YB1

Infrared Sensors for Rapid Identification of Select Microbial

Foodborne Contaminants• PI: Dr. L. Mauer• Co-PI’s: Dr. M. Cousin, Dr. B. Reuhs, Dr. S.

Krishnan, Dr. J. Gore, Dr. J. Bouldin• Research Associates: Dr. Sol Kim, Y. Burgula,

D. Khali, J. Stratton

Slide 1

YB1 Yash Burgula, 4/6/2004

Project Milestones1. Create a library of FT-IR spectra of bacterial cell wall

components and whole cells (from Salmonella, Campylobacter jejuni, and Escherichia coli O157:H7) needed for cell identification and differentiation.

2. Develop FT-IR methods for cell identification and quantification in water, cultural media, and foods.

3. Develop a limited wavelength approach for cell identification.

4. Build and validate an IR sensor based on the most promising few-wavelength algorithm developed using FT-IR techniques developed in the first two milestones.

FT-IR instrument

ThermoElectron, Nexus 670

Milestone 1: Library of cell wall components and whole cells

Progress: E. coli and Salmonella

Major Bacterial Mid-IR Peaks

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

0.55

Log(

1/R

)

1000 2000 3000 Wavenumbers (cm-1)

1

2

3

4

5

6 78-12

Peak #Wavenumber (cm-1) Assignment

1 3500 OH stretching

2 3200NH stretching (amide A) of proteins

3 2959-2852CH, CH2, CH3stretching in fatty acids

4 1655-1637Amide I of α-helical & β-pleated structures

5 1548 Amide II band

6 1515 Tyrosine band

7 1468CH2 bending of methylene

8 1310-1240Amide III band components of proteins

91250-1220, 1084-1088

PO2 stretching of phospodiesters

10 1200-900C-O-C, C-O of variouspolysaccharides

11 720CH2 rocking of methylene

12 900-600 "Fingerprint Region"

(Filtered Salmonella Typhimurium spectrum)

Select Bacteria

• Escherichia coli- E. coli DH5α, E. coli K12, E. coli O157:H7, E. coliO157:12, E. coli O157:19

• Salmonella- Salmonella enterica serovars

S. Typhimurium, S. Enteritidis, S. Thomasville,S. Brandenburg, S. Hadar, S. Seftenberg

- ~250 LPS isolates from J. Bouldin (USDA-ARS)

• Campylobacter- C. jujuni, C. lari, C. coli

Lipopolysaccharides (LPS)

• Lipid A- Highly conserved

• Core oligosaccharides (Core OS)- Non-repeating sugars- Inner/Outer

• O-polysaccharides (O-PS)- Repeating unit: 3-5 sugars- High variation

S. Typhimurium

S. Enteritidis

S. Thomasville

S. Brandenburg

S. Hadar

S. Seftenberg

DOC-PAGE (Salmonella LPS Extracts)

S. Typhimurium

S. Enteritidis

S. Thomasville

S. Brandenburg

Log

(1/R

)

S. Hadar

S. Seftenberg

500 1000 1500 2000 2500 3000 3500 4000

Wavenumbers (cm-1)

II III IV VIFTIR Spectra of FTIR Spectra of SalmonellaSalmonella LPS ExtractsLPS Extracts

CV1

CV2

S. Hadar

S. Brandenburg

S. Enteritidis

S. Typhimurium

S. Seftenberg

S. Thomasville

CVA-PCA (Salmonella LPS Extracts)

S. Enteritidis

S. Hadar

S. Seftenberg

S. Thomasville

Log

(1/R

)

S. Typhimurium

S. Brandenburg

1000 1500 2000 2500 3000 3500 Wavenumbers (cm-1)

I II III IV V

FTIR Spectra of intact FTIR Spectra of intact SalmonellaSalmonella cellscells

CVACVA--PCA (intact PCA (intact SalmonellaSalmonella cells) cells)

-15 CV1 5

CV2

5

S. Brandenburg

Summary of Results• FTIR spectra of crude LPS extracts in combination with

chemometrics successfully differentiated and classified between E. coli strains and between Salmonella serotypes

• Carbohydrate region (1,200-900 cm-1) was better suited for spectral analysis of crude bacterial LPS extracts than the entire mid-IR region (4,000-700 cm-1)

• FTIR spectra of intact cells of Salmonella serotypes failed to differentiate between Salmonella serotypes, with the exception of S. Brandenburg due in part to specific peaks in the amide region (1,800-1,500 cm-1)

• FTIR spectra of crude bacterial LPS extracts may facilitate taxonomical or epidemiological studies of microorganisms in a rapid, sensitive, and accurate way.

Milestone 2: FT-IR method development for cell identification

and quantification in water, cultural media, and foods

Progress: E. coli and Salmonella

Detection of whole cells of

O157:H7Salmonella Typhimurium and E. coli

IR Microscope Method

Contact ATR Objective

Membrane Filter

Pressure Sensitive Stage

Contact Indicator Panel

Filtration Apparatus

Vacuum

Membrane Filter

Membrane filtrationMembrane filtration Contact ATRContact ATR

Summary of Results• Filtration of bacterial samples is suitable for

sample preparation and analysis using the contact ATR accessory of the FT-IR ContinuµmMicroscope.

• The sensitivity of detection using the FT-IR was 500 CFU/mL following incubation of the Salmonella Typhimurium/ E. coli O157:H7 in TSB for 6 h.

• This FT-IR method required less time and is less expensive than previous FT-IR methods.

Detection of E. coli O157:H7 in Fruit Juices

Main Bench ATR Method• Samples concentrated on

Metricel® Filter using the filtration apparatus

• Spectral Collection (256 scans, 4 cm -1): FTIR main bench using ZnSe Flat Plate ATR accessory and Pressure Clamp

• Discriminant Analysis (DA): TQ Analyst to discriminate between E. coli O157:H7 and E. coli K12 in the amide II region

Flat plate multibounce ATR

With filter sample using pressure clamp

Summary of Results

• DA was able to classify the pathogenic E. coli O157:H7 from the non-pathogenic E. coli K-12 in all fruit juices analyzed without any misclassification

• Background matrix (juice type) did not significantly impact the analysis of the spectrum

Selectivity experiments• To improve specificity of detection of E. coli

O157:H7 by use of selective enrichment

Summary of Results

• The selective enrichment of modified EC broth is effective against non-targets such as Pseudomonas aeruginosa and E. coli O157:H7 is now detectable at the 6th dilution upon incubation.

• The sensitivity of detection using the FT-IR was 500 CFU/mL following selective enrichment and 7 h incubation of the E. coli O157:H7

Summary of Results

• Use of E. coli O157:H7 specific Dynabeads® to capture E. coli O157:H7 in a mixed broth system works

• Discriminant analysis was able to classify the pathogenic E. coli O157:H7 from the non-pathogenic E. coli K-12 without any misclassifications

• These findings can be further extended to liquid food systems and mixed culture systems

Milestone 3: Develop a limited wavelength approach for cell

identification

Algorithm

Where: C = Bacterial concentration ci = Calibration constantsIAλ,B1 = Integrated absorbance (sensor

signal) for first bacterial absorbance regionIAλ,B2 = Integrated absorbance (sensor signal)

for second bacterial (or interference) absorbance region

IAλ,bl = Integrated absorbance (sensor signal) for baseline absorbance region

bl

B

bl

B

IAIAR

cIAIA

c,

2,3

,

1,21c C

λ

λ

λ

λ ++=

⎟⎟⎠

⎞⎜⎜⎝

⎛⎟⎟⎠

⎞⎜⎜⎝

⎛+⎟

⎟⎠

⎞⎜⎜⎝

⎛+⎟

⎟⎠

⎞⎜⎜⎝

⎛+++=

bl

B

bl

B

bl

B

bl

B

bl

B

bl

B

IAIA

IAIA

cIAIA

cIAIA

cIAIA

cIAIA

c,

2,

,

1,6

2

,

2,5

2

,

1,4

,

2,3

,

1,21c C

λ

λ

λ

λ

λ

λ

λ

λ

λ

λ

λ

λ

Milestone 4: Build and validate an IR sensor based on the most

promising few-wavelength algorithm

ATR MicrobeSensor

Source modulator

Pulsable IR Source with reflector Quad detector with

filters

Signal conditioning and calibration

ATR Crystal

Sample

Presentations1. Y. Burgula, D. Khali, S. Kim, M. A. Cousin, B. R. Reuhs, and L. J. Mauer.

Development of a method to detect Salmonella Typhimurium using FT-IR spectroscopy. International Food Safety and Quality Conference. 2003. Abstract.

2. Y. Burgula, D. Khali, S. Kim, M.A. Cousin, B.R. Reuhs, and L.J. Mauer. Development of a method to improve the sensitivity and selectivity for detection of E. coli O157:H7 using FT-IR spectroscopy. Institute of Food Technologists’ Annual Meeting and Food Expo. 2004. Abstract.

3. Y. Burgula, D. M. Khali, S. Kim, M. A. Cousin, B. L. Reuhs, and L. J. Mauer. A rapid method for detection of E. coli O157:H7 using Dynabeads® and FTIR spectroscopy. International Association for Food Protection (IAFP) 91st Annual Conference. 2004. Abstract.

4. S. Kim, Y. Burgula, T. Ojanen-Reuhs, B.L. Reuhs, M.A. Cousin, and L.J. Mauer. Differentiation and classification of the crude lipopolysaccharides from Salmonellaspecies using Fourier transform infrared spectroscopy and canonical variate analysis. International Association for Food Protection (IAFP) 91st Annual Conference. 2004. Abstract

5. S. Kim, Y. Burgula, T. Ojaanen-Reuhs, B.R. Reuhs, M.A. Cousin, and L.J. Mauer. Differentiation and classification of the phenol-phase soluble crude lipopolysaccharidesfrom Eschericia coli strains using Fourier transform infrared spectroscopy and canonical variate analysis. Institute of Food Technologists’ Annual Meeting and Food Expo. 2004. Abstract.

Publications

1. Kim, S., B.L. Reuhs, and L.J. Mauer. Differentiation and classification of crude lipopolysaccharides from Salmonella strains using Fourier transform infrared spectroscopy and chemometrics. Journal of Applied Microbiology. (In press).

Questions?