masrs [ll/l Ill/ll / ll Willi/ll I fill This is to certify that the thesis entitled A Sero-epidemiological Study of the Causative Agent of Contagious Equine Metritis presented by Janice Marie Spence has been accepted towards fulfillment of the requirements for M.S . degree in 1811110ng)!— Major professor Date W 0-7 639 ;_ MSU RETURNING MATERIALS: Place in book drop to remove this checkout from LIBRARIES “ your record. FINES will be charged if book is ofiamma returned after the date stamped below. ‘m‘ 1023“? A SERO-EPIDEMIOLOGICAL STUDY OF THE CAUSATIVE AGENT 0F CONTAGIOUS EQUINE METRITIS By Janice Marie Spence A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Pathology 1981 ABSTRACT A SERO-EPIDEMIOLOGICAL STUDY OF THE CAUSATIVE AGENT 0F CONTAGIOUS EQUINE METRITIS By Janice Marie Spence This study examined several species, horses, cattle, sheep, pigs, dogs and humans, for serological evidence of exposure to the causative agent of contagious equine metritis. Serum samples are tested by sero- logical test methods developed for detection of the contagious equine metritis organism (CEMO) in horses. The methods used are a plate agglutination test, a complement fixation test and a passive hemagglutin- ation test. The serological methods are compared for their correlation of test results and technical merit as an epidemiological screen. The morphology of the streptomycin sensitive strain of the CEMO is studied by electron microscopy. Serological test results indicate that sheep, pigs and cattle may have exposure to CEMO or related organisms. There is a statistically significant increase in the incidence of antibodies to CEMO in human patients being tested or treated for venereal disease. 0f the serological tests evaluated, the passive hemagglutination method is the best suited for epidemiological studies. T0 JOHN Your love, understanding and encouragement has made this possible. 1'1 ACKNOWLEDGEMENTS I wish to express my appreciation to Dr. David G. Powell, my project advisor, for his patience, guidance and support during the com- pletion of my research. Special thanks go to Dr. Chris Brown for his advice and encourage- ment during the writing of my thesis. I also wish to thank Martha Thomas and Dr. J. D. Krehbiel, members of my guidance committee, for their support and counsel. Appreciation is also due to my co-workers for their assistance, understanding and moral support during the course of my study. Many people and groups provided assistance with the collection of the samples used in this project. I would like to acknowledge the Microbiology section of the Animal Health Diagnostic Laboratory of Michigan State University, the Michigan Department of Public Health and the Michigan Department of Agriculture--Geigley Laboratory for their aid. I would also like to thank Dr. Thomas Swerczek for providing bacterial cultures of CEMO and all supplies for the plate agglutination test. My appreciation is also extended to Dr. D. S. Fernie and Dr. J. T. Bryans for providing samples of CEMO antigen preparations. The study was supported by a Biomedical Research Support Grant from the Michigan State University College of Veterinary Medicine. TABLE OF CONTENTS CHAPTER Page LIST OF TABLES .......................................... vi LIST OF FIGURES ......................................... vii INTRODUCTION ............................................ l I. LITERATURE REVIEW ....................................... 2 1.1. Historical Review .............................. 2 l.2. Bacteriological Characteristics of CEMO ........ 4 l.3. Diagnostic Tests for CEM ....................... 9 l.4. Serological Techniques for CEM Detection ....... 12 1.5. Epidemiological Studies ........................ 20 l.6. Proposed Study ................................. 22 II. METHODS AND MATERIALS ................................... 24 2.1. Serum Samples .................................. 24 2.2. Contagious Equine Metritis Organism ............ 25 2.3. Serological Tests .............................. 26 2.4. Electron Microscopy Studies .................... 31 III. RESULTS AND DISCUSSION .................................. 32 3.l. Introduction ................................... 32 3.2. Description of Figures and Tables .............. 33 3.3. Evaluation of CEMO Exposure .................... 38 3.4. Correlation of Serological Test Results ........ 43 3.5. Comparison of Test Mechanics ................... 49 3.6. Electron Microscopy Studies .................... 53 IV. SUMMARY AND SUGGESTED FURTHER STUDIES ................... 59 4.l. Summary ........................................ 59 4.2. Suggested Further Studies ...................... 60 REFERENCES ..................................................... 61 iv APPENDICES A. ANTIGEN PREPARATION FOR SEROLOGICAL TESTS .............. A.l. CEMO Complement Fixation Reagent Preparation.. A.2. CEMO Passive Hemagglutination Antigen Prepara- tion ....................................... B. SCREENING DILUTION DETERMINATION FOR PHA ............... Table 3.1. Preliminary PHA Titer Results ........... C. DETERMINATION OF EXACT PROBABILITIES IN A 2x2 CON- TINGENCY TABLE ...................................... VITAE .......................................................... Page 64 64 65 68 68 69 71 TABLE common 3.]. LIST OF TABLES . Serological test results, percentage positive to CEMO antigen ................................................... . Distribution of complement fixation titers for CEMO antigens .................................................. . Distribution of passive hemagglutination titers for CEMO antigens .................................................. . Contingency tables for comparison of serological test results ................................................... . Comparison of human groups, all methods ................... Comparison of human groups, C-FIX test method ............. Comparison of human groups, PHA test method ............... Chi-squared values for testing independence of serological test methods .............................................. Preliminary PHA titer results ............................. vi Page 34 36 36 37 41 42 42 44 68 LIST OF FIGURES FIGURE Page l. Characteristics of CEMO and serologically similar bacterial genera ............... . ......................... 7 2. SEM photomicrograph of streptomycin-sensitive CEMO at 7,lOOX showing a mass of coccoid-shaped CEMO ............. 55 3. SEM photomicrograph of streptomycin-sensitive CEMO at l0,000X demonstrating pleomorphic rod-shaped CEMO ........ 55 4. SEM photomicrograph at streptomycin-sensitive CEMO at 23,000X demonstrating rod forms terminating in coccoid shaped organisms ......................................... 56 5. TEM photomicrograph of streptomycin-sensitive CEMO at 49,000X demonstrating a threadlike capsule (C) ........... 57 6. TEM photomicrograph of streptomycin-sensitive CEMO at 91,000X. Labelled features are capsule (C), outer membrane (0), dense intermediate layer (D) and cyto- plasmic membrane (P) ..................................... 57 vii INTRODUCTION Contagious equine metritis (CEM) is a recently discovered venereal disease of horses. Although CEM is a highly contagious dis- ease, it is not a serious infection for the horse itself. The primary impact of this disease is due to the serious economic effect that delays in successful breeding have on the Thoroughbred industry. Contagious equine metritis is caused by a previously unknown, fastidious, Gram—negative coccobacillus which has not yet been placed into a bacterial taxonomic classification. A foremost question when any new bacterium is isolated from one host, especially if pathogenic, is whether it is pathogenic for other animal species, especially man. The primary impetus for this research project is the identification of other animal species which might have had exposure to or have been a reservoir of the CEM organism. Serological test results indicate that sheep, pigs and cattle may have exposure to CEMO or related organisms. There is a statistically significant increase in the incidence of antibodies to CEMO in human patients being tested or treated for venereal disease. CHAPTER I LITERATURE REVIEW l.l. HISTORICAL REVIEW Contagious equine metritis (CEM) was first reported in 1977 by Crowhurst (1977). That outbreak involved Thoroughbreds at the National Stud in Newmarket, England. The highly contagious nature of this disease was soon evident and breeding was stopped at the National Stud in an effort to control the spread of infection. However, reports of other outbreaks in the Newmarket area soon appeared (Platt gt_al. l977, Ricketts gt_gl. 1977). Investigation revealed a clinically similar metritis had been present in Ireland in 1976 (O'Driscoll 1977). By 1980, CEM was confirmed in eight countries. The disease was spread mainly through the importation of clinically healthy Thoroughbred breeding stock. The clinical picture in CEM is an endometritis with an associated inflammation of the cervix and vagina. Most infected mares fail to con- ceive, although there is no permanent infertility. Mares have no signs of systemic involvement. Stallions show no signs of genital infections and are passive carriers (Platt.gt_al. l977, Ricketts gt_al. 1977). Transmission is primarily venereal; however CEM is highly contagious and has been transmitted during examination and handling by breeding shed personnel (David et al. 1977). Many difficulties were encountered in isolating the causative agent of CEM. Platt gt_al. (l977) succeeded in culturing a Gram-negative coccobacillus from cervical swabs taken from mares with acute infections. This organism was fastidious, requiring enriched media and increased CO2 for initial isolation, and was morphologically similar to organisms observed in the smears of uterine exudate from infected mares. Transmis- sion studies demonstrated that this newly isolated bacterium caused CEM (Platt‘gt_al. l977, Ricketts gt_al. 1977). The contagious equine metritis organism (CEMO) has not yet been placed into any bacterial taxonomic classification, although Taylor gt_gl. (1978) have proposed the name of Hemgphilus equigenitalis. In September of 1977, a code of practice was published in England which listed recommendations for the control of CEM (David gt_gl. 1977). This included the establishment of a bacteriological screening program for mares and stallions and specific improvements in hygiene standards on stud farms. It also contained recommendations for the treatment of infected horses. Other involved countries developed similar standards. The effectiveness of the code in England was substantiated by the reduction of cases in 1978. By the end of the 1977 breeding season, approximately 200 mares were infected at 29 breeding farms, involving 23 stallions (Powell 1978). In 1978 there were 2 outbreaks of CEM involving 4 mares and 2 stallions (Powell and Whitwell, 1979). The bacteriological screen detected 48 infected mares prior to breeding. 1.2. BACTERIOLOGICAL CHARACTERISTICS OF CEMO l.2.A. Morphology The CEMO is usually described as a Gram-negative coccobacillus, but they may appear as short rods with bipolar staining resembling members of the Brucella and Pasteurella genera (Eaglesome and Garcia 1979). Pleomorphic forms may appear after prolonged incubation. The organism is non~motile and no flagella were observed by electron micro- scopic studies. On 10% chocolate agar, the organism produced tiny, round and raised, grayish-white colonies approximately 0.5 mm in diameter after 48 hours incubation at 37°C (Eaglesome and Garcia 1979, Sahu and Dardiri 1980). 1.2.8 Cultural and Biochemical Characteristics The CEMO is a fastidious organism requiring enriched media and a 5-10% C02 atmosphere. Initial reports claimed a microaerophilic atmos- phere of 5-10% CO2 in hydrogen was required (Platt gt_gl. l977, Ricketts .££_El- 1977) but others obtained growth in 5-10% CO2 in air (Taylor .gt_gl. 1978, 1979). Growth occurred over a temperature range of 30-41°C with optimum growth at 37°C. Best growth of the CEMO was obtained on chocolate agar made with 10% horse blood and Eugon agar, incubated under 5-10% CO at 37°C (Taylor et al. 1978). 2 The bacterium is catalase, cytochrome oxidase and phosphatase posi- tive but unreactive in other conventional tests for biochemical characterization. The CEMO does not ferment carbohydrates nor requires X (hemin) or V (NAD) factors. The lack of dependence on X factor was confirmed by a positive result with the D-amino-levulinic acid test of Kiliam (Taylor_et_al. 1978). However, CEMO growth is stimulated by X factor. Taylor gt_gl. (1978) have determined the DNA base composition, as estimated by melting point temperature, to be 36.1 mole per cent guanine plus cytosine (G+C). The CEMO appears to be sensitive to many antibiotics including benzyl-penicillin, ampicillin, tetracyclines, trimethophrim, sulfa- methoxazole, clindamycin, polymyxin B, furazolidone, and trivetrin (Taylor gt_gl. 1978, Eaglesome and Garcia 1979). Strains of CEMO isolated in England and Ireland were resistant to streptomycin but a strain of CEMO isolated in Kentucky was found to be sensitive to streptomycin (Swerczek 1978a). 1.2.0. Serological Testing Cross-agglutination studies were performed to identify antigenically similar and possibly related bacteria. Serological tests were performed by several investigators, using antisera to other known Gram-negative bacteria and purified anti-CEMO antibodies induced in rabbits (Taylor gngl, 1978, Smith 1978). The CEMO did not react with various Brucella antisera in slide agglutination tests. A variety of known Gram—negative bacteria were tested against the anti-CEMO antisera. Smith (1978) reported consistent cross-reactivity with Moraxella and Mfimg_species. Taylor gt_gl. (1978) reported cross-reactivity with Hemophilus, Moraxella, Neisseria, and Pasteurella species. 1.2.0. Taxonomic Classification Establishment of taxonomic groups in bacteria is difficult. Bacteria multiply asexually and thereby lack the means for genetic exchange which leads to the evolution of distinct species. Instead, muta- tions accumulate which produce gradients of related types. Bacterial groups of related ecologic types are easily recognized. However, division of these groups into species becomes a decision as to what extent 2 types must differ before they are classified as distinct species. Molecular composition of DNA has now been added to the taxonomic criteria as an objective measure of the relationship between bacteria. Closely related bacteria must have similar G+C mole % ratios, although such similarities are not proof of a close relationship. Characteristics of CEMO and serologically similar bacteria are listed in Figure 1. Data for the CEMO is from Taylor gt_gl. (1978) while characteristics for other bacterial genera are compiled from Bergey's Manual of Determinative Bacteriology (Buchanan and Gibbons 1974). Taylor 53:31, (1978) proposed the name of Hemophilus eguigenitalis for CEMO. However, CEMO lacks dependency on either X or V factors, characteristics many consider primary requirements for inclusion into the Hemophilus group. The CEMO does not appear to be closely related to Pasteurella, which is much more biochemically active than CEMO, producing acids from sugars, H25 and reducing nitrates. The remaining bacterial groups, Neisseria, Branhamella, Moraxella and Acinetobacter all belong to the Neisseriaceae family, which is presently undergoing taxonomic revision (Baumann et al. 1968a, 1968b; Henriksen and dere 1968). The CEMO has many characteristics in common with the Neisseriaceae family, and C “I w 5 3 8 § .2: s d 5 a Q R $ ~l O 1‘ a to la a. j? £3 3? £5 3‘ i? CHARACTERI ST I c g $1.: a? £7 5 :9: g SHAPE CB CB CB c C CB CB PRODUCES ACIDS - _ - FROM SUGARS V T D D CYTOCHROME _ OXIDASE T "A T T T T REDUCES NITRATES - + + D + D .- G+CHOLEPERCENT 36° 38-4236-4347-52 uo-uauo-us 39-47 FACULTATIVE _. - ANAEROBE T T T T T PENICILLIN SENSITIVE T T T T T T ‘ PARASITE OF MAMMALIAN _ MUCOUS MEMBRANES T T T T T T DATA FOR CEMO FROM TAYLOR EI.AL. (1978) DATA FOR OTHER BACTERIAL GENERA COMPILED FROM BERGEXL$.UANUAL DE DEIERUINAILVE BACIEBIDLDGIJ 81H ED., BUCHANAN AND GIBBONS (1974) LEGEND NA INFORMATION NOT AVAILABLE + MOST (GREATER OR EOUAL TO 902) POSITIVE - MOST (GREATER OR EOUAL TO 901) NEGATIVE V STRAIN INSTABILITY. NOT DIFFERENCES BETHEEN STRAINS 0 SOME STRAINS POSITIVE, SOME STRAINS NEGATIVE C COCCOID SHAPE CB Cocco-BACILLARY SHAPE a G+C MOLE PER CENT DETERMINED BY THE TEMPERATURE MELTING POINT, ALL OTHERS DETERMINED BY THE BUOYANT DENSITY METHOD FIGURE 1. CHARACTERISTICS OF CEMO AND SEROLOGICALLY SIMILAR BACTERIAL GENERA I believe that it belongs in that group. The recognized species of Acinetobacter are separated from Moraxella based on their lack of cytochrome oxidase. Acinetobacter species are usually saprophytic and resistant to penicillin. It is doubtful that the CEMO belongs in this genus. Neisseria, Branhamella and Moraxella are known parasites of mammal- ian mucous membranes. They are biochemically inert and sensitive to penicillin. Both Neisseria and Moraxella are known agents of genital infections in man. The G+C ratio for Neisseria is 47-52%, which is significantly higher than that determined for the CEMO. However, the G+C ratio for Branhamella and Moraxella, 40-46%tn/the buoyant density method, is close to the CEMO ratio, 36% by the melting point method. This small difference in G+C ratios could arise from differences between methods used in determining the G+C mole per cent. Moraxella and Branhamella are phenotypically and genetically very similar and it has been proposed to fuse the two into one genus (Henriksen and B¢vre 1968). This proposal has not been universally accepted since the genera differ in their gross morphology. Although the significance of shape is unknown, it has a long-standing historical value. However, in some bacteria, morphology is dependent on the phase of the growth cycle. For example, Acinetobacter are definitely bacillary during their exponential growth phase, becoming shorter and more coccoid in appearance when viewed in the other growth phases. The CEMO also has this phase-dependent morphology, as demonstrated by finding predominantly bacillar forms on media which support good growth of the CEMO, and pre- dominantly coccoid forms on less supportive media (Swaney and Breese, 1980). Taxonomic division based on morphology is not suitable for such types of bacteria. Thus the division between Moraxella and Branhamella on the basis of morphology may be artificial. The known characteristics of CEMO do not allow placement into any established taxonomic classification. Many characteristics suggest a Close relationship with Branhamella and Moraxella, genera which are presently undergoing taxonomic revision. Thus final placement of the CEMO must be postponed. l.3. DIAGNOSTIC TESTS FOR CEM l.3.A. Bacteriological Screen The bacteriological screen, established by David gt_al. (1977) includes three sets of swabs taken from several sites at weekly intervals. At least one set must be taken from mares during estrus. Stallions and teasers have swabs taken from the urethra, urethral fossa and prepuce. Samples of pre-ejaculatory fluid are also cultured. Mares have swabs taken from the endometrium, cervix and clitoral fossa including the clitoral sinuses. All three sets of cultures must be negative before the horses can be bred. Careful screening has determined that CEMO can per- sist in the clitoral sinuses after treatment and eradication of the organism from the rest of the genital tract (Simpson and Eaton-Evans 1978). 1.3.8. Serological Screens Serological tests which have been developed for CEM detection in horses include: lO 1. A serum agglutination test with antiglobulin phase (SAT), developed by Benson gt_gl. (1978). 2. Two complement fixation methods (C-FIX), one developed by Croxton-Smith gt_a1. (1978) and the other by Bryans gt_gl. (1979). 3. A passive hemagglutination test (PHA), developed by Fernie _e_t_a_l_. (1979). 4. A plate agglutination test (PAT), developed by Swerczek (1979, 1980). 5. An enzyme-linked immunosorbent assay (ELISA), developed by Sahu M. (1979). All test methods found relatively high serum antibody titers in mares with active CEM infections and diagnostic titers have been established for each test method. Specific information on these tests will be provided later. 1.3.0. Problems in CEM Diagnosis The primary problem in the control of CEM is the detection of clinically healthy carrier mares. Bacteriological methods are not relia— ble for several reasons. The first problem is the low number of organisms present in the carrier state. More importantly, CEMO is a slow growing, fastidious bacterium, often requiring several days to a week for primary isolation on enriched media, allowing more rapidly growing normal genital flora to obscure other colonies on the culture plate. The orig- inal strains of CEMO isolated in England and Ireland were streptomycin resistant, and subsequently, streptomycin was used in the primary isola- tion media to inhibit the growth of most normal flora. However, in 1978, 11 a streptomycin sensitive strain of CEMO was found in Kentucky (Swerczek 1978a). It was also determined that certain organisms among the normal genital flora produce a metabolite which inhibits the growth of CEMO (Swerczek 1978b, Altherton 1978). These combined factors make the identification of a streptomycin sensitive CEMO carrier mare by bacterio- logical methods quite difficult. Serological tests demonstrated relatively high serum antibody titers in mares with active CEM infections. However, antibody levels in carrier mares were often below diagnostic levels. Stallions do not have demon- strable CEM titers because they do not become infected with CEMO but are simply passive carriers of the organism. Thus the present serological tests for CEM are not useful for detection of infected stallions or carrier mares (Benson gt_gl. 1978, Croxton-Smith gt_gl. 1978, Fernie et a1. 1979, Swerczek 1978a, 1979, 1980). 1.3.0. Comparison of Screening Methods There are many advantages in using serological over bacteriological screening tests. Serological samples are easier to obtain and are more stable. The samples can be obtained at any time, independent of the mare's reproductive cycle. Serological tests are less labor intensive than culture techniques and test results can be obtained much faster. Most serological tests can be completed within 2 days while culture results can be called negative only after 6 days incubation (Anon., Vet. Rec., 1980). The major disadvantage of serological tests is that they give only presumptive results. True confirmation of bacterial infection can only be 12 achieved by actual recovery of organisms from the infected site. The immune response of antibody production starts after antigen exposure and usually takes several days to occur. Thus a serum sample taken very early in an infection would yield negative results when culture tech- niques may have been able to detect the organism. Also serological tests may remain positive after resolution of the infection. 1.4. SEROLOGICAL TECHNIQUES FOR CEM DETECTION 1.4.A. General Review Serology is, by definition, the study of igdgitr9_antibody-antigen reactions. These reactions may be demonstrated by a variety of tech- niques, depending mainly on the nature and struCture of the antigen. The antibody-antigen reaction itself is affected by many factors, includ- ing time, temperature, electrolyte concentration, pH and ratio of antigen to antibody. Many techniques are available to measure antibody levels, with varying degrees of specificity and sensitivity. Antibody-antigen reac- tions are by their very nature specific. Serological test Specificity refers to the degree of accuracy with which one can detect the desired agent, be that antigen or antibody. Sensitivity refers to the level of detectable antibody—antigen complexes. Most antibody-antigen reactions occur in two distinct stages. The first is the specific combination of antibody and its corresponding antigen or hapten. The second stage is the formation of an observable reaction such as precipitation, agglutination or complement fixation. These two stages often overlap, each stage requiring a different time 13 span, temperature, electrolyte concentration or other specific condition. Antibody-antigen reactions may also occur without the development Of an observable reaction. 1.4.B. Diagnostic Test Considerations i. Antibody Detection. Serological tests used to measure the humoral immune response (antibody production) fall into three categories. The most sensitive are the primary binding tests that directly measure the amount of immune complex formed (Tizard 1977). Competitive radio- immunoassays are in this category and can detect antibody concentrations as low as 5x10.5 pg protein/m1. Secondary binding tests measure the consequences of immune complex formation in 31359, These tests are less sensitive than primary binding tests, but are considerably simpler to perform. The types of measurable antibody-antigen interaction include precipitation of soluble antigens, agglutination Of particulate antigens, and activation of complement. The sensitivity of secondary binding tests varies with the procedure, requiring from as little as 5x10'5 pg protein/m1 to as much as 3 x 101 pg protein/m1 for an observable reaction. Tertiary tests measure the consequences of the immune response 1_ vivo. An example is passive cutaneous anaphlyaxis, with a sensitivity of 2x10-2 pg protein/ml (Tizard 1977). ii. Antibody Quantification. Most serological tests are semi- quantitative. While a simple positive or negative result is sufficient for some tests, it is usually necessary to estimate the amount of anti- body present for diagnostic purposes. The serum titer, a common unit of comparison in serological tests, is defined as the reciprocal of the 14 highest dilution of serum yielding a positive reaction. The serum dilu- tion itself is also used as a comparison unit. A common problem in interpretation of both systems is whether the unit refers to the dilution of serum used in the test procedure or the actual final dilution of serum in the test reaction. The SAT method of Benson gt_§l. (1978) refers to the actual final dilution of serum in the test reaction while the PHA method of Ferniqut_gl. (1979), the ELISA method of Sahu gt_gl. (1979), and the two C-FIX methods of Byrans gtggfl, (1979) and Croxton-Smith_gt_§l. (1978) refer to the dilution of serum used in the test proper. While these methods of estimating antibody levels are adequate for diagnostic purposes, they are not a true measure of antibody concentra- tion. The demonstrable reactivity of a serum sample is dependent on the immunoglobulin Class of antibody present as well as the concentration. The immunoglobulin Class responsible for the reaction is especially critical in secondary binding tests since different classes have varying ability to form the secondary reactions of precipitation, agglutination or complement fixation. Thus, secondary binding methods can over- or underestimate antibody levels (Tizard 1977). iii. Immunoglobulins of Test Species. Five classes of immuno- globulins are known, designated as 196, IgM, 19A, 190, and IgE. All of the domestic animals possess IgG, IgM, and IgA. It is probable that most possess IgE, although it has not been isolated in all species. 190 has been found only in man, and its function is not well understood. The basic Characteristics of these immunoglobulin classes do not differ between species. However the number and types of subclasses do vary between species (Tizard 1977). 15 Immunoglobulin G. 196 is the immunoglobulin class found in the highest concentration in the serum (Tizard 1977). It plays a major role in antibody-mediated defense mechanisms. Its small Size allows passage through blood vessels to tissue spaces and body surfaces. IgG can opsonize, agglutinate and precipitate antigen. It can activate complement only if sufficient numbers have accumulated in a specific configuration on the antigen (Delaat 1974). Immunoglobulin M. IgM is found in the second highest concen- tration in the serum of most animals (Tizard 1977). It is the major immunoglobulin produced in a primary immune response. IgM is also produced in a secondary response, but is often masked by the massive IgG production. On a molar basis, IgM is considerably more efficient than IgG at complement fixation, opsonization, agglutination, and viral neutralization (Delaat 1974). Due to its large size, IgM is confined to the blood vascular system. Immunoglobulin A. 19A is found in the second highest concen- tration in man, though it is a minor component in animal serum (Tizard 1977). It is the major immunoglobulin found in external secretions. AS such, its primary function is to protect the exposed environs of the gastrointestinal, respiratory and genital tracts, the udder and eyes. IgA cannot activate complement or act as an opsonin. It can agglutinate particulate antigens and neutralize viruses. Immunoglobulin E. IgE is in extremely low concentrations in the serum and has not been found in all species (Tizard l977). IgE mediates Type I hypersensitivity reactions seen in allergies and anaphyl- axis. It plays no known role in serological testing. 16 1.4.C. Serological Tests for CEM Diagnosis i. Serum Agglutination Test (SAT). The SAT for CEM developed by Benson gt_al. (1978), relies on the visible agglutination of the CEMO antigen. The SAT antigen is a boiled saline suspension of CEM bacteria. Antigen and diluted serum are incubated together at 37°C and the degree of bacterial agglutination determined. Agglutination of more than 25% of the bacterial antigen is a positive test reaction. Species specific antiglobulin sera is added to the washed test mixture to allow detection of non-agglutinating antibodies bound to the CEM antigen. The use of antiglobulin sera increases test sensitivity and clarifies most incon- clusive SAT test results. Serum samples from unexposed horses had SAT titers up to 20. Titers greater than or equal to 80 were considered positive. Samples with titers of 40 were considered inconclusive with SAT and further tested with equine antiglobulin sera. If the titer with antiglobulin sera remained the same as the SAT result, the sample was considered negative. If the titer increased, the sample was considered positive. ii. Plate Agglutination Test (EAT). Plate agglutination tests rely on the rapid formation of visible agglutination patterns when the antigen preparation is mixed with serum containing the corresponding antibody. Swerczek (1978a, 1979, 1980) developed a PAT for CEM detection in horses. The antigen preparation is a stable suspension of intact CEM bacteria. Two test procedures are available, depending on the breed of horse being tested. The Thoroughbred procedure requires a stronger agglutination pattern for a significant positive test result and uses less serum than the non-Thoroughbred procedure. Test results of complete 17 agglutination (4+) with sera diluted 1:2 were considered positive for CEM. Lesser reactions were recorded but the significance of these reactions are unknown (Swerczek 1979). iii. Complement Fixation (C-FIX). Complement fixation tests are based on the ability of certain antibodies to bind and activate (fix) complement. Antibodies can fix complement only after binding to an antigen. Thus, the amount of activated complement is proportional to the amount of antigen bound antibody. Sheep red blood cells are readily lysed by activated complement and are used as an indicator of complement activity. Complement fixation tests consist of two distinct stages. In the first (test) stage, serum and the specific antigen are mixed together in the presence of complement. Interaction of serum antibody and specific antigen may fix complement. In the second (indicator) stage, antibody coated sheep erythrocytes (RBCs) are added to the test mixture. The antibody coating the sheep RBCs will activate any remaining comple- ment and cause red cell lysis. Thus. the amount of red cell lysis is proportional to the excess of complement over the amount of specific antibody present in the test stage. Antibody coated sheep RBCs and complement activity must be carefully balanced so that complement fixa- tion in the test stage will prevent lysis of the coated RBCs in the indicator stage. All reagents and serum samples are also tested for their ability to non-specifically absorb complement in the absence of antigen. The specific C-FIX methods developed by Bryans gt_gl. (1979) and Croxton-Smith (1978) are similar in that they both use a saline 18 suspension of intact CEM bacteria as the antigen. Bryans et_gl, (1979) showed that the reactivity of the CEMO antigen preparation is related to the proportion of organisms with intact bacterial envelopes. Both methods have established the test reaction of 4+ (no lysis) in the 1:4 dilution of serum as the minimal positive test result in horses. Croxton-Smith gt_al. (1978) had a high percentage of samples with anti- complementary activity, 26% AC, although Bryans gt_gl: (1979) did not report such findings. Anticomplementary activity invalidates C-FIX test results and decreases the usefulness of the method for diagnostic purposes. iv. Passive Hemagglutination (PHA). Erythrocytes that are treated with dilute tannic acid will adsorb proteins onto their surfaces (Boyden 1951, Stavitsky 1954). Using RBCs as inert carriers of various antigens aids in the visualization of antibody-antigen reactions in test systems. By this method, many precipitation reactions can be converted into more sensitive agglutination test systems (Boyden 1951, Stavitsky 1954, Delaat 1976). Serum samples used in PHA tests must have antibodies to the carrier system removed prior to testing for the added specific anti- gens. This is accomplished by incubating serum samples with tanned but uncoated carrier cells. This adsorbed serum is used in subsequent PHA tests. Fernie gt_gl. (1979) developed a passive hemagglutination test for CEM. This test method differs significantly from the previously described tests in that it uses soluble CEMO antigens rather than intact bacteria. A saline suspension of CEMO is disrupted by ultrasonic vibra- tions and the resulting cellular debris is removed by centrifugation. 19 Formolized (Sequeira and Eldridge, 1973) and tanned (Boyden 1951) turkey erythrocytes are used as carriers for the antigen preparation. Diagnostic PHA serum titers for CEM in horses have been established. Titers up to and including 32 were considered negative for CEM exposure. Titers of 256 or higher were considered positive. Results between 32 and 256 were suspicious for CEM exposure but inconclusive (Fernie gt_gl. 1979). v. Enzyme-linked Immunosorbent Assay (ELISA). Enzyme immuno- assay is a recently developed quantitative serological technique which is analogous to radioimmunoassay and quantitative immunofluorescence. Advantages of ELISA include sensitivity (ng/ml range), simplicity, stability of reagents, lack of radiation hazard and relatively inexpen- sive equipment. In the ELISA assay, antigen is fixed to a solid phase, incubated with test serum, then incubated with species-specified anti- immunoglobulin labeled with enzyme. Thus, enzyme adherent to the solid phase is related to the amount of antibody bound in the first incubation phase. Substrate is added and the enzyme activity is related to anti- body concentration. Sahu gt_al. (1979) developed an ELISA test for CEM. Soluble CEMO antigens, obtained from a sonicated saline suspension of CEMO, are fixed to U-bottom wells of microtiter plates. Anti-equine 196 is labeled with alkaline phosphatase and p-nitrophenylphosphate disodium is used as the indicator substrate. Positive serum samples have an Optical density (00) at 405 nm that is 2.5 times the average 00 of controls. Samples positive at a 1:20 or higher dilution of serum are considered positive for CEM. 20 1.5. EPIDEMIOLOGICAL STUDIES The purpose of epidemiological screening tests is to quickly and reliably separate the exposed population from the total test population. The suspect population is then subjected to additional specific testing. Since the function of a screening test is to eliminate only unexposed subjects, the tests used should have a low probability of false negative results, while a certain percentage of false positive results are quite acceptable. Serological tests can be excellent epidemiological screening tests. The test samples are easily obtained and stable at -20°C for prolonged storage. Serological tests for bacterial infections are easily standard— ized, relatively economical, and less labor intensive than bacteriologic- al culture techniques. Some Characteristics of serological tests, such as retention of seropositivity after resolution of infection, allow a better estimation of total population exposure to an agent, since bacteriological methods give only a presently infected population estimate. 1.5.A. Equine Studies Initial investigations of serological methods for CEM detection in horses found low-titered antibody levels against the CEMO in the pre- sumedly unexposed horse population. Low-titered, non-specific reactions are common in serological tests. These results may represent serological cross-reactions between antigens of commensal or environmental organisms and those of the CEMO. Benson gt_gl. (1978), using the SAT method, found titers of 20 and occasionally up to 40 in horses with no known contact with CEMO. 21 Croxton-Smith gt_gl. (1979) found 38% of their unexposed horse popula- tion had C-FIX test reactions at the 1:2 serum dilution and an additional 10% reacted at the 1:4 serum dilution. Fernie gt_gl. (1979) used serum samples from two groups of unexposed horses. One group of frozen serum samples was obtained in 1976, prior to the first outbreak of CEM in England. The second group of serum samples was taken from clinically healthy horses in 1977 and 1978. Using the PHA for CEM, they found only 2.3% of the samples from 1976 had titers greater than 8, with the highest titer being 32. The PHA results for the 1977 and 1978 samples showed that 40% had titers greater than 8, with the highest being 64. Similar results were found in serum samples taken from clinically healthy Thoroughbred mares from Kentucky during the 1978 CEM outbreak in the United States. Of these samples, 22% had PHA titers greater than 8, with the highest titer being 128 (Ferniqujgal, 1979). Croxton-Smith §t_gl. (1978) found that the incidence of low- titered results increased with the age of the horse. Fernie gt_gl. (1979) proposed that the increasing background titers found in Clinically healthy mares may be due to indirect exposure to CEMO and subsequent sub-clinical infections. The increased background titers against CEMO between the 1976 and 1977-78 clinically healthy horse populations corre- lates with the increased opportunity for CEMO exposure. 1.5.B. Human Studies Taylor and Rosenthal (1978a, 1978b, and Taylor 1979), using the SAT method of Benson et a1. (1978), found a higher percentage of agglu- tinins to CEMO in patients attending a genito-medical clinic than in 22 other test populations. The percentage of patients with agglutinins to CEMO, with titers greater than or equal to 20, varied from 2% among women attending an antenatal clinic to 22% among women attending a genito- medical clinic. There was a 7% rate among healthy adult females. 0f men attending a genito-medical clinic, 13% were positive compared with only 4% of healthy males. Additional studies performed on men with non- gonococcal urethritis found agglutinins to CEMO, with titers greater than or equal to 20, in 37.6% of those tested; 12.5% showed a four-fold or greater rise in titer during the course of their illness. These observations suggest that the CEMO or related organism may be involved in human venereal disease. Attempts to isolate CEMO from these patients have been unsuccessful. 1.5.C. Bovine Studies Corbel and Brewer (1980), using the SAT method of Benson gt_§l. (1978), found agglutinins to CEMO in a high percentage of bovine samples from the United Kingdom. However, 96.2% had titers less than or equal to 40, with 69.7% having titers less than or equal to 20. The signifi- cance of these results is not known, and they may simply represent a higher normal background titer to CEMO in cattle. l.6. PROPOSED STUDY The primary objective of this study is to identify species with a high probability of CEMO exposure. Another objective is to compare the serological methods for CEM detection, in terms of both test results and technical aspects of test mechanics. The final objective is to examine the morphology of the streptomycin sensitive strain of CEMO. 23 Species to be tested include sheep, horses, dogs, cattle, pigs and humans. Three serological tests for CEM detection will be performed on 200 serum samples from each test group. An additional 400 human samples will be tested to give three human population groups similar to those of Taylor and Rosenthal (1979a). The serological tests for CEM used in this study will be: 1. The PAT method of Swerczek (1979), using the more sensitive non-Thoroughbred procedure. 2. A C-FIX method based on the work of Bryans gtggfl, (1980). 3. The PHA method of Fernie gt_gl. (1979). All three tests will be performed on all serum samples if sample volume is sufficient. Samples will be screened at an initial dilution. All reac- tive samples at the screening dilution will be further titrated to deter- mine final serum titer. The streptomycin sensitive strain of CEMO will be examined using scanning and transmission electron microscopy techniques. It is known that the streptomycin resistant strain of CEMO is encapsulated (Swaney and Breese 1980), and that the capsule is responsible for some of the CEMO antigenic specificity (Bryans gt_gl. 1980). The goal of this portion of the study is to determine if the streptomycin sensitive strain is encapsulated and, if so, how that capsule compares to that of the strep- tomycin resistant strain. It is hoped that the results of this preliminary work will be use- ful as a guide for further research on the habitat and host-range of the CEMO. CHAPTER II METHODS AND MATERIALS This chapter describes in detail the various components and pro- cedures for the serological tests performed in this study. It also briefly covers the factors involved in the electron microscopy studies. Section 1 covers the obtaining and storage of serum samples for the epidemiological study. The next section deals with growth and mainte- nance of the two strains of CEMO. Actual procedures for the serological tests, including reagent preparation and necessary preliminary test pro- cedures, are covered in section 3. The final section deals with the preparation of the CEMO for electron microscopy studies. 2.1. SERUM SAMPLES Serum samples were obtained from the Animal Health Diagnostic Laboratory at Michigan State University, the Michigan Department Of Public Health, the Michigan Department of Agriculture (Giegley Labora- tory), the Pathology Department of Michigan State University, a local slaughter house and several area horse farms. All samples were drawn in 1979 and stored at -20°C. Minimum acceptable sample volume was set at one ml of serum. Grossly lipemic or hemolyzed samples were rejected. At least two-hundred samples were collected from each animal group. The criteria used in sample selection were based on the known distribu- tion of CEM antibodies among horses. Samples known to be from young or 24 25 neutered animals were rejected since the disease is sexually transmitted in horses. Samples known to be from male animals were rejected to avoid the problem of serological non-reactivity found in equine passive carriers of CEMO. Thus samples selected were primarily from sexually mature, unaltered females. Over six-hundred serum samples from both men and women were col- lected from the Michigan Department of Public Health. Samples were selected from those submitted for VDRL (syphilis) or rubella testing. Patient confidentiality was preserved by recording only the age and sex of the donor. These samples were divided into 3 groups of approximately two-hundred each. Groups were selected to facilitate comparison to a preliminary study conducted in England by Taylor gt_gl. (1978). Rubella samples selected were primarily from women Of Child-bearing age. This group was chosen to correspond to Taylor's antenatal group. The VDRL samples selected were divided into 2 groups based on the reason given for test submission. Pre-employment or pre-marital VDRL screening samples were considered representative of the normal population, while samples submitted for syphilis diagnosis or treatment were considered a high risk group, in the sense that they are more likely than the general population to have been exposed to venereal disease. The screening population and the high risk group correspond to Taylor's healthy and V0 patient group. 2.2. CONTAGIOUS EQUINE METRITIS ORGANISM Two strains of CEMO were Obtained, under USDA permit #8641, from Dr. Thomas Swerczek (University of Kentucky, Lexington, KY 40546). One strain, #78-188, was a streptomycin resistant strain and the other, 26 #79-314, was a streptomycin sensitive strain. The streptomycin resistant CEMO is considered the type culture, and as such was used in the C-FIX and PHA antigen preparations. The streptomycin sensitive CEMO was examined with electron microscopic techniques to compare to previous studies of the streptomycin resistant CEMO. Both strains were maintained on chocolate agar plates under micro- aerophilic conditions at 37°C. Plates were made with 10% sterile horse blood in Columbia agar base.1 MicroaerOphilic conditions of 5-10% CO2 in air were obtained with a C02 generating system (GasPak)1 without using a palladium catalyst. 2.3. SEROLOGICAL TESTS 2.3.A. Plate Agglutination Test (PAT) i. Reagents. All reagents for the PAT were supplied by Dr. T. Swerczek. The PAT antigen consisted of a stable suspension of intact CEMO. A PAT buffer was used to dilute samples to determine final titer. ii. Procedure. In the screening procedure, 0.060 ml of serum was mixed with 0.025 ml of PAT antigen on a glass plate. Results were read after a 10 minute room temperature incubation. Results were graded 0 to 4+, with 0 indicating no visible agglutination and 4+ consisting of flocculent agglutination, yielding a ring pattern with clearing of the test area. Plates were covered during incubation to prevent evaporation. Test results were read using a diffuse light source beneath the test plate. 1BBL, Div. of Becton, Dickinson and Co., Cockeysville, MD. 27 All samples with a result of 3+ or greater in the screening test were serially diluted in PAT buffer, heat inactivated at 56°C for 45 minutes and retested to determine final serum titer. 2.3.B. Complement Fixation (C-FIX) i. Reagents. The C-FIX reagents and CEMO antigen were prepared according to the method of Bryans gt_gl, (1979). The antigen was pre- pared from the streptomycin resistant strain of CEMO, #78-188. For more information on C-FIX antigen preparation see Appendix A. A tris buffered C-FIX diluent, pH 7.4 i .04 was used for all reagent and serum dilu- tions. Sheep RBCs, anti-sheep serum (hemolysin) and lyophilized guinea pig complement were obtained from commercial sources.1 All test and preliminary titrations were performed using Cooke Microtiter2 U-bottom plates, microdiluters and pipette droppers. ii. Procedures. Preliminary testing of diluent, hemolysin, comple- ment and antigen were performed according to the procedures in Delaat (1974). Checkerboard titrations of hemolysin and complement were used to determine the optimal working concentration of both reagents. The con- centration of complement used in the test procedure was 2.5 times that which caused 50% hemolysis of an equal volume of 1% sensitized sheep RBC suspension (2.5 HD Hemolytic and anti-complementary (AC) activity 50)° of the antigen was checked according to the procedure in Delaat (1974). Antigen reactivity was matched in a C-FIX antigen supplied through Dr. J. Bryans. 1 2 Colorado Serum Co., Denver, CO. Dynatech Laboratories, Inc., Alexandria, VA. 28 The test procedure followed was a standardized micromethod that used 0.025 ml volumes for all reagents (Delaat 1974). To perform the test, samples were diluted in C-FIX diluent and heat inactivated at 56°C for 30 minutes immediately prior to testing. Equal volumes of diluted serum, antigen and complement (2.5 H050) were placed in each test well. Plates were covered, mixed and incubated for 2 hours at room temperature. Following incubation, an equal volume of a 1% sensitized sheep RBC sus- pension was added to all wells. Plates were then incubated for 30 minutes at 37°C, mixing after 15 minutes and again at 30 minutes. Test results were recorded after the cells had settled. Serum titer is de- fined as the highest serum dilution to yield a 2+ or greater result, which is hemolysis of less the 50% of the sensitized sheep RBCs. Test controls included in each run were antigen hemolytic and AC controls, 50% and 100% lysis complement controls, diluent hemolytic controls, known positive and negative samples, and individual AC controls. Anticomplementary samples do not yield valid test results and were recorded simply as AC. Preliminary titrations were performed on approximately 15 samples from each animal group and the screening human group to determine the test screening dilution for each group. These presumed unexposed samples were selected from those which had no reaction with the PAT. Samples were diluted 1:4 in C-FIX diluent and heat inactivated. Using the micro- titer equipment, additional 2-fold serial dilutions were made so that serum dilutions of 1:4 through 1:256 were tested. Since no preliminary samples were reactive at the initial 1:4 dilution, all samples were screened at 1:4. Any samples with test results of 2+ or greater at this 29 screening dilution were serially diluted and retested to determine final serum titer. 2.3.C. Passive Hemagglutination Test (PHA) i. Reagents. The CEMO antigen preparation contained soluble anti— gens obtained from a sonicated CEMO suspension, strain #78-188. Cellular debris and intact bacteria were removed through ultracentrifugation. The antigen (supernatant) was stored at -20°C. Citrated blood (ACD) was collected from several 15 week-old turkeys. These cells were washed in saline and the buffy coat removed. Washed RBCS were subsequently preserved with formaldehyde using the method of Sequeira and Eldridge (1973). Portions of the formolized RBCS were re- suspended in a phosphate buffered saline (PBS), pH 7.2, to a 4% suspen- sion and treated with a 0.0005% w/v tannic acid solution. Tanned RBCS were sensitized with CEMO antigen by mixing 1 volume of 4% tanned RBCS with 4 volumes of PBS (pH 6.4) and 1 volume of CEMO antigen. The mixture was incubated at 37°C for 30 minutes, washed several times in saline, then reconstituted to a 1% suspension with 1% bovine serum albumin (BSA) in saline. For more information on PHA antigen preparation, including details on turkey RBC treatment, see Appendix A. All tests and preliminary titrations were performed using Cooke Microtiter U-bottom plates, microdiluters and pipette droppers. ii. Procedures. The concentration of CEMO antigen used to sensi- tize the tanned RBCS was determined through preliminary titrations. The selected antigen dilution yielded a clear cell sedimentation pattern with both positive and negative sera, and antigen reactivity was matched to a 3O PHA antigen prepared at Nellcome Laboratories and supplied through Dr. D. S. Fernie. Samples were heat inactivated for 30 minutes at 56°C, diluted in 1% BSA saline and adsorbed with an equal volume of packed formalized RBCS. Test wells contained 0.025 ml of diluted serum and 0.050 ml of antigen. Individual serum controls contained 0.025 ml of diluted serum with 0.050 ml of 1% formolized RBCS in 1% BSA saline. Additional con- trols included in each test run were known positive and negative serum samples and diluent controls. Plates were covered, mixed by swirling and incubated for 2 hours at room temperature. A negative sedimentation pat- tern was defined as a tight button of cells whose diameter was less than 1/2 the diameter of the bottom of the reaction well. Preliminary titrations were performed on approximately 20 samples from each animal group and the screening human group to determine the test screening dilution for each group. These presumed unexposed samples were selected from those that were non-reactive with the PAT and negative (excluding AC) with C-FIX. Preliminary titrations were performed in duplicate using both adsorbed and unadsorbed sera. Samples were initially diluted 1:8 and additional 2-fold serial dilutions were made with the microtiter equipment. Serum dilutions of 1:8 through 1:128 were tested. Test results were coded 0 to 5 with 0 representing a negative result at the 1:8 dilution and 5 being a positive result at the 1:128 dilution. The screening dilution for each species was set at the dilution at or above the group's average coded PHA result plus 1 SD. Mass screening was performed on unadsorbed serum samples only if no significant differ- ences were found between the preliminary titrations on adsorbed and 31 unadsorbed sera. All samples positive at the screening dilution were adsorbed with packed formolized turkey RBCS, serially diluted and retested to determine final serum titer. 2.4. ELECTRON MICROSCOPY STUDIES 2.4.A. Scanning Electron Microscopy (SEM) The streptomycin sensitive strain of CEMO, #79-314, was grown in Schaedler broth1 at 37°C under microaerophilic conditions. A 5 day-old culture was fixed in 1% gluteraldehyde overnight at 4°C, dehydrated with If 'TWWTTTVI a graded alcohol series and placed on glass coverslips. Samples were sputter coated with gold and examined in an 151 Super 111.2 2.4.8. Transmission Electron Microscopy (TEM) The streptomycin sensitive strain of CEMO, #79-314, was grown in Schaedler broth1 at 37°C under microaerophilic conditions. A 5 day-old culture was fixed with Karnovsky's solution, washed in Zetterqvist buffer and post-fixed in a Zetterqvist-osmium solution. The sample was dehydrated with a graded alcohol series, using propylene oxide in the final dehydration phase. The CEMO was embedded in a Epon-Araldite resin mixture. Ultra-thin sections were stained with uranyl acetate and lead citrate solutions and examined on a Philips EM-201.3 TBBL, Div. of Becton, Dickinson and Co., Cockeysville, MD. 2International Scientific Instruments, Inc., Santa Clara, CA. 3N. v. Philips, Gloeilampenfabrieken, Eindhoven, Netherlands. CHAPTER III RESULTS AND DISCUSSION 3.1. INTRODUCTION The primary objective of this study is the identification of species with a high probability of CEMO exposure. Secondary objectives include a comparison of test methods in terms of test results and the technical aspects of test mechanics, and a morphological study of the streptomycin sensitive strain of CEMO. Presumed exposure to CEMO was estimated from the strength and dis- tribution of titered test results in species with no defined normal group. Human results were compared to the presumed normal population to discern whether there was CEMO exposure. Contingency tests were used to check for correlation of test results. Only results from samples tested by all three test methods were used to generate the contingency tables for correlation testing. Test methods were also Compared on a technical basis. Points of comparison included serum requirements, reagent preparation and stabil- ity, technical skills required for test performance, and suitability for use as an epidemiological screening test. The morphological study of the streptomycin sensitive strain of CEMO included SEM examination of shape and surface detail, and TEM studies of internal structures. The EM work was compared to previous 32 33 studies performed on the streptomycin resistant strain of CEMO by Swaney and Breese (1980) and Sahu and Dardiri (1981). A description of the figures and tables of results is in section 3.2. The next section evaluates the serological test results concerning the possibility of CEMO exposure. Correlation of serological test results is examined in section 3.4. The final section consists of the morphological study of the streptomycin sensitive strain of CEMO. 3.2. DESCRIPTION OF FIGURES AND TABLES 3.2.A. Percentage Positive to CEMO Table 1 lists the percentage positive (reactive) to CEMO antigens by test method and species, positive referring to samples reactive at or above the screening dilution. The PAT used straight serum for screening, while the serum dilutions used for C-FIX and PHA screening were deter- mined on an individual species basis through preliminary titrations of selected samples from each group. With the C-Fix, no preliminary titrated samples from any group were positive at the initial dilution of 1:4. Therefore, all groups were screened at a 1:4 dilution of serum. The titered PHA results varied among the species. Human, sheep, dog and swine samples were screened at a 1:8 serum dilution. Horse and cattle samples had measurable PHA titers, and thus they were screened at a 1:16 and 1:32 serum dilution respectively. For preliminary PHA titration results, see Appendix 8. 3.2.8. Titered Results Samples reactive at the screening dilution were serially diluted to determine final serum titer. No samples were reactive beyond the .20....2129 ofzuwmum m1... w>om< mo h< m>_._.u:._mon_ c 34 A:.cfiv mm cam Ao.m~v am Aw.mv m amm Am.mv am cam m4hh_»_moa mu - —' - 201 FIG 45 164 -' - - - - 209 HORSE 20 178 -» 2 - — - 200 CATTLE S4 171 - S 4 - - 234 AC ' ANTI'COMPLEMENTARY TABLE 3. DISTRIBUTION OF PASSIVE HEMAGGLUTINATION TITERS FOR CEMO ANTIGENS. PASSIVE HEHAGGLUTINATION TITERS TR°UP TESTED nes° 8 15 32 64 128 256 512 1024 T°TAL "RCLseaoups 591 10 10 4 - - - - - 515 SCREENING 200 l 3 1 - - - - - 205 RUBELLA 203 - 1 2 - - -: - - 206 HIGH RISK 188 9 6 1 - - - — - 204 SHEEP 166 - - 5 19 3 3 2 - 198 nos 193 5 5 - - - - - - 205 P16 180 15 2 1 I 5 3 - 2 210 HORSE 169 MT 15 10 2 2 3 - - 201 CATTLE 215 NT NT 11 8 4 1 1 - 240 ' NEGATIVE AT INITIAL SCREENING DILUTION. NT ‘ NOT TESTED. HORSE SERA SCREENED AT 1:16 DILUTION. CATTLE SERA SCREENED AT 1:32 DILUTION OF SERA. ALL OTHER GROUPS SCREENED AT 1:8 DILUTION OF SERA. ( woN I _ mN + a3Uzmoz_.pzou 8N I _ c + 352 2.. _SN R 2 _ RN I _ S 33.2 5.... I I . a: N a RN o 8 n nu...“ ,3 + N c i a a + .. - .255: SN I a + 3:8 2.. _SN 8 2 _ n2 I _ a 32:: 5.6 a: I 2 a RH o a In «Juno-sic: m + a a N o c + .. - 9.: I — c + 35.2 2.. m2 mm 2 SH _ a 3:8 5.... 8‘ I mm a N: c a In Gianna“... m + c o a o N + m I Nam I — c 4 3:2 2.. Tom 3 .2 m? an 32.3.. 5.... 2m I S c mm... o R In 4.....qu :N + m a ON a m + a 2.55: 32 I _ a: + 32:: 2.. 3.: 8m 2 B: I Nm 358 x26 3.: I an N 32 N e. I n 33am.“ R: + mN a o: N N + .. 9,252.3 3.2.: I + I + I 2.. _ 2.25 , 2.... 2 x16— .: m3m

Qm¢<=awl_:u .m u3m