, J. Homola and R. A. Lieberman, 2009, vol. 7356. 34 35 83. J. H. Pai, C. T. Yang, H. Y. Hsu, A. B. Wedding and B. Thierry, Analytica Chimica Acta, 36 2017, 974, 87-92. 37 84. K. Wang, S. Li, Y. Liu, L. Jiang, F. Zhang, Y. Wei, Y. Zhang, Z. Qi, K. Wang and S. Liu, 38 Analytical Chemistry, 2017, 89, 7262-7268. 39 85. D. L. M. Rupert, G. V. Shelke, G. Emilsson, V. Claudio, S. Block, C. Lasser, A. Dahlin, J. O. 40 Lotvall, M. Bally, V. P. Zhdanov and F. Hook, Analytical Chemistry, 2016, 88, 9980-9988. 41 86. L. Grasso, R. Wyss, L. Weidenauer, A. Thampi, D. Demurtas, M. Prudent, N. Lion and H. 42 Vogel, Analytical and Bioanalytical Chemistry, 2015, 407, 5425-5432. 43 87. A. A. Ibn Sina, R. Vaidyanathan, S. Dey, L. G. Carrascosa, M. J. A. Shiddiky and M. Trau, 44 45 Scientific Reports, 2016, 6. 46 88. J. Park, H. Im, S. Hong, C. M. Castro, R. Weissleder and H. Lee, Acs Photonics, 2018, 5, 47 487-494. 48 89. G. Di Noto, A. Bugatti, A. Zendrini, E. L. Mazzoldi, A. Montanelli, L. Caimi, M. Rusnati, D. 49 Ricotta and P. Bergese, Biosensors & Bioelectronics, 2016, 77, 518-524. 50 90. X. J. Ding, Y. R. Yan, S. Q. Li, Y. Zhang, W. Cheng, Q. Cheng and S. J. Ding, Analytica 51 Chimica Acta, 2015, 874, 59-65. 52 91. A. Aviñó, C. S. Huertas, L. M. Lechuga and R. Eritja, Anal. Bioanal. Chem., 2016, 408, 885– 53 54 893 55 56 92. S. Schmieder, J. Weisspflog, N. Danz, M. Hubner, S. Kreth, U. Klotzbach and F. Sonntag, 57 Engineering in Life Sciences, 2017, 17, 1264-1270. 58 93. Y. L. Hu, L. Zhang, Y. Zhang, B. Wang, Y. W. Wang, Q. L. Fan, W. Huang and L. H. Wang, 59 Acs Applied Materials & Interfaces, 2015, 7, 2459-2466. 60 Page 46 of 47 Page 47 of 47 Analyst 1 2 3 94. L. Zhang, J. Wang, J. Zhang, Y. Liu, L. Wu, J. Shen, Y. Zhang, Y. Hu, Q. Fan, W. Huang and View Article Online DOI: 10.1039/C9AN00701F 4 L. Wang, Acs Sensors, 2017, 2, 1435-1440. 5 95. A. Esmaeili-bandboni, S. M. Amini, R. Faridi-majidi, J. Bagheri, J. Mohammadnejad and 6 7 E. Sadroddiny, Iet Nanobiotechnology, 2018, 12, 453-458. 8 96. H.-K. Na, J.-S. Wi, H. Y. Son, J. G. Ok, Y.-M. Huh and T. G. Lee, Biosensors & Bioelectronics, 9 2018, 113, 39-45. 10 97. C. S. Huertas, A. Aviñó, C. Kurachi, A. Piqué, J. Sandoval, R. Eritja, M. Esteller and L. M. Published on 01 October 2019. Downloaded by Centro de Investigacion y Desarrollo (CID) on 10/2/2019 10:58:32 AM. 11 Lechuga, Biosens. Bioelectron., Analytical and Bioanalytical Chemistry, 2018, 120, 47-54. 12 98. X. Li, T. Song and X. Guo, Analyst, 2015, 140, 6230-6233 13 99. Y. Xia, L. Wu, Y. Hu, Y. He, Z. Cao, X. Zhu, X. Yi and J. Wang, Biosensors & Bioelectronics, 14 2019, 126, 269-274. 15 16 17 100. T. Hiragun, Y. Yanase, K. Kose, T. Kawaguchi, K. Uchida, S. Tanaka and M. Hide, Biosensors 18 & Bioelectronics, 2012, 32, 202-207. Analyst Accepted Manuscript 19 101. A. R. Ferhan, J. A. Jackman, J. H. Park, N.-J. Cho and D.-H. Kim, Advanced Drug Delivery 20 Reviews, 2018, 125, 48-77. 21 102. Y. Hong, M. Ku, E. Lee, J. S. Suh, Y. M. Huh, D. S. Yoon and J. Yang, Journal of Biomedical 22 Optics, 2014, 19. 23 103. E. Mauriz, S. Carbajo-Pescador, R. Ordoñez, M. C. García-Fernández, J. L. Lechuga, L. M. 24 Lechuga and J. González-Gallego, Analyst, 2014, 139, 1426–1435 25 26 104. C. Liu, S. Alwarappan, H. A. Badr, R. Zhang, H. Liu, J.-J. Zhu and C.-Z. Li, Analytical 27 Chemistry, 2014, 86, 7305-7310. 28 105. J. Hendriks, I. Stojanovic, R. B. M. Schasfoort, D. B. F. Saris and M. Kaiperien, Analytical 29 Chemistry, 2018, 90, 6563-6571. 30 106. P. Y. Chen, M. T. Chung, W. McHugh, R. Nidetz, Y. W. Li, J. P. Fu, T. T. Cornell, T. P. Shanley 31 and K. Kurabayashi, Acs Nano, 2015, 9, 4173-4181. 32 107. Y. J. Song, P. Y. Chen, M. T. Chung, R. Nidetz, Y. Park, Z. H. Liu, W. McHugh, T. T. Cornell, 33 J. P. Fu and K. Kurabayashi, Nano Letters, 2017, 17, 2374-2380. 34 35 108. S. K. Vashist, E. M. Schneider and J. H. T. Luong, Analyst, 2015, 140, 4445-4452. 36 109. S. A. Vance and M. G. Sandros, Scientific Reports, 2014, 4. 37 110. B. Wu, R. Jiang, Q. Wang, J. Huang, X. Yang, K. Wang, W. Li, N. Chen and Q. Li, Chemical 38 Communications, 2016, 52, 3568-3571 39 111. Y.-H. Choi, H. Ko, G.-Y. Lee, S.-Y. Chang, Y. W. Chang, M.-J. Kang and J.-C. Pyun, Sensors 40 and Actuators B-Chemical, 2015, 207, 133-138. 41 112. M. Soler, M. C. Estevez, M. Alvarez, M. A. Otte, B. Sepulveda and L. M. Lechuga, Sensors 42 (Switzerland), 2014, 14, 2239-2258 43 113. M. Pawula, Z. Altintas and I. E. Tothill, Talanta, 2016, 146, 823-830. 44 45 114. Q. Wu, Y. Sun, D. Zhang, S. Li, Y. Zhang, P. Ma, Y. Yu, X. Wang and D. Song, Biosensors & 46 Bioelectronics, 2017, 96, 288-293. 47 115. P. Palladino, M. Minunni and S. Scarano, Biosensors & Bioelectronics, 2018, 106, 93-98. 48 116. T. Liyanage, A. Sangha and R. Sardar, Analyst, 2017, 142, 2442-2450. 49 117. M. Ashaduzzaman, S. R. Deshpande, N. A. Murugan, Y. K. Mishra, A. P. F. Turner and A. 50 Tiwari, Scientific Reports, 2017, 7. 51 118. V. Singh and S. Krishnan, Analyst, 2018, 143, 1544-1555. 52 119. W. Diao, M. Tang, S. Ding, X. Li, W. Cheng, F. Mo, X. Yan, H. Ma and Y. Yan, Biosensors & 53 54 Bioelectronics, 2018, 100, 228-234. 55 120. E. C. Peláez, M. C. Estevez, A. Portela, J. P. Salvador, M. P. Marco and L. M. Lechuga, 56 57 Biosensors & Bioelectronics, 2018, 119, 149-155. 58 59 60 Page 47 of 47