Perspective pubs.acs.org/ac

Isotachophoretic Phenomena in Electric Field Gradient Focusing: Perspectives for Sample Preparation and Bioassays Jos Quist, Paul Vulto, and Thomas Hankemeier* Division of Analytical Biosciences, Leiden Academic Centre for Drug Research (LACDR), Gorlaeus Laboratories, Einsteinweg 55, Leiden, 2333CC, The Netherlands Netherlands Metabolomics Centre (NMC), Leiden University, Einsteinweg 55, Leiden, South Holland 2333CC, The Netherlands ABSTRACT: Isotachophoresis (ITP) and electric field gradient focusing (EFGF) are two powerful approaches for simultaneous focusing and separation of charged compounds. Remarkably, in many EFGF methods, isotachophoretic hallmarks have been found, including observations of plateau concentrations and contiguous analyte bands. We discuss the similarities between ITP and EFGF and describe promising possibilities to transfer the functionality and applications developed on one platform to other platforms. Of particular importance is the observation that single-electrolyte isotachophoretic separations with tunable ionic mobility window can be performed, as is illustrated with the example of depletion zone isotachophoresis (dzITP). By exploiting the rapid developments in micro- and nanofluidics, many interesting combinations of ITP and EFGF features can be achieved, yielding powerful analytical platforms for sample preparation, biomarker discovery, molecular interaction assays, drug screening, and clinical diagnostics.

■

sotachophoresis (ITP) and electric field gradient focusing (EFGF) are two classes of methods which are capable of simultaneous separation and efficient focusing. Concentration factors often exceed thousand-fold and sometimes even millionfold for both ITP1−3 and EFGF methods.4−9 This Perspective presents a deeper look into both classes of methods, revealing similarities between ITP and EFGF at a fundamental level. As a result, the strengths of ITP and EFGF can be combined, providing powerful new possibilities for sample preparation and bioassays. Catalyzed by the rapid developments in the field of micro- and nanofluidics, this novel view can change the relative unpopularity of ITP and EFGF methods and may make them methods of first choice in many cases where high sensitivity and good specificity have to be combined. As ITP is one of the most difficult to understand among electrokinetic techniques, we will first discuss some basic principles of this powerful method. Next, EFGF will be briefly introduced. We present the observations of ITP-like phenomena in several EFGF methods including conductivity gradient focusing, electrocapture, bipolar electrode focusing, and micro/ nanofluidic concentration polarization devices. We discuss how insights from ITP can be used for the understanding of several phenomena in EFGF. The synergism between ITP and EFGF yields many promises for powerful new bioassays and other analytical platforms. For more comprehensive background information, we refer to the several useful introductory reviews on ITP10−12 Updates on recent developments in ITP have been frequently given by Gebauer et al.13−19 For EFGF technologies, the review by Shackmann and Ross provides an excellent introduction,20 but several more reviews are available.21−24

I

© 2014 American Chemical Society

ISOTACHOPHORESIS Isotachophoretic Condition. The word “isotachophoresis” contains the Greek words isos (equal) and takhos (speed), capturing the essence of the method: an ITP separation is forced toward a dynamic equilibrium in which all co-ions migrate with the same velocity. In this situation, the separation fulfills the so-called isotachophoretic condition. The electrophoretic velocity vi of an ion is determined by its ionic mobility μi and by the local electric field E: vi = μi E

(1)

Therefore, if ions with different ionic mobilities have the same velocity, they must be in regions with differing local electric field. Ions a, b, and c with ionic mobilities of μa < μb < μc will appear in zones A, B, and C in which μa,A EA = μ b,B E B = μc,C EC

(2)

Disturbances of the isotachophoretic condition (eq 2) are subject to a self-correcting mechanism (Figure 1). If any ion with mobility μb were transported to zone C (for example, by diffusion), the lower electric field would cause it to migrate with lower velocity than all the surrounding ions with higher mobility μc, until it returns in zone B. The inverse happens to the same ion placed in zone A. This self-correcting mechanism creates pure co-ion zones with sharply defined borders. The sharpness of these borders is determined by a balance of electromigration and dispersion: at lower electric fields, Received: November 19, 2013 Accepted: March 26, 2014 Published: March 26, 2014 4078

dx.doi.org/10.1021/ac403764e | Anal. Chem. 2014, 86, 4078−4087

Analytical Chemistry

Perspective

This equation tells us that the plateau concentration cb,B of ion b in zone B is in principle solely dependent on the concentration ca,A of ion a in zone A and on the mobilities of the ions involved. Equation 6 also may be derived from the Kohlrausch regulation function (KRF):25 zc KRF = ∑ i i μi (7) i where zi is the charge of ion i. The Kohlrausch regulation function is a conservation law: at each position in any electrokinetic separation, the value for KRF remains constant over time. For example, if LE is replaced by TE at a certain location in an ITP separation, TE concentration will adjust to match the local value of the Kohlrausch regulation function. A clear and more extensive discussion of the Kohlrausch regulation function and its limitations and of other conservation laws in electrophoresis has been provided by Hruška and Gaš.26 In many cases, analytes have insufficient quantities to reach plateau concentrations. Such analytes will form focused peaks between the zones of the ions with the nearest mobilities. This situation is often referred to as “peak mode ITP” and is contrasted to “plateau mode ITP”.27 Overlap between zones limits the capacities of ITP for separation purposes. For example, Kaniansky et al.28 demonstrated an ITP separation of 14 small molecules, which is a relatively high number. This can be contrasted to capillary zone electrophoresis, where electropherograms may show hundreds of peaks. To increase separation efficiency, ITP often is combined with capillary zone electrophoresis by means of transient ITP (see below). Zone overlap in ITP might be prevented using spacers, for example, nonfluorescing compounds which have intermediate mobility between fluorescent analytes. Forming plateau zones, these spacers can fully segregate adjacent analyte zones. Zone overlap also can be used very advantageously. For example, Bercovici et al. showed that hybridization of two different DNA strands could be accelerated over 10 000-fold by means of peak mode ITP because of the ability to concentrate the DNA molecules in a very confined volume.29 The plateau concentrations resulting from Kohlrausch regulation impose a limit to analyte preconcentration. Nevertheless, if analytes have very low starting concentrations and sufficient time is provided, peak mode ITP can be used to obtain extremely high concentration factors. Several articles report over 10 000-fold protein preconcentration,2,3,30 and under ideal conditions, million-fold preconcentration of a fluorophore has been achieved.1 The isotachophoretic condition only holds for completed ITP separations. There are several situations in which the dynamic equilibrium which is associated with the isotachophoretic condition is never reached. Incomplete ITP separations are characterized by the presence of mixed zones, which contain multiple co-ions with different mobilities. For example, sample might be injected continuously, resulting in continuously broadening zones. A well-known nonequilibrium ITP process is transient ITP. Several transient ITP alternatives exist.31 For example, analytes can be dissolved in a TE plug which is sandwiched between LE zones. Then, the analytes will be focused at the front end of the TE plug, while the back of the TE plug is dissolved by the faster LE ions. When the TE becomes completely dissolved, ITP focusing ceases and analytes will be separated by zone electrophoresis. tITP is widespread as a very useful method for inline sample

Figure 1. Self-correcting mechanism in ITP. Arrows represent electrophoretic velocities of ions. Profiles of electrical field and ion concentration profiles are indicated in the graph.

diffusion is more dominant, while higher electric fields lead to sharper zones. The isotachophoretic condition can be imposed on a separation by introducing a discontinuous electrolyte system consisting of a high-mobility leading electrolyte (LE) and a lowmobility trailing electrolyte (TE). Though they have lower mobility, TE ions cannot move slower than the LE in the sense that an electrolyte-free zone would form between the TE and the LE, because such a region would be rapidly filled with TE ions by the increased electric field. The TE and LE zones therefore must move with equal velocity, and the electric field and the TE concentration adjust accordingly. Analytes with intermediate mobility are focused between the LE and the TE. Usually, the TE and LE have a common counterion. The counterions travel in opposite directon and adjust to local TE, LE, and analyte concentrations to maintain electroneutrality. Ion concentrations directly influence conductivity and local electric field. Therefore, each isotachophoretic zone not only must have its own electric field but also must reach a corresponding plateau concentration. These plateau concentrations are dependent on LE concentration. ITP separations often can be readily recognized by a stair-like profile of contiguous plateau-shaped analyte zones. In the case of monovalent strong ions, a common counterion x, and pure isotachophoretic zones, plateau concentrations can be calculated assuming current continuity, electroneutrality, and the isotachophoretic condition. For separations in linear channels, the current in all zones is equal (IA = IB, considering only zones A and B). Therefore,

∑ Fc i,Aμi,A EA AA = ∑ Fc i,Bμi,BEBAB i

i

(3)

where F is the Faraday constant and A is the cross-section of the channel. As we deal with pure zones, we only have to sum the co-ion and the common counterion x. Therefore, assuming uniform cross-section, we have ca,Aμa,A EA + c x,Aμx,A EA = cb,Bμ b,B E B + c x,Bμx,B E B

(4)

Because monovalent strong ions are assumed, electroneutrality dictates equal concentrations of co- and counterions in each zone, simplifying eq 4 to ca,A(μa,A + μx,A )EA = cb,B(μ b,B + μx,B )E B

(5)

Invoking the isotachophoretic condition in eq 2 results in cb,B = ca,A

(μa,A + μx,A )μa,A (μ b,B + μx,B )μ b,B

(6) 4079

dx.doi.org/10.1021/ac403764e | Anal. Chem. 2014, 86, 4078−4087

Analytical Chemistry

Perspective

gradients ( μb. The figure shows how the ions obtain different focusing positions dependent on their ionic mobility. Analytes may have too low or too high ionic mobility to be focused on a gradient. These analytes are not trapped. EFGF Methods. Multiple techniques have been reported to induce an electric field gradient. An overview of the main EFGF methods is given in Figure 3. Koegler and Ivory32,33 focused and separated proteins in a dialysis tubing which was placed in a converging channel. The converging channel shaped the electric field while the dialysis tubing helped to maintain a pressure-driven counterflow with uniform velocity. Later, researchers have used a linear channel molded into a converging area filled with conductive polymer.8 We will refer to these and to similar approaches as “classical EFGF”. A variant of EFGF named conductivity gradient focusing (CGF) was shortly thereafter introduced by Greenlee and Ivory.34 This technique uses a salt gradient. Inglis et al. used a nanochannel which connected a high salt and a low salt reservoir for CGF.35 Dynamic field gradient focusing (DFGF) uses an array of individually controllable electrodes for detailed control of the electric field gradient.36 Temperature gradient focusing (TGF) is a versatile method combining a temperature-sensitive buffer with locally controlled heating to create an electric field gradient.9 Joule heating in a narrow channel section also can be used for TGF.937−39 The aforementioned EFGF methods can be characterized by the presence of shallow electric field gradients, typically spanning several millimeters. In several other methods, steep 4080

dx.doi.org/10.1021/ac403764e | Anal. Chem. 2014, 86, 4078−4087

Analytical Chemistry

Perspective

Figure 3. Overview of several EFGF methods. The schemes are examples; multiple embodiments have been described for most EFGF methods. Each of the methods is discussed in more detail in the text. Bulk flow direction is from right to left, and electrophoretic analyte migration is from left to right. The green bands indicate analyte focusing positions, where electrophoretic velocity is equal to bulk flow velocity. CP = concentration polarization.

inline labeling,65 and desalination.66 Moreover, massive parallelization of these devices also has been achieved.64,67

and to perform positioning of analyte zones. In DFGF, focused peaks could be positioned and the positions of focused proteins could even be swapped by locally changing the electric field gradient during the experiment.48 In TGF, Akbari et al. used a digital projector as a light source and heater which could be moved in order to position a focused fluorescein zone.53 In micro/nanofluidic concentration polarization devices, positioning is possible by tuning the balance of depletion zone growth and opposite bulkflow, as has been demonstrated in depletion zone isotachophoresis (dzITP).54 EFGF Applications. Practical demonstrations of EFGF applicability are limited in number but promising. For example, Balss et al. used TGF to perform two different DNA hybridization assays. In a first assay, a DNA target was focused and peptide nucleic acids (PNA) were introduced in the bulk flow, resulting in specific hybridization. In a second assay, bulk flow was varied to move focused PNA−DNA duplexes through the temperature gradient in order to measure their melting temperature.55 EC, which appears to be the only commercialized EFGF method,56 is capable of preconcentration, selective and sequential elution of trapped compounds, desalting, detergens removal, buffer exchange, and inline reactions, which are all very useful features with regard to the fact that this method can be online coupled with mass spectrometry (MS).57 Micro/nanofluidic concentration polarization devices have been used for enzyme assays,58−60 immunoassays,61−64

■

ITP PROCESSES IN EFGF METHODS EFGF and the Isotachophoretic Condition. In an EFGF experiment, analyte ions are focused on a fixed position on a gradient. This implies that upon completion of the focusing process all focused analytes have the same velocity. In the case of a stable gradient position, this velocity is equal to minus the bulk flow velocity. In other words, gradient focusing results in something to which the term iso-tacho-phoresis may be applied very literally. For EFGF, the isotachophoretic condition holds. We have seen that from this condition a number of notable characteristics can be derived: each co-ion will form a pure, sharply defined zone with a plateau concentration and with an electric field which is adjusted to the ionic mobility of the coion concerned. This would imply that analytes will form contiguous plateau-shaped zones similar to conventional ITP. Moreover, ITP tricks like spacer insertion and indirect detection by fluorescent tracers would be applicable to EFGF. One might remark that in iso-electric focusing (IEF), focused analytes also move with the same velocity (zero velocity). However, in IEF, analytes have zero net charge at the focusing points, while in ITP and EFGF the ions retain their charge when focused. ITP and EFGF are therefore subject to Kohlrausch regulation. In IEF, Kohlrausch regulation may be 4081

dx.doi.org/10.1021/ac403764e | Anal. Chem. 2014, 86, 4078−4087

Analytical Chemistry

Perspective

growing far beyond the range of the original gradient (Figure 4c). Evidence of ITP Phenomena in EFGF. We have analyzed the literature on EFGF for evidence on the hypothesis that ITP processes occur in EFGF. It turned out that important hallmarks of ITP have been reported in a surprisingly wide range of EFGF methods. For example, in an early paper on classical EFGF, Koegler and Ivory33 performed modeling of EFGF and showed that the electric field gradient evolved into a stair-like profile when sufficiently high concentrations of two model analytes were provided (Figure 5a). A stair-like electric field gradient is very characteristic for ITP. However, in their analysis, the corresponding analyte zones did not clearly have a plateau shape, though clearly deviate from a Gaussian profile. Nevertheless, in the experimental section of the same paper, an absorbance trace of separated myoglobins shows an analyte zone with a plateau-like shape. In conductivity gradient focusing, Greenlee and Ivory observed wide contiguous bands of BSA and hemoglobin. The border between these bands was very sharply defined, and the concentration throughout the bands appeared to be constant (Figure 5b). The authors noted that such bands are characteristic for ITP but viewed the isotachophoretic effect as undesired.34 Lin et al.69 studied finite sample effects in TGF (i.e., situations in which sample concentration is large enough to affect the electric field distribution) using a model based on a generalized Kohlrausch regulation function. Experimental results were used to verify some of their observations. In TGF, plateau shapes are not expected in concentration profiles, since the temperature gradient significantly affects analyte mobility. Velocity profiles however revealed a plateau-shaped zone which broadened over time. The study of Lin et al. did not include multiple analyte ions, which we predict to reveal multiple ITP-like zones.

a side effect near the focusing points, but it is not the principal focusing mechanism. Figure 4 schematically shows how ITP phenomena arise during an EFGF experiment according to our hypothesis. If

Figure 4. Transition from peak mode to plateau mode in EFGF. In peak mode (above), the electric field gradient is hardly affected by the focusing analytes. During the transition (middle), plateaus will form in the electric field gradient. After prolonged focusing in plateau mode, the analyte plateaus can grow far beyond the range of the original gradient (below).

analytes are present in low amounts, they will form focused peaks that cause negligible alterations of the electric field gradient. This situation corresponds to peak mode ITP (Figure 4a). When larger amounts of analyte are focused, they will locally alter the electric field. By local displacement of the electrolyte ions, the focusing analytes will reach a plateau concentration. This situation is similar to plateau mode ITP (Figure 4b). Continued focusing may result in plateau zones

Figure 5. Indications of ITP processes in several EFGF methods: (a) Koegler simulated the electric field yielding a stair-like pattern that is typical for ITP (image reprinted with permission from ref 33, Copyright 1996, John Wiley and Sons); (b) Greenlee found sharply defined contiguous analyte plateau zones in CGF (image reprinted with permission from ref 34, Copyright 1998, John Wiley and Sons); (c−e) In BEF, Laws found a separation of two focused zones by a continuously growing zone, indicating the insertion of a spacer (Adapted from the Supporting Information of ref 68. Copyright 2009 American Chemical Society). 4082

dx.doi.org/10.1021/ac403764e | Anal. Chem. 2014, 86, 4078−4087

Analytical Chemistry

Perspective

Figure 6. Functionality of dzITP. (a) Schematic representation of a H-shaped concentration polarization device as used for depletion zone isotachophoresis. (b) dzITP-separated zones at the border of a depletion zone. (c) Spatiotemporal plot of a dzITP separation using a discrete injection of fluorescein and 6-carboxyfluorescein and a continuous injection of acetate. The acetate acts as a nonfluorescent spacer, eluting focused 6carboxyfluorescein in upstream direction while fluorescein remains trapped at the border of the depletion zone. Images (a−c) reprinted from ref 54. Copyright 2011 American Chemical Society. (d) Selective enrichment of 6-carboxyfluorescein over fluorescein by ionic mobility filtering; acetate was used to establish the ionic mobility cutoff. (e) Selective trapping and indirect detection of urine compounds using fluorescein as a marker. Fl.: fluorescein. El.:electrolyte. The Roman numerals I−VI indicate putative analyte zones. Images (d) and (e) reprinted from ref 52. Copyright 2012 American Chemical Society. (f) Fluorescent aptamer assay for detection of several concentrations of IgE. The left fluorescent band arises from unbound aptamer; the band on the right arises from IgE-bound aptamer. The bands are spaced by nonfluorescent BSA protein. Image reprinted from ref 62. Copyright 2011 American Chemical Society.

For electrocapture, Astorga-Wells et al.70 compared flow velocities and measurements of local electric fields. If the background electrolyte is migrating with equal but opposite velocity with respect to the flow velocity, the ratio vflow/E should be equal to the mobility of the background electrolyte (see also eq 2). At low flow velocities, this was indeed the case for several background electrolytes, which points at stable zones of immobilized background electrolyte. Moreover, local electric field was independent of the externally applied voltage (using equal flow velocities), which is evidence of local electric field adjustment to the isotachophoretic condition: ions in plateau mode must have a constant value of the ratio vflow/E. Therefore, as long as the isotachophoretic condition holds and the flow velocity remains constant, the local electric field also must remain constant, even when increasing the external voltage. At higher flow velocities, the concentration polarization effects apparently broke down and the focusing condition was no longer present. In the same research, separation of analytes into adjacent zones was observed, although it was not clear whether these zones were peaks or plateaus. The authors explained their findings as being consistent with ITP.

In BEF, Hlushkou et al.71 observed both in experiments and simulations that a focusing analyte (bodipy disulfonate) reached a plateau concentration independent of the starting concentration. This plateau concentration was about five times lower than the electrolyte concentration (1 mmol/L TrisHCL). Simulated profiles of the electric field distribution revealed a growing plateau, which extended the electric field gradient. Although the authors do not mention ITP effects as a possible explanation, we do think their results are consistent with ITP. In another paper on BEF, Laws et al.68 used up to three fluorescent analytes, which they were able to separate. Although some of these separations appeared to occur in peak mode, one of the video’s in the Supporting Information of that paper clearly showed a rapidly broadening weakly fluorescent zone, which spaced some other analytes apart (Figure 5c−e). The length of this zone (>5 mm) stretched well beyond the predicted range of the electrode-induced electric field gradient. The fluorescence intensity in this zone appeared to be approximately constant. The authors did not comment on this zone in their paper, but it can be strongly suspected that it has an isotachophoretic nature. Finally, in a recent publication describing BEF focusing on the cationic species [Ru(bpy)3]2+, 4083

dx.doi.org/10.1021/ac403764e | Anal. Chem. 2014, 86, 4078−4087

Analytical Chemistry

Perspective

Table 1. Predicted and Demonstrated Features of ITP/EFGF Methodsa feature analyte plateau zones >10 000-fold preconcentration separation of proteins and peptides separation of nucleic acids separation of small molecules/ions dynamic control of analyte zone position dynamic control of ionic mobility window use of spacer compounds indirect analyte detection using tracers single electrolyte desalting enzyme assays immunoassays nucleic acid hybridization assays hyphenation with CZE hyphenation with MS

(transient) ITP d d d d d

d d d74 d75 d77 d d d

classical EFGF

CGF

DFGF

TGF

BEF

EC

d d d p d p d p p d

d p d p p p p p p

p p d p p d p p p d

p d d d d d d p p d

p p d d p

p p p p p

p p d p p

d d p p d d d73 p p d d79 p p p p p

p p p p p p

p p p p p

d p p d d d76 p p d78 d

CP devices and dzITP d d d p d d d d d d d d d p d p

a

EC = electrocapture. CP = concentration polarization. CZE = capillary zone electrophoresis. MS = mass spectrometry. d = demonstrated. p = predicted.

Cheow et al.62 used dzITP in ultrasensitive IgE and HIV-1 RT assays, using fluorescent aptamers for detection. BSA was used as a nonfluorescent spacer between bound and unbound aptamer, improving detection (Figure 6f).

the formation of a plateau-shaped analyte zone was also observed.72 dzITP. The clearest case of ITP effects during electric field gradient focusing is perhaps depletion zone isotachophoresis (dzITP). dzITP was recently developed in our lab using conventional H-shaped micro/nanofluidic CP devices (Figure 6a).54 At the upstream border of the depletion zone, analytes are focused and, if present in sufficient concentrations, plateau zones are readily formed (Figure 6b). A range of techniques that are typically associated with ITP have been implemented. For example, both discrete and continuous injections were performed for up to four fluorescent analytes, which were separated in ITP zones. Also the use of spacers was demonstrated. It was possible to elute a focused 6-carboxyfluorescein zone with a nonfluorescent acetate spacer toward the upstream reservoir, while keeping fluorescein focused at a stable position at the depletion zone border (Figure 6c). A number of relatively new operations was also demonstrated. For instance, we demonstrated the tunability of the ionic mobility window.52 Focused compounds could be released along the depletion zone in the downstream part of the channel. Similar to a valve that can be opened to several settings, the flux of released compound could be controlled. This was used for both pulsed and continuous release. In continuous mode, a balance of fluxes of released and supplied compound was used to establish filter action. A marker compound was partially released, defining the ionic mobility cutoff. Undesired compounds were coreleased while compounds in the desired ionic mobility window were trapped behind the marker compound zone. This principle was applied to selectively enrich 6-carboxyfluorescein over lower-mobility fluorescein despite having a 250× lower starting concentration; to achieve this, acetate was used as a nonfluorescent spacer for establishing the ionic mobility cutoff between fluorescein and 6carboxyfluorescein (Figure 6d). Additionally, using fluorescein as an ionic mobility cutoff marker, high-mobility metabolites in urine were selectively trapped and enriched; the fluorescein simultaneously allowed for indirect detection of plateau zones of the trapped analytes (Figure 6e).

■

IMPLICATIONS AND OPPORTUNITIES The above literature examples provide clear evidence of ITP effects in EFGF methods. Particularly in the case of dzITP, the two phenomena are strongly intertwined. One could consider ITP as a subclass of EFGF techniques or vice versa. Setting debates about definitions aside, it becomes obvious that the benefits of both classes of method are interchangeable. In the remainder of this paper, we will focus on the implications of this important observation. Apart from the typical leading/trailing electrolyte configuration, all hallmarks from the isotachophoretic principle have been observed in simulations and/or experimental research of EFGF methods. This includes the existence of analyte plateau concentrations that are independent of analyte starting concentrations but dependent on leading electrolyte concentration. It also includes adjustment of local electric fields to the isotachophoretic condition, the formation of contiguous analyte bands, and the observation of plateaus in electric field and analyte velocity. A thorough knowledge of ITP phenomena will therefore be of great help in understanding EFGF experiments. An important lesson from ITP is that zones can grow indefinitely if sufficient compound is present. Similarly, in EFGF, zones may grow beyond the range of the original gradient. Thus, if in EFGF plateau zones are allowed, peak capacity is not limited by the range of the gradient but rather by the length of the separation channel upstream from the gradient. A powerful ITP trick is the use of spacers. The baseline separation that may be obtained by the use of a spacer is not only very useful for detection but also for (bio)chemical assays. For example, undesired reactions may be prevented. Additionally, a reaction may be monitored by quantifying the amount of reaction product that is transferred across a spacer zone. Indirect detection by nonfocusing tracers, such as the counterspeeder and underspeeder concept originally developed for ITP,80 is also applicable to EFGF. 4084

dx.doi.org/10.1021/ac403764e | Anal. Chem. 2014, 86, 4078−4087

Analytical Chemistry

Perspective

hyphenation with mass spectrometry (MS). Electrocapture and concentration polarization devices have shown powerful performance in desalting of samples, an operation that is highly important to prevent ion suppression in electron-spray mass spectrometry. Furthermore, dzITP has shown that focused analyte bands can be selectively released, which is expected to enable analyte transport to the electrospray emitter on a bandby-band basis. This will also enhance sensitivity of MS-based detection. The clearest demonstration of this potential so far has been demonstrated in electrocapture.81 For specific applications, including the analysis of metabolites from single cells or from other ultrasmall samples, these advantages will make hyphenated EFGF/ITP-MS methods superior to CE-MS, HPLC-MS, and direct infusion MS techniques. For microfluidic setups, the biggest opportunities lie in the field of molecular interaction assays. For such type of assays, the fact that one can trap molecular species in solution reduces surface interactions that are a typical disturbance in solid-phase immobilized assays. The fact that molecular compounds of lower mobility or opposite charge can be flushed through focused analyte zones80 may increase binding and interaction efficiencies dramatically. Similar opportunities exist for multiple reagents that are concentrated in overlapping peak mode zones. Quantification through measurement of zone growth enables real-time determination of binding constants and enzyme activity. Such measurements are supported by the positional stability of many EFGF separations. A powerful example is the aptamer dzITP assay presented by Cheow et al.62 When adding functionality like selective ionic mobility filtering and the use of spacers and tracers for indirect detection, a molecular interaction platform is obtained which has unprecedented versatility compared to other separation methods. We envision many impactful biochemical assays, including protein/protein and protein/nucleic acid interaction assays for determining affinity interaction and binding constants; enzyme activity assays, among which protein kinase/inhibitor assays are important as many kinases play an important role in cancer and other diseases; immunoassays; metabolite/protein interaction platforms for drug discovery purposes; and nucleic acid hybridization assays. These assays, being made sensitive and specific, do not necessarily require complicated instrumentation like MS or NMR but will become available as stand-alone benchtop instruments for rapid diagnostics in, e.g., the clinic. By integrating microelectronics and miniaturized detection technology, ITP/EFGF methods will ultimately be available as hand-held analyzers77 that may be used for water and food quality monitoring as well as for point-of-care diagnostics.

Vice versa, EFGF offers important additional advantages to ITP-like separations. For example, the positional stability of analyte zones allows one to monitor zone broadening and to study physicochemical processes at ITP zone interfaces. Zone positioning may be used to scan analyte zones repeatedly along a sensor. The tunable ionic mobility window makes it possible to release analyte zones selectively for downstream analyses or to selectively enrich low-abundant species. Table 1 lists a number of features that is demonstrated or predicted in ITP and EFGF methods. Most demonstrated features have been discussed in the previous sections; if not so, they have been referenced. The predicted features have been extrapolated from observations of similar features in other EFGF methods and may require adjustments of existing methods. In some cases, a feature might not be available due to intrinsic limitations of a method. From Table 1, a number of important observations can be made. First, all features demonstrated in conventional ITP have also been demonstrated in one or more EFGF methods. Concentration polarization devices (including dzITP) stand out in this regard. Second, as discussed above, EFGF methods have several features that are not available in conventional ITP, providing important extra functionality and simplicity of use. Third, once the predicted features are realized, these features might be combined at will, making each EFGF method a very versatile toolbox for bioanalysis. The development of combined ITP/EFGF methods will face numerous challenges. First, in most cases, the required technology is not directly available. For miniaturized separations, custom-made chips have to be produced, and the classical problem of world-to-chip interfacing has to be addressed. Second, EFGF techniques are still awaiting optimization of experimental conditions such as pH, the type and concentration of electrolyte, and coating procedures. Conditions have to be benchmarked for different types of analytes. Third, quantitation experiments must take differences in ionic mobilities into account because these result in focusing rates which are different per analyte. Fourth, when EFGF is followed by a zone electrophoresis step or by a transport or flushing step, the transient electric field gradient may cause large conductivity changes in the separation system, resulting in complicated behavior. Most of these challenges will not be too difficult to overcome, because many clues can be found in the large body of electrophoresis literature. Moreover, the advantages for sensitivity and selectivity are significant. We expect therefore that ITP/EFGF methods will have many breakthrough applications in diverse fields of biology, including omics studies, biomarker discovery, drug discovery, and diagnostics. So far, all methods described above have been implemented either as a microfluidic setup, a capillary setup, or preparative scale setup. We expect particular benefit from the microfluidic and capillary setups in distinctive application fields. For a capillary ITP/EFGF setup, we foresee the highest impact in hyphenated settings. The use of transient ITP to enhance capillary zone electrophoresis separations has already been proven to be a powerful technique. Increased usability is expected as the leading/trailing electrolyte system is replaced with a single electrolyte system, simplifying sample preparation and injection procedures. In addition, the concentration and cleanup capabilities of such a system may add to the reproducibility and sensitivity of capillary zone electrophoresis separations. Another promising application will be direct

■

CONCLUSION A large number of EFGF and ITP techniques has been developed to-date, and a wide range of applications and functionality has been demonstrated. Although regarded in the field as two distinct classes of methods, many hallmarks of ITP have been observed in EFGF techniques. This includes the formation of plateau zones and contiguous analyte bands. We believe that a classical leading/trailing electrolyte setup is only one of the many implementation forms to achieve an ITP separation, which could in fact be achieved with any EFGF technique. Our view means a shift in convention that has important implications for the functionality of the two techniques. Integration of EFGF and ITP on a single platform provides a powerful combined toolbox which includes, among others, ultraefficient and selective preconcentration, separation, 4085

dx.doi.org/10.1021/ac403764e | Anal. Chem. 2014, 86, 4078−4087

Analytical Chemistry

Perspective

(29) Bercovici, M.; Han, C. M.; Liao, J. C.; Santiago, J. G. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 11127−11132. (30) Bottenus, D.; Jubery, T. Z.; Dutta, P.; Ivory, C. F. Electrophoresis 2011, 32, 550−562. (31) Bahga, S. S.; Santiago, J. G. Analyst 2013, 138, 735−754. (32) Koegler, W. S.; Ivory, C. F. J. Chromatogr., A 1996, 726, 229− 236. (33) Koegler, W. S.; Ivory, C. F. Biotechnol. Prog. 1996, 12, 822−836. (34) Greenlee, R. D.; Ivory, C. F. Biotechnol. Prog. 1998, 14, 300− 309. (35) Inglis, D. W.; Goldys, E. M.; Calander, N. P. Angew. Chem., Int. Ed. 2011, 50, 7546−7550. (36) Huang, Z.; Ivory, C. F. Anal. Chem. 1999, 71, 1628−1632. (37) Kim, S. M.; Sommer, G. J.; Burns, M. A.; Hasselbrink, E. F. Anal. Chem. 2006, 78, 8028−8035. (38) Ge, Z.; Yang, C.; Tang, G. Int. J. Heat Mass Transfer 2010, 53, 2722−2731. (39) Sommer, G. J.; Kim, S. M.; Littrell, R. J.; Hasselbrink, E. F. Lab Chip 2007, 7, 898−907. (40) Dhopeshwarkar, R.; Hlushkou, D.; Nguyen, M.; Tallarek, U.; Crooks, R. M. J. Am. Chem. Soc. 2008, 130, 10480−10481. (41) Perdue, R. K.; Laws, D. R.; Hlushkou, D.; Tallarek, U.; Crooks, R. M. Anal. Chem. 2009, 81, 10149−10155. (42) Pu, Q.; Yun, J.; Temkin, H.; Liu, S. Nano Lett. 2004, 4, 1099− 1103. (43) Schoch, R. B.; Han, J.; Renaud, P. Rev. Mod. Phys. 2008, 80, 839−883. (44) Zangle, T. A.; Mani, A.; Santiago, J. G. Chem. Soc. Rev. 2010, 39, 1014−1035. (45) Kim, S. J.; Song, Y. A.; Han, J. Chem. Soc. Rev. 2010, 39, 912− 922. (46) Quist, J.; Trietsch, S. J.; Vulto, P.; Hankemeier, T. Lab Chip 2013, 13, 4810−4815. (47) Kuo, C. H.; Wang, J. H.; Lee, G. B. Electrophoresis 2009, 30, 3228−3235. (48) Burke, J. M.; Huang, Z.; Ivory, C. F. Anal. Chem. 2009, 81, 8236−8243. (49) Knust, K. N.; Sheridan, E.; Anand, R. K.; Crooks, R. M. Lab Chip 2012, 12, 4107−4114. (50) Hoebel, S. J.; Balss, K. M.; Jones, B. J.; Malliaris, C. D.; Munson, M. S.; Vreeland, W. N.; Ross, D. Anal. Chem. 2006, 78, 7186−7190. (51) Astorga-Wells, J.; Vollmer, S.; Tryggvason, S.; Bergman, T.; Jörnvall, H. Anal. Chem. 2005, 77, 7131−7136. (52) Quist, J.; Vulto, P.; van der Linden, H.; Hankemeier, T. Anal. Chem. 2012, 84, 9065−9071. (53) Akbari, M.; Bahrami, M.; Sinton, D. Microfluid. Nanofluid. 2012, 12, 221−228. (54) Quist, J.; Janssen, K. G. H.; Vulto, P.; Hankemeier, T.; van der Linden, H. J. Anal. Chem. 2011, 83, 7910−7915. (55) Balss, K. M.; Ross, D.; Begley, H. C.; Olsen, K. G.; Tarlov, M. J. J. Am. Chem. Soc. 2004, 126, 13474−13479. (56) http://www.epizell.com/service_electrocapture.html. (57) Shariatgorji, M.; Astorga-Wells, J.; Ilag, L. Anal. Bioanal. Chem. 2011, 399, 191−195. (58) Lee, J. H.; Song, Y.-A.; Tannenbaum, S. R.; Han, J. Anal. Chem. 2008, 80, 3198−3204. (59) Lee, J. H.; Cosgrove, B. D.; Lauffenburger, D. A.; Han, J. J. Am. Chem. Soc. 2009, 131, 10340−10341. (60) Sarkar, A.; Han, J. Lab Chip 2011, 11, 2569−2576. (61) Wang, Y.-C.; Han, J. Lab Chip 2008, 8, 392−394. (62) Cheow, L. F.; Han, J. Anal. Chem. 2011, 83, 7086−7093. (63) Lee, J. H.; Han, J. Microfluid. Nanofluid. 2010, 9, 973−979. (64) Ko, S. H.; Kim, S. J.; Cheow, L. F.; Li, L. D.; Kang, K. H.; Han, J. Lab Chip 2011, 11, 1351−1358. (65) Wang, C.; Ouyang, J.; Ye, D.-K.; Xu, J.-J.; Chen, H.-Y.; Xia, X.H. Lab Chip 2012, 12, 2664−2671. (66) Kim, S. J.; Ko, S. H.; Kang, K. H.; Han, J. Nat. Nanotechnol. 2010, 5, 297−301. (67) Lee, J. H.; Song, Y.-A.; Han, J. Lab Chip 2008, 8, 596−601.

tunable analyte focusing windows, requirement of a single electrolyte only, and the ability to stably trap analytes and execute molecular interaction assays. Among the most promising applications are hyphenation with MS for -omics studies and biomarker discovery, as well as a wide range of molecular interaction assays for drug screening and clinical diagnostics.

■

AUTHOR INFORMATION

Corresponding Author

*Fax: 31 71 527 4565. Tel: +31 71 527 4220. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This project was financed by The Netherlands Metabolomics Centre (NMC), The Netherlands Genomics Initiative (NGI), and The Netherlands Organization for Scientific Research (NWO).

■

REFERENCES

(1) Jung, B.; Bharadwaj, R.; Santiago, J. G. Anal. Chem. 2006, 78, 2319−2327. (2) Bottenus, D.; Jubery, T. Z.; Ouyang, Y.; Dong, W.-J.; Dutta, P.; Ivory, C. F. Lab Chip 2011, 11, 890−898. (3) Wang, J.; Zhang, Y.; Mohamadi, M. R.; Kaji, N.; Tokeshi, M.; Baba, Y. Electrophoresis 2009, 30, 3250−3256. (4) Wang, Y.-C.; Stevens, A. L.; Han, J. Anal. Chem. 2005, 77, 4293− 4299. (5) Kim, S. M.; Burns, M. A.; Hasselbrink, E. F. Anal. Chem. 2006, 78, 4779−4785. (6) Anand, R. K.; Sheridan, E.; Knust, K. N.; Crooks, R. M. Anal. Chem. 2011, 83, 2351−2358. (7) Kim, M.; Jia, M.; Kim, T. Analyst 2013, 138, 1370−1378. (8) Humble, P. H.; Kelly, R. T.; Woolley, A. T.; Tolley, H. D.; Lee, M. L. Anal. Chem. 2004, 76, 5641−5648. (9) Ross, D.; Locascio, L. E. Anal. Chem. 2002, 74, 2556−2564. (10) Holloway, C. J.; Trautschold, I. Fresenius’ J. Anal. Chem. 1982, 311, 81−93. (11) Petr, J.; Maier, V.; Horáková, J.; Ševcík, J.; Stránský, Z. J. Sep. Sci. 2006, 29, 2705−2715. (12) Garcia-Schwarz, G.; Rogacs, A.; Bahga, S. S.; Santiago, J. G. J. Visualized Exp. 2012, 61, 3890. (13) Gebauer, P.; Boček, P. Electrophoresis 1997, 18, 2154−2161. (14) Gebauer, P.; Boček, P. Electrophoresis 2000, 21, 3898−3904. (15) Gebauer, P.; Boček, P. Electrophoresis 2002, 23, 3858−3864. (16) Gebauer, P.; Malá, Z.; Boček, P. Electrophoresis 2007, 28, 26−32. (17) Gebauer, P.; Malá, Z.; Boček, P. Electrophoresis 2009, 30, 29−35. (18) Gebauer, P.; Malá, Z.; Boček, P. Electrophoresis 2011, 32, 83−89. (19) Malá, Z.; Gebauer, P.; Boček, P. Electrophoresis 2013, 34, 19−28. (20) Shackman, J. G.; Ross, D. Electrophoresis 2007, 28, 556−571. (21) Ivory, C. Sep. Sci. Technol. 2000, 35, 1777−1793. (22) Kelly, R. T.; Woolley, A. T. J. Sep. Sci. 2005, 28, 1985−1993. (23) Meighan, M. M.; Staton, S. J. R.; Hayes, M. A. Electrophoresis 2009, 30, 852−865. (24) Vyas, C. A.; Flanigan, P. M.; Shackman, J. G. Bioanalysis 2010, 2 (4), 815−827. (25) Kohlrausch, F. Ann. Phys. 1897, 298, 209−239. (26) Hruška, V.; Gaš, B. Electrophoresis 2007, 28, 3−14. (27) Khurana, T. K.; Santiago, J. G. Anal. Chem. 2008, 80, 6300− 6307. (28) Kaniansky, D.; Masár, M.; Bielčiková, J.; Iványi, F.; Eisenbeiss, F.; Stanislawski, B.; Grass, B.; Neyer, A.; Jöhnck, M. Anal. Chem. 2000, 72, 3596−3604. 4086

dx.doi.org/10.1021/ac403764e | Anal. Chem. 2014, 86, 4078−4087

Analytical Chemistry

Perspective

(68) Laws, D. R.; Hlushkou, D.; Perdue, R. K.; Tallarek, U.; Crooks, R. M. Anal. Chem. 2009, 89, 8923−8929. (69) Lin, H.; Shackman, J. G.; Ross, D. Lab Chip 2008, 8, 969−978. (70) Astorga-Wells, J.; Vollmer, S.; Bergman, T.; Jörnvall, H. Anal. Chem. 2007, 79, 1057−1063. (71) Hlushkou, D.; Perdue, R. K.; Dhopeshwarkar, R.; Crooks, R. M.; Tallarek, U. Lab Chip 2009, 9, 1903−1913. (72) Sheridan, E.; Hlushkou, D.; Knust, K. N.; Tallarek, U.; Crooks, R. M. Anal. Chem. 2012, 84, 7393−7399. (73) Anand, R. K.; Sheridan, E.; Hlushkou, D.; Tallarek, U.; Crooks, R. M. Lab Chip 2011, 11, 518−527. (74) Shkolnikov, V.; Bahga, S. S.; Santiago, J. G. Phys. Chem. Chem. Phys. 2012, 14, 11534−11545. (75) Bruchelt, G.; Buedenbender, M.; Schmidt, K.-H.; Jopski, B.; Treuner, J.; Niethammer, D. J. Chromatogr., A 1991, 545, 407−412. (76) Astorga-Wells, J.; Bergman, T.; Jörnvall, H. Anal. Chem. 2004, 76, 2425−2429. (77) Kaigala, G.; Bercovici, M.; Behnam, M.; Elliott, D.; Santiago, J.; Backhouse, C. Lab Chip 2010, 10, 2242−2250. (78) Astorga-Wells, J.; Swerdlow, H. Anal. Chem. 2003, 75, 5207− 5212. (79) Knust, K. N.; Hlushkou, D.; Anand, R. K.; Tallarek, U.; Crooks, R. M. Angew. Chem., Int. Ed. 2013, 52, 8107−8110. (80) Chambers, R. D.; Santiago, J. G. Anal. Chem. 2009, 81, 3022− 3028. (81) Astorga-Wells, J.; Jörnvall, H.; Bergman, T. Anal. Chem. 2003, 75, 5213−5219.

4087

dx.doi.org/10.1021/ac403764e | Anal. Chem. 2014, 86, 4078−4087