Electrokinetic sandwich assay and DNA mediated charge amplification

This short review article is based on the research paper titled “Electrokinetic sandwich assay and DNA mediated charge amplification for enhanced sensitivity and specificity”, authored by Siddharth Sourabh Sahu, Christiane Stiller, Elizabeth Paz Gomero, Ábel Nagy,  Amelie Eriksson Karlström, Jan Linnros, Apurba Dev

The research paper was published in Biosensors and Bioelectronics Journal on the 15th of March, 2021. The study demonstrates the proof of principle of an electrokinetic sandwich assay and explores the possibility to enhance the sensitivity and specificity through DNA mediation.

A pressure-driven flow controller was used to generate periodic pressure pulses in the liquid flowing through the capillary. A flow sensor was necessary to measure the flow rate.

ABSTRACT

In this study, the authors proposed, for the first time, the proof of concept of an electrokinetic immuno-sandwich assay, employing the  streaming current method for signal transduction. This technique offers better target selectivity and a linear concentration dependent response for a target concentration within the range 0.2–100 nM. Further, the possibility to enhance the signal by DNA-conjugation was explored both theoretically and with experimental validation. A clear and consistent enhancement of detection signal was observed as a function of DNA lengths. An application of this technique was demonstrated by carrying out target detection from a complex E. Coli lysate medium.

You can also read our application note about medium recirculation for dynamic cell culture!

Watch the webinar by BeNjamin Garlan, PhD about Microfluidic immunoassay systems for diagnostics!

INTRODUCTION

Biological targets like antibodies, proteins, DNA, extracellular vesicles can be detected through electrokinetic methods such as streaming current/potential. However, electrokinetic principles have a limited application in high salinity buffers due to  Coulomb screening, [1].

The authors have shown in a previous study [2] that the limit of detection varies a lot due to the size and the charge parameters of adsorbing particles that may either counteract or assist in electrokinetic sensing. 

In this article, the authors use the streaming current method for specific biomolecule detection to realize an electrical sandwich immunoassay, which is considered to be a gold standard for highly selective detection of a target molecule. 

The use of microfluidics was important to create periodic pressure pulses in the liquid through the capillary. The authors used a microfluidic flow controller as well as a flow sensor to measure the flow rate in the capillary.

AIM & OBJECTIVES

In short, this paper shows :

• a proof of principle of an electrokinetic sandwich assay by using IgG as a model analyte and Z-domain as both the capture and the detection probes
• the enhancement of  the sensitivity and specificity through DNA mediation by conjugating single-stranded DNA oligonucleotides to the detection probe. The authors modulated their zeta potential and used them to demonstrate that a better detection sensitivity can be achieved.
• an experimental application of the proposed method by detecting the target from a complex medium.

MATERIALS & METHODS

• DNA-conjugated Z-domain was prepared : The unmodified Z-domain was expressed and purified from E. coli cultures outlined in the authors’ earlier work [3].

• The inner surface of each capillary was functionalized using the protocol similarly as published previously [4].
• Fluidic and electrical measurements : To create periodic pressure pulses in the liquid flowing through the capillary, the authors used ELVEFLOWⓇ OB1 Mk3 pressure controller , as the flow sensor (MSF3) allowed them to measure the flow rate.

Starting with the recording of the 1st baseline, followed by the injection of the target until an equilibrium is reached. Thereafter, the 2nd baseline is recorded for a fixed duration. After the injection of the detection probe, the recording of the 3rd baseline follows. The signal of the sandwich assay is the difference indicated by the double-arrow. 

• The model system: two Z-domains as probes binding to the target trastuzumab : For the proof of concept study presented here, the trastuzumab-Z-domain affinity pair was utilized (Fig. 3, top). The trastuzumab (T) molecule is symmetrical with two identical binding sites for the Z-domain in the Fc region [5].

• Measurement of interaction affinity between T and various detection probes

KEY FINDINGS

In immunoassays, the detection of a target molecule often suffers from a low selectivity due to various non-specific interactions and cross-reactivity [6,7,8,9] of the affinity probes with non-target molecules. This is of particular concern when the target is to be detected from a complex medium e.g. plasma or serum. This is a major motivation to opt for sandwich assays if the target has multiple specific binding sites available. However there can be other challenges when dealing with two specific interactions instead of one with the target.

Concentration dependent measurements (Fig 4b and  4c) reveal two main contrasts between the direct and the sandwich assay in this study: (1) the magnitude of the signal of the direct assay is higher in comparison to the sandwich assay case and (2) the signal of the sandwich assay is in the opposite direction with respect to the direct assay – which means that the baseline shifts in opposite directions as a result of these steps. These two aspects are also reflected in the simulations (Fig 4d) that were carried out based on the model developed in the authors’ earlier work [2], and originate from the nature of the electrostatic interactions taking place. 

However, this does not mean that we have to compromise the sensitivity if we are to increase selectivity. The simulations shed light on another key aspect. Despite the disadvantage of a lower value of the signal in case of the sandwich assay, it is possible to enhance this signal by increasing the charge contrast between the probes and the target. Since the target in this case is positively charged, in order to raise the charge contrast, the authors increased the negative charge on the detection probes by conjugating them with DNA. Upon comparing Fig 4c and 5a one notices a consistent enhancement in the signal over the entire concentration range of T investigated, when using Z conjugated with 15 nt DNA instead of only Z as the detection probe.

Panel (a) shows an exemplary experimental data corresponding to the measurement sequence illustrated in Fig. 2, for 100 nM T. The zeta potential in the 2nd baseline is raised by the binding of a positively charged target, while the zeta potential in the 3rd baseline is lowered by the binding of the negatively charged detection probe. The difference between the 1st and 2nd baseline represents the direct assay response, while the difference between the 2nd and 3rd baseline embodies the sandwich assay response. It is worth noting that the direct assay response is markedly higher than the sandwich assay response. Concentration dependent responses obtained from a (b) direct assay and a (c) sandwich assay are shown as bar plots. The negative controls are shown as bar plots in (b) and dotted line in (c). Simulations comparing the signals from the direct assay with the sandwich assay are shown in (d) and demonstrate the possibility of signal enhancement by modulating the charge of the detection probe.

(a) T concentration dependent signal for the sandwich assay step when Z-DNA15 is used as the detection probe, with each signal enhanced at least 20% compared to using Z as the detection probe. The negative control is represented by the dotted line. (b) Comparison of the signals when different lengths of the DNA-conjugated Z are used as detection probes for 100 nM T. (c) SPR sensorgram for Z binding to a T surface along with affinity comparison between various detection probes (inset). (d) Response and various negative controls for detection of T from E. coli cell lysate diluted 100 times with 1× PBS. Negative controls involved injecting the lysate (C1) and lysate spiked with T (C1+T) on a surface with no immobilized capture probes, and injecting the lysate (C2) and then the detection probe Z-DNA30 (C2+Z-DNA30) on a surface immobilized with capture probes. 

The influence of molecular charge becomes even more apparent in the results obtained with Z conjugated to DNA oligonucleotides with different lengths (Fig. 5b). Changing the length of the conjugated DNA from 5 to 30 oligonucleotides is accompanied by a progressive enhancement in the signal. The signal with Z-DNA5 detection probe however is lower in comparison to the case when Z is used as the detection probe. This goes against the simulations and is likely to be caused by a reduction of the affinity of Z to T in the presence of DNA conjugate and/or steric effects, The comparison of interaction affinities for the different probe-target combinations (Fig. 5c) shows that the value of KD does indeed tend to increase as a result of DNA conjugation to Z. This means that despite the DNA conjugate affecting the interaction between Z and T, we still see a signal enhancement when a larger DNA length is used, due to the charge amplification. In principle, it could be possible to conjugate a very long DNA strand to the detection probe to further increase the signal, perhaps even exceeding the sensitivity as obtained from the direct assay case. 

CONCLUSION

The authors show a label-free and electrical sandwich immunoassay that uses electrokinetic streaming current method for the signal transduction. Simulations showed the possibility to enhance the signal by increasing the charge contrast between the target and the detection probe. This was implemented experimentally by using detection probes conjugated with DNA that is negatively charged, to detect positively charged targets. It was also shown that as the length of the conjugated DNA was increased, the signal was progressively enhanced, reaching upto 100% enhancement with 30 nt long DNA. Moreover, the improved specificity of the electrokinetic implementation of the sandwich immunoassay also demonstrated by detecting the target in a complex medium – the E. coli cell lysate. The study therefore is a step forward in developing highly selective and sensitive electrokinetic assays for possible application in clinical investigations.

Streaming current based measurements are extremely sensitive to the applied pressure and the flow of the sample through the sensor. The pressure-driven flow controlled microfluidics allowed the authors to setup, monitor and have a very accurate control over the pressure drop and the flow rate of the fluid through the sensor and was vital to this investigation. 

References