Review
For reprint orders, please contact [email protected]
Systems for multiplexing homogeneous immunoassays
High-throughput multiplex protein biomarker assays continue to gain significance in the fields of biomarker discovery and drug development, due to their economical use of not only the precious clinical biological samples but also expensive reagents. Among these platforms, homogeneous multiplex systems have potential for short assay run times and cost-effective reagent consumptions. However, these systems must overcome challenges of signal cross talk and biochemical cross-reactivity. Despite these obstacles, several homogeneous multiplex immunoassays have been demonstrated. These include fluorescent polarization, fluorescent resonance energy transfer with quantum dots or graphene, luminescent oxygen-channeling immunoassay coupled with aqueous two-phase systems and DNA proximity assays. The balance between speed/simplicity and high multiplexing and robustness of these homogeneous multiplex immunoassays are discussed in this review.
Introduction to multiplex homogeneous immunoassays The NIH defined protein biomarkers as ‘key molecular or cellular events that link a specific environmental exposure to a health outcome.’ Understanding the interactions of multiple protein biomarkers of disease will play an integral role in drug screening, in biological pathway analysis and in disease diagnosis [1] . Single-analyte detection has provided high accuracy results, but in a limited scope. Multiplex detection methods, measurement and analysis of multiple biomarkers together, provide a view of the entire system. Current protein biomarker discovery techniques often use mass spectrometry on a few patient samples for screening large numbers of proteins without the need for preconceived protein targets [2] . At the other extreme, qualification of biomarker targets are often performed using relatively lowthroughput sandwich immunoassays, due to their high sensitivity and specificity [3] . In clinical practice, immunoassays are also preferred over mass spectroscopy due to lower cost and shorter time to result for single target detection. This process of approving
10.4155/BIO.15.78 © 2015 Future Science Ltd
biomarkers or biomarker panels from discovery to clinical practice presents a great challenge; the translation of discovered targets to clinical practice yields fewer than two new protein biomarkers approved by the US FDA per year [4,5] . Multiplex immunoassays could help bridge the gap between protein biomarker discovery and clinical use. Due to their high sensitivity, specificity and procedural convenience, this article will focus on homogeneous sandwich immunoassays. Sandwich immunoassays come in two formats: heterogeneous and homogeneous. The standard ELISA is heterogeneous, with capture antibodies (cAb) immobilized onto a microtiter plate. The addition of detection antibodies (dAb) forms the antibody–antigen–antibody sandwich complex [6] . The requirement for both the cAbs and dAbs to bind to the target reduces noise from nonspecific binding [7] , making the sandwich format more specific than label-free methods [8–15] , which only use one antibody. Unlike heterogeneous immunoassays, homogeneous sandwich immunoassays use both antibodies in solution, typically conjugated to
Bioanalysis (2015) 7(12), 1545–1556
Cameron D Yamanishi1,2, Joyce Han-Ching Chiu1,2 & Shuichi Takayama*,1,2,3,4 1 Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI 48109, USA 2 Biointerfaces Institute, University of Michigan, Ann Arbor, MI 48109, USA 3 Department of Macromolecular Science & Engineering, University of Michigan, Ann Arbor, MI 48109, USA 4 Michigan Center for Integrative Research in Critical Care, University of Michigan, Ann Arbor, MI 48109, USA *Author for correspondence: Tel.: +1 734 615 5539 [email protected]
part of
ISSN 1757-6180
1545
Review Yamanishi, Han-Ching Chiu & Takayama
Key terms Protein biomarker: Proteins in the blood whose concentration may be used to diagnose disease state. Multiplex detection: Simultaneous measurement of multiple targets in the same solution. Homogeneous sandwich immunoassay: Technique using a soluble antibody pair to measure concentration of proteins in solution. Fluorescence polarization immunoassay: Assay based on stabilization of molecular motions by antigen–antibody binding that will reduce the depolarization of excitation of fluorescent tags, which are excited by polarized light sources. Fluorescence resonance energy transfer: Measures the fluorescence signals by energy transfer from donor to acceptor fluorophore. Quantum dots: Nanocrystals with tunable fluorescent properties.
an identifiable (encoded) marker. Binding in solution, rather than at a surface, drastically reduces the distance that antigens must diffuse to reach the antibodies. Combined with the elimination of washing steps, this reduces assay time, making homogeneous assays more suitable for biosensors and assays used in field applications. Additionally, the absence of washing steps allows antibodies with relatively low affinity to be used, because unbinding events resulting from washing steps will be reduced, thereby retaining more antigens. The soluble format can also provide a more native reaction environment for the antibodies than the surface-bound format, as proteins have been shown to undergo conformational changes when they adsorb to nonnative surfaces [16] . Most homogenous sandwich immunoassays to date, however, have a major limitation. Because both capture and detection antibody reagents are freely diffusing in solution, it has been difficult to perform spatially segregated reagent multiplexing when performing homogeneous immunoassays. Additionally, multiplex immunoassays suffer from cross talk, the phenomenon resulting from one labeling molecule, such as a fluorescent marker, generating an overlapping signal that appears as a false positive for another labeling molecule. With higher number of antibodies and antigens used in multiplex assays, the chances of cross-reactivity between molecules increases. The goal of this article is to review the latest methods and techniques to overcome this limitation to enable multiplex homogenous sandwich immunoassays. Bead-based systems fit in between heterogeneous and homogeneous. These assays replace the planar surface of standard ELISA with microbeads, encoded with identifiers. Bead assays still have faster mix-
1546
Bioanalysis (2015) 7(12)
ing than standard ELISA, because diffusion time scales as: 2
T .
L D
Equation 1
where L is the characteristic length and D is the diffusivity; the characteristic length between particles in solution is much smaller than the length between the bulk solution and the surface. However, large beads diffuse more slowly than free antibodies. The Stokes– Einstein relation describes the diffusivity as inversely proportional to radius at constant temperature and viscosity. IgG antibodies have radius ~5 nm [17] . Adding a bead 1 μm in diameter will therefore reduce the diffusivity by 100-fold. For the purpose of this review, we strictly define homogeneous immunoassays as those containing all antibodies and antigens in solution, with no washing between antibody–antigen binding steps. This definition will exclude pseudohomogeneous bead-based assay formats, such as the Luminex xMAP system, which employs suspension microbeads as a solid-phase replacement for planar surfaces (bead sources). The Luminex system retains some of the time-to-result advantages of homogeneous systems, but the wash step limits the effectiveness of low affinity antibodies used in assays. This is not a problem for biomarkers with well-established antibody libraries, but finding high affinity antibodies for obscure or recently discovered proteins can be difficult. Our selection criteria also exclude microfluidic platforms despite their capacity for automation of washing steps or their shared advantages with homogeneous sandwich immunoassays, including reduced diffusion distances and small sample volume consumption [18–21] . In this report, we primarily review techniques for multiplexing homogeneous sandwich immunoassays, highlighting the merits and demerits of each, as summarized in Table 1. Fluorescent polarization immunoassay (FPIA) is not a sandwich assay, as it employs only one antibody, but it is included in this scope because it is an established multiplex homogeneous assay, demonstrated with two-plex [22] . Fluorescence resonance energy transfer (FRET) assay employs the photon energy transfer from the donor to the acceptor in a sandwich system to detect antibody and antigen binding reaction for up to two-plex assay [23] . It has been adapted to incorporate graphene and quantum dots to overcome the limitation of distance dependent behavior of an energy transfer system. The luminescent oxygenchanneling immunoassay (LOCI) uses a different type of energy transfer via singlet oxygen to supply the energy necessary for the acceptor chemiluminescent label for two-plex system, and higher order or multiplexing can
future science group
Systems for multiplexing homogeneous immunoassays
Review
Table 1. Multiplex homogeneous immunoassay parameters. Assay format
LLOQ
ULOQ
Advantages
Limitations
Highest reported multiplexing
Ref.
FPIA
360 pg/ml
~120 ng/ml
Rapid (40 ng/ml
Rapid (1.5 h)
Susceptible to false positives from cross-reactivity
4
[24,29]
DNA proximity
9.6 pg/ml
~120 ng/ml
High multiplexing, specific, sensitive
Complex readout, time-to-result of 1 day
96
[25,30]
FPIA: Fluorescence polarization immunoassay; FRET: Fluorescence resonance energy transfer; LOCI: Luminescent oxygen-channeling immunoassay.
be achieved by the use of aqueous two-phase systems (ATPS) [24] . DNA proximity assays use DNA oligomers as reporter molecules, employing their specificity and existing amplification techniques for effective multiplexing. These techniques have been demonstrated up to 96-plex [25] . To compare assay formats, tests should ideally be performed within the same lab and for the same targets, as antibodies and user skill can drastically alter assay results [26] . However, cross-platform comparisons between the homogeneous immunoassay techniques covered in this review are currently sparse, both for single target and multiplex formats. Nevertheless, comparisons of assay range between different labs can still be useful. To compare formats, we selected cTnI, as it is already an established biomarker of cardiac arrest. As summarized in Table 1, the quantitative ranges of FPIA [27] , FRET [28] and LOCI [29] have all been reported to be similar, while the DNA proximity assay [30] has been published with a comparatively lower limit of detection. However, it should be emphasized that these parameters are application and target specific, and the quantitative ranges may be tuned to fit the application needs. The techniques covered in this review have had only limited use in clinical trials. Between multiplex versions of FPIA, FRET and LOCI, no clinical trials have been reported in WHO participant countries. However, the single-plex versions of these assays are used in clinical diagnosis. Only two clinical trials have been reported using the multiplex DNA proximity immunoassays. We will discuss the potential and current obstacles to implement these multiplex techniques in clinical settings. Homogeneous multiplex sandwich immunoassay formats
polarized plane as the light source. Conversely, light from a rotating fluorescent molecule will be emitted in a rotated plane. In a free floating solution, large molecules experience less rotation and motion during excitation, and will therefore give off a smaller depolarization, whereas small molecules will have larger depolarization (Figure 1) . As fluorescent target molecules bind with the antibody, the difference in depolarization can be measured. The immunoassay uses antibodies alongside fluorescently labeled antigen, a.k.a. tracer. When the tracer binds with antibodies, the whole fluorophore complex is stabilized, thereby reducing motion and allowing for greater polarized emission [31–33] . When test samples such as patient plasma or serum are introduced to the system, the patient’s target antigen will compete with tracer molecules for the antibodies. The technique is by nature homogenous and compatible with automated detection for high-throughput analysis. However, it is limited to small-molecule applications because more dramatic changes in depolarization may be observed with small molecules. Multiplex FPIA can be achieved by utilizing multicolor quantum dots with narrow emission bandwidths that can be cleanly resolved. This format uses the same principle as single-plex FPIA, wherein the antibody stabilizes the quantum dot to polarize the light. Specifically, a two-plex assay was designed with red and green quantum dots conjugated with tumor biomarkers. The target protein can be identified by the color of the quantum dot, as shown in Figure 2. In this method, centrifugation with ultracentrifuge filters of serum samples prior to the assay was necessary to remove matrix proteins to reduce matrix interference. The results of this multiplex assay were shown to have clinical diagnostic value comparable to commercial ELISA [22] .
Fluorescence polarization immunoassay
FPIA is a form of competitive immunoassay. When incident light excites a fluorescently tagged stationary molecule, the resulting emission will be on the same
future science group
Fluorescence resonance energy transfer
FRET-based immunoassays utilize a donor–acceptor system, as illustrated in Figure 3. When a laser excites
www.future-science.com
1547
Review Yamanishi, Han-Ching Chiu & Takayama
A
Slow rotation High polarization
Fast rotation Low polarization
Fluorescent label
Fluorescent label
Antigen
Antigen Antibody
B
Add sample
Figure 1. Fluorescence polarization immunoassay. (A) Polarization intensity in relation to molecule complex size and rotation speed. (B) Schematic representation of multiplex fluorescence polarization. (A) Adapted with permission from [33] © Springer (2008). (B) Adapted with permission from [22] © Elsevier (2012).
the donor fluorophores, the resultant fluorescent energy is transferred from the donor fluorophore to the nearby acceptor molecule, which will subsequently release photons of a different wavelength for detection. The donor and acceptor molecules are both attached to antibodies specific to the target protein. If the target is present, the antibodies can form sandwich complexes with the target, keeping their respective donor and acceptor tags in close proximity. Because the efficiency of energy transfer decreases rapidly as the distance between the donor and acceptor pair increases, the fluorescent signal corresponds to the presence of the target. However, this distance dependence also limits the sensitivity of sandwich immunoassays, because the distance across the antibodies and antigens in a sandwich complex is large enough to diminish FRET efficiency [34] . Considerable progress has been made to improve FRET efficiency. The important parameters for FRET efficiency are the quantum yield, extinction coefficients
1548
Bioanalysis (2015) 7(12)
and spectral overlap [35] . The quantum yield and extinction coefficients of both the donor and the acceptor should be high to maximize efficiency. Using genetic engineering of fluorescent proteins, donor–acceptor pairs can be tuned with large overlap between donor emission and acceptor absorption spectra to increase energy transfer efficiency. On the other hand, undesirable overlap between the donor emission and the acceptor emission spectra can be minimized to avoid optical cross talk. Although FRET-based assays are traditionally built with fluorescent dyes, current research on FRET-based immunoassays is increasingly performed with nanostructures made up of nanorods, nanoshells and quantum dots. These nanostructures make use of the surface plasmon resonance effect to increase FRET signal for higher detection sensitivities. Zeng et al. engineered a quantum dot and gold nanorod two-plex assay to maximize FRET yield [23] . Gold nanorods have two distinct plasmon resonances, corresponding to their two dimensions,
future science group
Systems for multiplexing homogeneous immunoassays
A
B
Review
Acceptor fluorophore
Relaxation
S1
FRET
Fluorescence
Fluorescence
Excitation
S1
FRET
S0 Acceptor
S0 Donor
Donor fluorophore
Figure 2. Fluorescence resonance energy transfer. (A) Resonance energy transfer between donor and acceptor. (B) FRET applied to immunochemistry. FRET: Fluorescence resonance energy transfer. (B) Adapted with permission from [35] © Nature Publishing Group (2012).
longitudinal and transverse. Because both dimensions are tunable, gold nanorods are ideal acceptors for multiplex assays. Two different sizes of quantum dots (CdTe/ CdS core/shell) with two different emission wavelengths were tailored to maximize the overlap of quantum dot donor emission and self-assembled gold nanorods acceptor absorption spectra. This system exhibited high
specificity between the two targets, indicating the utility of the FRET pairs in the multiplex system. Fluorescent quantum dots-graphene immunoassay
Graphene quenching provides an alternative to FRET. Graphene sheets efficiently quench fluorescence from
Figure 3. Quantum dots and graphene sheets sensor. This platform utilizes the fluorescence quenching ability of the graphene sheets to overcome distance-dependent limitation of energy transfer techniques. Adapted with permission from [36] © Royal Society of Chemistry (2010).
future science group
www.future-science.com
1549
Review Yamanishi, Han-Ching Chiu & Takayama quantum dots when the two are in close proximity. Therefore, in constrast to FRET, a decrease in fluorescence corresponds to presence of targets in the sample. The primary disadvantage of assays based on energy transfer between quantum dots and nanostructure systems lies in the required donor and acceptor dipole– dipole distance, which is limited to 10–100 Å [34] . However, graphene sheets are shown to overcome this restriction [36] . Compared with a quantum dot-quantum dot pair, the 2D graphene sheets have more possible spatial arrangements in which parts of the sheet are close enough to the spherical quantum dots to cause quenching, as seen in Figure 4. This platform increased the effective distance between the quantum dot anchor and the graphene anchor from 100 Å up to 223 Å. This distance increase enables greater dynamic range in the sandwich assay formats that use the antigen as a bridge to bring the detection Ab-coated quantum dot close to the capture Ab-coated graphene surface.
complete the complex. The presence of the target protein thus determines how many donor–acceptor pairs are in close proximity, where they can produce the chemiluminescent signal. The LOCI can perform twocolor multiplex immunoassays using a second acceptor bead that is also excited by singlet oxygen and emits at 545 nm. The second donor bead is functionalized with the appropriate antibodies, but contains the same light-sensitive compound as the first donor bead. The LOCI format is fairly convenient for bench-top research. LOCI assays can be read using a microplate reader equipped with the proper fluorescent channels. For these readers, high laser power is recommended for optimal signal. Additionally, the LOCI format has intrinsic advantages for accuracy. Because LOCI employs a shift from a high wavelength excitation to a lower wavelength emission, it eliminates noise from autofluorescence. Compared with the other optical detection assays, the LOCI assay has a relatively wide dynamic range and low lower limit of quantification for the detection of cTnI, as shown in Table 1. Furthermore, the single-plex version of LOCI is available from Siemens for clinical diagnostics. Although the commercially available PerkinElmer system is currently limited to two-plex using chemiluminescence at two different emission wavelengths, higher multiplexing has been demonstrated by spatially arraying reagent solutions using aqueous two-phase system (ATPS) [24] . When two immiscible polymers are mixed in an aqueous solution at appropriate concentrations, they cause separation of the resulting aqueous solution into two phases. These systems are mild to biomolecules, making them appropriate for use in immunoassays. In a polyethylene glycol – DEX ATPS, antibodies and beads partition preferentially into the DEX phase, allowing spatial
Luminescent oxygen-channeling immunoassay
The LOCI, commercially available as AlphaLISA, from PerkinElmer, uses donor and acceptor beads (250–350 nm diameter) [37] . When the donor beads are excited by 680 nm light, they release reactive singlet oxygen. This activated singlet oxygen decays back to its ground state quickly in solution. However, if acceptor beads are nearby, the singlet oxygen molecules can transfer their energy to the chemiluminescent compound in the acceptor beads, triggering the acceptor beads to emit light at 615 nm [37] , as depicted in Figure 5 [38] . In a sandwich assay format, capture antibodies attached to acceptor beads bind to target proteins in solution. Biotinylated detection antibodies sandwich the target proteins. Streptavidin-coated donor beads then bind to the detection antibodies to
O2
1
Excitation 680 nm
Emission 615 nm
B Streptavidin-coated donor bead
Biotinylated anti-analyte
Analyte
Anti-analyte-conjugated acceptor bead
Figure 4. Luminescent oxygen-channeling immunoassay. In the luminescent oxygen-channeling immunoassay format, donor beads are excited by 680 nm light and release singlet oxygen. Acceptor beads in the vicinity emit 615 nm light in response to the singlet oxygen. Adapted under the Creative Commons Attribution license from [38] .
1550
Bioanalysis (2015) 7(12)
future science group
Systems for multiplexing homogeneous immunoassays
Conventional 1-color LOCI
Review
ATPS LOCI
Antibodies and beads bind to antigens
DE
PEG
PBS
X
Antibody–bead cross-reactions
Donor beads are excited
DE
PEG
PBS
X
No spatial separation
Acceptor beads emit signal
Figure 5. Aqueous two-phase system ELISA. ATPS ELISA co-localizes donor–acceptor pairs within dextran microdroplets, enabling spatially distinct multiplex detection with a single bead color. ATPS: Aqueous two-phase system; DEX: Dextran; LOCI: Luminescent oxygen-channeling immunoassay; PBS: Phosphate buffer serum; PEG: Polyethylene glycol. Adapted with permission from [24] © WorldScientific (2014).
confinement of distinct AlphaLISA bead pairs preallocated into individual phase-separated droplets within a single well, as demonstrated in Figure 6. The ATPS reagent localization strategy marks a promising development for multiplex assays. The technique is theoretically compatible with other homogeneous immunoassay formats that have not yet been multiplexed, including simple bead agglutination. Additionally, the spatial localization eliminates the possibility of cross talk between unmatched antibody pairs. Therefore, assay validation does not require the extensive combinatorial screening that many other multiplex techniques must undergo. However, the ATPS method has some drawbacks. The polymers involved are viscous, lowering the diffusivity of reagents and requiring slightly longer assay time. This method also requires specialty plates with topographical features within each well to position the reagent microdroplets from moving within the well. The published four-plex AlphaLISA assay plate was designed to be compatible with conventional 1536 well format AlphaLISA readers. DNA proximity
Proximity multiplex formats use DNA oligomers to encode antibodies. Taking advantage of existing DNA
future science group
sequencing techniques, the proximity extension assay (PEA) [25] uses antibody pairs attached to DNA oligonucleotides. As shown in when the antibodies sandwich the target protein, the DNA oligonucleotides come into close proximity, where the DNA strand from one antibody can hybridize with its complementary DNA strand on the partner antibody. This hybridization allows a DNA polymerase to extend the DNA at both ends. The resulting oligonucleotides can be subsequently amplified by qPCR. Only the completed DNA strands, i.e., those which had hybridized and been extended by the DNA polymerase, can be amplified by PCR. Thus, DNA detected by qPCR reflects the presence of the target protein. In multiplex PEA, distinct DNA sequences are used to identify each antibody pair. The PEA technology is available from Olink, and has been demonstrated in a 96-plex assay [25] . Key terms Singlet oxygen: Electronically excited oxygen. Aqueous two-phase systems: Aqueous two-phase systems comprising two immiscible polymers in water. Extension: Technique to replicate partially overlapping complementary DNA.
www.future-science.com
1551
Review Yamanishi, Han-Ching Chiu & Takayama
A
Incubation
B
Hybridization extension
C
Preamplification
D
E
Digestion
Microfluidic qPCR
Protein Expression (ddCq)
12 10 8 6 4 2 0
Figure 6. Proximity extension assay. (A) The proximity extension assay uses antibodies with oligonucleotides in a sandwich format. (B) DNA strands are hybridized with nearby complementary strands. (C–E) Through qPCR, DNA hybridization can be quantified. qPCR: Quantitative PCR. Reprinted under the Creative Commons Attribution license from [25] .
In a similar format to PEA, the proximity ligation assay (PLA) [39,40] also uses DNA oligomers to encode antibody pairs. Whereas the DNA oligomers in PEA are complementary to each other, the DNA oligomers in PLA do not hybridize with each other. Rather, a third DNA oligomer, complementary to both antibody-bound oligomers, is added to the solution to bridge the DNA on the antibody pair. The two antibody-bound oligomers may then be connected end-to-end. To increase the specificity of the assay, one study attached an antibody to the third DNA oligomer, making the assay require three matching epitopes to achieve a positive result [41] . Proximity multiplex immunoassays do not eliminate cross-reactions but avoid reading out such cross-reactions because only matching antibody pairs’ complementary oligonucleotide labels can hybridize to create signal. Although Olink offers a 96-plex cardiovascular disease panel, cTnI is not included. However, the PLA has been demonstrated to have a lower limit of quantification of 9.6 pg/ml for cTnI. The advantages of low limit of detection and high multiplexing in DNA proximity assays come at the cost of increased assay time and complexity. The 96-plex assay required an overnight sample incubation, followed by DNA hybridization and the qPCR steps. The Olink company has addressed many of the added steps by incorporating their qPCR process into the Fluidigm-automated microfluidic PCR reader, the BioMark HD. To date, PLA and PEA have been employed in one clinical trial each, as recorded by WHO countries. Multiplex PLA is currently being used in a clinical trial from National Taiwan University Hospital to quantify biomarkers that can aid in characterization of hepatocellular carcinoma and esophageal cancer. The group, led by J Chia-Hsien Cheng, aims to determine tumor response, prognosis and radiation-induced organ toxicity in the lung and liver by assaying 56 proteins in serum samples of hundreds of patients before, during and after radiation therapy [42] . The PEA clinical trial employed multiplex biomarker analysis as an evaluative tool,
1552
Bioanalysis (2015) 7(12)
rather than as a new disease diagnosis tool. A Miclescu’s group at Uppsala University Hospital tested ten patients in a drug trial to evaluate methylene blue as a treatment for neuropathic pain. This study monitored patient outcomes both by qualitative pain reports from the patients and by quantitative measurement of plasma and urine concentrations of a panel of 92 protein biomarkers [43] . Assay characterization Before implementation, new immunoassays must be characterized for their analytical performance, then for clinical validation [44] . The degree and type of characterization necessary depends on the desired application. Lee et al. outline some of these differences for pharmacokinetic studies, drug development biomarker assays and diagnostic biomarker assays. They also provide guidelines according to the developmental stage of the assay [45] . Such is true for single-plex assays, but there are additional considerations and specific applications for multiplex assays. Systems such as FPIA, FRET and LOCI are suitable for time-sensitive point-of-care diagnostic applications, because they provide rapid results. However, DNA proximity assays currently generate higher multiplexing at the cost of simplicity and timeto-result. Therefore, the DNA proximity assays are more suitable for early biomarker candidate screening, pharmacokinetic studies and evaluation during drug development. For single-plex immunoassay characterization, several informative papers have been published [45–50] . Therefore, we will focus on considerations specific to homogeneous and multiplex immunoassays. Potential cross-reactions between both endogenous and exogenous antibodies and proteins may result in over- and under-estimation of protein concentrations and should be screened out. Polyclonal antibodies recognize multiple epitopes and can compensate for low affinities, but the high specificity of monoclonal antibodies are preferred for reduced cross-reactivity [51–54] . Previous reports have characterized multiplex immunoas-
future science group
Systems for multiplexing homogeneous immunoassays
says by comparison with corresponding single-plex assays [7] . At high numbers of multiplex targets, alternative comparisons such as batch sampling have been employed [25] . However, these batch sampling methods do not account for cross-reactivity, and are only sufficient for proof-of-concept studies. Another important consideration for immunoassays is matrix interference. Matrix interference refers to changes in assay results caused by interactions between assay components and moieties from the sample other than the intended target. This effect can be observed by dilutions of samples with different dilution factor and diluents. Investigation of minimum dilution factor is necessary to avoid quantification errors [49,55,56] . This is especially true in no-wash sandwich immunoassays, where the hook effect, or prozone phenomenon occurs [50] . As higher concentrations are measured, the signal will increase, plateau, then decrease in a hook shape. The decrease results from target proteins saturating both the detection antibodies and the capture antibodies, preventing formation of the sandwich complex. The hook effect may be addressed by diluting the sample into the monotonic range of the assay [57] . Alternatively, Weinstock et al. have demonstrated that addition of eythylenediaminetetraacetic acid (EDTA) to serum can reduce the hook effect, by inhibiting the activity of calcium-dependent matrix proteins, such as the complement system [58] . This reduces antibody quenching from matrix effects, thus maintaining more available antibody binding sites and increasing the monotonic range of the assay. Multiplex immunoassays are particularly susceptible to false positives arising from cross-reactivity [59] . Cross-reaction can occur in several different ways: antibody–antibody, antibody–antigen and antigen– antigen. In single-plex ELISAs, antibody pairs are carefully screened and selected, but the number of necessary tests rapidly increases with plexing. Juncker and coworkers performed a combinatorial analysis of possible cross-reacting pairs in a sandwich multiplex format [7] , and found that the number of liable pairs scales as: 4N(N-1) Equation 2
in an N-plex assay. Although these types of reactions may be rigorously screened and eliminated, the upper limit of screening at reasonable cost has been estimated at 50 [60] . To address cross-reactivity in planar arrays, ATPS have been employed to confine detection antibody solutions to the volume directly above their corresponding capture site [24] . Some patient samples contain human heterophilic antibodies, i.e., antibodies that bind to many of the animal IgGs used in immunoassays [61] . These
future science group
Review
heterophilic antibodies interfere with the immunoassay to cause nonlinearity. Classically, this is detected by measuring serial dilutions to check for linearity. However, protections from heterophilic antibody interference can be built into the assay. These measures include polyethylene glycol (PEG) precipitation [62] , specialized antibodies [63] and blocking additives [62] . Conclusion In this review, we described the principles and state-ofthe-art of multiplexing homogeneous sandwich immunoassays: methods of encoding and quantification for FRET, AlphaLISA and DNA proximity assays. We discussed the need for multiplexing to increase throughput of immunoassays for pharmacokinetics/pharmacodynamics, drug evaluation and disease diagnosis, as well as the advantages multiplexing provides in sample and reagent consumption. On the other hand, we also acknowledged some of the remaining challenges in the field of multiplex immunoassays: high-throughput validation, matrix interference, cross-reactivity and some of the workarounds to address these issues. The growing field of multiplex immunoassays offers many promising methods for high-throughput protein screening, and we expect multiplex immunoassays to assume a significant role in biology and healthcare. Future perspective Despite the advantages of multiplexed homogeneous immunoassays, they have seen little use in clinical trials. The authors speculate that this is a consequence of the limited multiplexing capabilities of these assays, until recently. While FPIA, FRET and LOCI have not yet progressed beyond two- to four-plex, the DNA proximity assays are commercially available up to 96-plex. Accordingly, those assays have been employed in clinical trials. In the near future (2015–2025), the authors anticipate expanded multiplexing capabilities for FPIA, FRET and LOCI. Considering the capabilities of quantum dots, FPIA and FRET could be quickly extended to five-plex or more. However, FPIA must overcome its current target size limitations before seeing more widespread use. With the quantum dot-graphene technique, Key terms Ligation: Technique to connect two DNA strands end-to-end. Matrix interference: Undesired, nonspecific interactions between assay components and biological material in the sample. Cross-reaction: Undesired interactions between assay components.
www.future-science.com
1553
Review Yamanishi, Han-Ching Chiu & Takayama FRET has greater flexibility. Although LOCI cannot borrow from quantum dot technology, discovery of novel chemiluminescent molecules that emit light at additional wavelengths in response to singlet oxygen could propel LOCI toward five- or ten-plex. Furthermore, each of these techniques is compatible with the ATPS spatial array technology, which could further extend the multiplexing capabilities of each assay. The rapid time-to-result and simplicity of these assays make them attractive tools for diagnostic biomarker panels. Meanwhile, DNA proximity techniques are poised for expanded use applications such as PK/PD and novel drug evaluation, which are not as time-sensitive.
The high multiplexing capabilities of the DNA proximity techniques increase their value for screening applications. Financial & competing interests disclosure S Takayama holds shares of PHASIQ, Inc, a company working on related technologies. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. No writing assistance was utilized in the production of this manuscript.
Executive summary Multiplex immunoassays • Multiplex protein detection is useful in drug screening, biological pathway analysis and diagnostic biomarker validation.
Advantages of homogeneous immunoassays • Homogeneous sandwich immunoassays are fast, no wash, specific, sensitive, robust to low affinity antibodies and often have wide dynamic range.
Multiplex homogeneous immunoassays • Fluorescence polarization immunoassay measures the change in intensity of polarized light from antigen stabilization by antibody. • Fluorescence resonance energy transfer uses a donor–acceptor system to only fluoresce when the antibody pair is in close proximity. • Quantum dots are used for their distinct optical properties in both fluorescence resonance energy transfer and fluorescence polarization immunoassay techniques. • The luminescent oxygen-channeling immunoassay uses a donor–acceptor system with singlet oxygen transfer to emit chemiluminescence when the pair is in close proximity. This technique was coupled with aqueous two-phase systems to spatially encode antibodies. • Proximity extension assays encode antibodies using DNA oligonucleotides, which hybridize only when the antibodies are in close proximity. Quantification is assessed through quantitative PCR.
Limitations & future outlook • Many multiplex homogeneous sandwich immunoassays are limited in multiplexing, suffer cross talk, require specialized readers and experience the hooking effect.
References
6
Hornbeck P. Enzyme-linked immunosorbent assays. In: Current protocols in immunology. John Wiley and Sons, Inc., NJ, USA, Chapter 2, Unit 2 1 (2001).
Papers of special note have been highlighted as: • of interest; •• of considerable interest
1554
1
Lemoine J, Fortin T, Salvador A, Jaffuel A, Charrier JP, Choquet-Kastylevsky G. The current status of clinical proteomics and the use of mrm and mrm(3) for biomarker validation. Expert Rev. Mol. Diagn. 12(4), 333–342 (2012).
7
Li H, Bergeron S, Juncker D. Microarray-to-microarray transfer of reagents by snapping of two chips for crossreactivity-free multiplex immunoassays. Anal. Chem. 84(11), 4776–4783 (2012).
2
Boja ES, Jortani SA, Ritchie J et al. The journey to regulation of protein-based multiplex quantitative assays. Clin. Chem. 57(4), 560–567 (2011).
8
Markov DA, Swinney K, Bornhop DJ. Label-free molecular interaction determinations with nanoscale interferometry. J. Am. Chem. Soc. 126(50), 16659–16664 (2004).
3
Diamandis EP. The failure of protein cancer biomarkers to reach the clinic: why, and what can be done to address the problem? BMC Med. 10, 87 (2012).
9
Gordon R, Sinton D, Kavanagh KL, Brolo AG. A new generation of sensors based on extraordinary optical transmission. Acc. Chem. Res. 41(8), 1049–1057 (2008).
4
Rifai N, Gillette MA, Carr SA. Protein biomarker discovery and validation: the long and uncertain path to clinical utility. Nat. Biotechnol. 24(8), 971–983 (2006).
10
5
Anderson NL. The clinical plasma proteome: a survey of clinical assays for proteins in plasma and serum. Clin. Chem. 56(2), 177–185 (2010).
Washburn AL, Luchansky MS, Bowman AL, Bailey RC. Quantitative, label-free detection of five protein biomarkers using multiplexed arrays of silicon photonic microring resonators. Anal. Chem. 82(1), 69–72 (2010).
11
Ji J, O’connell JG, Carter DJ, Larson DN. Highthroughput nanohole array based system to monitor
Bioanalysis (2015) 7(12)
future science group
Systems for multiplexing homogeneous immunoassays
multiple binding events in real time. Anal. Chem. 80(7), 2491–2498 (2008). 12
13
Zheng G, Patolsky F, Cui Y, Wang WU, Lieber CM. Multiplexed electrical detection of cancer markers with nanowire sensor arrays. Nat. Biotechnol. 23(10), 1294–1301 (2005). Wang ZH, Jin G. A label-free multisensing immunosensor based on imaging ellipsometry. Anal. Chem. 75(22), 6119–6123 (2003).
26
Shaw LM, Vanderstichele H, Knapik-Czajka M et al. Qualification of the analytical and clinical performance of CSF biomarker analyses in adni. Acta Neuropathol. 121(5), 597–609 (2011).
27
Qiao Y, Tang H, Munske GR, Dutta P, Ivory CF, Dong WJ. Enhanced fluorescence anisotropy assay for human cardiac troponin I and T detection. J. Flouresc. 21(6), 2101–2110 (2011).
28
Pakkila H, Malmi E, Lahtinen S, Soukka T. Rapid homogeneous immunoassay for cardiac troponin I using switchable lanthanide luminescence. Biosens. Bioelectron.62, 201–207 (2014).
14
Wang ZH, Meng YH, Ying PQ, Qi C, Jin G. A labelfree protein microfluidic array for parallel immunoassays. Electrophoresis 27(20), 4078–4085 (2006).
15
Xue C-Y, Khan SA, Yang K-L. Exploring optical properties of liquid crystals for developing label-free and high-throughput microfluidic immunoassays. Adv. Mat. 21(2), 198–202 (2009).
29
Arrebola MM, Lillo JA, Diez De Los Rios MJ et al. Analytical performance of a sensitive assay for cardiac troponin I with LOCI technology. Clin. Biochem. 43(12), 998–1002 (2010).
16
Michael KE, Vernekar VN, Keselowsky BG, Meredith JC, Latour RA, Garcia AJ. Adsorption-induced conformational changes in fibronectin due to interactions with well-defined surface chemistries. Langmuir 19(19), 8033–8040 (2003).
30
Schallmeiner E, Oksanen E, Ericsson O et al. Sensitive protein detection via triple-binder proximity ligation assays. Nat. Meth. 4(2), 135–137 (2007).
31
Schwenzer KS, Anhalt JP. Automated fluorescence polarization immunoassay for monitoring streptomycin. Antimicrob. Agents Chemother. 23(5), 683–687 (1983).
32
Filburn BH, Shull VH, Tempera YM, Dick JD. Evaluation of an automated fluorescence polarization immunoassay for vancomycin. Antimicrob. Agents Chemother. 24(2), 216–220 (1983).
33
Smith DS, Eremin SA. Fluorescence polarization immunoassays and related methods for simple, highthroughput screening of small molecules. Anal. Bioanal. Chem. 391(5), 1499–1507 (2008).
17
18
Armstrong JK, Wenby RB, Meiselman HJ, Fisher TC. The hydrodynamic radii of macromolecules and their effect on red blood cell aggregation. Biophys. J. 87(6), 4259–4270 (2004). Diercks AH, Ozinsky A, Hansen CL, Spotts JM, Rodriguez DJ, Aderem A. A microfluidic device for multiplexed protein detection in nano-liter volumes. Anal. Biochem. 386(1), 30–35 (2009).
19
Park J, Sunkara V, Kim TH, Hwang H, Cho YK. Lab-ona-disc for fully integrated multiplex immunoassays. Anal. Chem. 84(5), 2133–2140 (2012).
••
20
Tenje M, Xia H, Evander M et al. Acoustic trapping as a generic non-contact incubation site for multiplex bead-based assays. Anal. Chim. Acta 853, 682–688 (2015).
Overview of the fluorescence polarization immunoassay (FPIA) technology.
34
Ng AH, Uddayasankar U, Wheeler AR. Immunoassays in microfluidic systems. Anal. Bioanal. Chem. 397(3), 991–1007 (2010).
Hussain SA. An introduction to fluorescence resonance energy transfer (FRET). http://arxiv.org
35
Lam AJ, St-Pierre F, Gong Y et al. Improving FRET dynamic range with bright green and red fluorescent proteins. Nat. Meth. 9(10), 1005–1012 (2012).
36
Liu M, Zhao H, Quan X, Chen S, Fan X. Distanceindependent quenching of quantum dots by nanoscalegraphene in self-assembled sandwich immunoassay. Chem. Commun. 46(42), 7909–7911 (2010).
•
Novel innovation to resolve distance-dependent behavior that may serve as template to overcome assay restrictions.
37
Ullman EF, Kirakossian H, Switchenko AC et al. Luminescent oxygen channeling assay (LOCI): sensitive, broadly applicable homogeneous immunoassay method. Clin. Chem. 42(9), 1518–1526 (1996).
38
Leister KP, Huang R, Goodwin BL, Chen A, Austin CP, Xia M. Two high throughput screen assays for measurement of TNF-alpha in thp-1 cells. Curr. Chem. Genomics 5, 21–29 (2011).
39
Fredriksson S, Gullberg M, Jarvius J et al. Protein detection using proximity-dependent DNA ligation assays. Nat. Biotechnol. 20(5), 473–477 (2002).
40
Fredriksson S, Dixon W, Ji H, Koong AC, Mindrinos M, Davis RW. Multiplexed protein detection by proximity
21
22
23
Tian J, Zhou L, Zhao Y, Wang Y, Peng Y, Zhao S. Multiplexed detection of tumor markers with multicolor quantum dots based on fluorescence polarization immunoassay. Talanta 92, 72–77 (2012). Zeng Q, Zhang Y, Liu X, Tu L, Kong X, Zhang H. Multiple homogeneous immunoassays based on a quantum dotsgold nanorods fret nanoplatform. Chem. Commun. 48(12), 1781–1783 (2012).
24
Simon AB, Frampton JP, Huang NT, Kurabayashi K, Paczesny S, Takayama S. Aqueous two-phase systems enable multiplexing of homogeneous immunoassays. Technology 2(2), 176 (2014).
••
Explanation of luminescent oxygen-channeling immunoassay (LOCI) and aqueous two-phase systems (ATPS).
25
Assarsson ELM, Holmquist G, Bjo Rkesten J et al. Homogenous 96-plex pea immunoassay exhibiting high sensitivity, specificity, and excellent scalability. PLoS ONE 9(4), e95192 (2014).
•
Explanation of DNA proximity assays.
future science group
www.future-science.com
Review
1555
Review Yamanishi, Han-Ching Chiu & Takayama ligation for cancer biomarker validation. Nat. Meth. 4(4), 327–329 (2007). 41
Darmanis S, Nong RY, Hammond M et al. Sensitive plasma protein analysis by microparticle-based proximity ligation assays. Mol. Cell. Proteomics 9(2), 327–335 (2010).
42
ClinicalTrials Database: NCT01957241. http://ClinicalTrials.gov
43
ClinicalTrials Database: NCT02303886. http://ClinicalTrials.gov
•
Detailed explanation of fluorescence resonance energy transfer (FRET) theory and assay design including its difficulties.
44
Ellington AA, Kullo IJ, Bailey KR, Klee GG. Antibodybased protein multiplex platforms: technical and operational challenges. Clin. Chem. 56(2), 186–193 (2010).
45
Lee JW, Devanarayan V, Barrett YC et al. Fit-for-purpose method development and validation for successful biomarker measurement. Pharmaceut. Res. 23(2), 312–328 (2006).
•
Detailed guidelines for immunoassay characterization in an application-specific format.
46
Desilva B, Garofolo F, Rocci M et al. 2012 White paper on recent issues in bioanalysis and alignment of multiple guidelines. Bioanalysis 4(18), 2213–2226 (2012).
Lipman NS, Jackson LR, Trudel LJ, Weis-Garcia F. Monoclonal versus polyclonal antibodies: distinguishing characteristics, applications, and information resources. ILAR J./National Research Council, Institute of Laboratory Animal Resources 46(3), 258–268 (2005).
53
Schwenk JM, Lindberg J, Sundberg M, Uhlen M, Nilsson P. Determination of binding specificities in highly multiplexed bead-based assays for antibody proteomics. Mol. Cell. Proteomics: MCP 6(1), 125–132 (2007).
54
Kim JW, You J. Protein target quantification decision tree. Int. J. Proteomics 2013, 701247 (2013).
55
Wood WG. “Matrix effects” in immunoassays. Scand. J. Clin. Lab. Invest. Suppl. 205, 105–112 (1991).
56
Mire-Sluis AR, Barrett YC, Devanarayan V et al. Recommendations for the design and optimization of immunoassays used in the detection of host antibodies against biotechnology products. J. Immunol. Methods 289(1–2), 1–16 (2004).
57
Ermens AaM, Bayens AJM, Crooymans A, Broekman-Van Hout AaM, Van Duijnhoven HLP. Dilution protocols for detection of hook effects/prozone phenomenon. Clin. Chem. 46(10), 1719–1720 (2000).
58
Weinstock C, Schnaidt M. The complement-mediated prozone effect in the luminex single-antigen bead assay and its impact on HLA antibody determination in patient sera. Int. J. Immunogenet. 40(3), 171–177 (2013).
47
Tighe P, Negm O, Todd I, Fairclough L. Utility, reliability and reproducibility of immunoassay multiplex kits. Methods 61(1), 23–29 (2013).
59
48
Li D, Chiu H, Gupta V, Chan DW. Validation of a multiplex immunoassay for serum angiogenic factors as biomarkers for aggressive prostate cancer. Clin. Chim. Acta 413(19–20), 1506–1511 (2012).
Juncker D, Bergeron S, Laforte V, Li H. Cross-reactivity in antibody microarrays and multiplexed sandwich assays: shedding light on the dark side of multiplexing. Curr. Opin. Chem. Biol. 18, 29–37 (2014).
60
49
Shah VP, Midha KK, Dighe S et al. Analytical methods validation: bioavailability, bioequivalence, and pharmacokinetic studies. J. Pharm. Sci. 81, 309–312 (1992).
Perlee L, Christiansen J, Dondero R et al. Development and standardization of multiplexed antibody microarrays for use in quantitative proteomics. Proteome Sci. 2(1), 9 (2004).
61
50
Schnaidt M, Weinstock C, Jurisic M, Schmid-Horch B, Ender A, Wernet D. HLA antibody specification using single-antigen beads‐‐a technical solution for the prozone effect. Transplantation 92(5), 510–515 (2011).
Bjerner J, Bormer OP, Nustad K. The war on heterophilic antibody interference. Clin. Chem. 51(1), 9–11 (2005).
62
Bolstad N, Warren DJ, Nustad K. Heterophilic antibody interference in immunometric assays. Best Pract. Res. Clin. Endocrinol. Metab. 27(5), 647–661 (2013).
63
Bjerner J, Nustad K, Norum LF, Olsen KH, Bormer OP. Immunometric assay interference: incidence and prevention. Clin. Chem. 48(4), 613–621 (2002).
51
1556
52
Poetz O, Ostendorp R, Brocks B et al. Protein microarrays for antibody profiling: specificity and affinity determination on a chip. Proteomics 5(9), 2402–2411 (2005).
Bioanalysis (2015) 7(12)
future science group
For reprint orders, please contact [email protected]
Systems for multiplexing homogeneous immunoassays
High-throughput multiplex protein biomarker assays continue to gain significance in the fields of biomarker discovery and drug development, due to their economical use of not only the precious clinical biological samples but also expensive reagents. Among these platforms, homogeneous multiplex systems have potential for short assay run times and cost-effective reagent consumptions. However, these systems must overcome challenges of signal cross talk and biochemical cross-reactivity. Despite these obstacles, several homogeneous multiplex immunoassays have been demonstrated. These include fluorescent polarization, fluorescent resonance energy transfer with quantum dots or graphene, luminescent oxygen-channeling immunoassay coupled with aqueous two-phase systems and DNA proximity assays. The balance between speed/simplicity and high multiplexing and robustness of these homogeneous multiplex immunoassays are discussed in this review.
Introduction to multiplex homogeneous immunoassays The NIH defined protein biomarkers as ‘key molecular or cellular events that link a specific environmental exposure to a health outcome.’ Understanding the interactions of multiple protein biomarkers of disease will play an integral role in drug screening, in biological pathway analysis and in disease diagnosis [1] . Single-analyte detection has provided high accuracy results, but in a limited scope. Multiplex detection methods, measurement and analysis of multiple biomarkers together, provide a view of the entire system. Current protein biomarker discovery techniques often use mass spectrometry on a few patient samples for screening large numbers of proteins without the need for preconceived protein targets [2] . At the other extreme, qualification of biomarker targets are often performed using relatively lowthroughput sandwich immunoassays, due to their high sensitivity and specificity [3] . In clinical practice, immunoassays are also preferred over mass spectroscopy due to lower cost and shorter time to result for single target detection. This process of approving
10.4155/BIO.15.78 © 2015 Future Science Ltd
biomarkers or biomarker panels from discovery to clinical practice presents a great challenge; the translation of discovered targets to clinical practice yields fewer than two new protein biomarkers approved by the US FDA per year [4,5] . Multiplex immunoassays could help bridge the gap between protein biomarker discovery and clinical use. Due to their high sensitivity, specificity and procedural convenience, this article will focus on homogeneous sandwich immunoassays. Sandwich immunoassays come in two formats: heterogeneous and homogeneous. The standard ELISA is heterogeneous, with capture antibodies (cAb) immobilized onto a microtiter plate. The addition of detection antibodies (dAb) forms the antibody–antigen–antibody sandwich complex [6] . The requirement for both the cAbs and dAbs to bind to the target reduces noise from nonspecific binding [7] , making the sandwich format more specific than label-free methods [8–15] , which only use one antibody. Unlike heterogeneous immunoassays, homogeneous sandwich immunoassays use both antibodies in solution, typically conjugated to
Bioanalysis (2015) 7(12), 1545–1556
Cameron D Yamanishi1,2, Joyce Han-Ching Chiu1,2 & Shuichi Takayama*,1,2,3,4 1 Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI 48109, USA 2 Biointerfaces Institute, University of Michigan, Ann Arbor, MI 48109, USA 3 Department of Macromolecular Science & Engineering, University of Michigan, Ann Arbor, MI 48109, USA 4 Michigan Center for Integrative Research in Critical Care, University of Michigan, Ann Arbor, MI 48109, USA *Author for correspondence: Tel.: +1 734 615 5539 [email protected]
part of
ISSN 1757-6180
1545
Review Yamanishi, Han-Ching Chiu & Takayama
Key terms Protein biomarker: Proteins in the blood whose concentration may be used to diagnose disease state. Multiplex detection: Simultaneous measurement of multiple targets in the same solution. Homogeneous sandwich immunoassay: Technique using a soluble antibody pair to measure concentration of proteins in solution. Fluorescence polarization immunoassay: Assay based on stabilization of molecular motions by antigen–antibody binding that will reduce the depolarization of excitation of fluorescent tags, which are excited by polarized light sources. Fluorescence resonance energy transfer: Measures the fluorescence signals by energy transfer from donor to acceptor fluorophore. Quantum dots: Nanocrystals with tunable fluorescent properties.
an identifiable (encoded) marker. Binding in solution, rather than at a surface, drastically reduces the distance that antigens must diffuse to reach the antibodies. Combined with the elimination of washing steps, this reduces assay time, making homogeneous assays more suitable for biosensors and assays used in field applications. Additionally, the absence of washing steps allows antibodies with relatively low affinity to be used, because unbinding events resulting from washing steps will be reduced, thereby retaining more antigens. The soluble format can also provide a more native reaction environment for the antibodies than the surface-bound format, as proteins have been shown to undergo conformational changes when they adsorb to nonnative surfaces [16] . Most homogenous sandwich immunoassays to date, however, have a major limitation. Because both capture and detection antibody reagents are freely diffusing in solution, it has been difficult to perform spatially segregated reagent multiplexing when performing homogeneous immunoassays. Additionally, multiplex immunoassays suffer from cross talk, the phenomenon resulting from one labeling molecule, such as a fluorescent marker, generating an overlapping signal that appears as a false positive for another labeling molecule. With higher number of antibodies and antigens used in multiplex assays, the chances of cross-reactivity between molecules increases. The goal of this article is to review the latest methods and techniques to overcome this limitation to enable multiplex homogenous sandwich immunoassays. Bead-based systems fit in between heterogeneous and homogeneous. These assays replace the planar surface of standard ELISA with microbeads, encoded with identifiers. Bead assays still have faster mix-
1546
Bioanalysis (2015) 7(12)
ing than standard ELISA, because diffusion time scales as: 2
T .
L D
Equation 1
where L is the characteristic length and D is the diffusivity; the characteristic length between particles in solution is much smaller than the length between the bulk solution and the surface. However, large beads diffuse more slowly than free antibodies. The Stokes– Einstein relation describes the diffusivity as inversely proportional to radius at constant temperature and viscosity. IgG antibodies have radius ~5 nm [17] . Adding a bead 1 μm in diameter will therefore reduce the diffusivity by 100-fold. For the purpose of this review, we strictly define homogeneous immunoassays as those containing all antibodies and antigens in solution, with no washing between antibody–antigen binding steps. This definition will exclude pseudohomogeneous bead-based assay formats, such as the Luminex xMAP system, which employs suspension microbeads as a solid-phase replacement for planar surfaces (bead sources). The Luminex system retains some of the time-to-result advantages of homogeneous systems, but the wash step limits the effectiveness of low affinity antibodies used in assays. This is not a problem for biomarkers with well-established antibody libraries, but finding high affinity antibodies for obscure or recently discovered proteins can be difficult. Our selection criteria also exclude microfluidic platforms despite their capacity for automation of washing steps or their shared advantages with homogeneous sandwich immunoassays, including reduced diffusion distances and small sample volume consumption [18–21] . In this report, we primarily review techniques for multiplexing homogeneous sandwich immunoassays, highlighting the merits and demerits of each, as summarized in Table 1. Fluorescent polarization immunoassay (FPIA) is not a sandwich assay, as it employs only one antibody, but it is included in this scope because it is an established multiplex homogeneous assay, demonstrated with two-plex [22] . Fluorescence resonance energy transfer (FRET) assay employs the photon energy transfer from the donor to the acceptor in a sandwich system to detect antibody and antigen binding reaction for up to two-plex assay [23] . It has been adapted to incorporate graphene and quantum dots to overcome the limitation of distance dependent behavior of an energy transfer system. The luminescent oxygenchanneling immunoassay (LOCI) uses a different type of energy transfer via singlet oxygen to supply the energy necessary for the acceptor chemiluminescent label for two-plex system, and higher order or multiplexing can
future science group
Systems for multiplexing homogeneous immunoassays
Review
Table 1. Multiplex homogeneous immunoassay parameters. Assay format
LLOQ
ULOQ
Advantages
Limitations
Highest reported multiplexing
Ref.
FPIA
360 pg/ml
~120 ng/ml
Rapid (40 ng/ml
Rapid (1.5 h)
Susceptible to false positives from cross-reactivity
4
[24,29]
DNA proximity
9.6 pg/ml
~120 ng/ml
High multiplexing, specific, sensitive
Complex readout, time-to-result of 1 day
96
[25,30]
FPIA: Fluorescence polarization immunoassay; FRET: Fluorescence resonance energy transfer; LOCI: Luminescent oxygen-channeling immunoassay.
be achieved by the use of aqueous two-phase systems (ATPS) [24] . DNA proximity assays use DNA oligomers as reporter molecules, employing their specificity and existing amplification techniques for effective multiplexing. These techniques have been demonstrated up to 96-plex [25] . To compare assay formats, tests should ideally be performed within the same lab and for the same targets, as antibodies and user skill can drastically alter assay results [26] . However, cross-platform comparisons between the homogeneous immunoassay techniques covered in this review are currently sparse, both for single target and multiplex formats. Nevertheless, comparisons of assay range between different labs can still be useful. To compare formats, we selected cTnI, as it is already an established biomarker of cardiac arrest. As summarized in Table 1, the quantitative ranges of FPIA [27] , FRET [28] and LOCI [29] have all been reported to be similar, while the DNA proximity assay [30] has been published with a comparatively lower limit of detection. However, it should be emphasized that these parameters are application and target specific, and the quantitative ranges may be tuned to fit the application needs. The techniques covered in this review have had only limited use in clinical trials. Between multiplex versions of FPIA, FRET and LOCI, no clinical trials have been reported in WHO participant countries. However, the single-plex versions of these assays are used in clinical diagnosis. Only two clinical trials have been reported using the multiplex DNA proximity immunoassays. We will discuss the potential and current obstacles to implement these multiplex techniques in clinical settings. Homogeneous multiplex sandwich immunoassay formats
polarized plane as the light source. Conversely, light from a rotating fluorescent molecule will be emitted in a rotated plane. In a free floating solution, large molecules experience less rotation and motion during excitation, and will therefore give off a smaller depolarization, whereas small molecules will have larger depolarization (Figure 1) . As fluorescent target molecules bind with the antibody, the difference in depolarization can be measured. The immunoassay uses antibodies alongside fluorescently labeled antigen, a.k.a. tracer. When the tracer binds with antibodies, the whole fluorophore complex is stabilized, thereby reducing motion and allowing for greater polarized emission [31–33] . When test samples such as patient plasma or serum are introduced to the system, the patient’s target antigen will compete with tracer molecules for the antibodies. The technique is by nature homogenous and compatible with automated detection for high-throughput analysis. However, it is limited to small-molecule applications because more dramatic changes in depolarization may be observed with small molecules. Multiplex FPIA can be achieved by utilizing multicolor quantum dots with narrow emission bandwidths that can be cleanly resolved. This format uses the same principle as single-plex FPIA, wherein the antibody stabilizes the quantum dot to polarize the light. Specifically, a two-plex assay was designed with red and green quantum dots conjugated with tumor biomarkers. The target protein can be identified by the color of the quantum dot, as shown in Figure 2. In this method, centrifugation with ultracentrifuge filters of serum samples prior to the assay was necessary to remove matrix proteins to reduce matrix interference. The results of this multiplex assay were shown to have clinical diagnostic value comparable to commercial ELISA [22] .
Fluorescence polarization immunoassay
FPIA is a form of competitive immunoassay. When incident light excites a fluorescently tagged stationary molecule, the resulting emission will be on the same
future science group
Fluorescence resonance energy transfer
FRET-based immunoassays utilize a donor–acceptor system, as illustrated in Figure 3. When a laser excites
www.future-science.com
1547
Review Yamanishi, Han-Ching Chiu & Takayama
A
Slow rotation High polarization
Fast rotation Low polarization
Fluorescent label
Fluorescent label
Antigen
Antigen Antibody
B
Add sample
Figure 1. Fluorescence polarization immunoassay. (A) Polarization intensity in relation to molecule complex size and rotation speed. (B) Schematic representation of multiplex fluorescence polarization. (A) Adapted with permission from [33] © Springer (2008). (B) Adapted with permission from [22] © Elsevier (2012).
the donor fluorophores, the resultant fluorescent energy is transferred from the donor fluorophore to the nearby acceptor molecule, which will subsequently release photons of a different wavelength for detection. The donor and acceptor molecules are both attached to antibodies specific to the target protein. If the target is present, the antibodies can form sandwich complexes with the target, keeping their respective donor and acceptor tags in close proximity. Because the efficiency of energy transfer decreases rapidly as the distance between the donor and acceptor pair increases, the fluorescent signal corresponds to the presence of the target. However, this distance dependence also limits the sensitivity of sandwich immunoassays, because the distance across the antibodies and antigens in a sandwich complex is large enough to diminish FRET efficiency [34] . Considerable progress has been made to improve FRET efficiency. The important parameters for FRET efficiency are the quantum yield, extinction coefficients
1548
Bioanalysis (2015) 7(12)
and spectral overlap [35] . The quantum yield and extinction coefficients of both the donor and the acceptor should be high to maximize efficiency. Using genetic engineering of fluorescent proteins, donor–acceptor pairs can be tuned with large overlap between donor emission and acceptor absorption spectra to increase energy transfer efficiency. On the other hand, undesirable overlap between the donor emission and the acceptor emission spectra can be minimized to avoid optical cross talk. Although FRET-based assays are traditionally built with fluorescent dyes, current research on FRET-based immunoassays is increasingly performed with nanostructures made up of nanorods, nanoshells and quantum dots. These nanostructures make use of the surface plasmon resonance effect to increase FRET signal for higher detection sensitivities. Zeng et al. engineered a quantum dot and gold nanorod two-plex assay to maximize FRET yield [23] . Gold nanorods have two distinct plasmon resonances, corresponding to their two dimensions,
future science group
Systems for multiplexing homogeneous immunoassays
A
B
Review
Acceptor fluorophore
Relaxation
S1
FRET
Fluorescence
Fluorescence
Excitation
S1
FRET
S0 Acceptor
S0 Donor
Donor fluorophore
Figure 2. Fluorescence resonance energy transfer. (A) Resonance energy transfer between donor and acceptor. (B) FRET applied to immunochemistry. FRET: Fluorescence resonance energy transfer. (B) Adapted with permission from [35] © Nature Publishing Group (2012).
longitudinal and transverse. Because both dimensions are tunable, gold nanorods are ideal acceptors for multiplex assays. Two different sizes of quantum dots (CdTe/ CdS core/shell) with two different emission wavelengths were tailored to maximize the overlap of quantum dot donor emission and self-assembled gold nanorods acceptor absorption spectra. This system exhibited high
specificity between the two targets, indicating the utility of the FRET pairs in the multiplex system. Fluorescent quantum dots-graphene immunoassay
Graphene quenching provides an alternative to FRET. Graphene sheets efficiently quench fluorescence from
Figure 3. Quantum dots and graphene sheets sensor. This platform utilizes the fluorescence quenching ability of the graphene sheets to overcome distance-dependent limitation of energy transfer techniques. Adapted with permission from [36] © Royal Society of Chemistry (2010).
future science group
www.future-science.com
1549
Review Yamanishi, Han-Ching Chiu & Takayama quantum dots when the two are in close proximity. Therefore, in constrast to FRET, a decrease in fluorescence corresponds to presence of targets in the sample. The primary disadvantage of assays based on energy transfer between quantum dots and nanostructure systems lies in the required donor and acceptor dipole– dipole distance, which is limited to 10–100 Å [34] . However, graphene sheets are shown to overcome this restriction [36] . Compared with a quantum dot-quantum dot pair, the 2D graphene sheets have more possible spatial arrangements in which parts of the sheet are close enough to the spherical quantum dots to cause quenching, as seen in Figure 4. This platform increased the effective distance between the quantum dot anchor and the graphene anchor from 100 Å up to 223 Å. This distance increase enables greater dynamic range in the sandwich assay formats that use the antigen as a bridge to bring the detection Ab-coated quantum dot close to the capture Ab-coated graphene surface.
complete the complex. The presence of the target protein thus determines how many donor–acceptor pairs are in close proximity, where they can produce the chemiluminescent signal. The LOCI can perform twocolor multiplex immunoassays using a second acceptor bead that is also excited by singlet oxygen and emits at 545 nm. The second donor bead is functionalized with the appropriate antibodies, but contains the same light-sensitive compound as the first donor bead. The LOCI format is fairly convenient for bench-top research. LOCI assays can be read using a microplate reader equipped with the proper fluorescent channels. For these readers, high laser power is recommended for optimal signal. Additionally, the LOCI format has intrinsic advantages for accuracy. Because LOCI employs a shift from a high wavelength excitation to a lower wavelength emission, it eliminates noise from autofluorescence. Compared with the other optical detection assays, the LOCI assay has a relatively wide dynamic range and low lower limit of quantification for the detection of cTnI, as shown in Table 1. Furthermore, the single-plex version of LOCI is available from Siemens for clinical diagnostics. Although the commercially available PerkinElmer system is currently limited to two-plex using chemiluminescence at two different emission wavelengths, higher multiplexing has been demonstrated by spatially arraying reagent solutions using aqueous two-phase system (ATPS) [24] . When two immiscible polymers are mixed in an aqueous solution at appropriate concentrations, they cause separation of the resulting aqueous solution into two phases. These systems are mild to biomolecules, making them appropriate for use in immunoassays. In a polyethylene glycol – DEX ATPS, antibodies and beads partition preferentially into the DEX phase, allowing spatial
Luminescent oxygen-channeling immunoassay
The LOCI, commercially available as AlphaLISA, from PerkinElmer, uses donor and acceptor beads (250–350 nm diameter) [37] . When the donor beads are excited by 680 nm light, they release reactive singlet oxygen. This activated singlet oxygen decays back to its ground state quickly in solution. However, if acceptor beads are nearby, the singlet oxygen molecules can transfer their energy to the chemiluminescent compound in the acceptor beads, triggering the acceptor beads to emit light at 615 nm [37] , as depicted in Figure 5 [38] . In a sandwich assay format, capture antibodies attached to acceptor beads bind to target proteins in solution. Biotinylated detection antibodies sandwich the target proteins. Streptavidin-coated donor beads then bind to the detection antibodies to
O2
1
Excitation 680 nm
Emission 615 nm
B Streptavidin-coated donor bead
Biotinylated anti-analyte
Analyte
Anti-analyte-conjugated acceptor bead
Figure 4. Luminescent oxygen-channeling immunoassay. In the luminescent oxygen-channeling immunoassay format, donor beads are excited by 680 nm light and release singlet oxygen. Acceptor beads in the vicinity emit 615 nm light in response to the singlet oxygen. Adapted under the Creative Commons Attribution license from [38] .
1550
Bioanalysis (2015) 7(12)
future science group
Systems for multiplexing homogeneous immunoassays
Conventional 1-color LOCI
Review
ATPS LOCI
Antibodies and beads bind to antigens
DE
PEG
PBS
X
Antibody–bead cross-reactions
Donor beads are excited
DE
PEG
PBS
X
No spatial separation
Acceptor beads emit signal
Figure 5. Aqueous two-phase system ELISA. ATPS ELISA co-localizes donor–acceptor pairs within dextran microdroplets, enabling spatially distinct multiplex detection with a single bead color. ATPS: Aqueous two-phase system; DEX: Dextran; LOCI: Luminescent oxygen-channeling immunoassay; PBS: Phosphate buffer serum; PEG: Polyethylene glycol. Adapted with permission from [24] © WorldScientific (2014).
confinement of distinct AlphaLISA bead pairs preallocated into individual phase-separated droplets within a single well, as demonstrated in Figure 6. The ATPS reagent localization strategy marks a promising development for multiplex assays. The technique is theoretically compatible with other homogeneous immunoassay formats that have not yet been multiplexed, including simple bead agglutination. Additionally, the spatial localization eliminates the possibility of cross talk between unmatched antibody pairs. Therefore, assay validation does not require the extensive combinatorial screening that many other multiplex techniques must undergo. However, the ATPS method has some drawbacks. The polymers involved are viscous, lowering the diffusivity of reagents and requiring slightly longer assay time. This method also requires specialty plates with topographical features within each well to position the reagent microdroplets from moving within the well. The published four-plex AlphaLISA assay plate was designed to be compatible with conventional 1536 well format AlphaLISA readers. DNA proximity
Proximity multiplex formats use DNA oligomers to encode antibodies. Taking advantage of existing DNA
future science group
sequencing techniques, the proximity extension assay (PEA) [25] uses antibody pairs attached to DNA oligonucleotides. As shown in when the antibodies sandwich the target protein, the DNA oligonucleotides come into close proximity, where the DNA strand from one antibody can hybridize with its complementary DNA strand on the partner antibody. This hybridization allows a DNA polymerase to extend the DNA at both ends. The resulting oligonucleotides can be subsequently amplified by qPCR. Only the completed DNA strands, i.e., those which had hybridized and been extended by the DNA polymerase, can be amplified by PCR. Thus, DNA detected by qPCR reflects the presence of the target protein. In multiplex PEA, distinct DNA sequences are used to identify each antibody pair. The PEA technology is available from Olink, and has been demonstrated in a 96-plex assay [25] . Key terms Singlet oxygen: Electronically excited oxygen. Aqueous two-phase systems: Aqueous two-phase systems comprising two immiscible polymers in water. Extension: Technique to replicate partially overlapping complementary DNA.
www.future-science.com
1551
Review Yamanishi, Han-Ching Chiu & Takayama
A
Incubation
B
Hybridization extension
C
Preamplification
D
E
Digestion
Microfluidic qPCR
Protein Expression (ddCq)
12 10 8 6 4 2 0
Figure 6. Proximity extension assay. (A) The proximity extension assay uses antibodies with oligonucleotides in a sandwich format. (B) DNA strands are hybridized with nearby complementary strands. (C–E) Through qPCR, DNA hybridization can be quantified. qPCR: Quantitative PCR. Reprinted under the Creative Commons Attribution license from [25] .
In a similar format to PEA, the proximity ligation assay (PLA) [39,40] also uses DNA oligomers to encode antibody pairs. Whereas the DNA oligomers in PEA are complementary to each other, the DNA oligomers in PLA do not hybridize with each other. Rather, a third DNA oligomer, complementary to both antibody-bound oligomers, is added to the solution to bridge the DNA on the antibody pair. The two antibody-bound oligomers may then be connected end-to-end. To increase the specificity of the assay, one study attached an antibody to the third DNA oligomer, making the assay require three matching epitopes to achieve a positive result [41] . Proximity multiplex immunoassays do not eliminate cross-reactions but avoid reading out such cross-reactions because only matching antibody pairs’ complementary oligonucleotide labels can hybridize to create signal. Although Olink offers a 96-plex cardiovascular disease panel, cTnI is not included. However, the PLA has been demonstrated to have a lower limit of quantification of 9.6 pg/ml for cTnI. The advantages of low limit of detection and high multiplexing in DNA proximity assays come at the cost of increased assay time and complexity. The 96-plex assay required an overnight sample incubation, followed by DNA hybridization and the qPCR steps. The Olink company has addressed many of the added steps by incorporating their qPCR process into the Fluidigm-automated microfluidic PCR reader, the BioMark HD. To date, PLA and PEA have been employed in one clinical trial each, as recorded by WHO countries. Multiplex PLA is currently being used in a clinical trial from National Taiwan University Hospital to quantify biomarkers that can aid in characterization of hepatocellular carcinoma and esophageal cancer. The group, led by J Chia-Hsien Cheng, aims to determine tumor response, prognosis and radiation-induced organ toxicity in the lung and liver by assaying 56 proteins in serum samples of hundreds of patients before, during and after radiation therapy [42] . The PEA clinical trial employed multiplex biomarker analysis as an evaluative tool,
1552
Bioanalysis (2015) 7(12)
rather than as a new disease diagnosis tool. A Miclescu’s group at Uppsala University Hospital tested ten patients in a drug trial to evaluate methylene blue as a treatment for neuropathic pain. This study monitored patient outcomes both by qualitative pain reports from the patients and by quantitative measurement of plasma and urine concentrations of a panel of 92 protein biomarkers [43] . Assay characterization Before implementation, new immunoassays must be characterized for their analytical performance, then for clinical validation [44] . The degree and type of characterization necessary depends on the desired application. Lee et al. outline some of these differences for pharmacokinetic studies, drug development biomarker assays and diagnostic biomarker assays. They also provide guidelines according to the developmental stage of the assay [45] . Such is true for single-plex assays, but there are additional considerations and specific applications for multiplex assays. Systems such as FPIA, FRET and LOCI are suitable for time-sensitive point-of-care diagnostic applications, because they provide rapid results. However, DNA proximity assays currently generate higher multiplexing at the cost of simplicity and timeto-result. Therefore, the DNA proximity assays are more suitable for early biomarker candidate screening, pharmacokinetic studies and evaluation during drug development. For single-plex immunoassay characterization, several informative papers have been published [45–50] . Therefore, we will focus on considerations specific to homogeneous and multiplex immunoassays. Potential cross-reactions between both endogenous and exogenous antibodies and proteins may result in over- and under-estimation of protein concentrations and should be screened out. Polyclonal antibodies recognize multiple epitopes and can compensate for low affinities, but the high specificity of monoclonal antibodies are preferred for reduced cross-reactivity [51–54] . Previous reports have characterized multiplex immunoas-
future science group
Systems for multiplexing homogeneous immunoassays
says by comparison with corresponding single-plex assays [7] . At high numbers of multiplex targets, alternative comparisons such as batch sampling have been employed [25] . However, these batch sampling methods do not account for cross-reactivity, and are only sufficient for proof-of-concept studies. Another important consideration for immunoassays is matrix interference. Matrix interference refers to changes in assay results caused by interactions between assay components and moieties from the sample other than the intended target. This effect can be observed by dilutions of samples with different dilution factor and diluents. Investigation of minimum dilution factor is necessary to avoid quantification errors [49,55,56] . This is especially true in no-wash sandwich immunoassays, where the hook effect, or prozone phenomenon occurs [50] . As higher concentrations are measured, the signal will increase, plateau, then decrease in a hook shape. The decrease results from target proteins saturating both the detection antibodies and the capture antibodies, preventing formation of the sandwich complex. The hook effect may be addressed by diluting the sample into the monotonic range of the assay [57] . Alternatively, Weinstock et al. have demonstrated that addition of eythylenediaminetetraacetic acid (EDTA) to serum can reduce the hook effect, by inhibiting the activity of calcium-dependent matrix proteins, such as the complement system [58] . This reduces antibody quenching from matrix effects, thus maintaining more available antibody binding sites and increasing the monotonic range of the assay. Multiplex immunoassays are particularly susceptible to false positives arising from cross-reactivity [59] . Cross-reaction can occur in several different ways: antibody–antibody, antibody–antigen and antigen– antigen. In single-plex ELISAs, antibody pairs are carefully screened and selected, but the number of necessary tests rapidly increases with plexing. Juncker and coworkers performed a combinatorial analysis of possible cross-reacting pairs in a sandwich multiplex format [7] , and found that the number of liable pairs scales as: 4N(N-1) Equation 2
in an N-plex assay. Although these types of reactions may be rigorously screened and eliminated, the upper limit of screening at reasonable cost has been estimated at 50 [60] . To address cross-reactivity in planar arrays, ATPS have been employed to confine detection antibody solutions to the volume directly above their corresponding capture site [24] . Some patient samples contain human heterophilic antibodies, i.e., antibodies that bind to many of the animal IgGs used in immunoassays [61] . These
future science group
Review
heterophilic antibodies interfere with the immunoassay to cause nonlinearity. Classically, this is detected by measuring serial dilutions to check for linearity. However, protections from heterophilic antibody interference can be built into the assay. These measures include polyethylene glycol (PEG) precipitation [62] , specialized antibodies [63] and blocking additives [62] . Conclusion In this review, we described the principles and state-ofthe-art of multiplexing homogeneous sandwich immunoassays: methods of encoding and quantification for FRET, AlphaLISA and DNA proximity assays. We discussed the need for multiplexing to increase throughput of immunoassays for pharmacokinetics/pharmacodynamics, drug evaluation and disease diagnosis, as well as the advantages multiplexing provides in sample and reagent consumption. On the other hand, we also acknowledged some of the remaining challenges in the field of multiplex immunoassays: high-throughput validation, matrix interference, cross-reactivity and some of the workarounds to address these issues. The growing field of multiplex immunoassays offers many promising methods for high-throughput protein screening, and we expect multiplex immunoassays to assume a significant role in biology and healthcare. Future perspective Despite the advantages of multiplexed homogeneous immunoassays, they have seen little use in clinical trials. The authors speculate that this is a consequence of the limited multiplexing capabilities of these assays, until recently. While FPIA, FRET and LOCI have not yet progressed beyond two- to four-plex, the DNA proximity assays are commercially available up to 96-plex. Accordingly, those assays have been employed in clinical trials. In the near future (2015–2025), the authors anticipate expanded multiplexing capabilities for FPIA, FRET and LOCI. Considering the capabilities of quantum dots, FPIA and FRET could be quickly extended to five-plex or more. However, FPIA must overcome its current target size limitations before seeing more widespread use. With the quantum dot-graphene technique, Key terms Ligation: Technique to connect two DNA strands end-to-end. Matrix interference: Undesired, nonspecific interactions between assay components and biological material in the sample. Cross-reaction: Undesired interactions between assay components.
www.future-science.com
1553
Review Yamanishi, Han-Ching Chiu & Takayama FRET has greater flexibility. Although LOCI cannot borrow from quantum dot technology, discovery of novel chemiluminescent molecules that emit light at additional wavelengths in response to singlet oxygen could propel LOCI toward five- or ten-plex. Furthermore, each of these techniques is compatible with the ATPS spatial array technology, which could further extend the multiplexing capabilities of each assay. The rapid time-to-result and simplicity of these assays make them attractive tools for diagnostic biomarker panels. Meanwhile, DNA proximity techniques are poised for expanded use applications such as PK/PD and novel drug evaluation, which are not as time-sensitive.
The high multiplexing capabilities of the DNA proximity techniques increase their value for screening applications. Financial & competing interests disclosure S Takayama holds shares of PHASIQ, Inc, a company working on related technologies. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. No writing assistance was utilized in the production of this manuscript.
Executive summary Multiplex immunoassays • Multiplex protein detection is useful in drug screening, biological pathway analysis and diagnostic biomarker validation.
Advantages of homogeneous immunoassays • Homogeneous sandwich immunoassays are fast, no wash, specific, sensitive, robust to low affinity antibodies and often have wide dynamic range.
Multiplex homogeneous immunoassays • Fluorescence polarization immunoassay measures the change in intensity of polarized light from antigen stabilization by antibody. • Fluorescence resonance energy transfer uses a donor–acceptor system to only fluoresce when the antibody pair is in close proximity. • Quantum dots are used for their distinct optical properties in both fluorescence resonance energy transfer and fluorescence polarization immunoassay techniques. • The luminescent oxygen-channeling immunoassay uses a donor–acceptor system with singlet oxygen transfer to emit chemiluminescence when the pair is in close proximity. This technique was coupled with aqueous two-phase systems to spatially encode antibodies. • Proximity extension assays encode antibodies using DNA oligonucleotides, which hybridize only when the antibodies are in close proximity. Quantification is assessed through quantitative PCR.
Limitations & future outlook • Many multiplex homogeneous sandwich immunoassays are limited in multiplexing, suffer cross talk, require specialized readers and experience the hooking effect.
References
6
Hornbeck P. Enzyme-linked immunosorbent assays. In: Current protocols in immunology. John Wiley and Sons, Inc., NJ, USA, Chapter 2, Unit 2 1 (2001).
Papers of special note have been highlighted as: • of interest; •• of considerable interest
1554
1
Lemoine J, Fortin T, Salvador A, Jaffuel A, Charrier JP, Choquet-Kastylevsky G. The current status of clinical proteomics and the use of mrm and mrm(3) for biomarker validation. Expert Rev. Mol. Diagn. 12(4), 333–342 (2012).
7
Li H, Bergeron S, Juncker D. Microarray-to-microarray transfer of reagents by snapping of two chips for crossreactivity-free multiplex immunoassays. Anal. Chem. 84(11), 4776–4783 (2012).
2
Boja ES, Jortani SA, Ritchie J et al. The journey to regulation of protein-based multiplex quantitative assays. Clin. Chem. 57(4), 560–567 (2011).
8
Markov DA, Swinney K, Bornhop DJ. Label-free molecular interaction determinations with nanoscale interferometry. J. Am. Chem. Soc. 126(50), 16659–16664 (2004).
3
Diamandis EP. The failure of protein cancer biomarkers to reach the clinic: why, and what can be done to address the problem? BMC Med. 10, 87 (2012).
9
Gordon R, Sinton D, Kavanagh KL, Brolo AG. A new generation of sensors based on extraordinary optical transmission. Acc. Chem. Res. 41(8), 1049–1057 (2008).
4
Rifai N, Gillette MA, Carr SA. Protein biomarker discovery and validation: the long and uncertain path to clinical utility. Nat. Biotechnol. 24(8), 971–983 (2006).
10
5
Anderson NL. The clinical plasma proteome: a survey of clinical assays for proteins in plasma and serum. Clin. Chem. 56(2), 177–185 (2010).
Washburn AL, Luchansky MS, Bowman AL, Bailey RC. Quantitative, label-free detection of five protein biomarkers using multiplexed arrays of silicon photonic microring resonators. Anal. Chem. 82(1), 69–72 (2010).
11
Ji J, O’connell JG, Carter DJ, Larson DN. Highthroughput nanohole array based system to monitor
Bioanalysis (2015) 7(12)
future science group
Systems for multiplexing homogeneous immunoassays
multiple binding events in real time. Anal. Chem. 80(7), 2491–2498 (2008). 12
13
Zheng G, Patolsky F, Cui Y, Wang WU, Lieber CM. Multiplexed electrical detection of cancer markers with nanowire sensor arrays. Nat. Biotechnol. 23(10), 1294–1301 (2005). Wang ZH, Jin G. A label-free multisensing immunosensor based on imaging ellipsometry. Anal. Chem. 75(22), 6119–6123 (2003).
26
Shaw LM, Vanderstichele H, Knapik-Czajka M et al. Qualification of the analytical and clinical performance of CSF biomarker analyses in adni. Acta Neuropathol. 121(5), 597–609 (2011).
27
Qiao Y, Tang H, Munske GR, Dutta P, Ivory CF, Dong WJ. Enhanced fluorescence anisotropy assay for human cardiac troponin I and T detection. J. Flouresc. 21(6), 2101–2110 (2011).
28
Pakkila H, Malmi E, Lahtinen S, Soukka T. Rapid homogeneous immunoassay for cardiac troponin I using switchable lanthanide luminescence. Biosens. Bioelectron.62, 201–207 (2014).
14
Wang ZH, Meng YH, Ying PQ, Qi C, Jin G. A labelfree protein microfluidic array for parallel immunoassays. Electrophoresis 27(20), 4078–4085 (2006).
15
Xue C-Y, Khan SA, Yang K-L. Exploring optical properties of liquid crystals for developing label-free and high-throughput microfluidic immunoassays. Adv. Mat. 21(2), 198–202 (2009).
29
Arrebola MM, Lillo JA, Diez De Los Rios MJ et al. Analytical performance of a sensitive assay for cardiac troponin I with LOCI technology. Clin. Biochem. 43(12), 998–1002 (2010).
16
Michael KE, Vernekar VN, Keselowsky BG, Meredith JC, Latour RA, Garcia AJ. Adsorption-induced conformational changes in fibronectin due to interactions with well-defined surface chemistries. Langmuir 19(19), 8033–8040 (2003).
30
Schallmeiner E, Oksanen E, Ericsson O et al. Sensitive protein detection via triple-binder proximity ligation assays. Nat. Meth. 4(2), 135–137 (2007).
31
Schwenzer KS, Anhalt JP. Automated fluorescence polarization immunoassay for monitoring streptomycin. Antimicrob. Agents Chemother. 23(5), 683–687 (1983).
32
Filburn BH, Shull VH, Tempera YM, Dick JD. Evaluation of an automated fluorescence polarization immunoassay for vancomycin. Antimicrob. Agents Chemother. 24(2), 216–220 (1983).
33
Smith DS, Eremin SA. Fluorescence polarization immunoassays and related methods for simple, highthroughput screening of small molecules. Anal. Bioanal. Chem. 391(5), 1499–1507 (2008).
17
18
Armstrong JK, Wenby RB, Meiselman HJ, Fisher TC. The hydrodynamic radii of macromolecules and their effect on red blood cell aggregation. Biophys. J. 87(6), 4259–4270 (2004). Diercks AH, Ozinsky A, Hansen CL, Spotts JM, Rodriguez DJ, Aderem A. A microfluidic device for multiplexed protein detection in nano-liter volumes. Anal. Biochem. 386(1), 30–35 (2009).
19
Park J, Sunkara V, Kim TH, Hwang H, Cho YK. Lab-ona-disc for fully integrated multiplex immunoassays. Anal. Chem. 84(5), 2133–2140 (2012).
••
20
Tenje M, Xia H, Evander M et al. Acoustic trapping as a generic non-contact incubation site for multiplex bead-based assays. Anal. Chim. Acta 853, 682–688 (2015).
Overview of the fluorescence polarization immunoassay (FPIA) technology.
34
Ng AH, Uddayasankar U, Wheeler AR. Immunoassays in microfluidic systems. Anal. Bioanal. Chem. 397(3), 991–1007 (2010).
Hussain SA. An introduction to fluorescence resonance energy transfer (FRET). http://arxiv.org
35
Lam AJ, St-Pierre F, Gong Y et al. Improving FRET dynamic range with bright green and red fluorescent proteins. Nat. Meth. 9(10), 1005–1012 (2012).
36
Liu M, Zhao H, Quan X, Chen S, Fan X. Distanceindependent quenching of quantum dots by nanoscalegraphene in self-assembled sandwich immunoassay. Chem. Commun. 46(42), 7909–7911 (2010).
•
Novel innovation to resolve distance-dependent behavior that may serve as template to overcome assay restrictions.
37
Ullman EF, Kirakossian H, Switchenko AC et al. Luminescent oxygen channeling assay (LOCI): sensitive, broadly applicable homogeneous immunoassay method. Clin. Chem. 42(9), 1518–1526 (1996).
38
Leister KP, Huang R, Goodwin BL, Chen A, Austin CP, Xia M. Two high throughput screen assays for measurement of TNF-alpha in thp-1 cells. Curr. Chem. Genomics 5, 21–29 (2011).
39
Fredriksson S, Gullberg M, Jarvius J et al. Protein detection using proximity-dependent DNA ligation assays. Nat. Biotechnol. 20(5), 473–477 (2002).
40
Fredriksson S, Dixon W, Ji H, Koong AC, Mindrinos M, Davis RW. Multiplexed protein detection by proximity
21
22
23
Tian J, Zhou L, Zhao Y, Wang Y, Peng Y, Zhao S. Multiplexed detection of tumor markers with multicolor quantum dots based on fluorescence polarization immunoassay. Talanta 92, 72–77 (2012). Zeng Q, Zhang Y, Liu X, Tu L, Kong X, Zhang H. Multiple homogeneous immunoassays based on a quantum dotsgold nanorods fret nanoplatform. Chem. Commun. 48(12), 1781–1783 (2012).
24
Simon AB, Frampton JP, Huang NT, Kurabayashi K, Paczesny S, Takayama S. Aqueous two-phase systems enable multiplexing of homogeneous immunoassays. Technology 2(2), 176 (2014).
••
Explanation of luminescent oxygen-channeling immunoassay (LOCI) and aqueous two-phase systems (ATPS).
25
Assarsson ELM, Holmquist G, Bjo Rkesten J et al. Homogenous 96-plex pea immunoassay exhibiting high sensitivity, specificity, and excellent scalability. PLoS ONE 9(4), e95192 (2014).
•
Explanation of DNA proximity assays.
future science group
www.future-science.com
Review
1555
Review Yamanishi, Han-Ching Chiu & Takayama ligation for cancer biomarker validation. Nat. Meth. 4(4), 327–329 (2007). 41
Darmanis S, Nong RY, Hammond M et al. Sensitive plasma protein analysis by microparticle-based proximity ligation assays. Mol. Cell. Proteomics 9(2), 327–335 (2010).
42
ClinicalTrials Database: NCT01957241. http://ClinicalTrials.gov
43
ClinicalTrials Database: NCT02303886. http://ClinicalTrials.gov
•
Detailed explanation of fluorescence resonance energy transfer (FRET) theory and assay design including its difficulties.
44
Ellington AA, Kullo IJ, Bailey KR, Klee GG. Antibodybased protein multiplex platforms: technical and operational challenges. Clin. Chem. 56(2), 186–193 (2010).
45
Lee JW, Devanarayan V, Barrett YC et al. Fit-for-purpose method development and validation for successful biomarker measurement. Pharmaceut. Res. 23(2), 312–328 (2006).
•
Detailed guidelines for immunoassay characterization in an application-specific format.
46
Desilva B, Garofolo F, Rocci M et al. 2012 White paper on recent issues in bioanalysis and alignment of multiple guidelines. Bioanalysis 4(18), 2213–2226 (2012).
Lipman NS, Jackson LR, Trudel LJ, Weis-Garcia F. Monoclonal versus polyclonal antibodies: distinguishing characteristics, applications, and information resources. ILAR J./National Research Council, Institute of Laboratory Animal Resources 46(3), 258–268 (2005).
53
Schwenk JM, Lindberg J, Sundberg M, Uhlen M, Nilsson P. Determination of binding specificities in highly multiplexed bead-based assays for antibody proteomics. Mol. Cell. Proteomics: MCP 6(1), 125–132 (2007).
54
Kim JW, You J. Protein target quantification decision tree. Int. J. Proteomics 2013, 701247 (2013).
55
Wood WG. “Matrix effects” in immunoassays. Scand. J. Clin. Lab. Invest. Suppl. 205, 105–112 (1991).
56
Mire-Sluis AR, Barrett YC, Devanarayan V et al. Recommendations for the design and optimization of immunoassays used in the detection of host antibodies against biotechnology products. J. Immunol. Methods 289(1–2), 1–16 (2004).
57
Ermens AaM, Bayens AJM, Crooymans A, Broekman-Van Hout AaM, Van Duijnhoven HLP. Dilution protocols for detection of hook effects/prozone phenomenon. Clin. Chem. 46(10), 1719–1720 (2000).
58
Weinstock C, Schnaidt M. The complement-mediated prozone effect in the luminex single-antigen bead assay and its impact on HLA antibody determination in patient sera. Int. J. Immunogenet. 40(3), 171–177 (2013).
47
Tighe P, Negm O, Todd I, Fairclough L. Utility, reliability and reproducibility of immunoassay multiplex kits. Methods 61(1), 23–29 (2013).
59
48
Li D, Chiu H, Gupta V, Chan DW. Validation of a multiplex immunoassay for serum angiogenic factors as biomarkers for aggressive prostate cancer. Clin. Chim. Acta 413(19–20), 1506–1511 (2012).
Juncker D, Bergeron S, Laforte V, Li H. Cross-reactivity in antibody microarrays and multiplexed sandwich assays: shedding light on the dark side of multiplexing. Curr. Opin. Chem. Biol. 18, 29–37 (2014).
60
49
Shah VP, Midha KK, Dighe S et al. Analytical methods validation: bioavailability, bioequivalence, and pharmacokinetic studies. J. Pharm. Sci. 81, 309–312 (1992).
Perlee L, Christiansen J, Dondero R et al. Development and standardization of multiplexed antibody microarrays for use in quantitative proteomics. Proteome Sci. 2(1), 9 (2004).
61
50
Schnaidt M, Weinstock C, Jurisic M, Schmid-Horch B, Ender A, Wernet D. HLA antibody specification using single-antigen beads‐‐a technical solution for the prozone effect. Transplantation 92(5), 510–515 (2011).
Bjerner J, Bormer OP, Nustad K. The war on heterophilic antibody interference. Clin. Chem. 51(1), 9–11 (2005).
62
Bolstad N, Warren DJ, Nustad K. Heterophilic antibody interference in immunometric assays. Best Pract. Res. Clin. Endocrinol. Metab. 27(5), 647–661 (2013).
63
Bjerner J, Nustad K, Norum LF, Olsen KH, Bormer OP. Immunometric assay interference: incidence and prevention. Clin. Chem. 48(4), 613–621 (2002).
51
1556
52
Poetz O, Ostendorp R, Brocks B et al. Protein microarrays for antibody profiling: specificity and affinity determination on a chip. Proteomics 5(9), 2402–2411 (2005).
Bioanalysis (2015) 7(12)
future science group
