<%BANNER%>

Enumeration of Total Airborne Bacteria, Yeast and Mold Contaminants and Identification of Escherichia coli O157:H7, List...


PAGE 1

ENUMERATION OF TOTAL AIRBORNE BACTERIA, YEAST AND MOLD CONTAMINANTS AND IDENTIFICATION OF Escherichia coli O157:H7, Listeria Spp., Salmonella Spp., AND Staphylococcus Spp. IN A BEEF AND PORK SLAUGHTER FACILITY By GABRIEL HUMBERTO COSENZA SUTTON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

ii ACKNOWLEDGMENTS The author is sincerely grateful to Dr S. K. Williams, associate professor and supervisory committee chairperson, for he r superb guidance and supervision in conducting this study and manuscript preparation. He also extends his gratitude to the other committee members, Dr. Dwain Johnson, Dr. Ronald H. Schmidt, Dr. David P. Chynoweth and Dr. Murat O. Balaban, for their collaboration and useful recommendations during this study. Special th anks are extended to the Institute of Food and Agricultural Sciences of the Universi ty of Florida for sponsoring the author throughout the program. Great appreciation a nd love are expressed to Adrienne and Humberto Cosenza, the author’s parents, and to his girlfriend Candi ce L. Lloyd for their support throughout the graduate program. Th e author wishes to thank Larry Eubanks, Byron Davis, Tommy Estevez, Doris Sart ain, Frank Robbins, Noufoh Djeri and fellow graduate students for their friendly support a nd assistance. Overall the author would like to thank God for His inspiration, love and caring.

PAGE 3

iii TABLE OF CONTENTS page ACKNOWLEDGMENTS..................................................................................................ii TABLE OF CONTENTS...................................................................................................iii LIST OF TABLES.............................................................................................................vi ABSTRACT.....................................................................................................................vi ii CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................5 Bioaerosols...................................................................................................................5 Bioaerosol Sampling Methods......................................................................................7 Sedimentation........................................................................................................8 Impaction...............................................................................................................9 Impingement........................................................................................................10 Filtration..............................................................................................................11 Centrifugation......................................................................................................12 Electrostatic Precipitation....................................................................................12 Thermal Precipitation..........................................................................................13 Beef and Pork Slaughter Process................................................................................13 Stunning or Immobilization.................................................................................14 Bleeding...............................................................................................................14 Scalding/Dehairing..............................................................................................14 Evisceration.........................................................................................................15 Splitting and Chilling..........................................................................................15 Bioaerosols in the Meat Industry................................................................................16 Bioaerosols in the Dairy Industry...............................................................................20 Microbiological Analysis of Bioaerosols...................................................................23 Environmental and Beef and Pork Associated Bacteria.............................................27 Gram Negative Bacteria.............................................................................................29 Enterobacteriaceae .....................................................................................................29 Escherichia coli ...................................................................................................30 Salmonella spp.....................................................................................................34 Shigella spp.........................................................................................................37

PAGE 4

iv Klebsiella spp......................................................................................................38 Serratia spp.........................................................................................................38 Enterobacter spp.................................................................................................39 Citrobacter spp....................................................................................................39 Pantoea spp.........................................................................................................40 Morganella spp....................................................................................................40 Kluyvera spp........................................................................................................41 Cedecea spp.........................................................................................................41 Leclercia spp.......................................................................................................41 Non Enterobacteriaceae .............................................................................................42 Acinetobacter spp................................................................................................42 Moraxella spp......................................................................................................42 Chryseomonas spp...............................................................................................42 Flavimonas spp....................................................................................................43 Pseudomonas spp................................................................................................43 Stenotrophomonas spp.........................................................................................43 Chryseobacterium indologenes ...........................................................................44 Gram Positive Bacteria...............................................................................................44 Listeria spp..........................................................................................................44 Staphylococcus spp..............................................................................................48 Bacillus spp.........................................................................................................51 Brevibacterium spp..............................................................................................52 Cellulomonas spp................................................................................................52 Brochothrix spp...................................................................................................52 Microbacterium spp.............................................................................................53 Micrococcus spp..................................................................................................53 3 MATERIALS AND METHODS...............................................................................55 Air Sampling System..................................................................................................55 Air Sample and Carcass Swab Collection..................................................................57 Preliminary Studies.............................................................................................57 Carcass Sample Collection..................................................................................58 Air and Carcass Microbiol ogical Sample Analysis....................................................59 Bacterial Identific ation of Air and Carcass Samples..................................................62 Microbiological Identification....................................................................................63 Statistical Analyses.....................................................................................................68 4 RESULTS AND DISCUSSION.................................................................................69 Preliminary Work: Air Samp ling Time Determination..............................................69 Microbiology of Bioaerosols Collected During Pork Slaughters...............................70 Total Airborne Bacteria.......................................................................................70 Total Airborne Yeast and Mold...........................................................................72 Isolation of Staphylococcus species....................................................................75 Isolation of Listeria species.................................................................................77 Isolation of Escherichia coli ................................................................................80

PAGE 5

v Isolation of Salmonella species...........................................................................83 Identification of Airborne Bacteria Collected from Pork Slaughters..................84 Carcass Swabs Collected During Pork Slaughters..............................................87 Microbiology of Bioaerosols Co llected During Beef Slaughters...............................89 Total Airborne Bacteria.......................................................................................89 Total Airborne Yeast and Mold...........................................................................91 Isolation of Staphylococcus species....................................................................93 Isolation of Listeria species.................................................................................95 Isolation of Escherichia coli ................................................................................97 Isolation of Salmonella Species...........................................................................99 Identification of Airborne Bacteria Collected from Beef Slaughters................101 Carcass Swabs Collected During Beef Slaughters............................................103 Comparison of Bioaerosols Collected from Pork and Beef Slaughters....................105 5 SUMMARY AND CONCLUSIONS.......................................................................108 APPENDIX ORGANISMS IDENTIFI ED IN AIR AND CARCASS SAMPLES......112 LIST OF REFERENCES.................................................................................................127 BIOGRAPHICAL SKETCH...........................................................................................132

PAGE 6

vi LIST OF TABLES Table page 1 Air sampling areas for beef and pork slaughtering processes..................................55 2 Sampling time determination for air sa mple collection during a beef slaughter......69 3 Mean total airborne bacteria collected on tryptic soy agar plates in eight areas during three pork slaughters.....................................................................................72 4 Mean total airborne yeast and mold colle cted on potato dextrose agar plates in eight areas during three pork slaughters...................................................................74 5 Mean airborne bacteria collected on ma nnitol salt agar plates in eight areas during three pork slaughters.....................................................................................76 6 Airborne bacteria isolated from mannito l salt agar plates collected from three pork slaughters.........................................................................................................77 7 Mean airborne bacteria collected on m odified oxford agar pl ates in eight areas during three pork slaughters.....................................................................................78 8 Airborne bacteria isolat ed from modified oxford agar plates collected from three pork slaughters.........................................................................................................79 9 Mean airborne bacteria collected on violet red bile ag ar plates in eight areas during three pork slaughters.....................................................................................81 10 Airborne bacteria isolated from violet red bile agar plates collected from three pork slaughters.........................................................................................................82 11 Mean airborne bacteria collected on xyl ose-lysine-tergitol 4 ag ar plates in eight areas during three pork slaughters............................................................................83 12 Airborne bacteria isolated from xylos e-lysine-tergitol 4 agar plates collected from three pork slaughters........................................................................................84 13 Airborne bacteria isolated be fore and during three pork slaughters........................86 14 Bacteria isolated from pork carcasses held at 00C for 24 hours...............................88

PAGE 7

vii 15 Mean total airborne bacter ia collected on tryptic soy agar plates in eight areas during three beef slaughters.....................................................................................90 16 Mean total airborne yeast and mold co llected on potato dextrose agar plates in eight areas during three beef slaughters...................................................................92 17 Mean airborne bacteria collected on ma nnitol salt agar plates in eight areas during three beef slaughters.....................................................................................94 18 Airborne bacteria isolated from mannito l salt agar plates collected from three beef slaughters..........................................................................................................95 19 Mean airborne bacteria collected on m odified oxford agar pl ates in eight areas during three beef slaughters.....................................................................................96 20 Airborne bacteria isolat ed from modified oxford agar plates collected from three beef slaughters..........................................................................................................97 21 Mean airborne bacteria collected on violet red bile ag ar plates in eight areas during three beef slaughters.....................................................................................97 22 Airborne bacteria isolated from violet red bile agar plates collected from three beef slaughters..........................................................................................................98 23 Mean airborne bacteria collected on xyl ose-lysine-tergitol 4 ag ar plates in eight areas during three beef slaughters..........................................................................100 24 Airborne bacteria isolated from xylos e-lysine-tergitol 4 agar plates collected from beef slaughters...............................................................................................100 25 Airborne bacteria isolated befo re and during three beef slaughters.......................101 26 Bacteria isolated from beef carcasses held at 00C for 24 hours.............................104

PAGE 8

viii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ENUMERATION OF TOTAL AIRBORNE BACTERIA, YEAST AND MOLD CONTAMINANTS AND IDENTIFICATION OF Escherichia coli O157:H7, Listeria Spp., Salmonella Spp., AND Staphylococcus Spp. IN A BEEF AND PORK SLAUGHTER FACILITY By Gabriel Humberto Cosenza Sutton August 2004 Chair: Sally K. Williams Major Department: Animal Sciences Environmental air monitoring programs can be employed to reduce unsanitary conditions in animal slaughterhouses due to susp ended bacterial particle s in the air. The main objectives of this study were to enumer ate total airborne bacteria and yeast and mold contaminants and determine the presence of Escherichia coli O157:H7, Listeria spp., Salmonella spp., and Staphylococcus spp. in bioaerosols ge nerated in a slaughter facility and on pork and beef carcasses. Air samples were taken before and during three separate pork and beef slaughter processes at the bleeding area, hide removal or dehairing area, back splitting area and hol ding cooler using an Andersen N6 single stage impactor. Pork and beef carcass surface bacterial swabs we re collected from five different carcass sides which had been held in the holding cooler at 00C for 12 hours. Total airborne bacteria l (TAB) counts (log CFU/m3 of air) were generally higher during slaughtering than before slaughtering. TAB counts were greater than three logs

PAGE 9

ix during slaughtering and less than three logs before slaughtering. The holding cooler had TAB counts less than or equal to two logs. Similar recovery rates for Staphylococcus Escherichia and Salmonella species were obtained through direct air and enriched air microbiological sample analysis methods. Most of the Gram negativ e airborne bacteria isolated during slaughtering were from the Enterobacteriaceae and Pseudomonadaceae family. The predominant Gram positive airbor ne bacteria isolated during slaughtering were Staphylococcus Microbacterium Bacillus and Micrococcus species. Potentially pathogenic Staphylococcus aureus Escherichia coli and Salmonella spp. were isolated from bioaerosols generated during slaughter ing and from pork and beef carcasses. Neither Listeria spp. nor Escherichia coli O157:H7 were insolated from air samples or pork and beef carcasses. The isolation of various microorganisms, including Staphylococcus Escherichia and Salmonella spp., from air samples and carca ss swabs support the theory that bioaerosols transport bacteria and contribute to contaminatio n of pork and beef carcasses. The determination of the levels and types of airborne bacterial contaminants present in a small scale slaughter facility has various imp lications. The effectiveness of a plant’s sanitation program can be evaluated and the sources of airborne contamination can be determined allowing for increased food safety.

PAGE 10

1 CHAPTER 1 INTRODUCTION In the past it was thought that food products were contaminated when they came in contact with contaminated surfaces, but now it is known that additional product contamination occurs from contact with air borne bacteria. Unsanitary environmental conditions in food processing plants can occur due to suspended bacter ial particles in the air. These biological particles are microscopi c, with a diameter of 0.5 to 50 m and are suspended in the air as an aerosol. Airborne contaminants are also known as bioaerosols and include bacteria, fungi, viruses and pollen. These may be present in the air as solid (dust) or as liquid (condensa tion and water). An aerosol is a two-phase system of gaseous phase (air) and particulate matter ( dust, pathogens), thus making an important bacterial vehicle. Pathogenic bacteria at tach to dust particle s and condensation, and travel around the processing facility. This contaminated air comes in contact with food products, containers, equipment and other food contact surfaces during processing. According to the Food and Drug Administra tion (FDA), the food industry must reduce product contamination by reduci ng airborne microorganisms (11, 18) Airborne contaminants cause human illness due to ingestion of contaminated foods and also reduce product shelf lif e resulting in an economic loss (18) Swabbing of equipment is typically used to determine th e sanitation level of food processing plants. This method does not always provide an effec tive enumeration of airborne contaminants. Air sampling is more effective because it co llects aerosols settled on equipment and food contact surfaces. Through air sampling, food processing facilities can identify airborne

PAGE 11

2 contamination due to air contact with food products (18) Ideally, an air sampler would be able to collect all of the viable microorganisms per unit vol ume of air, but this is not possible because not all airborne cells can be physically separated from the air without killing them during sampling (22) Methods for the detec tion of viable airborne microorganisms include sedimentation, impaction on solid surfaces, impingement in liquids, filtration, centrifugati on, electrostatic precipitati on and thermal precipitation. Impaction methods are usually used because th ey obtain higher recovery rates than other air sampling methods and can be used in situat ions where bioaerosol levels might be low (11, 22) It is impossible to keep airborne bacter ia, yeast and mold in food processing areas at a zero level. Some of the major sources of contamination in food processing facilities are wastewater, rinse water and spilled pr oduct that become aerosolized. Airborne bacteria, yeast and mold are generated in pr ocessing facilities by h eating, ventilation and air conditioning systems (HVAC). These sy stems contribute airborne microorganisms under normal operation because they provide fert ile areas for growth due to moisture. Worker activity, equipment operation, sink and floor drains, and high pressure spraying are also major sources of bioaerosols (18) Worker activity, talking, sneezing and coughing create dust particles and air disturba nces creating airborne microorganisms. The workers’ contribution of airborne bact eria depends on their health, condition of clothing, hygiene and locati on in processing facility (18) Equipment operation contributes to variations in mi croorganism levels. Airborne b acteria increase with the use of conveyor systems which cause bacterial aerosols that adhere to conveyor surfaces (38) The direction of airflow is important in th e control of bioaerosol contamination and it

PAGE 12

3 should always be counter current to that of the product flow. Barriers like walls and doors are used to separate clean and unclean areas (34) Sink and floor drains can harbor microorganisms because they are humid and contain nutrients from wastewater which provide a fertile growth environment. Fl ooding of drains causes microorganisms on the surface to become aerosolized and air disp erses them, causing increased levels of aerosolized bacteria in th e food processing facility (38) High pressure spraying also causes an increased level of aerosolized bact eria after spraying. The extent of this increase depends on the condition of floor, wate r pressure and the amount of water used. High temperature and humidity in the proces sing room increase micr obial growth, but if the environment is controlled, bacterial growth can be minimized (38) Intense husbandry practices and long term residence of cattle in feedlots and pens provide great opportunity for microorganisms to affix to hoofs and hides. Research regarding airborne contamination levels in m eat processing facilities indicates airborne microbes are a potential source of microbio logical contamination in various meat products. According to Rahkio and Korkeala (34) higher concentrations of airborne bacteria exist in the back-splitting area than in the weighing section in pork slaughterhouses. The skin or hide of slaught ered animals can be a source of airborne bacteria in slaughterhouses. According to Jericho et al. (21) many processes during cattle slaughtering are associated with the creation of bioaerosols. The attachment of specific pathogenic and nonpathogenic bacteria to carc ass surfaces after hide removal is usually immediate (21) Bacterial isolates may be identified using a variety of methods. The order in which these tests are performed is referred to as an identification scheme Culturable bacteria

PAGE 13

4 are usually first classified according to th eir microscopic morphological characteristics and Gram stain reaction. Microorganism identification can be accomplished based on phenotypic or genotypic characte ristics or both. Phenotypi c identification is based on observable physical or metabolic characteris tics of bacteria. Genotypic identification implicates the characterization of a portion of the bacterium’s genome using molecular techniques for DNA or RNA analysis (14) According to Mead et al. (30) Salmonella (31%), Listeria (28%), Toxoplasma (21%), Norwalk-like viruses (7%), Campylobacter (5%), and E. coli O157:H7 (3%) account for more than 90% of estim ated food-related fatalities. E. coli O157:H7, Salmonella spp., Listeria monocytogenes and Staphylococcus aureus are major foodborne bacteria pathogens that have an anim al reservoir and have been implicated in the contamination of various meat products. The main objective of this study was to e numerate total airborne bacteria and yeast and mold contaminants and determine the presence of Escherichia coli O157:H7, Listeria spp., Salmonella spp., and Staphylococcus spp. on pork and beef carcasses and in bioaerosols generated in a slaugh tering facility at the University of Florida, Gainesville, FL.

PAGE 14

5 CHAPTER 2 LITERATURE REVIEW Bioaerosols The study of airborne microorganisms and their effect on human health and the environment is known as aerobiology. In recent years research in this field has increased because of growing awareness of the variet y of health problems potentially caused by airborne microorganisms (5) An aerosol is a suspension of microscopic solid and/or liquid particles in air or gas. Biological aerosols (bioaero sols) are single microorganisms or clumps of microorganisms attached to so lid or liquid particles suspended in the air (11) Organisms present in bioaerosols can be b acteria, yeasts, molds, spores of bacteria and molds, microbial fragments, toxins, metabolites, viruses, parasites and pollen. Bioaerosols generally range in size from 0.5 m to 50m in diameter (5, 22) Microorganisms in bioaerosols may attach to dust particles or may survive as free floating particles surrounded by a coating of dried organic or inorganic material. They cannot multiply in bioaerosols due to a lack of nutrients, but these ae rosols can travel in the air for great distances. Location and environmental conditions such as humidity, density and temperature have a great effect on the type of population and amount of microorganisms in the air. Some of the ma jor sources of bioaerosols are humans (by sneezing, coughing and talking), animals, vegetation and dust particles (1) Microorganisms can become aerosolized from environmental sources such as worker activity, water spraying, sink a nd floor drains, air conditioni ng systems, and different food processing systems (11)

PAGE 15

6 Aerosols display intricate aerodynamic behavior resulting from various combinations of physical factors such as Br ownian motion, electric gradient, gr avitational field, inertia force, electromagnetic radiati on, particle density, temperature gradient and humidity. The behavior of bi oaerosols is dominated by physical and biological factors. Physical factors affect wher e and how many bioaerosol part icles will reach a specific surface. The most important biological factor is the ability of the bi oaerosol particle to withstand lethal or sublethal stress or damage as it is dispersed in the air. Some of these stresses are created during aerosol genera tion, dispersion and landing or collection. These stresses are usually sublethal, but when joined with other environmental stressors like temperature, dehydration, ir radiation, oxidation and pollution the effect is often lethal (22) Bioaerosol particle size is one of the main factors affecting aerodynamic behavior. Vegetative bacterial cells usually will not survive long in air unless they have a protective medium surrounding them or unless relative humidity and temperature are favorable. Vegetative bacteria are usually present in the air in lower numbers than bacterial and mold spores. Bioaerosols generated from th e environment are usually bacterial spores, yeasts and molds, but during food processi ng the main sources of contamination are vegetative bacteria like Staphylococcus spp. and Micrococcus spp. (11) According to Al-Dagal and Fung (1) Escherichia coli exhibits rapid death at low relative humidity (<50%) and temperatures between 150C and 300C. Aerosolization is stressful enough for vegetative bacteria so it is important to reduce additional stress caused by collection procedures and growth media used during ai r sampling. When bioaerosols have been subjected to mechanical or physical damage th eir recovery on selective media is reduced. Bioaerosol research includes generation, collec tion, storage and analysis of aerosols. In

PAGE 16

7 addition to cell injury some other factors th at will influence bio aerosol collection are strain of the microorganism, growth conditions, aerosol gene ration, aerosol particle size and collection method (22) Bioaerosol Sampling Methods In an ideal situation a bioaerosol sample r would be able to count the total number of viable microorganisms per unit volume of air. In reality this is not possible because 100% of airborne cells can not be physically separated from the air without killing them during sampling (22) Quantitative and qualitative gui delines that relate numbers and types of microorganisms per unit volume of ai r to acceptable levels of product samitation must be established. These guidelines should be established for each individual processing plant to ascertain possible sources of product contamination (air flow patterns, ventilation systems, personnel activity and density), therefore having the capacity to reduce airborne contamination. Various met hods for the detection of viable airborne microorganisms exist. The quantitative dete rmination of airborne microorganisms is possible by sedimentation, impaction on solid surfaces, impingement in liquids, filtration, centrifugation, electrostatic precip itation and thermal precipitation (11, 22) It is important to recognize that no singl e sampler can be used for collecting and analyzing all bioaerosols. F actors that must be considered during the selection of an appropriate sampling method in clude sampling environment, analysis methods used and monitoring objectives. An air sampler’s pe rformance is determined by physical and biological components. The inlet efficiency and particle collection efficiency are the physical parameters affecting an air sampler’s performance. Inlet e fficiency of the air sampler is its ability to extrac t particles from the environmen t without bias for shape, size or density, and collection effici ency is its ability to remove particles from the air and

PAGE 17

8 transfer them onto collection media. The biol ogical component of an air sampler is its ability to collect all microorganisms without a ffecting culturability wh ich is required for their detection and quantification. Sa mpling stress may injure the collected microorganism causing it to be in a viable bu t nonculturable state. The collection time may also affect the ability to obtain a repres entative bioaerosol sample. Sampling times that are too long may cause increased sampli ng stress or cause an overload of particles which may prevent enumeration. Air sampling collection periods are usually moderately short and a single monitoring result can be of little value due to the variability of bioaerosols. Duplicate air samples are hi ghly recommended and their measurement is expressed as an average of th e replicate data observations (5) Sedimentation Sedimentation sampling is a static method th at relies on the force of gravity and air currents to cause the settling of airborne microorganisms onto agar plates filled with general and selective media. Standard 90 mm diameter plates are placed throughout the processing facility for about 15 minutes. Afte r exposure the plates are incubated for an appropriate time and temperat ure. Sedimentation results are expressed as CFU (colony forming units) or particles per minute of exposur e. This technique is inexpensive, easy, and collects bioaerosols in thei r original state. The main disadvantage of this method is its inability to measure the number of viab le particles per volume of air. Other disadvantages of this method are long sampli ng times, great reliance on air currents, bias towards large particles and low correlation with counts obtained using other methods. This method is useful when fallout onto a specific surface is of particular interest (11, 22)

PAGE 18

9 Impaction The majority of air samplers used in th e food industry use im paction as the method for collecting bioaerosols. Impaction methods us e the inertia of partic les to separate them from the air currents (5) Impactors collect airborne mi croorganisms onto an agar surface or an adhesive coated surface with the use of a vacuum. An impactor consists of an air jet that is directed over the impaction surface causing the part icle to collide and stick to the surface. There are two types of impactor s: slit (i.e. STA, New Brunswick Sci. Co. Inc., Casella, BGI Inc.) or sieve (i.e. A ndersen sampler, Thermo Electron Corp.) samplers. A slit sampler is cylindrical in sh ape and has a tapered slit tube that creates a jet stream when an air samples is pulled by a vacuum. The air sample is collected onto an agar plate which is rotating on a turn table to create an even distri bution of particles. A slit sampler requires a vacuum to draw a c onstant flow rate of usually 28.3 liters per minute (11, 22) Sieve samplers function by drawing air (i .e. 28.3 l/min) through a metal plate with many small holes. Air particles impact on the agar surface which is a few millimeters below the metal sieve. Sieve samplers like the Andersen sampler may consist of a single stage or two, six or eight stages. The stages of a multiple stage sampler have decreasingly smaller holes causing increased par ticle velocity as the air travels through the sampler. Large particles are impacted on the first stages and smaller particles are carried until they are accelerat ed enough to impact the late r stages. Multiple stage impactors are not only used for the enumeration of viable particles pe r unit volume of air, but also yield a size profile of particles in th e bioaerosol. A two stage impactor is used when the differentiation between respirable particles (<5 m ) and nonrespirable particles (>5 m) is of interest. Multiple stage impactors are used more in health care settings than

PAGE 19

10 in food processing environments. Single st age impactors do not differentiate between particle sizes and are used wh en the total number of viable particles per unit volume of air is needed. Impaction methods obtain hi gher recovery rates th an other air sampling methods and are used when bioaerosol levels ar e expected to be low. This method results in a low sampling stress and after collection no further manipulation is needed because particles are on agar plates. Impactors possess relatively hi gh sampling efficiencies, are rugged and simple to operate. Some disadvantag es of this method are that these samplers are usually difficult and bulky to handle, expens ive, and cumbersome. Also the inside of the sampler and the outside of th e agar plates must remain st erile until sampling begins or else samples may become contaminated (11, 22) Impingement Impingement methods use a liquid medi um for the collection of airborne microorganisms. Air particles are entrapped in the liquid as air is dispersed through the liquid. Liquid impingers are either low ve locity or high velocity. Low velocity impingers do not efficiently collect small par ticles (<5 m) since they can be trapped in bubbles and carried out with the released air. High velocity impingers collect all particles sized greater than 1m but tend to destroy vegetative cells due to the high air velocities generated during sampling. Liquid impingers must use appropriate collection medium for the recovery of different microorganisms. The medium must pres erve the viability of the microorganisms sampled while inhibiting it s growth. Some of the common collection media include phosphate buffer, buffered gela tin, peptone water and nutrient broth. Airborne microorganism quantification is accomplished by serial diluting and plating an aliquot of the collection liqui d onto solid growth medium The total volume of air sampled and the volume of collection fluid must be measured to determine the number of

PAGE 20

11 cells collected. Liquid impingers are used in situations where bioaer osol concentrations are expected to be high. Th e All Glass Impinger (AGI-30, Ace Glass, Inc.) is a high velocity impinger that is commonly used for air sample collection in respiratory disease studies. It operates by drawing air through an inlet tube mimicking the nasal passage, at a rate of 12.5 liters per minute. Impingers are relatively inexpensive and simple to operate but viability loss may occur due to shear force exerted during collection. While particles impinge into the collection media, the air st ream approaches sonic speed which can cause destruction of vegetative cells. Overestimation of bacterial c ounts is also a problem with this sampling method since high air sampling ve locity can disperse dust particles, thus breaking up clumps of bacteria. Another limitation of the impingement method is its failure to collect particles smaller than 1 m (11, 22) Filtration The filtration method collects airborne microorganisms onto a filter which is mounted on a holder and connected to a vacuum source with a flow rate controller. The filter can consist of sodium alginate, cellulose fiber, glass fiber, gelatin membrane (pore size 3 m) or a synthetic membrane (pore si ze 0.45 m or 0.22 m). Gelatin membrane filters are water-soluble and can be placed dire ctly onto an agar surface for quantification or can be serially diluted in a liquid first. Synthetic membrane filters are agitated in a liquid to disperse the particles before bacteriological analysis is performed. Filter collection devices are used for the enumerati on of molds or bacteria l spores but are less effective for vegetative cells because of the stress associated with desiccation, although shorter sampling times could reduce this stre ss. Filtration devices are low in cost and simple to operate and their filters possess a large number of pores so that large volumes of air can be sampled during a short period of time (15)

PAGE 21

12 Centrifugation Centrifugation sampling methods create a ce ntrifugal force that propels airborne microorganisms onto an agar surf ace. An aerosol is spun in a circular path at a high velocity and the centrifugal force causes the pa rticles to impact against an agar surface. Microorganisms experience less stress compar ed to impaction or impingement sampling methods since no high velocity jet forces ar e created during centrifugal sampling. These devices can give more representative sample s since they can rapidly sample high volumes of air. Air sample results are expressed as CFU per liter of air sampled. Centrifugal samplers are simple to operate and are less expensive than most impactors. One of the limitations of this method is its inability to create enough centrifugal force to impact small particles onto an agar surface. The Re uter centrifugal air sampler (RCS Sampler, Biotests Diagnostics Co.) is portable, battery operated and easy to use. This sampler collects 100% of 15 m particles, 55% to 75% of 4 m to 6 m particles and does not effectively collect particles sizes smaller than 1 m (11) Electrostatic Precipitation Electrostatic precip itation methods capture microo rganisms by giving them an electrostatic charge and colle cting them on an oppositely ch arged rotating disk. This surface can be agar or glass. This method possesses a high sampling rate, high collection efficiency, and low resistance to air flow. A disadvantage of this method is that nitrogen oxide and ozone which may be toxic to mi croorganisms are produced during air sample ionization. Little is known a bout the effect of electrostatic charges on the viability and clumping of microorganisms. Electrostatic precipitation is rarely used for aerosol detection since its equipment is co mplex and requires careful handling (11)

PAGE 22

13 Thermal Precipitation Thermal precipitation methods are based on thermophoresis principles in which particles move away from hot surfaces toward cooler surfaces. The degree of particle movement will depend on the temperature gradie nt. This method is used to determine particle size distribution, especially when collecting particles sm aller than 1 m. Microorganisms are collected onto glass cover-slips and are sized and counted microscopically. This method is not commonl y used in industry si nce it requires very precise adjustments and its air sampling rate is considerably low (300 to 400 ml/ minute) (22) Beef and Pork Slaughter Process Slaughtering procedures di ffer among species, but lives tock slaughtering always includes bleeding, hair or hide removal and removal of the abdominal viscera (stomach(s), intestines, liver and reproduc tive glands). The head, pluck (trachea, esophagus, lungs and heart) and the feet (excep t in swine) are also removed. The kidneys and surrounding fat are removed from swine, but in the United States they remain on beef carcasses because of tradition. The head, edible offal (heart, liver, tongue and sweetbreads) and inedible offal (rest of vis cera and pluck) removed from the animal are kept with the matching carca ss for inspection purposes. St andard operating procedures for beef slaughtering consist of, immobilization, bleeding, head removal, hide removal, feet removal, evisceration, splitting, in spection, washing and chilling. The pork slaughtering procedure consists of stunni ng, bleeding, scalding/dehairing and toenail removal, head removal, evisceration, splitting, inspection, wa shing and chilling (42)

PAGE 23

14 Stunning or Immobilization All animals slaughtered under inspecti on must be stunned before bleeding according to the Humane Methods of Sla ughter Act of 1978. Cattle are immobilized mechanically via concussion (stunning hammer) or penetrating type devices (captive bolt stunner). A sharp blow is delivered midw ay between the eyes and halfway up the forehead since this is an area where the brain is closest to the skull. Pigs are electrically stunned using a hand-held device that dispenses electrical current to the skull producing a loss of consciousness (42) Bleeding When bleeding cattle, a knife is inserted at a 450 angle directly be low the brisket. A cut 10 to 15 inches long extending from th e brisket to the throat causes blood removal by severing the carotid arteries and jugular veins. In bleeding pork, a six inch knife is inserted in the middle of the ventral aspect extending midway from the sternum to the throat. The knife is inserted with the point upward and then with the point moved downward until it reaches the backbone. During bleeding the carotid arteries and jugular veins are severed. Approximately six to nin ce minutes are required to bleed out beef and pork completely during the slaughtering process (42) Scalding/Dehairing Hogs are immersed in 600C water until their hair begins to loosen from the flank area. Hot water causes collagen surrounding the hair follicles to change into gelatin which facilitates removal of the hair eith er manually by scraping or with a dehairing machine. Scalding for an excessive amount of time or temperature may cause the skin to cook resulting in large areas of skin and fat being removed from the carcass, therefore

PAGE 24

15 marring the carcass. Residual hair remaining after dehairing is removed by shaving the carcass with knives and singing with a gas-fired hand-held torch (42) Evisceration Removing the viscera from the abdominal cav ity and pluck from the thoracic cavity is performed during the evisceration process. First the anus is cut loose from the attachment under the tail and the head is rem oved at the occipito-atlantal space. It is critical that both ends of the gastrointestinal tract are secu red with string to prevent the contents leakage onto or into the carcass. Ne xt a cut is made from the crotch to the sternum without cutting the viscera or pluck. The sternum must be severed with a knife in pork and a saw in beef to be able to remove the pluck. During pork slaughtering the worker removes the viscera by cutting around the intestines, stomach, liver and spleen beginning at the detached anus. During beef slaughtering the worker pulls and cuts around the stomachs (rumen, reticulum, omasum absomasum), intestines, and spleen beginning at the loosened anus. The pluc k (heart, esophagus, lungs and trachea) is removed by cutting through the diaphragm membrane and down the backbone. During evisceration, extreme caution must be employed not to cut or tear the gastrointestinal tract to prevent ingesta or fecal material from contaminating the carcass (42) Splitting and Chilling Pork carcasses are split down the vertebra l column and the kidneys and leaf fat are removed in preparation for chilling. The carca ss is washed, blood clots and loose glands are trimmed off and the carcass is weighed a nd identified. Beef carcasses are split down the center of the backbone and the kidneys and leaf fat and are left on the carcass. Beef carcasses are carefully trimmed removing pieces of hide, brui ses, hair, feces or ingesta and are then washed. Carcasses are wei ghed, tagged and placed in a -2 to 00C rapid air

PAGE 25

16 movement cooler to facilitate swift carcass temperature re duction. Carcasses are usually chilled between 18 and 24 hours before being moved to a holding cooler at 00C to 10C for subsequent storage until fabrication (42) Bioaerosols in the Meat Industry Investigations regarding air borne contamination levels in meat processing facilities indicate airborne microbes are a potential s ource of microbiological contamination in various meat products. Limited information ex ists in illustrating th e association between beef and pork carcass contamin ation and airborne bacter ia during the slaughtering process. Rahkio and Korkeala (34) took air samples from the back splitting and weighing areas of beef and pork slaughtering lines at a height of 1 to 1.5 m fr om the floor and 1 m from the carcass. An Andersen two stage sa mpler, adjusted to a flow rate of 0.0283 m-3/min, was used to collect the air samples. Significantly higher (P < 0.05) mean log CFU/m3 of air were shown to exist in the back -splitting area (3.13 to 4.07) than in the weighing section (2.45 to 3.58) in pork slaughterhouses. Rahkio and Korkeala (34) concluded that higher levels of airborne bacteria present in the back-splitting area when compared to the weighing area might be caused by bacteria being carried by airflow from contaminated parts of the lines. Higher ba cterial counts in the b ack-splitting area may also be due to movement of the saw blade a nd water flowing from the cutting saw. The movement of personnel between clean and uncle an areas in the slaughtering line appears to be linked to higher car cass contamination levels (34) Intense husbandry practices and long term dwelling of cattle in feedlots and pens provides great opportunity for microorganisms to affix to hoofs and hides. Mesophilic bacteria of fecal and soil origin at tached to hoofs and hides may exceed 109 CFU/cm2 and represent the main food safety hazard in cattle slaughterhouses (21) Wilson et al. (53)

PAGE 26

17 reported that mostly Gram positive bacteria ( Bacillus spp., Corynebacterium spp., Micrococcus spp., Paenibacillus spp. and Yersinia spp.) were isolated from air samples taken at a cattle feedlot. Gram negative bact eria might have been present in the feedlot environment but were reduced in number by environmental conditions or may have been in a viable but nonculturable state. Butera et al. (6) reported that Gram positive cocci ( Staphylococcus Micrococcus Leuconostoc Streptococcus and Aerococcus ) made up 72% of the total bacteria isolated from a hog growing facility. Gram positive rods ( Bacillus ) made up 7.2% and Gram negative rods ( Enterobacteriaceae ) made up 20.8% of the total bacteria isolated in that study. According to Jericho et al. (21) the majority of processes in cattle slaughterhouses are associated with the creation of bioaerosols. The skin or hide of slaughtered animals can be a source of airborne bacteria in sl aughterhouses. The attachment of specific pathogenic and nonpathogenic bacteria to carc ass surfaces after hide removal is usually immediate. An undetermined amount of th is contamination comes from bioaerosols which can be easily recovered from carcass surfaces. Jericho et al. (21) used a slit air sampler (STA-203) to collect air samples from the hide removal floor, carcass dressing floor and cooler. The highest total viable counts were observe d in the hide removal floor. The presence of bacterial counts in air samp les collected before the slaughtering process commenced was attributed to in-process cl eanup operations using high water pressure hoses (21) High speed equipment and air circulation are widely used in slaughtering and processing practices. Therefore, it is possi ble for bacteria on the surface of animals, employees and equipment to become airbor ne during these processes. Kotula and

PAGE 27

18 Emswiler-Rose (25) used a Ross-Microban air sampler to collect air samples during evisceration, offal recover y, carcass cooling, carcass break ing and further processing. Average CFU per 0.028 m3 of air had a tendency to d ecrease during the process of converting pork carcasses to edible product. The variability in mean CFU per 0.028 m3 of air between the different locations alt hough significant (P < 0.05) was in most cases less than one log and therefore was of little importance. In that study, coliform growth on Petri plates was only observed in 15% of th e air samples. According to Kotula and Emswiler-Rose (25) the low incidence of airborne co liforms was surprising because of speculation that bioaerosols may be an impor tant cause of indiscriminate meat product contamination. Air currents can be influenced by a plan t’s configuration therefore affecting the airborne contamination of beef carcasses. Reduction of airborne b acterial contamination can be accomplished by building walls between clean and unclean areas or by allowing sufficient distance between these areas in sl aughtering processes. In the slaughter process, the unclean areas ar e considered from stunning the animal to the hide removal and the clean areas are considered from eviscer ation to the final wash of the carcass. Worfel et al. (54) used an Andersen single stage samp ler to collect air samples from various slaughtering processes in hide-on (unclean) and hide-off (clean) areas from three slaughter facilities with different plant la youts. Two plants whic h had a wall separating the clean and unclean areas had significantl y (P < 0.05) lower bacterial counts in the clean areas than the plant with no area separation (54) Whyte et al. (52) used a six stage Andersen air sampler to collect air samples from the defeathering, evisceration, air chills, packing, deboning an d further processing areas

PAGE 28

19 in three poultry processing plant. Various me dia were used in that study to enumerate total mesophilic and psychrophilic aerobic bacteria, Escherichia coli, Enterobacteriaceae thermophilic Campylobacter spp., and Salmonella spp. Total aerobic bacterial counts were significantly (P < 0.05) higher in the defeathering areas when compared to the other areas. Mean log CFU/m3 of air counts of Escherichia coli and Enterobacteriaceae were higher in air samples acqui red from the defeathering (1.85, 1.82) and evisceration (1.15, 0.97) areas and were not detected in the subsequent areas. Salmonella spp. was not positively identified in a ny of the air samples taken from the three poultry processing plants in th at study. According to Whyte et al. (52) that study identified air as a possible carrier of pathoge nic bacteria and demonstrated the need for separation of dirty and clean areas in poultry slaughtering. Ellerbroek (13) collected air samples from the re ception and evisceration areas in a poultry processing facility using an Andersen air sampler. The reception area to the processing facility consisted mostly of Gram positive bacteria ( Micrococcus spp., Corynebacterium spp., Staphylococcus spp.) and yeasts, probably coming from poultry skin and feathers. The evisceration area consisted mostly of Gram positive Staphylococcus spp., but Gram negative Acinetobacter and Moraxella spp. were also present in this area. Airborne bacterial counts in the reception area were similar to those in the evisceration area, however Enterobacteriaceae were lower in numbers in the air of the evisceration area (13) Lutgring et al. (27) took air samples from various areas in four poultry processing facility using an N6 single stag e Andersen viable sampler. Different media were inserted into the air sampler to enumerate mesophili c and psychrophilic bacteria and yeast and

PAGE 29

20 mold. In all four plants the highest concentration of mesophilic bacteria were observed in the receiving areas. Airborne bacterial con centrations were 100 to 1,000 times greater in the receiving area than outside the plant. Bi oaerosol levels progressively decreased as slaughtering went from receiving to car cass cutting and deboning. Psychrotrophic bacterial counts were about one log less than mesophilic bacterial c ounts for the majority of samples areas. Throughout that study tota l yeast and mold counts were far less than total bacteria counts. Differences in yeas t and mold counts between areas sampled were much less apparent than differences in bact erial counts. Accordi ng to Lutgring et al. (27) air flow should be controlled so that it tr avels from the finished products area to the receiving area, and when possi ble these areas should be phys ically separated. The food safety effect of bioaerosols on products th at require cooking for consumption is less important than on ready to eat food products (27) Airborne bacterial counts could offer an alternative method for determining carcasses contamination in sl aughterhouses. Even though ai r sampling employs the use of special equipment, it can be a more conve nient and rapid method for carcass sampling. Using this method there is no interruption of the slaughtering line and no need for personnel to touch the carcasse s while sampling. Nevertheless surface carcass swabs are essential for indicating pres ence of pathogenic bacteria due to slaughtering events producing fecal contamination (34) Bioaerosols in the Dairy Industry Bioaerosols may be a means for microbi al contamination of dairy products according to the U. S. Food and Drug Administ ration. Airborne contamination in dairy processing facilities can result in manufact uring of low quality pr oducts with a reduced shelf life (37) Aerosols in dairy plants may be created by HVAC (heating, ventilation,

PAGE 30

21 and air conditioning) systems, high pressure water spraying, floor drains, plant workers, supplies, openings between processing room s, spilled milk on the floor and conveyor systems. The main areas of bioaerosol cont amination in dairy pro cessing facilities are during filling of retail cont ainers and while filling of holding tanks with pasteurized products (36, 38) An effective bioaerosols sampli ng program can be used in a milk processing plant as a tool to control pathoge ns therefore increasing product shelf life. The use of packaging equipment that eliminat ed airborne contamination produced a 7 day increase in product shelf life (36) Ren and Frank (38) took air samples in two fluid milk and two ice cream plants using a centrifugal air sampler (RCS Sampler). Air samples were collected from the ventilation inlet and outlet, above drain in packaging area, processing/packaging area after water spraying, conveyor belt, pasteu rized product holding tank and product filling area. At each sampling point eight liter s of air were sampled onto media for the enumeration of total ae robic bacteria (TA), Staphylococcus spp. (ST) and total yeast and mold (YM). Air samples exiting the ventilati on system yielded similar levels of TA and ST as those entering the system. It was esta blished that the air filtration equipment used in these plants had been negl ected. A significant (P < 0.05) in crease in TA and ST in the air above the floor drains during processing was observed when comp ared to air without processing activities. Efficient floor drai n cleaning in these critical rooms is recommended to reduce the extent of bioaeros ol contamination. The use of high pressure water hoses during processing and packaging was linked to significant increases in TA and ST levels in all plants and YM levels in some plants when compared to levels during inactivity. Conveyer systems contain metal surfaces to which microbes can adhere.

PAGE 31

22 These microbes can become aerosolized when physical disturbances are created on the conveyer systems during product or case collisi on. A significant increase in TA, ST and YM levels were observed in all four plan ts when conveyer systems were operating. People are a major source of airborne cont amination through their speaking, sneezing and breathing. Worker activity at the filler equipm ent station produced a significant rise in TA and ST levels. Low levels of airbor ne contamination were observed inside the pasteurized holding tanks (38) Ren and Frank (36) took air samples at four fluid milk processing plants using an Andersen two stage sampler, a Ross-Microban sieve sampler and a Biotest RCS sampler. The areas sampled were the raw milk storag e area, processing, and filling area. Log mean viable particle counts obtained with the Andersen sampler were significantly greater than those obtained with the Biotes ts RCS and the Ross-Microban samplers. All of the samplers recovered the lowest amount of microbial aeroso ls in the raw milk storage area and the highest amount in the milk filling areas. Ren and Frank (37) monitored the pasteurized mix storage, processing and filling areas in two commercial ice cream plants using the same samplers as the previous study. The level of microbial aerosols recovered by the Andersen two stage sampler were similar to those recovered by the Biotest RCS sampler, but were significan tly (P < 0.05) greater than those recovered by the Ross-Microban sampler. Each sampler recovered more microbial aerosols in the filling areas than in the processing areas and si milar levels in the pasteurized mix storage and filling areas in both ice cream plants (37) Kang and Frank (23) used a tracer microorganism ( Serratia marcescens ) to prove that floor washing and drain fl ooding are sources of viable ae rosols in a dairy processing

PAGE 32

23 plant. The ability of biological aerosols to stay viable in the environment was also observed. The Andersen six stage sieve air sa mpler was used in that study due to its proven reliability thr ough previous research. Air samp les were collected from the non inoculated drain area during a break in processing to obtai n background bacter ial levels. Air samples were also collected 30 seconds af ter spraying water on the floor to verify the absence of Serratia marcescens Air samples were taken be fore processing, 30 seconds after spraying water on the floor, and at 10 minute intervals until reaching 40 minutes. Inoculated Serratia marcescens was recovered from air above the drain, therefore proving that microorganisms including pathogens can become aerosolized from drains by physical disruption. It took a bout 40 minutes or more for aer osol microbial levels to return to background levels. The greatest re duction in aerosol microbial levels was seen in the first 10 minutes after drain flooding (23) Microbiological Analysis of Bioaerosols Various sample analysis methods can be applied to air samples to provide information regarding concentration and compos ition of bioaerosols. First it is important to select the sample analysis methods bei ng used since not all air sampling systems are compatible. Impaction type sampling methods usually rely on microbiological methods like culturing and microscopy. Impingement and filtration methods are more adaptable with respect to sample analysis alternatives since the air samples are collected in liquid or on a filter. Limitations to tr aditional techniques have give n way to alternative methods like biochemical, immunological an d molecular assays. A vast majority of bioaerosol data generated has been attained by using cult ure analysis and most current air samplers are designed to collect particles on nutrient ag ar. A major hurdle in using culture analysis is that only those cells that survive sampling and that can be cultured will be enumerated.

PAGE 33

24 Those microorganisms that are subjected to sampling stress and environmental stress may not grow under artificial nutrien t conditions in a laboratory (5) Microorganisms have a wide range of nut ritional requirements which cannot all be met by one culturing media. During bioaerosol monitoring a general media that provides the conditions for the greatest number of microorganisms to grow should be employed. It may be necessary to perform replicate sampling and use a variety of sampling media in an attempt to enumerate the greatest number of microorganisms. For the culture of general bacteria several broa d spectrum media li ke tryptic soy agar nutrient agar, and casein soy peptone agar can be utilized. Af ter the correct incubation time, culturable microorganisms are determined by enumerati ng colony forming units (CFU). Culturable airborne microorganisms are calculated by di viding the number of total CFU per sample by the volume of air sampled. The identificatio n of fungal isolates is usually performed by microscopic determination of the morphological characteristics (5) The tests and the order in which these tests are performed to identify microorganisms are referred to as an identifi cation scheme. Bacterial isolates may be identified using a variety of methods. Cultu rable bacteria are usua lly first classified according to their microscopic morphological characteristics and Gram stain reaction. Microorganism identification can be accomplished based on phenotypic or genotypic characteristics or both. Phe notypic identification is base d on observable physical or metabolic characteristics of bacteria. The most commonly used phe notypic identification methods are macroscopic and microscopi c morphology, staining characteristics, environmental requirements for growth, resist ance or susceptibility to antimicrobial agents, and nutritional requirements and meta bolic capabilities. Other methods of

PAGE 34

25 characterization are based on the antigenic makeup of organisms and they involve antigen-antibody interactions. Some i mmunochemical identification methods are precipitin assays (immunodiffusi on), particle agglutination a ssays (latex agglutination), immunofluorescent assays, and enzyme-li nked immunosorbent assays (ELISA). Genotypic identification implicates the charac terization of a portion of the bacterium’s genome using molecular techniques for DNA or RNA analysis. The presence of a particular gene or specific nucleic acid sequence is in terpreted as a definitive identification of the microorga nism. Some molecular identifi cation methods used for the identification of microorganisms are nucleic acid hybridization me thods, amplification methods (polymerase chain reaction or PCR, reverse transcription-PCR or RT-PCR, Real Time PCR, repetitive extragenic palindromic PCR or REP-PCR), enzymatic digestion of nucleic acids (restriction frag ment length polymorphisms or RFLPs), and pulsed field gel electrophoresis (PFGE) (14) The identification of bacteria to their ge nus and species levels can be accomplished by the use of conventional biochemical methods or commercially available identification systems. API is one of the most widely us ed diagnostic tools for identification of the Enterobacteriaceae family of bacteria (11) Initially, its database was mostly oriented toward clinical isolates, but in recent year s it has grown to include food, industrial and environmental isolates also. The API testi ng system is made up of various small and elongated wells which contain different dehydr ated media, and is intended to perform various biochemical tests from a pure culture gr own on an agar plate. Advantages of the API system include an excellent data base long shelf life and a proven testing record throughout the past 25 years. API identification can be time consuming due to the

PAGE 35

26 inoculation of the numerous wells containing media. API 20E is used to identify species/subspecies of Enterobacteriaceae ( Escherichia coli and Salmonella spp.) and/or non-fastidious Gram negative rods. API Sta ph is used for overnight identification of Gram positive Staphylococci spp., Micrococci spp., and Kocuria spp. API Listeria, API Campy and API Coryn are used to identify Listeria spp., Campylobacter spp., and Corynebacterium spp. respectively (11) Short chain fatty acids analysis (vol atile fatty acids, VFA) using gas chromatography (GC) has been regularly used to identify anaerobic bacteria. The fatty acids between nine and 20 carbons long have been used to characterize genera and species of bacteria, especially nonferme ntative Gram negative organisms. The introduction of fused silica capillary columns makes it convenient to use gas chromatography of whole cell fatty acid methyl esters (FAME) to identify a wide range of organisms. These columns permit the recovery of hydroxy acids and resolution of many isomers. Around 300 fatty acids and related compounds have been found in bacteria analyzed in the MIDI Research and Development Laboratory (MIDI, Inc.). This information can be quantitatively used to differentiate among groups of bacteria; therefore giving great "naming" power to th e Sherlock MIS (Microbial Identification System). This system uses fatty acids 9-20 carbons in length and peaks are automatically named and quantified by the system. In so me Gram positive bacteria branched chain acids predominate, while in Gram negative bacteria the lipopolysaccharides are characterized by short chain hydroxy acids (41) The five steps to prepare sample extr acts for gas chromatography are harvesting bacterial cells, saponification, methylation, ex traction and sample cleanup. Samples are

PAGE 36

27 injected into an Ultra 2 colu mn which is a 25 m x 0.2 mm ph enyl methyl silicone fused silica capillary column. The GC temperatur e program heating the column goes from 1700 C to 2700C at 50C per minute. The flame ionization detector used by the Sherlock MIS allows for a large dynamic ra nge of detection and provides a good level of sensitivity. Hydrogen is the carrier gas and air is used to support the fl ame. The GC detector passes an electronic signal to the computer as organic compounds burn in the flame of the ionization detector. These elec tronic signals are integrated into peaks and the data is stored. The composition of the bacterial sample FAME is compared to a stored database using pattern recognition software (41) The Sherlock library is composed of more than 100,000 strains collected from around the world. This library is not limited by a fixed number of biochemical assays like biochemi cal and enzymatic identification methods are (41) Environmental and Beef and Pork Associated Bacteria Foodborne pathogens have been estimated to cause 6 to 81 million illnesses and as many as 90,000 deaths annually in the Unites States alone (30) In general, bacterial pathogens account for 60% of hospitalizatio ns and 72% of deaths attributable to foodborne transmission (30) Parasites and viruses account for 5% and 34% of hospitalizations along with 21% and 7% of deaths associated with foodborne transmission. Salmonella (31%), Listeria (28%), Toxoplasma (21%), Norwalk-like viruses (7%), Campylobacter (5%), and E. coli O157:H7 (3%) account for more than 90% of estimated f ood-related fatalities (30) Escherichia coli O157:H7, Salmonella spp., Listeria monocytogenes and Staphylococcus aureus are major foodborne bacterial pathogens that have an animal reservoir and have been impli cated in the contamination of various meat products. Meat is a substrate rich in nutrients and can support the growth of

PAGE 37

28 a great number of microorganisms. The readil y available water content of fresh muscle tissue is high, along with the usable glyc ogen, peptides, amino acids, metal ions and soluble phosphorous makes muscle tissue a good substrate for microbial growth. These pathogens have been associated with th e hide, skin, mucous membranes, and the intestinal tract of healthy animals and are ubiquitous in nature (9, 39) Microorganisms on the hide also include Micrococcus Pseudomonas species and yeasts and molds that are normally associated with skin, fecal material and soil. Animals entering a slaughter facility have a vast population of aerobic mi croorganisms on their hide, hooves and inside their intestinal tract, while the internal surface of carcasses is generally considered to be st erile. It is generally ag reed that the majority of microorganisms on a dressed red meat car cass are directly transferred from the contaminated hide and ruptured gastrointestin al tract. Bacterial tr ansfer to the carcass also occurs by aerosols, dust generation, worker s hands, and the contact of the hide with the exposed tissue. The majority of carca ss contamination occurs during slaughtering and dressing procedures (12, 16) The spoilage of meat at am bient temperature is a direct result of the growth of mesophiles like Clostridium perfringens and members of the Enterobacteriaceae family. Storage of meat at refrig eration temperatures restricts the growth of mesophiles and allows psychrotrophs to dominate th e spoilage microflora. The spoilage of high moisture meat surfaces that are held at high relativ e humidity is usually the result of bacteria l activity. The microflora that dominate meat spoilage under refrigerated aerobic conditions are the species Pseudomonas Alcaligenes Moraxella and Aeromonas (12, 16)

PAGE 38

29 Gram Negative Bacteria Enterobacteriaceae The most significant families of Gram negative bacteria are Enterobacteriaceae Vibrionaceae and Pasteurellaceae (19) No sole group of taxonomically classified microorganisms has accounted for greater not ice from the scientific and medical communities than the Enterobacteriaceae family. No other group has more selective and differential media created for the isolation of its various members from medical and environmental samples than the Enterobacteriaceae This group is the backbone of databases used in manual, semiautomated a nd automated microbial identification systems (20) The taxonomically defined Enterobacteriaceae family includes facultative anaerobic Gram negative straight bacilli th at are oxidase negative, glucose fermenting, usually catalase positive and nitrate reducing plus grow on MacConkey agar. The genera in this family usually have a cell diameter of 0.3 to 1.8 m and those genera that are motile employ peritrichous flagella. Most members of this family grow well at 370C and some grow better and are more active metabolically at 25 to300C. Although this family consists mainly of mesophilic genera some psychrotrophic strains of Enterobacter and Serratia can grow at 00C. Enterobacteriaceae are distributed worldwide and can be found in soil, water, fruits, vegetables, plants trees and animals from insects to humans. They inhabit a wide variety of niches whic h include human and animal gastrointestinal tracts and various environmental sites (11, 14, 19) Some species like Salmonella typhi, Shigella spp., and Yersinia pestis (overt pathogens) cause diarrheal diseases which can be life threatening like typhoid fever, bacillary dysentery and “black” plague. Ot her species not usually associated with

PAGE 39

30 diarrheal diseases (opportunistic pathogens), along with thos e that are associated with diarrhea, may cause many extraintestinal diseases like bacteremia, meningitis, urinary tract, respiratory, a nd wound infections. Escherichia coli Klebsiella Enterobacter Proteus Providencia and Serratia marcesens account for about 50% of nosocomial infections. Foodborne genera of the Enterobacteriaceae family consist of Escherichia Salmonella Klebsiella Serratia Enterobacter Citrobacter Yersinia Proteus Providencia Shigella and Erwinia (11, 14, 19) Escherichia coli Escherichia coli is a straight rod measuring 1.1 to 1.5 m by 2.0 to 6.0 m which occur singly or in pairs and has an optimum growth temperature of 370C. Capsules or microcapsules occur in many strains and some strains are motile by peritrichous flagella. Escherichia coli is part of the normal flora of the intestinal tract of humans and various animals. It can be classified as an ov ert or an opportunistic pathogen and usually constitutes about 1% of the to tal biomass of feces. Most Escherichia coli do not cause gastrointestinal illnesses, but some can cau se life threatening diarrhea and chronic sequelae or disability. E. coli is serologically classified on the basis of three major surface antigens: O (somatic), H (flagella) and K ( capsule). The serogroup of the strain is identified by the O antigen and its combination with the H antigen identifies the serotype. There are more than 170 different serogroups of E coli identified. Diarrhea causing E coli isolates are clas sified into specific groups based on virulence properties, pathogenicity mechanisms, clinical syndrom es, and specific O:H serotypes. These groups include enterotoxogenic E coli (ETEC), enteroaggregative E coli (EAEC), enteropathogenic E coli (EPEC), enterohemorrhagic E coli (EHEC), enteroinvasive E coli (EIEC), and diffuse-adhering E coli (DAEC) (11, 12, 19)

PAGE 40

31 ETEC strains adhere to the small intestin al mucosa by means of surface fimbriae (type 1 pili) and cause symptoms by producing two enterotoxins: heat labile (LT) and heat stable (ST) toxins Diarrhea is caused by these ente rotoxins acting on the intestinal mucosa cells and not by ETEC invading the mucosa. One ETEC binds to the host cell membrane the toxin is endocytosed and tran slocated through the cell and reaches its target (adenylate cyclase) on the basolateral membrane of the intestinal epithelial cells. cAMP dependent protein ki nase is activated leading to stimulation of Clsecretion and inhibition of NaCl absorption resulting in wate ry diarrhea. ETEC strains cause travelers diarrhea which can be severe and often fatal in infants. Contaminated foods and water are the most frequent vehicles of ETEC in fections. EAEC strains cause persistent diarrhea mostly in children. This type of E coli strain does not produce enterotoxins LT and ST but does adhere to Hep-2 cells in an aggregative adheren ce pattern. Symptoms include diarrhea, vomiting with persis tent diarrhea in some patients (11, 12, 32) EPEC strains cause severe diarrhea in young children and infants. The pathogen adheres to the intestinal mucosa though A/ E (attachment/effacing) lesions; causing an intimate attachment to intest inal cells with effacement of underlying microvilli and accumulation of filamentous actin (F-actin). The extensive loss of absorptive microvilli by the A/E lesion could lead to diarrhea by improper absorption. EPEC actively alters ion transport in epithelial cells causing an influx of positive ions or an efflux of negative ions across the membrane. EIEC strains are genetically and pathogenically closely related to Shigella spp. but do not produce Shiga toxin. These strains cause nonbloody diarrhea by invading, multiplying and eventually ca using death of colonic epithelial cells. High fever and watery diarrhea are symptoms of this infection. DAEC strains cause

PAGE 41

32 nonbloody diarrhea in children ages one to five. This strain produces a diffuse adherence to HEp-2 and HeLa cell lines, but do not pr oduce LT and ST enterotoxins or high levels of Shiga toxin. DAEC strains do not posses EPEC adherence factor plasmids or A/E lesions and do not inva de epithelial cells (11, 12, 32) No recently described enteric pathogen has received as much scientific and medical examination as Escherichia coli O157:H7. There are many serotypes belonging to the EHEC group, but serotype O157:H7 is the predominant food pathogen. It was first recognized as a pathogen in 1982 when it was identified as the cau sative agent of two outbreaks involving the consump tion of undercooked ground beef leading to hemorrhagic colitis. It is estimated that more than 700 i ndividuals were affected and 4 deaths occurred between 1992 and 1993 due to this pathogen. According to CDC estimates there are more than 20,000 cases and 250 deaths annually in the U.S. due to Escherichia coli O157:H7 infections. Cattle are considered to be the main reservoirs of this pathogen and many outbreaks are associated with the c onsumption of undercooked ground beef and unpasterurized milk (11, 20, 30) This pathogen can be encountered through th e fecal oral route or through human to human transmission. It is highl y acid resistant and has a low infectious dose of less than 100 cells and possibly as few as 10 cells. Escherichia coli O157:H7 colonizes the large intestine (cecum and colon) by A/E lesions. Th ese lesions are an intimate attachment of bacteria to intestinal cells causing an effacement of underlying microvilli and accumulation of filamentous actin. These le sions are mediated by binding of intimin (gene: eae) to its receptor (Tir ) on the host cell. Both genes are located in LEE (locus of Enterocyte Effacement). Intimin ( , and ) is the sole adherence factor of E. coli

PAGE 42

33 O157:H7. Using type III secretion system EHEC secrete proteins (EspA, EspB, EspD) responsible for signal transduction ev ents during AE lesion formation. Escherichia coli O157:H7 causes damage by the production of on e or two Shiga toxins (Stx1, Stx2) which are potent cytotoxins. This factor leads to death and many other symptoms in patients infected with EHEC. The Stx A-B subunit stru cture is conserved ac ross all members of the Shiga toxin (Stx) family. The B pentamer binds the toxin to a specific glycolipid receptor (Gb3) present on the surface of euka ryotic cells. After binding, the holotoxin is endocytosed through coated pits and is transpor ted to the Golgi apparatus and then to the endoplasmic reticulum. The A subunit is transl ocated to the cytoplasm where it acts on the 60S ribosomal subunit. The A1 peptide is an N-glycosidase that removes a single adenine residue from the 28S rRNA of euka ryotic ribosomes causing the inhibition of protein synthesis. The resulting disruption of protein synthesis leads to the death of renal endothelial cells, intestinal ep ithelial cells, Vero or HeLa cells, or any cells which possess the Gb3 (or Gb4 for Stx2e) receptor (32, 43) Escherichia coli O157:H7 induces damage by local cytokine response in clonic mucosa and profuse bleeding of endotheli al cells caused by the interaction of inflammatory cytokines and Stx (blood vessels damaged in lamina propria). A/E lesions are sufficient to cause nonbloody diarrhea but Stx is essential for the development of bloody diarrhea and hemorrhagic colitis. The ga strointestinal illness hemorrhagic colitis (HC) is characterized by severe cramping, a bdominal pain, watery diarrhea followed by grossly bloody diarrhea, and little or no fever. Hemolytic-uremic syndrome (HUS) is associated with Stx producing E coli HUS is the damage to glomerular endothelial cells which leads to narrowing of capillary lumina and clogging of the glomerular

PAGE 43

34 microvasculature with platelets and fibrin. The decreased glomerular filtration rate is responsible for the acute renal failure th at is typical of HUS. HC and HUS are responsible for damage to bot h intestinal and renal tissue (32, 43) Escherichia coli O157:H7 was first recognized in 1982 as a foodborne pathogen in ground beef and achieved great notoriety in 1993 after a multi-state outbreak resulting in four deaths. In 1994 the Food Safety Inspection Service (FSIS) declared E coli O157:H7 as an adulterant in beef and caused it to be the first microorganisms to have such status under the Federal Meat inspection Act. E coli O157:H7 has been reported in feces, rumen contents and on the hide of cattle during slaughter. Ac cording to BarkocyGallagher et al. (3) E coli O157:H7 was less prevalent in feces (12.9%) than on the hide (60.6%) or on preevisceration carcasses (31.7%) in three large beef processing plants in the U.S. E coli O157:H7 was recovered from 1.2% of postintervention beef carcasses. Cattle hides are a source of E coli O157:H7 and can be transferred from hide to carcass during processing. The previous study suggests that hides are a more significant source of contamination than feces. The prevalence of E coli O157:H7 on hides may reflect many sources of contamination like soil, feces from other animals and lairage (3) According to McEvoy et al. (29) E coli O157:H7 can be transferred to the carcass during hide removal operations. The bung ty ing operation has been found to contribute to E coli O157:H7 contamination indicating that this operation may not prevent transfer from the anus to the carcass. Carcass chilli ng has been shown to reduce the prevalence of this pathogen due to the synergisti c effect of low water activity (aw) and temperature (29) Salmonella spp. Salmonella species are straight ro ds measuring 0.7 to 1.5 m by 2 to 5 m, motile by peritrichous flagella and have an optimum growth temperature of 370C. The

PAGE 44

35 nomenclature of Salmonella has changed through the years and presently is comprised of only two species ( enterica and bongori ). S enterica consists of six subspecies and each one contains multiple serotypes. Some Salmonella serotypes, dublin and typhimurium affect cattle and some, cholerasuis and typhimurium affect pigs and others, pullorum and gallinarum affect poultry. Salmonella enterica subsp. enterica serotype typhi and paratyphi A or B are human specific and can cause typhoid fever. The genus Salmonella is composed of more than 2300 serotypes. The main antigens used to distinguish between its serotypes are the somatic (O ), flagellar (H), and capsular (K). Salmonella spp. has a wide occurrence in the natural envi ronment. Intense husbandry practices in the meat, poultry, fish and shellfish industry, along with the recycling of offal into animal feed have favored the presence of this pat hogen in the global food ch ain. This pathogen is a part of the microflora of many anim als like chicken, cattle and reptiles. The predominance of Salmonella spp. in the poultry and egg indu stry has overshadowed its importance in meat, such as pork, beef and mutton. As a means to address this, the USDA FSIS published the “Final Rule on Pa thogen Reduction and Hazard Analysis and Critical Control Points (HACCP) in July 1996. As part of this rule, FSIS will conduct Salmonella testing to ensure the HACCP plan is helping to control this pathogen in finished products (11, 12, 43) This pathogen is encountered through the fecal oral rout and requires a high infectious dose of 10 to 100 million organisms. Human infection with Salmonella can lead to many clinical conditi ons which include enteric (typh oid) fever, uncomplicated enterocolitis, and systemic infections by non-typhoid microorganisms. Uncomplicated gastroenteritis caused by Salmonella enterica subsp. enterica caused secretory diarrhea,

PAGE 45

36 nausea and vomiting. Salmonella enterica subsp. enterica serotype cholerasuis and typhimurium cause focal infection of the vascular endothelium resulting in damage of the intestinal mucosa. Enteric (typhoid) fever is a serious human disease which is associated with typhoid and paratyphoid serotypes. The pathogen penetrates the mucosal barrier potentially at the M-cell surface (Peyers Patch) the apical membrane of the intestinal epithelial cells or the junction between the cells. A ruffling effect of the plasma membrane is observed and a cytoskeletal re arrangement leads to the uptake of the organisms into a phagocytic vesicle. Salmonella travels to the basal membrane and the organisms are released into the lamina propr ia. Bacteremia may result leading to the invasion of the gallbladder, kidneys and reinvasion of the gut mucosa (43) Cases of Salmonella in cattle have been widely re ported. Infected animals can shed the organism without showing clinical signs of disease and can carry this organism into the slaughtering facility. The presence of Salmonella in or on the cattle at the slaughter facility along with the possibility of cross co ntamination of edible carcass is a significant food safety hazard. Salmonella enterica subsp. enterica serotype dublin and typhimurium are the most common serotypes in cattle. The prevalence of Salmonella in feces, rumen and carcass samples was assessed by McEvoy et al. (29) at a commercial beef slaughtering plant in Ireland. Salmonella was found in 2% of feces, 2% of rumen and from 7 to 6% of carcass samples. Salmonella enterica subsp. enterica serotype dublin represented 72% of the positive isolates and Salmonella enterica subsp. enterica serotype typhimurium represented 14% of the positive samples (29) According to RiveraBetancourt et al. (39) hide removal and evisceration are procedures where pathogens may be transferred onto carcasses. Salmonella has been found on carcasses after hide

PAGE 46

37 removal and other slaughte ring and dressing processes (39) According to BarkocyGallagher et al. (3) Salmonella was less prevalent in feces (4.72%) than on the hide (71.0%) or on preevisceration carcasses (13.38%) in three large beef processing plants in the U.S. Salmonella had a prevalence of 0.1% on postintervention beef carcasses. Salmonella prevalence rates of 15.4 to 86.9% on b eef hides have been obtained in previous U.S. studies (3) Shigella spp. Shigella species are facultative anaerobic nonmotile straight rods with an optimum growth temperature of 370C. This genus consists of four species ( dysenteriae flexneri boydii and sonnei ) which are serologically distingui shed by the O antigen of their lipopolysaccharide (LPS). Shigella dysenteriae causes the most serious illness (dysentery) and Shigella sonnei causes a mild illness (water diarrhea). The transmission is from human to human although contamin ation through food or water contaminated with feces also occurs. The infective dose is very small (100 to 1000 cells) and can easily be transmitted from human to human. Shigella spp. is not commonly associated with any specific foods, but some outbreaks have occurr ed in potato salad, chicken and shellfish. This pathogen is usually in troduced into the food supply by an infected person like food handlers with poor personal hygiene (12, 19, 43) Shigellosis is characterized by the produc tion of watery to bloody diarrhea or dysentery. The dysentery stage of disease is characterized by the extensive bacteria colonization of the colonic mucosa. The orga nism invades the epithelial cells at the M cells (Peyer Patch). The bacteria are rele ase into the lamina propria where they are ingested by macrophages. They then reente r the mucosal epithelial cells by rearranging actin and cytoskeletal elements. Once in side they escape the phagosomal vesicle and

PAGE 47

38 multiply. As the membrane breaks the organisms is released into the host’s cytoplasm and the cell to cell invasion c ontinues. Abscesses and muco sal ulcerations are produced at infected sites, therefore causing blood, pus and mucus in stools (12, 43) Klebsiella spp. Klebsiella species are facultative anaerobic nonmotile encapsulated straight rods (0.3 to 1.0 m by 0.6 to 6.0 m) that are arranged singly, in pairs or in short chains. Klebsiella species have an optimum growth temperature of 370C. This organism can be found in animal and human feces, soil, water, sewage, grains, fruits and vegetables. One to 6% of healthy people carry Klebsiella in their throat or nose. Klebsiella species include ornithinolytica oxytoca planticola pneumoniae pneumoniae subsp. pneumoniae pneumoniae subsp. ozaenae pneumoniae subsp. rhinoscleromatis and terrigena Klebsiella pneumoniae and oxytoca and sometimes other species are opportunistic pathogens which can cause bacteremia, pneumonia, urinary tract infections and other human infections. The ability of Klebsiella to cause gastroen teritis is unclear (19, 20, 28) Serratia spp. Serratia species are facultative anaerobic straight rods (0.5 to 0.8 m by 0.9 to 2.0 m). Serratia species are usually motile by peritric hous flagella and have an optimum growth temperature of 30 to 370C. This organism can be found in human clinical specimens, soil, water, plant surfaces, environmental sites, and the digestive tract of rodents and insects. Serratia species include entomophila ficaria fonticola grimesii liquefaciens marcesans odorifera plymuthica rubidaea In domestic animals Serratia has been isolated from pigs, cows, horses, rabbits and poultry. Serratia marcesans and liquefaciends are prominent opportunistic pathogens causing septicemia and urinary tract

PAGE 48

39 infections in hosp italized humans. Serratia in humans are nearly exclusively acquired nosocomially and infection usually only involves Serratia marcesans This organism causes mastitis in cows a nd other animal infections (19, 20, 28) Enterobacter spp. Enterobacter species are facultative anaerobic straight rods (0.6 to 1.0 m by 1.2 to 3.0 m) that are usually motile by peritrichous flagella (except: asburiae ) and have an optimum growth temperature of 30 to 370C. This organism is widely distributed in nature and can be found in animal and huma n feces, soil, water, sewage, plants and vegetables. Enterobacter species include agglomerans aerogenes amnigenus asburiae cancerogenes cloacae dissolvens gergoviae hormaechei intermedius kobei nimipressuralis pyrinus and sakazakii This organism may also be isolated from farm animals, primates, fish and insects. Enterobacter is frequently isolated from foods like meats, dairy products, poultry and vegetables since it is common to plants (crops) and animals. Beef and pork products like ground beef, pork sausage, pork loin, and beef steaks are common reservoirs of this organism. Enterobacter cloacae aerogenes and agglomerans are present in 2.6% to 31.2% of ground beef, packaged pork sausage, vacuum packaged beef trim and primal cuts just before retail manipulation. Enterobacter cloacae sakazakii aerogenes agglomerans and gergoviae are opportunistic pathogens that cause burn, wound, and urinary tract inf ections and sporadically cause septicemia and meningitis (19, 20, 28) Citrobacter spp. Citrobacter species are facultative anaerobic straight rods (2 to 6 m in length) that are arranged singly and in pairs. Citrobacter are usually motile by peritrichous flagella and have an optimum growth temperature of 370C. This organism occurs in soil, water,

PAGE 49

40 sewage, foods, and in the intestin al tract of various animals. Citrobacter has been recovered from the intestinal tract of cows horses, dogs, cats and birds. It is an increasingly important pathogen in fish. Citrobacter species include amalonaticus braakii farmeri freundii koseri rodentium sedlakii werkmanii and youngae Most infections with this organism are nosocomia lly acquired. Approximately 1% to 2% of bacteremias, pneumonias, and wound and ur inary tract infections are caused by Citrobacter species. Citrobacter freundii and koseri have the potenti al to colonize humans and are associated with infection (19, 20, 28) Pantoea spp. Pantoea species are facultative anaerobic straight rods (0.5 to 1 m by 1 to3 m) motile by peritrichous flagella that have an optimum growth temperature of 300C. This organism has been isolated from soil, water, seeds, and from animal and human wounds, blood and urine. Pantoea species include agglomerans ananatis citrea dispersa punctata stewartii and terrea Pantoea agglomerans is considered an opportunistic human pathogen. The genus name Pantoea means all sorts of sources or from diverse geographical samples (19, 20, 28) Morganella spp. Morganella species are facultative anaerobic straight rods (0.6 to 0.7 m by 1 to1.7 m) motile by peritrichous flagella that ha ve an optimum growth temperature of 370C. This organism occurs in the intestinal tract of dogs, other mammals, and reptiles. Morganella species include morganii subsp. morganii and morganii subsp. sibonii Morganella morganii subsp. morganii is an opportunistic seconda ry invader that has been associated with bacteremia, respiratory tr act, wound, and urinary tract infections in

PAGE 50

41 humans. Limited evidence does not support the role of Morganella morganii subsp. morganii in diarrheal diseases (19, 20, 28) Kluyvera spp. Kluyvera species are facultative anaerobic straight rods (0.5 to 0.7 m by 2 to 3 m) motile by small peritrichous flagella that have an optimum growth temperature of 370C. This organism occurs in food, soil, sewage, human respiratory tract, feces and urinary tract. Kluyvera species include ascorbata cochleae cryocrescens and georgiana This organism is an infr equent opportunistic pathogen (19, 20, 28) Cedecea spp. Cedecea species are motile facultative anaerobic round shaped rods (0.5 to 0.6 m by 2 to 2 m) that have an optimum growth temperature of 370C. This organism has been isolated from humans. More than 50% of the cases involving Cedecea have been isolated from the respiratory tract. Cedecea species include davisae lapagei and neteri This organism is an infrequent opportuni stic pathogen. Little epidemiological information about this organism is currently available (19, 20, 28) Leclercia spp. Leclercia species are facultative anaerobic straight rods that are motile by peritrichous flagella and have an optimum growth temperature of 370C. This organism has been isolated from a variety of f oods, water, and environmental sources. Leclercia adecarboxylata is an occasional opportunistic human pathogen. This organism has been classified with the Enterobacteriaceae family as Escherichia adecarboxylata and Enterobacter agglomerans before being given its own genus with a single species. A battery of conventional biochemical tests a nd computer assisted probability matrix has clearly dismissed its association with Escherichia and Enterobacter (19, 20, 28)

PAGE 51

42 Non Enterobacteriaceae Acinetobacter spp. Acinetobacter species are nonsporing aerobic/mi croaerophilic rods (0.9 to 1.6 m by 1.5 to 2.5 m) that become spherical in their stationary phase of growth. They commonly occur in pairs and in chains of diffe rent length. This organism stains Gram negative but can be rather difficult to destai n. The optimum growth temperature is 33 to 350C but Acinetobacter species can grow between 20 and 300C. This organism grows in soil, water and sewage. This spoilage micr oorganism is able to grow on aerobically stored refrigerated muscle foods. Acinetobacter species include baumanii calcoaceticus haemolyticus johnsonii junii lwoffii and dioresistens Acinetobacter is considered to be nonpathogenic but has been known to cause sept icemia, meningitis, endocarditis, pneumonia, and urinary tract infections in humans (12, 19, 26, 28) Moraxella spp. Moraxella species are aerobic/microaerophilic rods (1 to 1.5 m by 1.5 to 2.5 m) or cocci (0.6 to 1 m in diameter) that commonly occur in pairs and in short chains and can present capsule. This organism stains Gr am negative but can re sist decolorization. The optimum growth temperature for this organism is 33 to 350C. Moraxella species include atlantae boevrei bovis caprae equi lacunata lincolnii nonliquefaciens osloensis ovis and saccharolytica This spoilage microor ganism grows on aerobically stored refrigerated muscle foods. It can be found in the mucus membranes of humans and animals (12, 19, 26, 28) Chryseomonas spp. Chryseomonas species are aerobic/microaerophilic rods with parallel sides and rounded ends. They are motile by multitrichous polar flagella and grow in a temperature

PAGE 52

43 range of 18 to 480C. This organism is not know to be present in the general environment but can be a saprophyte or commensal of hum ans and animals in which they may be pathogenic. The single Chryseomona species is luteola (19, 28) Flavimonas spp. Flavimonas species are aerobic/microaerophilic rods with pa rallel sides and rounded ends. They grow in a temp eratures ranging from 18 to 480C and are know to be present in the genera l environment. They can be saprophytes or commensals of humans and animals in which they may be pathogenic. The single Flavimonas species is oryzihabitans (19, 28) Pseudomonas spp. Pseudomonas species are aerobic/microaerophilic st raight or slight ly curved rods (0.5 to 1 m by 1.5 to 5.0 m) that are motile by one or several polar flagella. There are more than 76 species which are widely di stributed in nature and some species are pathogenic to humans, animals or plant. Pseudomonas species are normally associated soil and fecal mater and theref ore are found on the hide of b eef and pork. This spoilage microorganism is able to successfully grow on aerobically stored refrigerated muscle foods, but its growth is suppressed under vacuum and modified atmosphere storage (12, 28) Stenotrophomonas spp. Stenotrophomonas maltophilia is an aerobic/microaerophilic short to medium sized rod that is nonsporing and nonencapsulated. This organism is motile by polar tuft flagella and has an optimum gr owth temperature of 35 to 370C. This organism used to be classified as a Pseudomonas and is an organism of low viru lence with limited ability to

PAGE 53

44 cause infection in humans. Stenotrophomonas maltophilia is a water organism that can survive and multiply in aquatic environment for extended periods (12, 19, 28) Chryseobacterium indologenes Chryseobacterium indologenes is also known as Flavobacterium indologenes which is an aerobic/microaerophilic rod with parallel sides and r ounded ends. It is nonsporing and nonencapsulated and has an optimum growth temperature of 370C (19, 28) Gram Positive Bacteria Listeria spp. The genus Listeria belongs to the Bacilli class along with Bacillus Staphylococcus Streptococcus Lactobacillus and Brochothrix It is a Gram positive non-spore forming, facultative anaerobic rod that has no capsule. It has regular shaped short rods measuring 0.4 to 0.5 m by 0.5 to 2 m with rounded ends. Listeria spp. is motile by a few peritrichous flagella when grown at 20 to 250C. This organism’s optimum growth temperature is 30 to 370C but can grow in temperature ranging from 0 to 450C. The Listeria genus consists of six species but Listeria monocytogenes (LM) and Listeria ivanovii are the only potential pathogens. Listeria seeligeri is considered nonpathogenic, although it has been implicated in so me cases of human listeriosis. Listeria spp. can be isolated from a vast amount of environmental s ources like: soil, water, effluents, a variety of foods, and feces of humans and animals. F oods most often implicat ed in listeriosis are soft cheeses and other dairy pr oducts, meat products, salads and ready to eat products not necessarily reheated before eating (12, 19, 51) Listeriosis is an illness of major public h ealth concern, because of the severity of the disease resulting in abortion, meningiti s and septicemia. Very few people are

PAGE 54

45 infected annually, about two to 10 cases per million population in Europe and United States. LM seems to have a lower pathogeni c potential than other food-borne pathogens with a suspected minimum dose requirement of 106 cells. It has a high case-fatality rate of approximately 20 to 30%, presents a l ong incubation time in the host, and has an affinity for individuals who have underlying co nditions that lead to impairment of T-cell mediated immunity. The susceptibility of the host plays an im portant part in the presentation of clinical diseases upon expos ure to LM. Those groups most at risk are pregnant women and neonates, immunocomprom ised adults, and the elderly (>60 years old). The incidence of listeriosis in pre gnant women is estimated to be 12 per 100,000, which is high compared to 0.7 per 100,000 in the general population. Listeriosis has a fondness for expectant mothers, showing an estim ated 17-fold increase in incidence. The severity and high mortality rate (20 to 30%) of LM infected neonates encourages rapid isolation, diagnosis, treatment and overall unde rstanding of this bacterium in order to reduce the likelihood of infection of pregnant women, their babies, and immunocompromised and healthy people (12, 31, 51) Fetal and neonatal listeriosis results from the invasion of the fetus via the placenta and progresses into chorioamnionits, or ascending from the colonized vaginal canal. It is an inflammatory process invol ving the chorion, fetal blood ve ssels, the umbilical cord, and the amnion by extension of the inflammation. The inflammatory process is potentially fatal to mother and fetus. The c onsequences are abortion, usually five or more months into gestation, a stillborn fetus or th e birth of a baby with a generalized infection characterized by pyogranulomatous microabsces ses all over the body. The infection is usually asymptomatic in the pregnant mother but may present mild flu-like symptoms

PAGE 55

46 like chills, fatigue, fever, headaches and muscle and joint pain about 2 to 14 days before the miscarriage (31, 51) The listeriosis infection most often associated with non-pregnant adults is that of the central nervous system (CNS). Infection normally evolves into a meningoencephalitis, which is accompanied by drastic changes in consciousness, movement disorders, and paralysis. The mort ality rate of CNS liste riosis is about 20% but can be as high as 40 to 60% in imm unocompromised adults. Bacteremia or septicemia is another form of listeriosis whic h can have a 70% mortality rate if associated with debilitated conditions. Other less fr equent clinical forms are endocarditis, myocarditis, arteritis, pn eumonia, pleuritis, hepatitis and peritonitis. Listeria food-borne outbreaks are associated with febrile gastroenteritis whic h includes: diarrhea, vomiting, and fever. The cutaneous form of Listeria infection is ch aracterized by a pyogranulomatous rash (51) The major source of infecti on is through the ingestion of contaminated foods. The primary site of entry of Listeria species into the host is throug h the gastrointestinal tract. Ingestion of LM is thought to be very common, because of its ubiquitous distribution and its probable presence in proce ssed foods. Ingested bacteria must resist acid conditions in the stomach before they can reach the intes tine. Gastric acidity kills a considerable number of Listeria organisms that are ingested in contaminated foods. LM goes through the intestinal epithelium by penetrating th rough the apical surface of polarized Caco-2 cells with a brush border. Other studies ha ve shown that LM penetrates the host via M cells overlying the Peyer’s patch. Organisms th at penetrate the intestinal wall proceed to invade neighboring enterocytes wh ich leads to enteritis. LM survives intracellularly by

PAGE 56

47 preventing phagosome maturation. It is an in vasive pathogen that becomes internalized into many different types of cells that ar e not phagocytic such as epithelial and endothelial cells, hepatocytes, fibroblasts and nerve cells. Once internalized it escapes the double membrane of the phagosome, by mean s of hemolysin and phospholipases, into the cytosol of intestinal cells where it multip lies. During multiplication in the cytosol, Listeria is surrounded by actin filaments which propel it to move from cell to cell where a new round of intracellular prol iferation commences. LM organisms that cross the intestinal barrier are internalized by resident macrophages, in which they can survive and replicate, and are carried thr ough the lymph or blood to the spl een and liver. The initial stage of host tissue colonization is rapid, and it is followed by long a incubation period in which LM establishes a systemic infection. Bacteria are cleared from the bloodstream by macrophages in the spleen and liver. More th an 90% of the bacteria are accumulated in the liver Kupffer cells (16, 24, 31, 51) The entry of LM into slaughtering faci lities occurs through soil on workers’ clothing and shoes, equipment transport, feces and contaminated hides of animals, and human carriers. LM growth is favored by mois t environments and is most often detected in floor drains, condensed and stagnant wate r, floors and processing equipment. This pathogen can readily attach to many surfaces and create biofilms on various meat and dairy processing equipment. LM has been known to survive on workers hands after washing and also in aerosols. The pres ence of this organism on carcasses can be attributes to contamination by fecal matter during slaughtering. Between 18 to 52% of animals are healthy fecal carrier s of LM. About 24% of cattle have contaminated internal retropharyngeal nodes and 45% of pigs carry this organism in their tonsils. LM has been

PAGE 57

48 recovered in clean and unclean zones in sl aughterhouses, and the most contaminated areas are the dehiding, and th e stunning and hoisting areas of porcine slaughtering. Postprocessing contamination is the most prev alent route of contamination of processed foods. It is particularly hard to eliminate this organism from sla ughtering plants because of its ability to form biofilms on food contact surfaces. Refrigerated and moist environments provided in slaughter facili ties are good growth environments for L.M (12) Chasseignaux et al. (7) identified LM in 38% of the samples from three pork and two poultry processing plants in France. Af ter cleaning, the contamination was lowered to 7.7% in the pork plants and 13.1% in the poultry plants. Autio et al. (2) detected LM in 9% of the pluck samples and Listeria innocua in 2% of the pluck samples in ten low capacity pork slaughterhouses in Finland. Th e highest prevalence of this organism was detected in tongue (14%) a nd tonsil (12%) samples and at lower amounts in hearts, kidneys and livers (6%). This organism was detected in eight of the 10 slaughterhouses and LM was detected in 7% of the environmental samples (2) Staphylococcus spp. Staphylococcus species are Gram positive nonmotile, nonspore forming, facultative anaerobic cocci occurr ing in pairs or irregular cluste rs. Its cells measure 0.5 to 1.5 m in diameter. This organism’s optimum growth temperature is 30 to 370C and is one of the hardiest nonspore forming bacteria that can survive extended periods of time on dry and inanimate objects. This organism is relatively heat resistant which allows it to survive in just about every environment in which humans coexist. The genus Staphylococcus is divided into 23 species and subspecies ( arlettae aureus auricularis capitis chromogenes cohnii epidermidis gallinarum haemolyticus hominis hyicus intermedius, kloosii lentus lugdunensis saprophyticus schleiferi sciuri simulans

PAGE 58

49 warneri and xylosus ). The most ubiquitous of these species is Staphylococcus epidermidis which is found on the skin of humans a nd animals and rarely causes disease. The growth of Staphylococcus aureus in foods is a potential p ublic safety hazard since many of its strains produce enterotoxins that cause f ood poisoning when ingested (Staphylococcal food poisoning). Staphylococcus intermedius and hyicus have also been shown to produce enterotoxins in food. Staphylococcus saprophyticus is known only to cause urinary tract infections. The coa gulase test is used to differentiate Staphylococcus aureus from the other species sin ce it is the only one to produce the coagulase enzyme (11, 12, 19) Humans are the main reservoirs of staphylococci involved in human disease, especially Staphylococcus aureus The other species are considered to be normal inhabitants of the exte rnal parts of the body. Staphylococcus aureus usually colonize the external nostrils and are found in about 30% of healthy individua ls. This species also can be found on the skin, oropharynx and feces of humans and animals. Staphylococcus aureus can be isolated from fomites produced by food processing such as meat grinders, knives, saw blades and cu tting boards or tables (11, 12,) Staphylococcal food poisoning (SFP) is th e fourth most common cause of food poisoning when ranked according to outbr eaks and the most common when ranked according to the number of aff ected individuals. SFP is th e classic example of short incubation time food poisoning cause by a t oxin produced in food. Staphylococci produce many metabolites, but the enterotoxins pos e the greatest risk to consumer health. Staphylococcus aureus produces five distinct enterotoxi ns (type A through E) which are single polypeptide proteins with a molecula r size from 22 to 28 kDa. The growth of

PAGE 59

50 Staphylococcus aureus in foods may lead to the produc tion of sufficient enterotoxins which may cause illness when these contamin ated foods are consumed. These toxins cause disease even in the absence of the or ganism. In the majority of cases SFP is associated with food being contaminated by the food handler who might have a minor Staphylococcus aureus infection such as a boil or cut. The contaminated food must be permitted to sit at an adequate temperature that will allow the Staphylococci to multiply and produce the toxin. Reheating the food be fore eating may kill the organism but does not eliminate the heat stable toxin. Some of the foods commonly linked with SFP are meat (beef, pork ad poultry), meat products (sausages, hotdogs, ham), salads ham, chicken, potato), cream filled ba ked products and dairy products (8, 11, 12,) The most common symptoms associated with the ingestion of Staphylococcal enterotoxins are vomiting, diarrhea, abdominal cramps, headaches, muscle cramping and prostration which usually occurs two to six hour s after the toxin is ingested. The diarrhea is usually watery but may also contain blood. Enterotoxins cause intensive peristalsis by working directly with the vomiting control area of the brain. The patie nt usually does not experience fever since there is no infection. The illness is relatively mild and usually lasts only a couple of hours to one day. St aphylococcal enterotoxins display pyrogenic (fever causing), shock inducing, and supe rantigen activity when entering the bloodstream. The toxin, as it ente rs into the bloodstream, acts as a superantigen causing the nonspecific activation of T-cells by binding to the variable region of the chain of the T cell receptor. This causes the proliferation of cytokine secretion leading to septic shock (8, 11, 12,)

PAGE 60

51 In raw food products (beef and pork carcasses) Staphylococcus aureus contamination may not only originate from hu mans but also from animal hides, skin, lesions and bruised tissue (11) Coagulase positive staphylococci (CPS) were found in 39.1% of the samples collected by Desmarchelier et al. (9) in three beef slaughterhouses located in Australia. CPS was isolated from 68.3% of the hides, 3.3% on preevisceration carcasses, 38.9% posteviscerati on carcasses, and 43% of pos tintervention carcasses. After overnight chilling the incidence of C PS decreased to 33% but rose to 83% after three days of chilling. The incidence of CPS when sampling the flank, brisket, and round of twenty carcasses was 40% 35% and 5% respectively. It was observed that the workers manipulated the areas of the flank a nd brisket extensively during the removal of the hide and internal organs. CPS were is olated from seven of ten air samples taken during that study (9) Bacillus spp. Bacillus species are endospore forming rod shaped cells (0.5 to 2.5 m by 1.2 to 1.0 m) and often arrange in pairs or chains. The cells are motile by peritrichous flagella and there is only one spore produced per cell where the exposure to oxygen does not repress sporulation. This organism is a common soil saprophyte and can easily contaminate many types of food like pasta, ri ce, meat, eggs and dairy products. The major food pathogen is Bacillus cereus which causes two types of foodborne illnesses: the diarrheal type and the emetic type. Th e diarrheal type is caused by an enterotoxin produced by Bacillus cereus during its vegetative state, in the small intestine. The emetic type of foofborne illness is caused by the organism growing in the food and producing the toxin. Spores that survive the food being heated are the source of the food poisoning. Heat treatment causes the spores to germinate and in the absence of

PAGE 61

52 competitive flora Bacillus cereus grows very well. Paenibacillus popilliae used to be classified as Bacillus popilliae but now is a taxonomic dissection of the Bacillus genus (12, 19) Brevibacterium spp. Young cells of Brevibacterium species are irregular rods (0.6 to 1.2 m by 1.6 to 5 m) often arranged singly or in pairs and of ten have V formations. Older cultures yield rods that segment into small cocci. This organism is nonsporing and nonmotile and its optimum growth temperature is 20 to 350C. Brevibacterium species include casei epidermidis frigoritolerans halotolerans iodinum linens liquefaciens mcbrellneri otitidis and stationis This organism is widely distri buted in dairy products and is also found on human skin (19, 28) Cellulomonas spp. Young cells of Cellulomonas species are slender irre gular rods (0.5 to 0.6 m by 2 to 5 m) often arranged in pairs or have V forma tions. Older cultures yield rods that are short and may segment into small cocci. This organism is often motile by single flagella and is nonsporing, having and optimum growth temperature of 300C. There are about ten Cellulomonas species that are widely distributed in soils and decaying plant matter (19, 28) Brochothrix spp. Brochothrix species are nonsporing nonmotile nonencapsulated regular unbranched rods (0.6 to 0.7 m by 1 to 2 m) that occur singly, in chains or in long filamentous chains that form knotted masses. This species occurs mainly in meat products and is widely distributed in the environment. Brochothrix optimal growth temperature is 20 to 250C, but can grow from 0 to 300C. Brochothrix thermosphacta is a meat spoilage

PAGE 62

53 microorganism that grows during refrigerati on. This spoilage microorganism, along with lactic acid bacteria, predominates in v acuum packaged and modified atmosphere conditions (12, 19) Microbacterium spp. Young cells of Microbacterium species are slender irre gular rods (0.4 to 8.0 m by 1 to 4 m) often arranged singly or in pairs and may have V formations. Older cultures yield rods that are short and may segment into small cocci. This organism is nonmotile or motile by one to three flagella, nonsporing and has and optimum growth temperature of 300C. Microbacterium species are found in dairy products, sewage and insects (19, 28) Micrococcus spp. Micrococcus species are strictly aerobic spherical cells (0.5 to 2 m in diameter) that occur in pairs, tetrads, or irregular clusters, but not in chains. This organism is usually not motile and nonsporing and its optimum growth temperature is 25 to 370C. Micrococcus species occur mostly on mammalian skin and in soil but are frequently isolated from food prod ucts and the air. Kocuria kristinae and Nesterenkonia halobia used to be classified as Micrococcus kristinae and Micrococcus halobia but now are a taxonomic dissection of the Micrococcus genus. This organism is frequently isolated from beef hides along with Staphylococcus and Pseudomonas (12, 19) Bioaerosol research in the meat slaughter and processing area has been limited to determining the levels of airborne bacteria and yeast and mold contaminants. Limited information exists about the bacterial compos ition of bioaerosols created in a beef and pork slaughtering facility. Information about the existence of pathogenic microorganisms in bioaerosols and the possibili ty of their transfer to food products is of great importance

PAGE 63

54 to food safety. The main objectives of th is study were to enumerate total airborne bacteria and yeast and mold contamin ants and determine the presence of Escherichia coli O157:H7, Listeria spp., Salmonella spp., and Staphylococcus spp. in bioaerosols generated in a small scale slaughter faci lity and on pork and beef carcasses.

PAGE 64

55 CHAPTER 3 MATERIALS AND METHODS Air Sampling System Air samples collected in this study we re obtained from the United States Department of Agriculture (USDA) inspected slaughter facility at the University of Florida, Gainesville, FL. Duplicate air sa mples were taken before and during pork and beef slaughtering processes at the bleeding area, hide removal/dehairing area, back splitting area and the holding cooler. Air samples were collected during three se parate pork and beef slaughtering days using the N6 single stage impactor (Thermo Andersen Corp., Smyrna, GA). Each beef slaughter consisted of five to 10 steers and/ or cows and each pork slaughter consisted of 20 to 25 pigs. The air sampling areas selected for beef and pork sla ughtering processes in this study are presented in Table 1. Table 1. Air sampling areas for beef and pork slaughtering processes Sampling Area Beef Pork Before Slaughter A Bleeding Pit Bleeding Pit B Hide Removal Scalding Tank C Back Splitting Back Splitting D Cooler (hot box) Cooler (hot box) During Slaughter E Bleeding Pit Bleeding Pit F Hide Removal Scalding Tank G Back Splitting Back Splitting H Cooler after 24 hr. Cooler after 24 hr.

PAGE 65

56 The N6 single stage impactor consists of an aluminum inlet cone, jet stage, a base plate held together by three sp ring clamps and sealed with O-ring gaskets, and a vacuum pump. The jet stage or sampling stage consis ts of 400 precision machined orifices. The vacuum pump pulls air from the environmen t through the inlet cone and jet stage and onto a Petri dish containing agar nutrient media. This single stage viable impactor requires a flow rate of exactly 28.3 liters/minute (1 cubic f oot/ minute) to achieve a cutoff diameter of 0.65 m (46) The impactor was shipped to Thermo Andersen Corporation (Smyrna, GA) for adjustment to a flow rate of 28.3 0.1 liters/minute before sampling. According to Lutgring et al. (27) the single stage viable impactor sampler should be calibrated with a primary flow mete r and adjusted to a flow rate of 28.3 0.1 liters/minute on a regular basis. Sampling ti me can range from 10 seconds to 10 minutes according to the level of airborne cont amination in the slaughter facility (27) The impactor was washed at the beginning of each slaughter with soapy water and disinfected with 1000C distilled water containing a ch lorine concentration of 100 parts per million (7.1 ml per 3.8 liters of water). The jet stage was checked for clogged holes and the impactor was then air dried. At the beginning of each slaughter the inlet cone, jet stage, and base plate were disinfected with alcohol swabs (Fisher Scientific, Pittsburgh, PA). The impactor was also disinfecte d with alcohol swabs when changing from sampling point to sampling point (A through H). The sampler was turned on for two minutes prior to sampling to allow the alcoho l to evaporate and not affect the amount of bacteria recovered. Air sample s were taken at a height of 1.5 meters from the floor and within one meter from the carcass. Steril e Petri dishes contai ning non-selective media (TSA, PDA) (Difco Inc., Detroit, MI) a nd selective media (VRBA, MOX, XLT4, MSA)

PAGE 66

57 (Difco Inc., Detroit, MI) were inserted, without th eir top, into the disinfected N6 single stage impactor and the vacuum was turned on for one minute at a flow rate of 28.3 0.1 liters/minute. After one minute the sampler was turned off and the Petri dish was removed and inverted in its cover. Petri dish es were stored in st erile bags to prevent contamination before further analysis. Direct air samples were coll ected in duplicates for each sampling point (A through H). Air Sample and Carcass Swab Collection Preliminary Studies Work was conducted to determine the time and volume of air to be sampled with the N6 single stage impactor to attain appr opriate growth on Petri plates and achieve a representative sample. Sampling times of 30 seconds, two, five, seven and 10 minutes were tested. It was determined that samp ling times between 30 seconds and two minutes with a flow rate of 28.3 0.1 liters/minute yielded countable bact erial growth on Petri plates (i.e. less than 250 CFU). Petri dishes measuring 15 mm by 95 mm (Fisher Scientific, Pittsburgh, PA) were filled with 20 ml of media yielding a plate height of 7 mm. The air, in the areas listed in Table 1, was sampled for total aerobic bacteria and total yeast and mold using tryptic soy agar (TSA) and potato dextrose agar (PDA), respectively. The ability to culture Escherichia coli O157:H7, Listeria spp., Salmonella spp., and Staphylococcus spp. onto their respective selectiv e media and enrichment broths was determined. Confirmed cultures of Escherichia coli O157:H7, Listeria spp., Salmonella spp., and Staphylococcus spp. were streaked onto violet red bile agar (VRBA), modified oxford agar (MOX), xyl ose-lysine-tergitol 4 agar (XLT4) and mannitol salt agar (MSA), respectively and typical colony growth was confirmed. In an

PAGE 67

58 attempt to isolate these bacteria from air samples, they were collected onto prepoured plates containing the previously discussed selective media. Carcass Sample Collection Pork and beef carcass surface bacterial samples were collected using a sponging technique. Swabs were taken from five different carcass sides after pork and beef slaughters. Carcasses were swabbed 12 hour s after being in the holding cooler at 00C. Beef and swine carcass samp ling kits (Solar Biologicals Inc., Ogdensburg, NY) were used to collect microbial surface samples. E ach individual sterile ki t consisted of a premoistened sponge, 15 ml Butterfield’s phosphate buffer, disposable template, sample bags and gloves. Butterfield’s buffered phos phate stock solution (BBPS) consisted of 34 g of potassium phosphate monobasic and 500 ml of distilled water. Beef carcass sides were sampled at flank, brisket and rump. Po rk carcass sides were sampled at the belly, jowl and ham. Samples were taken in the specific order stated above. Carcass swab samples were collected by wiping the 100 cm2 area inside the template with a sterile premoistened sponge 10 times horizontally and 10 times vertically. The wiping force was sufficient to remove dried blood. The flank and brisket (beef) areas and the belly and jowl (pork) areas were swabbed with one side of the sponge first. Next, the other side of the sponge was used to swab the rump (beef ) and the ham (pork) 10 times horizontally and 10 times vertically. Individual sponges we re placed into sample bags along with 15 ml Butterfield’s phosphate buffer. The bags were closed and kneaded several times. After samples were collected they were st ored in a walk in cooler at 5 to 60C until they were analyzed (48)

PAGE 68

59 Air and Carcass Microbiological Sample Analysis Procedures for culturing and isolating b acteria from food and water samples were conducted as outlined in the Microbiology Laboratory Guidebook (MLG) (49) and Bacteriological Analytical Manual (BAM) (50) Since there were no AOAC International (Association of Analytical Communities) accepted methods for the culturing and isolating bacterial pathogens from air samp les, the above manuals for food and water analysis were adapted for air samp le analysis. Air samples collected on Petri dishes filled with non-selective media (T SA, PDA) and selective media (VRBA, MOX, XLT4, MSA) were incubated for specific tim es and temperatures. TSA plates were incubated at 35oC for 24 hours and PDA plates were incubated at 25oC for 3 days. VRBA, MOX, XLT4 and MSA plates were incubated at 35oC for 24 hours, 35oC for 24 to 48 hours, 35oC for 24 hours and 35C for 24 hours, respectively (49) The USDA through the MLG and FDA through the BAM suggest incubating TSA for 48 hours at 350C. In preliminary work, air samples coll ected on TSA plates were incubated for 48 hours at 350C and some colony drying and hard ening was observed. This colony hardening and drying was negatively affecting further analysis of the air samples. In this study sufficient colony growth was attained after incubating TSA plates for 24 hours at 350C. Direct enumeration of TSA, PDA, VRBA, MOX, XLT4 and MSA plates was conducted after incubation. Sterile conditions were maintained during all microbiological analyses in order to prevent the contamination of samples and is olates obtained from the air samples. Previously enumerated TSA plates were then used for detection of Escherichia coli O157: H7, Listeria spp., Salmonella spp., and Staphylococcus spp. Air samples were enriched by flooding TSA plates with 0.1% pe ptone water (1.0 g pept one per 1 liter of

PAGE 69

60 distilled water) and then transferring 1ml of liquid to enrichment broths. Ten milliliters of 0.1% peptone water (Difco Inc., Detroit, MI) were added to each TSA plate with a sterile pipette. These plates sat approximat ely five minutes for b acterial colonies to soften and separate from the media. Peptone flooded plates were spun and cell spreaders were used to separate ba cterial colonies from the media and create a homogeneous bacterial mixture. This liquid bacterial slu rry was then transferred to the various preenrichment and enrichment broths. One m illiliter of the liquid bacterial slurry was transferred to duplicate test tubes co ntaining 9 ml of enrichment broths. Enrichment broths used to increase bacter ial concentrations were Modified EC plus Novobiacin (mEC) for Escherichia coli O157: H7; University of Vermont Broth (UVM) and Fraser broth (FB) for Listeria spp.; Lactose broth (LB) and Rappaport Vassiladis broth (RV) for Salmonella spp., and Brain heart infusion broth (BHI) for Staphylococcus spp. The enrichment broths mEC, UVM, FB, LB, RV, and BHI (Difco Inc., Detroit, MI) were incubated at 350C for 24 hours, 300C for 24 hours, 350C for 24 to 48 hours, 370C for 18 hours, 370C for 18 hours and at 350C for 24 hours, respectively (49) Pre-enrichment and enrichment steps were conducted for Listeria spp. and Salmonella spp. analyses for pork and beef slaughters one and two. Only the final enrichment step was conducted for slaughters three through six. Pre-enrichment steps for Listeria spp. and Salmonella spp. were omitted in slaughters three through si x because of the lengthy time involved in sample analysis. There were no notable differences in the recovery rates of Listeria spp. and Salmonella spp. when their pre-enrichment steps were omitted. In slaughters one and two, after the pre-enrichment br oths had been incubated, the test tubes were vortexed and 1 ml was transferred to 9 ml of enrichment broths and thes e were vortexed and incubated.

PAGE 70

61 After concluding the enrichment step, test tubes were vortexed and an inoculation loop was used to transfer a loop full of broth ont o duplicate selective media plates containing VRBA, MOX, XLT4 and MSA in an attempt to isolate Escherichia coli O157:H7, Listeria spp., Salmonella spp., and Staphylococcus spp. respectively. A three-way-streak was conducted in order to achieve optimum co lony isolation. Selective media plates were incubated for the previously specified temperatures and times. Bags containing carcass swab samples were removed from the cooler (5 to 60C) less than one hour after sample collection and an alyzed for presence of Escherichia coli O157: H7, Listeria spp., Salmonella spp., and Staphylococcus spp. The sponge and contents of the bags were kneaded several times to thoroughly mix samples. One milliliter was aseptically transferred from the bags to test tubes containing 9 ml of the enrichment broths: Modified EC plus N ovobiacin (mEC), Fraser broth (FB), Rappaport Vassiladis broth (RV), and Brain Heart Infusion broth (BHI). The above enrichment step was conducted in duplicate and incubated at th e previously discussed temperatures and times. Incubated test tubes containing the va rious enrichment brot hs were vortexed and an inoculation loop was used to transfer a loop full of broth onto duplicate selective media plates previously discussed to isolate Escherichia coli O157:H7, Listeria spp., Salmonella spp., and Staphylococcus spp. A three-way-streak was conducted to obtain individual colonies. Selective media plates were incubated for the previously specified temperatures and times. Individual colonies from the enriched ai r samples, enriched carcass samples and colonies from direct air samples collect ed on selective media (VRBA, MOX, XLT4, MSA) using the N6 single stage impactor we re picked for further analysis. Colonies

PAGE 71

62 were selected on the basis of their macrosc opic or visual qualitie s. A representative sample of individual colonies possessing differe nt colors, shapes or sizes were selected from specific selective media plates. Strict records were kept regarding sampling point and the selective media plates these colonies came from. Colonies were picked from selective media plates that contained air sa mples from as many sampling points (A-H) as possible. The overall objective was to pick visually unlike colonies from different selective media plates from various sampling poi nts for further identification. Due to the vast number of colonies being picked for furt her identification, some colonies had to be preserved. Colonies were pr eserved in TSB broth containi ng 40% glycerol (Fisher, Pittsburgh, PA). A 40% solution of glycerol in TSB was made and then autoclaved at 1210C for 15 minutes. Individual colonies were picked with a ster ile inoculation loop and stored in 2 ml CryoVials (Fisher Scien tific, Pittsburgh, PA) c ontaining 1 ml of 40% glycerol TSB. CryoVials were vor texed, labeled and stored at -840C for a period of time that did not exceed six months. According to Gorman and Adley (15) 40 to 50% glycerol stocks are extensively employed in th e preservation of bacteria such as Salmonella typhimurium Escherichia coli and Staphylococcus aureus Campylobacter spp. maintained its viability for up to seven mont hs when stored in 10% glycerol at -200C (15) Bacterial Identification of Air and Carcass Samples Colonies preserved in TSB broth containing 40 % glycerol were taken out of the -840C freezer for further analysis. These CryoVials were submerged in room temperature water to thaw. Once the liquid inside the Cr yoVials reached room temperature they were incubated at 350C for 48 hours. After 48 hours, CryoVials were checked for growth (turbidity), and those exhibiting no growth we re incubated for an additional 24 hours.

PAGE 72

63 Incubated CryoVials exhibiting growth were vortexed and an inoculation loop was used to transfer a loop full of brot h onto duplicate selective media plates previously discussed to isolate Escherichia coli O157:H7, Listeria spp., Salmonella spp., and Staphylococcus spp. A three-way-streak was c onducted to obtain indi vidual pure colonies that could be used for further analysis. Selective media plates were incubated for the previously specified temperatures and times. One m illiliter was transferred from individual CryoVials exhibiting growth to test tubes containing 9 ml of their appropriate enrichment broths. These test tubes were incubated unde r the same conditions as discussed earlier. An inoculation loop was used to transfer a loop full of broth onto duplicate selective media plates previously di scussed. A three-way-str eak was conducted to obtain individual pure colonies that co uld be used for further analysis. Isolated colonies were observed and colo r, size, texture and color changes in the media were recorded. Gram stains (Fishe r, Pittsburgh, PA) were performed and colony morphology was determined. Samples were classi fied as coccoid, bac illi or spherical in shape (47) It was necessary to have a Gram iden tification and colony sh ape to be able to select the appropriate API identification stri p (bioMrieux Inc., Marcy l'Etoile, France). Microbiological Identification The API testing system was used to identify Escherichia spp., Salmonella spp., Listeria spp., and Staphylococcus spp. API 20E was used to identify species/subspecies of Enterobacteriaceae ( Escherichia coli and Salmonella spp.) and/or non -fastidious Gram negative rods. API Staph was used for overnight identification of Gram positive Staphylococci spp., Micrococci spp., and Kocuria spp. API is one of the most used diagnostic tools for identification of the Enterobacteriacae family of bacteria. Initiall y, its database was mostly oriented toward

PAGE 73

64 clinical isolates, but in recent years it has grown to include food, industrial and environmental isolates also. The API testi ng system is made up of various small and elongated wells which contain different dehydr ated media, and is intended to perform various biochemical tests from a pure culture grown on an agar plate (11) A 24-hour isolated colony was placed into 5 ml of sterile suspension media (NaCl 0.85%) for API 20E or 6 ml of API Staph medi um for the API Staph strip and vortexed to achieve a homogeneous bacteria suspension. Cultures being identified were never more than 24 hours old. It is recommended to use young cultures betw een 18 and 24 hours old (4) Strips were placed in moist incubation boxes to prevent dehydration and incubated at 350C for 18 to 24 hours. All the wells reacti ons of the API 20E and API Staph strips were recorded after the incubation period. We lls requiring the addition of reagents were revealed and reactions of thes e strips were read by comparing to color change tables provided for each test. Numerical values were obtained for each bacterial colony analyzed and identification was obtaine d through the analytical profile index (4) MIDI’s Microbial Identification System (M IDI Inc., Newark, DE) which identifies bacterial cells by gas chromatographic (GC) anal ysis of cellular fatty acids was used to identify colonies growing on VRBA, MOX, XL T4 and MSA plates that had also been identified by the API method. MIDI’s Micr obial Identification Sy stem (MIS) was also used to identify atypical colonies growing on MOX plates. API Lister ia can only be used to identify Listeria species and will not identify any other bacteria growing in Listeria enrichment broths or on selective media. Gr owth conditions of sample bacteria must be carefully regulated to obtain valid and reproducible cellular fatty acid profiles (41) All

PAGE 74

65 bacterial samples being identified were st reaked onto TSA using the three-way-streak method. Plates were incubated at 350C for 24 hours. Five steps involved in preparing samples for gas chromatographic analysis were: harvesting, saponification, methylation, extractio n and base wash. Approximately 40 mg of bacterial cells from the third quadrant of the three-way-streaked TSA plate were harvested using a 4 mm inoculation loop, and placed in a clean 13x100 culture tube. One milliliter of Reagent 1 (Saponification, 45 g sodium hydroxide, 150 ml methanol, and 150 ml distilled water) was added to each tube containing bacterial cells; they were securely sealed with Teflon lined caps, vortexed briefly and heated in a boiling water bath for five minutes. Tubes were then vigorously vorte xed for 5-10 seconds and returned to the boiling water bath for another 30 minutes. Th e tubes were cooled and uncapped and 2ml of Reagent 2 (Methylation-325 ml certified 6.0 N hydrochloric acid and 275 ml methyl alcohol) were added in order to lower the pH of the solution below 1.5 causing methylation of the fatty acid. At this point the fatty acid me thyl ester were inadequately soluble in the aqueous phase. Tubes were capped and briefly vorte xed and heated for 10 minutes at 800C. A 1.25 ml aliquot of Reagent 3 (Extraction-200 ml hexane and 200 ml methyl tert-butyl ether) was added to the cooled tubes which were recapped and gently tumbled using a clinical rotator for about 10 mi nutes. This reagent was added to extract the fatty acid methyl esters into the organi c phase for use with the gas chromatograph. The tubes were later uncapped and the aque ous (lower) phase was pipetted out for disposal. A base wash was performed by adding approximately 3 ml of Reagent 4 (Cleanup-10.8 g sodium hydroxide dissolved in 800 ml distilled wate r) to the organic phase left behind in the tubes. Tubes were recapped and tumbled for five minutes. After

PAGE 75

66 tumbling, about 2/3 of the organic phase was pipetted into a GC vial which was capped and ready for analysis. This procedure reduced contamination of the injection port liner, the column, and the detector (41) Prepared samples were injected into an Ultra 2 column which is a 25 m x 0.2 mm phenyl methyl silicone fuse d silica capillary column. The GC temperature program heating the column went from 1700 C to 2700C at 50C per minute. A flame ionization detector was used which allowed for a larg e dynamic range of detection and provided a good level of sensitivity. Hydrogen was used as the carrier gas and air was used to support the flame. The GC detector passed an electronic signal to the computer as organic compounds were burning in the flam e of the ionization detector. These electronic signals were integrated into peaks and the data were stored on the hard disk. The composition of the bacterial sample fatty ac id methyl ester was compared to a stored database using pattern recognition software (41) Samples positively identified as Escherichia coli by API 20E and/or MIDI’s Microbial Identification System (MIS) were analyzed to determine if they were pathogenic serotype O157:H7. Two methods based on serological identification were used to detect the presence of Escherichia coli O157:H7. Bacto E. coli O antiserum O157 and Bacto E. coli H antiserum H7 (Difco Inc., Detroit, MI) were used to identify possible O257:H7 Escherichia coli serotypes. Escherichia coli samples being serotyped were streaked onto TSA using the three-wa y-streak method and were incubated at 350C for 24 hours. Reagents were allowed to reach room temperature before opening and using. One drop of Bacto E. coli O antiser um O157 was placed on the left side of a clean slide and on the right a drop of 0.85% saline solution. An inoculation loop was

PAGE 76

67 used to transfer a loop full of the bacteria l colony being serotyped to both sides of the slide. A homogenate was created by blendi ng the bacteria and the liquids, being careful to keep them apart on the slide. The slide was placed under the light microscope and viewed through the 10X, 40X and eventually th e 100 X ocular lens. If agglutination of bacterial cells was observed in the presence of E. coli O antiserum O157 when compared to those cells in the 0.85% saline so lution then the sample was a positive Escherichia coli O157 serotype. The same procedure was followed for the H7 antiserum (10) The second method used was the Reveal Escherichia coli O157:H7 test system (Neogen Corp., Lansing, MI). A portion of enrichment culture was placed into the sample area and passed through a reagent zone which contains specific antibodies (antiEscherichia coli O157:H7) conjugated to colloidal gold pa rticles. If the sample contains antigens they will bind to the gold-conjuga ted antibodies, thus producing an antigenantibody complex that leaves the reagent zone and travel s through the nitrocellulose membrane and displays a visible line (33) Escherichia coli samples being serotyped were streaked onto TSA using the three-wa y-streak method and were incubated at 350C for 24 hours. A loop full of colonies was tran sferred to test tube s containing Modified EC plus Novobiacin (mEC) and incubated at 350C for 24 hours. Test strips were allowed to reach room temperature (250C) before opening their pouches. The appropriate number of test strips required were removed from the foil pouches, and placed on a flat surface and labeled with appropriate sample informati on. A transfer pipette was used to deliver 5 ml of enrichment broth containing culture into the port of the test strip. Results were recorded after 15 minutes.

PAGE 77

68 Statistical Analyses A log transformation of the bacterial count s was performed since the raw data was not normally distributed. Bacterial c ounts were expressed as log10 CFU per m3 of air. The statistical analysis for this study was performed using SAS for Windows (40) The experimental design utilized in this st udy was a randomized complete three-factor factorial. The factors tested in this design were sampling points (8 levels), slaughters (3 levels), and species (2 levels). The depende nt variables for each of the factors tested were types of media (6 variables). Each ai r sample collected in this study consisted of two subsamples. The total number of observatio ns in this experimental design was 576. All data collected in this study were statistically analyzed us ing the analysis of variance method for General Linear M odel Procedures (Proc GLM) (40) Any significant differences were analyzed by the multiple comparisons procedure of LSD (least significant difference), using a le vel of significance of alpha = 0.05 (40)

PAGE 78

69 CHAPTER 4 RESULTS AND DISCUSSION Preliminary Work: Air Sampling Time Determination Preliminary work was conducted to determine the time and volume of air to be sampled with the N6 single stage impactor to achieve appropriate growth on Petri plates and attain a representative sample. Air samp les were collected on TSA filled Petri plates during the slaughter of six steers for 30 sec onds, two, five, seven and 10 minutes (Table 2). It was determined that sampling times between 30 seconds and two minutes with a flow rate of 28.3 0.1 liters/minute yielde d countable bacterial growth on TSA filled Petri plates (i.e. less than 250 CFU). A sa mpling time of five minutes or more yielded bacterial growth on Petri plat es that was too numerous to count (TNTC). A sampling time of one minute during the slaughter pro cess and two minutes when sampling before the slaughter process yielded countable and representative bacterial growth. Table 2. Sampling time determination for air sample collection during a beef slaughter Total airborne bacteria: CFU/ft3 of air a Sampling Time (minutes) Sampling Areab 0.5 2 5 7 10 E 33 107 TNTC TNTC TNTC F 30 155 200 TNTC TNTC G 153 223 TNTC TNTC TNTC H 4 11 12 15 13 a 1 CFU/ft3 of air = 0.02832 CFU/m3 of air, b E = bleeding pit; F = hide removal; G = back splitting; H = cooler after 24 hr.

PAGE 79

70 Microbiology of Bioaerosols Coll ected During Pork Slaughters The results for air samples collected on non selective and selective media during three separate pork slaughter processes are presented in Tables 3 through 13. Total Airborne Bacteria The collection of total airborne bacteria (TAB) on TSA Petri plates for three pork slaughters resulted in no si gnificant differences (P > 0.05) between slaughters (replications). There were however signifi cant difference (P < 0.05) between particular sampling areas within each pork slaughter. During the first pork slaughter sampling area E had the highest log CFU per m3 of air counts and along with areas F and G had significantly higher (P < 0.05) counts than sa mpling areas A, B and C, while sampling area H had the lowest counts. In slaughter two sampling area F ha d the highest counts but was not significantly diffe rent from areas E and G which were not significantly different from areas A and C. In the last pork slaughter sampling areas E, F and G had significantly higher counts than areas A, B, C and D which were significantly higher than area H. Average data revealed significantly higher (P < 0.05) TAB counts in areas E, F, and G when compared to areas A, B, C, D and H. Areas A, B, C and D had similar (P > 0.05) TAB counts. Areas E, F and G also had similar (P > 0.05) TAB counts. The TAB counts for area H were significantly lower (P < 0.05) than the othe r areas sampled. TAB counts were greater than three logs for areas E, F and G and less than three logs for areas A, B, C, D and H. Area D had TAB counts around two logs and ar ea H had counts that were less than one log (Table 3). The existence of non-zero TAB counts before slaughtering (A, B, C, D) could be attributed to the presence of environmental bacteria. The significantly higher counts observed in areas E, F and G when compared to areas A, B, C and D could be attributed

PAGE 80

71 to the ongoing pork slaughteri ng process (Table 3). The presence of animals on the slaughtering floor along with th e slaughtering processes may have greatly increased the amount of bacteria circulati ng in the environment. Animals from feedlots and pens entering a slaughtering facility have a vast amount of microorganisms affixed to their hoofs and hides (>109 CFU/cm2). These microorganisms represent the main food safety hazard in cattle slaughterhouses (21) The majority of processes in slaughterhouses are associated with the creation of bioaerosols. Hide spraying and removal cause bacteria on the surface of animals to become airbor ne. The attachment of pathogenic and nonpathogenic bacteria to car cass surfaces after hide re moval is usually immediate (21) The use of high speed equipment during sl aughtering creates bioaerosols which may cause an increase in carcass contamination (25) Higher bacterial counts in the back splitting area may have been due to wate r flowing from the cutting saw, therefore increasing the creation of bi oaerosols. The manipulation of carcasses by personnel during slaughtering may also have increased the TAB counts in areas E, F, and G. Theoretically the lowest log CFU/m3 counts should have been observed in sampling areas D and H sin ce these areas were at 00C. The growth of psychrotrophic microorganisms is favored at this low temp erature therefore reducing environmental and carcass bacteria. The fact that sampling ar ea H had significantly lower (P < 0.05) counts than area D was conflicting since carcasses were present in area H (Table 3). Theoretically the presence of carcasses inside the cooler should increase the amount of environmental bacteria and therefore increase the counts of area H compared to area D. One reason why sampling area H had lower coun ts may have been because the presence of carcasses disrupted air flow inside the c ooler causing a smaller amount of bioaerosols

PAGE 81

72 to reach the Andersen air sampler. Anot her reason may have been the presence of organic acids in the air and on the carcasses causing a decrease in viable microorganisms. Table 3. Mean total airborne bacteria collected on tryptic soy agar plates in eight areas during three pork slaughters Total airborne bacteria (log CFU/m3 of air) Sampling Areaa Slaughter 1 Slaughter 2 Slaughter 3 Average d SEMb A 2.30CD c 2.09BCD 2.24B 2.21B 0.08 B 2.37C 1.16DE 1.94B 1.82B 0.68 C 2.27CD 2.33BCD 2.21B 2.27B 0.12 D 2.09D 1.94CD 2.33B 2.12B 0.06 E 3.70A 3.34AB 3.36A 3.47A 0.20 F 3.35B 3.77A 3.71A 3.61A 0.07 G 3.55AB 3.32ABC 3.43A 3.43A 0.01 H 0.00E 0.00E 0.77C 0.26C 0.44 Average 2.45 2.24 2.50 2.40 SEM 0.08 0.43 0.29 a A = bleeding pit; B = scalding tank; C = back splitting; D = cooler; E = bleeding pit; F = scalding tank; G = back splitting; H = cooler after 24 hr., A-D are before slaughter and E-H are during slaughter, b Standard error of the mean, c Means within a column followed by different letters (A, B, C, D, E) are significantly different (P < 0.05), d n = 6 observations. Rahkio and Korkeala (34) reported log CFU/m3 of air counts ranging from 3.13 to 4.07 in the back splitting ar ea and counts ranging from 2.45 to 3.58 in the weighing area of a pork slaughter line. Log CFU/m3 of air counts ranging from 2.93 to 3.61 were observed in the evisceration area and counts as low as 2.13 were recorded inside the holding cooler in a pork slaughtering facility (25) Total Airborne Yeast and Mold Significant differences (P < 0.05) between slaughters (replica tions) were observed for some of the sampling areas during three po rk slaughters (Table 4). Sampling areas A and C had significantly higher (P < 0.05) log yeast and mold/m3 of air counts in slaughter one than in slaughter three which had higher c ounts than in slaughter two. No differences

PAGE 82

73 in counts were observed between slaughters two and three at sampling areas B and E, but these slaughters had significantly lower count s than slaughter one. Total yeast and mold counts were similar during all three slaughters for sampling areas D, F, and H (Table 4). These differences in yeast and mold counts be tween slaughters for some of the sampling areas could be attributed to experimental error. Yeast a nd mold collected on PDA media were incubated from three to five days. During this incubation time overgrowth of some yeast and molds could have caused an unde restimation of counts in some plates. Total airborne yeast and mold collected on PDA for three pork slaughters yielded significant differences (P < 0.05) between sampling areas within each slaughter. In slaughter one sampling areas E and F had si gnificantly higher (P < 0.05) log yeast and mold/m3 of air counts than areas A, B, C and G, while areas D and H had the lowest yeast and mold counts. During the second sla ughter sampling areas E, F and G had higher counts than areas A, B and C which had higher counts than areas D and H. In the third pork slaughter sampling areas E, F and G count s were not significantly higher (P > 0.05) than those in areas A, B and C. Sampling area H had the lowest yeast and mold counts for the third pork slaughter. Average data revealed significantl y higher (P < 0.05) log yeast and mold/m3 of air counts in area E when compared to areas A, B, C, D, F and H. Areas A, B, C, F and G had similar (P > 0.05) yeast and mold counts. Yeast and mold counts for areas D and H were significantly lower (P < 0.05) than the other areas sampled. Sampling area H had the lowest c ounts of all the areas sampled. Yeast and mold counts were greater than three logs for areas E, F and G and le ss than three logs for areas A, B, C, D and H. Areas D and H ha d yeast and mold counts that were around two logs (Table 4).

PAGE 83

74 Table 4. Mean total airborne y east and mold collected on potato dextrose agar plates in eight areas during three pork slaughters Total yeast and mold (log yeast and mold/m3 of air) Sampling Areaa Slaughter 1 Sl aughter 2 Sla ughter 3 Average d SEMb A 3.15CX c 2.66BZ 2.88ABY 2.90B 0.05 B 3.17CX 2.57BY 2.81BCY 2.85B 0.06 C 3.09CDX 2.57BZ 2.84ABY 2.83B 0.01 D 2.66E 2.00C 2.52C 2.39C 0.12 E 3.72AX 3.10AY 3.13AY 3.32A 0.03 F 3.33B 2.75A 2.92AB 3.00B 0.17 G 2.95DY 2.98AY 3.10ABX 3.01AB 0.02 H 1.55F 1.85C 1.79D 1.73D 0.14 Average 2.95 2.56 2.75 2.75 SEM 0.05 0.12 0.10 a A = bleeding pit; B = scalding tank; C = back splitting; D = cooler; E = bleeding pit; F = scalding tank; G = back splitting; H = cooler after 24 hr., A-D are before slaughter and E-H are during slaughter, b Standard error of the mean, c Means within a column followed by different letters (A, B, C, D, E) and means within a row followed by different letters (X, Y, Z) are significantly different (P < 0.05), d n = 6 observations. Few differences in yeast and mold count s were seen between air samples taken before and during slaughtering. Both the pr esence of animals on the slaughtering floor and the slaughtering process di d not greatly increase the amount of yeast and mold circulating in the environment. Yeast and mold present before and during slaughtering were probably from environmental sources and not associated with animals and the slaughtering process. The lower yeast and mold counts observed in areas D and H could be attributed to the cooler temperature (00C). Sampling area H may have had lower yeast and mold counts because of the presence of carcasses in the cooler. This could have caused an air flow disruption leading to less yeast and mold being collected by the Andersen air sampler. Kotula et al. (25) reported log yeast and mold/m3 of air counts ranging from 1.92 to 2.84 in the eviscerati on area and counts ranging from 1.57 to 2.52 inside the holding cooler during a pork slaughter.

PAGE 84

75 Isolation of Staphylococcus species Airborne bacteria collected on MSA filled Petri plates during three pork slaughters resulted in no significant differences (P > 0.05) between slaughters (T able5). Significant differences (P < 0.05) were observed between pa rticular sampling areas within each pork slaughter. In the first pork slaughter, sampling area F had th e highest log CFU/m3 of air counts of all the areas. Sampling areas E, F, and G had significantly higher (P < 0.05) counts than areas A, B, C, D, and H for the first and second pork sla ughter. In the last pork slaughter sampling areas E, F, G, and D had significant higher c ounts than areas A, B, C and H. Average data revealed signifi cantly higher (P < 0.05) counts of airborne bacteria collected on MSA media in areas E, F and G when compared to areas A, B, C, D and H. Areas A, B, C, D and H had similar (P > 0.05) counts. Areas E, F and G also had similar (P > 0.05) counts. The bacterial coun ts were around three logs for areas E, F and G and less than one log for areas A, B, C, D and H (Table 5). The significantly higher counts observed in areas E, F and G when compared to areas A, B, C, D and H may be attributed to the presence of animals on the slaughtering floor along with the ongoing sla ughtering processes. Most of the processes during pork slaughtering are associated with the creation of bioaerosols (21) Bacteria on the surface of animals may become airborne during hide spraying and removal (21) Higher bacterial counts in the back splitting area may have b een caused by the blade movement and water flowing from the cutting saw. The presence of Staphylococcus aureus on animal hides, skin, lesions and bruised tissue along with personnel manipulation during slaughtering may also have increased the bacteria counts in areas E, F, and G. Microorganisms collected on MSA filled Petri plates during three pork slaughters are presented in Table 6.

PAGE 85

76 Table 5. Mean airborne bacteria collected on mannitol salt agar plates in eight areas during three pork slaughters Airborne bacteria (log CFU/m3 of air) Sampling Areaa Slaughter 1 Slaught er 2 Slaughter 3 Average d SEMb A 1.55D c 0.00B 0.00B 0.52B 0.00 B 0.00E 0.77B 0.00B 0.26B 0.45 C 0.00E 0.77B 0.00B 0.26B 0.45 D 0.00E 0.00B 1.55A 0.52B 0.00 E 3.23B 3.49A 2.41A 3.05A 0.12 F 3.71A 3.13A 2.51A 3.12A 0.56 G 3.19C 3.02A 2.42A 2.87A 0.01 H 0.00E 0.00B 0.00B 0.00B 0.00 Average 1.46 1.40 1.11 1.32 SEM 0.01 0.40 0.35 a A = bleeding pit; B = scalding tank; C = back splitting; D = cooler; E = bleeding pit; F = scalding tank; G = back splitting; H = cooler after 24 hr., A-D are before slaughter and E-H are during slaughter, b Standard error of the mean, c Means within a column followed by different letters (A, B, C, D, E) are significantly different (P < 0.05), d n = 6 observations. Most of the airborne bacteria collecte d on MSA media during three pork slaughters were Staphylococcus species. The rarely pathogenic Staphylococcus epidermidis is the most ubiquitous of the various Staphylococcus species found on the skin of humans and animals. Staphylococcus aureus is a potential food safety haza rd since many of its strains produce enterotoxins which cause food pois oning when ingested (Staphylococcal food poisoning). Staphylococcus aureus contamination of pork car casses does not only come from human contact but also from animal hides, skin, lesions and bruised tissue (11) Staphylococcus hyicus has also been shown to produce en terotoxins in food. Other Gram positive bacteria isolated from MSA filled Petri plates dur ing three pork slaughters were Bacillus megaterium and Kocuria kristinae The presence of Staphylococcus Micrococcus and Bacillus species in bioaerosols collected during three pork slaughters

PAGE 86

77 could be attributed to their association with mammalian skin, soil and animal feces (Table 6). In this study a total of 171 bacterial colonies were isol ated from various selective media used to collect air samples and carca ss swab samples during three pork and beef slaughters. Of these bacterial isolated a total of 115 were from air samples. Staphylococcus species were identified from 46.7% of the bacterial colonies isolated from air samples during three pork slaughters. Staphylococcus aureus was found in 15% of the colonies isolated from air sa mples during three pork slaughters. Table 6. Airborne bacteria isolated from ma nnitol salt agar plates collected from three pork slaughters Organism (Genus Species) Bacillus megaterium Staphylococcus hominis Kocuria kristinae Staphylococcus hyicus Staphylococcus arlettae Staphylococcus kloosii Staphylococcus aureus Staphylococcus lugdunensis Staphylococcus auricularis St aphylococcus saprophyticus Staphylococcus capitis Staphylococcus sciuri Staphylococcus chromogenes Staphylococcus warneri Staphylococcus cohnii cohnii Staphylococcus xylosus Staphylococcus epidermidis Staphylococcus gallinarum Isolation of Listeria species No significant differences (P > 0.05) between slaughter s resulted from airborne bacteria collected on MOX filled Petri plates during three pork slaughters (Table 7). Significant differences (P < 0.05) were obs erved between individual sampling areas within each pork slaughter. Sampling areas E and F had significantly higher (P < 0.05) log CFU/m3 of air counts than areas A, B, C, D, and H in the first pork slaughter. In the second pork slaughter sampling areas E, F a nd G had significantly higher counts (mean log CFU/m3 of air) than areas A, B, C and H. Du ring the last pork slaughter, counts from

PAGE 87

78 sampling areas E and G were similar to those of areas A, B, D and H. Average data revealed significantly higher (P < 0.05) c ounts of airborne bacteria collected on MOX media in areas E, F and G when compared to ar eas A, B, C, D and H. Areas A, B, C, D and H had similar (P > 0.05) counts. Areas E, F and G also had similar (P > 0.05) counts (Table 7). The bacterial counts were le ss than three logs fo r all sampling areas. Table 7. Mean airborne bacteria collected on modified oxford agar plates in eight areas during three pork slaughters Airborne bacteria (log CFU/m3 of air) Sampling Areaa Slaughter 1 Slaughter 2 Slaughter 3 Average d SEMb A 0.92BC c 0.00C 0.92BC 0.62BC 0.76 B 0.77BC 0.00C 0.92BC 0.57BC 0.70 C 0.00C 0.77C 0.00C 0.26C 0.45 D 0.00C 1.85B 1.55ABC 1.13B 0.17 E 2.56A 3.01A 2.68AB 2.75A 0.06 F 2.93A 2.60AB 3.01A 2.85A 0.08 G 2.00AB 2.40AB 2.59AB 2.33A 0.27 H 0.00C 0.00C 0.92BC 0.31C 0.53 Average 1.15 1.33 1.57 1.35 SEM 0.46 0.30 0.57 a A = bleeding pit; B = scalding tank; C = back splitting; D = cooler; E = bleeding pit; F = scalding tank; G = back splitting; H = cooler after 24 hr., A-D are before slaughter and E-H are during slaughter, b Standard error of the mean, c Means within a column followed by different letters (A, B, C) are significantly different (P < 0.05), d n = 6 observations. The significantly higher counts observed in areas E, F and G when compared to areas A, B, C, D and H could be attribut ed to the ongoing pork slaughtering process. Animal washing, hide spraying, bleeding and hi de removal are processes often associated with the aerosolization of bacteria. Higher bacterial counts in the back splitting area may have been due to water flowing from the cut ting saw, therefore increasing the creation of bioaerosols. Carcass mani pulation by personnel during sl aughtering may also have increased counts in areas E, F, and G. No Listeria species were isolated from air samples

PAGE 88

79 during this study. Microorganisms collected on MOX filled Petri plates during three pork slaughters are presented in Table 8. The majority of the airborne bacteria collected on MOX media during three pork slaughters were Gram positive species with the exception of Escherichia coli Salmonella bongori and Shigella boydii (Table 8). The Gram positive species collected on MOX media have a wide environmental distribution and are associated with the contamination of meat and dairy products (12) The genus Listeria belongs to the bacilli class along with Bacillus Paenibacillus Staphylococcus Streptococcus Lactobacillus and Brochothrix Although bacteria from the bacilli class were isolated, Listeria species were not identified from a ny of the 171 bacterial colonies collected from air and carcass samples. The presence of Bacillus Brevibacterium Brochothrix Cellulomonas Microbacterium and Micrococcus species in bioaerosols co llected during three pork slaughters could be attributed to their predominance in so il, mammalian skin and animal feces. These bacteria could have entered the slaughtering facility through contaminated hides of animals, feces, soil on workers’ clothing and shoes, equipment transport and human carriers. Table 8. Airborne bacteria isolated from modi fied oxford agar plates collected from three pork slaughters Organism (Genus Species) Bacillus pumilus Microbacterium barkeri Brevibacterium casei Microbacterium saperdae Brevibacterium epidermidis Nesterenkonia halobia Brochothrix campestris Salmonella bongori Cellulomonas fimi Shigella boydii Escherichia coli Staphylococcus aureus Kocuria kristinae Staphylococcus epidermidis Staphylococcus haemolyticus

PAGE 89

80 Ren (35) isolated Listeria spp. from specific environmental surfaces of dairy plants but could not recover it from air samples collected close to these surfaces. It is hypothesized that Listeria species suspended in air may loose their colony forming capability even though they retain overall vi ability. Air sample collection using the impaction method has been demonstrated to cause bacteria to enter in to a viable but non culturable (VBNC) state (17) Impaction stress create d during aerosol generation, dispersion and landing or collect ion, along with environmental stressors like temperature, dehydration, irradiation, oxidati on and pollution often prove to be lethal to airborne bacteria (22, 45) Bacteria that are sublethally inju red during sample collection onto agar, using the impaction method, will not form co lonies and, therefore cannot be accounted for (45) It is important to include repair and resuscitation steps using enrichment broths to increase the culturability of sublethally injured bacteria (21) Listeria species may have become lethally injured by sample co llection and/or environmental conditions, or sublethally injured induc ing the VBNC state. Isolation of Escherichia coli The collection of airborne bacteria on VRBA filled Petri plates during three pork slaughters resulted in no si gnificant differences (P > 0.05) between slaughters. Significant differences (P < 0.05) between pa rticular sampling areas were only observed during the third pork slaughter. Sampling ar eas E and F had significantly higher (P < 0.05) counts than area G which had higher counts than areas A, B, C, D and H where no growth was obtained. Average data revealed significantly higher (P < 0.05) counts of airborne bacteria collected on VRBA media in areas E, F and G when compared to areas A, B, C, D and H were no growth was obtaine d. Areas E, F and G also had similar (P > 0.05) counts. The bacterial counts were less than one log for sampling areas E, F and G

PAGE 90

81 (Table 9). The presence of airborne bacter ia collected on VRBA filled Petri plates in areas E, F and G compared to the absence of ba cteria in areas A, B, C, D and H could be attributed to the ongoing pork sl aughtering process. Processe s like animal washing, hide spraying, bleeding and hair removal are of ten associated with the production of bioaerosols. Table 9. Mean airborne bacteria collected on vi olet red bile agar pl ates in eight areas during three pork slaughters Airborne bacteria (log CFU/m3 of air) Sampling Areaa Slaughter 1 Sla ughter 2 Slaught er 3 Averaged SEMb A 0.00 0.00 0.00C c 0.00B 0.00 B 0.00 0.00 0.00C 0.00B 0.00 C 0.00 0.00 0.00C 0.00B 0.00 D 0.00 0.00 0.00C 0.00B 0.00 E 0.77 0.00 2.05A 0.94A 0.46 F 0.00 0.92 1.85A 0.92A 0.53 G 0.00 1.01 1.55B 0.85A 0.58 H 0.00 0.00 0.00C 0.00B 0.00 Average 0.10 0.24 0.68 0.34 SEM 0.27 0.49 0.07 a A = bleeding pit; B = scalding tank; C = back splitting; D = cooler; E = bleeding pit; F = scalding tank; G = back splitting; H = cooler after 24 hr., A-D are before slaughter and E-H are during slaughter, b Standard error of the mean, c Means within a column followed by different letters (A, B, C) are significantly different (P < 0.05), d n = 6 observations. The majority of airborne bacteria co llected on VRBA media during three pork slaughters were Gram negative bacteria from the Enterobacteriaceae family. Enterobacteriaceae have a worldwide distribution and are found in soil, water, fruits, vegetables, plants, trees and animals. They inhabit a wide variety of niches which include human and animal gastrointestinal tracts and various environmental sites (11, 14, 19) Other Gram negative bacteria present like Pseudomonas spp., Chryseobacterium

PAGE 91

82 indologenes, Chryseomonas lute ola, Flavimonas oryzihabitans and Stenotrophomonas maltophilia are all members of the Pseudomonadaceae family (Table 10). These species are normally associated with soil and fecal mater and therefore ar e found on the hides of animals. Gram positive bacteria isolated from VRBA plates were Kocuria kristinae Microbacterium barkeri and Bacillus pumilus which are frequently isolated from meat and dairy products (11, 19) The presence of Enterobacteriaceae species in bioaerosols collected during three pork sl aughters could be attributed to their association with mammalian skin, soil and animal feces. These bacteria could have entered the slaughtering facility on the hides of animal s, feces and by the soil on workers’ shoes. Table 10. Airborne bacteria isolated from violet red bile agar plates collected from three pork slaughters Organism (Genus Species) Cedecea neteri Microbacterium barkeri Chryseobacterium indologenes Pantoea spp. Chryseomonas luteola Pseudomonas aeruginosa Enterobacter aerogenes Pseudomonas fluorescens Enterobacter cloacae Pseudomonas putida Escherichia coli Salmonella bongori Escherichia fergusonii Salmonella choleraesuis choleraesuis Flavimonas oryzihabitans Salmonella typhi Klebsiella pneumoniae ozaenae Serratia liquefaciens Klebsiella pneumoniae pneumoniae Serratia rubidaea Klebsiella terrigena Shigella boydii Kluyvera spp. Shigella flexneri Kocuria kristinae Shigella sonnei Leclercia adecarboxylata Stenotrophomonas maltophilia Generic Escherichia coli was identified from 8.3% of the 115 airborne bacterial colonies isolated during three pork slaughters. Escherichia coli O157:H7 was not isolated from any of the air samples take n from the various sampling areas. The low airborne concentration of Escherichia coli O157:H7 may be one reason this organism

PAGE 92

83 was not isolated. Impaction stress created during aerosol co llection along with environmental factors like temperature and de hydration may have caused lethal injury to this organism or induced the VBNC state, therefore contributing to the inability of isolating this organism from air samples (17) Isolation of Salmonella species Airborne bacteria collected on XLT4 filled Petri plates during three pork slaughters yielded no significant differences (P > 0.05) between slaughters. No significant differences (P > 0.05) were observed between individual sampling areas within each pork slaughter. Average data reve aled significantly higher (P < 0.05) counts of airborne bacteria collected on XLT4 media in area F when compared to the rest of the areas were no growth was obtained. The bacterial counts we re less than one log for sampling area E (Table 11). Table 11. Mean airborne bacteria collected on xy lose-lysine-tergitol 4 agar plates in eight areas during three pork slaughters Airborne bacteria (log CFU/m3 of air) Sampling Areaa Slaughter 1 Sla ughter 2 Sla ughter 3 Average d SEMb A 0.00 0.00 0.00 0.00B c 0.00 B 0.00 0.00 0.00 0.00B 0.00 C 0.00 0.00 0.00 0.00B 0.00 D 0.00 0.00 0.00 0.00B 0.00 E 0.00 0.00 0.00 0.00B 0.00 F 1.01 0.00 0.77 0.60A 0.73 G 0.00 0.00 0.00 0.00B 0.00 H 0.00 0.00 0.00 0.00B 0.00 Average 0.13 0.00 0.10 0.07 SEM 0.36 0.00 0.28 a A = bleeding pit; B = scalding tank; C = back splitting; D = cooler; E = bleeding pit; F = scalding tank; G = back splitting; H = cooler after 24 hr., A-D are before slaughter and E-H are during slaughter, b Standard error of the mean, c Means within a column followed by different letters (A, B) ar e significantly different (P < 0.05), d n = 6 observations.

PAGE 93

84 Most of the airborne bacteria collected on XLT4 media during th ree pork slaughters were Gram negative bacteria from the Enterobacteriaceae family (Table 12). Salmonella species were not isolated from air sample s using XLT4 media during the three pork slaughters. Airborne Salmonella species, Escherichia coli and Shigella species were readily isolated from the less selective VRBA media. Salmonella enterica subsp. enterica serotype typhi choleraesuis houtenae and typhimurium and Salmonella bongori were isolated from VRBA media (T ables 12). The poor isolation of Salmonella from air samples collected on XLT4 may have been due to this media’s high selectivity. Environmental factors along with impaction stress may have re duced the ability to isolate Salmonella species from air samples (17) Salmonella species were identified from 10% of the total airborne bacterial colonies isolated during thr ee pork slaughters. Table 12. Airborne bacteria isol ated from xylose-lysine-terg itol 4 agar plates collected from three pork slaughters Organism (Genus Species) Enterobacter aerogenes Pseudomonas aeruginosa Enterobacter asburiae Pseudomonas fluorescens Enterobacter cancerogenus Pseudomonas putida Enterobacter cloacae Serratia liquefaciens Escherichia coli Serratia marcescens Escherichia fergusonii Shigella flexneri Klebsiella ornithinolytica Shigella sonnei Klebsiella pneumoniae pneumoniae Staphylococcus aureus Morganella morganii Staphylococcus epidermidis Identification of Airborne Bacteri a Collected from Pork Slaughters Most of the airborne bacteria recovered be fore the pork slaughters (A, B, C, and D) were Gram negatives from the Enterobacteriaceae family with the exception of Flavimonas oryzihabitans from the Pseudomonadaceae family. Staphylococcus

PAGE 94

85 Micrococcus and Bacillus species were the main Gram positive airborne bacteria recovered before the pork sl aughters (Table 13). The majority of the Gram negative airborne bacteria isolated during the three pork slaughters (E, F, and G) were from the Enterobacteriaceae and Pseudomonadaceae family. Most of the Gram positive airborne bacteria isolated during the three pork slaughters were Staphylococcus Microbacterium Brevibacterium Bacillus and Micrococcus species. These species are found in the soil and are associated with mammalian skin and are therefore frequen tly isolated from food products and the environment. Another Gram positive organism isolated was Brochothrix thermosphacta which is a meat spoilage microorganism that grows at refrigeration temperatures and predominates in vacuum packaged and modified atmosphere conditions (12, 19 ). Fewer microorganisms were isolated before the pork slaughters (A, B, C, D) when compared to the number of microorganisms isol ated during the slaughters (E, F, and G) (Tables 13). Higher numbers of microorganism s isolated from areas E, F and G could be attributed to the ongoing pork slaughtering process. The presence of animals on the slaughtering floor along with th e slaughtering processes may have greatly increased the number of airborne microorganisms. Most of the microorganisms that were isol ated before the pork slaughters were also isolated during the slaughters. The presence these airborne bacteria during the slaughter process could be attributed to their associ ation with the animals or to their general environmental presence. Bacillus megaterium Staphylococcus auricularis capitis and lugdunensis were only isolated before the three pork slaughters. These microorganisms isolated before the pork slaughters could be regarded as environmental or background

PAGE 95

86 bacteria (Table 13). The least amount of airborne bacteria was isolated from the 00C cooler (D and H). Kocuria kristinae Staphylococcus capitis auricularis and cohnii were isolated from the empty cooler (D) and Staphylococcus cohnii and sciuri were isolated from the cooler after carcasses had be en stored for at least 24 hours at 00C (H). Table 13. Airborne bacteria isolated before and duri ng three pork slaughters Organism (Genus Species) Before Slaughter During Slaughter Gram negative Gram negative Escherichia coli Cedecea neteri Flavimonas oryzihabitans Enterobacter aerogenes Salmonella bongori Enterobacter cloacae Shigella boydii Escherichia coli Gram positive Escherichia fergusonii Bacillus megaterium Flavimonas oryzihabitans Kocuria kristinae Klebsiella ornithinolytica Staphylococcus auricularis Klebsiella pneumoniae ozaenae Klebsiella pneumoniae pneumoniae Staphylococcus capitis Staphylococcus chromogenes Klebsiella terrigena Staphylococcus cohnii cohnii Kluyvera spp. Staphylococcus epidermidis Pseudomonas aeruginosa Staphylococcus hominis Pseudomonas fluorescens Staphylococcus lugdunensis Staphylococcus sciuri Pseudomonas putida Salmonella bongori Staphylococcus warneri Salmone lla choleraesuis choleraesuis Salmonella typhi Serratia liquefaciens Serratia marcescens Serratia rubidaea Shigella boydii Stenotrophomonas maltophilia Gram positive Bacillus pumilus Brevibacterium casei Brevibacterium epidermidis Brochothrix campestris Cellulomonas fimi Kocuria kristinae Microbacterium barkeri Microbacterium saperdae Nesterenkonia halobia Staphylococcus arlettae

PAGE 96

87 Table 13. Continued. Organism (Genus Species) During Slaughter Gram positive Staphylococcus aureus Staphylococcus chromogenes Staphylococcus cohnii cohnii Staphylococcus epidermidis Staphylococcus gallinarum Staphylococcus haemolyticus Staphylococcus hominis Staphylococcus hyicus Staphylococcus kloosii Staphylococcus saprophyticus Staphylococcus sciuri Staphylococcus warneri Staphylococcus xylosus Carcass Swabs Collected During Pork Slaughters The microorganisms listed in Table 14 were isolated from swabs taken from pork carcasses which had been in a cooler at 00C for 24 hours. The majority of bacteria isolated from pork carcass swabs were Gram negative bacteria from the Enterobacteriaceae family. Other Gram negative spoilage microorganisms present were Pseudomonas spp., Chryseobacterium indologene s, Chryseomonas luteola which are members of the Pseudomonadaceae family. The only Gram positive bacteria isolated from pork carcass swabs were Staphylococcus species. A total of 56 out of the 1 71 bacterial colonies isolated during this study were from carcass swabs. Staphylococcus species were identified fr om 52.4% of the bacterial colonies isolated from pork carcass swabs. Staphylococcus aureus was identified from 33.3% of the bacterial colonies is olated during pork carcass sampling. Escherichia coli and Salmonella species were identified from 14. 3% and 4.8% respectively, of the bacterial colonies isolated from pork carcass swabs. Listeria species and Escherichia coli

PAGE 97

88 O157:H7 were not isolated from any of the por k carcass swabs. Acetic acid spraying and carcass washing along with low temperature st orage of pork carcasses may have greatly reduced the growth of these microorganisms. Some organisms like Enterobacter cloacae Escherichia coli and fergusonii Klebsiella pneumoniae pneumoniae Pseudomonas aeruginosa Salmonella choleraesuis choleraesuis Shigella species, Staphylococcus aureus epidermidis and other Staphylococcus species were isolated from air sa mples and pork carcass swabs (Tables 13 and 14). None of the airborne bacteria isolat ed from the empty cooler (D) and the cooler containing carcasses (H) were isolated from the pork carcasses. The isolation of various organisms, including Staphylococcus Escherichia and Salmonella species, from air samples taken during slaughtering and carcass sw abs, supports the theory that bioaerosols transport bacteria and contribute to the contamination of pork carcasses. Table 14. Bacteria isolated fr om pork carcasses held at 00C for 24 hours Organism (Genus Species) Gram negative Gram positive Chryseobacterium indologenes Staphylococcus aureus Chryseomonas luteola Staphylococcus chromogenes Enterobacter asburiae Staphylococcus epidermidis Enterobacter cancerogenus Staphylococcus hominis Enterobacter cloacae Staphylococcus hyicus Escherichia coli Staphylococcus warneri Escherichia fergusonii Staphylococcus xylosus Klebsiella pneumoniae pneumoniae Leclercia adecarboxylata Morganella morganii Pantoea spp. Pseudomonas aeruginosa Salmonella choleraesuis choleraesuis Shigella flexneri Shigella sonnei

PAGE 98

89 Microbiology of Bioaerosols Coll ected During Beef Slaughters The results for air samples collected on non selective and selective media during three separate beef slaughter processe s are presented in Tables 15 through 25. Total Airborne Bacteria Total airborne bacteria (TAB) collected on TSA Petri plates for three beef slaughters resulted in no si gnificant differences (P > 0.05) between slaughters (replications) (Table 15). Significant differe nces (P < 0.05) between specific sampling areas within each beef slaughter were observe d for slaughters one and two. Log CFU per m3 of air counts for sampling areas during beef slaughtering (E, F, G, H) and sampling areas before slaughtering (A, B, D) were not significantly different in slaughter one. Counts for sampling area C were significantly lower than the counts for the rest of the sampling areas. Sampling areas F and G had the highest TAB counts and sampling area D had the lowest counts during the second beef slaughter. There were no significant differences (P > 0.05) between sampling areas fo r the third beef slaughter. Average data revealed significantly higher (P < 0.05) TAB counts in areas F and G when compared to areas A, B, C, D and H. Areas A, B, C, D and H had similar (P > 0.05) TAB counts. Areas E, F and G also had similar (P > 0.05) TAB counts. Sampling area E had similar counts to those of areas A, B and C. Sa mpling areas H and D had the lowest counts although not significantly different (P > 0.05) from those of the sampling areas before slaughter (A, B, C). TAB counts were around three logs for areas E, F and G and less than three logs for areas A, B, C, D and H. Areas D and H had TAB counts that were less than two logs (Table 15). The existence of non-zero TAB counts before slaughtering (A, B, C, D) may have been caused by the presence of environmental bacteria. The signifi cantly higher counts

PAGE 99

90 observed in areas F and G when compared to areas A, B, C a nd D could be attributed to the ongoing beef slaughtering process (Table 15). Th e presence of cattle on the slaughtering floor, combined with the slaugh tering process, may have greatly increased the amount of environmental bacteria. The spraying of cattle and hide removal could have caused surface bacteria to become air borne. The attachment of pathogenic and nonpathogenic bacteria to car cass surfaces after hide re moval is usually immediate (21) Higher levels of TAB present in the back sp litting area may have been caused by water flowing from the cutting saw or due to bioaer osol transfer from contaminated parts of the line like the hide removal area. The manipul ation of carcasses by personnel during hide removal, evisceration and back splitting may have also increased the TAB counts in areas F and G. Table 15. Mean total airborne bact eria collected on tryptic soy agar plates in eight areas during three beef slaughters Total airborne bacteria (log CFU/m3 of air) Sampling Areaa Slaughter 1 Slaught er 2 Slaughter 3 Averaged SEMb A 2.20A c 2.50B 2.35 2.35BC 0.06 B 2.07 A 2.14BC 2.63 2.28BC 0.10 C 0.92 B 2.76AB 3.25 2.31BC 0.54 D 2.78 A 0.77D 1.16 1.57C 0.81 E 3.02 A 2.78AB 2.69 2.83AB 0.16 F 2.94 A 3.61A 2.99 3.18A 0.20 G 3.09 A 3.44A 2.94 3.16A 0.15 H 2.22 A 1.55CD 1.55 1.77C 0.04 Average 2.40 2.44 2.44 2.43 SEM 0.35 0.29 0.44 a A = bleeding pit; B = hide removal; C = back splitting; D = cooler; E = bleeding pit; F = hide removal; G = back splitting; H = c ooler after 24 hr., A-D are before slaughter and E-H are during slaughter, b Standard error of the mean, c Means within a column followed by different letters (A, B, C, D) are significantly different (P<0.05), d n = 6 observations.

PAGE 100

91 Rahkio and Korkeala (34) reported log CFU/m3 of air counts ranging from 2.21 to 3.70 in the back splitting ar ea and counts ranging from 2.21 to 3.59 in the weighing area of a beef slaughter line. According to Jericho et al. (21) a count of 4.03 log CFU/m3 of air was observed at the hide removal stati on during beef slaughtering and a count of 3.25 log CFU/m3 of air was recorded at the hide re moval station before slaughtering. According to Jericho et al. (21) the presence of non-zero TAB counts before slaughtering was attributed to previo us clean-up operation usi ng high pressure hosing. Total Airborne Yeast and Mold Significant differences (P < 0.05) between slaughters (replica tions) were observed for some of the sampling areas during the three beef slaughters (Table 16). Log yeast and mold/m3 of air counts were similar in slaught ers one and two which were significantly higher (P < 0.05) than the counts observed dur ing the third beef slaughter for sampling areas C, E, F and G (Table 16). These di fferences in yeast and mold counts between slaughters for some of the sampling areas coul d be attributed to experimental error. During slaughter three yeast and mold counts may have been underestimated due to their overgrowth on PDA Petri plates. Total airborne yeast and mold collected on PDA for three beef slaughters yielded significant differences (P < 0.05) between sampling areas within each slaughter. Sampling areas E, F, and G had significantl y higher (P < 0.05) log yeast and mold/m3 of air counts than areas A, B and C, while areas D and H had the lowest counts in the first beef slaughter. In slaughter two sampling ar eas E, F and G did not have significantly higher (P > 0.05) counts than areas A, B and C. Sampling area H had the lowest counts during the second beef slaughter. No signi ficant differences (P > 0.05) were observed between the sampling areas in the third beef slaughter. Overall during the three beef

PAGE 101

92 slaughters the sampling areas E, F and G had similar yeast an d mold counts as the areas A, B, and C while areas D and H had the lowest counts (Table 16). Average data revealed sampling areas E, F and G did not ha ve significantly highe r (P > 0.05) log yeast and mold/m3 of air counts than areas A, B and C. Sampling areas H and D had the lowest counts (P < 0.05) of all the areas sampled. Y east and mold counts were greater than three logs for areas E, F and G and less than three logs for areas A, B, C, D and H. Areas D and H had yeast and mold counts that we re less than two l ogs (Table 16). Table 16. Mean total airborne yeast and mold co llected on potato dextrose agar plates in eight areas during three beef slaughters Total yeast and mold (log yeast and mold/m3 of air) Sampling Areaa Slaughter 1 Sla ughter 2 Slaughter 3 Averaged SEMb A 2.76B c 2.92AB 2.45 2.71A 0.12 B 2.84B 2.82AB 2.40 2.68A 0.10 C 2.90BX 2.77ABX 2.27Y 2.64A 0.08 D 1.55D 2.09B 1.01 1.55B 0.59 E 3.59AX 3.67AX 2.60Y 3.29A 0.09 F 3.49AX 3.66AX 2.69Y 3.28A 0.09 G 3.60AX 3.42AX 2.56Y 3.19A 0.11 H 1.87C 0.77C 0.77 1.14B 0.64 Average 2.82X 2.76X 2.09Y 2.56 SEM 0.07 0.28 0.47 a A = bleeding pit; B = hide removal; C = back splitting; D = cooler; E = bleeding pit; F = hide removal; G = back splitting; H = c ooler after 24 hr., A-D are before slaughter and E-H are during slaughter, b Standard error of the mean, c Means within a column followed by different letters (A, B, C, D) and means within a row followed by different letters (X, Y, Z) are signifi cantly different (P < 0.05), d n = 6 observations. No yeast and mold count differences were seen between air samples taken before and during slaughtering. The am ount of airborne yeast and mo ld was not affected by the presence of animals on the slaughtering floor or the slaughtering process. There seems to be little or no association between cattle a nd the level of yeast and molds. Yeast and mold present during slaughtering are proba bly not associated with cattle and the

PAGE 102

93 slaughtering process but are environmental in nature. The lower yeast and mold counts observed in areas D and H could be attri buted to the low cooler temperature (00C). Isolation of Staphylococcus species Airborne bacteria collected on MSA filled Petri plates during three beef slaughters resulted in no significant differences (P > 0.05) between slaughters. Significant differences (P < 0.05) between certain sampli ng areas within each beef slaughter were observed for slaughters two and three. Duri ng the second slaughter, counts for sampling areas E, F, and G were not significantly higher (P > 0.05) than those of area A, but were greater than (P < 0.05) th ose of areas B, C, D, and H. In the third beef slaughter all the sampling area counts were similar except for ar eas D and H which had the lowest counts. Average data revealed significantly higher (P < 0.05) counts of airborne bacteria collected on MSA media in areas E and F when compared to areas A, B, C, D and H. Areas A, B, C, D and H had similar (P > 0.05) counts. Areas E, F and G also had similar (P > 0.05) counts. The bact erial counts were around 2.5 logs for areas E and F and around one log for areas G, A, B, C, D and H (Table 17). The significantly higher counts observed in areas E and F when compared to areas A, B, C, D and H may be attributed to th e presence of cattle on the slaughtering floor along with the ongoing slaughter ing practices. Processes like cattle washing, hide spraying, bleeding and hide removal cause surf ace bacteria to become aerosolized. The presence of Staphylococcus aureus on animal hides, skin, lesions and bruised tissue along with personnel manipulation du ring slaughtering may also ha ve increased the bacteria counts in areas E and F. Microorganisms collected on MSA filled Petri plates during three beef slaughters ar e presented in Table 18.

PAGE 103

94 Table 17. Mean airborne bacteria collected on mannitol salt agar pl ates in eight areas during three beef slaughters Airborne bacteria (log CFU/m3 of air) Sampling Areaa Slaughter 1 Slaught er 2 Slaughter 3 Averaged SEMb A 0.77 1.70A c 1.55AB 1.34B 0.46 B 1.85 0.00B 1.94A 1.26B 0.05 C 1.01 0.00B 2.25A 1.09B 0.59 D 2.29 0.00B 0.77BC 1.02B 0.45 E 2.60 2.59A 2.37A 2.52A 0.13 F 2.90 2.59A 2.15A 2.55A 0.08 G 0.92 2.27A 1.79A 1.66AB 0.56 H 1.90 0.77B 0.00C 0.89B 0.49 Average 1.78 1.24 1.60 1.54 SEM 0.57 0.28 0.30 a A = bleeding pit; B = hide removal; C = back splitting; D = cooler; E = bleeding pit; F = hide removal; G = back splitting; H = c ooler after 24 hr., A-D are before slaughter and E-H are during slaughter, b Standard error of the mean, c Means within a column followed by different letters (A, B, C) are significantly different (P < 0.05), d n = 6 observations. The majority of the airborne bacteria collected on MSA media during three beef slaughters were Staphylococcus species. Other Gram positive bacteria isolated from MSA filled Petri plates during three beef slaughters were Bacillus atrophaeus and Micrococcus species. The association of Staphylococcus Micrococcus and Bacillus species with mammalian skin, soil and anim al feces could explain their presence in bioaerosols collected during thr ee beef slaughters. Some Gram negative bacteria isolated from this media were Moraxella nonliquefaciens Enterobacter aerogenes Salmonella bongori and Shigella flexneri (Table 18). These Gram negative species are from the Enterobacteriaceae family which has a worldwide dist ribution and can be found in soil, water, plants and animal feces. A total of 171 bacterial colonies were isol ated from various selective media used to collect air samples and carcass swab sample s during three pork and beef slaughters.

PAGE 104

95 Staphylococcus species were identified from 41.8% of the bacterial colonies isolated from air samples during three beef slaughters. Staphylococcus aureus was found in 1.8% of the colonies isolated from air sa mples during three beef slaughters. Table 18. Airborne bacteria isol ated from mannitol salt agar plates collected from three beef slaughters Organism (Genus Species) Bacillus atrophaeus Staphylococcus gallinarum Enterobacter aerogenes Staphylococcus hominis Micrococcus spp. Staphylococcus hyicus Moraxella nonliquefaciens Staphylococcus kloosii Salmonella bongori Staphylococcus lentus Shigella flexneri Staphylococcus saprophyticus Staphylococcus aureus Staphylococcus schleiferi Staphylococcus capitis Staphylococcus sciuri Staphylococcus chromogenes Staphylococcus simulans Staphylococcus cohnii cohnii Staphylococcus xylosus Staphylococcus epidermidis Isolation of Listeria species Overall significant differences (P < 0.05) between slaughters (replications) were observed during three beef slaughters (Table 19 ). Slaughter one had significantly higher (P < 0.05) log CFU/m3 of air counts than sl aughter two, which had si milar counts to those of the last beef slaughter. During the fi rst and third beef sl aughter there were no significant differences (P > 0.05) in bacterial counts between individual sampling areas. Sampling areas F and G had significantly higher (P < 0.05) counts than areas E, H, A, B, C and D in the second slaughter. Average bact erial counts were similar (P > 0.05) for all sampling areas. The bacterial counts were le ss than three logs for all sampling areas (Table 19). The absence of differences in counts before and during slaughtering could be attributed to the beef slau ghtering process not significan tly increasing the amount of

PAGE 105

96 bacteria collectable on MOX medi a, or could be attributed to the large variability of the air sample counts. No Listeria species were isolated from air samples during this study. Microorganisms collected on MOX filled Petr i plates during three beef slaughters are presented in Table 20. Table 19. Mean airborne bacteria collected on modified oxford agar plates in eight areas during three beef slaughters Airborne bacteria (log CFU/m3 of air) Sampling Areaa Slaughter 1 Slaught er 2 Slaughter 3 Averaged SEMb A 0.77 0.00B c 1.79 0.85 0.47 B 1.85 0.00B 1.97 1.27 0.24 C 1.55 0.00B 2.15 1.23 0.00 D 2.39 0.00B 0.77 1.05 0.47 E 2.09 1.01B 2.45 1.85 0.59 F 1.20 2.83A 1.12 1.72 0.95 G 2.94 2.95A 1.90 2.59 0.24 H 2.74 0.77B 0.00 1.17 0.45 Average 1.94X 0.95Y 1.52XY 1.47 SEM 0.52 0.45 0.53 a A = bleeding pit; B = hide removal; C = back splitting; D = cooler; E = bleeding pit; F = hide removal; G = back splitting; H = c ooler after 24 hr., A-D are before slaughter and E-H are during slaughter, b Standard error of the mean, c Means within a column followed by different letters (A, B) and mean s within a row followed by different letters (X, Y) are significantly different (P < 0.05), d n = 6 observations. All of the airborne bacter ia collected on MOX media dur ing three beef slaughters were Gram positive. These Gram positive species are widely distributed in the environment and are associated with the c ontamination of meat and dairy products. Listeria species were not identified from any of the 171 bacterial colonies collected from air and carcass samples, although bacteria fr om the bacilli class were isolated. The presence of Bacillus Microbacterium and Micrococcus species in bioaerosols collected during three beef slaughters coul d be attributed to their pr esence in soil, mammalian skin and animal feces (Table 20). These bacteria could have entered the slaughtering facilities

PAGE 106

97 through contaminated hides of animals, feces, soil on workers’ clothing and shoes, equipment transport and human ca rriers. The non recovery of Listeria species from bioaerosols could be attribut ed to them becoming lethally injured by sample collection and/or environmental conditions, or subletha lly injured causing the VBNC state. Table 20. Airborne bacteria is olated from modified oxford agar plates collected from three beef slaughters Organism (Genus Species) Bacillus atrophaeus Microbacterium esteraromaticum Bacillus coagulans Microbacterium saperdae Bacillus pumilus Microbacterium schleiferi Kocuria kristinae Nesterenkonia halobia Microbacterium barkeri Paenibacillus popilliae Isolation of Escherichia coli Table 21. Mean airborne bacteria collected on vi olet red bile agar pl ates in eight areas during three beef slaughters Airborne bacteria (log CFU/m3 of air) Sampling Areaa Slaughter 1 Sla ughter 2 Slaught er 3 Averaged SEMb A 0.00 0.00B c 0.77 0.26 0.45 B 0.00 0.77B 0.00 0.26 0.45 C 0.00 0.00B 0.77 0.26 0.45 D 0.00 0.00B 0.00 0.00 0.00 E 0.77 0.00B 0.77 0.52 0.63 F 1.12 2.00A 0.00 1.04 0.66 G 2.03 0.00B 0.00 0.68 0.00 H 0.00 0.00B 0.00 0.00 0.00 Average 0.49 0.35 0.29 0.38 SEM 0.48 0.28 0.47 a A = bleeding pit; B = hide removal; C = back splitting; D = cooler; E = bleeding pit; F = hide removal; G = back splitting; H = c ooler after 24 hr., A-D are before slaughter and E-H are during slaughter, b Standard error of the mean, c Means within a column followed by different letters (A, B) are significantly different (P<0.05), d n = 6 observations. Airborne bacteria collected on VRBA filled Petri plates during three beef slaughters resulted in no significan t differences (P > 0.05) between slaughters (Table 21).

PAGE 107

98 Significant differences (P < 0.05) between i ndividual sampling areas were only observed during the second beef slaughter. Sampli ng area F had significantly higher (P < 0.05) counts than the rest of the areas. Average ba cterial counts were similar (P > 0.05) for all sampling areas. The bacterial counts were less than one log for all sampling areas except for area F (Table 21). The high variability of the counts obtained from the air samples could have been a reason why no differ ences were observed before and during slaughtering. Table 22. Airborne bacteria isolated from violet red bile agar plates collected from three beef slaughters Organism (Genus Species) Acinetobacter calcoaceticus Morganella morganii Acinetobacter haemolyticus Pantoea spp. Bacillus pumilus Pseudomonas aeruginosa Cedecea davisae Pseudomonas fluorescens Chryseomonas luteola Pseudomonas putida Citrobacter freundii Salmonella choleraesuis houtenae Enterobacter cloacae Salmonella typhi Escherichia coli Salmonella typhimurium Escherichia fergusonii Shigella flexneri Flavimonas oryzihabitans Shigella sonnei Kluyvera ascorbata Stenotrophomonas maltophilia Kluyvera cryocrescens Most of the airborne bacteria coll ected on VRBA media during three beef slaughters were Gram negative bacteria from the Enterobacteriaceae family (Table 22). Other Gram negative bacteria present, belonging to the Pseudomonadaceae family, were Pseudomonas spp., Chryseomonas luteola, Flavimonas oryzihabitans and Stenotrophomonas maltophilia (Table 22). These species are normally associated with soil and fecal mater and therefore are found on the hides of cattle. The only Gram positive bacterium isolated from VRBA plates was Bacillus pumilus which is frequently

PAGE 108

99 isolated from meat and dairy products (12) The Enterobacteriaceae and Pseudomonadaceae species found in the bioaerosols collected may have entered the slaughtering facility through the hides of animals, animal feces or the workers’ shoes. Generic Escherichia coli was identified from 7.3% of the 115 airborne bacterial colonies isolated during three beef slaughters. Escherichia coli O157:H7 was not isolated from any of the air samples taken from various sampling areas. Escherichia coli O157:H7 may not have been isolated from the air due to its low air borne concentration. Environmental factors like temperature and dehydration along with impaction stress created during aerosol collection may have cau sed lethal injury to this organism or induced the VBNC state, therefore contributing to the inability of isolating this organism from air samples (17) Isolation of Salmonella Species Airborne bacteria collected on XLT4 filled Petri plates during three beef slaughters yielded no significant differen ces (P > 0.05) between slaugh ters (Table 23). No Significant differences (P > 0.05) were observed between individual sampling areas within each beef slaughter. Average data re vealed significantly hi gher (P < 0.05) counts of airborne bacteria collected on XLT4 media in area E when compared to the rest of the areas where no growth was obtained. The bact erial counts were less than one log for sampling area E (Table 23). A ccording to Jericho et al. (21) the number of CFU of surviving target microorganisms like Salmonella typhimurium may be underestimated by up to six log units if enrichment steps are not used. Whyte et al. (52) did not isolate Salmonella species from any of the air samples obtained during the slaughtering process in three poultry plants.

PAGE 109

100 Table 23. Mean airborne bacteria collected on xy lose-lysine-tergitol 4 agar plates in eight areas during three beef slaughters Airborne bacteria (log CFU/m3 of air) Sampling Areaa Slaughter 1 Sla ughter 2 Slaught er 3 Averaged SEMb A 0.00 0.00 0.00 0.00B c 0.00 B 0.00 0.00 0.00 0.00B 0.00 C 0.00 0.00 0.00 0.00B 0.00 D 0.00 0.00 0.00 0.00B 0.00 E 1.55 0.77 0.00 0.77A 0.45 F 0.00 0.00 0.00 0.00B 0.00 G 0.00 0.00 0.00 0.00B 0.00 H 0.00 0.00 0.00 0.00B 0.00 Average 0.19 0.10 0.00 0.10 SEM 0.00 0.27 0.00 a A = bleeding pit; B = hide removal; C = back splitting; D = cooler; E = bleeding pit; F = hide removal; G = back splitting; H = c ooler after 24 hr., A-D are before slaughter and E-H are during slaughter, b Standard error of the mean, c Means within a column followed by different letters (A, B) ar e significantly different (P < 0.05), d n = 6 observations. Table 24. Airborne bacteria isol ated from xylose-lysine-terg itol 4 agar plates collected from beef slaughters Organism (Genus Species) Bacillus megaterium Escherichia coli Bacillus pumilus Kluyvera ascorbata Bacillus sphaericus Salmonella choleraesuis houtenae Cedecea neteri Salmonella typhimurium Enterobacter asburiae Shigella flexneri Enterobacter cancerogenus Shigella sonnei Enterobacter cloacae Most of the airborne bacteria collected on XLT4 media during th ree beef slaughters were Gram negative bacteria from the Enterobacteriaceae family (Table 24). Salmonella spp., Escherichia coli and Shigella spp. were readily isolated from both VRBA and XLT4 filled Petri plates. Salmonella species identified include Salmonella enterica subsp. enterica serotype typhi choleraesuis houtenae choleraesuis choleraesuis and typhimurium and Salmonella bongori (Tables 22 and 24). Environmental factors and

PAGE 110

101 sampling stress may have redu ced the ability to isolate Salmonella species from air samples (17) Salmonella species were identified from 5.5% of the total airborne bacterial colonies isolated during three beef slaughters. Identification of Airborne Bacteri a Collected from Beef Slaughters Table 25. Airborne bacteria isolated before and du ring three beef slaughters Organism (Genus Species) Before Slaughter During Slaughter Gram negative Gram negative Acinetobacter calcoaceticus Cedecea davisae Acinetobacter haemolyticus Chryseomonas luteola Chryseomonas luteola Enterobacter cloacae Escherichia coli Escherichia coli Flavimonas oryzihabitans Flavimonas oryzihabitans Pseudomonas aeruginosa Kluyvera cryocrescens Shigella flexneri Pantoea spp. Stenotrophomonas maltophilia Pseudomonas aeruginosa Gram positive Pseudomonas fluorescens Bacillus atrophaeus Pseudomonas putida Bacillus pumilus Salmonella choleraesuis houtenae Micrococcus spp. Salmonella typhi Nesterenkonia halobia Salmonella typhimurium Staphylococcus capitis Shigella flexneri Staphylococcus chromogenes Gram positive Staphylococcus epidermidis Bacillus atrophaeus Staphylococcus gallinarum Bacillus coagulans Staphylococcus hominis Bacillus pumilus Staphylococcus hyicus Bacillus sphaericus Staphylococcus sciuri Kocuria kristinae Staphylococcus xylosus Microbacterium barkeri Microbacterium esteraromaticum Microbacterium schleiferi Micrococcus spp. Nesterenkonia halobia Staphylococcus aureus Staphylococcus capitis Staphylococcus chromogenes Staphylococcus cohnii cohnii Staphylococcus epidermidis Staphylococcus hominis Staphylococcus hyicus Staphylococcus kloosii

PAGE 111

102 Table 25. Continued. Organism (Genus Species) During Slaughter Gram positive Staphylococcus lentus Staphylococcus saprophyticus Staphylococcus schleiferi Staphylococcus xylosus Most of the Gram negative airborne bacter ia isolated before the beef slaughtering process (A, B, C, and D) were non Enterobacteriaceae from the Pseudomonadaceae and Moraxellaceae family with the exception of Escherichia coli and Shigella flexneri Staphylococcus Micrococcus and Bacillus species were the main Gram positive airborne bacteria recovered before th e beef slaughters (Table 25). The majority of the Gram negative airborne bacteria isolated during the three beef slaughters (E, F, and G) were from the Enterobacteriaceae and Pseudomonadaceae family. Most of the Gram positive microorganisms isolated from the three beef slaughters were Staphylococcus Bacillus Microbacterium and Micrococcus species. These species have a wide environmental distribution and found in the soil and on mammalian skin (12, 19 ). A lot fewer microorganisms were isolated be fore the slaughters (A, B, C, D) when compared to the number of microorganisms isol ated during the slaughters (E, F, and G) (Table 25). The ongoing b eef slaughtering process coul d be the main reason why a greater number of microorganisms were isolated from areas E, F and G. Animals present on the slaughtering floor along with the sl aughtering processes may have greatly increased the number of airborne microorganisms. Most of the microorganisms that were isolat ed before the beef slaughters were also isolated during the slaughters. The presen ce of these airborne bacteria during the

PAGE 112

103 slaughter process could be attrib uted to their association with the cattle or to their general environmental presence. Acinetobacter spp., Stenotrophomonas maltophilia and Staphylococcus sciuri were only isolated before the beef slaughters. Organisms isolated before the beef slaughters could be regarded as environmental or background bacteria (Table 25). The least am ount of airborne bacteria was isolated from the 00C cooler (D and H). Staphylococcus sciuri and xylosus were isolated from th e empty cooler (D) and Escherichia coli Shigella flexneri and Micrococcus spp. were isolated from the cooler after carcasses had been stored for at least 24 hours at 00C (H). Carcass Swabs Collected During Beef Slaughters The microorganisms listed in Table 26 were isolated from swabs taken from beef carcasses which had been in a cooler at 00C for 24 hours. All of the Gram negative bacteria isolated from beef carcass swab s during three slaughters were from the Enterobacteriaceae family, except for Moraxella nonliquefaciens This spoilage microorganism grows on aerobically st ored refrigerated muscle foods (12) Similar amounts of Gram positive and Gram negative b acteria were isolated from beef carcass swabs. The majority of Gram positive bacter ia isolated from beef carcass swabs were Bacillus Microbacterium and Staphylococcus species. Of the 171 bacterial colonies isolated dur ing this study, a tota l of 56 were from carcass swabs. Staphylococcus species were identified fr om 37.1% of the bacterial colonies isolated from beef carcass swabs. Staphylococcus aureus was identified from 14% of the bacterial colonies is olated during beef carcass sampling. Escherichia coli and Salmonella species were identified from 17.1% and 8.6% of the bacterial colonies isolated from beef carcass swabs. Listeria species and Escherichia coli O157:H7 were not isolated from any of the beef car cass swabs. Barkoc y-Gallagher et al. (3) recovered

PAGE 113

104 E coli O157:H7 from 1.2% of postintervention beef carcasses. Carcass washing and acetic acid spraying, along with low temperatur e storage, may have inhibited the growth of these microorganisms on beef carcasses. Table 26. Bacteria isolated from beef carcasses held at 00C for 24 hours Organism (Genus Species) Gram negative Gram positive Cedecea neteri Bacillus megaterium Citrobacter freundii Bacillus pumilus Enterobacter aerogenes Kocuria kristinae Enterobacter asburiae Microbacterium esteraromaticum Enterobacter cancerogenus Microbacterium saperdae Enterobacter cloacae Paenibacillus popilliae Escherichia coli Staphylococcus aureus Escherichia fergusonii Staphylococcus capitis Kluyvera ascorbata Staphylococcus chromogenes Kluyvera cryocrescens Staphylococcus cohnii-cohnii Moraxella nonliquefaciens Staphylococcus epidermidis Morganella morganii Staphylococcus hominis Salmonella bongori Staphylococcus hyicus Salmonella typhimurium Staphylococcus lentus Shigella sonnei Staphylococcus saprophyticus Shigella flexneri Staphylococcus simulans Enterobacter cloacae Escherichia coli and fergusonii Kluyvera cryocrescens Salmonella typhimurium Shigella flexneri Bacillus pumilus Kocuria kristinae Microbacterium esteraromaticum Staphylococcus aureus epidermidis and other Staphylococcus species were isolated from both air samples and beef carcass swabs (Tables 25 and 26). None of the airborne bacteria isolat ed from the empty cooler (D) were isolated from the beef carcasses. Escherichia coli Shigella flexneri and Micrococcus spp. were isolated from air samples ta ken in the cooler containing carcasses (H) and from the carcasses themselves. The presence of Escherichia coli Shigella flexneri and Micrococcus spp. in the cooler after carcasse s had been stored for 24 hours

PAGE 114

105 and not in the empty cooler suggests that these organisms were introduced by the beef carcasses. The isolation of various organisms, including Staphylococcus Escherichia and Salmonella species, from both air and carcass swabs supports the theory that bioaerosols transport bacteria and contribute to the cont amination of beef carcasses. Comparison of Bioaerosols Collected from Pork and Beef Slaughters A significant (P < 0.05) interaction wa s observed between sampling area and species for airborne bacteria collected on TSA, PDA, MSA, MOX and XLT4. This interaction indicates that the comparison of different sampling areas was affected by the species. There were no significant differences (P < 0.05) in airborne bacteria collected on TSA, PDA, MSA, MOX, VRBA and XLT4 media when comparing sampling areas between species. Similar recovery rates for Staphylococcus Escherichia and Salmonella species were obtained through di rect air and enriched air mi crobiological sample analysis methods. Staphylococcus species were the main airborne b acteria isolated from MSA media during the slaughter of both species (Tables 6 and 18). Airborne Bacillus Microbacterium and Micrococcus species were isolated from MOX media during pork and beef slaughters (Tables 8, 20). Similar microorganisms from the Enterobacteriaceae family, isolated fr om VRBA and XLT4, were found in air samples collected from pork and beef slaughters (Tables 10, 12, 22 and 24). More airborne Enterobacteriaceae were isolated during the pork slaughter s than during the beef slaugh ters. Comparable types of airborne Gram positive bacteria and Pseudomonadaceae were isolated during the slaughtering of both species (Tables 13 and 25). Mostly Enterobacteriaceae were isolated from pork and beef carcasses. Far more Gram positive species were isolated

PAGE 115

106 from beef carcasses than from pork carcasses held in a cooler at 00C for 24 hours (Tables 14 and 26). The airborne and carcass bacteria identifie d during the pork and beef slaughters are presented in Appendix A. This raw data is presented throughout this paper and has been included in various tables. The data is orga nized by the slaughter number (1-6), type of selective media on which samples were colle cted (MSA, XLT4, VRBA, MOX), type of sample (direct or enriched air sample, carca ss samples), sampling area (A, B, C, D, E, F G, H), and similarity index. Slaughters one three and six were beef slaughters and slaughters two, four and five were pork sla ughters. Colonies from air samples were either identified directly (direct air samples) or they were enriched in selective media to increase their numbers prior to identificati on (enriched air samples). Carcass samples were taken using swabs and enriched before identification. Sample number refers to the number given to the specific colony isolated, and similarity index is the pr obability that the colony is a specific genus and species. Or ganisms with a similarity index of 1, 2 or 3 were identified using the API method, and organisms with a similarity index of 0.01 to 0.99 were identified using the MIDI system. A sample nu mber that appears only once has been identified as a single organism a nd those that appear more than once are accompanied by the list of possible organisms a nd their probability of being that specific organism. A similarity index of 1, 2 and 3 was used for organisms identified by the API method since no actual probability is given after identification. A similarity index value of 3 represents the highest proba bility that the sample colony is the specified organism. Samples identified by the MIDI system have a percent probability attached to each possible organism. The last column of th is table provides the API identification of

PAGE 116

107 samples that were also identified by the MIDI system. In the last column when more than one organism is given for a specific sample, then the first one listed is accompanied with the highest probability of being the correctly identified organism.

PAGE 117

108 CHAPTER 5 SUMMARY AND CONCLUSIONS The main objectives of this study were to enumerate total airb orne bacteria and yeast and mold contaminants and determine the presence of Escherichia coli O157:H7, Listeria spp., Salmonella spp., and Staphylococcus spp. in bioaerosols generated in a slaughtering facility and on pork and beef carcasses. Air sampling results revealed significant increases (P < 0.05) in TAB counts in the hide removal (F ) and back splitting (G) areas for pork and beef slaughters when compared to the same areas before slaughtering. Few differences in yeast and mold counts were observed between air samples taken before and during slaughtering. The overall higher c ounts observed during slaughtering (E, F, G) when compared to be fore slaughtering (A, B, C, D, H) were attributed to the presence of animal s on the slaughtering floor and the ongoing slaughtering process. Most of the slaught ering processes were associated with the creation of bioaerosols. Similar recovery rates for Staphylococcus Escherichia and Salmonella species were obtained th rough direct air and enri ched air microbiological sample analysis methods. The majority of the Gram negative airbor ne bacteria isolated during slaughtering from the bleeding pit (E), hide removal (F) and back splitting (G) areas were from the Enterobacteriaceae and Pseudomonadaceae family. Most of the Gram positive airborne bacteria isolated dur ing slaughtering were Staphylococcus Microbacterium Bacillus and Micrococcus species. The potentially pathogeni c microorganisms found in the air sampling areas and on the carcasses were Escherichia coli Salmonella spp., Shigella

PAGE 118

109 spp., Staphylococcus spp. and Bacillus spp. and the spoilage microorganisms were Moraxella spp., Pseudomonas spp., Acinetobacter spp., Brochothrix spp. and Micrococcus spp. The isolation of various organisms, including Staphylococcus Escherichia and Salmonella species, from both air samples and carcass swabs supported the theory that bioaerosols transport bacteria and contribute to the contamination of pork and beef carcasses. The specific levels of airborne bacteria a nd yeast and mold contaminants present in a small scale slaughter facility were determ ined in this study. The types of airborne bacteria present in this pork and beef slaughter facility were also determined. This study revealed that an effective use of air sampling in a small scale slaughter facility is to create guidelines that relate numbers and types of microorganisms per unit volume of air to acceptable levels of plant sanitation. These guidelines should be established for each individual processing plant to ascertain possible sources of product contamination, therefore, having the capacity to increase food safety. The realization that bioaerosols trans port bacteria and contribute to the contamination of pork and beef carcasses validat es the importance of controlling airborne contamination. The determination of the levels and types of airborne bacterial contaminants present in a small scale slaught er facility has various implications. The effectiveness of a plant’s sanitation program can be evaluated by collecting air samples after cleaning and sanitizing a nd comparing the bacterial coun ts to those obtained during slaughtering. Sources of contamination can also be determined a nd appropriate changes in the slaughter processes and plant layout can be made. Vari ations and changes in plant layout, air flow patterns, ventilation system s, personnel activity and animal density can

PAGE 119

110 be evaluated by comparing air borne bacterial counts obtaine d before changes to those obtained after changes. If th ere is a significant reduction in airborne contamination in the slaughter facility after change s, provided that these are ec onomically feasible, then these can be made to improve food safety. In this study the collection of bi oaerosols was accomplished by using the impaction method which is widely used in the food industry. Some di sadvantages of this method were its inability to recover organisms present at low levels in the air along with the creation of impaction stress during bioaer osol collection. This may have reduced the ability to isolate organisms like Listeria spp. and Escherichia coli O157:H7 from air samples. Methods employed in this st udy to enumerate, isolate and identify microorganisms from air samples were ade quate. However, highly selective media should not be used to isolate specific microorganisms from air samples. The isolation of specific microorganisms from air samples can be improved by using media with low selectivity combined with at le ast two enrichment steps. In this study, the use of MIDI to identify bacterial cells was a more accurate and rapid method than the biochemical assays (API) and the microbiological methods (p late count and selective media) used. Additional research in this area should be performed in a larger scale slaughtering facility. Air samples should be taken throughout the year to evaluate seasonal effects. A greater number of air samples (replications) must be taken at each sampling point to increase data reliability. Va rious selective media should be evaluated for the collection of specific microorganisms using the impacti on method. Further stud ies in this area could include the use of liquid impingement collection methods and the use of PCR and DNA or RNA probes for microorganism iden tification. These ot her collection and

PAGE 120

111 bacterial analysis methods could yield bette r recovery rates and increase levels of bacterial identification.

PAGE 121

APPENDIX ORGANISMS IDENTIFIED IN AIR AND CARCASS SAMPLES

PAGE 122

113Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 1 MSA 17 Direct Air B Staphylococcus epidermidis 1 1 MSA 19 Direct Air C Staphylococcus epidermidis 3 1 MSA 19 Direct Air C Staphylococcus hominis 2 1 MSA 19 Direct Air C Staphylococcus capitis 1 1 MSA 66 Enriched Air D Staphylococcus sciuri 2 1 MSA 66 Enriched Air D Staphylococcus xylosus 1 1 XLT4 14 Direct Air E Bacillus sphaericus 0.338 Escherichia coli 1 XLT4 14 Direct Air E Bacillus sphaericus 0.111 1 MSA 52 Enriched Air E Staphylococcus aureus 0.371 Staphylococcus aureus 1 VRBA 5 Direct Air F Pseudomonas putida 0.626 1 VRBA 5 Direct Air F Pseudomonas putida 0.572 1 MOX 11 Direct Air F Bacillus coagulans 0.590 1 MOX 11 Direct Air F Microbacterium schleiferi 0.448 1 MOX 12 Direct Air F Kocuria kristinae 0.719 1 MOX 12 Direct Air F Nesterenkonia halobia 0.656 1 MOX 12 Direct Air F Microbacterium barkeri 0.648 1 MSA 22 Direct Air F Staphylococcus cohnii cohnii 0.747 Staphylococcus chromogenes 1 MSA 22 Direct Air F Staphylococcus xylosus 0.529 Staphylococcus hyicus

PAGE 123

114Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 1 MSA 22 Direct Air F Staphylococcus epidermidis 0.505 1 VRBA 3 Direct Air G Flavimonas oryzihabitans 0.855 1 VRBA 3 Direct Air G Pseudomonas aeruginosa 0.841 1 VRBA 3 Direct Air G Chryseomonas luteola 0.520 1 MOX 78 Enriched Air G Microbacterium barkeri 0.713 1 MOX 78 Enriched Air G Kocuria kristinae 0.701 1 MOX 78 Enriched Air G Microbacterium esteraromaticum 0.663 1 MSA 24 Carcass S Staphylococcus aureus 1 1 MSA 26 Carcass S Staphylococcus aureus 1 1 MSA 27 Carcass S Staphylococcus aureus 2 1 MSA 27 Carcass S Staphylococcus hominis 1 1 MSA 29 Carcass S Staphylococcus chromogenes 3 1 MSA 29 Carcass S Staphylococcus simulans 2 1 MSA 29 Carcass S Staphylococcus aureus 1 1 MSA 30 Carcass S Staphylococcus aureus 0.380 Staphylococcus aureus 1 VRBA 31 Carcass S Enterobacter cloacae 1 1 VRBA 34 Carcass S Shigella sonnei 0.861 Escherichia coli 1 VRBA 34 Carcass S Escherichia coli 0.762 1 VRBA 34 Carcass S Morganella morganii 0.655 1 VRBA 35 Carcass S Shigella sonnei 0.794 Escherichia coli 1 VRBA 35 Carcass S Escherichia fergusonii 0.559 1 VRBA 35 Carcass S Salmonella typhimurium 0.543 1 MOX 37 Carcass S Microbacterium esteraromaticum 0.797 1 MOX 37 Carcass S Kocuria kristinae 0.773 1 MOX 37 Carcass S Microbacterium saperdae 0.710

PAGE 124

115Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 1 MOX 38 Carcass S Paenibacillus popilliae 0.112 1 XLT4 39 Carcass S Shigella flexneri 0.836 Escherichia coli 1 XLT4 39 Carcass S Shigella sonnei 0.681 1 XLT4 39 Carcass S Shigella sonnei 0.639 1 XLT4 41 Carcass S Shigella flexneri 0.854 Escherichia coli 1 XLT4 41 Carcass S Shigella sonnei 0.746 1 XLT4 41 Carcass S Shigella sonnei 0.692 2 MSA 83 Enriched Air A Staphylococcus capitis 3 2 MSA 83 Enriched Air A Staphylococcus warneri 2 2 MSA 83 Enriched Air A Staphylococcus hominis 1 2 MSA 84 Enriched Air A Staphylococcus epidermidis 1 2 MOX 82 Enriched Air B Shigella boydii 0.868 2 MOX 82 Enriched Air B Escherichia coli 0.800 2 MOX 82 Enriched Air B Salmonella bongori 0.750 2 VRBA 47 Direct Air E Escherichia coli 0.810 Enterobacter spp. 2 VRBA 47 Direct Air E Salmonella bongori 0.789 2 VRBA 47 Direct Air E Salmonella typhi 0.781 2 VRBA 48 Direct Air E Microbacterium barkeri 0.686 2 VRBA 48 Direct Air E Kocuria kristinae 0.642 2 VRBA 49 Direct Air E Enterobacter cloacae 1 2 MSA 71 Enriched Air E Staphylococcus sciuri 0.844 Staphylococcus-xylosus 2 MSA 71 Enriched Air E Staphylococcus gallinarum 0.507 2 MSA 45 Direct Air F Staphylococcus hyicus 0.737 Staphylococcus chromogenes 2 MSA 45 Direct Air F Staphylococcus hyicus 2 VRBA 50 Direct Air F Flavimonas oryzihabitans 1

PAGE 125

116Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 2 VRBA 68 Enriched Air F Cedecea neteri 0.623 Serratia rubidaea 2 VRBA 68 Enriched Air F Salmonella choleraesuis choleraesuis 0.591 2 VRBA 68 Enriched Air F Shigella boydii 0.591 2 MSA 72 Enriched Air F Staphylococcus kloosii 0.721 Staphylococcus cohnii cohnii 2 MSA 72 Enriched Air F Staphylococcus arlettae 0.687 Staphylococcus saprophyticus 2 MSA 72 Enriched Air F Staphylococcus sciuri 2 MSA 44 Direct Air G Staphylococcus chromogenes 1 2 VRBA 59 Carcass S Shigella sonnei 0.810 Escherichia coli 2 VRBA 59 Carcass S Shigella sonnei 0.764 2 VRBA 59 Carcass S Shigella flexneri 0.705 2 MSA 60 Carcass S Staphylococcus aureus 0.759 Staphylococcus chromogenes 2 MSA 60 Carcass S Staphylococcus aureus 0.702 Staphylococcus aureus 2 MSA 61 Carcass S Staphylococcus aureus 0.618 Staphylococcus aureus 2 MSA 61 Carcass S Staphylococcus warneri 0.419 Staphylococcus hominis 2 MSA 62 Carcass S Staphylococcus aureus 1 2 XLT4 80 Carcass S Enterobacter asburiae 0.797 Enterobacter cloacae 2 XLT4 80 Carcass S Enterobacter cancerogenus 0.760 2 XLT4 80 Carcass S Enterobacter cancerogenus 0.732 2 XLT4 81 Carcass S Staphylococcus epidermidis 0.834 2 XLT4 81 Carcass S Staphylococcus epidermidis 0.704 2 XLT4 81 Carcass S Staphylococcus aureus 0.539 3 MSA 93 Direct Air A Micrococcus spp. 1 3 VRBA 123 Enriched Air A Chryseomonas luteola 0.937 Pseudomonas aeruginosa 3 VRBA 123 Enriched Air A Flavimonas oryzihabitans 0.851 3 MSA 131 Enriched Air A Staphylococcus epidermidis 3

PAGE 126

117Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 3 MSA 131 Enriched Air A Staphylococcus hominis 2 3 MSA 131 Enriched Air A Staphylococcus capitis 1 3 MSA 132 Enriched Air A Staphylococcus epidermidis 1 3 MSA 133 Enriched Air B Staphylococcus hyicus 0.506 Staphylococcus hyicus 3 MSA 133 Enriched Air B Staphylococcus chromogenes 0.474 3 MSA 133 Enriched Air B Staphylococcus gallinarum 0.399 3 MOX 86 Direct Air E Kocuria kristinae 0.797 3 MOX 86 Direct Air E Microbacterium esteraromaticum 0.772 3 MSA 100 Direct Air E Staphylococcus kloosii 0.591 Staphylococcus saprophyticus 3 MSA 100 Direct Air E Staphylococcus cohnii cohnii 0.468 3 MSA 100 Direct Air E Bacillus atrophaeus 0.456 3 MSA 101 Direct Air E Staphylococcus chromogenes 2 3 MSA 101 Direct Air E Staphylococcus hyicus 1 3 MSA 136 Enriched Air E Staphylococcus schleiferi 1 3 VRBA 90 Direct Air F Flavimonas oryzihabitans 0.934 3 VRBA 90 Direct Air F Pseudomonas aeruginosa 0.824 3 VRBA 90 Direct Air F Chryseomonas luteola 0.634 3 MSA 95 Direct Air F Staphylococcus chromogenes 1 3 MOX 120 Enriched Air F Bacillus coagulans 0.103 3 VRBA 127 Enriched Air F Cedecea davisae 0.730 Pantoea spp. 3 VRBA 127 Enriched Air F Kluyvera cryocrescens 0.730 3 VRBA 127 Enriched Air F Salmonella typhi 0.730 3 VRBA 128 Enriched Air F Salmonella choleraesuis houtenae 0.850 Enterobacter cloacae 3 VRBA 128 Enriched Air F Salmonella typhimurium 0.839 3 VRBA 128 Enriched Air F Enterobacter cloacae 0.834

PAGE 127

118Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 3 XLT4 130 Enriched Air F Salmonella choleraesuis houtenae 0.844 Enterobacter cloacae 3 XLT4 130 Enriched Air F Enterobacter cloacae 0.833 3 XLT4 130 Enriched Air F Salmonella typhimurium 0.811 3 MSA 138 Enriched Air F Staphylococcus chromogenes 0.651 Staphylococcus chromogenes 3 MSA 138 Enriched Air F Staphylococcus cohnii cohnii 0.642 Staphylococcus hyicus 3 MSA 138 Enriched Air F Staphylococcus saprophyticus 0.580 3 MSA 139 Enriched Air F Staphylococcus chromogenes 2 3 MSA 139 Enriched Air F Staphylococcus hyicus 1 3 MSA 97 Direct Air G Staphylococcus cohnii cohnii 0.901 Staphylococcus chromogenes 3 MSA 97 Direct Air G Staphylococcus xylosus 0.574 Staphylococcus hyicus 3 MSA 141 Enriched Air G Staphylococcus epidermidis 3 3 MSA 141 Enriched Air G Staphylococcus capitis 2 3 MSA 141 Enriched Air G Staphylococcus hominis 1 3 MSA 103 Carcass S Staphylococcus epidermidis 1 3 MSA 104 Carcass S Staphylococcus sciuri 1 3 MSA 105 Carcass S Staphylococcus sciuri 0.903 Staphylococcus cohnii cohnii 3 MSA 105 Carcass S Staphylococcus sciuri 3 MSA 106 Carcass S Staphylococcus hyicus 1 3 MSA 112 Carcass S Staphylococcus saprophyticus 1 3 MOX 113 Carcass S Kocuria kristinae 0.815 3 MOX 113 Carcass S Microbacterium esteraromaticum 0.788 3 MOX 113 Carcass S Microbacterium saperdae 0.703 3 XLT4 115 Carcass S Enterobacter asburiae 0.836 Enterobacter cloacae 3 XLT4 115 Carcass S Enterobacter cancerogenus 0.836 3 XLT4 115 Carcass S Kluyvera ascorbata 0.677

PAGE 128

119Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 3 XLT4 116 Carcass S Bacillus pumilus 0.727 3 XLT4 116 Carcass S Bacillus megaterium 0.504 3 XLT4 117 Carcass S Shigella flexneri 0.770 Escherichia coli 3 XLT4 117 Carcass S Shigella sonnei 0.757 4 MSA 163 Enriched Air A Bacillus megaterium 0.906 4 VRBA B1 Enriched Air B Flavimonas oryzihabitans 1 4 MOX 156 Enriched Air E Brevibacterium casei 0.956 4 MOX 156 Enriched Air E Brevibacterium epidermidis 0.674 4 MSA E1 Enriched Air E Staphylococcus warneri 3 4 MSA E1 Enriched Air E Staphylococcus hominis 2 4 MSA E1 Enriched Air E Staphylococcus aureus 1 4 VRBA E2 Enriched Air E Enterobacter aerogenes 2 4 VRBA E2 Enriched Air E Klebsiella terrigena 1 4 MSA SR Direct Air E Staphylococcus sciuri 2 4 MSA SR Direct Air E Staphylococcus hominis 1 4 MSA SY Direct Air E Staphylococcus warneri 3 4 MSA SY Direct Air E Staphylococcus hominis 2 4 MSA SY Direct Air E Staphylococcus aureus 1 4 VRBA 154 Enriched Air F Escherichia coli 1 4 VRBA 154 Enriched Air F Klebsiella pneumoniae ozaenae 0.854 4 VRBA 154 Enriched Air F Escherichia coli 0.848 4 VRBA 154 Enriched Air F Salmonella typhi 0.840 4 MOX 157 Enriched Air F Bacillus pumilus 0.803 4 VRBA F1 Direct Air F Salmonella spp 2 4 XLT4 F1 Enriched Air F Pseudomonas aeruginosa 2

PAGE 129

120Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 4 VRBA F1 Direct Air F Escherichia coli 1 4 XLT4 F1 Enriched Air F Pseudomonas fluorescens 1 4 VRBA 155 Enriched Air G Klebsiella pneumoniae pneumoniae 0.944 4 VRBA 155 Enriched Air G Enterobacter aerogenes 0.864 4 VRBA 155 Enriched Air G Salmonella bongori 0.796 4 MOX 158 Enriched Air G Microbacterium barkeri 0.810 4 MOX 158 Enriched Air G Kocuria kristinae 0.706 4 MOX 158 Enriched Air G Nesterenkonia halobia 0.600 4 XLT4 161 Enriched Air G Escherichia fergusonii 0.850 Klebsiella ornithinolytica 4 XLT4 161 Enriched Air G Klebsiella pneumoniae pneumoniae 0.842 4 XLT4 161 Enriched Air G Enterobacter aerogenes 0.823 4 MSA 166 Enriched Air G Staphylococcus warneri 0.854 4 VRBA G1 Direct Air G Pseudomonas fluorescens 2 4 XLT4 G1 Enriched Air G Serratia marcescens 2 4 XLT4 G1 Enriched Air G Serratia liquefaciens 1 4 VRBA G1 Direct Air G Pseudomonas aeruginosa 1 4 XLT4 144 Carcass S Shigella sonnei 0.866 4 XLT4 144 Carcass S Morganella morganii 0.615 4 XLT4 144 Carcass S Shigella flexneri 0.588 4 MSA 146 Carcass S Staphylococcus aureus 0.762 4 MSA 147 Carcass S Staphylococcus aureus 0.818 4 MSA 147 Carcass S Staphylococcus aureus 0.617 4 MSA 148 Carcass S Staphylococcus chromogenes 2 4 MSA 148 Carcass S Staphylococcus hyicus 1 4 VRBA 149 Carcass S Leclercia adecarboxylata 0.878 Pantoea spp.

PAGE 130

121Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 4 VRBA 149 Carcass S Salmonella choleraesuis choleraesuis 0.822 Pantoea spp. 4 VRBA 149 Carcass S Enterobacter cloacae 0.726 4 VRBA 151 Carcass S Pseudomonas aeruginosa 0.788 Pseudomonas aeruginosa 4 VRBA 151 Carcass S Chryseomonas luteola 4 XLT4 4A Carcass S Escherichia coli 1 4 XLT4 4B Carcass S Pseudomonas aeruginosa 1 5 MSA 202 Enriched Air B Staphylococcus lugdunensis 0.886 Staphylococcus lugdunensis 5 MSA 202 Enriched Air B Staphylococcus chromogenes 5 MSA 202 Enriched Air B Staphylococcus hominis 5 MSA 203 Enriched Air C Staphylococcus epidermidis 1 5 MSA 204 Enriched Air D Staphylococcus auricularis 2 5 MSA 204 Enriched Air D Staphylococcus cohnii cohnii 1 5 MSA 5D1 Direct Air D Kocuria kristinae 2 5 MSA 5D1 Direct Air D Staphylococcus capitis 1 5 MSA 5D2 Direct Air D Kocuria kristinae 2 5 MSA 5D2 Direct Air D Staphylococcus capitis 1 5 MOX 167 Direct Air E Staphylococcus aureus 0.698 5 MOX 167 Direct Air E Staphylococcus haemolyticus 0.625 5 MOX 167 Direct Air E Staphylococcus epidermidis 0.546 5 MSA 173 Direct Air E Staphylococcus sciuri 1 5 MSA 174 Direct Air E Staphylococcus warneri 2 5 MSA 174 Direct Air E Staphylococcus hominis 1 5 VRBA 178 Direct Air E Serratia rubidaea 1 5 VRBA 179 Direct Air E Stenotrophomonas maltophilia 0.983 Kluyvera spp. 5 MOX 194 Enriched Air E Microbacterium barkeri 0.708

PAGE 131

122Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 5 XLT4 200 Enriched Air E Pseudomonas putida 0.715 5 XLT4 200 Enriched Air E Pseudomonas putida 0.596 5 XLT4 201 Enriched Air E Pseudomonas aeruginosa 0.927 5 MSA 205 Enriched Air E Staphylococcus hominis 2 5 MSA 205 Enriched Air E Staphylococcus sciuri 1 5 MSA 206 Enriched Air E Staphylococcus chromogenes 2 5 MSA 206 Enriched Air E Staphylococcus aureus 1 5 VRBA 212 Enriched Air E Pseudomonas putida 0.830 5 VRBA 212 Enriched Air E Pseudomonas putida 0.653 5 MOX 170 Direct Air F Brochothrix campestris 0.037 5 MOX 171 Direct Air F Kocuria kristinae 0.483 5 MOX 171 Direct Air F Microbacterium saperdae 0.444 5 MOX 171 Direct Air F Cellulomonas fimi 0.377 5 MSA 176 Direct Air F Staphylococcus aureus 0.752 Staphylococcus aureus 5 MSA 176 Direct Air F Staphylococcus aureus 0.604 5 MSA 176 Direct Air F Staphylococcus aureus 0.452 5 VRBA 180 Direct Air F Pseudomonas aeruginosa 0.806 Serratia liquefaciens 5 XLT4 181.5 Direct Air F Pseudomonas putida 0.343 5 MOX 196 Enriched Air F Staphylococcus haemolyticus 0.623 5 MOX 196 Enriched Air F Staphylococcus epidermidis 0.608 5 MOX 196 Enriched Air F Staphylococcus aureus 0.502 5 MSA 207 Enriched Air F Staphylococcus aureus 1 5 VRBA 215 Enriched Air F Pseudomonas putida 0.759 5 MSA 175 Direct Air G Staphylococcus chromogenes 2 5 MSA 175 Direct Air G Staphylococcus aureus 1

PAGE 132

123Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 5 MOX 197 Enriched Air G Microbacterium barkeri 0.759 5 MOX 197 Enriched Air G Kocuria kristinae 0.639 5 MOX 197 Enriched Air G Microbacterium saperdae 0.556 5 MOX 199 Enriched Air G Microbacterium saperdae 0.759 5 MOX 199 Enriched Air G Kocuria kristinae 0.742 5 MOX 199 Enriched Air G Microbacterium barkeri 0.652 5 MSA 209 Enriched Air G Staphylococcus epidermidis 0.697 5 MSA 209 Enriched Air G Staphylococcus epidermidis 0.645 Staphylococcus aureus 5 MSA 209 Enriched Air G Staphylococcus aureus 0.590 5 MSA 210 Enriched Air G Staphylococcus chromogenes 1 5 VRBA 217 Enriched Air G Pseudomonas putida 0.493 5 VRBA 217 Enriched Air G Pseudomonas putida 0.383 5 MSA 211 Enriched Air H Staphylococcus cohnii cohnii 2 5 MSA 211 Enriched Air H Staphylococcus sciuri 1 5 VRBA 183 Carcass S Escherichia fergusonii 0.883 5 VRBA 183 Carcass S Escherichia coli 0.822 5 VRBA 183 Carcass S Klebsiella pneumoniae pneumoniae 0.798 5 VRBA 184 Carcass S Chryseobacterium indologenes 1 5 VRBA 185 Carcass S Pseudomonas aeruginosa 0.705 5 MSA 188 Carcass S Staphylococcus xylosus 1 5 MSA 189 Carcass S Staphylococcus xylosus 1 5 MSA 190 Carcass S Staphylococcus aureus 0.734 Staphylococcus xylosus 5 MSA 190 Carcass S Staphylococcus aureus 0.682 5 MSA 190 Carcass S Staphylococcus aureus 0.505 5 MSA 191 Carcass S Staphylococcus epidermidis 2

PAGE 133

124Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 5 MSA 191 Carcass S Staphylococcus hominis 1 6 MSA 231 Direct Air A Micrococcus spp. 1 6 MSA 267 Enriched Air A Micrococcus spp. 1 6 MOX 222 Direct Air B Bacillus atrophaeus 0.442 6 MOX 222 Direct Air B Nesterenkonia halobia 0.438 6 MSA 232 Direct Air B Staphylococcus xylosus 1 6 MSA 269 Enriched Air B Staphylococcus xylosus 1 6 VRBA 219 Direct Air C Acinetobacter haemolyticus 0.744 Stenotrophomonas maltophilia 6 VRBA 219 Direct Air C Acinetobacter calcoaceticus 0.737 6 VRBA 279 Enriched Air C Bacillus pumilus 0.839 Escherichia coli 6 VRBA 279 Enriched Air C Escherichia coli 6 VRBA 220 Direct Air E Flavimonas oryzihabitans 1 6 MSA 235 Direct Air E Staphylococcus lentus 1 6 MSA 236 Direct Air E Staphylococcus lentus 0.583 Staphylococcus hyicus 6 MSA 236 Direct Air E Staphylococcus hyicus 0.527 6 MSA 273 Enriched Air E Micrococcus spp. 1 6 MSA 274 Enriched Air E Micrococcus spp. 1 6 MSA 237 Direct Air F Staphylococcus epidermidis 1 6 MOX 264 Enriched Air F Bacillus pumilus 0.777 6 MOX 265 Enriched Air F Bacillus pumilus 0.815 6 MSA 275 Enriched Air F Shigella flexneri 0.503 Micrococcus spp. 6 VRBA 280 Enriched Air F Chryseomonas luteola 3 6 VRBA 280 Enriched Air F Pseudomonas aeruginosa 2 6 VRBA 280 Enriched Air F Pseudomonas fluorescens 1 6 VRBA 281 Enriched Air F Chryseomonas luteola 3

PAGE 134

125Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 6 VRBA 281 Enriched Air F Pseudomonas aeruginosa 2 6 VRBA 281 Enriched Air F Pseudomonas fluorescens 1 6 MOX 230 Direct Air G Bacillus pumilus 0.836 6 MSA 238 Direct Air G Staphylococcus xylosus 1 6 MSA 276 Enriched Air G Micrococcus spp. 1 6 VRBA 282 Enriched Air G Chryseomonas luteola 3 6 VRBA 282 Enriched Air G Pseudomonas aeruginosa 2 6 VRBA 282 Enriched Air G Pseudomonas fluorescens 1 6 XLT4 260 Enriched Air H Escherichia coli 1 6 XLT4 261 Enriched Air H Shigella flexneri 0.513 Escherichia coli 6 MSA 277 Enriched Air H Micrococcus spp. 1 6 MSA 241 Carcass S Staphylococcus capitis 1 6 MSA 243 Carcass S Moraxella nonliquefaciens 0.919 6 MSA 243 Carcass S Salmonella bongori 0.873 6 MSA 243 Carcass S Enterobacter aerogenes 0.867 6 MSA 245 Carcass S Staphylococcus lentus 1 6 VRBA 247 Carcass S Kluyvera cryocrescens 0.929 Citrobacter freundii 6 VRBA 247 Carcass S Kluyvera ascorbata 0.800 6 VRBA 247 Carcass S Salmonella typhimurium 0.754 6 VRBA 248 Carcass S Citrobacter freundii 1 6 VRBA 249 Carcass S Citrobacter freundii 1 6 VRBA 250 Carcass S Shigella flexneri 0.628 Enterobacter cloacae 6 VRBA 250 Carcass S Shigella sonnei 0.527 6 VRBA 251 Carcass S Enterobacter cloacae 1 6 VRBA 252 Carcass S Enterobacter cloacae 1

PAGE 135

126Slaughter # Media Sample # Sample type Area Organism (genus species) API and MIDI ID Sim. index Organism (genus species) API ID 6 XLT4 254 Carcass S Enterobacter cloacae 1 6 XLT4 255 Carcass S Enterobacter asburiae 0.787 Enterobacter cloacae 6 XLT4 255 Carcass S Enterobacter cancerogenus 0.704 6 XLT4 255 Carcass S Cedecea neteri 0.703 6 XLT4 257 Carcass S Escherichia coli 1 6 XLT4 258 Carcass S Enterobacter cloacae 1 6 XLT4 259 Carcass S Enterobacter cloacae 1

PAGE 136

127 LIST OF REFERENCES 1. Al-Dagal, M., and D. Y. C. F ung. 1990. Aeromicrobiology: a review. Crit Rev Food Sci Nutr 29:333-340. 2. Autio, T., T. Steri, M. Fredriksson-Ahomaa, M. Rahkio, J. Lundn, and H. Korkeala. 2000. Listeria monocytogenes contamination pattern in pig slaughterhouses. J Food Prot 63:1438-1442. 3. Barkocy-Gallagher, G. A., T. M. Arthur, M. Rivera-Betancourt, X. Nou, S. D. Shackelford, T. L. Wheeler, and M. Koohmaraie. 2003. Seasonal prevalence of shiga toxin-producing Escherichia coli including O157:H7 and non-O157 serotypes, and Salmonella in commercial beef processing plants. J Food Prot 66:1978-1986. 4. bioMrieux. 2002. API 20E, API Staph, and API Listeria identification manuals. bioMrieux Inc., Marcy l'Etoile, France. 5. Bitton, G. 2002. Encyclopedia of environm ental microbiology. Wiley Interscience Publication, New York, NY. 6. Butera, M., J. H. Smith, W. D. Morrison, R. R. Hacker, F. A. Kains, and J. R. Ogilvie. 1991. Concentration of respirable dust and bio aerosols and identification of certain microbial types in a hog growing facility. Can J Anim Sci 71:271-277. 7. Chasseignaux, E., P. Grault, M. T. Toquin, G. Salvat, P. Colin, G. Ermel. 2002. Ecology of Listeria monocytogenes in the environment of raw poultry meat and raw pork meat processing plants. FEMS Microbiol Lett 210:271-275. 8. Crane, J. K. 1999. Preformed bacterial toxins. Clin Lab Med 19:583-599. 9. Desmarchelier, P. M., G. M. Higgs, L. Mills, A. M. Sullivan, and P. B. Vanderlinde. 1999. Incidence of coagulase positive Staphylococcus on beef carcasses in three Australian abattoirs. Int J Food microbiol 47:221-229. 10. Difco. 1986. Serological identification of Escherichia coli : Bacto E coli O antiserum O157 and Bacto E coli H antiserum H7. Difco Inc., Detroit, MI. 11. Downes, F. P., and K. Ito (4th ed.). 2001. Compendium of methods for the microbiological examination of foods. Ameri can Public Health Association Press, Washington, D.C.

PAGE 137

128 12. Doyle, M. P., L. R. Beuchat, and T. J. Montville (2nd ed.). 2001. Food microbiology: fundamentals and frontie rs. ASM press, Washington, D.C. 13. Ellerbroek, L. 1997. Airborne microflora in poultry slaughtering establishments. Food microbiol 14:527-531. 14. Forbes, B. A., D. F. Sahm, and A. S. Weissfeld (11th ed.). 2002. Bailey & Scott’s diagnostic microbiology. Mosby Press, St. Louis, MO. 15. Gorman, R., and C. C. Adley. 2004. Characterization of Salmonella enterica serotype typhimurium isolates from human, food, and animal sources in the Republic of Ireland. J Clin Microbiol 42: 2314-2316. 16. Hansson, I. B. 2001. Microbiological m eat quality in high and low capacity slaughterhouses in Sweden. J Food Prot 64:820-825. 17. Heidelberg, J. F., M. Shahamat, M. Levin, I. Rahman, G. Stelma C. Grim, and R. R. Colwell. 1997. Effect of aerosolization on culturability and viability of gramnegative bacteria. Appl Environ Microbiol 63:3585-3588. 18. Heldman, D. R. 1974. Factors influencing airborne contamination of foods: a review. J Food Sci 39:962-969. 19. Holt, J. G., N. R. Krieg, P. H. A. Sneath, J. T. Staley, and S. T. Williams (9th ed.). 1994. Bergey’s manual of determinative bacteriology. Lippincott Williams & Wilkins, Baltimore, MD. 20. Janda, J. M., and S. L. Abbott. 1998. Th e enterobacteria. Lippincott-Raven Publishers, Philadelphia, PA. 21. Jericho, K. W. F., J. Ho, and G. C. Ko zub. 2000. Aerobiology of a high-line speed cattle abattoir. J Food Prot 63:1523-1528. 22. Kang, Y. J., and J. F. Frank. 1989. Biologi cal aerosols: a revi ew of airborne contamination and its measurement in dairy processing plants. J Food Prot 52: 512-524. 23. Kang, Y. J., and J. F. Frank. 1990. Characteri stics of biological aerosols in dairy processing plants. J Dairy Sci 73: 621-626. 24. Karunasager, I., B. Senghaas, G. Krohne, W. Goebel. 1994. Ultrastructural study of Listeria monocytogenes entry into cultured human colonic epithelial cells. Infect Immun 62:3554-3558. 25. Kotula, A. W., and B. S. Emswiler-Rose. 1988. Airborne microorganisms in a pork processing establishment. J Food Prot 51:935-937.

PAGE 138

129 26. Krieg, N. R., J. G. Holt. 1984 Bergey’s manual of systematic bacteriology. Williams & Wilkins Press, Baltimore, MD. 27. Lutgring, K. R., R. H. Linton, N. J. Zimmerman, M. Peugh, and A. J. Heber. 1997. Distribution and quantifica tion of bioaerosols in p oultry slaughtering plants. J Food Prot 60: 804-810. 28. MacFaddin, J. F. 2000. Biochemical tests fo r identification of medical bacteria. Lippincott Williams & Wilkins, Baltimore Press, MD. 29. McEvoy, J. M., A. M. Doherty, J. J. Sher idan, F. M. Thomson-Carter, P. Garvey, L. McGuire, I. S. Blair, and D. A. McDowell. 2003. The prevalence of Escherichia coli O157:H7 at a commercial beef abattoir. J Appl microbiol 95:256-266. 30. Mead, P. S., L. Slutsker, V. Dietz, L. F. McCaig, J. S. Bresee, C. Shapiro, P. M. Griffin, and R. V. Tauxe. 1999. Food-related il lness and death in the United States. Emerg Inf Dis 5:607-625. 31. Mylonakis, E., M. Paliou, E. L. Hohm ann, S. B. Calderwood, E. J. Wing. 2002. Listeriosis during pregnancy, a cas e series review of 222 cases. Medicine 81:26069. 32. Nataro, J. P., and J. B. Kaper. 1998. Diarrheagenic Escherichia coli Clin Microbiol Rev 11:142-201. 33. Neogen. 2003. Reveal microbial screening test: E coli O157:H7 test system. Neogen Corp., Lansing, MI. 34. Rahkio, T. M., and H. J. Korkeala. 1997. Airborne bact eria and carcass contamination in slaughterhouses. J Food Prot 60:38-42. 35. Ren, T. J. 1991. Microbial analysis of ai r, environmental surfaces, and retail packages in ice cream and fluid milk plan ts. Ph.D. Dissertation, Univ. of Georgia, Athens, GA. 36. Ren, T. J., and J. F. Frank. 1992. A survey of four fluid milk processing plants for airborne contamination usi ng various sampling methods. J Food Prot 55:38-42. 37. Ren, T. J., and J. F. Frank. 1992. Measurem ent of airborne contamination in two commercial ice cream plants. J Food Prot 55:43-47. 38. Ren, T. J., and J. F. Frank. 1992. Sampli ng of microbial aerosols at various locations in fluid milk and ice cream plants. J Food Prot 55: 279-283.

PAGE 139

130 39. Rivera-Betancourt, M., S. D. Shackelfor d, T. M. Arthur, K. E. Westmoreland, G. Bellinger, M. Rossman, J. O. Reagan, and M. Koohmaraie. 2004. Prevalence of Escherichia coli O157:H7, Listeria monocytogenes and Salmonella in two geographically distinct commercial beef processing plants in the United States. J Food Prot 67:295-302. 40. SAS. 1998. SAS user’s guide: statistics, ve rsion 8. SAS Institute Inc., Cary, NC. 41. Sasser, M., 2001. Identification of bacteria by gas chromatography of cellular fatty acids. Technical note # 101. Av ailable at: http://www.midiinc.com/pages/literature.h tml. Accessed 4 April 2004. 42. Savell, J. W., and G. C. Smith (7th ed.). 2000. Meat science laboratory manual. American Press, Boston, MA. 43. Schaechter, M., N. C. Engleberg, B. I. Eisenstein, and G. Medoff (3rd ed.). 1999. Mechanisms of microbial disease. Lippinc ott Williams & Wilkins, Baltimore Press, MD. 44. Sneath, P. H. A., N. S. Mair, M. E. Sh arpe, and J. G. Holt. 1986 Bergey’s manual of systematic bacteriology. Williams & Wilkins Press, Baltimore, MD. 45. Stewart, S. L., S. A. Grinshpun, K. Willeke, S. Terzieva, V. Ulevicius, and J. Donnelly. 1995. Effect of impact stress on microbial recovery on an agar surface. Appl Environ Microbiol 61:1232-1239. 46. Thermo Andersen. 2001. Operator manual: single stage / N6 microbial sampler. Thermo Andersen Corp., Smyrna, GA. 47. Tortora, G. J., B. R. Funke, and C. L. Case (6th ed.). 1998. Microbiology: an introduction. Addison Wesley L ongman Press, Menlo Park, CA. 48. U.S. Department of Agriculture. 1996. Ru les and regulations. USDA, Food Safety and Inspection Service. Fe deral Register 61: 13917-13944. 49. U.S. Department of Agriculture. 1998. Microbiology La boratory Guidebook. Available at: http://www.fsis.us da.gov/ophs/microlab/mlgbook.htm#change. Accessed May 14, 2004. 50. U.S. Food & Drug Administration. 2001. Th e Bacteriological Analytical Manual Online. Available at: http://www.cfsan.fda.gov/~ebam/bam-toc.html. Accessed 14 May 2004. 51. Vazques-Boland, J.A., M. Kuhn, P. Berche, T. Chakraborty, G. Domingues-Bernal, W Goebel, B. Gonzales-Zorn, J. Wehland, and J. Kreft. 2001. Listeria pathogenesis and molecular virulence determinants. Clin Microbiol Rev 14:584-640.

PAGE 140

131 52. Whyte, P., J. D. Collins, K. Mc Grill, C. Monahan, and H. O’Mahony. 2001. Distribution and prevalence of airborne microorganisms in three commercial poultry processing plants. J Food Prot 64:388-391. 53. Wilson, S.C., J. Morrow-Tesch, D. C. Stra us, J. D. Cooley, W. C. Wong, F. M. Mitlhner, J. J. McGlone. 2002. Airborne microbial flora in a cattle feedlot. Appl Environ Microbiol 68:3238-3242. 54. Worfel, R. C., J. N. Sofos, G. C. Smith, and G. R. Schmidt. 1996. Airborne bacterial contamination in beef slaughteri ng-dressing plants with different layouts. Dairy Food Environ Sanit 16:440-433.

PAGE 141

132 BIOGRAPHICAL SKETCH Gabriel Humberto Cosenza Sutton was bor n in Basil, Switzerland, on November 27, 1975, and has lived in Tegucigalpa, Honduras since the age of three. In 1997, he graduated from the Escuela Agricola Panamericana, in Zamorano, Honduras. In 1999, he received his Bachelor of Science degree with honors from the Department of Food Science and Human Nutrition, University of Flor ida. In 2001, he received his Master of Science degree from the Department of Anim al Sciences. In 2001, he received an IFAS graduate fellowship for doctoral studies in the Department of Animal Sciences. He will earn his Doctor of Philosophy degree in A ugust 2004. Upon graduation, Gabriel plans to return to Honduras and work for a family owned biotechnology company which works closely with various American companies.


Permanent Link: http://ufdc.ufl.edu/UFE0006613/00001

Material Information

Title: Enumeration of Total Airborne Bacteria, Yeast and Mold Contaminants and Identification of Escherichia coli O157:H7, Listeria spp., Salmonella spp., and Staphylococcus spp. in a Beef and Pork Slaughter Facility
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0006613:00001

Permanent Link: http://ufdc.ufl.edu/UFE0006613/00001

Material Information

Title: Enumeration of Total Airborne Bacteria, Yeast and Mold Contaminants and Identification of Escherichia coli O157:H7, Listeria spp., Salmonella spp., and Staphylococcus spp. in a Beef and Pork Slaughter Facility
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0006613:00001


This item has the following downloads:


Full Text












ENUMERATION OF TOTAL AIRBORNE BACTERIA, YEAST AND MOLD
CONTAMINANTS AND IDENTIFICATION OF Escherichia coli 0157:H7, Listeria
Spp., Salmonella Spp., AND Staphylococcus Spp. IN A BEEF AND PORK
SLAUGHTER FACILITY















By

GABRIEL HUMBERTO COSENZA SUTTON


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2004















ACKNOWLEDGMENTS

The author is sincerely grateful to Dr. S. K. Williams, associate professor and

supervisory committee chairperson, for her superb guidance and supervision in

conducting this study and manuscript preparation. He also extends his gratitude to the

other committee members, Dr. Dwain Johnson, Dr. Ronald H. Schmidt, Dr. David P.

Chynoweth and Dr. Murat 0. Balaban, for their collaboration and useful

recommendations during this study. Special thanks are extended to the Institute of Food

and Agricultural Sciences of the University of Florida for sponsoring the author

throughout the program. Great appreciation and love are expressed to Adrienne and

Humberto Cosenza, the author's parents, and to his girlfriend Candice L. Lloyd for their

support throughout the graduate program. The author wishes to thank Larry Eubanks,

Byron Davis, Tommy Estevez, Doris Sartain, Frank Robbins, Noufoh Djeri and fellow

graduate students for their friendly support and assistance. Overall the author would like

to thank God for His inspiration, love and caring.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S .................................................................................................. ii

TABLE OF CONTENTS.............................................. iii

LIST OF TABLES.... ......... .. .... ....... ..........................vi

ABSTRAC T ............ ........... .............. ...................... viii

CHAPTER

1 INTRODUCTION ................... ............................ .......... .. .......... 1

2 LITERATURE REVIEW .................................................. ...............5

Bioaerosols ............................ .. ............ ........ 5
Bioaerosol Sampling Methods............................... ...............7
Sedim entation ....................................................................... .......8
Im action ........................................................................................9
Impingement......................................................... ... ....... .........10
Filtration ...................................... .................. ........ ....... ...............1
Centrifugation...................................... ........................................12
Electrostatic Precipitation................................ .......................... ... ....12
Thermal Precipitation .............................. ...............13
Beef and Pork Slaughter Process............................................ ........13
Stunning or Immobilization ............... ................ ...........14
Bleeding.........................................14
Scalding/D ehairing ............... ............................. ......... ...... ... .............. 14
Evisceration ................................................................ ... ... ........ 15
Splitting and Chilling .................. .... .... .. ................ ..............15
Bioaerosols in the M eat IndustryP.......................................................... .. ... 16
Bioaerosols in the D airy Industry ............................................................... 20
Microbiological Analysis of Bioaerosols ........................................23
Environmental and Beef and Pork Associated Bacteria...........................................27
Gram Negative Bacteria .................................................. ........29
Enterobacteriaceae.............. ......... ... ................ .... .............. .....29
Escherichia coli .................... ...................... ....................30
Salmonella spp.............................................34
e// spp. ........................................................................ 37










Klebsiella spp. ........................... .....................38
Serratia spp. ................................................................38
Enterobacter spp. ................................ .. ............. ........ 39
Citrobacter spp...................................................... ........39
Pantoea spp. .............................................40
Morganella spp.................................................40
Kluyvera spp. ........................... ......................41
Cedecea spp.............. ..... ......... ................41
Leclercia spp. ..................................................... ........ 41
Non Enterobacteriaceae .................. ......... .......... ........42
Acinetobacter spp. .............................................................42
Moraxella spp. ...................................................... ........42
Chryseomonas spp...................................... ........................... .........42
Flavimonas spp.............................................43
Pseudomonas spp. ................................................... ........43
Stenotrophomonas spp................................................43
Chryseobacterium indologenes ............................... ........44
Gram Positive Bacteria............................................... .............. ......44
Listeria spp. .............................................44
Staphylococcus spp....................................... ........48
Bacillus spp. ................................................................51
Brevibacterium spp........................................ ........52
Cellulomonas spp. .................................................. ........ 52
Brochothrix spp. .................................................52
M icrobacterium spp............................................... ............... 53
Micrococcus spp .............................. ..............53

3 MATERIALS AND METHODS ..................................................... 55

Air Sampling System ............................ .......... ...................... 55
Air Sample and Carcass Swab Collection .............. ......... ...................... 57
Preliminary Studies ...........................................................57
Carcass Sample Collection .......................................58
Air and Carcass Microbiological Sample Analysis........................ ..........59
Bacterial Identification of Air and Carcass Samples................................................62
Microbiological Identification ....... ..... .......................63
Statistical Analyses ......................................................68

4 RESULTS AND DISCUSSION ...................... ...........69

Preliminary Work: Air Sampling Time Determination ....... ..... ...... .........69
Microbiology of Bioaerosols Collected During Pork Slaughters..............................70
Total Airborne Bacteria.................................. ..70
Total Airborne Yeast and M old...................................................... ............. 72
Isolation of Staphylococcus species ................................... .................75
Isolation of Listeria species......................................... ......... 77
Isolation of Escherichia coli................................................... 80


iv










Isolation of Salmonella species .................. ............................83
Identification of Airborne Bacteria Collected from Pork Slaughters..................84
Carcass Swabs Collected During Pork Slaughters............................................87
Microbiology of Bioaerosols Collected During Beef Slaughters..............................89
Total A airborne B acteria................................................... 89
Total A airborne Y east and M old...................................................... .............91
Isolation of Staphylococcus species .................................... ...............93
Isolation of Listeria species.................................... ............... 95
Isolation of Escherichia coli.................................... .................. 97
Isolation of Salmonella Species.................................99
Identification of Airborne Bacteria Collected from Beef Slaughters ..............101
Carcass Swabs Collected During Beef Slaughters....................103
Comparison of Bioaerosols Collected from Pork and Beef Slaughters....................105

5 SUM M ARY AND CONCLUSIONS ............................................ ............... 108

APPENDIX ORGANISMS IDENTIFIED IN AIR AND CARCASS SAMPLES......112

L IST O F R E F E R E N C E S ........................................................................................... 127

BIOGRAPHICAL SKETCH .............................................................. ...............132
















LIST OF TABLES


Table page

1 Air sampling areas for beef and pork slaughtering processes............... ...............55

2 Sampling time determination for air sample collection during a beef slaughter......69

3 Mean total airborne bacteria collected on tryptic soy agar plates in eight areas
during three pork slaughters .............................................................. .............72

4 Mean total airborne yeast and mold collected on potato dextrose agar plates in
eight areas during three pork slaughters................................................. ...... 74

5 Mean airborne bacteria collected on mannitol salt agar plates in eight areas
during three pork slaughters ............... ............... ...............76

6 Airborne bacteria isolated from mannitol salt agar plates collected from three
pork slaughters ....................................................77

7 Mean airborne bacteria collected on modified oxford agar plates in eight areas
during three pork slaughters ............... ............... ...............78

8 Airborne bacteria isolated from modified oxford agar plates collected from three
pork slaughters ....................................................79

9 Mean airborne bacteria collected on violet red bile agar plates in eight areas
during three pork slaughters ............... ............... ...............81

10 Airborne bacteria isolated from violet red bile agar plates collected from three
pork slaughters ....................................................82

11 Mean airborne bacteria collected on xylose-lysine-tergitol 4 agar plates in eight
areas during three pork slaughters................ ....................... 83

12 Airborne bacteria isolated from xylose-lysine-tergitol 4 agar plates collected
from three pork slaughters.................................... ......... 84

13 Airborne bacteria isolated before and during three pork slaughters .....................86

14 Bacteria isolated from pork carcasses held at 00C for 24 hours ..............................88









15 Mean total airborne bacteria collected on tryptic soy agar plates in eight areas
during three beef slaughters ................. ............ ...................90

16 Mean total airborne yeast and mold collected on potato dextrose agar plates in
eight areas during three beef slaughters ...................................... ........... ....92

17 Mean airborne bacteria collected on mannitol salt agar plates in eight areas
during three beef slaughters ................. ................................94

18 Airborne bacteria isolated from mannitol salt agar plates collected from three
beef slaughters ............... ................... .............. .95

19 Mean airborne bacteria collected on modified oxford agar plates in eight areas
during three beef slaughters ................. ................................96

20 Airborne bacteria isolated from modified oxford agar plates collected from three
beef slaughters ...................................... ................................ ......... 97

21 Mean airborne bacteria collected on violet red bile agar plates in eight areas
during three beef slaughters ................. ............ ...................97

22 Airborne bacteria isolated from violet red bile agar plates collected from three
beef slaughters ...................................... ................................ ......... 98

23 Mean airborne bacteria collected on xylose-lysine-tergitol 4 agar plates in eight
areas during three beef slaughters .......................... ......... .......... 100

24 Airborne bacteria isolated from xylose-lysine-tergitol 4 agar plates collected
from beef slaughters .......................... .. ............. .... ............ 100

25 Airborne bacteria isolated before and during three beef slaughters ..................101

26 Bacteria isolated from beef carcasses held at 00C for 24 hours ............................104















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

ENUMERATION OF TOTAL AIRBORNE BACTERIA, YEAST AND MOLD
CONTAMINANTS AND IDENTIFICATION OF Escherichia coli 0157:H7, Listeria
Spp., Salmonella Spp., AND Staphylococcus Spp. IN A BEEF AND PORK
SLAUGHTER FACILITY

By

Gabriel Humberto Cosenza Sutton

August 2004

Chair: Sally K. Williams
Major Department: Animal Sciences

Environmental air monitoring programs can be employed to reduce unsanitary

conditions in animal slaughterhouses due to suspended bacterial particles in the air. The

main objectives of this study were to enumerate total airborne bacteria and yeast and

mold contaminants and determine the presence of Escherichia coli 0157:H7, Listeria

spp., Salmonella spp., and Staphylococcus spp. in bioaerosols generated in a slaughter

facility and on pork and beef carcasses. Air samples were taken before and during three

separate pork and beef slaughter processes at the bleeding area, hide removal or dehairing

area, back splitting area and holding cooler using an Andersen N6 single stage impactor.

Pork and beef carcass surface bacterial swabs were collected from five different carcass

sides which had been held in the holding cooler at 00C for 12 hours.

Total airborne bacterial (TAB) counts (log CFU/m3 of air) were generally higher

during slaughtering than before slaughtering. TAB counts were greater than three logs









during slaughtering and less than three logs before slaughtering. The holding cooler had

TAB counts less than or equal to two logs. Similar recovery rates for Staphylococcus,

Escherichia and Salmonella species were obtained through direct air and enriched air

microbiological sample analysis methods. Most of the Gram negative airborne bacteria

isolated during slaughtering were from the Enterobacteriaceae and Pseudomonadaceae

family. The predominant Gram positive airborne bacteria isolated during slaughtering

were Staphylococcus, Microbacterium, Bacillus and Micrococcus species. Potentially

pathogenic Staphylococcus aureus, Escherichia coli and Salmonella spp. were isolated

from bioaerosols generated during slaughtering and from pork and beef carcasses.

Neither Listeria spp. nor Escherichia coli 0157:H7 were isolated from air samples or

pork and beef carcasses.

The isolation of various microorganisms, including Staphylococcus, Escherichia,

and Salmonella spp., from air samples and carcass swabs support the theory that

bioaerosols transport bacteria and contribute to contamination of pork and beef carcasses.

The determination of the levels and types of airborne bacterial contaminants present in a

small scale slaughter facility has various implications. The effectiveness of a plant's

sanitation program can be evaluated and the sources of airborne contamination can be

determined allowing for increased food safety.














CHAPTER 1
INTRODUCTION

In the past it was thought that food products were contaminated when they came in

contact with contaminated surfaces, but now it is known that additional product

contamination occurs from contact with airborne bacteria. Unsanitary environmental

conditions in food processing plants can occur due to suspended bacterial particles in the

air. These biological particles are microscopic, with a diameter of 0.5 to 50 .im and are

suspended in the air as an aerosol. Airborne contaminants are also known as bioaerosols

and include bacteria, fungi, viruses and pollen. These may be present in the air as solid

(dust) or as liquid (condensation and water). An aerosol is a two-phase system of

gaseous phase (air) and particulate matter (dust, pathogens), thus making an important

bacterial vehicle. Pathogenic bacteria attach to dust particles and condensation, and

travel around the processing facility. This contaminated air comes in contact with food

products, containers, equipment and other food contact surfaces during processing.

According to the Food and Drug Administration (FDA), the food industry must reduce

product contamination by reducing airborne microorganisms (11, 18).

Airborne contaminants cause human illness due to ingestion of contaminated foods

and also reduce product shelf life resulting in an economic loss (18). Swabbing of

equipment is typically used to determine the sanitation level of food processing plants.

This method does not always provide an effective enumeration of airborne contaminants.

Air sampling is more effective because it collects aerosols settled on equipment and food

contact surfaces. Through air sampling, food processing facilities can identify airborne









contamination due to air contact with food products (18). Ideally, an air sampler would

be able to collect all of the viable microorganisms per unit volume of air, but this is not

possible because not all airborne cells can be physically separated from the air without

killing them during sampling (22). Methods for the detection of viable airborne

microorganisms include sedimentation, impaction on solid surfaces, impingement in

liquids, filtration, centrifugation, electrostatic precipitation and thermal precipitation.

Impaction methods are usually used because they obtain higher recovery rates than other

air sampling methods and can be used in situations where bioaerosol levels might be low

(11, 22).

It is impossible to keep airborne bacteria, yeast and mold in food processing areas

at a zero level. Some of the major sources of contamination in food processing facilities

are wastewater, rinse water and spilled product that become aerosolized. Airborne

bacteria, yeast and mold are generated in processing facilities by heating, ventilation and

air conditioning systems (HVAC). These systems contribute airborne microorganisms

under normal operation because they provide fertile areas for growth due to moisture.

Worker activity, equipment operation, sink and floor drains, and high pressure spraying

are also major sources of bioaerosols (18). Worker activity, talking, sneezing and

coughing create dust particles and air disturbances creating airborne microorganisms.

The workers' contribution of airborne bacteria depends on their health, condition of

clothing, hygiene and location in processing facility (18). Equipment operation

contributes to variations in microorganism levels. Airborne bacteria increase with the use

of conveyor systems which cause bacterial aerosols that adhere to conveyor surfaces (38).

The direction of airflow is important in the control of bioaerosol contamination and it









should always be counter current to that of the product flow. Barriers like walls and

doors are used to separate clean and unclean areas (34). Sink and floor drains can harbor

microorganisms because they are humid and contain nutrients from wastewater which

provide a fertile growth environment. Flooding of drains causes microorganisms on the

surface to become aerosolized and air disperses them, causing increased levels of

aerosolized bacteria in the food processing facility (38). High pressure spraying also

causes an increased level of aerosolized bacteria after spraying. The extent of this

increase depends on the condition of floor, water pressure and the amount of water used.

High temperature and humidity in the processing room increase microbial growth, but if

the environment is controlled, bacterial growth can be minimized (38).

Intense husbandry practices and long term residence of cattle in feedlots and pens

provide great opportunity for microorganisms to affix to hoofs and hides. Research

regarding airborne contamination levels in meat processing facilities indicates airborne

microbes are a potential source of microbiological contamination in various meat

products. According to Rahkio and Korkeala (34) higher concentrations of airborne

bacteria exist in the back-splitting area than in the weighing section in pork

slaughterhouses. The skin or hide of slaughtered animals can be a source of airborne

bacteria in slaughterhouses. According to Jericho et al. (21) many processes during cattle

slaughtering are associated with the creation of bioaerosols. The attachment of specific

pathogenic and nonpathogenic bacteria to carcass surfaces after hide removal is usually

immediate (21).

Bacterial isolates may be identified using a variety of methods. The order in which

these tests are performed is referred to as an identification scheme. Culturable bacteria









are usually first classified according to their microscopic morphological characteristics

and Gram stain reaction. Microorganism identification can be accomplished based on

phenotypic or genotypic characteristics or both. Phenotypic identification is based on

observable physical or metabolic characteristics of bacteria. Genotypic identification

implicates the characterization of a portion of the bacterium's genome using molecular

techniques for DNA or RNA analysis (14).

According to Mead et al. (30) Salmonella (31%), Listeria (28%), Toxoplasma

(21%), Norwalk-like viruses (7%), Campylobacter (5%), and E. coli 0157:H7 (3%)

account for more than 90% of estimated food-related fatalities. E. coli 0157:H7,

Salmonella spp., Listeria monocytogenes, and Staphylococcus aureus are major

foodborne bacteria pathogens that have an animal reservoir and have been implicated in

the contamination of various meat products.

The main objective of this study was to enumerate total airborne bacteria and yeast

and mold contaminants and determine the presence of Escherichia coli 0157:H7, Listeria

spp., Salmonella spp., and Staphylococcus spp. on pork and beef carcasses and in

bioaerosols generated in a slaughtering facility at the University of Florida, Gainesville,

FL.














CHAPTER 2
LITERATURE REVIEW

Bioaerosols

The study of airborne microorganisms and their effect on human health and the

environment is known as aerobiology. In recent years research in this field has increased

because of growing awareness of the variety of health problems potentially caused by

airborne microorganisms (5). An aerosol is a suspension of microscopic solid and/or

liquid particles in air or gas. Biological aerosols (bioaerosols) are single microorganisms

or clumps of microorganisms attached to solid or liquid particles suspended in the air

(11). Organisms present in bioaerosols can be bacteria, yeasts, molds, spores of bacteria

and molds, microbial fragments, toxins, metabolites, viruses, parasites and pollen.

Bioaerosols generally range in size from 0.5 [im to 50[im in diameter (5, 22).

Microorganisms in bioaerosols may attach to dust particles or may survive as free

floating particles surrounded by a coating of dried organic or inorganic material. They

cannot multiply in bioaerosols due to a lack of nutrients, but these aerosols can travel in

the air for great distances. Location and environmental conditions such as humidity,

density and temperature have a great effect on the type of population and amount of

microorganisms in the air. Some of the major sources of bioaerosols are humans (by

sneezing, coughing and talking), animals, vegetation and dust particles (1).

Microorganisms can become aerosolized from environmental sources such as worker

activity, water spraying, sink and floor drains, air conditioning systems, and different

food processing systems (11).









Aerosols display intricate aerodynamic behavior resulting from various

combinations of physical factors such as Brownian motion, electric gradient, gravitational

field, inertia force, electromagnetic radiation, particle density, temperature gradient and

humidity. The behavior of bioaerosols is dominated by physical and biological factors.

Physical factors affect where and how many bioaerosol particles will reach a specific

surface. The most important biological factor is the ability of the bioaerosol particle to

withstand lethal or sublethal stress or damage as it is dispersed in the air. Some of these

stresses are created during aerosol generation, dispersion and landing or collection.

These stresses are usually sublethal, but when joined with other environmental stressors

like temperature, dehydration, irradiation, oxidation and pollution the effect is often lethal

(22). Bioaerosol particle size is one of the main factors affecting aerodynamic behavior.

Vegetative bacterial cells usually will not survive long in air unless they have a protective

medium surrounding them or unless relative humidity and temperature are favorable.

Vegetative bacteria are usually present in the air in lower numbers than bacterial and

mold spores. Bioaerosols generated from the environment are usually bacterial spores,

yeasts and molds, but during food processing the main sources of contamination are

vegetative bacteria like Staphylococcus spp. and Micrococcus spp. (11). According to

Al-Dagal and Fung (1), Escherichia coli exhibits rapid death at low relative humidity

(<50%) and temperatures between 150C and 300C. Aerosolization is stressful enough for

vegetative bacteria so it is important to reduce additional stress caused by collection

procedures and growth media used during air sampling. When bioaerosols have been

subjected to mechanical or physical damage their recovery on selective media is reduced.

Bioaerosol research includes generation, collection, storage and analysis of aerosols. In









addition to cell injury some other factors that will influence bioaerosol collection are

strain of the microorganism, growth conditions, aerosol generation, aerosol particle size

and collection method (22).

Bioaerosol Sampling Methods

In an ideal situation a bioaerosol sampler would be able to count the total number

of viable microorganisms per unit volume of air. In reality this is not possible because

100% of airborne cells can not be physically separated from the air without killing them

during sampling (22). Quantitative and qualitative guidelines that relate numbers and

types of microorganisms per unit volume of air to acceptable levels of product samitation

must be established. These guidelines should be established for each individual

processing plant to ascertain possible sources of product contamination (air flow patterns,

ventilation systems, personnel activity and density), therefore having the capacity to

reduce airborne contamination. Various methods for the detection of viable airborne

microorganisms exist. The quantitative determination of airborne microorganisms is

possible by sedimentation, impaction on solid surfaces, impingement in liquids, filtration,

centrifugation, electrostatic precipitation and thermal precipitation (11, 22).

It is important to recognize that no single sampler can be used for collecting and

analyzing all bioaerosols. Factors that must be considered during the selection of an

appropriate sampling method include sampling environment, analysis methods used and

monitoring objectives. An air sampler's performance is determined by physical and

biological components. The inlet efficiency and particle collection efficiency are the

physical parameters affecting an air sampler's performance. Inlet efficiency of the air

sampler is its ability to extract particles from the environment without bias for shape, size

or density, and collection efficiency is its ability to remove particles from the air and









transfer them onto collection media. The biological component of an air sampler is its

ability to collect all microorganisms without affecting culturability which is required for

their detection and quantification. Sampling stress may injure the collected

microorganism causing it to be in a viable but nonculturable state. The collection time

may also affect the ability to obtain a representative bioaerosol sample. Sampling times

that are too long may cause increased sampling stress or cause an overload of particles

which may prevent enumeration. Air sampling collection periods are usually moderately

short and a single monitoring result can be of little value due to the variability of

bioaerosols. Duplicate air samples are highly recommended and their measurement is

expressed as an average of the replicate data observations (5).

Sedimentation

Sedimentation sampling is a static method that relies on the force of gravity and air

currents to cause the settling of airborne microorganisms onto agar plates filled with

general and selective media. Standard 90 mm diameter plates are placed throughout the

processing facility for about 15 minutes. After exposure the plates are incubated for an

appropriate time and temperature. Sedimentation results are expressed as CFU (colony

forming units) or particles per minute of exposure. This technique is inexpensive, easy,

and collects bioaerosols in their original state. The main disadvantage of this method is

its inability to measure the number of viable particles per volume of air. Other

disadvantages of this method are long sampling times, great reliance on air currents, bias

towards large particles and low correlation with counts obtained using other methods.

This method is useful when fallout onto a specific surface is of particular interest (11,

22).









Impaction

The majority of air samplers used in the food industry use impaction as the method

for collecting bioaerosols. Impaction methods use the inertia of particles to separate them

from the air currents (5). Impactors collect airborne microorganisms onto an agar surface

or an adhesive coated surface with the use of a vacuum. An impactor consists of an air

jet that is directed over the impaction surface causing the particle to collide and stick to

the surface. There are two types of impactors: slit (i.e. STA, New Brunswick Sci. Co.

Inc., Casella, BGI Inc.) or sieve (i.e. Andersen sampler, Thermo Electron Corp.)

samplers. A slit sampler is cylindrical in shape and has a tapered slit tube that creates a

jet stream when an air samples is pulled by a vacuum. The air sample is collected onto

an agar plate which is rotating on a turn table to create an even distribution of particles.

A slit sampler requires a vacuum to draw a constant flow rate of usually 28.3 liters per

minute (11, 22).

Sieve samplers function by drawing air (i.e. 28.3 1/min) through a metal plate with

many small holes. Air particles impact on the agar surface which is a few millimeters

below the metal sieve. Sieve samplers like the Andersen sampler may consist of a single

stage or two, six or eight stages. The stages of a multiple stage sampler have

decreasingly smaller holes causing increased particle velocity as the air travels through

the sampler. Large particles are impacted on the first stages and smaller particles are

carried until they are accelerated enough to impact the later stages. Multiple stage

impactors are not only used for the enumeration of viable particles per unit volume of air,

but also yield a size profile of particles in the bioaerosol. A two stage impactor is used

when the differentiation between respirable particles (<5 [im) and nonrespirable particles

(>5 [m) is of interest. Multiple stage impactors are used more in health care settings than









in food processing environments. Single stage impactors do not differentiate between

particle sizes and are used when the total number of viable particles per unit volume of

air is needed. Impaction methods obtain higher recovery rates than other air sampling

methods and are used when bioaerosol levels are expected to be low. This method results

in a low sampling stress and after collection no further manipulation is needed because

particles are on agar plates. Impactors possess relatively high sampling efficiencies, are

rugged and simple to operate. Some disadvantages of this method are that these samplers

are usually difficult and bulky to handle, expensive, and cumbersome. Also the inside of

the sampler and the outside of the agar plates must remain sterile until sampling begins or

else samples may become contaminated (11, 22).

Impingement

Impingement methods use a liquid medium for the collection of airborne

microorganisms. Air particles are entrapped in the liquid as air is dispersed through the

liquid. Liquid impingers are either low velocity or high velocity. Low velocity

impingers do not efficiently collect small particles (<5 [im) since they can be trapped in

bubbles and carried out with the released air. High velocity impingers collect all particles

sized greater than 1 [tm but tend to destroy vegetative cells due to the high air velocities

generated during sampling. Liquid impingers must use appropriate collection medium

for the recovery of different microorganisms. The medium must preserve the viability of

the microorganisms sampled while inhibiting its growth. Some of the common collection

media include phosphate buffer, buffered gelatin, peptone water and nutrient broth.

Airborne microorganism quantification is accomplished by serial diluting and plating an

aliquot of the collection liquid onto solid growth medium. The total volume of air

sampled and the volume of collection fluid must be measured to determine the number of









cells collected. Liquid impingers are used in situations where bioaerosol concentrations

are expected to be high. The All Glass Impinger (AGI-30, Ace Glass, Inc.) is a high

velocity impinger that is commonly used for air sample collection in respiratory disease

studies. It operates by drawing air through an inlet tube mimicking the nasal passage, at a

rate of 12.5 liters per minute. Impingers are relatively inexpensive and simple to operate

but viability loss may occur due to shear force exerted during collection. While particles

impinge into the collection media, the air stream approaches sonic speed which can cause

destruction of vegetative cells. Overestimation of bacterial counts is also a problem with

this sampling method since high air sampling velocity can disperse dust particles, thus

breaking up clumps of bacteria. Another limitation of the impingement method is its

failure to collect particles smaller than 1 [tm (11, 22).

Filtration

The filtration method collects airborne microorganisms onto a filter which is

mounted on a holder and connected to a vacuum source with a flow rate controller. The

filter can consist of sodium alginate, cellulose fiber, glass fiber, gelatin membrane (pore

size 3 [m) or a synthetic membrane (pore size 0.45 [tm or 0.22 pm). Gelatin membrane

filters are water-soluble and can be placed directly onto an agar surface for quantification

or can be serially diluted in a liquid first. Synthetic membrane filters are agitated in a

liquid to disperse the particles before bacteriological analysis is performed. Filter

collection devices are used for the enumeration of molds or bacterial spores but are less

effective for vegetative cells because of the stress associated with desiccation, although

shorter sampling times could reduce this stress. Filtration devices are low in cost and

simple to operate and their filters possess a large number of pores so that large volumes

of air can be sampled during a short period of time (15).









Centrifugation

Centrifugation sampling methods create a centrifugal force that propels airborne

microorganisms onto an agar surface. An aerosol is spun in a circular path at a high

velocity and the centrifugal force causes the particles to impact against an agar surface.

Microorganisms experience less stress compared to impaction or impingement sampling

methods since no high velocity jet forces are created during centrifugal sampling. These

devices can give more representative samples since they can rapidly sample high volumes

of air. Air sample results are expressed as CFU per liter of air sampled. Centrifugal

samplers are simple to operate and are less expensive than most impactors. One of the

limitations of this method is its inability to create enough centrifugal force to impact

small particles onto an agar surface. The Reuter centrifugal air sampler (RCS Sampler,

Biotests Diagnostics Co.) is portable, battery operated and easy to use. This sampler

collects 100% of 15 .im particles, 55% to 75% of 4 .im to 6 C.m particles and does not

effectively collect particles sizes smaller than 1 im (11).

Electrostatic Precipitation

Electrostatic precipitation methods capture microorganisms by giving them an

electrostatic charge and collecting them on an oppositely charged rotating disk. This

surface can be agar or glass. This method possesses a high sampling rate, high collection

efficiency, and low resistance to air flow. A disadvantage of this method is that nitrogen

oxide and ozone which may be toxic to microorganisms are produced during air sample

ionization. Little is known about the effect of electrostatic charges on the viability and

clumping of microorganisms. Electrostatic precipitation is rarely used for aerosol

detection since its equipment is complex and requires careful handling (11).









Thermal Precipitation

Thermal precipitation methods are based on thermophoresis principles in which

particles move away from hot surfaces toward cooler surfaces. The degree of particle

movement will depend on the temperature gradient. This method is used to determine

particle size distribution, especially when collecting particles smaller than 1 jim.

Microorganisms are collected onto glass cover-slips and are sized and counted

microscopically. This method is not commonly used in industry since it requires very

precise adjustments and its air sampling rate is considerably low (300 to 400 ml/ minute)

(22).

Beef and Pork Slaughter Process

Slaughtering procedures differ among species, but livestock slaughtering always

includes bleeding, hair or hide removal and removal of the abdominal viscera

(stomach(s), intestines, liver and reproductive glands). The head, pluck (trachea,

esophagus, lungs and heart) and the feet (except in swine) are also removed. The kidneys

and surrounding fat are removed from swine, but in the United States they remain on beef

carcasses because of tradition. The head, edible offal (heart, liver, tongue and

sweetbreads) and inedible offal (rest of viscera and pluck) removed from the animal are

kept with the matching carcass for inspection purposes. Standard operating procedures

for beef slaughtering consist of, immobilization, bleeding, head removal, hide removal,

feet removal, evisceration, splitting, inspection, washing and chilling. The pork

slaughtering procedure consists of stunning, bleeding, scalding/dehairing and toenail

removal, head removal, evisceration, splitting, inspection, washing and chilling (42).









Stunning or Immobilization

All animals slaughtered under inspection must be stunned before bleeding

according to the Humane Methods of Slaughter Act of 1978. Cattle are immobilized

mechanically via concussion (stunning hammer) or penetrating type devices (captive bolt

stunner). A sharp blow is delivered midway between the eyes and halfway up the

forehead since this is an area where the brain is closest to the skull. Pigs are electrically

stunned using a hand-held device that dispenses electrical current to the skull producing a

loss of consciousness (42).

Bleeding

When bleeding cattle, a knife is inserted at a 450 angle directly below the brisket.

A cut 10 to 15 inches long extending from the brisket to the throat causes blood removal

by severing the carotid arteries and jugular veins. In bleeding pork, a six inch knife is

inserted in the middle of the ventral aspect, extending midway from the sternum to the

throat. The knife is inserted with the point upward and then with the point moved

downward until it reaches the backbone. During bleeding the carotid arteries and jugular

veins are severed. Approximately six to nince minutes are required to bleed out beef and

pork completely during the slaughtering process (42).

Scalding/Dehairing

Hogs are immersed in 600C water until their hair begins to loosen from the flank

area. Hot water causes collagen surrounding the hair follicles to change into gelatin

which facilitates removal of the hair either manually by scraping or with a dehairing

machine. Scalding for an excessive amount of time or temperature may cause the skin to

cook resulting in large areas of skin and fat being removed from the carcass, therefore









marring the carcass. Residual hair remaining after dehairing is removed by shaving the

carcass with knives and singing with a gas-fired hand-held torch (42).

Evisceration

Removing the viscera from the abdominal cavity and pluck from the thoracic cavity

is performed during the evisceration process. First the anus is cut loose from the

attachment under the tail and the head is removed at the occipito-atlantal space. It is

critical that both ends of the gastrointestinal tract are secured with string to prevent the

contents leakage onto or into the carcass. Next a cut is made from the crotch to the

sternum without cutting the viscera or pluck. The sternum must be severed with a knife

in pork and a saw in beef to be able to remove the pluck. During pork slaughtering the

worker removes the viscera by cutting around the intestines, stomach, liver and spleen

beginning at the detached anus. During beef slaughtering the worker pulls and cuts

around the stomachs (rumen, reticulum, omasum, absomasum), intestines, and spleen

beginning at the loosened anus. The pluck (heart, esophagus, lungs and trachea) is

removed by cutting through the diaphragm membrane and down the backbone. During

evisceration, extreme caution must be employed not to cut or tear the gastrointestinal

tract to prevent ingesta or fecal material from contaminating the carcass (42).

Splitting and Chilling

Pork carcasses are split down the vertebral column and the kidneys and leaf fat are

removed in preparation for chilling. The carcass is washed, blood clots and loose glands

are trimmed off and the carcass is weighed and identified. Beef carcasses are split down

the center of the backbone and the kidneys and leaf fat and are left on the carcass. Beef

carcasses are carefully trimmed removing pieces of hide, bruises, hair, feces or ingesta

and are then washed. Carcasses are weighed, tagged and placed in a -2 to 00C rapid air









movement cooler to facilitate swift carcass temperature reduction. Carcasses are usually

chilled between 18 and 24 hours before being moved to a holding cooler at 00C to 1oC for

subsequent storage until fabrication (42).

Bioaerosols in the Meat Industry

Investigations regarding airborne contamination levels in meat processing facilities

indicate airborne microbes are a potential source of microbiological contamination in

various meat products. Limited information exists in illustrating the association between

beef and pork carcass contamination and airborne bacteria during the slaughtering

process. Rahkio and Korkeala (34) took air samples from the back splitting and weighing

areas of beef and pork slaughtering lines at a height of 1 to 1.5 m from the floor and 1 m

from the carcass. An Andersen two stage sampler, adjusted to a flow rate of 0.0283 m-

3/min, was used to collect the air samples. Significantly higher (P < 0.05) mean log

CFU/m3 of air were shown to exist in the back-splitting area (3.13 to 4.07) than in the

weighing section (2.45 to 3.58) in pork slaughterhouses. Rahkio and Korkeala (34)

concluded that higher levels of airborne bacteria present in the back-splitting area when

compared to the weighing area might be caused by bacteria being carried by airflow from

contaminated parts of the lines. Higher bacterial counts in the back-splitting area may

also be due to movement of the saw blade and water flowing from the cutting saw. The

movement of personnel between clean and unclean areas in the slaughtering line appears

to be linked to higher carcass contamination levels (34).

Intense husbandry practices and long term dwelling of cattle in feedlots and pens

provides great opportunity for microorganisms to affix to hoofs and hides. Mesophilic

bacteria of fecal and soil origin attached to hoofs and hides may exceed 109 CFU/cm2 and

represent the main food safety hazard in cattle slaughterhouses (21). Wilson et al. (53)









reported that mostly Gram positive bacteria (Bacillus spp., Corynebacterium spp.,

Micrococcus spp., Paenibacillus spp. and Yersinia spp.) were isolated from air samples

taken at a cattle feedlot. Gram negative bacteria might have been present in the feedlot

environment but were reduced in number by environmental conditions or may have been

in a viable but nonculturable state. Butera et al. (6) reported that Gram positive cocci

(Staphylococcus, Micrococcus, Leuconostoc, Streptococcus, and Aerococcus) made up

72% of the total bacteria isolated from a hog growing facility. Gram positive rods

(Bacillus) made up 7.2% and Gram negative rods (Enterobacteriaceae) made up 20.8%

of the total bacteria isolated in that study.

According to Jericho et al. (21) the majority of processes in cattle slaughterhouses

are associated with the creation of bioaerosols. The skin or hide of slaughtered animals

can be a source of airborne bacteria in slaughterhouses. The attachment of specific

pathogenic and nonpathogenic bacteria to carcass surfaces after hide removal is usually

immediate. An undetermined amount of this contamination comes from bioaerosols

which can be easily recovered from carcass surfaces. Jericho et al. (21) used a slit air

sampler (STA-203) to collect air samples from the hide removal floor, carcass dressing

floor and cooler. The highest total viable counts were observed in the hide removal floor.

The presence of bacterial counts in air samples collected before the slaughtering process

commenced was attributed to in-process cleanup operations using high water pressure

hoses (21).

High speed equipment and air circulation are widely used in slaughtering and

processing practices. Therefore, it is possible for bacteria on the surface of animals,

employees and equipment to become airborne during these processes. Kotula and









Emswiler-Rose (25) used a Ross-Microban air sampler to collect air samples during

evisceration, offal recovery, carcass cooling, carcass breaking and further processing.

Average CFU per 0.028 m3 of air had a tendency to decrease during the process of

converting pork carcasses to edible product. The variability in mean CFU per 0.028 m3

of air between the different locations although significant (P < 0.05) was in most cases

less than one log and therefore was of little importance. In that study, coliform growth on

Petri plates was only observed in 15% of the air samples. According to Kotula and

Emswiler-Rose (25) the low incidence of airborne coliforms was surprising because of

speculation that bioaerosols may be an important cause of indiscriminate meat product

contamination.

Air currents can be influenced by a plant's configuration therefore affecting the

airborne contamination of beef carcasses. Reduction of airborne bacterial contamination

can be accomplished by building walls between clean and unclean areas or by allowing

sufficient distance between these areas in slaughtering processes. In the slaughter

process, the unclean areas are considered from stunning the animal to the hide removal

and the clean areas are considered from evisceration to the final wash of the carcass.

Worfel et al. (54) used an Andersen single stage sampler to collect air samples from

various slaughtering processes in hide-on (unclean) and hide-off (clean) areas from three

slaughter facilities with different plant layouts. Two plants which had a wall separating

the clean and unclean areas had significantly (P < 0.05) lower bacterial counts in the

clean areas than the plant with no area separation (54).

Whyte et al. (52) used a six stage Andersen air sampler to collect air samples from

the defeathering, evisceration, air chills, packing, deboning and further processing areas









in three poultry processing plant. Various media were used in that study to enumerate

total mesophilic and psychrophilic aerobic bacteria, Escherichia coli,

Enterobacteriaceae, thermophilic Campylobacter spp., and Salmonella spp. Total

aerobic bacterial counts were significantly (P < 0.05) higher in the defeathering areas

when compared to the other areas. Mean log CFU/m3 of air counts of Escherichia coli

and Enterobacteriaceae were higher in air samples acquired from the defeathering (1.85,

1.82) and evisceration (1.15, 0.97) areas and were not detected in the subsequent areas.

Salmonella spp. was not positively identified in any of the air samples taken from the

three poultry processing plants in that study. According to Whyte et al. (52) that study

identified air as a possible carrier of pathogenic bacteria and demonstrated the need for

separation of dirty and clean areas in poultry slaughtering.

Ellerbroek (13) collected air samples from the reception and evisceration areas in a

poultry processing facility using an Andersen air sampler. The reception area to the

processing facility consisted mostly of Gram positive bacteria (Micrococcus spp.,

Corynebacterium spp., Staphylococcus spp.) and yeasts, probably coming from poultry

skin and feathers. The evisceration area consisted mostly of Gram positive

Staphylococcus spp., but Gram negative Acinetobacter and Moraxella spp. were also

present in this area. Airborne bacterial counts in the reception area were similar to those

in the evisceration area, however Enterobacteriaceae were lower in numbers in the air of

the evisceration area (13).

Lutgring et al. (27) took air samples from various areas in four poultry processing

facility using an N6 single stage Andersen viable sampler. Different media were inserted

into the air sampler to enumerate mesophilic and psychrophilic bacteria and yeast and









mold. In all four plants the highest concentration of mesophilic bacteria were observed in

the receiving areas. Airborne bacterial concentrations were 100 to 1,000 times greater in

the receiving area than outside the plant. Bioaerosol levels progressively decreased as

slaughtering went from receiving to carcass cutting and deboning. Psychrotrophic

bacterial counts were about one log less than mesophilic bacterial counts for the majority

of samples areas. Throughout that study total yeast and mold counts were far less than

total bacteria counts. Differences in yeast and mold counts between areas sampled were

much less apparent than differences in bacterial counts. According to Lutgring et al. (27)

air flow should be controlled so that it travels from the finished products area to the

receiving area, and when possible these areas should be physically separated. The food

safety effect of bioaerosols on products that require cooking for consumption is less

important than on ready to eat food products (27).

Airborne bacterial counts could offer an alternative method for determining

carcasses contamination in slaughterhouses. Even though air sampling employs the use

of special equipment, it can be a more convenient and rapid method for carcass sampling.

Using this method there is no interruption of the slaughtering line and no need for

personnel to touch the carcasses while sampling. Nevertheless, surface carcass swabs are

essential for indicating presence of pathogenic bacteria due to slaughtering events

producing fecal contamination (34).

Bioaerosols in the Dairy Industry

Bioaerosols may be a means for microbial contamination of dairy products

according to the U. S. Food and Drug Administration. Airborne contamination in dairy

processing facilities can result in manufacturing of low quality products with a reduced

shelf life (37). Aerosols in dairy plants may be created by HVAC (heating, ventilation,









and air conditioning) systems, high pressure water spraying, floor drains, plant workers,

supplies, openings between processing rooms, spilled milk on the floor and conveyor

systems. The main areas of bioaerosol contamination in dairy processing facilities are

during filling of retail containers and while filling of holding tanks with pasteurized

products (36, 38). An effective bioaerosols sampling program can be used in a milk

processing plant as a tool to control pathogens therefore increasing product shelf life.

The use of packaging equipment that eliminated airborne contamination produced a 7 day

increase in product shelf life (36).

Ren and Frank (38) took air samples in two fluid milk and two ice cream plants

using a centrifugal air sampler (RCS Sampler). Air samples were collected from the

ventilation inlet and outlet, above drain in packaging area, processing/packaging area

after water spraying, conveyor belt, pasteurized product holding tank and product filling

area. At each sampling point eight liters of air were sampled onto media for the

enumeration of total aerobic bacteria (TA), Staphylococcus spp. (ST) and total yeast and

mold (YM). Air samples exiting the ventilation system yielded similar levels of TA and

ST as those entering the system. It was established that the air filtration equipment used

in these plants had been neglected. A significant (P < 0.05) increase in TA and ST in the

air above the floor drains during processing was observed when compared to air without

processing activities. Efficient floor drain cleaning in these critical rooms is

recommended to reduce the extent of bioaerosol contamination. The use of high pressure

water hoses during processing and packaging was linked to significant increases in TA

and ST levels in all plants and YM levels in some plants when compared to levels during

inactivity. Conveyer systems contain metal surfaces to which microbes can adhere.









These microbes can become aerosolized when physical disturbances are created on the

conveyer systems during product or case collision. A significant increase in TA, ST and

YM levels were observed in all four plants when conveyer systems were operating.

People are a major source of airborne contamination through their speaking, sneezing and

breathing. Worker activity at the filler equipment station produced a significant rise in

TA and ST levels. Low levels of airborne contamination were observed inside the

pasteurized holding tanks (38).

Ren and Frank (36) took air samples at four fluid milk processing plants using an

Andersen two stage sampler, a Ross-Microban sieve sampler and a Biotest RCS sampler.

The areas sampled were the raw milk storage area, processing, and filling area. Log

mean viable particle counts obtained with the Andersen sampler were significantly

greater than those obtained with the Biotests RCS and the Ross-Microban samplers. All

of the samplers recovered the lowest amount of microbial aerosols in the raw milk

storage area and the highest amount in the milk filling areas. Ren and Frank (37)

monitored the pasteurized mix storage, processing and filling areas in two commercial ice

cream plants using the same samplers as the previous study. The level of microbial

aerosols recovered by the Andersen two stage sampler were similar to those recovered by

the Biotest RCS sampler, but were significantly (P < 0.05) greater than those recovered

by the Ross-Microban sampler. Each sampler recovered more microbial aerosols in the

filling areas than in the processing areas and similar levels in the pasteurized mix storage

and filling areas in both ice cream plants (37).

Kang and Frank (23) used a tracer microorganism (Serratia marcescens) to prove

that floor washing and drain flooding are sources of viable aerosols in a dairy processing









plant. The ability of biological aerosols to stay viable in the environment was also

observed. The Andersen six stage sieve air sampler was used in that study due to its

proven reliability through previous research. Air samples were collected from the non

inoculated drain area during a break in processing to obtain background bacterial levels.

Air samples were also collected 30 seconds after spraying water on the floor to verify the

absence of Serratia marcescens. Air samples were taken before processing, 30 seconds

after spraying water on the floor, and at 10 minute intervals until reaching 40 minutes.

Inoculated Serratia marcescens was recovered from air above the drain, therefore

proving that microorganisms including pathogens can become aerosolized from drains by

physical disruption. It took about 40 minutes or more for aerosol microbial levels to

return to background levels. The greatest reduction in aerosol microbial levels was seen

in the first 10 minutes after drain flooding (23).

Microbiological Analysis of Bioaerosols

Various sample analysis methods can be applied to air samples to provide

information regarding concentration and composition of bioaerosols. First it is important

to select the sample analysis methods being used since not all air sampling systems are

compatible. Impaction type sampling methods usually rely on microbiological methods

like culturing and microscopy. Impingement and filtration methods are more adaptable

with respect to sample analysis alternatives since the air samples are collected in liquid or

on a filter. Limitations to traditional techniques have given way to alternative methods

like biochemical, immunological and molecular assays. A vast majority of bioaerosol

data generated has been attained by using culture analysis and most current air samplers

are designed to collect particles on nutrient agar. A major hurdle in using culture analysis

is that only those cells that survive sampling and that can be cultured will be enumerated.









Those microorganisms that are subjected to sampling stress and environmental stress may

not grow under artificial nutrient conditions in a laboratory (5).

Microorganisms have a wide range of nutritional requirements which cannot all be

met by one culturing media. During bioaerosol monitoring a general media that provides

the conditions for the greatest number of microorganisms to grow should be employed. It

may be necessary to perform replicate sampling and use a variety of sampling media in

an attempt to enumerate the greatest number of microorganisms. For the culture of

general bacteria several broad spectrum media like tryptic soy agar, nutrient agar, and

casein soy peptone agar can be utilized. After the correct incubation time, culturable

microorganisms are determined by enumerating colony forming units (CFU). Culturable

airborne microorganisms are calculated by dividing the number of total CFU per sample

by the volume of air sampled. The identification of fungal isolates is usually performed

by microscopic determination of the morphological characteristics (5).

The tests and the order in which these tests are performed to identify

microorganisms are referred to as an identification scheme. Bacterial isolates may be

identified using a variety of methods. Culturable bacteria are usually first classified

according to their microscopic morphological characteristics and Gram stain reaction.

Microorganism identification can be accomplished based on phenotypic or genotypic

characteristics or both. Phenotypic identification is based on observable physical or

metabolic characteristics of bacteria. The most commonly used phenotypic identification

methods are macroscopic and microscopic morphology, staining characteristics,

environmental requirements for growth, resistance or susceptibility to antimicrobial

agents, and nutritional requirements and metabolic capabilities. Other methods of









characterization are based on the antigenic makeup of organisms and they involve

antigen-antibody interactions. Some immunochemical identification methods are

precipitin assays (immunodiffusion), particle agglutination assays (latex agglutination),

immunofluorescent assays, and enzyme-linked immunosorbent assays (ELISA).

Genotypic identification implicates the characterization of a portion of the bacterium's

genome using molecular techniques for DNA or RNA analysis. The presence of a

particular gene or specific nucleic acid sequence is interpreted as a definitive

identification of the microorganism. Some molecular identification methods used for the

identification of microorganisms are nucleic acid hybridization methods, amplification

methods (polymerase chain reaction or PCR, reverse transcription-PCR or RT-PCR, Real

Time PCR, repetitive extragenic palindromic PCR or REP-PCR), enzymatic digestion of

nucleic acids (restriction fragment length polymorphisms or RFLPs), and pulsed field gel

electrophoresis (PFGE) (14).

The identification of bacteria to their genus and species levels can be accomplished

by the use of conventional biochemical methods or commercially available identification

systems. API is one of the most widely used diagnostic tools for identification of the

Enterobacteriaceae family of bacteria (11). Initially, its database was mostly oriented

toward clinical isolates, but in recent years it has grown to include food, industrial and

environmental isolates also. The API testing system is made up of various small and

elongated wells which contain different dehydrated media, and is intended to perform

various biochemical tests from a pure culture grown on an agar plate. Advantages of the

API system include an excellent data base, long shelf life and a proven testing record

throughout the past 25 years. API identification can be time consuming due to the









inoculation of the numerous wells containing media. API 20E is used to identify

species/subspecies of Enterobacteriaceae (Escherichia coli and Salmonella spp.) and/or

non-fastidious Gram negative rods. API Staph is used for overnight identification of

Gram positive Staphylococci spp., Micrococci spp., and Kocuria spp. API Listeria, API

Campy and API Coryn are used to identify Listeria spp., Campylobacter spp., and

Corynebacterium spp. respectively (11).

Short chain fatty acids analysis (volatile fatty acids, VFA) using gas

chromatography (GC) has been regularly used to identify anaerobic bacteria. The fatty

acids between nine and 20 carbons long have been used to characterize genera and

species of bacteria, especially nonfermentative Gram negative organisms. The

introduction of fused silica capillary columns makes it convenient to use gas

chromatography of whole cell fatty acid methyl esters (FAME) to identify a wide range

of organisms. These columns permit the recovery of hydroxy acids and resolution of

many isomers. Around 300 fatty acids and related compounds have been found in

bacteria analyzed in the MIDI Research and Development Laboratory (MIDI, Inc.). This

information can be quantitatively used to differentiate among groups of bacteria;

therefore giving great "naming" power to the Sherlock MIS (Microbial Identification

System). This system uses fatty acids 9-20 carbons in length and peaks are automatically

named and quantified by the system. In some Gram positive bacteria branched chain

acids predominate, while in Gram negative bacteria the lipopolysaccharides are

characterized by short chain hydroxy acids (41).

The five steps to prepare sample extracts for gas chromatography are harvesting

bacterial cells, saponification, methylation, extraction and sample cleanup. Samples are









injected into an Ultra 2 column which is a 25 m x 0.2 mm phenyl methyl silicone fused

silica capillary column. The GC temperature program heating the column goes from 1700

C to 2700C at 50C per minute. The flame ionization detector used by the Sherlock MIS

allows for a large dynamic range of detection and provides a good level of sensitivity.

Hydrogen is the carrier gas and air is used to support the flame. The GC detector passes

an electronic signal to the computer as organic compounds burn in the flame of the

ionization detector. These electronic signals are integrated into peaks and the data is

stored. The composition of the bacterial sample FAME is compared to a stored database

using pattern recognition software (41). The Sherlock library is composed of more than

100,000 strains collected from around the world. This library is not limited by a fixed

number of biochemical assays like biochemical and enzymatic identification methods are

(41).

Environmental and Beef and Pork Associated Bacteria

Foodborne pathogens have been estimated to cause 6 to 81 million illnesses and as

many as 90,000 deaths annually in the Unites States alone (30). In general, bacterial

pathogens account for 60% of hospitalizations and 72% of deaths attributable to

foodbome transmission (30). Parasites and viruses account for 5% and 34% of

hospitalizations along with 21% and 7% of deaths associated with foodbome

transmission. Salmonella (31%), Listeria (28%), Toxoplasma (21%), Norwalk-like

viruses (7%), Campylobacter (5%), and E. coli 0157:H7 (3%) account for more than

90% of estimated food-related fatalities (30). Escherichia coli 0157:H7, Salmonella

spp., Listeria monocytogenes, and Staphylococcus aureus are major foodbome bacterial

pathogens that have an animal reservoir and have been implicated in the contamination of

various meat products. Meat is a substrate rich in nutrients and can support the growth of









a great number of microorganisms. The readily available water content of fresh muscle

tissue is high, along with the usable glycogen, peptides, amino acids, metal ions and

soluble phosphorous makes muscle tissue a good substrate for microbial growth. These

pathogens have been associated with the hide, skin, mucous membranes, and the

intestinal tract of healthy animals and are ubiquitous in nature (9, 39).

Microorganisms on the hide also include Micrococcus, Pseudomonas species and

yeasts and molds that are normally associated with skin, fecal material and soil. Animals

entering a slaughter facility have a vast population of aerobic microorganisms on their

hide, hooves and inside their intestinal tract, while the internal surface of carcasses is

generally considered to be sterile. It is generally agreed that the majority of

microorganisms on a dressed red meat carcass are directly transferred from the

contaminated hide and ruptured gastrointestinal tract. Bacterial transfer to the carcass

also occurs by aerosols, dust generation, workers hands, and the contact of the hide with

the exposed tissue. The majority of carcass contamination occurs during slaughtering

and dressing procedures (12, 16). The spoilage of meat at ambient temperature is a direct

result of the growth of mesophiles like Clostridium perfringens and members of the

Enterobacteriaceae family. Storage of meat at refrigeration temperatures restricts the

growth of mesophiles and allows psychrotrophs to dominate the spoilage microflora. The

spoilage of high moisture meat surfaces that are held at high relative humidity is usually

the result of bacterial activity. The microflora that dominate meat spoilage under

refrigerated aerobic conditions are the species Pseudomonas, Alcaligenes, Moraxella and

Aeromonas (12, 16).









Gram Negative Bacteria

Enterobacteriaceae

The most significant families of Gram negative bacteria are Enterobacteriaceae,

Vibrionaceae, and Pasteurellaceae (19). No sole group of taxonomically classified

microorganisms has accounted for greater notice from the scientific and medical

communities than the Enterobacteriaceae family. No other group has more selective and

differential media created for the isolation of its various members from medical and

environmental samples than the Enterobacteriaceae. This group is the backbone of

databases used in manual, semiautomated and automated microbial identification systems

(20).

The taxonomically defined Enterobacteriaceae family includes facultative

anaerobic Gram negative straight bacilli that are oxidase negative, glucose fermenting,

usually catalase positive and nitrate reducing plus grow on MacConkey agar. The genera

in this family usually have a cell diameter of 0.3 to 1.8 [tm and those genera that are

motile employ peritrichous flagella. Most members of this family grow well at 370C and

some grow better and are more active metabolically at 25 to300C. Although this family

consists mainly of mesophilic genera some psychrotrophic strains of Enterobacter and

Serratia can grow at 00C. Enterobacteriaceae are distributed worldwide and can be

found in soil, water, fruits, vegetables, plants, trees and animals from insects to humans.

They inhabit a wide variety of niches which include human and animal gastrointestinal

tracts and various environmental sites (11, 14, 19).

Some species like Salmonella typhi, .\/nge//a spp., and Yersiniapestis (overt

pathogens) cause diarrheal diseases which can be life threatening like typhoid fever,

bacillary dysentery and "black" plague. Other species not usually associated with









diarrheal diseases (opportunistic pathogens), along with those that are associated with

diarrhea, may cause many extraintestinal diseases like bacteremia, meningitis, urinary

tract, respiratory, and wound infections. Escherichia coli, Klebsiella, Enterobacter,

Proteus, Providencia, and Serratia marcesens account for about 50% of nosocomial

infections. Foodborne genera of the Enterobacteriaceae family consist of Escherichia,

Salmonella, Klebsiella, Serratia, Enterobacter, Citrobacter, Yersinia, Proteus,

Providencia, .i/ngell// and Erwinia (1], 14, 19).

Escherichia coli

Escherichia coli is a straight rod measuring 1.1 to 1.5 [tm by 2.0 to 6.0 [tm which

occur singly or in pairs and has an optimum growth temperature of 370C. Capsules or

microcapsules occur in many strains and some strains are motile by peritrichous flagella.

Escherichia coli is part of the normal flora of the intestinal tract of humans and various

animals. It can be classified as an overt or an opportunistic pathogen and usually

constitutes about 1% of the total biomass of feces. Most Escherichia coli do not cause

gastrointestinal illnesses, but some can cause life threatening diarrhea and chronic

sequelae or disability. E. coli is serologically classified on the basis of three major surface

antigens: 0 (somatic), H flagellaa) and K (capsule). The serogroup of the strain is

identified by the 0 antigen and its combination with the H antigen identifies the serotype.

There are more than 170 different serogroups of E. coli identified. Diarrhea causing E.

coli isolates are classified into specific groups based on virulence properties,

pathogenicity mechanisms, clinical syndromes, and specific O:H serotypes. These

groups include enterotoxogenic E. coli (ETEC), enteroaggregative E. coli (EAEC),

enteropathogenic E. coli (EPEC), enterohemorrhagic E. coli (EHEC), enteroinvasive E.

coli (EIEC), and diffuse-adhering E. coli (DAEC) (11, 12, 19).









ETEC strains adhere to the small intestinal mucosa by means of surface fimbriae

(type 1 pili) and cause symptoms by producing two enterotoxins: heat labile (LT) and

heat stable (ST) toxins. Diarrhea is caused by these enterotoxins acting on the intestinal

mucosa cells and not by ETEC invading the mucosa. One ETEC binds to the host cell

membrane the toxin is endocytosed and translocated through the cell and reaches its

target (adenylate cyclase) on the basolateral membrane of the intestinal epithelial cells.

cAMP dependent protein kinase is activated leading to stimulation of CY secretion and

inhibition of NaCl absorption resulting in watery diarrhea. ETEC strains cause travelers

diarrhea which can be severe and often fatal in infants. Contaminated foods and water

are the most frequent vehicles of ETEC infections. EAEC strains cause persistent

diarrhea mostly in children. This type of E. coli strain does not produce enterotoxins LT

and ST but does adhere to Hep-2 cells in an aggregative adherence pattern. Symptoms

include diarrhea, vomiting with persistent diarrhea in some patients (11, 12, 32).

EPEC strains cause severe diarrhea in young children and infants. The pathogen

adheres to the intestinal mucosa though A/E (attachment/effacing) lesions; causing an

intimate attachment to intestinal cells with effacement of underlying microvilli and

accumulation of filamentous actin (F-actin). The extensive loss of absorptive microvilli

by the A/E lesion could lead to diarrhea by improper absorption. EPEC actively alters

ion transport in epithelial cells causing an influx of positive ions or an efflux of negative

ions across the membrane. EIEC strains are genetically and pathogenically closely

related to ./uge//i spp. but do not produce Shiga toxin. These strains cause nonbloody

diarrhea by invading, multiplying and eventually causing death of colonic epithelial cells.

High fever and watery diarrhea are symptoms of this infection. DAEC strains cause









nonbloody diarrhea in children ages one to five. This strain produces a diffuse adherence

to HEp-2 and HeLa cell lines, but do not produce LT and ST enterotoxins or high levels

of Shiga toxin. DAEC strains do not posses EPEC adherence factor plasmids or A/E

lesions and do not invade epithelial cells (11, 12, 32).

No recently described enteric pathogen has received as much scientific and medical

examination as Escherichia coli 0157:H7. There are many serotypes belonging to the

EHEC group, but serotype 0157:H7 is the predominant food pathogen. It was first

recognized as a pathogen in 1982 when it was identified as the causative agent of two

outbreaks involving the consumption of undercooked ground beef leading to hemorrhagic

colitis. It is estimated that more than 700 individuals were affected and 4 deaths occurred

between 1992 and 1993 due to this pathogen. According to CDC estimates there are

more than 20,000 cases and 250 deaths annually in the U.S. due to Escherichia coli

0157:H7 infections. Cattle are considered to be the main reservoirs of this pathogen and

many outbreaks are associated with the consumption of undercooked ground beef and

unpasteurized milk (11, 20, 30).

This pathogen can be encountered through the fecal oral route or through human to

human transmission. It is highly acid resistant and has a low infectious dose of less than

100 cells and possibly as few as 10 cells. Escherichia coli 0157:H7 colonizes the large

intestine cecumm and colon) by A/E lesions. These lesions are an intimate attachment of

bacteria to intestinal cells causing an effacement of underlying microvilli and

accumulation of filamentous actin. These lesions are mediated by binding of intimin

(gene: eae) to its receptor (Tir) on the host cell. Both genes are located in LEE (locus of

Enterocyte Effacement). Intimin (0, 8, Y, and 5) is the sole adherence factor of E. coli









0157:H7. Using type III secretion system EHEC secrete proteins (EspA, EspB, EspD)

responsible for signal transduction events during AE lesion formation. Escherichia coli

0157:H7 causes damage by the production of one or two Shiga toxins (Stxl, Stx2) which

are potent cytotoxins. This factor leads to death and many other symptoms in patients

infected with EHEC. The Stx A-B subunit structure is conserved across all members of

the Shiga toxin (Stx) family. The B pentamer binds the toxin to a specific glycolipid

receptor (Gb3) present on the surface of eukaryotic cells. After binding, the holotoxin is

endocytosed through coated pits and is transported to the Golgi apparatus and then to the

endoplasmic reticulum. The A subunit is translocated to the cytoplasm where it acts on

the 60S ribosomal subunit. The Al peptide is an N-glycosidase that removes a single

adenine residue from the 28S rRNA of eukaryotic ribosomes causing the inhibition of

protein synthesis. The resulting disruption of protein synthesis leads to the death of renal

endothelial cells, intestinal epithelial cells, Vero or HeLa cells, or any cells which possess

the Gb3 (or Gb4 for Stx2e) receptor (32, 43).

Escherichia coli 0157:H7 induces damage by local cytokine response in clonic

mucosa and profuse bleeding of endothelial cells caused by the interaction of

inflammatory cytokines and Stx (blood vessels damaged in lamina propria). A/E lesions

are sufficient to cause nonbloody diarrhea but Stx is essential for the development of

bloody diarrhea and hemorrhagic colitis. The gastrointestinal illness hemorrhagic colitis

(HC) is characterized by severe cramping, abdominal pain, watery diarrhea followed by

grossly bloody diarrhea, and little or no fever. Hemolytic-uremic syndrome (HUS) is

associated with Stx producing E. coli. HUS is the damage to glomerular endothelial cells

which leads to narrowing of capillary lumina and clogging of the glomerular









microvasculature with platelets and fibrin. The decreased glomerular filtration rate is

responsible for the acute renal failure that is typical of HUS. HC and HUS are

responsible for damage to both intestinal and renal tissue (32, 43).

Escherichia coli 0157:H7 was first recognized in 1982 as a foodborne pathogen in

ground beef and achieved great notoriety in 1993 after a multi-state outbreak resulting in

four deaths. In 1994 the Food Safety Inspection Service (FSIS) declared E. coli 0157:H7

as an adulterant in beef and caused it to be the first microorganisms to have such status

under the Federal Meat inspection Act. E. coli 0157:H7 has been reported in feces,

rumen contents and on the hide of cattle during slaughter. According to Barkocy-

Gallagher et al. (3), E. coli 0157:H7 was less prevalent in feces (12.9%) than on the hide

(60.6%) or on preevisceration carcasses (31.7%) in three large beef processing plants in

the U.S. E. coli 0157:H7 was recovered from 1.2% of postintervention beef carcasses.

Cattle hides are a source of E. coli 0157:H7 and can be transferred from hide to carcass

during processing. The previous study suggests that hides are a more significant source

of contamination than feces. The prevalence of E. coli 0157:H7 on hides may reflect

many sources of contamination like soil, feces from other animals and lairage (3).

According to McEvoy et al. (29), E. coli 0157:H7 can be transferred to the carcass

during hide removal operations. The bung tying operation has been found to contribute

to E. coli 0157:H7 contamination indicating that this operation may not prevent transfer

from the anus to the carcass. Carcass chilling has been shown to reduce the prevalence of

this pathogen due to the synergistic effect of low water activity (aw) and temperature (29).

Salmonella spp.

Salmonella species are straight rods measuring 0.7 to 1.5 [tm by 2 to 5 [am, motile

by peritrichous flagella and have an optimum growth temperature of 370C. The









nomenclature of Salmonella has changed through the years and presently is comprised of

only two species (enterica and bongori). S. enterica consists of six subspecies and each

one contains multiple serotypes. Some Salmonella serotypes, dublin and typhimurium

affect cattle and some, cholerasuis and typhimurium affect pigs and others, pullorum and

gallinarum affect poultry. Salmonella enterica subsp. enterica serotype typhi and

paratyphi A or B are human specific and can cause typhoid fever. The genus Salmonella

is composed of more than 2300 serotypes. The main antigens used to distinguish

between its serotypes are the somatic (0), flagellar (H), and capsular (K). Salmonella

spp. has a wide occurrence in the natural environment. Intense husbandry practices in the

meat, poultry, fish and shellfish industry, along with the recycling of offal into animal

feed have favored the presence of this pathogen in the global food chain. This pathogen

is a part of the microflora of many animals like chicken, cattle and reptiles. The

predominance of Salmonella spp. in the poultry and egg industry has overshadowed its

importance in meat, such as pork, beef and mutton. As a means to address this, the

USDA FSIS published the "Final Rule on Pathogen Reduction and Hazard Analysis and

Critical Control Points (HACCP) in July 1996. As part of this rule, FSIS will conduct

Salmonella testing to ensure the HACCP plan is helping to control this pathogen in

finished products (11, 12, 43).

This pathogen is encountered through the fecal oral rout and requires a high

infectious dose of 10 to 100 million organisms. Human infection with Salmonella can

lead to many clinical conditions which include enteric (typhoid) fever, uncomplicated

enterocolitis, and systemic infections by non-typhoid microorganisms. Uncomplicated

gastroenteritis caused by Salmonella enterica subsp. enterica caused secretary diarrhea,









nausea and vomiting. Salmonella enterica subsp. enterica serotype cholerasuis and

typhimurium cause focal infection of the vascular endothelium resulting in damage of the

intestinal mucosa. Enteric (typhoid) fever is a serious human disease which is associated

with typhoid and paratyphoid serotypes. The pathogen penetrates the mucosal barrier

potentially at the M-cell surface (Peyers Patch), the apical membrane of the intestinal

epithelial cells or the junction between the cells. A ruffling effect of the plasma

membrane is observed and a cytoskeletal rearrangement leads to the uptake of the

organisms into a phagocytic vesicle. Salmonella travels to the basal membrane and the

organisms are released into the lamina propria. Bacteremia may result leading to the

invasion of the gallbladder, kidneys and reinvasion of the gut mucosa (43).

Cases of Salmonella in cattle have been widely reported. Infected animals can shed

the organism without showing clinical signs of disease and can carry this organism into

the slaughtering facility. The presence of Salmonella in or on the cattle at the slaughter

facility along with the possibility of cross contamination of edible carcass is a significant

food safety hazard. Salmonella enterica subsp. enterica serotype dublin and typhimurium

are the most common serotypes in cattle. The prevalence of Salmonella in feces, rumen

and carcass samples was assessed by McEvoy et al. (29) at a commercial beef

slaughtering plant in Ireland. Salmonella was found in 2% of feces, 2% of rumen and

from 7 to 6% of carcass samples. Salmonella enterica subsp. enterica serotype dublin

represented 72% of the positive isolates and Salmonella enterica subsp. enterica serotype

typhimurium represented 14% of the positive samples (29). According to Rivera-

Betancourt et al. (39), hide removal and evisceration are procedures where pathogens

may be transferred onto carcasses. Salmonella has been found on carcasses after hide









removal and other slaughtering and dressing processes (39). According to Barkocy-

Gallagher et al. (3) Salmonella was less prevalent in feces (4.72%) than on the hide

(71.0%) or on preevisceration carcasses (13.38%) in three large beef processing plants in

the U.S. Salmonella had a prevalence of 0.1% on postintervention beef carcasses.

Salmonella prevalence rates of 15.4 to 86.9% on beef hides have been obtained in

previous U.S. studies (3).

Shigella spp.

.\/nge//A species are facultative anaerobic nonmotile straight rods with an optimum

growth temperature of 370C. This genus consists of four species (dysenteriae,flexneri,

boydii, and sonnei) which are serologically distinguished by the 0 antigen of their

lipopolysaccharide (LPS). .\li/nge/ dysenteriae causes the most serious illness

(dysentery) and.\/nge//t sonnei causes a mild illness (water diarrhea). The transmission

is from human to human although contamination through food or water contaminated

with feces also occurs. The infective dose is very small (100 to 1000 cells) and can easily

be transmitted from human to human. .l/nge/la spp. is not commonly associated with any

specific foods, but some outbreaks have occurred in potato salad, chicken and shellfish.

This pathogen is usually introduced into the food supply by an infected person like food

handlers with poor personal hygiene (12, 19, 43).

Shigellosis is characterized by the production of watery to bloody diarrhea or

dysentery. The dysentery stage of disease is characterized by the extensive bacteria

colonization of the colonic mucosa. The organism invades the epithelial cells at the M

cells (Peyer Patch). The bacteria are release into the lamina propria where they are

ingested by macrophages. They then reenter the mucosal epithelial cells by rearranging

actin and cytoskeletal elements. Once inside they escape the phagosomal vesicle and









multiply. As the membrane breaks the organisms is released into the host's cytoplasm

and the cell to cell invasion continues. Abscesses and mucosal ulcerations are produced

at infected sites, therefore causing blood, pus and mucus in stools (12, 43).

Klebsiella spp.

Klebsiella species are facultative anaerobic nonmotile encapsulated straight rods

(0.3 to 1.0 [tm by 0.6 to 6.0 [tm) that are arranged singly, in pairs or in short chains.

Klebsiella species have an optimum growth temperature of 370C. This organism can be

found in animal and human feces, soil, water, sewage, grains, fruits and vegetables. One

to 6% of healthy people carry Klebsiella in their throat or nose. Klebsiella species

include ornithinolytica, oxytoca, planticola, pneumoniae, pneumoniae subsp.

pneumoniae, pneumoniae subsp. ozaenae, pneumoniae subsp. rhinoscleromatis, and

terrigena. Klebsiella pneumoniae and oxytoca and sometimes other species are

opportunistic pathogens which can cause bacteremia, pneumonia, urinary tract infections

and other human infections. The ability of Klebsiella to cause gastroenteritis is unclear

(19, 20, 28).

Serratia spp.

Serratia species are facultative anaerobic straight rods (0.5 to 0.8 [im by 0.9 to 2.0

[tm). Serratia species are usually motile by peritrichous flagella and have an optimum

growth temperature of 30 to 370C. This organism can be found in human clinical

specimens, soil, water, plant surfaces, environmental sites, and the digestive tract of

rodents and insects. Serratia species include entomophila,ficaria,fonticola, grimesii,

liquefaciens, marcesans, odorifera, plymuthica, rubidaea. In domestic animals Serratia

has been isolated from pigs, cows, horses, rabbits and poultry. Serratia marcesans and

liquefaciends are prominent opportunistic pathogens causing septicemia and urinary tract









infections in hospitalized humans. Serratia in humans are nearly exclusively acquired

nosocomially and infection usually only involves Serratia marcesans. This organism

causes mastitis in cows and other animal infections (19, 20, 28).

Enterobacter spp.

Enterobacter species are facultative anaerobic straight rods (0.6 to 1.0 [im by 1.2 to

3.0 [tm) that are usually motile by peritrichous flagella (except: asburiae) and have an

optimum growth temperature of 30 to 370C. This organism is widely distributed in

nature and can be found in animal and human feces, soil, water, sewage, plants and

vegetables. Enterobacter species include ',,'///in'ians, aerogenes, amnigenus, asburiae,

cancerogenes, cloacae, dissolvens, gergoviae, hormaechei, intermedius, kobei,

nimipressuralis, pyrinus and sakazakii. This organism may also be isolated from farm

animals, primates, fish and insects. Enterobacter is frequently isolated from foods like

meats, dairy products, poultry and vegetables, since it is common to plants (crops) and

animals. Beef and pork products like ground beef, pork sausage, pork loin, and beef

steaks are common reservoirs of this organism. Enterobacter cloacae, aerogenes and

cip'h1vinci ii\ll are present in 2.6% to 31.2% of ground beef, packaged pork sausage,

vacuum packaged beef trim and primal cuts just before retail manipulation. Enterobacter

cloacae, sakazakii, aerogenes, agg1hiinciin\, and gergoviae are opportunistic pathogens

that cause burn, wound, and urinary tract infections and sporadically cause septicemia

and meningitis (19, 20, 28).

Citrobacter spp.

Citrobacter species are facultative anaerobic straight rods (2 to 6 [tm in length) that

are arranged singly and in pairs. Citrobacter are usually motile by peritrichous flagella

and have an optimum growth temperature of 370C. This organism occurs in soil, water,









sewage, foods, and in the intestinal tract of various animals. Citrobacter has been

recovered from the intestinal tract of cows, horses, dogs, cats and birds. It is an

increasingly important pathogen in fish. Citrobacter species include amalonaticus,

braakii, farmer, freundii, koseri, rodentium, sedlakii, werkmanii, and youngae. Most

infections with this organism are nosocomially acquired. Approximately 1% to 2% of

bacteremias, pneumonias, and wound and urinary tract infections are caused by

Citrobacter species. Citrobacter freundii and koseri have the potential to colonize

humans and are associated with infection (19, 20, 28).

Pantoea spp.

Pantoea species are facultative anaerobic straight rods (0.5 to 1 [tm by 1 to3 [tm)

motile by peritrichous flagella that have an optimum growth temperature of 300C. This

organism has been isolated from soil, water, seeds, and from animal and human wounds,

blood and urine. Pantoea species include (uglunwi'i/an\, ananatis, citrea, dispersa,

punctata, stewartii, and terrea. Pantoea (ugh1inci.lu an. is considered an opportunistic

human pathogen. The genus name Pantoea means all sorts of sources or from diverse

geographical samples (19, 20, 28).

Morganella spp.

Morganella species are facultative anaerobic straight rods (0.6 to 0.7 [tm by 1 tol.7

[tm) motile by peritrichous flagella that have an optimum growth temperature of 370C.

This organism occurs in the intestinal tract of dogs, other mammals, and reptiles.

Morganella species include morganii subsp. morganii and morganii subsp. sibonii.

Morganella morganii subsp. morganii is an opportunistic secondary invader that has been

associated with bacteremia, respiratory tract, wound, and urinary tract infections in









humans. Limited evidence does not support the role of Morganella morganii subsp.

morganii in diarrheal diseases (19, 20, 28).

Kluyvera spp.

Kluyvera species are facultative anaerobic straight rods (0.5 to 0.7 [tm by 2 to 3

[tm) motile by small peritrichous flagella that have an optimum growth temperature of

370C. This organism occurs in food, soil, sewage, human respiratory tract, feces and

urinary tract. Kluyvera species include ascorbata, cochleae, cryocrescens and georgiana.

This organism is an infrequent opportunistic pathogen (19, 20, 28).

Cedecea spp.

Cedecea species are motile facultative anaerobic round shaped rods (0.5 to 0.6 [tm

by 2 to 2 [tm) that have an optimum growth temperature of 370C. This organism has

been isolated from humans. More than 50% of the cases involving Cedecea have been

isolated from the respiratory tract. Cedecea species include davisae, lapagei and neteri.

This organism is an infrequent opportunistic pathogen. Little epidemiological

information about this organism is currently available (19, 20, 28).

Leclercia spp.

Leclercia species are facultative anaerobic straight rods that are motile by

peritrichous flagella and have an optimum growth temperature of 370C. This organism

has been isolated from a variety of foods, water, and environmental sources. Leclercia

adecarboxylata is an occasional opportunistic human pathogen. This organism has been

classified with the Enterobacteriaceae family as Escherichia adecarboxylata and

Enterobacter (1-.h's1',in'/i//\ before being given its own genus with a single species. A

battery of conventional biochemical tests and computer assisted probability matrix has

clearly dismissed its association with Escherichia and Enterobacter (19, 20, 28).









Non Enterobacteriaceae

Acinetobacter spp.

Acinetobacter species are nonsporing aerobic/microaerophilic rods (0.9 to 1.6 [im

by 1.5 to 2.5 [tm) that become spherical in their stationary phase of growth. They

commonly occur in pairs and in chains of different length. This organism stains Gram

negative but can be rather difficult to destain. The optimum growth temperature is 33 to

350C but Acinetobacter species can grow between 20 and 300C. This organism grows in

soil, water and sewage. This spoilage microorganism is able to grow on aerobically

stored refrigerated muscle foods. Acinetobacter species include baumanii,

calcoaceticus, haemolyticus, johnsonii, junii, iwoffii, and dioresistens. Acinetobacter is

considered to be nonpathogenic but has been known to cause septicemia, meningitis,

endocarditis, pneumonia, and urinary tract infections in humans (12, 19, 26, 28).

Moraxella spp.

Moraxella species are aerobic/microaerophilic rods (1 to 1.5 [im by 1.5 to 2.5 [im)

or cocci (0.6 to 1 [tm in diameter) that commonly occur in pairs and in short chains and

can present capsule. This organism stains Gram negative but can resist decolorization.

The optimum growth temperature for this organism is 33 to 350C. Moraxella species

include atlantae, boevrei, bovis, caprae, equi, lacunata, lincolnii, nonliquefaciens,

osloensis, ovis, and saccharolytica. This spoilage microorganism grows on aerobically

stored refrigerated muscle foods. It can be found in the mucus membranes of humans

and animals (12, 19, 26, 28).

Chryseonmonas_spp.

Chryseomonas species are aerobic/microaerophilic rods with parallel sides and

rounded ends. They are motile by multitrichous polar flagella and grow in a temperature









range of 18 to 480C. This organism is not know to be present in the general environment

but can be a saprophyte or commensal of humans and animals in which they may be

pathogenic. The single Chryseomona species is luteola (19, 28).

Flavimonas spp.

Flavimonas species are aerobic/microaerophilic rods with parallel sides and

rounded ends. They grow in a temperatures ranging from 18 to 480C and are know to be

present in the general environment. They can be saprophytes or commensals of humans

and animals in which they may be pathogenic. The single Flavimonas species is

oryzihabitans (19, 28).

Pseudomonas spp.

Pseudomonas species are aerobic/microaerophilic straight or slightly curved rods

(0.5 to 1 [tm by 1.5 to 5.0 [tm) that are motile by one or several polar flagella. There are

more than 76 species which are widely distributed in nature and some species are

pathogenic to humans, animals or plant. Pseudomonas species are normally associated

soil and fecal mater and therefore are found on the hide of beef and pork. This spoilage

microorganism is able to successfully grow on aerobically stored refrigerated muscle

foods, but its growth is suppressed under vacuum and modified atmosphere storage (12,

28).

Stenotrophomonas spp.

Stenotrophomonas maltophilia is an aerobic/microaerophilic short to medium sized

rod that is nonsporing and nonencapsulated. This organism is motile by polar tuft

flagella and has an optimum growth temperature of 35 to 370C. This organism used to be

classified as a Pseudomonas and is an organism of low virulence with limited ability to









cause infection in humans. Stenotrophomonas maltophilia is a water organism that can

survive and multiply in aquatic environment for extended periods (12, 19, 28).

Chryseobacterium indologenes

Chryseobacterium indologenes is also known as Flavobacterium indologenes

which is an aerobic/microaerophilic rod with parallel sides and rounded ends. It is

nonsporing and nonencapsulated and has an optimum growth temperature of 370C (19,

28).

Gram Positive Bacteria

Listeria spp.

The genus Listeria belongs to the Bacilli class along with Bacillus, Staphylococcus,

Streptococcus, Lactobacillus, and Brochothrix. It is a Gram positive non-spore forming,

facultative anaerobic rod that has no capsule. It has regular shaped short rods measuring

0.4 to 0.5 [m by 0.5 to 2[tm with rounded ends. Listeria spp. is motile by a few

peritrichous flagella when grown at 20 to 250C. This organism's optimum growth

temperature is 30 to 370C but can grow in temperature ranging from 0 to 450C. The

Listeria genus consists of six species but Listeria monocytogenes (LM) and Listeria

ivanovii are the only potential pathogens. Listeria seeligeri is considered nonpathogenic,

although it has been implicated in some cases of human listeriosis. Listeria spp. can be

isolated from a vast amount of environmental sources like: soil, water, effluents, a variety

of foods, and feces of humans and animals. Foods most often implicated in listeriosis are

soft cheeses and other dairy products, meat products, salads and ready to eat products not

necessarily reheated before eating (12, 19, 51).

Listeriosis is an illness of major public health concern, because of the severity of

the disease resulting in abortion, meningitis and septicemia. Very few people are









infected annually, about two to 10 cases per million population in Europe and United

States. LM seems to have a lower pathogenic potential than other food-borne pathogens

with a suspected minimum dose requirement of 106 cells. It has a high case-fatality rate

of approximately 20 to 30%, presents a long incubation time in the host, and has an

affinity for individuals who have underlying conditions that lead to impairment of T-cell

mediated immunity. The susceptibility of the host plays an important part in the

presentation of clinical diseases upon exposure to LM. Those groups most at risk are

pregnant women and neonates, immunocompromised adults, and the elderly (>60 years

old). The incidence of listeriosis in pregnant women is estimated to be 12 per 100,000,

which is high compared to 0.7 per 100,000 in the general population. Listeriosis has a

fondness for expectant mothers, showing an estimated 17-fold increase in incidence. The

severity and high mortality rate (20 to 30%) of LM infected neonates encourages rapid

isolation, diagnosis, treatment and overall understanding of this bacterium in order to

reduce the likelihood of infection of pregnant women, their babies, and

immunocompromised and healthy people (12, 31, 51).

Fetal and neonatal listeriosis results from the invasion of the fetus via the placenta

and progresses into chorioamnionits, or ascending from the colonized vaginal canal. It is

an inflammatory process involving the chorion, fetal blood vessels, the umbilical cord,

and the amnion by extension of the inflammation. The inflammatory process is

potentially fatal to mother and fetus. The consequences are abortion, usually five or more

months into gestation, a stillborn fetus or the birth of a baby with a generalized infection

characterized by pyogranulomatous microabscesses all over the body. The infection is

usually asymptomatic in the pregnant mother, but may present mild flu-like symptoms









like chills, fatigue, fever, headaches and muscle and joint pain about 2 to 14 days before

the miscarriage (31, 51).

The listeriosis infection most often associated with non-pregnant adults is that of

the central nervous system (CNS). Infection normally evolves into a

meningoencephalitis, which is accompanied by drastic changes in consciousness,

movement disorders, and paralysis. The mortality rate of CNS listeriosis is about 20%

but can be as high as 40 to 60% in immunocompromised adults. Bacteremia or

septicemia is another form of listeriosis which can have a 70% mortality rate if associated

with debilitated conditions. Other less frequent clinical forms are endocarditis,

myocarditis, arteritis, pneumonia, pleuritis, hepatitis and peritonitis. Listeria food-borne

outbreaks are associated with febrile gastroenteritis which includes: diarrhea, vomiting,

and fever. The cutaneous form of Listeria infection is characterized by a

pyogranulomatous rash (51).

The major source of infection is through the ingestion of contaminated foods. The

primary site of entry of Listeria species into the host is through the gastrointestinal tract.

Ingestion of LM is thought to be very common, because of its ubiquitous distribution and

its probable presence in processed foods. Ingested bacteria must resist acid conditions in

the stomach before they can reach the intestine. Gastric acidity kills a considerable

number of Listeria organisms that are ingested in contaminated foods. LM goes through

the intestinal epithelium by penetrating through the apical surface of polarized Caco-2

cells with a brush border. Other studies have shown that LM penetrates the host via M

cells overlying the Peyer's patch. Organisms that penetrate the intestinal wall proceed to

invade neighboring enterocytes which leads to enteritis. LM survives intracellularly by









preventing phagosome maturation. It is an invasive pathogen that becomes internalized

into many different types of cells that are not phagocytic such as epithelial and

endothelial cells, hepatocytes, fibroblasts and nerve cells. Once internalized it escapes

the double membrane of the phagosome, by means of hemolysin and phospholipases, into

the cytosol of intestinal cells where it multiplies. During multiplication in the cytosol,

Listeria is surrounded by actin filaments which propel it to move from cell to cell where a

new round of intracellular proliferation commences. LM organisms that cross the

intestinal barrier are internalized by resident macrophages, in which they can survive and

replicate, and are carried through the lymph or blood to the spleen and liver. The initial

stage of host tissue colonization is rapid, and it is followed by long a incubation period in

which LM establishes a systemic infection. Bacteria are cleared from the bloodstream by

macrophages in the spleen and liver. More than 90% of the bacteria are accumulated in

the liver Kupffer cells (16, 24, 31, 51).

The entry ofLM into slaughtering facilities occurs through soil on workers'

clothing and shoes, equipment transport, feces and contaminated hides of animals, and

human carriers. LM growth is favored by moist environments and is most often detected

in floor drains, condensed and stagnant water, floors and processing equipment. This

pathogen can readily attach to many surfaces and create biofilms on various meat and

dairy processing equipment. LM has been known to survive on workers hands after

washing and also in aerosols. The presence of this organism on carcasses can be

attributes to contamination by fecal matter during slaughtering. Between 18 to 52% of

animals are healthy fecal carriers of LM. About 24% of cattle have contaminated internal

retropharyngeal nodes and 45% of pigs carry this organism in their tonsils. LM has been









recovered in clean and unclean zones in slaughterhouses, and the most contaminated

areas are the dehiding, and the stunning and hoisting areas of porcine slaughtering.

Postprocessing contamination is the most prevalent route of contamination of processed

foods. It is particularly hard to eliminate this organism from slaughtering plants because

of its ability to form biofilms on food contact surfaces. Refrigerated and moist

environments provided in slaughter facilities are good growth environments for L.M (12).

Chasseignaux et al. (7) identified LM in 38% of the samples from three pork and

two poultry processing plants in France. After cleaning, the contamination was lowered

to 7.7% in the pork plants and 13.1% in the poultry plants. Autio et al. (2) detected LM

in 9% of the pluck samples and Listeria innocua in 2% of the pluck samples in ten low

capacity pork slaughterhouses in Finland. The highest prevalence of this organism was

detected in tongue (14%) and tonsil (12%) samples and at lower amounts in hearts,

kidneys and livers (6%). This organism was detected in eight of the 10 slaughterhouses

and LM was detected in 7% of the environmental samples (2).

Staphylococcus spp.

Staphylococcus species are Gram positive nonmotile, nonspore forming, facultative

anaerobic cocci occurring in pairs or irregular clusters. Its cells measure 0.5 to 1.5 [tm in

diameter. This organism's optimum growth temperature is 30 to 370C and is one of the

hardiest nonspore forming bacteria that can survive extended periods of time on dry and

inanimate objects. This organism is relatively heat resistant which allows it to survive in

just about every environment in which humans coexist. The genus Staphylococcus is

divided into 23 species and subspecies (arlettae, aureus, auricularis, capitis,

chromogenes, cohnii, epidermidis, gallinarum, haemolyticus, hominis, hyicus,

intermedius, kloosii, lentus, lugdunensis, saprophyticus, schleiferi, sciuri, simulans,









warneri, and xylosus). The most ubiquitous of these species is Staphylococcus

epidermidis which is found on the skin of humans and animals and rarely causes disease.

The growth of Staphylococcus aureus in foods is a potential public safety hazard since

many of its strains produce enterotoxins that cause food poisoning when ingested

(Staphylococcal food poisoning). Staphylococcus intermedius and hyicus have also been

shown to produce enterotoxins in food. Staphylococcus saprophyticus is known only to

cause urinary tract infections. The coagulase test is used to differentiate Staphylococcus

aureus from the other species since it is the only one to produce the coagulase enzyme

(11, 12, 19).

Humans are the main reservoirs of staphylococci involved in human disease,

especially Staphylococcus aureus. The other species are considered to be normal

inhabitants of the external parts of the body. Staphylococcus aureus usually colonize the

external nostrils and are found in about 30% of healthy individuals. This species also can

be found on the skin, oropharynx and feces of humans and animals. Staphylococcus

aureus can be isolated from fomites produced by food processing such as meat grinders,

knives, saw blades and cutting boards or tables (11, 12,).

Staphylococcal food poisoning (SFP) is the fourth most common cause of food

poisoning when ranked according to outbreaks and the most common when ranked

according to the number of affected individuals. SFP is the classic example of short

incubation time food poisoning cause by a toxin produced in food. Staphylococci

produce many metabolites, but the enterotoxins pose the greatest risk to consumer health.

Staphylococcus aureus produces five distinct enterotoxins (type A through E) which are

single polypeptide proteins with a molecular size from 22 to 28 kDa. The growth of









Staphylococcus aureus in foods may lead to the production of sufficient enterotoxins

which may cause illness when these contaminated foods are consumed. These toxins

cause disease even in the absence of the organism. In the majority of cases SFP is

associated with food being contaminated by the food handler who might have a minor

Staphylococcus aureus infection such as a boil or cut. The contaminated food must be

permitted to sit at an adequate temperature that will allow the Staphylococci to multiply

and produce the toxin. Reheating the food before eating may kill the organism but does

not eliminate the heat stable toxin. Some of the foods commonly linked with SFP are

meat (beef, pork ad poultry), meat products (sausages, hotdogs, ham), salads ham,

chicken, potato), cream filled baked products and dairy products (8, 11, 12,).

The most common symptoms associated with the ingestion of Staphylococcal

enterotoxins are vomiting, diarrhea, abdominal cramps, headaches, muscle cramping and

prostration which usually occurs two to six hours after the toxin is ingested. The diarrhea

is usually watery but may also contain blood. Enterotoxins cause intensive peristalsis by

working directly with the vomiting control area of the brain. The patient usually does not

experience fever since there is no infection. The illness is relatively mild and usually

lasts only a couple of hours to one day. Staphylococcal enterotoxins display pyrogenic

(fever causing), shock inducing, and superantigen activity when entering the

bloodstream. The toxin, as it enters into the bloodstream, acts as a superantigen causing

the nonspecific activation of T-cells by binding to the variable region of the 0 chain of the

T cell receptor. This causes the proliferation of cytokine secretion leading to septic shock

(8, 11, 12,).









In raw food products (beef and pork carcasses) Staphylococcus aureus

contamination may not only originate from humans but also from animal hides, skin,

lesions and bruised tissue (11). Coagulase positive staphylococci (CPS) were found in

39.1% of the samples collected by Desmarchelier et al. (9) in three beef slaughterhouses

located in Australia. CPS was isolated from 68.3% of the hides, 3.3% on preevisceration

carcasses, 38.9% postevisceration carcasses, and 43% of postintervention carcasses.

After overnight chilling the incidence of CPS decreased to 33% but rose to 83% after

three days of chilling. The incidence of CPS when sampling the flank, brisket, and round

of twenty carcasses was 40%, 35% and 5% respectively. It was observed that the

workers manipulated the areas of the flank and brisket extensively during the removal of

the hide and internal organs. CPS were isolated from seven often air samples taken

during that study (9).

Bacillus spp.

Bacillus species are endospore forming rod shaped cells (0.5 to 2.5 [tm by 1.2 to

1.0 [tm) and often arrange in pairs or chains. The cells are motile by peritrichous flagella

and there is only one spore produced per cell where the exposure to oxygen does not

repress sporulation. This organism is a common soil saprophyte and can easily

contaminate many types of food like pasta, rice, meat, eggs and dairy products. The

major food pathogen is Bacillus cereus which causes two types of foodborne illnesses:

the diarrheal type and the emetic type. The diarrheal type is caused by an enterotoxin

produced by Bacillus cereus, during its vegetative state, in the small intestine. The

emetic type of foofborne illness is caused by the organism growing in the food and

producing the toxin. Spores that survive the food being heated are the source of the food

poisoning. Heat treatment causes the spores to germinate and in the absence of









competitive flora Bacillus cereus grows very well. Paenibacilluspopilliae used to be

classified as Bacillus popilliae but now is a taxonomic dissection of the Bacillus genus

(12, 19).

Brevibacterium spp.

Young cells of Brevibacterium species are irregular rods (0.6 to 1.2 [im by 1.6 to 5

[tm) often arranged singly or in pairs and often have V formations. Older cultures yield

rods that segment into small cocci. This organism is nonsporing and nonmotile and its

optimum growth temperature is 20 to 350C. Brevibacterium species include case,

epidermidis, frigoritolerans, halotolerans, iodinum, linens, liquefaciens, mcbrellneri,

otitidis and stations. This organism is widely distributed in dairy products and is also

found on human skin (19, 28).

Cellulomonas spp.

Young cells of Cellulomonas species are slender irregular rods (0.5 to 0.6 [im by 2

to 5 [tm) often arranged in pairs or have V formations. Older cultures yield rods that are

short and may segment into small cocci. This organism is often motile by single flagella

and is nonsporing, having and optimum growth temperature of 300C. There are about ten

Cellulomonas species that are widely distributed in soils and decaying plant matter (19,

28).

Brochothrix spp.

Brochothrix species are nonsporing nonmotile nonencapsulated regular unbranched

rods (0.6 to 0.7 [tm by 1 to 2 [tm) that occur singly, in chains or in long filamentous

chains that form knotted masses. This species occurs mainly in meat products and is

widely distributed in the environment. Brochothrix optimal growth temperature is 20 to

250C, but can grow from 0 to 300C. Brochothrix thermosphacta is a meat spoilage









microorganism that grows during refrigeration. This spoilage microorganism, along with

lactic acid bacteria, predominates in vacuum packaged and modified atmosphere

conditions (12, 19).

Microbacterium spp.

Young cells of Microbacterium species are slender irregular rods (0.4 to 8.0 [im by

1 to 4 [tm) often arranged singly or in pairs and may have V formations. Older cultures

yield rods that are short and may segment into small cocci. This organism is nonmotile

or motile by one to three flagella, nonsporing and has and optimum growth temperature

of 300C. Microbacterium species are found in dairy products, sewage and insects (19,

28).

Micrococcus spp.

Micrococcus species are strictly aerobic spherical cells (0.5 to 2 [im in diameter)

that occur in pairs, tetrads, or irregular clusters, but not in chains. This organism is

usually not motile and nonsporing and its optimum growth temperature is 25 to 370C.

Micrococcus species occur mostly on mammalian skin and in soil but are frequently

isolated from food products and the air. Kocuria kristinae and Nesterenkonia halobia

used to be classified as Micrococcus kristinae and Micrococcus halobia but now are a

taxonomic dissection of the Micrococcus genus. This organism is frequently isolated

from beef hides along with Staphylococcus, and Pseudomonas (12, 19).

Bioaerosol research in the meat slaughter and processing area has been limited to

determining the levels of airborne bacteria and yeast and mold contaminants. Limited

information exists about the bacterial composition of bioaerosols created in a beef and

pork slaughtering facility. Information about the existence of pathogenic microorganisms

in bioaerosols and the possibility of their transfer to food products is of great importance






54


to food safety. The main objectives of this study were to enumerate total airborne

bacteria and yeast and mold contaminants and determine the presence of Escherichia coli

0157:H7, Listeria spp., Salmonella spp., and Staphylococcus spp. in bioaerosols

generated in a small scale slaughter facility and on pork and beef carcasses.














CHAPTER 3
MATERIALS AND METHODS

Air Sampling System

Air samples collected in this study were obtained from the United States

Department of Agriculture (USDA) inspected slaughter facility at the University of

Florida, Gainesville, FL. Duplicate air samples were taken before and during pork and

beef slaughtering processes at the bleeding area, hide removal/dehairing area, back

splitting area and the holding cooler.

Air samples were collected during three separate pork and beef slaughtering days

using the N6 single stage impactor (Thermo Andersen Corp., Smyrna, GA). Each beef

slaughter consisted of five to 10 steers and/or cows and each pork slaughter consisted of

20 to 25 pigs. The air sampling areas selected for beef and pork slaughtering processes in

this study are presented in Table 1.

Table 1. Air sampling areas for beef and pork slaughtering processes
Sampling Area Beef Pork

Before Slaughter
A Bleeding Pit Bleeding Pit
B Hide Removal Scalding Tank
C Back Splitting Back Splitting
D Cooler (hot box) Cooler (hot box)
During Slaughter
E Bleeding Pit Bleeding Pit
F Hide Removal Scalding Tank
G Back Splitting Back Splitting
H Cooler after 24 hr. Cooler after 24 hr.









The N6 single stage impactor consists of an aluminum inlet cone, jet stage, a base

plate held together by three spring clamps and sealed with O-ring gaskets, and a vacuum

pump. The jet stage or sampling stage consists of 400 precision machined orifices. The

vacuum pump pulls air from the environment through the inlet cone and jet stage and

onto a Petri dish containing agar nutrient media. This single stage viable impactor

requires a flow rate of exactly 28.3 liters/minute (1 cubic foot/ minute) to achieve a cut-

off diameter of 0.65 rm (46). The impactor was shipped to Thermo Andersen

Corporation (Smyrna, GA) for adjustment to a flow rate of 28.3 0.1 liters/minute before

sampling. According to Lutgring et al. (27), the single stage viable impactor sampler

should be calibrated with a primary flow meter and adjusted to a flow rate of 28.3 0.1

liters/minute on a regular basis. Sampling time can range from 10 seconds to 10 minutes

according to the level of airborne contamination in the slaughter facility (27).

The impactor was washed at the beginning of each slaughter with soapy water and

disinfected with 1000C distilled water containing a chlorine concentration of 100 parts

per million (7.1 ml per 3.8 liters of water). The jet stage was checked for clogged holes

and the impactor was then air dried. At the beginning of each slaughter the inlet cone, jet

stage, and base plate were disinfected with alcohol swabs (Fisher Scientific, Pittsburgh,

PA). The impactor was also disinfected with alcohol swabs when changing from

sampling point to sampling point (A through H). The sampler was turned on for two

minutes prior to sampling to allow the alcohol to evaporate and not affect the amount of

bacteria recovered. Air samples were taken at a height of 1.5 meters from the floor and

within one meter from the carcass. Sterile Petri dishes containing non-selective media

(TSA, PDA) (Difco Inc., Detroit, MI) and selective media (VRBA, MOX, XLT4, MSA)









(Difco Inc., Detroit, MI) were inserted, without their top, into the disinfected N6 single

stage impactor and the vacuum was turned on for one minute at a flow rate of 28.3 0.1

liters/minute. After one minute the sampler was turned off and the Petri dish was

removed and inverted in its cover. Petri dishes were stored in sterile bags to prevent

contamination before further analysis. Direct air samples were collected in duplicates for

each sampling point (A through H).

Air Sample and Carcass Swab Collection

Preliminary Studies

Work was conducted to determine the time and volume of air to be sampled with

the N6 single stage impactor to attain appropriate growth on Petri plates and achieve a

representative sample. Sampling times of 30 seconds, two, five, seven and 10 minutes

were tested. It was determined that sampling times between 30 seconds and two minutes

with a flow rate of 28.3 0.1 liters/minute yielded countable bacterial growth on Petri

plates (i.e. less than 250 CFU). Petri dishes measuring 15 mm by 95 mm (Fisher

Scientific, Pittsburgh, PA) were filled with 20 ml of media yielding a plate height of 7

mm. The air, in the areas listed in Table 1, was sampled for total aerobic bacteria and

total yeast and mold using tryptic soy agar (TSA) and potato dextrose agar (PDA),

respectively. The ability to culture Escherichia coli 0157:H7, Listeria spp., Salmonella

spp., and Staphylococcus spp. onto their respective selective media and enrichment broths

was determined. Confirmed cultures ofEscherichia coli 0157:H7, Listeria spp.,

Salmonella spp., and Staphylococcus spp. were streaked onto violet red bile agar

(VRBA), modified oxford agar (MOX), xylose-lysine-tergitol 4 agar (XLT4) and

mannitol salt agar (MSA), respectively and typical colony growth was confirmed. In an









attempt to isolate these bacteria from air samples, they were collected onto prepoured

plates containing the previously discussed selective media.

Carcass Sample Collection

Pork and beef carcass surface bacterial samples were collected using a sponging

technique. Swabs were taken from five different carcass sides after pork and beef

slaughters. Carcasses were swabbed 12 hours after being in the holding cooler at 00C.

Beef and swine carcass sampling kits (Solar Biologicals Inc., Ogdensburg, NY) were

used to collect microbial surface samples. Each individual sterile kit consisted of a pre-

moistened sponge, 15 ml Butterfield's phosphate buffer, disposable template, sample

bags and gloves. Butterfield's buffered phosphate stock solution (BBPS) consisted of 34

g of potassium phosphate monobasic and 500 ml of distilled water. Beef carcass sides

were sampled at flank, brisket and rump. Pork carcass sides were sampled at the belly,

jowl and ham. Samples were taken in the specific order stated above. Carcass swab

samples were collected by wiping the 100 cm2 area inside the template with a sterile pre-

moistened sponge 10 times horizontally and 10 times vertically. The wiping force was

sufficient to remove dried blood. The flank and brisket (beef) areas and the belly and

jowl (pork) areas were swabbed with one side of the sponge first. Next, the other side of

the sponge was used to swab the rump (beef) and the ham (pork) 10 times horizontally

and 10 times vertically. Individual sponges were placed into sample bags along with 15

ml Butterfield's phosphate buffer. The bags were closed and kneaded several times.

After samples were collected they were stored in a walk in cooler at 5 to 60C until they

were analyzed (48).









Air and Carcass Microbiological Sample Analysis

Procedures for culturing and isolating bacteria from food and water samples were

conducted as outlined in the Microbiology Laboratory Guidebook (MLG) (49) and

Bacteriological Analytical Manual (BAM) (50). Since there were no AOAC

International (Association of Analytical Communities) accepted methods for the

culturing and isolating bacterial pathogens from air samples, the above manuals for food

and water analysis were adapted for air sample analysis. Air samples collected on Petri

dishes filled with non-selective media (TSA, PDA) and selective media (VRBA, MOX,

XLT4, MSA) were incubated for specific times and temperatures. TSA plates were

incubated at 350C for 24 hours and PDA plates were incubated at 250C for 3 days.

VRBA, MOX, XLT4 and MSA plates were incubated at 350C for 24 hours, 35C for 24

to 48 hours, 35C for 24 hours and 350C for 24 hours, respectively (49). The USDA

through the MLG and FDA through the BAM suggest incubating TSA for 48 hours at

350C. In preliminary work, air samples collected on TSA plates were incubated for 48

hours at 350C and some colony drying and hardening was observed. This colony

hardening and drying was negatively affecting further analysis of the air samples. In this

study sufficient colony growth was attained after incubating TSA plates for 24 hours at

350C. Direct enumeration of TSA, PDA, VRBA, MOX, XLT4 and MSA plates was

conducted after incubation.

Sterile conditions were maintained during all microbiological analyses in order to

prevent the contamination of samples and isolates obtained from the air samples.

Previously enumerated TSA plates were then used for detection of Escherichia coli

0157: H7, Listeria spp., Salmonella spp., and Staphylococcus spp. Air samples were

enriched by flooding TSA plates with 0.1% peptone water (1.0 g peptone per 1 liter of









distilled water) and then transferring Iml of liquid to enrichment broths. Ten milliliters

of 0.1% peptone water (Difco Inc., Detroit, MI) were added to each TSA plate with a

sterile pipette. These plates sat approximately five minutes for bacterial colonies to

soften and separate from the media. Peptone flooded plates were spun and cell spreaders

were used to separate bacterial colonies from the media and create a homogeneous

bacterial mixture. This liquid bacterial slurry was then transferred to the various pre-

enrichment and enrichment broths. One milliliter of the liquid bacterial slurry was

transferred to duplicate test tubes containing 9 ml of enrichment broths.

Enrichment broths used to increase bacterial concentrations were Modified EC plus

Novobiacin (mEC) for Escherichia coli 0157: H7; University of Vermont Broth (UVM)

and Fraser broth (FB) for Listeria spp.; Lactose broth (LB) and Rappaport Vassiladis

broth (RV) for Salmonella spp., and Brain heart infusion broth (BHI) for Staphylococcus

spp. The enrichment broths mEC, UVM, FB, LB, RV, and BHI (Difco Inc., Detroit, MI)

were incubated at 350C for 24 hours, 300C for 24 hours, 350C for 24 to 48 hours, 370C for

18 hours, 370C for 18 hours and at 350C for 24 hours, respectively (49). Pre-enrichment

and enrichment steps were conducted for Listeria spp. and Salmonella spp. analyses for

pork and beef slaughters one and two. Only the final enrichment step was conducted for

slaughters three through six. Pre-enrichment steps for Listeria spp. and Salmonella spp.

were omitted in slaughters three through six because of the lengthy time involved in

sample analysis. There were no notable differences in the recovery rates of Listeria spp.

and Salmonella spp. when their pre-enrichment steps were omitted. In slaughters one and

two, after the pre-enrichment broths had been incubated, the test tubes were vortexed and

1 ml was transferred to 9 ml of enrichment broths and these were vortexed and incubated.









After concluding the enrichment step, test tubes were vortexed and an inoculation loop

was used to transfer a loop full of broth onto duplicate selective media plates containing

VRBA, MOX, XLT4 and MSA in an attempt to isolate Escherichia coli 0157:H7,

Listeria spp., Salmonella spp., and Staphylococcus spp. respectively. A three-way-streak

was conducted in order to achieve optimum colony isolation. Selective media plates

were incubated for the previously specified temperatures and times.

Bags containing carcass swab samples were removed from the cooler (5 to 60C)

less than one hour after sample collection and analyzed for presence of Escherichia coli

0157: H7, Listeria spp., Salmonella spp., and Staphylococcus spp. The sponge and

contents of the bags were kneaded several times to thoroughly mix samples. One

milliliter was aseptically transferred from the bags to test tubes containing 9 ml of the

enrichment broths: Modified EC plus Novobiacin (mEC), Fraser broth (FB), Rappaport

Vassiladis broth (RV), and Brain Heart Infusion broth (BHI). The above enrichment step

was conducted in duplicate and incubated at the previously discussed temperatures and

times. Incubated test tubes containing the various enrichment broths were vortexed and

an inoculation loop was used to transfer a loop full of broth onto duplicate selective

media plates previously discussed to isolate Escherichia coli 0157:H7, Listeria spp.,

Salmonella spp., and Staphylococcus spp. A three-way-streak was conducted to obtain

individual colonies. Selective media plates were incubated for the previously specified

temperatures and times.

Individual colonies from the enriched air samples, enriched carcass samples and

colonies from direct air samples collected on selective media (VRBA, MOX, XLT4,

MSA) using the N6 single stage impactor were picked for further analysis. Colonies









were selected on the basis of their macroscopic or visual qualities. A representative

sample of individual colonies possessing different colors, shapes or sizes were selected

from specific selective media plates. Strict records were kept regarding sampling point

and the selective media plates these colonies came from. Colonies were picked from

selective media plates that contained air samples from as many sampling points (A-H) as

possible. The overall objective was to pick visually unlike colonies from different

selective media plates from various sampling points for further identification. Due to the

vast number of colonies being picked for further identification, some colonies had to be

preserved. Colonies were preserved in TSB broth containing 40% glycerol (Fisher,

Pittsburgh, PA). A 40% solution of glycerol in TSB was made and then autoclaved at

1210C for 15 minutes. Individual colonies were picked with a sterile inoculation loop

and stored in 2 ml CryoVials (Fisher Scientific, Pittsburgh, PA) containing 1 ml of 40%

glycerol TSB. CryoVials were vortexed, labeled and stored at -840C for a period of time

that did not exceed six months. According to Gorman and Adley (15) 40 to 50% glycerol

stocks are extensively employed in the preservation of bacteria such as Salmonella

typhimurium, Escherichia coli and Staphylococcus aureus. Campylobacter spp.

maintained its viability for up to seven months when stored in 10% glycerol at -200C

(15).

Bacterial Identification of Air and Carcass Samples

Colonies preserved in TSB broth containing 40% glycerol were taken out of the

-840C freezer for further analysis. These CryoVials were submerged in room temperature

water to thaw. Once the liquid inside the CryoVials reached room temperature they were

incubated at 350C for 48 hours. After 48 hours, CryoVials were checked for growth

(turbidity), and those exhibiting no growth were incubated for an additional 24 hours.









Incubated CryoVials exhibiting growth were vortexed and an inoculation loop was used

to transfer a loop full of broth onto duplicate selective media plates previously discussed

to isolate Escherichia coli 0157:H7, Listeria spp., Salmonella spp., and Staphylococcus

spp. A three-way-streak was conducted to obtain individual pure colonies that could be

used for further analysis. Selective media plates were incubated for the previously

specified temperatures and times. One milliliter was transferred from individual

CryoVials exhibiting growth to test tubes containing 9 ml of their appropriate enrichment

broths. These test tubes were incubated under the same conditions as discussed earlier.

An inoculation loop was used to transfer a loop full of broth onto duplicate selective

media plates previously discussed. A three-way-streak was conducted to obtain

individual pure colonies that could be used for further analysis.

Isolated colonies were observed and color, size, texture and color changes in the

media were recorded. Gram stains (Fisher, Pittsburgh, PA) were performed and colony

morphology was determined. Samples were classified as coccoid, bacilli or spherical in

shape (47). It was necessary to have a Gram identification and colony shape to be able to

select the appropriate API identification strip (bioMerieux Inc., Marcy 1'Etoile, France).

Microbiological Identification

The API testing system was used to identify Escherichia spp., Salmonella spp.,

Listeria spp., and Staphylococcus spp. API 20E was used to identify species/subspecies

of Enterobacteriaceae (Escherichia coli and Salmonella spp.) and/or non-fastidious Gram

negative rods. API Staph was used for overnight identification of Gram positive

Staphylococci spp., Micrococci spp., and Kocuria spp.

API is one of the most used diagnostic tools for identification of the

Enterobacteriacae family of bacteria. Initially, its database was mostly oriented toward









clinical isolates, but in recent years it has grown to include food, industrial and

environmental isolates also. The API testing system is made up of various small and

elongated wells which contain different dehydrated media, and is intended to perform

various biochemical tests from a pure culture grown on an agar plate (11).

A 24-hour isolated colony was placed into 5 ml of sterile suspension media (NaCl

0.85%) for API 20E or 6 ml of API Staph medium for the API Staph strip and vortexed to

achieve a homogeneous bacteria suspension. Cultures being identified were never more

than 24 hours old. It is recommended to use young cultures between 18 and 24 hours old

(4). Strips were placed in moist incubation boxes to prevent dehydration and incubated at

350C for 18 to 24 hours. All the wells reactions of the API 20E and API Staph strips

were recorded after the incubation period. Wells requiring the addition of reagents were

revealed and reactions of these strips were read by comparing to color change tables

provided for each test. Numerical values were obtained for each bacterial colony

analyzed and identification was obtained through the analytical profile index (4).

MIDI's Microbial Identification System (MIDI Inc., Newark, DE) which identifies

bacterial cells by gas chromatographic (GC) analysis of cellular fatty acids was used to

identify colonies growing on VRBA, MOX, XLT4 and MSA plates that had also been

identified by the API method. MIDI's Microbial Identification System (MIS) was also

used to identify atypical colonies growing on MOX plates. API Listeria can only be used

to identify Listeria species and will not identify any other bacteria growing in Listeria

enrichment broths or on selective media. Growth conditions of sample bacteria must be

carefully regulated to obtain valid and reproducible cellular fatty acid profiles (41). All









bacterial samples being identified were streaked onto TSA using the three-way-streak

method. Plates were incubated at 350C for 24 hours.

Five steps involved in preparing samples for gas chromatographic analysis were:

harvesting, saponification, methylation, extraction and base wash. Approximately 40 mg

of bacterial cells from the third quadrant of the three-way-streaked TSA plate were

harvested using a 4 mm inoculation loop, and placed in a clean 13x100 culture tube. One

milliliter of Reagent 1 (Saponification, 45 g sodium hydroxide, 150 ml methanol, and 150

ml distilled water) was added to each tube containing bacterial cells; they were securely

sealed with Teflon lined caps, vortexed briefly and heated in a boiling water bath for five

minutes. Tubes were then vigorously vortexed for 5-10 seconds and returned to the

boiling water bath for another 30 minutes. The tubes were cooled and uncapped and 2ml

of Reagent 2 (Methylation-325 ml certified 6.0 N hydrochloric acid and 275 ml methyl

alcohol) were added in order to lower the pH of the solution below 1.5 causing

methylation of the fatty acid. At this point the fatty acid methyl ester were inadequately

soluble in the aqueous phase. Tubes were capped and briefly vortexed and heated for 10

minutes at 800C. A 1.25 ml aliquot of Reagent 3 (Extraction-200 ml hexane and 200 ml

methyl tert-butyl ether) was added to the cooled tubes which were recapped and gently

tumbled using a clinical rotator for about 10 minutes. This reagent was added to extract

the fatty acid methyl esters into the organic phase for use with the gas chromatograph.

The tubes were later uncapped and the aqueous (lower) phase was pipetted out for

disposal. A base wash was performed by adding approximately 3 ml of Reagent 4

(Cleanup-10.8 g sodium hydroxide dissolved in 800 ml distilled water) to the organic

phase left behind in the tubes. Tubes were recapped and tumbled for five minutes. After









tumbling, about 2/3 of the organic phase was pipetted into a GC vial which was capped

and ready for analysis. This procedure reduced contamination of the injection port liner,

the column, and the detector (41).

Prepared samples were injected into an Ultra 2 column which is a 25 m x 0.2 mm

phenyl methyl silicone fused silica capillary column. The GC temperature program

heating the column went from 1700 C to 2700C at 50C per minute. A flame ionization

detector was used which allowed for a large dynamic range of detection and provided a

good level of sensitivity. Hydrogen was used as the carrier gas and air was used to

support the flame. The GC detector passed an electronic signal to the computer as

organic compounds were burning in the flame of the ionization detector. These

electronic signals were integrated into peaks and the data were stored on the hard disk.

The composition of the bacterial sample fatty acid methyl ester was compared to a stored

database using pattern recognition software (41).

Samples positively identified as Escherichia coli by API 20E and/or MIDI's

Microbial Identification System (MIS) were analyzed to determine if they were

pathogenic serotype 0157:H7. Two methods based on serological identification were

used to detect the presence of Escherichia coli 0157:H7. Bacto E. coli 0 antiserum

0157 and Bacto E. coli H antiserum H7 (Difco Inc., Detroit, MI) were used to identify

possible 0257:H7 Escherichia coli serotypes. Escherichia coli samples being serotyped

were streaked onto TSA using the three-way-streak method and were incubated at 350C

for 24 hours. Reagents were allowed to reach room temperature before opening and

using. One drop of Bacto E. coli 0 antiserum 0157 was placed on the left side of a

clean slide and on the right a drop of 0.85% saline solution. An inoculation loop was









used to transfer a loop full of the bacterial colony being serotyped to both sides of the

slide. A homogenate was created by blending the bacteria and the liquids, being careful

to keep them apart on the slide. The slide was placed under the light microscope and

viewed through the 10X, 40X and eventually the 100 X ocular lens. If agglutination of

bacterial cells was observed in the presence of E. coli 0 antiserum 0157 when compared

to those cells in the 0.85% saline solution then the sample was a positive Escherichia coli

0157 serotype. The same procedure was followed for the H7 antiserum (10).

The second method used was the Reveal Escherichia coli 0157:H7 test system

(Neogen Corp., Lansing, MI). A portion of enrichment culture was placed into the

sample area and passed through a reagent zone which contains specific antibodies (anti-

Escherichia coli 0157:H7) conjugated to colloidal gold particles. If the sample contains

antigens they will bind to the gold-conjugated antibodies, thus producing an antigen-

antibody complex that leaves the reagent zone and travels through the nitrocellulose

membrane and displays a visible line (33). Escherichia coli samples being serotyped

were streaked onto TSA using the three-way-streak method and were incubated at 350C

for 24 hours. A loop full of colonies was transferred to test tubes containing Modified

EC plus Novobiacin (mEC) and incubated at 350C for 24 hours. Test strips were allowed

to reach room temperature (250C) before opening their pouches. The appropriate number

of test strips required were removed from the foil pouches, and placed on a flat surface

and labeled with appropriate sample information. A transfer pipette was used to deliver 5

ml of enrichment broth containing culture into the port of the test strip. Results were

recorded after 15 minutes.









Statistical Analyses

A log transformation of the bacterial counts was performed since the raw data was

not normally distributed. Bacterial counts were expressed as loglO CFU per m3 of air.

The statistical analysis for this study was performed using SAS for Windows (40). The

experimental design utilized in this study was a randomized complete three-factor

factorial. The factors tested in this design were sampling points (8 levels), slaughters (3

levels), and species (2 levels). The dependent variables for each of the factors tested

were types of media (6 variables). Each air sample collected in this study consisted of

two subsamples. The total number of observations in this experimental design was 576.

All data collected in this study were statistically analyzed using the analysis of variance

method for General Linear Model Procedures (Proc GLM) (40). Any significant

differences were analyzed by the multiple comparisons procedure of LSD (least

significant difference), using a level of significance of alpha = 0.05 (40).














CHAPTER 4
RESULTS AND DISCUSSION

Preliminary Work: Air Sampling Time Determination

Preliminary work was conducted to determine the time and volume of air to be

sampled with the N6 single stage impactor to achieve appropriate growth on Petri plates

and attain a representative sample. Air samples were collected on TSA filled Petri plates

during the slaughter of six steers for 30 seconds, two, five, seven and 10 minutes (Table

2). It was determined that sampling times between 30 seconds and two minutes with a

flow rate of 28.3 + 0.1 liters/minute yielded countable bacterial growth on TSA filled

Petri plates (i.e. less than 250 CFU). A sampling time of five minutes or more yielded

bacterial growth on Petri plates that was too numerous to count (TNTC). A sampling

time of one minute during the slaughter process and two minutes when sampling before

the slaughter process yielded countable and representative bacterial growth.

Table 2. Sampling time determination for air sample collection during a beef slaughter
Total airborne bacteria: CFU/ft3 of air a

Sampling Sampling Time (minutes)
Areab 0.5 2 5 7 10
E 33 107 TNTC TNTC TNTC

F 30 155 200 TNTC TNTC
G 153 223 TNTC TNTC TNTC
H 4 11 12 15 13
a 1 CFU/ft3 of air = 0.02832 CFU/m3 of air, bE = bleeding pit; F = hide removal; G =
back splitting; H = cooler after 24 hr.









Microbiology of Bioaerosols Collected During Pork Slaughters

The results for air samples collected on non selective and selective media during

three separate pork slaughter processes are presented in Tables 3 through 13.

Total Airborne Bacteria

The collection of total airborne bacteria (TAB) on TSA Petri plates for three pork

slaughters resulted in no significant differences (P > 0.05) between slaughters

replicationss). There were however significant difference (P < 0.05) between particular

sampling areas within each pork slaughter. During the first pork slaughter sampling area

E had the highest log CFU per m3 of air counts and along with areas F and G had

significantly higher (P < 0.05) counts than sampling areas A, B and C, while sampling

area H had the lowest counts. In slaughter two sampling area F had the highest counts

but was not significantly different from areas E and G which were not significantly

different from areas A and C. In the last pork slaughter sampling areas E, F and G had

significantly higher counts than areas A, B, C and D which were significantly higher than

area H. Average data revealed significantly higher (P < 0.05) TAB counts in areas E, F,

and G when compared to areas A, B, C, D and H. Areas A, B, C and D had similar (P >

0.05) TAB counts. Areas E, F and G also had similar (P > 0.05) TAB counts. The TAB

counts for area H were significantly lower (P < 0.05) than the other areas sampled. TAB

counts were greater than three logs for areas E, F and G and less than three logs for areas

A, B, C, D and H. Area D had TAB counts around two logs and area H had counts that

were less than one log (Table 3).

The existence of non-zero TAB counts before slaughtering (A, B, C, D) could be

attributed to the presence of environmental bacteria. The significantly higher counts

observed in areas E, F and G when compared to areas A, B, C and D could be attributed









to the ongoing pork slaughtering process (Table 3). The presence of animals on the

slaughtering floor along with the slaughtering processes may have greatly increased the

amount of bacteria circulating in the environment. Animals from feedlots and pens

entering a slaughtering facility have a vast amount of microorganisms affixed to their

hoofs and hides (>109 CFU/cm2). These microorganisms represent the main food safety

hazard in cattle slaughterhouses (21). The majority of processes in slaughterhouses are

associated with the creation of bioaerosols. Hide spraying and removal cause bacteria on

the surface of animals to become airborne. The attachment of pathogenic and

nonpathogenic bacteria to carcass surfaces after hide removal is usually immediate (21).

The use of high speed equipment during slaughtering creates bioaerosols which may

cause an increase in carcass contamination (25). Higher bacterial counts in the back

splitting area may have been due to water flowing from the cutting saw, therefore

increasing the creation of bioaerosols. The manipulation of carcasses by personnel

during slaughtering may also have increased the TAB counts in areas E, F, and G.

Theoretically the lowest log CFU/m3 counts should have been observed in

sampling areas D and H since these areas were at 00C. The growth of psychrotrophic

microorganisms is favored at this low temperature therefore reducing environmental and

carcass bacteria. The fact that sampling area H had significantly lower (P < 0.05) counts

than area D was conflicting since carcasses were present in area H (Table 3).

Theoretically the presence of carcasses inside the cooler should increase the amount of

environmental bacteria and therefore increase the counts of area H compared to area D.

One reason why sampling area H had lower counts may have been because the presence

of carcasses disrupted air flow inside the cooler causing a smaller amount of bioaerosols









to reach the Andersen air sampler. Another reason may have been the presence of

organic acids in the air and on the carcasses causing a decrease in viable microorganisms.

Table 3. Mean total airborne bacteria collected on tryptic soy agar plates in eight areas
during three pork slaughters

Total airborne bacteria (log CFU/m3 of air)
Sampling Slaughter 1 Slaughter 2 Slaughter 3 Averaged SEM
Area'
A 2.30cDc 2.09BCD 2.24B 2.21B 0.08
B 2.37c 1.16DE 1.94B 1.82B 0.68
C 2.27CD 2.33BCD 2.21B 2.27B 0.12
D 2.09D 1.94CD 2.33B 2.12B 0.06
E 3.70A 3.34AB 3.36A 3.47A 0.20
F 3.35B 3.77A 3.71A 3.61A 0.07
G 3.55AB 3.32ABC 3.43A 3.43A 0.01
H 0.OOE 0.OOE 0.77c 0.26c 0.44
Average 2.45 2.24 2.50 2.40
SEM 0.08 0.43 0.29
aA = bleeding pit; B = scalding tank; C = back splitting; D = cooler; E = bleeding pit;
F = scalding tank; G = back splitting; H = cooler after 24 hr., A-D are before slaughter
and E-H are during slaughter, b Standard error of the mean, c Means within a column
followed by different letters (A, B, C, D, E) are significantly different (P < 0.05), dn = 6
observations.

Rahkio and Korkeala (34) reported log CFU/m3 of air counts ranging from 3.13 to

4.07 in the back splitting area and counts ranging from 2.45 to 3.58 in the weighing area

of a pork slaughter line. Log CFU/m3 of air counts ranging from 2.93 to 3.61 were

observed in the evisceration area and counts as low as 2.13 were recorded inside the

holding cooler in a pork slaughtering facility (25).

Total Airborne Yeast and Mold

Significant differences (P < 0.05) between slaughters replicationss) were observed

for some of the sampling areas during three pork slaughters (Table 4). Sampling areas A

and C had significantly higher (P < 0.05) log yeast and mold/m3 of air counts in slaughter

one than in slaughter three which had higher counts than in slaughter two. No differences









in counts were observed between slaughters two and three at sampling areas B and E, but

these slaughters had significantly lower counts than slaughter one. Total yeast and mold

counts were similar during all three slaughters for sampling areas D, F, and H (Table 4).

These differences in yeast and mold counts between slaughters for some of the sampling

areas could be attributed to experimental error. Yeast and mold collected on PDA media

were incubated from three to five days. During this incubation time overgrowth of some

yeast and molds could have caused an underestimation of counts in some plates.

Total airborne yeast and mold collected on PDA for three pork slaughters yielded

significant differences (P < 0.05) between sampling areas within each slaughter. In

slaughter one sampling areas E and F had significantly higher (P < 0.05) log yeast and

mold/m3 of air counts than areas A, B, C and G, while areas D and H had the lowest yeast

and mold counts. During the second slaughter sampling areas E, F and G had higher

counts than areas A, B and C which had higher counts than areas D and H. In the third

pork slaughter sampling areas E, F and G counts were not significantly higher (P > 0.05)

than those in areas A, B and C. Sampling area H had the lowest yeast and mold counts

for the third pork slaughter. Average data revealed significantly higher (P < 0.05) log

yeast and mold/m3 of air counts in area E when compared to areas A, B, C, D, F and H.

Areas A, B, C, F and G had similar (P > 0.05) yeast and mold counts. Yeast and mold

counts for areas D and H were significantly lower (P < 0.05) than the other areas

sampled. Sampling area H had the lowest counts of all the areas sampled. Yeast and

mold counts were greater than three logs for areas E, F and G and less than three logs for

areas A, B, C, D and H. Areas D and H had yeast and mold counts that were around two

logs (Table 4).









Table 4. Mean total airborne yeast and mold collected on potato dextrose agar plates in
eight areas during three pork slaughters
Total yeast and mold (log yeast and mold/m3 of air)
Sampling Slaughter 1 Slaughter 2 Slaughter 3 Averaged SEM
Area'
A 3.15cxc 2.66Bz 2.88ABY 2.90B 0.05
B 3.17cx 2.57BY 2.81BCY 2.85B 0.06
C 3.09CDX 2.57Bz 2.84ABY 2.83B 0.01
D 2.66E 2.00c 2.52c 2.39c 0.12
E 3.72AX 3.10AY 3.13AY 3.32A 0.03
F 3.33B 2.75A 2.92AB 3.00B 0.17
G 2.95DY 2.98AY 3.10ABX 3.01AB 0.02
H 1.55F 1.85c 1.79D 1.73D 0.14
Average 2.95 2.56 2.75 2.75
SEM 0.05 0.12 0.10
aA = bleeding pit; B = scalding tank; C = back splitting; D = cooler; E = bleeding pit;
F = scalding tank; G = back splitting; H = cooler after 24 hr., A-D are before slaughter
and E-H are during slaughter, b Standard error of the mean, c Means within a column
followed by different letters (A, B, C, D, E) and means within a row followed by
different letters (X, Y, Z) are significantly different (P < 0.05), dn = 6 observations.

Few differences in yeast and mold counts were seen between air samples taken

before and during slaughtering. Both the presence of animals on the slaughtering floor

and the slaughtering process did not greatly increase the amount of yeast and mold

circulating in the environment. Yeast and mold present before and during slaughtering

were probably from environmental sources and not associated with animals and the

slaughtering process. The lower yeast and mold counts observed in areas D and H could

be attributed to the cooler temperature (00C). Sampling area H may have had lower yeast

and mold counts because of the presence of carcasses in the cooler. This could have

caused an air flow disruption leading to less yeast and mold being collected by the

Andersen air sampler. Kotula et al. (25) reported log yeast and mold/m3 of air counts

ranging from 1.92 to 2.84 in the evisceration area and counts ranging from 1.57 to 2.52

inside the holding cooler during a pork slaughter.









Isolation of Staphylococcus species

Airborne bacteria collected on MSA filled Petri plates during three pork slaughters

resulted in no significant differences (P > 0.05) between slaughters (TableS). Significant

differences (P < 0.05) were observed between particular sampling areas within each pork

slaughter. In the first pork slaughter, sampling area F had the highest log CFU/m3 of air

counts of all the areas. Sampling areas E, F, and G had significantly higher (P < 0.05)

counts than areas A, B, C, D, and H for the first and second pork slaughter. In the last

pork slaughter sampling areas E, F, G, and D had significant higher counts than areas A,

B, C and H. Average data revealed significantly higher (P < 0.05) counts of airborne

bacteria collected on MSA media in areas E, F and G when compared to areas A, B, C, D

and H. Areas A, B, C, D and H had similar (P > 0.05) counts. Areas E, F and G also had

similar (P > 0.05) counts. The bacterial counts were around three logs for areas E, F and

G and less than one log for areas A, B, C, D and H (Table 5).

The significantly higher counts observed in areas E, F and G when compared to

areas A, B, C, D and H may be attributed to the presence of animals on the slaughtering

floor along with the ongoing slaughtering processes. Most of the processes during pork

slaughtering are associated with the creation of bioaerosols (21). Bacteria on the surface

of animals may become airborne during hide spraying and removal (21). Higher bacterial

counts in the back splitting area may have been caused by the blade movement and water

flowing from the cutting saw. The presence of Staphylococcus aureus on animal hides,

skin, lesions and bruised tissue along with personnel manipulation during slaughtering

may also have increased the bacteria counts in areas E, F, and G. Microorganisms

collected on MSA filled Petri plates during three pork slaughters are presented in Table 6.









Table 5. Mean airborne bacteria collected on mannitol salt agar plates in eight areas
during three pork slaughters
Airborne bacteria (log CFU/m3 of air)
Sampling Slaughter 1 Slaughter 2 Slaughter 3 Averaged SEM
Area'
A 1.55Dc 0.OOB 0.OOB 0.52B 0.00
B 0.OOE 0.77B 0.OOB 0.26B 0.45
C 0.OOE 0.77B 0.OOB 0.26B 0.45
D 0.OOE 0.OOB 1.55A 0.52B 0.00
E 3.23B 3.49A 2.41A 3.05A 0.12
F 3.71A 3.13A 2.51A 3.12A 0.56
G 3.19c 3.02A 2.42A 2.87A 0.01
H 0.00E 0.00B O.OOB 0.00B 0.00
Average 1.46 1.40 1.11 1.32
SEM 0.01 0.40 0.35
aA = bleeding pit; B = scalding tank; C = back splitting; D = cooler; E = bleeding pit;
F = scalding tank; G = back splitting; H = cooler after 24 hr., A-D are before slaughter
and E-H are during slaughter, b Standard error of the mean, c Means within a column
followed by different letters (A, B, C, D, E) are significantly different (P < 0.05), dn = 6
observations.

Most of the airborne bacteria collected on MSA media during three pork slaughters

were Staphylococcus species. The rarely pathogenic Staphylococcus epidermidis is the

most ubiquitous of the various Staphylococcus species found on the skin of humans and

animals. Staphylococcus aureus is a potential food safety hazard since many of its strains

produce enterotoxins which cause food poisoning when ingested (Staphylococcal food

poisoning). Staphylococcus aureus contamination of pork carcasses does not only come

from human contact but also from animal hides, skin, lesions and bruised tissue (11).

Staphylococcus hyicus has also been shown to produce enterotoxins in food. Other Gram

positive bacteria isolated from MSA filled Petri plates during three pork slaughters were

Bacillus megaterium and Kocuria kristinae. The presence of Staphylococcus,

Micrococcus and Bacillus species in bioaerosols collected during three pork slaughters









could be attributed to their association with mammalian skin, soil and animal feces (Table

6).

In this study a total of 171 bacterial colonies were isolated from various selective

media used to collect air samples and carcass swab samples during three pork and beef

slaughters. Of these bacterial isolated a total of 115 were from air samples.

Staphylococcus species were identified from 46.7% of the bacterial colonies isolated

from air samples during three pork slaughters. Staphylococcus aureus was found in 15%

of the colonies isolated from air samples during three pork slaughters.

Table 6. Airborne bacteria isolated from mannitol salt agar plates collected from three
pork slaughters
Organism (Genus Species)
Bacillus megaterium Staphylococcus hominis
Kocuria kristinae Staphylococcus hyicus
Staphylococcus arlettae Staphylococcus kloosii
Staphylococcus aureus Staphylococcus lugdunensis
Staphylococcus auricularis Staphylococcus saprophyticus
Staphylococcus capitis Staphylococcus sciuri
Staphylococcus chromogenes Staphylococcus warneri
Staphylococcus cohnii cohnii Staphylococcus xylosus
Staphylococcus epidermidis
Staphylococcus gallinarum

Isolation of Listeria species

No significant differences (P > 0.05) between slaughters resulted from airborne

bacteria collected on MOX filled Petri plates during three pork slaughters (Table 7).

Significant differences (P < 0.05) were observed between individual sampling areas

within each pork slaughter. Sampling areas E and F had significantly higher (P < 0.05)

log CFU/m3 of air counts than areas A, B, C, D, and H in the first pork slaughter. In the

second pork slaughter sampling areas E, F and G had significantly higher counts (mean

log CFU/m3 of air) than areas A, B, C and H. During the last pork slaughter, counts from









sampling areas E and G were similar to those of areas A, B, D and H. Average data

revealed significantly higher (P < 0.05) counts of airborne bacteria collected on MOX

media in areas E, F and G when compared to areas A, B, C, D and H. Areas A, B, C, D

and H had similar (P > 0.05) counts. Areas E, F and G also had similar (P > 0.05) counts

(Table 7). The bacterial counts were less than three logs for all sampling areas.

Table 7. Mean airborne bacteria collected on modified oxford agar plates in eight areas
during three pork slaughters
Airborne bacteria (log CFU/m3 of air)
Sampling Slaughter 1 Slaughter 2 Slaughter 3 Averaged SEM
Area'
A 0.92Bcc 0.OOc 0.92BC 0.62BC 0.76
B 0.77BC 0.OOc 0.92BC 0.57BC 0.70
C 0.OOc 0.77c 0.OOc 0.26c 0.45
D 0.OOc 1.85B 1.55ABC 1.13B 0.17
E 2.56A 3.01A 2.68AB 2.75A 0.06
F 2.93A 2.60AB 3.01A 2.85A 0.08
G 2.00AB 2.40AB 2.59AB 2.33A 0.27
H 0.OOc 0.OOc 0.92BC 0.31c 0.53
Average 1.15 1.33 1.57 1.35
SEM 0.46 0.30 0.57
aA = bleeding pit; B = scalding tank; C = back splitting; D = cooler; E = bleeding pit;
F = scalding tank; G = back splitting; H = cooler after 24 hr., A-D are before slaughter
and E-H are during slaughter, b Standard error of the mean, c Means within a column
followed by different letters (A, B, C) are significantly different (P < 0.05), dn =6
observations.

The significantly higher counts observed in areas E, F and G when compared to

areas A, B, C, D and H could be attributed to the ongoing pork slaughtering process.

Animal washing, hide spraying, bleeding and hide removal are processes often associated

with the aerosolization of bacteria. Higher bacterial counts in the back splitting area may

have been due to water flowing from the cutting saw, therefore increasing the creation of

bioaerosols. Carcass manipulation by personnel during slaughtering may also have

increased counts in areas E, F, and G. No Listeria species were isolated from air samples









during this study. Microorganisms collected on MOX filled Petri plates during three pork

slaughters are presented in Table 8.

The majority of the airborne bacteria collected on MOX media during three pork

slaughters were Gram positive species with the exception of Escherichia coli, Salmonella

bongori and .\/nge/la boydii (Table 8). The Gram positive species collected on MOX

media have a wide environmental distribution and are associated with the contamination

of meat and dairy products (12). The genus Listeria belongs to the bacilli class along

with Bacillus, Paenibacillus, Staphylococcus, Streptococcus, Lactobacillus, and

Brochothrix. Although bacteria from the bacilli class were isolated, Listeria species were

not identified from any of the 171 bacterial colonies collected from air and carcass

samples. The presence of Bacillus, Brevibacterium, Brochothrix, Cellulomonas,

Microbacterium and Micrococcus species in bioaerosols collected during three pork

slaughters could be attributed to their predominance in soil, mammalian skin and animal

feces. These bacteria could have entered the slaughtering facility through contaminated

hides of animals, feces, soil on workers' clothing and shoes, equipment transport and

human carriers.

Table 8. Airborne bacteria isolated from modified oxford agar plates collected from three
pork slaughters
Organism (Genus Species)
Bacillus pumilus Microbacterium barker
Brevibacterium casei Microbacterium saperdae
Brevibacterium epidermidis Nesterenkonia halobia
Brochothrix campestris Salmonella bongori
Cellulomonas fimi 1/nge/'al boydii
Escherichia coli Staphylococcus aureus
Kocuria kristinae Staphylococcus epidermidis
Staphylococcus haemolyticus









Ren (35) isolated Listeria spp. from specific environmental surfaces of dairy plants

but could not recover it from air samples collected close to these surfaces. It is

hypothesized that Listeria species suspended in air may loose their colony forming

capability even though they retain overall viability. Air sample collection using the

impaction method has been demonstrated to cause bacteria to enter into a viable but non

culturable (VBNC) state (17). Impaction stress created during aerosol generation,

dispersion and landing or collection, along with environmental stressors like temperature,

dehydration, irradiation, oxidation and pollution often prove to be lethal to airborne

bacteria (22, 45). Bacteria that are sublethally injured during sample collection onto agar,

using the impaction method, will not form colonies and, therefore cannot be accounted

for (45). It is important to include repair and resuscitation steps using enrichment broths

to increase the culturability of sublethally injured bacteria (21). Listeria species may

have become lethally injured by sample collection and/or environmental conditions, or

sublethally injured inducing the VBNC state.

Isolation of Escherichia coli

The collection of airborne bacteria on VRBA filled Petri plates during three pork

slaughters resulted in no significant differences (P > 0.05) between slaughters.

Significant differences (P < 0.05) between particular sampling areas were only observed

during the third pork slaughter. Sampling areas E and F had significantly higher (P <

0.05) counts than area G which had higher counts than areas A, B, C, D and H where no

growth was obtained. Average data revealed significantly higher (P < 0.05) counts of

airborne bacteria collected on VRBA media in areas E, F and G when compared to areas

A, B, C, D and H were no growth was obtained. Areas E, F and G also had similar (P >

0.05) counts. The bacterial counts were less than one log for sampling areas E, F and G









(Table 9). The presence of airborne bacteria collected on VRBA filled Petri plates in

areas E, F and G compared to the absence of bacteria in areas A, B, C, D and H could be

attributed to the ongoing pork slaughtering process. Processes like animal washing, hide

spraying, bleeding and hair removal are often associated with the production of

bioaerosols.

Table 9. Mean airborne bacteria collected on violet red bile agar plates in eight areas
during three pork slaughters
Airborne bacteria (log CFU/m3 of air)
Sampling Slaughter 1 Slaughter 2 Slaughter 3 Averaged SEM
Area'
A 0.00 0.00 0.00cc 0.00B 0.00
B 0.00 0.00 0.OOc 0.OOB 0.00
C 0.00 0.00 0.OOc 0.OOB 0.00
D 0.00 0.00 0.OOc 0.OOB 0.00
E 0.77 0.00 2.05A 0.94A 0.46
F 0.00 0.92 1.85A 0.92A 0.53
G 0.00 1.01 1.55B 0.85A 0.58
H 0.00 0.00 0.OOc 0.OOB 0.00
Average 0.10 0.24 0.68 0.34
SEM 0.27 0.49 0.07
aA = bleeding pit; B = scalding tank; C = back splitting; D = cooler; E = bleeding pit;
F = scalding tank; G = back splitting; H = cooler after 24 hr., A-D are before slaughter
and E-H are during slaughter, b Standard error of the mean, c Means within a column
followed by different letters (A, B, C) are significantly different (P < 0.05), dn = 6
observations.

The majority of airborne bacteria collected on VRBA media during three pork

slaughters were Gram negative bacteria from the Enterobacteriaceae family.

Enterobacteriaceae have a worldwide distribution and are found in soil, water, fruits,

vegetables, plants, trees and animals. They inhabit a wide variety of niches which

include human and animal gastrointestinal tracts and various environmental sites (11, 14,

19). Other Gram negative bacteria present like Pseudomonas spp., Chryseobacterium









indologenes, Chryseomonas luteola, Flavimonas oryzihabitans, and Stenotrophomonas

maltophilia are all members of the Pseudomonadaceae family (Table 10). These species

are normally associated with soil and fecal mater and therefore are found on the hides of

animals. Gram positive bacteria isolated from VRBA plates were Kocuria kristinae,

Microbacterium barker and Bacillus pumilus which are frequently isolated from meat

and dairy products (11, 19). The presence of Enterobacteriaceae species in bioaerosols

collected during three pork slaughters could be attributed to their association with

mammalian skin, soil and animal feces. These bacteria could have entered the

slaughtering facility on the hides of animals, feces and by the soil on workers' shoes.

Table 10. Airborne bacteria isolated from violet red bile agar plates collected from three
pork slaughters
Organism (Genus Species)
Cedecea neteri Microbacterium barker
Chryseobacterium indologenes Pantoea spp.
Chryseomonas luteola Pseudomonas aeruginosa
Enterobacter aerogenes Pseudomonas fluorescens
Enterobacter cloacae Pseudomonas putida
Escherichia coli Salmonella bongori
Escherichiafergusonii Salmonella choleraesuis choleraesuis
Flavimonas oryzihabitans Salmonella typhi
Klebsiella pneumoniae ozaenae Serratia liquefaciens
Klebsiella pneumoniae pneumoniae Serratia rubidaea
Klebsiella terrigena .l/nge//la boydii
Kluyvera spp. .'1ige/ll Iflexneri
Kocuria kristinae *\/nge/al sonnei
Leclercia adecarboxylata Stenotrophomonas maltophilia

Generic Escherichia coli was identified from 8.3% of the 115 airborne bacterial

colonies isolated during three pork slaughters. Escherichia coli 0157:H7 was not

isolated from any of the air samples taken from the various sampling areas. The low

airborne concentration of Escherichia coli 0157:H7 may be one reason this organism









was not isolated. Impaction stress created during aerosol collection along with

environmental factors like temperature and dehydration may have caused lethal injury to

this organism or induced the VBNC state, therefore contributing to the inability of

isolating this organism from air samples (17).

Isolation of Salmonella species

Airborne bacteria collected on XLT4 filled Petri plates during three pork slaughters

yielded no significant differences (P > 0.05) between slaughters. No significant

differences (P > 0.05) were observed between individual sampling areas within each pork

slaughter. Average data revealed significantly higher (P < 0.05) counts of airborne

bacteria collected on XLT4 media in area F when compared to the rest of the areas were

no growth was obtained. The bacterial counts were less than one log for sampling area E

(Table 11).

Table 11. Mean airborne bacteria collected on xylose-lysine-tergitol 4 agar plates in eight
areas during three pork slaughters
Airborne bacteria (log CFU/m3 of air)
Sampling Slaughter 1 Slaughter 2 Slaughter 3 Averaged SEM

A 0.00 0.00 0.00 0.00Bc 0.00
B 0.00 0.00 0.00 0.OOB 0.00
C 0.00 0.00 0.00 0.OOB 0.00
D 0.00 0.00 0.00 0.OOB 0.00
E 0.00 0.00 0.00 0.OOB 0.00
F 1.01 0.00 0.77 0.60A 0.73
G 0.00 0.00 0.00 0.OOB 0.00
H 0.00 0.00 0.00 0.OOB 0.00
Average 0.13 0.00 0.10 0.07
SEM 0.36 0.00 0.28
aA = bleeding pit; B = scalding tank; C = back splitting; D = cooler; E = bleeding pit;
F = scalding tank; G = back splitting; H = cooler after 24 hr., A-D are before slaughter
and E-H are during slaughter, b Standard error of the mean, c Means within a column
followed by different letters (A, B) are significantly different (P < 0.05), dn 6
observations.










Most of the airborne bacteria collected on XLT4 media during three pork slaughters

were Gram negative bacteria from the Enterobacteriaceae family (Table 12). Salmonella

species were not isolated from air samples using XLT4 media during the three pork

slaughters. Airborne Salmonella species, Escherichia coli and .\lnge//l species were

readily isolated from the less selective VRBA media. Salmonella enterica subsp.

enterica serotype typhi, choleraesuis houtenae and typhimurium and Salmonella bongori

were isolated from VRBA media (Tables 12). The poor isolation of Salmonella from air

samples collected on XLT4 may have been due to this media's high selectivity.

Environmental factors along with impaction stress may have reduced the ability to isolate

Salmonella species from air samples (17). Salmonella species were identified from 10%

of the total airborne bacterial colonies isolated during three pork slaughters.

Table 12. Airborne bacteria isolated from xylose-lysine-tergitol 4 agar plates collected
from three pork slaughters
Organism (Genus Species)
Enterobacter aerogenes Pseudomonas aeruginosa
Enterobacter asburiae Pseudomonas fluorescens
Enterobacter cancerogenus Pseudomonas putida
Enterobacter cloacae Serratia liquefaciens
Escherichia coli Serratia marcescens
Escherichiafergusonii .\ngge/A llflexneri
Klebsiella ornithinolytica .\/nglie/a sonnei
Klebsiella pneumoniae pneumoniae Staphylococcus aureus
Morganella morganii Staphylococcus epidermidis

Identification of Airborne Bacteria Collected from Pork Slaughters

Most of the airborne bacteria recovered before the pork slaughters (A, B, C, and D)

were Gram negatives from the Enterobacteriaceae family with the exception of

Flavimonas oryzihabitans from the Pseudomonadaceae family. Staphylococcus,









Micrococcus and Bacillus species were the main Gram positive airborne bacteria

recovered before the pork slaughters (Table 13).

The majority of the Gram negative airborne bacteria isolated during the three pork

slaughters (E, F, and G) were from the Enterobacteriaceae and Pseudomonadaceae

family. Most of the Gram positive airborne bacteria isolated during the three pork

slaughters were Staphylococcus, Microbacterium, Brevibacterium, Bacillus and

Micrococcus species. These species are found in the soil and are associated with

mammalian skin and are therefore frequently isolated from food products and the

environment. Another Gram positive organism isolated was Brochothrix thermosphacta

which is a meat spoilage microorganism that grows at refrigeration temperatures and

predominates in vacuum packaged and modified atmosphere conditions (12, 19).

Fewer microorganisms were isolated before the pork slaughters (A, B, C, D) when

compared to the number of microorganisms isolated during the slaughters (E, F, and G)

(Tables 13). Higher numbers of microorganisms isolated from areas E, F and G could be

attributed to the ongoing pork slaughtering process. The presence of animals on the

slaughtering floor along with the slaughtering processes may have greatly increased the

number of airborne microorganisms.

Most of the microorganisms that were isolated before the pork slaughters were also

isolated during the slaughters. The presence these airborne bacteria during the slaughter

process could be attributed to their association with the animals or to their general

environmental presence. Bacillus megaterium, Staphylococcus auricularis, capitis, and

lugdunensis were only isolated before the three pork slaughters. These microorganisms

isolated before the pork slaughters could be regarded as environmental or background









bacteria (Table 13). The least amount of airborne bacteria was isolated from the 00C

cooler (D and H). Kocuria kristinae, Staphylococcus capitis, auricularis and cohnii were

isolated from the empty cooler (D) and Staphylococcus cohnii and sciuri were isolated

from the cooler after carcasses had been stored for at least 24 hours at 00C (H).

Table 13. Airborne bacteria isolated before and during three pork slaughters
Organism (Genus Species)
Before Slaughter During Slaughter
Gram negative Gram negative
Escherichia coli Cedecea neteri
Flavimonas oryzihabitans Enterobacter aerogenes
Salmonella bongori Enterobacter cloacae
.\/nlge/lt boydii Escherichia coli
Gram positive Escherichia fergusonii
Bacillus megaterium Flavimonas oryzihabitans
Kocuria kristinae Klebsiella ornithinolytica
Staphylococcus auricularis Klebsiella pneumoniae ozaenae
Staphylococcus capitis Klebsiella pneumoniae pneumoniae
Staphylococcus chromogenes Klebsiella terrigena
Staphylococcus cohnii cohnii Kluyvera spp.
Staphylococcus epidermidis Pseudomonas aeruginosa
Staphylococcus hominis Pseudomonas fluorescens
Staphylococcus lugdunensis Pseudomonas putida
Staphylococcus sciuri Salmonella bongori
Staphylococcus warneri Salmonella choleraesuis choleraesuis
Salmonella typhi
Serratia liquefaciens
Serratia marcescens
Serratia rubidaea
.\1/nge/1a boydii
Stenotrophomonas maltophilia
Gram positive
Bacillus pumilus
Brevibacterium casei
Brevibacterium epidermidis
Brochothrix campestris
Cellulomonas fimi
Kocuria kristinae
Microbacterium barker
Microbacterium saperdae
Nesterenkonia halobia
Staphylococcus arlettae









Table 13. Continued.
Organism (Genus Species)
During Slaughter
Gram positive
Staphylococcus aureus
Staphylococcus chromogenes
Staphylococcus cohnii cohnii
Staphylococcus epidermidis
Staphylococcus gallinarum
Staphylococcus haemolyticus
Staphylococcus hominis
Staphylococcus hyicus
Staphylococcus kloosii
Staphylococcus saprophyticus
Staphylococcus sciuri
Staphylococcus warneri
Staphylococcus xylosus

Carcass Swabs Collected During Pork Slaughters

The microorganisms listed in Table 14 were isolated from swabs taken from pork

carcasses which had been in a cooler at 00C for 24 hours. The majority of bacteria

isolated from pork carcass swabs were Gram negative bacteria from the

Enterobacteriaceae family. Other Gram negative spoilage microorganisms present were

Pseudomonas spp., Chryseobacterium indologenes, Chryseomonas luteola, which are

members of the Pseudomonadaceae family. The only Gram positive bacteria isolated

from pork carcass swabs were Staphylococcus species.

A total of 56 out of the 171 bacterial colonies isolated during this study were from

carcass swabs. Staphylococcus species were identified from 52.4% of the bacterial

colonies isolated from pork carcass swabs. Staphylococcus aureus was identified from

33.3% of the bacterial colonies isolated during pork carcass sampling. Escherichia coli

and Salmonella species were identified from 14.3% and 4.8% respectively, of the

bacterial colonies isolated from pork carcass swabs. Listeria species and Escherichia coli









0157:H7 were not isolated from any of the pork carcass swabs. Acetic acid spraying and

carcass washing along with low temperature storage of pork carcasses may have greatly

reduced the growth of these microorganisms.

Some organisms like Enterobacter cloacae, Escherichia coli andfergusonii,

Klebsiella pneumoniae pneumoniae, Pseudomonas aeruginosa, Salmonella choleraesuis

choleraesuis, .\/nge//Ai species, Staphylococcus aureus, epidermidis and other

Staphylococcus species were isolated from air samples and pork carcass swabs (Tables 13

and 14). None of the airborne bacteria isolated from the empty cooler (D) and the cooler

containing carcasses (H) were isolated from the pork carcasses. The isolation of various

organisms, including Staphylococcus, Escherichia, and Salmonella species, from air

samples taken during slaughtering and carcass swabs, supports the theory that bioaerosols

transport bacteria and contribute to the contamination of pork carcasses.

Table 14. Bacteria isolated from pork carcasses held at 00C for 24 hours
Organism (Genus Species)
Gram negative Gram positive
Chryseobacterium indologenes Staphylococcus aureus
Chryseomonas luteola Staphylococcus chromogenes
Enterobacter asburiae Staphylococcus epidermidis
Enterobacter cancerogenus Staphylococcus hominis
Enterobacter cloacae Staphylococcus hyicus
Escherichia coli Staphylococcus warneri
Escherichiafergusonii Staphylococcus xylosus
Klebsiella pneumoniae pneumoniae
Leclercia adecarboxylata
Morganella morganii
Pantoea spp.
Pseudomonas aeruginosa
Salmonella choleraesuis choleraesuis
.l/Ige/ll Iflexneri
*\/ne//t sonnei









Microbiology of Bioaerosols Collected During Beef Slaughters

The results for air samples collected on non selective and selective media during

three separate beef slaughter processes are presented in Tables 15 through 25.

Total Airborne Bacteria

Total airborne bacteria (TAB) collected on TSA Petri plates for three beef

slaughters resulted in no significant differences (P > 0.05) between slaughters

replicationss) (Table 15). Significant differences (P < 0.05) between specific sampling

areas within each beef slaughter were observed for slaughters one and two. Log CFU per

m3 of air counts for sampling areas during beef slaughtering (E, F, G, H) and sampling

areas before slaughtering (A, B, D) were not significantly different in slaughter one.

Counts for sampling area C were significantly lower than the counts for the rest of the

sampling areas. Sampling areas F and G had the highest TAB counts and sampling area

D had the lowest counts during the second beef slaughter. There were no significant

differences (P > 0.05) between sampling areas for the third beef slaughter. Average data

revealed significantly higher (P < 0.05) TAB counts in areas F and G when compared to

areas A, B, C, D and H. Areas A, B, C, D and H had similar (P > 0.05) TAB counts.

Areas E, F and G also had similar (P > 0.05) TAB counts. Sampling area E had similar

counts to those of areas A, B and C. Sampling areas H and D had the lowest counts

although not significantly different (P > 0.05) from those of the sampling areas before

slaughter (A, B, C). TAB counts were around three logs for areas E, F and G and less

than three logs for areas A, B, C, D and H. Areas D and H had TAB counts that were

less than two logs (Table 15).

The existence of non-zero TAB counts before slaughtering (A, B, C, D) may have

been caused by the presence of environmental bacteria. The significantly higher counts









observed in areas F and G when compared to areas A, B, C and D could be attributed to

the ongoing beef slaughtering process (Table 15). The presence of cattle on the

slaughtering floor, combined with the slaughtering process, may have greatly increased

the amount of environmental bacteria. The spraying of cattle and hide removal could

have caused surface bacteria to become airborne. The attachment of pathogenic and

nonpathogenic bacteria to carcass surfaces after hide removal is usually immediate (21).

Higher levels of TAB present in the back splitting area may have been caused by water

flowing from the cutting saw or due to bioaerosol transfer from contaminated parts of the

line like the hide removal area. The manipulation of carcasses by personnel during hide

removal, evisceration and back splitting may have also increased the TAB counts in areas

F and G.

Table 15. Mean total airborne bacteria collected on tryptic soy agar plates in eight areas
during three beef slaughters

Total airborne bacteria (log CFU/m3 of air)
Sampling Slaughter 1 Slaughter 2 Slaughter 3 Averaged SEM
Area'
A 2.20AC 2.50B 2.35 2.35BC 0.06
B 2.07 A 2.14BC 2.63 2.28BC 0.10
C 0.92 B 2.76AB 3.25 2.31BC 0.54
D 2.78 A 0.77D 1.16 1.57c 0.81
E 3.02 A 2.78AB 2.69 2.83AB 0.16
F 2.94A 3.61A 2.99 3.18A 0.20
G 3.09A 3.44A 2.94 3.16A 0.15
H 2.22 A 1.55CD 1.55 1.77c 0.04
Average 2.40 2.44 2.44 2.43
SEM 0.35 0.29 0.44
aA = bleeding pit; B = hide removal; C = back splitting; D = cooler; E = bleeding pit;
F = hide removal; G = back splitting; H = cooler after 24 hr., A-D are before slaughter
and E-H are during slaughter, b Standard error of the mean, c Means within a column
followed by different letters (A, B, C, D) are significantly different (P<0.05), dn = 6
observations.









Rahkio and Korkeala (34) reported log CFU/m3 of air counts ranging from 2.21 to

3.70 in the back splitting area and counts ranging from 2.21 to 3.59 in the weighing area

of a beef slaughter line. According to Jericho et al. (21), a count of 4.03 log CFU/m3 of

air was observed at the hide removal station during beef slaughtering and a count of 3.25

log CFU/m3 of air was recorded at the hide removal station before slaughtering.

According to Jericho et al. (21) the presence of non-zero TAB counts before slaughtering

was attributed to previous clean-up operation using high pressure hosing.

Total Airborne Yeast and Mold

Significant differences (P < 0.05) between slaughters replicationss) were observed

for some of the sampling areas during the three beef slaughters (Table 16). Log yeast and

mold/m3 of air counts were similar in slaughters one and two which were significantly

higher (P < 0.05) than the counts observed during the third beef slaughter for sampling

areas C, E, F and G (Table 16). These differences in yeast and mold counts between

slaughters for some of the sampling areas could be attributed to experimental error.

During slaughter three yeast and mold counts may have been underestimated due to their

overgrowth on PDA Petri plates.

Total airborne yeast and mold collected on PDA for three beef slaughters yielded

significant differences (P < 0.05) between sampling areas within each slaughter.

Sampling areas E, F, and G had significantly higher (P < 0.05) log yeast and mold/m3 of

air counts than areas A, B and C, while areas D and H had the lowest counts in the first

beef slaughter. In slaughter two sampling areas E, F and G did not have significantly

higher (P > 0.05) counts than areas A, B and C. Sampling area H had the lowest counts

during the second beef slaughter. No significant differences (P > 0.05) were observed

between the sampling areas in the third beef slaughter. Overall during the three beef