Quantitation without calibration: response profile as an indicator of

Jun 19, 2018 - One of the major challenges in bi-omarker's quantitation is the need to have a calibration for correlat-ing a measured signal to a targ...
0 downloads 0 Views 655KB Size
Subscriber access provided by TUFTS UNIV

Letter

Quantitation without calibration: response profile as an indicator of target amount Mrittika Debnath, Jessica M Farace, Kristopher D Johnson, and Irina Nesterova Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b02053 • Publication Date (Web): 19 Jun 2018 Downloaded from http://pubs.acs.org on June 19, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Quantitation without calibration: response profile as an indicator of target amount Mrittika Debnath, Jessica M. Farace, Kristopher D. Johnson, and Irina V. Nesterova* Department of Chemistry and Biochemistry, Northern Illinois University, DeKalb, IL 60115, USA * Corresponding author: 433 LaTourette Hall, Northern Illinois University, DeKalb, IL 601115, USA; Phone: +1-815-753-6843; e-mail: [email protected]

ABSTRACT: Quantitative assessment of biomarkers is essential in numerous contexts from decision-making in clinical situations to food quality monitoring to interpretation of life-science research findings. However, appropriate quantitation techniques are not as widely addressed as detection methods. One of the major challenges in biomarker’s quantitation is the need to have a calibration for correlating a measured signal to a target amount. The step complicates the methodologies and makes them less sustainable. In this work we address the issue via a new strategy: relying on position of response profile rather than on an absolute signal value for assessment of a target’s amount. In order to enable the capability we develop a target-probe binding mechanism based on a negative cooperativity effect. A proof-of-concept example demonstrates that the model is suitable for quantitative analysis of nucleic acids over a wide concentration range. The general principles of the platform will be applicable towards a variety of biomarkers such as nucleic acids, proteins, peptides and others.

Accurate information on amounts of biomolecules is a valuable asset in a variety of situations; however, their quantitation is not as widely addressed as detection. While there are some established approaches (i.e. quantitative PCR for nucleic acids or ELISA for proteins), they are, typically, time consuming, expensive, and require professional personnel. One of the reasons behind such complexity is that in order to quantify a target, a signal magnitude generated by an unknown sample needs to be correlated to some kind of calibration (Figure 1A). However, any calibration comes limitations such as an increased cost, need of a pure quantitatively characterized standard of a target (or its proxy), matching the calibration conditions/environment to the target assay to mitigate matrix effects, and/or others. Herein we report a new general concept for the quantitative assessment of biomolecules that obviates the need for calibration on example of nucleic acids as a proof-of-concept model. Nucleic acids per se are indispensable biomarkers for assessing variety of conditions ranging from routine food quality monitoring to clinical diagnostics to life sciences research.1-6 In those contexts, not only qualitative but also quantitative information on specific oligonucleotides has a practical significance: for example, concentrations of specific DNA/RNAs are important clinical indicators;4, 7-9 quantitative information on certain pathogen genotypes is essential for public health risk assessment,10-12 accurate quantitation of genetically-modified organisms is a subject of legislative regulations.13 Therefore, the availability of reliable quantitative methods that are specific, robust, time-efficient and inexpensive is essential.

A number of techniques for detection of nucleic acids has been established over the recent decades;4 however, quantitation of those remain much more challenging. 14 The golden standard for specific nucleic acid quantitation is qPCR,12 a method that does provide information on a target amount; however, is rather expensive, time-consuming, requires specific instrumentation, and, after all, suffers from matrix effects affecting quantitation reliability.13 While less expensive, other approaches (i.e. microarrays, fluorescent barcodes, in-situ hybridizations) are less reliable, especially, when it concerns discerning accurate quantitative information.4

Figure 1. (A) A regular approach for obtaining a quantitative information on a target involves assessment of a signal from a sample and correlating it to the target’s amount via some kind of an established calibration (Approach 1). (B, C) An alternative approach (Approach 2) can be based on assessing an unknown target’s response profile and obtaining quantitative information from a position of a response profile feature such as peak maximum (B) or inflection point (C).

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 5

Figure 2. In the quantitation approach, two probes bind the same target with similar (or identical) affinity but with negative cooperativity effect. The negative cooperativity causes binding of one probe to decrease target’s affinity towards another probe (A). (B) If an unknown amount of target is exposed to a varying concentrations of probes (present equimolarly), 1:1 target:probe hybrids will preferentially form. The 1:2 target:probe hybrids will form only when the total probe concentration exceeds target concentration (i.e. probe concentration above 6 in the example). Assuming signal transduction mechanism that is capable to discern 1:2 hybrids, the signal inflection on response profile (“inflection point”) with respect to probe concentration indicates the target concentration. (C) Probes for an oligonucleotide target are designed the way that while a target can bind both probes, the binding of each individual probe decreases binding affinity of another probe (due to binding region overlap, blue shaded box). Oligonucleotide sequences are included in Table S1. We hypothesized that relying the target’s quantitative assessment on the position of its response profile (Figure 1B) instead of an absolute signal value (Figure 1A) should improve the overall reliability of the measurements. For example, in case of a target at the concentration 2 in Figure 1B, even if experimental factors affect the signal magnitude, the target’s quantity can be reliably determined from the position of its response profile. However, developing a methodology that positions response profile as a function of target’s quantity is rather challenging. In order to do it, one needs to establish a target-probe binding model that yields a response profile that uniquely defines the target amounts. To the best of our knowledge, no one realized such capability up-to-date. In this work, we establish an approach for absolute quantitation of nucleic acids based on position of inflection point as an indicator of target amount (Figure 1C). In order to unable the capability we deliberately designed a sensing system based on a mechanism for negative cooperativity with limited receptor concentration.15 In this system, two probes bind target with similar affinity. Each probe can bind the target separately or both probes can bind it simultaneously. However, by design, the binding of one probe to target decreases affinity of another (negative cooperativity effect) (Figure 2A). It means that in the presence of target and both probes, formation of 1:1 target:probe hybrid is a more favorable than formation of 1:2 target:probe hybrid. Therefore, if the target is exposed to various amounts of probes present equimolarly, 1:1 target:probe hybrids preferentially form when the total concentration of the probes is below than the concentration of target (Figure 2B, total probe concentration of 6 and below). However, when the total concentration of probes exceeds concentration of target, 1:2 target: probe hybrids start forming (Figure 2B, total probe concentration above 6). In this case, if there is a signal transduction pathway that discerns 1:2 target: probe hybrids over the tar-

get, unbound probes and 1:1 target:probe hybrids, the change in the signal (inflection point) indicates that the total concentration of probes exceeds target’s concentration (Figure 2B, graph, theoretical considerations in Figure S5). Detection of this point manifests the unknown target concentration. For the proof-of-concept, we designed a sensing system for detection of single stranded DNA (based on the nucleic acid system reported in ref. 15). Overlapping binding regions of the two probes imposes the negative cooperativity effect: when one probe is bound to the target (Figure 2C), binding affinity of another probe decreases. To enable separation-free signal transduction, both probes are labeled with fluorophores (FRET pair) the way that FRET is efficient only when both probes are in proximity (i.e. in 1:2 target:probe hybrid, Figure 3B). In this case, the inflection point in signal (donor/acceptor emission ratio) indicates that the probes’ total concentration exceeds the concentration of target. First, we assessed fluorescent signal from systems containing different amounts of target over ranges of probes concentrations (Figures 3A and S1, experimental details in SI). As expected,15 we do observe sigmoid response curves over the total probe concentration overlapping target’s. By the approach design (Figure 2B), the inflection point between initial plateau and slope regions of the curve should correlate with target concentration. Clearly, the positions of the inflection points for different target concentrations do visually correlate with target quantities. We observe the pattern for the range of target concentrations exceeding two orders of magnitude (Figures 3A and S1). Importantly, there are no changes in signal in the absence of target and when random DNA instead of target is present (control experiments in Figures S2-S4).

2 ACS Paragon Plus Environment

Page 3 of 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 3. (A) Response profiles of target exposed to changing probe concentrations are characteristic sigmoidal curves with inflection point visually corresponding to the target concentration (Figure S1 contains the same response profiles without signal normalization). (B) We use changes in FRET rates as an indicator of 1:2 target/probe hybrid formation. (C) Two methods for discerning value of inflection point yield similar results. The results agree well with target contents over a wide concentration range. %RSDs over multiple measurements of inflection point (n = 3 − 8) are ≤ 15%. Colors in table in panel C are synchronized with curves in panel A. Further, beyond the visual confirmation of correlation between position of response profile and target amount, we looked into a more formal ways to establish the inflection point and, consequently, target quantity. While the data can be fit with sigmoidal fits, the fit does not yield the inflection point and, therefore, the quantitation information. Therefore, we have to establish a procedure for deriving inflection point. To do it, we evaluated two procedures based on the inflection point’s physical meaning. Method 1 defines the point as a concentration at which the signal exceeds “initial plateau − standard error” value (Figure 3C, left column. Method 2 defines the point as an intersection of a line extrapolating initial plateau with an extrapolation of transition’s slope (Figure 3C, right column). As a result, our data yield accurate values for target quantities over a concentration range exceeding two orders of magnitude regardless which method is used to discern the inflection point (Figure 3C). We observe the highest deviations from expected values for the lowest (10 nM) target concentration. We speculate that the need to measure very low signals (close to limit of regular spectrofluorimeter) is the reason for inaccuracies at this level. The conclusion is based on the assumption that if the inaccuracy originated from too low concentrations of probes and/or target to ensure 1:2 target:probe hybrid formation, the inflection point would be shifted towards higher total probe concentrations; however, the situation is completely opposite: the inflection point is shifted towards lower concentration. Therefore, at this moment, our ability to evaluate small target concentrations (< 20 nM) are limited by FRET-based signal transduction platform. The actual limitations of the approach with respect to equilibrium parameters are currently under investigation. Overall, it is worth mentioning that while both methods for establishing inflection point provide similar results, both of them have limitations: method 1 requires collecting data to cover the whole sigmoid curve and performing sigmoid fit (to determine the value of initial plateau and its standard error via a fitting procedure) that may deem impractical in certain situations while method 2 involves subjective choice of plateau vs. slope points.

Further, we established that the sensing system is capable of quantitating the target in the presence of random DNA. Particularly, we evaluated response profiles of the target in the presence of different amounts of a random oligonucleotide (one that does not bind probes). The results indicate that the approach yields accurate target quantification in the presence of exceeding amounts of a random background (Figure 4).

Figure 4. The target can be reliably quantitated in presence of excessive amounts of random DNA. Red points/curves: target only, no random DNA; Black points/curves: target + equimolar amounts of random DNA; blue points/curves: target + 2 × amounts of random DNA, green points/curve: target + 3.8 × amounts of random DNA. Solid symbols/solid lines: target at 20 nM; empty symbols/dashed lines: target at 40 nM. * “np” means “not performed”. Method 2 (Figure 3C) was used to determine target concentration. Colors in table are synchronized with colors of data points/curves. In conclusion, we report a new approach for quantitative analysis of oligonucleotide targets based on a position of response profile. The capability is enabled via deliberate incorporation of negative cooperativity interference into a target-probe binding model. We demonstrate that the approach allows direct separation-less assessment of oligonucleotide target amounts over a

3 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

concentration range of 10-8 – 10-6 M. It is important that the methodology does not require a target standard and/or a calibration. Importantly, the approach has a broader utility for quantitative assessment of other biomolecules as the major design principles are applicable to any target that can bind probes capable to form “negative cooperativity” pair. The sensitivity of the approach is limited by target/probe binding affinities and by availability of signal transduction platforms capable to discern 1:2 target:probe hybrids. Besides, we expect the described platform to be capable to accurately quantify specific nucleic acids on background of native DNA’s since, in general, binary probes yield better selectivity.16 Our current research is focused on evaluating the capabilities of the platform to quantitate SNP-containing targets on the background of wild type species as well as comparison of the quantitation accuracy to traditional methodologies. Overall, considering the simplicity along with availability of various signal transduction mechanisms for nucleic acids,17 we expect the approach to integrate into platforms developed for the point-of-care and limited resource settings

ASSOCIATED CONTENT Supporting Information Experimental protocols, supplementary Figures S1-S5 as referenced throughout the text, Tables S1 (oligonucleotide sequences) and S2 (representative data). The Supporting Information is available free of charge on the ACS Publications website.

AUTHOR INFORMATION Corresponding Author [email protected].

Notes The authors declare no competing financial interests.

ACKNOWLEDGMENT The funding for this project is provided by NIU’s College of Liberal Arts and Sciences. We express gratitude to Professor Jim Horn from NIU’s Department of Chemistry and Biochemistry for providing the opportunity to use a spectrofluorimeter in his lab.

Page 4 of 5

REFERENCES 1 Bellassai, N.; Spoto, G. Anal. Bioanal. Chem. 2016, 408, 7255-7264. 2 Gubala, V.; Harris, L. F.; Ricco, A. J.; Tan, M. X.; Williams, D. E. Anal. Chem. 2012, 84, 487-515. 3 Kalman, T. Z.; Khalandovsky, R.; Tenenbaum Gonikman, E.; Bercovici, M. Angew. Chem., Int. Ed. 2018, 57, 3343-3348. 4 Khodakov, D.; Wang, C.; Zhang, D. Y. Adv. Drug Delivery Rev. 2016, 105, 3-19. 5 Schwarzenbach, H. Breast Cancer Res. 2013, 15, 211. 6 Wang, L. X.; Ji, C. J. Bioanalysis 2016, 8, 143-155. 7 Kessler, H. H.; Preininger, S.; Stelzl, E.; Daghofer, E.; Santner, B. I.; Marth, E.; Lackner, H.; Stauber, R. E. Clin. Diagn. Lab. Immunol. 2000, 7, 298-300. 8 Yang, S.; Rothman, R. E. Lancet Infect. Dis. 2004, 4, 337-348. 9 Kawaguchi, T.; Komatsu, S.; Ichikawa, D.; Morimura, R.; Tsujiura, M.; Konishi, H.; Takeshita, H.; Nagata, H.; Arita, T.; Hirajima, S.; Shiozaki, A.; Ikoma, H.; Okamoto, K.; Ochiai, T.; Taniguchi, H.; Otsuji, E. Br. J. Cancer 2013, 108, 361-369. 10 Ishii, S.; Segawa, T.; Okabe, S. Appl. Environ. Microbiol. 2013, 79, 2891-2898. 11 Donhauser, S. C.; Niessner, R.; Seidel, M. Anal. Chem. 2011, 83, 3153-3160. 12 Ramírez-Castillo, F. Y.; Loera-Muro, A.; Jacques, M.; Garneau, P.; Avelar-González, F. J.; Harel, J.; GuerreroBarrera, A. L. Pathogens 2015, 4, 307-334. 13 Cankar, K.; Štebih, D.; Dreo, T.; Žel, J.; Gruden, K. BMC Biotechnol. 2006, 6, 37. 14 Wang, Z.; Luo, Y.; Xie, X.; Hu, X.; Song, H.; Zhao, Y.; Shi, J.; Wang, L.; Glinsky, G.; Chen, N.; Lal, R.; Fan, C. Angew. Chem., Int. Ed. 2018, 57, 972-976. 15 Ha, S. H.; Ferrell, J. E. Science 2016, 352, 990-993. 16 Kolpashchikov, D. M. Chem. Rev. 2010, 110, 47094723. 17 Li, D.; Song, S. P.; Fan, C. H. Acc. Chem. Res. 2010, 43, 631-641.

4 ACS Paragon Plus Environment

Page 5 of 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

TOC graphics

ACS Paragon Plus Environment

5