Rapid Optical Detection Methods for Foodborne Pathogens

Bosoon Park*, Seung-Chul Yoon, Gary Gamble, Kurt Lawrence

*Contact: bosoon.park@usda.gov

USDA, ARSU.S. National Poultry Research Center

Quality & Safety Assessment Research UnitAthens, Georgia

mailto:bosoon.park@usda.gov

Research ObjectivesObjective 1: Develop high-speed imaging methods for rapid detection of pathogens in live poultry flocks, and foodborne hazards, including foreign materials in processed poultry products

Objective 2: Develop rapid methods and protocols for early detection, identification, and quantification of pathogens in poultry products (foods) using imaging spectroscopy

Objective 3: Develop methods to detect biofilms in poultry processing facilities with optical technologies. In developing the methods assess if any biomarkers can be identified to enhance or improve the detection sensitivity or specificity

Research Goals• Hypothesis (1A): Live layer flocks, highly infected with Salmonella, can be identified while in

poultry housing with an early-warning imaging system that detects bile in their fecal droppings.

• Hypothesis (1B): HI can rapidly differentiate normal chicken products from foreign materials during poultry further processing

• Hypothesis (2A): Foodborne pathogenic bacteria can be identified without culturing by HMI and multivariate data analysis (MVDA) methods.

• Hypothesis (2B): A molecular method using multiplex FISH microscopy can rapidly and simultaneously identify whole cells of different foodborne pathogenic bacteria with high specificity.

• Goal (2C): SERS-active nanoparticles conjugated with DNA aptamers can detect specific target bacteria.

• Hypothesis (2D): BPW can be chemically modified to neutralize a wide variety of poultry processing sanitizer solutions to eliminate their residual effect on microbial analysis.

• Hypothesis (2E): HI can classify E. coli O157:H7, O26, O45, O103, O111, O121, and O145 serovars growing in modified chromogenic agar media.

• Hypothesis (3): An optical imaging system can be developed to detect the presence of biofilm in the poultry processing environment.

Rapid optical detection methods for foodborne pathogens (1)

Rapid optical detection methods for foodborne pathogens (2)

Accomplishments- Biofilm (Obj. 3) -

• Desirable to detect both living and disinfected biofilms: dead cells and EPS may serve as growth platforms

• Cost- and time-effective detection over relatively large surface areas:• Goal is to enhance pre-operational organoleptic

inspection by qualified personnel to insure effectiveness of SSOPs

Process Flow for Poultry Processing

Food Grade Dyes

• Food grade dyes (GRAS and/or FDA approved) were evaluated for their ability to bind with attached cells (live or dead) and EPS components, using various acceptable solvent systems, to a level perceptible to the human eye using reflectance or fluorescence

• Dyes included riboflavin, quinine, thiamine, pyridoxine, erioglaucine, erythrosine, allura red, brilliant blue, indigo carmine, fast green, tartrazine, sunset yellow, beet extract, grape extract, annatto, and chlorophyll.

Erythrosin B(aka FD&C Red 3)

• Procedure• Grow Pseudomonas putida or

Listeria innocua• biofilm on stainless steel

coupon• Expose biofilm to various

sanitizer/cleaners• Pre-treat resulting coupon with

QAC• Dye coupon using Erythrosin B

solution• Photograph coupons• Analyze color difference from

background using method simulating human perception (CIELAB L*a*b*: ΔE*)

Results• Erythrosin B method non-specifically dyes all

biofilm components • Common methods to sanitize biofilms do not

necessarily remove the components• Ongoing and Future Work• Single-stage food grade solution of erythrosin

B is under development to provide a simple, cost effective, and efficient method to visualize biofilm

• The method highlights the resistance of some biofilms to cleaning: new alternatives to current cleaning/sanitizing methods are under evaluation

• Multispecies biofilms grown in lab or obtained from processing environment will be included in future studies

Accomplishments - Intervention Carryover Mitigation (Obj. 2D) -

nBPW Origin and Optimization

• nBPW was originally developed to overcome potential false negative reporting of Salmonella

• Neutralizing agents (lecithin, sodium thiosulfate, sodium bicarbonate) were added to standard BPW to mitigate carry-over of commonly used chemical interventions

• Collaboration with USNPRC: BEAR and PPMS Units

Steps for Detecting Contaminants

Effect of Carry-over of Antimicrobials

Ongoing and Future Work on nBPW

• The efficacy of nBPW for recovery of Campylobacter and other pathogens is not yet optimized:• Campy is sensitive to residual neutral pH sodium chlorite, which is not

neutralized by thiosulfate present in nBPW (Berrang et al. J. Appl. Poult. Res., 2018, 27: 299– 303.)

• a universal antioxidant is currently being developed to replace thiosulfate

• Prevalence of Campylobacter on parts decreased with nBPW vs. BPW (Williams, et al. J. Food Prot., 2018, 81:1851-1863.)

• a change in FSIS assay from direct plating to enrichment procedure (as with Salmonella) may provide clarification

• nBPW without challenge by antimicrobial intervention carry-over can be optimized to provide better conditions for survival of Campylobacter

Campylobacter survival in Bolton broth and nBPW during simulated shipment

from plant to lab

Rapid Optical Detection Methods for Foodborne Pathogens (3)

Accomplishments- Hyperspectral Microscope Imaging (Obj. 2A) -

Eady, Park, Choi. J. Food Prot. 78(4): 668-674, 2015Eady, Park. J. Microscopy. 263(1): 10-19, 2016Eady, Park. J. Spectral Imaging 5(1): 1-10, 2016Park et al. J. Annals Clinic. Pathol. 5(2): 1108-1116, 2017Eady, Park, Yoon, Lawrence, Haidekker. Trans. ASABE61(2): 437-448, 2018Eady, Park. J. Spectral Imaging. 2018. https://doi.org/10.1255/jsi.2018.a6Eady, Setia, Park. Talanta 195: 313-319, 2019

https://doi.org/10.1255/jsi.2018.a6

HMI for Presumptive Screening- Goal: Culture-free, Label-free -

Immunomagnetic Separation (IMS) of Foodborne Pathogens

Preliminary results• IMS recovery of S. Enteritidis

and non-specific binding rates of E. coli are influenced by:

• IMB size – Recovery decreases as IMB size increases;

• Initial bacterial concentration spiked into the sample –Recovery at low spiking levels is higher;

• The IMB size effect is more noticeable at higher spiking levels.

3-D Scattering Intensity Distribution of Bacterial Cells

HMI from Pathogen Samples with ROI for Analysis

Classification ResultsA Quadratic discriminant analysis (QDA) was used to classify bacteria at cellular level

Bacteria Cells Accuracy (%)

Bifidobacter 76 100

Campylobacter 68 100

Clostridium 52 100

Enterobacter 82 100

Lactobacillus 31 93.5

Shigella 51 100

Overall 360 99.4

Impact and Future Research

• Hyperspectral Microscope Imaging Technology– $120,000 Incoming Agreement (No. 58-6040-8-031F) with

Korea Food Research Institute (KFRI)– CRADA with TruTag Technologies, Inc. in progress– Tech transfer to UCD (Ireland) and Stellenbosch (South

Africa)– New FPI-based hyperspectral microscope imaging platform

for bacterial detection at cellular level– New Near-infrared hyperspectral microscope imaging

system for biochemical constituents of bacterial cells– Develop portable (miniaturized) HMI system

Accomplishments- Surface Enhanced Raman Spectroscopy (SERS) (Obj. 2C) -

• Chen, Park, Eady. J. Food Analy. Methods 10(9):3181-3194, 2017• Chen, Park, Huang, Zhao, Kwon. J. Food Measur. & Charact.11(4):1773-1779, 2017• Chen, Park. J. Food Microbiol. 267:1-8, 2018

Label-free SERS Detection for Salmonella Typhimurium (ST)

• 10 times better sensitivity than existing methods

• Able to differentiate serotypes

400 600 800 1000 1200 1400 1600 18000.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

16451233953

10271077

blank

buffer

EC

EF

SE

ST

Norm

alized

Intens

ity

Wavenumber (cm-1)

998

725785

1315-1330

551617

683852 1568

1455-1470

• Chen, Park, Huang, Zhao, Kwon. J. Food Measur. & Charact.11(4):1773-1779, 2017

Accomplishments- Surface Plasmon Resonance Imaging (SPRi) -

• Chen, Park. Analyt. Bioanalyt. Chem. 410:5455-5464, 2018.• Chen, Park. J. Food Prot. 79(6):1055-1069, 2016

Surface Functionalization of SPRi Sensor Chips

• Automatic arraying of ligand spots: Spot size: 500 µm Average deposition volume: 5 nL/spot Sensor surface area: 0.96 cm x 0.64 cm

• Optimization of ligand immobilization condition was conducted on a single chip. Optimization parameters: pH (4.6, 5.6, and 7.4) Concentration of antibody (10-250 µg/mL)

Optimizing Ligand Immobilization• Biochip image before SPRi

Ligand spots are visible at concentration ≥ 100 µg/mL

Spot size may exceed 500 µm at high concentrations

• SPR signal intensity after sample injection Max signal is obtained at pH 4.6

and 250 µg/mL

SPRi Image Analysis

• Flow cell images at the beginning of each injection is set as reference images

• Difference images are generated by subtracting real-time flow cell images during each injection by the reference image

Biochip image

Multiplex and Label-Free Screening of Two Pathogens(S. Typhimurium and O157)

Impact and Future Research (2)

• Nanotechnology– $60,000 Incoming Agreement (No. 293-0210-110) with Korea

Rural Development Administration (KRDA)– Collaboration with HORIBA France for upgrade SPRi– Webinar for “SPRi for foodborne pathogen screening” 44

attendees from 21 countries around the world. The webinar is posted on youtubehttps://www.youtube.com/watch?v=1SKBgVIlsR0&t=233s)

– High-throughput multiplexed SPRi sensing for the detection and identification of several bacteria species simultaneously in about 2 hours

– Develop portable (miniaturized) SPRi system - Horiba France– Optimize sampling protocol with IMS from matrix (poultry

carcass rinse)

https://www.youtube.com/watch?v=1SKBgVIlsR0&t=233s

Rapid Optical Detection Methods for Foodborne Pathogens (4)

Components of Agar Media

SerogroupFalse

positivea(%)

False negative

(%)

Sensitivityc(%)

Specificityd(%)

MCCe(%)

O26 1.0 24.5 75.5 98.9 0.80

O45 0.0 5.4 94.6 100 0.97

O103 4.6 2.1 97.9 95.4 0.81

O111 0.0 0.0 100 100 1.00

O121 0.0 1.8 98.2 100 0.98

O145 0.3 21.9 78.0 99.6 0.85

UKB 0.5 12.6 87.4 99.5 0.83

Multiplex fluorescence in-situ hybridization and spectral unmixing

• m-FISH to detect several (hopefully >5) microorganisms simultaneously by one FISH reaction.

• Hyperspectral imaging is a technique that has potential to overcome the spectral cross-talk among emission channels in standard optical fluorescence.

• Probe design and optimization– m-FISH probes should show as

little mutual complementarity as possible.

– m-FISH probes should be optimized to employ the same hybridization temperatures and stringency conditions.

Hyperspectral Cell Imaging forSimultaneous Detection of Foodborne Pathogens

Hyperspectral Fluorescence In Situ Hybridization (FISH) Multiplexing: Mixed Pseudomonas aeruginosa A27853 & Campylobacter jejuni 4993

Hyperspectral Cell Imaging:Cell Counting via Accurate & High-Throughput Image Analysis

Result Counts

Rapid Optical Detection Methods for Foodborne Pathogens (5)

High-speed Hyperspectral Imaging

• Technology & system development– Wavelength-configurable hyperspectral imaging

systems (Vis-NIR pushbroom line scanners)• Multispectral imaging with 2-3 wavelengths: ~500 Hz• Hyperspectral imaging with full spectra: 300-600 Hz

– RGB color-based hyperspectral imaging • Color-to-hyperspectral image reconstruction: camera

frame rate• Multispectral-to-hyperspectral image reconstruction:

camera frame rate

High-Speed Hyperspectral ImagingOne Single Universal System for Real-Time Disease and/or

Fecal Detection System

High-Speed Hyperspectral Imaging (2)Hyperspectral Machine Vision Guided-Real-Time Foreign Material

Detection System

Spectral Library: Multispectral Data (7 bonds)

Applications of Hyperspectral Image Reconstruction

Test Set: Original Spectrum vs. Recovered Spectrum: Ingesta (Fecal Mat.)

Fecal Materials and Normal SkinMean Reflectance Spectra

Big Six non-O157 STECsMean Reflectance Spectra

Ongoing and Future Research

• Agar media: Additional STEC media types and additives

• FISH: Probe optimization• High-speed hyperspectral imaging: poultry parts

inspection & foreign material detection

Publications• Gamble, G.R., Berrang, M.E., Buhr, R.J., Hinton Jr, A., Bourassa, D.V., Johnston, J.J., Ingram, K.D., Adams, E.S., Feldner, P.W. 2016. Effect of simulated sanitizer carryover on recovery of

salmonella from broiler carcass rinsates. Journal of Food Protection. 79(5):710-714.• Park, B., Yoon, S.C. 2015. Real-time hyperspectral imaging for food safety applications. Book Chapter. Chapter 13: pp 305-329.• Park, B. 2015. AOTF hyperspectral microscope imaging for foodborne bacteria detection. Book Chapter. Chapter 15: pp 359-390.• Yoon, S.C., Lawrence, K.C., Park, B. 2015. Automatic counting and classification of bacterial colonies using hyperspectral imaging. Food and Bioprocess Technology. 8:2047-2065.• Yoon, S.C., Shin, T., Lawrence, K.C., Heitschmidt, G.W., Park, B., Gamble, G.R. 2015. Hyperspectral imaging using RGB color for foodborne pathogen detection. Journal of Electronic

Imaging. 24(4):043008.• Chu, X., Wang, W., Yoon, S.C., Ni, X., Heitschmidt, G.W. 2017. Detection of aflatoxin B1 (AFB1) in individual maize kernels using short wave infrared (SWIR) hyperspectral imaging.

Biosystems Engineering. 157:13-23.• Gamble, G.R., Berrang, M.E., Buhr, R.J., Hinton Jr, A., Bourassa, D.V., Johnston, J.J., Ingram, K.D., Adams, E.S., Feldner, P.W. 2017. Neutralization of bactericidal activity related to

antimicrobial carry-over in broiler carcass rinse samples. Journal of Food Protection. 80:(4)685-591.• Park, B., Seo, Y., Eady, M.B., Yoon, S.C., Hinton Jr, A., Lawrence, K.C., Gamble, G.R. 2017. Classification of Salmonella serotypes with hyperspectral microscope imagery. Annals of

Clinical Pathology. ACS. • Wang, B., Park, B., Xu, B., Kwon, Y. 2017. Label-free biosensing of Salmonella enterica serovars at single-cell level. Journal of Nanobiotechnology (Biomed Central Open Access).

15:40.• Chen, J., Park, B., Eady, M.B. 2017. Simultaneous detection and serotyping of Salmonellae by immunomagnetic separation and label-free surface enhanced Raman spectroscopy.

Journal of Food Analytical Methods 10(9): 3181-3193.• Chen, J., Park, B., Huang, Y., Zhao, Y., Kwon, Y. 2017. Label-free SERS detection of Salmonella Typhimurium on DNA aptamer modified AgNR substrates. Journal of Food Measurement

and Characterization.• Eady, M.B., Park, B. 2016. Rapid identification of Salmonella serotypes through hyperspectral microscopy with different lighting sources. Journal of Spectral Imaging.• Berrang, M.E., Gamble, G.R., Hinton, A. Jr., Johnston, J.J. 2017. Neutralization of residual antimicrobial processing chemicals in broiler carcass rinse for improved detection of

Campylobacter. Journal of Applied Poultry Research. https://doi:10.3382/japr/pfx071.• Berrang, M.E., Harrison, M., Meinersman, R., Gamble, G.R. 2017. Self-contained chlorine dioxide generation and delivery pods for decontamination of floor drains. Journal of Applied

Poultry Research 26(3):410-415.• Chen, J., Park, B. 2017. Label-free screening of foodborne Salmonella using surface plasmon resonance imaging. Analytical and Bioanalytical Chemistry.

https://doi.org/10.1007/s00216-017-0810-z• Chen, J., Park, B. 2018. Effect of immunomagnetic bead size on recovery of foodborne pathogenic bacteria. International Journal of Food Microbiology 267: 1-8.• Eady, M., Park, B., Yoon, S.C., Lawrence, K.C., Haidekker, M. 2018. Methods for hyperspectral microscope calibration and spectra normalization from images of bacteria cells,

Transactions of the ASABE 61(2): 437-448.• Eady, M., Park, B. 2018. Unsupervised classification of individual foodborne bacteria from a mixture of bacteria cultures within a hyperspectral microscope image. Journal of Spectral

Imaging. https://doi.org/10.1255/jsi.2018.a6• Jiang, H., Yoon, S.C., Zhuang, H., Wang, W., Yang, Y. 2017. Evaluation of factors in development of Vis/NIR spectroscopy models for discriminating PSE, DFD and normal broiler breast

meat. British Poultry Science 58(6): 673-680• Landrum, M.A., Cox, N.A., Wilson, J.L., Cosby, D.E., Berrang, M.E., Gamble, G.R., Da Costa, M.J., Pesti, G.M., Hinton, A. Jr. 2018. Low Ph PROCESSING AID TO LOWER THE PRESENCE OF

NATURALLY OCCURRING Campylobacter on whole broiler carcasses. Advanced Food and Nutrition Science 3:7-13.• Nam, S., Park, B., Condon, B. 2018. Water-based binary polyol process for the controllable synthesis of silver nanoparticles inhibiting human and foodborne pathogen bacteria. RSC

Advances 8:21937-21947.• Seo, Y.W., Park, B, Yoon, S.C., Lawrence, K.C., Gamble, G. 2018. Hyperspectral microscopic morphological image analysis for foodborne pathogenic bacteria identification.

Transactions of the ASABE 61(1): 5-13.

https://doi:10.3382/japr/pfx071https://doi.org/10.1007/s00216-017-0810-zhttps://doi.org/10.1255/jsi.2018.a6

Rapid Optical Detection Methods for Foodborne PathogensResearch ObjectivesResearch GoalsRapid optical detection methods for foodborne pathogens (1)Rapid optical detection methods for foodborne pathogens (2)Accomplishments�- Biofilm (Obj. 3) -Process Flow for Poultry ProcessingFood Grade DyesErythrosin B�(aka FD&C Red 3)ResultsAccomplishments �- Intervention Carryover Mitigation (Obj. 2D) -Steps for Detecting ContaminantsEffect of Carry-over of AntimicrobialsOngoing and Future Work on nBPWCampylobacter survival in Bolton broth and nBPW during simulated shipment from plant to labRapid Optical Detection Methods for Foodborne Pathogens (3)Accomplishments�- Hyperspectral Microscope Imaging (Obj. 2A) -HMI for Presumptive Screening- Goal: Culture-free, Label-free -Immunomagnetic Separation (IMS) of Foodborne Pathogens3-D Scattering Intensity Distribution of Bacterial CellsHMI from Pathogen Samples with ROI for AnalysisClassification Results�A Quadratic discriminant analysis (QDA) was used to classify bacteria at cellular levelImpact and Future ResearchAccomplishments�- Surface Enhanced Raman Spectroscopy (SERS) (Obj. 2C) -Label-free SERS Detection for Salmonella Typhimurium (ST)Accomplishments�- Surface Plasmon Resonance Imaging (SPRi) -Surface Functionalization of SPRi Sensor ChipsOptimizing Ligand ImmobilizationSPRi Image AnalysisMultiplex and Label-Free Screening of Two Pathogens�(S. Typhimurium and O157)Impact and Future Research (2)Rapid Optical Detection Methods for Foodborne Pathogens (4)Components of Agar MediaMultiplex fluorescence in-situ hybridization and spectral unmixingHyperspectral Cell Imaging forSimultaneous Detection of Foodborne PathogensHyperspectral Cell Imaging:Cell Counting via Accurate & High-Throughput Image AnalysisResult CountsRapid Optical Detection Methods for Foodborne Pathogens (5)High-speed Hyperspectral ImagingHigh-Speed Hyperspectral Imaging�One Single Universal System for Real-Time Disease and/or Fecal Detection SystemHigh-Speed Hyperspectral Imaging (2)�Hyperspectral Machine Vision Guided-Real-Time Foreign Material Detection SystemSpectral Library: Multispectral Data (7 bonds) Applications of Hyperspectral Image ReconstructionTest Set: Original Spectrum vs. �Recovered Spectrum: Ingesta (Fecal Mat.)Fecal Materials and Normal Skin�Mean Reflectance SpectraBig Six non-O157 STECs�Mean Reflectance SpectraOngoing and Future ResearchPublications