A smartphone-based rapid telemonitoring system for Ebola and

Publication Date (Web): December 7, 2018. Copyright © 2018 American Chemical Society. Cite this:ACS Sens. XXXX, XXX, XXX-XXX ...
0 downloads 0 Views 665KB Size
Subscriber access provided by YORK UNIV

Article

A smartphone-based rapid telemonitoring system for Ebola and Marburg disease surveillance Mohan Natesan, Sz-Wei Wu, Chieh-I Chen, Stig MR Jensen, Neven Karlovac, Beverly K Dyas, Onur Mudanyali, and Robert G Ulrich ACS Sens., Just Accepted Manuscript • DOI: 10.1021/acssensors.8b00842 • Publication Date (Web): 07 Dec 2018 Downloaded from http://pubs.acs.org on December 9, 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

1 1

A smartphone-based rapid telemonitoring system for Ebola and

2

Marburg disease surveillance

3

Mohan Natesan†*, Sz-Wei Wu‡, Chieh-I Chen‡, Stig M. R. Jensen†, Neven Karlovac‡, Beverly K.

4

Dyas†, Onur Mudanyali‡, and Robert G. Ulrich†

5

†Division

6

Institute of Infectious Diseases, Frederick MD 21702, USA

7

‡Cellmic

of Molecular and Translational Sciences, United States Army Medical Research

LLC, Los Angeles, CA 90024, USA

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 2 of 22

2 8

ABSTRACT:

9

We have developed a digital and multiplexed platform for the rapid detection and

10

telemonitoring of infections caused by Ebola and Marburg filoviruses. The system includes a flow

11

cell assay cartridge that captures specific antibodies with microarrayed recombinant antigens

12

from all six species of filovirus, and a smartphone fluorescent reader for high-performance

13

interpretation of test results. Multiplexed viral proteins, which are expandable to include greater

14

numbers of probes, were incorporated to obtain highest confidence results by cross-correlation,

15

and custom smartphone application was developed for data analysis, interpretation and

16

communication. The smartphone reader utilizes an opto-electro-mechanical hardware attachment

17

that snaps at the back of a Motorola smartphone which provides a user interface to manage the

18

operation, acquire test results and communicate with cloud service. The application controls the

19

hardware attachment to turn on LEDs and digitally record the optically enhanced images. Assay

20

processing time is approximately 20 min for microliter amounts of blood, and test results are

21

digitally processed and displayed within 15 seconds. Furthermore, a secure cloud service was

22

developed for the telemonitoring of test results generated by the smartphone readers in the field.

23

Assay system results were tested with sera from non-human primates that received a live

24

attenuated EBOV vaccine. This integrated system will provide a rapid, reliable, and digital

25

solution to prevent the rapid overwhelming of medical systems and resources during EVD or MVD

26

outbreaks. Further, this disease-monitoring system will be useful in resource limited countries

27

where there is a need for dispersed laboratory analysis of recent or active infections.

28 29

Keywords:

30

Ebola; Point-of-care; Smartphone reader; Disease surveillance; Telemonitoring

31

ACS Paragon Plus Environment

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

3 32

Healthcare decisions t h a t a r e based on rapid, reliable, and digital data are a high priority

33

measure to prevent the inundation of medical systems during episodes of infectious disease

34

outbreaks. The largest Ebola disease outbreak to date occurred in West Africa (2013-2016) and

35

resulted in more than 11,200 deaths, with approximately 28,000 suspected cases in Guinea,

36

Liberia and Sierra Leone.1 This recent and unanticipated Ebola epidemic exposed many

37

weaknesses in existing healthcare monitoring and management systems.2 The regional and

38

international responses were compromised by inaccessible, incomplete and underutilized health

39

information.3 Accessible methods to rapidly monitor biological fluids for detection of suspected

40

infections will result in actionable data for professional healthcare workers. These methods must

41

not only provide accurate diagnosis by untrained personnel, but also overcome the difficulties

42

encountered in the field, and storage of a high volume of c o n f i d e n t i a l test records. However,

43

there are no truly portable diagnostic devices that can be used by care providers with minimal

44

training, especially those designed for detecting Ebola and Marburg virus emergence within

45

the community.4 Polymerase chain reaction (PCR),5 - 6 and antigen detection tests7 - 9 are

46

performed and interpreted only in the clinic, and thus rely on movement of patients or

47

specimens to operational laboratories. While PCR methodologies demonstrate excellent

48

sensitivity and specificity, data can only be obtained during the first few days of viremia, using

49

dedicated equipment and highly trained personnel. Similarly, antigen detection tests7-9 are only

50

useful during the first few days of infection, and few assays are commercially available. Enzyme

51

linked immunosorbent assays (ELISA) typically measure antibody against only a single antigen

52

from one species of filovirus, usually Ebolavirus10 (EBOV). Further, ELISAs are time consuming,

53

require costly instrumentation that is unwieldly, and trained personnel to perform, thus restricting

54

use in resource limited countries.

55

Smartphones are finding increased use in biosensor applications, especially for optical

56

readers.11 The camera, computing power, and the communication capability to remotely report

57

results make smartphones, an attractive choice for integration12-14 into diagnostic and surveillance

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 4 of 22

4 58

assays. We previously reported15-16 the development of a filovirus protein microarray that

59

used a desktop laser scanner to characterize antibody responses from human and

60

non-human primate survivors of Ebola Viral (EVD) and Marburg Viral Diseases (MVD).

61

Here, we describe a highly portable filovirus disease detection assay that is interfaced with a

62

smartphone reader. The reader system was used to analyze antibody binding to 12 essential

63

antigens from all six species of filoviruses that are arranged in a microarray on a disposable

64

microfluidic chip. Performance of the system prototype was demonstrated with sera from non-

65

human primates that were immunized with a recombinant Vesicular Stomatitis Virus- Ebola Virus

66

(rVSV-EBOV) vaccine containing EBOV glycoprotein.17

67 68

EXPERIMENTAL SECTION

69

Materials and Methods.

70

Recombinant filovirus proteins. Full-length genes for nucleoprotein (NP), and the

71

glycoprotein (GP) mucin-like domain fragments for the filovirus species of Reston (RESTV),

72

Bundibugyo (BDBV), Zaire (EBOV), Sudan (SUDV), and Taï Forest ebolavirus (TAFV), and also

73

Marburg marburgvirus (MARV), were cloned into pENTR™/TEV/D-TOPO® vector (Life

74

Technologies, Grand Island, NY, USA) and sequence verified. All entry vector clones were

75

shuttled into destination bacterial expression vectors via LR reaction (LR Clonase® II, Life

76

Technologies, Carlsbad, CA, USA). Specifically, GP mucin open reading frames (ORF) were

77

shuttled into pDESTHisMBP (Addgene plasmid 11085) containing an N-terminal HisMBP tag,

78

while all NP ORFs were shuttled into pDEST17 (Life Technologies) containing an N-terminal His

79

tag. Expression of recombinant GP mucin protein domains were accomplished with BL21-AI™

80

Escherichia coli (Life Technologies) transformed with the pDEST17 constructs. Transformed

81

bacterial were cultured in LB supplemented with 0.1% glucose and 100 μg/mL, and induced at

82

0.4 OD (600 nm) with 0.8% L-arabinose (SIGMA) for 4 hours at 30°C. The bacteria were pelleted

ACS Paragon Plus Environment

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

5 83

by centrifugation (20,000 g, 10 min, 4°C), and lysed using BPER lysing reagent (Thermo

84

Scientific, Rockford, IL, USA) supplemented with 1 x Halt™ protease and phosphatase inhibitors

85

cocktail, EDTA-free (Thermo Scientific), 20 mg/mL lysozyme, and 75 U/mL DNase I (Thermo

86

Scientific). The proteins were purified by immobilized metal ion affinity chromatography (IMAC)

87

FPLC using 1 mL HisTrap™ FF columns (GE Healthcare Lifesciences, Pittsburg, PA, USA).

88

Binding and washing steps were conducted with 25 mM HEPES, 140 mM NaCl, 25 mM imidazole,

89

pH 7.5, and for protein elution the imidazole concentration was raised to 500 mM. Eluted proteins

90

were examined both by gel-electrophoresis, and by mass spectrometry. Expression of

91

recombinant NP proteins were similarly expressed from transformed BL21-AI™ E. coli bacteria,

92

but due to poor IMAC purification capability of these recombinant proteins, expressions conditions

93

were optimized to form inclusion bodies. Induction was done at 0.6 OD (600 nm) with 0.2% L-

94

arabinose (SIGMA-Aldrich, St. Louis, MO, USA) overnight at 18°C. Pelleting and lysis of the

95

bacteria were performed as described above. Inclusion bodies were washed 2x in 140 mM NaCl,

96

20 mM Tris-HCl pH 7.5 to remove soluble impurities, at which point >90% of remaining protein

97

was recombinant NP proteins. The proteins were then solubilized in 25 mM HEPES, 140 mM

98

NaCl, 3 M urea, and refolded on the microarray surface. The purified proteins were analyzed by

99

Agilent Bioanalyzer system (Agilent Technologies, Santa Clara, CA, USA). All purified proteins

100

were stored at -20°C in respective elution buffers, with glycerol added to a final concentration of

101

25%. For quality control purposes and to validate our assay design, printed microarrays were

102

probed with a panel of purified antibodies directed towards filovirus proteins. The panel included

103

anti-EBOV GP, anti-Sudan GP, anti-RESTV GP antibodies (IBT Biosciences, Gaithersburg, MD,

104

USA). The bound antibodies were detected by Alexa Fluor 647-conjugated goat anti-mouse IgG,

105

and goat anti-rabbit IgG antibodies (Life Technologies). Arrays were scanned with Genepix laser

106

scanner (Molecular Devices, San Jose, CA, USA) and analyzed using Genepix Pro 7 software as

107

described previously.16

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 6 of 22

6 108 109

Prototype cartridge. A prototype cartridge (Figure 1) was assembled with the following

110

components. The base substrate was made up of cyclic olefin copolymer (COC) with the following

111

dimensions 25 x 25 x 1 mm. A polyester backed 0.22 μm nitrocellulose membrane (15 x 15 mm)

112

was attached (Bio-Rad, Hercules, CA, USA) to the COC substrate (Fig 1a) using 3MTM polyester

113

double sided diagnostic microfluidic medical tape, coated with pressure sensitive acrylate

114

adhesive (3M, St. Paul, MN, USA). The recombinant proteins were diluted in PBS at 100 μg/mL

115

concentration and printed on the nitrocellulose membrane in a 4 x 8 array format by continuous

116

flow microprinting (CFM, Carterra, Salt Lake City, UT, USA). The protein samples were delivered

117

to the surface by the printer using 48 micro-channels, which allowed the samples to cycle across

118

the nitrocellulose surfaces bi-directionally for 5 min. The spot size of the printed proteins were

119

400 µm each with a pitch of 500 µm. An adhesive-backed acrylic cover plate (Figure 1b) with a

120

dimension of 7 x 4 x 1 mm (Wainamics, Inc., Pleasanton, CA, USA) was pressed over the printed

121

array to form a flow cell with a volume of 100 μL (Figure 1b). The flow cell had one inlet and three

122

outlets to facilitate uniform flow across the membrane.

123

Smartphone reader. The smartphone reader consisted of three main components: a Motorola

124

Nexus 6 phone with an opto-electro-mechanical hardware attachment that snaps at the back of

125

the smartphone, a compartment to house the prototype cartridge, and a custom application (App)

126

to run the assay. The hardware attachment was fabricated by 3-D printing, using fused deposition

127

modeling and stereolithography technologies (Stratasys Ltd., Eden Prairie, MN, USA). The 3D

128

printed attachment also had a custom-designed tray to hold the prototype cartridge (Figure 2).

129

The Android App provided the user interface to manage the operation, acquire test results, and

130

communicate with cloud service. It also controlled the hardware attachment via flash pulses to

131

turn on specific excitation LEDs and digitally record optically enhanced assay images through the

ACS Paragon Plus Environment

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

7 132

emission filter. The smartphone, which was concealed within the hardware attachment, served

133

as the user interface, digital image acquisition device, and system processer.

134

Detection of Ebola and Marburg specific antibodies. The printed microarray chips were

135

blocked for 1 hour in blocking buffer (Grace Bio-Labs, Bend, OR, USA), probed with serum diluted

136

at 1:150 in phosphate buffered saline (PBS), and washed three times. Bound antibody was

137

detected by incubation (1 hour, RT) with Alexa Fluor 647 or 488-conjugated Goat anti-human IgG

138

(Invitrogen, Carlsbad, CA, USA) diluted at 1 μg/mL. The microarrays were rinsed with purified

139

water, dried and scanned. A similar process was applied for the prototype cartridge-based flow

140

cell assay. The reagents were injected by pipetting a volume of 100 μL through the inlet using

141

PIPETMAN P200 (Gilson Inc., Middleton, WI, USA) and incubated for 5 min (22°C). The waste

142

was collected at the outlet with blotting paper.

143 144

RESUTLS AND DISCUSSION

145

We previously showed that human antibody responses to GP protein were predominantly directed

146

towards the polypeptide region of the highly glycosylated mucin-like domain15. Hence, in this study

147

only GP mucin proteins were produced for the microarray. The recombinant proteins were

148

physically characterized for molecular size and purity by using a protein chip analyzer (Agilent

149

Technologies). The functional activity of the proteins was confirmed by probing with a panel of

150

purified antibodies directed towards filovirus proteins. The EBOV GP mucin protein was

151

specifically recognized by the rabbit anti-EBOV GP antibodies (Supporting Information Figure S-

152

1a). Similarly, RESTV GP mucin was also detected specifically by mouse anti-RESTV GP

153

antibodies (Figure S-1b). Minor cross-reactivity between MARV NP and BDBV NP was observed

154

for microarrays probed with rabbit anti-SUDV NP (Figure S-1c). Combined, binding data from

155

these control antibodies indicate that the filovirus microarrays were functional and that the protein

156

probes reacted specifically with the correct antibodies.

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 8 of 22

8 157

Alexa Fluor 350, 430, 488, 647, and Quantum Dot 525 fluorophores were evaluated with the

158

smartphone camera to obtain an optimal image that was capable of differentiating between

159

positive and negative samples. The average Stokes shifts of these fluorophores ranged from 17-

160

100 nm. Figure S-2 shows the excitation/emission spectra of these fluorophores. Proteins

161

conjugated to the fluorescent tags were printed in a microarray format, and the fluorescent images

162

were scanned by a high-resolution laser scanner (Genepix Scanner 4400a) and the smartphone

163

camera (Figure S-3). Although, Alexa Flour 488 had smaller average Stokes shift (30 nm)

164

compared to other fluorophores tested, it produced the strongest signal with the smartphone

165

camera, and was selected for incorporation into the assay.

166

The proteins were printed on a 15 x 15 mm polyester backed nitrocellulose membrane fixed on

167

a plastic (COC) support. Because proteins bind irreversibly to nitrocellulose, no additional

168

chemistry was needed for attachment. An adhesive backed acrylic cover plate was carefully

169

pressed over the COC slide to seal the flow cell. The cover plate had one inlet and three outlets

170

for even flow of reagents across the flow cell and membrane. We examined a cartridge prepared

171

with a microarray containing human IgG, rabbit IgG and mouse IgGs. Anti-human IgG-Alexa Fluor

172

488 conjugate was injected, and the fluorescent intensity of the spots were measured by Genepix

173

laser scanner. No fluid leaks were found during the assay. Little or no binding was observed with

174

control IgGs (rabbit and mouse), whereas high binding noted with human IgG (Figure S-4).

175

Keeping the cover plate in place reduced assay signals compared to signals produced without

176

the cover (Figure S-4c). This reduction may be attributed to the change in refractive index created

177

by air trapped between the cover and the microarray. An index matching solution may be injected

178

at the last step to overcome this problem. The results demonstrated that immunoassays could be

179

performed successfully in the prototype cartridge.

180

An opto-electro-mechanical hardware attachment (Figure 2) that snaps on the back of a

181

Motorola Nexus smartphone device was fabricated by 3D printing (Dimensions Elite (FDM) and

182

Object30 Pro (PolyJet), Stratasys Ltd.) with FDM (Fused Deposition Modeling), and PolyJet

ACS Paragon Plus Environment

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

9 183

technologies. The 3D model of the device was constructed using computer aided design (CAD).

184

The mechanical body isolated the imaging environment and blocked the ambient light entering

185

the imaging area, which was important to obtain reproducible results (Figure 2a). A special tray

186

held the prototype cartridge for imaging by the smartphone camera. Two LEDs (LED Engin, San

187

Jose, CA, USA) on each side of the microarray served as fluorescent dye excitation sources

188

(Figure 2b). The Android App controlled all the steps involved in the operation of the reader device

189

including the acquisition and processing of the image (Figure S-5). The algorithm in the App

190

enhanced the captured images, based on the calibration and cut-off values, and determined if

191

spots present were positive or negative. The results were displayed within 15 seconds after the

192

image was taken by the smartphone. The use of the dedicated smartphone camera was a critical

193

part of the reader development, but this also limited the ability to manipulate captured images.

194

The CMOS imaging sensors and firmware on smartphones were typically designed and optimized

195

for digital photography purposes. Hence, these devices presented certain limitations for direct use

196

with the microfluidic assay. The digital images collected by the sensors were automatically post-

197

processed and compressed by the phone operating system to provide the best-looking photos.

198

Furthermore, the camera settings had limited pre-set options that were designed to simplify use.

199

Therefore, total control over the imaging parameters and supporting firmware on smartphone

200

electronic products was not possible. The raw images (i.e., unprocessed data collected by the

201

imaging sensor) could not be acquired on these devices, limiting the ability to control the camera

202

settings (e.g., ISO, aperture and length of exposure). Similarly, the precision and tolerances were

203

low in the 3-D fabrication method used for the opto-electro-mechanical attachment, but were,

204

however, selected for this proof-of-concept study due to ease of production and cost. Injection

205

molding, which requires significant cost and time, will be required for large-scale production.

206

A Health Insurance Portability and Accountability Act (HIPAA)-compliant cloud service for

207

telemonitoring the test results was also developed. The cloud service was accessed through the

208

SSL (Secure Sockets Layer)-protected web page. The primary function of this service was to

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 10 of 22

10 209

report the test results in a secure manner using the smartphone. To this end, we designed the

210

front-end interface with a third-party HIPAA-compliant database provided by TrueVault, a service

211

provider for small businesses. From these readers, each user could register and securely access

212

their own cloud service test results that were stored in the TrueVault database. Users had access

213

to the database through a secure cloud website, which displayed the test results as a table and

214

as pins on Google map. The cloud site also enabled study directors and select administrators to

215

upload calibration files to the specific end-user groups in the field.

216

Protein microarrays were printed on the cartridge at various X-Y coordinates to align with the

217

field of view of the smartphone reader. The X-Y coordinates of 7.3 x 13.75 mm on the

218

nitrocellulose membrane provided the best results. Figure 3 shows the images of a sample array

219

obtained using with reference Genepix laser scanner as well as smartphone reader. The array

220

(Rows A and B) contained serial dilutions of goat anti-mouse IgG-Alexa Fluor 488 printed in

221

duplicates for measuring linear response. Rows C and D were printed with a fixed concentration

222

of goat anti-mouse IgG-Alexa Fluor 488 and used for illumination correction and calibration. The

223

App utilized Row D elements for the correction of illumination non-uniformity over the chip.

224

Although the laser excitation provided a stronger emission signal than the LED source in the

225

smartphone reader (Figure 3), there was ~95% correlation between the results.

226

To test the performance of the integrated assay, we evaluated antibodies from non-human

227

primates that were immunized with rVSV-EBOV, a vaccine that protects against EBOV and has

228

undergone human clinical trials.17-18 Sera from Cynomolgus macaque (Macaca fascicularis) that

229

received intramuscular injection of rVSV-EBOV were collected before and 36 days after

230

immunization. Filoviral and control proteins were printed in a 4 x 8 microarray format. Rows A, B,

231

and C contained the filoviral proteins and Row D the calibration spots (Figure 4). The microarray

232

included GP mucin and NP proteins from all six species of filovirus family, human IgG for positive

233

control, buffer control, and goat anti-mouse IgG-Alexa Flour 488 for smartphone reader calibration

ACS Paragon Plus Environment

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

11 234

(spots D1, D3, D6 and D8). The assay was completed in 20 min, and the fluorescent intensities

235

of the spots were measured with the smartphone reader device. We analyzed six different

236

samples that demonstrated quantitative comparisons for vaccinated and pre-vaccinated NHP

237

sera (spots B3 and B4), as shown in Figure 4a. The relative signal values were calculated based

238

on the local background signal values. Spots B3 and B4 in Figure 4a were printed with EBOV-GP

239

mucin domain proteins. The two NHP pre-vaccinated sera showed no binding, whereas the

240

vaccinated sera samples produced strong fluorescent signals that indicated recognition of the

241

EBOV GP mucin proteins. The cross-sectional intensity profiles of B3 and B4 spots are shown in

242

Figure 4b. The signal intensities of pre-vaccinated samples were significantly lower than the

243

vaccinated samples. Thus, the integrated smartphone reader system can clearly distinguish

244

between sera that contains anti-EBOV antibodies and non-vaccinated controls. The microarray

245

contained GP and NP proteins from all six species for broad disease coverage, and high probe

246

specificity was demonstrated with anti-EBOV sera. For this proof-of-concept study, we used sera

247

from NHP vaccinated with a single species (EBOV) of filovirus, however we have previously used

248

human sera exposed to three different filoviral species and demonstrated the specificity and

249

cross-reactivity.15 By including multiple probes from all six species of filovirus family the specificity

250

can be more precisely determined than by using the single antigens. Antibody levels for filovirus

251

infections are usually described in arbitrary units that do not reference absolute serum

252

concentrations19-20, making comparisons between methods very difficult. Pathogen-specific IgG

253

concentrations in serum are typically in μM (μg/ml) or higher range and IgM levels, which peak

254

first and are important for disease diagnosis,21 may be a log lower (serum concentration of total

255

IgG 10-16 mg/ml and IgM 1-2 mg/ml). Although IgM was not evaluated here, microarray

256

technology platforms have sensitivities in the low-to-mid picomolar (pM ~ ng/ml)22 range, and thus

257

should be able to detect relevant antibodies in disease sera. While the system described here

258

detects only filovirus specific IgG, future modifications may allow evaluation of IgM responses to

259

open up the possibility of diagnostic applications.

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

12 260

The main objective of this research was to develop a point-of-care platform which will monitor

261

and remotely report data that reveal the emergence and progression of infections caused by

262

Ebola (EVD) and the closely related Marburg (MVD) filoviruses. This goal was accomplished by

263

integrating a microfluidic microarray assay15 with a portable reader that capitalizes on the ease of

264

use and widespread availability of smartphones. The reader capitalizes on widespread availability

265

of smartphones and ease of use. The reader utilizes the computational power of the built-in

266

smartphone and software to guide and monitor proper operation by untrained users especially

267

applicable to resource limited countries. An easy to follow, menu-driven user guide is displayed

268

on the smartphone application, and test results obtained with the device are encrypted and

269

wirelessly transmitted. A HIPAA-compliant secure telemonitoring service provides test results

270

access to care personnel for rapid clinical assessment. The fluorescent-based readout is highly

271

sensitive, and the assay requires only 1 to 2 µl of serum or plasma. The image capturing system

272

employed to maintain image quality, high fluorophore excitation and low background, probe

273

separation limits, and other factors determine the assay multiplexing capacity. The current

274

configuration will accommodate up to 60 individual probes that are printed in 400 µm diameter

275

spots, and the number of probes can be expanded with further refinements of the detection

276

system. Thus, the multiplexing capacity of the assay can be extended to include additional

277

proteins23-25 from viruses that may co-circulate or co-infect along with filoviruses. For example,

278

yellow fever, Lassa, Rift valley fever, dengue and other infections can detrimentally affect the

279

survival of EVD and MVD patients. Further, the infection histories from such studies can be used

280

for developing informed diagnosis and understand epidemiology of EVD and MVD. Thus, the

281

platform described here will address important gaps in resources to monitor and remotely report

282

data that reveal the emergence and progression of infections caused by Ebola and Marburg

283

viruses.

284 285

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

13 286

CONCLUSIONS

287

We developed a simple, low cost, smartphone- based system for the serological detection of

288

Ebola or Marburg diseases in resource limited countries. The antibody detection system

289

developed can monitor and report data r e m o t e l y from infections caused by all six species

290

of virus at the point of use, and will use only a drop of blood, such as that required for a

291

standard glucose analysis by diabetics. The device will also be able to acquire antibody data

292

from both symptomatic and non-symptomatic cases, and thus enable an additional application

293

and unmet need for disease surveillance. The next generation device in development

294

incorporates a cartridge design for plasma separation to enable direct blood sampling, injection

295

molded opto-electro-mechanical component, and a universal scanner that can be combined with

296

any cell phone.

297 298

ASSOCIATED CONTENT

299

Additional figures illustrating the specificity of recombinant proteins, selection of fluorophores,

300

flow cell assay, smartphone App interface, and cloud service for telemonitoring are presented in

301

the supporting information.

302

AUTHOR INFORMATION

303

Corresponding Author

304

*E-mail: [email protected]. Tel +1 301 619 8187

305

Author Contributions

306

Production of recombinant proteins, cartridge and assay were developed by USAMRIID team.

307

Smartphone mechanical attachment, App and set up of secure cloud service were developed by

308

Cellmic team. All authors have given approval to the final version the manuscript.

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 14 of 22

14 309

Notes

310

Research was conducted under an IACUC approved protocol in compliance with the Animal

311

Welfare Act, PHS Policy, and other Federal statutes and regulations relating to animals

312

and experiments involving animals. The facility where this research was conducted is accredited

313

by the Association for Assessment and Accreditation of Laboratory Animal Care, International

314

and adheres to principles stated in the Guide for the Care and Use of Laboratory Animals,

315

National Research Council, 2011.

316

Opinions, interpretations, conclusions, and recommendations are those of the authors and are

317

not necessarily endorsed by the United States Army, the National Institutes of Health, or the

318

United States Government.

319 320

ACKNOWLEDGMENTS

321

This work was funded by National Institutes of Allergy and Infectious Diseases-Small Business

322

Innovation Research Grant HHSN272201600031C/ CELLGEN1608.

323

ACS Paragon Plus Environment

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

15 324

REFERENCES

325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369

1. Team, W. H. O. E. R.; Agua-Agum, J.; Allegranzi, B.; Ariyarajah, A.; Aylward, R.; Blake, I. M.; Barboza, P.; Bausch, D.; Brennan, R. J.; Clement, P.; Coffey, P.; Cori, A.; Donnelly, C. A.; Dorigatti, I.; Drury, P.; Durski, K.; Dye, C.; Eckmanns, T.; Ferguson, N. M.; Fraser, C.; Garcia, E.; Garske, T.; Gasasira, A.; Gurry, C.; Hamblion, E.; Hinsley, W.; Holden, R.; Holmes, D.; Hugonnet, S.; Jaramillo Gutierrez, G.; Jombart, T.; Kelley, E.; Santhana, R.; Mahmoud, N.; Mills, H. L.; Mohamed, Y.; Musa, E.; Naidoo, D.; Nedjati-Gilani, G.; Newton, E.; Norton, I.; Nouvellet, P.; Perkins, D.; Perkins, M.; Riley, S.; Schumacher, D.; Shah, A.; Tang, M.; Varsaneux, O.; Van Kerkhove, M. D., After Ebola in West Africa--Unpredictable Risks, Preventable Epidemics. N Engl J Med 2016, 375 (6), 587-96. 2. Kouadio, K. I.; Clement, P.; Bolongei, J.; Tamba, A.; Gasasira, A. N.; Warsame, A.; Okeibunor, J. C.; Ota, M. O.; Tamba, B.; Gumede, N.; Shaba, K.; Poy, A.; Salla, M.; Mihigo, R.; Nshimirimana, D., Epidemiological and Surveillance Response to Ebola Virus Disease Outbreak in Lofa County, Liberia (March-September, 2014); Lessons Learned. PLoS Curr 2015, 7. doi: 10.1371/currents.outbreaks.9681514e450dc8d19d47e1 724d2553a5. 3. Cancedda, C.; Davis, S. M.; Dierberg, K. L.; Lascher, J.; Kelly, J. D.; Barrie, M. B.; Koroma, A. P.; George, P.; Kamara, A. A.; Marsh, R.; Sumbuya, M. S.; Nutt, C. T.; Scott, K. W.; Thomas, E.; Bollbach, K.; Sesay, A.; Barrie, A.; Barrera, E.; Barron, K.; Welch, J.; Bhadelia, N.; Frankfurter, R. G.; Dahl, O. M.; Das, S.; Rollins, R. E.; Eustis, B.; Schwartz, A.; Pertile, P.; Pavlopoulos, I.; Mayfield, A.; Marsh, R. H.; Dibba, Y.; Kloepper, D.; Hall, A.; Huster, K.; Grady, M.; Spray, K.; Walton, D. A.; Daboh, F.; Nally, C.; James, S.; Warren, G. S.; Chang, J.; Drasher, M.; Lamin, G.; Bangura, S.; Miller, A. C.; Michaelis, A. P.; McBain, R.; Broadhurst, M. J.; Murray, M.; Richardson, E. T.; Philip, T.; Gottlieb, G. L.; Mukherjee, J. S.; Farmer, P. E., Strengthening Health Systems While Responding to a Health Crisis: Lessons Learned by a Nongovernmental Organization During the Ebola Virus Disease Epidemic in Sierra Leone. J Infect Dis 2016, 214 (suppl 3), S153-S163. 4. Broadhurst, M. J.; Brooks, T. J.; Pollock, N. R., Diagnosis of Ebola Virus Disease: Past, Present, and Future. Clin Microbiol Rev 2016, 29 (4), 773-93. 5. Weller, S. A.; Bailey, D.; Matthews, S.; Lumley, S.; Sweed, A.; Ready, D.; Eltringham, G.; Richards, J.; Vipond, R.; Lukaszewski, R.; Payne, P. M.; Aarons, E.; Simpson, A. J.; Hutley, E. J.; Brooks, T., Evaluation of the Biofire FilmArray BioThreat-E Test (v2.5) for Rapid Identification of Ebola Virus Disease in Heat-Treated Blood Samples Obtained in Sierra Leone and the United Kingdom. J Clin Microbiol 2016, 54 (1), 114-9. 6. Semper, A. E.; Broadhurst, M. J.; Richards, J.; Foster, G. M.; Simpson, A. J.; Logue, C. H.; Kelly, J. D.; Miller, A.; Brooks, T. J.; Murray, M.; Pollock, N. R., Performance of the GeneXpert Ebola Assay for Diagnosis of Ebola Virus Disease in Sierra Leone: A Field Evaluation Study. PLoS Med 2016, 13 (3), e1001980. 7. Boisen, M. L.; Cross, R. W.; Hartnett, J. N.; Goba, A.; Momoh, M.; Fullah, M.; Gbakie, M.; Safa, S.; Fonnie, M.; Baimba, F.; Koroma, V. J.; Geisbert, J. B.; McCormick, S.; Nelson, D. K.; Millett, M. M.; Oottamasathien, D.; Jones, A. B.; Pham, H.; Brown, B. L.; Shaffer, J. G.; Schieffelin, J. S.; Kargbo, B.; Gbetuwa, M.; Gevao, S. M.; Wilson, R. B.; Pitts, K. R.; Geisbert, T. W.; Branco, L. M.; Khan, S. H.; Grant, D. S.; Garry, R. F., Field Validation of the ReEBOV Antigen Rapid Test for Point-of-Care Diagnosis of Ebola Virus Infection. J Infect Dis 2016, 214 (suppl 3), S203-S209. 8. Boisen, M. L.; Oottamasathien, D.; Jones, A. B.; Millett, M. M.; Nelson, D. S.; Bornholdt, Z. A.; Fusco, M. L.; Abelson, D. M.; Oda, S.; Hartnett, J. N.; Rowland, M. M.; Heinrich, M. L.; Akdag, M.; Goba, A.; Momoh, M.; Fullah, M.; Baimba, F.; Gbakie, M.; Safa, S.; Fonnie, R.; Kanneh, L.; Cross, R. W.; Geisbert, J. B.; Geisbert, T. W.; Kulakosky, P. C.; Grant, D. S.; Shaffer, J. G.; Schieffelin, J. S.; Wilson, R. B.; Saphire, E. O.; Branco, L. M.; Garry, R. F.; Khan, S. H.; Pitts, K. R.; Viral Hemorrhagic Fever, C., Development of Prototype Filovirus Recombinant Antigen Immunoassays. J Infect Dis 2015, 212 Suppl 2, S359-67. 9. Broadhurst, M. J.; Kelly, J. D.; Miller, A.; Semper, A.; Bailey, D.; Groppelli, E.; Simpson, A.; Brooks, T.; Hula, S.; Nyoni, W.; Sankoh, A. B.; Kanu, S.; Jalloh, A.; Ton, Q.; Sarchet, N.; George, P.; Perkins, M. D.;

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 16 of 22

16 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416

Wonderly, B.; Murray, M.; Pollock, N. R., ReEBOV Antigen Rapid Test kit for point-of-care and laboratorybased testing for Ebola virus disease: a field validation study. Lancet 2015, 386 (9996), 867-74. 10. Vu, H.; Shulenin, S.; Grolla, A.; Audet, J.; He, S.; Kobinger, G.; Unfer, R. C.; Warfield, K. L.; Aman, M. J.; Holtsberg, F. W., Quantitative serology assays for determination of antibody responses to Ebola virus glycoprotein and matrix protein in nonhuman primates and humans. Antiviral Res 2016, 126, 55-61. 11. Quesada-Gonzalez, D.; Merkoci, A., Mobile phone-based biosensing: An emerging "diagnostic and communication" technology. Biosens Bioelectron 2017, 92, 549-562. 12. Lee, S.; Kim, G.; Moon, J., Development of a Smartphone-based reading system for lateral flow immunoassay. J Nanosci Nanotechnol 2014, 14 (11), 8453-7. 13. Vashist, S. K.; Marion Schneider, E.; Zengerle, R.; von Stetten, F.; Luong, J. H., Graphene-based rapid and highly-sensitive immunoassay for C-reactive protein using a smartphone-based colorimetric reader. Biosens Bioelectron 2015, 66, 169-76. 14. Mudanyali, O.; Dimitrov, S.; Sikora, U.; Padmanabhan, S.; Navruz, I.; Ozcan, A., Integrated rapiddiagnostic-test reader platform on a cellphone. Lab Chip 2012, 12 (15), 2678-86. 15. Natesan, M.; Jensen, S. M.; Keasey, S. L.; Kamata, T.; Kuehne, A. I.; Stonier, S. W.; Lutwama, J. J.; Lobel, L.; Dye, J. M.; Ulrich, R. G., Human Survivors of Disease Outbreaks Caused by Ebola or Marburg Virus Exhibit Cross-Reactive and Long-Lived Antibody Responses. Clin Vaccine Immunol 2016, 23 (8), 717-24. 16. Kamata, T.; Natesan, M.; Warfield, K.; Aman, M. J.; Ulrich, R. G., Determination of specific antibody responses to the six species of ebola and Marburg viruses by multiplexed protein microarrays. Clin Vaccine Immunol 2014, 21 (12), 1605-12. 17. Henao-Restrepo, A. M.; Camacho, A.; Longini, I. M.; Watson, C. H.; Edmunds, W. J.; Egger, M.; Carroll, M. W.; Dean, N. E.; Diatta, I.; Doumbia, M.; Draguez, B.; Duraffour, S.; Enwere, G.; Grais, R.; Gunther, S.; Gsell, P. S.; Hossmann, S.; Watle, S. V.; Konde, M. K.; Keita, S.; Kone, S.; Kuisma, E.; Levine, M. M.; Mandal, S.; Mauget, T.; Norheim, G.; Riveros, X.; Soumah, A.; Trelle, S.; Vicari, A. S.; Rottingen, J. A.; Kieny, M. P., Efficacy and effectiveness of an rVSV-vectored vaccine in preventing Ebola virus disease: final results from the Guinea ring vaccination, open-label, cluster-randomised trial (Ebola Ca Suffit!). Lancet 2017, 389 (10068), 505-518. 18. Gsell, P. S.; Camacho, A.; Kucharski, A. J.; Watson, C. H.; Bagayoko, A.; Nadlaou, S. D.; Dean, N. E.; Diallo, A.; Diallo, A.; Honora, D. A.; Doumbia, M.; Enwere, G.; Higgs, E. S.; Mauget, T.; Mory, D.; Riveros, X.; Oumar, F. T.; Fallah, M.; Toure, A.; Vicari, A. S.; Longini, I. M.; Edmunds, W. J.; Henao-Restrepo, A. M.; Kieny, M. P.; Keita, S., Ring vaccination with rVSV-ZEBOV under expanded access in response to an outbreak of Ebola virus disease in Guinea, 2016: an operational and vaccine safety report. Lancet Infect Dis 2017, 17 (12), 1276-1284. 19. Logue, J.; Tuznik, K.; Follmann, D.; Grandits, G.; Marchand, J.; Reilly, C.; Sarro, Y. D. S.; Pettitt, J.; Stavale, E. J.; Fallah, M.; Olinger, G. G.; Bolay, F. K.; Hensley, L. E., Use of the Filovirus Animal Non-Clinical Group (FANG) Ebola virus immuno-assay requires fewer study participants to power a study than the Alpha Diagnostic International assay. J Virol Methods 2018, 255, 84-90. 20. Bower, H.; Glynn, J. R., A systematic review and meta-analysis of seroprevalence surveys of ebolavirus infection. Sci Data 2017, 4, 160133. 21. Macneil, A.; Reed, Z.; Rollin, P. E., Serologic cross-reactivity of human IgM and IgG antibodies to five species of Ebola virus. PLoS Negl Trop Dis 2011, 5 (6), e1175. 22. Ingvarsson, J.; Larsson, A.; Sjoholm, A. G.; Truedsson, L.; Jansson, B.; Borrebaeck, C. A.; Wingren, C., Design of recombinant antibody microarrays for serum protein profiling: targeting of complement proteins. J Proteome Res 2007, 6 (9), 3527-36. 23. Keasey, S.; Pugh, C.; Tikhonov, A.; Chen, G.; Schweitzer, B.; Nalca, A.; Ulrich, R. G., Proteomic basis of the antibody response to monkeypox virus infection examined in cynomolgus macaques and a comparison to human smallpox vaccination. PLoS One 2010, 5 (12), e15547.

ACS Paragon Plus Environment

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

17 417 418 419 420 421 422 423

24. Keasey, S. L.; Pugh, C. L.; Jensen, S. M.; Smith, J. L.; Hontz, R. D.; Durbin, A. P.; Dudley, D. M.; O'Connor, D. H.; Ulrich, R. G., Antibody Responses to Zika Virus Infections in Environments of Flavivirus Endemicity. Clin Vaccine Immunol 2017, 24 (4). doi: 10.1128/CVI.00036-17 25. Smith, J. L.; Pugh, C. L.; Cisney, E. D.; Keasey, S. L.; Guevara, C.; Ampuero, J. S.; Comach, G.; Gomez, D.; Ochoa-Diaz, M.; Hontz, R. D.; Ulrich, R. G., Human Antibody Responses to Emerging Mayaro Virus and Cocirculating Alphavirus Infections Examined by Using Structural Proteins from Nine New and Old World Lineages. mSphere 2018, 3 (2). doi: 10.1128/mSphere.00003-18.

424

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 18 of 22

18 Figure 1 Prototype cartridge for the detection of filoviral antibodies (a) A polyester backed nitrocellulose membrane (0.2 µm) was affixed on a COC slide; (b) an acrylic cover plate with a dimension of 7 x 4 x 1 mm was pressed on tightly to form a flow cell with a volume of 100 µl. The cartridge had one inlet and three outlets for uniform flow across the microarray. The schematics (c) show how the cartridge was assembled for the immunoassay.

(c) Cartridge Schematics STEP 1: Cut COC to 25 x 25 x 1 mm

(a)

STEP 2: Attach 15 x 15 mm NC membrane

(b) Outlet

Nitrocellulose membrane

COC

Flow cell

Inlet

STEP 3: Print filoviral proteins using CFM STEP 4: Affix cover plate to form the flow cell

STEP 5: Run assay Add serum Add anti-human IgGAlexa Fluor 488 Read results

ACS Paragon Plus Environment

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

19 Figure 2 The opto-electro-mechanical device attachment. The 3D printed attachment with the assay chip in panel (a), LEDs for excitation of the fluorophores, the camera aperture with interference filter and external imaging lens in panel (b), the back, front and side view with the smartphone are shown in panel (c), (d) and (e) respectively.

(b) (c)

LED with bandpass filters

(a)

Chip

Camera aperture with interference filter and lens LED with bandpass filters

(d)

(e)

ACS Paragon Plus Environment

ACS Sensors

20 Figure 3 Comparison of microarray fluorescent signals produced by the Genepix laser scanner and by the smartphone camera. Serial dilutions of goat anti-mouse IgG-Alexa Fluor 488 in duplicates was printed in Row A and B. Fixed concentration of Goat anti-mouse IgG-Alexa Fluor 488 was printed in Row C and D for image correction and calibration. Panel (a) shows image outputs by the two readers. Panel (b) shows comparison of normalized signal values in relative fluorescence units (RFU), where each row was individually normalized. Panel (c) shows correlation between the normalized RFUs of the two readers.

Laser scanner

(a)

Smartphone reader

Row

A B C D 1

2

3

4

2

RFU

(b)

5

6

7

8

1

2

3

2

Row A

Laser scanner Smartphone reader

1

4

5

6

7

Row B

8

Laser scanner Smartphone reader

1

0

0 100

200

400

800

1600

IgG Dilution

3200

6400

IgG Dilution

Correlation between normalized signal values

(c) RFU

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 22

1.6 1.2 0.8 0.4 0

1.6 1.2 0.8 0.4 0

Row A

1

2

3

4

5

6

7

8

Row B

1

2

3

Spot Position Laser scanner

Smartphone reader

Laser scanner

ACS Paragon Plus Environment

4 5 6 Spot Position

7

Smartphone reader

8

Page 21 of 22

21 Figure 4 Evaluation of the smartphone reader with NHPs vaccinated with rVSV-EBOV vaccine. The microarray was printed with GP mucin proteins from BDBV (A1, A2), MARV (A3, A4), RESTV (A5, A6), SUDV (A7, A8), TAFV (B1, B2) and EBOV (B3, B4); NP proteins from BDBV (B5, B6), MARV (B7, B8), RESTV (C1, C2), SUDV (C3, C4), TAFV (C5, C6) and EBOV (C7, C8). Row D was printed with goat anti-mouse IgG- Alexa Flour 488 as calibration spots (D1, D3, D6, D8), positive controls of human IgG (D2, D4), and negative controls of buffer (D5, D7). Sera from four vaccinated NHPs and two controls (pre-vaccinated) were tested in the reader system. Goat antihuman IgG-Alexa Fluor 488 was used for detection of antibodies to microarray. The red windows show the location of EBOV-GP proteins in the array (a). The cross-sectional intensity profiles of B3 and B4 spots are shown in panel (b). The fluorescent signals (RFU) from vaccinated NHP samples are higher than the pre-immune NHP controls.

(b) Intensity profiles

(a) Smart phone images A B C D

Vac Vac

A B C D

Vac Vac Vac

RFU

Pre-vaccinated

Spot B3

Pre Pre Pre

A B C D

Spot B4

A B C D

Vac Vac

A B C D

Vac

RFU

Vaccinated

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

Vac Pre Pre

A B C D 1

2

3

4

5

6

7

8

Cross-sectional position

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 22 of 22

22 For TOC only

425

ACS Paragon Plus Environment