The diagnosis of an infection depends on the presence of specific signs and symptoms, typical findings in medical imaging, or the detection of specific biomarkers or other biomolecules resulting from a bacterial or viral infection. In many cases, the identification of an infection takes place after the patient notices that he or she is feeling sick and decides to visit a medical doctor. Due to incubation time, patient delays and doctor’s delays, this may be early or relatively late in the time course of an infection. In resource-rich countries, the majority of infections can be cured or at least be turned into a chronic asymptomatic disease state with good medication and health care, as we see, for example, with human immuno-deficiency virus (HIV) positive patients. This is in contrast to the developing world. Here, access to good medication and health care is still difficult and infectious diseases are still a major cause of illness and mortality. Furthermore, even though initial outbreaks may be geographically localized, globalization and international travel can lead to rapid spread of infections around the world and can therefore represent a global threat as we saw for example with H1N1 influenza virus (1) or the HIV virus (2) (see also Figure 1).
Availability of accurate and inexpensive point-of-care tests for the spectrum of infectious diseases occurring in developing countries would improve the access to medical treatment. Important here is the ability to rapidly and correctly identify and differentiate diseases with similar symptoms such as malaria, dengue and chikungunya infection which all have fever in common. As a consequence, clinicians would be able to effectively treat and isolate patients without the need of expensive laboratory facilities and reduce pathogen spread and alleviate illness and mortality (3).
The development of new rapid point-of-care tests for developing countries poses new challenges: criteria, which are pivotal for such applications are specificity and sensitivity, time consumption, robustness, price, independence from expensive equipment, ease of use, and ease of interpretation for healthcare workers with limited training. We define the term nano-diagnostics as the spectrum of clinical diagnostic tests that exploit nanotechnology, e.g., to allow highly parallelized analyses, to reduce time, cost, consumption, and to improve the sensitivity and the specificity. Nano-diagnostics are a part of the nanomedicine field, which includes the application of nanotechnology to medical prevention, diagnosis and therapy. In general, the use of nanotechnology can help to extend the limits of the current diagnostic tools in various ways. Therefore, this review will focus on basic principles, illustrated by various examples of current work and developments in the field of infectious disease diagnosis. The analytical targets include both biomarkers originating from the infectious agents and antibody responses. Frequently used samples include blood, urine or saliva. Tests that do not need a sample preparation step are particularly beneficial, especially if the goal is to provide a result within few minutes up to 1 h, a time window, which is also suited for outpatient clinics and medical care in the field. Lateral flow-based strip tests currently represent the most frequently used point-of-care tests, while microfluidic devices are becoming increasingly popular. Both, strip tests, often based on color-based detection of nanoparticles, and microfluidic devices, allowing to perform diagnosis on nanoliter sized samples, are part of the evolving nanodiagnostic field.
This review starts with summarizing the principles of the lateral flow tests to then introduce microfluidic devices and explore their capabilities and limits by providing recent examples of their application for multiplexed assays. Finally, further developments in nanotechnology suited for diagnostic applications will be discussed. In an upcoming review, we will separately introduce important pressing infectious diseases, with an emphasis on frequent infections that manifest as febrile syndromes such as malaria, dengue, chikungunya, typhoid fever, influenza, and pneumonia, respectively. HIV is included as this is an important infection with a high number of non-diagnosed patients in developing countries that profoundly influences the general health status of its carriers. In this separate review, we will also refer to the current reference diagnostic method and the latest developments in lateral flow tests, microfluidic devices as well as the relevant developments in other nanotechnology-based tools will be presented. As a comprehensive review of the huge body of work that is being performed is out of scope of this paper, we aim to provide an update of the latest achievements in the field. The cited articles were collected by cross-referencing online keyword searches in citation and database searching. The emphasis of the literature citations is on the evolution of lateral flow and microfluidic tests for point-of-care applications.
From lateral flow assays to microfluidic devices
This brief introduction to the principles of lateral flow assays and immunoassays as currently used in strip tests forms the starting point for discussing the principles, benefits and limitations of microfluidic devices.
Lateral flow assays on membranes
In lateral flow (LF) assays, for which a widely known example is the pregnancy strip test, the fluid flow is typically capillary-force driven and controlled by both the wetting properties and the structure of a porous membrane. The chemicals required for the test are pre-stored within the strip. The readout of such a test is usually done optically, e.g., by visual detection of a color change in the detection area, optionally supplemented by an objective optical readout. LF tests are used for qualitative, semi-quantitative and in few cases, quantitative monitoring in resource-poor or non-laboratory environments. In most cases, they are relatively inexpensive and provide a result within a few minutes. The two most frequently used test principles are the direct (non-competitive) and the inhibition or competitive assay (see Figure 2). The non-competitive test is well suited when testing for larger size analyte molecules with multiple antigenic sites and when it is not feasible to provide an excess of the sample analyte. The competitive reaction test is typically used when testing for small molecules with single antigenic determinants, which cannot bind to two antibodies simultaneously (4).
Commonly used membranes are nitrocellulose (high protein binding), cellulose acetate (low protein binding) and glass fiber membranes (non-protein binding), whereby choice of the membrane type is dictated by the specific test requirements. In addition to the chemical properties, other variables are also crucial, such as possible flow rate of the sample with a specific viscosity through the membrane, defined by the membrane porosity, and membrane capacity. The porosity of the membrane is important for the determination of the amount of bound antibodies per unit area of the membrane (5).
Antibodies are used in the stationary phase in the control lines and test lines and can be used in the mobile phase for initial analyte binding. Antibody color labeling by micro/nanoparticles is discussed more comprehensively elsewhere (6, 7). Colored micro/nanoparticles are commercially available and the dimensions are typically up to 1/10th of the pore size of the membrane. Hydrophobic interaction between membrane and particle can be minimized by suited pretreatments, e.g., with sucrose or water soluble inert polymers such as polyvinylpyrrolidone (PVP) (5).
Reaction kinetics are critical. The test must allow sufficient time for optimal antibody-antigen binding, but the overall test should consume as little time as necessary. Several parameters have to be optimized: (1) with increasing flow rate, the reaction rate, the assay time and the sensitivity decrease; (2) with increasing flow rate, the reagent consumption will increase, (3) the flow rate decreases as the distance from the origin increases, and (4) when nitrocellulose membranes are used, the amount of protein bound decreases with increasing pore size (5).
Lateral flow has the advantage that test strips can be easily produced in almost any standard laboratory. However, sensitivity is typically relatively low. One reason for limited sensitivity is that the detectable signal stems from the top layer of the membrane, which is roughly the first 10 μm of the membrane. The thickness of such a membrane is usually 100 m and therefore part of the signal below may be missed (8). Various methods to improve sensitivity including signal amplification, can be found in the literature but are beyond the scope of this review. Nanotechnology strategies to increase sensitivity will be discussed later in this review in the section “Nanotechnology for in vitro diagnostics.
Microfluidic devices for immunoassays
Many microfluidic devices fall in the category of micro total analysis systems, also known as lab-on-a-chip (LOC) devices, denoting microsystems that carry out a sequence of analytic steps sequentially. The use of nano- and microtechnology has led to a range of new miniaturized total analysis devices for many different purposes in the last decade [for further reading see ref. (9–12)]. In this review, also microfluidic devices are cited that do not fulfill all the characteristics of a total analysis microdevice. Microfluidic devices are powerful platforms that can be used for the detection of multiple analytes and the use of more sophisticated manufacturing techniques opens the door to the integration of additional functions on the same device. These systems often combine a number of building blocks discussed below.
Sample injection takes place at a macro-to-micro interface allowing application of tiny amounts (microliters down to nL) of a fluid from a macroscopic sample to the microsystem. One simple solution for injection of the sample into the microfluidic device is the use of a syringe and a high pressure and precision syringe pump as presented by Salieb-Beugelaar et al. (13). Here, tiny amounts of the sample can be pumped into the channels of the microdevice. Alternatively, capillary pressure and evaporation effects can be employed as done by Xu et al. (14) or Zimmermann and coworkers (15). They developed a controllable on-chip pump that enables the valve free sample injection at nL level. Here, the advantage is that it does not require peripheral equipment. A proof of principle was provided by the injection and detection of amounts of dsDNA (10 ng/μL) stained with intercalated ethidiumbromide. In 2009, Sato and coworkers (16) presented a nL sample injector that enables the injection of 1.9 nL with a volume accuracy of 2%. The device was manufactured from the fluoroplastic material NEOFLON EFEP that is suitable for biological assays. In addition, this work is an example of a high-performance-low-cost microfluidic device (16). Another possibility is the use of an electrical field to inject charged molecules into a device, which is similar to the injection of molecules into a capillary electrophoresis system, already known for many years. Scott et al. (17), presented microdialysis sampling coupled to an electrophoresis microfluidic chip by using a flow gated interface. The samples are electrochemically detected with a platinum electrode and one can use either the positive or the negative polarity in the device, using an injection time of 1 s and the total analysis time of the sample was between 40 and 60 s (17). For further reading about sample injection, see, e.g., Lee et al. (18).
For fluid pumping, capillary forces, optionally enhanced by controlled evaporation, are well suited (14, 15). In a similar but not continuous way, the capillary forces can be used to pump the fluid through the microfluidic channels. Fluids can also be pumped through the channels by deflection of the elastic membranes as presented by Cui and coworkes (19). Their device is manufactured from polydimethylsiloxane (PMDS) and includes microfluidic channels and actuation membranes. A vacuum source regulated by an electromagnetic valve induces the sequential deflection of the elastic membranes that results in a flow rate of maximal 600 μL/min. Yang et al. (20) presented a pneumatic micropump integrated with a normally closed microvalve. For the control of the fluid flow in the device, only one electromagnetic valve is used. This microvalve is activated by the hydrodynamic pressure generated by the pump itself. The applied pressure and driving frequency of the electromagnetic valve is basically determining the pumping rate. Pumping rates of 900 μL/min were provided and high back pressures can be generated of up to 20 psi (1.378×105 Nt/m2). Valves based on surface tension effects can also be exploited for directional flow (21). In microfluidic devices on paper platforms, fluid flow is limited by capillary forces. Godino and coworkers (22) used paper in a polymethyl methacrylate (PMMA) microfluidic disc. An interplay of the centrifugal and capillary forces was used to demonstrate the capability of blood separation, fluid routing and valving. With this design, Godino et al. (22) presented a user-friendly, low-cost device that is capable of dry reagent storage due to the integration of paper. Another possibility is the use of electro osmotic pumps (EOP) where the application of an electrical field (AC or DC or a combination of both) results in a flow. A non-uniform AC field introduces the possibility to separate (bio)molecules (or particles) based on their di-electrical properties, the shape and size and the applied AC frequency. When using an electrical DC field only, the charged molecules will separate inside a device and as a result the flow rate will decrease, an obstacle that can be addressed by use of an AC field as reported by Huang et al. (23), who developed an AC-EOP. When used at low AV voltages (1 V rms, 1 kHz), power (5 mW) and current (4.5 mA), the pump is capable to produce a 1.3 kPa head pressure and an effective slip velocity of 1.3 mm/s over the electrodes in the device. The AC-EOP can be used to pump biological samples of 100 mM strength in a separate microfluidic device by using distilled water as the working solution. The proof of principle was presented by the DNA hybridization in a microfluidic microarray.
In laminar flow, as is typically present in microfluidic devices, mixing depends on diffusion. To accelerate mixing, chip geometries such as grooves in the microchannel wall or zigzag shaped channels can be used, which result in flow vortices, thereby enhancing the (passive) mixing of the fluids. The advantage of such on-chip incorporated structures is that they are reliable and low at cost as these do not require peripheral equipment such as pumps. Alternatively, one can also actively mix fluidics, e.g., by a periodic electric fields. Lim et al. (24) developed a T-shaped mixer driven by an AC field. At the inlets, two AC fields with a phase difference of 180° with respect to each other are induced. This results in a fluid oscillation at the junction of the two inlets and the formation of alternative plugs at the constriction of the micro-channel. Half circle shaped waves emanate after arrival in the much wider micro-channel, thereby increasing the contact area and thus, the mixing between the two fluids (24). Acoustic waves can also be used for active mixing: Liu et al. (25) demonstrated the use of air bubbles on a solid phase to generate circulatory flows induced by sound vibrations, which led to mixing of two dyes in a microfluidic chamber of 50 μL in only 6 s. El Moctar and coworkers (26) developed an electro-hydrodynamic mixer for fluids that have different electrical properties and identical viscosity and density. The electrodes are placed such that they generate an electric field perpendicular to the fluids in the microchannel, resulting in a secondary, transversal flow and leading to excellent mixing in <1 s over a short distance. With an eye on simplified designs that can be used in resource poor environments, active mixing appears usually not necessary. For more about microfluidic mixing, the review of Lee et<nonbrspace;al. is recommended (27).
Microreactors and reaction sites
Microreactors or reaction sites are locations inside a microfluidic device where various physical, chemical, biochemical events take place, affinity assays can be performed and biological events can be observed such as single cell receptor analysis. Grünberger and coworkers (28) developed a disposable pL bioreactor that can be used to cultivate and investigate bacteria on single cell level. With this reactor, they showed 1.5-fold increased grow rates of the Corynebacterium glutamicum (wild type) bacteria when compared to the conventional lab-scale batch cultivation. After inducing artificial starving conditions, they could analyze and quantify the morphological changes of this organism. Remarkable is that only 1 h was enough to provide the information that is usually collected in a 1-week experiment. They also presented a 24 h timelapse experiment of the cultivation of an arginine producing strain that encoded a fluorescence sensor. Single cell productivity and growth could be followed. Such a device could be very useful in healthcare for the rapid investigation on the response of the patient’s bacteria on the specifically chosen and dosed antibiotics prior to the actual start of the treatment. Of course, this is related to the patient’s condition and the state of the infection, but will contribute to limiting the spread of drug resistance. Freifeld et al. (29) used a single-use microreactor for the preparation of stool samples and the extraction of DNA for the rapid diagnostis of Clostridium. difficile infection that is followed by a rapid polymerase chain reaction (PCR) step. Agarose gel electrophoresis showed that spiked stool samples with only 11 genomic copies/mL could be reliably detected with total analysis time of only 20 min. This work is promising for future rapid and reliable POC tests. For further reading about enzyme-immobilized microfluidic process reactors, we refer you to the work of Asanomi and coworkers (30).
Depending of the final purpose of the device, several detection modalities are available. Fluorescent labels and the coupling to a proper microscope setup or a portable readout system are well known. Wolf and coworkers (31) coupled their micromosaic immunoassay device to a fluorescence microscope and a CCD camera. The limit of sensitivity was ∼0.03 μg/mL for the cardiac marker C-reactive protein (CRP).
The microfluidic device might also be coupled to a portable mass spectrometer. Other detection possibilities lay in the field of plasmonics that paves the way to detect molecular interactions on almost a single molecule level and without the use of fluorescent labels. This field is currently rapidly evolving, as, for example, indicated by the work of Henley et al. (32), who developed a laser-patterning method to fabricate plasmonic nanostructures on glass that could be used for microfluidic devices. A suitable method in this field to detect antibody-antigen binding is surface plasmon resonance (SPR). Here, changes in the local refraction upon adsorption are measured. For further reading about SPR, see the review of Helmerhorst et al. (33). For POC devices, visual detection and readout is desirable. In a recent work of Souza et al. (34), a toner-based device with colorimetric detection was presented. This design is based on a computer design that is laser-printed on a polyester film. After alignment of its mirror image, the films were thermally laminated at 150°C and 100 μm deep microchannels were obtained. The advantage of this manufacturing technique is that in 5 min, 35 single polyester pieces were obtained at 10 cents each and at total device costs of around 15 cents. The sample volume is 40 μL and the limits of detection were 0.2, 0.3 and 8 mg/mL for cholesterol, glucose, and protein, respectively. They also showed that the device is capable of performing bioassays without loss of activity after 5 days of storage at different temperatures and thus the device might be a good candidate for the use in resource-poor environments. The disadvantage was, however, the difficulty to produce more complicated designs, the lack of uniformity along the microchannel and the poor device-to-device reproducibility. Nevertheless, this device offers possibilities for further improvement such as the integration of paper technology on chip. Another creative example of a colorimetric detection method was presented by Zhou et al. (35). They developed a sandwich immunoassay on cotton threads in a microfluidic cartridge. The design allows multiple tests to be conducted in parallel with a limit of detection in the picomolar range and was comparable to other lateral flow tests. The antigen was first bound to the detection antibodies coupled on the surface of the golden nanoparticles and subsequently to the capture antibodies in the capture zone. The accumulation of gold nanoparticles in the capture zone on the thread, results in the visualization of the result. The intensity of the color is proportional to the concentration. By using a flatbed scanner instead of a more expensive densitometer, signals could be quantified. The capacity of the device was shown by performing an immunoassay for the detection of CRP in diluted serum, for which the limit of detection was 377 pM with results becoming visible to the eye within minutes. Zhou and coworkers (35) suggested that the device might be capable of using other physiological fluids like saliva or tears. For further reading about detection for in situ diagnosis on chip, see the work of Lee and Lee (18).
The continuous developments in microtechnology and microfabrication have resulted in a large number of innovations for microfluidic POC devices in the last decade. Materials used to manufacture such devices include borosilicate glass, fused silica, silicon, various polymers (e.g., methacrylate), PMDS and even paper. The chemical properties of the materials and the geometry of the device determine the flow of the fluid and physico-chemical processes occurring in the channel and at the channel walls. The developments and challenges in this field are beyond the scope of this review but were recently reviewed by Nge and coworkers (12).
A microfluidic system is a complex, engineered structure, where manifold physical and biochemical processes occur simultaneously or sequentially, and are a prime example of the need for interdisciplinary expertise. Full understanding of such systems is often beyond experimental exploration and may profit from complementary computer modeling. This allows the investigation of relevant design parameters, which influence the design performance and allow optimizing a system even before manufacturing begins. Zimmermann and coworkers (36) modeled microfluidic flows in combination with an immunoassay on the surface of microchannels. The applied finite difference algorithm was used to elucidate the influence of the fluid transport of the analyte, of analyte diffusion, of binding kinetics between antigen and antibody and the surface. Moreover, optimal capture site antibody densities were explored, focusing on the alternative scenarios ‘maximum sensitivity’, ‘minimum sample volume’, and ‘minimal assay time’. This resulted in a set of design parameters leading to picomolar sensitivity within 2 min test time, or alternatively, picomolar sensitivity using a few tens of a nanoliter sample. They also show that the use of microfluidic networks might enable the multiplexed detection of analytes in a single drop of samples within minutes. Interconnected 3D thread (Nylon) networks were used as a novel template method. The flow inside the channel network was investigated prior manufacturing.
Simulations can also be applied to optimize specific microfluidic building blocks. Sabotin et al. (37) used simulations to design patterned groove micromixers. This was done in two steps. First one groove of the staggered heringbone micromixer (SHM) was designed such that the transversal fluid movement was maximal at the end of a single groove and at a fixed channel aspect ratio. By using simulations, it was concluded that the favorable geometry of the groove is dependent of the Reynolds (Re) number, i.e., a dimensionless quantity used in fluid mechanics. It describes the ratio of inertial forces to viscous forces of a fluid and predicts whether a fluid is laminar or turbulent. The laminar flow normally encountered in microfluidics retards mixing of fluids as mentioned above. Sabotin et al. (37) showed by computational modeling that for the Re regimes 0.5≤Re≥20, the suitable groove depth stayed constant. When the Re number is higher, a wider groove is required. In the second step, the adequate groove was further investigated in six groove configurations. It was concluded from the simulations that mixing depends on the configuration layout and that the experimental results performed with the manufactured groves were in agreement with the simulations. These findings support the validity and the utility of computer simulations prior manufacturing of microfluidic devices. For further reading, see, e.g., the work of Bai et al. (38), describing the simulation of cell adhesion and detachment in microfluidics, or the work of Yuan and coworkers (39), which presents the optimization of a planar interdigitated microelectrode array that can be used for biofluid transport by AC electrothermal effects.
Examples of diagnostic assays in microfluidics
The recent developments in microfluidics for (multiplexed) immunoassays are very promising, which will be illustrated by the following examples. Yang and coworkers (40) developed a capillary microreactor for chemiluminescent immunoassays. The wall of the capillary is coated with streptavidin. This was proven to be useful for the support for highly efficient immobilization of antibodies in a flow cell for flow-through immunoassays. The performance was tested using α-fetoprotein (AFP) as model analyte, resulting in a detection limit of 0.1 ng/mL and a linear range from 0.5 to 200 ng/mL. White and coworkers (41) developed a platform for the multiplexed detection of antibodies. This wash-free electrochemical method enables the quantitative detection in undiluted and unprocessed blood serum in sub-nanomolar detection limits. For the detection of anti-protein antibodies, they developed redox-tagged, electrode-bound nucleic acid probes. The anchor strand is modified with the redox reporter methylene blue at the 3′ terminus and attached to an electrode via the 5′ terminus (see also Figure 3A). The recognition strand is coupled to the relevant antigen at the 5′ terminus. When no antibody is attached, the methylene blue reaches the electrode surface and electrons are transferred. When an antibody is bound the epitope, the electron transfer is reduced (Figure 3B) followed by a detectable change in faradaic current (Figure 3C and 3D). Five selected anti-HIV antibodies and an anti-FLAG antibody were successfully detected in the multiplex test set with sub-nanomolar detection limits (see Figure 3D). This nearly universal approach opens doors to new microfluidic diagnostic and point-of-care devices. Many microfluidic POC devices, including lateral flow strip tests, are fabricated using standard film lamination techniques [see, e.g., (42) or (43)]. Using pressure sensitive adhesive, different materials can be bonded to each other. An example is the multilayered microfluidic immunoassay card presented by Lafleur and coworkers (44). Proof of principle was provided with the detection of the IgM antibodies against Salmonella typhi LPS and malarial antigen HRPII from Plasmodium falciparum. This microfluidic POC device uses four different types of membranes used for filtration and protein binding, including blood filtration, bead filtration, a nitrocellulose membrane for the assay and an air-permeable membrane. They were sandwiched between laminate layers. In total, the device consists of 23 layers. Fluid flow inside the card is controlled by a pneumatic system. It is well suited for small amounts of fluids. However, flow rate cannot be controlled and there is a risk of air bubble formation on the air-liquid interface. Fluid flow control on the card required a clever design. For the S. typhi assay, the IgM antibodies were the target and therefore an additional step was required, namely the dilution of the sample and the removal of the IgG antibodies. This was also performed on-chip by the selective binding of the IgG molecules on protein G-coated beads. IgM-specific antibodies were coupled to gold nanoparticles, so that, upon IgM-binding to the immobilized typhoid lipopolysaccharide (LPS) antigen, gold nanoparticles accumulate and lead to a visible signal. The assay was tested by clinical plasma samples. For the HRPII assay, IgM antibodies were bound on the surface of the membrane and IgG antibodies were coupled to gold nanoparticles. When the immunosandwich is formed (see Figure 4), a visible signal readout is also achieved by gold nanoparticle accumulation. The limit of detection for the HRPII antigen was between 10–20 ng/mL, but was not fully clear for S. typhi due to the limited number of samples. The generic design of this device and these assays renders them potentially useful for other immunoassays (44).
Nanotechnology for in vitro diagnostics
Nanotechnology comprises research and development (including design and manufacturing) of nano-size structures. This field thus ranges from nanovesicles to mechanically nano-structured surfaces but also includes the use of nanometer-sized tips to scan and modify surfaces. In this part of the review, we will present relevant recent progress in nanotechnology methods for diagnostics, illustrated, where possible, by applications suited for diagnosis of infectious diseases.
Nanobeads or nanoparticles are increasingly used in strip tests or microfluidic devices. In particular, signal enhancement by gold nanobeads, a manifestation of surface plasmons, is already widely applied. New applications with various types of particles, such as super-paramagnetic nanoparticles (SPN) are still being developed. An example of this is the recent work of Yu and coworkers (45), who describe a magnetic field-controllable microfluidic chip that is capable to rapidly detect dual cancer biomarkers in serum. A nickel pattern was manufactured on-chip, enabling the generation of high magnetic field gradients to increase the magnetic force on the SPN. The non-functionalized nanoparticles were ∼500 nm, growing to ∼600 nm after surface functionalization. A sandwich immunoassay method was developed, allowing detection of carcinoma embryonic antigen (CEA) and α-fetoprotein (AFP) within 40 min, significantly faster compared to the conventional ELISA method. To specifically detect the biomarkers, streptavidin-modified fluorescent quantum dot antibodies were used. The linear range was from 10–800 ng/mL with a detection limit of 3.5 ng/mL for CEA and 3.9 ng/mL for AFP. In another investigation, Zheng et al. (46) used SPN in a lateral flow assay for the rapid detection of the major fish allergen parvalbumin. The linear range was from 0.01 to 100 μg/mL and the limit of detection for qualitative detection was 5 μg/mL and for the quantitative detection 0.046 μg/mL. To detect the signals, a magnetic plate reader was used, and total test time was 20 min, less than the 5 h needed for the conventional method (Western blot). Zheng and coworkers (46) bound antibodies to the SPN through the amino group of the antibody and the carboxyl surface of the SPN. It appears that covalent binding of antibodies to nanoparticles improves the availability of binding sites. Orientation of the antibody is important for optimal antigen binding for steric reasons. In 2010, Puertas et<nonbrspace;al. (8) compared two different methods for the immobilization of antibodies on magnetic nanoparticles. The first, conventional procedure used the amino group of the antibody and the carboxylic groups of the nanoparticles. The second procedure exploited the carbohydrate chains located in the Fc region of the antibody of the antibodies for linkage to aminated nanoparticles. Figure 5 presents the type of orientation of the antibody on the bead after using the carbohydrate chains (5B) or the amine moieties (5C).
Wang and coworkers (47) used silica nanobeads to capture cells under continuous flow and showed that the nanotopography of closely packed arrays of the nanobeads is relevant for capture efficiency. Bead diameters from 100 to 1150 nm where used. Capture efficiency at low flow rates generally increased when nanoparticle size increased. Such dense packing of beads on a surface might increase the binding capacity of antigens or specific proteins from serum or blood in a similar way.
Serrate et al. (48) developed a device, which measures the local magnetization of nanoparticles on strips (nitrocellulose membrane). The device is capable of detecting 7.9 ng of SPN (200 nm), corresponding to ∼4×105 particles. Signal strength of the sensor of the device was proportional to the magnetic field in a wide concentration range. An assay for the detection of chorionic gonadotropin hormone was developed, achieving a limit of detection of 5.5 ng/mL. Detection can be performed in two different ways, the “friction free” and the “pulling method”, respectively. The first method does not damage the used chip, it, however, prolongs measurement time to about 10 min, whereas the second method damages the chip, but scanning can be done within 80 s only with an excellent signal-to-noise ratio. In a related study, Marquina and coworkers (49) explored the magnetoresistive response of a spin-valve giant sensor. This sensor is placed such that it is in vicinity of the particles that are present in the strip. They suggested that the use of tunnel magnetoresistance as published by Cardoso et al. (50) might improve the output signal by a factor of 50. In addition, they communicated that, according to their experience with several commercial manufacturers, nanoparticles in the market were far from being optimal.
Signal amplification is a valuable capability of gold nanoparticles and already widely applied. The advantage of gold nanoparticles for detection is that it allows relatively simple, low-cost assays, which do not require special facilities. An example is the CEA sandwich ELISA based on a gold nanoparticle layer as presented by Zhou et al. (51, 52). Gold nanoparticle layers in a microtiter plate format were associated with a limit of detection of 2 ng/mL, communicated to be significantly higher compared to commercial solutions, although limited by long assay times, namely overnight incubations. Mak and coworkers (6) produced organic indigo-nanoparticles, on which antibodies can be coupled. After binding of the immunocomplexes in the test zone, a signal-developing reagent is added to visualize the signal. The lowest visible signal reported was 1.25 ng/mL IgG. The provided sensitivity (signal-to-noise ratio) was even better when compared to the gold nanoparticle label. In another investigation, Juntunen and coworkers (53) reported the possibilities of fluorescent nanoparticles and presented the use Eu(III)-nanoparticles on test strips. The use of these nanoparticles in an immunosandwich method allowed detection of the prostate-specific antigen (PSA) with a concentration limit of around 80 pg/mL.
Recently, Zong and coworkers (54) presented a multiplex immunoassay with a limit of detection (LOD) of 0.1 pg/mL. The method uses scanning surface-enhanced Raman scattering (SERS)-fluorescence dual mode nanoprobes combined with magnetic nanobeads. Nanoprobes were generated by assembling nanorods and quantum dots onto silica nanospheres. The nanorods are AU and Ag core shell nanorods and serve as the SERS substrate. The quantum dots act as fluorescent indicators. Antibodies are coupled on both systems and when the specific antigen is present, an immune complex is formed. This structure generates both a strong SERS and a fluorescence signal. The target molecules were human IgG, mouse IgG, bovine IgG, while on the nanoprobe and magnetic nanobeads, goat anti-human IgG or goat anti mouse were coupled. The two different nanoprobes were labeled with 5,5 dithiobis (2-nitrobenzoic) (DNTB) or 4-mercaptobenzoic acid (4MBA) to generate additional SERS signals. The DTNB nanoprobes were used for the goat-anti human IgG- and the 4MBA-labeled nanoprobes for the goat anti mouse IgG. Immune complexes were only formed upon addition of the proper IgG molecules. Even though some non-specific binding was found, the LOD was 0.1 pg/mL. As multiplexing is possible with this test, it may lead to more sensitive and specific immunoassays.
A final example is the work of Noguera and coworkers (55) using carbon nanoparticles (CNP) in a lateral flow setup for the detection and identification genes encoding for different Shiga toxin virulence factors of the Escherichia coli bacteria. The method is based on the earlier mentioned nucleic acid lateral flow immunoassay (NALFIA). Prior to detection, the samples are amplified with tagged primers, resulting in a limit of detection of 0.1–0.9 ng/μL DNA. Such carbon nanoparticles are able to bind proteins non-covalently, such that neutravidin can be bound to the surface, PCR primers are labeled at both ends with a specific tag and a biotin label. In this way each amplicon can be captured to the neutravidin coated CNP with the biotin and the specific antibody for the other tag, which is immobilized on the strip (nitrocellulose membrane). For further reading about carbon nano particles as signaling labels in rapid diagnostic arrays, we would like to refer to the review of Postuma-Trumpie (56) and for the use of nano-sized labels to the review of Yu and coworkers (57). In general, we can conclude that the use of nanoparticles can improve the sensitivity of immunoassays.
The use of nanoporous membranes for in vitro diagnosis is likewise rapidly developing and promising. In 2010, Chai et al. (58) presented a method to immobilize antibodies on a nanoporous aluminum surface, which can be used as an immunosensor. The aluminum surface was silanized with 3-aminopropyltryethoxysilane (APTES) and antibodies were covalently bound to the surface in the presence of glutaraldehyde. Monoclonal antibodies against Staphylococcus aureus enterotoxin B were used investigate the functionality of the surface. Time resolved electrochemical impedance spectroscopy was used to transduce the physico-chemical changes upon antibody-antigen binding into analytical signals. Antigens could be measured in real-time and with concentrations <10 pg/mL. In 2012, Chai et al. (58) demonstrated the immobilization of antibodies on a nanoporous surface and concluded that with these surfaces more antibodies could be bound. While an increase in the fluorescent intensity was achieved, an improvement of the sensitivity of the immunosensor could not be not achieved. Prasad and coworkers (59) used nanoporous alumina integrated in a silicon microchip. The nano membrane was embedded on the surface of the metal electrodes. The pores formed nanowells with an average diameter of 200 nm and a depth of 250 nm. Due to the capillary forces, antibodies flowed to the bottom of the well. To immobilize the antibodies within the nanowells, dithiobis succinimidyl propionate (DSP) was used. Under the conditions used in this experiment, the electrical double layer was higher than 50 nm. Thus, as a result of the dimensions of the pore size and the antibody, the charge distribution was changed upon binding of both the antibody and later antigen on the surface of the gold working electrode, resulting in a measurable change in capacitance at the interface. The signal is the sum of all the individual pores that can be seen as an array of multiple parallel capacitors. The signal is therefore an amplification provided from all these pores and enables the detection of <10 ng/mL. To validate these nanosensors, immunoassays were developed for two cardiac biomarkers, C-reactive protein (CRP) and NT-pro-brain natriuretic peptide (BNP). Prasad et al. (59) achieved a near real-time detection with a sensitivity of 1 ag/mL for BNP and 1 fg/mL for CRP in human serum. From these investigations, it can be concluded that nanopores are a fascinating new option to develop POC devices for very sensitive and rapid diagnosis.
Cheng et al. (60) used nanoporous alumina of 300–500 nm in thickness, placed onto a sensing platinum electrode. Pore sizes were 10–200 nm in which the antibodies were adsorbed. When the immune complexes are formed within the nanopores, the Faradaic current of the electrode towards the redox probe changes and can be measured. The total analysis time is 50 min with a limit of detection of 1 plaque forming units (PFU)/mL for the dengue virus type 2. The linear detection range was 1–103 PFU/mL. This is similar to the results found by Nguyen et al. (61). In another study of Peh and coworkers (62), the detection of the dengue virus was demonstrated using impedance measurements across a nanoporous alumina membrane. The LOD was 0.230 PFU/mL and 0.710 PFU/mL for the dengue virus type 2 and 3, respectively, with a total analysis time of 40 min. This work presents a simplified electrochemical setup that can be incorporated into disposable POC diagnostic tools. Nanopores can also be used to preconcentrate samples as was done by using elektrokinetic trapping in a nanofilter (63) or by the use polycarbonate nanopores (64). The applied elektrokinetic fluid flow resulted in a highly localized accumulation of the proteins with an accumulation factor of 105–108 in both investigations. It can therefore generally be concluded from that the use of nanomembranes holds promise for future developments for POC tests.
Nanowires and nanotubes
Already in 2001, Ciu and coworkers (65) presented the capabilities of boron-doped silicon nanowires. They showed that the biotinylated nanowires could detect streptavidin in the picomolar range. Patolsky and coworkers (66) presented single virus detection by the use of silicon nanowires in 2004. Nanowires were modified with specific antibodies for the simultaneous detection of influenza or adenovirus and it was shown that multiple viruses could be selectively detected in parallel. Silicon nanowires have already been used in microdevices for the multiplexed detection of biomarkers. In these devices, the nanowires function as field effect transistors (FET). In brief, a FET is composed of a gate (e.g., an aluminumoxide layer), a drain and a source. Two types of FETS exist, an n-channel or a p-channel, which conduct the current in electrons or holes, respectively. Changes in surface charge on the gate, e.g., by binding of molecules, lead to changes in FET conductance form drain and source, enabling label-free detection of an analyte. pH and the presence of other ions will influence this effect. Therefore, desalting the sample might be required, which complicates the design of POC tests. Tarasov and coworkers (67) developed a reference FET, of which the proton sensitivity is suppressed by two orders of magnitude. This was achieved by the vapor deposition of a silane layer with a long alkyl chain on the highly pH sensitive layer of the alumina surface of the silicon nanowire FET. In 2010, Gong et al. (68) showed that proteins could be detected down to the attomolar range by nanowires and an alternating current (AC). The use of an electrical AC field results in the capability to trap particles based on their dielectric properties, their shape and size. This is called di-electrophoresis (DEP) and can be optimized for each protein or particle. Gong et al. (68) further showed that the presence of an AC field resulted in an almost 104-fold increase in sensitivity. The huge advantage of using DEP is that the samples do not need to be purified nor desalted. As proof of principle, the detection of prostate specific antigen (PSA) and cholera toxin (CTB) was presented, although not in a multiplex setup. The LOD for PSA was ∼2 fM and for CTB ∼18 aM. This work shows that rapid highly sensitive diagnosis can be possible in the attomolar range. For further reading about dielectrophoresis, see for example the work of Abonnenc et al. (69) who developed a dielectrophoresis-based microarray chip for the manipulation and immunophenotyping of cells.
In the last decade, the use of cantilevers for various purposes has become popular in many different ways. In 2001, Hansen and coworkers (70) presented the capability to discriminate between mismatched and matched DNA strands. Single stranded DNA (20–25-mer) was bound to the cantilever and acted as a receptor for the target strands (10-mer DNA strands). When the target is completely complementary, than they will hybridize and provide a mechanical signal, whereas a non-complementary target, with 1 or 2 internal mismatches will produce a net negative deflection. In this investigation, it was for the first time shown that a stable and detectable hybridization occurs between complementary sequences of four basepairs. Wu and coworkers (71) showed the capability of using nanometer thick cantilevers as tools for diagnostics. The deflection experiments were performed with various geometries of cantilevers and using the specific and sensitive detection of the prostate specific antigen (PSA). For the longest cantilever (600 μm and 650 nm thick) a LOD of 0.2 ng/mL PSA was achieved, which was below the clinical threshold of 4 ng/mL (see Figure 6).
In 2003, a device was developed for multiple real-time label-free protein detection, which was presented by Arntz and coworkers (72). The cantilever sensors were functionalized with two different antibodies specific for the (label-free) detection of two cardiac biomarkers, creatin kinase and myoglobin. The deflections of the cantilevers were studied and proteins were detected in real-time in <10 min. Experiments showed that multiple biomarker proteins could be detected in an unspecific background and without cross-talk between the antibodies on the cantilever sensors. The device achieved a sensitivity of better than 20 μg/mL. The work reflects the reliability and the specificity of the developed cantilever array sensors. In addition, this approach permitted the use of up to seven different antigen-antibody reactions. Note that the correct bonding of the antibody, i.e., the orientation, on the cantilever is important and has to be controlled if optimal sensitivity has to be achieved. In this investigation, antibodies were covalent coupled in random orientation via their primary amines. An alternative might be the use of the carbohydrate chains of the antibody, as described above (8) or the use of nanobodies (also called single domain antibodies). The latter will be described in the next paragraph.
Atomic force microscopy (AFM) and cantilever based microscopy, was used to “see and count” individual antigens in an investigation presented by Roy and coworkers (73). They presented the visualization of the captured prostate specific antigens on a spot, which was microarrayed and showed the capability to count the number of antibodies on the surface. Interestingly, they were also able to measure the force specific for the bond between antibody and antigen. The number of captured PSA antigens was 12 in an area of 4.32×104 nm2 when the concentration was 100 fM. Gfeller and coworkers (74) used cantilevers to detect the active growth of E. coli cells within 1 h. The increasing mass on the cantilever array alters the resonance frequency. The sensitivity of this sensor is calculated to be ∼50 pg/Hz, which corresponds to approximately 100 E. coli cells. This approach enables to detect bacteria and their response to antibiotics within only 2 h. In 2008, Park and coworkers (75) presented the design, production and testing of the living cantilever array. This has a similar principle, but their approach is capable to measure mass of single adherent living cells (HeLa cells) in fluid. The cells were attracted to the cantilevers by using positive dielectrophoresis and then cultured on the cantilevers in a microfluidic environment. The resonance frequency shifts by the changes in mass. The mass of a single cell they found was close to the value found in the literature, which was calculated from cell density, and close to the volume acquired from confocal microscopy. This type of investigation opens doors to the study of cell mass during division or to a closer look on the physiological functions of cells versus their mass under various conditions. Cantilevers can also be manufactured such that they contain microchannels through which liquids can be pumped. This type of cantilevers is called suspended microchannel resonators (SMR). When pumping the liquid, the masses of liquid and suspended materials can be measured. The changes in mass are translated into changes in resonance frequency. Burg and coworkers (76) demonstrated that by using this technique, single bacterial cells, single nanoparticles and sub-monolayers of adsorbed proteins in water could be measured in sub-femtogram resolution. In Figure 7, an experiment with such a cantilever is presented. First, poly(ethyleneglycol)biotin grafted poly-L-lysine is electrostatically coupled to the surface. Then neutravidin is bound and finally the biotinylated anti-goat monoclonal antibodies are coupled. When the proteins are accumulated inside the cantilever, the resonance frequency shifts. This technique is suitable for the analysis of small amounts of liquid. The detector described in this study has a volume of 10 pL and the volume is swept more that 100 times per second at a flow rate of 100 nL/min (76). In addition, the large surface-to-volume ratio provides a high capture efficiency. However, the limitation in this technique is the high flow resistance inside the microchannels, which will prolong the detection time when larger fluid amounts need to be analyzed (76). In addition, cantilever biosensors allow direct monitoring of patients in an intensive care setting by continuous exhaled air monitoring, showing the generic nature of cantilevers as array sensors (77). For further reading about biosensing and the use of dynamic cantilever sensors, we refer to the reviews of Fritz (78) and Johnson and Mutharasan (79).
As already touched above, the orientation of the antibody is important for the specificity of the assay (8). Monoclonal antibodies are relatively large molecules (±12 nm in length). In particular, when also coupled to quantum dots (∼30 nm diameter), steric hindrance will be of influence before and after binding to a surface. Sukhanova and coworkers (80) presented the development of single domain antibodies coupled to quantum dots also called “nanoprobes”. The advantage here is that only the highly specific part of the monoclonal antibody (sdAb) is used and that the sizes are 12 times smaller when compared to an antibody-QD complex. In this work, it was shown that four copies of sdAbs could be coupled to a QD in a highly oriented manner. The proof of principle was provided by the development of sdAbs-QDs with an excellent specificity for the carcinoembryonic antigen (CEA) as demonstrated by the quantitative discrimination of CEA-positive and CEA negative cells by flow cytometry. In addition, the sdAbs-QDs were also used for the immunohistochemical labeling of biopsy samples and compared to gold standard methods. Here, the results were comparable or even superior. (80). These nanoprobes might be useful for (on-chip) sandwich immunoassays, where they might replace the detection antibody. The single domain antibodies described in the previous work are also mentioned in other investigations and are called nanobodies. In general, most antibodies are composed of two heavy and two light chains. Both chains contribute to the antigen binding and they are found in all vertebrates. Llamas, other camelids and sharks produce antibodies composed of only heavy chains. Here, the binding of these chains with the antigen in formed by a single domain and is termed VNAR in shark antibodies and VHH in camelid antibodies. Nanobodies have interesting properties such as thermal stability, high solubility, refolding capacity and good tissue penetration in vivo (81). In particular, the thermal stability renders them attractive as an alternative reagent in immunoassays in tropical, resource-poor countries. Kim et al. (82) developed a VHH-based immunoassay using 3-phenoxybenzoic acid, which is the major metabolite of pyrethroid insecticides, as a model system. They developed several immunoassays, including bacteriophage-born VHH, and compared the results. The best immunoassay showed a LOD of 0.01 ng/mL. They also showed that the VHH-ELISA (VELISA) was highly specific for 3-phenoxybenzoic acid and had a 150% cross-reactivity towards the hydroxylated derivate (4-hydroxy 3-phenoxybenzoic acid). The recovery from spiked urine samples ranged from 80% to 112%. We conclude that nanobodies offer promising possibilities for future POC devices under both resource-poor and conventional circumstances.
In recent work, Han and coworkers (83) presented a chip capable of detecting a common food-borne toxin, Staphylococcal enterotoxin B in attomolar concentrations. Here, photonic crystals were constructed in nanoscale wells on a chip. The nanostructure of the photonic crystals amplifies the fluorescent signal as a result of the guided mode resonance (83). To immobilize antibodies, nanoparticles were used. Subsequently, the nanoparticles were entrapped in individual wells by using dielectrophoresis. The photons of the emitted light the immunocomplexes were detected by using a single photon counting avalanche photodiode. This diode generates for each photon a single pulse. The device with nanowells presents a linearity >6 orders of magnitude and is reaching attomolar concentrations with a LOD of 1 fg/mL (∼35 aM) when compared to a reference ELISA plate. This is in a similar range as published by Maraldo et al. (84), who detected the enterotoxin B in apple juice and milk at 2.5 fg/mL using millimeter-sized cantilever sensors.
Proteinticles are nanoscale particles formed by self-assembly of certain proteins inside cells. The structure and surface topology is constant. Lee and coworkers (85) presented a method for the genetic engineering of these proteinticles. By genetic modification, a wide spectrum of specified proteins or peptides could be added or inserted to the internal region or the N- or C-terminus of the protein constituent. As a proof of concept, they engineered proteinticles with proteins/peptides on the surface that recognize disease-specific antibodies. For the detection of autoantibodies (multiple sclerosis), human ferritin-based proteinticles displaying the extracellular domain of myelin oligodendrocyte glycoprotein (MOG) (native conformation) were engineered, which enabled the discrimination between autoantibodies to native or denatured MOG and lead to the reliable diagnosis of multiple sclerosis with improved accuracy. The limit of detection was ∼ pM. Human ferritin-based proteinticles, for the highly sensitive detection of the antibodies against (non) structural proteins of the hepatitis C virus were also engineered. The heterogenous peptide surface of the proteinticles enables the detection of the anti hepatitus C antibodies in patient sera with 100% accuracy. In addition, Lee et al. (85) also immobilized proteinticles in a hydrogel, which resulted in a highly sensitive 3D bioassay.
Conclusion and perspectives
The rapid evolution in nano- and microtechnology enables the development of clever designed microdevices for various purposes, including point-of-care diagnostics. In POC devices as well as in lateral flow tests, nanobeads are already integrated and widely used. New applications are under development and will result in exciting capabilities such as measurements in attomolar concentration ranges as illustrated by the discussed work (83). The question relevant in this context is how these technological advances can be translated to clinical applications in the pressing field of infectious diseases. It will be tried to answer this, using selected infectious diseases and the currently used diagnostics as well as the challenges and pitfalls for (new) nanodiagnostics in a consecutive review.
This review has been written to contribute to the DiscoGnosis project that has the core objective to develop a platform that would allow the detection of malaria and similar pathogenic diseases in a rapid, multiplexed and non-invasive way (www.discognosis.eu). This project is supported by the European Commission through the 7th Framework Programme on Research and Technological Development within the Objective FP7 ICT-2011.3.2 and under Grant Agreement No. 318408.
Diamandis PE, Christopoulos TK. Immunoassay. Academic Press. ISBN-10: 0122147308, ISBN-13: 978-0122147302.Google Scholar
Wong R, Tse H. Lateral Flow Immunoassay. USA, NY: Humana Press, ISBN: 978-1-58829-908-6.Google Scholar
Mak WC, Sin KK, Chan CP, Wong LW, Renneberg R. Biofunctionalized indigo-nanoparticles as biolabels for the generation of precipitated visible signal in immunodipsticks. Biosensors Bioelectronics 2011;26:3148–53.CrossrefPubMedGoogle Scholar
Mashayekhi F, Le AM, Nafisi PM, Wu BM, Kamei DT. Enhancing the lateral-flow immunoassay for detection of proteins using an aqueous two-phase micellar system. Anal Bioanal Chem 2012;404:2057–66.Google Scholar
Puertas S, Moros M, Fernández-Pacheco R, Ibarra MR, Grazú V, de la Fuente JM. Designing novel nano-immunoassays: antibody orientation versus sensitivity. J Phys D: Appl Phys 2010;43:474012.CrossrefGoogle Scholar
Salieb-Beugelaar GB, Teapal J, van Nieuwkasteele J, Wijnperlé D, Tegenfeldt JO, Lisdat F, et al. Field-dependent DNA mobility in 20 nm High nanoslits. Nano Lett 2008;8:1785–90.Google Scholar
Xu ZR, Zhong CH, Guan YX, Chen XW, Wang JH, Fang ZL. A microfluidic flow injection system for DNA assay with fluids driven by an on-chip integrated pump based on capillary and evaporation effects. Lab Chip 2008;8:1658–63.CrossrefPubMedGoogle Scholar
Sato T, Kawai K, Kanai M, Shoji S. Development of an all-fluoroplastic microfluidic device applied as a nanoliter sample injector Jpn. J Appl Phys 2009;48:06FJ03.Google Scholar
Scott DE, Grigsby RJ, Lunt SM. Microdialysis sampling coupled to microchip electrophoresis with integrated amperometric detection on an all-glass substrate. ChemPhysChem 2013;14:2288–94.PubMedCrossrefGoogle Scholar
Yang YN, Hsiun SK, Lee GB. A pneumatic micropump incorporated with a normally closed valve capable of generating a high pumping rate and a high back pressure. Microfluid Nanofluid 2009;6:823–33.CrossrefGoogle Scholar
Godino N, Comaskey E, Gorkin III R, Ducré J. Centrifugally enhanced paper microfluidics. 978-1-4673-0325-5/12, MEMS 2012, Paris, FRANCE, 29 January–2 February 2012.Google Scholar
Grünberger A, Paczia N, Probst C, Schendzielorz G, Eggeling L, Noack S, et al. A disposable picolitre bioreactor for cultivation and investigation of industrially relevant bacteria on the single cell level. Lab Chip 2012;12:2060–8.CrossrefPubMedGoogle Scholar
Freifeld AG, Simonsen KA, Booth CS, Zhao X, Whitneu SE, Karre T, et al. A new rapid method for clostridium difficile DNA extraction and detection in stool. Toward point-of-care diagnostic testing. Journal of Molecular Diagnostics 2012;14:274–9.CrossrefGoogle Scholar
Wolf M, Juncker D, Michel B, Hunziker P, Delamarche E. Simultaneous detection of C-reactive protein and other cardiac markers in human plasma using micromosaic immunoassays and self-regulating microfluidic networks. Biosens Bioelectron 2004;19:1193–202.CrossrefGoogle Scholar
Helmerhorst E, Chandler DJ, Nussio M, Mamotte CD. Real-time and label-free bio-sensing of molecular interactions by surface plasmon resonance: a laboratory medicine perspective. Clin Biochem Rev 2012;33:161–73.PubMedGoogle Scholar
Zimmermann M, Delamarche E, Wolf M, Hunziker P. Modeling and optimization of high-sensitivity, low-volume microfluidic-based surface immunoassays. Biomed Microdevices 2005;7:99–110.CrossrefGoogle Scholar
Yang Z, Zong C, Ju H, Yan F. Streptavidin-functionalized capillary immune microreactor for highly efficient chemiluminescent immunoassay. Anal Chim Acta 2011;706:143–8.Google Scholar
White RJ, Kallewaard HM, Hsieh W, Patterson AS, Kasehagen JB, Cash KJ, et al. A Wash-free, electrochemical platform for the quantitative, multiplexed detection of specific antibodies. Anal Chem 2012; 84:1098–103.CrossrefGoogle Scholar
Lafleur L, Stevens D, McKenzie K, Ramachandran S, Spicar-Mihalic P, Singhal M, et al. Progress toward multiplexed sample-to-result detection in low resource settings using microfluidic immunoassay cards. Lab Chip 2012;12:1119–27.CrossrefGoogle Scholar
Yu X, Xia HS, Sun ZD, Lin Y, Wang K, Yu J, et al. On-chip dual detection of cancer biomarkers directly in serum based on self-assembled magnetic bead patterns and quantum dots. Biosens Bioelectron 2013;41:129–36.CrossrefGoogle Scholar
Zheng C, Wang X, Lu Y, Liu Y. Rapid detection of fish major allergen parvalbumin using superparamagnetic nanoparticle-based lateral flow immunoassay. Food Control 2012;26:446–52.CrossrefGoogle Scholar
Serrate D, De Teresa JM, Marquina C, Marzo J, Saurel D, Cardoso FA, et al. Quantitative biomolecular sensing station based on magnetoresistive patterned arrays. Biosens Bioelectron 2012;35:206–12.CrossrefGoogle Scholar
Marquina C, de Teresa JM, Serrate D, Marzo J, Cardoso FA, Saurel D, et al. GMR sensors and magnetic nanoparticles for immuno-chromatographic assays. Biosens Bioelectron. J Mag and Mag Mat 2012;324:3495–8.Google Scholar
Cardoso FA, Germano J, Ferreira R, Cardoso S, Martins VC, Freitas PP, et al. Detection of 130nm magnetic particles by a portable electronic platform using spin valve and magnetic tunnel junction sensors. J App Phys 2008;103:07A310.Google Scholar
Juntunen E, Myyryläinen T, Salminen T, Soukka T, Pettersson K. Performance of fluorescent europium(III) nanoparticles and colloidal gold reporters in lateral flow bioaffinity assay. Anal Biochem 2012;428:31–8.Google Scholar
Zong S, Wang Z, Zhang R, Wang C, Xu S, Cui Y. A multiplex and straightforward aqueous phase immunoassay protocol through the combination of SERS-fluorescence dual mode nanoprobes and magnetic nanobeads. Biosens Bioelectron 2013;41:745–51.CrossrefGoogle Scholar
Noguera P, Posthuma-Trumpie GA, van Tuil M, van der Wal FJ, de Boer A, Moers AP, et al. Carbon nanoparticles in lateral flow methods to detect genes encoding virulence factors of Shiga toxin-producing Escherichia coli. Anal Bioanal Chem 2011;399:831–8.Google Scholar
Posthuma-Trumpie GA, Wichers JH, Koets M, Berendsen LB, van Amerongen A. Amorphous carbon nanoparticles: a versatile label for rapid diagnostic (immuno)assays. Anal Bioanal Chem 2012;402:593–600.CrossrefGoogle Scholar
Goryacheva IY, Lenain P, de Saeger S. Nanosized labels for rapid immunotests. Trends Anal Chem 2013;46:30–43.Google Scholar
Cheng MS, Ho JS, Tan CH, Wong JP, Ng LC, Toh CS. Development of an electrochemical membrane-based nanobiosensor for ultrasensitive detection of dengue virus. Anal Chim Acta 2012;725:74–80.Google Scholar
Nguyen BT, Peh AE, Chee CY, Fink K, Chow VT, Ng MM, et al. Electrochemical impedance spectroscopy characterization of nanoporous alumina dengue virus biosensor. ioelectrochemistry 2012;88:15–21.CrossrefGoogle Scholar
Cui Y, Wei Q, Park H, Lieber CM. Nanowire nanosensors for highly sensitive and selective detection of biological and chemical species. Science 2001;293:1289–92.Google Scholar
Abonnenc M, Manaresi N, Borgatti M, Medoro G, Fabbri E, Romani A, et al. Programmable interactions of functionalized single bioparticles in a dielectrophoresis-based microarray chip. Anal Chem 2013;85:8219–24.CrossrefGoogle Scholar
Hansen KM, Ji HF, Wu G, Datar R, Cote R, Majumdar A, et al. Cantilever-based optical deflection assay for discrimination of dna single-nucleotide mismatches. Anal Chem 2001;73: 1567–71.CrossrefGoogle Scholar
Schmid D, Lang H, Marsch S, Gerber C, Hunziker P. Clinical studies: diagnosing disease by nanomechanical olfactory sensors–system design and clinical validation. Eur J Nanomed 2008;1:44–7.Google Scholar
Sukhanova A, Even-Desrumeaux K, Kisserli A, Tabary T, Reveil B, Millot JM, et al. Oriented conjugates of single-domain antibodies and quantum dots: toward a new generation of ultrasmall diagnostic nanoprobes. Nanomedicine 2012;8: 516–25.CrossrefGoogle Scholar
Wesolowski J, Alzogaray V, Reyelt J, Unger M, Juarez K, Urrutia M, et al. Single domain antibodies: promising experimental and therapeutic tools in infection and immunity. Med Microbiol Immunol 2009;198:157–74.CrossrefGoogle Scholar
Kim HJ, McCoy MR, Majkova Z, Dechant JE, Gee SJ, Tabares-da Rosa S, et al. Isolation of alpaca anti-hapten heavy chain single domain antibodies for development of sensitive immunoassay. Anal Chem 2012;84:1165–71.CrossrefGoogle Scholar
Han JH, Kim HJ, Sudheendra L, Gee SJ, Hammock BD, Kennedy IM. Photonic crystal lab-on-a chip for detecting staphylococcal enterotoxin b at low attomolar concentration. Anal Chem 2013;85:3104–9.CrossrefGoogle Scholar
Maraldo D, Mutharasan R. Detection and confirmation of staphylococcal enterotoxin b in apple juice and milk using piezoelectric-excited millimeter-sized cantilever sensors at 2.5 fg/mL. Anal Chem 2007; 79:7636–43.Google Scholar
Lee JH, Seo HS, Song JA, Kwon KC, Lee EJ, Kim HJ, et al. Proteinticle engineering for accurate 3D diagnosis. ACS Nano 2013;7:10879–86.Google Scholar
About the article
Georgette B. Salieb-Beugelaar
Georgette B. Salieb-Beugelaar began her professional life in the fields of clinical genetics studying DNA research and diagnostics at the Academic Medical Centre in Amsterdam (The Netherlands) in 1996. She studied in parallel Chemistry at the University of Utrecht (The Netherlands) between 2000 and 2003. In 2005, Georgette changed her professional field into the microfluidic and nanofluidic world at the University of Twente in Enschede (The Netherlands) and investigated single DNA molecules in nanoconfined environments on chips. For this she received my PhD in 2009, and continued the next 2 years at the Korean Institute of Science and Technology (Saarbrücken, Germany) with nanodropled pseudocrystals in microfluidic chips. Meanwhile, she was also working for the Mesa Institute for Nanotechnology (Twente University), which was then setting up their BioNano Laboratory. Since November 2012, Georgette has been a member of the multidisciplinary NanoMedicine Group of Prof. Patrick Hunziker, and works for the DiscoGnosis project (www.discognosis.eu). She is involved in this journal as an assistant communicating and managing editor.
Patrick R. Hunziker
Patrick Hunziker studied Medicine at the University of Zurich, Switzerland. He received a doctoral degree based on thesis work in Experimental Immunology from the University of Zurich and did further research in Experimental Hematology at University Hospital in Zurich, Switzerland. He earned specialist degrees in Internal Medicine, Cardiology and Intensive Care Medicine. As a fellow of the Massachusetts General Hospital, Harvard Medical School, he worked on cardiac imaging in a joint project with the Massachusetts Institute of Technology, Cambridge. His professional activities in Europe, the USA, Africa and China gave him a broad insight into the needs for the medicine of the future in a variety of settings. Patrick Hunziker became involved in medical applications of nanoscience in the late 1990s and has been the pioneer physician in nanomedicine in Switzerland since then. With improved prevention, diagnosis and cure of cardiovascular disease as his main research topic, he worked in the nanoscience fields of atomic force microscopy, nano-optics, micro/nanofluidics, nanomechanical sensors and polymer nanocarriers for targeting. He is the founding president of the European Society of Nanomedicine, cofounder of the European Foundation for Clinical Nanomedicine and coinitiator of the European Conference for Clinical Nanomedicine and is clinically active as deputy head of the Clinic for Intensive Care Medicine at the University Hospital Basel, Switzerland. In November 2008 he became Professor for Cardiology and Intensive Care Medicine at the University of Basel.
Published Online: 2014-03-06
Published in Print: 2014-03-01