USF Libraries
USF Digital Collections

Development of ganglioside-based assays for the identification of botulinum and cholera toxins utilizing an evanescent w...

MISSING IMAGE

Material Information

Title:
Development of ganglioside-based assays for the identification of botulinum and cholera toxins utilizing an evanescent wave biosensor
Physical Description:
Book
Language:
English
Creator:
Bedenbaugh, Crystal M
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Vibrio cholerae
Detection
Clostridium botulinum
Oysters
Bioweapon
Dissertations, Academic -- Biology -- Doctoral -- USF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: An evanescent wave fiber-optic biosensor was used in an effort to develop an assay for the rapid detection of two biological toxins: cholera toxin and botulinum toxin. The Analyte 2000 fiber-optic biosensor utilizes a sandwich immunoassay format. Gangliosides or liposomes are directly adsorbed to the surface of the fiber-optic waveguide through hydrophobic interactions. The waveguide is exposed to a sample containing the toxin of interest, then subsequently exposed to a polyclonal detection antibody conjugated to the fluorophore cyanine 5. Excitation light from a 635nm laser diode is propagated through the waveguide and fluorescent molecules within approximately 100nm of the waveguide are excited. The emission light from the excited cyanine 5 molecules reverberates into the waveguide and is quantitated in pico Amperes and displayed on a computer. The exotoxins of Vibrio cholerae and Clostridium botulinum, cholera and botulinum toxin, respectively, were used for pote ntial assay development. Assay development utilizing the biosensor was attempted for the detection of botulinum toxin in buffer. The limit of detection remained too high to generate a positive signal for the detection of botulinum toxin. Biosensor assays were developed to detect cholera toxin in buffer, oyster homogenate, pure culture and induction media. A cholera toxin standard curve was generated with a limit of detection of 1 ng/ml. The values were normalized by setting 100 ng/ml of cholera toxin to a value of 100. Signals were detected in oyster homogenate spiked at 5 ug/ml as well as unspiked oyster homogenate. A Western blot showed that there were cross reactive proteins in the oyster matrix at molecular weights different from those of the cholera toxin. Cholera toxin production by three strains of Vibrio cholerae with values estimated to range from 100 pg --^ 100 ng was detected with the biosensor. Additionally, oysters were harvested from Tampa Bay and placed in a 10 gallon tank filled with different types of induction media. The tank was inoculated with Vibrio cholerae and the oysters and induction medium were analyzed at varying times for the presence of cholera toxin. Vibrio cholerae cells were viable through 24 hours but no toxin was detectable.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2006.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Crystal M. Bedenbaugh.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 140 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001796865
oclc - 156930906
usfldc doi - E14-SFE0001617
usfldc handle - e14.1617
System ID:
SFS0025935:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001796865
003 fts
005 20070724100728.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 070724s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001617
040
FHM
c FHM
035
(OCoLC)156930906
049
FHMM
090
QH307.2 (ONLINE)
1 100
Bedenbaugh, Crystal M.
0 245
Development of ganglioside-based assays for the identification of botulinum and cholera toxins utilizing an evanescent wave biosensor
h [electronic resource] /
by Crystal M. Bedenbaugh.
260
[Tampa, Fla] :
b University of South Florida,
2006.
3 520
ABSTRACT: An evanescent wave fiber-optic biosensor was used in an effort to develop an assay for the rapid detection of two biological toxins: cholera toxin and botulinum toxin. The Analyte 2000 fiber-optic biosensor utilizes a sandwich immunoassay format. Gangliosides or liposomes are directly adsorbed to the surface of the fiber-optic waveguide through hydrophobic interactions. The waveguide is exposed to a sample containing the toxin of interest, then subsequently exposed to a polyclonal detection antibody conjugated to the fluorophore cyanine 5. Excitation light from a 635nm laser diode is propagated through the waveguide and fluorescent molecules within approximately 100nm of the waveguide are excited. The emission light from the excited cyanine 5 molecules reverberates into the waveguide and is quantitated in pico Amperes and displayed on a computer. The exotoxins of Vibrio cholerae and Clostridium botulinum, cholera and botulinum toxin, respectively, were used for pote ntial assay development. Assay development utilizing the biosensor was attempted for the detection of botulinum toxin in buffer. The limit of detection remained too high to generate a positive signal for the detection of botulinum toxin. Biosensor assays were developed to detect cholera toxin in buffer, oyster homogenate, pure culture and induction media. A cholera toxin standard curve was generated with a limit of detection of 1 ng/ml. The values were normalized by setting 100 ng/ml of cholera toxin to a value of 100. Signals were detected in oyster homogenate spiked at 5 ug/ml as well as unspiked oyster homogenate. A Western blot showed that there were cross reactive proteins in the oyster matrix at molecular weights different from those of the cholera toxin. Cholera toxin production by three strains of Vibrio cholerae with values estimated to range from 100 pg --^ 100 ng was detected with the biosensor. Additionally, oysters were harvested from Tampa Bay and placed in a 10 gallon tank filled with different types of induction media. The tank was inoculated with Vibrio cholerae and the oysters and induction medium were analyzed at varying times for the presence of cholera toxin. Vibrio cholerae cells were viable through 24 hours but no toxin was detectable.
502
Dissertation (Ph.D.)--University of South Florida, 2006.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 140 pages.
Includes vita.
590
Adviser: Daniel V. Lim, Ph.D.
653
Vibrio cholerae.
Detection.
Clostridium botulinum.
Oysters.
Bioweapon.
690
Dissertations, Academic
z USF
x Biology
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1617



PAGE 1

Development of Ganglioside-Based Assays for the Identification of Botulinum and Cholera Toxins Utilizing an Evanescent Wave Biosensor by Crystal M. Bedenbaugh A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Biology College of Arts and Sciences University of South Florida Major Professor: Daniel V. Lim, Ph.D. Valerie J. Harwood, Ph.D. My Lien Dao. Ph.D. Andrew Cannons, Ph.D. Date of Approval: July 10, 2006 Keywords: Vibrio cholerae detection, Clostridium botulinum oysters, bioweapon Copyright 2006, Crystal M. Bedenbaugh

PAGE 2

i TABLE OF CONTENTS LIST OF TABLES………………………………………………………………………. iv LIST OF FIGURES……………………………………………………………………... vii ABSTRACT ……………………………………………………………………………… x INTRODUCTION………………………………………………………………………... 1 Vibrio cholerae …………………………………………………………………... 1 Background……………………………………………………………….. 1 Historical Background……………………………………………………. 2 Classification………………………………………………………4 Transmission……………………………………………………… 6 Cholera toxin……………………………………………………… 6 Iron………………………………………………………………. 10 ToxR regulon……………………………………………………. 10 Disease Control………………………………………………………….. 11 Ecology and Marine Adaptations……………………………………….. 12 Survival in the environment……………………………………... 13 Association with marine organisms……………………………... 15 Detection of Vibrio cholerae and Cholera Toxin……………………….. 19 Vibrio cholerae …………………………………………………. 19 CT……………………………………………………………….. 22 Clostridium botulinum …………………………………………………………. 23 Background……………………………………………………………… 23 Virulence factors………………………………………………… 24 Types of botulism……………………………………………….. 25 Detection of Clostridium botulinum and Botulinum Toxin…………….. 27 Immunoassays…………………………………………………………………... 30 Background…………………………………………………………….. 30 Detection Format……………………………………………………….. 31 Antibodies………………………………………………………………. 32 Labels……………………………………………………………………. 35 Biosensors………………………………………………………………………. 36 Background……………………………………………………………… 36 Types of Optical Biosensors…………………………………………….. 38 Evanescent Wave Fiber-Optic Biosensor……………………………….. 39 Affinity-Based Elements………………………………………………… 40

PAGE 3

ii Summary………………………………………………………………… 42 MATERIALS AND METHODS……………………………………………………….. 45 Bacterial Strains…………………………………………………………………. 45 Media and Culture Conditions…………………………………………………... 45 Stock cultures……………………………………………………………. 45 Enrichment and selective media………………………………………… 45 Toxins ……………………………………………………………………………46 Gangliosides…………………………………………………………………….. 46 Liposomes……………………………………………………………………….. 46 Antibodies and Labeling………………………………………………………… 47 Sources………………………………………………………………….. 47 Cyanine 5 labeling………………………………………………………. 47 Biotin labeling…………………………………………………………… 48 Viable Counts ……………………………………………………………………49 Oysters…………………………………………………………………………... 49 Toxin Assays……………………………………………………………………. 50 Bicinchoninic acid protein assay……………………………………….. 50 SDS-PAGE……………………………………………………………… 51 Western blot…………………………………………………………….. 51 Ganglioside-based ELISA……………………………………………… 52 Indirect ELISA………………………………………………………….. 52 Toxin induction…………………………………………………………. 53 Toxin spiked oysters…………………………………………………….. 54 Tank inoculation………………………………………………………… 54 API 20E Identification…………………………………………………………... 55 PCR……………………………………………………………………………… 55 Ganglioside-Based Biosensor Assay for the Detection of Cholera Toxin……… 56 Analyte 2000 biosensor…………………………………………………. 56 Fiber preparation………………………………………………………… 57 Background readings……………………………………………………. 58 Sample assay……………………………………………………………. 58 Cholera toxin standard preparation……………………………………… 60 Liposome-Based Biosensor Assay for the Detection of Botulinum Toxin……… 60 Fiber preparation………………………………………………………… 60 Background readings……………………………………………………. 61 Sample assay…………………………………………………………….. 61 RESULTS………………………………………………………………………………. 62 Biosensor Immunoassay Development for the Detection of Cholera Toxin……. 62 Cholera toxin purity……………………………………………………... 62 BCA protein assay………………………………………………………. 64 Comparison of Affinity and Specificity for GM1 and Polyclonal Antibody Binding to V. cholerae Whole Cells and Toxin……………... 64 CT Standard Curve ……………………………………………………... 66

PAGE 4

iii Sensitivity of Antibody Versus Ganglioside as Capture Molecule……... 68 Signal Amplification…………………………………………………….. 69 Toxin Induction…………………………………………………………. 71 ELISA and Analyte 2000 utilizing CT………………………….. 71 CAYE as induction medium…………………………………….. 73 Alternate induction media……………………………………….. 75 Toxin Spiked Oysters…………………………………………………… 77 Cross-Reactive Proteins…………………………………………………. 78 Affinity for polyclonal anti-CTrabbit antibody………………. 78 Affinity for IgG antibody……………………………………….. 79 Specificity for GM1……………………………………………... 80 Removal of cross-reactive proteins from oyster matrix………… 82 Tank Inoculation………………………………………………………… 84 Instant Ocean as induction medium…………………………….. 84 SWC-DI as induction medium………………………………….. 88 CAYE-IO as induction medium………………………………… 90 Biosensor Immunoassay Development for the Detection of Botulinum Toxin ……………………………………………………………………………92 BCA Protein Assay……………………………………………………… 92 ELISA for Detection of Botulinum Toxoids Using Anti-Botulinum Antibody……………………………………………………………….. 93 Detection of Botulinum Toxoids Using Sandwich ELISA with GT1b Capture………………………………………………………….. 94 GT1b-and GD1b-Based ELISA…………………………………………. 96 Biosensor Assay Detecting Toxoid Type A Using Liposomes…………. 97 Biosensor Assay Utilizing Anti body as Captur e Molecu le…………… ... 99 Varying liposome concentrations as detection molecule for the detection of botulinum toxoid type A…………………….. 99 Varying liposome concentrations as detection molecule for the detection of botulinum toxoid type B……………………. 105 DISCUSSI ON............................................................................................................. 109 REFERENC ES............................................................................................................ 123 ABOUT THE AUTHOR....................................................................................End Page

PAGE 5

iv LIST OF TABLES Table 1 Classification of epidemic and non-epidemic associated Vibrio cholerae ……………………………………………………………………4 Table 2 Antigenic determinants of Vibrio cholerae O1 and O139 serotypes…….. 5 Table 3 Biochemical tests used to differentiation between classical and El Tor biotypes of Vibrio cholerae serogroup O1………………………... 5 Table 4 Factors affecting the survival and growth of V. cholerae in the environment …………………………………………………………….. 14 Table 5 PCR primers for identification of Vibrio cholera ………………………. 56 Table 6 Extrapolated CTconcentrations from BCA standard curve…………. 64 Table 7 Average normalized pA values and standard deviations for CTstandard curve using the Analyte 2000………………………………… 67 Table 8 Average normalized pA values and standard deviations for signal amplification using the Analyte 2000………………………………….. 71 Table 9 Average normalized pA values and standard deviations for CT standard curve using the Analyte 2000………………………………… 73 Table 10 Cholera toxin induction values utilizing an ELISA and the Analyte 2000…………………………………………………………………….. 75 Table 11 Viable cell counts for CT induction utilizing alternative media ……….. 77

PAGE 6

v Table 12 Toxin spiked oyster homogenate samples analyzed with the Analyte 2000…………………………………………………………….. 78 Table 13 Analyte 2000 values for oyster homogenate fractions………………….. 84 Table 14 Isolates presumptively identified by API 20E for Instant Ocean tank............................................................................................................ 85 Table 15 Cell counts for Vibrio cholerae recovered from oyster and Instant Ocean……………………………………………………………. 86 Table 16 Isolates presumptively identified by API 20E for SWC-DI tank………. 89 Table 17 Isolates presumptively identified by API 20E recovered from CAYE-IO with 2% glucose tank…………………………………………91 Table 18 Extrapolated botulinum toxoid protein concentrations…………………. 93 Table 19 Biosensor values for toxoid A adsorbed directly to the waveguide…….. 98 Table 20 Sandwich biosensor assay using anti-toxoid A antibody for capture and liposomes at 1:1,000 as the detection reagent for the detection of botulinal toxoid type A ………………………………………………100 Table 21 Biosensor assay with liposome concentrations of 1:2,500 and 1:5,000 for the detection of botulinum toxoid type A…………………. 101 Table 22 Sandwich biosensor assay using anti-toxoid A antibody for capture and liposomes at 1:10,000 and 1:20,000 as the detection reagent for the detection of botulinal toxoid type A ……………………103 Table 23 Sandwich biosensor assay with varying liposome concentrations for the detection of botulinum toxoid type A……………………………104

PAGE 7

vi Table 24 Biosensor assay with liposome concentration at 1:1,000 for the detection of botulinum toxoid type B………………………………….. 106 Table 25 Biosensor assay with liposome concentrations of 1:10,000 and 1:20,000 for the detection of botulinum toxoid type B………………… 107

PAGE 8

vii LIST OF FIGURES Figure 1 Mode of action of cholera toxin………………………………………… 8 Figure 2 ToxR regulon…………………………………………………………… 11 Figure 3 Abiotic and biotic factors which influence Vibrio cholerae growth and transmission ………… ...................................... ..................................16 Figure 4 Copepod with egg sac…………………………………………………… 17 Figure 5 Procedure for recovery of Vibrio cholerae from fecal specimens……… 20 Figure 6a Direct detection with sandwich immunoassay………………………….. 31 Figure 6b Indirect immunoassay detection with direct adsorption………………… 32 Figure 7 General principles of a biosensor……………………………………….. 37 Figure 8 Total internal reflection in optical fiber…………………………………. 40 Figure 9 Liposome construct…………………………………………………….. 47 Figure 10 Map of Tampa Bay, Florida……………………………………………. 50 Figure 11 Tank set-up…………………………………………………………….. 55 Figure 12 Analyte 2000 evanescent wave, fiber-optic biosensor………………….. 57 Figure 13 Assembly of Analyte 2000 cuvette and polystyrene waveguide used in biosensor assays………………………………………………… 58 Figure 14 Illustration of sandwich immunoassay utilized with Analyte 2000…………………………………………………………………….. 59 Figure 15 SDS-PAGE of CT and CT-B……………………………………………. 63

PAGE 9

viii Figure 16 Western blot of CT and CT-B…………………………………………... 63 Figure 17 Sandwich ELISA with GM1 as capture molecule and anti-CT-B antibody as detection molecule for detection of CT-B and V. cholerae serotypes and anti-CT antibody……………………………. 65 Figure 18 Mean value for ELISA CT-B standard curve …………………………... 66 Figure 19 Analyte 2000 normalized standard curve of CT-B ……………………... 67 Figure 20 Comparison of anti-CTantibody ( ) versus GM1 gangliosidebased ( ) capture assay using the Analyte 2000………………………… 69 Figure 21 Comparison of biosensor assay using Cy5-anti-CT-B antibody ( ) or a mixture of anti-CT-B antibody and Cy5-anti-rabbit IgG ( ) as the detection reagent …………………..…………………….. 70 Figure 22 ELISA values for CT standard curve assay …………………………..... 72 Figure 23 Standard curve for CT using Analyte 2000…………………………….. 72 Figure 24 Toxin induction during V. cholerae growth in four types of media ……………………………………………………………………. 76 Figure 25 Western blot of four unspiked oyster samples.…………………………. 79 Figure 26 Western blot of oyster homogenate utilizing goat anti-rabbit IgG conjugated to alkaline phosphatase only ………………………………... 80 Figure 27 ELISA values for fractions of oyster sample adsorbed directly to ELISA wells …………………………………………………………… 81 Figure 28 ELISA values for GM1 based assay……………………………………. 82 Figure 29 ELISA values for oyster homogenate fractions ………………………... 83

PAGE 10

ix Figure 30 ELISA values for samples from tank inoculation with Instant Ocean …………………………………………………………………… 86 Figure 31 PCR of isolates using its primers ………………………………………. 87 Figure 32 PCR of isolates using ctxA primers …………………………………….. 88 Figure 33 ELISA for the detection of CT after V. cholerae inoculation of tank containing SWC-DI ……………………………………………….. 90 Figure 34 ELISA detecting presence of CT from tank containing CAYE-IO with 0.2% glucose, inoculated with V. cholerae ...................................... 92 Figure 35 Detection of botulinum toxoids adsorbed to plate using anti-botulinum toxin antibodies ………………………………………….. 94 Figure 36 Detection of toxoids A and B using sandwich ELISA with GT1b capture ………………………………………………………………….. 95 Figure 37 Detection of botulinum toxoids A or B using GD1b capture in ELISA ………………………………………………………………….. 97

PAGE 11

x Development of Ganglioside-Based Assays for the Identification of Botulinum and Cholera Toxins Utilizing an Evanescent Wave Biosensor Crystal M. Bedenbaugh ABSTRACT An evanescent wave fiber-optic biosensor was used in an effort to develop an assay for the rapid detection of two biological toxins: cholera toxin and botulinum toxin. The Analyte 2000 fiber-optic biosensor utilizes a sandwich immunoassay format. Gangliosides or liposomes are directly adsorbed to the surface of the fiber-optic waveguide through hydrophobic interactions. The waveguide is exposed to a sample containing the toxin of interest, then subsequently exposed to a polyclonal detection antibody conjugated to the fluorophore cyanine 5. Excitation light from a 635nm laser diode is propagated through the waveguide and fluorescent molecules within approximately 100nm of the waveguide are excited. The emission light from the excited cyanine 5 molecules reverberates into the waveguide and is quantitated in pico Amperes and displayed on a computer. The exotoxins of Vibrio cholerae and Clostridium botulinum cholera and botulinum toxin, respectively, were used for potential assay development. Assay development utilizing the biosensor was attempted for the detection of botulinum toxin in buffer. The limit of detection remained too high to generate a positive signal for the detection of botulinum toxin. Biosensor assays were developed to detect cholera toxin in buffer, oyster homogenate, pure culture and induction media.

PAGE 12

xi A cholera toxin standard curve was generated with a limit of detection of 1 ng/ml. The values were normalized by setting 100 ng/ml of cholera toxin to a value of 100. Signals were detected in oyster homogenate spiked at 5 g/ml as well as unspiked oyster homogenate. A Western blot showed that there were cross reactive proteins in the oyster matrix at molecular weights different from those of the cholera toxin. Cholera toxin production by three strains of Vibrio cholerae with values estimated to range from 100 pg – 100 ng was detected with the biosensor. Additionally, oysters were harvested from Tampa Bay and placed in a 10 gallon tank filled with different types of induction media. The tank was inoculated with Vibrio cholerae and the oysters and induction medium were analyzed at varying times for the presence of cholera toxin. Vibrio cholerae cells were viable through 24 hours but no toxin was detectable.

PAGE 13

1 INTRODUCTION Vibrio cholerae Background Vibrio cholerae is a facultative anaerobic Gram-negative curved bacillus that is motile by a single flagellum. Vibrio cholerae is autochthonous in brackish waters and marine environments. Vibrio cholerae is the most clinically significant member of the genus. It is the etiological agent of cholera, an acute bacterial enteric disease. The organism is ingested with contaminated food or water and, due to the potentially destructive acidity in the stomach, requires a microbial load of 108 to 1011 for infection to occur (20). The Vibrio cells will colonize the intestinal tract but they do not penetrate the intestinal epithelium (20). The pathogenesis of V. cholerae is related to the production of an enterotoxin, the cholera toxin (CT), which inhibits fluid absorption. The majority of cholera cases are mild; however, the severest form of the disease is characterized by rapid loss of body fluids leading to dehydration and shock (20). Initial symptoms (watery stools, nausea and vomiting) begin to appear 12 to 72 hours after ingestion of the organism (20). Historical records indicate that cholera may have first appeared in India around 500 B.C. (24). Since that time, seven pandemics of Vibrio cholerae O1 have occurred. The most recent pandemic began in Sulawesi, Indonesia in 1961 (24). The emergence of a new serotype, Vibrio cholerae O139, occurred in October of 1992 in India and

PAGE 14

2 Bangladesh. Vibrio cholerae O139 has spread from South Asia to Near-East and Southeast Asia and has the potential to begin the eighth pandemic (167). Historical Background Cholera has been a devastating disease for centuries. Robert Koch was initially credited for isolating and discovering the vibrio bacillus in May of 1884 (95). Filippo Pacini studied the organism for 20 years but his research was ignored. Pacini was posthumously credited for being the first to isolate the bacillus, V. cholerae in 1854 (95). In 1965 the International Committee on Nomenclature adopted the formal name Vibrio cholerae Pacini 1854 in honor of Filippo Pacini and his discovery (95). Cholera was originally endemic to the Indian subcontinent with the Ganges river serving as a reservoir (24, 25, 26). Between 1817 and 1961 the world witnessed six pandemics with 1961 marking the beginning of the seventh pandemic, which is still ongoing (24). The first pandemic began in Bengal in 1817, spread across India and extended as far as China and the Caspian Sea before ending in 1823 (24). The second pandemic (1829-1851) is believed to have begun in Russia and then reached Europe and the Americas (24). London, Paris, Quebec, Ontario and New York were all affected by the second cholera pandemic (24). An outbreak in 1849 in New York claimed the life of the U.S. President James K. Polk on June 15th (45). During the second pandemic in 1854, a second outbreak occurred in London. Dr. John Snow lived in London at the time. In 1849 he had proposed that cholera was associated with contaminated water and food, but was unable to prove his hypothesis (158). With the emergence of the second outbreak, Dr. Snow demonstrated that the incidence of cholera was associated with contaminated food and water sources. Dr.

PAGE 15

3 Snow conducted one of the first epidemiological studies and plotted the locations where cholera related deaths occurred throughout the city. He is credited with relating the spread of the disease to the mixing of drinking water and sewage in Broad Street (150, 158). Russia and Europe were primarily affected by the third, fourth, fifth and sixth cholera pandemics with over 1 million deaths occurring in Russia (24). Advances in protecting the public water supply, chlorination, water mains and the addition of sewer systems, led to the reduction in the number of cholera cases presented after the sixth pandemic. The seventh pandemic began in Indonesia in 1961 and was associated with a new biotype – the El Tor biotype (24). The classical biotype was responsible for the fifth and sixth pandemics and may have been connected with earlier pandemics. There is no definitive evidence available to prove that the classical strain caused the first through fourth cholera pandemics but it has been associated with their occurrence. In 1993, a new serogroup of cholera, Vibrio cholerae O139, was identified after a large outbreak of cholera-like disease in Bangladesh and India (130). The strain of V. cholerae responsible for the outbreak could not agglutinate in O1 specific antiserum. It was determined that the organism belonged to a new serogroup (130). In 1994, 94 countries reported incidence of cholera to the World Health Organization (WHO) with over 10,000 fatalities (64). Twenty-one countries in Latin America reported over 1 million cases with almost 12,000 deaths at the end of 1996 (183). By the end of 2004, 56 countries officially reported cases to the WHO (183, 185). Of the 101,383 cases of cholera reported during that year, there were only 2,345 deaths (2.3%) (185). Africa

PAGE 16

4 reported a total of 95,560 cases in 2004 representing 94% of the global cases of cholera (185). The number of cholera cases reported to the WHO may be inaccurate with individuals globally failing to identify cases and officially reporting those confirmed as cholera. Classification. Vibrio cholerae is a waterborne bacterium measuring 0.3 microns in diameter and 1.3 microns in length. Different strains of Vibrio cholerae can be identified by agglutination with O-specific antiserum. The O antigen is a cell wall associated component of the lipidpolysaccharide and is used to distinguish between strains of vibrio. Of the more than 193 currently recognized strains of V. cholerae V. cholerae O1 and V. cholerae O139 are the only serotypes pathogenic to humans (18). The flagellar H antigen is common to many vibrios and all O groups and, therefore, is not useful in distinguishing one strain of V. cholerae from another (40). The classification scheme for epidemic and nonepidemic strains of Vibrio cholerae is listed in Table 1. TABLE 1. Classification of epidemic and non-epidemic associated Vibrio cholerae (78) Classification Serogroup Biotype Serotype CT production Epidemic O1 Classical and El Inaba, Ogawa Yes associated Tor and Hikojima O139 Not applicable Inaba, Ogawa Yes and Hikojima Not epidemic All other Not applicable Not applicable Usually not associated strains

PAGE 17

5 The O1 serogroup can be subdivided based on antigenic differences in the B and C components of the O antigen (147). The A antigen is common to all three serotypes and it is the B or C specific antigen that determines the serotype (147). The antigenic combinations give rise to the three serotypes: Ogawa, Inaba and Hikojima (Table 2). V. cholerae O1 strains are able to convert or switch between the three serotypes both in vivo and in vitro (147). The O139 strain has a unique O antigen and cannot be classified into one of the three serotypes. TABLE 2. Antigenic determinants of Vibrio cholerae O1 and O139 serotypes (147). Serotype Major O-antigenic determinant Ogawa A, B Inaba A, C Hikojima A, B, C The O1 serogroup is divided into two biotypes, classical and El Tor, on the basis of phenotypic characteristics. Biochemical tests differentiate the two biotypes (Table 3). Both biotypes cause cholera and their differentiation is useful for epidemiological studies. TABLE 3. Biochemical tests used to differentiation between classical and El Tor biotypes of Vibrio cholerae serogroup O1 (78). Biotype VP test Zone around Agglutination (modified polymyxin B of chicken, with 1% (50 U) goat or sheep erythrocytes Classical + El Tor + +

PAGE 18

6 Transmission. Cholera transmission occurs via the fecal-oral route through the ingestion of cells in contaminated food or water. Poor, lacking or non-existant sanitation systems in impoverished areas allow for contamination of water systems or surface waters. Use of contaminated water for drinking or in food preparation spreads the disease. Raw vegetables (109, 122, 183), molluscan shellfish (40, 109, 122, 123), crustaceans (40, 109, 122) and rice (109) have been implicated in foodborne transmission of the disease. Filterfeeding shellfish, such as oysters, clams and mussels, will concentrate V. cholerae from water into its tissues. These types of shellfish are often consumed raw, which poses a higher risk of causing illness (40, 77, 98, 122). The first documented case of cholera in the United States attributed to consumption of raw oysters occurred in 1973 in Texas (180). Since 1973, 65 cases of cholera caused by V. cholerae O1 have occurred through 1991 in the United States; most cases were related to the consumption of contaminated shellfish (12). Sixty-eight cases of cholera were reported to the CDC between 19952003 and 9% were acquired through the consumption of seafood harvested from the Gulf Coast (15). After the hurricanes Katrina and Rita struck Louisiana in October of 2005, 2 cases of cholera were reported (16). Cholera is rarely seen in the United States and there is a low incidence associated with shellfish consumption (11, 15, 16). Cholera toxin. The pathogenesis of cholera is associated with the production of an exotoxin, the cholera toxin (CT) (192). Bacterial exotoxins are produced by a variety of bacteria and are a type of virulence mechanism used by microorganism to assist in host

PAGE 19

7 invasion and establishment (148). Exotoxins are soluble, heat-labile proteins that can be toxic in small quantities (93). Exotoxins may be classified into one of three categories based on structure and mechanism of action: (1) A-B type, (2) membrane disrupting type and (3) superantigen (148). CT is a heterohexameric AB5 enterotoxin. A-B toxins are composed of two subunits. The A subunit is responsible for the enzymatic activity of the toxin, while the B subunit binds specifically to a host cell surface molecule (148). A-B toxins are classified as simple or compound (148). Simple A-B toxins, e.g., botulinum toxin, are produced as a single chain and are protealytically cleaved to produce one A subunit and one B subunit (148). Compound A-B toxins, such as CT, are composed of multiple B subunits joined to the A subunit by a disulfide bond (148). The B subunit of an A-B toxin binds to a glycolipid or a protein on the host cell surface (148). Following binding of the B subunit, the A subunit is cleaved and gains entry into the host cell (148). CT is an 84 kD protein composed of one A subunit (27 kD) and five identical B subunits (11.6 kD each) (103, 191, 192). The B subunits of the CT bind a glycolipid cellular receptor, ganglioside GM1, on intestinal epithelial cells (33, 60, 143, 145). The A subunit enters the cytoplasm of the target cell and transfers an ADP-ribose from NAD+ to the subunit of the Gs protein. The ADP-ribosylated subunit subsequently activates adenylate cyclase which leads to an increase in the intracellular concentration of cAMP (Figure 1) (20, 74, 93). The increase in cAMP concentration causes uncontrolled fluid and electrolyte flow into the intestine.

PAGE 20

8 FIGURE 1. Mode of action of cholera toxin. A,B cholera toxin subunits; GM1 ganglioside receptor; Gs protein; AC adenylate cyclase; cAMP cyclic AMP (74) The symptoms of the disease are severe diarrhea and dehydration. Of those infected, approximately 75% are asymptomatic and 25% present symptoms (183). Of those 25% with illness, 5% have moderate illness requiring medical attention and 2% face a potentially life-threatening disease (183). Among those infected, initial symptoms begin to appear 12 to 72 hours after ingestion of the organism (20). The continued secretion of watery feces, also called secretory diarrhea or rice-water stool, can cause an infected individual to lose up to 200 ml per kg of body weight per day (20). Secretory diarrhea may cause death within three hours if left untreated (143). Treatment of the disease consists mainly of oral rehydration with electrolyte replacement (12). Antibiotics can be administered to shorten the duration of the illness. Epithelial cell GM1 A B Gs AC cAMP ATP A N a+, H2O Vibrio cholerae Cholera toxi n A B

PAGE 21

9 The LD50 in mice for cholera toxin is 250 g/kg, which when extrapolated out would equal 313 mg/100 lb (53). Consequently, a high concentration of CT is required to cause illness in humans. Studies with American volunteers demonstrated that oral administration of up to 1011 viable virulent V. cholerae led to virtually no signs or symptoms of cholera. Furthermore, cells could rarely be recovered from feces (20, 49). However, if stomach acid was neutralized with sodium bicarbonate, then 103 viable organisms could to cause the disease (49). Another study of American volunteers revealed that an oral dose of 2.5 g of purified CT caused no illness, whereas a 5 g dose resulted in 1-6 L of diarrhea in 80% of volunteers and a 25 g dose resulted in 22 L of diarrhea in 100% of volunteers tested (88). The pathogenesis of cholera is almost exclusively dependant on cholera toxin production with 95% of O1 and O139 strains possessing CT producing capabilities (18, 20, 50). The two most important virulence factors of V. cholerae are the presence or production of CT and the toxin–coregulated pilus (TCP) (18, 20, 34, 58, 120, 179). The ctxAB genes-the genes encoding cholera toxinare located on the CTX element, which is part of the genome of a filamentous bacteriophage, CTX (18, 20, 34, 58, 120, 149, 179). The receptor for this phage is TCP, which it uses to infect V. cholerae cells (18, 20, 34, 58, 120, 179). Once a cell is infected, the CTX can become a lysogen or replicate extrachromosomally as a plasmid (34). The tcp gene-the gene encoding the toxincoregulated pilusis part of the V. cholerae pathogenicity island (VPI), (18, 20, 34, 58, 120, 149, 179). Research indicates that the VPI is of bacteriophage origin as well (18, 20, 34, 58, 120, 149, 179). The current hypothesis is that nontoxigenic strains of V. cholerae acquire the VPI through horizontal gene transfer, thus providing nonvirulent

PAGE 22

10 strains with the ability to acquire the CTX and toxin producing capabilities (18, 20, 34, 58, 120, 179). The acquisition of the ctxAB and tcp genes makes it possible for the evolution of pathogenic strains from previously nonpathogenic strains of V. cholerae (18, 20, 34, 58, 120, 179). Iron. Most bacterial organisms require iron to synthesize cytochromes and need to sequester iron from the environment. In natural waters, V. cholerae must utilize a complex system for iron acquisition and in vivo the organism is forced to compete against animal hosts for iron. Humans main utilization of iron is in the proteins hemoglobin and lactoferrin. Iron can be a limiting factor in microbial growth. In order to chelate iron from host and environmental water systems, V. cholerae utilizes a variety of ironchelating siderophores to take up insoluble iron (49, 79, 100). It is still questionable whether siderophores are required for increased expression of CT. ToxR regulon. The regulation of the ctx and tcp genes are controlled by the ToxR regulon (Figure 2) (34, 58, 131, 149, 179). Expression of these genes is transcriptionally regulated by a cascade system of regulatory proteins: ToxR, ToxS and ToxT (58, 149, 179). In 1987, M iller et al. ( 107) presented the currently accepted model that indicates that the ToxR regulon senses environmental signals and undergoes a conformational change that activates transcription of the other genes ctxAB, tcpA and toxT in the regulon (20, 58, 131, 149). ToxR is stabilized through interactions with the ToxS protein and activates transcription of the ctxAB operon and toxT gene (20, 58, 131, 149). ToxT activates the transcription of tcp and ctxAB (20, 58, 131, 149). External environmental conditions that influence virulence include temperature, osmolarity, medium composition, pH and oxygen tension (20, 58, 131, 149, 179). It is not clearly understood

PAGE 23

11 how the environmental signals and sensing occurs, but studies indicate that the ToxR regulon is influenced by environmental factors including the unknown in vivo conditions of the human host (20, 131). Further analysis is needed with respect to the role of environmental signals and their influence on the CT virulence genes and their expression under the control of the ToxR regulon (20, 58, 107, 131). FIGURE 2. ToxR regulon. Arrows indicate transcription activation (148). Disease control. Cholera is strictly a human disease and is not pathogenic to animals (5). The WHO’s current recommendations for disease control focuses on preventing fecal contamination of water, implementing personal hygiene practices and managing cases with oral and intravenous rehydration therapy (184). An effective sanitation system is paramount for the control of cholera transmission. In underdeveloped countries where sanitation systems are not at the forefront of public ToxR ctxABtoxT A + B ToxT tcpA

PAGE 24

12 health concerns, cholera becomes epidemic. Transmission is normally associated with contaminated water, however in areas where there is a lack of clean water the disease can also be spread through households via contaminated foods. A parenteral cholera vaccine can be administered to those living in or traveling to an area where cholera is endemic (22). However, the vaccine is not very effective, provides limited immunity and is short-lived – approximately 6 months (22). An orallyadministered inactivated cholera vaccine has been field tested in Bangladesh and offers immunity substantially longer than the parenteral vaccine (22). However, the CDC does not currently recommend the use of a vaccine and suggests those traveling to areas with prevalent cholera cases to seek the advice of their doctor (12). In confirmed, severe cases of cholera tetracycline, doxycycline, furazolide, trimethoprim-sulfamethoxazole, erythromycin and chloramphenicol therapy can be administered to shorten the length of infection (183, 184). Oral rehydration therapy of water, sugars and salts is normally recommended by the WHO (183, 184). With prompt and adequate care, fatalities due to cholera can be minimized to less than 1% (183). Ecology and Marine Adaptations Vibrio cholerae is a ubiquitous aquatic organism of brackish and marine environments. It was believed that humans were the natural reservoir of V. cholerae ; however, the ocean may actually be the reservoir for the organism (25, 26). V. cholerae is found in association with phytoplankton and zooplankton (95). A set of complex ecological interactions exists between cholera, phytoplankton and zooplankton that are required for the growth, proliferation, expression of virulence and transmission of V. cholerae in the ocean environment (95).

PAGE 25

13 Survival in the environment. The ability of the organism to survive in food and water is dependent upon several variables. Growth of V. cholerae can be suppressed by the presence of other microflora (77). Other organisms can out compete V. cholerae for nutrients and required growth factors therefore causing limited or no growth. There are various abiotic factors that affect the survival of the organism, including temperature, salinity, sunlight, pH and other microflora (Table 4). The most important physical factor that has a direct effect on the ecology and growth of V. cholerae is temperature (95). During the warmer months, higher levels of V. cholerae occur in the water column, which makes it is easier to isolate the organism (95). Changes in salinity, water temperatures and possible seroconversion contribute to unfavorable growth conditions (95). Vibrio cholerae grows optimally within a salinity range of 2-14 g per L (26). The ability to tolerate fluctuations in salinity allows V. cholerae to survive and grow in brackish and marine waters.

PAGE 26

14 TABLE 4. Factors affecting the survival and growth of V. cholerae in the environment (137). Factor Effect Temperature Growth of V. cholerae above 10C Salinity Growth of V. cholerae best survival at 0.25-3 %; optimum at 2% Sunlight (UV light) Reduces survival pH Optimum 7.0-8.5 Other microflora Survival suppressed by competitors The role of climate change in the survival and growth of V. cholerae is still under investigation. Colwell used remote sensing imagery systems to study the seasonal fluctuations in V. cholerae growth in marine waters (25). Studies in Bangladesh indicated that cholera outbreaks occurred after sea surface temperatures and heights peaked (25). Temperature models indicate an increase in global temperatures may occur in the next 100 years. Without efficient public health measures an increase in global temperature could lead to an increase in the range and prevalence of V. cholerae (95). Lipp et al proposed that CTX induction and propagation could increase with increased UV intensity (95). It is also important to take into account natural weather patterns; however, it is difficult to predict the impact systems such as the El Nino Southern Oscillation (ENSO) have on V. cholerae growth (95). When environmental conditions are unfavorable, Vibrio cholerae enters into a viable but nonculturable (VBNC) state with altered cell morphology (7, 27, 31, 73, 187). Not only do cells retain viability, but the genes responsible for pathogenicity and

PAGE 27

15 virulence remain intact and can be expressed as well (17, 28). VBNC cells are detectable by molecular methods or fluorescent microscopy when conventional laboratory methods fail (8, 54, 66). During nutrient deprivation, elevated salinity and/or reduced temperature, the organism enters into a state of dormancy. The shape normally associated with V. cholerae, a bacillus, becomes smaller with an ovoid or coccoid morphology (73). The VBNC state in Gram-negative organisms has been equated to the spore state in Gram-positive cells (141). Association with marine organisms. Cholera has been found in association with a variety of aquatic organisms. It has been suggested that there is a complex relationship between the growth of phytoplankton, algal plants, zooplankton and the growth of V. cholerae in an aquatic environment (Figure 3).

PAGE 28

16 FIGURE 3. Abiotic and biotic factors which influence Vibrio cholerae growth and transmission (95). Vibrio cholerae has been found in association with a variety of marine organisms including cyanobacteria (69, 95), diatoms (101), green algae (69), oysters (77, 98) and copepods (65, 67, 95). The abiotic factors listed in Figure 6, temperature, pH, Fe3+, sunlight and salinity, directly affect the growth of plankton and aquatic plants. (95). Plankton growth is stimulated by temperature, nutrients and sunlight. Proliferation of Abiotic conditions (Temperatur e pH, Fe3+, salinity, sunlight) favor growth of V. cholerae and/or plankton Aquatic plants and algae grow, allowing growth of plankton which helps to promote growth and survival of V. cholerae Zooplankton, copepods an d other crustaceans feed off of plankton and aquatic plants. V. cholerae grows in association with copepods and proliferate as commensals with the zooplankton V. cholerae is transmitted to humans with the ingestion of water contaminated with V. cholerae or V. cholerae attached to zooplankton Other factors which influence survival, proliferation and transmission of V. cholerae: Changing weather patterns Sanitation systems Endemic V. cholerae

PAGE 29

17 algae and aquatic plants increase the survival and proliferation of cholera cells and provides a food source for zooplankton communities. Research indicates that an association with green algae increases CT expression in V. cholerae (69). Zooplankton feeds off of the plankton; therefore, an increase in plankton causes an increase in zooplankton. Vibrio cholerae can colonize different substrates due the production of multiple enzymes (31). In environmental waters, V. cholerae produces the enzyme chitinase which assists in its growth on chitin surfaces such as those associated with copepods (26, 31, 46, 127). V. cholerae will colonize copepods (Figure 4) at the oral region and the egg sac (67). In the human host, mucinases are produced which enable the organism to penetrate the mucous barrier covering the gastrointestinal epithelium (31, 127). V. cholerae will multiply until bacterial cells cover the entire egg sac. V. cholerae feeds FIGURE 4. Copepod with egg sac (144). Egg sac ~ 1 mm

PAGE 30

18 off of the chitin substrate of the egg sac and persists on the surface of the copepod for extended periods of time. As the eggs reach maturity V. cholerae synthesizes chitinase (113, 114). The chitinase dissolves the egg sac causing the copepod eggs and V. cholerae to be released into the aquatic environment. V. cholerae can also colonize the gut of copepods. It has been proposed by Singleton et al. that during its association of the copepod CT production by V. cholerae serves an environmental role in osmoregulation of the copepod (157). While the bacteria are attached to the copepod, they are provided with a large surface area, substrate, nutrients, protection from low temperatures and acidic conditions, and a means of transport through the environmental water system throughout the life cycle of the copepod (26). There is a direct correlation with the number of copepods and the number of V. cholerae in the environment. Large populations of copepods will aid in the proliferation of large numbers of V. cholerae With high numbers of V. cholerae in the water, filter feeding organisms, e.g., oysters can and will concentrate the organisms (98, 122, 164). In impoverished areas where cholera is endemic and sanitation conditions are poor, there is a direct relationship between zooplankton populations and cholera incidence (95). In areas of endemnicity, the V. cholerae and the copepods remain in the water due to weather and circulation patterns. In areas of endemnicity such as Bangladesh, due to seasonal weather patterns of this region, seasonal increases in zooplankton are followed by increased numbers of cholera cases (95). Individuals in this area drink contaminated water containing the copepods harboring the V. cholerae cells. Huq et al demonstrated that filtering copepod contaminated water through a sari or nylon

PAGE 31

19 cloth significantly reduced the number of cholera cases by approximately 52% (65). In areas such as Japan where there are adequate sanitation facilities and cases of cholera appear sporadically, there is no direct correlation between increases in zooplankton and cholera cases (176). The ecology of cholera is related directly to human disease. One of the initial problems with understanding epidemics of cholera was identifying V. cholerae’s natural reservoir. Researchers were initially troubled because V. cholerae could not be cultured from the environment between periods of cholera outbreak. With the discovery of the VBNC state, scientists began to understand the cholera paradigm. Using remote sensing technology, Colwell has studied the ecological relationships between V. cholerae and marine organisms (25). Filter-feeding organisms in water with high concentration of cholera cells concentrate the organisms in their tissues. Humans harvest and often eat these filter-feeding organisms raw thereby leading to illness. CT production has been shown to occur in association with algae, in the gut of the copepod, or in the intestine of the human host. Production of CT in each relationship ultimately benefits V. cholerae : allowing its survival, growth and release into the environment. Detection of Vibrio cholerae and Cholera Toxin Vibrio cholerae The most common laboratory procedure for recovery of V. cholerae from environmental samples and stool cultures involves conventional laboratory techniques (Figure 5) (41, 78). Optimum growth occurs at 37C and the organism can survive at ambient temperatures for approximately 5 days (11, 137). The organism grows quickly at a pH of 7.0-8.5 but can survive alkaline conditions in the range of 6.0-10.0 (137). Many vibrios collectively all species of Vibrio are halophilic and require salt or

PAGE 32

20 are stimulated by the addition of it to growth medium (132). Cells are typically enriched in alkaline peptone water (APW, pH 8.5) for 6-8 hours at 37C or plated directly onto thiosulfate citrate bile sucrose agar (TCBS). TCBS, a selective medium used for the isolation of vibrios, offers several benefits: (1) commercial availability, (2) does not require autoclaving, (3) is highly selective for Vibrio spp. and (4) contains sucrose for the differentiation of sucrose fermenters ( V. cholerae ) from nonfermeters (78). After isolation on TCBS, isolates are streaked onto a nonselective medium, such as tryptic soy agar (TSA), followed by agglutination reactions with specific antiserum and other biochemical tests. FIGURE 5. Procedure for recovery of Vibrio cholerae from fecal specimens (78). Optional Screening Tests String Oxidase KIA or TSI Arginine or Lysine Direct Enrichment TCBS Nonselective medium V. Cholera e O1 Polyvalent antiserum Ogawa and Inaba Anitsera

PAGE 33

21 The most probable number technique (MPN) is often used in association with biochemical tests to estimate population numbers in complex matrices, such as soil and food (122). Cells are enriched in APW, plated on TCBS and estimated population sizes determined from a standard MPN table. Conventional laboratory techniques are inadequate for detection of vibrios in the VBNC state. Molecular methods and fluorescent microscopy are used to detect VBNC cells (8, 66, 96). Rapid detection methods such as PCR are commonly used for the detection of viable and VBNC V. cholerae cells (21). Nucleic acid-based methods can be nonquantitative (42, 48, 64, 75, 80, 152) and require visualization of PCR products through electrophoresis (98). Alternative PCR methods such as TaqMan PCR allow for near real-time detection and are quantitative (98). The inherent weaknesses with this molecular method includes the inability to determine culturable cells from VBNC cells, PCR inhibition and the ability to only detect four genes in a single reaction (98, 122). DNA microarrays (182) and multiplex PCR (136) are also used to identify V. cholerae Detection of cells does not give an indication of virulence. The Y-1 mouse adrenal cell assay, ELISAs, rabbit ileal loop assay and the commercially available immunoassay (VET-RPLA, Oxoid, Inc. Ogdensburg, NY) are used to test for enterotoxigenicity of V. cholerae (41, 108, 112, 126). The Y-1 mouse adrenal assay is a tissue culture based method which relies on morphological changes in cell shape of the mouse cells for determining the presence of CT (99). It is important to note that cholera toxin production is different in vivo versus in vitro (112). A strain of V. cholerae that may produce toxin in the human small intestine or the mouse adrenal cells may not do so under a different set of conditions (112). Unfortunately, the perfect assay does not exist

PAGE 34

22 with an ideal set of cultural conditions that allows different strains of V. cholerae to maximize CT production (112). ELISAs with different biological recognition elements – antibodies and gangliosidesare commonly used to detect CT (60, 61, 145). The limit of detection (LOD) utilizing a strictly antibody based technique is 0.09 g/ml (61). Results using gangliosides as a capture molecule and antibodies as detection molecules have produced limits of detection in the range of 40 ng/ml to 1 g/ml in sandwich immunoassays using an array biosensor with antibodies directed against rabbit and goat respectively (143). Dawson reported a limit of detection of 100 ng/ml when using GM1 as a capture molecule in a sandwich ELISA format (35). The rabbit ileal loop assay is a time-consuming and cumbersome biological assay (36, 76). Typically, adult rabbits are starved for a period of 24 hours, then while still alive a loop of bowel is ligated and injected with live V. cholerae At autopsy 24 hours later, the ileal loop is measured in length and volume for the production of CT. CT The commercially available reverse passive latex agglutination assay is simple to use and has a limit of detection of 1-2 ng/ml of CT (112). The assay requires induction of CT production in an enrichment broth followed by subsequent detection of the toxin. If the goal is to rapidly detect CT, then the method employed should detect CT and not the cells. Very few assays directly assay for CT in a sample matrix in lieu of V. cholerae cells. One assay, the bead-ELISA can directly detect CT in stool samples at a limit of detection between 26 pg/ml and > 100 ng/ml (129). The beads are the solid

PAGE 35

23 phase component of the assay and are coated with an anti-CT antibody. The assay is performed in a test tube in a sandwich immunoassay format. The methods described previously for the direct detection of cholera toxin are especially valuable because it is the toxin that is responsible for the pathogenicity of the organism. However, conventional methods used to enrich for cells and then assay for CT can be time consuming and can give inaccurate results with respect to toxigenicity due to the absence of expression of necessary genes (112). When the cells are taken from the environment and then enriched and grown in the laboratory under ideal conditions, specific genes can be turned on, off, rearranged or lost in the process (126). A rapid, sensitive and specific assay is necessary for the direct detection of CT. Clostridium botulinum Background Clostridium botulinum spores are ubiquitous in soils and aquatic sediments worldwide (159). The bacterium is a Gram-positive obligate-anaerobic sporulating rod that measures 0.5-2.0 m in width by 1.6-22.0 m in length. Clostridium botulinum is the etiological agent of food-borne botulism, an intoxication caused by the ingestion of a potent exotoxin, botulinum toxin (148). Botulinum toxin is a neurotoxin and is classified into seven different serotypes, A-G, with types A, B, E and F causing the most serious disease in humans: botulism (68, 84, 159, 163, 181, 190). Botulism is a paralytic illness caused by the action of the neurotoxin at the vertebrate neuromuscular junction (23). The neurotoxin blocks the release of the neurotransmitter, acetylcholine, at the peripheral cholinergic nerve endings leading to respiratory and musculoskeletal paralysis (23, 102).

PAGE 36

24 There are three main kinds of botulism: (1) foodborne botulism caused by the ingestion of preformed toxin; (2) wound botulism caused by toxin produced from a wound infected with C. botulinum ; and (3) infant botulism caused by consuming the spores of the bacterium which sporulate in the intestines and release toxin (68, 84, 102, 124, 140, 159, 163). Virulence factors. The neurotoxin produced by C. botulinum is the most potent biological toxin known with a minimum lethal dose in monkeys of 0.5 ng when administered orally (53), 0.00625 ng in mice when administered intraperitoneal (59) and an estimated LD50 in humans of 1 ng/kg for type A toxin (53, 63). All seven neurotoxins are similar in their mode of action and are simple A-B toxins. The toxin is produced as a single protein chain and then protealytically cleaved to produce one A subunit and one B subunit (148). The complete toxin has a molecular weight of approximately 150 kD and is composed of one heavy chain (B subunit) and one light chain (A subunit) linked by a disulfide bond (84). The active toxin heavy chain (approximately 100 kD) is responsible for the recognition of specific neuron receptors and mediates the internalization of the light chain into the cytosol (68, 124, 140, 163). The light chain (approximately 50 kD) is responsible for the intracellular activity; blocking the release of acetylcholine at the neuromuscular junction leading to flaccid paralysis (68, 124, 140). The action of the neurotoxin is caused by a series of events. The heavy chain binds a co-receptor complex, the glycoplipid cellular receptor, (GT1b, GD1b, GQ1b or GD1a), and the protein synaptotagmin II (19, 84, 116, 117, 156, 190). After binding, the receptor-ligand complex is internalized via endocytosis, the light chain is translocated

PAGE 37

25 into the cytosol and acts on nerve endings by inhibiting the release of the neurotransmitter, acetylcholine (68, 84). The continued inhibition of acetylcholine release causes flaccid paralysis that may eventually lead to respiratory failure if the paralysis extends to the respiratory muscles (59). Symptoms begin 18-24 hours after ingestion of the toxin and include double vision, blurred vision, slurred speech, difficulty swallowing and muscle weakness (53). The intoxication is characterized by cranial nerve palsies followed by symmetrical descending flaccid paralysis of motor and autonomic nerves (157, 163). Patients remain fully alert until death (53, 159). If detected early enough after the ingestion of the toxin, (usually within the first 24 hours of symptoms appearing), botulism can be treated with trivalent antitoxin (types A, B and E) which halts the progression of paralysis and shortens the length of the illness (148, 159). The antitoxin acts by neutralizing any free toxin in circulation; thereby, inhibiting the toxin from binding to nerve endings (148, 159). In the United States, more than 80% of patients are treated with antitoxin (13). If the presence of the toxin is detected after the neurotoxin has begun to effect nerve endings and cause damage, supporative care is given to the patient (13, 53, 148, 159). Types of botulism. Botulism is commonly acquired through ingestion of toxin in food that has been improperly prepared prior to its consumption (53, 102, 159). Foodborne botulism is rare in the United States and is usually associated with homecanned vegetables, fruits and meat products (13, 148, 159). Spores survive in foods that are incorrectly processed and are only inactivated by heating to 121C under pressure of 15-20 lb/in2 for at least 20 minutes (159). Toxin production only occurs in an anaerobic environment with a low salt concentration and a minimum pH of 4.6 (13, 159). Between

PAGE 38

26 1950 and 1996, the CDC reported 444 cases of foodborne botulism with a case-fatality ratio of 15.5% (13). A total of 263 cases were reported to the CDC between the years 1990-2000 (159). Of those cases, there was a case-fatality ratio of 4% (159). In 2003, a total of 8 cases of foodborne botulism were reported to the CDC with a case-fatality ratio of 25% (15). Infant botulism is the most common form of botulism in the United States (14, 85, 163). Infant botulism was first recognized in the United States in 1976 when the toxin and/or cells were detected in the feces of infants (163). This type of botulism is an infection with spores obtained from contaminated honey (163). The spores temporarily colonize and germinate in the intestine of the child and produce toxin. It is recommended that infants who are still nursing should not be given honey (163). Symptoms include constipation, trouble sucking, swallowing, or crying and progressive muscle weakness (14). The CDC reports the annual incidence of infant botulism is 2 per 100,000 live births (13). Wound botulism results from the colonization of a wound with C. botulinum (133). When the skin is compromised and contact with contaminated soil or gravel occurs, the tissue may become infected with C. botulinum. This form of botulism is very rare and disease results in the production of neurotoxin, which travels through the bloodstream and to the nerves (133). Symptoms are similar to those associated with foodborne botulism without the gastrointestinal involvement: blurred vision, sore throat, trouble swallowing and the eventual culmination in paralysis and death. Wound botulism was first recognized in 1943 and through 1990, only 47 cases were reported to the CDC (59, 133). Since 1990, the majority of wound botulism cases have been associated with

PAGE 39

27 IV drug use (9, 110, 121, 133). Between 1990-2002, 210 of 217 cases reported to the CDC were associated with IV drug use (133). Due to the potency of the neurotoxin, botulism is a major threat as a bioweapon (3). The toxin can be aerosolized and cause the same disease as foodborne botulism through inhalation. One gram of aerosolized toxin has the potential to kill 1.5 million people (3, 121). The 1972 Biological and Toxin Weapons Convention prohibited the offensive development of bioweapons (94); however, Iraq continued to develop their weapons program (3). Iraq disclosed to the UN Security Council after the Gulf War of 1991 that over 19,000 L of concentrated botulinum toxin had been produced (3, 193). Of the 19,000 L, 10,000 L had been loaded into military weapons for use (3). Due to the serious threat botulinum causes as a bioweapon, the CDC continues to monitor any reported cases. Detection of Clostridium botulinum and Botulinum Toxin Botulism is most effectively treated soon after the appearance of symptoms. Rapid diagnosis is important due to the damage exerted on nerve endings. Currently, the mouse bioassay is the gold standard for the detection of botulinum toxin or for differentiation of serotypes (1, 47, 53, 59, 118, 181). Mice are injected with an intraperitoneal dose of botulinum toxin and observed for a period of four days (4, 47, 181). Animals are observed for symptoms of botulism; those injected with higher doses show symptoms within 8 hours while those injected with lower doses can take a few days to develop symptoms (4). The limit of detection for the most active serotype is 10-20 pg/ml (4, 181). The mouse bioassay has several disadvantages: (1) it takes several days for results; (2) many animals are required for the assay; (3) special facilities are required;

PAGE 40

28 (4) there are inherent hazards associated with injecting the mice; (5) it is expensive; and (6) additional testing is required to differentiate neurotoxin serotypes (4, 47, 181). ELISAs have been developed in lieu of the mouse bioassay for the detection of the botulinum toxin (4, 47, 59, 181). The ELISA allows for the screening of a larger sample size with more rapid results without the need for animals. Ferreira et al reported a limit of detection of 10 minimal lethal dose (MLD) per ml (47). The current MLD currently is 0.02 ng for type A toxin (47). Wictome et al reported a limit of detection of 0.5 MLD per ml when utilizing a colorimetric immunoassay for type B toxin (181). Signal amplification was used in conjunction with an ELISA by Sharma et al in several different types of food matrices including liquids, solids and semi-solids and generated extremely sensitive limits of detection for toxin types A, B, E, and F (153). The limits of detection for the four different serogroups were as follows: type A toxin was 60 pg/ml, type B was 176 pg/ml, type E was 163 pg/ml and type F was 117 pg/ml (153). The use of affinity molecules other than antibodies in association with botulinum toxin is also being employed in conjunction with an ELISA. Liposomes containing the natural cellular receptor for botulinum toxin, (GT1b, GD1b or GQ1b), and fluorophorelabeled lipids have been used as detection molecules in a sandwich fluoroimmunoassay (156). Concentrations as low as 1nM of botulinum toxin could be detected with fluorescently labeled liposomes (156). Another receptor immunoassay was developed with GT1b inserted into a liposome (1). In this assay, botulinum toxin was detected as a colored band on a nitrocellulose membrane either visually or with a densitometer (1). The limit of detection for this assay with type A toxin was 15 pg/ml (1). Current ELISA assays are not sensitive enough to replace the mouse bioassay and make take up to 5 days

PAGE 41

29 for the cultures to produce toxin; therefore, they do not produce results in a near-real time fashion (118). There is a need for a rapid and sensitive test and assays utilizing molecular techniques and biosensors are being developed as an alternative to the ELISA and mouse bioassay. PCR can be used to identify toxin genes in 24 h botulinal cultures as type A, B, E or F neurotoxin producers (32, 161). PCR uses nucleic acid probes which, under controlled conditions, hybridize with the complementary nucleic acid sequences of interest. Clostridium botulinum isolates are enriched anaerobically for 24 hours to obtain vegetative cells. A simultaneous PCR reaction for all 4 toxin serotypes can then be performed. The PCR may be used in conjunction with the mouse bioassay as a confirmatory test (32, 161). This type of PCR assay does not produce real-time results but merely shortens the time interval with which a diagnosis could be provided. Realtime PCR assays have been developed for the detection of type A, B and E toxins from purified DNA samples and crude DNA extracted from broths (2). Using real-time PCR, seven out of eight botulinum cases were identified which provided a faster preliminary diagnosis than conventional bioassays (2). A fiber optic-based biosensor assay utilizing a sandwich immunoassay produced a limit of detection of 5 ng/ml for botulinum toxin type A (118). Immobilized antibodies were used as a capture molecule and fluorescently labeled antibodies were used as a detector. The toxin detection occurred in one minute with previously prepared fibers. The assay provided sensitive results with no cross-reactivity to the similar toxin, tetanus toxin.

PAGE 42

30 Immunoassays Background Immunoassays have been in use since the 1960s when Rosalyn Yalow and Solomon Berson used a radioimmunoassay to detect and quantify insulin in plasma samples (188). These assays combine the principles of biochemistry and immunology, enabling scientists to detect very low concentrations of a specific antigen. The principle of the test involves using antibodies as reagents. In an immunoassay, the antibody at a specific concentration either a primary or secondary is attached to a label: an enzyme (enzyme immunoassay; EIA), a radioisotope (radioimmunoassay; RIA) a fluorescent dye (fluorescent immunoassay; FIA) or a luminescent label (luminescence immunoassay; LIA). The conjugation of the antibody to one of these labels provides the means for determining the concentration of an unknown antigen. Depending on the type of label conjugated to the antibody, a color change, emission of light, the amount of radioactivity or some other signal is produced and measured after the immunological binding reaction. Special visualization systems including spectrophotometer, fluorometer and luminometer are required to quantitate the amount of label or end product present. Enzyme immunoassays (EIA) are the most widely used type of immunoassay. A large number of samples and complex antigens such as bacteria can be assayed using EIA (139). Sensitivity, the ability to detect low levels of antigen, and specificity, the ability to differentiate between antigens, are two ways assays can be described. Sensitivity and specificity are both dependent on the antibody antigen interaction. EIAs are easy to perform and involve the basic principle of antibody-antigen interaction: typical interactions are highly sensitive and very specific. Thus, immunoassays are practical

PAGE 43

31 when an unknown concentration of an antigen needs to be quantified. In a sandwich assay, an antigen is sandwiched between two antibodies: a capture antibody immobilized on a solid surface and a detection antibody with a label. In a direct assay, cells or antigen are adsorbed to wells of a microtitre plate. The method of detection with different assay formats can be divided into two categories: direct and indirect. Detection Format The direct sandwich immunoassay (Figure 6a) uses two layers of antibodies to detect antigen. Antigen of interest must have at least two antigenic sites: one to bind capture antibody and one to bind labeled detection antibody. Antibodies are attached to a solid surface to capture antigen. Labeled primary antibody binds directly to the antigen and is detected by the label. If the label is an enzyme, such as peroxidase or alkaline phosphatase, substrate is added and the unknown concentration of the antigen is determined. If the label is a fluorophore, such as Cyanine 5 or rhodamine, under the appropriate wavelength of light, it fluoresces and antigen concentration is quantified. FIGURE 6a. Direct detection with sandwich immunoassay. label Detector antibody antigen Capture antibody label Detector antibody antigen Capture antibody

PAGE 44

32 FIGURE 6b. Indirect immunoassay detection with direct adsorption. Passive adsorption allows ELISA reagents, antibody or antigen, to attach passively to the solid phase of a surface during an incubation step. In an indirect assay, antigen binds to capture antibody followed by the addition of primary antibodies (Figure 6b). A binding event occurs between the primary antibodies that have an affinity for the antigen of interest. The binding of the primary antibody is followed by incubation with labeled secondary antibodies directed against the species in which the primary antibody was made. Upon binding of the labeled secondary antibody, antigen concentration is determined. Given these two detection formats, it is imperative to have high quality antibodies with good specificity and an advanced detection system in order to develop a quality immunoassay. Antibodies Immunoglobulins, or antibodies, are a group of secreted glycoproteins found in the serum of all animals. The production of antibodies occurs at the culmination of a series of events between B lymphocytes and effector cells of the immune system. When a mature B cell encounters a specific antigen, it will undergo activation, proliferation and label Secondary antibody Primary antibody antigen label Secondary antibody Primary antibody antigen

PAGE 45

33 differentiation leading to the eventual production of antibodies (139). Antibodies perform two key functions; they bind antigens and some mediate effector functions throughout the body: fixation of complement and binding to various cell types (139). Antibodies are extremely specific thus making them an ideal molecule for the use in assays to detect a variety of substances. There are five classes of antibodies, IgG, IgA, IgM, IgD and IgE, determined by amino acid sequence differences in the constant region of the heavy chain. The antibodies of the class IgG are the most prevalent antibody in serum and the most commonly used in immunoassays. IgG antibodies have a molecular weight of approximately 150 kD and are composed of two light and two heavy chains. B lymphocytes are capable of making over 18 billion antibodies through a variety of mechanisms: site-specific recombinations, mutations, random gene splicing, splicing inaccuracies and class switch recombination (139). Antibodies need to be specific with a high affinity for the target antigen for use in immunoassays (97, 134, 166). Affinity is the strength of the bond between the antigen and antibody. In order to detect antigen in low concentrations, the affinity of the antibody must be high. The antibodies must be stable and have low cross reactivity with other molecules. If an antibody has a high degree of cross-reactivity with other molecules aside from the antigen of interest, then quantitation of the antigen could be inaccurate. When selecting an antibody for a particular application, it is important to choose one that provides the greatest amount of sensitivity with the least amount of crossreactivity.

PAGE 46

34 The antigenic determinant, or epitope, is a single antigenic site to which the antibody binds. The size of an epitope is normally 6-8 amino acid residues. A single antigen can have multiple epitopes. In an immune response, if the antigen is large (e.g., a bacterium), animals will produce a large number of antibodies from different B cells. Each B cell will produce its own type of antibody. These antibody producing cells manufacture antibodies that recognize different epitopes on the same foreign molecule. This heterogeneous mixture of antibodies is found in the serum of animals. The natural collection of antibodies produced from different B cell lines is called polyclonal antibodies. Polyclonal antibodies used for research purposes are normally made from goat, rabbit or sheep (57). Polyclonal antibodies are the most commonly used antibodies for immunoassays (57). These antibodies remain stable for approximately one year at 4C when in suitable buffers and held at high protein concentrations of 1 mg/ml (57). Monoclonal antibodies are produced by only one type of B cell and can recognize only a single type of epitope on a complex antigen. In 1975, Khler and Milstein developed the first hybridomas by fusing together mouse myeloma cells and lymphocytes from the spleens of mice that had been immunized with a specific antigen (81). The hybridomas combine the immortality of the myeloma cell with the antibody producing ability of the lymphocyte. Monoclonal antibodies are directed against a single epitope and therefore can function as a capture or detection molecule within the constraints of the same assay. Monoclonal antibodies generated from the same clone would compete against itself for the same epitope if it were used as both the capture and detection molecule in the same immunoassay.

PAGE 47

35 Labels In order to detect and quantify an antigen, antibodies must be conjugated to a label. The presence of labeled directly correlates with the concentration of antigen. The original label used for immunoassays was the radioisotope (125). Since the development of the immunoassay, different enzyme labels with a matching substrate or fluorophores have been used to produce measurable end products. Enzymes used for colorimetric assays include horseradish peroxidase and alkaline phosphatase. Chromogenic, flurogenic and chemiluminescent substrates can be used with either enzyme. Fluorophores, such as rhodamine, can be used as an antibody label. Also, enzymes for the detection of chemiluminescence (peroxidase-catalyzed luminol assays) are used in immunoassays. In order to utilize the immunoassay without the constraints of radioactivity, enzymatic, fluorescent and chemiluminescent labels are replacing raidioisotopes (125). Most enzymatic detection systems offer an equal or better detection capability than the radioactive systems without the dangers associated with radioactivity. When using an enzymatic tag, samples can be read with a spectrophotometer, luminometer or fluorometer. In order to get faster and better results, fluorescent and chemiluminescent systems are employed (125). Fluorescent labels provide a powerful label in immunoassays. Typically, the results produced from a fluorescent label are more sensitive and can be obtained quicker than an enzymatic or colorimetric assay; however, their potential can be limited if dyes are not stable and there is sample interference (57). Also, a complex and expensive

PAGE 48

36 instrument for the detection of fluorescence, a fluorometer, is required if a fluorescent label is chosen. Chemiluminescent tags are the most sensitive and require the fewest amount of reagents of the four labels mentioned (89, 106, 189). Unlike colorimetric or fluorescent assays, samples assayed with chemiluminescent tags contribute little to no inherent background interference (106). Regardless of the label and detection system employed, all immunoassays are based on the same antigen-antibody interactions. Therefore, the specific interaction of antigen and antibody is very important for low limits of detection-the smallest concentration measurable with a particular assay-making immunoassays applicable to a wide variety of compounds of interest. Colorimetric, fluorescent and chemiluminescent assays are safer and more sensitive than radioimmunoassays and have therefore become more popular in use and application. Biosensors Background There is an increased demand for detection systems capable of producing sensitive and specific results within a relatively short period of time. The need for this type of detection system spans different types of industries: environmental, agricultural, pharmaceutical, food and public health (87). Biosensors are analytical detection devices which utilize biological molecules to detect other biomolecules or chemical substances. All biosensors exploit the ability of biomolecules to specifically recognize a target analyte, a substance an assay aims to detect. Biosensors combine this specific interaction or affinity between a biomolecule and its target with signal generation near or directly at

PAGE 49

37 the surface of a transducer (Figure 7) (30). The transducer takes one type of energy and converts it to another. After the molecular recognition between biomolecule and analyte occurs, a signal is produced that can be coupled and converted by the transducer into a signal that can be measured (30). The signal produced by the transducer, (e.g. electrochemical, optical or thermal), corresponds to the concentration of the target (30). Biosensors have existed since the 1970s but the interest in developing sensors for near real-time/real-time detection has lead to advances in their development (186). Different types of affinity systems can be used in conjunction with a biosensor, including the following: enzyme-substrate, antibody-antigen, receptor-ligand, lectin-sugar or nucleic acid for complementary sequence hybridization (30). FIGURE 7. General principles of a biosensor (138). Immunoassays are frequently used in conjunction with biosensors. As described previously, antibodies are most commonly used; however, other biomolecules are also used to recognize and bind target. These biomolecules can be immobilized to the solid surface platform through passive adsorbption, linked through a biotin-streptavidin bridge Enzyme Antibody Receptor Lectin Nucleic acid Electroactive Electrode substance pH change semiconductor pH electrode Light Photon counter Mass change Pizoelectric device Biological Recognition Element Signal Transducers Electric signal

PAGE 50

38 or covalently attached to serve as capture molecules (154). A detection molecule, which also recognizes the target, is tagged with an appropriate label and is used to generate a measurable signal (92). Biosensors are not without their disadvantages: some are expensive, produce inconsistent results or are bulky and not practical for in the field use. However, biosensors offer several advantages over conventional laboratory techniques. Biosensor assays are generally easier to perform and do not require trained personnel (92). For example, home pregnancy tests and glucose monitoring systems are bioassays. A home pregnancy test utilizes monoclonal antibodies directed against HCG to produce a colorimetric change. The over the counter home pregnancy test is a notable example of an effective biosensor: it is inexpensive, disposable, simple and accurate. Due to the specific recognition required between biomolecule and analyte, biosensors generate high degrees of sensitivity and specificity. Near real-time/real-time detection in complex matrices is also an advantage frequently derived from the use of a biosensor (37, 38, 39, 92). Finally, fully automated biosensor assays that can operate unattended have the potential to be integrated into on-line monitoring systems in water treatment plants or food processing facilities. Types of Optical Biosensors Optical biosensors utilize optical grade glass, plastic or silicon fiber to transmit light from one position to another. Light is propagated through the interior core of the fiber and reacts with reagents that are placed near the surface of the fiber. The light emission is converted into a quantifiable signal and the intensity correlates with the concentration of the unknown target. The transducer (a monochromator, lenses and photomultiplier tube) receives a signal and collectively converts it into an electrical signal

PAGE 51

39 that can be recorded (87). Types of optical biosensors include surface plasmon resonance, fiber optic and evanescent wave. Evanescent wave Fiber-Optic Biosensor One type of optical biosensor is based on evanescent wave technology. An evanescent wave fiber-optic biosensor uses the principle of total internal light reflection (Figure 8). Electromagnetic waves of light are propagated within the waveguide and a portion of it travels outside the core and is referred to as an evanescent wave. When biomolecules labeled with fluorophores fall within the range of the evanescent wave (approximately 100-1000 nm), the fluorophores become excited by the light source and the fluorescent signal is coupled back through the fiber. A photodiode allows for quantitation of the light emitted from the fluourophores within the range of the evanescent wave. One type of fiber-optic evanescent wave biosensor detection system is the Analyte 2000 (Research International, Monroe, WA), which uses an optical fiber as the waveguide (37, 38, 39, 43, 85, 86, 91, 92, 155, 170, 171). The Analyte 2000 utilizes a sandwich immunoassay based on the specificity of antigen-antibody or receptor-ligand binding. An immobilized, capture biomolecule is attached to a fiber-optic waveguide through a streptavidin-biotin bridge or passively adsorbed. This waveguide is exposed to a substance containing the suspect antigen and then subsequently exposed to a detection biomolecule conjugated to a fluorophore. Excitation light from a 635 nm laser diode is propagated through the waveguide and fluorescent molecules within approximately 100 nm of the fiber are excited (92, 178). The emission light from the excited fluorophore

PAGE 52

40 reverberates into the waveguide and is quantitated in picoamps and displayed on a computer. FIGURE 8. Total internal reflection in optical fiber (154). The propagation of electromagnetic waves in optical fiber by total internal reflection. A portion of the light travels just outside the core, approximately 100 nm, and is referred to as the evanescent wave or field. Affinity-Based Elements There are a wide variety of biomolecules that can function as capture and/or detection molecules in conjunction with biosensors including antibodies, receptors or nucleic acids (70). When selecting an affinity-based recognition element for use with the biosensor, it is important to consider the application of the assay. The biomolecule should be selected based on its ability to minimize non-specific binding and increase specificity and sensitivity (70, 162). Antibodies, either monoclonal or polyclonal, are the most commonly used recognition molecule in conjunction with biosensor assays (70, 97). Theoretically, polyclonal antibodies would not compete for the same epitope and could be used as both a capture and a detection molecule. Monoclonal antibodies from the same clone would theoretically compete for the same epitope if used as a capture and Evanescent wave travels outside the core Light guided in core Evanescent wave travels outside the core Light guided in core

PAGE 53

41 detection molecule. Therefore, it would seem advantageous to use polyclonal antibodies as both detection and capture molecule or a combination of polyclonal and monoclonal antibodies. The receptor/ligand relationship offers an alternative to antibodies in developing biosensor assays (162). Receptors are cell surface molecules, either cytoplasmic proteins (steroid receptors) or transmembrane receptors (gangliosides), which exhibit specificity for a particular effector molecule (105, 135, 143, 174). As in antibody interactions, receptors may bind their respective ligands with low or high affinities. High affinity interactions are desirable in biosensor assays. Ganglioside receptors are glycosphingolipids widely distributed in all tissues but heavily concentrated in the central nervous system of animals (105, 174). Gangliosides are natural cellular receptors for toxins and like antibodies, rely on specific interactions with ligands. Common gangliosides include GD1a, GD1b, GD2, GD3, GM1, GM2, GM3 and GT1b (82). Viral and bacterial pathogens exploit the natural relationship between ganglioside/ligand interactions to gain entry into host cells (105, 143, 164, 174). Binding of these pathogens to ganglioside receptors can be utilized for detection. Gangliosides can be used as capture molecules when immobilized on solid surface assay platforms or labeled with a tag and used as a detection element (105, 143, 162, 165, 174). Another non-antibody based biological interaction used with biosensor assay development is the nucleic acid probe that can specifically hybridize with its complementary sequence. A nucleic acid probe is a segment of DNA designed specifically to a nucleic acid target. Unlike antibodies or receptors, nucleic acid recognition elements can only bind and interact with molecules containing RNA or DNA.

PAGE 54

42 These probes lack the ability to detect proteins, chemicals or other biological molecules. Recognition of the target nucleic acid is dependent on the formation of stable hydrogen bonds between the complementary nucleic acid probe and its target (70). Nucleic acid probes can be used for either capture or detection. Antibodies, receptors and nucleic acid probes, can be used in a variety of assay formats. One type of element may be used as both the capture and detection molecule or a combination of recognition molecules can be used in unison with one another. Using more than one type of element per assay could limit the amount of competition that occurs for binding sites as long as the two elements do not bind to the same epitope. If, in the same assay, a ganglioside and an antibody are used together as capture and detection molecule, respectively, than theoretically they are not competing against one another for the same recognition site. This type of assay could then potentially increase the specificity of the assay due to the double recognition of the target analyte. Summary Millions have suffered and thousands have died from the virulent toxin released from the cells of V. cholerae Cholera is transmitted through the fecal-oral route and there is a higher incidence of infection and endemnicity in areas with poor sanitation. The cells of V. cholerae are best adapted to marine and brackish waters and exist in either a commensal or symbiotic role with marine organisms. Cholera occurs most often in populations with low socioeconomic status: sanitation facilities are poor or nonexistent, personal hygiene practices are wanting and the people lack the education to understand the disease. Eradication of the disease may be too big of a task for national or international control programs. However, measures can be taken for quick detection

PAGE 55

43 followed by containment of the disease. The most critical element of cholera control during an epidemic is the early identification of the illness, proper notification of health officials, adequate healthcare facilities and the disposal of raw sewage (7). The neurotoxin produced by C. botulinum is one of the most lethal biological toxins known. Botulism is rarely seen manifested in society; however, it is possible to suffer lethal food poisoning with improperly processed foods. Recent world events have focused the world’s attention and raised awareness to the potential use of this toxin as a biological agent against our military and civilian populations. Botulinum toxin is relatively easy to obtain or produce and the intentional contamination of food or water supplies remains a viable threat. The purpose of this research was to develop a sensitive and specific assay utilizing gangliosides in conjunction with the fiber-optic evanescent wave biosensor to detect biological toxins. The biosensor immunoassay is a detection system that has the capability to produce rapid, sensitive and specific results in near real-time. Gangliosides, recognition elements other than antibodies, were used as capture or detection molecules. Assay development was performed using CT as the model protein. Once a rapid, sensitive and specific assay was developed to detect CT in buffer using the biosensor, the method was applied to the detection of the toxin in oysters. Shellfish, specifically oysters when ingested raw or undercooked, have been implicated in the transmission of cholera (40, 109, 122, 123). The pathogenicity of cholera is directly related to the production of CT. A rapid, sensitive and specific assay is necessary for the detection of CT in contaminated water and food. The majority of assays used to assess contamination of food and water with V.

PAGE 56

44 cholerae rely on conventional laboratory techniques: enrichment, isolation, serological and/or biochemical tests. The presence of CT genes can be detected through PCR; however, enrichment is often needed and the assay fails to determine if the protein is being expressed. Vibrio cholerae cells recovered from contaminated food or water can be induced in vitro to produce CT. This type of assay can assess the virulence of the strain but is time consuming. Enrichment followed by inoculation into induction media is needed to induce the organism to produce the CT. For these reasons, utilizing the evanescent wave, fiber-optic biosensor is an innovative means of detecting CT directly from matrices. After the CT biosensor immunoassay was developed in combination with gangliosides, the protocol was used in assay development for the detection of botulinum toxin. The mouse assay is the accepted standard for the detection of botulinum toxin. This assay requires the use of many animals, and requires several days to determine the amount and type of toxin present (1, 47, 53, 59, 118, 181). Due to the potency of the botulinum neurotoxins, it is important to rapidly detect and differentiate between the toxins in food, clinical or environmental samples. A rapid, sensitive and specific assay is needed for the detection and differentiation of botulinum neurotoxins. The ability to rapidly and accurately detect contaminated food and water would serve as a preventive measure and aid those involved in public health, quality control specialists and possibly first responders in the reduction of the spread of disease.

PAGE 57

45 MATERIALS AND METHODS Bacterial strains Vibrio cholerae Pacini 569B ATCC 25870 and Vibrio cholerae Pacini El Tor ATCC 39050 were obtained from the American Type Culture Collection (Manassas, VA). Vibrio cholerae O1 El Tor Inaba was obtained from the Centers for Disease Control and Prevention (CDC, Atlanta, GA). Bacteria were suspended in tryptic soy broth with 20% glycerol (Remel, Lenexa, KA) and stored in a -80C freezer until needed. Media and Culture Conditions Stock cultures. All cultures were grown on tryptic soy agar plates (TSA, Remel) for 18 hours in a 37C incubator and serially diluted in 0.01 M phosphate-buffered saline, 0.85% NaCl, pH 7.4 (PBS) for use in assays. Enrichment and selective media. Alkaline peptone broth (1% NaCl, 1% peptone pH 8.5) was used for the enrichment of all V. cholerae isolates from seawater and oyster matrices. Casamino acids yeast extract medium (CAYE, 3% casamino acids, 0.3% yeast extract, 0.05% K2HPO4, pH 7.0, CAYE with 0.2% glucose, Instant Ocean (Aquarium Systems, Inc. Mentor, OH.), Salt water culture medium (0.05 % peptone, 0.03 % yeast extract, 0.03 % glycerol, 0.01 % CaCO3), Salt water culture medium with 0.2% glucose, Salt water culture medium made in Instant Ocean and Salt water culture medium with 0.2% glucose made in Instant Ocean were used in the development of enrichment assays (108).

PAGE 58

46 Thiosulfate Citrate Bile Salts agar (TCBS, Becton Dickinson & Co., Sparks, MD) was used for the selective plating of V. cholerae Plates were grown at 37 C in an incubator for 18 hours. Suspect yellow colonies were tested using API 20E (Biomerieux, Durham, NC) and PCR for confirmation. Toxins Lyophilized cholera toxin beta subunit (CT-B) and cholera toxin (CT) were purchased from Sigma Chemical Co. (Saint Louis, MO) and reconstituted in 0.1M phosphate buffered saline, pH 7.4. Two botulinum toxoids were provided by the CDC. Type-A botulinal toxoid was at a 1:10,000 dilution and type-B botulinal toxoid at a 1:1,000 dilution. Gangliosides The lyophilized gangliosides GM1, GD1b and GT1b were purchased from Sigma Chemical Co. The GM1 was reconstituted in PBS. Liposomes Liposomes were made to specification by Dr. Anup Singh at the Sandia National Laboratories (Livermore, CA). The liposomes (Figure 9) were composed of Ldistearroylphosphatidylcholine (DSPC), cholesterol, Alexa-DHPE (Alexafluor 647-1,2dihexadecanoylsn –glycerol-3-phospoethanolamine, Molecular Probes, Eugene, Oregon) and GT1b (Sigma Chemical Co.) ( 151 ). The mole ratio used was 42.5:42.5:10:5 (DSPC/cholesterol/Alexa Fluor/GT1b) (156 ).

PAGE 59

47 FIGURE 9. Liposome construct (156). Antibodies and Labeling Sources. Lyophilized rabbit anti-cholera toxin beta subunit antibody was purchased from Biogenesis (Kingston, NH) and reconstituted in PBS. Lyophilized rabbit antiV. cholerae antibody was purchased from Difco (Sparks, MD). Lyophilized rabbit anti-botulinum toxin antibody was purchased from Accurate Chemical and Scientific Corporation (Westbury, NY) and reconstituted in PBS. Goat anti-botulinum type A, rabbit anti-botulinum type B and goat anti-botulinum type B antibodies were a gift from the CDC. Affinity purified peroxidase goat anti-rabbit IgG was purchased from Kirkegaard & Perry Laboratories Inc. (KPL, Gaithersburg, MD). Goat anti-rabbit IgG conjugated to alkaline phosphatase was purchased from Sigma Chemical Co. Affinity purified peroxidase rabbit anti-goat IgG was purchased from Jackson Immunologicals (West Grove, PA). Cyanine 5 labeling. Cholera toxin and botulinum toxin antibodies were labeled directly with the cyanine 5 fluorophore (FluoroLink Cy5 Reactive Dye 5-pack Amersham Biosciences, Piscataway, NJ). Labeling was performed according to the protocol Gan g lioside Fluorescein-li p i d Toxin Antibod y

PAGE 60

48 described previously by Demarco et al (37). The antibody to be conjugated was diluted to 1mg/ml in 500 l of conjugation buffer (0.1 M carbonate-bicarbonate buffer, pH 9.3). The antibody solution was added to the cyanine 5 dye pack vial. The dye pack vial was placed in the original foil packaging and placed at 4C for 18 hours. Free dye and labeled antibody were separated by filtration on a Bio-Gel P10 column (exclusion limit of 1,500 to 20,000 Da; Bio-Rad, Hercules, CA). A 10 ml bed volume was poured and equilibrated with PBS-0.1% sodium azide. Labeled-antibody fractions were collected and the molar concentration of the labeled antibody was determined using a DU-64 spectrophotometer (Beckman, Fullerton, CA) set at 280 nm and 650 nm. Cy5 labeled antibodies were stored at 4C until needed. Biotin labeling Antibodies were labeled with biotin using succinimidyl-6(biotinamido) hexanoate (EZ-Link NHS-LC-Biotin; Pierce, Rockford, IL). Labeling was performed according to the protocol described previously by DeMarco et al (37). One mg of EZ-Link NHS-LC-Biotin was dissolved in 1 ml of N,N -dimethylformamide (DMF); 75 l of this solution was added to 425 l of antibody that had been dissolved in 0.1 M sodium carbonate, pH 8.5, at a concentration of 2 mg/ml. The solution was then placed on ice for 2 hours. Free biotin and labeled antibody were separated by filtration on a Bio-Gel P10 column. Labeled-antibody fractions were collected in half ml fractions and the protein concentration of the labeled antibody was determined using a DU-64 spectrophotometer set at 280 nm. Biotinylated labeled antibodies were stored at 4C until needed.

PAGE 61

49 Viable Counts Cells were resuspended in 0.1 M PBS, pH 7.4 (100 dilution). Ten-fold serial dilutions were made, 10-1 to 10-7. Nine hundred micrliters of sterile PBS, pH 7.4, in sterile microfuge tubes was used as the diluent. One-hundred microliters of appropriate dilutions (usually 10-5, 10-6, 10-7) were duplicate plated onto TSA and incubated at 37C for 18 hours. Oysters Unshelled oysters processed by Hilton’s Willapoint (South Bend, WA) were purchased from Publix (Lakeland, FL). Live oysters were harvested from the northwest side of the Gandy Bridge at Tampa Bay, Fl. (Figure 10). Live oysters were collected and immediately placed on ice. They were transported back to the laboratory within an hour and placed in a 10 gallon tank containing Instant Ocean (20 ppt NaCl) as described by Murphree and Tamplin (111).

PAGE 62

50 FIGURE 10. Map of Tampa Bay, Florida. Live oysters were harvested during low tide from the northwest side of the Gandy Bridge on the St. Petersburg side of the Tampa Bay. Toxin Assays Bicinchoninic acid protein assay. Fifty parts of BCA reagent A (Sigma Chemical Co.) were mixed with 1 part BCA reagent B (Sigma Chemical Co.) to make the working reagent. A standard curve was made using Bovine Serum Albumum (BSA, Fisher, Pittsburgh, PA). For each concentration, BSA was mixed with PBS to a final volume of 50 l. One ml of the working reagent was added to each BSA concentration and incubated at room temperature for 1 hour. At the same time CTwas serially diluted and then one ml of working reagent was added and incubated at room temperature for 1 hour. The absorptions were determined using a DU-64 spectrophotometer set at

PAGE 63

51 562nm. The concentration of the protein was determined by extrapolating from the standard curve line. SDS-PAGE. 12% Bis Tris precast gels purchased from Invitrogen (Carlsbad, CA) were used. Twenty microliters of the protein sample, 10 l of sample buffer (Invitrogen, Carlsbad, CA), 4 l of reducing agent (Invitrogen) and 6 l of DI water were mixed and heated at 70C for 10 minutes. Twenty microliters of each protein sample mixture was used per well. One well contained 10 l of the Novex Color Marker 12 (Sigma Chemical Co.). MES-DES running buffer (Invitrogen) was used and the gel was electrophoresed at 200V for 40 minutes. The gel was then stained with Coomassie Blue R250 (Sigma Chemical Co.) for one hour. The gel was destained with a 7.5% methanol, 10% acetic acid solution for 30 minutes and then placed in DI water for 18 hours. Western blot. An SDS-PAGE gel was removed from the gel box and transferred to a nitrocellulose membrane at 14V, 4 for 18 hours using Towbin buffer, pH 8.3 (173). The membrane was removed and incubated on a shaker with 5% skim milk/PBS, pH 7.4, containing 0.1% Tween 20 (PBST, pH 7.4) at 24C for 10 minutes. The gel was rinsed once with PBST and then incubated with the primary antibody, polyclonal anti-CTrabbit (Biogenesis) at a 1:500 dilution for one hour at 24C. This was followed by three 45 second washes with 10 mM EDTA/PBS (pH 8.5). The membrane was rinsed three times with PBST for 20 seconds each and then incubated with the secondary antibody, goat anti-rabbit IgG conjugated to alkaline phosphatase at a 1:5,000 dilution for an hour at 24C. This was followed by three 30 second washed with PBST. Substrate (50 ml borate buffer, ph 9.7, 13 mg o-dianisidine, 13 mg -napthyl acid phosphate) was added and the membrane was kept in the dark until bands appeared.

PAGE 64

52 Ganglioside-based ELISA. All volumes were at 100 l and all incubations were at 24C unless otherwise noted. Ganglioside was dissolved in PBS at a concentration of 1.5 g/ml. One hundred microliters were added to the wells of a 96-well Nunc Maxi Sorp microtitre plate and incubated at 37C for 18 hours. This mixture was aspirated and wells were washed once with PBST. Blocking buffer (PBS, 2 mg/ml casein, 2 mg/ml BSA) was added and incubated at 24C for 30 minutes. This mixture was aspirated and wells were washed once with PBST. The antigens were added into their respective wells and incubated for 10 minutes. This solution was aspirated and wells were washed three times with PBST. Polyclonal antibody at a concentration of 10 g/ml was added to each well and incubated for 30 minutes. This mixture was aspirated and wells were washed three times with PBST. Horseradish peroxidase-labeled secondary antibody, anti-rabbit IgG or anti-goat IgG (KPL) was added to each well at a 1:500 dilution and incubated for 30 minutes. This mixture was aspirated and wells were washed three times with PBST. QuantaBlue substrate (Pierce Biotechnology, Rockford, IL) was added and incubated for 25 minutes. QuantaBlue stop solution (Pierce Biotechnology) was added and the relative fluorescence was determined at 325 nm excitation and 420 nm emission with a Spectra Max Gemini XS fluorometer (Molecular Devices, Sunnyvale, CA). A signal to noise ratio of 2 was positive detection. Indirect ELISA. All volumes were at 100 l and all incubations were at 24C unless otherwise noted. Serial dilutions of protein were made in sterile 0.01 M PBS with 0.1% BSA, pH 7.4. Protein was added to the wells of a 96-well Nunc Maxi Sorp microtitre plate and incubated at 4C for 18 hours. This mixture was aspirated and wells were washed once with PBST. Blocking buffer was added and incubated at 24C for 30

PAGE 65

53 minutes. This solution was aspirated and wells were washed once with PBST. Primary antibody at 10 g/ml was added to each well and incubated for 30 minutes. This mixture was aspirated and wells were washed three times with PBST. Horseradish peroxidaselabeled secondary antibody, anti-rabbit IgG or anti-goat IgG was added to each well at a 1:500 dilution and incubated for 30 min. This mixture was aspirated and wells were washed three times with PBST. After washing, QuantaBlue substrate (Pierce Biotechnology) was added and incubated for 25 minutes. QuantaBlue stop solution (Pierce Biotechnology) was added and the relative fluorescence was determined at 325 nm excitation and 420 nm emission with a Spectra Max Gemini XS fluorometer (Molecular Devices). A signal to noise ratio of 2 was positive detection. Toxin induction. Vibrio cholerae O1 569B ATCC 25870, Vibrio cholerae O1 El Tor ATCC 39050 and Vibrio cholerae O1 El Tor CDC were grown on TSA. Strains were induced to produce CT by following the protocol of Minami et al. (108). Ten ml of modified CAYE broth were placed into sterile 100 mm petri dishes. Each dish was inoculated with two colonies from each strain. After stationary incubation for 18 hours at 30C, a loopful of the culture was transferred to a new petri dish containing 10 ml of the modified CAYE medium with 0.2% glucose. Salt water culture medium, Salt water culture medium made in Instant Ocean (Aquarium Systems) and CAYE made in Instant Ocean were also used as enrichment media for the production of CT. This culture was incubated under stationary conditions for 18 hours at 30C. Five ml of each culture was stored at 4C and 5 ml was centrifuged (1,600 g, 30 min, 4C). The supernatant fluids were filtered through sterile 0.22 m-pore-size membrane filters. CT in the filtered

PAGE 66

54 supernatant fluid was detected using the Analyte 2000 (Research International) and an ELISA. Toxin spiked oysters. Four grams of oyster (Hilton’s Willapoint, South Bend, WA) and 36 ml of Alkaline Peptone Water (APW, pH 8.5) were added into a 50 ml conical tube and spiked with CT (168). The sample was homogenated for 90 seconds at the highest speed using a PowerGen 125 homogenizer (Fisher) allowed to settle at 24C for 15 minutes and then separated by filtration on a Bio-Gel P10 column. A 4 ml bed volume was poured and equilibrated with PBS with 0.1% sodium azide. Ten oyster fractions were collected in 500 ul volumes. The CT was detected using the Analyte 2000 (Research International) and an ELISA. Tank inoculation. A 10 gallon aquarium tank with 10 L of medium was set up and the liquid was recirculated with an AquaClear aquarium pump (Wal-Mart) (Figure 11). The temperature in the tank was maintained at varying temperatures with a VisiTherm Automatic Aquarium Heater (Wal-Mart). On day one, oysters harvested from the Tampa Bay were introduced into the tank and allowed to sit for 24 hours, these were designated the control oysters, tc. On day two, the tank was inoculated with V. cholerae 569B ATCC 25870 to a final concentration of 104 cfu/ml and the oysters collected prior to inoculation were designated t0. Beginning on day two, oyster and broth medium samples were collected at the following time intervals: t0, t6, t12, t24, t48 and t72 hours. Following the method of Murphree and Tamplin, ten oysters from each tank were scrubbed and their meat pooled (111). The meat was homogenized 1:1 in PBS followed by serial dilution in PBS. A five tube MPN was prepared by enriching dilutions of oyster homogenate or tank medium in APW (pH 8.5) at 42C for 6 hours in a water bath. Each

PAGE 67

55 tube in the MPN series was then streaked onto TCBS (Difco) (41). Plates were incubated at 37 C for 18 hours. Suspect V. cholerae isolates were confirmed using API 20E and PCR. FIGURE 11. Tank set-up. API 20E Identification Yellow colonies that were 2-3 mm in diameter, gram negative rods and oxidase positive were resuspended in PBS and identified using the API 20E (Biomerieux, Durham, NC). Colonies identified as V. cholerae were further confirmed using PCR. PCR PCR targeted two genes for the identification of Vibrio cholerae : the 564 base pair region of the cholera toxin A subunit ( ctxA ) and the 300 base pair region of the 16S23S rRNA intergenic spacer region ( its ). Primers were made by IDT (Coralville, IA) and their sequences are listed in Table 5. The forward and reverse primers were diluted to a 10 m concentration. Each 50 l reaction volume contained the following: 1X PCR

PAGE 68

56 TABLE 5. PCR primers for identification of Vibrio cholerae. Target Sequence Amplicon Reference size (bp) ctxA F 5’ CGG GCA GAT TCT AGA CCT CCT G 3’ 564 48 R 5’ CGA TGA TCT TGG AGC ATT CCC AC 3’ i ts F 5’ TTA AGC ATT TTC TCT GAG AAT G 3’ 300 21 R 5’ AGT CAC TTA ACC ATA CAA CCC G 3’ buffer, 200 M each of dNTP mixture, 0.25 M of the forward and reverse primer, 2 l of whole V. cholerae cells resuspended in sterile DI water, 2.5U of TaKaRa ExTaq polymerase (TaKaRa, Shuzo, Otsu, Japan) and 37.25 l sterile water. Samples were run on a Bio-Rad iCycler (Hercules, CA) with the following conditions: 95 C for 2 minutes; 30 cycles of 95C for 1 minute, 58C for 1 minute, and 72C for 2 minute; and extension at 72C for 10 minutes. PCR products were visualized by 1% agarose (Amresco Inc., Solon, OH) gel electrophoresis. The volume per lane was 8 l of each PCR reaction mixed with 2l of loading dye (Bio-Rad, Hercules, CA). Five microliters of each PCR/dye mixture was electrophoresed on a 1.5 % agarose gel. A 1 kb PCR marker (Promega, Madison, WI) was used. Amplicon bands were visualized using GelStar nucleic acid stain (BioWhittaker Molecular Applications, Rockland, MD) at a concentration of 3 l stain/100 ml agarose in TBE buffer and a transilluminator. Ganglioside-Based Biosensor Assay for the Detection of Cholera Toxin Analyte 2000 biosensor. The Analyte 2000 (Research International) is a fiberoptic biosensor that utilizes the properties of an evanescent wave in conjunction with a fluoroimmunoassays to detect a wide variety of molecules (Figure 12). Utilizing four

PAGE 69

57 different fiber-optic probes and a single wavelength, the Analyte 2000 can analyze up to four channels simultaneously. Results are produced in near real-time, approximately 20 minutes with previously prepared waveguides. The Analyte 2000 can assay for target analyte in dirty matrices (37, 38, 39, 86, 171). FIGURE 12. Analyte 2000 evanescent wave, fiber-optic biosensor. Fiber preparation. A polystyrene waveguide (Research International) was sonicated for 30 seconds in isopropanol. The waveguide was rinsed in DI water and black ink was placed on the distal end of the waveguide to prevent the escape of light. After the paint dried, the waveguide was added to a glass capillary tube to form a reaction chamber. The waveguide was incubated with 1.5 g/ml of GM1 in PBS and incubated 18 h at 37C. The waveguide was removed from the glass capillary tube and placed into an Analyte 2000 cuvette as shown in Figure 13. The waveguide was rinsed with 1 ml of PBST.

PAGE 70

58 Cuvette Waveguide Cuvette Cap Assay Chamber FIGURE 13. Assembly of Analyte 2000 cuvette and polystyrene waveguide used in biosensor assays. Background readings. Waveguides were rinsed with 1 ml PBST, and then three background readings were taken by repeating the following steps: each waveguide was incubated for 5 minutes with 200 l Cy-5 labeled anti-CT-subunit IgG (10 g/ml) in blocking buffer (2mg/ml casein, 2 mg/ml BSA in PBS), followed two 1 ml PBST rinses. The laser was turned on and the value, in picoAmperes (pA), was recorded after the final rinse. For each reading, the value recorded after the last PBST rinse was designated the baseline reading. The pA value for each baseline reading was subtracted from the subsequent baseline reading, and this calculated value was designated as the change in previous signal for baseline readings 2-4. The average of change in previous signal for the baseline readings was calculated. The detection limit was calculated as three times the standard deviation (SD) of the change in previous signal for the baseline readings plus the average of change in previous signal. When samples were tested, each previous sample reading was subtracted from the next sample reading. Sample assay. Figure 14 illustrates the immunoassay used with the Analyte 2000. Assays were performed by adding 1 ml of sample to the reaction chamber containing the

PAGE 71

59 fiber waveguide and incubating at 24C for 10 minutes. The sample was rinsed with 1 ml PBST and the waveguide was incubated for 5 minutes at 24C with 200 l Cy5 anti-CTB-subunit IgG in blocking buffer. The waveguide was rinsed twice with PBST and the laser was turned on. The signal was recorded in pA. The change in signal ( pA) was FIGURE 14. Illustration of sandwich immunoassay utilized with Analyte 2000. The target antigen is captured by a bound biomolecule in a standard immunoassay. The captured antigen is detected by a Cy5-labeled detection antibody. IIIIIIIIIIII IIIII Ganglioside Sample applied, toxin binds to ganglioside Toxin IIIIIIIIIIII IIIII IIIIIIIIIIII IIIII Other cells and debris washed away Detection antibody binds to analyte and fluoresces when laser turned on IIIIIIIIIIII IIIII Laser on Y Y Y Y Y Y Y Y Y IIIIIIIIIIII IIIII Ganglioside Sample applied, toxin binds to ganglioside Toxin IIIIIIIIIIII IIIII IIIIIIIIIIII IIIII Other cells and debris washed away IIIIIIIIIIII IIIII Ganglioside IIIIIIIIIIII IIIII IIIIIIIIIIII IIIII Ganglioside Sample applied, toxin binds to ganglioside Toxin IIIIIIIIIIII IIIII Sample applied, toxin binds to ganglioside Toxin IIIIIIIIIIII IIIII Toxin Toxin IIIIIIIIIIII IIIII IIIIIIIIIIII IIIII IIIIIIIIIIII IIIII IIIIIIIIIIII IIIII IIIIIIIIIIII IIIII Other cells and debris washed away IIIIIIIIIIII IIIII IIIIIIIIIIII IIIII IIIIIIIIIIII IIIII Other cells and debris washed away Detection antibody binds to analyte and fluoresces when laser turned on IIIIIIIIIIII IIIII Laser on Y Y Y Y Y Y Y Y Y Detection antibody binds to analyte and fluoresces when laser turned on IIIIIIIIIIII IIIII Laser on Detection antibody binds to analyte and fluoresces when laser turned on IIIIIIIIIIII IIIII Laser on Detection antibody binds to analyte and fluoresces when laser turned on IIIIIIIIIIII IIIII IIIIIIIIIIII IIIII Laser on Laser on Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y

PAGE 72

60 calculated as the value of the sample reading after the second rinse minus the value of the previous sample reading. In order to account for variation among each waveguide, the values were normalized by using the following equation (85). pA value for sample being tested 100 pA value for 100 ng/ml CT A sample was considered positive if the change in signal ( pA) was above the limit of detection and the normalized value was above zero. A standard curve was generated using known CT concentrations. CT concentrations of the samples were determined by extrapolation from the CT standard curve. The assay for signal amplification follows the same biosensor protocol with the following exception. The secondary antibody, Cy5 conjugated anti-rabbit IgG, was incubated with the primary antibody, anti-CT antibody, in a 1:2 ratio, respectively, to a final protein concentration of 10 g/ml. The antibodies were incubated at 24C for 5 min before incubation on the waveguide. Cholera toxin standard preparation. The stock solution was prepared by diluting CT to 100 ng/ml in PBS containing 0.1% BSA. This was serially diluted tenfold in PBSBSA 0.1 %. Liposome-Based Biosensor Assay for the Detection of Botulinum Toxin Fiber preparation. Waveguides were prepared as previously described. The waveguide was incubated in the glass capillary reaction chamber with 100 l of 100 g/ml streptavidin solution prepared in PBS for 18 hours at 4C. The waveguide was removed from the glass capillary tube and placed into an Analyte 2000 cuvette as shown in Figure 13. The waveguide was rinsed with 5 ml PBST to remove any unbound

PAGE 73

61 streptavidin. One hundred microliters of 100 g/ml of biotinylated polyclonal antibotulinum toxin was prepared in PBS and this was incubated on the waveguide for one hour at 24C. Following incubation with the biotinylated antibody, the waveguide was rinsed with 1 ml of PBST. Background readings. Fibers were rinsed with 1 ml PBST, and then three background readings were taken by repeating the following steps: each waveguide was incubated for 5 minutes at 24C with 200 l of liposome at varying dilutions followed two 1 ml PBST rinses. The laser was turned on and the value, in picoAmperes (pA), was recorded after the final rinse. For each reading, the value recorded after the last rinse was the baseline reading. The standard deviation was calculated as previously described. Sample assay. Assays were performed by adding 1 ml of sample to the reaction chamber containing the fiber waveguide and incubating at 24C for 10 minutes. The sample was rinsed with 1 ml PBST and the waveguide was incubated with 200 l of liposome for 5 minutes at 24C. The waveguide was rinsed twice with PBST. The laser was turned on and the signal was recorded in pA. The change in signal was calculated as previously described. A sample was considered positive if the change in signal was above the limit of detection and the normalized value was above zero.

PAGE 74

62 RESULTS Biosensor Immunoassay Development for the Detection of Cholera Toxin Cholera Toxin Purity CT-B and CT purchased from Sigma were assayed for the presence of other proteins. An SDS-PAGE (Figure 15) and a Western blot (Figure 16) were performed to assess protein purity. The cholera toxin is composed of two subunits with molecular weights of 27.2 kD A subunit and 11.6 kD B subunit. CT-B is a polymerized form of the B subunit that, when subjected to the reducing conditions of an SDS-PAGE, should dissociate into CTmomomers. CTmonomers produced a band at approximately 11.5 kD corresponding to the molecular weight of the B subunit. There were no visible band on either the SDS-PAGE or the Western blot that correlated to the molecular weight of subunit A. The holotoxin has an A to B ratio of 1:5 and, therefore, it is possible that there was a smaller concentration of the A subunit. The polyclonal anti-CT-B antibody used for the Western would not detect the A subunit. The holotoxin dissociated when subjected to the reducing agent used in the SDS-PAGE. Therefore, the presence of a band with approximately the same molecular weight corresponding to the B subunit, 11.5 kD, on both the gel and membrane (Figure 15 and 16) and the lack of any other protein bands reaffirmed that the commercially purchased toxins were not contaminated.

PAGE 75

63 FIGURE 15. SDS-PAGE of CT and CT-B. Lane 1, 10 ng CT; Lane 2, 10 ng CT-B; Lane 3, marker. The bands visible in lanes 1 and 2 corresponded to the molecular weight of CTB. FIGURE 16. Western blot of CT and CTB. Lane 1, 10 ng CT; Lane 2, 10 ng CTB; Lane 3, marker. Filter was probed with anti-CTB antibody. 14.4 kD 6 kD Predicted 11.5 kD band 1 2 3 14.4 kD 6 kD Predicted 11.5 kD band 1 2 3 1 2 3 Predicted 11.5 kD band 1 2 3 Predicted 11.5 kD band

PAGE 76

64 BCA protein assay. A BCA protein assay was done in order to determine if the concentration of the CTB protein listed on the Sigma label was accurate (Table 6). The concentration of the protein was listed at 1 mg/ml. TABLE 6. Extrapolated CT-B concentrations from BCA standard curve. Protein concentration Absorbance @ 562 nm Extrapolated concentration 0 mg/ml BSA 0 0.2 mg/ml BSA 0.165 0.4 mg/ml BSA 0.366 0.6 mg/ml BSA 0.513 0.8 mg/ml BSA 0.655 1 mg/ml BSA 0.780 0.2 mg/ml CT-B 0.188 0.216 mg/ml CT-B 0.6 mg/ml CT-B 0.443 0.538 mg/ml CT-B The correlation coefficient (r) of the line was 0.997. It was determined that dilutions made from this commercial source of protein would correspond with the expected calculated value of that dilution. The label concentration of protein in Sigma CTB was considered accurate. Comparison of Affinity and Specificity of GM1 and Polyclonal Antibody Binding to V. cholerae Whole Cells and Toxin A sandwich ELISA using GM1 for capture and anti-CTB for detection was used to test the affinity of the ganglioside GM1 and the anti-CTB antibody binding to V. cholerae serotypes O1 and O139 and to CTB (Figure 17). Two sets of ten-fold serial dilutions were made for both V. cholerae O1 and V. cholerae O139. One set of dilutions

PAGE 77

65 was boiled for 10 minutes, whereas the other set was composed of viable cells. Ten-fold serial dilutions were also made of CT-B (initial concentration at 1 mg/ml). 0 2 4 6 8 10 12 123456 Dilution (10-x)Signal to noise ratio FIGURE 17. Sandwich ELISA with GM1 as capture molecule and anti-CT-B antibody as detection molecule for detection of CTB and V. cholerae serotypes. Tenfold dilutions of V. cholerae serotype O1 ( ; 1.62 x 109 cfu/ml), V. cholerae serotype O1 boiled ( ;1.62 x 109 cfu/ml), V. cholerae serotype O139 ( ; 7.3 x 108 cfu/ml), V. cholerae serotype O139 boiled (x; 7.3 x 108 cfu/ml) and CT( ; 1mg/ml) were done in duplicate. Error bars ( ) indicate the standard deviation of the means. When the GM1 was used for capture and the anti-CTB antibody was used for detection, there is no cross-reactivity to cells. The sandwich ELISA was capable of detecting CT-B at a dilution of 10-5 which corresponds to concentration of 10 ng/ml. The sandwich ELISA shows no reactivity to V. cholerae cells. The cells themselves will not be a source of false positives, they must produce CTB for positive detection.

PAGE 78

66 CT Standard Curve A standard curve was generated using the sandwich ELISA as previously described (Figure 18). Assays using the Analyte 2000 (Figure 19) are also shown. GM1coated wells or waveguides were prepared at a concentration of 1.5 g/ml. Serial dilutions of CTwere added to the wells or waveguides and incubated for 10 minutes. After washing, anti-CTB antibody was added to wells or waveguides. Antibody was detected using secondary antibody for ELISA or by a direct Cy5 label for the Analyte 2000. Sample values for the Analyte 2000 were normalized (as previously described) using 100 ng/ml as the standard. For all ELISA assays, a signal to noise value greater than 2 was treated as a positive signal. The sensitivity of the ELISA was approximately 0.625 ng/ml CT-B. 0.0 2.0 4.0 6.0 8.0 10.0 0.3130.6251.252.5510 CT-B concentration in ng/mlSignal to noise ratio FIGURE 18. Mean value for ELISA CT-B standard curve. Sandwich ELISA with GM1 as capture molecule and anti-CT-B antibody as detection molecule. Mean value for three separate assays.

PAGE 79

67 0 20 40 60 80 100 120 0.1110100 CT-B concentration in ng/mlNormalized change in pA FIGURE 19. Analyte 2000 normalized standard curve of CT-B. Mean normalized values for 21 waveguides. The average normalized values and standard deviations for the three serial dilutions of 1 mg/ml CT-B are listed in Table 7. The limit of detection for the Analyte 2000 assay was 1 ng/ml. The Analyte 2000 assay has a greater dynamic range of detection. It detects 1 ng/ml to 100 ng/ml of CT-B while the ELISA becomes saturated at CT-B of 2.5 ng/ml. TABLE 7. Average normalized pA values and standard deviations for CT-B standard curve using the Analyte 2000. CT-B concentration Average normalized value Standard deviation 0.1 ng/ml 1.6 1.53 1 ng/ml 4.2 0.95 10 ng/ml 24.8 3.83

PAGE 80

68 Sensitivity of Antibody Versus Ganglioside as Capture Molecule An Analyte 2000 assay comparison was done using ant-CT-B antibody as the capture molecule versus the ganglioside, GM1, as the capture molecule. The assay was performed in order to determine if one of the biomolecules was more effective at capturing CTB. Two waveguides were incubated with GM1 for 18 hours at 1.5 g/ml in a 37C incubator. Two waveguides were incubated with 100 l of 100 g/ml streptavidin for 18 hours at 4C. The following day the waveguides were rinsed with 5 ml PBST and 100 l of 50 g/ml of biotin-labeled anti-CTantibody was incubated with the two waveguides containing streptavidin at 24 C for one hour. The standard biosensor assay protocol using Cy5-anti-CTB antibody for detection was followed. Figure 20 illustrates that the ganglioside GM1 as a capture molecule produced the same levels of detection sensitivity as the anti-CTB antibody: 1 ng/ml.

PAGE 81

69 0 5 10 15 20 25 30 0.010.1110 CT-B concentration in ng/mlNormalized value FIGURE 20. Comparison of anti-CTB antibody ( ) versus GM1 ganglioside-based ( ) capture assay using the Analyte 2000. Cy5-anti-CTB antibody was used for detection in both assays. Error bars are present, but not visible with some data points because they are so small. Signal Amplification A secondary antibody that would bind the Fc portion of the anti-CTantibody was used to amplify the pA signal of the Analyte assay (Figure 21). Amplification of the signal may allow detection of lower concentrations of CTB. The secondary antibody, Cy5-labeled anti-rabbit-IgG, was preincubated with the primary antibody (anti-CTB) at a ratio of 1:2 prior to incubation on each waveguide. The same biosensor assay protocol was followed, but the antibody mixture was used as the detection antibody.

PAGE 82

70 0 20 40 60 80 100 120 0.1110100 CT-B concentration in ng/mlNormalized value FIGURE 21. Comparison of biosensor assay using Cy5-anti-CTB antibody ( ) or a mixture of anti-CTB antibody and Cy5-anti-rabbit IgG ( ) as the detection reagent. The average normalized pA values and standard deviations for the three serial dilutions of 1 mg/ml CTB are listed Table 8. The trends in variability in normalized values for signal amplification were the same as those seen in the CT standard curve. Adding a secondary antibody did not increase the sensitivity of the assay. The limit of detection for the assay was determined to be the same as the assay with no signal amplification: 1 ng/ml CT-B.

PAGE 83

71 TABLE 8. Average normalized pA values and standard deviations for signal amplification using the Analyte 2000. CT-B concentration Average normalized value Standard deviation 0.1 ng/ml 0.3 0.3 1 ng/ml 2.3 0.8 10 ng/ml 20.0 5.1 The signal amplification did not cause an increase in sensitivity of the assay. The average normalized values were lower for signal amplification than with a primary antibody only (Table 7 and 8). Figure 21 demonstrates that there is no advantage to using a secondary antibody. Toxin Induction ELISA and Analyte 2000 utilizing CT. Vibrio cholerae cells produce the holotoxin CT, not just CT-B. In previous ELISA and Analyte 2000 assays, a commercially modified CT-B was used to generate data. There was a concern that ELISA and biosensor results would change when testing the holotoxin, CT. With the ELISA, it became necessary to lower the anti-CTB antibody concentration to 310 ng/ml. The concentration used in previously reported assays was 10 g/ml, but the range of detection was low (0.1-5 ng/ml) because the antibody saturated the signal. The detection antibody concentration remained 10 g/ml for the Analyte 2000 assays. Detection of CT instead of CTB produced results with a sensitivity similar to the one obtained with the ELISA (Figure 22) and the Analyte 2000 (Figure 23): 1 ng/ml.

PAGE 84

72 0 2 4 6 8 10 12 14 16 18 20 0.0010.010.1110100 CT concentration in ng/ml Signal to noise ratio FIGURE 22. ELISA values for CT standard curve assay. Standard curve values are the mean of four assays. 0 20 40 60 80 100 120 0.1110100 CT concentration in ng/mlNormalized cchange in pA FIGURE 23. Standard curve for CT using Analyte 2000. The pA values are the mean from eight waveguides.

PAGE 85

73 The average normalized pA values were higher and the standard deviations were lower when CT was utilized in the assay instead of CT-B (Table 9). However, the limit of detection when using CT was the same as when testing CTB: 1 ng/ml. TABLE 9. Average normalized pA values and standard deviations for CT standard curve using the Analyte 2000. CT concentration Average normalized value Standard deviation 0.1 ng/ml 2.5 1.2 1 ng/ml 5.9 1.4 10 ng/ml 28.2 3.9 CAYE as induction medium. Using the Miniami et al. protocol, three strains of V. cholerae were induced in vitro to produce cholera toxin: Vibrio cholerae Pacini 569B ATCC 25870, Vibrio cholerae Pacini El Tor ATCC 39050 and Vibrio cholerae O1 El Tor Inaba CDC (108). Cholera toxin was detected by ELISA and Analyte 2000 without signal amplification (Table 10). Cell counts of 108 cfu/ml or higher were needed to induce CT production. Vibrio cholerae O1 Classical 569B ATCC 25870 is considered to be the ideal CT producing strain (44). Based on the values in Table 10, this organism produced the greatest amount of CT. Both V. cholerae El Tor strains produced lower levels of CT and were not ideal toxin producers. Signal saturation occurred at approximate CT concentrations of 5 ng/ml in ELISAs for CT. Analyte 2000 normalized values greater than 5.9 (Table 9), the limit of detection for 1 ng/ml from the CT standard curve assay, were treated as positives.

PAGE 86

74 V. cholerae O1 Classical 569B ATCC 25870 was the only strain that had positive values for all of the assays utilizing both methods of detection. High concentrations (>108 cfu/ml) of V. cholerae O1 El Tor ATCC 39050 never produced positive signals with the Analyte 2000 and had 60% (3/5) positive ELISA tests. The V. cholerae O1 El Tor (received from the CDC) was not any more efficient at CT production than the ATCC El Tor strain. The CDC El Tor strain never generated positive signals with the Analyte 2000 and had 80% (4/5) positive ELISA tests for CT production. The concentration of primary antibody used for an ELISA was 310 ng/ml, while 10 g/ml of detection antibody was used with the Analyte 2000. The difference in antibody concentration could account for positive signals generated with the ELISA and not the Analyte 2000. Based on the higher values shown for CT production in Table 10, V. cholerae O1 Classical 569B ATCC 25870 was selected as the strain to use in future induction assays.

PAGE 87

75 TABLE 10. Cholera toxin induction values utilizing an ELISA and the Analyte 2000. Viable count Analyte 2000 ELISA normalized value signal to noise ratio Assay 1 V. cholerae O1 El Tor (ATCC 39050) 5.9 x 108 cfu/ml 1.6 3.7 V. cholerae O1 El Tor (CDC) 3.0 x 109 cfu/ml 3.4 3.9 V. cholerae O1 Classical 569B 4.2 x 109 cfu/ml 53.5 4 (ATCC 25870) Assay 2 V. cholerae O1 El Tor (ATCC 39050) 1.85 x 109 cfu/ml 0.4 3.4 V. cholerae O1 El Tor (CDC) 1.36 x 109 cfu/ml 1.6 4.4 V. cholerae O1 Classical 569B 3.9 x 109 cfu/ml 51.3 4.4 (ATCC 25870) Assay 3 V. cholerae O1 El Tor (ATCC 39050) 2.89 x 109 cfu/ml 0.7 2.6 V. cholerae O1 El Tor (CDC) 2.95 x 109 cfu/ml 2.7 3.8 V. cholerae O1 Classical 569B 1.07 x 1010 cfu/ml 65.4 5.4 (ATCC 25870) Assay 4 V. cholerae O1 El Tor (ATCC 39050) 2.02 x 109 cfu/ml 0.8 1.0 V. cholerae O1 El Tor (CDC) 8.9 x 108 cfu/ml 2.2 1.5 V. cholerae O1 Classical 569B TNTCa 44.8 3.4 (ATCC 25870) Assay 5 V. cholerae O1 El Tor (ATCC 39050) 3.7 x 109 cfu/ml 0.7 1.0 V. cholerae O1 El Tor (CDC) 7.4 x 108 cfu/ml 1.7 2.0 V. cholerae O1 Classical 569B 4.7 x 109 cfu/ml 55.3 3.0 (ATCC 25870) aToo numerous too count. Alternate induction media. Vibrio cholerae O1 classical 569B was induced for CT production using a petri dish and a beaker with a similar surface area to volume ratio of a 10 gallon tank. For subsequent assays, oysters were harvested from Tampa Bay and placed in a 10 gallon tank. The medium in the tank needed to serve three essential functions: 1) maintain the oysters viability 2) allow V. cholerae growth and 3) stimulate CT production. Four types of media were tested in petri dish and beaker trials: salt water

PAGE 88

76 culture medium made in DI water (SWC-DI), salt water culture medium made in Instant Ocean (SWC-IO), CAYE made in Instant Ocean (CAYE-IO) and CAYE in DI water (CAYE-DI) as the control (Figure 24). The induction protocol previously described was used to induce CT production. 0 2 4 6 8 10 12 14 SWC in D ISWC in IOCAYE in D ICAYE in IO MediumSignal to noise ratio FIGURE 24. Toxin induction during V. cholerae growth in four types of media. A petri dish ( ) and a beaker ( ) with the same surface to volume ratio of a 10 gallon tank were used for these assays. SWC in DI, salt water culture medium in deionized water; SWC in IO, salt water culture medium in instant ocean; CAYE in DI, casamino acids yeast extract in deionized water; CAYE in IO, casamino acids yeast extract in instant ocean. All four types of media supported CT production in a petri dish. As seen in Table 11, three of the four petri dish assays produced cell counts that correlated with those in Table 10. The petri dish culture with the highest cfu/ml (SWC-DI) generated the highest

PAGE 89

77 signal to noise ratio while the culture that had the lowest cfu/ml (SWC-IO) generated the lowest signal to noise ratio. When running the beaker assays, two types of media allowed for toxin production: SWC-DI and CAYE-IO. The SWC-IO tank simulation failed to support the growth of any V. cholerae (Table 11). Cell counts for the beaker assays (tank simulation) never reached those of the petri dishes. TABLE 11. Viable cell counts for CT induction utilizing alternative media. Culture conditions Cell count Petri dish – SWC-DI 4.0 109 cfu/ml Petri dish – SWC-IO 1.1 107 cfu/ml Petri dish – CAYE-DI (control) 3.6 109 cfu/ml Petri dish – CAYE-IO 3.1 109 cfu/ml Beaker assays – SWC-DI 9.0 107 cfu/ml Beaker assays– SWC-IO 0 cfu/ml Beaker assays CAYE-DI (control) 3.8 107 cfu/ml Beaker assays – CAYE-IO 3.5 108 cfu/ml Toxin Spiked Oysters The detection of CT in an oyster matrix was tested using the Analyte 2000. Store purchased oysters were processed according to Tamplin and Capers (168). Oysters were spiked with 10 ng/ml of CT and waveguides were prepared with 1.5 g/ml of GM1 as previously described. The assay was done in a progressive manner: buffer was assayed, followed by assays for an oyster, a spiked oyster and then the CT standard. The limit of detection was determined for each Analyte 2000 channel. All four of the channels generated false positives with the unspiked oyster matrix (Table 12). All of the channels

PAGE 90

78 were positive for the spiked oyster matrix and the standard. Based on the false positives obtained for the unspiked oyster matrix, it was hypothesized that there were proteins present that were cross reacting with the detection antibody. TABLE 12. Toxin spiked oyster homogenate samples analyzed with the Analyte 2000. This is a representative sample of waveguides tested. Channel a 1 2 3 4 LOD 13 12 23 38 Change in signal (pA) for Buffer -3 -4 -6 -10 Oyster 40.0 87 111 153 Spiked oyster 383 386 625 375 100 ng/ml CT 1781 1306 2260 1377 anumbers in bold are positive signals. Cross-Reactive Proteins Affinity for polyclonal anti-CT-B rabbit antibody A Western blot was performed to determine if the Analyte 2000 values shown in Table 12 for the unspiked oyster homogenate were false positives. Four different oysters were processed and analyzed for cross-reactive proteins. Electrophoresed proteins transferred onto nitrocellulose filter were probed with polyclonal anti-CT-B antibody (Figure 25). Two protein bands reacted with the anti-CT-B (Figure 25). One band had a molecular weight of approximately 100 kD and one had a molecular weight of approximately 60 kD. The CT holotoxin has a

PAGE 91

79 molecular weight of 85 kD, the A subunit has a molecular weight of 27.2 kD and the B subunit has a molecular weight of 11.6 kD. CT and CT-B molecular weights do not correspond with 100 and 60 kD; therefore, the presence of these proteins suggests that the false positives may be due to cross-reactive proteins FIGURE 25. Western blot of four unspiked oyster samples. Lane 1, marker; Lane 2, buffer; Lane 3, 75 ng CT; Lane 4, unspiked oyster A; Lane 5, unspiked oyster B; Lane 6, unspiked oyster C; Lane 7, unspiked oyster D; Lane 8, 75 ng CT. Affinity for IgG antibody. The Western Blot was repeated in order to determine if the two cross reactive proteins (100 kD and 60 kD) seen in Figure 25 were binding specifically to the polyclonal anti-CT-B rabbit antibody (Biogenesis) or just binding nonspecifically to IgG antibody (Figure 26). Incubation with a primary antibody was 1 2 3 4 5 6 7 8 135 98 47 33 26 15 7 100 kD 60 kD

PAGE 92

80 eliminated and the membrane was incubated directly with anti-rabbit IgG conjugated to alkaline phosphatase. The only visible bands were those of the molecular weight marker, indicating the 100 kD and 60 kD cross-reactive proteins were binding specifically to the anti-CT-B antibody. FIGURE 26. Western blot of oyster homogenate utilizing goat anti-rabbit IgG conjugated to alkaline phosphatase only. Lane 1, marker; Lane 2, 75 ng CT; Lane 3, buffer; Lane 4, oyster homogenate spiked with 10 ng/ml CT; Lane 5, buffer; Lane 6, oyster homogenate; Lane 7, buffer; Lane 8, buffer; Lane 9, 75 ng CT; Lane 10, buffer. Specificity for GM1. ELISAs were performed to determine if the oyster matrix was binding to GM1. Two types of samples were tested: (1) APW spiked with 100 ng/ml of CT and (2) unspiked oyster homogenate. Samples were fractionated by size using a 1 2 3 4 5 6 7 8 9 10 135 98 47 33 26 15 7

PAGE 93

81 P10 column and fractions were collected in 500 l volumes. Samples were directly adsorbed to wells (Figure 27) or captured using 1.5 g/ml of GM1 per well (Figure 28). All of the fractions from both samples were tested. Figures 27 and 28 illustrate that the oyster homogenate did not exhibit nonspecific binding to the wells of the 96-well microtitre plate. The fractionated oyster homogenate displayed no specificity or binding activity with the ganglioside GM1. All 10 fractions had a signal to noise ratio of less than 2 and, therefore, were considered negative. ELISA results would also indicate the level of nonspecific binding of the oyster matrix to the plate and to GM1. It was concluded based on Figures 27 and 28 that the cross-reactive proteins bind to anti-CT-B antibody and do not bind to GM1. 0 5 10 15 20 12345678910 Fraction numberSignal to noise ratio FIGURE 27. ELISA values for fractions of oyster sample adsorbed directly to ELISA wells. Ten fractions from oyster homogenate ( ) and APW spiked with 100 ng/ml CT ( ) were tested for binding to GM1.

PAGE 94

82 0 5 10 15 20 12345678910 Fraction numberSignal to noise ratio FIGURE 28. ELISA values for GM1 based assay. Ten fractions from oyster homogenate ( ) and APW spiked with 100 ng/ml CT ( ) were tested for binding to GM1. Removal of cross-reactive proteins from oyster matrix. The cross-reactive proteins in the oyster were removed by passage through a P10 gel bed. Five hundred microliters of oyster homogenate spiked with 5 g/ml of CT and unspiked oyster homogenate were passed through two separate P10 columns with a 10 ml bed volume. The columns were equilibrated with PBS with 0.1% sodium azide. Ten 500 l fractions were collected and analyzed using an ELISA (Figure 29) and the Analyte 2000 (Table 13). Fractions 6 through 10 contained the cholera toxin from the spiked sample. None of the fractions from the unspiked sample contained cholera toxin or other proteins that would generate a positive ELISA signal. According to ELISA results, fractions 7 and 8 should have high concentrations of CT. Fractions 7 and 8, were analyzed using the Analyte 2000 (Table 13). Positive signals were generated when fractions 7 and 8 of the

PAGE 95

83 spiked sample were assayed. Negative signals were obtained for the nonspiked fractions 7 and 8. Fractionating the oyster matrix through the gel column may have removed the cross-reactive proteins or diluted them to undetectable concentrations. 0 5 10 15 20 25 30 35 40 45 50 12345678910 Fraction numberSignal to noise ratio FIGURE 29. ELISA values for oyster homogenate fractions. Oyster homogenate spiked with 5 g/ml of CT ( ) and unspiked oyster homogenate ( ) were fractionated. All ten fractions from each sample were tested.

PAGE 96

84 TABLE 13. Analyte 2000 values for oyster homogenate fractions. Channela 1 2 3 4 Fraction 7 Fraction 8 Fraction 7 Fraction 8 spiked spiked unspiked unspiked LOD 0 4 7 36 Change in signal for APW -1 0 -2 -1 Sample 899 645 0 -1 anumbers in bold are positive signals. Tank Inoculation Instant Ocean as induction medium. A 10 gallon aquarium was filled with 10 L of Instant Ocean. The temperature of the tank was maintained at 30 C. This is the optimum temperature to induce CT production. Oysters were harvested from Tampa Bay during low tide. The oysters were scrubbed to remove loose particles and placed in the aquarium. Oyster shells remained closed confirming viability. After 24 hours, V. cholerae O1 569B ATCC 25870, was inoculated into the tank at a final concentration of 7.27 104 cfu/ml. Organisms were enumerated with alkaline peptone broth by the five tube MPN enrichment method. After 6 hours of incubation in a 42C waterbath, all turbid tubes were streaked on TCBS agar for the isolation of V. cholerae (41, 111). V. cholerae was identified using PCR and API 20E. V. cholerae and other bacteria were recovered from the oyster and Instant Ocean at 6 and 12 hours and presumptively identified by API 20E (Table 14). Table 15 lists the cfu/g (oyster matrix) or cfu/ml (tank

PAGE 97

85 medium) determined by a 5 tube MPN. V. cholerae was no longer recovered from the oyster or Instant Ocean after 12 hours of incubation. Cell counts listed in Table 15 were at least 1000-times lower than those observed using the petri plate toxin induction method. The lower cell counts suggested that V. cholerae cells would not produce toxin. The ELISA values (Figure 30) confirmed the lack of CT production. TABLE 14. Isolates presumptively identified by API 20E for Instant Ocean tank. tc, control oysters processed at time of harvest; t0, samples placed in tank, processed 24 hours after harvest and immediately prior to tank inoculation; t6, t12, t24, t48, and t72, samples processed 6, 12, 24, 48, and 72 hours after tank inoculation respectively; oyster or Instant Ocean, type of sample. Time point and sample at collection Isolate identification Control oyster tc Aeromonas hydrophila, Klebsiella ozaenae, Vibrio parahaemolyticus Oyster – t0 Klebsiella ozaenae Instant Ocean – t0 Vibrio parahaemolyticus Oyster – t6 Escherichia coli, Vibrio cholerae Instant Ocean – t6 Vibrio parahaemolyticus, Vibrio fluvialis, Vibrio cholerae Oyster – t12 Vibrio parahaemolyticus, Vibrio cholerae Instant Ocean – t12 Vibrio fluvialis, Vibrio cholerae Oyster – t24 Vibrio fluvialis Instant Ocean – t24 Vibrio fluvialis Oyster – t48 Aeromonas hydrophila Instant Ocean – t48 Vibrio fluvialis Oyster – t72 unidentifiable Instant Ocean – t72 Escherichia coli

PAGE 98

86 TABLE 15. Cell counts for Vibrio cholerae recovered from oyster and Instant Ocean. t6, and t12, samples processed 6 and 12 hours after tank inoculation; oyster or Instant Ocean, type of sample. Time point and sample at collection Cell count Oyster – t6 4.0 104 cfu/g Instant Ocean – t6 2.5 104 cfu/ml Oyster – t12 6.5 103 cfu/g Instant Ocean – t12 6.5 104 cfu/ml 0 2 4 6 8 10 12345678910 Fraction numberSignal to noise ratio FIGURE 30. ELISA values for samples from tank inoculation with Instant Ocean. Oyster homogenate was fractionated as described previously and assayed for the presence of CT. Control oysters ( ), oysters at t0 ( ), oysters at t6 ( ) and oysters at t12 ( ) were all negative for CT.

PAGE 99

87 Isolates presumptively identified by API 20E as V. cholerae were confirmed by PCR using two sets of primers and predicted lengths: its (Figure 31) and ctxA (Figure 32). The its primers were designed to amplify the 16S-23S rRNA intergeneic spacer region of V. cholerae (21) The ctxA primers were designed to amplify the genes encoding the A subunit of CT (48). Figure 31 shows positive amplification of its gene from V. cholerae isolates at 6 and 12 hour time points. Figure 32 shows positive amplification of ctxA gene from V. cholerae isolates at both time points. 1 2 3 4 5 6 7 predicted 300 bp amplicon FIGURE 31. PCR of isolates using its primers. Lane1, ladder; Lane 2, positive control V. cholerae ; Lane 3, negative control E. coli; Lane 4, presumptive V. cholerae from oyster at t6; Lane 5, presumptive V. cholerae from Instant Ocean at t6; Lane 6, presumptive V. cholerae from oyster at t12; Lane 7, Presumptive Vibrio cholerae from Instant Ocean at t12.

PAGE 100

88 FIGURE 32. PCR of isolates using ctxA primers. Lane 1, ladder; Lane 2, positive control V. cholerae; Lane 3, negative control E. coli; lane 4, presumptive V. cholerae from oyster at t6; Lane 5, presumptive V. cholerae from Instant Ocean at t6; Lane 6, presumptive V. cholerae from oyster at t12; Lane 7, presumptive V. cholerae from Instant Ocean at t12; SWC-DI as induction medium. SWC-DI simulates the previously described petri dish protocol. A 10 gallon aquarium was filled with 10 L of SWC-DI. The temperature of the tank was maintained at 30C. Oysters harvested from Tampa Bay were placed in the aquarium for 24 hours. Vibrio cholerae O1 569B ATCC 25870 was then inoculated into the tank at a final concentration of 4.56 104 cfu/ml. After 0, 6, 12, 24, and 48 hour time points organisms were recovered from the oyster matrix and the tank medium. Table 16 lists those isolates presumptively identified by API 20E. Vibrio cholerae was not recovered from the oyster matrix or the SWC medium at any time interval. The inability to recover any V. cholerae leads to the conclusion that it would be highly unlikely that CT production would have occurred. The ELISA testing for the presence of 1 2 3 4 5 6 7 p redicted 564 bp amplicon

PAGE 101

89 CT (Figure 33) confirmed the lack of CT production within the oyster matrix or tank medium. TABLE 16. Isolates presumptively identified by API 20E for SWC-DI tank. Oyster and tank medium samples were collected at varying time intervals. Ten oysters from each tank were scrubbed, their meat pooled and homogenized 1:1 in PBS followed by serial dilution in PBS (111). A five tube MPN was prepared by enriching dilutions of oyster homogenate or tank medium in APW (pH 8.5) at 42C for 6 hours in a water bath. Each tube in the MPN series was then streaked onto TCBS (41). Plates were incubated at 37 C for 18 hours. Individual colonies were resuspended in PBS for presumptive identification by API 20E. tc, control oysters processed at time of harvest; t0, samples placed in tank, processed 24 hours after harvest and immediately prior to tank inoculation; t6, t12, t24, and t48, samples processed 6, 12, 24, and 48 hours after tank inoculation, respectively; oyster or Instant Ocean, type of sample. Time point and sample at collection Isolate identification Control oyster tc Vibrio parahaemolyticus Oyster – t0 Vibrio parahaemolyticus SWC-DI – t0 Vibrio parahaemolyticus Oyster – t6 Vibrio alginolyticus SWC-DI – t6 unidentifiable Oyster – t12 Vibrio parahaemolyticus SWC-DI – t12 Escherichia coli Oyster – t24 Vibrio alginolyticus SWC-DI – t24 Vibrio alginolyticus Oyster – t48 No growth SWC-DI – t48 Vibrio alginolyticus

PAGE 102

90 0 2 4 6 8 10 6122448 Time interval (h)Signal to noise ratio FIGURE 33. ELISA for the detection of CT after V. cholerae inoculation of tank containing SWC-DI. Oyster samples ( ) and SWC-DI ( ) were assayed at different time intervals for the presence of CT. The ganglioside GM1 was used as the capture molecule and the primary antibody was the anti-CT antibody. A signal to noise ratio greater than 2 was a positive value. CAYE-IO as induction medium. CAYE-IO simulates the previously described petri dish protocol. A 10 gallon aquarium was filled with 10 L of CAYE-IO and was inoculated with V. cholerae The temperature of the tank was maintained at 30C. At 24 hours post inoculation, ten mls of the liquid was transferred to a new tank containing CAYE-IO with 0.2% glucose. The in vitro induction assays described previously utilized fresh CAYE medium with 0.2% glucose for CT induction at 24 hours post inoculation. Oysters were harvested from Tampa Bay and then added to the aquarium. After 24 hours, V. cholerae O1 569B ATCC 25870, was inoculated into the tank at a final concentration of 4.56 104 cfu/ml. At 0, 6, 12, 24, and 48 hour time points organisms

PAGE 103

91 were recovered from the oyster matrix and the tank medium. Table 17 lists those isolates presumptively identified by API 20E. Vibrio cholerae was not recovered from the oyster or SWC at any time interval. It was unlikely that CT was produced without the growth of any viable V. cholerae ELISA detecting the presence of CT (Figure 34) confirmed the absence of CT production. TABLE 17. Isolates presumptively identified by API 20E recovered from CAYE-IO with 0.2% glucose tank. tc, control oysters processed at time of harvest; t0, samples placed in tank, processed 24 hours after harvest and immediately prior to tank inoculation; t6, t12, t24, and t48, samples processed 6, 12, 24, and 48 hours after tank inoculation, respectively; oyster or Instant Ocean, type of sample. Time point and sample at collection Isolate identification Control oyster tc Vibrio parahaemolyticus Oyster – t0 Vibrio parahaemolyticus CAYE – t0 Vibrio parahaemolyticus Oyster – t6 Vibrio alginolyticus, Vibrio parahaemolyticus CAYE – t6 Vibrio alginolyticus, Vibrio parahaemolyticus Oyster – t12 Vibrio alginolyticus CAYE – t12 Vibrio alginolyticus Oyster – t24 Vibrio alginolyticus CAYE – t24 Vibrio alginolyticus Oyster – t48 No growth CAYE with 0.2% glucose – t48 Vibrio alginolyticus

PAGE 104

92 0 2 4 6 8 10 6122448 Time interval (h)Signal to noise ratio FIGURE 34. ELISA detecting presence of CT from tank containing CAYE-IO with 0.2% glucose, inoculated with V. cholerae Oyster samples ( ) and CAYE-IO with 0.2% glucose ( ) were assayed at different time intervals for the presence of CT. All samples assayed were negative for CT. Biosensor Immunoassay Development for the Detection of Botulinum Toxin BCA Protein Assay No concentrations were provided with the botulinum toxoids received from the CDC. Only dilution factors were given: type A @ 1:10,000 and type B @ 1:1,000. A BCA protein assay was performed to determine the actual protein concentration of each toxoid. The correlation coefficient (r) of the line was 0.997. The extrapolated values are listed in Table 18 but their validity is questionable. The absorbance readings for each of the toxoids were below the reading of the lowest known BCA concentration and, therefore, determined to be invalid. Thus, it was determined that the dilutions to be tested would be based on the CDC stock suspension.

PAGE 105

93 TABLE 18. Extrapolated botulinum toxoid protein concentrations. Protein concentration Absorbance @ 562 nm Extrapolated concentration 0 mg/ml BSA 0 0.1 mg/ml BSA 0.039 0.2 mg/ml BSA 0.079 0.4 mg/ml BSA 0.137 0.6 mg/ml BSA 0.181 0.8 mg/ml BSA 0.240 1 mg/ml BSA 0.296 Toxoid A 0.019 0.018 mg/ml Toxoid B 0.016 0.028 mg/ml ELISA for Detection of Botulinum Toxoids Using Anti-Botulinum Antibody Anti-botulinum toxoid antibodies obtained commercially and from the CDC (Figure 35) were tested for their ability to bind to botulinum toxoids A and B using an ELISA. Serial dilutions of botulinum toxoids A and B were directly adsorbed to the wells of a 96-well microtitre plate. Antibodies to toxoid type A and type B were used as primary antibodies at a concentration of 10 g/ml. Secondary antibody, anti-rabbit IgG conjugated to HRP or anti-goat IgG conjugated to HRP was used at a dilution of 1:500. The polyclonal anti-A botulinum toxin antibody produced in goat and the polyclonal antiB antibody produced in rabbit, both received from the CDC, bound to their respective toxoids (Figure 35). The CDC anti-toxin A antibody bound toxoid A and could be detected at dilutions of 1:100,000 and 1:1,000,000, whereas the CDC ant-toxin B antibody bound to toxoid B and could be detected when the toxoid was diluted 1:10,000. The Accurate Chemical and Scientific Corporation antibody binding to toxoids A and B

PAGE 106

94 could not be detected. The CDC polyclonal anti-B goat antibody binding to toxoid B could not be detected. 0 1 2 3 4 5 1234567 Dilution 10-xSignal to noise ratio FIGURE 35. Detection of botulinum toxoids adsorbed to plate using anti-botulinum toxin antibodies. Antibody affinity for toxoid A and B was tested using an ELISA. Botulinum toxoid A and B were detected using antibodies provided by the CDC or purchased from Accurate Chemical and Scientific Corporation. Initial concentration of toxoid A was 1:10,000 and type B was 1:1,000. Combinations of Toxoid A with the CDC goat anti-A antibody ( ), Toxoid A with the Accurate antibody ( ), Toxoid B with the CDC goat anti-B antibody ( ), Toxoid B with the CDC rabbit anti-B antibody ( ) and Toxoid B with the Accurate antibody () were all assayed using an ELISA. Detection of Botulinum Toxoids Using Sandwich ELISA with GT1b Capture A sandwich ELISA was performed with the ganglioside, GT1b in PBS (1.5 g/ml) (Figure 36), as a capture molecule. GT1b was directly adsorbed to the wells of a

PAGE 107

95 mirotitre plate. Serial dilutions of toxoid A and B were captured by the GT1b. Commercial or CDC antibodies to toxoid type A and type B were used as primary antibodies to detect either toxoids. Secondary antibody diluted 1:500 was either antirabbit IgG or anti-goat IgG–HRP conjugated. The signal to noise ratio never reached a value of 2. The ganglioside GT1b showed no specificity for either botulinum toxoids. 0 1 2 3 1234567 Dilution 10-x of botulinum toxoidsSignal to noise ratio FIGURE 36. Detection of toxoids A and B using sandwich ELISA with GT1b capture. GT1b (1.5 g/ml) affinity for toxoid A and B was tested using an ELISA. Botulinum toxoid A and B were detected using antibodies provided by the CDC or purchased from Accurate Chemical and Scientific Corporation. Initial concentration of toxoid A was 1:10,000 and type B was 1:1,000. Combinations of Toxoid A with the CDC goat anti-A antibody ( ), Toxoid A with the Accurate antibody ( ), Toxoid B with the CDC goat anti-B antibody ( ), Toxoid B with the CDC rabbit anti-B antibody ( ) and Toxoid B

PAGE 108

96 with the Accurate antibody () were all assayed using an ELISA. All five combinations produced results that overlapped when graphed. GT1b-and GD1b-Based ELISA Two of the natural cellular receptors for botulinum toxins are the gangliosides GT1b and GD1b. These receptors were used as capture molecules in the development of an ELISA. Since no positive signal was obtained when the GT1b was used as a capture molecule, another ganglioside, GD1b was tested as a capture molecule. An ELISA was performed to determine if there was any specificity between the ganglioside GD1b for botulinum toxoid A and/or B (Figure 37). GD1b was dissolved in PBS at a concentration of 1.5 g/ml. The primary antibodies used were the CDC goat anti-A and CDC rabbit anti-B at a concentration of 10 g/ml. The secondary antibody was diluted 1:500 and was either anti-rabbit IgG or anti-goat IgG–HRP conjugated. Neither toxoid could be detected using the GD1b capture/antibody detection ELISA format. Signal to noise ratios were below 2 for botulinum toxoid A at 1:100,000 dilution and toxoid B at a 1:10,000 dilution. GD1b had no affinity for either botulinal toxoid A or B (Figure 37).

PAGE 109

97 0 2 4 6 8 10 Botulinum toxoid ABotulinum toxoid B Toxoid at 1:10 dilutionSignal to noise ratio FIGURE 37. Detection of botulinum toxoids A or B using GD1b capture in ELISA. GD1b (1.5 g/ml) affinity for toxoid A and B was tested using an ELISA. Botulinum toxoid A and B were diluted 1:10 and detected using antibodies provided by the CDC. Initial concentration of toxoid A was 1:10,000 and type B was 1:1,000. Biosensor Assay Detecting Toxoid Type A Using Liposomes Singh et al. (156) demonstrated that the incorporation of the ganglioside into a liposome may allow successful detection of botulinum toxin (156). Liposomes containing Alexafluor 647 were synthesized following the specifications of Singh et al. (156) for use as a detection molecule in conjunction with the Analyte 2000. Due to the fluorophore that was incorporated into the liposome being out of the spectral range of the fluorometer, the normal progression in the development of a biosensor assay could not be followed. The assay development was initiated with the Analyte 2000. Toxoid A (1:10 or 1:100) was adsorbed directly to the waveguide and allowed to incubate for 18 hours at 4C. The waveguides were rinsed with PBS and then incubated in blocking buffer (PBS, 2 mg/ml casein, 2 mg/ml BSA) and allowed to incubate 18 hours at 4C. The

PAGE 110

98 waveguides were then rinsed with PBS and liposomes at a dilution of 1:1,000 were incubated on the waveguides for 10 minutes. Table 19 shows representative results from a single waveguide assay. There was a significant increase in pA signal after incubation with the liposome followed by PBS rinses for all five channels including the negative control (Table 19), TABLE 19. Biosensor values for toxoid A adsorbed directly to the waveguide. Waveguides were incubated with toxoid A for 18 hours, followed by blocking buffer for 18 hours and then incubated with liposome at a 1:1,000 dilution. Fluorescent readings were recorded for each step of the assay. Numbers in parentheses indicate incubation times. Channel 1 2 3 4 Negative control Toxoid concentration 1:100 1:100 1:10 1:10 PBS Initial waveguide fluorescence 891 1093 1196 1277 670 Signal (pA) after assay step 0.2 ml liposome (5 min) 1393 2930 20431 6945 1168 Two 1 ml PBS wash 5271 7563 22578 6522 2981 0.2 ml PBST (5 min) 8440 15395 22577 17349 4254 Two 1 ml PBS wash (final reading) 4975 834 5763 4618 1284 Singh et al (156) reported that incubation with a detergent (such as PBST or Triton X) after addition of liposome disrupts the liposomes and alleviates dequenching of

PAGE 111

99 the fluorophore, maximizing fluorescent readings. Following incubation with PBST for 5 min, signals increased dramatically with the exception of channel 3, which had reached the upper limitations of the machine’s pA signal readout capabilities. After incubation with PBST, two 1 ml rinses with PBS removed any unbound liposomes causing the pA signal to decrease. However, for channels 1, 3, 4, and 5, the pA signal after the final PBS rinse was still higher than that prior to liposome incubation. There was a three-to-fivefold increase in the final fluorescence compared with the initial reading. The negative control illustrates a substantial amount of liposomes non-specifically sticking to the waveguide. The final reading obtained for the negative control waveguide was double the initial fluorescence reading. A direct sandwich assay format utilizing antibody for capture and the liposomes as a detection molecule was attempted. Biosensor Assay Utilizing Antibody as a Capture Molecule Varying liposome concentrations as de tection molecule for the detection of botulinal toxoid type A. Multiple sandwich biosensor assays were performed utilizing different concentrations of liposome as the detection molecule. Following 18 hours incubation with streptavidin, biotinylated goat anti-A antibody (100 g/ml) was incubated for 18 hours at 4C. On the following day, the waveguides were rinsed with PBS and then incubated for 18 hours at 4C in blocking buffer. On the final day, the waveguides were rinsed with PBS and an assay was completed with the liposome as the detection molecule. Initially liposomes at 1:1,000 to 1:5,000 dilutions were used as the detection reagent (Table 20 and 21). Biosensor assays using these dilutions of liposome produced very high LODs ranging from 117 to 3162. Table 22 shows results from

PAGE 112

100 biosensor assay for the detection of toxoid A using liposomes diluted 1:10,000 and 1:20,000 for the detection reagent. TABLE 20. Sandwich biosensor assay using anti-toxoid A antibody for capture and liposomes at 1:1,000 as the detection reagent for the detection of botulinal toxoid type A. Fluorescent readings were recorded for each step of the assay. Numbers in parentheses indicate incubation times. The limit of detection remained high and the botulinum toxoid type A was never utilized in the assay. Channela 1 2 3 4 Liposome concentration 1:1000 1:1000 1:1000 1:1000 Signal (pA) after assay step 0.2 ml liposome (5 min) 2 1 ml PBS wash 5995 2656 6104 3032 0.2 ml PBST (5 min) 7449 7913 11139 3869 2 1 ml PBS wash 6036 4172 2075 2265 0.2 ml liposome (5 min) 2 1 ml PBS wash 6782 3549 5937 3175 0.2 ml PBST (5 min) 6941 4909 10143 3880 2 1 ml PBS wash 5880 2984 3965 2874 0.2 ml liposome (5 min) 2 1 ml PBS wash 5622 1856 2876 3202 0.2 ml PBST (5 min) 5844 3259 3984 3411 2 1 ml PBS wash 5371 1775 2009 1776 0.2 ml liposome (5 min) 2 1 ml PBS wash 5870 1519 2098 2142 0.2 ml PBST (5 min) 6130 1871 3021 2580 2 1 ml PBS wash 5465 1224 1462 1582 LOD 624 2958 3162 1502 a numbers in italics are limits of detection.

PAGE 113

101 TABLE 21. Sandwich biosensor assay using anti-toxoid A antibody for capture and liposomes at 1:2,500 and 1:5,000 as the detection reagent for the detection of botulinal toxoid type A. Fluorescent readings were recorded for each step of the assay. Numbers in parentheses indicate incubation times. The limit of detection remained high and the botulinum toxoid type A was never utilized in the assay. Channela 1 2 3 4 Liposome concentration 1:2500 1:2500 1:5000 1:5000 Signal (pA) after assay step 0.2 ml liposome (5 min) 2 1 ml PBS wash 1529 1976 2013 1356 0.2 ml PBST (5 min) 2206 3102 2650 1715 2 1 ml PBS wash 953 1485 1790 1426 0.2 ml liposome (5 min) 2 1 ml PBS wash 881 1843 1671 1137 0.2 ml PBST (5 min) 1359 2335 1970 1311 2 1 ml PBS wash 934 1429 1632 1207 0.2 ml liposome (5 min) 2 1 ml PBS wash 1008 1491 1488 1111 0.2 ml PBST (5 min) 1197 1635 1674 1173 2 1 ml PBS wash 903 1257 1372 1120 0.2 ml liposome (5 min) 2 1 ml PBS wash 848 1106 1254 1015 0.2 ml PBST (5 min) 949 1236 1344 1071 2 1 ml PBS wash 832 1037 1196 973 LOD 117 455 596 417 a numbers in italics are limits of detection.

PAGE 114

102 The LODs (Table 20 and 21) were too high to continue with either assay. Although the assay for cholera toxin utilized the ganglioside as a capture molecule, the LOD obtained never exceeded values of 40. When a capture antibody was ut ilized in the botulinum assay, a trend in signals across all four channels was observed: the fluorescence continued to decrease with the addition of liposome followed by subsequent washes. A great deal of variability remained in the background signals and a low LOD was never attained.

PAGE 115

103 TABLE 22. Sandwich biosensor assay using anti-toxoid A antibody for capture and liposomes at 1:10,000 and 1:20,000 as the detection reagent for the detection of botulinal toxoid type A. Fluorescent readings were recorded for each step of the assay. Numbers in parentheses indicate incubation times. The limit of detection remained high and the botulinum toxoid type A was never utilized in the assay. Channela 1 2 3 4 Liposome concentration 1:10000 1:10000 1:20000 1:20000 Signal (pA) after assay step 0.2 ml liposome (5 min) 2 1 ml PBS wash 828 858 933 964 0.2 ml PBST (5 min) 807 933 951 957 2 1 ml PBS wash 770 721 871 884 0.2 ml liposome (5 min) 2 1 ml PBS wash 846 804 929 937 0.2 ml PBST (5 min) 885 844 949 937 2 1 ml PBS wash 789 695 850 872 0.2 ml liposome (5 min) 2 1 ml PBS wash 798 766 926 895 0.2 ml PBST (5 min) 805 790 926 903 2 1 ml PBS wash 750 691 852 863 0.2 ml liposome (5 min) 2 1 ml PBS wash 815 693 864 890 0.2 ml PBST (5 min) 845 693 883 887 2 1 ml PBS wash 839 683 861 868 LOD 133 150 79 53 anumbers in italics are limits of detection.

PAGE 116

104 A 1:10,000 or 1:20,000 dilution of liposome may have eliminated the nonspecific binding or it is possible that a smaller concentration of liposome eliminated the signal completely. An assay was performed with toxoid type A diluted 1:10 in PBS-BSA 0.1% (Table 23). One hundred ug/ml of goat anti-A antibody was used as the capture molecule. Varying liposome concentrations were utilized for the detection of the protein. TABLE 23. Sandwich biosensor assay with varying liposome concentrations for the detection of botulinal toxoid type A. Fluorescent readings were recorded for each step of the assay. Numbers in parentheses indicate incubation times. Buffer was assayed followed by botulinum toxoid type A. Channela 1 2 3 4 liposome 1:1000 1:1250 1:2500 1:5000 LOD 64 196 73 74 Signal (pA) after 5 minute incubation with liposome PBS-BSA 0.1% 30 46 -77 -50 1:10 Type A Toxoid 33 -110 -11 -18 anumbers in italics are limits of detection. Changes in signal values higher than the calculated LOD were treated as positive values. All of the buffer samples generated negative signals (Table 23). Unfortunately,

PAGE 117

105 subsequent incubation with the toxoid yielded negative signals as well. The assay for botulinum toxin with the Analyte 2000 produced LODs that were too high for valid results or negative signals for the protein of interest. The inherent nature of the liposome to bind nonspecifically to the surface of the polystyrene waveguide made it an impractical detection molecule for the botulinum A toxin. Varying liposome concentrations as det ection molecule for the detection of botulinal toxoid type B. Toxoid type A had a higher degree of probability of generating positive signals compared to type B toxoid in the Analyte 2000 assay. All of the initial assays were performed using toxoid type A and the CDC goat anti-A antibody. The results indicate that the utilization of the liposomes as the detection molecule in conjunction with the Analyte 2000 was not successful (Table 20, 21, and 22). The synthesized liposomes were developed for the intended detection of toxoid type B and type A, since both botulinal toxoids utilize the same receptor, GT1b. Considering the results for toxoid type A, it would be improbable that the detection of toxoid type B would be successful. Bearing this in mind, biosensor assays for the detection of toxoid type B (Tables 24 and 25) were attempted utilizing concentrations of liposome at lower and higher concentrations of liposome dilution. The same procedure was followed as that used for toxoid type A with a different capture antibody (biotinylated CDC rabbit anti-B antibody).

PAGE 118

106 TABLE 24. Biosensor assay with liposome concentration at 1:1,000 for the detection of botulinal toxoid type B. Fluorescent readings were recorded for each step of the assay. Numbers in parentheses indicate incubation times. The limit of detection remained high and the botulinum toxoid type B was never utilized in the assay. Channela 1 2 3 4 Liposome concentration 1:1000 1:1000 1:1000 1:1000 Signal (pA) after assay step 0.2 ml liposome (5 min) 2 1 ml PBS wash 4261 2975 10755 10724 0.2 ml PBST (5 min) 6466 4072 11955 11737 2 1 ml PBS wash 2036 1964 7242 8513 0.2 ml liposome (5 min) 2 1 ml PBS wash 3875 3364 7915 8768 0.2 ml PBST (5 min) 4747 4256 8258 9073 2 1 ml PBS wash 1505 1976 6806 8208 0.2 ml liposome (5 min) 2 1 ml PBS wash 2369 2355 6988 8331 0.2 ml PBST (5 min) 3144 2068 7179 8396 2 1 ml PBS wash 998 12466 6027 8248 0.2 ml liposome (5 min) 2 1 ml PBS wash 1385 1605 6792 8158 0.2 ml PBST (5 min) 1509 1458 6858 8153 2 1 ml PBS wash 1004 1087 6635 8076 LOD 3124 1130 1024 123 anumbers in italics are limits of detection

PAGE 119

107 TABLE 25. Biosensor assay with liposome concentrations of 1:10,000 and 1:20,000 for the detection of botulinal toxoid type B. Fluorescent readings were recorded for each step of the assay. Numbers in parentheses indicate incubation times. The limit of detection remained high and the botulinum toxoid type B was never utilized in the assay. Channela 1 2 3 4 Liposome concentration 1:10000 1:10000 1:20000 1:20000 Signal (pA) after assay step 0.2 ml liposome (5 min) 2 1 ml PBS wash 1426 1212 3680 1283 0.2 ml PBST (5 min) 2259 1557 4148 1344 2 1 ml PBS wash 1396 919 2095 946 0.2 ml liposome (5 min) 2 1 ml PBS wash 2346 1816 1684 1101 0.2 ml PBST (5 min) 2903 2017 1888 1268 2 1 ml PBS wash 1255 1171 1205 995 0.2 ml liposome (5 min) 2 1 ml PBS wash 1420 990 1100 1043 0.2 ml PBST (5 min) 1562 1064 1121 1084 2 1 ml PBS wash 937 802 964 904 0.2 ml liposome (5 min) 2 1 ml PBS wash 895 915 958 921 0.2 ml PBST (5 min) 903 940 975 955 2 1 ml PBS wash 665 707 896 887 LOD 741 672 1258 164 anumbers in italics are limits of detection

PAGE 120

108 Low limits of detection (<40 based on cholera assays) were never achieved (Table 24 and 25). Due to the inability to reduce the background fluorescence when utilizing liposomes as the detection molecule, this concluded any further attempt to detect botulinal toxoid.

PAGE 121

109 DISCUSSION The exotoxins of V. cholerae and C. botulinum have different cell specificities and modes of action. Yet, both of these biological toxins can cause serious human disease. V. cholerae has the potential to cause pandemic illness. C. botulinum produces one of the deadliest biological toxins known and can kill within 24 hours. Botulism, and cholera, are rarely seen in the United States and the chances of naturally acquiring one of these illnesses is low; however, both toxins can be purposefully disseminated and cause mass disease. In October of 2001, the world witnessed the response of both public and private sector agencies to the intentional release of Bacillus anthracis spores through the mail system. Public health agencies learned with the release of the spores, they would need access to rapid and sensitive detection methods for the protection of the public health. Conventional laboratory test methods for the identification of potential bioterrorism agents may not provide public health care officials with the rapid results needed to prevent the dissemination of the potential agent. Conventional methods typically require multiple steps: enrichment, isolation and serological and/or biochemical tests. These methods are time consuming, resulting in delayed detection and, more importantly, are difficult to apply in the field. The ability to rapidly and accurately detect contaminated food and water would serve to reduce the spread of disease. The goal of this research

PAGE 122

110 was development of a rapid and sensitive ganglioside-based assay for the identification of cholera and/or botulinum toxins utilizing an evanescent wave biosensor. Conventional serological tests including co-agglutination tests (72, 128), enzymelinked immunosorbent assays (ELISA) (104, 151) membrane-based assays (172) and PCR (52) may be used for rapid identification of the etiological agent. The need for special equipment and trained employees are not the only inherent weaknesses associated with these methods; problems with cross-reactivity can also hinder the assay (104). The polymerase chain reaction (PCR) may also be used to detect bacterial pathogens (52); however, special equipment and trained employees are needed as well as clean facilities (91). Also, PCR lacks the capability of detecting a protein product from the cell. Therefore, PCR would be inefficient at detecting either protein from V. cholerae or C. botulinum Enrichment from a complex matrix can be used to obtain the necessary number of organisms to perform an ELISA or PCR; however, this may take an additional one to two days, thereby negating the test as a rapid method of identification (52, 151). These shortcomings coupled with the realization that positive identification of the organism when using an ELISA or PCR does not give an indication of the toxicity associated with the serotype can make these methods less advantageous than the detection of the toxic protein (64, 91). Prompt diagnosis of cholera or botulism is essential to the patient and the health care worker. Botulism is most effectively treated early after the appearance of symptoms (13, 148). Of the more than 193 recognized Vibrio cholerae serotypes, only the O1 and O139 serotypes are capable of causing epidemic and pandemic cholera (20). Rapid diagnosis of the serotype or the virulence factor is important due to the virulence of the

PAGE 123

111 O1 and O139 serotypes and their potential to cause disease. Identification of the characteristic rice-water stool can allow one to make a presumptive diagnosis; however, a definitive diagnosis relies on the isolation of the organism from either feces or vomit, followed by serological and/or chemical tests (78, 112). The Analyte 2000 fiber optic-based biosensor assay provides an innovative alternative to conventional methods for obtaining near real-time (20-min) results. Fiberoptic biosensors have been utilized in the detection of the fraction 1 antigen of Yersinia pestis (10), Clostridium botulinum toxin A (118), pseudexin and ricin toxin (71, 119) trinitrotoluene (TNT) (90), staphylococcal enterotoxin (169) and bacterial pathogens (37, 38, 43, 86, 92). The biosensor can analyze four samples simultaneously and use a variety of biomolecules for either capture or detection of the Analyte (92). One advantage of the Analyte 2000 is the ability to produce specific results in a complex matrix (37, 38, 39, 86, 171). The CDC currently lists Vibrio cholerae as a category B critical agent (142). A study done by Levine et al with American volunteers showed that different dosages of purified oral CT elicited different responses (88). An oral dose of at least 5 g was required to initiate diarrheal symptoms. Based on the LD50 in mice, 313 mg of CT/100 lb of body weight would be required to cause illness (53). Rowe-Taitt et al utilized ganglioside-based assays for the detection of cholera toxin in an indirect immunoassay using an array biosensor and generated limits of detection of 40 ng/ml and 1 g/ml depending on the species the detection antibody was directed against (143). The sensitivity of the ganglioside-based assay developed in this research was 1 ng/ml. The assay is sensitive and specific but more importantly, repeatable.

PAGE 124

112 The Analyte 2000 uses a polystyrene waveguide as the sample platform. There was always variability with the signals between different waveguides. In order to account for variation among each waveguide due to the manufacturing process, the signals from different waveguides were normalized. Normalization was performed by dividing the sample change in signal by the change in signal of the 100 ng/ml CT standard and multiplying by 100 (85). Normalization allowed for the comparison of values between individual waveguides throughout the entire study. The biosensor assay relies on detection of the analyte based on the distance of the evanescent wave from the surface of the waveguide. Currently, the evanescent wave is propagated approximately 100-1000 nm from the surface of the polystyrene waveguide. Typically, the fluorescence of any fluorophore falling within this distance is reverberated back into the waveguide and the signal is recorded by a computer in picoAmperes (pA). The CT toxin assay developed within this project was consistent and reproducible based primarily on the small size of the protein. Bacteria are large (average E. coli is 1 m wide by 6 m long). A typical sandwich immunoassay consists of an antibody bound to a solid surface, followed by the binding of the analyte and then the attachment of an antibody conjugated to a fluorophore. When the analyte is a bacterium, it has a higher potential for the cell to fall outside of the evanescent wave. In contrast, a small protein like CT typically falls within the evanescent wave, therby produceing repeatable and accurate results. The CT assay developed was successful whether the source of the toxin was commercially purchased or produced naturally by the V. cholerae cells. The ELISA and Analyte 2000 were able to detect both sources of CT.

PAGE 125

113 Environmental sources of cholera toxin include contaminated food and water. Identification of the organism V. cholerae from contaminated water followed by assaying for either cholera toxin or the ctx gene has been reported (96, 108, 175). Most research has focused on assaying for the presence of cells in contaminated food and not the cholera toxin (51, 62, 77, 96, 98, 122, 164). The method developed in this project relies on assaying for CT in a complex matrix, an oyster. Initially, two cross-reactive proteins (with molecular weights of approximately 60 and 100 kD) presented problems in accurately detect the presence of CT. However, fractionating the oyster homogenate via a P10 column eliminated the cross-reactive proteins. The Analyte 2000 was able to detect the presence of CT at a concentration of 5 g/ml in artificially-contaminated oysters. The only information known about the two cross-reactive proteins is their approximate molecular weight. It was unexpected to find two proteins within the oyster matrix that had an affinity for the anti-CT antibody. A future research project could include sequencing these two proteins as an initial step in identifying them. With the ability to induce toxin production in vitro and detect the protein in the oyster tissue using the Analyte 2000, a tank with live oysters was inoculated with a toxinproducing strain of V. cholerae The aim of the assay was to simulate the ocean environment with live oysters, introduce V. cholerae into the system, allow the oysters to concentrate the organism via filter feeding and assay for the production of CT within the oyster. The Analyte 2000 assay would be utilized to detect the toxin. The success of the assay was dependent on the ability of the V. cholerae to grow and proliferate within the tank and the oysters to remain viable. Three different types of media were used to support the growth and viability of the cells and the oysters: Instant Ocean, SWC-DI and

PAGE 126

114 CAYE-IO. Instant Ocean is used to maintain a saltwater tank, SWC is used for enrichment of marine microorganisms, and CAYE was used in this research study for in vitro toxin production. Previous assays developed as part of this project showed that in order for in vitro toxin production to occur, V. cholerae required 24 hours for growth and needed to reach a concentration of approximately 109 cfu/ml. The Instant Ocean tank was able to support V. cholerae growth for 12 hours and the oysters remained viable for up to 72 hours. The maximum cell count obtained from either the oysters or the medium was approximately 104 cfu/ml. Cholera toxin was not detectable in either the oyster or the tank medium. These results were not unexpected in as much as the cell count was five logs lower than needed and the cells were not detectable past 12 hours of inoculation with V. cholerae The tank with the SWC-DI was less capable of supporting V. cholerae growth and CT production than the tank with the Instant Ocean. Vibrio cholerae was undetectable at any time in either the oyster or the tank medium. The oysters in this tank only remained viable for 24 hours while the tank medium itself turned into a viscous form with little to no resemblance of the original medium. CT was not detected in the oysters or the tank medium. The tank with the CAYE-IO has results similar to the SWC-DI tank. Vibrio cholerae and CT were undetectable at any time in either the oyster or the tank medium. This tank also turned into a viscous medium and the oysters were dead by 24 hours. Like the SWC-DI tank, CT went undetected as well. While V. cholerae was not detected by culturing in two of the three tanks, cells may have been present. Extensive research has demonstrated that these cells enter the VBNC state under stressful conditions (7, 26, 66, 73, 157, 187). The physical appearance

PAGE 127

115 of the tank, the death of the oysters and other small crustaceans, as well as the proliferation of other Vibrio spp. and other nonbacterial microorganisms may have placed the cells in a stressful environment. It is also important to recognize the impact on an isolate when removing it from laboratory growth conditions and placing it into an environmental simulation: specific genes may be turned on, off, rearranged or lost (109). It is possible to have the same impact on the genetics of the microorganism when going from environmental conditions to laboratory conditions. Vibrio parahaemolyticus and Vibrio alginolyticus were recovered from both tanks in instances where V. cholerae was not recoverable; these other bacteria could have outcompeted V. cholerae for available nutrients. Molecular methods for the ctxAB gene or fluorescent microscopy could not be used to determine if the V. cholerae cells were present in the tank but possibly entered into the VBNC state because the cells were intentionally inoculated into the tank. Molecular or immunological methods would detect the DNA or antigenic determinants of the cells that were present as a result of deliberate inoculation. It is also important to consider the realistic application of assaying for the toxin in seawater. The dilution factor the ocean would provide must be taken into consideration. The toxin may be secreted into the ocean by toxigenic cells but it would be extraordinarily diluted so that it may never be detected by available conventional methods. Seawater samples could be concentrated to recover CT in low concentrations. The inability to detect CT was not a function of the Analyte 2000 but a function of the tank simulation itself. There were unknown variables in the tank system that did not

PAGE 128

116 allow for the growth of V. cholerae cells. Without the growth of the cells, there is no CT expression. There are many complex interactions required for the growth, proliferation, expression of virulence and transmission of V. cholerae in the ocean environment (95). Research by Raskin et al. indicated that the in vivo signals that affect the ToxR regulon in control of virulence factors in Vibrio cholerae were unknown (131). The regulon can be induced in vitro by different environmental signals such as pH, osmolarity and temperature (131). Taking into account the complex environmental interactions and the lack of knowledge of the signals needed to activate the ToxR regulon, the lack of in vivo toxin induction in the tank simulation was not totally unexpected. All of the tank simulations were performed without a filtration system. The oysters were harvested directly from Tampa Bay and were contaminated with other bacterial and mycotic organisms. The lack of filtration allowed for the accumulation of these microorganisms and waste products in the tank environment. If a filter had been used, then the V. cholerae cells would have been removed from the tank. However, other cells were allowed to grow and proliferate. The proliferation of these organisms may have out-competed the V. cholerae for nutrients, thereby making CT production difficult, if not impossible under these conditions. The initial in vitro induction assays were performed in pure culture. It may be advantageous to return to the pure culture petri dish assays in order to determine the limiting factors in the tank simulation. Future assays could test the effect growth of other organisms found in the tank simulation have on the toxin-producing capabilities of V. cholerae

PAGE 129

117 One of the first assays attempted with the live oysters was to inoculate a single oyster in a large beaker (data not reported) and then assay for CT. The experiment failed because the oyster was unable to survive longer than 6 hours. The development of this assay was dependent on the survival of the oyster and the growth of V. cholerae In spite of the fact that toxin production could not be elicited in the tank simulation, it was induced in vitro in three separate strains of V. cholerae, which provided a natural source of the toxin. A VET-RPLA detection kit (Oxoid) exists on the market that requires 24 hours enrichment of a pure culture of V. cholerae before the agglutination assay can be performed (146, 177). The limit of detection for the kit is currently 1-2 ng/ml of CT (177). The assay developed in this research has the ability to detect CT in the presence of cells without enrichment within a 20 minute time period with a limit of detection of 1 ng/ml. Conventional methods and VET-RPLA require the growth of the organism. Any assay requiring enrichment would take longer than the 20 minute required for the Analyte 2000. Molecular methods such as PCR can be used to assay for virulent V. cholerae (21, 96, 98, 122). However, the inhibitory substances present in complex matrices such as food require sample processing before any testing can be performed (39, 96). The assay developed in this research was not inhibited by any substances which may be found in the food matrix. The only problem encountered with the oyster was the presence of two cross reactive proteins, which was overcome with fractionization of the oyster homogenate. The ability of the Analyte 2000 assay to detect CT in a complex food matrix could be applied to other matrices as well. Fecal material and vomit are two other

PAGE 130

118 complex matrices in which CT may be found. The utilization of the P10 gel column for sample fractionation could possibly allow for the detection of cholera toxin in these matrices. The assay developed in this study showed that, if CT is present at 10 ng/ml or higher in the oyster matrix, the Analyte 2000 has the capability to detect its presence. The ganglioside-based assay is a comparable alternative to an antibody-based assay, (sensitivity of 1 ng/ml). Use of the ganglioside as the capture molecule and the antibody as the detection molecule conferred double specificity to the assay: receptortoxin specificity and toxin-antibody specificity. The causative agent of cholera is ultimately CT itself. Toxin detection in lieu of detecting the organism is a more direct measure of pathogenicity. The assays developed within this project successfully detected the presence of commercial and natural sources of cholera toxin in buffer, in the presence of cells and within the matrix of an oyster. The CDC currently lists Clostridium botulinum toxin as a category A critical agent (142). With an LD50 of 1 ng/kg of body weight in humans (53, 59, 163), a human oral lethal dose for type A toxin at 1 g/kg (47, 163), botulinum toxin is one of the most potent biological toxins available (29, 53, 59, 163). Unlike the cholera toxin assay developed in this research, the botulinum assay would need to be extremely sensitive. Research has shown that detection is possible with a colorimetric ELISA at 20 pg/ml for type A and B toxin ( 1), with the mouse bioassay at a MLD of 10-30 pg (47, 181) and with a ganglioside-liposome immunoassay at a toxin concentration of 15 pg/ml (1). Consequently, the botulinum assay must have a limit of detection on the order of at least 2-3 logs more sensitive than the cholera toxin assay to be competitive with other assays available.

PAGE 131

119 Both the antibodies and type A and B botulinum toxins used for the development of this assay were obtained from the CDC. Exact protein concentrations of the toxins were not available. Although a BCA protein assay was performed to determine toxin protein concentrations, the results were inconclusive. Two of the three antibodies received from the CDC had an affinity for the toxins and the commercial antibody purchased from Accurate displayed no affinity for either toxin. An ELISA verified antibody-toxin specificity; it was therefore determined that the inability to generate a positive signal with the ELISA involved ganglioside-toxin specificity. Initially, indirect ELISAs with the two gangliosides specific for botulinum toxin, GT1b and Gd1b, passively adsorbed to the wells of a microtitre plate were performed like those used to detect CT: neither showed any specificity for the toxin. Detecting botulinum toxins with the use of a ganglioside, GT1b or GD1b, had been achieved when it was incorporated in a liposome (1, 156). Creation of a liposome assay was attempted. Several attempts to prepare liposomes incorporated with the ganglioside failed. Custom liposomes were synthesized for the development of this assay by Dr. Anup Singh at Sandia National Laboratories in Livermore, California (156). GT1b was selected as the ganglioside for incorporation into the liposome. A fluorophore conjugated to a phospholipid was required for the liposome. Cyanine 5 conjugated to a phospholipid was not available commercially; therefore, a fluorophore with similar spectral characteristics to Cyanine 5 was selected. The Cyanine 5 fluorophore has an excitation at 635 nm with an emission at 670 nm and the Alexfluor 647 fluorophore has an excitation at 633/635 nm and an emission at 668 nm. Alexaflou647 conjugated to a phospholipid was available commercially from Molecular Probes and

PAGE 132

120 was selected for incorporation into the custom liposomes. The emission of the Alexafluor 647 was out of the range of the Gemini fluorometer; therefore, ELISAs typically used in assay development were not used to assay for ganglioside-toxin specificity. Initial work on the botulinum assay utilizing the Analyte 2000 was performed using toxoid type A and the CDC-goat anti-type A antibody. Of the five possible antigen-antibody combinations, the CDC-goat anti-type A antibody displayed the highest affinity for its respective toxoid. The protein was passively adsorbed to the polystyrene waveguide followed by incubation with the liposome. It appeared that this assay had some potential in that there was a significant increase, of threeto five fold, in signal for three out of four of the Analyte 2000 channels; however, the negative control channel fluorescence signal doubled. The increase in signal for the negative control was attributed to liposome sticking non-specifically to the polystyrene waveguide. Further attempts were made in assay development with type A toxoid utilizing a biotinylated capture antibody bound to the waveguide via a streptavidin bridge. Assays utilizing this sandwich immunoassay format were unsuccessful as well: the LODs ranged in values from 53-3100 pA. In order to try and reduce the LODs and obtain positive signals several variables were manipulated: the concentration of the liposome was changed, the toxin concentration was varied, specificity was tested using a different ligand, blocking buffers and times were varied and different wash buffers were used. Regardless of the variable altered, the assay continued to produce high LODs and negative results. With the limitations of the assay utilizing type A toxoid, attempts were made to try and develop the assay using type B toxoid. As illustrated previously, the antibody for

PAGE 133

121 the type B toxoid had a lower signal to noise ratio with the ELISA than the one used to capture the type A toxoid. The assay followed the exact same format as the one used for the type A toxoid: a sandwich immunoassay with the liposome as the detection molecule. The assays were unsuccessful with LODs ranging from 163-3100 pA. Limitations on the successful development of this assay for botulinum toxin could not be resolved. The nonspecific binding of the liposomes to the polystyrene waveguide and the inability to reduce the LODs contributed to the failed attempts to develop this assay. Even though the assay itself was not successful, valuable information was obtained from the study. It is not advisable for future assay development with the Analyte 2000 to ut ilize liposomes as detection molecules. The liposomes are expensive compared to most antibodies and they do not produce practical results. The research conducted in this study with liposomes and the Analyte 2000 will hopefully influence others not to utilize them as detection molecules. Even though gangliosides could not be utilized in the botulinum assay, this does not rule out the possibility of using strictly antibodies with this platform as capture and detector molecules. Bioterrorism is a serious threat to the United States and to the world and cholera has re-emerged as a major infectious disease worldwide. Documented events of bioterrorism have occurred since 184 B.C. and, with the advances in science and technology, transportation systems and the availability of select agents, such events remain a possibility today (29). It is with the safety of the public health in mind that both public and private sector entities are focusing on the development of technology to detect biological agents (94). The evanescent wave fiber-optic biosensor represents one form of technology with the capability to produce rapid, sensitive and specific assays for the

PAGE 134

122 detection of CT. This research has shown that gangliosides can be used in a biosensor assay to rapidly detect CT. The Analyte 2000 has the capability to produce near real-time results with minimal sample preparation and the biosensor represents an alternative method for the detection of bacterial products such as toxins.

PAGE 135

123 REFERENCES 1. Ahn-Yoon, S., T. R. DeCory, and R. A. Durst. 2004. Ganglioside-liposome immunoassay for the detection of botulinum toxin. Anal. Bioanal. Chem. 378: 68 75. 2. Akbulut, D., K. A. Grant, and J. McLaughlin. 2004. Development and application of real-time PCR assays to detect fragments of the Clostridium botulinum types A, B, and E neurotoxin genes for investigation of human foodborne and infant botulism. Foodborne Pathog. Dis. 1: 247-257. 3. Arnon, S. S., R. Schechter, T. V. Inglesby, D. A. Henderson, J. G. Bartlett, M. S. Ascher, E. Eitzen, A. D. Fine, J. Hauer, M. Layton, S. Lillibridge, M. T. Osterholm, T. O’Toole, G. Parker, T. M. Perl, P. K. Russell, D. L. Swerdlow, and K. Tonat. 2001. Botulinum toxin as a biological weapon. JAMA. 285: 10591070. 4. Barr J.R., H. Moura, A. E. Boyer, A. R. Woolfitt,S. R. Kalb, and A. Pavlopoulos. 2005. Botulinum neurotoxin detection and differentiation by mass spectrometry. Emerg Infect Dis. Oct [5-13-06]. Available from http://www.cdc.gov/ncidod/EID/ vol11no10/04-1279.htm. 5. Beneson, A. 1991. Cholera, p. 207-225. In A. S. Evans and P. S. Brachman (ed.), Bacterial Infections of Humans Plenum Medical Book Co., New York, N. Y. 6. Bennish, M. K. 1994. Cholera: pathophysiology, clinical features, and treatment, p. 229-256. In I. K. Wachsmuth, P.A. Blake and O. Olsvik (ed.), Vibrio cholerae and Cholera: Molecular to Global Perspectives. ASM Press,Washington, D.C. 7. Binsztein, N., M. C. Costagliola, M. Pichel, V. Jurquiza, F. C. Ramfrez, R. Akselman, M. Vacchino, A. Huq, and R. R. Colwell. 2004. Viable but nonculturable Vibrio cholerae O1 in the aquatic environment of Argentina. Appl. Environ. Microbiol. 70: 7481-7486. 8. Brayton, P. R., and R. R. Colwell. 1987. Fluorescent antibody staining method for enumeration of viable environmental Vibrio cholerae O1. J. Microb. Methods 6: 309-314. 9. Brett, M. M., G. Hallas, and O. Mpamugo. 2004. Wound botulism in the UK and Ireland. J. Med. Microbiol. 53: 555-561.

PAGE 136

124 10. Cao, L. K., G. P. Anderson, F. S. Ligler, and J. Ezzel. 1995. Detection of Yersinia pestis fraction 1 antigen with a fiber-optic biosensor. J. Clin. Microbiol. 33: 336-341. 11. Centers for Disease Control and Prevention. 1991. Cholera associated with imported frozen coconut Milk. Morbid. Mortal. Weekly Rep. 40: 844-845. 12. Centers for Disease Control and Prevention. 1991. Current trends update: cholerawestern hemisphere and recommendations for treatment of cholera. Morbid. Mortal. Weekly Rep. 40: 562-565. 13. Centers for Disease Control and Prevention. 1998. Botulism in the United States, 1899-1996. Handbook for Epidemiologists, Clinicians, and Laboratory Workers Atlanta,Georgia. 14. Centers for Disease Control and Prevention. 2003. Infant botulism – New York City, 2001-2002. Morbid. Mortal. Weekly Rep. 52: 21-24. 15. Centers for Disease Control and Prevention. 2005. Summary of notifiable diseasesUnited States, 2003. Morbid. Mortal. Weekly Rep. 52: 1-85. 16. Centers for Disease Control and Prevention. 2006. Two cases of toxigenic Vibrio cholerae O1 infection after hurricanes Katrina and Rita-Louisiana, October 2005. Morbid. Mortal. Weekly Rep. Synopsis for January 19, 2006. 17. Chaiyanan, S., Chaiyanan, S., A. Huq, T. Maugel, and R. R. Colwell. 2001. Viability of the nonculturable Vibrio cholerae O1 and O139. Syst. Appl. Microbiol. 24: 331-41. 18. Chakraborty, S., A. K. Mukhopadhyay, R. K. Bhadra, A. N. Ghosh, R.Mitra, T. Shimada, S. Yamasaki, S. M. Faruque, Y. Takeda, R. R. Colwell, and G. B. Nair. 2000. Virulence genes in environmental strains of Vibrio cholerae Appl. Environ. Microbiol. 66: 4022-4028. 19. Chapman, E. R. 2002. Synaptotagmin: a Ca2+ sensor that triggers exocytosis? Mol. Cell Biol. 3: 1-3. 20. Chiang, S. L., and J. J. Mekalanos. 1999. Horizontal gene transfer in the emergence of virulent Vibrio cholerae p. 156-164. In E. Rosenberg (ed.), Microbial Ecology and Infectious Disease. ASM Press, Washington, D.C.

PAGE 137

125 21. Chun, J., A. Huq, and R. R. Colwell. 1999. Analysis of 16S-23S rRNA intergenic spacer regions of Vibrio cholerae and Vibrio mimicus Appl. Environ. Microbiol. 65: 2202-2208. 22. Clemens, J., D. Spriggs, and D. Sack. 1994. Public health considerations for the use of cholera vaccines in cholera control programs, p. 425-442. In I. K. Wachsmuth, P.A. Blake and O. Olsvik (ed.), Vibrio cholerae and Cholera: Molecular to Global Perspectives. ASM Press, Washington, D.C. 23. Coffield, J. A., N. M. Bakry, A. B. Maksymowych, and L. L. Simpson. 1999. Characterization of a vertebrate neuromuscular junction that demonstrates selective Resistance to botulinum toxin. J. Pharmacol. Exp. Ther. 289: 1509-1516. 24. Colwell, R. R. 1996. Global climate and infectious disease: the cholera paradigm. Science. 274: 2025-2031. 25. Colwell, R. R. 2002. A voyage of discovery: cholera, climate and complexity. Environ. Microbiol. 4: 67-69. 26. Colwell, R. R. 2004. Infectious disease and environment: cholera as a paradigm for waterborne disease. Intern. Microbiol. 7: 285-289. 27. Colwell, R. R., P. R. Brayton, D. J. Grimes, D. R. Roszak, S. A. Huq, and L. M. Palmer. 1985. Viable, but not-culturable Vibrio cholerae and related pathogens in the environment: implication for release of genetically engineered microorganisms. Bio.Technology 3: 817-820. 28. Colwell, R. R., P. R. Brayton, and D. Herrington. 1996. Viable but nonculturable Vibrio cholerae O1 revert to a cultivable state in the human intestine. World J. Microbiol. Biotechnol. 12: 28-31. 29. Committee on Science and Technology for Countering Terrorism. 2002. Making the nation safer. National Research Council, Washington, D.C. 30. Cooper, J., and T. Cass. 2004. Biosensors 2nd Ed. Oxford University Press, Oxford. 31. Cottingham, K. L., D. A. Chiavelli, and R. K. Taylor. 2003. Environmental microbe and human pathogen: the ecology and microbiology of Vibrio cholerae Front. Ecol. Environ. 1: 80-86. 32. Craven, K. E., Ferreira, J. L., Harrison, M. A., and P. Edmonds. 2002. Specific detection of Clostridium botulinum types A, B, E and F using the polymerase chain reaction. J AOAC Int. 85: 1025-1028.

PAGE 138

126 33. Cuatrecasas, P. 1973. Gangliosides and membrane receptors for cholera toxin. Biochem. 12: 3558-3566. 34. Dalsgaard, A., O. Serichantalergs, A. Forslund, W. Lin, J. Mekalanos, E. Mintz, T. Shimada, and J. G. Wells. 2001. Clinical and environmental isolates of Vibrio cholerae serogroup O141 carry the CTX phage and the genes encoding the toxin-coregulated pilus. J. Clin. Microbiol. 39: 4086-4092. 35. Dawson, R. M. 2005. Characterization of the binding of cholera toxin to ganglioside GM1 immobilized on microtiter plates. J. Appl. Toxicol. 25: 30-8. 36. De, S. N., and D. N. Chatterjee. 1953. An experimental study of the mechanism of action of Vibrio cholerae on the intestinal mucous membrane. J. Pathol. Bacteriol. 66: 559-562. 37. DeMarco, D. R., E. W. Saaski, D. A. McCrae, and D. V. Lim. 1999. Rapid detection of Escherichia coli O157:H7 in ground beef using a fiber-optic biosensor. J. Food Prot. 62: 711-716. 38. DeMarco, D. R. and D. V. Lim. 2001. Direct detection of Escherichia coli O157:H7 in unpasteurized apple juice with an evanescent-wave biosensor. J. Rapid Methods Automat. Microbiol. 9: 241-257. 39. DeMarco, D. R. and D. V. Lim. 2002. Detection of Escherichia coli O157:H7 in 10 and 25-gram ground beef samples with an evanescent-wave biosensor with silica and polystyrene waveguides. J Food Prot. 65: 596-602. 40. DePaola. A. 1981. Vibrio cholerae in marine foods and environmental waters: a literature review. J. Food Sci. 46: 66-70. 41. DePaola. A., and C. A. Kaysner. 2004. Vibrio, p. 1-40. In Bacteriological Analytical Manual Online. FDA/CFSAN. Rockville, MD. 42. DePaola. A., and G. C. Hwang. 1995. Effect of dilution, incubation time, and temperature of enrichment on cultural and PCR detection of Vibrio cholerae obtained from the oyster Crassostrea virginica. Mol. Cell. Probes. 9: 75-81. 43. Donaldson, K. A., M. F. Kramer, and D. V. Lim. 2004. A rapid detection method for vaccinia virus, the surrogate for smallpox virus. Biosensor. Bioelect. 20: 322-327. 44. Dubey, R. S., M. Lindblad, and J. Holmgren. 1990. Purification of El Tor cholera enterotoxins and comparisons with classical toxins. J. Gen. Microbiol. 136: 1839-1847.

PAGE 139

127 45. Dusinberre, W. 2003. Slavemaster President: The Double Career of James Polk. Oxford University Press, Oxford. 46. Epstein, P. R., T. E. Ford, and R. R. Colwell. 1993. Marine Ecosystems. Lancet. 342: 1216-1219. 47. Ferreira, J. L., S. J. Eliasberg, P. Edmonds, and M. A. Harrison. 2004. Comparison of the mouse bioassay and the enzyme-linked immunosorbent assay procedures for the detection of type A botulinal toxin in food. J. Food Prot. 67: 203-206. 48. Fields, P. I., T. Popovic, K. Wachsmuth, and O. Olsvik. 1992. Use of polymerase chain reaction for detection of toxigenic Vibrio cholerae O1 strains from the Latin America cholera epidemic. J. Clin. Microbiol. 30: 2118-2121. 49. Finkelstein, R.A. 1988. Cholera, the cholera enterotoxins, and the cholera enterotoxin-related enterotoxin family, p. 85-100. In P. Owen and T. J. Foster (ed.), Immunolgical and Molecular Genetic Analysis of Bacterial Pathogens. Elsevier Science Publishers, St. Loius. 50. Ghosh, C., R. K. Nandy, S. K. Dasgupta, G. B. Nair, R. H. Hall, and A. C. Ghose. 1997. A search for cholera toxin (CT), toxin coregulated pilus (TCP), the regulatory element ToxR and other virulence factors in non-O1/non-O139 Vibrio cholerae Microb. Pathogen. 22: 199-208. 51. Gil, A. I., V. R. Louis, I. N. G. Rivera, E. Lipp, A. Huq, C. F. Lanata, D. N. Taylor, E. Russek-Cohen, N. Choopun, R. B. sack, and R. R. Colwell. 2004. Occurrence and distribution of Vibrio cholerae in the coastal environment of Peru. Environ. Microbiol. 6: 699-706. 52. Gilbert, C., D. Winters, A. O’Leary, and M. Slavik. 2003. Development of a triplex PCR assay for the specific detection of Campylobacter jejuni, Salmonella spp. and Escherichia coli O157:H7. Mol. Cell. Probes 17: 135-138. 53. Gill, D. M. 1982. Bacterial toxins: a table of lethal amounts. Micro. Rev. 46: 8694. 54. Gonzales-Escalona, N., A. Fey, M. G. Hofle, R. T. Espejo, and A. Guzman. 2006. Quantitative reverse transcription polymerase chain reaction analysis of Vibrio cholerae cells entering the viable but non-culturable state and starvation in response to cold shock. Environ. Microbiol. 8: 658-66. 55. Gugnani, H.C., and S.C. Pal. 1974. Variation in polymyxin sensitivity among colonies in primary plate cultures of Vibrio El Tor. J. Med. Microbiol. 7: 535-536.

PAGE 140

128 56. Han, G.K., and T.S. Khie. 1963. A new method for the differentiation of Vibrio comma and Vibrio El Tor. Am. J. Hyg. 77: 184-186. 57. Harlow, E., and D. Lane. 1999. Using Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory Press, New York., N.Y. 58. Hase, C., and J. J. Mekalanos. 1998. TcpA protein is a positive regulator of virulence gene expression in Vibrio cholerae. Proc. Natl. Acad. Sci. USA. 95: 730-734. 59. Hatheway, C. L. 1990. Toxigenic clostridia. Clin. Microbiol. Rev. 3: 66-98. 60. Holmgren, J. 1973. Comparison of the tissue receptors for Vibrio cholerae and Escherichia coli enterotoxins by means of gangliosides and natural cholera toxoid. Infect. Immun. 8: 851-859. 61. Holmgren, J., and A.-M. Svennerholm. 1973. Enzyme-linked immunosorbent assays for cholera serology. Infect. Immun. 7: 759-763. 62. Hood, M. A., G. E. Ness, and G. E. Rodrirk 1981. Isolation of Vibrio cholera serotype O1 from the eastern oyster, Crassostrea virginica Appl. Environ. Microbiol. 41: 559-560. 63. Horowitz, B. Z. 2005. Botulinum toxin. Crit. Care Clin. 21: 825-39. 64. Hoshino, K., S. Yamasaki, A. K. Mukhopadhyay, S. Chakraborty, A. Basu, S. K. Bhattacharya, G. B. Nair, T. Shimada, and Y. Takeda. 1998. Development and evaluation of a multiplex PCR assay for rapid detection of toxigenic Vibrio cholerae O1 and O139. FEMS Immunol. Med. Microbiol. 20: 201-207. 65. Huq, A. R. R. Colwell, M. A. R. Chowdhury, B. Xu, and R. Montilla. 1996. A simple filtration method for removal of Vibrio cholerae associated with planktonic copepods. Appl. Environ. Microbiol. 62: 2508-2512. 66. Huq, A., R. R. Colwell, R. Rahman, A. Ali, M. A. R. Chowdhury, S. Parveen, D. A. Sack, and R. Russek-Cohen. 1990. Detection of Vibrio cholerae O1 in the aquatic environment by fluorescent-monoclonal antibody and culture methods. Appl. Environ. Microbiol. 56: 2370-2373. 67. Huq, A., E. B. Snall, P. A. West, M. I. Huq, R. Rahman, and R. R. Colwell. 1983. Ecological relationships between Vibrio cholerae and planktonic copepods. Appl. Environ. Microbiol. 45: 275-283.

PAGE 141

129 68. Ihara, H., T. Kohda, F. Morimoto, K. Tsukamoto, T. Karasawa, S. Nakamura, M. Mukamoto, and S. Kozaki. 2003. Sequence of the gene for Clostridium botulinum Type B neurotoxin associated with infant botulism, expression of the C-terminal half of heavy chain and its binding activity. Biochem. Biophys. Acta. 1625: 19-26. 69. Islam, M. S., B. S. Draser, and D. J. Bradley. 1989. Attachment of toxigenic Vibrio cholerae to various freshwater plants and survival with a filamentous green alga, Rhizoclonium fontanum. J. Trop. Med. Hyg. 93: 133-139. 70. Ivnitski, D., I. Abdel-Hamid, P. Atanasov, E. Wilkins, and S. Stricker. 2000. Application of electrochemical biosensors for detection of food pathogenic bacteria. Electroanal. 12: 317-325. 71. James, E. A., K. Schmeltzer, and F. S. Ligler. 1996. Detection of endotoxin Using an evanescent wave fiber-optic biosensor. Appl. Biochem. Biotechnol. 60: 189-202. 72. Jesudason, M. V., C. P. Thangavelu, and M. K. Lalitha. 1984. Rapid screening of fecal samples for Vibrio cholerae by a coagglutination technique. J. Clin. Microbiol. 19: 712-713. 73. Johnston, M. D., and M. H. Brown. 2002. An investigation into the changed physiological state of Vibrio bacteria as a survival mechanism in response to cold temperatures and studies on their sensitivity to heating and freezing. J. Appl. Microbiol. 92: 1066-1077. 74. Kaper, J. B., A. Fasano, and M. Trucksis. 1994. Toxins of Vibrio cholerae. P. 145-176. In I. K. Wachsmuth, P.A. Blake and O. Olsvik (ed.), Vibrio cholerae and Cholera: Molecular to Global Perspectives. ASM Press, Washington, D.C. 75. Karunasagar, I., G. Sugumar, I. Karunasagar, and A. Reilley. 1995. Rapid detection of Vibrio cholerae contamination of seafood by polymerase chain reaction. Mol. Mar. Biol. Biotechnol. 4: 365-368. 76. Kasai, G. J., and W. Burrows. 1966. The titration of cholera toxin and antitoxin in the rabbit ileal loop. J. Infect. Dis. 116: 606-614. 77. Kaysner, C. A., and W. E. Hill. 1994. Toxigeneic Vibrio cholerae O1 in food and water. p. 27-39. In I. K. Wachsmuth, I.K., P.A. Blake and O. Olsvik (ed.), Vibrio cholerae and Cholera: Molecular to Global Perspectives. ASM Press, Washington, D.C.

PAGE 142

130 78. Kay, B. A., C. A. Bopp, and J. G. Wells. 1994. Isolation and identification of Vibrio cholerae O1 from fecal specimens, p. 3-25. In I. K. Wachsmuth, P.A. Blake and O. Olsvik (ed.), Vibrio cholerae and Cholera: Molecular to Global Perspectives. ASM Press, Washington, D.C. 79. Keating, T. A., C. G. Marshall, and Ct. Walsch. 2000. Reconstitution and characterization of the Vibrio cholerae vibriobactin synthetase from VibB, VibE, VibF, and VibH. Biochem. 39: 15522-30. 80. Koch, W. C., W. L. Payne, B. A. Wentz, and T. A. Cebula. 1993. Rapid polymerase chain reaction method for detection of Vibrio cholerae in foods. Appl. Environ. Microbiol. 59: 556-560. 81. Khler, G., and C. Milstein. 1975. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature. 256: 495-497. 82. Kolter, T., R. L. Proia, and K. Sandhoff. 2002. Combinatorial ganglioside biosynthesis. J Biol. Chem. 277: 25859-25862. 83. Kozaki, S., Y. Kamata, T. I. Nishiki, H. Kakinuma, H. Maruyama, H. Takahashi, T. Karasawa, K. Yamakawa, and S. Nakamura. 1998. Characterization of Clostridium Botulinum Type B neurotoxin associated with infant botulism in Jpn. Infect. Immun. 66: 4811-4816. 84. Kozaki, S., Y. Kamata, S. Watarai, T. I. Nishiki, and S. Mochida. 1998. Ganglioside GT1b as a complementary receptor component for Clostridium botulinum neurotoxins. Microb Pathog. 25: 91-99. 85. Kramer, M. F., T. B. Tims, D. R. DeMarco, and D.V. Lim. 2002. Recovery of Escherichia coli O157:H7 from fiber optic waveguides used for rapid biosensor detection. J. Rapid Methods Automat. Microbiol. 10: 93-106. 86. Kramer, M. F., and D. V. Lim. 2004. A rapid and automated fiber optic-based biosensor assay for the detection of Salmonella in spent irrigation water used in the sprouting of sprout seeds. J. Food Prot. 67: 46-52. 87. Kress-Rogers, E. 1996. Handbook of Biosensors and Electronic Noses: Medicine, Food and the Environment. CRC Press, Boca Raton. 88. Levine, M. M., J. B. Kaper, R. E. Black, and M. L. Clements. 1983. New knowledge on pathogenesis of bacterial enteric infections as applied to vaccine development. Microbiol. Rev. 47: 510-550.

PAGE 143

131 89. Lewkowich, I. P., J. D. Campbell, and K. T. Hayglass. 2001. Comparison of chemiluminescent assays and colorimetric ELISAs for quantification of murine IL-12, human IL-4 and murine IL-4: chemiluminescent substrates provide markedly enhanced sensitivity. J. Immunol. Methods 247: 111-8. 90. Ligler, F. S., Golden, J. P., Shriver-Lake, L. C., Ogert, R. A., Wijesuria, D., and G. P. Anderson. 1993. Fiber-optic biosensor for the detection of hazardous materials. Immuno. Meth. 3: 122-127. 91. Lim, D. V. 2000. Rapid pathogen detection in the new millennium. Nat. Food Process. Assoc. J. 2: 13-17. 92. Lim, D. V., 2003. Detection of microorganisms and toxins with evanescent wave fiber-optic biosensors. Proc. IEEE 91: 902-907. 93. Lim, D. V. 2003. Microbiology, 3rd Ed. Kendall/Hunt, Dubuque, IA. 94. Lim, D. V., J. M. Simpson, E. A. Kearns, and M. F. Kramer. 2005. Current and developing technologies for monitoring agents of bioterrorism and biowarfare. Clin. Microbiol. Rev. 18: 583-607. 95. Lipp, E. K., A. Huq, and R. R. Colwell. 2002. Effects of global climate on infectious disease: the cholera model. Clin. Microbiol. Rev. 15: 757-770. 96. Lipp, E. K., I. N. G. Rivera, A. I. Gil, E. M. Espeland, N. Choopun, V. R. Louis, E. Russek-Cohen, A. Huq, and R. R. Colwell. 2003. Direct detection of Vibrio cholerae and ctxA in Peruvian coastal water and plankton by PCR. Appl. Environ. Microbiol. 69: 3676-3680. 97. Lindholm, L., J. Holmgren, M. Wikstrm, U. Karlsson, K. Anderson, and N. Lycke. 1983. Monoclonal antibodies to cholera toxin with special references to cross-reactions with Escherichia coli heat-labile enterotoxin. Infect. Immun 40: 570-576. 98. Lyon, W.J. 2001. TaqMan PCR for detection of Vibrio cholerae O1, O139, nonO1 and non-O139 in pure cultures, raw oysters, and synthetic seawater. Appl. Environ. Micrbiol 67: 4685-4693. 99. Maneval, D. R., R. R. Colwell, S. W. Joseph, R. Grays and S. T. Donata. 1980. A tissue culture method for the detection of bacterial enterotoxins. Methods Cell Science. 6: 85-90. 100. Marshall, C. G., N. J. Hilson, and C. T. Walsch. 2002. Catalytic mapping of the vibriobactin biosynthetic enzyme VibF. Biochem. 41(1): 244-50.

PAGE 144

132 101. Martin, Y. P., and M. A Bianchi. 1980. Structure, diversity and catabolic potentialities of aerobic heterotrophic bacterial populations associated with continous cultures of natural marine phytoplankton. Microbial Ecol. 5: 265. 102. Maksymowych, A. B., M. reinhard, C. J. Malizio, M. C. Goodnough, E. A. Johnson, and L. L. Simpson. 1999. Pure botulinum neurotoxin is adsorbed from the stomach and small intestine and produces peripheral neuromuscular blockade. Infect. Immun. 67: 4708-4712. 103. Maulik, R. A. Reed, and G. C. Shipley. 1995. The 2.4 crystal structure of cholera toxin B subunit penatmer: choleragenoid. J. Mol. Biol. 25: 251-254 104. Meer, R. R., and D. L. Park. 1995. Immunochemical detection methods for Samonella spp., Escherichia coli O157:H7 and Listeria monocytogenes in foods. Rev. Environ. Contam. Toxicol. 142: 1-12. 105. Meldolesi, M. F., P. H. Fishman, S. M. Aloj, L. D. Kohn, and R. O. Brady. 1976. relationship of gangliosides to the structure and function of thyrotropin receptors: their absence on plasma membranes of a thyroid tumor defective in thryotropin receptor activity. P. Natl. Acad. Sci. USA. 73: 4060-4064. 106. Miller, C. A., and P. O. Vogelhut. 1978. Chemiluminescent detection of bacteria: and theoretical limits. Appl. Environ. Micrbiol 35(4): 813-816. 107. Miller, V. L., R. K. Taylor, and J. J. Mekalanos. 1987. Cholera toxin transcriptional activator ToxR is a transmembrane DNA binding protein. Cell. 48: 271-279. 108. Minami, A., S. Hashimoto, H. Abe, M. Arita, T. Taniguchi, T. Honda, T. Miwatani, and M. Nishibuchi. 1991. Cholera enterotoxin production in Vibrio cholerae O1 strains isolated from the environment and from humans in Japan. Appl. Environ. Microbiol. 57: 2152-2157. 109. Mintz, E. D., T. Popovic, and P. A. Blake. 1994. Transmission of Vibrio cholerae O1, p. 345-356. In I. K. Wachsmuth, P.A. Blake and O. Olsvik (ed.), Vibrio cholerae and Cholera: Molecular to Global Perspectives ASM Press, Washington, D.C. 110. Mitchell, P. A., and P. T. Pons. 2001. Wound botulism associated with black tar heroin and lower extremity cellulitis. J. Emerg. Med. 20: 371-375. 111. Murphree, R. L., and M. L. Tamplin. 1995. Uptake and retention of Vibrio cholerae O1 in the eastern oyster, Crassostrea virginica Appl. Environ. Microbiol. 61: 3656-3660.

PAGE 145

133 112. Nair, G. B., and Y. Takeda. 1994. Detection of toxins of Vibrio cholerae O1 and non-O1, p. 53-67. In I. K. Wachsmuth, P.A. Blake and O. Olsvik (ed.), Vibrio cholerae and Cholera: Molecular to Global Perspectives ASM Press, Washington, D.C. 113. Nalin, D. R. 1976. Cholera, copepods and chitinase. Lancet. 2: 958. 114. Nalin, D. R., V. Daya, A. Reid, M. M. Levine, and L. Cisneros. 1979. Adsorption and growth of Vibrio cholerae on chitin. Infect. Immun. 25: 768-770. 115. Nandy, R.K., T.K. Sengputa, S. Mukhopadhyay, and A.C. Ghose. 1995. A comparative study of of the properties of Vibrio cholerae O139, O1 and other non-O1 strains. J. Med. Microbiol. 42: 251-257. 116. Nishiki, T., Y. Tokuyama, Y. Kamata, Y. Nemoto, A. Yoshida, K. Sato, M. Sekiguchi, M. Takahashi, and S. Kozaki. 1996. The high-affinity binding of Clostridium botulinum type B neurotoxin to synaptotagmin II associated with Gangliosides Gt1b/GD1a. FEBS Letters. 378: 253-257. 117. Nishiki, T., Y. Kamata, Y. Nemoto, A. Omori, T. Ito, M. Takahashi, and S. Kozaki. 1994. Identification of protein receptor for Clostridium botulinum type B neurotoxin in rat brain synaptosomes. J. Biol. Chem. 269: 10498-10503. 118. Ogert, R. A., J. E. Brown, B. R. Singh, L. C. Shriver-Lake, and F. S. Ligler 1992. Detection of Clostridium botulinum toxin A using a fiber optic-based biosensor. Anal. Biochem. 205: 306-312. 119. Ogert, R. A., L. C. Shriver-Lake, and F. S. Ligler 1993. Toxin detection using a fiber optic-based biosensor. SPIE. 1885: 11-17. 120. O’Shea, Y. A., F. J. Reen, A. M. Quirke, and E. F. Boyd. 2004. Evolutionary genetic analysis of the emergence of epidemic Vibrio cholerae isolates on the basis of comparative nucleotide sequence analysis and multilocus virulence gene profiles. J. Clin. Microbiol. 42: 4657-4671. 121. O’Sullivan, J. M., and G. McMahon. 2005. Descending polyneuropathy in an intravenous drug user. Eur. J. Emerg. Med. 12: 248-250. 122. Panicker, G., D. R. Call, M. J. Krug, and A. K. Bej. 2004. Detection of pathogenic Vibrio spp. In shellfish by using multiplex PCR and DNA microarrays. Appl. Environ. Microbiol 70: 7436-7444. 123. Pavia, A. T., J. F. Campbell, P. A. Blake, J. D. Smith, T. W. McKinley, and D. I. Martin. 1987. Cholera from raw oysters shipped interstate. JAMA 258: 2374.

PAGE 146

134 124. Pellizzari, R., O. Rossetto, G. Schiavo, and C. Montecucco. 1999. Tetanus and botulinum neurotoxins: mechanism of action and therapeutic uses. Phil. Trans R. Soc. Land. B 354: 259-268. 125. Pelizzola, D., E. Bombardieri, A. Brocchi, G. Cappelli, A. Coli, M. Federghini, M. Giganti, M. Gion, G. Madeddu, and M.L. Maussieri. 1995. How alternative are immunoassay systems employing non-radioisotopic labels? A comparative appraisal of their main analytical characteristics. Quart. J. Nucl. Med. 39: 251-63. 126. Popovic, T., P. I. Fields, and O. Olsvik. 1994.Detection of cholera toxin genes, p. 41-52. In I. K. Wachsmuth, P.A. Blake and O. Olsvik (ed.), Vibrio cholerae and Cholera: Molecular to Global Perspectives. ASM Press, Washington, D.C. 127. Pruzzo, C., R. Tarsi, M. M. Lleo, C. Signoretto, M. Zampini, L. Pane, R. R. Colwell, and P. Canepari. 2003. Persistence of adhesive properties in Vibrio cholerae after long-term exposure to sea water. Environ. Microbiol. 5: 850-858. 128. Rahman, M., D. A. Sack, S. mahmood, and A. Hossain. 1987. Rapid diagnosis of cholera by coagglutination test using 4-hour fecal enrichment cultures. J. Clin. Microbiol. 25: 2204-2206. 129. Ramamurthy, T., S. K. Bhattacharya, Y. Uesaka, K. Horigome, M. Paul, D. Sen, S. C. Pal, T. Takeda, Y. Takeda, and G. B. Nair. 1992. Evaluation of the bead enzyme-linked immunosorbent assay for the detection of cholera toxin directly from stool specimens. J. Clin. Microbiol. 30: 1783-1786. 130. Ramamurthy, T., S. Garg, R. Sharma, S. K. Bhattacharya, G. B. Nair, T. Shimada, T. Takeda, T. Karasawa, H. Kurazano, A. Pal, and Y. Takeda. 1993. Emergence of novel strains of Vibrio cholera with epidemic potential in southern and eastern India. Lancet. 341: 703-704. 131. Raskin, D., J. Bina, and J. J. Mekalanos. 2004. Genomic and genetic analysis of Vibrio cholerae p. 57-62. In ASM News (M. I. Goldberg), ASM Press, Washington, D.C. 132. Reichelt, J.L., and P. Baumann. 1974. Effect of sodium chloride on growth of heterotrophic marine bacteria. Arch. Microbiol. 97: 329-345. 133. Reller, M. E., R. W. Douce, S. E. Maslanka, D. S. Torres, S. R. Manock, and J. Sobel. 2006. Wound botulism acquired in the amazonian rain forest of Ecuador. Am. J. Trop. Med. Hyg. 74: 628-631.

PAGE 147

135 134. Remmers, E. F., R. R. Colwell, and R. A. Goldsby. 1982. Production and characterization of monoclonal antibodies to cholera toxin. Infect. Immun 37: 7076. 135. Rich, R. L., L. R. Hoth, K. F. Geoghegan, T. A. Brown, P. K. LeMotte, S. P. Simons, P. Hensley, and D. G. Myszka. 2002. Kinetic analysis of estrogen receptor/ligand interactions. Proc. Natl. Acad. Sci. USA. 99: 8562-7. 136. Rivera, I. N., E. K. Lipp, A. Gil, N. Choopun, A. Juq, and R. R. Colwell. 2003. Method of DNA extraction and application of multiplex polymerase chain reaction to detect toxigenic Vibrio cholerae O1 and O139 from aquatic ecosystems. Environ. Microbiol. 5: 599-606. 137. Roberts, D. 1992. Growth and survival of Vibrio cholerae in foods. Microbiol. Dig. 9: 24-31. 138. Rogers, K. R., and C.L. Gerlach. 1996. Environmental Biosensors November. Environ Science and Technol. Washington D.C. 139. Roitt, I., J. Brostff, and D. Male. 2001. Immunology, 6th Ed. Mosby, London. 140. Rossetto, O., M. Seveso, P. Caccin, G. Schiavo, and C. Montecucco. 2001. Tetanus and botulinum neurotoxins: turning bad guys into good by research. Toxicon. 39: 27-41. 141. Roszak, D. B., and R. R. Colwell. 1987. Survival strategies of bacteria in the natural environment. Microbiol. Rev. 51: 365-379. 142. Rotz, L.D., Khan, A. S., Lillibridge, S. R., Ostroff, S. M., and J. M. Hughes. 2002. Public health assessment of potential biological terrorism agents. Emerg. Infect. Dis. 8: 225-230. 143. Rowe-Taitt, C. A., J. J. Cras, C. H. Patterson, J. P. Golden, and F. S. Ligler. 2000. A ganglioside-based assay for cholera toxin using an array biosensor. Anal. Biochem. 281: 123-133. 144. Ruppert, E. E., and R. D. Barnes. Invertebrate Zoology 6th Ed. Harcourt College Publishers. Fort Worth, Tx. 145. Sack, D. A., S. Huda, P. K. B. Neogi, R. R. Daniel, and W. M. Spira. 1980. Micortiter ganglioside enzyme-linked immunosorbent assay for Vibrio and Escherichia coli heat-labile enterotoxins and antitoxin. J. Clin. Microbiol. 11: 3540.

PAGE 148

136 146. Said, B., S. M. Scotland, and B. Rowe. 1994. The use of gene probes, immunoassays and tissue culture for the detection of toxin in Vibrio cholerae nonO1. J. Med. Microbiol. 40: 31-36. 147. Sakazaki, R., and K. Tamura. 1971. Somatic antigenic variation in Vibrio cholerae Jpn. J. Med. Sci. Biol. 24: 93-100. 148. Salyers, A. A., and D. D. Whitt. 1994. Bacterial Pathogeneis: A Molecular Approach. ASM Press. Washington, D.C. 149. Sarkar, A., R. K. Nandy, G. B. Nair, and A. C. Ghose. 2002. Vibrio pathogenicity island and cholera toxin genetic element-associated virulence genes and their expression in non-O1 non-O139 strains of Vibrio cholerae Infect. Immun. 70: 4735-4742. 150. Schoenberg, B. E. 1974. Snow on the water of London. Mayo Clin. Proc. 49: 680-684. 151. Sewell, A. M., D. W. Warburton, A. Boville, E. F. Daley, and K. Mullen. 2003. The development of an efficient and rapid enzyme-linked fluorescent assay method for the detection of Listeria spp. from foods. Int. J. Food Microbiol. 81: 123-129. 152. Shangkuan, Y. H., C. M. Tsao, and H. C. Lin. 1997. Comparison of Vibrio cholerae O1 isolates by polymerase chain reaction fingerprinting and ribotyping. J. Med. Microbiol. 46: 941-948. 153. Sharma, S. K., J. L. Ferreira, B. S. Eblen, and R. C. Whiting. 2006. Detection of type A, B, E, and F Clostridium botulinum neurotoxins in foods by using an amplified enzyme-linked immunosorbent assay with digoxigenin-labeled antibodies. Appl. Environ. Microbiol. 72: 1231-1238. 154. Shriver-Lake, L. C., G. P. Anderson, J. P. Golden, R. A. Ogert, D. Wijesuria, L. Cao, J. M. Mauro, and F. S. Ligler. 1994. Fiber-optic sensor detection of biological agents, p. 47-54. In K. C. Bailey (ed.), Director’s Series on Proliferation. Seiten. Springfield, VA. 155. Simpson, J. M., and D.V. Lim. 2005. Rapid PCR confirmation of E. coli O157:H7 after evanescent wave fiber optic biosensor detection. Biosensor. Bioelect. 21: 881-887. 156. Singh, A. K., S. H. Harrison, and J. S. Schoeniger. 2000. Gangliosides as receptors for biological toxins: development of sensitive fluoroimmunoassays using ganglioside-bearing liposomes. Anal. Chem. 72: 6019-6024.

PAGE 149

137 157. Singleton, F. L., R. W. Attwell, M. S. Jangi, and R. R. Colwell. 1982. Effects of temperature and salinity on Vibrio cholerae growth. Appl. Environ. Microbiol. 44: 1047-1048. 158. Snow, J. 1849. On the pathology and mode of communication of cholera. London Medical Gazette. 44: 745-52;923-29. 159. Sobel, J., N. Tucker, A. Sulka, J. McLaughlin, and S. Maslanka. 2004. Foodborne botulism in the United States, 1990-2000. Emerg. Infect. Dis. 10: 1606-1611. 160. Soelberg, S. D., T. Chinowsky, G. Geiss, C. B. Spinelli, R. Stevens, S. Near, P. Kaufmann, S. Yee, and C. E. Furlong. 2005. A portable surface plasmon resonance sensor system for real-time monitoring of small to large analytes. J. Ind. Microbiol. Biotechnol. 32(11-12): 669-74. 161. Solomon, H. M., and T. Lilly. 2001. Clostridium botulinum, In Bacteriological Analytical Manual Online FDA/CFSAN. Rockville. 162. Song, X., and B. I. Swanson. 1999. Direct, ultrasensitive, and selective optical detection of protein toxins using multivalent interactions. Anal. Chem. 71: 20972107. 163. Sugiyama, H. 1980. Clostridium botulinum neurotoxin. Microbial. Rev. 44: 419448. 164. Sousa, O. V., R. H. S. Vieira, F. G. R. Menezes, C. M. F. Reis, and E. Hofer. 2004. Detetction of Vibrio parahaemolyticus and Vibrio cholerae in oyster, Crassostrea rhizophorae collected from a natural nursery in the Coco river estuary, Fortaleza, Ceara, Brazil. Rev. Inst. Med. Trop. S Paulo. 46: 59-62. 165. Svennerholm, A., and J. Holmgren. 1978. Identification of Escherichia coli heat-labile enterotoxin by means of a ganglioside immunosorbent assay (GM1ELISA) procedure. Current Microbiol. 1: 19-23. 166. Svennerholm, A.-M., M. Wikstrm, M. Lindblad, and J. Holmgren. 1986. Monoclonal antibodies to Escherichia coli heat-labile enterotoxins: neutralizing activity and differentiation of human and porcine LTs and cholera toxin. Med. Biol 64: 23-30. 167. Swerdlow, D. L., and A. A. Ries. 1993. Vibrio cholerae non-O1-the eighth pandemic? Lancet. 342: 382-383.

PAGE 150

138 168. Tamplin, M. L., and G. M. Capers. 1992. Persistence of Vibrio vulnificus in tissues of gulf coast oysters, Crassostrea virginica exposed to seawater disinfected with UV light. Appl. Environ. Microbiol. 58: 1506-1510. 169. Templeman, L., K. D. King, G. P. Anderson, and F. S. Ligler. 1996. Quantitating staphylococcal enterotoxin B in diverse media using a portable fiber-optic biosensor. Anal. Biochem. 233: 50-57. 170. Tims, T. B., and D. V. Lim. 2003. Confirmation of viable E. coli O157:H7 by enrichment and PCR after rapid biosensor detection. J. Microbiol. Methods. 55: 141-147. 171. Tims, T. B., and D. V. Lim. 2004. Rapid detection of Bacillus anthracis spores directly from powders with an evanescent wave fiber-optic biosensor. J. Microbiol. Methods. 59: 127-130. 172. Toma, C., L. Sisavath, and M. Iwanaga. 1997. Reversed passive latex agglutination for detection of toxigenic Corynebacterium diphtheriae J. Clin. Microbiol. 35: 147-149. 173. Towbin, H. T., Staehlin, T., and J. Gordon. 2005 Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheet: procedures and some applications. Proc. Natl. Acad. Sci. USA. 76: 4350-4354. 174. Tsai, B., J. M. Gilbert, T. Stehle, W. Lencer, T. L. Benjamin, and T. A. Rapoport. 2003. Gangliosides are receptors for murine polyoma virus and SV40. EMBO. 22: 4346-4355. 175. Twedt, R. M., J. M. Madden, J. M. Hunt, D. W. Francis, J. T. Peeler, A. P. Duran, W. O., Hebert, S. G. McCay, C. N. Roderick, G. T. Spite, and T. J. Wazenski. 1981. Characterization of Vibrio cholerae isolated from oysters. Appl. Environ. Microbiol. 41: 1475-1478. 176. Venkateswaran, K., T. Takai, I. M. Navarro, H. Nakano, H. Hasimoto, and R. J. Siebeling. 1989. Ecology of Vibrio cholerae non-O1 and Salmonella spp. and role of zooplankton in their seasonal distribution in Fukuyama coastal waters, Jpn. Appl. Environ. Microbiol. 55: 1591-1598. 177. VET-RPLA Toxin Detection Kit. Oxoid. Basingstoke. Available at http://www.oxoid.com/uk. 178. Wadkins, R. M., J. P. Golden, and F. S. Ligler. 1995. Calibration of biosensor response using simultaneous evanescent wave excitation of cyanine-labeled captured antibodies and antigens. Anal Biochem. 232: 73-78.

PAGE 151

139 179. Waldor, M. K., and J. J. Mekalanos. 1994. ToxR regulates virulence gene expression in non-O1 strains of Vibrio cholerae that cause epidemic cholera. Infec. Immun. 62: 72-78. 180. Weissman, J. B., W. E. DeWitt, J. Thompson, C. N. Muchnick, B. L. Portnoy, J. C. Feeley, and E. J. Gangarosa. 1974. A case of cholera in Texas, 1973. Am. J. Epidemiol. 100: 487-490. 181. Wictome, M., K. A. Newton, K. Jameson, P. Dunnigan, S. Clarke, J. Gaze, A. Y. Tauk, K. A. Foster, and C. C. Shone. 1999. Development of in vitro assays for the detection of botulinum toxins in foods. FEMS Immunol. Med. Microbiol. 24: 319-323. 182. Wilson, W. J., C. L. Strout, T. Z. DeSantis, J. L. Stilwell, A. V. Carrano, and G. L. Anderson. 2002. Sequence-specific identification of 18 pathogenic microorganisms using microarray technology. Mol. Cell. Probes 16: 119-127. 183. World Health Organization. Manual for the laboratory methods for the diagnosis of epidemic dysentery and cholera. Division of Bacterial and Mycotic Diseases. http://www.cdc.gov/ncidod/dhmd/diseaseinfo/cholera-lab-manual.htm 184. World Health Organization. 1992. Guidelines for Cholera Control (rev. 1992). WHO/CDD/SER 80.4 Rev. 3. World Health Organization, Geneva. 185. World Health Organization. 2005. Cholera, 2004. Weekly epidemiological record. 31: 261-268. 186. Wyvill, J. C., and D. Gottfried. 2004. Innovative biosensors are opening up new frontiers. PoultryTech 16(3): 1-2. 187. Xu, H.-S., N. Roberts, F. L. Singleton, R. W. Atwell, D. J. Grimes and R. R. Colwell. 1982. Survival and viability of non-culturable Escherichia coli and Vibrio cholerae in the estuarine marine environment. Microb. Ecol. 8: 313-323. 188. Yalow, R.S., and S. A. Berson. 1960. Immunoassay of endogenous plasma insulin in man. J. Clin. Invest. 39: 1157-1175. 189. Yamashoji, S., A. Asakawa, S. Kawasaki, and S. Kawamoto. 2004. Chemiluminescent assay for detection of viable microorganisms. Anal. Biochem 333: 303-308. 190. Yowler, B. C., R. D. Kensinger, and C. L. Schengrund. 2002. Botulinum neurotoxin A activity is dependane upon the presence of specific gangliosides in neuroblastoma cells expressing synaptotagmin I. J. Biol. Chem. 277: 3281532819.

PAGE 152

140 191. Zhang, R. G., D. L. Scott, M. L. Westbrook, S. Nance, B. D. Spangler, G. C. Shipley, and E. M. Westbrook. 1995. The three dimensional structure of cholera toxin. J. Mol. Biol. 251: 563-573. 192. Zhang, R. G., M. L. Westbrook, E. M. Westbrook, D. L. Scott, Z. Otwinowski, P. R. Maulik, R. A. Reed, and G. G. Shipley. 1995. The 2.4 structure of cholera toxin B subunit pentamer: choleragenoid. J. Mol. Biol. 251: 550-562. 193. Zilinskas, R. A. 1997. Iraq’s biological weapons. The past as future? JAMA. 278: 418-424.

PAGE 153

About the Author Crystal Bedenbaugh received a Bachelor’s Degree in Microbiology and Cell Science from the University of Florida in 1992 and a M.A. in Secondary Science Education from the University of South Florida in 2000. She entered the Ph.D. program at the University of South Florida in 2000. She started teaching in 1994 and while in the Ph.D. program at the University of South Florida she continued on as a teaching assistant in the biology department. She has presented several research presentations at regional and national meetings of the American Society for Microbiology.