Insulin detection using a corona phase molecular recognition site on

Subscriber access provided by READING UNIV

Article

Insulin detection using a corona phase molecular recognition site on single-walled carbon nanotubes Gili Bisker, Naveed Ali Bakh, Michael A. Lee, Jiyoung Ahn, Minkyung Park, Ellen B O'Connell, Nicole Iverson, and Michael S. Strano ACS Sens., Just Accepted Manuscript • DOI: 10.1021/acssensors.7b00788 • Publication Date (Web): 23 Jan 2018 Downloaded from http://pubs.acs.org on January 29, 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 free 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 accessible to all readers and 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.

ACS Sensors 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 31 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

ACS Sensors

Insulin detection using a corona phase molecular recognition site on single-walled carbon nanotubes Gili Bisker,1, ‡ Naveed A. Bakh, 1, ‡ Michael A. Lee,1 Jiyoung Ahn,1 Minkyung Park,1 Ellen B. O'Connell,2 Nicole M. Iverson,3 and Michael S. Strano1* 1

Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge,

Massachusetts 02139, United States 2

Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge,

Massachusetts 02139, United States 3

Department of Biological Systems Engineering, University of Nebraska-Lincoln, 223 L.W.

Chase Hall, Lincoln, Nebraska 68583, USA * Corresponding author: [email protected]

KEYWORDS: Single-walled carbon nanotubes; Fluorescent nanosensors; Molecular recognition; Insulin; High-throughput screening;

ACS Paragon Plus Environment

1

ACS Sensors 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 31

ABSTRACT

Corona Phase Molecular Recognition (CoPhMoRe) is a technique whereby an external, adsorbed phase around a colloidal nanoparticle is selected such that its molecular conformation or interaction recognizes a specific target analyte. In this work, we employ a high-throughput screening of a library of poly(ethylene glycol) (PEG)-conjugated lipids adsorbed onto nearinfrared fluorescent single-walled carbon nanotubes to discover a corona phase selective for insulin.

We find that a C16-PEG(2000kDa)-Ceramide causes a 62% fluorescent intensity

decrease of the (10,2) chirality nanotube in 20 µg/ml insulin concentration. The insulin protein has no prior affinity towards the C16-PEG(2000kDa)-Ceramide molecules in free solution, verified by isothermal titration calorimetry, and the interaction occurs only upon their adsorption onto the single-walled carbon nanotube scaffolds. Testing a panel of proteins originated from human blood as well as short, 7 amino-acid, fragments of the insulin peptide, rules out nonselective recognition mechanisms such as molecular weight, isoelectric point, and hydrophobicity based detection. Interestingly, longer fragments of isolated α- and β-peptide chains of insulin are detected by the construct, albeit with lower affinity compared to the intact insulin protein, suggesting that the construct recognizes insulin in its native form and conformation. Finally, the insulin recognition and the quantification of its solution concentration were demonstrated both in buffer and in blood serum, showing that the CoPhMoRe construct works in this complex environment despite the presence of potential non-specific adsorption. Our results further motivate the search for non-biological synthetic recognition sites and open up a new path for continuous insulin monitoring in vivo with the hope of improving glycemic control in closed-loop artificial pancreas systems.


2

Page 3 of 31 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

ACS Sensors

Insulin is a peptide hormone, synthesized and secreted by the pancreas, responsible for stimulating glucose uptake from the blood and the synthesis of lipids, as well as inhibiting glucose generation, ketone bodies production, and the breakdown of proteins, glycogen, and lipids.1 Insulin deficiency and/or insulin resistance characterize the chronic disease diabetes mellitus,2 from which more than 400 million people suffer world-wide.3 In type 1 diabetes, the body cannot produce insulin due to a complete destruction of the insulin producing cells. Consequently, patients must rely on exogenous insulin to maintain their blood glucose levels.4 One of the more effective ways to maintain glycemic control is through the use of a continuous glucose monitoring device coupled to an insulin infusion pump with the appropriate control algorithms.5 The ultimate goal for this technology is the development of a closed-loop artificial pancreas system.6 One complication with insulin treatment in general is that insulin dosing must necessarily be conservative because its existing concentration in vivo is unknown. Treatment protocols tend to conservatively under-dose insulin due to the potential of induced hypoglycemia, a condition characterized by low blood glucose levels.7 The lack of information regarding the concentration of circulating insulin already present in blood before the administration of exogenous insulin necessitates this conservative dosing.8 Unlike glucose, whose blood levels can be monitored continuously or sampled on demand easily by the patient,9 insulin levels are not yet accessible with a continuous monitor. Current analysis requires a blood test and laboratory equipment for analysis.10 Hence, the need for improved glycemic control gives rise to the necessity of a new technology for measuring insulin concentrations in vivo to improve algorithms for closed-loop systems.8, 11 We address this challenge by developing a novel


3

ACS Sensors 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 4 of 31

synthetic recognition site for insulin on fluorescent nanoparticles. Such a system is a critical advancement towards real-time detection of insulin concentration in vivo owing to the possibility of non-invasive optical readout in the near-infrared tissue transparency window. One of the most prominent bottlenecks in detecting and monitoring bio-analytes in vivo is the molecular recognition of the target analyte.12 In the case of insulin, nature has provided a variety of recognition options such as the insulin receptor, insulin antibody, insulin binding aptamer, and insulin binding peptide. The insulin receptor (IR) is a transmembrane tyrosine kinase receptor which is composed of two α-subunits and two β-subunits.13 The IR is activated upon the binding of insulin to the α-subunit, leading to a cascade of biochemical events that results in multiple effects, including the promotion of glucose influx.14 Insulin antibodies, on the other hand, should not be found in the blood of healthy individuals and their production is often the result of exogenous insulin treatment.15 The presence of such antibodies may cause insulin resistance or hypoglycemia16 and in other cases can indicate an early diabetic stage17 or allergy to insulin.18 The insulin-binding peptide (IBP) [Cys-Val-Glu-Glu-Ala-Ser] was designed by translating the complementary sequence of the gene that codes for the IR insulin binding region. This peptide interacts with the carboxyl terminus of the insulin β-chain.19 Aptamers, which are oligonucleotides with a unique three-dimensional conformation that allows for strong interactions with a target protein, are usually discovered by in vitro combinatorial evolution (Systematic evolution of ligands by exponential enrichment, or SELEX).20 The molecular recognition of aptamers has been utilized in the development of sensors, which use a variety of techniques such as electrochemistry and fluorescence to transduce the binding event to an observable signal.21 A natural aptamer for insulin was found in the insulin gene promoter region that regulates insulin transcription22, suggesting that insulin regulates its own transcription. This


4

Page 5 of 31 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

ACS Sensors

region, referred to as the insulin-linked polymorphic region (ILPR) has repeats of a G-rich sequence that were shown to have strong affinity towards the insulin protein. Based on this finding, further in vitro selection processes have led to the discovery of additional insulin aptamers with even stronger affinity.23 The main shortcomings of the natural molecular recognition systems described here include their high cost, lack of binding tunability, and limited lifetime.24 Synthetic systems,25 on the other hand, can benefit from both chemical and thermal stability, which enable specific detection of analytes in harsh environments, such as in vivo.26 Recently, our group has introduced the concept of corona phase molecular recognition, or CoPhMoRe.27 In this approach, a heteropolymer is selected from a library and used to suspend fluorescent nanoparticles, such that its pinned configuration around the nanoparticles, referred to as the corona phase, enables the selective recognition of a target analyte. We have used single-walled carbon nanotubes (SWNTs) as the transducing fluorescent nanoparticles owing to their high photo-stability, lack of photobleaching, and bright fluorescent emission in the near-infrared (nIR) part of the spectrum, which overlaps with the tissue transparency window.28 Further, SWNT can be rendered bio-compatible and thus can be used for in vivo applications.8, 26, 29 Historically, we first demonstrated the concept of CoPhMoRe for small molecules detection such as, riboflavin, L-thyroxine, estradiol, and dopamine.27a, 30 Recently, we have extended the concept to larger biological macromolecules, with the recognition and detection of the protein fibrinogen being successfully demonstrated.31 Selective CoPhMoRe phases have been discovered using a high-throughput screening process where libraries of polymers were wrapped around nanotubes, thus creating a variety of corona phases. These corona phases were then tested against a library of relevant analytes. The nIR fluorescence of the SWNT enables immediate feedback


5

ACS Sensors 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 6 of 31

by monitoring emission in response to molecular docking or binding to the corona phase. This molecular interaction has generally been observed as fluorescence intensity changes or shifts in the peak emission wavelengths.27b The interaction between a target molecule and the SWNT at the surface of the corona phase can be complex and is affected by multiple factors such as the wrapping polymer composition,32 the nanotube chirality,33 the valency of the corona phase,34 the lipophilicity of the target,35 and the redox-potential of the target.36 The inverse problem of CoPhMoRe design, which is defined as choosing the corona phase a priori for a given target molecule has been addressed theoretically for helically wrapped SWNT,37 whereas experimental demonstration of such predictions remains the subject of future research. In this work we have implemented an extended protein CoPhMoRe screen with a library of PEGylated lipid wrappings (listed in Chart 1) targeting insulin recognition. We have screened this library of nanotube suspensions against a protein panel constructed previously from abundant and clinically significant proteins,31 and found a specific C16-PEG2000-Ceramide – SWNT complex that detects insulin with high specificity. With this wrapping (N-palmitoylsphingosine-1-succinyl[methoxyPEG2000]), the (10,2) chirality shows over a 60% decrease in the fluorescence emission intensity upon the interaction with 20 µg/ml insulin. According to isothermal titration calorimetry (ITC) measurements, the insulin has no affinity to the free lipid off of the nanotube, but rather to its specific configuration when wrapped around the nanotube sensor. The fluorescent response of the sensor is not correlated to the protein molecular weight, hydrophobicity, or isoelectric point, which suggests that the response is due to molecular recognition rather than a measurement of other physical parameters. We note that larger end groups attached to the PEG-chain were shown to hinder the fluorescent response. Testing the sensor against long insulin fragments of its two chains resulted in reduced response, whereas


6

Page 7 of 31 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

ACS Sensors

shorter insulin peptides were not detected at all, underscoring the specificity of the recognition. Finally, the local insulin concentration can be inferred from the fluorescent response both in buffer and in serum environments, opening the promising possibility of continuous insulin monitoring in vivo.


7

ACS Sensors 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 8 of 31

Chart 1. Library of PEGylated lipids used in this study


8

Page 9 of 31 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

ACS Sensors

PE stands for phosphorylethanolamine. The number adjacent to PEG is its molecular weight in Daltons. The preceding numbers stand for the ratio between saturated to unsaturated carbon bonds.

EXPERIMENTAL SECTION Reagents and Materials. Reagents were purchased from Sigma-Aldrich, unless otherwise noted. Single-walled carbon nanotubes were purchased from NanoIntegris, Inc. PEGylated-lipids were purchased from Avanti Polar Lipids. Fetal Bovine Serum (FBS) was purchased from American Type Culture Collection (ATCC). Short peptides were synthesized by the Koch Institute for Integrative Cancer Research, at Massachusetts Institute of Technology. SWNT suspensions. HiPCO SWNT were initially suspended with 2 wt% sodium cholate (SC) using direct ultrasonication (12 W for 60 minutes, QSonica Q125) followed by ultracentrifugation (150,000 RCF for 4 hours) to separate the individually suspended nanotubes from aggregates and other impurities.31 Subsequently, a solution of 40 mg/L SC-SWNT was mixed with 2 mg/ml PEGylated-lipids (Chart 1) and dialyzed against water with multiple water exchanges, such that the PEG-lipid derivatives exchanged the nanotube wrapping by adsorbing onto the nanotube surface and replacing the small surfactant molecules.38 Out of the 18 PEGylated-lipids used in this work, we have included 6 from our previous study,31 labeled with squared-parenthesis in Chart 1. CoPhMoRe screening. A solution of 1 mg/L SWNT suspended with PEGylated-lipid was added to the wells of a 96-well plate, to which the protein analytes were added to a final concertation of 20 µg/ml in PBS. The fluorescence spectra of the various samples were acquired following 30 minutes incubation using an inverted near-infrared (nIR) fluorescence microscope


9

ACS Sensors 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 10 of 31

coupled to a nitrogen-cooled InGaAs detector through an Acton SP2500 spectrometer (Princeton Instruments). A 785 nm photodiode laser (B&W Tek Inc.) was used for fluorescence excitation. Isothermal titration calorimetry. Samples of PEGylated lipids (5 mg/ml) were injected into an isothermal titration calorimetry cell (VP-ITC MicroCalorimeter, MicroCal) containing either an insulin solution (0.5 mg/ml) or phosphate-buffered saline (PBS) for control. The temperature of the experimental cell was kept constant at 25ᵒC, and the experiment consisted on serial injections, of 10 µl volume each, every 4 minutes into a 2.5 ml volume cell. The data were analyzed using the MicroCal Analysis software. Circular dichroism. Samples were analyzed in the wavelength range of 190 – 260 nm (Aviv Model 202) in 1 nm intervals, using a 1 mm path length quartz cuvette (Hellma).

RESULTS AND DISCUSSION SWNT suspension and initial characterization. The initial SC-SWNT suspension showed bright fluorescence emission, with clearly distinguishable peaks corresponding to the various chiralities in the mixture (Figure 1a). Following dialysis to exchange the SC with the PEGylatedlipids, the suspensions were characterized by UV-vis-nIR photoabsorption, where stable suspensions demonstrated distinct photoabsorption peaks ranging from the ultraviolet (UV) to the near-infrared (nIR) corresponding to SWNT interband transitions (Supporting information, Fig S1). All the suspensions showed bright fluorescence with distinct emission peaks in the nIR (Figure 1b).


10

Page 11 of 31 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

ACS Sensors

The emission peak wavelengths vary between the different suspensions and the corresponding corona phases, an effect referred to as a solvatochromism shift. Sovatochromatic shifts are attributed to the different effective (composite) dielectric constants felt by the fluorescent nanoparticles due differences in their corona phases.39 The semi-empirical functional form that relates this shift to the diameter of the nanotube is given by31: 2ሺߝ − 1ሻ 2ሺ݊ଶ − 1ሻ 1 ߛ ሺ‫ܧ‬ଵଵ ሻଶ∆‫ܧ‬ଵଵ = −‫ ݇ܮ‬ቈ − ቉ ସ= ସ ଶ 2݊ + 1 ݀ 2ߝ + 1 ݀ #ሺ1ሻ ଶ 2ሺߝ − 1ሻ 2ሺ݊ − 1ሻ ߛ ≡ −‫ ݇ܮ‬ቈ − ቉ 2݊ଶ + 1 2ߝ + 1 Where ‫ܧ‬ଵଵ is the first inter-band optical transition energy, ∆‫ܧ‬ଵଵ is the energy difference between the optical transition of the wrapped-nanotubes and the optical transition of a bare nanotube in vacuum or a low dielectric constant medium39 ‫ ܮ‬is a fluctuation factor, ݇ is a scaling constant, ߝ is the dielectric constant, ݊ is the refractive index, ݀ is the nanotube diameter, and ߛ

is a parameter grouping, characterizing the corona phase The fluorescent spectra of the various corona phase – SWNT complexes used in this study were deconvoluted to identify the contributions of the individual chiralities in the SWNT mixture. For each chirality, the peak emission wavelength was found and the solvatochromism shift compared to a pristine SWNT was calculated. The scaling of the solvatochromic shift, ሺ‫ܧ‬ଵଵ ሻଶ ∆‫ܧ‬ଵଵ which is the left hand side of Equation 1, is plotted against ݀ ିସ in Figure 1c and shows a linear trend in the data, as per our work in Choi et al.39 The various suspensions are ranked according to the slope of the fit, ߛ (Figure 1d), where higher values correspond to higher surface coverage.39


11

ACS Sensors 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 12 of 31


12

Page 13 of 31 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

ACS Sensors

Figure 1. (a) Initial fluorescent emission of the SC-SWNT suspensions. The intensity profile is deconvoluted into the various chiralities, which are labeled by the corresponding chiral indexes. Adapted with permission from Ref31. Copyright 2016 Nature Publishing group. (b) Normalized fluorescence spectra of the various SWNT suspensions used in this study. (c) The solvatochromic (wavelength) shift plotted against the diameter to the power of negative four (blue dots) as per our work in Choi et al.39 A linear fit from the analysis of each corresponding spectrum is plotted in red, with identical ordering as in (a). (d) A bar chart showing the slope of the fit for the various SWNT suspensions in ascending order. Results labeled with squaredparenthesis were adapted with permission from Ref31. Copyright 2016 Nature Publishing group. Protein library screening. We screened our nanosensors against a protein library constructed as a part of a previous study.31 Briefly, 14 proteins were selected due to their high abundance or clinical significance in human whole blood, including albumin, Immunoglobulin G (IgG), fibrinogen, α1-antitrypsin, transferrin, haptoglobin, α2-macroglobulin, IgA, IgM, α1acidglycoprotein, apolipoprotein A-I, insulin, human chorionic gonadotropin (hCG), and Creactive protein (CRP). The fluorescence spectra of all the SWNT suspensions, each with a distinct corona phase, were measured at a concentration of 1 mg/L following a 30 minute incubation with the proteins in the library at a concentration of 20 µg/ml in phosphate buffered saline (PBS). The resulting spectra were deconvoluted to assess the contributions of the different chiralities as described in previous studies.27a, 31 The results for the (10,2) chirality are presented in Figure 2a as a heat-map, with yellow-red color for fluorescence intensity increase, and greenwhite color for fluorescence intensity decrease. The intensity changes are normalized by the initial intensity, Io such that ∆I/Io can be tracked. The PEGylated lipid wrapped SWNTs in the heat-map are rank ordered according to their ∆I/Io response towards insulin. The strongest insulin


13

ACS Sensors 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 14 of 31

response (∆I/Io = -74%) was observed for the 18:0 PEG(2000) - carboxylic acid wrapping; however, this suspension showed a comparable response to fibrinogen (∆I/Io = -79%), and a smaller response to albumin (∆I/Io = -24%). The second strongest intensity change for insulin was observed for C16-PEG(2000kDa)-Ceramide (∆I/Io = -62%, Supporting information, Figure S2); however, in this case no other protein induced an intensity decrease of over 5%, and only a single protein, apolipoprotein A-I, induced an intensity increase of ∆I/Io = +25%. Hence, we chose to focus on the C16-PEG(2000kDa)-Ceramide corona phase for further study as an insulin sensor. We chose to focus on the (10,2) chirality since it was easily distinguishable among the various chiralities in the mixture, and the fluorescence response provided the best selectivity and specificity towards insulin. For example, the (7,5) chirality showed only half of the response for insulin with the C16-PEG(2000kDa)-Ceramide, where other corona phases were not selective towards insulin (Supporting information, Figure S3). The (9,4) chirality also showed a much weaker response towards insulin, with many additional non-specific responses to other proteins. Finally, the (11,3) chirality showed a comparable decrease in fluorescence in response to insulin for one of the corona phases; however, apolipoprotein A-I also induced a decrease of ∆I/Io = 23% thus interfering with insulin recognition. Other chiralities were not as easily distinguishable, and thus were not used in the analysis. In conclusion, the (10,2) chirality of the C16PEG(2000kDa)-Ceramide provides the optimal detection of insulin in terms of the relative magnitude of the response, selectivity, and sensitivity. Examining the fluorescence intensity changes in response to insulin versus the solvatochromism slope, ߛ, from Equation 1, no correlation is found between these two


14

Page 15 of 31 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

ACS Sensors

parameters (Figure 2b). However, a statistically significant anti-correlation of -0.52 (‫ < ݌‬0.05) is found between the insulin response and the molecular weight of the end group attached to the PEG chain (Figure 2c). This suggests that larger end groups appear to hinder the interaction of proteins with the nanotube surface and thus prevent fluorescence modulation. Despite this, the insulin response induced by wrappings with no side group at all, which includes the chosen insulin sensor C16-PEG(2000kDa)-Ceramide, vary significantly in their responses ranging between -62% to +2%. As in our previous study of fibrinogen31, an inspection of simple protein parameters such as their hydrophobicity (Figure 2d) or isoelectric point (Figure 2e) yields no correlation with the fluorescence intensity change of the C16-PEG(2000kDa)-Ceramide – SWNT conjugate. Notably, the insulin change appears to be an outlier in these plots. As a side note, due to having a larger data set than the previous study31, we have found a correlation of 0.67 (‫ < ݌‬0.05ሻ between the molecular weight of the PEG chain and the response towards fibrinogen (Supporting information, Figure S4).


15

ACS Sensors 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 16 of 31


16

Page 17 of 31 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

ACS Sensors

Figure 2. (a) Heat map showing the increase or decrease in normalized nIR fluorescence intensity of the (10,2) chirality SWNT to the protein library with the CoPhMoRe phases listed along the x-axis and the protein panel listed along the y-axis. Unless otherwise indicated, all proteins were tested in the presence of SWNT sensors at final concentration of 20 µg/mL. Results labeled with squared-parenthesis were adapted with permission from Ref.31 Copyright 2016 Nature Publishing group. (b) The relative fluorescence intensity change of the (10,2) SWNT chirality is plotted against the slope of the fit for the solvatochromic shift (labeled ‘slope ߛ’) and the (c) molecular weight of the end group of the corona phase. The black dotted line is used as a guide to the eye. (d) The relative fluorescence intensity change of the (10,2) chirality of the C16-PEG(2000kDa)-Ceramide – SWNT is plotted against the hydrophobic surface area of the proteins and (e) their isoelectric point. The two outliers, insulin (intensity decrease) and apoliporprotein-AI (intensity increase) are identified. In order to rule out the possibility that the C16-PEG(2000kDa)-Ceramide – SWNT simply responds to the adsorption of the smallest protein in the library, which happens to be insulin, we tested the response of the insulin sensor to insulin-fragments as peptides of a much smaller molecular weight. Insulin is composed of two chains, α and β, linked by two disulfide bonds (Figure 3a). The α-chain is of 21 amino acids and the β-chain has 30 amino acids. To construct a peptide panel, we first divided the insulin chains into 9 peptides of 7 amino acids: GIVEQCC, TSICSLY, QLENYCN, FVNQHLC, GSHLVEA, LYLVCGE, RGFFYTP, FFYTPKT (Figure 3b). In addition, we included a short sequence of an insulin binding peptide,19 CVEEAS, two amino acid sequences that were showed to induce insulin fibrillation,40 LVEALYL and RGFFYT, and finally, 2 longer fragments of the insulin: GIVEQCCTSICSL (α-chain 1-13), and SHLVEALYLVCGERG (β-chain 9-23). Besides the 1-13 amino acids of the α-chain and the 9-


17

ACS Sensors 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 18 of 31

23 amino acids of the β-chain, that induced 33% and 39% decrease in the fluorescence intensity, respectively, the rest of the peptides showed negligible or non-detectable changes compared to the response to the intact insulin protein (Figure 3c). Hence, there appears to be no correlation between the molecular weight of the analyte and the fluorescence modulation of the insulin sensor (Figure 3d). These results also rule out the small possibility that the C16-PEG(2000kDa)Ceramide – SWNT recognizes a particular amino acid sequence along the insulin protein. The evidence suggests a recognition mechanism that is a more complex than the recognition of a simple peptide sequence. Furthermore checking the circular dichroism (CD) of the peptides shows no structural motif appearing in the data (Supporting information, Figure S5), and the signals represent mostly random coil structures.41 The insulin CD spectrum matches literature reports.42 Based on these findings, we conclude that it is the 3-dimensional conformation of the intact native protein that is crucial for the molecular recognition.


18

Page 19 of 31 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

ACS Sensors


19

ACS Sensors 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 20 of 31

Figure 3. (a) The three-dimensional structure of insulin43 based on the Protein Data Base (PDB) entry 3I3Z reproduced here to highlight the specific fragments that were tested. The α-chain appears in orange and the β-chain in green. The yellow lines represent disulfide bonds. Reproduced with permission of the International Union of Crystallography. (b) The α (blue) and β (green) chains of insulin. The various colors along the amino-acid sequence are used to denote the fragments tested. (c) Relative fluorescence intensity change of the (10,2) chirality of the C16PEG(2000kDa)-Ceramide – SWNT to the peptide panel. (d) Relative fluorescence intensity change of the (10,2) chirality of the C16-PEG(2000kDa)-Ceramide – SWNT to the protein and peptide panels plotted against their molecular weight. The outliers are identified in the plot: insulin, residues 1-13 of the α–chain, and residues 9-23 of and β–chain (all decrease in intensity), as well as apoliporprotein-AI (increase in intensity). Insulin calibration curve. The fluorescence intensity changes of C16-PEG(2000kDa)Ceramide – SWNT in response to insulin concentrations ranging between 180 pM to 3.5 µM were measured. Subsequently, the fluorescence spectra were deconvoluted to assess the contributions of the various chiralities in the mixture, such that each can be analyzed independently. These are plotted for the (6,5), (7,5), (10,2), and (9,4) SWNT chiralities in Figure 4a against the insulin concentration. Assuming that the fluorescence intensity change is linearly proportional to the relative coverage of binding sites of the nanotube surface,27a, 30-31, 34 the data were fitted using the Hill isotherm model44: ሺ‫ܥ‬ூ ሻ௡ ‫ ܫ‬− ‫ܫ‬଴ ߠ௕௢௨௡ௗ = −ߚ = −ߚ #2 ሺ‫ܭ‬ௗ ሻ௡ + ሺ‫ܥ‬ூ ሻ௡ ‫ܫ‬଴ ߠ௧௢௧௔௟ where ‫ܫ‬଴ and ‫ ܫ‬are the initial and final fluorescence intensity respectively, ߚ is a

proportionality factor, ߠ௧௢௧௔௟ and ߠ௕௢௨௡ௗ are the concentration of the total and occupied binding ACS Paragon Plus Environment

20

Page 21 of 31 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

ACS Sensors

sites respectively, ‫ܭ‬ௗ is the dissociation constant, ‫ܥ‬ூ is the insulin concentration, and ݊ is the cooperativity Hill coefficient. The three fit-parameters and their 95% confidence intervals can be seen in Figure 4b. The dissociation constants range between 0.2-0.9 µM, which is within two order of magnitude of previously reported CoPhMoRe targets (0.4 µM for neurotransmitters30, 2-25 µM for the small molecules estradiol, L-thyroxine, and riboflavin27a, 0.003-0.1 µM for fibrinogen45, and 0.01-0.1 µM for lipids35). Note that the Hill coefficients are smaller than 1 indicating negative cooperativity, and the proportionality factor, ߚ, ranges between 0.28 to 0.69 indicating the maximal relative fluorescence response in saturation. Previous work discussed the correlations between the relative fluorescence intensity change to the surface coverage of a specific analyte−corona pair46 as manifested by the proportionality factor ߚ. Future work will relate this parameter to accessible surface area in a more rigorous manner. As anticipated, calibrating the insulin sensor against various concentrations of the insulin α-chain (1-13) and the insulin β-chain (9-23) resulted in dissociation constants of 12.5 µM and 27 µM, respectively, indicating weaker affinity compared to insulin (Supporting information, Figure S6). In order to challenge our sensor in a complex environment, we tested it in the presence of bovine serum (10% Fetal bovine serum, FBS, in PBS) (Figure 4c). Background insulin levels in the serum sample do not exceed 6 pM and thus were ignored. Using similar analysis, we obtained calibration curves for two chiralities, the (6,5) and (10,2), which were well described by Equation 2. Dissociation constants were larger, ranging between 11-13 µM, indicative of the weaker affinity in the serum environment, the Hill coefficients were 0.95 and 1.4 respectively, indicating either small negative cooperativity or positive cooperativity respectively, and the proportionality factor, ߚ, was reduced to 0.26 for both chiralities, which could be due to ACS Paragon Plus Environment

21

ACS Sensors 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 22 of 31

screening by other factors in the serum. For example, cholesterol, which is found in the FBS sample in comparable molar concentrations to the added insulin, has been shown to interact with SWNT surfaces35 and can thus be one of the sources of non-specific binding that competes with insulin and causes the reduced response. In addition, these results may also represent the limitations in using Equation 2 to describe what is ultimately a complex adsorption process. Our previous work on a related CoPhMoRe phase sensitive to fibrinogen found agreement with a three-step adsorption process, suggesting that a more complex mechanism could exist for the insulin sensor described in this work as well.31 The successful demonstration of this synthetic recognition site for insulin, using the corona phase of single-walled carbon nanotubes, is a crucial milestone in the path towards continuous insulin monitoring in vivo. Identifying a robust recognition site is often the bottleneck in the detection of biological molecules,12 mainly due to the inherent limitations of natural recognition systems24 and the search for new technologies for this purpose is an active area of research.27a, 29c, 47

Our approach of corona phase molecular recognition was proven useful for the recognition and

quantification of insulin both in buffer and in serum environments. One of the challenges is the detection range of our sensor being higher than physiological insulin concentration of 57-79 pM (0.33 – 0.46 ng/ml) during fasting or up to 430 pM (2.5 ng/ml) after meals.48 Improving the detection limit will be the subject of future research and will consist of screening derivatives of the C16-PEG(2000kDa)-Ceramide corona with input from theoretical design tools.8, 37 An in vivo detection platform would consist of a biocompatible hydrogel encapsulating the SWNT sensors for implantation in a location of interest. The advantages of a hydrogel implant include accurate placement with a minimally invasive procedure, ease of fluorescent excitation and collection from an external device, prevention of systemic distribution of the nanoparticle sensors, and


22

Page 23 of 31 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

ACS Sensors

reducing non-specific interaction by engineering the hydrogel properties such as pore size, available functional groups, surface charge, and pH. Optimizing the sensor properties has been reported previously8 and our group’s previous work has successfully demonstrated the fluorescent detection of such hydrogel implants in vivo and in tissue phantoms.26, 28d Insulin and corona phase affinity. To verify that the CoPhMoRe phase itself, when bound to the nanotube surface, is responsible for the binding, and not the free lipid, the independent binding of insulin to the free lipid was explored. The binding affinity between insulin and the C16-PEG(2000kDa)-Ceramide, without the nanotubes, was tested using isothermal titration calorimetry (ITC).49

The ITC instrument (MicroCal VP-ITC, Malvern) monitors the heat

released or absorbed during a series of consecutive injections of 10 µL of 5 mg/mL of C16PEG(2000kDa)-Ceramide in PBS, into 2.5 mL of 0.5 mg/mL insulin in PBS (Figure 4d) or just PBS without insulin for control (Figure 4e). According to the binding isotherms, which are calculated by integrating the heat pulses of each injection with respect to time (Figure 4f), the heat changes of the C16-PEG(2000kDa)-Ceramide injections into an 85 µM insulin solution almost overlap with the heat changes of the C16-PEG(2000kDa)-Ceramide injections into PBS, indicating negligible binding between the C16-PEG(2000kDa)-Ceramide and insulin.50-51 These finding support the necessity of the CoPhMoRe phase, resulting from the pinned configuration the C16-PEG(2000kDa)-Ceramide corona adopted when adsorbed around the nanotube scaffold, to successfully detecting the insulin protein.


23

ACS Sensors 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 24 of 31

Figure 4. (a) Calibration curves of sensor response in PBS environment versus insulin concentration for (6,5), (7,5), (10,2), and (9,4) SWNT chiralities and the corresponding fits using Equation 2. (b) The three fit parameters according to Equation 2 used to fit the data in (a). (c) Calibration curves of sensor response in 10% FBS environment. (d) Isothermal Titration Calorimetry of C16-PEG(2000kDa)-Ceramide into a 85 µM insulin solution or (e) PBS. (f) Binding isotherms for the titration of C16-PEG(2000kDa)-Ceramide into insulin solution (circles) or PBS (squares) plotted against the molar ratio of C16-PEG(2000kDa)-Ceramide to insulin. The overlapping curves of the injections into insulin or PBS indicate that the released heat in both cases is identical, and we conclude that insulin does not interact significantly with C16PEG(2000kDa)-Ceramide without the presence of the nanotube interface.


24

Page 25 of 31 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

ACS Sensors

CONCLUSIONS In summary, we have demonstrated a synthetic, chemically stable, antibody analog of insulin. Using the corona phase molecular recognition (CoPhMoRe) approach, we have discovered a complex of a fluorescent single-walled carbon nanotube suspended by a PEGylated lipid amphiphilic hetero-polymer that recognizes insulin, as shown by a concentration-calibrated fluorescence intensity decrease. The analysis of insulin fragments emphasizes the importance of the three-dimensional conformation of the insulin protein, identifying its structure as the underlying motif of the recognition. Future work will develop such nanosensors into implantable devices by encapsulation in a bio-compatible hydrogel for screening interferents and precise implant positioning, allowing them to be excited and queried non-invasively from an external device.8, 26, 28a, 28c, d, 29a, 47, 52 Our discovery points to the continuous monitoring of insulin in vivo, owing to the high stability of the synthetic recognition site and the optical signal transduction in the near infrared portion of the spectrum, which enables signal transmission through human tissue.28d28d, 53

Supporting Information. Supporting Information Available: The following files are available free of charge. Supplementary figures. Absorption spectra, heat maps, fluorescent response, circular dichroism spectra, and calibration curves. AUTHOR INFORMATION Corresponding Author *E-mail: [email protected].


25

ACS Sensors 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 26 of 31

Author Contributions M.S.S and G.B conceived the original concept, designed the experiments, and wrote the manuscript with N. B.. G.B and N. B. performed the experiments, with the assistance of E.B.O, and analyzed the data. All authors discussed the results and commented on the manuscript. ‡These authors contributed equally.

Funding Sources This research was partly supported by the Juvenile Diabetes Research Foundation. G.B. acknowledges the Technion-MIT postdoctoral fellowship. The Biophysical Instrumentation Facility for the Study of Complex Macromolecular Systems (NSF-0070319) is gratefully acknowledged.

Notes The authors declare no competing financial interest.

ABBREVIATIONS SWNT, Single-walled carbon nanotube; CoPhMoRe, Corona phase molecular recognition.


26

Page 27 of 31 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

ACS Sensors

REFERENCES 1. Sonksen, P.; Sonksen, J., Insulin: understanding its action in health and disease. British Journal of Anaesthesia 2000, 85 (1), 69-79. 2. Pankaj, M., Diabetes Beyond Insulin: Review of New Drugs for Treatment of Diabetes Mellitus. Current Drug Discovery Technologies 2007, 4 (1), 39-47. 3. Global report on diabetes. World Health Organization, Geneva 2016. 4. Daneman, D., Type 1 diabetes. The Lancet 2006, 367 (9513), 847-858. 5. (a) Choudhary, P.; Ramasamy, S.; Green, L.; Gallen, G.; Pender, S.; Brackenridge, A.; Amiel, S. A.; Pickup, J. C., Real-Time Continuous Glucose Monitoring Significantly Reduces Severe Hypoglycemia in Hypoglycemia-Unaware Patients With Type 1 Diabetes. Diabetes Care 2013, 36 (12), 4160-4162; (b) Jacobi, J.; Bircher, N.; Krinsley, J.; Agus, M.; Braithwaite, S. S.; Deutschman, C.; Freire, A. X.; Geehan, D.; Kohl, B.; Nasraway, S. A., Guidelines for the use of an insulin infusion for the management of hyperglycemia in critically ill patients. Critical care medicine 2012, 40 (12), 3251-3276; (c) Doyle, E. A.; Weinzimer, S. A.; Steffen, A. T.; Ahern, J. A. H.; Vincent, M.; Tamborlane, W. V., A randomized, prospective trial comparing the efficacy of continuous subcutaneous insulin infusion with multiple daily injections using insulin glargine. Diabetes Care 2004, 27 (7), 1554-1558. 6. (a) Steil, G. M.; Panteleon, A. E.; Rebrin, K., Closed-loop insulin delivery—the path to physiological glucose control. Advanced Drug Delivery Reviews 2004, 56 (2), 125-144; (b) Breton, M.; Farret, A.; Bruttomesso, D.; Anderson, S.; Magni, L.; Patek, S.; Dalla Man, C.; Place, J.; Demartini, S.; Del Favero, S., Fully integrated artificial pancreas in type 1 diabetes. Diabetes 2012, 61 (9), 2230-2237; (c) Kovatchev, B.; Tamborlane, W. V.; Cefalu, W. T.; Cobelli, C., The artificial pancreas in 2016: a digital treatment ecosystem for diabetes. Diabetes Care 2016, 39 (7), 1123-1126. 7. Maynard, G.; Wesorick, D. H.; O’Malley, C.; Inzucchi, S. E.; Force, S. o. H. M. G. C. T., Subcutaneous insulin order sets and protocols: effective design and implementation strategies. J Hosp Med 2008, 3 (5 Suppl), 29-41. 8. Bisker, G.; Iverson, N. M.; Ahn, J.; Strano*, M. S., A Pharmacokinetic Model of a Tissue Implantable Insulin Sensor. Advanced Healthcare Materials 2015, 4 (1), 87-97. 9. Newman, J. D.; Turner, A. P., Home blood glucose biosensors: a commercial perspective. Biosensors and bioelectronics 2005, 20 (12), 2435-2453. 10. Katz, A.; Nambi, S. S.; Mather, K.; Baron, A. D.; Follmann, D. A.; Sullivan, G.; Quon, M. J., Quantitative Insulin Sensitivity Check Index: A Simple, Accurate Method for Assessing Insulin Sensitivity In Humans. The Journal of Clinical Endocrinology & Metabolism 2000, 85 (7), 2402-2410. 11. Bakh, N. A.; Bisker, G.; Lee, M. A.; Gong, X.; Strano, M. S., Rational Design of Glucose-Responsive Insulin Using Pharmacokinetic Modeling. Adv Healthc Mater 2017, 1700601-1700610. 12. (a) Iqbal, S. S.; Mayo, M. W.; Bruno, J. G.; Bronk, B. V.; Batt, C. A.; Chambers, J. P., A review of molecular recognition technologies for detection of biological threat agents. Biosensors and bioelectronics 2000, 15 (11), 549-578; (b) Rebek, J., Molecular recognition with model systems. Angewandte Chemie International Edition in English 1990, 29 (3), 245-255; (c) Brooijmans, N.; Kuntz, I. D., Molecular recognition and docking algorithms. Annual review of biophysics and biomolecular structure 2003, 32 (1), 335-373.


27

ACS Sensors 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 28 of 31

13. Ullrich, A.; Bell, J. R.; Chen, E. Y.; Herrera, R.; Petruzzelli, L. M.; Dull, T. J.; Gray, A.; Coussens, L.; Liao, Y. C.; Tsubokawa, M.; et al., Human insulin receptor and its relationship to the tyrosine kinase family of oncogenes. Nature 1985, 313 (6005), 756-761. 14. Kasuga, M.; Karlsson, F. A.; Kahn, C. R., Insulin stimulates the phosphorylation of the 95,000-dalton subunit of its own receptor. Science 1982, 215 (4529), 185-187. 15. Schernthaner, G., Immunogenicity and allergenic potential of animal and human insulins. Diabetes Care 1993, 16 Suppl 3, 155-165. 16. Hattori, N.; Duhita, M. R.; Mukai, A.; Matsueda, M.; Shimatsu, A., Development of insulin antibodies and changes in titers over a long-term period in patients with type 2 diabetes. Clin Chim Acta 2014, 433, 135-138. 17. Yalow, R. S.; Berson, S. A., Plasma Insulin Concentrations in Nondiabetic and Early Diabetic Subjects: Determinations by a New Sensitive Immuno-assay Technic. Diabetes 1960, 9 (4), 254-260. 18. Fineberg, S. E.; Kawabata, T. T.; Finco-Kent, D.; Fountaine, R. J.; Finch, G. L.; Krasner, A. S., Immunological responses to exogenous insulin. Endocr Rev 2007, 28 (6), 625-652. 19. Knutson, V. P., Insulin-binding peptide. Design and characterization. Journal of Biological Chemistry 1988, 263 (28), 14146-14151. 20. Stoltenburg, R.; Reinemann, C.; Strehlitz, B., SELEX—a (r) evolutionary method to generate high-affinity nucleic acid ligands. Biomolecular engineering 2007, 24 (4), 381-403. 21. (a) Song, K. M.; Lee, S.; Ban, C., Aptamers and Their Biological Applications. SensorsBasel 2012, 12 (1), 612-631; (b) Landry, M. P.; Ando, H.; Chen, A. Y.; Cao, J. C.; Kottadiel, V. I.; Chio, L.; Yang, D.; Dong, J. Y.; Lu, T. K.; Strano, M. S., Single-molecule detection of protein efflux from microorganisms using fluorescent single-walled carbon nanotube sensor arrays. Nat Nanotechnol 2017, 12 (4), 368-377. 22. Connor, A. C.; Frederick, K. A.; Morgan, E. J.; McGown, L. B., Insulin Capture by an Insulin-Linked Polymorphic Region G-Quadruplex DNA Oligonucleotide. Journal of the American Chemical Society 2006, 128, 4986-4991. 23. Yoshida, W.; Mochizuki, E.; Takase, M.; Hasegawa, H.; Morita, Y.; Yamazaki, H.; Sode, K.; Ikebukuro, K., Selection of DNA aptamers against insulin and construction of an aptameric enzyme subunit for insulin sensing. Biosensors & bioelectronics 2009, 24, 1116-1120. 24. (a) Vlatakis, G.; Andersson, L. I.; Müller, R.; Mosbach, K., Drug assay using antibody mimics made by molecular imprinting. Nature 1993, 361 (6413), 645-647; (b) Wulff, G., Molecular imprinting in cross‐linked materials with the aid of molecular templates—a way towards artificial antibodies. Angewandte Chemie International Edition in English 1995, 34 (17), 1812-1832; (c) Mosbach, K.; Ramström, O., The emerging technique of molecular imprinting and its future impact of biotechnology. Bio/technology 1996, 14 (2), 163-170. 25. Schirhagl, R.; Latif, U.; Podlipna, D.; Blumenstock, H.; Dickert, F. L., Natural and Biomimetic Materials for the Detection of Insulin. Analytical Chemistry 2012, 84 (9), 39083913. 26. Iverson, N. M.; Barone, P. W.; Shandell, M.; Trudel, L. J.; Sen, S.; Sen, F.; Ivanov, V.; Atolia, E.; Farias, E.; McNicholas, T. P.; Reuel, N.; Parry, N. M. A.; Wogan, G. N.; Strano, M. S., In vivo biosensing via tissue-localizable near-infrared-fluorescent single-walled carbon nanotubes. Nat Nano 2013, 8 (11), 873-880. 27. (a) Zhang, J.; Landry, M. P.; Barone, P. W.; Kim, J.-H.; Lin, S.; Ulissi, Z. W.; Lin, D.; Mu, B.; Boghossian, A. A.; Hilmer, A. J.; Rwei, A.; Hinckley, A. C.; Kruss, S.; Shandell, M. A.; Nair, N.; Blake, S.; Sen, F.; Sen, S.; Croy, R. G.; Li, D.; Yum, K.; Ahn, J.-H.; Jin, H.; Heller, D.


28

Page 29 of 31 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

ACS Sensors

A.; Essigmann, J. M.; Blankschtein, D.; Strano, M. S., Molecular recognition using corona phase complexes made of synthetic polymers adsorbed on carbon nanotubes. Nat Nanotechnol 2013, 8 (12), 959-968; (b) Landry, M.; Kruss, S.; Nelson, J.; Bisker, G.; Iverson, N.; Reuel, N.; Strano, M., Experimental Tools to Study Molecular Recognition within the Nanoparticle Corona. Sensors-Basel 2014, 14 (9), 16196-16211. 28. (a) Barone, P. W.; Baik, S.; Heller, D. A.; Strano, M. S., Near-infrared optical sensors based on single-walled carbon nanotubes. Nature Materials 2005, 4 (1), 86–92; (b) Boghossian, A. A.; Zhang, J. Q.; Barone, P. W.; Reuel, N. F.; Kim, J. H.; Heller, D. A.; Ahn, J. H.; Hilmer, A. J.; Rwei, A.; Arkalgud, J. R.; Zhang, C. T.; Strano, M. S., Near-Infrared Fluorescent Sensors based on Single-Walled Carbon Nanotubes for Life Sciences Applications. Chemsuschem 2011, 4 (7), 848-863; (c) Kruss, S.; Hilmer, A. J.; Zhang, J.; Reuel, N. F.; Mu, B.; Strano, M. S., Carbon nanotubes as optical biomedical sensors. Advanced Drug Delivery Reviews 2013, 65 (15), 1933-1950; (d) Iverson, N. M.; Bisker, G.; Farias, E.; Ivanov, V.; Ahn, J.; Wogan, G. N.; Strano, M. S., Quantitative Tissue Spectroscopy of Near Infrared Fluorescent Nanosensor Implants. Journal of Biomedical Nanotechnology 2016, 12 (5), 1035-1047. 29. (a) Giraldo, J. P.; Landry, M. P.; Faltermeier, S. M.; McNicholas, T. P.; Iverson, N. M.; Boghossian, A. A.; Reuel, N. F.; Hilmer, A. J.; Sen, F.; Brew, J. A.; Strano, M. S., Plant nanobionics approach to augment photosynthesis and biochemical sensing. Nature materials 2014, 13 (4), 400-408; (b) Oliveira, S. F.; Bisker, G.; Bakh, N. A.; Gibbs, S. L.; Landry, M. P.; Strano, M. S., Protein functionalized carbon nanomaterials for biomedical applications. Carbon 2015, 95, 767-779; (c) Kwak, S. Y.; Wong, M. H.; Lew, T. T. S.; Bisker, G.; Lee, M. A.; Kaplan, A.; Dong, J.; Liu, A. T.; Koman, V. B.; Sinclair, R.; Hamann, C.; Strano, M. S., Nanosensor Technology Applied to Living Plant Systems. Annu Rev Anal Chem (Palo Alto Calif) 2017, 10 (1), 113-140; (d) Wong, M. H.; Giraldo, J. P.; Kwak, S. Y.; Koman, V. B.; Sinclair, R.; Lew, T. T.; Bisker, G.; Liu, P.; Strano, M. S., Nitroaromatic detection and infrared communication from wild-type plants using plant nanobionics. Nat Mater 2017, 16 (2), 264-272. 30. Kruss, S.; Landry, M. P.; Vander Ende, E.; Lima, B. M. A.; Reuel, N. F.; Zhang, J.; Nelson, J.; Mu, B.; Hilmer, A.; Strano, M., Neurotransmitter Detection Using Corona Phase Molecular Recognition on Fluorescent Single-Walled Carbon Nanotube Sensors. Journal of the American Chemical Society 2013, 136 (2), 713-724. 31. Bisker, G.; Dong, J.; Park, H. D.; Iverson, N. M.; Ahn, J.; Nelson, J. T.; Landry, M. P.; Kruss, S.; Strano, M. S., Protein-targeted corona phase molecular recognition. Nature Communications 2016, 7, 10241-10254. 32. Landry, M. P.; Vuković, L.; Kruss, S.; Bisker, G.; Landry, A. M.; Islam, S.; Jain, R.; Schulten, K.; Strano, M. S., Comparative Dynamics and Sequence Dependence of DNA and RNA Binding to Single Walled Carbon Nanotubes. The Journal of Physical Chemistry C 2015, 119 (18), 10048-10058. 33. Salem, D. P.; Landry, M. P.; Bisker, G.; Ahn, J.; Kruss, S.; Strano, M. S., Chirality dependent corona phase molecular recognition of DNA-wrapped carbon nanotubes. Carbon 2016, 97, 147-153. 34. Nelson, J. T.; Kim, S.; Reuel, N. F.; Salem, D. P.; Bisker, G.; Landry, M. P.; Kruss, S.; Barone, P. W.; Kwak, S.; Strano, M. S., Mechanism of Immobilized Protein A Binding to Immunoglobulin G on Nanosensor Array Surfaces. Analytical Chemistry 2015, 87 (16), 81868193. 35. Jena, P. V.; Roxbury, D.; Galassi, T. V.; Akkari, L.; Horoszko, C. P.; Iaea, D. B.; Budhathoki-Uprety, J.; Pipalia, N.; Haka, A. S.; Harvey, J. D.; Mittal, J.; Maxfield, F. R.; Joyce,


29

ACS Sensors 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 30 of 31

J. A.; Heller, D. A., A Carbon Nanotube Optical Reporter Maps Endolysosomal Lipid Flux. ACS Nano 2017, 10689–10703. 36. Polo, E.; Kruss, S., Impact of Redox-Active Molecules on the Fluorescence of PolymerWrapped Carbon Nanotubes. The Journal of Physical Chemistry C 2016, 120 (5), 3061-3070. 37. Bisker, G.; Ahn, J.; Kruss, S.; Ulissi, Z. W.; Salem, D. P.; Strano, M. S., A Mathematical Formulation and Solution of the CoPhMoRe Inverse Problem for Helically Wrapping Polymer Corona Phases on Cylindrical Substrates. The Journal of Physical Chemistry C 2015, 119 (24), 13876-13886. 38. Welsher, K.; Liu, Z.; Sherlock, S. P.; Robinson, J. T.; Chen, Z.; Daranciang, D.; Dai, H., A route to brightly fluorescent carbon nanotubes for near-infrared imaging in mice. Nat Nanotechnol 2009, 4 (11), 773-780. 39. Choi, J. H.; Strano, M. S., Solvatochromism in single-walled carbon nanotubes. Applied Physics Letters 2007, 90 (22), 223114-223116. 40. Chiang, H.-L.; Ngo, S. T.; Chen, C.-J.; Hu, C.-K.; Li, M. S., Oligomerization of Peptides LVEALYL and RGFFYT and Their Binding Affinity to Insulin. PLoS ONE 2013, 8 (6), e65358e65368. 41. Greenfield, N. J.; Fasman, G. D., Computed circular dichroism spectra for the evaluation of protein conformation. Biochemistry 1969, 8 (10), 4108-4116. 42. Pocker, Y.; Biswas, S. B., Conformational dynamics of insulin in solution. Circular dichroic studies. Biochemistry 1980, 19 (22), 5043-5049. 43. Timofeev, V. I.; Chuprov-Netochin, R. N.; Samigina, V. R.; Bezuglov, V. V.; Miroshnikov, K. A.; Kuranova, I. P., X-ray investigation of gene-engineered human insulin crystallized from a solution containing polysialic acid. Acta Crystallographica Section F 2010, 66 (3), 259-263. 44. Foo, K. Y.; Hameed, B. H., Insights into the modeling of adsorption isotherm systems. Chemical Engineering Journal 2010, 156 (1), 2-10. 45. Bisker, G.; Dong, J.; Park, H. D.; Iverson, N. M.; Ahn, J.; Nelson, J. T.; Landry, M. P.; Kruss, S.; Strano, M. S., Protein-targeted corona phase molecular recognition. Nat Commun 2016, 7. 46. Ulissi, Z. W.; Zhang, J.; Sresht, V.; Blankschtein, D.; Strano, M. S., 2D equation-of-state model for corona phase molecular recognition on single-walled carbon nanotube and graphene surfaces. Langmuir 2015, 31 (1), 628-636. 47. Alvarez, M. M.; Aizenberg, J.; Analoui, M.; Andrews, A. M.; Bisker, G.; Boyden, E. S.; Kamm, R. D.; Karp, J. M.; Mooney, D. J.; Oklu, R.; Peer, D.; Stolzoff, M.; Strano, M. S.; Trujillo-de Santiago, G.; Webster, T. J.; Weiss, P. S.; Khademhosseini, A., Emerging Trends in Micro- and Nanoscale Technologies in Medicine: From Basic Discoveries to Translation. ACS Nano 2017, 11 (6), 5195-5214. 48. Iwase, H.; Kobayashi, M.; Nakajima, M.; Takatori, T., The ratio of insulin to C-peptide can be used to make a forensic diagnosis of exogenous insulin overdosage. Forensic Science International 2001, 115 (1), 123-127. 49. Freire, E.; Mayorga, O. L.; Straume, M., Isothermal titration calorimetry. Analytical Chemistry 1990, 62 (18), 950A-959A. 50. Velázquez‐Campoy, A.; Ohtaka, H.; Nezami, A.; Muzammil, S.; Freire, E., Isothermal titration calorimetry. Current Protocols in Cell Biology 2004, 17.8.1–17.8.24. 51. Leavitt, S.; Freire, E., Direct measurement of protein binding energetics by isothermal titration calorimetry. Current opinion in structural biology 2001, 11 (5), 560-566.


30

Page 31 of 31 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

ACS Sensors

52. (a) Giraldo, J. P.; Landry, M. P.; Kwak, S.-Y.; Jain, R. M.; Wong, M. H.; Iverson, N. M.; Ben-Naim, M.; Strano, M. S., A Ratiometric Sensor Using Single Chirality Near-Infrared Fluorescent Carbon Nanotubes: Application to In Vivo Monitoring. Small 2015, 11 (32), 39733984; (b) Lee, M. A.; Bakh, N.; Bisker, G.; Brown, E. N.; Strano, M. S., A Pharmacokinetic Model of a Tissue Implantable Cortisol Sensor. Adv Healthc Mater 2016, 5 (23), 3004-3015. 53. Tao, Z.; Hong, G.; Shinji, C.; Chen, C.; Diao, S.; Antaris, A. L.; Zhang, B.; Zou, Y.; Dai, H., Biological imaging using nanoparticles of small organic molecules with fluorescence emission at wavelengths longer than 1000 nm. Angewandte Chemie International Edition 2013, 52 (49), 13002-13006.

For TOC only


31

Insulin detection using a corona phase molecular recognition site on

Recommend Documents