Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Aug 1.
Published in final edited form as: J Phys Org Chem. 2018 Mar 23;31(8):e3827. doi: 10.1002/poc.3827

Self-Propagating Amplification Reactions for Molecular Detection and Signal Amplification: Advantages, Pitfalls, and Challenges

Xiaolong Sun a, Doron Shabat b,✉, Scott T Phillips c,✉, Eric V Anslyn a,✉
PMCID: PMC6205521  NIHMSID: NIHMS963992  PMID: 30386006

Abstract

Self-propagating cascade reactions are a recent development for chemo-sensing protocols. These cascade reactions, in principle, offer low limits of detection by virtue of exponential signal amplification, and are initiated by a specific, pre-planned molecular detection event. This combination of selectivity for a detection event followed by in situ signal amplification is achieved by exploitation of mechanistic organic chemistry, and thus has resulted in various chemo-sensing protocols that employ one or more reagents to achieve the desired selectivity and sensitivity for an assay. Species such as hydrogen peroxide, thiols, and fluoride, have been used as active reagents to initiate the first examples of self-propagating signal amplification reactions, although many other active reagents should be compatible with the approaches. A common feature of the reagents that support the self-propagating signal amplification reactions is the involvement of quinonemethide intermediates resulting from elimination of optical reporters and/or active reagents, where the latter propagates the signal amplification reaction. The early examples of these amplification sequences, however, are slow to reach full signal, thus leaving time for background reactions to generate non-specific signals. This issue of background has limited practical applications of these self-propagating signal amplification reactions, as has challenging synthetic routes to the reagents, as well as the potential for other chemical species to interfere with the detection and signal amplification processes. Thus, the goal of this review is to summarize the progress of self-propagating signal amplification technology, identify the pitfalls of current designs, and by doing so, to stimulate future studies in this growing and promising research area.

Keywords: auto-inductive cascades, molecular self-propagation, signal amplification, self-immolative, dendritic chain reaction

INTRODUCTION

Traditional chemo-sensing protocols provide a 1-to-1 relationship between the signal output and the detection event.[1] This 1-to-1 relationship defines the sensitivity of the chemo-sensing assay, thus limiting traditional chemo-sensors to applications in which the target analyte is abundant. Employing chemo-sensors in applications for which the target analyte is scarce requires amplification, either target or signal amplification. As a consequence, many amplification strategies have been developed to improve the sensitivity of chemo-sensing detection protocols,[2] including those that employ catalytic and self-propagating cascade reactions.[3] Other approaches for enhancing the sensitivity of chemo-sensing protocols include the use of chemosensory conjugated polymers,[2b, 4] analyte-induced activation of an organometallic catalyst,[2c, 5] as well as leveraging the polymerase chain reaction (PCR) to amplify signal for specific detection events.[6]

Each of these amplification approaches offer advantages and disadvantages, with some of the latter including (i) the requirement for user input to achieve amplification, or (ii) use of reagents (e.g., enzymes) that must be stored and handled carefully. Another feature to consider when choosing an amplification protocol is the kinetics of the amplification process. Depending on the chosen amplification procedure, the signal will increase linearly with time, exponentially, or will reach a pre-programmed level and stop increasing. Exponential signal amplification, in particular, offers the potential to realize the most sensitive assays by lowering limits of detection (LOD) for a target analyte.[7] Exponential signal amplification is inherent in most self-propagating signal amplification reactions, thus this review focuses on the strengths and weaknesses of this emerging and promising type of amplification reaction.

GENERAL DESCRIPTION OF A SELF-PROPAGATING SIGNAL AMPLIFICATION REACTION

Self-propagating signal amplification reactions require one or more reagents to simultaneously generate signal and an active reagent, or multiple copies of an active reagent. In the illustration in Scheme 1, the single amplification reagent contains four key parts. The first piece is functionality that responds to a specific target signal, which may be the analyte or an active reagent that propagates the signal amplification reaction. This functionality is termed “trigger” in Scheme 1, but it also has been referred to as a detection unit, and even an activity-based detection unit. The second key part of the reagent is the core functionality that enables release of pendant molecules once the trigger (Scheme 1) is activated by the target signal. This core functionality is called a “dendritic adaptor” in Scheme 1. Attached to the core functionality are the other two key pieces of the reagent, which include reagents (Scheme 1) and a reporter (or reporters). Both the reagents and the reporter are inactive when attached to the amplification reagent, but become active once released from the core functionality. In the example in Scheme 1, the released reagents become “active reagents” that then react with triggers on other copies of the signal amplification reagent. In this general depiction, the analyte and the active reagents are the same entities.

Scheme 1.

Scheme 1

Open in a new tab

Schematic illustration of one type of self-propagating signal amplification reaction. Reproduced with permission from ref 2a. Copyright 2016 American Chemical Society.

The key to the design in Scheme 1 is that for every one analyte, two active reagents are released along with one reporter, which usually are colorimetric and fluorometric probes. The two active reagents react with two additional copies of the signal amplification reagent to generate two more reporters and four active reagents. This cycle of reactions continues until all of the signal amplification reagent is consumed, or until an assay is stopped. Because the concentration of the active reagents increases exponentially over time, the rate of the reaction increases exponentially, as does the quantity of the reporter.

The first example of this type of self-propagating signal amplification reaction was reported in 2009 by Doron Shabat for the detection of hydrogen peroxide.[8] Shortly thereafter, a series of related self-propagating reactions were described by Scott Phillips, including reagents that respond to H2O2, fluoride and thiol.[9]

DEFINITIONS: AUTOINDUCTIVE, DENDRITIC CHAIN REACTIONS, SELF-IMMOLATIVE

As this field emerged, different terms were used to describe various examples of self-propagating amplification reactions. To improve clarity moving forward, we wish to define and coalesce terminology here. We also propose the decision tree in Scheme 2 for use when selecting nomenclature for new self-propagating amplification reactions.

Scheme 2.

Scheme 2

Open in a new tab

Decision tree for use when describing self-propagating amplification reactions.

We believe that the descriptor with the broadest scope is self-propagating amplification reaction. This phrase encompasses a wide variety of reaction mechanisms and reagent architectures, yet it signifies two important attributes: i.e., (i) self-propagating reactions, and (ii) amplification. In this review, we discuss self-propagating signal amplification reactions, but self-propagating target amplification may be of interest in future systems.

Secondary descriptors also are necessary for clearly articulating the type of self-propagating amplification reaction. These secondary descriptors focus on whether an active reagent is consumed when it reacts with the amplification reagent, as well as on the architecture of the amplification reagent. In the former case, if the active reagent is consumed, then the self-propagating reaction is autoinductive, because the active reagent induces the propagation reaction. If the active reagent is not consumed, then it operates as a catalyst, and thus the self-propagating reaction is autocatalytic.

Further, we believe that defining the architecture of the amplification reagent helps create an accurate mental image of the amplification reagent as well as the self-propagating amplification reaction. Hence, if the active reagent(s) and/or reporter(s) are attached to the amplification reagent in a dendritic-type architecture, then we further denote the self-propagating amplification reaction as dendritic chain reaction. [2a] If a dendritic architecture is not employed, then no additional descriptor is needed.

Lastly, the term self-immolative has been used to describe the dendritic chain reactions, auto-inductive cascades, as well as spontaneous depolymerizations.[10] Self-immolation strictly means “to kill oneself as a sacrifice”, often by setting oneself on fire. Albeit the religious connotation of a sacrifice has been removed in the chemical context, the concept is similar. The dendritic structure, or linear polymer, is decomposed (i.e. killed) by a trigger that is created by other molecules of the very same dendrimer or polymer.

MEASURING THE DEGREE OF AMPLIFICATION

To quantitate the molecular reagents generated from the auto-inductive cascades, mathematical calculations based on the types of cascades facilitates an understanding of the signal amplification. For an AB-type dendrimer where A is the self-immolative moieties and B is the molecular reagent released, only one active reagent will be released and will be consumed by another AB molecule in the cycling no matter how many cycles in the process. Thus, no more than one reagent B can exist in the system after all the AB molecules reacted. While for AB2 dendrimer, number of 2m reagents B (where m is the number of cycles) can be released during the self-propagating process and (N+1) number of reagent molecules, where N is number of AB2 molecule (N = nNa, where N = number of molecules, n = number of moles in substance, and Na is Alvogadros constant) can be generated after complete disassembly. Further for AB3 dendrimer, 3m reagents B accumulated in the auto-induction, and eventually (2N+1) number of molecules can be produced after full induction.

To measure the amplification from the self-propagating cascades over conventional probes, an amplification factor (α) can be employed, which is defined as (Iamplification - Ibackground)/Iinitial, where Iamplification is the intensity of optical signal produced after amplification, Iinitial is the intensity without amplification, and Ibackground is the signal arising from spontaneous self-propagation without added analytes.[9a] Thus, to evaluate any auto-amplification diagnostic assay, these amplification factors are a critical index for researchers in this area to determine.

QUANTITATION OF THE ANALYTE

In terms of analyte quantitation, because each self-propagating cascade will disassemble to release all reporters over an extended period irrespective of the amount of trigger, the optical response must be measured at a set time to distinguish differing levels of trigger. Each cascade generates different signals with time, and thus a calibration curve using known analytes and concentrations must first be generated for this set time point, and subsequently the signal from unknown amounts of the analyte at the same time-point results in a quantitative assay. The optical response could be UV-Vis,[11] fluorescence[12] or photography.[13] Also, we are currently exploring a novel strategy by employing photography coupled with self-propagating cascades to quantitate and differentiate G and V nerve agents surrogates based on fluorescence chromaticity. We intend to design a portable device for recognition and quantitation of unknown chemical warfare in field.

MECHANISTIC FEATURES

A unifying feature of many previous self-propagating cascades is the involvement of quinomethide intermediates resulting from the elimination of optical reporters and/or active reagents (Eq. 1).[14] The key step is generation of an electron rich aromatic species (phenol or aniline) with an ortho- or para-benzyl-leaving group (nucleofuge). As shown, arrow pushing reveals a pathway for first-order elimination (an E1CB reaction) that generates a reporter or chain-propagating species (the leaving group, either directly or via a decarboxylation). The rate of signal accumulation resulting from such cascades depends on this release of the leaving groups concomitant with generation of the quinomethide-like intermediates. Such intermediates are high in energy, in that the aromaticity of the phenyl rings has been lost, and commonly electronegative groups are attached to the exo-cyclic alkenes, which is further destabilizing.

graphic file with name nihms963992e1.jpg Eq. 1

Thus, while signal amplification is indeed achieved with these systems, the total signal accumulation from complete disassembly requires hours to a couple of days.[9a, 9f, 15] In addition, due to the slowness of the reactions, the possibility of extremely low but naturally abundant levels of the analytes, or the presence of analyte analogs, can lead to a background due to adventitious initiation of the cascades (see below). In other words, given enough time, every auto-inductive cascade yet reported releases all reagents and reporters over an extended time period.

HYDROGEN PEROXIDE SELF-PROPAGATING CASCADES

Hydrogen peroxide (H2O2) is the simplest peroxide, and it is a strong oxidizing agent, which has been widely employed as a bleach and as a cleaning reagent to reduce BOD[16] and COD[17] from industrial wastewater. In addition, as one of reactive oxygen species (ROS), H2O2 plays an important role as a signalling molecule in the regulation of a variety of biological processes, such as immune response, cell signalling, Alzheimer’s diseases and cancer.[18] Thus, a series of boronic acid-based self-propagating cascades (1–4, Figure 1) were developed by Shabat which integrated phenylboronic acid triggers, two reagents, colorimetric reporters, together with a 2,4,6-/2,6-hydroxymethyl phenolate. Oxidation reaction between boron and H2O2 induced the self-immolation through subsequent 1,4-/1,6-quinone methide elimination and spontaneous decarboxylation to release reporters and active reagents, which were then reacted with choline oxidase (COX), alcohol oxidase (AOX), glucose oxidase (GOX) and sarcosine oxidase (SOx), respectively.[8, 9e, 19] Exponential growth took place that progresses through a growing number of diagnostic cycles. Shabat also reported structure 5 (Figure 1) which integrated a boronic ester with hydroquinone for hydrogen peroxide self-propagation.[20] The mechanism also initiates by oxidation of boron by peroxide and subsequent oxidation to hydroquinone appended with electron-donating group, which then is involved in another cycle to decompose 5, each time creating a planar push-pull fluorophore. Compound 6 (Figure 1), developed by Phillips, for glycosylase enzyme detection also employed a selective oxidative cleavage reaction between peroxide and an aryl boronic acid, as well as transformation of 1,2,4-trihydroxybenzene through auto-oxidation to generate hydrogen peroxide to feed back into the cycle and to generate a colorimetric signal.[13] To summarize these examples, the unifying features are the involvement of boronic acid/ester oxidation and hydrolysis of quinomethide intermediates resulting from eliminations, to offer more reagents for auto-inductive cycling.

Figure 1.

Figure 1

Open in a new tab

Structures of compounds for hydrogen peroxide self-propagation.

FLUORIDE SELF-PROPAGATING CASCADES

Fluoride anion plays a role in dental health[21] and has potential use for the treatment of osteoporosis[18b] Fluoride is also a product of the hydrolysis of sarin and soman (G-type nerve agents), and consequently, the ability to monitor fluoride in victims and surrounding environments after a terrorist incident using this nerve gas is of great value.[22] Thus, fluoride chemosensors are in high demand due to their importance in a variety of healthcare and environmental contexts. In the development of fluoride self-propagating cascades, Shabat and Phillips created designs that incorporate fluoride-initiated TBS/TBDPS deprotections to generate phenol derivatives that lead through eliminations to quinone methide intermediates, and subsequent colorimetric signal amplifications. In particular, these researchers and others have reported three fluoride self-amplification systems, which release two fluorides from each fluoride trigger (compound 7, 8 and 9, Figure 2).[9a, 9b, 9f, 15] The Anslyn group recently reported a six-fold release of fluoride for each fluoride trigger, amplifying both a colorimetric and fluorescence response (compound 10).[23] Phillips also designed self-powered polymeric materials (compounds 12 and 13, Figure 2) by incorporating a fluoride self-propagating auto-inductive reaction directly onto polymers.[9c, 9d] TentaGel microsphere 12 converted UV light signals into initiating the pumping of the surrounding fluid, and responded autonomously and continuously after removing the fleeting stimuli. Noticeably, in example 12, they introduced a 2-methoxy group into the 4-aminobenzaldehyde to speed up the formation of quinomethides, as indicated above has been a detriment to these kinds of systems. Similarly, hydrophobic polymer 13 under a 300-nm light trigger, converts to hydrophilic (blue) via self-propagating reactions that continue even in the absence of the signal.

Figure 2.

Figure 2

Open in a new tab

Structures of compounds for fluoride self-propagation.

Differing from the hydrogen peroxide self-propagating cascades, fluoride cascades release reagent units with the same chemical without additional stimulus, thereby initiating amplification cycles, and the probe component generates a chromogenic signal. However, the rate of signal accumulation resulting from these cascades depends on the hydrolysis of aromatic mono- or di-fluoromethyl groups, that slowly release fluoride through quinomethide-like intermediates. Hence, while signal amplification is achieved with these systems, the total signal accumulation from complete disassembly requires hours to a couple of days. High background signals without any input of analytes are the consistent weaknesses both for H2O2 and fluoride auto-inductive cascades.

In an effort to overcome the problems mentioned above, a new auto-inductive cascade using 11 employing benzoyl fluoride as a latent source of fluoride was reported by the Anslyn group.[12] The auto-induction leads to a maximum four-fold signal enhancement for each fluoride generated, as well as a self-propagating cycle that generates three fluorophores for each single fluoride released. A two-step integrated protocol creates a more rapid auto-inductive cascade, as well as a highly sensitive diagnostic assay for the ultratrace quantitation of a phosphoryl fluoride nerve agent surrogate. Even though the process is sped up, the system still suffers from a background initiation with time.

THIOL SELF-PROPAGATING CASCADES

Thiols are important biomarkers that participate in several physiological functions and their plasma level can indicate the diagnosis of disease states.[24] Thiol containing bio-species, such as cysteine, homocysteine, glutathione, etc., play roles in the regulation of physiological processes, In addition, thiol containing V-series nerve agents are threats to modern society as terrorist tools, and thus are important targets for monitoring by self-propagating systems.[25] Shabat reported a self-immolative compound 14 for thiol self-propagation through a dendritic chain reaction (Figure 3). His probe used a sulfhydryl-triggered benzoquinone reduction and the analyte is penicillin-G-amidase (PGA). PGA is an enzyme that catalyzes a chemical reaction to release free mercaptoacetic acid as the target, thus initiating another round of the cycle. [26] Phillips described a chemically responsive polymer film 15 that is capable of detecting low levels of thiols and subsequently initiating a self-propagating reaction wherein the material converts from a nonfluorescent film into a globally fluorescent material.[27] The thiol initiates a deprotection of the dinitrophenylsulfonyl group to generate a phenol, that then cyclizes on the thioester to release another equivalent of thiol. Recently, Anslyn developed an auto-inductive protocol which employs a Meldrum’s-acid based conjugate acceptor 16 as a latent source of thiol for signal amplification, as well as colorimetric detection of thiols. Using 16 in a two-step integrated protocol yields a rapid, sensitive, and precise diagnostic assay for the ultratrace quantitation of a thiophosphate nerve agent surrogate. To date, only three thiol auto-inductive cascades have been reported. Thus, new designs and improvements would be welcome compliments to these systems.

Figure 3.

Figure 3

Open in a new tab

Structures of compounds for sulfuryl self-propagation.

ADVANTAGES AND PITFALLS

The advantage of using self-propagating cascades have been accentuated above, and can be embodied in the ability to achieve exponential signal amplification. While not necessarily an advantage, an attractive feature of these systems is the challenge they inspire for mechanistic organic chemists. Generating these systems takes a knowledge of organic reaction mechanisms, clever insights into triggering routines, and the ability to recognize approaches for generation of propagating species that repeatedly create further copies of themselves, to generate a cascade that cycles and exponentially generates signals. Thus, the area naturally fits within the realm of physical organic chemistry.

Some of the pitfalls of the methods have also been introduced above. One is that a large fraction of the reaction cascades proceed via quinomethide intermediates, which are naturally of high energy because of the loss of aromaticity, and routine placement of electronegative elements on olefinic carbons. A challenge for those individuals entering this field is to move away from, or significantly stabilize, these intermediates as a means of speeding up the kinetics of the reactions.

One aspect of these reactions that has not been touched upon is that each generates the propagation species via elimination of leaving groups, i.e. nucleofuges. Inherently, all nucleofuges are also nucleophiles. Similarly, the triggers for the cascades are all nucleophiles (with the exception of H2O2, which can be nucleophilic and electrophilic): i.e. fluoride and thiols. Imaginative mechanistic insights are required to generate systems that propagate via electrophiles, or are generally initiated by electrophilic analytes. Many important analytes are highly electrophilic, such as nerve agents, but triggering mechanisms for electrophilic agents are generally lacking.

Yet, by far the limiting aspects of self-propagating cascades are the background reactions. In theory, the autoinductive amplification approach could lead to analyte detection at the level of a single molecule. Practically, this is not possible because any non-specific decomposition of the probe (background reactions, e.g. hydrolysis etc.) activates the amplification cycle of the system and results in a background signal. All cascades that have yet been reported ultimately self-initiate. Adventitious hydrolysis of silyl-protecting groups, slow but spontaneous release of leaving groups, or disulfide bond-exchange, can lead to unwanted initiation of the cascades. It is only a matter of time that they all eventually start the cycling, and once started the cycles cannot be broken or reset. To us, this is the biggest challenge facing the field, and we call to the sensing community to take up this challenge to solve. Thus, to further improve the signal-to-noise obtained so for (and thus the detection sensitivity), there is a need to develop molecular self-propagating systems with better stability and faster disassembly cycles.

SUMMARY AND OUTLOOK

Herein, we summarized the new technology of self-propagating amplification reactions, including auto-inductive cascades, self-immolative systems, and the dendritic chain reaction. These methods offer exponential signal amplification and molecule self-replication to achieve a lower limit of detection and high sensitivity. To improve and expand the scope of self-propagating cascades, various targets beyond H2O2, fluoride and thiols are needed both in bio- and chemo-sensing. For example, in addition to nucleophile substances, electrophiles need to be employed as the targets. In addition, to accelerate the self-propagating speed, adapting different mechanism for the system design beyond quinonemethide formation should be approached. Further, background signals due to trivial amount of analytes or interferences should be reduced and eliminated to improve the amplification factors and limit of detection. Ease of synthesis should be considered of high value for mass production and further practical applications. In bio-application of cell imaging, cascades with fluorescence embodying ratiometric signal changes, along with fluorophores offering long wavelength emission, low background, water-solubility, also non-interferences from other species, are clearly a frontier to be explored. In environmental analysis, simple and efficient portable devices, cooperating with self-propagating cascades for signal amplification, potentially for in-field detection applications will be possible once the pitfalls have been addressed. Further, as early results indicate, there is high potential for incorporating autonomous inductive cascades with stimuli-responsive polymers, which translates molecule detection events into macroscopic responses. While many challenges are needed to be solved to overcome the current drawbacks, and potential for self-propagating cascades is enormous. It is a young but pioneering research field, offering large signal amplifications and massive molecule accumulation with low levels of stimulus. These advantages will continuously inspire more researchers to explore this expanding area of research.

Acknowledgments

X.L.S. would like to thank University of Texas at Austin for the postdoc opportunity. D.S. thanks the Israel Science Foundation (ISF), the Binational Science Foundation (BSF) and the German Israeli Foundation (GIF) for financial support. S.T.P. appreciates the support of the National Institutes of Health: R01GM105686. EVA thanks the Welch Regents Chair (F-0045), and the Defense Threat Reduction Agency-Joint Science and Technology Office for Chemical and Biological Defense (Grant no: HDTRA1-16-1-0001).

Footnotes

DEDICATION

This paper was submitted to be part of a special issue honouring the life and scientific accomplishments of Dr. John (Jack) Roberts of Caltech. One of the authors, EVA, has extremely fond memories as a graduate student at Caltech of Jack in seminars asking probing questions, showing a depth and breadth of knowledge in many fields that was unprecedented. Then, during visits back to Caltech as an invited seminar speaker, EVA recalls meetings in Jack’s office discussing science, that revealed his passion and dedication to undergraduate education, the results of which were the primary points of his discussions. Finally, EVA was honoured to be the Saul Winstein Lecturer at UCLA in 2014, and at the speaker’s banquet Jack was also honoured by a celebration of his 90th birthday. This was a very special occasion, that EVA will think of as a highlight of his career long into the future.

References

  • 1.a Chan J, Dodani SC, Chang CJ. Nat. Chem. 2012;4:973–984. doi: 10.1038/nchem.1500. [DOI] [PMC free article] [PubMed] [Google Scholar]; b De Silva AP, Gunaratne HN, Gunnlaugsson T, Huxley AJ, McCoy CP, Rademacher JT, Rice TE. Chem. Rev. 1997;97:1515–1566. doi: 10.1021/cr960386p. [DOI] [PubMed] [Google Scholar]; c Sun X, James TD. Chem. Rev. 2015;115:8001–8037. doi: 10.1021/cr500562m. [DOI] [PubMed] [Google Scholar]
  • 2.a Roth ME, Green O, Gnaim S, Shabat D. Chem. Rev. 2016;116:1309–1352. doi: 10.1021/acs.chemrev.5b00372. [DOI] [PubMed] [Google Scholar]; b Swager TM. Acc. Chem. Res. 1998;31:201–207. [Google Scholar]; c Zhu L, Anslyn EV. Angew. Chem. Int. Ed. 2006;45:1190–1196. doi: 10.1002/anie.200501476. [DOI] [PubMed] [Google Scholar]
  • 3.Goggins S, Frost CG. Analyst. 2016;141:3157–3218. doi: 10.1039/c6an00348f. [DOI] [PubMed] [Google Scholar]
  • 4.a Marsella MJ, Swager TM. J. Am. Chem. Soc. 1993;115:12214–12215. [Google Scholar]; b Zhou Q, Swager TM. J. Am. Chem. Soc. 1995;117:7017–7018. [Google Scholar]
  • 5.a Wu Q, Anslyn EV. J. Am. Chem. Soc. 2004;126:14682–14683. doi: 10.1021/ja0401038. [DOI] [PubMed] [Google Scholar]; b Zhu L, Lynch VM, Anslyn EV. Tetrahedron. 2004;60:7267–7275. [Google Scholar]
  • 6.a Saghatelian A, Guckian KM, Thayer DA, Ghadiri MR. J. Am. Chem. Soc. 2003;125:344–345. doi: 10.1021/ja027885u. [DOI] [PMC free article] [PubMed] [Google Scholar]; b Gianneschi NC, Ghadiri MR. Angew. Chem. Int. Ed. 2007;119:4029–4032. [Google Scholar]
  • 7.a Swiderska MA, Reymond J-L. Nat. Chem. 2009;1:527–528. doi: 10.1038/nchem.385. [DOI] [PubMed] [Google Scholar]; b Mohapatra H, Schmid KM, Phillips ST. Chem. Commun. 2012;48:3018–3020. doi: 10.1039/c2cc17566e. [DOI] [PubMed] [Google Scholar]
  • 8.Sella E, Shabat D. J. Am. Chem. Soc. 2009;131:9934–9936. doi: 10.1021/ja903032t. [DOI] [PubMed] [Google Scholar]
  • 9.a Baker MS, Phillips ST. J. Am. Chem. Soc. 2011;133:5170–5173. doi: 10.1021/ja108347d. [DOI] [PubMed] [Google Scholar]; b Baker MS, Phillips ST. Org. Bio. Chem. 2012;10:3595–3599. doi: 10.1039/c2ob25363a. [DOI] [PubMed] [Google Scholar]; c Baker MS, Yadav V, Sen A, Phillips ST. Angew. Chem. Int. Ed. 2013;52:10295–10299. doi: 10.1002/anie.201304333. [DOI] [PubMed] [Google Scholar]; d Kim H, Baker MS, Phillips ST. Chem. Sci. 2015;6:3388–3392. doi: 10.1039/c5sc00701a. [DOI] [PMC free article] [PubMed] [Google Scholar]; e Yoshii T, Onogi S, Shigemitsu H, Hamachi I. J. Am. Chem. Soc. 2015;137:3360–3365. doi: 10.1021/ja5131534. [DOI] [PubMed] [Google Scholar]; f Gu J-A, Mani V, Huang S-T. Analyst. 2015;140:346–352. doi: 10.1039/c4an01723d. [DOI] [PubMed] [Google Scholar]
  • 10.a Amir RJ, Pessah N, Shamis M, Shabat D. Angew Chem Int Edit. 2003;42:4494–4499. doi: 10.1002/anie.200351962. [DOI] [PubMed] [Google Scholar]; b Sagi A, Weinstain R, Karton N, Shabat D. J. Am. Chem. Soc. 2008;130:5434–5435. doi: 10.1021/ja801065d. [DOI] [PubMed] [Google Scholar]; c Yeung K, Kim H, Mohapatra H, Phillips ST. J. Am. Chem. Soc. 2015;137:5324–5327. doi: 10.1021/jacs.5b02799. [DOI] [PubMed] [Google Scholar]; d Kim H, Mohapatra H, Phillips ST. Angewandte Chemie. 2015;127:13255–13259. doi: 10.1002/anie.201506511. [DOI] [PubMed] [Google Scholar]
  • 11.Sun X, Anslyn EV. Angew. Chem. Int. Ed. 2017;129:9650–9654. [Google Scholar]
  • 12.Sun X, Dahlhauser SD, Anslyn EV. J. Am. Chem. Soc. 2017;139:4635–4638. doi: 10.1021/jacs.7b01008. [DOI] [PubMed] [Google Scholar]
  • 13.Yeung K, Schmid KM, Phillips ST. Chem. Commun. (Camb.) 2013;49:394–396. doi: 10.1039/c2cc36861g. [DOI] [PubMed] [Google Scholar]
  • 14.Gnaim S, Shabat D. Acc. Chem. Res. 2014;47:2970–2984. doi: 10.1021/ar500179y. [DOI] [PubMed] [Google Scholar]
  • 15.Perry-Feigenbaum R, Sella E, Shabat D. Chem. Eur. J. 2011;17:12123–12128. doi: 10.1002/chem.201101796. [DOI] [PubMed] [Google Scholar]
  • 16.Biochemical oxygen demand (BOD) is the amount of dissolved oxygen needed by aerobic biological organisms in a body of water to break down organic material present in a given water sample at certain temperature over a specific time period.
  • 17.In environmental chemistry, the chemical oxygen demand (COD) test is commonly used to indirectly measure the amount of organic compounds in water
  • 18.a Sun X, Xu SY, Flower SE, Fossey JS, Qian X, James TD. Chem. Commun. (Camb.) 2013;49:8311–8313. doi: 10.1039/c3cc43265c. [DOI] [PubMed] [Google Scholar]; b Kleerekoper M. Endocrinol. Metab. Clin. 1998;27:441–452. doi: 10.1016/s0889-8529(05)70015-3. [DOI] [PubMed] [Google Scholar]
  • 19.a Avital-Shmilovici M, Shabat D. Biorg. Med. Chem. 2010;18:3643–3647. doi: 10.1016/j.bmc.2010.02.038. [DOI] [PubMed] [Google Scholar]; b Sella E, Lubelski A, Klafter J, Shabat D. J. Am. Chem. Soc. 2010;132:3945–3952. doi: 10.1021/ja910839n. [DOI] [PubMed] [Google Scholar]; c Karton-Lifshin N, Shabat D. New J. Chem. 2012;36:386–393. [Google Scholar]
  • 20.Sella E, Shabat D. Org. Biomol. Chem. 2013;11:5074–5078. doi: 10.1039/c3ob40962g. [DOI] [PubMed] [Google Scholar]
  • 21.Kirk KL. Biochemistry of the Elemental Halogens and Inorganic Halides. Springer; 1991. pp. 19–68. [Google Scholar]
  • 22.Zhou Y, Zhang JF, Yoon J. Chem. Rev. 2014;114:5511–5571. doi: 10.1021/cr400352m. [DOI] [PubMed] [Google Scholar]
  • 23.Sun X, Reuther JF, Phillips ST, Anslyn EV. Chem. Eur. J. 2017;23:3903–3909. doi: 10.1002/chem.201604474. [DOI] [PubMed] [Google Scholar]
  • 24.a McCully KS. Nat. Med. 1996;2:386–389. doi: 10.1038/nm0496-386. [DOI] [PubMed] [Google Scholar]; b Michelet F, Gueguen R, Leroy P, Wellman M, Nicolas A, Siest G. Clin. Chem. 1995;41:1509–1517. [PubMed] [Google Scholar]
  • 25.Wood ZA, Schröder E, Robin Harris J, Poole LB. Trends Biochem. Sci. 2003;28:32–40. doi: 10.1016/s0968-0004(02)00003-8. [DOI] [PubMed] [Google Scholar]
  • 26.Sella E, Weinstain R, Erez R, Burns NZ, Baran PS, Shabat D. Chem. Commun. (Camb.) 2010;46:6575–6577. doi: 10.1039/c0cc02195d. [DOI] [PubMed] [Google Scholar]
  • 27.Mohapatra H, Kim H, Phillips ST. J. Am. Chem. Soc. 2015;137:12498–12501. doi: 10.1021/jacs.5b08582. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES