Louisiana State University LSU Digital Commons
LSU Historical Dissertations and Theses Graduate School
1974 Identification of Pathogenic Microorganisms in Seafood by Gas Chromatography. William Earnest Mccullough Louisiana State University and Agricultural & Mechanical College
Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses
Recommended Citation Mccullough, William Earnest, "Identification of Pathogenic Microorganisms in Seafood by Gas Chromatography." (1974). LSU Historical Dissertations and Theses. 2621. https://digitalcommons.lsu.edu/gradschool_disstheses/2621
This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. 74-24,790 NcCULLOUGH, William Earnest, 1937- IDENTIFICATION OF PATHOGENIC MICROORGANISMS IN SEAFOOD BY GAS CHROMATOGRAPHY. The Louisiana State University and Agricultural and Mechanical College, Ph.D., 1974 Food Technology
University Microfilms, A XEROX Company, Ann Arbor, Michigan
© 1974
WILLIAM EARNEST McCULLOUGH
ALL RIGHTS RESERVED IDENTIFICATION OF PATHOGENIC MICROORGANISMS IN SEAFOOD BY GAS CHROMATOGRAPHY
A Dissertation
Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy
in
The Department of Food Science
by William Earnest McCullough B.S., Louisiana College, 1963 M.S., Louisiana Polytechnic University, 1965 May 1974 ACKNOWLEDGMENT
The author wishes to express sincere appreciation to his Major
Professor, Dr. M.R. Ramachandra Rao, for his direction, counsel, and assistance during this investigation and during the preparation of this dissertation. Appreciation is also expressed to Dr. Arthur F.
Novak, Head of the Department of Food Science, for the opportunity
to perform this investigation in the department.
Gratitude is expressed to Dr. Arthur F. Novak, Dr. Ronald J.
Siebeling, Dr. William H. James, Dr. Joseph A. Liuzzo, Dr. Robert
M. Grodner and Dr. Fred H. Hoskins for serving on his examining
committee.
He wishes to thank Dr. Robert M. Grodner, Dr. James Rutledge,
The National Shrimp Dreaders and Processors, Inc., and his fellow
graduate students, John Kendall, Mike Moody, Dill Hopkins, Dill Hurst
and Sidney Crow, for their helpful suggestions during the course of
this research.
The author wishes to express thanks to the Office of Naval
Research and to the U.S. Public Health Service for financial support
during this study. Deep appreciation is expressed to his wife,
Jeannette, for her financial support and counsel throughout his
education.
ii TABLE OF CONTENTS
ACKNOWLEDGMENT...... ii
LIST OF TABLES...... vi
LIST OF FIGURES...... vii
ABSTRACT...... ix
INTRODUCTION...... 1
REVIEW OF LITERATURE...... 3
I. Breading as a Source of Contamination...... 4
II. Staphlococcus in Seafood...... 7
III. E. coli as an Indicator...... II
IV. Various Tests for Contamination...... 13
V. Detection of Salmonella...... 23
VI. Gas Chromatography...... 27
MATERIALS AND METHODS...... 33
I. Materials...... 33
A. Source of Sample...... 33
B. Glassware...... 34
C. Reagents...... 36
D. Bacto-Brain Heart Infusion Agar...... 36
E. Bacto-Brain Heart Infusion Broth...... 36
F. Tryptone Glucose Extract Agar...... 36
G. Butterfield's Phosphate Buffer...... 36
H. Nutrient Agar...... 3 ?
I. Eugon Agar...... 37
iii J. Tryptic Mannitol Meat Broth...... 37
K. Trypticase Soy Broth...... 37
L. Vogel and Johnson Agar...... 37
M. Coagulase Plasma...... 38
N. Bacto-Lauryl Tryptone Broth...... 38
0. Simmons Citrate Agar...... 38
P. Bacto-Brilliant Green Bile 2 % ...... 39
Q. Levine Eosin Methylene Blue Agar...... 39
R. M.R.-V.P. Medium...... 40
S. Tryptone...... 40
T. SIM Medium...... 40
U. Bacto-Tetrathionate Broth Base ...... 41
V. Bacto-Triple Sugar Iron Agar...... 41
W. Bacto-Salmonella Shigella Agar...... 41
X. Bacto-Urea Broth...... 42
Y. Bacto-Bismuth Sulfite Agar...... 42
Z. Growth Media...... 4^
AA. Gas Chromatograph...... 43
II. Methods...... 44
A. Sterilization...... 44
B. Shrimp Homogenate...... 44
C. Total Plate Count...... 45
D. Isolation of E. coli from Shrimp...... 46
E. Completed Test on E. coli...... 4^
a. Indole Test...... 48
b. Methyl Red Test...... 48
c. Vogus-Proskauer Teat...... 49
(I. Citrate Test...... 49
iv F. Standardization of Inoculum-Staphylocoecus aureus...... 49
G. Most Probable Number Technique for Coagulase- Positive S. aureus...... 51
H. Standardization of Inoculum-Salmonella...... 53
I. Isolation of Salmonella...... 53
J. Gas Chromatography...... 56
RESULTS...... 60
DISCUSSION...... 95
SUMMARY...... 101
LITERATURE CITED...... 103
VITA...... 113
v LIST OF TABLES
TABLE Page
1. Characteristics of Test Organisms...... 35
2. Food Sample Retention Times...... 61
3. Peak Retention Times...... 94
vi LIST OF FIGURES
FIGURE Page
1. Coliform-Fecal coliform AOAC method...... 47
2. MPN technique for coagulase-positive S. aureus...... 52
3. Gas chromatograms of Media I and Media II on Machine 1...... 63
4. Gas chromatograms of Media I containing n-butanol, Media II containing n-butanol and Media II containing redistilled Anisole...... 65
5. Gas chromatograms of E. coli in Media I and Media II on Machine I and Machine II...... 66
6. Gas chromatograms of E. coli in Media II on Machine II after I hr, 2 hr, and 10 hr incubation at 35°G...... 68
7. Gas chromatograms of Enterobacter aerogenes in Media II on Machine II...... 69
8. Gas chromatograms of Enterobacter aerogenes in Media II on Machine II...... 70
9. Gas chromatograms of Salmonella typhimurium in Media II on Machine II...... 71
10. Gas chromatograms of Salmonella thompson in Media II on Machine II after 18 hr and 48 hr incubation at 35°C...... 72
11. Gas chromatograms of Salmonella senftenberg in Media II developed on Machine II...... 73
12. Gas chromatograms of Salmonella anatum in Media II on Machine II...... 75
13. Gas chromatograms of a mixture of pure cultures of E. coli, Salmonella typhimurium, and Enterobacter aerogenes in Media II on Machine II...... 76
14. Gas chromatograms of Staphylococcus aureus in Media II with 1 % dextrose and Media II w i t h l k mannitol on Machine I I...... 77 IS. Gas chromatograms of green headless shrimp in Media II on Machine II...... 78
16. Gas chromatograms of green headless shrimp sample, 401-1, inoculated with Enterobacter aerogenes in Media II on Machine II...... 79
17. Gas chromatograms of green headless shrimp sample, 401-1M, inoculated with Enterobacter aerogenes in Media II con taining 1 % mannitol on Machine II...... 81
18. Gas chromatograms of green headless shrimp sample, 402-1, inoculated with Salmonella anatum in Media II on Machine II..82
19. Gas chromatograms of breaded shrimp sample, 500, in Media II on Machine II...... 83
20. Gas chromatograms of peeled and deveined shrimp sample, 502, in Media II on Machine II...... 84
21. Gas chromatograms of breaded shrimp sample, 503, in Media II on Machine II...... 85
22. Gas chromatograms of breaded shrimp sample, 600, in Media II on Machine II...... 87
23. Gas chromatograms of breaded shrimp sample, 603, in Media II on Machine II...... 88
24. Gas chromatograms of fresh shrimp sample, 700, in Media II on Machine II...... 89
25. Gas chromatograms of breaded fish sample, BKW, in Media II on Machine II...... 90
26. Gas chromatograms of breaded fish sample, BKC, in Media II on Machine II...... 91
27. Gas chromatograms of commercial oyster sample in Media II on Machine II...... 93
viii ABSTRACT
Gas-liquid partitioning of head-space vapors over dextrose fermenting bacterial cultures were performed on samples of frozen green headless, frozen peeled and deveined and frozen breaded shrimp as well as frozen breaded fish portions and fresh canned oysters.
The samples of seafood were supplied by commercial processors and the National Shrimp Breaders and Processors, Inc.
Organisms representing the following genera were studied in pure cultures: Salmonella, Enterobacter, Escherichia, and
Staphylococcus. The metabolites produced in head-space vapors over synthetic media cultures were analyzed by gas-liquid-chromatography in an effort to obtain "signatures" differentiating the microorganisms.
Seafood materials were inoculated with test organisms and examined by chromatographic procedures.
Analysis of volatile materials produced during growth of dextrose fermenting bacterial cultures produced patterns differen tiating various members of the family Enterobacteriaceae. The shrimp samples inoculated with test organisms presented patterns similar to those produced by Enterobacter aerogenes and Salmonella anaturn. Gas chromatographic analysis of uninoculated seafood samples indicated the presence of members of the family Enterobacter but failed to demonstrate the presence of a specific organism due to interference by contaminating organisms. The inability to produce chromatographic patterns for
Staphylococci over dextrose and the inability to determine total plate counts from chromatograms are two limitations of this procedure.
This investigation was performed as a part of a cooperative
study between the seafood industry, the U.S. Food and Drug Admin
istration and members of the Louisiana State University Department of Food Science to screen rapid procedures which could be used for the development of microbial standards in commercial seafood products.
This method gives promise as a reproducible method and can be applied.to routine mcirobiological examination of seafood.
x INTRODUCTION
Fresh excellent quality seafood taken from its natural habitat, rapidly loses its natural characteristics with handling and storage.
Thus, it is necessary to process seafood rapidly to preserve its distinct qualities for the consumer. This rapid processing now necessitates development of rapid methodology for microbial analysis.
The seafood industry, the U.S. Food and Drug Administration and the
Department of Food Science, Louisiana State University have under taken a cooperate study to develop and test rapid methodology for the detection of microorganisms of public health significance.
Methods of determining the qualitative and quantitative existence of microorganisms of public health importance in foods have remained essentially the same for decades. These ancient methods are rapidly becoming obsolete due to the decrease in time required with today's rapid processing techniques.
Bassette et al. (1966) have shown that microbial metabolites of fermentable substrates can be fingerprinted by gas liquid chroma tography to determine the species of microorganisms. Quantitative determinations are made by the comparison of total plate counts with total peak area calculated from a strip chart recorder.
The purpose of this study was to demonstrate the utilization of "fingerprinting" in the detection of low levels of microbial contamination in seafood. Parameters of food microbiology and chemistry affecting food safety are rapidly being explored. The refinement of these parameters to produce the necessary application and accuracy is an ever challenging job for the food scientist. REVIEW OF LITERATURE
Spoilage by bacterial contamination in both fresh and stored foods is a continuing problem for food processors. By the time contamination is detected by visual or olfactory inspection, the contaminated product is beyond the point where remedial action can be taken to preserve its fitness for human consumption. Large quan- tites of convenience foods are now being mass produced and widely distributed. Many of these foods require only defrosting, rehydrating or warming prior to serving. Because of these circumstances, the marketing of a single contaminated lot could endanger the health of thousands of people.
The public health record of the frozen food industry is re assuring, but it is no guarantee against future difficulties, especially where the potential exists for abuse by the distributors of purchaser.
In recent years many new frozen pre-cooked fishery products and specialities have found a ready market in the United States as a "convenience food." From a microbiological viewpoint, it is highly possible that these commercially prepared fishery products may suffer from excessive handling, improper processing, and generally unsanitary conditions.
Shrimp undergo a variety of processing techniques depending on the type product produced. After receipt from the trawler, the shrimp are deheaded, washed, and frozen. Peeled and deveined shrimp are deheaded, peeled, deveined, washed, blanched in boiling water for approximately 50 seconds, and then frozen individually. Breaded shrimp are deheaded, peeled, deveined, coated with a liquid batter, and sprinkled with a dry breading material. They are then frozen
(Surkiewicz et al. 1967).
The term breading commonly includes bread or cracker crumbs containing such products as nonfat dry milk solids, toasted cereals, eggs, leavening, shortening, dextrose, monosodium glutamate and malt extract.
Larkin et al. (1956) and Kern (1957) have suggested that bread ing may be a major contributor of bacterial contamination in pre cooked frozen fishery products. Bemarde (1958) obtained breading from three sources: from the manufacturer, at retail outlets, and from processors of fish and shellfish products as they were being breaded. The samples were subjected to a series of microbial examinations to discover the number and types of contaminating microflora.
Bemarde (1958) concluded that breadings contribute to bacterial contamination of pre-cookod frozen fishery products after the bread ings have been contaminated by handlers during processing. The pre-cooked process destroys the bacteria which contaminate the bread ing during processing as evidenced by very low bacterial populations of the retail samples.
Several authors have surveyed frozen raw breaded fishery products with alarming results. Foster (1887) was the first to describe bacteria which grew at temperature near the freezing point of water.
Griffiths and Stansby (1934), Kiser (1944), Reay (1935) and Shewan
(1944) found that the bacterial count steadily increased throughout ice storage with subsequent product decomposition in fish flesh.
Griffiths and Stansby (1934) found that when the concentration of 7 the microflora of the fish reached 1.0x10 per g the fish were no longer marketable. Fitzgerald and Conway (1937), Harrison et al.
(1926), Hunter (1920), Hunter and Linden (1923), Reag (1935) and
Shewan (1944) stated that when the bacterial count on market fish exceed 1.0x10^ per g, the product quality is poor.
Fieger and Novak (1961) stated that shelf life of refrigerated seafoods is closely associated with its microbial population. They said that a quantitative measurement of the total microbial popula tion closely correlates with the degree of spoilage that has taken place and by making this determination initially, one can estimate closely the shelf life of a product. Velankar and Govindon (1957,
1958) stated that the rapid spoilage in shellfish, as compared to fish, was caused by the greater amounts of free amino acids found in shellfish. Fieger and Novak (1961) said that spoilage in shellfish resulted from enzymatic action from the tissues as well as from the microorganisms present. Green (1949) found the average bacterial count on 26 samples of market shrimp, purchased from retail dealers, to be 3.3xloVg.
In an analysis of 27 samples of frozen raw breaded shrimp,
Gunderson et al. (1954) found 9.5x10^-14x10^ organisms per g and
150-14x10^ colifonn organisms per g. Although all samples contained coagulase-positive staphylococci, salmonellae were absent from all samples,
Kachikian et al. (1959) surveyed 144 commercial samples of frozen and breaded shrimp from 24 producers and reported that total aerobic 3 6 counts varied from 22x10 to 54x10 organisms per g; 68% of the samples contained less than 100 coliform bacteria per g. Bacillus,
Micrococcus. Achromobacter. and Aerobacter were the dominant groups of bacteria present while the Salmonella-Shigella group was not found in any of the samples.
Silverman et al. (1961) examined 91 samples of commercial frozen shrimp for total bacteria count, coliforms, coagulase-positive staphylococci, and members of the Salmonella-Shigella-typhoid group.
The samples tested were raw or cooked and had either been headed, shelled or left unshelled. Total bacterial counts ranged from 50 organisms per g to over 9x10^ g. No coliforms were detected in the precooked samples tested. Seventy five percent of the frozen shrimp samples contained coagulase-positive staphylococci, present as a small percentage of the total microbial flora.
Isolations of 1 sample were positive to Salmonella polyvalent sera, while 4 were weakly positive. Sliigella organisms were presump tively present in 16 samples tested.
Studies on the microbiology of fishery products have proceeded on two avenues. Of paramount importance is the detection and control of pathogenic microorganisms. The other general area relates to the microbiological quality of commercial food products. Because it is manifestly impossible to examine a food product for every possible pathogenic microorganism, the concept of "indicator" organisms has been used since the early days of food microbiology. Inherent in this concept is the recognition that sanitation is inseparably linked to the wholesomeness and microbiological safety of foods, Leininger et al. (1971). Those organisms established by the National Shrimp Breaders and
Processors, Inc. as of public health importance in shrimp are the
Salmonella and coagulase-positive staphylococci. As indicators of
possible contamination with pathogenic organisms, the total coliform
group, including Escherichia coli, have been used.
Some strains of staphylococci, notably the coagulase-positive
strains, are capable of producing a protein enterotoxin. Several antigenically distinct types of enterotoxin (A, B, C, C2 , D) are
known at this time, Casman et al. (1963). Dack in 1930 proved that
ingestion of the enterotoxin resulted in acute gastrointestinal up set. The enterotoxin is heat stable and can withstand boiling for
30 minutes without loss of its ability to cause vomiting.
Coagulase-positive £5. aureus has been incriminated as the causative agent of food poisoning in a wide variety of foods (Hodge
1960; Dack 1956). Hodge (1960) reported that the vast majority of food poisoning cases were caused by cooked, high protein foods.
Furthermore, lie reported that 94% of the outbreaks that he studied, leftover foods were responsible. Some of the foods incriminated in the outbreaks of staphylococcal intoxication are ham and other meat products, poultry and poultry dressing, sauces and gravies, potato salad, bread pudding, cheese and milk (Bryan 1968).
Staphylococcus aureus is widely distributed in nature. Between
40 and 50% of humans are nasal carriers of the organism and it can be isolated from the air. These two sources rate as the main vectors of food contamination, with humans being far more important.
Bryan (1968) summarized the conditions necessary for occurence of staphylococcal intoxication as follows: (l) there must be a reservoir for the enterotoxigenic strain of aureus (the nose or
hands of food handlers); (2) a mode of dissemination of the organism
(handling of food); (3) contamination of a food that is capable of
supporting the growth of the organism (food must not be too acid and
must be relatively free of competing organisms); (4) a temperature
level for a length or time sufficient to permit adequate multipli
cation of the organism and toxin production and (5) consumption of
a sufficient amount of enterotoxin by a susceptible person.
Kendall (1969) demonstrated the presence of coagulase-positive
S. aureus in shrimp after irradiation of 0.1 Mrad destroyed over
90% of normal microbial flora of shrimp.
The microflora of Gulf Coast shrimp was studied by Green (1949)
and Williams et al. (1952). They reported that the predominant types of organisms present in fresh shrimp were Achromobacter, Bacillus,
Micrococcus, and Pseudomonas. After ice storage, the Achromobacter were predominant.
The enumeration of S^. aureus in food products is complicated
by the presence of large numbers of other organisms. For this reason, media proposed for isolation must selectively allow the growth of
aureus while supressing the growth of other organisms. The three
types of selective agents that have been used in S. aureus media are as follows: (l) sodium chloride in high concentration (Chapman
1946); (2) various chemicals, such as tellurite and lithium chloride
(Ludlam 1949), triphenyl tetrazolium (Kennedy and Barbaro 1952),
glycine, tellurite and azide (Zebovitz et al. 1955) and sorbic acid
(Raj and Liston 1961); (3) antibiotics, such as polymyxin (Finegold
and Sweeney 1,961) and sulphamezathine (Smith and Baird-Parker 1964). Certain media have an added advantage in that the organism being isolated has a distinctive appearance on that media. The in corporation of tellurite into selective media has been useful in that coagulase-positive S. aureus produces striking black colonies.
Gillespie and Alder (1952) observed that most coagulase-positive strains produce opacity when grown in media containing egg yolk.
Subsequently, selective agents and differential agents such as tellurite and egg yolk have been combined in a large number of media for the isolation and enumeration of staphylococci. The multiplicity of media formulated indicated the difficulty in finding a wholly satisfactory method of differentiating S_. aureus. Further evidence of this problem are the reports of numerous workers advocating the superiority of different coagulase-positive S. aureus isolating media for different foods (Crisley et al. 1956; Giolitti and Cantoni
1966; Jay 1963; Raj 1966; Ranunell and Howic'k 3 967).
Baer et al. (1966) of the U.S. Food and Drug Administration reported that Vogel and Johnson Agar showed the best results for the isolation of coagulase-positive S. aureus from a variety of foods.
Recently Hobbs et al. (1968) reported that the International Committee on Microbiological Specifications for Foods recommended five solid media for the isolation and enumeration of coagulase-positive S_. aureus from foods. They are (l) Telluri.te-Polymyxin-Egg Yolk Agar;
(2) Baird-Parker Medium; (3) Vogel and Johnson Agar; (4) Egg Yol'k-
Sodium-Azide; (5) Milk Salt Agar.
The choice of media is not the only controversy in coagulase- positive S^. aureus isolation. Carter (1960) reported that a direct plating technique was efficient in isolating and enumerating coagulase- positive S. aureus from frozen chicken pot pies. However, Silverman et al. (1961) reported that for isolation of coagulase-positive
S. aureus from shrimp, direct plating was inadequate and that a selective enrichment technique was superior. Gilden et al. (1966) reported that enrichment isolation was clearly more efficient than direct plating. Consequently, Baer et al. (1966) reported that Tryp- ticase Soy Broth containing 10% sodium chloride was the most efficient enrichment medium and suggested its use with Vogel and Johnson Agar as a plating medium. Later, Baer (1968) recommended adoption of a three-tube Most Probable Number enrichment broth technique utilizing
Trypticase Soy Broth with 10% sodium chloride for enrichment and Vogel and Johnson Agar as a selective plating medium. Testing for coagulase production was carried out with reconstituted Rabbit Coagulase
Plasma with EDTA (Bayliss and Hull 1965).
Various methods for the rapid enumeration of S. aureus have been proposed by Grodner (1978), Woolverton (1973), and Meyers (1971).
However, none have achieved popularity since there are false reactions which detract from their usefulness.
Grodner et al. (1973) reported that although Tryptic Mannitol
Meat Broth (Difco) at 18 hours correlated 90 to 95% when compared at A.O.A.C. procedures, false positives produced by mannitol- utilizing bacilli confused the interpretations. Woolverton et al.
(1973), using Woolverton's Medium (Difco), reported similar difficulties as well as numerous false negative reactions. One objective of this study will be to develop simple, rapid procedures for use with gas chroma to graphy. The coliform group, including E. coli have been used as indicators of unsanitary conditions and possible fecal contamination, respectively, for many decades in the food industry (Clark and Kabler
1964). Their adoption as indicator organisms by the seafood industry is well founded in research and practical application. Pedraja
(1972) speaking for the National Shrimp Breaders in rebuttal to
"Consumer Reports" stated that E. coli has been considered by lead ing health authorities as an indicator bacteria, because its presence may indicate the possible presence of enteric pathogens (Shigella,
Salmonella, etc.).
The coliform group of organisms is known to originate in the intestinal tract of man and warm blooded animals (Millipore 1969;
Mundt and Rai 1963). Their presence is not a health hazard in itself, although some special types may cause illness if present in sufficient number.
Since methods of testing for these pathogenic organisms are complicated and time consuming, sanitary standards are based on the concentration of coliforms present in the sample (Millipore 1969).
Detection of Escherichia coli in food products does not necessarily mean that it should be condemned because of fecal contamination. Escherichia coli will not only be found in foods that have been in contact with soil (Figueiredo 1970) but sometimes its presence in foods is an indication of poor manufacturing practices rather than fecal contamination (Figueiredo 1970). Often in milk products, a high count of E. coli is indicative of the sanitary procedures in the plant (Swcnarton 1927; Alexander 1936).
Standard Methods for Examination of Water and Wastewater (12th
Ed.) classifies coliforms only as fecal or non-fecal. Focal coliforms 12 were defined as ,Tthe portion of the coliform group which is present
in the gut or the feces of warm blooded animals,,f generally includes organisms which are capable of producing gas from lactose in a suitable culture medium at 44.5° i 0.5°C.
Coliforms of non-fecal origin are Aerobacter (Enterobacter) and the intermediates, although E. coli is included in the organisms of feces (Koser 1926). The occurrence of either strain outside its normal habitat is accidental, and would indicate contamination as reported by Griffin and Stewart (1940).
Perry (1939) found that coliform organisms could be present in large numbers in oysters and shellfish which had not been subjected to any fecal contamination. Coliforms were found to be present naturally and in sufficient numbers to indicate contamination in orange juice (Wagman 1960) and frozen vegetables (Splittstoessur
1964). Splittstoesscr and Wottergreen (1964) contended that a low count of coliforms in frozen vegetables might indicate a poorer quality product because of the germicidal action of time and low temperature. I'll ere fore, the coliform group per sc should not bo used as an indicator of contamination in food products (Figueiredo
1970; Walford et al. 1948).
Sutton and McFiirlane (1947) doubt the usefulness of E. coli as a danger sign in egg powder, because Salmonella organisms were isolated from samples which contained no E. coli. Winter et al.
(1948) noted that E. coli could be destroyed easier than Salmonella in pasteurization of liquid egg products. Therefore, it should bo remembered that the absence of E. coli in foods does not mean that pathogens are not present. Although the use of E. coli as an index of fecal contamination is often questioned, it remains the best
indicator of inferior quality in many foods (Bartram et al. 1937;
Brown 1971).
Williams et al. (1952, Part I) have shown that shrimp tissue itself is not a source of bacteria, but that bacteria are found mainly on the surface of the shrimp and that the numbers depend upon the handling and washing procedures of the shrimp. The external microflora was composed mainly of organisms from the
Achromobacter, Micrococcus, Pseudomonas and Bacillus groups (1952,
Part II). Their examinations of the intestinal flora of shrimp showed no members of the coliform group to be present (1952, Part
III). Therefore, the presence of these organisms on shrimp indicates contamination L:rom an exogenous source and may be of sanitary significance.
Because most processed shrimp, (doheaded, peeled and deveined, breaded) are frozen and the freezing process Is physiologically debilitating to the microorganisms present, media used in micro biological testing should be enriching for these organisms to grow well enough to be easily detected. The media should also be highly selective in inhibiting the growth of extraneous organisms not being sought (Raj 1969). The testing method which has met with the most success for coliforms is the dilution or multi tube method.
The multitube method resulted from the definition of coliforms as organisms which torment lactose with the production of gas.
Serial dilutions of the sample are inoculated into lactose broth and an ’'indicated number” of coliforms can be calculated from the reciprocals of the dilutions in which positive results, or gas production, occur (McCarthy et al. 1958). McCrady (1915) introduced
the practice of using the expression "most probable number” (MPN).
This technique was adopted as compulsory in the eighth edition of
Standard Methods of Water Analysis (1936).
Modifications of this method have been made and the procedure
is now approved by the United States Public Health Service (Millipore
Corp. 1969) and is an official method of the Association of Official
Analytical Chemists (Surkiewicz 1966). The use of Lauryl Tryptose
Broth (LST) (Difco) has replaced the use of lactose broth in the presumptive test. Tryptose was found by Mailman and Darby (1941) to cause many ”slow lactose fermenters” to produce gas in greater quantities in a shorter period of time, and sodium lauryl sulphate was found to provide a medium selective for the coliform group.
The addition of these two substances provided a more complete and accurate presumptive test.
Hall et al. (1967) found that LST detected coliform organisms in many instances where other media did not, especially in fish and seafoods. They found this medium to be more sensitive, possibly because the coliform organisms in frozen products have been injured, and thus can initiate growth easier in this relatively noninhibitory medium.
In the total coliform analysis, gas production within 48 hours in LST is followed by inoculation into Brilliant Green Lactose
Bile 2 % (BGLB) (Surkiewicz 1966; APHA 1940). This has been accepted universally as a confirmatory test and the observations of Hall et al. (1967) have supported its reliability. Both the brilliant green dye and the bile salt content make the medium selective for coliforms. This method of confirming the presence of coliforms is simple, efficient, and practical (Hajna 1943).
The use of BGLB, or the confirmatory test, depends upon the accuracy of counts desired. Where no great accuracy is desired, the use of the presumptive test alone is satisfactory. For moderately accurate results, both the presumptive and confirmed tests are used.
For extreme accuracy however, a completed test which consists of isolation of the coliform organisms must be employed. The completed test is rarely used due to complexity; furthermore the confirmed test is sufficiently accurate for routine analysis (McCrady 1937;
Hajna 1943).
The test for fecal coliforms is primarily a test for Escherichia coli. At the same time tubes of BGLB were inoculated from the positive LST tubes for the total coliform analysis, EG Medium (EC) is inoculated. However the EG tubes are incubated at a temperature of 45.5°G. Hajna et al. (1939) found a tomj)eraturo greater than
45°G to be essential for the supression of coliform types other than
E. coli as they usually fail to grow and produce gas. Perry and
Hajna (1944) found the combination of EC Medium and 45.5°C highly specific for E. coli identification. This medium and temperature are recommended by the AOAC (Surkiowiez 1966). l’ishbein and
Surkiewicz (1964) demonstrated the overall advantages of this method.
Hall et al. (1967) reported that this method alone gives acceptably accurate E. coll counts.
A completed test for E. coli involves streaking of positive tubes of EG onto Levine’s Eosin Methylene Blue (EMB) Agar and performing a tost for the production of indole from tryptono broth, for acid production as demonstrated by the indicator methyl red, for production of acetyl-methyl-carbinol from dextrose (Voges-
Proskauer Reaction), and for the utilization of citrate as the sole carbon source on the isolated colonies. This is the classical
IMViC biochemical series. A characteristic green sheen is usually shown by E. coli in EMB by reflected light, but Raj and Liston
(1961) reported that this is not a reliable criterion for isolating this organism. In order to insure the presence of E. coli the bio chemical IMViC tests must be performed. Typical isolated colonies on EMB agar are picked and incubated on agar slants to increase the number of cells. The IMViC series is performed on cells from these slants. Cultures which produce the results of indole pro duction positive or negative, methyl red positive, Voges-Proskauer negative, and citrate negative are considered positive in the tost for fecal coliforms (Surkiewicz 196G).
These methods are involved and quite time consuming. Faster methods of determining total and fecal coliform contents of foods must be developed.
Methods used for differential identification of enteric and related grain-negative bacilli have been dependent mostly on bio chemical test. Many of these test arc time consuming, and 72 hours may elapse before the organism can be properly identified.
Investigators have constantly sought simpler and more rapid pro cedures to differentiate’ gram-negative bacilli. Until recently, the vast majority of biochemical tests have been diverse and separate in media and methods. Today a number of rapid biochemical test systems are available to the microbiologist. One such test system is the Ortho-nitropheyl-beta-D-galacto- pyranoside (ONPG) described by Seidman and Link (1950). Butler
(1970) performed total coliform and fecal coliform analysis on fresh frozen headless, frozen peeled and deveined, and frozen breaded shrimp employing two experimental ONPG procedures, and the standard
AOAC methods. Comparison of the three methods showed that the results obtained by the ONPG procedures approximated those of the AOAC method.
A comparison of the time required for completion favored the two ONPG procedures, since they gave a count after 24 hours incubation, while the AOAC method required 48 to 96 hours.
Both ONPG procedures, disc and powder, gave several false positives in the fecal coliform analysis due to the presence of
Bacillus liclienifbrmis.
It was recommendod that throe presumptive tests incorporate an inhibitor reducing the growth of tin's extraneous organism.
Amsterdam ot al. (1968), Liston (1969) and Matson et al. (1969) have reported on the use of reagent-impregnated paper strips as a more rapid substitute for conventional methods.
A modified R/B enteric differential system which uses two tubes of culture media and various reagents, has been developed and evaluated by Smith et al. (1971). The R/B system correctly identified 95.5% of the 200 cultures tested. Some false negative reactions were noted including misidentification of Salmonella cultures as Arizona.
Washington et al. (1971) evaluated the Analytab system of miniaturized sterile plastic tubes and found that 88% of Entero- bacteriaceac were correctly speeiated. Its principal disadvantage 18 was the time required to prepare and inoculate the 20 test and the care required in the tedious task of filling each tube. An individual colony or pure culture is required as an inoculum.
Another newly formulated test system, Enterotube, is a compartmental tube containing a number of differential types of bio chemical test media used for the identification of Enterobacteriaceae
(Grunberg 1969), In a parallel evaluation of the Enterotube system and conventional test, Elston et al. (1971) reported 15 strains of
242 were incorrectly identified including Shigella strains frequently misidentified as Escherichia. Primary isolates or pure cultures are required as inoculum for the Enterotube system.
Although the previously discussed methods would have application in diagnostic laboratories, their use would encounter major obstacles in the modem food qual ity assurance laboratory. Rapid, practical methods must be developed for the differentiation of Enteric-gram negative organisms.
Increased consumption of frozen foods in recent years has focused attention on establishing bacteriological standards for these products.
Today, modem attitudes on the bacteriological safety of frozen sea foods have recognized that food poisoning organisms can gain access to foods via the raw materials or during processing, and that constant attention must be given to factors controlling the possible transmission and survival of those organisms to and in such foods.
The microbial examination of frozen foods is influenced by the fact that the processes of freezing, frozen storage, or thawing are capable of altering the microbial population at rates that depend on the organisms present, the product, the rate of cooling or thawing, 19 and the storage temperature. The basis of the frozen food industry then is to use temperatures low enough to inhibit bacterial growth and enzyme action. According to Georgala and Hurst (1963), none of the common food poisoning microorganisms is Known to grow below zero degrees centigrade, therefore, survival in frozen foods implies survival without growth. Numerous workers have reported their studies in which microorganisms in frozen foods remained viable to different extents over prolonged storage periods.
McCleskey and Christopher (1941) reported that Staphylococcus aureus, Salmonella typhimurium, Salmonella paratyphi A, Salmonella paratyphi B, and Salmonella typhi survived deep-freeze storage in unsliced strawberries after 14 months storage at minus 18°C.
Hartscll (1949) demonstrated that salmonellae survived ten months when .inoculated into liquid egg and stored at -17.8°C. His experiment demonstrated the fact that salmonellae survive better at
-17.8°C than at -1°C, a finding he later substantiated with beef and peas inoculated with salmonellae (Hartscll 1951).
Raj and Liston (1961) reported the prolonged survival of salmonellae in frozen fish, with 10% inoculum surviving over a year at -17.8°C.
In the particular case of seafoods, there may be survival of pathogenic or potentially pathogenic bacteria in numbers sufficient to constitute a hazard to health after long storage in the frozen state.
Other investigators have reported substantial protection of food poisoning pathogens by food materials such as chicken chow mein (Gungerson and Rose 1948), creamed salmon and shrimp (Proctor and Phillips 1948), and by egg white and c o m syrup (Woodbura and
Strong I960). All studies indicate that the food poisoning bacteria die more
rapidly when stored at 0 to -10°C than at -17 to -2o°C. Death is
greatest during the actual freezing process, but is slower during
subsequent cold storage. The frozen food industry is based on the
use of temperatures in the region of -17°C, therefore, commercial
freezing processes cannot be relied upon to destroy any pathogens
present in foods, especially when foods such as meat, fish, and
eggs are known to protect bacteria against injury at low temperatures
(Georgala and Hurst 1963),
From a bacteriological viewpoint, it is highly possible that
these commercially prepared fishery products may suffer from excessive handling, improper processing, and generally unsanitary conditions.
In particular, frozen shrimp present an additional problem in that
they can be sold cooked or raw, peeled or unpeeled, and with or without breading.
The numbers of reported human cases of salmonellosis (other than typhoid fever) in the United States have steadily been increasing over the last 25 years. In 1946 only 720 Salmonella infections in
man were reported. During the subsequent years 1951, 1961, 1963,
and 1964, respectively 1,733; 8,500; 18,000; and 21,113 cases were
reported (l’rost and Riemann 1967). In 1967, a total of 273 outbreaks
of food poisoning were reported to the National Communicable Disease
Center from 37 states. The outbreaks involved 22,171 cases, with
12,836 cases or 58"„ having boon caused by Salmonella spp. (NCDO 1968).
Some authorities feel that as many as 2 million human cases of sal
monellosis may occur each year in the United States alone. As with
staphylococcal food poisoning, the largest outbreaks of salmonellosis
usually occur at banquets. The increasing incidence of human food-borne salmonellosis
appearing during recent years, especially in countries with a high
standard of living, has not been satisfactorily explained.
Kampelinacher (1963) has offered the following explanations for the
increase: (l) the increase in mass food preparation, which favors
spread of Salmonella, (2) inappropriate methods of storing food
which, because of modem living conditions, is sometimes accumulated
in excessive amounts, (3) the increasing habit of eating raw or
insufficiently heated foods, partly because of ever reliance on food
inspection, and (4) decreased resistance in infection resulting from
improved standards of general hygiene.
Thatcher and Clark (1969) states that contamination of foods can
occur from three main causes: (l) exposure of foods to pollution of disease, (2) poor manufacturing practices while processing, and
(3) storage under improper conditions.
The primary habitat of Salmonella spp. is the intential tract of man and animals, particularly farm animals, reptiles, birds
(fowl), etc. The frequency with which salmonellae are isolated
from domestic fowl indicates that these animals probably constitute the largest single reservoir of the bacteria (NCDC 1965).
As intestinal forms, these organisms are excreted in feces and subsequently are transmitted by flics, cockroaches, and other insects world-wide. Sources of contamination include slaughter houses
food processing plants, and institutional kitchens with the food handler being the most implicated in the transmission of salmonellosis
Fox* this reason, hygienic practices must bo used in the preparing, cooking, holding, storing and serving of foods in Institutional 22
kitchens, schools, and restaurants. Other environmental reservoirs
of Salmonella include drugs and animal feeds.
The widespread distribution of salmonellae in animals used for
food frequently results in contamination of human foods derived from
these sources. The most common food vehicles of salmonellosis in
man are poultry, meat, and meat products, eggs, milk and milk products,
and fish. In a study of 61 outbreaks of Salmonella food poisoning
for the period 1963-65, Steele and Galton (1967) found that eggs and
egg products accounted for 23, chicken and turkey for 16, beef and
pork for 8, ice cream for 3, potato salad for 2, and other miscel
laneous foods for 9. Adinarayanan et al. (1965) examined some
prepared and packaged foods which contained contaminated ingredients.
A subsequent study revealed that cake mixes, cookie douj^is, broad
crumbs, rolls, mixes, ami dried cereals contained 7 Salmonella
serotypes.
Animal feeds especially those derived from meat and fish by products, may carry salmonellae to poultry, swine, and cattle. In a study of animal feeds for the years 1958-60, Morehouse and Weidman
(1961) examined 5,712 samples of bone meal, feather meal, tankage,
fish meal, egg products, and complete feeds. They found that 718
(l3/o) of the samples contained 59 Salmonella serotypes, Salmonella senftenberg was isolated most frequently followed by S^ anaturn,
£>. typhimurium, S. cubana, and _S. montivideo. The presence of salmonellae in animal feeds has also been shown by Coring (1958).
In this case 9 of 16 samples of domestic fish meal from various com mercial sources yielded 11 serotypes of Salmonella.
Freshly cooked fish is rarely a vehicle of Salmonella food poisoning, however, processed fish products such as fish pies and 23
fish cakes may readily be contaminated if improper storage, handling and sanitary conditions are maintained at the processing plant.
Shellfish normally are not vehicles of salmonellosis but oysters, shrimp, etc. may become contaminated if they are exposed to polluted waters. Between 1919-34, Hobbs (1953) estimated that more than
100,000 cases of typhoid fever were due to the consumption of shell
fish from uncontrolled beds occurred in France and of these 25,000 were fatal. Gulasekharam et al. (1956) reported that of 39 fish of 24 species examined, 5 samples of fish washing yielded Salmonella.
The genus Salmonella consists of small gram-negative nonsporing rods that are indistinguishable from Escherichia coli under the microscope or on ordinary nutrient media. These organisms are generally unable to ferment lactose, sucrose, or salicin, although glucose and certain other monosaccliaridos are fermented, with the production of gas. As with other gratn-nogative bacteria, Salmonella strains are able to grow on a large number of culture media and produce visible colonies within 24 hours at 37°C.
Kauffmann (1966) defines the genus Salmonella as motile bacteria that conform to the criteria of the family Cnterobacteriaceae and the tribe. Salmonella.
In view of recent federal regulations prohibiting the occurrence of Salmonella spp. in foods, the recovery of these organisms from food presents a di f l.’icult problum to food scientists and food micro biologists, The analytical problems of detection and enumeration of
Salmonella in foods possess certain peculiarities not commonly encountered in the clinical laboratory. First, the organisms have been subjected to physiologically debilitating processes such as freezing, desiccation, curing ingredients, and extremes of pH, heat,
and osmotic pressures during the manufacture or storage of the
product. Second, salmonellae in foods usually comprise an exceedingly
small component of the total microbial population, and in addition,
are almost invariably outnumbered by physiologically similar competi
tive and antagonistic bacteria. For these reasons, therefore, both
the clincal bacteriologist and food analyst should recognize that the
media used for detection and enumeration of Salmonella in clinical
laboratories are not necessarily applicable for the use in food
products.
It has become increasing apparent, through the work in many
laboratories, that not all selective agars are efficacious in the
detection of Salmonella spp. and their enumeration from foods.
The United States Food and Drug Administration has recommended
the use of Salmonolla-Sh.i golla Agar (Difco), and Brilliant Ore on Agar
(Difco) us the selective plating media lor the; isolation of Salmonella
from foods.
Bacto-Salmonella-Shigclla (SS) Agar is a highly selective
medium recommended for the isolation of Salmonella from stools and
other suspect materials. The medium is essentially a modification of the Deoxycholatc-Citrate Agar (Difco) described by Leifson (1935).
It is designed to provide excellent differentiation of lactose
fennenters from non-fomienters, and to give maximum inhibition
of coliform organisms without restricting Hie growth of pathogenic
grani-negative organisms. Due to the highly selective nature of SS
agar, a heavy inoculum must bo used in order to increase the chances
of Salmonella isolation, and plates must be incubated 24-48 hours at 37°G. In addition, a non-selective, general enteric medium should be used in conjunction with the selective SS agar to insure recovery of the most fastidious enteric pathogens.
McCullough and Byrne (1952) used specimens obtained from humans inoculated with known Salmonella spp. to compare the isolation abilities of Salmonella-Shigella (SS) Agar, Bismuth Sulfite (BS)
Agar, Eosin methylene Blue (EMB) Agar, MacConkey Agar, and Deoxy- cholate-Citrate-Lactose-Sucrose (DCLS) Agar. SS agar was found to be superior to the other agars for the isolation of Salmonella by the direct streak technique.
Bacto-Bismuth Sulfite (BS) Agar, a modification of the Wilson and Blair formula, is a highly selective medium designed especially for the isolation of Salmonella typhosa from feces, sewage, and other materials. 'Hie early history of the development of BS agar started when Wilson (1923) pointed out that in a medium containing sodium sulfite, glucose, and iron salts, reduction of sulfite to sulfide was accomplished by members of the typhoid and paratyphoid groups, with the exception of Bacillus (Salmonella) paratyphosus A.
In the past many different techniques and selective media have been developed to recover salmonellae from foods, feeds, focal speci mens, and natural waters. However, complete identification of salmonellae is often a time-consuming procedure in the diagnostic laboratory.
Since most salmonellae are motile, Hajna and Damon (1950) suggested that polyvalent Salmonella flagellar agglutination could be used as a rapid screening test for the presence of Salmonella.
Spicer (1956) developed a simplified method for identifying the 26 flagellar antigens of the most commonly encountered salmonellae.
Edwards and Ewing (1962) modified the Spicer antisera to achieve greater specificity and to minimize the incidence of cross-reactions.
Silliker et al. (1965) used Spicer-Edwards H antisera as a primary screening procedure for detecting salmonellae in food products.
MacFarlane et al. (1970) described a two phased microimmuno diffusion test for the detection of Salmonella in foods. The detection of Salmonella, with the maximum time for the complete test set at
24 hours, was the established aim of the research. The test is capable of detecting as little as 1 salmonella per 100 g of food in less than 24 hours. False positives have not been encountered.
Dockstadler (1970) described as accelerated detection of
Salmonella utilizing direct plating of pre-onriched materials and overlaying with Padron-Dockstadler Salmonella Agar (PDS Agar).
It is reported that positive reactions occurred in as little as .'S hours i ncubation.
The present investigation was undertaken to establish an accelerated Salmonella detection method for use in fresh "green" headless as well as processed shrimp and shellfish products.
Robert Koch could enter a modem microbiology laboratory and find very little that would be new or strange. Microorganisms today are cultured, purified, plated and counted much as they were 100 years ago.
The millions of routine microbiological analyses conducted annually for clinical, vitality control, or regulatory purposes are performed by tedious, laborious and ancient methods. Automation of microbiological technicpies are urgently needed. 27
The feasibility of using gas chromatography as a sensitive and rapid method for the classification of microorganisms by analysis of chemical composition was proposed by Able et al. (1962). Extending the concept of Wolochow (1959) that microorganisms, both living and non-living, could be differentiated from higher forms of life on the basis of chemical compounds which are unique to microorganisms.
Abel (1963) used gas chromatography to provide patterns representing the carboxylic acid methyl esters of the lipids from selected bacteria.
Extraction and transesterification of the lipid was necessary.
Chromatography is a term that covers a great deal of ground, and therefore is a difficult one to define. Basically it is a separation technique; a sample containing a variety of materials is separated into its components which are then measured individually.
This is doiu; by passing the sample over a material which holds iiack some of tlie components and lots others puss through, so that the end result is that the components appear one after another, i.e. separated in time.
Gas chromatography as a method consists of carrying a mixture of components through a column by means of a carrier gas. The components arrive separately at the exit where they are measured by a suitable detector.
Recordings obtained by measuring the detector output thus result in a series of peaks. 'Hie peak location or characteristic component speed provides the qualitative information. The peak heights or areas can be used successfully for determining percent concentration.
Chromatography essentially consists of passing a mixture through a column. It represents a two-phase system in which advantage 28
is taken of the different equilibria existing between the two phases
to separate components differing in equilibrium constants. The sample
and carrier are generally called the mobile phase whereas the column material is referred to as the stationary phase.
The mobile phase can be a gas, liquid, or solid in solution whereas the stationary phase can only be a liquid or solid. From this, the following combination can result and are generally dealt with as separate fields in chromatography:
1. Gas-liquid chromatography (partition)
2. Gas-solid chromatography (absorption)
3. Liquid-liquid chromatography partition (counter-current)
4. Liquid-solid chromatography (conventional chromatography)
(ion exchange chromatography)
5. Solid-liquid chromatography. Same as 3 and 4, except
that sample :is in solution.
6. Solid-solid chromatography
In gas-liquid or partition chromatography, the liquid which represents the effective stationary phase is spread as a thin film over a largo surface area solid (such as flux calcined diatomaceous earth). This latter only acts as a support and should preferably have no absorptive properties at all.
Components present in the mobile phase will partition between the vapor phase and substrate. The difference in the partition coefficient; for each component provides the basis for separation
(Littlewood 1970).
Reiner (1965) described the first account of the application of pyrolysis-gas-liquid chromatography to the classification of natural 29 products which are largely macroraolecular in character. Ninety five analyses were performed and gaseous, thermally-degraded products were obtained by pyrolyzing microorganisms as well as other forms of cellular matter in an inert atmosphere on a nickel filament at
850°C for ten seconds.
Methyl sulfide, a volatile chemical compound producing defects described as cowy or feedy, produced by Aexobacter aerogenes was detected and levels determined by Toan et al. (1965) using gas chromatographic analysis of head-space gas.
Payne et al. (1965) described a procedure to determine the response to varying conditions of the activity of primary alcohol sulfatase in an ammonium sulfate-precipitated fraction of cell-free extracts from Pseudomonas and to reveal the distinction of this enzyme from other sulfatase enzymes, Gas-liquid chromatographic analyses were obtained for the C^-alcohol moiety after release of sulfate.
Pseudomonas species was shown to grow on and oxidize linear primary alcohols with even and odd numbered carbon chains ranging from Ct) to ('-q ' Analysis by gas liquid chromatography of hexane extracts of filtrates of cultures containing mixtures of even-carbon numbered alcohols from to C^g revealed that decanol was rapidly utilized, whereas the remainder were slowly dissimilated up to 19 hours and then were rapidly degraded (Williams et al. 1966).
Henis et al. (1966) reported 32 species and strains of bacteria
(Bacillus, Escherichia, Aerobacter (Enterobacter), Pseudomonas) were grown in a synthetic medium (Proom and Knight) supplemented with peptone. After growth, metabolic products were extracted from the medium and examined in a gas chromatograph by use of flame 30
ionization and electron capture detectors. It was found that different
chromatograms (signatures) were obtained for each of the species and strains studied.
Metabolic products produced by washed bacterial cells were
extraced with ether and analyzed by gas chromatography (O’Brien
1967). Organisms tested showed different chromatograms of the metabolic products with the exception of Pseudomonas and Vibrio which did not produce any detectable products.
Differentiation of species of Aerobacter (Enterobacter) and
Escherichia by gas liquid chromatography of some metabolic products by head-space gas sampling of milk cultures was described by Bawdon et al. (1966).
Bassetto et al. (1966) described detection of volatile materials in milk as produced by several species of bacteria. As an extension of the work of Bassetto and Olaydon (1965), species of Aerobacter,
Escherichia, Streptococcus, Achromobacter, Lactobacillus and
Pseudomonas were compared using gas chromatographic patterns of volatile materials.
Mitruka and Alexander (1967) utilized the electron capture detector (ECD) to detect certain common microbial metabolites. It was found that the ECD was sensitive to conjugated carbonyl, halogens, sulphur, or ni.tro groups found in the bacterial metabolites.
Mitruka and Alexander (1968) found that gas chromatographic methods could be useful for the rapid detection of small numbers of organisms growing in vitro and also of viral activity in vivo.
Henis et al. (1968) using concentrated ether extracts of distilled
culture fluids of Streptomyccs strains on dual-channel chromatography
found each extract gave a distinctly different chromatogram. Mitruka et al. (1968, 1968, 1969) using the electron capture detector have shown the presence of small numbers of virus in horses and dogs. Ether extracts of culture fluids were observed for similar and dissimilar metabolites.
Ether extracts of culture broths of 13 species of Clostridium were subjected to gas chromatographic analysis. The Clostridium could be separated on the basis of amines, alochols, and fatty acid content analyzed by a temperature-programmed gas chromatograph equipped with fLame ionization detectors (Crooks and Moore 1969).
Analysis of head-space vapors over dextrose fermenting bacterial cultures by gas-liquid partitioning produced chromatographic patterns enabling Guarino et al. (1969) to differentiate members of the family
Enterobacteriaceae.
Vegetative colls and spores of ten strains of Clostridium botulinuin types A, 11, and E were differentiated by Cone and
Lichowicli (1969) employing pyrolysis-gas-liquid chromarography while
Farshy and Muss (1970) differentiated ten species of Clostridium on the basis of trimethylsilyl derivatives of whole-cell hydrolysates subjected to gas liquid chromatography employing hydrogen flame ionization.
Simmonds (1970) employing whole-colls compared the pyrolysates of two species of microorganisms and comparable pyrolysates of meteorite and fossil organic matter. The studies were carried out using pyrolysis-gas chromatography mass spectrometry to aid in the interpretation of a soil, organic analysis experiment to be performed on the surface of Murs j.n 1975.
The present investigation was designed to determine the feasi- 32 bility of using gas-liquid chromatography for the early detection of organisms of the family Enterobacteriaceae and other microorganisms associated with food-borne diseases or incriminated in food poisoning outbreaks in seafood. MATERIALS
All materials and media used in the experimental procedures
are generally accepted for the designated purpose. All materials and media were used for the intended purpose. Methods of preparation are given. Although the use of some of the media has been modified, the underlying principles are applied for identification, cultiva tion, or examination of the organism.
Samples
The samples of shrimp used were supplied by a commercial processor and consisted of fresh frozen headless shrimp (shell-on), frozen breaded shrimp, and frozen peeled and deveined shrimp. Upon receipt they were stored in the freezer at -23°C until used.
A twenty-five gram sample of the unthawed shrimp was placed in a sterile Waring blendor jar containing 225 ml of sterile Butter field's Phosphate Buffer (Butterfield 1932), achieving a 1 in 10 dilution. The sample was then blended for 2 minutes. All dilutions were prepared with 90 ml of sterile phosphate buffer plus 10 ml of the previous dilution. Dilutions of 1 in 100 and 1 in 1000 were made in this manner, using sterile 10 ml pipettes.
The coagulase-positive S. aureus strain used in this study was
American Type Culture Collection 9664, which was isolated from a food poisoning incident. The identity and coagulase-positive characteristic of the S. aureus were verified by biochemical test- The stock culture was maintained on Nutrient Agar slants. The slants were streaked from the stock culture and stored in an incu bator at 35°C,
Brain Heart Infusion (Difco) agar slant cultures of Salmonella typhimurium. Salmonella senftenberg. Salmonella thompson and
Salmonella paratyphi A were obtained from the Poultry Science
Department, Louisiana State University. The identity of each
Salmonella serotype was verified by performing both biochemical and serological tests.
Stock cultures of each serotype were maintained on Brain Heart
Infusion Agar slants. Freshly prepared Brain Heart Infusion Agar slants were streaked from the stock cultures and incubated at 35°C for 24 hours.
The Escherichia coli, Enterobacter aerogenes, and Pseudomonas aeruginosa strains used in this study were obtained from the Depart ment of Microbiology, Louisiana State University. The identity of the strains was confirmed by biochemical testing, as shown in Table
1.
Stock cultures were maintained on Brain Heart Infusion Agar slants incubated at 35°G.
Glassware
All glass containers and pipettes were sterilized by autoclaving at 121°C for 30 minutes, and dried. Glass bottles calibrated to
100 ml were used to prepare water dilution blanks. Pipettes were selected to deliver only the designated volume of the highest calibration on the pipette. TABLE 1 CHARACTERISTICS OF TEST ORGANISMS
Organisms ______Test____ T S I A h 2s IMViC Coagulase
Escherichia coli Acd,Acd,Gas Neg ++__
Enterobacter aerogenes Acd,Acd,Gas Neg _++
Salmonella typhimurium Alk,Acd,Gas Pos
Salmonella thompson Alk,Acd,Gas Pos
Salmonella senftenberg Alk,Acd,Gas Neg
Salmonella anatum Alk,Acd,Gas Pos
Staphylococcus aureus Positive 36
Reagents
The chemical composition of all reagents and media used in identification tests are listed in the appendix.
Media
Bacto-Brain Heart Infusion Agar
Brain Heart Infusion Agar (Difco) is a solid medium recommended for the cultivation of fastidious pathogenic bacteria. The organisms for standardization of inoculum and inoculation into the experimental sample were grown on this medium. All stock cultures were maintained and stored at 25°C on this medium.
Bacto-Brain Heart Infusion Broth
Brain Heart Infusion Broth (Difco) is a liquid medium used to maintain virulence, antigenicity, and other serological characteristics of organisms.
Tryptone Glucose Extract Agar
Tryptone Glucose Dxtract Agar (Difco) is recommended for use in determining the standard plate count of milk and other food products. This medium was used in determining the total plate count in non-sterile samples and to verify MPN results in standardizing the inoculum.
Butterfield’s Phosphate Buffer
All dilutions throughout this investigation were made with
Butterfield's Phosphate Buffer. A stock solution was prepared by dissolving 24 g of potassium acid phosphate (lG^PO^) in 500 ml of distilled water. The pH was adjusted to 7.2 with N NaOH and the mixture diluted to one liter with distilled water. This stock solution was stored in the refrigerator at 40°F until needed. 37
Dilutions were prepared by adding 1,25 ml of the stock solution to one liter of distilled water and adjusting the pH to 7.2 with 0.1 N
NaOH.
Nutrient Agar
Nutrient Agar (Difco) is recommended as a general culture medium for the cultivation of the majority of less fastidious organisms.
This medium was also the recommended stock medium by American Type
Culture Collection. Nutrient Agar was utilized for growth of organisms prior to microscopic examination in the MPN technique for coagulase-positive aureus and E. coli.
Eugon Agar
Eugon agar is a medium designed to obtain luxuriant cultures of bacteria, including many organisms ordinarily regarded as difficult to cultivate. It is recommended for examination of meats and control of food processing and for obtaining total plate counts.
Tryptic Mannitol Meat liroth (T.M.M.)
T.M.M. (Difco) is a selective broth medium recommended for use in the MPN technique for coagulase-positive S, aureus. The medium is inoculated in 3 tube MPN series as per A.O.A.C. and incubated for 24 hours at 44°C. Any color change produced by the utilization of mannitol is recorded as presumptive coagulase-positive staph.
Trypt.lease Soy broth
Tryptjease Soy Broth (BHL) is recommended as a good general purpose medium. The U.S. Food and Drug Administration recommends the addition of 10% sodium chloride to inhibit the growth of most organisms, but not S. aureus, in the MPN technique.
Vogel and Johnson Agar
Vogel and Johnson Agar (BBL), or Tellurite Glycine Red Agar 38
Base, is the selective plating medium recommended by the U.S. Food and Drug Administration for use in its MPN technique for coagulase- positive S. aureus. Positive colonies on this medium are black surrounded by a yellow zone. The black colonies are due to the reduction of potassium tellurite which is added at the rate of
2 ml of a 1% solution per 100 ml of Vogel and Johnson Agar. The yellow zone is due to the utilization of mannitol in the medium by the organisms.
Coagulase Plasma
Reconstituted Rabbit Coagulase Plasma with EDTA (BBL) was used in testing for the coagulase reaction in isolates which are presumptively coagulase-positive £>. aureus. Clot formation within
4 hours at 37°C is considered positive confirmation of the presence of coagulase-positive J3. aureus.
Bacto-Lauryl Tryptose Broth
Lauryl Tryptose Broth (Difco) is recommended for use in the detection of coliform organisms according to American Public Health
Association (APHA) Standard Methods for the examination of waters, dairy products, and other foods. The formation of gas in fermentation tubes within 48-3 hours constitutes a positive presumptive test.
Simmons Citrate Agar
To differentiate between members of the coliform group, the utilization of citrate is determined. Simmons Citrate Agar (Difco) is one medium that is employed, and was prepared by adding 24.2 g of the powder to 1 liter of water. This was boiled until all the agar was dissolved and then the medium was autoclaved. After cooling to 50°G, the hot agar was poured into sterile plastic petri dishes, cooled and dried. Brom thymol blue was incorporated into the medium as an indicator for the production of an alkaline zone which accompanies growth after 24 hours. Escherichia coli does not utilize citrate as its sole carbon source and therefore will not grow on the medium.
Bacto-Brilliant Green Bile 2%
Brilliant Green Bile 2% Broth (Difco) is recommended by the
APHA Standard Methods for the examination of water and wastewater and Recommended Microbiological Examination of Foods. It is used to confirm the presence of coliform organisms by the production of gas within 48-3 hours, the bile and brilliant green being inhibitory for most other bacteria.
Levine Rosin Methylene Blue Agar
The purpose of this agar was to differentiate between members of the coliform group (lactose fermontors) and is recommended in
’’Standard Methods for the Examination of Water and Wastewater” for the confirmatory test for bacteria of the coliform group (12th ed.).
The role of this agar :in the experiment was for isolation of colonies and a differential medium. In order to rehydrate properly, 37.5 g of the powder was mixed in 1 liter of cold distilled watur and heated to boiling to dissolve the agar. The hot solution was distributed into bottles and autoclaved. AiH:or cooling down to 50°C, the agar was poured into sterile petri dishes and dried before used. Typical
Escherichia coli colonies exhibited a metallic, green sheen, usually with dark centers after incubating for 21 hours at 37°C (Difco
Manual, 9tli ed.). M.R.-V.P. Medium
This medium is recommended in "Standard Methods for the Examina
tion of Water and Wastewater" (12th ed.) and was used as a substrate
to grow bacteria for the employment of the Methyl Red and Voges-
Proskauer test in the identification of E. coli. Rehydration was
conducted by adding 17 g of M.R.-V.P. Medium (Difco) to 1 liter of
cold distilled water and agitating until completely dissolved.
Inocula of 10 ml were dispersed in each tube and, in turn, this was sterilized by autoclaving. E. coli will be methyl red positive and acetylmethylcarbinol negative (Voges-Proskauer) when tested properly
by this medium.
Tryptone
Tryptone (Difco) is recommended in "Standard Methods for the
Examination of Water and Wastewater" (12th ed.) and used in the performance of the indole test for the identification of E. coli.
A 1 percent tryptone solution was made by dissolving 10 g of powdered
tryptone in 1 liter of cold distilled water. Tubes containing 10 ml were autoclaved for sterilization. E. coli gives a positive
indole test.
SIM Medium
SIM Medium (Difco) is a semisolid medium which was used for the
detection of H^S and indole production and mortality. The medium was prepared by mixing 3 6 g of the dehydrated powder in 1 liter of
cold distilled water, boiling to dissolve and autoclaving after
dispersing 10 ml portions into test tubes (Difco Manual, 9th ed.).
The suspected organism was stabbed deeply into the medium by means of
a needle with one smooth stroke, and incubated at 37°CI for 24 hours. 41
Bacto-Tetrathionate Broth Base
Tetrathionate Broth Base (Difco), with iodine-iodide solution and 1:1000 aqueous brilliant green dye solution added, is recommended by the FDA as a selective enrichment medium for the isolation of
Salmonella from foodstuffs. Inhibition of many other bacteria is obtained by means of the incorporated salts and iodine and brilliant green dye.
Bacto-Salmonella-Shigella Agar
Salmonella-Shigella Agar (Difco) is recommended by the FDA as a selective plating medium for the isolation of Salmonella from foods. Gram-positive and coliform organisms are inhibited chiefly by means of the bile salts mixture. Salmonella, Shigella, and other non-laetoso fermenting organisms form opaque, transparent or trans lucent uncolored colonies, while the few lactose fermenting organisms which may develop on the medium form red, pink, or sometimes black colonies.
Bacto-Triple Sugar Iron Agar
Triple Sugar Iron Agar (Difco) is recommended by FDA as the medium to which typical Salmonella colonies are transferred after picking from selective agar plates. This medium is designed to differentiate the Gram-negative enteric bacilli, by means of their ability to attack dextrose, lactose, and sucrose and their ability to liberate sulfides thus forming hydrogen sulfide gas and precip itating iron sulfide. A presumptive positive reaction for salmonellae consists of all Triple Sugar Iron Agar slant tubes having alkaline
(red) slants and acid (yellow) butts, with or without the formation of gas (H.,S) and a black iron sulfide precipitate. 42
Bacto-Urea Broth
Urea Broth (Difco) is recommended by the FDA in the biochemical
verification of salmonellae. It may be used for the idaitification of organisms based upon urea utilization. It is a means of differ
entiation between the Proteus and Salmonella-Shigella group.
Bacto-Bismuth Sulfite Agar
Bismuth Sulfite Agar (Difco) is recommended by the FDA as a
selective plating medium for the isolation of Salmonella from
foods. This medium is particularly designed for detection of
Salmonella typhosus in foods. Colonies of S. typhosa are flat, dry,
and black, sometimes exhibiting a green metallic sheen. Proteus spp. either fail to grow or produce greenish colonies and Escherichia
coli from brown sticky raised colonies.
Growth Media - G.L.C.
Media I
Tript.ic Soy Broth 30 g/1
Peptone 10 g/l
Glucose 30 g/l
Media I-A
Triptic Soy Broth 30 g/l
Peptone 10 g/l
Manni tol 10 g/l
Wright et al. (I960) broth (Difco) was prepared from a dehydrated
medium as directed and dispensed in 15 ml amounts into 20x150 mm
screw capped tubes. The medium containing those amino acids found
essential for £. aureus and S. typhosa was recommended by Guarino
et al. (1969) as the growth medium for members of the family Entero-
bacteriaceae and associated food-borne disease producing organisms. Gas Chromatograph
Gas Liquid Chromatography analyses were made on two different
gas chromatographs. Initial studies were carried out on a Microtex,
model 2000 R (Microtex, Inc., Baton Rouge, La.) with a hydrogen flame
detector and a model LS 11A/MAZ-11 Westronic strip chart recorder
(Westronic Inc., Fort Worth, Tex.). The columns were 8 ft. by \
in. stainless steel, packed with 20% carbowax 20M on 60/80 mesh
DMCS treated acid washed Chromosorb W.
Final analyses were made on a Perkin-Elmer 990 (Norwich, Conn.)
equipped with a hydrogen detector and a Westronic LS 11A/MAZ recorder.
An Infitronic model GR 208 Automatic Digital Integrator was also used. The columns were 10 ft. by 1/8 inch (0.3 cm.) in outer
diameter stainless steel, and packed with 20% carbowax 20M on Chromo sorb W-AW DMCS 60/80 mesh (Wilkins Instrument and Research, Inc.).
A 2,5 ml gas-tight syringe with fixed needle (Hamilton, Reno,
Nevada) was used to withdraw samples of metabolite vapors from a
50 ml Erlenmeyer flask (Pyrex no. 4980) fitted with a # 1 rubber serum bottle cap. METHODS
Sterilization
All media in the form of agar and broth, and phosphate buffer were sterilized by autoclaving at 121°C for 15 minutes, unless otherwise specified.
Shrimp Homogenate
A shrimp homogenate was used in this study because it has many advantages over the use of whole shrimp and is widely used in studies involving inoculation of specific organisms into foods.
Some of the advantages of homogenate in this type of study are:
(l) a sample can be uniformly inoculated in homogenate while in solid
foods difficulty in spreading the inoculum uniformly over the surface of the food is encountered, especially when a small volume is being
inoculated; (2) in surface inoculated foods, part of the inoculum and subsequent growth is rubbed off on the inner surface of the storage container and thereby removed from the food, while in homogenates, the walls of the container are constantly washed with media; and
(3) homogenates are easily sterilized in glass test tubes and can
remain in the same tubes for storage, while sterilization of solid
food involves either use of larger, more expensive glassware for sterilization and storage or transfer of the sterile solid from the sterilization vessel to another container for storage. The use
of homogenate is logical, more efficient, cheaper, less cumbersome, and gives more uniform results. 45
Shrimp homogenates were made by adding 25 g of shrimp to 225 ml of buffer and blending in sterile screw capped Waring Blendor jars for 2 minutes at low speed to minimize foaming. The resulting
1 in 10 homogenate contained the greatest concentration of shrimp
that could be pipetted with precision.
Total Plate Count
A total plate count was determined initially on the breaded shrimp sample to ascertain both the number of mesophilic and psychrophilic organisms present. In this procedure 25 g of breaded shrimp were selected at random from the package sample and added to 225 ml of Butterfield's phosphate buffer.
The mixture was blended in a sterile screw cap Waring Blendor jar for 2 minutes. The resulting 1:10 shrimp homogenate was allowed
to stand at room temperature for 5 minutes until all foam settled.
Afterwards, serial, dilutions were made using 1 ml of the shrimp homogenate and 9 ml sterile buffered dilution blanks. From each serial dilution a 1 ml aliquot was dispensed in two duplicate sets of sterile petri plates and about 20 ml of tempered (40°G) Nutrient agar were poured into each plate. The plates were swirled and allowed to solidify. One duplicate set was incubated at 35°C in an inverted
position for 48 hours. The other set of duplicate plates was incubated
at 15°C in an inverted position for 72 hours. After incubation those
plates which had between 30 and 300 organisms per plate were counted
using a (Juebuc colony counter. The average of the two countable
plates for each set was multiplied by the reciprocal of the dilution
factor to ascertain the total number of both viable aerobic mesophilic
and psychrophilic bacteria/ml of breaded shrimp homogenate. 46
Isolation of E. coli from Shrimp
Shrimp were obtained from a commercial source and stored at
-23°C until experimental procedures were initiated. Aseptically,
25.0 g of raw, headed, shrimp were analytically weighed and transferred into a sterile Waring Blendor. The addition of 225 ml of phosphate buffer to the blender was followed by homogenizing for 2 minutes at high speed. After 5 minutes, when the homogenate liquid settled from the foam, the dilutions were made. All dilutions were originally made from the initial 1:10 dilution into sterile 90 ml buffered solution blanks in such a manner as to give the following dilutions:
1:10, 1:100, and 1:1000. One milliliter from each of these dilutions was transferred into 10 ml of Lauryl Tryptose Broth. Triplicate tubes were inoculated at each dilution. The nine tubes were incubated in a water bath at 35°C. After 24 hours, the positive tubes with growth and gas, were streaked on Levine EMB agar plates and incubated for 24 hours. Typical E. coli colonies, exhibiting a metallic, green sheen, were picked from the plates and streaked on nutrient agar slants. After incubating for 24 hours, the organisms were subjected to the IHViC test series. Typical reactions lor E. coli are ++— , and for Enterobacter are — ++ (Standard Methods for the Examination of Water and Wastewater, 12th ed.), illustrated in Figure 1.
Further tests were conducted on the E. coli suspects for definite identification. The cultures on the nutrient agar slants were streaked onto MacConkey, Desoxycholate, and Levine EMB Agar plates. The former two served as primary plating media for the identification of coliform bacteria, while the latter media served both as identification media, and for re-isolating the organism to insure against contamination. 47
Figure 1
Collform-Fecal Collform AOAC Method
ml ml
1*10 lilOO lilOOO dllut1on LST with duram tubes ho U m m
rapid „ 2k hrs collfornr / All tubes with gas and turbidity J reading here are total collforms
111 BOB'* 2% with duram tubes *35 C for 48 hrs
all tubes with gas and turbidity
readings here are fecal E. coll
iflJ E.C. medium with duram tubes 44.5°C for 2k hrs
confirm In LST or EMB agar
O 48
Typical E. coli colonies were picked and transferred to nutrient agar slants. The IMViC tests were repeated along with detection of motility, indole and in SIM medium, and fermentation of lactose in Brilliant Green Bile 2% and Lauryl Tryptose borth. The identified fecal E. coli on the nutrient agar slants were used throughout all further experimentation for comparison with stock E. coli cultures.
The organism was tested for contamination periodically by inoculating into Lauryl Tryptose Broth at 45°C for 24 hours and restreaking on EMB agar. The typical isolated colony demonstrating a green sheen was picked and restreaked onto a nutrient agar slant.
The IMViC test series was conducted to verify E. coli. The stock cultures were stored at 40°G.
Completed Test on E. coli
Indole Test
A tryptone (Difco) solution was the substrate used to grow the bacteria Cor detecting the production of indole from tryptophan.
Two tubes were inoculated with the test organism, while an uninocu lated tube served as the control. The tubes were incubated for 24 hours at 37°G. Each tube was tested for indole by adding 0.3 ml of Kovac’s reagent and shaking. The violet-red color in the alcohol layer was a positive test for indole, while the yellow alcohol layer in the control indicated a negative result for indole.
Methyl Red Test
M.R.-V.P. Medium (Difco) is recommended for the methyl red test.
Two tubes were inoculated, and an uninoculated tube served as the control. The tubes were incubated at 37°C, and after incubating for
96 hours, 5 drops of methyl red solution wore added. The organism 49 demonstrated an acid reaction by developing a red color. The control reamined yellow.
Voges-Proskauer Test
This test is used to detect the presence of acetylmethycarbinol.
The same M.R.-V.P. Medium (Difco) used for the methyl red test was also used for the Voges-Proskauer test. Two tubes were inoculated with the test organism, while a tube inoculated with E. aerogenes served as a control. The tubes were incubated 48 hours at 37°G.
One ml was extracted from each tube, deposited into another test tube, and 0.6 ml of a-naphthol solution and 0.2 ml of IN KQH were added. The test organism did not develop any color after 3 hours, while the control developed an eosin pink color.
Citrate Test
This test was used primarily to differentiate between E. coli and coliform of the aerogenes group. El. coli does not utilize citrate as its sole carbon source and does not grow on citrate agar; however, E. aerogenes utilizes citrate and grows on the agar. The test organism, along with E. aerogenes as a control, was spotted on a Simmons Citrate Agar (Difco) plate and incubated 24 hours at 37°C.
The agar is initially dark green, but upon utilization of the citrate, it turns blue due to the production of alkaline conditions.
The test culture failed to grow on the agar and caused no color change, while the control organism grew and produced a blue colored zone around the colonies.
Standardization of Inoculum-Staphylococcus aureus
For the purpose of comparison, standard inocula of 10x10^ and
10x10^ coagulase-positive _S. aureus organisms/g of sample were desired. The total inocula in 25 g samples were 250xl06 and 2xl06 organisms, respectively.
The concentration of bacteria was obtained by comparing the
optical density or percent transmittance to a standard curve (number of organisms plotted against various optical densities at a given wavelength).
It was determined to use a wavelength of 525 mu on a Bausch and Lomb Spectrophotometer. The spectrophotometer was allowed to warm up and a sterile cuvette containing sterile Butterfield's
Phosphate buffer to be used as a standard was inserted. The necessary adjustment to 100% transmittance was carried out. <
The stock culture of coagulase-positive S. aureus was inoculated into a 50 ml bottle of Brain Heart Infusion Broth. After 24 hours incubation at 37°C, the broth culture was used to streak slants of
Tryptone Glucose Extract Agar which were in turn incubated for 24 hours at 37°C. Following incubation, each slant was washed with approximately 5 ml of sterile Butterfield's Phosphate Buffer. The resulting suspension was transferred with a sterile pipette to a sterile 50 ml screw top test tuba containing approximately 15 sterile glass beads. This tube was then vibrated vigorously for 5 minutes using a Matheson Scientific Super-Mixer to break up clumps of cells.
One milliliter of the suspension was then aseptically pipetted into a sterile spectrophotometer cuvette. A transmittance of 50% (0.3-0D) was obtained by adding sterile phosphate buffer in 1 ml amounts with intermittent shaking of the cuvette to insure thorough mixing.
This coagulase-positive S^. aureus suspension was serially diluted and duplicate counts were taken at each decimal dilution using the Most Probable Number technique and a pour plate technique with Tryptone Glucose Yeast Extract Agar. The direct plating technique was used to verify the Most Probable Number counts. The procedure was carried out 10 separate times to insure consistent technique.
It was determined that 1 ml from the standardized cuvette containing the suspension gave 9.8x10^ organisms/g in a 25 g sample and 1 ml 2 4 from the 10 dilution gave 9.8x10 organisms/g of a 25 g sample.
Most Probable Number Technique for Coagulase-Positive S. aureus
The same decimal dilutions used for the total plate count were used in determining the number of coagulase-positive S. aureus present. Care was taken that enough dilutions were made to reach an end point. Illustrated in Figure 2.
Three tubes of Trypticase Soy Broth with 10% NaCl were inoculated at each dilution level. These tubes were incubated for 48 hours at 35-37°C.
One of each visibly different colony types which had reduced tellurite was picked from all sample dilutions, tested and transferred to Nutrient Agar slants. Incubation of the slants at 35-37°C, was carried out until growth was evident. The cultured organisms were then microscopically examined for coccal forms and all non-coccal forms were discarded.
Tubes containing 0.2 ml Brain Heart Infusion Broth were inocu lated with small amounts of growth from the Nutrient Agar slants.
Three tubes were inoculated at each dilution level. Following incubation at 35-37°C for 18-24 hours, 0.5 ml of reconstituted
Coagulase Plasma was then mixed thoroughly. The tubes were then placed in a 37“0 water bath and examined periodically over a 4 hour 52
MPN Technique for Coagulase-Positive S. aureus (Kendall,1969) Figure 2
1 ml from previous 1 ml to ne^t dilution tube dilution tube
1 ml DILUTION TUBE
Streak Streak 1 loopful loopful J W TRYPTICASE SOY BBOTH WITH 10# NaCl
Streak VOGEL AND JOHNSON AGAR PLATES
Streak
NUTRIENT AGAR SLANTS
1 Microscopic Examination of Nutrient ' Agar Slants for ICoccal Forms
0.5 ml ^ 0.5 ml 0.5 ml Coagulase Coagulase Coagulase Plasma Plasma Plasma
0.2 ml TUBES OF BRAIN HEART INFUSION 53 interval for clot formation. Tubes showing a positive coagulase reaction were counted and the proper code number obtained from a
Most Probable Number Table (Prescott et al. 1946).
The Most Probable Number of coagulase-positive S. Aureus was calculated by dividing the code number from the MPN table by 100 and multiplying it by the dilution factor of the middle tube used in determining the MPN code number. The resulting figure was the
Most Probable Number of coagulase-positive Sk aureus per gram of sample.
Standardization of Inoculum - Salmonella
A Bausch and Lomb Spectrophotometer 20 was used to standardize inocula of Salmonella typhimurium and Salmonella senftenberg. The wavelength of maximum transmittance was found to be 510 mm. The spectrophotometer was allowed to warm up and then a sterile cuvette containing sterile buffer used as the standard was inserted. The necessary adjustment was made to achieve 100% transmittance.
A stock culture of Salmonella typhimurium was inoculated into a sterile tube of 13HI Broth and incubated at 35°C for 24 hours.
The broth culture was used to streak a slant of Salmonella typhimurium.
The entire procedure was repeated similarly in triplicate for
Salmonella senftenberg. The results indicated that 1 ml contained 7 4.2x10 cells of Salmonella senftenberg.
It can be assumed from the standardization of S_. typhimurium and senftenberg that suspension of S. thompson and S. paratyphi 7 A would be in the range of 4.0 to 4.2x10 cells per ml.
Isolation of Salmonella
The procedure as prescribed by the FDA was followed for the 54
detection of Salmonella in the experimental samples. Ten ml aliquots
of the previously prepared shrimp homogenate were pipetted into 90 ml bottles of Selinite-Cystine and Tetrathionate enrichment broths
respectively. After shaking to insure even distribution of the
inoculum, the bottles were incubated at 35°C for 24 hours. Following
incubation, a loopful of each selective enrichment broth was streaked onto triplicate dry plates of Salmonella-Shigella Agar, and Bismuth
Sulfite Agar, respectively. All selective agar plates were incubated at 35°C for 24 hours. Two to four typical suspect colonies were picked from each selective agar plate and inoculated by streaking the slant and stabbing the butt of Triple Sugar Iron Agar (TSI) slants.
These agar slants were incubated at 35°C for 24-48 hours with the screw-caps loosened. Following incubation, all TSIA slants having alkaline (red) slants and acid (yellow) butts were considered presumptively positive for slamonella. Also, the presence of gas
(cracking or gas pockets in the agar) and hydrogen sulfide production
(black) was noted.
Biochemical and serological tests were done on growth from all presumptive positives in the following manner.
Growth from positive TSIA slant tubes was inoculated into Urea
Broth and incubated at 35°C for 24 hours in a water bath. A positive reaction was a pink color while no color change (yellow) was a negative reaction. All negative tubes were retained and positive tubes dis carded and recorded as negative for salmonellae.
Tryptone Broth was .inoculated with growth from TSIA slants and incubated at 3 5°G for 24 hours. The presence of indole (no red color) constituted a positive test for tryptone. 55
Malonate Broth was inoculated with growth from TSIA slants.
After incubating the slants at 35°C for 48 hours, the utilization of malonate was indicated by a change of the indicator to blue
(alkaline).
Lactose, Sucrose, and Dulcitol Broths were respectively inoculated with growth from TSIA slants. They were incubated at 35°C for 48 hours. The presence or absence of acid and gas was recorded.
At least two tubes of MR-VP Broth was inoculated with growth
from TSIA slants. One tube was incubated at 35°C for 48 hours after which it was checked for the presence of acetylmethylcarbinol
(VP test). The other tube was incubated at 35°C for 96 hours. It was checked for the presence of acid (MR test).
Brain heart Infusion Broth was inoculated with growth from TSIA
slants. The tubes were incubated at 35°G for 8 hours in order to maintain physiological properties necessary for serological testing.
To check for flagellar agglutination, 2.5 ml of formalized saline was added to 5 ml Brain Heart Infusion (BHI) culture. Then 0.5 ml of polyvalent H antiserum was dispensed into a 10 x 75 mm tube and
0.5 ml of the formalized Bill culture was added. The tube was then
incubated in a water bath at 50 to 52°G for at least 60 minutes.
At the end of this time, precipitated agglutinated cells indicated
the presence of polyvalent H antigen. In addition, all cultures were tested for agglutination with polyvalent 0 antiserum, group 0
serum, and Vi serum. All cultures which exhibited typical biochemical
reactions for slamonellac and that gave individual 0 group reactions
as well as positive polyvalent H antiserum agglutination were
recorded as being members of the genus Salmonella. 56
Gas Chromatography
The chromatograms shown in Figures 3 through 5 were obtained
with a Microtex 2000R dual column Gas Chromatograph equipped with a
hydrogen flame detector. For the chromatograms illustrated the column
temperatures were programmed from 60°C to 185°C at 8°/min after a
three minute lag. Injection port temperature was 150°C and the flame
ionization detector temperature was set at 275°G. Sensitivity was 2 noted at 10 and attenuated by 4. Nitrogen as the carrier gas was used at 40 p.s.i. Chart speed was 1 in per minute.
Media I and Media IA were prepared from the dehydrated medium
components as directed and dispensed in 15 ml amounts into 20x150 mm
screw capped tubes. The tubed broth was retained under refrigeration at 5°C until needed.
The broths were inoculated from nutrient agar slants of each of
the test organisms. Tubes of uninoculated broth were included as blank. The inoculated broths were incubated for 18 hours - 1 hour at 35°C and transferred to refrigeration at 5°C. This was done to arrest growth and prevent loss of volatile metabolic components.
The broth cultures were submitted to gas chromatographic analysis after cooling.
A fifty milliliter Erlenmeyer flask containing 10.0 g anhydrous sodium sulfate and stoppered with rubber serum stopper was used to receive the cooled broth cultures. The flask was placed in a 60°C water bath for 10 minutes and was then removed and swirled for 1 minute prior to sampling.
The needle of a 2.5 ml Hamilton gas-tight syringe was inserted through the rubber stopper cap into the liead-spaco of the flask. A two and five tenth milliliter sample of vapor was withdrawn after the sample was agitated by mixing the vapors in the flask. The needle was inserted through the injection port septum, the vapor injected into the sample column and the needle withdrawn after a
15 sec lag.
The syringe was purged with distilled water and checked for contamination by injecting an air sample at a column temperature of
210°C.
The chromatograms shown in Figure 4 through 26 were obtained with a Perkin Elmer 990 Gas Chromatograph operating under the following program.
Instrument: Perkin-Elmer 990 Gas Chromatograph
Carrier Flow:
Tank 60 p.s.i.
Internal 40 p.s.i.
Photometer 25 p.s.i.
Hydrogen: 20 p.s.i.
Air: 30 p.s.i.
Manifold 275°C
Injector 150° C
Column 60-175°C
Rate 8°/min
Initial time 3 min
Cool rate fast
Final time 4 min
Attenuation x 128
Ampli L'i er x 1 58
The chromatographic response was recorded on a Westronics strip
recroder through an Infitronics Automatic Digital Integrator Model
C R 208 operating under the following program.
Track rate 300/min
Mark on
Recorder Linear
Minimum peak 30 sec
Maximum peak 100 sec
Recorder Attenuation 10 x
Peak Sensor gain 10 x
Wright and Mundy broth was prepared from the dehydrated media as directed and dispensed in 15 ml amounts as described previously
for Media I and Media IA.
Each of the broth cultures was analyzed by the method described previously to obtain gas chromatographic patterns of head space vapor.
Gas Chromatographic Analysis of Standard Reference Solution
The chromatograms shown in Figure 4 were standardized using a 5.0 ppm aqueous solution of reagent grade n-butanol. The solution was substituted for the 15 ml broth culture and the procedure as described previously was employed.
The chromatograms shown in Figure 4 were standardized using a
5.0 ppm aqueous solution of reagent grade anisole (boiling point
150°G - 154°C).
Inoculation of Seafood Materials and Identification of Organisms
homogenates of the seafood material were made by adding 25 g of the product to 225 ml of buffer and blending in a sterile screw 59 capped Waring Blendor jar for 2 minutes at low speed. Samples were
inoculated with cultures of each organism used in these experiments.
An 18 hour Wright and Mundy broth culture of each organism was used
for this purpose. The final density of the sample to be analyzed 3 was approximately 12x10 organisms prior to incubation for 18 hours
at 3 5°C.
Uninoculated controls were retained with the inoculated samples.
Each of the broth cultures was analyzed by the method described previously to obtain gas chromatographic patterns of head-space vapors. RESULTS
Chromatograms of head-space vapors of microbial metabolites were interpreted by visual examination. Retention times, the presence or absence of peaks, and the ratio of peak heights with relationship to each other were particularly useful. No quantitation or identification of peaks was attempted in these studies.
Duplicate samples were grown for each culture, and multiple replicates of each sample were examined by Flame Ionization Detector
Chromatography. For purposes of simplifying the material presented
in this paper, only representative programs of each sample are shown, since results wore consistent between both duplicate cultures and replicate chromatographic determinations. Duplicate chromatograms of organisms showing utilization of different media types are pre sented.
A signature or fingerprint for each bacterium was established by arranging the various peaks, designating metabolites in an order corresponding to increasing retention times. Such signatures for the seven bacteria are listed in Table 2. For each peak, the designated
R^ value is the mean of the corrected retention time, in seconds,
for all test bacteria. In the tabular presentation, a blank signifies that no product with the designated retention time was observed.
A signature or fingerprint for each bacterium was established using the two media described previously and carried out on the
60 TABLE 2 FOOD SAMPLE RETENTION TIMES
Sample Code Retention times minutes
Shrimp, Green Headless 2-1 2.0— 5.2-
Shrimp, Green Headless 401-1 1.7— -5.2-7. 7-13—
Shrimp, Green Headless 401-1M 2.0-- 5.2-7.7-.0.2-14
Shrimp, Green Headless 402-1 — 2. 7-5.2-7. 7-- 13----
Shrimp, Breaded 500 1.7--- 5.2----11.3----
Shrimp, Peeled & Deveinod 502 ------
Shrimp, Breaded 503 ------6.2------
Shrimp, Breaded 600 1. 7------6. 7-9-11-13-
Shrimp, Breaded 603 ------7 . 4 - 1 3 -
Shrimp, Fresh 700 - _ -6.9-9.5-11-
Fish, Whiting BKW -3.4 4.7 6.7-9.5-11-12.5
Fish, Cod HKC ----- 5.2— 7.0— 12.0---
Oysters, Natural Wet Pak P.J. -- 2.7 5.2------62 two gas chromatographs previously identified. Such signatures for the seven organisms tested are listed in Figures 5 throu^i 14.
By means of the signatures thus obtained, the organisms could be distinguished from one another by considering the presence or absence of peaks. Alternatively, the bacteria could be distinguished by considering the quantities of the various products formed. Thus a comparison was made among the seven bacteria to determine the FID- positive metabolites which could be used to distinguish among the organisms on the basis of significant differences in product yield.
For this purpose, the bacteria were considered to be distinguishable if any of the following applied, (i) The two strains being compared yielded peaks having identical retention times but different areas.
(ii) One of the two organisms produced the metabolite, either in significant or trace amounts, whereas a substance with the same value was not found in any replicates of the second organism.
(iii) One of the organisms produced the compound in significant quantities, whereas the other produced only traces.
All cultures used in this study were pure and fitted closely the description of the represented organism as described in Bergey’s
Manual. Typical G.L.C. Chromatograms shown in Figures 5 through 14 illustrates differences between them.
Organisms commonly referred to as "Enterics" consistently evi denced the presence of the same group of volatiles but with differing magnitudes of concentration. Among the organisms tested, volatile metabolite concentrations varied from infintesimal to large quantities.
The relative amounts of particular metabolites were relatively constant for a given organism during multiple analysis over a three year Figure 3. Gas chromatograms of Media I (o) and Media II (c>) on Machine I. 10
9
Tine (Minutes) period. Differences in peak size were detected over the duration of
the program but few of significance were found throughout the course
of these analyses.
Uninoculated broth samples of each media yielded consistent
insignificant peaks (Figure 3). The peaks were insignificant in
size in relation to other peaks present. These peaks were eluted at
times different from major peaks and thus did not interfere with the
chromatographic patterns of the sample.
The standard solution of 5 ppm aqueous n-butanol was used as a standard reference for preliminary studies and consistently yielded
a peak with a retention time of 12 minutes (132°G) (Figure 4). The standard solution was analyzed frequently during the preliminary studies to determine if the column characteristics had changed.
The reference standard was changed to redistilled Anisolc,
Rt-14. 6 min, temperature-152.8°G, as the original standard was felt to interfere with late eluting peaks found in some food sample
chromatograms. Anisole was used at 5 ppm and proved to maintain
consistent retention times and peak heights throughout the last two years of this study (Figure 4).
Escherichia coli cultures produces chromatograms similar to those in Figure 5. The metabolically originated volatiles produced peaks eluting after 7 min-92°C on both Media I and Media II chromato
graphed on Machine I, Figure 5. When Wright-Mundy broth was used as
culture media and the volatiles chromatographed on the Perkin-Eliner
900, two peaks were demonstrated. One major peak eluted at 7 min,
92°G, and another at 9.5 min, H2°G. This chromatographic representa
tion continued through the completion of this research (Figure 5).
In an effort to determine if clear signatures can be developed Peak He If, hr. 10 6 7 __ - I iue4 Gaschromatograms of IMediacontaining Figure n-butanol, 4.MediaII containing n-butanol Q Pi ;i t' .1 ©
ei I (Anisole)IIMedia (*»). n ei Icnann eitle nsl. ybl: ei ( , (o), MediaII (I Media Symbols:andMediaII containing redistilled Anisole. 6 7 ie (Minutes) Time 8 9 10 1 2 13 12 11 4 15 14
C/1 O' Peak Hivl.r,lit, 10 •i i i § i i i \ i i • T i i j I j iue5 Gaschromatograms of coli E.in Figure IMedia and Media II5.on MachineIand MachineII. Symbols: Media I, (o), Media IMachine MediaII, (',>), Symbols: MachineI II,Media (c). Machine ie (Minutes)Tine 10 12 o ' O in samples incubated for very short time periods, chromatograms were developed for E. coli after 1 hr, 2 hr and 10 hr incubations. Peaks
for all three incubation times eluted after 7 minutes. The magnitude of the peak increased with incubation time and produced characteristic signatures for that particular incubation time (Figure 6).
Enterobacter aerogenes produced peaks that eluted after 5.2 min,
7.7 min and 13 min* The 7.7 min and 13 min peaks were similar to that of E. coli. The 5.2 min peak was of smellier magnitude than the other two peaks which gave it a rather unique signature as seen in Figure 7. Some strains of E. aerogenes isolated from seafood samples gave chromatographic patterns similar to Figure 8. This pattern was not detected in all samples isolated.
The chromatographic patterns of Salmonella typhimurium were similar in retention times with those of li. aerogenes but differed markedly in magnitude of each peak (Figure 9). The 5.2 min and 7.7 min peaks were of similar magnitude with the 13 min peak being very small.
The heights of the 5.2 and 13 min peaks differentiated the two organisms.
Salmonella thompson showed chromatographic patterns characterized by peaks eluting after 5.2 and 7.7 min. In cultures incubated for
18 hours at 35°C, the 7.7 min peak was the larger of the two metabolites.
In cultures incubated for 48 hours at 35°C, the 5.2 min j)eak was the larger of the two (Figure 10).
Salmonella senftenberg was characterized by peaks of 8.2 and
10.7 min retention time. These times were unique among the Salmonella tested as shown in Figure 11.
Chromatographs of Salmonella anaturn evidenced an early peak Figure 6 . Gas chromatograms of Eh coli in Media II on Machine II after 1 hr (o),
2 hr (3>), and 10 hr (^) incubation at 35°C.
r - o
5 6 7 8 9 10 11 12 13 14 15 Time (Minutes)
ooO' Peak Height 10 4_ 5_ 3 iue7 Gaschromatograms ofFigure Enterobacter 7. aerogenes in MediaII on MachineII. 7 9 8 7 6 ie (Minutes) Time 1 2 31 15 14 13 12 11 'O ' O Figure 8 . Gas chromatograms of Enterobacter aerogenes in Media II on Machine II.
Time (Minutes) Peak Height 5 -- iue . Gaschromatograms Salmonellaof Figure 9.typhimurium in MediaII on MachineII. ie (Minutes)Time 11 Figure 10. Gas chromatograms of Salmonella thompson in Media II on Machine II after 18 hrs
and 48 hrs incubation at 35°C. Symbols: 18 hrs (-3), 48 hrs (O) •
2 31 4 5 6 7 8 9 11 12 13 14 1510 Time (Minutes)
-j t o Peak Height 6 4 5 2 __ iueI. Gaschromatograms ofSalmonella Figure IX.senftenberg in MediaII developed on MachineII. ie (Minutes)Time 11 •■j 9 0 74 after only 2.9 min (Figure 12). This peak, insignificant in size on all of the other chromatographic patterns, was exceptionally large and served to characterize this organism.
Staphylococcus aureus (Figure 13), grown in Wright-Mundy Broth containing IX dextrose, failed to elute a significant peak. One small peak, attributed to media and air, eluted after one minute.
Staphylococcus aureus, grown in Wright-Mundy Broth containing 1X mannitol, eluted peaks at 5 and 11 min. The 5 min peak was greater in magnitude than the 11 min peak (Figure 13).
A mixed culture of Escherichia coli, Enterobacter aerogenes and Salmonella typhimurium incubated 18 hours at 35°C produced peaks as demonstrated in Figure 14.
The retention times of 2.5, 4.2, 6.5, 9, 10 and 12.5 min did not correspond to those retention times found in chromatographs of pure cultures of E. coli and Salmonella typhimurium previously described.
These retention times correspond to the chromatogram of Enterobacter aerogenes as seen in Figure 8.
Samples of food materials collected from multiple sources were chromatographed to determine microbial flora fermenting dextrose.
Shrimp sample 2-1 (Figure 15) chromatographed using Media II and Machine II demonstrated no appreciable metabolic response.
Peaks of little significance appeared at 2.0 and 5.2 min. This sample had been held in frozen storage at -20°C for 2 years prior to examina tion and was for the purpose of this experiment devoid of significant numbers of microorganisms or were of the type that failed to ferment dextrose.
Shrimp sample 401-1 (Figure 16) showed peaks with retention Peak Height 5 __ iue1. Gaschromatograms of SalmonellaFigure 12. anatum in MediaII on MachineII. ie (Minutes)Time 11 Peak Height 10 5 __ iue1. Gaschromatograms Figureof a 13. mixtureof pure cultures of typhimurium, and Enterobacter aerogenesin MediaII on MachineII. ie (Minutes)Time E. coli. Salmonella 11 / o -j Figure 14. Gas chromatograms of Staphylococcus aureus in Media II with 1% dextrose and
Media II with 1% mannitol on Machine II. Symbols: dextrose (a), mannitol (a ).
i ! 1 2 3 4 5 6 7 8 9 10 11 13 14 1512 Time (Minutes)
Si •si Peak Height 10 2 3 4 5. 8 8 7 9 iue1. Gaschromatograms Figureof green 15. headlessshrimp in XIMedia on MachineII. . K 7 9 8 7 6 ie (Minutes)Time 01 2 13 12 11 10 4 15 14 Peak Height iue1. Gaschromatograms Figureofgreen 16. headlessshrimp sample, 401-1, inoculated with Enterobacteraerogenes in IIMedia on MachineII. ie (Minutes)Time times of 1.7, 5.2, 7.7 and 13 min. This sample inoculated with
Enterobacter aerogenes demonstrated typical gas chromatographic patterns for this organism.
The inoculated sample 401-1M (Figure 17) exhibited chromatograph peaks with retention times of 2.0, 5.2, 7.7, 10.2 and 14 min. The sample was inoculated onto Wright-Mundy Broth containing 1 % mannitol and chromatographed on the Perkin-Elmer 900 (Machine II).
The inoculated sample 402-1 (Figure 18) was analyzed using
Media II and Machine II. Retention times of peaks were 2.7, 5.2,
7.7 and 11.3 min. These times correspond favorably to the times recorded for the pure culture of the inoculum, Salmonella anatum
(Figure 12).
The uninoculated breaded shrimp sample, code 500, demonstrated peaks at 1.7, 5.2 and 11.3 min retention times. The 1.7 and 11.3 min peaks are of small magnitude as compared to the 5.2 min peak
(Figure 19).
Peeled and deveined shrimp (sample 502) did not demonstrate chromatographic peaks when it was submitted to analysis. This would indicate that the level of microorganisms was below the detection limits of the instrument or that the microorganisms were of the type that failed to ferment dextrose. The total plate count recorded for this sample was 93xl0^/g with negative counts for total coliform and E. coli. Staphylococcus-coagulase + counted 3.6/g (Figure 20).
One peak was demonstrated on the chromatograph of breaded shrimp, code 503. The lone peak seen at 6.2 min (Figure 21) was of medium magnitude and "tailed” despite refining techniques applied to the instrument. The T.P.C. of the sample was 1x10^ Peak Height 5 __ iue1. Gaschromatograms of Figuregreen17. headlessshrimp sample, 401-1M, inoculated with Enterobacter aerogenesin IIMedia containing 12 mannitolon MachineII. ie (Minutes)Time 11 oo Peak Height 5 __ iue1. Gaschromatograms Figure of green 18. headlessshrimp sample, 402-1, Inoculated with Salmonella anatumin IIMedia on MachineII. ie (Minutes)Time 112 11 Peak Height 5 __ iue1. Gaschromatograms Figureof breaded19. shrimp sample, 500, in MediaII on MachineII. ie (Minutes)Time 11 oo W Figure 20 Gas chromatograms of peeled and deveined shrimp sample, 502, in Media II on
10 Machine II.
9
8
7
6
JJ J= c 60 5, •H 01 ac
0) Pu 3
2
1
6 7 8 9 10 11 12 13 14 15 Time (Minutes) Peak Height 5 __ iue2. Gaschromatograms Figureof breaded 21.shrimp sample, 503, in MediaII on MachineII. ie (Minutes)Time 86 with coliform and E. coli being absent. Coagulase positive Staphylo coccus were present at 9.l/g.
Peaks with retention times of 1.7, 6.7, 9, 11 and 13 min were demonstrated in the breaded shrimp sample 600 (Figure 22). Major peaks were found at 1.7, 9 and 13 min. The total plate count of
94,000 included 460 coliform, less than 3 E. coli and 9.1 coagulase positive Staphylococcus.
Sample 603 demonstrated (Figure 23) large peaks at 7.4 and 13 min retention times. The sample of frozen breaded shrimp of Indian origin contained a total of 4x10^ organisms/g with greater than 1100/g coliforms present. E. coli, Salmonella and coagulase positive
Staphylococcus were not detected in the sample run by A.O.A.G. methods.
Fresh Gulf Coast shrimp (pink) were analyzed and found to give retention times of 6.9, 9.5 and 11 min (Figure 24). Total aerobic counts of 65xl0*$ were found in association with 290/g coliform. E. coli was present at 160/g and coagulase-positive
Staphylococcus and Salmonella were not detected in the sample.
Atlantic Whiting samples yielded chromatograms with peaks at
3,4, 4.7, 6.7, 9.5, 11 and 12.5 min retention times. A.O.A.C. pro cedure results yielded a total plate count of 4xl05 organisms per gram. Coliforms numbered 240/g and E. coli was found at 120/g.
Staph and Salmonella could not be detected (Figure 25).
Breaded cod fish portions were analyzed and found to produce peaks with retention times of 5.2, 7 and 12 min. £. coli were found to represent 95/g organisms in the total coliform count of
210/g. Salmonella and coagulase-positive Staph were not detected in the standard microbiological procedures (Figure 26). Peak Height 5 __ iue 2 Gaschromatograms of breadedFigure 22.shrimp sample, 600, in MediaII on MachineII. ie (Minutes)Time 11 Figure 23. Gas chromatograms of breaded shrimp sample, 603, in Media II on Machine II.
1 2 34 5 6 7 8 9 10 12 1311 14 15 Time (Minutes) Peak Height iue2. Gaschromatograms offreshFigure shrimp24. sample, 700, in IIMedia on MachineII. ie (Minutes) Time Peak Height 10 Figure 25. Figure Gas chromatograms of breaded fish sample, BKW in Media II on Machine II. on Machine II in Media BKW sample, fish breaded of chromatograms Gas ie (Minutes)Time Peak Height: 10 iue 6 Gaschromatograms of breaded 26.Figurefish sample, BKC, in MediaII on MachineII. ie (Minutes)Time 11 Fresh oysters packed in their natural juices were submitted to analysis (Figure 27). Peaks of 2.7 and 5.2 min were observed.
Standard microbiological procedures were not carried out on this sample.
Although many of the chromatographic patterns represented in this study were heterogeneous in regard to coliform organism, members of the family Enterobacteriaceae could be detected (Table 3). As has been described previously, this method would be useful as a screening procedure in determining whether members of the family
Enterobacteriaceae were present. Application of this method for seafood analysis to determine the presence of specific organisms would require the absence of interfering organisms. Peak Height 5 __ iue2. Gaschromatograms ofcommercial 27.Figure oyster sample in MediaII on MachineII. ie (Minutes)Time 11 TABLE 3 PEAK RETENTION TIMES
Test Culture Retention Times
Escherichia colj ------7------9.5-
Enterobacter aero genes------5.2----- 7.7— ■--- 13---
Salmonella typhimurium — — ---5.2------7.7----— ---- 13--
Salmonella thompson ------5.2----- 7.7------13--
Salmonella senftenberg------8.2— 10.7------
Salmonella anatum------2.9---5.2------7.7------11.5----
Staphylococcus aureus ( M ) -- 5.0------— ----- 11.0----
Blank Broth (Media I ) ------
Blank Broth (Media II) 1.5------n-butanol Standard------1.2,0-
Anisolu Standard ------14.6 DISCUSSION
The seafood processor is concerned with both the safety and the spoilage potential of his product.
In most cases food poisoning incidents have been caused by one or more of the following: contaminated ingredients, unsanitary conditions in the production plant, processing failure, cross contamination after processing, and mishandling either by the processor, consumer, or during shipment, distribution and retail handling.
A very strong, well structured quality control program that uses bacterial indicators to measure safety or spoilage potential of foods must be an inherent part of any seafood processing industry. The sensitivity and diversity of the methods used should realistically reflect the purpose intended.
The seafood industry has through its own volition established tests and procedures designed to reflect those indicators of spoilage or public health significance that are applicable to the products being produced.
Depending on the kind of information required, different tests and organisms may be used.
In an attempt to determine the total viable aerobic population of a sample, Total Plate Counts are used by the industry with knowl
edge of its limitation ever mindful.
Total count does not indicate the safety of a food. A high
95 96
count does not mean that a seafood product is a public health risk, nor does a low count mean positive safety regarding the consumption of a seafood product.
The presence of small amounts of toxin or small numbers of pathogens such as Salmonella cannot be predicted just by knowing the total viable count in the seafood product.
The need is to study the bacterial ecology of a particular food based on its point of origin, process and packaging and their effects on the safety and spoilage potential of this seafood product.
In an attempt to determine total plate counts using the techniques associated with gas chromatography, total aerobic plate counts were obtained using A.O.A.C. approved techniques. This count was compared to the total peak area, in counts/sec/inv, of the specific chromatogram.
Figure 21 depicting sample 503 of breaded shrimp has a total peak area of 37,603 counts/sec/mv and a total plate count of 1x10^ organisms/g or 11.4 organisms per coimt/sec/mv. However, shrimp sample 600 (Figure 22) produced a chromatogram containing a total peak area of 380,493 counts/sec/mv and a total plate count of
94x10^ organisms/g or 4.0 counts per organism.
This type of correlation was found throughout the investigation and points out dramatically that it is not the total number of organisms that is important in gas chromatography but the type of organisms present. Sample 503 ( Figure 21) contained 10^ organisms/ g but contained less than 3 coliforms and Lh coli, 9.1 Staphylococci and 0 Salmonella per gram. Sample 600 (Figure 22), however, con tained only 94x10^ organisms/g. Important to note is that 460 coliform per gram were also found to be present. From this obsor-
* vation, the total counts/sec/mv would tend to indicate the total number of organisms fermenting dextrose and not the total viable count of organisms present in a given sample.
The coliform group has been widely used to indicate, to a certain degree, the quality of seafood products. The presence of these organisms can be an indication of inadequate sanitation and of product mishandling.
One should recognize that the presence of these organisms is not an ipso facto indication of fecal contamination. The presence of the more specific bacterium in this group, Escherichia coli, has been associated with fecal pollution because this organism is found in the intestinal tract of man and animals.
Significant numbers of E. coli in cooked frozen seafoods indicate contamination after cooking. If handling or packaging equipment is improperly cleaned and sanitized, E. coli could grow on the equipment and contaminate the finished cooked product.
Escherichia coli here indicates not fecal contamination, but poor manufacturing practices.
When Escherichia coli is found, it is important that the source be determined and its significance be judged.
The sea (bod industry recognizes the importance of not only
E. coli but the entire coliform group as indicators of multiple quality problems and includes tests and techniques to determine their presence in the complete quality program.
These indicator organisms are used, of course, to indicate the real or potential hazard associated with contamination of a seafood product with Salmonella. As the industry is well aware, tests 98
for Salmonella are not only exacting but time consuming, sometimes
involving one week or more for completion.
In an attempt to determine the feasibility of using G.L.C. as
a rapid screening mechanism for this large group of organisms referred
to as "Enterics," G.L.C. profiles on pure and mixed cultures of several
of these organisms were obtained (Figures 5-14).
By the selection of proper growth media, appropriate column
packing material, temperature programming and a detector capable of
distinguishing the various metabolic compounds produced, it has been
possible to obtain clearly defined and differentiated "fingerprints"
for all the genera and to a certain degree the species of organisms
tested. From those "fingerprints" listed, the retention times of the
various metabolic components are shown in Table 3. As can be observed, many of the organisms produce metabolites with similar retention times. By observing the corresponding chromatograms it can
be seen that the peaks will vary in magnitude and it is this combined with the R required to produce a differential fingerprint.
Staphylococci have been proposed as indicators of a potential hazard in food. Since staphylococci grow or are present on hands and nasal passages, excessive numbers of these organisms in a food should
be investigated. The nature of the corrective measures used should
depend upon the food product itself and the mode of entry of staphylococci into this product.
Development of gas chromatographic patterns for this organism would have been desirable but the method was ineffective for measuring
the volatilos present (Figure 13). Staphylococci grown in Wright-
Mundy broth containing IX mannitol replacing L?o dextrose produced 99 chromatograms as seen in figure 13. In order to obtain the chromato grams it is necessary to prepare a different medium which defeats the idea of a quick, simple screening test. It may, however, be possible in future experiments to produce chromatograms of all the organisms used here using mannitol as the fermentable substrate.
The seafood samples inoculated with Salmonella anaturn and Entero- bacter aerogenes displayed typical pattern for these organisms when analyzed in broth cultures (Figure 16 and 18).
The uninoculated seafood samples developed a myriad of chromato graphic patterns indicating the possible presence of several
Enterobacteriaceae (Figure 15 to 27). Peaks characteristic of
Enterobacter and E. coli were observed. It could not be determined from the chromatograms if other members of this family were present.
Tests of the samples confirmed the presence of coliform and E. coli but the presence of Salmonella could not be demonstrated.
The methods and procedures as described offer a feasible alternative in the effective rapid screening of seafood for those public health organisms capable of using dextrose as a fermentable substrate. Procedures for screening of samples for fingerprints after one hour incubation can be developed (Figure 5).
It is likely that the ability to distinguish between organisms will be increased with technology, and direct examination of a food for a speci fic organism, such as Salmonella. would not require the absence of other contaminating organisms. The method at the present state of technology should bo used only as a rapid screening technique and must be used in conjunction with routine procedures until a history or data file is assembled for the specific machine and machine location. 100
Because the method is still of an empirical nature, there must
be strict adherence to the procedures and materials employed in the
program. If results are to be duplicated, one must show consistency
in the preparation of media, incubation times and temperatures, gas
sampling procedures, syringe injection procedures, and machine "set-up11
procedures. Glassware must be free of volatile contaminants, reagents
must always be free of contamination and the machine housed in a
stable environment free of radical temperature and relative humidity
changes.
With all of its limitations taken into account, the procedure presented here may be considered as a new procedure for a more rapid means of screening for the presence of microorganisms of public health
significance in seafood products. The groat sensitivity to the
detection of microorganisms decreases very appreciably as the time
loss entailed prior to detection and identification. Further
investigations of this procedure may make it more rapid and more
specific. SUMMARY
Analyses were made to differentiate genera of the family
Enterobacteriaceae by gas-liquid chromatography (GLC) of the volatile
compounds produced over dextrose fermenting bacteria during growth.
Organisms of the genera Salmonella, Enterobacter, Escherichia, and
Staphylococcus were studied.
Escherichia coli and E. aerogenes gave distinctly different head space gas chromatographic patterns while members of the Salmonella gave diverse and overlapping gas chromatographic patterns. Staphylo coccus aureus failed to produce a chromatographic response over dextrose but responded well when grown in synthetic media (AOAC) containing 1% mannitol.
Analyses of seafood samples inoculated with Enterobacter aerogenes and Salmonella anaturn indicated the presence of each organism in chromatographic patterns similar to those detected over broth cultures.
Uninoculated seafood materials including frozen green headless,
frozen peeled and dcveincd, frozen raw breaded, and fresh shrimp were analyzed for Hie presence of bacteria of sanitary and public health significance.
Peaks with retention times similar to those of E. coli and
E. aerogenes were found in some samples. Salmonella was not characterized by any of the peaks detected in these chromatograms.
1.01 102
Interfering organisms increased the difficulty of identifying specific organisms of tihe Enterobacteriaceae due to the production of extraneous volatile materials.
Gas samples drawn from above samples of Staphylococcus aureus failed to produce chromatographic patterns when grown in dextrose but they produced chromatographic patterns when this organism was grown in mannitol.
Gas-liquid partitioning of volatile compounds produced during growth offers possibility as a quality control procedure and can be applied to routine microbiological examination of seafood. LITERATURE CITED
Abel, K., DeSchmertzing and Peterson, J. I. 1962. Classification of microorganisms by analysis of chemical composition. J. Bact. 85:1039-1044.
Adinarayanan, N., Foltz, V. D. and McKinley, F. 1965. Incidence of Salmonella in prepared and packaged foods. J. Inf. Dis. 115:19-26.
Alexander, W. F. and Milone, N. A. 1936. The significance of the Escherichia-Aerobacter group in pasteurized milk. J. Bact. 31:574.
Amsterdam, D. and Wolfe, M. W. 1968. Comparison of reagent- impregnated paper strips and conventional test for distinguish ing Escherichia from Ajerobacter:Correlation with colonial morphology. Appl. Micro. 16:1460-1464.
Annual Summary, Salmonella Surveillance, 1965. National Center For Disease Control Publication, Atlanta, Ca.
Annual Summary, Salmonella Surveillance, 1968. National Center for Disease Control Publication, Atlanta, Ca.
Baer, E. F., Franklin, M. K. and Gildun, M. M. 1966. Efficiency of several selective media for isolation of coagulase-positivo Staphylococci from food products. J. A.O.A.C. 49:267-269.
Bartam, M. T. and Black, L. A. 1937. Detection and significance of the coli form group in milk. II. Identi fication of speci es isolated. Food Res. 2:21-26.
Bassette, R. and Claydon, T. J. 1965. Characterization of some bacteria by gas chromatographic analysis of head-space vapors from milk. J. Dairy Sci. 48:775.
Bassette, R., Bawdon, R. E. and Claydon, T. J. 1966. Production of volatile materials in milk by some species of bacteria. J. Dairy Sci. 50:167-171.
Bawdon, R. E. and Bassette, R. 1966. Differentiation of Escherichia coli and Aerobacter aerogenes by gas liquid chromatography. J. Dairy Sci. 49:624-6277
103 104
Bayliss, B. G. and Hall, E. R. 1965. Plasma coagulation by organisms other than Staphylococcus aureus. J. Bact. 89:101-105.
Bemade, M, A. 1968. Breading contributes to the microbial popula tion of frozen breaded products. Comm. Fisheries Review 20(3): 6-10.
Boring, J. R. 1958. Domestic fish meal as a source of various Salmonella types. Vet. Med. 53:311.
Brooks, J. B. and Moore, W. E. C. 1969. Gas chromatographic analysis of amines and other compounds produced by several species of Clostridium. Can. Jour. Micro. 15:1433-1447.
Brown, D. F. 1971. Examination of food for coliform organisms. "Food Microbiology”. U.S. Dept, Health, Education and Welfare. Cincinnati, Ohio.
Bryan, F. L. 1968. What the sanitarian should know about Staphylococci and Salmonella in non-dairy products. I. Staphylococci. J. Milk Food Tech. 31:110-116.
Butler, Boyce Hodge. 1970. Comparison of quantitative tests for coliform in shrimp. A Thesis. Louisiana State University.
Butterfield, C. T. 1932. The selection of a dilution water for bacteriological examinations. J. Bact. 23:355-368.
Carter, C, H. 1960. Egg yolk agar for isolation of coagulase- positivc Staphylococci. J. Bact. 79:753-754.
Casman, E. P., Berdoll, M. S. and Robertson, J. 1963. Designation of Staphylococcal enterotoxin. J. Bact. 85:715-716.
Chapman, G. 11. 1946. A single culture medium for selective isolation of plasma-coagulating Staphylococci and. for improved testing of chromogenesis, plasma coagulation, Mannitol fermentation and the stone reaction. J. Bact. 51:409-410.
Clark, H. F. and Kabler, P. W. 1964. Reevaluation of the significance of the coliform bacteria. J.A.W.W.A. 56:931-936.
Cone, R. D. and Lichowich, R. V. 1969. Differentiation of Clostridium botullnum types A, B, and E by pyrolysi.s-gas-liquid chromatography. Appl. Micro. 19:138-145.
Crisley, F. D., Peller, J. T. and Amgelotti, R. 1965. Comparative evaluation of five selective and differential media for the detection and enumeration of coagulase-positive Staphylococci in foods. Appl. Micro. 13:140-156.
Dack, 0. M, 1956. Food Poisoning. 3rd ed. University of Chicago Press. 105
Difco Laboratories. Difco Manual of Dehydrated Culture Media and Reagents for Microbiological and Clinical Laboratory Procedures. 9th ed. Detroit, Mich.
Dockstadler, W. B. 1970. Bacteriological Branch, Division of Microbiology, Office of Food and Nutritional Sciences, Food and Drug Administration. Personal communication.
Edwards, P. R. and Ewing, W. H. 1962. Identification of Entero bacteriaceae. 2nd ed. Burgess Publishing Co., Minneapolis, Minn. p. 116-128.
Elston, H. R. 1969. Comparative study of the paper strip technique with the micro-spot plate and liquid medium methods for Niacin. Amer. Rev. Resp. Dis. 100:729-731.
Elston, Harry R., Baudo, Judith A., Stanek, JoAnn P. and Schaab, Mercedes. 1971. Multi-biochemical test system for dis tinguishing enteric and other gram-negative bacilli. Appl. Micro. 22:408-414.
Farshy, David C. and Moss, C. Wayne. 1970. Chararacterization of Clostridia by gas chromatography:Differentiation of species by trimethylsilyl derivatives of whole-cell hydrolysates. Appl. Micro. 20:78-84.
Fieger, E. A. and Novak, A. I'. 1961. Microbiology of shellfish deterioration. I’ish as Food. 1:561.
Figueiredo, M. P. de. 1970. Microbial indices for quality assurance. Food Tech. 24:51-55.
Finegold, S. M. and Sweeney, E. E. 1961. New selective and differ ential medium for coagulase-positive Staphylococci allowing rapid growth and strain differentiation. J. Bact. 81:636-641.
Fishbein, M. and Surkiewicz, B. F. 1964. Comparison of the recovery of Escherichia coli from frozen foods and nutmeats by confirmatory incubation in E.C. medium at 44.5 and 45.5°C. Appl. Micro. 12: 127-131.
Fitzgerald, G. A. and Conway, W. S. 1937. Sanitation and quality control in the fishery industries. Am. J. Pub. Health 27:1094.
Georgala, D. L. and Hurst, A. 1963. The survival of food poisoning bacteria in frozen foods. J. Appl. Bact. 26:346-358.
Gilden, M. M., Baer, E. F. and Franklin, M. K. 1966. Comparative evaluation of a direct plating procedure detecting coagulase- positive Staphylococci in foods. J. A.O.A.C. 49:273-275.
Gillespie, W. A. and Alder, V. G. 1952. Production of opacity in egg yolk media by coagulase-positive Staphylococci. J. Path. Bact. 64:187-200. 106
Giolitti, G. and Cantoni, G. 1966. A medium for the isolation of Staphylococci for foodstuffs. J. Appl. Bact. 29:395-398.
Green, M. 1949. Bacteriology of shrimp. II. Quantitative studies on freshly caught and iced shrimp. J. Food Res. 14:372-383.
Griffin, A. M. and Stuart, C. A. 1940. An ecological study of the coliform bacteria. J. Bact. 40:83-100.
Griffiths, F. F. and Stansby, M. C. 1934. Significance of the bacterial count and the chemical tests in determining the relative freshness of haddock. Trans. Am. Fisheries Soc. 64:401.
Grodner, R. M., McCullough, W. E. and Woolverton, J. 1973. In Publication.
Grunberg, E., Titsworth, E., Beskid, G., Cleelend, R., Jr. and Delorenzo, W. F, 1969. Efficiency of a multitest system (Enterotube) for rapid identification of Enterobacteriaceae. Appl. Micro. 18:207-213.
Guarino, Phillip A. and Kramer, Amihud. 1969. Gas chromatographic analysis of head-space vapors to identify microorganisms in foods. J. Food Sci. 34:31-37.
Gulasekharam, J. T. Vclaudapillai and Niles, G. R. 1956. The isolation of Salmonella organisms from fresh fish sold in a Colombo fish market. J. llyg. 54:581-684.
Gunderson, M. F. find Rose, K. 1). 1948. The survival of bacteria in a precooked fresh frozen fish. Food Res. 13:254-263.
Hajna, A. A. and Damon, S. R. 1950. Polyvalent Salmonella ,fH,r agglutination as a rapid screening test for Salmonella organisms. U.S. Public Health Report 65:116-118.
Hajna, A. A. and Perry, A. 1939. Optimum temperature for differ entiation of Escherichia coli from other coliform bacteria. J. Bact. 39:275-283.
Hall, H. E., Brown, D. F. and Lewis, K. H. 1967. Examination of market foods for coliform organisms. Appl. Micro. 15:1062- 1069.
Harrison, F. C., Perry, M. and Smith, P. W. P. 1926. The bacteri ology of certain sea fish. National Research Council of Canada 19:1.
Hartsell, S. E. 1949. The testing of frozen eggs for pathogens. J. Milk Tech. 12:107.
Hartsell, S. E. 1951, The longevity and behavior of pathogenic bacteria in frozen foods:The influence of plating media. Am. J. Pub. Health 41:1072. Henis, Y., Gould, J. R. and Alexander, M. 1966. Detection and identification of bacteria by gas chromatography, Appl. Micro. 14:513-524.
Henis, Y. and Alexander, M. 1968. Differentiation of Streptomyces strains by gas chromatography. Antonie Van Leeuwenhoek 34:159-164.
Hobbs, B. C. 1953. Food Poisoning and Food Hygiene. Edward Arnold and Co. London.
Hobbs, B. C., Kendall, M. and Gilbert, R. J. 1968. Use of phenol- phthalein diphosphate agar with polymixin as a selective medium for the isolation and enumeration of coagulase-positive Staphylo cocci from foods. Appl. Micro. 16:535.
Hodge, B. E. 1960. Control of Staphylococcal food poisoning. Public Health Reports 75:355-361,
Hunter, A. C. 1920. Bacterial decomposition of salmon. J. Bact. 5:353.
Hunter, A. C. and Linden, B. R. 1923. An investigation of oyster spoilage. Am. Food J. 18:538.
Jay, J. M. 1963. The relative efficiency of six selective media in isolating coagulase-positive Staphylococci from meats. J. Appl. Bact. 26:69-74.
Kachikian, R., Fellers, C. R. and Litsky, W. 1959. A bacterial survey of commercial frozen breaded shrimp. J. Milk Food Tech. 22:310-312.
Kampelmacher, E. H. 1963. The role of Salmonellae in food-borne diseases. Microbiological Quality of Foods. L. W. Slantz et al., Editors, Academic Press, New York. 84-100.
Kauffman, F. 1966. The Bacteriology of Enterobacteriaceae. Williams and Williams. Baltimore, Md.
Kendall, John Hugh. 1969. Low dose gamma irradiation of coagulase- positive Staphylococcus aureus in Gulf Shrimp. Louisiana State Universi ty.
Kennedy, E. R. and Barbaro, J. F. 1952. The inhibitory effect of tetrazolium on some strains of inicrococci. J. Bact. 63:297- 302.
Kern, J. 1957. Bacterial studies of frozen raw breaded shrimp. Comm. Fisheries Review 19(12):11-14,
Kiser, J. S. 1944. Effects of temperature, approximately 0°C, upon tlx; growtli and biochemical activities of bacteria isolated from mackeral. Food Research 9:257. 108
Koser, S. A. 1926. Differential tests and their relation to habitat of the Coli-Aerogenes group. J. Bact. 11:77-78.
Larkin, E, P., Litsky, W. and Fuller, J. F. 19S6. Incidence of fecal streptococci and Coliform bacteria in frozen fish products. Am. J. Pub. Health 46:464-468.
Leifson, E. 1935. New culture media based on sodium deoxycholate for the isolation of intestinal pathogens and for the enumeration of colon bacilli in milk and water. J. Path. Bact. 40:581-599.
Leininger, H. V., Shelton, L. R. and Lewis, K. H. 1971. Micro biology of frozen cream pies, frozen cooked-peeled shrimp and dry food-grade gelatin. Food Tech. 25:224-233.
Littlewood, A. B. 1970. Gas Chromatography, Principles, Techniques and Application. 2nd ed. Academic Press. New York.
Ludham, G. B. 1949. A selective medium for the isolation of Staphylococcus aureus for heavily contaminated material. Monthly Bull. Min Health Great Britain 8:15-20.
MacFarlane, J. 0., Bailey, C. A., Ackerson, Diana, Ringle, D. A. and Herndon, B. L. 1970. A unique two-phase rapid technique for Salmonella detection in foods. For presentation at Amer. Soc. for Micro., Boston, Mass. April 28, 1970.
McCarthy, J. A., Thomas, H. A. and Delaney, J. E. 1958. Evaluation of the reliability of coliform density tests. Am. J. Pub. Health 48:1628-1635.
McCleskey, C. S. and Christopher, W. N. 1941. Some factors in fluencing the survival of pathogenic bacteria in cold-packed strawberries. Food Res. 6:327-333.
McCrady, M. 11. 1937. Practical study of procedures for the detection of the presence of coliform organisms in water. Am. J. Pub. Health 27:1243-1258.
McCrady, M. H. 1915. The numerical interpretation of fermentation tube results. J, Infect. Dis. 17:18.
McCullough, N. B. and Byrne, A. F. 1952. Relative efficiency of different culture mediums in isolation of certain members of the Salmonella group. J. Inf. Dis. 90:71-75.
Mailman, W. L. and Darby, C. W. 1941. Uses of a lauryl sulfate tryptose broth for the detection of coliform organisms. Am. .1. Pub. Health 31:127-134.
Matsen, J. M. and Sherris, J. C. 1969. Comparative study of seven paper-reagent strips and conventional biochemical test in identifying gram-negative organisms. Appl. Micro. 18:542-547. 109
Meyers, Richard. Personal Communication. 1971.
Millipore Corp. 1969. Microbiological Analysis of Water. Appli cation report AR-81. Bedford, Mass.
Mitruka, B. M. and Alexander, M. 1967. Ultrasensitive detection of certain microbial metabolites by gas chromatography. Anal. Biochem. 20:548-550.
Mitruka, B. M., Carmichael, L. E. and Alexander, M. 1968. Detection of activities of pathogenic microorganisms by gas chromatography. Bacteriol. Proceedings M37.
Mitruka, B. M., Carmichael, L. E. and Alexander, M. 1969. Gas chromatographic detection of in vitro and jin vivo activities of certain canine viruses. J. In^. Dis. 119:6^5-634.
Morehouse, L. G. and Wediman, E. F. 1961. Salmonella and other disease-producing organisms in animal by-products:A survey. J. Am. Vet. Med. Assoc. 139:989.
Mundt, J. 0. and Rai, B. N. 1963. Rapid detection of fecal coliform bacteria in food processing. J. Milk Food Tech. 26(2):46-49.
O ’Brien, R. T. 1967. Differentiation of bacteria by gas chromato graphic analysis of products of glucose catabolism. Food Tech. 21:78-80.
Payne, W. J., Williams, .Joy P. and Mayberry, W, R. 1965, Primary alcohol sulfatase in a Pseudomonas species. Appl. Micro. 13: 698-701.
Pedraja, Raphael. 1972. Presentation before technical committee. National Shrimp Breaders and Processors, Inc. Baton, Rouge, La.
Perry, C. A. 1939. A summary of studies on pollution in shellfish. Escherichia coli versus the coliform group as an index of fecal pollution and the value of a modified Eijkman test. Food Res. 4:381-395.
Perry, C. A. and Hajna, A. A. 1944, Further evaluation of EC medium for the isolation of coliform bacteria and Escherichia coli. Am. J. Pub. Health 34:735-738.
Prescott, S. C., Winslow, C. E. and McCrady, M. 1946. Water Bacteriology. 6th ed. London:Chapman and Hull, Ltd.
Proctor, It. E. and Phillips, A. W., Jr. 1948. Frozen precooked foods. Am. J. Pub. Health 38:44-49.
Prost, E. and Riemann, H. 1967. Food-borne Salmonellosis. Ann. Rev. Micro. 21:495-528. 110
Rammell, C. G. and Howick, J. M. 1967. Enumeration of coagulase- positive Staphylococci in cheese. J. Appl. Bact. 30:382-388.
Raj, H. 1966. A new procedure for the detection and enumeration of coagulase-positive Staphylococci for frozen seafoods. Can. J. Micro. 12:191-198.
Raj, H. D. 1969. Solutions of some problems in food bacteriology. Lab. Pract. 18(2):157-160.
Raj, H. and Liston, J. 1961. Survival of bacteria of public health significance in frozen seafoods. Food Tech. 15:429-434.
Reay, G. A. 1935. Some observations on methods of estimating the degree of preservation of white fish. Chem. and Ind. 54:145.
Reiner, E. 1965. Identification of bacterial strains by pyrolysis- gas-liquid chromatography. Nature 206:1272-1274.
Shewan, J. M. 1944. The bacterial flora of some species of marine fish and its relation to spoilage. Proc. Soc. Agr. Bact. 15:56.
Silliker, J. H., Fagan, P. T., Chiu, J. Y. and Williams, A. 1965. Polyvalent H agglutination as a rapid means of screening non- plactose-fermenting colonies for Salmonella organisms. Amer. J. Clin. Path. 43:548-554.
Silverman, G. J., Davis, N. S., Nickerson, J. T. R., Duncan, D. W., Tizcan, I. and Johnson, M. 1961. Microbial analysis of frozen raw and cooked shrimp. II. Certain characteristics of Staphylococcus isolates. Food Tech. 15:458-464.
Simmonds, Peter G. 1970. Whole microorganisms studied by pyrolysis- gas-chromatography-inass spectrometry :Si gni ficance for extra terrestrial life detection experiments. Appl. Micro. 20:567-572.
Smith, B. A. and Baird-Parker, A. C. 1964. The use of sulpha- mezathine for inhibiting Proteus sup. on Baird-Parkers isolation medium for Staphylocoecus aureus. J. Bact. 27:78-82.
Smith, P. B., Tomfohrde, K. M., Rhoden, D. L. and Balows, A. 1971. Evaluation of the modified R/B system for identification of Enterobacteriaceae. Appl. Micro. 22:928-929.
Spicer, C. C. 1956. A quick method of identifying Salmonella 11 antigens. J. Clin. Path. 9:378-379.
Splittstocsser, D. F. and Wettergreen, W. P. 1964. The significance of coliforms in frozen vegetables. Food Tech. 18(3):134-136.
Standard Methods for the Examination of Water and Wastewater. 12th ed. Amer. Public Health Ass. Inc. New York. Ill
Standard Methods of Water Analysis. 8th ed. 1936. Amer. Public Health Assoc. New York.
Steele, J. H. and Galton, M. M. 1967. Epidemiology of food-borne Salmonellosis. Health Lab. Sci. 4:207-212.
Surkiewicz, B. F. 1966. Microbiological methods for examination of frozen and/or prepared foods. J. A.O.A.G. 49:276-281.
Surkiewicz, B. F., Hyndman, J. B. and Yancy, M. V. 1967. Bacterio logical survey of the frozen prepared food industry. II. Frozen breaded raw shrimp. Appl. Micro. 15:1-9.
Sutton, R. R. and McFarlane, V. H. 1947. Microbiology of spray- dried whole egg. III. Escherichia coli. Food Res. 12:474-483.
Swenarton, J. C. 1927. Can E. coli be used as an index of the proper pasteurization ofmills? J. Bact. 13:419-429.
Thatcher, F. S. and Clark, D. S. 1969. Microorganisms in Foods. University of Toronto Press.
Toan, T. T., Bassette, R. and Claydon, T. J. 1965. Methyl sulfide production by Aerobacter aerogenes in milk. J. Dairy Sci. 48: 1174.
Velankar, N. K. and Govindan, T. K. 1957. The free amino acid nitrogen content of the skeletal muscle of some marine fishes and invertebrates. Current Sci. (India) 26:285.
Velanker, N. K. and Govindan, T. K. 1958. A preliminary study of the distribution of non-protein nitrogen in some marine fishes and invertebrates. Proe. Indian Acad. Sci. II 47(4) :202.
Wagman, J. 1960. Evidence of cytoplasmic membrane injury in the drying of bacteria. J. Bact. 80:558-564.
Walford, E. R. and Berry, J. A. 1948. Bacteriology of slime in a citrus processing plant with special reference to coliforms. Food Res. 13:340-346.
Washington, John A. II, Yu, Pauline, K. W. and Martin, William Jeffery. 1971. Evaluation of accuracy of multitest micro method system for identification of Enterobacteriaceae. Appl. Micro. 22:267-269.
Williams, Joy P., Mayberry, W. R. and Payne, W. J. 1966. Metabolism of linear alcohols with various chain lengths by a Pseudomonas species. Appl. Micro. 14:156-160.
Williams, 0. 11., Rees, II. B., Jr. and Campbell, L. L. , Jr. 1952, The bacteriology of gulf coast shrimp. I. Experimental procedures and quantitative results. Tex. J. Sci. 4(l):49-52. 112
Williams, 0. B., Campbell, L. L., Jr. and Rees, H. B., Jr. 1952. The bacteriology of gulf coast shrimp. II. Qualitative observations on the external flora. Tex. J. Sci. 4(l):53-54.
Williams, 0. B. and Rees, H. B., Jr. 1952. The bacteriology of gulf coast shrimp. III. The intestinal flora. Tex. J. Sci. 4(l): 55-58.
Wilson, W. J. 1923. Reduction of sulphites by certain bacteria in media containing a fermentable carbohydrate and metallic salts. J. Hyg. 21:392-398.
Winter, A. R., Stewart, G. F. and Wilkin, M. 1948. Pasteurization of liquid-egg products. IV. Destruction of coliforms. Food Res. 13:11-18.
Wolochow, H. 1959. Detection of airborne microorganisms through their unique compounds. Armed Services Technical Information Agency. Rept. # 211170, p. 2-18.
Woodbum, M. J. and Strong, D. H. 1960. Survival of Salmonella typhimurium, Staphylococcus aureus, and Streptococcus faecalis frozen in simplified food substrates. Appl. Micro. 8:109-113.
Woolverton, Jack. 1973. Personal communication.
Wright, E. B. and Mundy, R. A. 1960. Defined medium for phenol coefficient tests with Salmonella typhosa and Staphylococcus aureus. J. Bact. 80:279.
Zebovitz, E., Evans, J. C. and Niven, C. F. 1955. Tellurite- glycine agar:A selective plating medium for the quantitative detection of coagulase-positive Staphylococci. J. Bact. 70: 686-690. VITA
The author was b o m in Alexandria, Louisiana on March 5, 1937.
Upon graduation from Bolton High School in May, 1955, he entered
Louisiana College in September 1955, and after an interruption for
military service, he received the Bachelor of Science Degree in
Biology in May, 1963. The author was accepted into graduate school
at Louisiana Polytechnic University in June, 1963, and was awarded
a Master of Science Degree in Microbiology in January, 1965.
The author is married to the former Jeannette Fonner of
Pineville, Louisiana. They have two children, William Daniel, who was b o m in December, 1963, and John Stephen, who was b o m in
June, 1970.
The author entered the graduate school of Louisiana State
University in August, 1968. He is currently a candidate for the
degree of Doctor of Philosophy in the Department of Food Science with a minor in Microbiology.
113 EXAMINATION AND THESIS REPORT
Candidate: William Earnest McCullough
Major Field: Food Science
Title of Thesis: Identification of Pathogenic Microorganisms in Seafood by Gas Chromatography Approved:
Major Professor and Chairman
Dean of the Graduate School
EXAMINING COMMITTEE:
/Q. 2 YlmjaJL-
t4i
Date of Examination:
February 15, 1974