The Case for Human Serum as a Highly Preferable Sample Matrix for

Oct 23, 2018 - Jennifer H. Granger and Marc D. Porter .... Congratulations to 2018 Nobel Prize in Chemistry Winners Frances H. Arnold, George P. Smith...
0 downloads 0 Views 945KB Size
Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Article

The Case for Human Serum as a Highly Preferable Sample Matrix for Detection of Anthrax Toxins Jennifer H. Granger, and Marc D. Porter ACS Sens., Just Accepted Manuscript • DOI: 10.1021/acssensors.8b00566 • Publication Date (Web): 23 Oct 2018 Downloaded from http://pubs.acs.org on October 24, 2018

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

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

Page 1 of 22 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 Case for Human Serum as a Highly Preferable Sample Matrix for Detection of Anthrax Toxins Jennifer H. Granger1,* and Marc D. Porter1, 2* 1Nano

Institute of Utah and 2Departments of Chemistry and Chemical Engineering University of Utah, Salt Lake City, UT 84112

ABSTRACT: This paper describes preliminary results on the surprising impact of human serum as a sample matrix on the detectability of protective antigen (PA) and lethal factor (LF), two antigenic protein markers of Bacillus anthracis, in a heterogeneous immunometric assay. Two sample matrices were examined: human serum and physiological buffer. Human serum is used as a specimen in the diagnostic testing of potentially infected individuals. Physiological buffers are often applied to the recovery of biomarkers dispersed in suspicious white powders and other suspect specimens, and as a serum diluent in order to combat contributions to the measured test response from nonspecific adsorption. The results of these experiments using a sandwich immunoassay read out by surface-enhanced Raman scattering (SERS), yielded estimates for the limit of detection (LOD) for both markers when using spiked human serum that were remarkably lower when compared to spiked physiological buffer (~70,000× for PA and ~25,000× for LF). The difference in LODs is attributed to a degradation in the effectiveness of the capture and/or labeling steps in the immunoassay due to the known propensity for both proteins to denature in buffer. These findings indicate that the use of physiological buffer for serum dilution or recovery from a powdered matrix is counter to the low-level detection of these two antigenic proteins. The potential implications of these results with respect to the ability to detect markers of other pathogenic agents are briefly discussed. Keywords: Anthrax toxins, Surface-enhanced Raman Scattering (SERS), Immunoassay, Serum Corresponding author email: [email protected] 1 ACS Paragon Plus Environment

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

This paper describes the preliminary results and potential implications from a comparison of immunometric measurements for two anthrax markers, protective antigen (PA) and lethal factor (LF), when spiked into either human serum or physiological buffer. Tests using serum are part of a protocol for identifying individuals infected with the disease.1, 2 Measurements in buffer reflect procedures employed for biomarker analysis in suspicious white powders and other suspect specimens.3 Buffers are also employed for the dilution of serum samples as a means to minimize the background response due to nonspecific adsorption.4, 5 Anthrax is a devastating disease caused by the endospore-forming bacterium Bacillus anthracis.6 Its use as a potential threat to public health was reinforced in 2001 when letters containing a white powder laced with anthrax spores were dispatched through the U.S. Postal system to offices of news media and U.S. Congress members.7 Based on its lethality, ease of dissemination, and need for enhanced preparedness in the event of a biological attack, B. anthracis has been classified by the U.S. National Institute of Allergy and Infectious Diseases as a Category A priority pathogen.8, 9 This designation has been adopted by a number of other countries around the world.10, 11 The most likely route for anthrax exposure in biowarfare is inhalation. When inhaled, anthrax spores begin to germinate in the lymph nodes of the chest. Germination leads to the emergence of vegetative bacteria in the circulatory system of the host and the release of PA, LF, and edema factor (EF). Individually, these proteins are nontoxic.12 However, it is the combination of PA with LF, called lethal toxin (LeTx),13 or PA with EF, termed edema toxin (EdTx), that produces the etiological agents for the disease.12, 14 The clinical symptoms of early-stage infection – fever, cough, and malaise – are generic to many common ailments, resulting in high rates of misdiagnosis and failure to begin treatment when it can be the most effective.15 It may only take

2 ACS Paragon Plus Environment

Page 2 of 22

Page 3 of 22 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 day before the infection spreads from the lymph nodes to the lungs, potentially leading to septic shock, coma, and death.6, 13 The high lethality of this infection makes it imperative that exposure to anthrax be confirmed as soon as possible. This places a premium on the creation of tests with low LODs and rapid turnaround times that can readily be used to analyze anthrax in suspicious white powders, soil samples, and body fluid specimens. One of the major challenges faced by first responders and health care providers at the time of the "Anthrax letters" was the lack of a diagnostic test that could be used to analyze suspect samples with the immediacy needed to quickly and reliably assess whether a threat was real. In 2001, anthrax testing relied heavily on culturing, which could take up to 5 days to reach a confirmatory response.16 7 This situation, as recently reviewed,17, 18 triggered the development of a number of immunoassays [e.g., ELISA,19, 20 lateral flow architectures (LFA),20 surface plasmon resonance (SPR),21 quantum dots (QDs),22 and surface-enhanced Raman scattering (SERS)23-27] designed for the low-level detection of LF, PA, and other25, 27-30 antigenic markers for B. anthracis. As part of our interests in exploring the strengths of SERS in the diagnostic testing arena,31-33 we recently completed a first set of immunometric studies targeted at the detection of LF and PA spiked into human serum. These experiments also encompassed running SERS immunoassays after spiking each marker in HEPES buffer augmented with 1% bovine serum albumin (BSA), i.e., HEPES-B, to ascertain how the measurement of these markers may be affected by a complex sample matrix (serum) as opposed to a seemingly more innocuous one (buffer). HEPES-B is the dissolution matrix recommended by the commercial source that we used for LF and PA (List Biological Laboratories), which follows the procedures described in the early purification work on these toxins by Fish et al.34 Our results show that sample matrix has a

3 ACS Paragon Plus Environment

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

profound impact on the detectability of both markers; but these results are strongly counter to the expectation that measurements in buffer should prove more effective than those in serum. Estimates from these data of the limit of detection (LOD) for both markers when using spiked human serum were remarkably lower (~70,000× for PA and ~25,000× for LF) when compared to spiked physiological buffer. This paper describes the design and results of these experiments, along with evidence in the literature that supports these findings,35 and provides possible insights into their mechanistic origin.36, 37 It also briefly discusses the implications of these findings with respect to the ability to detect other markers of pathogenic agents; meaning, are there other cases in which serum is preferable for sample dissolution? The design of the SERS immunoassay38 for the detection of LF and PA, denoted as "anthrax antigen," is shown in Figure 1. It uses SERS for readout and anti-LF (LF) or anti-PA (PA) antibodies adsorbed onto separate gold-coated well bottoms in a 96-well microplate, which were applied to capture either LF or PA spiked into either 5.0 mM HEPES-buffered saline (pH = 7.5) containing 1% bovine serum albumin (HEPES-B) or undiluted human serum. The captured antigens were subsequently tagged with a bi-functional gold nanoparticle label, termed an extrinsic Raman label (ERL), which was modified with a monolayer of a Raman reporter molecule (RRM)38, 39 followed by a layer of LF or PA antibodies. Spectrophotometric extinction measurements40 indicate that these anti-LF and anti-PA ERLs, which were modified and then dispersed in 2.0 mM borate buffer (BB) at pH 8.5 with 1 % BSA, are stable for a few weeks (longer times have not yet been tested). These particles can form a loose sediment41 when left standing for 24 h or so, but can be fully resuspended with gentle agitation. These augmented BB suspensions were used to tag either LF or PA after capture.

4 ACS Paragon Plus Environment

Page 4 of 22

Page 5 of 22 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

After challenging the immobilized capture antibody to an either LF or PA for 6 h, the wells were washed three times with 100 μL of 1% BSA in 2.0 mM BB. Finally, the capture antigen was tagged with the appropriately modified ERLs (30 μL/well) for ~8 h, followed by two rinses with 100 μL BB and two rinses with 100 μL deionized water. The plate was allowed to dry in air for ~1 h before readout. After labeling, each well was interrogated with SERS to determine the amount of captured antigen. The next sections provide more detail on the architecture of the assay and procedural steps. Further details on materials used, methods for the fabrication, characterization, and testing of each component of the immunoassays, and the procedures for running the immunoassays are provided in the Supporting Information (SI). The SI also includes a brief description of the approaches to, and results from, screening studies that identified the antibodies used in this work from a small set of candidates. The bottoms of the wells in a 96-well microplate were coated with ~200 nm of gold by electron beam deposition. This thickness ensures that the dielectric properties of bulk gold are retained42, 43 and, therefore, exhibit reproducibility in the plasmonic coupling of the gold core of the ERLS and the underlying gold capture substrate.44, 45 An aluminum mask was constructed as an insert to avoid coating the sidewalls of the wells with gold. The next step modified the gold addresses by exposure to 100 μL of a 1.0 mM ethanolic solution of dithiobis(succinimidyl propionate) (DSP) for 4 h. The addresses were rinsed three times with 100 μL of fresh ethanol to remove residual materials and dried under a directed stream of high purity nitrogen. This was followed by treating the coated addresses with 40 μL of a 10.0 μg/mL solution of either LF (List Biological Laboratories, Lot #772B) or PA (ViroStat, Lot #7825) in PBS for 3 h. Each

5 ACS Paragon Plus Environment

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

well was then washed three times with 100 μL of PBS and blocked by adding 40 μL of 1% BSA in 2.0 mM borate buffer (BB). After 1 h, the blocking solution was carefully removed by aspiration, and 30 μL of LF or PA calibration standards were pipetted into separate wells. These standards were prepared from asreceived aliquots of 0.1 mg lyophilized LF (List Biological Laboratories, Lot# 1692A1A) or PA (List Biological Laboratories Lot# 171D) that were reconstituted per manufacturer recommendations in 100 μL of 1.0 mg/mL BSA to a stock concentration of 1.0 mg/mL, dispensed into 5 μL aliquots, and stored at -20 o C until use. Solutions for generating calibration curves were prepared by serial dilutions of the thawed stocks with HEPES-B as recommended by the manufacturer or control human serum (Randox Laboratories). The diluents were also used as sample blanks. Upon completion of the antigen capture step (3 h), the calibrant solution was carefully removed with a pipette. Finally, captured antigen was tagged by treatment with the appropriate ERLs (30 μL/well) for ~18 h, followed by two rinses with 100 μL of BB and two rinses with 100 μL of deionized water. The ERL suspensions were prepared by the chemisorption of a monolayer of 5, 5’-dithiobis(succinimidyl-2-nitrobenzoate) (DSNB)38 onto 60 nm (nominal diameter) AuNPs (Ted Pella, Inc.), followed by the adsorption of a layer of either LF or PA. DSNB has an intrinsically strong Raman-active symmetric nitro stretch [S(NO2)] at 1336 cm-1, which we used to quantify the amount of captured antigen and construct dose-response plots. The plate was dried in air at room temperature for ~1 h before readout with a Raman Microscope equipped with a HeNe laser (633 nm) at a power of 8 mW at the sample and a 5× objective (~1 μm spot size). Spectra were collected at 16 separate spots per well, with 2 exposures at a 1 s integration time

6 ACS Paragon Plus Environment

Page 6 of 22

Page 7 of 22 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

per exposure. The relative standard deviations (RSDs) at the limit of detection for both PA and LF are 14%, which are comparable to typical RSDs observed in ELISA.46 All work with the two antigens were carried out at room temperature (25 o C) in a class II biosafety cabinet located in a biological safety level 2 (BSL2) laboratory, and procedures established by the Institutional Biosafety Committee of the University of Utah were rigorously followed. Figure 2 presents a set of SERS spectra for LF (Figure 2A) and PA (Figure 2B) spiked over a range of concentrations (0-10 ng/mL) into HEPES-B. These results will serve as a comparator for the measurements using pooled human serum. An examination of the two sets of spectra leads to three important points. First, all but one of the spectral features evident throughout the two sets of data are associated with the adlayer formed by the RRM DSNB on the AuNP core of the ERLs: ν(C=C) at 1558 cm-1; ν(C-C) at 1459 cm-1; νs(NO2) at 1336 cm-1; ν(C-O) at 1154 cm-1; and ν(N-C-O) at 1079 cm-1, which overlaps with aromatic ring modes at ~1060 cm-1.38 We suspect that the weak band at ~900 cm -1 is due to a small and varied amount of carbon contamination47, 48 from the electron beam deposition process. Second, there are no observable bands for the layer of antibodies that are part of the ERL architecture. This is attributed to an insufficient level of plasmonically generated signal enhancement to realize detection. Third, the spectra for the blank specimens points to a small, but detectable, level of nonspecific ERL adsorption. The responses measured for serum blanks in the absence of the BSA blocking agent were as much as 20× larger. An assessment of the detectability of the two antigens in HEPES-B can be developed from these experiments by generating dose-response plots from the average intensities of νs(NO2) with respect to their spike-in concentrations. Figure 3 presents these results, along with linear least

7 ACS Paragon Plus Environment

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

squares fits to the data. The LOD is defined as the average signal of the blank plus three times its standard deviation (i.e., 3 + Nblank, where  is the standard deviation of the blank measurement with an average signal of Nblank). Using this definition, the estimated LODs for PA and LF in HEPES-B are 160 ng/mL and 3,200 ng/mL, respectively. We note that these estimates, which are reported in Table 1, are 100-10,000× higher than those reflective of our past experiences in using this immunoassay format.49-51 The next set of experiments ran assays for LF and PA spiked into pooled human serum, following the same procedures used with HEPES-B as the sample matrix. Figure 4 presents a set of SERS spectra for serum blanks, and for LF (Figure 4A) and PA (Figure 4B) spiked into human serum at levels up to 10 ng/mL. Like the spectra obtained using HEPES-B as the sample matrix, the intensity of the S(NO2) of the RRM coating on the ERLs shows clear increases as the spike-in concentrations of both antigens in serum increase. There is, however, a significant difference in the strength of the signal responses observed for the two types of sample matrices. The responses for the samples at the same spike-in concentrations in serum are strikingly stronger than those with HEPES-B as the sample matrix. For example, the strength of the νs(NO2) for LF spiked into serum at 0.125 ng/mL (127 cts/s) is close to that for LF at 10,000 ng/mL in HEPES-B (180 cts/s), a difference in antigen concentration of 80,000. Similarly, PA spiked in serum has a response of 500 cts/s for 0.1 ng/mL, compared to 781 cts/s at 1000 ng/mL when spiked into HEPES-B, which represents a concentration difference of 10,000. This difference in detectability in sample matrix is brought out more fully by examining the dose-response curves for LF and PA in Figures 5A and 5B, respectively, which were constructed by plotting νs(NO2) intensity vs. standard concentration. Each data point represents the average of sixteen readings across the sample surface from three replicate sets of samples. In the case of

8 ACS Paragon Plus Environment

Page 8 of 22

Page 9 of 22 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

LF, a linear least squares fit to the lower portion of the dose-response plot (shown in the inset of Figure 5A) yields an estimated LOD of 125 pg/mL (~1.4 pM). As summarized in Table 1, this represents an improvement in the LOD for the measurement of LF in serum with respect to that in HEPES-B of more than 25,000×. The same type of fitting to the data presented in the inset of Figure 5B for PA spiked into serum translates to a projected LOD of 2.3 pg/mL. This approaches a 70,000× improvement in LOD over that in HEPES-B. The results of this study show that there is a significant improvement in the ability to detect PA and LF spiked in serum samples over those spiked in a HEPES-B matrix. This points to the need to reconsider the practice of using physiological buffers like HEPES-B and potentially other buffers like PBS (see below) as diluents in the mitigation of the negative impact of nonspecific adsorption in the analysis of serum specimens. It also suggests the application of these buffers in the analysis of suspicious powders should be reexamined. There is evidence in the immunometric assay literature on these markers that support our observations. The results of these studies can be grouped into experiments using serum (undiluted) and those using buffered (PBS) matrix. Those experiments on the direct detection in serum reported LODs down to femtomolar levels.52, 53 In contrast, the investigations using PBS to serially dilute serum have only reached LODs in the micro- to nano-molar range.54, 55 56 19, 57, 58 These results, while likely to have small contributions from differences in the binding affinities of the antibodies used in the capture and/or labeling steps, provide further confirmation for the observations herein and show that the differences extend beyond HEPES-B. The potential impact of sample matrix on detection led us to look into reports on the structural stability of LF and PA in different matrices. LF is a zinc metalloenzyme,12, 59 which has a structure unusually sensitive to changes in pH, ionic strength, and buffer composition36, 60

9 ACS Paragon Plus Environment

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

as shown by studies using catalytic activity34, 36, 61 and differential scanning calorimetry.36 Characterizations by circular dichroism60 and differential scanning calorimetry36 have also found that both PA and LF are thermolabile, but can be stabilized to differing degrees by salts, sugars, and other additives often used to prolong the storage of proteins and other large biomolecules.35 These reports suggest that the inferior levels of detection in buffer arise from the structural denaturation of LF and PA, which, in turn, severely compromises the effectiveness of the capture and/or labeling steps in a sandwich immunoassay. In summary, we have found that the detectability of LF and PA is markedly better in undiluted serum matrices and postulate that common proteins such as human serum albumin35 and globulins may have a stabilizing effect on proteinaceous antigens with differential thermal unfolding capacities close to room temperature, which may or may not be coupled to pHdependent structure in close proximity to the epitopes on the antigen.62 Possibilities include markers of C. jejuni, C. coli63, and Y. pestis.64 We suspect that the findings herein may have an impact on other assay formats, e.g., Surface Plasmon Resonance (SPR)65, electrochemical detection66, and quartz crystal microbalance (QCM),67 with those designed in a sandwich format more likely being effected than those in a label-less format. We are currently designing experiments to begin to identify these serum constituents and their role to test this hypothesis. We are also working through the process of transferring the platform in order to carry out tests on live agents in a Biosafety Level 3 laboratory.

10 ACS Paragon Plus Environment

Page 10 of 22

Page 11 of 22 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

Associated Content Supporting Information The Supporting Information is available free of charge on the ACS Publications website: Chemicals and reagents, methods, ligand binding curve analysis, and bar graphs for SERS responses in buffer and serum.

Author Information Corresponding Author E-mail: [email protected] Tel.: 801-587-8325 Acknowledgements This research was supported by the Partnerships for Biodefense program of the National Institutes of Health (NIH) and National Institute of Allergy and Infectious Diseases (NIAID) under grant number R01AI111495. The authors would like to thank Dr. Michael Granger for fabrication of the evaporation mask for the 96-well plates. Drs. Angelo Madonna, Brian Bennett, Jeffery Hogan, Shawn Slater and Mark Gunnell (Dugway Proving Ground) are acknowledged for their insightful discussions. Table 1. Comparison of estimated LODs for LF and PA in HEPES-B and pooled human serum.

antigen

limit of detection HEPES-B

LODHEPES-B/LODserum

human serum

lethal factor

3200 ng/mL

36 µM

125 pg/mL

1.4 pM

25,800

protective antigen

160 ng/mL

2 nM

2.3 pg/mL

28 fM

69,600

MW of LF =90 kDa12 ; MW of PA = 83 kDa12

11 ACS Paragon Plus Environment

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 22

Figure 1. Top: Schematic representation of the SERS Immunoassay for LF and PA. Bottom: Gold-coated well bottoms of a 96-well plate.

12 ACS Paragon Plus Environment

Page 13 of 22 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. SERS spectra for (A) LF and (B) PA spiked into HEPES-B at concentrations of (a) 10,000 ng/mL, (b) 1,000 ng/mL, (c) 100 ng/mL, (d) 10 ng/mL, (e) 1 ng/mL, (f) 0.1 ng/mL, (g) 0.01 ng/mL, and (h) 0 ng/mL.

13 ACS Paragon Plus Environment

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

Figure 3. Linear regions of the calibration plots for lethal factor and protective antigen in HEPES-B. Equations for linear least squares fit are y=0.0102x+19.3, R2 = 0.948 for LF, and y=0.538x + 35.9, R2 = 0.974 for PA.

14 ACS Paragon Plus Environment

Page 14 of 22

Page 15 of 22 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 4. SERS spectra for (A) LF spiked into pooled human serum at concentrations of (a) 10 ng/mL, (b) 1 ng/mL, (c) 0.5 ng/mL, (d) 0.25 ng/mL, (e) 0.125 ng/mL and (f) 0 ng/mL; and (B) PA spiked into pooled human serum at concentrations of (a) 10 ng/mL, (b) 1 ng/mL, (c) 0.1 ng/mL, (d) 0.01 ng/mL, and (e) 0 ng/mL.

15 ACS Paragon Plus Environment

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

Figure 5. (A) Calibration plot for lethal factor in human serum (linear region was used to estimate LOD). Equation of best fit line is y=543x+75.6, R2 = 0.997. (B) Calibration plot for protective antigen in human serum. A single-site ligand binding model was used to calculate a best fit of the data since a linear region is not observed within the concentration range studied (more details regarding data fitting are provided in the supporting information).

16 ACS Paragon Plus Environment

Page 16 of 22

Page 17 of 22 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

For TOC only

17 ACS Paragon Plus Environment

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

References 1. World Health Organization, Anthrax in Humans and Animals. 2008; p 219. 2. Centers for Disease Control and Prevention, Anthrax: Collecting, Preparing, and Shipping Serum Samples to CDC for Serology Testing. https://www.cdc.gov/anthrax/specificgroups/lab-professionals/cdcspecimens.html. 3. Silvestri, E. E.; Perkins, S. D.; Feldhake, D.; Nichols, T.; III, F. W. S., Recent Literature Review of Soil Processing Methods for Recovery of Bacillus anthracis Spores. Ann. Microbiol. 2015, 65, 1215-1226. 4. Waritani, T.; Chang, J.; McKinney, B.; Terato, K., An ELISA Protocol to Improve the Accuracy and Reliability of Serological Antibody Assays. MethodsX 2017, 4, 153-165. 5. Thakur, K.; Sharma, S.; Prabhakar, S.; Gupta, P.; Anand, A., Revisiting the Dilution Factor as Vital Parameter for Sensitivity of ELISA Assay in CSF and Plasma. Ann. Neurosci. 2015, 22, 37-42. 6. Centers for Disease Control and Prevention, Anthrax. http://www.cdc.gov/anthrax/. 7. Johnston, W. R. Review of Fall 2001 Anthrax Bioattacks; 2005. 8. National Institute of Allergy and Infectious Diseases, NIAID Emerging Infectious Diseases/Pathogens. https://www.niaid.nih.gov/topics/biodefenserelated/biodefense/pages/cata.aspx. 9. Inglesby, T. V.; O'Toole, T.; Henderson, D. A.; Bartlett, J. G.; Ascher, M. S.; Eitzen, E.; Friedlander, A. M.; Gerberding, J.; Hauer, J.; Hughes, J.; McDade, J.; Osterholm, M. T.; Parker, G.; Perl, T. M.; Russell, P. K.; Tonat, K., Anthrax as a Biological Weapon, 2002. J. Am. Med. Assoc. 2002, 287, 2236-2253. 10. Europen Biological Network, http://www.europeanbiosafetynetwork.eu/. 11. Pastorino, B.; Lamballerie, X. d.; Charrel, R., Biosafety and Biosecurity in European Containment Level 3 Laboratories: Focus on French Recent Progress and Essential Requirements. Frontiers in Public Health 2017, 5, 1-11. 12. Collier, R. J.; Young, J. A. T., Anthrax Toxin. Annu. Rev. Cell Dev. Biol. 2003, 19, 4570. 13. Prince, A. S., The Host Response to Anthrax Lethal Toxin: Unexpected Observations. J. Clin. Invest. 2003, 112, 656-658. 14. Young, J. A. T.; Collier, R. J., Anthrax Toxin: Receptor Binding, Initialization, Pore Formation and Translocation. Annu. Rev. Biochem. 2007, 76, 243-265. 15. Shadomy, S. V.; Traxler, R. M.; Marston, C. K., Infectious Diseases Related to Travel. In CDC Health Information for International Travel 2015, Brunette, G. W., Ed. 2015. 16. Centers for Disease Control and Prevention, Confirming Anthrax Through the Laboratory Response Network. http://www.cdc.gov/anthrax/lab-testing/isitanthrax.html. 17. Edwards, K. A.; Clancy, H. A.; Baeumner, A. J., Bacillus anthracis: Toxicology, Epidemiology and Current Rapid-detection Methods. Anal. Bioanal. Chem. 2006, 384, 73-84. 18. Kim, J.; Gedi, V.; Lee, S.-C.; Moon, J.-Y.; Yoon, M.-Y., Advances in Anthrax Detection: Overview of Bioprobes and Biosensors. Appl. Biochem. Biotechnol. 2015, 176, 957977. 19. Zai, X.; Zhang, J.; Liu, J.; Liu, J.; Li, L.; Yin, Y.; Fu, L.; Xu, J.; Chen, W., Quantitative Determination of Lethal Toxin Proteins in Culture Supernatent of Human Live Anthrax Vaccine Bacillus anthracis A16R. Toxins (Basel) 2016, 8, 1-17. 18 ACS Paragon Plus Environment

Page 18 of 22

Page 19 of 22 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

20. Gates-Hollingsworth, M. A.; Perry, M. R.; Chen, H.; Needham, J.; Houghton, R. L.; Raychaudhuri, S.; Hubbard, M. A.; Kozel, T. R., Immunoassay for Capsular Antigen of Bacillus anthracis Enables Rapid Diagnosis in a Rabbit Model of Inhalational Anthrax. PLoS One 2015, 10, e0126304. 21. Wang, D.-B.; Bi, L.-J.; Zhang, Z.-P.; Chen, Y.-Y.; Yang, R.-F.; Wei, H.-P.; Zhou, Y.F.; Zhang, Z.-E., Label-free Detection of B. anthracis Spores Using a Surface Plasmon Resonance Biosensor. Analyst 2009, 134, 738-742. 22. Ryu, J.; Lee, E.; Lee, K.; Jang, J., A Graphene Quantum Dots Based Fluorescent Sensor for Anthrax Biomarker Detection and its Size Dependence. J. Mat. Chem. B, 2015, 3, 4865-4870. 23. Sengupta, A.; Shende, C.; Farquharson, S.; Inscore, F., Detection of Bacillus anthracis Spores Using Peptide Functionalized SERS-active Substrates. Int. J. Spectrosc. 2012, 2012, 1-6. 24. Farquharson, S.; Shende, C.; Smith, W.; Huang, H.; Inscore, F.; Sengupta, A.; Sperry, J.; Sickler, T.; Prugh, A.; Guicheteau, J., Selective Detection of 1000 B. anthracis Spores Within 15 Minutes Using a Peptide Functionalized SERS Assay. Analyst 2014, 139, 6366-6370. 25. Zhang, X.; Young, M. A.; Lyandres, O.; Duyne, R. P. V., Rapid Detection of an Anthrax Biomarker by Surface-Enhanced Raman Spectroscopy. J. Am. Chem. Soc. 2005, 127, 4484-4489. 26. Zhang, X.; Yonzon, C. R.; Duyne, R. P. V., An Electrochemical Surface-Enhanced Raman Spectroscopy Approach to Anthrax Detection. Proceedings of SPIE 2003, 5221, 82-91. 27. Gao, R.; Ko, J.; Cha, K.; Jeon, J. H.; Rhie, G.-e.; Choi, J.; deMello, A. J.; Choo, J., Fast and Sensitive Detection of an Anthrax Biomarker Using SERS-based Solenoid Microfluidic Sensor. Biosensors Bioelectron. 2015, 72, 230-236. 28. Ko, J.; Cha, K.; Jeon, J. H.; Rhie, G.-e.; Choi, J.; Choo, J., Surface-enhanced Raman Scattering-based Competitive Immunoassay of Poly-D-Glutamic Acid for Highly Sensitive Detection of Anthrax Biomarker. Bull. Korean Chem. Soc. 2015, 36, 822-826. 29. Li, P.; Ang, A. N.; Feng, H.; Li, S. F. Y., Rapid Detection of an Anthrax Biomarker Based on the Recovered Fluorescence of Carbon Dot-Cu(II) Systems. J. Mat. Chem. C, 2017, 5, 6962-6972. 30. Bell, S. E. J.; Mackle, J. N.; Sirimuthu, N. M. S., Quantitative Surface-enhanced Raman Spectroscopy of Dipicolinic Acid - Towards Rapid Anthrax Endospore Detection. Analyst 2005, 130, 545-549. 31. Granger, J. H.; Schlotter, N. E.; Crawford, A. C.; Porter, M. D., Prospects for Point-ofCare Pathogen Diagnostics Using Surface-enhanced Raman Scattering (SERS). Chem. Soc. Rev. 2016, 45, 3865-3882. 32. Granger, J. H.; Granger, M. C.; Firpo, M. A.; Mulvihill, S. J.; Porter, M. D., Toward Development of a Surface-enhanced Raman Scattering (SERS)-based Cancer Diagnostic Immunoassay Panel. Analyst 2013, 138, 410-416. 33. Porter, M. D.; Lipert, R. J.; Siperko, L. M.; Wang, G.; Narayanan, R., SERS as a Bioassay Platform: Fundamentals, Design, and Applications. Chem. Soc. Rev. 2008, 37, 10011011. 34. Fish, D. C.; Mahlandt, B. G.; Dobbs, J. P.; Lincoln, R. E., Purification and Properties of In Vitro-produced Anthrax Toxin Components. J. Bacteriol. 1968, 95, 907-918. 35. Finn, T. E.; Nunez, A. C.; Sunde, M.; Easterbrook-Smith, S. B., Serum Albumin Prevents Protein Aggregation and Amyloid Formation and Retains Chaperone-like Activity in the Presence of Physiological Ligands. J. Biol. Chem. 2012, 287, 21530-21540.

19 ACS Paragon Plus Environment

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

36. Gupta, P.; Singh, S.; Tiwari, A.; Bhat, R.; Bhatnagar, R., Effect of pH on Stability of Anthrax Lethal Factor: Correlation Between Denaturation and Activity. Biochem. Biophys. Res. Commun. 2001, 284, 568-573. 37. Tonello, F.; Ascenzi, P.; Montecucco, C., The Metalloproteolytic Activity of the Anthrax Lethal Factor Is Substrate-inhibited. J. Biol. Chem. 2003, 10, 40075-40078. 38. Grubisha, D. S.; Lipert, R. J.; Park, H.-Y.; Driskell, J.; Porter, M. D., Femtomolar Detection of Prostate-Specific Antigen: An Immunoassay Based on Surface-Enhanced Raman Scattering and Immunogold Labels. Anal. Chem. 2003, 75 (21), 5936-5943. 39. Zhang, H.; Harpster, M. H.; Park, H. J.; Johnson, P. A., Surface-enhanced Raman Scattering Detection of DNA Derived from the West Nile Virus Genome Using magnetic capture of Raman-active gold nanoparticles. Anal. Chem. 2011, 83, 254-260. 40. Driskell, J. D.; Kwarta, K. M.; Lipert, R. J.; Porter, M. D.; Neill, J. D.; Ridpath, J. F., Low-Level Detection of Viral Pathogens by a Surface-Enhanced Raman Scattering Based Immunoassay. Anal. Chem. 2005, 77 (19), 6147-6154. 41. Crawford, A. C.; Young, C. C.; Porter, M. D., Gold Nanoparticle Labels and Heterogeneous Immunoassays: The Case for the Inverted Substrate. Anal. Chem. 2018, 90, 86658672. 42. Yakubovsky, D. I.; Arsenin, A. V.; Stebunov, Y. V.; Fedyanin, D. Y.; Volkov, V. S., Optical Constants and Structural Properties of Thin Gold Films. Opt. Express 2017, 25, 2557425587. 43. Pradhan, S. K., Observation of Optical Properties of Gold Thin Films Using Spectroscopic Ellipsometry. J. Mat. Sci. Eng. 2016, 6, 1-5. 44. Driskell, J.; Lipert, R.; Porter, M., Labeled gold nanoparticles immobilized at smooth metallic substrates: systematic investigation of surface plasmon resonance and surface-enhanced Raman scattering. J. Phys. Chem. B 2006, 110 (35), 17444-17451. 45. Driskell, J. D.; Larrick, C.; Trunell, C., Effect of Hydration on Plasmonic Coupling of Bioconjugated Gold Nanoparticles Immobilized on a Gold Film Probed by Surface-Enhanced Raman Spectroscopy. Langmuir 2014, 30, 6309-6313. 46. Hayashi, Y.; Matsuda, R.; Maitani, T.; Imai, K.; Nishimura, W.; Ito, K.; Maeda, M., Precision, Limit of Detection and Range of Quantitation in Competitive ELISA. Anal. Chem. 2004, 76, 1295-1301. 47. Wang, Y.; Alsmeyer, D. C.; McCreery, R. L., Raman Spectroscopy of Carbon Materials: Structural Basis of Observed Spectra. Chem. Mater. 1990, 2, 557-563. 48. Casari, C. S.; Bassi, A. L.; Baserga, A.; Ravagnan, L.; Piseri, P.; Lenardi, C.; Tommasini, M.; Milani, A.; Fazzi, D.; Bottani, C. E.; Milani, P., Low-frequency Modes in the Raman Spectrum of sp-sp2 Nanostructured Carbon. Phys. Rev. B 2008, 77, 1-7. 49. Ni, J.; Lipert, R. J.; Dawson, G. B.; Porter, M. D., Immunoassay Readout Method Using Extrinsic Raman Labels Adsorbed on Immunogold Colloids. Anal. Chem. 1999, 71 (21), 49034908. 50. Park, H. Y.; Driskell, J. D.; Kwarta, K. M.; Lipert, R. J.; Porter, M. D.; Schoen, C.; Neill, J. D.; Ridpath, J. F., Ultrasensitive Immunoassays Based on Surface-enhanced Raman Scattering by Immunogold Labels. In Surface-Enhanced Raman Scattering: Physics and Applications, Kneipp, K.; Moskovits, M.; Kneipp, H., Eds. Springer-Verlag: Berlin, 2006; Vol. 103, pp 427-446.

20 ACS Paragon Plus Environment

Page 20 of 22

Page 21 of 22 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

51. Yakes, B., J.; Lipert, R., J.; Bannantine, J., P.; Porter, M., D., Detection of Mycobacterium avium subsp. paratuberculosis by a Sonicate Immunoassay Based on Surfaceenhanced Raman Scattering. Clin. Vaccine Immunol. 2008, 15, 227-234. 52. Ryu, K.; Haes, A. J.; Park, H.-Y.; Nah, S.; Kim, J.; Chung, H.; Yoon, M.-Y.; Han, S.H., Use of Peptide for Selective and Sensitive Detection of an Anthrax Biomarker via Peptide Recognition and Surface-enhanced Raman Scattering. J. Raman Spectrosc. 2009, 41, 121-124. 53. Dragan, A. I.; Albrecht, M. T.; Pavlovic, R.; Keane-Myers, A. M.; Geddes, C. D., Ultra-fast pg/ml Anthrax Toxin (Protective Antigen) Detection Assay Based on Microwaveaccelerated Metal-enhanced Fluorescence. Anal. Biochem. 2012, 425, 54-61. 54. Quinn, C. P.; Semenova, V. A.; Elie, C. M.; Romero-Steiner, S.; Green, C.; Li, H.; Stamey, K.; Steward-Clark, E.; Schmidt, D. S.; Mothershed, E.; Pruckler, J.; Schwartz, S.; Benson, R. F.; Helsel, L. O.; Holder, P. F.; Johnson, S. E.; Kellum, M.; Messme, T.; Thacker, W. L.; Besser, L.; Plikaytis, B. D.; Thomas H. Taylor, J.; Freeman, A. E.; Wallace, K. J.; Dull, P.; Sejvar, J.; Bruce, E.; Moreno, R.; Schuchat, A.; Lingappa, J. R.; Marano, N.; Martin, S. K.; Walls, J.; Bronsdon, M.; Carlone, G. M.; Bajani-Ari, M.; Ashford, D. A.; Stephens, D. S.; Perkins, B. A., Specific, Sensitive, and Quantitative Enzyme-Linked Immunosorbent Assay for Human Immunoglobulin G Antibodies to Anthrax Toxin Protective Antigen. Emerging Infect. Dis. 2002, 8, 1103-1110. 55. Elliot, J. L.; Mogridge, J.; Collier, R. J., A Quantitative Study of the Interactions of Bacillus anthracis Edema Factor and Lethal Factor with Activated Protective Antigen. Biochemistry 2000, 39, 6706-6713. 56. Mabry, R.; Brasky, K.; Geiger, R.; Ricardo Carrion, J.; Hubbard, G. B.; Leppla, S.; Patterson, J. L.; Georgiou, G.; Iverson, B. L., Detection of Anthrax Toxin in the Serum of Animals Infected with Bacillus anthracis by Using Engineered Immunoassays. Clin. Vaccine Immunol. 2006, 13, 671-677. 57. Rucker, V. C.; Havenstrite, K. L.; Herr, A. E., Antibody Microarrays for Native Toxin Detection. Anal. Biochem. 2005, 339, 262-270. 58. Tang, S.; Moayeri, M.; Chen, Z.; Harma, H.; Zhao, J.; Hu, H.; Purcell, R. H.; Leppla, S. H.; Hewlett, I. K., Detection of Anthrax Toxin by an Ultrasensitive Immunoassay Using Europium Nanoparticles. Clin. Vaccine Immunol. 2009, 16, 408-413. 59. Bagramyan, K.; Kalkum, M., Ultrasensitive Detection of Botulinum Neurotoxins and Anthrax Lethal Factor in Biological Samples by ALISSA. In Microbial Toxins: Methods and Protocols, Methods in Molecular Biology, 2011; Vol. 739, pp 23-26. 60. Radha, C.; Salotra, P.; Bhat, R.; Bhatnagar, R., Thermostabilization of Protective Antigen - The Binding Component of Anthrax Lethal Toxin. J. Biotechnol. 1996, 50, 235-242. 61. Kim, J.; Chai, Y.-G.; Yoon, M.-Y., Implication of pH in the Catalytic Properties of Anthrax Lethal Factor. Biochem. Biophys. Res. Commun. 2004, 313, 217-222. 62. Thai, R.; Moine, G.; Desmadril, M.; Servent, D.; Tarride, J.-L.; Menez, A.; Leonetti, M., Antigen Stability Controls Antigen Presentation. J. Biol. Chem. 2004, 279, 50257-50266. 63. Lior, H., Serotyping of Campylobacter jejuni and C. coli by Slide Agglutination Based on Heat-labile Antigenic Factors. Butzler, J., Ed. CRC Press: 1984; pp 61-76. 64. Sansonetti, P., Host-pathogen Interactions: The Seduction of Molecular Cross Talk. Gut 2002, 50, iii2-iii8. 65. Masson, J.-F., Surface Plasmon Resonance Clincial Biosensors for Medical Diagnostics. ACS Sensors 2017, 2, 16-30.

21 ACS Paragon Plus Environment

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

66. Sharma, R.; Deacon, S. E.; Nowak, D.; George, S. E.; Szymonik, M. P.; Tang, A. A. S.; Tomlinson, D. C.; Davies, A. G.; McPherson, M. J.; Walti, C., Label-free Electrochemical Impedance Biosensor to Detect Human Interleukin-8 in Serum With sub-pg/mL Sensitivity. Biosensors Bioelectron. 2016, 80, 607-613. 67. Phan, H. T. M.; Bartelt-Hunt, S.; Rodenhausen, K. B.; Schubert, M.; Bartz, J. C., Investigation of Bovine Serum Albumin (BSA) Attachment onto Self-Assembled Monolayers (SAMs) Using Combinatorial Quartz Crystal Microbalance with Dissipation (QCM-D) and Spectroscopic Ellipsometry (SE). PLoS One 2015, 10, e0141282.

22 ACS Paragon Plus Environment

Page 22 of 22