Citation
Evaluation of Anaplasma marginale major surface protein 3 (MSP3) as a diagnostic test antigen

Material Information

Title:
Evaluation of Anaplasma marginale major surface protein 3 (MSP3) as a diagnostic test antigen
Creator:
Alleman, Arthur Rick, 1954-
Publication Date:
Language:
English
Physical Description:
xii, 124 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Anaplasma marginale ( jstor )
Antibodies ( jstor )
Antigens ( jstor )
Antiserum ( jstor )
Cattle ( jstor )
DNA ( jstor )
Gels ( jstor )
Genomics ( jstor )
Membrane proteins ( jstor )
Reactivity ( jstor )
Anaplasma -- immunology ( mesh )
Anaplasmosis -- diagnosis ( mesh )
Anaplasmosis -- veterinary ( mesh )
Antigens, Surface -- diagnostic use ( mesh )
Department of Pathobiology thesis Ph.D ( mesh )
Dissertations, Academic -- College of Veterinary Medicine -- Department of Pathobiology -- UF ( mesh )
Immunologic Tests ( mesh )
Laboratory Techniques and Procedures ( mesh )
Membrane Proteins -- diagnostic use ( mesh )
City of Gainesville ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1995.
Bibliography:
Includes bibliographical references (leaves 115-122).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Arthur Rick Alleman.

Record Information

Source Institution:
University of Florida
Rights Management:
Copyright Arthur Rick Alleman. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
49846151 ( OCLC )
028216649 ( ALEPH )

Downloads

This item has the following downloads:


Full Text













EVALUATION OF ANAPLASMA MARGINALE
MAJOR SURFACE PROTEIN 3 (MSP3) AS A DIAGNOSTIC TEST ANTIGEN




















By

ARTHUR RICK ALLEMAN



















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 1995


































Copyright 1995

by

Arthur Rick Alleman


































This document is dedicated to my beautiful wife Mary, my wonderful children, Arthur and Grace, and to God who we love and serve.















ACKNOWLEDGMENTS


I would like to acknowledge my committee members, Dr. John Dame, Dr. John Harvey, Dr. Paul Gulig, and Dr. Rose Raskin, for their guidance and assistance in the completion of this project. I would like to express a special appreciation to my major professor, Dr. Tony Barbet. He has been a great instructor, mentor, guidance counselor, and friend. He has always been there for me as well as for his other graduate students and staff. His devotion of his time in a completely unselfish manner has been an inspiration to me on many occasions, particularly in difficult times. His weekly individual meetings, and monthly lab meetings were a lesson well learned and something I hope to continue during my professional career. His guidance was most valuable to the completion of these experiments.



I would also like to express my appreciation to some of my coworkers, Dr. Bill Whitmire, Annie Moreland, Renee Blentlinger, Anna Lundgren, and Michael Bowie. We all worked together harmoniously in the lab, each willing to take time out to assist the other, sharing valuable work experiences. I am glad friendships last longer than Ph.D. projects.



iv
















TABLE OF CONTENTS



ACKNOWLEDGMENTS .................. iv

LIST OF FIGURES .................. vii

ABBREVIATIONS .................. viii

ABSTRACT ................... .... xi

CHAPTER 1 INTRODUCTION . . .... .. . . .. 1

CHAPTER 2 LITERATURE REVIEW ............. 5
General Methods of Serologic Diagnosis ... . 5
Methods for Detection of A. marginale Infected
Cattle ..... .. ............ 9
The Rapid Card Agglutination Test (CA) . 11 The Complement Fixation Test (CF) ..... 13 The Indirect Immunofluorescence Test (IIF) 14 The Radioimmunoassay (RIA) ....... 15 Enzyme-Linked Immunosorbent Assays (ELISA) 16 Nucleic Acid Probe Hybridization ..... 21
Previous Experiments ......... ..... 22

CHAPTER 3 MATERIALS AND METHODS ... .. . 26
Anaplasma marginale Strains . . . . . 26
Initial Body Preparation ........... 27
Babesia bovis and Babesia bigemina Antigen
Preparation. ........ . . . 28
Antisera used for Immunoblots ...... . 29 SDS-PAGE . ............ .. ... . 30
Two-Dimensional Gel Electrophoresis ...... 31 Immunoblots with Antisera . .......... 32
Purification of A. marginale Genomic DNA ... 33 MSP3 Clones . . . . . . . . 33
Verification of Recombinant MSP3 .. .... . 34 Digoxigenin Labeling of pBluescript MSP3-12 . 37 Representation of pBluescript MSP3-12 ..... 38
Restriction Enzyme Digestion .... ... 38 Southern Blots ........... . .. . 39
Presence of Multiple MSP3 Gene Copies . .. . 42
Distribution of MSP3 Copies in the A. marginale
Chromosome . . . ..... . ... 43


v









Restriction Enzyme Digestion in Agarose
Plugs . . . . . . . . 43
CHEF Gel Electrophoresis ......... 44

CHAPTER 4 RESULTS ... ............... 46
Specificity Experiments ............ 46
Conservation of MSP3 Between Different
Geographic Isolates of A. marginale .... 53
Immune Response to MSP3 ... ......... 56
MSP3 Nucleotide Sequence and Translation ... 62
MAb and Immune Sera Reactivity to Recombinant
MSP3 . . . . . . . . . 74
Representation of pBluescript MSP3-12 in the
A. marginale Genome ............ 79
Presence of Multiple MSP3 Gene Copies .... .. 79
Distribution of MSP3 Copies in the A. marginale
Chromosome ... ... . ..... . 84

CHAPTER 5 DISCUSSION . . . . . ..... 92
Rational for Studying the MSP3 Antigen . . . 92
The Specificity of MSP3 in Detecting A. marginale
Infection . . . . . . . . . 94
Size Polymorphism of MSP3 Among Various
Strains of A. marginale ... ....... 98
Immune Response to MSP3 ........... 100
Identification of pBluescript MSP3-12 as a
Recombinant Form of an 86 kDa Antigen . 103
pBluescript MSP3 is an Accurate Representation of
a Genomic Copy of MSP3 ......... 104
Multiple MSP3 Copies in the A. marginale Genome 105
Distribution of MSP3 Copies in the A. marginale
Genome ........... ...... 110
Summary and Conclusions ........ ... 111

REFERENCE LIST ... ... . ........... 115

BIOGRAPHICAL SKETCH .... . . . . . . 123

















vi















LIST OF FIGURES


Figure page
1 Diagram of MSP3 clones ... . . . ... 35
2 Representation of clone MSP3-12 in the A. marginale
genome ................... .. 40
3 Specificity experiments using immunoblots . .. 47 4 Specificity experiments using immunoblots . . 49 5 Specificity experiments using immunoblots . .. 51 6 Size polymorphism of MSP3 ........... 54
7 2-D gel electrophoresis of A. marginale proteins 58 8 2-D gel electrophoresis of A. marginale proteins 60 9 Gene sequence of MSP3-11 ............ 63
10 Gene sequence of MSP3-12 .......... . 65
11 Gene sequence of MSP3-19 ............ 67
12 Diagram of MSP3 clones with MSP2 homology . . 70 13 Comparison of MSP3 and MSP2 protein sequences . 72 14 Immunoblots of expressed MSP3 clones ..... 75 15 Immunoblots of expressed MSP3-12 clones ... 77 16 Representation of clone MSP3-12 in the A. marginale
genome . . . . . . . . . . . 80
17 Genomic representation of MSP3-12 ....... 82 18 Presence of multiple copies of MSP3 .. . . 85 19 Not I and Sfi I digestion of the A. marginale
genome . . . . . . . . .. .. 88
20 Distribution of MSP3 in the A. marginale genome 90




















vii















ABBREVIATIONS


bp base pair (s) BSA bovine serum albumin oC degrees Celsius CD card agglutination test CF complement fixation test CHEF clamped homogeneous electric field electrophoresis CT capillary tube agglutination test DNA deoxyribonucleic acid ECL enhanced chemiluminescence EDTA ethylenediaminetetraacetic acid ELISA enzyme-linked immunosorbent assay Fig. figure (s) FL Florida g gram (s) HCl hydrochloric acid HRP horseradish peroxidase IFA indirect fluorescent antibody IIF indirect immunofluorescence test IPTG isopropylthio-o-galactosidase kbp kilobase pair (s) kDa kilodalton



viii









M molar MAb monoclonal antibody (s) MAP1 major antigenic protein 1 mg milligram (s) min. minute (s) ml milliliter (s) mM millimolar MSPla major surface protein la MSP10 major surface protein 10 MSP2 major surface protein 2 MSP3 major surface protein 3 N normal NaCl sodium chloride NaOH sodium hydroxide ng nanogram (s) NP-40 nonidet P-40 PBS phosphate buffered saline PCV packed cell volume PI post-infection PMSF phenylmethylsufonly fluoride RFLP restriction fragment length polymorphism RIA radioimmunoassay rRNA ribosomal ribonucleic acid SDS-PAGE sodium dodecyl sulfate, polyacrylamide gel

electrophoresis sec. seconds


ix









SI South Idaho TBE tris-borate-EDTA buffer TE tris-EDTA buffer Tris tris hydroxymethyl aminoethane TWEEN 20 polyoxyethylene-sorbitan monolaurate TX Texas V volts VA Virginia w/v weight/volume WA Washington Ag microgram (s) 1 microliter (s)






























x















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

EVALUATION OF ANAPLASMA MARGINALE MAJOR SURFACE
PROTEIN 3 (MSP3) AS A DIAGNOSTIC TEST ANTIGEN By

Arthur Rick Alleman

August 1995


Chairperson: John Dame, PhD
Major Department: Veterinary Medicine

The immunodominant surface protein, MSP3, has been proposed as an antigen suitable for the diagnosis of bovine anaplasmosis. In this study we further characterized MSP3 to examine its potential as a test antigen for the serological detection of carrier cattle. The specificity of MSP3 was evaluated by probing immunoblots of A. marginale proteins with immune sera from animals infected with related organisms. Similarly, we used polyacrylamide gel electrophoresis (SDSPAGE) and immunoblots to evaluate the conservation of MSP3 between 4 different geographic isolates of A. marginale. In addition, proteins from a FL isolate were separated by 2dimensional gel electrophoresis, and immunoblotted with immune sera from cattle infected with one of 4 different geographic isolates of A. marginale. Genomic A. marginale DNA was digested with restriction endonucleases, transferred to nylon xi









membranes, and probed with a digoxigenin-labeled, cloned gene of MSP3 (courtesy of Dr. G. Palmer, W.S.U.).

Immunoblots demonstrated cross reactivity between MSP3 and sera from animals infected with A. ovis, E. risticii, and E. ewingii. Size polymorphism of MSP3 was seen between different geographic isolates of A. marginale. Twodimensional gel electrophoresis revealed at least 3 different antigens migrating at the 86kDa molecular mass, and sera from animals infected with different isolates reacted with different 86kDa antigens. Hybridization studies with a cloned MSP3 gene identified multiple copies of the gene in the genome. These results indicate MSP3 is, 1) cross reactive with A. ovis and some Ehrilichia sp., 2) not conserved between different isolates of A. marginale, and 3) in at least the FL isolate, MSP3 is actually a group of 3 or more 86kDa proteins with different isoelectric points. These data also suggest A. marginale may antigenically vary this immunodominant protein by use of a complex multigene family. The variability of MSP3 between isolates, the multiple 86kDa antigens in the FL isolate, and the multiple copies of the MSP3 gene indicate a single recombinant form of MSP3 may not be a suitable diagnostic test antigen. To be used as a diagnostic test antigen, conserved epitopes between copies of MSP3 genes may need to be identified and tested for reactivity with immune sera.




xii















CHAPTER 1
INTRODUCTION


Anaplasma marginale is an arthropod-borne, rickettsial hemoparasite which invades the red blood cells of cattle causing a clinical disease known as anaplasmosis. This organism is the representative species of the genus Anaplasma, in the family Anaplasmataceae, order Rickettsiales (Ristic and Krier, 1974). The genus name, Anaplasma, refers to the appearance of being devoid of cytoplasm, and the species name, marginale, refers to the location of the organism in the margin of infected erythrocytes. A. marginale was originally discovered in cattle experimentally infected with Babesia bigemina, and was thought to be a developmental stage of this organism (Smith and Kilborne 1893). However, several years later these marginal inclusions noted in the erythrocytes of cattle infected with B. bigemina were conclusively identified as agents belonging to another genus, Anaplasma (Theiler 1910).

A. marginale is now known to have a global distribution which includes the United States of America. World-wide economic losses are difficult to calculate, but losses in the U.S. alone are estimated to be over 100 million dollars annually (Goodger et al., 1979; McCallon, 1973). In certain


1









2

areas of the U.S., the incidence of infected cattle may be as high as 37% (Maas et al., 1986; McCallon, 1973).

Acute anaplasmosis is usually seen in cattle over 1 year of age and is characterized by a severe hemolytic anemia, resulting in weight loss, abortion, decreased milk production, and often death in infected animals over 3 years of age (Wanduragala & Ristic, 1993). In acute stages, the disease is easily diagnosed by finding organisms on routine blood smear evaluation. However, animals that survive the infection will remain carriers and maintain a low level of parasitemia which cannot be detected microscopically (Richey, 1981; Zaugg et al., 1986). These carrier cattle serve as a perpetual source of infection for susceptible cattle (Swift & Thomas, 1983). Cyclic rickettsemia has been detected and quantitated in carrier cattle using nucleic acid probe hybridization (Eriks et al., 1993; Kieser et al., 1990). These studies demonstrated rickettsemia levels in persistently infected cattle fluctuated at approximately 5 week intervals from a low of 104 to a high of 107 infected erythrocytes per ml of blood. Although the level of parasitemia was too low to be detected microscopically, uninfected Dermacentor andersoni ticks were able to acquire infection from the cattle at infectivity rates of up to 80% during the higher rickettsemia levels (107 infected erythrocytes per ml of blood) (Eriks et al., 1993). Even at extremely low levels of parasitemia 27% of the male ticks became infected. In addition, once ticks acquire the









3

infection, replication of the organism in the vector allowed easy transmission of the disease with only a few infected ticks regardless of the initial infecting dose. This firmly establishes the important role persistently infected carrier cattle play in the transmission of the disease. In order to reduce economic losses associated with anaplasmosis, control efforts must include an effective way of identifying and decreasing transmission from carrier cattle.

An inexpensive, sensitive, and specific field test for the identification of A. marginale infected carriers would have a tremendous impact on limiting the spread and consequently the economic losses associated with this disease. Entire herds could be easily tested and identified carriers removed or treated with oxytetracycline, thus eliminating the source of infection for susceptible cattle. A test such as this would also provide an accurate means of identifying carrier animals being shipped into nonendemic or noninfected areas. Ideally, animals infected with the less virulent species, Anaplasma centrale, should not contain antibodies that will cross react with the antigen used in this test. This would be a marked improvement over other serological tests which cannot distinguish between the two infections. In addition, in areas where vaccination with A. centrale is used as a means of prevention of A. marginale, this test may distinguish vaccinated from infected animals.









4

The purpose of this study was to evaluate the MSP3 protein of A. marginale, and determine if a recombinant form of this protein would be a suitable diagnostic test antigen to detect A. marginale infection in carrier cattle.















CHAPTER 2
LITERATURE REVIEW


General Methods of Serologic Diagnosis


Serology is the science of detection of specific antibodies in body fluids, particularly, though not exclusively, serum. There are 3 broad categories of serologic techniques, the primary binding tests, secondary binding tests, and tertiary binding tests. Primary binding tests allow antigen and antibody to combine, and the resulting immune complexes are measured using radioisotopes, fluorescent dye, or enzyme labels. Examples of the primary binding tests are radioimmunoassays, immunofluorescence assays, and enzymelinked immunosorbent assays. Primary binding tests are the most sensitive of the serologic techniques in terms of ability to detect smaller amounts of specific antibodies (Tizard, 1992).

Radioimmunoassays are widely used primarily because of their extreme sensitivity and ability to detect small amounts of antigen or antibody. In these assays radioactive isotopes such as 125I are used to label antigens or antibodies, and the level of radioactivity is used for quantitation. Disadvantages of this test are the dangers and restrictions



5









6

regarding the use of radioactivity, the complexity of the test procedure, and the need for specialized equipment. Because of their extreme sensitivity, these tests are frequently used to measure trace amounts of drugs in urine or other body fluids (Tizard, 1992).

Immunofluorescence assays employ the use of fluorescent dyes such as fluorescein isothiocyanate (FITC) to measure the formation of immune complexes. FITC is easily conjugated to immunoglobulins for detection using dark field microscopy with an ultraviolet light source, or flow cytometry. Specific immunoglobulins may be labeled directly and allowed to react with an antigen as in direct fluorescent antibody tests, or species specific antiglobulins may be labeled with FITC and used to bind to antibody/antigen complexes in indirect fluorescent assays. The indirect tests are usually more sensitive since each antibody molecule bound to the antigen may bind several labeled anti-globulin molecules (Tizard, 1992).

Enzyme-linked immunosorbent assays (ELISA) employ the use of enzymes for the detection of antigen/antibody interactions. The three most commonly used enzymes are alkaline phosphatase, horseradish peroxidase, and 0-galactosidase. In these techniques the enzymes are chemically linked to immunoglobulins or anti-globulins. Detection is accomplished by the addition of an enzyme substrate to the reaction,









7

producing a colored product. The color change may be estimated visually or determined spectrophotometrically.



There are several variations of the ELISA test; the direct ELISA which utilizes enzyme-linked immunoglobulins, the indirect ELISA which uses enzyme-linked anti-globulins, and the competitive ELISA which employs the use of a labeled monoclonal antibody to compete with specific antibodies in the test sample for a single epitope on an antigen molecule. Some ELISAs bind antibody to a solid phase media to capture a particular antigen (antigen capture ELISA), while other techniques bind antigen to the solid phase in order to capture and detect antibody (antibody capture ELISA) (Harlow & Lane, 1988). Various solid phase media may be used such as nitrocellulose membranes or polystyrene microtiter plates, depending on the purpose of the test and the nature of the material to be tested. The extreme versatility of this technique, its excellent sensitivity (comparable to RIA), and its simplicity make it one of the most widely used immunodiagnostic techniques for many viral, bacterial, and parasitic infections. In addition, it does not involve the use of hazardous radioactive isotopes. Because of its extreme sensitivity, specificity may be a problem with these tests, particularly if unpurified test antigens are used to detect specific antibodies in polyclonal sera.









8

Secondary binding tests measure the interaction of antigen and antibody by in vitro visualization of a secondary event which occurs as a result of immune complex formation. Such events include the precipitation of soluble antigens in solution by immune complex formation, agglutination of particulate antigen on the surface of bacteria or erythrocytes, and the activation of the complement pathway, resulting in cell lysis. Many of these tests require optimal concentration of antigen and antibody to allow visualization, and false results may occur in cases of antibody or antigen excess (Nakamura et al., 1988). As a result of this, and the gross visual detection required for many of these tests, secondary binding tests lack the sensitivity of primary binding assays. Examples of these tests include immunoprecipitation, immunodiffusion, agglutination, and complement fixation.

Tertiary binding tests actually measure the in vivo protective effects specific antibodies may have in an animal. These test measure the biological activity of the antibody to determine their ability to protect a susceptible animal from an infectious agent or neutralize the effects of an antigen. These tests are most useful in experimental trials or in treatment of certain diseases by passive transfer of antibody. They are not practical or suitable for the diagnosis of disease processes.









9

The suitability of a particular type of immunodiagnostic test depends on many factors such as the use of test (field versus laboratory use), the sensitivity and specificity required, the prevalence of the disease, the number of samples to be tested at any one time, and the speed at which a test must be performed. In general, selection of a diagnostic test often involves a compromise trade off between sensitivity, specificity, and ease of performance. As previously mentioned, the indirect ELISA is a very sensitive test, easy to perform, and may be adapted to field use where multiple samples may be tested simultaneously. It is a primary binding assay and requires no specialized equipment to perform the test or interpret the results. This may prove to be an ideal test for the field detection of A. marginale infected carrier cattle. Because of the extreme sensitivity of the indirect ELISA, specificity can be a problem when detecting specific antibodies in a polyclonal sera. To achieve acceptable specificity, great care must be taken in selection and purification of an appropriate test antigen.


Methods for Detection of A. marginale Infected Cattle


Currently an effective vaccine against A. marginale is not available, and the inability of present serologic tests to accurately detect persistently infected, carrier cattle severely compromises efforts to establish disease free herds and reduce economic losses through testing, isolation, and









10

treatment (Luther et al., 1980; Richey, 1981). Several types of serological tests have been described for the diagnosis of A. marginale, including complement fixation (CF), capillary tube agglutination (CT), card agglutination (CD), indirect immunofluorescence (IIF), and ELISA (Amerault & Roby, 1968; Barry et al., 1986; Duzgun et al., 1988; Gonzalez et al., 1978; Kuttler, 1981). Error rate with these tests is high, primarily because antigens used in these tests are a crude mixture of A. marginale and erythrocyte material (Kocan et al., 1978; Kocan et al., 1978). False positives as high as 20% are seen due to poor specificity of test antigen (Amerault & Roby, 1968; Amerault et al., 1973; Barry et al., 1986; Duzgun et al., 1988; Gonzalez et al., 1978; Luther et al., 1980; Todorovic et al., 1977). Poor sensitivity of these tests results in false negatives as high as 21% (Amerault et al., 1973; Barry et al., 1986; Goff et al., 1990; Gonzalez et al., 1978; Luther et al., 1980; Maas et al., 1986). Current technology, which was not available when previous tests were developed, allows us to economically produce a sensitive and specific ELISA test for the rapid field diagnosis of A. marginale using purified, recombinant, A. marginale protein.

As previously mentioned, several types of tests have been developed for the diagnosis of anaplasmosis. The tests which have received most attention have been the rapid card agglutination test (CA), the complement fixation test (CF),









11

indirect immunofluorescence test (IIF), the radioimmunoassay (RIA), and various enzyme-linked immunosorbent assays (ELISA).


The Rapid Card Agglutination Test (CA)


The CA is one of the most widely used in the field because it is easy to run, requires minimal equipment, and uses unheated sera or heparinized plasma (Amerault & Roby, 1968). Being a secondary binding test, this assay lacks the sensitivity of primary binding assays. Initial studies using the CA test reported 100% sensitivity in detecting carrier cattle and 86% sensitivity in detecting all known infected cattle (Amerault & Roby, 1968). However, the number of positive cattle tested was small (22), and subsequent studies comparing sensitivity and specificity of the CA test with other serological tests indicates sensitivity to be 84% (Gonzalez et al., 1978). Yet another study indicated the CA test was able to detect only 6 of 9 (66.6%) known carrier cattle (Luther et al., 1980). In addition, 89.3% of noninfected, A. marginale vaccinated cattle had positive reactions with the CA test for up to 15 months after vaccination (Luther et al., 1980).

Specificity of the CA test was initially reported to be 100% with none of the 24 normal cattle sera testing positive (Amerault & Roby, 1968). This test uses a crude preparation of A. marginale and erythrocyte antigen, and problems of specificity are likely to be encountered. Nonspecific









12

agglutination did occur when fresh serum was used; however, this problem did not occur when using heparinized plasma. Additional studies involving large numbers of serum samples (380) established 98% specificity for the CA test (Gonzalez et al., 1978). However, none of these studies investigated the potential for cross reactivity with sera from cattle infected with other hemoparasites. Current technology, which was not available when previous tests were developed, allows the phylogenetic relatedness of various species and genera to be determined. Using 16s rRNA sequence analysis investigators have determined A. marginale to be closely related to other Anaplasma sp., Ehrlichia sp., and Cowdria ruminantium (Dame et al., 1992; Van Vliet et al., 1992). To accurately determine the specificity of a test antigen, it is essential to test the ability of the antigen to distinguish between infections of related organisms. This was not done for any of the antigens currently used in serologic diagnosis of A. marginale. Therefore estimations of specificity for these tests are unreliable.

Although the CA test is well adapted for field use, its poor sensitivity (probably because it is a secondary binding test) and its questionable specificity (because of its use of a crude test antigen preparation) make it an undesirable test for the rapid field identification of carrier cattle.









13

The Complement Fixation Test (CF)


The CF is another widely used test for the detection of antibodies to A. marginale in infected cattle. This test relies on the ability of A. marginale antibodies in the serum to fix complement and lyse target cells on which the antigen is associated (Kuttler & Winward, 1984). The antigen used in this test is again a crude mixture of A. marginale and erythrocyte proteins, and since it is a secondary binding test, sensitivity is not as high as with primary binding assays. Sensitivity of this assay is reported to be low, varying from 10% to 79% (Gonzalez et al., 1978; Goff et al., 1990). In addition, it has been shown that CF antibodies decline rapidly after natural infections with A. marginale (Todorovic et al., 1977). CF antibodies decreased to very low levels (1:10) as soon as 10 weeks post-infection (PI), and below the sensitivity level of the test (1:5) by 14 weeks PI (Gonzalez et al., 1978). This may be because CF relies on the presence of IgM antibodies, which are better able to fix complement than IgG. IgM levels are elevated in acute infections, but may decline after acute episodes subside. These results would indicate this test is unreliable in detecting carrier cattle. In another study, 65 cattle from a known A. marginale infected herd were tested by CF and DNA probe hybridization (Goff et al., 1990). Stained blood smears prepared from these cattle detected no organisms. However, 64 of the 65 cattle blood samples tested positive by DNA probe









14

hybridization and 60 of the 65 sera tested positive by IIF, while only 5 of the 65 tested positive by CF.

The specificity of the CF test in detecting non-infected cattle is very high, up to 100% (Gonzalez et al., 1978). This is not surprising given the low test sensitivity. However, the study again looked at specificity involving the detection of CF antibodies in normal, non-infected cattle, not cattle which may be infected with other closely related hemoparasites. Therefore, cross reactivity with cattle infected with other rickettsial or hematoprotozoal agents may occur. In addition, as with the CA test, the CF test could not distinguish between infected animals and animals vaccinated with the killed A. marginale vaccine (Luther et al., 1980).

The CF test is a tedious test to perform and requires considerable technical skill and knowledge. Therefore, the complexity of the test and its poor sensitivity, particularly in detecting infected carriers makes it a poor choice for the field identification of infected animals.


The Indirect Immunofluorescence Test (IIF)


The indirect immunofluorescence test (IIF), being a primary binding assay, is much more sensitive than the CA or CF tests. This test uses A. marginale infected erythrocytes fixed to a microscope slide and permits reaction to them using a patient's serum. Antibodies to A. marginale on infected









15

erythrocytes are then visualized using a fluoresceinconjugated anti-bovine, rabbit immunoglobulin (Gonzalez et al., 1978). Sensitivity in detecting subclinically infected animals is reported to be 97% with lower limits of sensitivity not being reached by 18 weeks PI. In another study involving 64 cattle naturally infected with A. marginale and confirmed by DNA probe hybridization, 94% of cattle were positively identified by IIF (Goff et al., 1990). In this study circulating organisms were not seen in stained blood smears, but length of infection was undetermined.

Although sensitivity with IIF is much improved over CA and CF tests, the specificity is somewhat reduced with 10% of normal, non-infected cattle testing positive by IIF (Gonzalez et al., 1978). In addition, cross reactivity in cattle infected with related organisms may also present a problem. The sensitivity of this test is satisfactory, however, its questionable specificity, labor intensive nature, and need for specialized reagents and equipment make it undesirable as a field test for routine diagnosis and identification of A. marginale infected carriers.


The Radioimmunoassay (RIA)


Recently a radioimmunoassay (RIA) was developed for the detection of A. marginale antibodies in sera of infected cattle (Schuntner & Leatch, 1988). This test initially demonstrated high specificity and sensitivity (98.8% for each)









16

when testing a large number of A. marginale infected cattle, normal cattle, and cattle infected with B. bigemina, B. bovis, and Theileria orientalis. This test utilizes a crude A. marginale antigen preparation isolated from infected erythrocytes. Reactants are identified using '2SI-labeled, anti-bovine IgG, rabbit immunoglobulin and an automated gamma counter. As expected, sensitivity with this primary binding assay is high; however, substantial numbers of false positives (up to 37%) occurred unless sera was pre-absorbed with normal bovine erythrocytes and sonicated B. bovis antigen. Controls for cross reactivity using pre-absorption should not be necessary if purified antigens are used, even with tests as sensitive as RIAs.

Even though the RIA is highly sensitive and specific, it is too labor intensive to be used as a practical field test where large numbers of samples need to be processed. In addition, the use of radioactive material and the need for specialized equipment limits its use to reference or research laboratories.


Enzyme-Linked Immunosorbent Assays (ELISA)


Several ELISAs for measuring antibody to A. marginale have been developed over recent years (Barry et al., 1986; Duzgun et al., 1988; Montenegro-James et al., 1990; Nakamura et al., 1988; Trueblood et al., 1991; Winkler et al., 1987). Because these are primary binding assays, the sensitivity of









17

most of these tests are good. However, none of these tests use a single, purified A. marginale antigen, all use a crude mixture of A. marginale antigens prepared from initial bodies. Therefore, test specificity has been a problem, as is frequently encountered using a mixture of proteins as a test antigen in a highly sensitive technique.

Barry et al., 1986 developed one of the early ELISAs for diagnosis of A. marginale infection in cattle. Using a crude mixture of A. marginale initial bodies and cell membranes from ghost RBC's as a test antigen, the test reportedly was able to accurately distinguish cattle free from infection from recently infected animals (up to 8 months duration). However, no measures were taken to insure the positive or negative status of most of the animals tested. Also, test accuracy for identifying carrier cattle was questionable having identified only 2 of 3 cattle infected > 3 years. In addition, the test appeared to be fairly insensitive, necessitating the testing of sera at a dilution of 1:100. Using concentrated serum for testing may cause a problem with test specificity and 14.6% of cattle inoculated twice with B. bovis-infected erythrocytes were positive for A. marginale using this ELISA. Crossreactions were attributed to antibodies produced against RBC antigens in the inoculum. Experiments investigating cross reactivity with other closely related rickettsial or hematoprotozoal agents were not performed.









18

Another diagnostic test utilized the CF test antigen in particulate and SDS-solubilized form as a test antigen in an ELISA (Winkler et al., 1987). This study indicated SDSsolubilization of the proteins decreased background reactivity in the test and increased test sensitivity in detecting positive reactants. Complete correlation between the ELISA and CF test was found when using solubilized test antigen and CF test positive and negative reference sera. In testing other infected cattle, false negatives were not observed; however, cross reactivity ranged from 50% to 80% when testing sera from cattle infected with or immunized against different infectious agents.

Nakamura et al., 1988, developed an ELISA test antigen by nitrogen decompression of infected cells to isolate initial bodies, and solubilized them in triton X-100. Test specificity and sensitivity was determined by comparison of results with the CF test. Specificity of this ELISA was satisfactory having 100% agreement with sera tested negative by CF and no cross reactivity with serum from animals infected with Babesia sp., Theileria sp., or Eperythrozoon sp. Cross reactivity was detected with A. centrale. Cross reactivity with other rickettsial agents such as Ehrlichia sp. or Cowdria ruminantium was not determined. Sensitivity of this test was not adequately determined since the results were compared to a test (CF test) which itself has poor sensitivity, particularly in detecting carrier cattle. No effort was made









19

to test the sensitivity of this ELISA in detecting infected carriers shown to be positive by other proven sensitive tests such as IIF or calf inoculations.

Another ELISA used a 2 antigen technique to enhance test specificity; a negative antigen prepared from a cow prior to infection, and a positive antigen derived from A. marginale infected cells (Duzgun et al., 1988). Reactants were identified using net absorbance values obtained by subtracting the absorbance value of sera with negative antigen from the absorbance value of sera with positive antigen. Specificity of this test was good with only 3% of negative sera, 2% of sera from animals infected with B. bovis, and 4% of sera from animals infected with B. bigemina giving positive results. Other related rickettsial agents were not tested. Sensitivity also appeared to be good with no false negatives noted in 100 animals confirmed negative by IIF or calf inoculations. A small number of infected cattle had positive ELISAs for up to 3 years later. This test provided the best sensitivity and specificity thus far, and was the first ELISA that demonstrated the ability to detect long term carrier cattle. However, the crude nature of the test antigen necessitates the use of the 2 antigen system which is more cumbersome, complex, and time consuming. If this 2 antigen system had not been used, 37% of the negative sera would have given false positive results. In addition, the 2 antigen system requires the use of a spectrophotometer to identify reactants. This does not









20

allow for eventual visual identification as would be needed for rapid field diagnosis.

Direct visualization of positive reactants was accomplished using a Dot-ELISA (Montenegro-James et al., 1990). In this test whole initial body preparations were solubilized in SDS and dotted on to nitrocellulose disks. Test sera were reacted with the antigen and antigen/antibody complexes were visualized with alkaline phosphatase-conjugated protein A. Test specificity was good (95%) and cross reactivity to Babesia or Trypanosome vivax was not observed. However, other rickettsial agents or related hemoparasites were not tested. Test specificity was increased by using protein A-conjugated alkaline phosphatase versus an antiglobulin-conjugated molecule. This reduced nonspecific binding of antibodies to the nitrocellulose disk. However, the use of whole initial bodies isolated from infected erythrocytes still lends itself to false positive reactions.

Test sensitivity with the Dot-ELISA was fair (92.9%) with 19 false negatives out of 269 true positives (Montenegro-James et al., 1990). In addition, no effort was made to determine the ability of this test to detect chronically infected, long term carrier cattle. A good diagnostic test for identification of infected cattle for importation or herd management should have a sensitivity of at least 95% or higher. Therefore, while this test is easy and convenient to









21

perform, its lack of sensitivity and questionable specificity are certainly areas of needed improvement.

Recently, an antigen capture ELISA was developed for the detection of A. marginale infection (Trueblood et al., 1991). This ELISA utilizes 2 monoclonal antibodies (MAb) which recognize 2 different epitopes on the A. marginale surface protein MSP-la. In this test one MAb is bound to a polystyrene microtiter plate, antigen (ie. infected whole blood) is incubated with this monoclonal, and a second monoclonal conjugated to horseradish peroxidase is added to the reaction mixture for visualization of any captured antigen. This assay was sensitive enough to detect infected animals with parasitemias of <1.0%, however, it is not sensitive enough to detect carrier cattle which have parasitemias as low as 104 infected cells per ml of blood (0.0001%) (Eriks et al., 1993). Thus, although the detection of A. marginale antigen in the blood of infected cattle would be the most specific way to determine infection, the sensitivity of most primary binding assays will not likely be sufficient to detect the low level of parasitemia encountered in carrier cattle.


Nucleic Acid Probe Hybridization


One test which has shown excellent results in the ability to detect A. marginale organisms in infected cattle, even at a level of sensitivity sufficient to identify chronically









22

infected carriers, is a nucleic acid probe derived from a fragment within the gene coding for an A. marginale surface protein (Goff et al., 1988; Eriks et al., 1989). This probe can detect infected animals with parasitemias as low as 0.000025% (Eriks et al., 1989). Sensitivity and specificity of the test surpasses all previously developed serological tests, including the IIF (Goff et al., 1990). Although this test is too complex to be useful for routine field diagnosis, it has tremendous potential for identification of true carriers which in the past was done by subinocculation of splenectomized calves. This could be useful for the identification of cattle for reference sera, establishing the effects treatment with tetracycline has on carrier status, or identification of infected ticks. Considering the alternative of calf inoculation, a nucleic acid probe is a much more convenient and economical tool when accurate identification of infection is essential.


Previous Experiments


The ideal test antigen to detect A. marginale infected carriers must be antigenic enough to have antibodies present in sera of cattle during all stages of infection. It must be species specific so as not to cross react with sera from animals infected with related organisms, and it must be conserved between isolates of A. marginale, enough to be recognized by sera from cattle infected with various









23

geographic isolates. An antigen which shows promise for meeting the above requirements is the 86kDa surface protein of A. marginale, MSP3.

In previous experiments, the most antigenic surface proteins of A. marginale were identified by immunoprecipitation of radiolabeled initial body proteins from a Florida (FL) isolate of A. marginale with immune sera from infected cattle (Palmer et al., 1986). The FL isolate of A. marginale was used for antigen isolation because it has been found by adsorption studies to contain antigens common to both morphologic types of A. marginale, the tailed and non-tailed forms (Goff and Winward, 1985; Kreier and Ristic, 1963). An 86kDa protein, MSP3, was identified as the most immunodominant in all stages of infection from as early as 30 days postinfection (PI) to 255 days PI. Similar reactivity was observed when initial body preparations were immunoprecipitated with immune sera from cattle infected with either of 3 different isolates of A. marginale, FL, Virginia

(VA), and Texas (TX) isolates (Palmer et al., 1986). This suggested MSP3 from a FL strain was conserved enough to be recognized by immune sera from animals infected with different isolates. However, the conservation of MSP3 was not firmly established since these studies were performed using single dimensional gel electrophoresis.

These experiments identified for the first time antigenic surface proteins of A. marginale which could be investigated









24

for potential use as a diagnostic test antigen or vaccine candidate. Use of a single antigen in a diagnostic test could markedly increase test specificity over currently available tests. Since MSP3 was the most immunodominant protein, further experiments were done to determine its potential as a diagnostic test antigen. Monoclonal antibodies to MSP3 were produced and used in a sepharose bead affinity column to isolate purified MSP3 from FL isolate initial bodies (McGuire et al., 1991). Affinity purified MSP3 was injected into rabbits and rabbit-anti-MSP3 immune serum was used on dot blots to identify epitopes of MSP3 in at least 8 different geographic isolates of A. marginale (Mcguire et al., 1991). Using immune sera from infected cattle, purified MSP3 accurately identified long term carriers for up to 5 years PI. Immune sera from cattle infected with B. bovis, B. bigemina, or an unidentified rickettsial agent did not cross react with purified MSP3 (McGuire et al., 1991). However, sera from animals infected with organism now known to be phylogenetically related to A. marginale were not tested in these experiments.

Since MSP3 fulfilled many of the criteria for a good diagnostic test antigen, attempts were made to clone the MSP3 gene and produce a recombinant protein which might be used in an ELISA test. A genomic library made from a FL isolate of A. marginale DNA was screened using a pool of anti-MSP3 MAbs. Three clones of the MSP3 gene were identified and sequenced.









25

However, proteins expressed by these clones were inconsistent in their reactivity with immune cattle sera. In addition, in depletion experiments, anti-MSP3 MAbs were reacted with FL A. marginale initial bodies. The initial bodies still contained an 86 kDa antigen when reacted with immune cattle sera.

The MSP3 antigen of A. marginale appears to be a strong candidate as a diagnostic test antigen to detect infected carrier cattle. However, these previous experiments propose questions which must be answered to confirm this hypothesis. Questions must be addressed regarding the specificity of this protein in detecting A. marginale infection, the conservation of the MSP3 protein between various geographic isolates of A. marginale, and explanations for inconsistent reactivity of the 3 MSP3 clones. The purpose of our investigation is to further characterize the MSP3 protein and determine if it is a suitable candidate for a diagnostic test antigen to detect A. marginale infected, carrier cattle.















CHAPTER 3
MATERIALS AND METHODS


Anaplasma marginale Strains


Four isolates of A. marginale were used in this study, FL, VA, South Idaho (SI), and Washington (WA). These isolates are designated by their original location of isolation (McGuire et al., 1984). Isolates were stored in liquid nitrogen as cryopreserved stabilates (Love, 1972) before being used to infect splenectomized calves. Thawed stabilate (20ml) from each isolate was injected intramuscularly into 6-monthold, male Holstein calves. Calves were monitored daily for percent parasitemia by blood smear evaluation, and packed cell volume (PCV). Infected, whole blood was collected in EDTA from calves during periods of peak parasitemias (FL=70%, VA=36%, SI=42%, and WA=40%), centrifuged at 10,000 x g for 15 min., and the serum and buffy coat was removed. Packed erythrocytes were washed 3 times in phosphate buffered saline (PBS) (0.14M NaCl, 2.68 mM KC1, 8.26 mM K2HPO4, pH 7.4), resuspended to a PCV of 50%, and stored at -700C.

A. marginale strains used in clamped homogeneous electric field electrophoresis (CHEF) studies were prepared as previously described (Alleman et al., 1993). Briefly, whole



26









27

blood from cattle infected with the previously mentioned isolates was collected with sodium heparin used as an anticoagulant and washed 3 times in PBS. Erythrocytes were separated from bovine leukocytes by passing washed blood over an a-cellulose/microcrystalline cellulose column (Sigmacell type 50, Sigma Chemical Co., St. Louis, MO) as described by Beutler (Beutler, 1984), except that 1% bovine serum albumin (BSA) was included in the wash buffer. Washed cells were resuspended to a concentration of 5.0 x 106 cells/il, or approximately 35% packed cell volume (PCV). Blood films were prepared from isolated erythrocytes for microscopic detection of bovine leukocyte contamination and level of parasitemia.

Intact erythrocytes were embedded in 0.7% agarose by mixing 1 part erythrocyte suspension with 2 parts of 1% FMC InCert Agarose (FMC Bioproducts, Rockland, ME) in PBS (0.14M NaCl, 2.68 mM KC1, 8.26 mM K2HPO4, pH 7.4), 0.125 M EDTA. Mixtures of blood and agarose were kept at 370C while pipetting into molds. Molds containing plugs were placed on ice to set. Plugs were incubated in 0.5 M EDTA (pH 9.5), 1% N-lauroylsarcosine, and 2 mg/ml Proteinase K for 48 hours at 370C, then stored at 4C in fresh Proteinase K solution.


Initial Body Preparation


A. marginale initial bodies were isolated from infected erythrocytes as previously described (Palmer and McGuire, 1984). Briefly, frozen blood was quick-thawed at 370C and









28

washed 5 times in PBS with each centrifugation at 16,000 x g for 25 min. at 40C. After each centrifugation, an upper layer containing both leukocytes and erythrocytes was removed. The pellets were resuspended in PBS, sonicated for 2 min. on ice at 50 W and centrifuged as before. Pelleted material was again resuspended in PBS, and sonicated for 30 sec. on ice at 50 W and centrifuged a final time. Intact initial bodies were visualized by Wright-Giemsa stain. The pellets of initial bodies were resuspended in equal volumes of PBS for use in SDS-PAGE.

Initial bodies used in 2-D gel electrophoresis were resuspended in equal volumes of lysis buffer containing 9.5 M urea, 2% nonidet P-40, 1.6% Ampholyte 5/7 (Bio-Lyte 5/7, BioRad Laboratories, Richmond, CA), 0.4% ampholyte 3/10 (Bio-Lyte 3/7, Bio-Rad Laboratories, Richmond, CA), and 5.0% 0mercaptoethanol.

Protein concentrations were determined spectrophotometrically using the Micro BCA Protein Assay (Pierce, Rockford, Illinois). Initial body preparations were stored in small aliquots at -700C. Anaplasma centrale initial bodies were also prepared as described above.


Babesia bovis and Babesia bigemina Antigen Preparation


Babesia bovis and B. bigemina antigens were prepared from organisms maintained in microaerophilic stationary phase culture as previously described (Levy and Ristic, 1980).









29

Briefly, infected erythrocytes were centrifuged at 10,000 x g and the supernate was removed. Packed cells were resuspended to 20 times the volume in 10 mM sodium phosphate. The solution was centrifuged and the supernate was removed. The pellets were then resuspended in an equal volume of 10 mM sodium phosphate.


Antisera used for Immunoblots


Six Holstein calves were infected with blood stabilate containing a FL isolate of A. marginale. Two other Holstein calves were each infected with either a VA or a SI isolate. Antisera from all calves were collected at 50 and 70 days PI. Antiserum from a cow experimentally infected with a WA isolate, rabbit-anti-MSP3 polyclonal sera (RB-955), an antiMSP3 monoclonal antibody (MAb) (AMG75C2), and an anti-MSP2 MAb (ANAFl9E2) were supplied to us courtesy of Travis McGuire and Guy Palmer (Washington State University, Pullman). The reactivities of the MAbs (McGuire et al., 1984; McGuire et al., 1991; Palmer et al., 1988; Palmer et al., 1994) and the rabbit-anti-MSP3 polyclonal serum (McGuire et al., 1991) have been previously described. Sera obtained from calves prior to infection with A. marginale, and a MAb specific for the variable surface glycoprotein of Trypanosoma brucei (TRYPlEl), were used as negative controls in immunoblot experiments.

A. centrale and A. ovis antisera with indirect fluorescent antibody (IFA) titers of 1:4,000 were supplied









30

courtesy of Susan Oberle (The Salk Institute, San Diego, CA). Hyperimmune sera from cattle infected experimentally with B. bovis and B. bigemina were supplied by David Allred (University of Florida, Gainesville). Sera from cattle experimentally infected with Cowdria ruminantium were supplied by Michael Bowie (University of Florida, Gainesville). Equine sera from animals infected with Ehrlichia equi and Ehrlichia risticii had IFA titers of 1/1,600, and were obtained from Ibulaimu Kakoma (University of Illinois, Urbana). Serum from a dog with canine ehrlichiosis was obtained from Rose Raskin (University of Florida, Gainesville). This dog was infected with E. ewingii, but serum from this animal had an IFA titer of 1/160 for Ehrlichia canis, and 1/64 for E. chaffeensis.


SDS-PAGE


Initial body preparations containing 6.0 10.0 gg of protein were solubilized in one half their volume of a 3x sample buffer containing 0.1 M Tris pH 6.8, 5% SDS (w/v), 50% glycerol, 7.5% 0-mercaptoethanol, and 0.00125% bromophenol blue, and heat denatured at 1000C for 3 min. Proteins were electrophoresed on 7.5% to 17.5% (w/v) gradient polyacrylamide gels. Gels were fixed in 25 mM Tris, 191.8 mM Glycine and 20% Methanol, and electrophoretically transferred to nitrocellulose (Hybond ECL, Amersham International plc, Buckinghamshire, England).









31

Two-Dimensional Gel Electrophoresis


Isoelectric focusing gels were prepared per manufacturer's instructions (Protean II Slab Cell Instruction manual, Bio-Rad Laboratories, Richmond, CA). Briefly, an acrylamide\N,N'-methylene-bis-acrylamide solution containing 9.5 M urea, 2.0% nonidet P-40(v/v), 4.1% acrylamide/bis (30.8% T/2.6% C), 10 mM (3-[(3-cholamidopropyl)dimethylammonio]ipropanesulfonate, CHAPS), and 5.8% Bio-Lyte 5/7 (v/v) was polymerized in glass tubing 3 mm x 140 mm. Initial body preparations were incubated for 2 hours at room temperature in 4 times their volume of the previously described lysis buffer and 5 times their volume of sample buffer containing 9.5 M urea, 2.0% Triton X-100, 5% 0-mercaptoethanol, 1.6% Bio-Lyte 5/7, and 0.4% Bio-Lyte 3/10. The solution was then centrifuged at 100,000 x g for 2 hrs. at 250C. A volume of the supernatant containing 20gg of protein was loaded onto each tube gel and overlayed with 50il of overlay buffer containing 9.5 M urea, 0.8% Bio-Lyte 5/7, 0.2% Bio-Lyte 3/10, and 0.0025% bromophenol blue. Tube gels were electrophoresed at 400 volts for 16 hrs. and 800 volts for 2 hrs. using the BRL model V16 vertical gel electrophoresis system (Bethesda Research Laboratories, Gaithersburg, MD). Tube gels were extruded from the glass tubes and equilibrated for 5 min. in buffer containing 0.0625 M Tris-HC1, pH 6.8, 10.0% glycerol, 2.0% SDS (w/v), 5.0% 0-mercaptoethanol, and 0.00125%









32

bromophenol blue. Focused proteins were then electrophoresed on 7.5% to 17.5% (w/v) gradient polyacrylamide gels and treated and transferred to nitrocellulose membranes as described above.


Immunoblots with Antisera


Nitrocellulose membranes containing transferred proteins were blocked with 5% milk (w/v) in PBS with 0.25% polyoxyethylene-sorbitan monolaurate (Tween 20) to inhibit non-specific binding of primary and secondary antibodies. The membranes were washed with 1% milk (w/v) in PBS with 0.25% Tween 20, and probed with antisera from animals infected with one of the following organisms; A. marginale, A. centrale, A. ovis, B. bovis, B. bigemina, C. ruminantium, E. equi, E. risticii, or E. ewingii. Normal sera from respective uninfected species were used as negative controls. Serum dilutions (in PBS with 1% milk and 0.25% Tween 20) of 1/100 or greater were used. Rabbit-anti-MSP3 polyclonal sera was used at a dilution of 1/5,000. Normal rabbit sera was used at the same dilution as a negative control. Anti-MSP3, anti-MSP2, and negative control anti-trypanosome MAbs were used in concentrations of 5 pg/ml. The membranes were again washed with 1% milk (w/v) in PBS with 0.25% Tween 20 and probed with either species-specific anti-IgG-Horseradish peroxidase (HRP) conjugated antibody at a dilution of 1/2,000 (Sigma Immuno Chemicals, St. Louis, MO), or HRP-conjugated Protein G at a









33

dilution of 1/15,000 (Sigma Immuno Chemicals, St. Louis, MO). Membranes were processed for enhanced chemiluminescence (ECL) with detection reagents containing luminol as a substrate (ECL Western Blotting detection reagents), (Amersham International, plc, Buckinghamshire, England). The membranes were exposed to Hyperfilm MP (Amersham International, plc, Buckinghamshire, England) to visualize bound antibody.


Purification of A. marginale Genomic DNA


Florida, South Idaho, and Virginia isolates of A. marginale genomic DNA used in hybridization studies were purified by phenol/chloroform extraction and ethanol precipitation as previously described (Barbet et al., 1987; Barbet and Allred, 1991). A. marginale genomic DNA from FL, SI and VA isolates used in CHEF studies was prepared in agarose plugs as previously described above.


MSP3 Clones


Three clones of the MSP3 gene, MSP3-11, MSP3-12, and MSP3-19 were cloned, sequenced, and supplied to us courtesy of T. McGuire, G. Palmer, and T. McElwain (Washington State University, Pullman). These clones were prepared from a genomic library composed of mechanically sheared A. marginale DNA, ligated with Eco RI adaptors, and inserted into pBluescript SK(-) plasmids (Stratagene, Lajolla, CA). Clones were identified by screening the library with anti-MSP3 MAbs









34

(AMG75C2, AMG76B1, AMG43/19, & AMG43/23), one of which, AMG75C2, has been previously described (McGuire et al., 1991). One of these genes, MSP3-12, contains the N-terminus and 633 bp upstream to the open reading frame (Fig. 1). The Cterminus is not present in this clone. The remaining 2 clones, MSP3-11 and MSP3-19, are missing the N-terminal sequence of the open reading frame (Fig. 1). They both contain the C-terminal end of the gene as well as 1,473 bp and 2,480 bp respectively, downstream to the open reading frame.


Verification of Recombinant MSP3


To verify a cloned MSP3 gene represented a gene which produced one or more of the 86 kDa antigens seen on 2-D immunoblots, E. coli cells (Epicurian Coli XL1-Blue) (Stratagene, Lajolla, CA) were transformed with pBluescript containing each of the 3 MSP3 genes. This was done according to manufacturer's instructions. E. coli cells were transformed with nonrecombinant pBluescript as a negative control. Transformed cells were plated on Luria agar containing 50 pg/ml of ampicillin. Transformants were selected by blue/white screening and selected colonies of each transformant were grown overnight in Luria broth containing 50 Ag/ml of ampicillin. Transformed E. coli cells were suspended in PBS, lysed in one half their volume of a 3x sample buffer containing 0.1 M Tris pH 6.8, 5% SDS (w/v), 50% glycerol, 7.5% 8-mercaptoethanol, and 0.00125% bromophenol blue, and heat


























Fig. 1 Diagram of MSP3 clones.
A schematic representation of clones MSP3-11, MSP3-12, and MSP3-19. Homologous regions are indicated by like shaded areas. Nucleotide numbers are indicated on the bottom. Top numbers of clones MSP3-11 and MSP3-19 indicate corresponding nucleotides in clone MSP3-12.





















MSP3 CLONES



MSP3-12





1 633 1191 1306 1472 te1818 1953 200 2090 2280 233







MSP3-11
06 1472 1818 200 2090 2280 S



1 172 435 623 1042 1235 1590 1790 3253







MSP3-19
1191 1306 1472 1818 1953 StI.p




1 176 278 952 1124 14s 170s ON8










01









37

denatured at 1000C for 3 min. Bacterial lysates or similarly prepared FL A. marginale lysates were separated by SDS-PAGE and transferred to nitrocellulose as previously described above. Membranes were reacted with anti-MSP3 MAb AMG75C2 or immune sera from animals infected with a FL or VA isolate of A. marginale as described above. Nonimmune cattle sera and an anti-T. brucei MAb were used as negative controls. Antigen antibody reactions were visualized by ECL with detection reagents containing luminol as a substrate (ECL Western Blotting detection reagents), (Amersham International, plc, Buckinghamshire, England). The membranes were exposed to Hyperfilm MP (Amersham International, plc, Buckinghamshire, England) to visualize bound antibody.


Diqoxigenin Labeling of pBluescript MSP3-12


Empty pBluescript DNA and pBluescript MSP3-12 DNA were grown in previously prepared transformants. Plasmid DNA was isolated from bacterial DNA by ion exchange chromatography using the QIAGEN Plasmid Midi Kit (Qiagen Inc., Chatsworth, CA) according to the manufacturer's instructions. The purified plasmid DNA was precipitated in ethanol, dried, and dissolved in TE (10 mM Tris, pH 7.5, 1 mM EDTA).

A digoxigenin-labeled probe of pBluescript MSP3-12 was prepared by digestion of 5 yg of pBluescript MSP3-12 with Eco RI according to manufacturer's instructions (Boehringer Mannheim Corp., Indianapolis, IN). Empty pBluescript was









38

digested identically for size comparison of digested plasmids. The 2.3 kbp insert of the MSP3-12 gene, which was previously inserted using Eco RI adaptors, was separated from plasmid DNA by electrophoresis on a 1% agarose gel in Tris Borate EDTA buffer (TBE) (45 mM Tris, 45 mM Boric acid, 1 mM EDTA). The gel was stained with 0.5 pg/ml ethidium bromide for 20 min. and photographed. The 2.3 kbp band representing clone MSP3-12 was cut from the gel and DNA was extracted from the agarose plug by ion exchange chromatography using QIAquick Gel Extraction Kit (Qiagen Inc., Chatsworth, CA).

A probe was made by random prime labeling 200 ng of MSP312 DNA with digoxigenin using the Genius System Nonradioactive DNA Labeling Kit according to the manufacturer's instructions (Boehringer Mannheim Corp., Indianapolis, IN).


Representation of pBluescript MSP3-12 in the A. marginale Genome


Restriction Enzyme Digestion


To verify the cloned pBluescript MSP3 was an accurate representation of genomic MSP3, multiple restriction sites of pBluescript MSP3-12 and genomic A. marginale DNA from a FL isolate were compared using restriction enzymes which cut within the MSP3 gene. Restriction enzymes Nco I (Boehringer Mannheim Corp., Indianapolis, IN), Bsp M (New England BioLabs, Beverly, MA) and Eae I (New England BioLabs, Beverly, MA), were chosen because they produced large fragments in different









39

areas of the MSP3 gene (Fig. 2). Digestions of genomic A. marginale DNA (1.0 Ag) and pBluescript MSP3-12 (0.1 Ag) were performed according to the manufacturer's specifications. Digested genomic and plasmid DNA were separated by gel electrophoresis on 1% agarose gels containing 0.1 Ag ethidium bromide and photographed.


Southern Blots


Prior to transfer, the gel was incubated at room temperature for 30 min. in 0.4 N NaOH, 0.6 M NaCI then 30 min. in 1.5 M NaCl, 0.5 M tris HCL, pH 7.5. Digested DNA was then transferred to a positively charged, molecular biology nylon membrane (Boehringer Mannheim Corp., Indianapolis, IN) by capillary diffusion using 10x SSC (ix SSC is 0.15 M NaCl plus 0.015 M sodium citrate). The filter was washed for 30 sec. in 0.4 N NaOH, then in 0.2 M Tris HC1, pH 7.5, 2X SSC for 2 min. After air drying, the DNA was cross linked by ultraviolet radiation and incubated for 3 hours at 650C in prehybridization solution containing 6x SSC, 0.5% SDS (w/v) 200 Ag/ml herring sperm DNA, 5x Denhart's [(x Denhart's is 0.02% Ficoll (w/v), 0.02% polyvinylpyrrolidone (w/v), 0.02% BSA (w/v)]. The digoxigenin-labeled MSP3-12 probe was added to fresh prehybridization solution at a concentration of 15 ng/ml, and the filter was hybridized overnight at 650C. Bound probe was detected by enhanced chemiluminescence using alkaline phosphatase conjugated, anti-digoxigenin IgG

























Fig. 2 Representation of clone MSP3-12 in the A. marginale genome. Diagram of clone MSP3-12 indicating the restriction enzyme sites and the resulting fragments used to map MSP3-12 to the genome. Nucleotide cleavage site for each restriction enzyme is illustrated diagrammatically (top) and by nucleotide number(bottom). Size of resulting fragments in base pairs is also provided.






















2,337





Restriction Cleavage site in Size of co-migrating Enzyme pBluescript MSP3-12 plasmid and genomic fragments


Nco I 12/1,188 1,176 Eae I 52/1,762 1,710 Bsp M 193/2,076 1,886



IIH









42

according to the manufacturer's recommendations (The Genius System Luminescent Detection Kit, Boehringer Mannheim, Indianapolis, IN). The membranes were exposed to Hyperfilm MP (Amersham International, plc, Buckinghamshire, England) to visualize bound antibody. The molecular sizes of comigrating cloned and genomic fragments were determined by comparison to Lambda DNA/Hind III fragments (GIBCO BRL, Gaithersburg, MD) and 1 kbp DNA Ladder (GIBCO BRL, Gaithersburg, MD) molecular size standards.


Presence of Multiple MSP3 Gene Copies


A. marginale genomic DNA from either the FL, VA, or SI isolate was extracted as described above and aliquots of 1 pg of DNA were digested with Hinc II (New England Biolabs, Beverly, MA), Sac I (Boehringer Mannheim Corp., Indianapolis, IN), Nde I (Boehringer Mannheim Corp., Indianapolis, IN), or Sph I (Boehringer Mannheim Corp., Indianapolis, IN). These enzymes were chosen because they do not cut within the known sequence of MSP3-12 gene. Calf thymus DNA was digested identically as a control. Empty pBluescript and pBluescript MSP3-12 DNA (0.1 xg) were digested with Eco RI and used as negative and positive controls, respectively for probe hybridization. Digested fragments were separated by gel electrophoresis with 1% agarose gels containing 0.1 Ag ethidium bromide and photographed. Southern blotting was performed under prehybridization and hybridization conditions









43

as previously described above and blots were probed with digoxigenin-labeled pBluescript MSP3-12. Bound probe was detected by ECL as described above. Lamda DNA/Hind III fragments (GIBCO BRL,Gaithersburg, MD) and 1 kbp DNA Ladder (GIBCO BRL,Gaithersburg, MD) were used for molecular size standards.


Distribution of MSP3 Copies in the A. marginale Chromosome


Restriction Enzyme Digestion in Agarose Plugs


The locations of multiple MSP3 copies in the chromosome were determined by Southern blotting of large A. marginale genomic fragments separated by CHEF. Infected erythrocytes containing intact genomic DNA from FL, SI, and VA strains of A. marginale were embedded in 0.7% agarose plugs and stored in 0.5 M EDTA (pH 9.5), 1% N-lauroylsarcosine, and 2 mg/ml Proteinase K at 4 oC as previously described above. Agarose plugs were digested with Not I (Boehringer Mannheim Corp. Indianapolis, IN) and Sfi I (New England BioLabs, Beverly, MA) as previously described (Alleman et al., 1993). Briefly, each plug was washed in 10 times their volume of T.E (10 mM tris, pH 7.5, 0.1 mM EDTA) over several hours with fresh T.E replaced each hour, then incubated in T.E plus 1 mM phenylmethylsulphonyl fluoride (PMSF) for 2 hours at 370C. Agarose plugs were then equilibrated on ice for 1 hour in 10 volumes ix restriction enzyme buffer plus 0.1 mg/ml BSA. A









44

solution of 100 units of restriction enzyme in ix restriction enzyme buffer was added and DNA digestions were performed according to the manufacturer's instructions. Reactions were stopped by the addition of 0.25 total reaction volume of 0.5 M EDTA, pH 8.0, 0.1% N-lauroylsarcosine (w/v), 1 mg/ml proteinase K (w/v), incubating reaction mixture at 40C for 1 hour, then 370C for 15 min. Uncut M. bovis chromosomal DNA (Promega Corp., Madison, WI) was digested identically to serve as a control for restriction endonuclease activity.


CHEF Gel Electrophoresis


This was done on the CHEF DRII system (Bio-Rad Laboratories, Richmond, CA). Plugs of digested DNA were electrophoresed in 1% agarose gels in 0.5x TBE buffer at 140C. Electrophoretic conditions were set at 180 V, 10 sec. switch rate, with a 16 hour run time. Delta 39 Lambda Ladders (Promega Corp., Madison, WI) and Lambda DNA/Hind III Fragments (GIBCO BRL, Gaithersburg, MD) were used as size standards. The gels were stained for 30 min. in 0.5 1g/ml ethidium bromide and photographed. Southern blotting and hybridization of the separated bands were performed as previously described except that the gel was depurinated in 0.25 M HCl for 15 min. prior to washing and transfer to nylon membranes. Prehybridization and hybridization conditions were performed identically to those described above. The membranes were probed with digoxigenin-labeled pBluescript MSP3-12, and bound









45

probe was detected by ECL as before. The membranes were exposed to Hyperfilm MP (Amersham International, plc, Buckinghamshire, England) to visualize bound antibody.















CHAPTER 4
RESULTS


Specificity Experiments


SDS-PAGE separated A. marginale initial body proteins were transferred to nitrocellulose, and probed with antisera from animals infected with related rickettsial agents or protozoal hemoparasites. This was done to determine if animals infected with these organisms contain antibodies which cross react with the 86 kDa protein (MSP3) of A. marginale. As a negative control, normal sera from various species were reacted with A. marginale proteins. Where available, protein preparations from related organisms were used in a homologous reaction with respective antisera to serve as a positive control.

Sera from a sheep infected with A. ovis (Fig. 3), a horse infected with E. risticii (Fig. 4), and a dog infected with E. ewingii (Fig. 4) showed strong reactivity with MSP3 of A. marginale. These sera as well as sera from animals infected with E. equi and C. ruminantium showed reactivity against other A. marginale antigens as well. Sera from animals infected with A. ovis, C. ruminantium, E. ewingii, and E. equi, showed reactivity against a 36 kDa antigen, possibly MSP2 (Figs. 3, 4, & 5). Sera from animals infected with C.

46
























Fig. 3 Specificity experiments using immunoblots. A. marginale (AM) or A. centrale (AC) initial body preparations reacted with normal sera from non-infected sheep (NSS), non-infected cattle (NBS), or antisera from animals infected with A. ovis (AO), A. centrale (AC), or A. marginale (AM). Labeling above each lane indicates the serum used (top), the initial body preparation used (center), and the dilution of the serum (bottom). Molecular size standards in kilodaltons are illustrated on the left.
















NSS AO AM NBS AC AC AM AM AM AM AM AC AM AM
1/500 1/500 1/500 1/100 1/100 1/100 1/300 200 -97.4 -69 -






46








30 -






























Fig. 4 Specificity experiments using immunoblots. A. marginale (AM) initial body preparations reacted with normal serum from a non-infected horse (NHS) or anti-sera from animals infected with A. marginale (AM), E. risticii
(ER), E. equi (EE), or E. ewingii (EC). Labeling above each lane indicates the serum used (top), the initial body preparation used (center), and the dilution of the serum (bottom). Molecular size standards in kilodaltons are illustrated on the left.































9v








69


'L6




-- ooz oo0/ ool oos/i ool oo/0

















O
























Fig. 5 Specificity experiments using immunoblots. A. marginale (AM) initial body preparations, or B. bovis (Bbov) or B. bigemina (Bbig) antigen preparations reacted with normal serum from a non-infected cow (NBS), or antisera from animals infected with A. marginale (AM), C. ruminantium
(CR), B. bigemina (Bbig), or B. bovis (Bbov). Labeling above each lane indicates the serum used (top), the initial body preparation used (center), and the dilution of the serum (bottom). Molecular size standards in kilodaltons are illustrated on the left.

















NBS AM CR Bbig Bbig Bbov Bbov AM
AM AM AM Bbig AM Bbov AM AM
1/100 1/500 1/100 1/100 1/100 1/100 1/100 1/500


200 -97.4 -69 -




46 -




30 -
.......... i'ii~iii!:,,,:iii~to









53

ruminantium (Fig. 5) and E. equi (Fig. 4) reacted with an antigen of slightly smaller molecular weight than MSP3. Sera from animals infected with A. centrale (Fig. 3) or either Babesia sp. (Fig. 5) failed to react with any A. marginale antigens although strong reactivity was demonstrated in reactions involving homologous preparations.


Conservation of MSP3 Between Different
Geographic Isolates of A. marginale


The conservation of MSP3 between different geographic isolates of A. marginale was evaluated by SDS-PAGE and immunoblots using initial body preparations from FL, VA, SI, and WA isolates of A. marginale. Preparations were separated on 7.5% to 17.5% (w/v) gradient polyacrylamide gels as described above. They were then transferred to nitrocellulose, and reacted with varying dilutions of antiserum from an animal experimentally infected with a FL isolate and a MAb to MSP3, AMG75C2 (McGuire et al., 1991), for definitive identification of the MSP3 antigen. An optimal dilution of this antiserum was established for demonstration of the immunodominant MSP3 protein (Data not shown). Preinfection bovine sera and a MAb to a trypanosome surface protein were used as negative controls and showed no reactivity to MSP3.

Initial body preparations from these same isolates were then probed with antiserum at a single dilution previously established (Fig. 6). The side by side location of the
































Fig. 6 Size polymorphism of MSP3. Immunoblots of initial body preparations from a Florida
(FL), a Washington (WA), a South Idaho (SI), and a Virginia (VA) isolate of A. marginale (AM), reacted with normal serum from a non-infected cow, Lane 1 (Pre), or anti-sera from a cow infected with a FL isolate of AM, Lanes 2-5. Serum was diluted to 1/300. Molecular size standards in kilodaltons are illustrated on the left.









55










M.W. FL WA SI FL VA kD 200











97.4





69







46



Pre Post
Anti-Fl A.M. sera (1/300)









56

various isolates clearly demonstrates variation in size of the MSP3 proteins. The 86 kDa MSP3 antigen is seen in the FL isolate (Fig. 6). An antigen of similar size is seen when this same antiserum is reacted with proteins from a VA isolate (Fig. 6). However, in reactions with the SI isolate, an antigen of slightly smaller molecular mass is seen, and the WA isolate has an antigen with a molecular mass greater than 86 kDa (Fig. 6).


Immune Response to MSP3


Realizing MSP3 is not conserved between different geographic isolates, we then investigated the possibility that animals infected with different isolates may contain immune sera that varies in reactivity to MSP3 from a single isolate. This is an important consideration in attempting to develop a diagnostic test antigen derived from a single isolate. Initial body preparations from a FL isolate were separated by 2-D gel electrophoresis as described above in order to determine if co-migration of antigens of similar molecular size occurs. Electrophoretically separated proteins were transferred to nitrocellulose and probed with antisera from cattle infected with a FL, VA, SI, or WA isolate of A. marginale as well as an anti-MSP3 MAb (AMG75C2), rabbit-antiMSP3 polyclonal serum, and an anti-MSP2 MAb (ANFl9E2). Preinfection bovine sera, normal rabbit sera, and a MAb to a









57

trypanosome surface protein were used as negative controls and showed no reactivity to MSP3 (Data not shown).

In a homologous reaction with anti-FL serum, 2 major areas of reactivity were seen with a molecular mass of 86 kDa, one with an apparent isoelectric point (pI) of 6.5, and the other with a pI of approximately 6.2 (Fig. 7). There was slight reactivity with an antigen at a pI of approximately 5.6. When the initial body preparation from the FL isolate was reacted with the anti-MSP3 MAb, major reactivity was seen in the 5.6 area of the pH gradient, but no reactivity was noted with antigens at a pI of 6.5 or 6.2 (Fig. 7).

Serum from an animal infected with a VA isolate showed similar reactivity as the MAb. However, serum from an animal infected with a WA isolate reacted with 2 antigens in an entirely different area of the pH gradient, having pIs of approximately 5.1 and 5.3 (Fig. 7). When these same initial body preparations were reacted with antiserum from an animal infected with a SI isolate reactivity was noted in all 3 areas of the pH gradient, with approximate pIs of 6.5 to 6.2, 5.6 and 5.3 to 5.1 (Fig. 7). When a rabbit-anti-MSP3 polyclonal sera was used, reactivity was noted in areas of the pH gradient having pIs of 6.5 to 6.2 and 5.6 (Fig. 8).

Although conditions seen here were not optimized to separate the 36 kDa proteins of A. marginale (MSP2), multiple spots were visualized in that apparent molecular size (Fig. 7). Some variation in reactivity of the different antisera to




















Fig. 7 2-D gel electrophoresis of A. marginale proteins. Immunoblots using initial body preparation of a FL isolate of A. marginale separated by 2-D gel electrophoresis. Letters centered above each immunoblot indicate the antibody used in the reaction, FL = antiserum from a cow infected with a Florida isolate of AM, VA = antiserum from a cow infected with a Virginia isolate of AM, WA = antiserum from a cow infected with a Washington isolate of AM, SI = antiserum from a cow infected with a South Idaho isolate of AM, and MAb AMG75C2 = anti-MSP3 MAb. Antisera are all used at a dilution of 1/300. Concentration of MAb = 5g/ml. Numbers above each immunoblot indicate the pH taken at 1 cm distances along the length of the tube gel. Arrows indicate the isoelectric point of each 86 kilodalton band. On the far right side of each gel, initial body preparation from a FL isolate of A. marginale was electrophoresed in a single dimension and immunoblotted along with 2-D focused initial bodies to indicate the 86 kDa, MSP3. Molecular size standards in kilodaltons are illustrated on the left of each immunoblot.





















pl FL pl M..b AMG75CZ i VA

T T11" 1. C' Co
lo ,oV
4.495 4.42 43 5*54453 13 452 49 2.9 6a 242 61S 5*2 54 5.3 5123 4.5 4.' 49 2.0 442 4123 53 5" 5.3 153 44

















9- 7u' V 7
D I2W p ICD





























0 0 0
1" &.22.13 S 3"3 . 35. iii 35

























17.22
4& WA p1 SI

2.9 4.5 ta 4.42 4.23 .5 5.22 53 5.23 4.5 2.57 4.5 Ca .42 423 53 56 5.3 .13 44
Uw MAC
U U


920 97.





25 .






22s 42.












277. 27 723 74.3


























Fig. 8 2-D gel electrophoresis of A. marginale proteins.
Immunoblots using initial body preparation of a FL isolate of A. marginale separated by 2-D gel electrophoresis. Letters centered above each immunoblot indicate the antibody used in the reaction, RB-955 = rabbit anti-MSP3 polyclonal sera and MAb ANAFl9E2 = anti-MSP2 MAb. Rabbit anti-MSP3 was used at a dilution of 1/5,000. Concentration of MAb = 5gg/ml. Numbers above each immunoblot indicate the pH taken at 1 cm distances along the length of the tube gel. Arrows indicate the isoelectric point of each 86 kilodalton band. On the far right side of each gel, initial body preparation from a FL isolate of A. marginale was electrophoresed in a single dimension and immunoblotted along with 2-D focused initial bodies to indicate the 86 kDa, MSP3 and 36 kDa MSP2. Molecular size standards in kilodaltons are illustrated on the left of each immunoblot.















61














pl RB-955



6.97 6.9 6.68 6.42 6.13 5.82 5.66 5.36 5.13 4.16 M.W.
kD

200



97.4 69 46






30






21.5


pI MAb ANAF19E2



6.97 6.9 6.68 6.42 6.13 5.82 5.66 536 5.13 4.16 M.W.


200 97.4 69



46





30






21.5 -









62

MSP2 was seen. For example, in a homologous reaction there was reactivity with antigens with pIs from 6.42 to 5.66. However, when reacted with sera from animals infected with WA or SI isolates, reactivity with antigens with pIs of 6.42 to 5.36 was observed. These spots were confirmed to be the MSP2 antigen by use of an anti-MSP2 MAb (ANFl9E2), the reactivity of which has been previously described (McGuire et al., 1984; Palmer et al., 1988; Palmer et al., 1994). Multiple spots with a molecular mass of 36 kDa were recognized by this MAb (Fig. 8).


MSP3 Nucleotide Sequence and Translation


Drs. Travis McGuire, Terry McElwain, and Guy Palmer (Washington State University, Pullman) identified and sequenced 3 clones of the MSP3 gene by screening a genomic library of a FL strain of A. marginale DNA with MAbs to MSP3. These clones were designated MSP3-11, MSP3-12, and MSP3-19. The 3 clones in pBluescript vectors, along with their nucleotide and deduced amino acid sequences, were supplied to us courtesy of the above individuals. The entire nucleotide sequence of each clone is illustrated in Figs. 9, 10, & 11. A complete gene sequence is lacking in each clone. However, MSP3-12 contains the N-terminus of the gene and 633 bp upstream to the open reading frame. This area upstream to the 5' end of the open reading frame is likely to contain
































Fig. 9 Gene sequence of MSP3-11. The nucleotide sequence of pBluescript MSP3-11 is illustrated. Numbers on the left indicate the number of the first nucleotide in each row. Termination codon is underlined.










64



Msp311.Corentry:1 Length:3263 May29, 1991 11:34 Check:1828

1 GGTGGGGAGATGGTAGGAGTTGAT GAAGGCTAGT TATACAAGAACT GAOCAGAC CAGAAGAAT TAGAAAAGCTACAACATGAACTAGCAAAGCAAGTAA 101 GTAAAT TAGCT GAACT T GGAAACT TAAGTGGT TAGAGCMCT T GAGACACT GGAGACT CAGGAGT T GGAGAAGGT GGCGAAAGAGCCACACAGAAGCT
201 CAGTATATTMGAGGAGCAGCAGGATACAUTCACTAGAGAGCTGGAGAGAAGCTGAAGGAGATAGAGGAAATAAGGMACTGAAGGACCGGAGGCA 301 AT CAGAAAGCT TAGAGACT TGGAGAGGCTGGAAGC T GGAGMGT T GGAGGAGGT GAAGAAGAAGGTAAAGGGTTCAGAGC TTGCAGAGCATT T GGACAAAA 401 CGTGGTTGCTACGCAGGATTGGAAGAGTAGCAGGAACTAGCAGCGATMGGGAGCTGAAGGAATTGGGGCTAGAGGAACGGCTCAGGGTACTAGCTGA 501 GAT TAAGGAAGT TAGAGCATT GGCAGAGAAGCGGAAGGCTGGGGGAC TAGAGAT T CAGGAGGGGT TACAGCT GACTGAGAAGAT TAGAGCAT TGGGTGGA 601 CAGCT GGAT CT GCT GGAGGCACGAATAT TAT T GGGGCTAGAAGC T GAGAAGAT GAAGGACGT GGAAGCGGCAAAGGAGGMGT GGAGATTT T CCTGGGAT 701 TGCTCCACAACAATGACACAGAGACCAAGAAGATAGAGGAGGTGAAGACTCTGGTAATGCGGGTGGGTGACAGTCAGGGACTAAGGGGTOCGT T GGTCAA 801 GAAGTTAGAGGAGCTGGCTAGAAGT T GGAGCCAAAGGTGGGTGGCAATACAGGGT T T CAGGGT CAGGTAGGGCTAGT GMAAGGATCTTAAGAAAAAGCTA 901 GAAGAACTAGCAGCGATAAGGGAAGCAT T GGAGCAAC T GAAGGT GGAACCAGCGC TGCTAGAAGGGATAAAGAGGGAGGTGCTAACAOCAGAGTACTC 1001 AAGTTACACGTAGGGAGAATATCCACAAACAAGGCAGMGGTAAAAGAAGAGATAAGAGCGCTTAGAAAGCTAGGACAGTTGGATGAGCTAATACC 1101 AGAAGAGAGAT TGAGAGAATT GT CCAAGGTTAAGGAAACACT GAAGCGG TT GAAGGGTAAGGAGCTAGTAGACCTAGAGAGGAAGCTAGAAACTAT GGGG 1201 GAACTAAAGCAGGAGATAGAGAAAATCAAGGGGCAGGAAGAC T T GAAGCAGCTAGAGGCTAGGAAGT T GGAGGAGGTGAAGAAGGCAAAMAAGGGAAGTGT 1301 TTAT T CTGAACAAAGGAGTCTGGAGAAAGAGGGGAAGCTGGAGGAGTTAAGGGGCTATAAACTCAAGAAGGCGA AGCTG AACCTTGAT 1401 AGAGAAGATACGAGGGGTC CACGCTGAAGGAACAGT GGGAGCCAAAAGTTGAGAAGCTCAAGAGCAAACTAGAAACACTAGCT CCATAAGGGAGCTG 1501 AAGAAAT T GGGGT T TAAAGATT GGT TGAGACT CAGGACACTAGAGGAGCTTACAGAGATAGCT GAGCAGOGTGGAGT TGCAACAGTGAT G AGGCAGCAT 1601 TAGCTAGTGCGATGGAGATAGCTAAGAACAGGGG TTGGACTGATTAT T TAAACAGTCTAGATGTAAGTGAGAGAGCTAATGCTGCAAGGGAATTAATTGC 1701 TGCTGAGAAGAT TAGAAAATGGGCTAGGGATATTAATAACTTGGA T GCTGATGAACGGGC4ATGGTCGCTGGGGCCCTAACCCCTTCTACAACAGTGTGA 1801 CTCACCACTCTCCCCCGAAACATCACTCACACTTCCAACTCTACTGGGCACCACCAAGGGCAGCCAGCCTCTCTTGGCACCTCATCTCGTACCCTCTTGG 1901 CCTACCAAATCTCGACCCTGCCACCACAGCCCTCT GCACCGGTGTCACCTGCATAACCCGT CT CCCAGTCTCTTGTACCTCAACACCTGCATCTCATCTT 2001 CCTTACCCAATGTGGCATTCACAATCT TTGTCCTCTCTACTTACCCCCTACTTCTTACCGGCACCACTCACTGCTTTCCCAGGTCAACTAGTCACCCAA 2101 CCTCTAGCACCTTCAGCATCTCAGCCCAACTCACCACAGCTCTCTGCACCTTCAACAGCTCTAGCAAAAGCCCCGGCAACAACTGCCTTATCCTCCTTAC
2201 TCAGGTTATTAATGTTAT T CGCCATCCCCTGCAAGTTGATGGTGTCT T TGT T CCCCTCAACAGAAAGCAGTGTGTCTGTACCCTCAGTGAACACCTCACT 2301 AATCTTGGTTGGCAGTGCTATCGGTGGTACCACACT TCTTAGTGCCATTACTGCCTTCGCCCTTGCCACACATTCTGGCTCACCTTCTCACCATTA 2401 GTAGTACCCTCTACTGCTTTACCCCACTTCTTGGCTTCACCCTTAGTCATCTTACCTAAAGCAGTGGCAAGACGGTCTACCTGACCTCTTGCTGTATCAT
2501 ATGCTAACTCCTTTCCTAATAAGAATACTGAAGCTGTAT CCTCATTAGACTTCTTACCTCCCT TAATGACAAACCTTTCATACCCTACTTCTACT TCAAC 2601 CCTTGCTCCTCCAATACTATACCCAATGCTCCCCTCTAGGCTGTGAGCATGCTATGGTAGCCCTGATATGGTTATTCATCACAGCAGGTTCTACTAATG 2701 CCGATACTATCAGCAGTACTAACACTGCTCGATACAGCTACCAAGGT TAT T T GTGACGTCATTGTCTTCGTGCAGAACTGGGATTACATCATGAC 2801 AGGGGTGGTACTAGGT TCTGGCGT CATGGCTTGGCACACTTGCATGGCCTGCCATGGTAATGCTGGTTGTGT GTACAGCCTAGGTAAGCT TATGGCCAC 2901 TGTATTGAGGGTAATACCAGTTAGTGCAATGGACGTAGCGGTTTGGGATCCGGTATGTGCT ACTTAGATTCTTACTCATTGCTGTAGTAT 3001 TAGCATTTGGTATCTTACATGGTGTACCTGCTGGGGCTAGTGCCAGCGAGGGTGCTCAGACTGCTGATATACAGCGACTGTTGTTATATGTAACGT 3101 TATACGGTTTGTGCAGAAGCTGTGGACTACATCATGACTGGGGTGATACGTCAT CTACATTATGGCTATCTTTGGTAAGCTTGCATCCTGCTATT 3201 GTAATGCTGGTTGTAT TTACAGCTATATTCTTTGGTGCTGGTAAGCT TATGGCTAAATTCCCG
































Fig. 10 Gene sequence of MSP3-12. The nucleotide sequence of pBluescript MSP3-12 is illustrated. Numbers on the left indicate the number of the first nucleotide in each row. Start codon is underlined.











66





Nsp312.Corentry:1 Length:2337 August21, 1991 15:43 Type:N Check:1077

1 GGGGGCCT GCCATGGTAACTGGTTGTATGTACAGCCAAGGTAAGCTTATGGCCACTGTATTGAGGGTAATACCAGTTAGTGCAAGGGTGGACGTAGA 101 GGTTTTGGGAGAGT GGTGATGTTGCTAAAACTTAGATTCTTACTCAT TGCT GTAGTATTAGCATTGGGTATCTTACATGGTGTACCTGCTGGGGCTAGTA 201 AGTCCAGCGAGGGTGCTCAGACTGCTGATGTACAGCGACTATTGTTATATGTAACGTTATACGGTTTGTGCAGAAGCTGGGACTACCCATCATGACTGG 301 AGTGATACTAGGTT CTAGCATTATGGCTATCTTTGGTAAGCTTGCATGGCT GCTATTGTGATGTTGGTTGTATTTACAGCTATATTCTTTGGTGCTGGT 401 AAGCTTATGGTAAGTTTGCCTGCTGGGTTGAGTGGTGAGGTGTGAAGGATGCCGGTAGCTTTAGTTCCAGTTATAMGGCACTGCTAGGCAGTGAT 501 GGTGCAGAGAGCTGTGGTGGTCGTGCCAAGTTAGACTGGGATGGCCTTTGTTATGTCGGTGGTACCAAGAACTTCAAGTGT 601 GTTGGTGCCCAGTAGAGT T GGAAGTGTGAGGATGT TTCGAGGGAGAGTTGATGATGCCAAAACTTAGTTCTTACTATTGCTGGGGATCAGCATT 701 GGGTATCCTOCATAGTATGTCT GCGGGGGCTGCCGATAGMGTAAGTCACGCACACAGAGGTTGGCTAGTGGACCGGCTTT GGGGAAAGGCAACGGGAGC 801 T T CTACATAGGCCTAGACTATAACCCAACTT T CAACGGTAT CAAGGACCTGAAAAT CAT OGGCGAAACCGAT GAGGATGMAT GGATGT TCTCACCGGTG 901 CCAGGGGCCTGTTCCCGATGAACGCT CT TGCTAGCAACGTCACCGATTTTAACTCATACCACTTCGAC T AGTACCCCACTGCCTGGGCTAGAATTTGG 1001 GAACAGTACCCTGGCTCT TAGGGAGCATTGGGTACAGAATTGGAGGAGCCAGGGTTGAGTAGGGATAGGACATGAGAGTTTGTTATTAAGGGGGGA 1101 GATGATGCAGCATTCCTACTAGGTAGGGAACTAGCATTGGATACAGCGAGGGGTCAGT TACTATCCAGTGCAT TGGGTAGGATGTCCATGGTGATGTAC
1201 ACAGATTAAAGAAGGAAGTAGTTGATAGTATAGGAAGAGGAACAGCTAGTCCTGTAAGGGAATGTTTAGTAGAGAGATCTCAGATGGGMTACATTACT 1301 TGCTGGGGAG4TGGTAGGAGTTGATGAAGGACTAGTTATACAAGAACTGAGCAGACCAGMGAATTAGAMAGCTACAACATGAACTAGCAAAGCAAGTA 1401 AGTAAATTACTGAACTTGGAACTTAAGTGGTGAACTTG GAAAACTGGAGACTGAGGAGCTGGCGCAAGGACTCGAAGGAGCGCTAAAAGCTT 1501 TGGGCGTGGMGCATCAGTGCAGGAGCTGGTACAGAGATTCAAGAAGGAMTTGCTGATGTAAGACACCGAAGAGATMAGCTGGAGTGGATCAAGGA 1601 GATAGAAGCTAAGAAGT TAGCGGAGGTGCAGGAGCAGGCTGAGAAAAGCTGGCGGAGCTGGAACTAGCGGATAAAGGAGGAGCATTGGGkACTAAAGGC 1701 GAAATACGGGAAGTTAGGCMCTCAAGGAACTAGCGGATMAGGAGGAGCGTTAGGAATCATGGCCGAGMGCTCAAGAAGCAGGAAAGTCTCAAGGGGC 1801 TAGGAGGAACAGTAGAAGAATTAGCAGCGATAAGGGAACTGAAGGAAT TGGGGCTAGAAGAGCAGCTGAGATGCTGGCTGAGATTAAGGAAGTTAGAG 1901 ATTGGCAGAG4AGCAGAAATCACAAGGATTAGAGGCTCATGAGGGAT TACAACTGACTGAGAAGATTAGAGCATTGGGAGGACAGCTGGATCTGCTGGAA 2001 GAGAGAT T GAGT GAAAAGT TTAAGAAC TACAGCAGAAT CAGCAAGGTATACGTAAGGATCT GAAT CCACAAGCTATAACAGCAGT GGCAGGTAAACTAG 2101 ATGAGAT T TGGGTACT GGGGCCGCTAGGACAGT TAGATGAGC TAGT GCC TGGGAAGAGAGC T CAGACACTGGGGGAAATGAAGGCAACGCT GAAGGAGCT 2201 CGAGAAAATACGGAAGTTAGTAGACCTAGAGAAGAAGCTAGAAACTAT GGAGAACTAAAGAAGGAGATAGAGAAAATCAACGAGGAGGGGAGCTTGTG 2301 AGGTTTACAOGCAAAGAAGTCTCGGAGATATCACCCG
































Fig. 11 Gene sequence of MSP3-19. The nucleotide sequence of pBluescript MSP3-19 is illustrated. Numbers on the left indicate the number of the first nucleotide in each row. The termination codon is underlined.











68



Msp319.Conentry:1 Length:4185 October21, 1991 15:01 Type:N Check:2553
1 GGTGATGTACACAGATTAAAGAAGGAAGTAGTTGATAGTATAGGAAGAGGAACAGCTAGTCCTGTAAGGGCAATGT TTAGTAGAGAGATCTCAGATGGGA 101 ATACATTACTTGCTGGGGAATGGTAGGAGT T GATGAAGGACTAGTTATACAAGAACTGAGCAGACCAGMGAATTAGAMAGCTACAAATGAACTAGC 201 AAAGCAAGTAAGTAAAT TAGCTGAACT TGGAGAACT TAAGT GGT TAGAAGAACT T GAAAAGCT GGAGACTGAGGAGT T GGGGGAGCT GTTGAGACTGAAA 301 GCGAGAAAAGCCTCACAGGAACTTAGGACGTTGGCGGAAAGTAAAGGACAGCACCTAAATGCTGATGAGATAGAGGGGGACTAAAGAGCGCTGGGAT 401 TGGAAAGACTGGAAAAGT T GGCAGAGCTGGAGTTGGTCAMCAGCGGCTTGAGTCGATGAGGAGCTGGAGAAGAAGATAGAGGAGGCGCAGTTGACAGC 501 AGAAGAGCT CGGGAAATGAGGGAGAAGCTTAAGGAGT T GGCAGGAAGGGGGGAGGATCCTGCGGGGCTG AGAAAGCGATAGAGGAGGC AATGCTGAC 601 GAACT CAAGMGGGGCTGAAGGAGGTGCTG4AGACGCTGGAGGGAAAGAAGTCAGAGCT GTGAGGGAAGTAGGAAGGTTGAAGAAGGGAGCCTTGGAGG 701 AGATGAAGGCAGGAAAGGGGT T GGTAGAGGAGGTAAAGGGAGAGCTAGGAAAGCTAGAGC TAAGAAGATAGAGGAGGT AAGGCGGCAAAGGGGGAAGT 801 GTCG T T T T TCCCAGGAATAAGAGAACTCAAAAACTGGAAAGGAGTCGAGTTGCTGGAAAGGCTTGAGGAGGGGCTGAGGCGGCTAGAAGATAAAG 901 AAGGTGGAGGGCCAGGAGAAGCTGTTACAATCTCAAGACAAGCTAGMGAACTAGCAGCGATAAGGGAGCTGAAGGMGGGCCAGAGCTGGTACAGA 1001 T GCT CAAGGAGATAAGACT GGAGCT GGAAGAGCAGCT GGGGAT GC TAGCTGAGAT TAAGGAAGT TAAGGCACT GGCAGAGAGCGGCAGGCT GGAGGACT 1101 GGAACTTCAGAGGTATTACAGCTTACAACTAGAACTGTGGCATTGGAGAGGAAGCTGGTAGAAAAGGTAGTAAGCTAGATAGGAAGAGCCTGGAAAGT 1201 CAGAAGAAGGTACTGGAGGAGCATAAGAAGGAGCTGGAGAGGAAAAGCTTAAGCTAACTAAGGAGCTGGCTGAGAAGCGCAAAGCCAAAGGGCTCTTTG 1301 GAGCAGTACTAGACCTGGGACT GGT GGCGCTGGGT GTAGGGAAACAAAGGACT GGGAGTTAGT GATACTGGAGAAAAAACTAGCCAAGAT CAACGCACT 1401 CCTAGAGGGTGGAGATATAGCAAAACTGGGkAGACAGATAAATAGGATCAAGTGGCTAGAGGATCTAGCTGT TAGTAAGAAGCT TAAGGCAGCATTAGCT 1501 AGTGCGATGGAGATAGCTAAGAACAGGGGTT GGACTGATTATTTAAACAGTCTAGATGTMGTGAGAGAGCTAAT GCTGCAAAGGAATTMTTGCTGCTG 1601 AGAAAATTAG4AAATGGGCTAGGGATAT TAATAACT T GGAT GCTGATGAACGGGCAATGGTCGCTGGGGCCTAACCCTTTTCCTAACCOCTTCTACAAC 1701 AGTGTGACTCACCACTCTTTCGCAACATCGCTCACACTCCTAACCGGCCACTGTACTGAGGGTAATACAGTTAGTGCAAGGGTGGATAGAGGTTT 1801 TGGGAGAGTGGTGATGTTGCTAAAACTTAGAT TCTTACTCATTGCTGGAGTATTAGCATTGGGTATCTTACATGGTGTACTGCAGGGGCTGCTGCTCAG 1901 GCGAGCCCTAGTACTACTGGTACTGAAGCTGATGA CTGTTGTATATGCAATGTTATACGGTTTGTGCAGAAGCTGGGACTACCCATCATGA 2001 CTGGGGTGATACTAGGTTCTAGCATTATGTATCTTTGGTAAGCT T GCATGGCCTGCTATTGTAATGCTGGTTGTAT TTACAGCTATATTCTTTGGTGC 2101 TGGTAAGCTTATGGCTAAATTTGCTGCTGGGTTGAGTGGTGATGGCATAGGT GATGCCAGTAAGTT TGACTGTTCACAGATACAGCGGTCAACACCAGC 2201 CAAAGTAACCATAGAAGATTAGGTGCTGCTGGGCAGTGATGGTGCAGAGAGCTGTGTTGGCATTGGTTGGGAGATGAGGCAGCGGGAGTTCGGAGATAG 2301 GGACTAGGTGCATGGACGATGCAGGTGTGGGCAGAGAGCTGTGTTGGCCTTGGTAGTGAGT TGGGCTGAGGTGCCAAGGGCAGGAGGAGGTACCAGAAGT 2401 TGGGAGATGCTGGTTACCGTTGAAGTGAAOCGGTGCAGAGAGCTGTGGCCTTGGGGCCCGTAGAGTTGGCTGAGGGGCAGGGGTAGTGAAGGACACG 2501 ATTTGACTGTTCCAGTTATAAGGAAGTCAAGACCGGCACCAAACCAAGCTAAAGGTTAGGAAATGAAGTGCCGCGTGATGTGCAGAGAGCTGTGGTGGT 2601 AGTACAGGGGTGCTGGTGATAGT T T CTGGTGCTGCAGGGGTCAGAGGACTCCTACTCATGCTGCGGATTATCTCCTTTTGTGTTCCCACAGGTTCACAT 2701 AATTAACCGTACATGGGTAMTGTGCTGGTAGTGCCGGGAACCGCAAGTGGTAGATGGTGGAGGGGCACATTGGGGTGGTGTCCTAG4ATCCGAACC 2801 TGATTCCTAGTTCTACCCCTATAT TGCGTACCCGAAGGATGCTGAGACCTTCTCCTTGT T TTTCCTATGGGCTGGCGTCATCGACTGACGGTGGGC 2901 AGGGGTGTTGTCGTAGTTTGTGCCTAGTGTT T TGCTGGCAGTGCCGCCAATGAATGCTGAGGCTGGGTGTGATGCTGTAACTCACACCGCCTTGACC 3001 TGGCAGGTAAGCTGTGGCTGAAATCGTTGCGGTCAGTCCTACAAAGCTTGCACCCAACCCTGCGCAGGCGTAGGGAGACATTCTCCAGTCAACTAGTA 3101 ACCCCACCT GTGGGAAATCATAGCATGCACTCAGGATCGCTGAAACTACOCTGACCCCACTTAT T T CCACkATCTCCATrGCCCTCTGTGCTTGTGGCAAT 3201 TGCCCGACCMTGACGCCTTTTGCGAGCCGTGAGAGGCCMCAATATCCGT GATAGCTTGATTCCGCTTGCTCTCTTGCGTGTTCTGTGCAGAGGTATCA 3301 GCGCTACCAGCACCGGTTGCAGCAGCACGTTCAGTTTTGGCCTCCTCCCCAGGTATGATGGCATCAAACTTACTTATAACTTGTCAGCAGCGCTTTGC 3401 TTTCCCACGCTTCGTCATGGT CT CTAATATCTTCCAAACG CTTGATATCTCAAAGCGGCT TGGACAACTCCTTCTGCAGCGCGGCCCGAAACCT 3501 TCGACGTCAGCGTCGCACTGGCAGATATTCTTTTCACCAGGGCAAAAGTCGCACT CCCGCCGCTCAGTMGGCATAATTTTGCGATTTGAATCATAGT 3601 GCTCATGTCOGAATTCTAACTCAAGCCTGGTGCCGCGTACCACGCAACCCACGCTTCCCCTGAGGCCCAACAGACTGCTGT CTCAAACCCATGGCCGG 3701 ACTCTCCGTAGATTCCCCCGCCCAATCGTAGT T GGCGGAAGAAAGCTCAGCAGAGCCCAACTTCCCAATGTAGGGCAGTATCGCGACGGTTTCTTTTGCA 3801 GCATTAAGCCTAAACCCGCTTATCT TCCCTATGGTGGGCCCGTACCCAAAACTCACGTAAAAATCTTCAOGACCGT TCATCCCATGGGAACTCGCTGGGA 3901 CCAGATAACCAATCGTACTCAGCAATGCAGCAGCTATACATACTACGGGCTTCATGACGCACCCAATGACACCGACAACCAACAGGAGGCTAAACGCCCC 4001 TCACTGCACGCAGTGCTATACAACTGTTMAAAGGTCTTAAAATGAGTTGCAT T T T TGGGGCT TGCMCTATGCCTAGCTAGAGAATAAGGATTCCG 4101 GGCTGGCGGGCCAAACGGCGCACACTCAGGTTGCCCTTCGTGTGCATGCATCATCAGCGTGGCACTACCTACTTTTTCAGACCG









69

regulatory sequences for gene expression. Clones MSP3-11 and MSP3-19 lack the N-terminus but each contains the C-terminus and 1,473 bp and 2480 bp respectively, downstream to the 3' end of the open reading frame. A schematic representation of each clone shows areas of homology between the 3 genes as well as areas which are homologous to the MSP2 gene of A. marginale (Fig. 12).

Homologous areas are distributed throughout the 3 MSP3 genes. Areas common to all 3 genes are at bases 1191 to 1472 and bases 1818 to 2008 of the MSP3-12 clone. An area at the 3' end of MSP3-12 between bases 2090 to 2280 is also found in MSP3-11, but is absent in MSP3-19. The last 200 bp at the Cterminal end of genes MSP3-11 and MSP3-19 are homologous. This area is unavailable for comparison in MSP3-12.

The MSP2 gene encodes a 36 kDa surface protein, and is known to be expressed by a polymorphic, multigene family (Palmer et al., 1994). Areas of homology between the Nterminus of the MSP2 gene and MSP3-12 are indicated in Fig. 12. The N- terminus is absent in the MSP3-11 and MSP3-19 clones, so comparisons cannot be made. The areas of amino acid sequence homology between MSP3-12 and MSP2 (amino acids 55 through 176) shows 65.6% similarity and 54.9% identity (Fig. 13). There is a homologous area of over 500 bp between MSP3-11 and MSP2 (Fig. 12). However, this area is outside of the open reading frame, in a different reading frame, and a

























Fig. 12 Diagram of MSP3 clones with MSP2 homology. A schematic representation of clones MSP3-11, MSP3-12, and MSP3-19. Homologous regions are indicated by like shaded areas. Regions homologous to the MSP2 gene are indicated by MSP2->. Arrows indicate the direction of the reading frame. Nucleotide numbers are indicated on the bottom. Top numbers of clones MSP3-11 and MSP3-19 indicate corresponding nucleotides in clone MSP3-12.

























MSP3 CLONES1




MSP3-12
Stn





633 794 1141 1191 1306 1472 1818 1953 208 2090 220 2337 MSP2-> MSP3-11
1306 1472 a1818 200 2090 2280 Stop I 172 3 1042 1226 1190 17n 214 2654 3263 <-MSP2 MSP3-19
1191 1306 1472 1818 1953 Slop I 176 278 92 1124 1485 1705 3049 3077 418 <-MSP2











C0.































Fig. 13 Comparison of MSP3 and MSP2 protein sequences. BestFit alignment of amino acid sequences of MSP3 (top) and MSP2 (bottom). Identical amino acids are indicated by a vertical line. Conservative substitutions are indicated by an asterisk (*). The symbol ...... is used to denote a gap used to achieve optimal alignment between the sequences. Amino acid numbers are indicated at the beginning and end of each line.










73














BestFit Allignment of MSP3-12 (top)
and MSP2 (bottom)



55 GSFYIGLDYNPT FNGI K DL K IIGETDEDE 83

1IIIIIIIII II III 1 I
50 GSF YIGLDYSPA FGSI KD FKV. Q EAGGTT 77


84 MDVLTGARG L F PMNALASNVTD F NSY H FD 112

II. II 1. I I ** II
78 ... . . R G V F P Y K R D AAGRV D F KVHN F D 99


113 WST P L P GLE FGN S T L AL G GSIGyRIG I G GA 140

I I 1 1 I I III l I III
100 WSAPEP K ISF K DSMLTALEGSIG Y SI GGA 128


141 RVEVGIG HERFVI KG GDDAA ...... F L L 163

III I, IIIIIIII III
129 RVEVE VGYERFVI KG GK KSNEDTAS V FLL 157


164 GRELALDTARGQ L 176

I I I I I I I I
158 GKELAY HTARGQ V 170









74

different direction. A similar, but much smaller area is seen downstream to the open reading frame of MSP3-19 (Fig. 12).


MAb and Immune Sera Reactivity to Recombinant MSP3


Lysates from E. coli cells transformed with pBluescript containing each of the 3 MSP3 genes were separated by SDSPAGE, transferred to nitrocellulose and reacted with anti-MSP3 MAb (AMG75C2) or immune sera from cattle infected with a VA or FL strain of A. marginale. This was done to verify that a cloned MSP3 gene represented a gene which produced one or more of the 86 kDa antigens seen on 2-D immunoblots. Initial bodies from a FL strain of A. marginale were used as a positive control for MAb (AMG75C2) and immune cattle sera. Lysate from E. coli transformed with empty pBluescript, antiTrypanosome brucei-MAb (TYRP1El), and pre-immune cattle sera were used as negative controls.

MAb AMG75C2 reacts only with recombinant protein expressed by clone MSP3-12 (Fig. 14). The recombinant MSP3 has a molecular size of approximately 65.0 kDa. This corresponds to the calculated size of the protein expressed by the open reading frame of MSP3-12 which is 1,707 bp (Fig. 10). No reactivity was observed between MAb AMG75C2 and recombinant MSP3-11, MSP3-19, or empty pBluescript. Isotype control MAb TRYP1E1 showed no reactivity as well.

Immune serum from a cow infected with a VA isolate of A. marginale reacted with recombinant MSP3-12 (Fig. 15).
























Fig. 14 Immunoblots of expressed MSP3 clones. The reactivity of expressed proteins from clones MSP3-11, MSP3-12, and MSP3-19 are compared using an anti-MSP3 monoclonal antibody (AMG75C2) and a negative isotype control (TRP1E1). pBlue is expressed proteins from empty pBluescript plasmid used as negative control. FL is A. marginale initial body preparation from a FL isolate as a positive control. Numbers on the left indicate molecular size markers in kilodaltons.

















Antigen FL NSP3-11 NSP3-11 32 SP3-12 SP33--12 SP3-19 SP3-1 tue pu Antibody ANG Trp ANG Trp ANG Trp ANG Trp ANG
75C2 1E 75C2 1El 75C2 1E1 75C2 1E 75C2



kD
200.0 -97.4 -69.0 -46.0-30.0 -21.5 -14.3 --J
























Fig. 15 Immunoblots of expressed MSP3-12 clones. The reactivity of pre and post-infection sera from animals infected with a Florida isolate of A. marginale (Pre and Post-FL) or a Virginia isolate (Pre and Post-VA) are compared using expressed proteins from clones MSP3-12 and empty pBluescript plasmid (pBlue). Antigens are listed on top and anti-sera is listed on the bottom. In lane 1, initial body preparation from a FL isolate of A. marginale (FL) is reacted with immune sera from an animal infected with a FL isolate as a positive control. Numbers on the left indicate molecular size markers in kilodaltons.












co






-- S'LZ






-- O'O










-- 0"69


-- 9'L6 A-- 0'00Z VA VA VA VA 11 "1 "1-1 11 'RN
-aad -aid -sod -asod -aid -id -:SOd -4S0d -4SOd APOqtuV ZL-EdSH anIOa iZL-EdSN anliP ZL-EdSN anl8a ZL-'dSN anl8a 11 UOFluv









79

However, when this same recombinant protein was reacted with immune sera from an animal infected with a FL isolate, no reactivity above that which was seen with empty pBluescript was observed.


Representation of pBluescript MSP3-12 in the A. marginale Genome


To verify that the cloned pBluescript MSP3-12 faithfully represented the genomic copy, genomic DNA from a FL isolate of A. marginale and pBluescript MSP3-12 DNA were digested with restriction enzymes to yield predicted fragments of specified lengths. The enzymes used, the restriction sites in pBluescript MSP3-12, and the predicted size of the fragments are illustrated (Fig. 16). DNA fragments were Southern blotted to nylon filters and probed with whole digoxigeninlabeled MSP3-12. All enzymes used produced pBluescript MSP312 fragments of the predicted size which comigrated with a fragment in the genomic DNA (Fig. 17). Comigration of large fragments was seen in digests extending from the 5' terminal region to the 3' terminal region of the gene.


Presence of Multiple MSP3 Gene Copies


A. marginale genomic DNA from either FL, SI, or VA isolates was digested with restriction enzymes, Southern blotted to nylon membranes, and probed with whole, digoxigenin-labeled MSP3-12. Enzymes Sph I, Nde I, Sac I, and


























Fig. 16 Representation of clone MSP3-12 in the A. marginale genome. Diagram of clone MSP3-12 indicating the restriction enzyme sites and the resulting fragments used to map MSP3-12 to the genome. Nucleotide cleavage site for each restriction enzyme is illustrated diagramatically (top) and by nucleaotide number(bottom). Size of resulting fragments in base pairs is also provided.






















2,337





Restriction Cleavage site in Size of co-migrating Enzyme pBluescript MSP3-12 plasmid and genomic fragments


Nco I 12/1,188 1,176 Eae I 52/1,762 1,710 Bsp M 193/2,076 1,886

























Fig. 17 Genomic representation of MSP3-12. Southern blot of genomic DNA from a Florida (FL) isolate of A. marginale or pBluescript MSP3-12 (CL), digested with restriction enzymes Eae I, Bsp M, or Nco I and probed with digoxigenin-labeled MSP3-12. Molecular size markers in base pairs (bp) are indicated to the right and left.
















EaeI BSp M NcoI

FL CL FL CL FL CL

bp-- 23,130 5,00-- 9,416 5,090 -- -- 6,577 4,072 ---- 4,361 3,050 --2,332 2,036 -- 3 --2,057 1,636 -1,018 -506-- 564 506 -396 -344 -298 --









84

Hinc II were chosen because they do not cut within the sequence of the MSP3-12 gene. This should produce a single band if only one genomic copy of MSP3 is present. Sac I cuts once within the open reading frame of MSP3-19 (at nucleotide 498), potentially producing 2 observable bands for a single copy gene. None of the enzymes cut within the MSP3-11 gene.

Multiple bands were observed on Southern blot hybridizations, indicating multiple partially homologous MSP3 copies (Fig. 18). The exact number of copies cannot be determined since restriction site polymorphism may exist in other MSP3 copies, resulting in more than one band from a single copy. Similar intensities of many of the bands within a single isolate may indicate extensive homology exists between some of the copies. Restriction fragment length polymorphism (RFLP) is seen when comparing enzyme digests of each of the 3 isolates of A. marginale.


Distribution of MSP3 Copies in the A. marginale Chromosome


Intact, A. marginale genomic DNA from 3 isolates (FL, SI, & VA) was digested into large fragments with restriction enzymes Sfi I and Not I, and separated by CHEF electrophoresis. The gel was stained and photographed (Fig. 19). These fragments are nonoverlapping and have been previously shown to represent the entire 1,250 kbp A. marginale genome (Alleman et al., 1993). The Sfi I and Not I
























Fig. 18 Presence of multiple copies of MSP3. Genomic DNA from a Virginia (VA), South Idaho (SI), or Florida (FL) isolate of A. marginale was digested with the restriction enzymes indicated above the lanes and Southern blotted. EcoR I digested pBluescript MSP3-12 (pB 12) and undigested bovine calf thymus DNA (CT) were used as controls. The blots were hybridized with digoxigenin-labeled pBluescript MSP3-12. Molecular size markers in base pairs (bp) are indicated to the right and left.




















Sph I Nde I Sac I Hinc II EcoR I
bp
VA SI FL VA SI FL VA SI FL VA SI FL pB 12 CT

-- 23,130
bp .
m -- 9,416 s is.. of-- 6,577


,072 --4,361




-- 2,332 2,036 -- -- 2057 1,636




1,018 -564

506 -396 344 -298 -00
O%









87

fragments produced here are identical to those previously reported (Alleman, et al., 1993). Sfi I digestion of the FL isolate produced 12 bands ranging in size from 170 to 14 kbp.

This same gel was then Southern blotted to nylon filters and probed with whole, digoxigenin-labeled MSP3-12 (Fig. 20). The probe hybridized to multiple fragments of Not I and Sfi I digests of all 3 isolates. Most of the gene copies in the FL isolate appear to be contained within 2 large Sfi I fragments with previously reported band sizes of 170 kbp and 137.5 kbp (Alleman et al., 1993). However, the results do indicate the MSP3 gene is widely distributed throughout the genome in all 3 isolates tested. Gene copies were present on 6 of the FL and VA Sfi I digested fragments and 5 of the fragments from the SI isolate (Fig. 20). Not I digestions yielded 8 bands in the FL isolate which hybridized to the MSP3-12 probe, 7 bands in the VA isolate, and 5 bands in the SI isolate (Fig. 20). RFLP is also observed between isolates on CHEF gels.





























Fig. 19 Not I and Sfi I digestion of the A. marginale genome.
Genomic DNA from a Florida (FL), Virginia (VA), or South Idaho (SI) isolate of A. marginale was digested with the restriction enzymes indicated above the lanes, separated by clamped homogeneous electrical field gel electrophoresis, and stained with ethidium bromide. Lanes 1 and 2 contain lambda DNA-HindIII fragments and Promega Delta 39 markers, respectively, as size markers. Molecular size markers in kilobase pairs (kbp) are indicated to the left.




Full Text
91
9.4


Fig. 2 Representation of clone MSP3-12 in the A. margnale genome.
Diagram of clone MSP3-12 indicating the restriction enzyme sites and the resulting
fragments used to map MSP3-12 to the genome. Nucleotide cleavage site for each
restriction enzyme is illustrated diagrammatically (top) and by nucleotide
number(bottom). Size of resulting fragments in base pairs is also provided.


Fig. 18 Presence of multiple copies of MSP3.
Genomic DNA from a Virginia (VA), South Idaho (SI), or Florida (FL) isolate of A.
margnale was digested with the restriction enzymes indicated above the lanes and
Southern blotted. EcoR I digested pBluescript MSP3-12 (pB 12) and undigested
bovine calf thymus DNA (CT) were used as controls. The blots were hybridized with
digoxigenin-labeled pBluescript MSP3-12. Molecular size markers in base pairs (bp)
are indicated to the right and left.


68
Msp319.Corentry:1 Length:4185 October21, 1991 15:01 Type:N Check:2553
1 GGT GATGTACACAGATT AAAGAAGGAAGTAGTTGATAGT AT AGGAAGAGGAACAGCT AGTCCTGT AAGGQCAATGTTTAGT AGAGAGATCTCAGAT GGGA
101 ATACATTACTT GCT GGGGAGAT GGT AGGAGTTGAT GAAGGACT AGT T ATACAAGAACT GAGCAGACCAGAAGAATT AGAAAAGCTACAACAT GAACT AGC
201 AAAGCAAGTAAGTAAATTAGCTGAACTTGGAGAACTTAAGTGGTTAGAAGAACTTGAAAAGCTGGAGACTGAGGAGTTGGGGGAGCTGTTGAGACTGAAA
301 GCGAGAAAAQCCTCACAGGAACT T AGGACGTTGGCGGAAAGT AAAGGACAGCACCT AAATGCT GAT GAGATAGAGGGGGAACTAAAGAGCGCGCT GGGAT
401 TGGAAAGACTGGAAAAGTTGGCAGAGCTGGAGTTGGTCAAGCAGCGGCTTGAGTCGATGAAGGAGCTGGAGAAGAAGATAGAGGAGGCGCAGTTGACAGC
501 AGAAGAGCTCCGGGAAATGAGGGAGAAGCTTAAGGAGTTGGCAGGAAGGGGGGAGGATCCTGCGGGGCTGAAGAAAGCGATAGAGGAGGCAAATGCTGAC
601 GAACTCAAGAAGGGGCTGAAGGAGGTGCTGAAGACGCTGGAGGGAAAGAAGTCAGAGCTGGTGAGGGAAGTAGGAAGGTTGAAGAAGGGAGCCTTGGAGG
701 AGATGAAGGCAGGAAAGGGGTTGGTAGAGGAGGTAAAGGGAGAGCTAGGAAAGCTAGAGQCTAAGAAGATAGAGGAGGTGAAGGCGGCAAAGGGGGAAGT
801 GTCGTTTTTCCCAGGAATAAGAGAACTCACCAAAACTGGAAAGGAGTCGGAGTTGCTGGAAAGGCTTGAGGAGGGGCTGCAGGCGGCTAAGAAGATAAAG
901 AAGGTGGAGGGCCAGGAGAAGCTGTTACAAAATCTCAAGAGCAAGCTAGAAGAACTAGCAGCGATAAGGGAGCTGAAGGAAGGGCCAGAGCTGGTACAGA
1001 TGCTCAAGGAGATAAGACTGGAGCTGGAAGAGCAGCTGGGGATGCTAGCTGAGATTAAGGAAGTTAAGGCACTGGCAGAGAAGCGGCAGGCTGGAGGACT
1101 GGAACTTCAGGAGGTATTACAGCTTACAACTAGAACTGTGGCATTGGAGAGGAAGCTGGTAGAAAAGGTAAGTAAGCTAGATAGGAAGAGCCTGGAAAGT
1201 CAGAAGAAGGTACTGGAGGAGCATAAGAAOGAGCTGGAGCAGGAAAAGCTTAAGCTAACTAAGGAGCTGOCTGAGAAGCGAAAAGCCAAAGGGCTCTTTG
1301 GAGCAGTACTAGACCT GGGACTGGT GGCGCT GGGT GTAGGAGAAACAAAGGACTGGGAGTT AGT GAT ACTGGAGAAAAAACT AGCCAAGAT CAACGCACT
1401 CCTAGAGGGTGGAGATATAGCAAAACTGGGAAGACAGATAAATAGGATCAAGTGGCTAGAGGATCTAGCTGTTAGTAAGAAGCTTAAGGCAGCATTAGCT
1501 AGTGCGATGGAGATAGCTAAGAACAGGGGTTGGACTGATTATTTAAACAGTCTAGATGTAAGTGAGAGAGCTAATGCTGCAAAGGAATTAATTGCTGCTG
1601 AGAAAATTAGAAAATGGGCTAGGGATATTAATAACTTGGATGCTGATGAACGGGCAATGGTCGCTGGGGCCCTAACCCTTTTCCTAACCCCTTCTACAAC
1701 AGTGTGACTCACCACTCTTCCTCGCAACATCGCTCACACTCCTAACCGGCCACTGTACTGAGGGTAATACCAGTTAGTGCAAGGGTGGACGTAGAGGTTT
1801 TGGGAGAGTQGTGATGTTGCTAAAACTTAGATTCTTACTCATTGCTGGAGTATTAGCATTGGGTATCTTACATGGTGTACCTGCAGGGGCTGCTGCTCAG
1901 GCGAGCCCTAGTACTACTGGTACTGAAGCTGATGATACAQCGACTGTTGTTATATGCAATGTTATACGGTTTGTGCAGAAGCTGGGACTACCCATCATGA
2001 CTGGGGTGATACTAGGTTCTAGCATTATGGCTATCTTTGGTAAGCTTGCATGGCCTGCTATTGTAATGCTGGTTGTATTTACAGCTATATTCTTTGGTGC
2101 TGGTAAGCTTATGGCTAAATTTGCTGCTGQGTTGAGTGGTGATGGCATAGGTGATGCCAGTAAGTTTGACTGTTCACAGCATACAGCGGTCAACACCAGC
2201 CAAAGTAACCGATAGAAGATTAGGTGCTGCTGGGCAGTGATGGTGCAGAGAGCTGTGTTGGCATTGGTTQGGAGATGAGQCAGCGGGAGTTCGGAGATAG
2301 GGACTAGGTQCATGGACGATGCAGGTGTGGGCAGAGAGCTGTGTTGGCCTTGGTAGTGAGTTGGGCTGAGGTGCCAAGGQCAGGAGGAGGTACCAGAAGT
2401 TGGGAGATGCTGGTTACCGTTGAAGTGAACCGGTGCAGAGAGCTGTGGCCrTGGGGCCCAGTAGAGTTGGGCTGAGGGGCCAGGGGTAGTGAAGGACACG
2501 ATTTGACTGTTCCAGTTATAAGGAAGTCAAGACCGGCACCAAACCAAGCTAAAGGTTAGGAAATGAAGTGCCGCGTGATGGTGCAGAGAKTGTGGTGGT
2601 AGTACAGGGGTGCTGGTGATAGTTTCTGGTGCTGCAGGGGATCAGAGGACTCCTACTCATGCTGCGGATTATCTCCTTTTGTGTTCCCACAGGTTCACAT
2701 AATTAACCGTACATGGGTAAATGTGCTGGTAGTGCCGGGAACCGCAAGTGGGTAGATGGTGGAGGGGCACCATTGGGGTGGGTGTCCTAGAATCCGAACC
2801 TGATTCCTAGTTCTACCCCTATATTGCGTAGCCCGAAGGATGCTGAGACCTTCTCCTTTGTTTTTCCTAATGGGCTGGCGTCATCGACTQCACGGTGGGC
2901 AGGGGTGTTGTCGTAGTTTGTGCCTAGTGTTTTGCTGGCAGTGCCGCCAATGAATGCTGOCAGGCTGGGTGTGATGCTGTAACTCACACCAGCCTTGACC
3001 TGGCAGGTAAGCTGTGGCTGAAATCGTTGGGCGGTCAGTCCTACAAAGCTTGCACCCAACCCTGCGCAGQCGTAGGGAGACATTCTCCAGTCAACTAGTA
3101 ACCCCACCTGTGGGAAATCATAGCATGCACTCAGGATCGCTGAAACTACCCTGACCCCACTTATTTCCACAATCTCCATGCCCTCTGTGCTTGTGGCAAT
3201 TGCCCGACCAATGACGCCTTTTGCGAGCCGTGAGAGGCCAACAATATCCGTGATAGCTTGATTCCGCTTGCTCTCTTGCGTGTTCTGTGCAGAGGTATCA
3301 GCGCTACCAGCACCGGTTGCAGCAGCACGTTCAGTTTTGGCCTCCTCCCCAGGTATGATQGCATCAAACTTACTTATAAOCTTGTCAGCAAGCGCTTTGC
3401 TTTCCCACGCTTCGTCATGGTCTCTAATATCTTCCAAACGCTTCTTGATGGATCTCAAAOCCGGCTTGGACAACTCCTTCTGCAGCGCGGCCCGAAACCT
3501 TCGACGTCACGCGTCGCACTGGCAGATATTCTTTTCACCAGGGCAAAAGTCGCACTCCCQCCGCTCAGTAAGGCATAATTTTGCGATTTGAAATCATAGT
3601 GCTCATGTCCGAATTCTAACTCAAGCCTGGTGCCGCGTAOCACGCAACCCACGCTTCCCCTGAGGCCCAACAGACTGCTGTTCTCAAACCCCATGGCCGG
3701 ACTCTCCGTAGATTCCCCCQCCCAATCGTAGTTGGCGGAAGAAAGCTCAGCAGAGCCCAACTTCCCAATGTAGGGCAGTATCGCGACGGTTTCTTTTGCA
3801 GCATTAAGCCTAAACCCGCTTATCTTCCCTATGGTGGGCCEGTACCCAAAACTCACGTAAAAATCTTCAOGACCGTTCATCCCATGGGAACTCGCTGGGA
3901 CCAGATAACCAATCGTACTCAGCAATGCAGCAGCTATACATACTACGGGaTCATGACGCACCCAATGACACCGACAACCAACAGGAGGCTAAACGCCCC
4001 TCACTGCACGACAGTGCTATACAACTGTTAAAAAGGTCTTAAAATGAGTTGCATTTTTGAAGGGCTTGCAACTATGCCTAGCTAGAGAATAAGGATTCCG
4101 GGCT GGCGGGCCAAACGGCGCACACT CAGGTT GCCCTTCGTGT GCAT GCAT CAT CAGCGTGGCACTACCTACTTTTTCAGCACCG


79
However, when this same recombinant protein was reacted with
immune sera from an animal infected with a FL isolate, no
reactivity above that which was seen with empty pBluescript
was observed.
Representation of pBluescript MSP3-12 in the
A. margnale Genome
To verify that the cloned pBluescript MSP3-12 faithfully
represented the genomic copy, genomic DNA from a FL isolate of
A. margnale and pBluescript MSP3-12 DNA were digested with
restriction enzymes to yield predicted fragments of specified
lengths. The enzymes used, the restriction sites in
pBluescript MSP3-12, and the predicted size of the fragments
are illustrated (Fig. 16) DNA fragments were Southern
blotted to nylon filters and probed with whole digoxigenin-
labeled MSP3-12. All enzymes used produced pBluescript MSP3-
12 fragments of the predicted size which comigrated with a
fragment in the genomic DNA (Fig. 17). Comigration of large
fragments was seen in digests extending from the 5' terminal
region to the 3' terminal region of the gene.
Presence of Multiple MSP3 Gene Copies
A. margnale genomic DNA from either FL, SI, or VA
isolates was digested with restriction enzymes, Southern
blotted to nylon membranes, and probed with whole,
digoxigenin-labeled MSP3-12. Enzymes Sph I, Nde I, Sac I, and


33
dilution of 1/15,000 (Sigma Immuno Chemicals, St. Louis, MO).
Membranes were processed for enhanced chemiluminescence (ECL)
with detection reagents containing luminol as a substrate (ECL
Western Blotting detection reagents) (Amersham International,
pic, Buckinghamshire, England). The membranes were exposed to
Hyperfilm MP (Amersham International, pic, Buckinghamshire,
England) to visualize bound antibody.
Purification of A margnale Genomic DNA
Florida, South Idaho, and Virginia isolates of A.
margnale genomic DNA used in hybridization studies were
purified by phenol/chloroform extraction and ethanol
precipitation as previously described (Barbet et al., 1987;
Barbet and Allred, 1991). A. margnale genomic DNA from FL,
SI and VA isolates used in CHEF studies was prepared in
agarose plugs as previously described above.
MSP3 Clones
Three clones of the MSP3 gene, MSP3-11, MSP3-12, and
MSP3-19 were cloned, sequenced, and supplied to us courtesy of
T. McGuire, G. Palmer, and T. McElwain (Washington State
University, Pullman). These clones were prepared from a
genomic library composed of mechanically sheared A. margnale
DNA, ligated with Eco RI adaptors, and inserted into
pBluescript SK(-) plasmids (Stratagene, Lajolla, CA). Clones
were identified by screening the library with anti-MSP3 MAbs


87
fragments produced here are identical to those previously
reported (Alleman, et al., 1993). Sfi I digestion of the FL
isolate produced 12 bands ranging in size from 170 to 14 kbp.
This same gel was then Southern blotted to nylon filters
and probed with whole, digoxigenin-labeled MSP3-12 (Fig. 20).
The probe hybridized to multiple fragments of Not I and Sfi I
digests of all 3 isolates. Most of the gene copies in the FL
isolate appear to be contained within 2 large Sfi I fragments
with previously reported band sizes of 170 kbp and 137.5 kbp
(Alleman et al., 1993). However, the results do indicate the
MSP3 gene is widely distributed throughout the genome in all
3 isolates tested. Gene copies were present on 6 of the FL
and VA Sfi I digested fragments and 5 of the fragments from
the SI isolate (Fig. 20) Not I digestions yielded 8 bands in
the FL isolate which hybridized to the MSP3-12 probe, 7 bands
in the VA isolate, and 5 bands in the SI isolate (Fig. 20).
RFLP is also observed between isolates on CHEF gels.


CHAPTER 4
RESULTS
Specificity Experiments
SDS-PAGE separated A. margnale initial body proteins
were transferred to nitrocellulose, and probed with antisera
from animals infected with related rickettsial agents or
protozoal hemoparasites. This was done to determine if
animals infected with these organisms contain antibodies which
cross react with the 86 kDa protein (MSP3) of A. margnale.
As a negative control, normal sera from various species were
reacted with A. margnale proteins. Where available, protein
preparations from related organisms were used in a homologous
reaction with respective antisera to serve as a positive
control.
Sera from a sheep infected with A. ovis (Fig. 3), a horse
infected with E. risticii (Fig. 4), and a dog infected with E.
ewingii (Fig. 4) showed strong reactivity with MSP3 of A.
margnale. These sera as well as sera from animals infected
with E. equi and C. ruminantium showed reactivity against
other A. margnale antigens as well. Sera from animals
infected with A. ovis, C. ruminantium, E. ewingii, and E.
equi, showed reactivity against a 36 kDa antigen, possibly
MSP2 (Figs. 3, 4, & 5). Sera from animals infected with C.
46


24
for potential use as a diagnostic test antigen or vaccine
candidate. Use of a single antigen in a diagnostic test could
markedly increase test specificity over currently available
tests. Since MSP3 was the most immunodominant protein,
further experiments were done to determine its potential as a
diagnostic test antigen. Monoclonal antibodies to MSP3 were
produced and used in a sepharose bead affinity column to
isolate purified MSP3 from FL isolate initial bodies (McGuire
et al., 1991). Affinity purified MSP3 was injected into
rabbits and rabbit-anti-MSP3 immune serum was used on dot
blots to identify epitopes of MSP3 in at least 8 different
geographic isolates of A. marginale (Mcguire et al., 1991).
Using immune sera from infected cattle, purified MSP3
accurately identified long term carriers for up to 5 years PI.
Immune sera from cattle infected with B. bovis, B. bigemina,
or an unidentified rickettsial agent did not cross react with
purified MSP3 (McGuire et al., 1991). However, sera from
animals infected with organism now known to be
phylogenetically related to A. marginale were not tested in
these experiments.
Since MSP3 fulfilled many of the criteria for a good
diagnostic test antigen, attempts were made to clone the MSP3
gene and produce a recombinant protein which might be used in
an ELISA test. A genomic library made from a FL isolate of A.
marginale DNA was screened using a pool of anti-MSP3 MAbs.
Three clones of the MSP3 gene were identified and seguenced.


120
50. Richey, E.J. 1981. Bovine anaplasmosis, p. 767-772. In
R.J. Howard (ed.), Current Veterinary Therapy: Food
Animal Practice. The Sanders Co., Philadelphia, PA.
51. Ristic, M and J. Holland. 1993. Canine Ehrlichiosis,
p.172. In Z. Woldehiwet and M. Ristic (ed.), Rickettsial
and Chlamydias Diseases of Domestic Animals. Pergamon
Press, Inc., Tarrytown, New York.
52. Ristic, M. and J.P. Krier. 1974. Family Anaplasmataceae,
p. 906 In: Buchanan, R.E. and N.E. Gibbons (eds.),
Manual of Determinative Bacteriology. Williams and
Wilkins, Baltimore, MD.
53. Schuntner, C.A. and G. Leatch. 1988. Radioimmunoassay for
Anaplasma margnale antibodies in cattle. Am. J. Vet.
Res. 49 (4) :504-507.
54. Shkap, V., H. Bin, H. Ungar-Waron, and E. Pipano. 1990.
An enzyme-linked immunosorbent assay (ELISA) for the
detection of antibodies to Anaplasma margnale and
Anaplasma cntrale. Vet. Microbiol. 25:45-53.
55. Shkap, V., E. Pipano, T.C. McGuire, and G.H. Palmer.
1991. Identification of immunodominant polypeptides
common between Anaplasma cntrale and Anaplasma
margnale. Vet. Immunol. Immunopathol. 29:31-40.
56. Smith, T. and F.L. Kilborne. 1893. Investigations into
the nature, causation, and prevention of Texas or
southern cattle fever. U.S. Dept. Agr., Bur. Animal Ind.
Bull. 1:1-301.
57. Sparling, P.F., J. Tsai, and C.N. Cornelissen. 1990.
Gonococci are survivors. Scand. J. Infect. Dis, Suppl.
69:125-136.
58. Swift, B.L. and G.M. Thomas. 1983. Bovine anaplasmosis:
Elimination of the carrier state with injectable long-
acting oxytetracycline. J. Am. Vet. Med. Assoc. 183:63-
65.
59. Theiler, A. 1910. Anaplasma margnale. The marginal
points in the blood of cattle suffering from a specific
disease, p. 6-64. In: Theilre, A. (ed.), Report of the
Government Veterinary Bacteriologist 1908-1909 Transvaal
Department of Agriculture, Transvaal, South Africa.
60. Tizard, I., 1992. Serology: The detection and measurement
of antibodies. In: Veterinary Immunology. 4th edition,
W.B. Saunders Co., Philadelphia, PA, p 214-236.


UNIVERSITY OF FLORIDA
3 1262 08554 8179


118
30. Kuttler, K.L. and L.D. Winward. 1984. Serologic
comparisons of 4 Anaplasma isolates as measured by the
complement fixation test. Vet. Microbiol. 9:181-186.
31. Levy, M.G., and M. Ristic. 1980. Babesia bovis:
continuous cultivation in a microaerophilus stationary
phase culture. Science. 207:1218-1220.
32. Love, J.N. 1972. Cryogenic preservation of Anaplasma
margnale with dimethylsulf oxide. Am. J. Vet. Res.
33:2557-2560.
33. Luther, D.G., H.U. Cox, and W.O. Nelson. 1980.
Comparisons of Serotests with calf inoculations of
anaplasmosis-vaccinated cattle. Am.J. Vet. Res. 41:2085-
2086.
34. Maas, J., S.D. Lincoln, M.E. Croan, K.L. Kuttler, J.L.
Zaugg, and D. Stiller. 1986. Epidemiologic aspects of
bovine anaplasmosis in semiarid range conditions of south
central Idaho. Am. J. Vet. Res., 47:528-533.
35. Mahan, S.M., T.C. McGuire, S.M. Semu, M.V. Bowie, F.
Jongejan, F.R. Rurangirwa, and A.F. Barbet. 1994.
Molecular cloning of a gene encoding the immunogenic 21
kDa protein of Cowdria ruminantium. Microbiol., 140:2135-
2142.
36. McCallon, B.R. 1973. Prevalence and economic aspects of
anaplasmosis p. 1-3. In E.W. Jones (ed.), Proceedings
of the Sixth National Anaplasmosis Conference. Heritage
Press, Stillwater, Oklahoma.
37. McGuire, T.C., G.H. Palmer, W.L. Goff, M.I. Johnson, and
W.C. Davis. 1984. Common and isolate restricted antigens
of Anaplasma margnale detected with monoclonal
antibodies. Infect. Immun. 45:697-700.
38. McGuire, T.C., W.C. Davis, A.L. Brassfield, T.F.
McElwain, and G.H. Palmer. 1991. Identification of
Anaplasma margnale long-term carrier cattle by detection
of serum antibody to isolated MSP-3. J. Clin. Microbiol.
29:788-793.
39. Meyer, T.F., C.P. Gibbs, and R. Haas. 1990. Variation and
control of protein expression in Neisseria. Annu. Rev.
Microbiol. 44:451-471.
40. Montenegro-James, S., A.T. Guillen, S.J. Ma, P. Tapang,
A. Abel-Gawad, M. Toro, and M. Ristic. 1990. Use of the
dot enzyme-enzyme linked immunosorbent assay with
isolated Anaplasma marginale initial bodies for


Fig. 4 Specificity experiments using immunoblots.
A. margnale (AM) initial body preparations reacted with
normal serum from a non-infected horse (NHS) or anti-sera
from animals infected with A. margnale (AM), E. rstc
(ER), E. equi (EE), or E. ewingii (EC). Labeling above
each lane indicates the serum used (top) the initial
body preparation used (center), and the dilution of the
serum (bottom). Molecular size standards in kilodaltons
are illustrated on the left.


This document is dedicated to my beautiful wife Mary, my
wonderful children, Arthur and Grace, and to God who we love
and serve.


NBS AM CR
AM AM AM
1/100 1/500 1/100
200
97.4
69
46
30
Bbig
Bbig
1/100
Bbig Bbov Bbov AM
AM Bbov AM AM
1/100 1/100 1/100 1/500
U1
tsj


3
infection, replication of the organism in the vector allowed
easy transmission of the disease with only a few infected
ticks regardless of the initial infecting dose. This firmly
establishes the important role persistently infected carrier
cattle play in the transmission of the disease. In order to
reduce economic losses associated with anaplasmosis, control
efforts must include an effective way of identifying and
decreasing transmission from carrier cattle.
An inexpensive, sensitive, and specific field test for
the identification of A. margnale infected carriers would
have a tremendous impact on limiting the spread and
consequently the economic losses associated with this disease.
Entire herds could be easily tested and identified carriers
removed or treated with oxytetracycline, thus eliminating the
source of infection for susceptible cattle. A test such as
this would also provide an accurate means of identifying
carrier animals being shipped into nonendemic or noninfected
areas. Ideally, animals infected with the less virulent
species, Anaplasma cntrale, should not contain antibodies
that will cross react with the antigen used in this test.
This would be a marked improvement over other serological
tests which cannot distinguish between the two infections. In
addition, in areas where vaccination with A. cntrale is used
as a means of prevention of A. margnale, this test may
distinguish vaccinated from infected animals.


Fig. 6 Size polymorphism of MSP3.
Immunoblots of initial body preparations from a Florida
(FL) a Washington (WA) a South Idaho (SI) and a
Virginia (VA) isolate of A. margnale (AM), reacted with
normal serum from a non-infected cow, Lane 1 (Pre), or
anti-sera from a cow infected with a FL isolate of AM,
Lanes 2-5. Serum was diluted to 1/300. Molecular size
standards in kilodaltons are illustrated on the left.


This dissertation was submitted to the Graduate Faculty
of the College of Veterinary Medicine and to the Graduate
School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Pidioso}
August, 1995
Dean ^College of Veterinary
Medicine
Dean, Graduate School


29
Briefly, infected erythrocytes were centrifuged at 10,000 x g
and the supernate was removed. Packed cells were resuspended
to 2 0 times the volume in 10 mM sodium phosphate. The
solution was centrifuged and the supernate was removed. The
pellets were then resuspended in an egual volume of 10 mM
sodium phosphate.
Antisera used for Immunoblots
Six Holstein calves were infected with blood stabilate
containing a FL isolate of A. margnale. Two other Holstein
calves were each infected with either a VA or a SI isolate.
Antisera from all calves were collected at 50 and 70 days PI.
Antiserum from a cow experimentally infected with a WA
isolate, rabbit-anti-MSP3 polyclonal sera (RB-955), an anti-
MSP3 monoclonal antibody (MAb) (AMG75C2), and an anti-MSP2 MAb
(ANAF19E2) were supplied to us courtesy of Travis McGuire and
Guy Palmer (Washington State University, Pullman). The
reactivities of the MAbs (McGuire et al., 1984; McGuire et
al., 1991; Palmer et al., 1988; Palmer et al., 1994) and the
rabbit-anti-MSP3 polyclonal serum (McGuire et al., 1991) have
been previously described. Sera obtained from calves prior to
infection with A. margnale, and a MAb specific for the
variable surface glycoprotein of Trypanosoma brucei (TRYP1E1),
were used as negative controls in immunoblot experiments.
A. cntrale and A. ovis antisera with indirect
fluorescent antibody (IFA) titers of 1:4,000 were supplied


CHAPTER 2
LITERATURE REVIEW
General Methods of Serologic Diagnosis
Serology is the science of detection of specific
antibodies in body fluids, particularly, though not
exclusively, serum. There are 3 broad categories of serologic
techniques, the primary binding tests, secondary binding
tests, and tertiary binding tests. Primary binding tests
allow antigen and antibody to combine, and the resulting
immune complexes are measured using radioisotopes, fluorescent
dye, or enzyme labels. Examples of the primary binding tests
are radioimmunoassays, immunofluorescence assays, and enzyme-
linked immunosorbent assays. Primary binding tests are the
most sensitive of the serologic techniques in terms of ability
to detect smaller amounts of specific antibodies (Tizard,
1992) .
Radioimmunoassays are widely used primarily because of
their extreme sensitivity and ability to detect small amounts
of antigen or antibody. In these assays radioactive isotopes
such as 123I are used to label antigens or antibodies, and the
level of radioactivity is used for quantitation.
Disadvantages of this test are the dangers and restrictions
5


2
areas of the U.S., the incidence of infected cattle may be as
high as 37% (Maas et al., 1986; McCallon, 1973).
Acute anaplasmosis is usually seen in cattle over 1 year
of age and is characterized by a severe hemolytic anemia,
resulting in weight loss, abortion, decreased milk production,
and often death in infected animals over 3 years of age
(Wanduragala & Ristic, 1993). In acute stages, the disease is
easily diagnosed by finding organisms on routine blood smear
evaluation. However, animals that survive the infection will
remain carriers and maintain a low level of parasitemia which
cannot be detected microscopically (Richey, 1981; Zaugg et
al., 1986). These carrier cattle serve as a perpetual source
of infection for susceptible cattle (Swift & Thomas, 1983).
Cyclic rickettsemia has been detected and quantitated in
carrier cattle using nucleic acid probe hybridization (Eriks
et al., 1993; Kieser et al., 1990). These studies
demonstrated rickettsemia levels in persistently infected
cattle fluctuated at approximately 5 week intervals from a low
of 104 to a high of 107 infected erythrocytes per ml of blood.
Although the level of parasitemia was too low to be detected
microscopically, uninfected Dermacentor andersoni ticks were
able to acquire infection from the cattle at infectivity rates
of up to 80% during the higher rickettsemia levels (107
infected erythrocytes per ml of blood) (Eriks et al., 1993).
Even at extremely low levels of parasitemia 27% of the male
ticks became infected. In addition, once ticks acquire the


4
The purpose of this study was to evaluate the MSP3
protein of A. margnale, and determine if a recombinant form
of this protein would be a suitable diagnostic test antigen to
detect A. margnale infection in carrier cattle.


CHAPTER 5
DISCUSSION
Rational for Studying the MSP3 Antigen
The current serologic tests available for the diagnosis
of A. margnale infection are based on antigens which are a
crude mixture of A. margnale and erythrocyte proteins. This
reduces both the sensitivity and specificity of various tests
to unacceptable levels, particularly in detecting carrier
cattle (Goff et al., 1990; Luther et al., 1980; Maas et al.,
1986; Todorovic et al., 1977). Because of this, and because
A. margnale cannot be successfully maintained in erythrocyte
culture, efforts using molecular techniques have been
attempted to identify an immunodominant protein with
acceptable sensitivity and specificity to be used in
recombinant form as purified test antigen. The ideal test
antigen would be one which is not cross reactive with antigens
from related organisms, and could detect acutely infected
animals as well as carrier cattle infected with any of several
different geographic strains of A. margnale. This antigen
should preferably be conserved in all isolates of A. margnale
and be from a single copy gene to limit genetic recombination
and antigenic variation.
92


89
S/7 I Not I
1 2 FL VA SI FL VA SI
kbp
i W I
195
156
117
9
*
__
78
* *
if?
39
23.1
n
9.4 --
m
6.6 --
o
4.4 --
to


Antiqen
FL
MSP3-11
MSP3-11
MSP3-12
MSP3-12
MSP3-19
MSP3-19
pel Lie
pel ue
Antibody
AMG
Trp
AMG
Trp
AMG
Trp
AMG
Trp
AMG
75C2
1E1
75C2
1E1
75C2
1E1
75C2
1E1
75C2
M.U.
Id)
200.0 --
97.4
46.0 --
30.0
21.5 --
14.3
O)


16
when testing a large number of A. margnale infected cattle,
normal cattle, and cattle infected with B. bigemina, B. bovis,
and Theileria orientalis. This test utilizes a crude A.
margnale antigen preparation isolated from infected
erythrocytes. Reactants are identified using 125I-labeled,
anti-bovine IgG, rabbit immunoglobulin and an automated gamma
counter. As expected, sensitivity with this primary binding
assay is high; however, substantial numbers of false positives
(up to 37%) occurred unless sera was pre-absorbed with normal
bovine erythrocytes and sonicated B. bovis antigen. Controls
for cross reactivity using pre-absorption should not be
necessary if purified antigens are used, even with tests as
sensitive as RIAs.
Even though the RIA is highly sensitive and specific, it
is too labor intensive to be used as a practical field test
where large numbers of samples need to be processed. In
addition, the use of radioactive material and the need for
specialized equipment limits its use to reference or research
laboratories.
Enzyme-Linked Immunosorbent Assays (ELISA)
Several ELISAs for measuring antibody to A. margnale
have been developed over recent years (Barry et al., 1986;
Duzgun et al., 1988; Montenegro-James et al., 1990; Nakamura
et al., 1988; Trueblood et al., 1991; Winkler et al., 1987).
Because these are primary binding assays, the sensitivity of


42
according to the manufacturer's recommendations (The Genius
System Luminescent Detection Kit, Boehringer Mannheim,
Indianapolis, IN) The membranes were exposed to Hyperfilm -
MP (Amersham International, pic, Buckinghamshire, England) to
visualize bound antibody. The molecular sizes of comigrating
cloned and genomic fragments were determined by comparison to
Lambda DNA/Hind III fragments (GIBCO BRL, Gaithersburg, MD)
and 1 kbp DNA Ladder (GIBCO BRL, Gaithersburg, MD) molecular
size standards.
Presence of Multiple MSP3 Gene Copies
A. margnale genomic DNA from either the FL, VA, or SI
isolate was extracted as described above and aliquots of 1 /g
of DNA were digested with Hie II (New England Biolabs,
Beverly, MA) Sac 1 (Boehringer Mannheim Corp., Indianapolis,
IN), Nde I (Boehringer Mannheim Corp., Indianapolis, IN), or
Sph 1 (Boehringer Mannheim Corp., Indianapolis, IN). These
enzymes were chosen because they do not cut within the known
sequence of MSP3-12 gene. Calf thymus DNA was digested
identically as a control. Empty pBluescript and pBluescript
MSP3-12 DNA (0.1 ig) were digested with Eco RI and used as
negative and positive controls, respectively for probe
hybridization. Digested fragments were separated by gel
electrophoresis with 1% agarose gels containing 0.1 /g
ethidium bromide and photographed. Southern blotting was
performed under prehybridization and hybridization conditions


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
EVALUATION OF ANAPLASMA MARGINALE MAJOR SURFACE
PROTEIN 3 (MSP3) AS A DIAGNOSTIC TEST ANTIGEN
By
Arthur Rick Allemn
August 1995
Chairperson: John Dame, PhD
Major Department: Veterinary Medicine
The immunodominant surface protein, MSP3, has been
proposed as an antigen suitable for the diagnosis of bovine
anaplasmosis. In this study we further characterized MSP3 to
examine its potential as a test antigen for the serological
detection of carrier cattle. The specificity of MSP3 was
evaluated by probing immunoblots of A. margnale proteins with
immune sera from animals infected with related organisms.
Similarly, we used polyacrylamide gel electrophoresis (SDS-
PAGE) and immunoblots to evaluate the conservation of MSP3
between 4 different geographic isolates of A. margnale. In
addition, proteins from a FL isolate were separated by 2-
dimensional gel electrophoresis, and immunoblotted with immune
sera from cattle infected with one of 4 different geographic
isolates of A. margnale. Genomic A. margnale DNA was
digested with restriction endonucleases, transferred to nylon
xi


74
different direction. A similar, but much smaller area is seen
downstream to the open reading frame of MSP3-19 (Fig. 12).
MAb and Immune Sera Reactivity to Recombinant MSP3
Lysates from E. coli cells transformed with pBluescript
containing each of the 3 MSP3 genes were separated by SDS-
PAGE, transferred to nitrocellulose and reacted with anti-MSP3
MAb (AMG75C2) or immune sera from cattle infected with a VA or
FL strain of A. margnale. This was done to verify that a
cloned MSP3 gene represented a gene which produced one or more
of the 86 kDa antigens seen on 2-D immunoblots. Initial
bodies from a FL strain of A. margnale were used as a
positive control for MAb (AMG75C2) and immune cattle sera.
Lysate from E. coli transformed with empty pBluescript, anti-
Trypanosome brucei-KAb (TYRP1E1), and pre-immune cattle sera
were used as negative controls.
MAb AMG75C2 reacts only with recombinant protein
expressed by clone MSP3-12 (Fig. 14) The recombinant MSP3
has a molecular size of approximately 65.0 kDa. This
corresponds to the calculated size of the protein expressed by
the open reading frame of MSP3-12 which is 1,707 bp (Fig. 10).
No reactivity was observed between MAb AMG75C2 and recombinant
MSP3-11, MSP3-19, or empty pBluescript. Isotype control MAb
TRYP1E1 showed no reactivity as well.
Immune serum from a cow infected with a VA isolate of A.
margnale reacted with recombinant MSP3-12 (Fig. 15).


116
10. Barbour, A.G. 1991. Molecular biology of antigenic
variation in Lyme borreliosis and relapsing fever: a
comparative analysis. Scand. J. Infect. Dis Suppl.
77:88-93.
11. Barry, D.N., R.J. Parker, A.J. De Vos, P. Dunster, and
B.J. Rodwell. 1986. A microplate enzyme-linked
immunosorbent assay for measuring antibody to Anaplasma
margnale in cattle serum. Aust. Vet. J., 63:76-79.
12. Beutler, E. 1984. The preparation of red cells for assay.
Red Cell Metabolism: A Manual of Biochemical Methods. 3rd
ed. pp 8-19. Edited by Ernest Beutler, MD. Orlando,
Florida: Grue and Stratton, Inc.
13. Dame, J.B., S.M. Mahan, and C.A. Yowell. 1992.
Phylogenetic relationship of Cowdria ruminantium, agent
of heartwater, to Anaplasma margnale and other members
of the order Rickettsiales determined on the basis of 16S
rRNA sequence. Int. J. of Syst. Bacteriol., 42:270-274.
14. Duzgun, A., C.A. Schuntner, I.G. Wright, G. Leatch, and
D.J. Waltishbuhl. 1988. A sensitive ELISA technique for
the diagnosis of Anaplasma margnale infections. Vet.
Parasitol., 29:1-7.
15. Eriks, I.S., G.H. Palmer, T.C. McGuire, D.R. Allred, and
A.F. Barbet. 1989. Detection and quantitation of
Anaplasma margnale in carrier cattle by using a nucleic
acid probe. J. Clin. Microbiol. 27:279-284.
16. Eriks, I.S., D. Stiller, and G.H. Palmer. 1993. Impact of
persistent Anaplasma margnale rickettsemia on tick
infection and transmission. J. Clin. Microbiol. 31:2091-
2096.
17. Goff, W.L., A.F. Barbet, D. Stiller, G.H. Palmer, D.P.
Knowles, K.M. Kocan, J.R. Gorham, and T.C. McGuire. 1988.
Detection of Anaplasma margnale tick vectors by using a
cloned DNA probe. Proc. Natl. Acad. Sci. 85:919-923.
18. Goff, W.L., D. Stiller, R.A. Roeder, L.W. Johnson, D.
Falk, J.R. Gorham, and T.C. McGuire. 1990. Comparison of
a DNA probe, complement fixation, and indirect
immunofluorescence tests for diagnosing Anaplasma
margnale in suspected carrier cattle. Vet. Microbiol.
24:381-390.
19. Goff, W.L., and L.D. Winward. 1985. Detection of
geographic isolates of Anaplasma margnale, using bovine
polyclonal anti-sera and microfluorometry. Am. J. Vet.
Res. 46:2399-2403.


30
courtesy of Susan Oberle (The Salk Institute, San Diego, CA).
Hyperimmune sera from cattle infected experimentally with B.
bovis and B. bigemina were supplied by David Allred
(University of Florida, Gainesville). Sera from cattle
experimentally infected with Cowdria ruminantium were supplied
by Michael Bowie (University of Florida, Gainesville). Equine
sera from animals infected with Ehrlichia equi and Ehrlichia
risticii had I FA titers of 1/1,600, and were obtained from
Ibulaimu Kakoma (University of Illinois, Urbana). Serum from
a dog with canine ehrlichiosis was obtained from Rose Raskin
(University of Florida, Gainesville). This dog was infected
with E. ewingii, but serum from this animal had an IFA titer
of 1/160 for Ehrlichia canis, and 1/64 for E. chaffeensis.
SDS-PAGE
Initial body preparations containing 6.0 10.0 /xg of
protein were solubilized in one half their volume of a 3x
sample buffer containing 0.1 M Tris pH 6.8, 5% SDS (w/v), 50%
glycerol, 7.5% /3-mercaptoethanol, and 0.00125% bromophenol
blue, and heat denatured at 100C for 3 min. Proteins were
electrophoresed on 7.5% to 17.5% (w/v) gradient polyacrylamide
gels. Gels were fixed in 25 mM Tris, 191.8 mM Glycine and 20%
Methanol, and electrophoretically transferred to
nitrocellulose (Hybond ECL, Amersham International pic,
Buckinghamshire, England).


7
producing a colored product. The color change may be
estimated visually or determined spectrophotometrically.
There are several variations of the ELISA test; the
direct ELISA which utilizes enzyme-linked immunoglobulins, the
indirect ELISA which uses enzyme-linked anti-globulins, and
the competitive ELISA which employs the use of a labeled
monoclonal antibody to compete with specific antibodies in the
test sample for a single epitope on an antigen molecule. Some
ELISAs bind antibody to a solid phase media to capture a
particular antigen (antigen capture ELISA), while other
techniques bind antigen to the solid phase in order to capture
and detect antibody (antibody capture ELISA) (Harlow & Lane,
1988). Various solid phase media may be used such as
nitrocellulose membranes or polystyrene microtiter plates,
depending on the purpose of the test and the nature of the
material to be tested. The extreme versatility of this
technique, its excellent sensitivity (comparable to RIA), and
its simplicity make it one of the most widely used
immunodiagnostic techniques for many viral, bacterial, and
parasitic infections. In addition, it does not involve the
use of hazardous radioactive isotopes. Because of its extreme
sensitivity, specificity may be a problem with these tests,
particularly if unpurified test antigens are used to detect
specific antibodies in polyclonal sera.


23
geographic isolates. An antigen which shows promise for
meeting the above requirements is the 86kDa surface protein of
A. margnale, MSP3.
In previous experiments, the most antigenic surface
proteins of A. margnale were identified by
immunoprecipitation of radiolabeled initial body proteins from
a Florida (FL) isolate of A. margnale with immune sera from
infected cattle (Palmer et al.f 1986). The FL isolate of A.
margnale was used for antigen isolation because it has been
found by adsorption studies to contain antigens common to both
morphologic types of A. margnale, the tailed and non-tailed
forms (Goff and Winward, 1985; Kreier and Ristic, 1963). An
86kDa protein, MSP3, was identified as the most immunodominant
in all stages of infection from as early as 30 days post
infection (PI) to 255 days PI. Similar reactivity was
observed when initial body preparations were
immunoprecipitated with immune sera from cattle infected with
either of 3 different isolates of A. margnale, FL, Virginia
(VA) and Texas (TX) isolates (Palmer et al., 1986). This
suggested MSP3 from a FL strain was conserved enough to be
recognized by immune sera from animals infected with different
isolates. However, the conservation of MSP3 was not firmly
established since these studies were performed using single
dimensional gel electrophoresis.
These experiments identified for the first time antigenic
surface proteins of A. margnale which could be investigated


56
various isolates clearly demonstrates variation in size of the
MSP3 proteins. The 86 kDa MSP3 antigen is seen in the FL
isolate (Fig. 6) An antigen of similar size is seen when
this same antiserum is reacted with proteins from a VA isolate
(Fig. 6) However, in reactions with the SI isolate, an
antigen of slightly smaller molecular mass is seen, and the WA
isolate has an antigen with a molecular mass greater than 86
kDa (Fig. 6).
Immune Response to MSP3
Realizing MSP3 is not conserved between different
geographic isolates, we then investigated the possibility that
animals infected with different isolates may contain immune
sera that varies in reactivity to MSP3 from a single isolate.
This is an important consideration in attempting to develop a
diagnostic test antigen derived from a single isolate.
Initial body preparations from a FL isolate were separated by
2-D gel electrophoresis as described above in order to
determine if co-migration of antigens of similar molecular
size occurs. Electrophoretically separated proteins were
transferred to nitrocellulose and probed with antisera from
cattle infected with a FL, VA, SI, or WA isolate of A.
margnale as well as an anti-MSP3 MAb (AMG75C2), rabbit-anti-
MSP3 polyclonal serum, and an anti-MSP2 MAb (ANF19E2). Pre
infection bovine sera, normal rabbit sera, and a MAb to a


Fig. 11 Gene sequence of MSP3-19.
The nucleotide sequence of pBluescript MSP3-19 is
illustrated. Numbers on the left indicate the number of
the first nucleotide in each row. The termination codon
is underlined.


LIST OF FIGURES
Figure page
1 Diagram of MSP3 clones 35
2 Representation of clone MSP3-12 in the A. margnale
genome 40
3 Specificity experiments using immunoblots .... 47
4 Specificity experiments using immunoblots .... 49
5 Specificity experiments using immunoblots .... 51
6 Size polymorphism of MSP3 54
7 2-D gel electrophoresis of A. margnale proteins 58
8 2-D gel electrophoresis of A. margnale proteins 60
9 Gene sequence of MSP3-11 63
10 Gene sequence of MSP3-12 65
11 Gene sequence of MSP3-19 67
12 Diagram of MSP3 clones with MSP2 homology .... 70
13 Comparison of MSP3 and MSP2 protein sequences 72
14 Immunoblots of expressed MSP3 clones 75
15 Immunoblots of expressed MSP3-12 clones 77
16 Representation of clone MSP3-12 in the A. margnale
genome 8 0
17 Genomic representation of MSP3-12 82
18 Presence of multiple copies of MSP3 85
19 Not 1 and Sf I digestion of the A. margnale
genome 88
20 Distribution of MSP3 in the A. margnale genome 90
vii


119
serodiagnosis of anaplasmosis in cattle. Am. J. Vet. Res.
51(10):1518-1521.
41. Nakamura, Y., S. Shimizu, T. Minami, and S. Ito. 1988.
Enzyme-linked immunosorbent assay using solubilized
antigen for detection of antibody to Anaplasma margnale.
Trop. Anim. Hlth. Prod. 20:259-266.
42. Ndung'u, L.W., C. Aguirre, F.R. Rurangirwa, T.F.
McElwain, T.C. McGuire, D.P. Knowles, and G.H. Palmer.
1995. Detection of Anaplasma ovis infection in goats by
major surface protein 5 competitive inhibition enzyme-
linked immunosorbent assay. J. Clin. Microbiol. 33:675-
679.
43. Oaks, E.V., C.K. Stover, and R.M. Rice. 1987. Molecular
cloning and expression of Rickettsia tsutsugamishi genes
for two major protein antigens in E. coli. Infect. Immun.
55:2428-2435.
44. Oberle, S.M., G.H. Palmer, A.F. Barbet, and T.C. McGuire.
1988. Molecular size variations in an immunoprotective
protein complex among isolates of A. marginale, Infect.
Immun. 56:1567-1573.
45. Palmer, G.H., and T.C. McGuire. 1984. Immune serum
against Anaplasma marginale initial bodies neutralizes
infectivity for cattle. J. Immunol. 133:1010-1015.
46. Palmer, G.H., A.F. Barbet, K.L. Kuttler, and T.C.
McGuire. 1986. Detection of Anaplasma marginale common
surface proteins present in all stages of infection. J.
Clin. Microbiol. 23:1078-1083.
47. Palmer, G.H., A.F. Barbet, A.J. Musoke, J.M. Katende, F.
Rurangirwa, V. Shkap, E. Pipano, W.C. Davis, and T.C.
McGuire. 1988a. Recognition of conserved surface protein
epitopes on Anaplasma cntrale and Anaplasma marginale
isolates from Israel, Kenya, and the United States. Int.
J. Parasitol. 18:33-38.
48. Palmer, G.H., S.M. Oberle, A.F. barbet, W.L. Goff, W.C.
Davis, and T.C. McGuire. 1988b. Immunization of cattle
with a 36-kilodalton surface protein induces protection
against homologous and heterologous Anaplasma marginale
challenge. Infect. Immun. 56:1526-1531.
49. Palmer, G.H., G. Eid, A.F. Barbet, T.C. McGuire, and T.F.
McElwain. 1994. The immunoprotective Anaplasma marginale
major surface protein 2 is encoded by a polymorphic
multigene family. Infect. Immun. 62:3808-3816.


94
rickettsial agent isolated from an aborted calf fetus (McGuire
et al., 1991). However, using 16s rRNA sequencing analysis,
A. margnale has been shown to phylogenetically be more
closely related to C. ruminantium and the Ehrlichia sp.,
particularly E. risticii, E. equi, and E. canis (Dame et al.,
1992; Van Vliet et al., 1992).
Initial experiments suggested FL MSP3 was conserved
enough to be recognized by sera from animals infected with
different isolates (Palmer et al., 1986). However, since this
study used single dimension gel electrophoresis, reactivity of
immune sera with different proteins of comigrating molecular
size could not be ruled out.
The specificity of MSP3 in detecting A. marginale
infected carriers, the conservation of MSP3 between various
strains of A. marginale, the reactivity of sera from animals
infected with different strains of A. marginale to MSP3 from
a FL isolate, and the gene(s) which encode the MSP3 antigen
needed to be investigated in order to determine if a
recombinant form would make a suitable diagnostic test
antigen.
The Specificity of MSP3 in Detecting A. marginale Infection
Antigenic similarities between Anaplasma sp. ,
particularly A. marginale and A. cntrale have previously been
identified (Palmer et al., 1988a; Shkap et al., 1990; Shkap et
al., 1991). In our study, A. cntrale antiserum reacted with


14
hybridization and 60 of the 65 sera tested positive by IIF,
while only 5 of the 65 tested positive by CF.
The specificity of the CF test in detecting non-infected
cattle is very high, up to 100% (Gonzalez et al., 1978). This
is not surprising given the low test sensitivity. However,
the study again looked at specificity involving the detection
of CF antibodies in normal, non-infected cattle, not cattle
which may be infected with other closely related
hemoparasites. Therefore, cross reactivity with cattle
infected with other rickettsial or hematoprotozoal agents may
occur. In addition, as with the CA test, the CF test could
not distinguish between infected animals and animals
vaccinated with the killed A. margnale vaccine (Luther et
al., 1980).
The CF test is a tedious test to perform and reguires
considerable technical skill and knowledge. Therefore, the
complexity of the test and its poor sensitivity, particularly
in detecting infected carriers makes it a poor choice for the
field identification of infected animals.
The Indirect Immunofluorescence Test (IIF)
The indirect immunofluorescence test (IIF), being a
primary binding assay, is much more sensitive than the CA or
CF tests. This test uses A. margnale infected erythrocytes
fixed to a microscope slide and permits reaction to them using
a patient's serum. Antibodies to A. margnale on infected


38
digested identically for size comparison of digested plasmids.
The 2.3 kbp insert of the MSP3-12 gene, which was previously
inserted using Eco RI adaptors, was separated from plasmid DNA
by electrophoresis on a 1% agarose gel in Tris Borate EDTA
buffer (TBE) (45 mM Tris, 45 mM Boric acid, 1 mM EDTA). The
gel was stained with 0.5 ¡j.g/ml ethidium bromide for 20 min.
and photographed. The 2.3 kbp band representing clone MSP3-12
was cut from the gel and DNA was extracted from the agarose
plug by ion exchange chromatography using QIAquick Gel
Extraction Kit (Qiagen Inc., Chatsworth, CA).
A probe was made by random prime labeling 200 ng of MSP3-
12 DNA with digoxigenin using the Genius System Nonradioactive
DNA Labeling Kit according to the manufacturer's instructions
(Boehringer Mannheim Corp., Indianapolis, IN).
Representation of pBluescript MSP3-12 in
the A. margnale Genome
Restriction Enzyme Digestion
To verify the cloned pBluescript MSP3 was an accurate
representation of genomic MSP3, multiple restriction sites of
pBluescript MSP3-12 and genomic A. margnale DNA from a FL
isolate were compared using restriction enzymes which cut
within the MSP3 gene. Restriction enzymes Neo I (Boehringer
Mannheim Corp., Indianapolis, IN), Bsp M (New England BioLabs,
Beverly, MA) and Eae 1 (New England BioLabs, Beverly, MA) ,
were chosen because they produced large fragments in different


Ill
other isolates. However, judging from the intensities of the
bands, it appears most of the copies in the FL isolate are
located on two large Sfi I fragments approximately 160 and 130
kbp. The smaller band has been shown to contain a doublet of
comigrating fragments (Alleman, et al., 1993). Therefore,
these two hybridizing bands could represent as much as 25% to
37% of the genome size. Because this is such a large area of
the genome, conclusions regarding the proximity of the genes
are difficult to make. The Not I digest of the FL isolate,
and the Not I and Sfi I digests of the other isolates
indicates a more even distribution of copy numbers throughout
the genome.
This suggests MSP3 copies are widely distributed
throughout the A. marginale genome, similar to the pattern
seen with the MSP2 multigene family (Palmer et al., 1994).
These copies are likely not the result of simple duplication
of a single MSP3 gene since the copies are not present in
tandem along a single stretch of DNA. This assumption is
supported by the partial gene sequences available for 3 of the
MSP3 genes. This would indicate that any coordinated
regulation of MSP3 copies would involve trans-regulation.
Summary and Conclusions
We have determined that the MSP3 antigen of A. marginale
is of questionable specificity as a diagnostic test antigen
with potential for cross reactivity with sera from animals


113
produced by the expressed copies could then be evaluated.
Alternatively, conserved epitopes on these genes could be
identified and recombinant or synthetic peptides derived from
gene seguences could be tested with immune cattle sera to
determine their reactivity. However, considering the
apparent ability of A. margnale to antigenically vary this
protein, we feel these attempts may not be practical,
particularly since another surface antigen (MSP5) has shown
some promise for use as a diagnostic test antigen (Ndung'u et
al., 1995). This 19 kDa protein is encoded by a single copy
gene and appears to be conserved between all recognized
Anaplasma species (Visser et al., 1992).
Even though MSP3 may not be an ideal test antigen, the
results of these experiments may provide valuable information
regarding the use of this antigen in a subunit vaccine. It is
not known if response to any of the MSP3 antigens provides
protective immunity to cattle. However, the demonstration of
multiple 86 kDa proteins, and antigenic variation between
these proteins, would indicate that multiple expressed copies
of MSP3 may need to be evaluated for potential use in vaccine
trials. The need for the organism to antigenically vary this
surface protein suggests it serves an important function for
survival in the host. Immune response to an area conserved
between all expressed copies may prove beneficial in
neutralizing infectivity.


73
BestFit Allignment of MSP3-12 (top)
and MSP2 (bottom)
55 GSFYIGLDYNPTFNGIKDLKIIGETDEDE 83
I I I I I 11 11 I I III I- I
50 GSFYIGLDYSPAFGSIKDFKV.QEAGGTT 77
84 MDVLTGARGLFPMNALASNVTDFNSYHFD 112
78 RGVFPYKRDAAGRVDFKVHNFD 99
113 WSTPLPGLEFGNSTL ALGGSIGYRIGGA 140
100 WSAPEPKISFKDSMLTALEGSIGYSIGGA 128
141 RVEVGIGHERFVIKGGDDAA FLL 163
129 RVEVEVGYERFVIKGGKKSNEDTASVFLL 157
164 GRELALDTARGQL 176
158 GKELAYHTARGQV 170


MSP3 CLONES
MSP3-12
Start
MSP3-11
1306 1472 1818 2008 2090 2280 Stop
u>


124
at the molecular level. He plans to pursue a career in
academia where he will continue his laboratory investigation
of rickettsial agents as well as provide instructional
services to students and clinical services to the teaching
hospital.


13
The Complement Fixation Test (CF)
The CF is another widely used test for the detection of
antibodies to A. margnale in infected cattle. This test
relies on the ability of A. margnale antibodies in the serum
to fix complement and lyse target cells on which the antigen
is associated (Kuttler & Winward, 1984) The antigen used in
this test is again a crude mixture of A. margnale and
erythrocyte proteins, and since it is a secondary binding
test, sensitivity is not as high as with primary binding
assays. Sensitivity of this assay is reported to be low,
varying from 10% to 79% (Gonzalez et al., 1978; Goff et al.,
1990) In addition, it has been shown that CF antibodies
decline rapidly after natural infections with A. margnale
(Todorovic et al., 1977). CF antibodies decreased to very low
levels (1:10) as soon as 10 weeks post-infection (PI), and
below the sensitivity level of the test (1:5) by 14 weeks PI
(Gonzalez et al., 1978) This may be because CF relies on the
presence of IgM antibodies, which are better able to fix
complement than IgG. IgM levels are elevated in acute
infections, but may decline after acute episodes subside.
These results would indicate this test is unreliable in
detecting carrier cattle. In another study, 65 cattle from a
known A. margnale infected herd were tested by CF and DNA
probe hybridization (Goff et al., 1990). Stained blood smears
prepared from these cattle detected no organisms. However, 64
of the 65 cattle blood samples tested positive by DNA probe


Fig. 15 Immunoblots of expressed MSP3-12 clones.
The reactivity of pre and post-infection sera from animals infected with a Florida
isolate of A. margnale (Pre and Post-FL) or a Virginia isolate (Pre and Post-VA)
are compared using expressed proteins from clones MSP3-12 and empty pBluescript
plasmid (pBlue). Antigens are listed on top and anti-sera is listed on the bottom.
In lane 1, initial body preparation from a FL isolate of A. margnale (FL) is
reacted with immune sera from an animal infected with a FL isolate as a positive
control. Numbers on the left indicate molecular size markers in kilodaltons.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EXMLKBH7W_ZZWYSK INGEST_TIME 2015-03-25T19:12:11Z PACKAGE AA00029762_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


114
The information presented in this study may also be
valuable in studying antigenic variation of A. margnale in
persistently infected, carrier cattle. Copy-specific epitopes
on an MSP3 molecule could be defined, and variations in these
epitopes could be monitored in cyclic rickettsemias of carrier
cattle. This would provide much needed information regarding
the mechanisms by which this organism, and possibly other
rickettsial agents, evade the host immune system. Basic
information regarding the means by which organisms adapt to
their host helps establish better ways to diagnose, prevent,
and eventually eradicate these diseases.


20
allow for eventual visual identification as would be needed
for rapid field diagnosis.
Direct visualization of positive reactants was
accomplished using a Dot-ELISA (Montenegro-James et al.,
1990) In this test whole initial body preparations were
solubilized in SDS and dotted on to nitrocellulose disks.
Test sera were reacted with the antigen and antigen/antibody
complexes were visualized with alkaline phosphatase-conjugated
protein A. Test specificity was good (95%) and cross
reactivity to Babesia or Trypanosome vivax was not observed.
However, other rickettsial agents or related hemoparasites
were not tested. Test specificity was increased by using
protein A-conjugated alkaline phosphatase versus an anti-
globulin-conjugated molecule. This reduced nonspecific
binding of antibodies to the nitrocellulose disk. However,
the use of whole initial bodies isolated from infected
erythrocytes still lends itself to false positive reactions.
Test sensitivity with the Dot-ELISA was fair (92.9%) with
19 false negatives out of 269 true positives (Montenegro-James
et al., 1990). In addition, no effort was made to determine
the ability of this test to detect chronically infected, long
term carrier cattle. A good diagnostic test for
identification of infected cattle for importation or herd
management should have a sensitivity of at least 95% or
higher. Therefore, while this test is easy and convenient to


> > ^ N S ^
^cV ^ ^ < Restriction Cleavage site in Size of co-migrating
Enzyme pBIuescript MSP3-12 plasmid and genomic
fragments
Neo I 12/1,188 1,176
Eae I 52/1,'762 1,710
BspM 193/2,076 1,886
oo


44
solution of 100 units of restriction enzyme in lx restriction
enzyme buffer was added and DNA digestions were performed
according to the manufacturer's instructions. Reactions were
stopped by the addition of 0.25 total reaction volume of 0.5
M EDTA, pH 8.0, 0.1% N-lauroylsarcosine (w/v), 1 mg/ml
proteinase K (w/v), incubating reaction mixture at 4C for 1
hour, then 37C for 15 min. Uncut M. bovis chromosomal DNA
(Promega Corp., Madison, WI) was digested identically to serve
as a control for restriction endonuclease activity.
CHEF Gel Electrophoresis
This was done on the CHEF DRII system (Bio-Rad
Laboratories, Richmond, CA) Plugs of digested DNA were
electrophoresed in 1% agarose gels in 0.5x TBE buffer at 14C.
Electrophoretic conditions were set at 180 V, 10 sec. switch
rate, with a 16 hour run time. Delta 39 Lambda Ladders
(Promega Corp., Madison, WI) and Lambda DNA/Hind III Fragments
(GIBCO BRL, Gaithersburg, MD) were used as size standards.
The gels were stained for 30 min. in 0.5 ig/ml ethidium
bromide and photographed. Southern blotting and hybridization
of the separated bands were performed as previously described
except that the gel was depurinated in 0.25 M HC1 for 15 min.
prior to washing and transfer to nylon membranes.
Prehybridization and hybridization conditions were performed
identically to those described above. The membranes were
probed with digoxigenin-labeled pBluescript MSP3-12, and bound


53
ruminantium (Fig. 5) and E. equi (Fig. 4) reacted with an
antigen of slightly smaller molecular weight than MSP3. Sera
from animals infected with A. cntrale (Fig. 3) or either
Babesia sp. (Fig. 5) failed to react with any A. margnale
antigens although strong reactivity was demonstrated in
reactions involving homologous preparations.
Conservation of MSP3 Between Different
Geographic Isolates of A. marginale
The conservation of MSP3 between different geographic
isolates of A. marginale was evaluated by SDS-PAGE and
immunoblots using initial body preparations from FL, VA, SI,
and WA isolates of A. marginale. Preparations were separated
on 7.5% to 17.5% (w/v) gradient polyacrylamide gels as
described above.
They
were
then transferred
to
nitrocellulose, and
reacted
with
varying dilutions
of
antiserum from an animal experimentally infected with a FL
isolate and a MAb to MSP3, AMG75C2 (McGuire et al., 1991), for
definitive identification of the MSP3 antigen. An optimal
dilution of this antiserum was established for demonstration
of the immunodominant MSP3 protein (Data not shown). Pre
infection bovine sera and a MAb to a trypanosome surface
protein were used as negative controls and showed no
reactivity to MSP3.
Initial body preparations from these same isolates were
then probed with antiserum at a single dilution previously
established (Fig. 6). The side by side location of the


Fig. 10 Gene sequence of MSP3-12.
The nucleotide sequence of pBluescript MSP3-12 is
illustrated. Numbers on the left indicate the number of
the first nucleotide in each row. Start codon is
underlined.


102
kDa antigens are likely the result of either post-
translational modification of a protein from a single copy
gene, or transcription and translation of several closely
related genes from a multigene family.
Previous studies have shown multigene families encoding
major surface proteins of A. margnale. Between 7 to 10
similar gene copies encode the MSP2 gene of A. margnale
(Palmer et al., 1994). MSP2 is a 36 kDa protein, and our
results illustrate multiple MSP2 antigens using 2-D gel
electrophoresis and immunoblots with an anti-MSP2 MAb
(ANAF19E2). These results support the findings of the MSP2
multigene family, and suggest that transcription and
translation of this family produces variations in the MSP2
protein. In addition, the MSP1/3 subunit of the MSP1 gene of
A. margnale also is encoded by a partially homologous
multigene family (Viseshakul et al., 1994). Genes in this
family were shown to differ from each other by extensive
deletions, insertions and rearrangements of sequences
(Viseshakul et al., 1994).
These previous experiments, along with data presented in
this study, prompted us to investigate the possibility that
the MSP3 antigens could be encoded by a multigene family. If
this is true, it would further demonstrate the ability of the
organism to use multigene families to vary important
immunogenic surface proteins. In this case, immunogenic
epitopes conserved in all strains of A. margnale may need to


95
several A. cntrale proteins, however, this same antiserum
showed no reactivity against A. margnale antigens. In
previous experiments, strong reactivity was seen with the 36
kDa protein (MSP2) of an Israel isolate of A. margnale when
reacted with immune sera from a cow infected with A. cntrale
(Shkap et al., 1991). In addition, epitopes common to both
species were identified using MAbs against the 36 kDa and 105
kDa surface proteins of A. margnale (Palmer et al., 1988a).
Discrepancy between the results of this study and previously
reported data may be explained by differences in geographic
isolates of A. margnale used. Antigenic differences between
different isolates of A. margnale have been well documented
(McGuire et al., 1984). In our study, a FL isolate was used,
whereas an Israel isolate of A. margnale was used in a
previous study which identified common immunodominant proteins
between the 2 species (Shkap et al., 1991). In addition, it
is difficult to compare results of immunoprecipitation
techniques used in previous experiments (Palmer et al., 1988a)
with immunoblots used here. Immunoblotting may alter epitopes
during the process of denaturation and electrophoretic
transfer prior to interaction with antibodies. Our results do
indicate however, that certain epitopes on the MSP3 protein of
the FL isolate of A. margnale are distinctly different from
those recognized by cattle infected with A. cntrale. This
could be a useful distinction since A. cntrale is


109
active copies of the vmp genes are located on linear,
extrachromosomal, DNA plasmids. The promotor and active copy
of the genes are located at telomeric ends of the linear
plasmids. Switching of a silent copy to the active locus just
downstream from the promotor causes expression of an
antigenically different Vmp and conversion of the organism
from one serotype to another. Antigenic variation allows
Borrelia hermsii to evade the host immune system and avoid
complete clearance from the blood stream. This causes a
persistent illness with a cyclic rise and fall in body
temperature every 4 to 7 days (Barbour, 1991).
Neisseria gonorrhoeae uses multigene families to
antigenically vary 2 important adherence ligands, the pili and
outer membrane opacity proteins (Opa) (Meyer et al., 1990;
Sparling et al., 1990). Variations in pili are accomplished
by multiple, silent, incomplete copies (over 20) of pil genes
termed minicassettes (Sparling et al., 1990). Insertion of
one of these incomplete copies into an expression site can
result in the expression of an antigenically different pilus.
In contrast, 10 12 complete opa genes are present in
Neisseria gonorrhoeae, and more than one may be expressed
simultaneously (Meyer et al., 1990). Expression of individual
opa genes is dependent on a repetitive sequence, the coding
repeat, which encodes the Opa signal peptide. The number of
5-mer repeats present determines if the opa gene is
translationally in frame. Genes in which 5-mer repeats occur


Fig. 5 Specificity experiments using immunoblots.
A. margnale (AM) initial body preparations, or B. bovis (Bbov) or B. bigemina
(Bbig) antigen preparations reacted with normal serum from a non-infected cow
(NBS), or antisera from animals infected with A. margnale (AM), C. ruminantium
(CR) B. bigemina (Bbig), or B. bovis (Bbov). Labeling above each lane indicates
the serum used (top), the initial body preparation used (center), and the dilution
of the serum (bottom). Molecular size standards in kilodaltons are illustrated on
the left.


106
uncoded region of clone MSP3-12, which was included in the
probe, could cause hybridization to multiple bands not related
to the MSP3 if this sequence was repeated in the genome. This
region is likely to contain regulatory sequences of the MSP3
gene, but the possibility of sequence homology between this
region and other unnkown multigene families of A. marginale
does exist. To confirm this is not occurring, probes made
from an internal sequence of clone MSP3-12 will need to be
produced and used to probe digested, genomic, A. marginale
DNA.
These data do suggest a copy number of at least 10 to 15
MSP3 genes in the FL and SI isolates with slightly less, 7 to
10, in the VA isolate. With the gene size of MSP3 being
approximately 2.6 kbp, we estimate MSP3 occupies as much as
3.0% of the 1,250 kbp genome of A. marginale (Alleman et al.,
1993). The exact function of this major immunodominant
surface protein is unknown, however, its prevalence in the
small genome of A. marginale suggest a need to antigenically
vary this very immunogenic protein in response to stress from
the host immune system. The A. marginale genome is estimated
to contain 7 to 10 copies of MSP2, another immunodominant A.
marginale surface protein encoded by a multigene family
(Palmer et al., 1994). This gene family occupies > 1% of the
genome. The exact function of this protein is also unknown,
however, immunization of cattle with affinity purified MSP2


98
some surface proteins, and the fact that C. ruminantium
antiserum reacts with several A. margnale antigens on
immunoblots, caution must be taken in the development of a
diagnostic test which can distinguish between these 2
pathogens.
Sera from animals infected with B. bovis or B. bigemina
showed no evidence of cross reactivity with A. margnale
preparations. These organisms are not considered to be
closely related phylogenetically (Dame et al., 1992; Van Vliet
et al., 1992), nor has any similarity in antigens or DNA been
detected in other experiments (Eriks et al., 1989; McGuire et
al., 1984; McGuire et al., 1991; Shkap et al., 1990).
Based on our results, there is no strong evidence which
would exclude MSP3 as a diagnostic test antigen based solely
on the specificity of this protein. However, because there is
strong cross reactivity with sera from animals infected with
Ehrlichia sp. which do not infect cattle, the specificity of
MSP3 still remains questionable, particularly in areas where
E. bovis and A. marginale coexist.
Size Polymorphism of MSP3 Among Various
Strains of A. margnale
Size polymorphism is demonstrated in the MSP3 proteins in
3 of the 4 geographic isolates studied. Marked size
polymorphism has been recognized in other major surface
antigens of A. marginale, the MSPla protein (Oberle et al.,
1988), and the MSP2 protein (Palmer et al., 1988b). Size


105
in cloned and genomic DNA when cut with restriction enzymes
Neo I, Bsp M, and Eae I. For example, Nci I digestion of
MSP3-12 produces a 1,176 bp fragment which hybridizes with the
digoxigenin-labeled MSP3-12 probe. Digestion of genomic DNA
from a FL isolate of A. margnale produces a fragment of
identical size which also hybridizes with the MSP3-12 probe.
Digestion of genomic and cloned DNA with Eae 1 and Bsp M
produces similar results. These enzymes cut various places
within MSP3-12 clone, ranging from nucleotides 12 to 2076.
This range covers almost the entire 2337 nucleotide sequence
of the cloned gene.
Multiple MSP3 Copies in the A. margnale Genome
Hybridization studies using digoxigenin-labeled MSP3-12
identified multiple copies of partially homologous MSP3 genes
in the genome of FL, SI, and VA strains of A. margnale.
Genomic DNA was digested with restriction enzymes selected to
cut outside of the MSP3-12 sequence. Hybridization of the
probe with a single fragment would be seen if MSP3 was encoded
by a single copy gene. Multiple fragments homologous to the
MSP3-12 sequence are identified. However, the exact number of
copies cannot be determined because restriction sites may be
polymorphic in other copies of MSP3, causing an exaggerated
estimation of the number of gene copies. In addition, more
than one gene copy may be present on large fragments of
genomic DNA. Although unlikely, it is possible that the


200
97.4
69
46
30
NSS
AO
AM
NBS
AC
AC
AM
AM
AM
AM
AM
AC
AM
AM
1/500
1/500
1/500
1/100
1/100
1/100
1/300
.P



CHAPTER 1
INTRODUCTION
Anaplasma margnale is an arthropod-borne, rickettsial
hemoparasite which invades the red blood cells of cattle
causing a clinical disease known as anaplasmosis. This
organism is the representative species of the genus Anaplasma,
in the family Anaplasmataceae, order Rickettsiales (Ristic and
Krier, 1974) The genus name, Anaplasma, refers to the
appearance of being devoid of cytoplasm, and the species name,
margnale, refers to the location of the organism in the
margin of infected erythrocytes. A. margnale was originally
discovered in cattle experimentally infected with Babesia
bigemina, and was thought to be a developmental stage of this
organism (Smith and Kilborne 1893) However, several years
later these marginal inclusions noted in the erythrocytes of
cattle infected with B. bigemina were conclusively identified
as agents belonging to another genus, Anaplasma (Theiler
1910).
A. margnale is now known to have a global distribution
which includes the United States of America. World-wide
economic losses are difficult to calculate, but losses in the
U.S. alone are estimated to be over 100 million dollars
annually (Goodger et al., 1979; McCallon, 1973). In certain
1


37
denatured at 100C for 3 min. Bacterial lysates or similarly
prepared FL A. margnale lysates were separated by SDS-PAGE
and transferred to nitrocellulose as previously described
above. Membranes were reacted with anti-MSP3 MAb AMG75C2 or
immune sera from animals infected with a FL or VA isolate of
A. margnale as described above. Nonimmune cattle sera and an
anti-T. brucei MAb were used as negative controls. Antigen
antibody reactions were visualized by ECL with detection
reagents containing luminol as a substrate (ECL Western
Blotting detection reagents), (Amersham International, pic,
Buckinghamshire, England). The membranes were exposed to
Hyperfilm MP (Amersham International, pic, Buckinghamshire,
England) to visualize bound antibody.
Diqoxiqenin Labeling of oBluescript MSP3-12
Empty pBluescript DNA and pBluescript MSP3-12 DNA were
grown in previously prepared transformants. Plasmid DNA was
isolated from bacterial DNA by ion exchange chromatography
using the QIAGEN Plasmid Midi Kit (Qiagen Inc., Chatsworth,
CA) according to the manufacturer's instructions. The
purified plasmid DNA was precipitated in ethanol, dried, and
dissolved in TE (10 mM Tris, pH 7.5, 1 mM EDTA).
A digoxigenin-labeled probe of pBluescript MSP3-12 was
prepared by digestion of 5 fig of pBluescript MSP3-12 with Eco
RI according to manufacturer's instructions (Boehringer
Mannheim Corp., Indianapolis, IN). Empty pBluescript was


22
infected carriers, is a nucleic acid probe derived from a
fragment within the gene coding for an A. margnale surface
protein (Goff et al., 1988; Eriks et al., 1989). This probe
can detect infected animals with parasitemias as low as
0.000025% (Eriks et al., 1989). Sensitivity and specificity
of the test surpasses all previously developed serological
tests, including the IIF (Goff et al., 1990). Although this
test is too complex to be useful for routine field diagnosis,
it has tremendous potential for identification of true
carriers which in the past was done by subinocculation of
splenectomized calves. This could be useful for the
identification of cattle for reference sera, establishing the
effects treatment with tetracycline has on carrier status, or
identification of infected ticks. Considering the alternative
of calf inoculation, a nucleic acid probe is a much more
convenient and economical tool when accurate identification of
infection is essential.
Previous Experiments
The ideal test antigen to detect A. margnale infected
carriers must be antigenic enough to have antibodies present
in sera of cattle during all stages of infection. It must be
species specific so as not to cross react with sera from
animals infected with related organisms, and it must be
conserved between isolates of A. margnale, enough to be
recognized by sera from cattle infected with various


Fig. 19 Not I and Sfi I digestion of the A. margnale
genome.
Genomic DNA from a Florida (FL), Virginia (VA), or South
Idaho (SI) isolate of A. margnale was digested with the
restriction enzymes indicated above the lanes, separated
by clamped homogeneous electrical field gel
electrophoresis, and stained with ethidium bromide. Lanes
1 and 2 contain lambda DNA-tfindlll fragments and Promega
Delta 39 markers, respectively, as size markers.
Molecular size markers in kilobase pairs (kbp) are
indicated to the left.


Fig. 3 Specificity experiments using immunoblots.
A. margnale (AM) or A. cntrale (AC) initial body preparations reacted with normal
sera from non-infected sheep (NSS), non-infected cattle (NBS), or antisera from
animals infected with A. ovis (AO), A. cntrale (AC), or A. margnale (AM) .
Labeling above each lane indicates the serum used (top), the initial body
preparation used (center), and the dilution of the serum (bottom). Molecular size
standards in kilodaltons are illustrated on the left.


Fig. 8 2-D gel electrophoresis of A. margnale
proteins.
Immunoblots using initial body preparation of a FL
isolate of A. margnale separated by 2-D gel
electrophoresis. Letters centered above each immunoblot
indicate the antibody used in the reaction, RB-955 =
rabbit anti-MSP3 polyclonal sera and MAb ANAF19E2 =
anti-MSP2 MAb. Rabbit anti-MSP3 was used at a dilution
of 1/5,000. Concentration of MAb = 5/Ltg/ml. Numbers
above each immunoblot indicate the pH taken at 1 cm
distances along the length of the tube gel. Arrows
indicate the isoelectric point of each 86 kilodalton
band. On the far right side of each gel, initial body
preparation from a FL isolate of A. margnale was
electrophoresed in a single dimension and immunoblotted
along with 2-D focused initial bodies to indicate the 86
kDa, MSP3 and 36 kDa MSP2. Molecular size standards in
kilodaltons are illustrated on the left of each
immunoblot.


100
may be conserved, particularly if epitopes are contained
within tandem repeats as with MSPla. If a complex family of
closely related genes is responsible for size variations,
similar to the MSP2 protein of A. margnale, antigenic
variations are more likely to exist as well. Size
polymorphism in MSP3 could pose a problem to its use as a
diagnostic test antigen, if it results in variation of
antigenically important epitopes. This may result in poor
reactivity between the MSP3 antigen of one isolate and immune
sera from an animal infected with a heterologous strain.
Immune Response to MSP3
This study demonstrates variations in reactivity of
immune sera from cattle infected with different geographic
isolates when reacted to the FL MSP3. Multiple 86 kDa
antigens are seen using immunoblots of 2-D separated
preparations. Reactivity of antisera with these antigens
varied depending on which geographic isolate the cattle were
infected with. In a homologous reaction with anti-FI serum
2 major areas of reactivity at pis 6.5 and 6.2 are identified.
These areas are distinctly different from the antigen
recognized by the anti-MSP3 MAb and the anti-VA serum (pi
5.6). The anti-WA serum recognizes 2 entirely different 86
kDa antigens (pi 5.3 and 5.1). Only serum from SI infected
cattle reacts with all the 86 kDa antigens identified with
other antisera and the MAb. Although the anti-MSP3 MAb


Fig. 20 Distribution of MSP3 in the A. margnale
genome.
Genomic DNA from a South Idaho (SI), Virginia (VA), or
Florida (FL) isolate of A. margnale was digested with
the restriction enzymes indicated above the lanes, and
separated by clamped homogeneous electrical field gel
electrophoresis. The fragments were then Southern
blotted and probed with digoxigenin-labeled pBluescript
MSP3-12. Molecular size markers in kilobase pairs (kbp)
are indicated to the left.


43
as previously described above and blots were probed with
digoxigenin-labeled pBluescript MSP3-12. Bound probe was
detected by ECL as described above. Lamda DNA/Hind III
fragments (GIBCO BRL,Gaithersburg, MD) and 1 kbp DNA Ladder
(GIBCO BRL,Gaithersburg, MD) were used for molecular size
standards.
Distribution of MSP3 Copies in the A. margnale Chromosome
Restriction Enzyme Digestion in Agarose Plugs
The locations of multiple MSP3 copies in the chromosome
were determined by Southern blotting of large A. margnale
genomic fragments separated by CHEF. Infected erythrocytes
containing intact genomic DNA from FL, SI, and VA strains of
A. margnale were embedded in 0.7% agarose plugs and stored in
0.5 M EDTA (pH 9.5), 1% N-lauroy lsarcosine, and 2 mg/ml
Proteinase K at 4 C as previously described above. Agarose
plugs were digested with Not I (Boehringer Mannheim Corp.
Indianapolis, IN) and Sfi I (New England BioLabs, Beverly,
MA) as previously described (Alleman et al., 1993). Briefly,
each plug was washed in 10 times their volume of T.E (10 mM
tris, pH 7.5, 0.1 mM EDTA) over several hours with fresh T.E
replaced each hour, then incubated in T.E plus 1 mM
phenylmethylsulphonyl fluoride (PMSF) for 2 hours at 37C.
Agarose plugs were then equilibrated on ice for 1 hour in 10
volumes lx restriction enzyme buffer plus 0.1 mg/ml BSA. A


pi WA
T
7 6 9 6 68 6.42 6.13 5.82 5.66 5 36 5.13 4.56
14.3
VD


Fig. 7 2D gel electrophoresis of A. margnale proteins.
Immunoblots using initial body preparation of a FL isolate of A. margnale
separated by 2-D gel electrophoresis. Letters centered above each immunoblot
indicate the antibody used in the reaction, FL = antiserum from a cow infected with
a Florida isolate of AM, VA = antiserum from a cow infected with a Virginia isolate
of AM, WA = antiserum from a cow infected with a Washington isolate of AM, SI =
antiserum from a cow infected with a South Idaho isolate of AM, and MAb AMG75C2
= anti-MSP3 MAb. Antisera are all used at a dilution of 1/300. Concentration of
MAb = 5/ng/ml. Numbers above each immunoblot indicate the pH taken at 1 cm
distances along the length of the tube gel. Arrows indicate the isoelectric point
of each 86 kilodalton band. On the far right side of each gel, initial body
preparation from a FL isolate of A. margnale was electrophoresed in a single
dimension and immunoblotted along with 2-D focused initial bodies to indicate the
86 kDa, MSP3. Molecular size standards in kilodaltons are illustrated on the left
of each immunoblot.


50
30


REFERENCE LIST
1. Alleman, A.R., S.M. Hamper, N. Viseshakul, and A.F.
Barbet. 1993. Analysis of the Anaplasma margnale genome
by pulsed-field electrophoresis. J. Gen. Microbiol.,
139:2439-2444.
2. Allred, D.R., T.C. McGuire, G.H. Palmer, S.R. Leib, T.M.
Harkins, T.F. McElwain, and A.F. Barbet. 1990. Molecular
basis for surface antigen size polymorphisms and
conservation of a neutralization-sensitive epitope in
Anaplasma margnale. Proc. Natl. Acad. Sci. 87:3220-3224.
3. Amerault, T.E., and T.O. Roby. 1968. A rapid card
agglutination test for bovine anaplasmosis. J. Am. Vet.
Med. Assoc., 153:1828-1834.
4. Amerault, T.E., J.E. Rose, and T.O. Roby. 1973. Modified
card agglutination test for bovine anaplasmosis:
evaluation with serum and plasma from experimental and
natural cases of anaplasmosis. Proceedings of the Annual
Meeting of the U.S. Animal Health Assoc. 76:737.
5. Barbet, A.F. and D.R. Allred. 1991. The msplb multigene
family of Anaplasma margnale: nucelotide sequence
analysis of an expressed copy. Infect. Immun. 59:971-976.
6. Barbet, A.F., L.W. Anderson, G.H. Palmer, and T.C.
McGuire. 1983. Comparison of proteins synthesized by two
different isolates of Anaplasma margnale. Infect. Immun.
40:1068-1074.
7. Barbet, A.F., G.H. Palmer, P.J. Myler, and T.C. McGuire.
1987. Characterization of an immunoprotective protein
complex of Anaplasma margnale by cloning and expression
of the gene coding for polypeptide AM105L. Infect.
Immun. 55:2428-2435.
8. Barbet, A.F., S.M. Semu, N. Chigagure, P.J. Kelly, F.
Jongejan, and S.M. Mahan. 1994. Size variation of the
major immunodomiant protein of Cowdria ruminantium. Clin.
Diagn. Lab. Immunol. 1:744-746.
9. Barbour, A.G. 1990. Antigenic variation of a relapsing
fever Borrela species. Annu. Rev. Microbiol. 44:155-171.
115


Fig. 17 Genomic representation of MSP3-12.
Southern blot of genomic DNA from a Florida (FL) isolate of A. margnale or
pBluescript MSP3-12 (CL), digested with restriction enzymes Eae I, Bsp M, or Nco
I and probed with digoxigenin-labeled MSP3-12. Molecular size markers in base
pairs (bp) are indicated to the right and left.


122
males of Dermacentor andersoni stiles fed on an Idaho
field infected, chronic carrier cow. Am. J. Vet. Res.
47:2269-2271.


12
agglutination did occur when fresh serum was used; however,
this problem did not occur when using heparinized plasma.
Additional studies involving large numbers of serum samples
(380) established 98% specificity for the CA test (Gonzalez et
al., 1978). However, none of these studies investigated the
potential for cross reactivity with sera from cattle infected
with other hemoparasites. Current technology, which was not
available when previous tests were developed, allows the
phylogenetic relatedness of various species and genera to be
determined. Using 16s rRNA sequence analysis investigators
have determined A. margnale to be closely related to other
Anaplasma sp., Ehrlichia sp., and Cowdria ruminantium (Dame et
al., 1992; Van Vliet et al., 1992). To accurately determine
the specificity of a test antigen, it is essential to test the
ability of the antigen to distinguish between infections of
related organisms. This was not done for any of the antigens
currently used in serologic diagnosis of A. margnale.
Therefore estimations of specificity for these tests are
unreliable.
Although the CA test is well adapted for field use, its
poor sensitivity (probably because it is a secondary binding
test) and its questionable specificity (because of its use of
a crude test antigen preparation) make it an undesirable test
for the rapid field identification of carrier cattle.


108
multiple MSP3 antigens. In addition, the epitope recognized
by the anti-MSP3 MAb is present on only one of the 86 kDa
antigens.
It has been proposed that antigenic variation plays a
role in the cyclic rickettsemia and persistent infection
recognized in carrier cattle infected with A. margnale
(Kieser, et al., 1990). The level of rickettsemia varied
markedly at bimonthly intervals from <103 to >105 infected
erythrocytes per ml of blood (Eriks et al., 1989) The number
of infected erythrocytes gradually increased over a 10 to 14
day period, then precipitously decreased (Kieser, et al.,
1990). The length and consistency of the cycles suggests
recurrence is due to continual antigenic variation by the
organism and development of a primary immune response by the
host. Further work is needed to determine if antigenic
variation of MSP2 or MSP3 occurs during rickettsemia cycles in
persistent carriers. This could be done by identifying copy
specific epitopes on expressed MSP3 antigens and monitoring
changes in parasite antigens during these cycles.
Bacteria in the genera Borrelia and Neisseria have been
shown to use multigene families to vary important surface
antigens and aid in the evasion of the host immune system
(Meyer et al., 1990; Barbour, 1990). The genome of Borrelia
hermsii, the causative agent of relapsing fever, contains a
large repertoire of genes encoding variable major proteins
(Vmps) (Barbour, 1990; Barbour, 1991). Multiple silent and


8
Secondary binding tests measure the interaction of
antigen and antibody by in vitro visualization of a secondary
event which occurs as a result of immune complex formation.
Such events include the precipitation of soluble antigens in
solution by immune complex formation, agglutination of
particulate antigen on the surface of bacteria or
erythrocytes, and the activation of the complement pathway,
resulting in cell lysis. Many of these tests require optimal
concentration of antigen and antibody to allow visualization,
and false results may occur in cases of antibody or antigen
excess (Nakamura et al., 1988). As a result of this, and the
gross visual detection required for many of these tests,
secondary binding tests lack the sensitivity of primary
binding assays. Examples of these tests include
immunoprecipitation, immunodiffusion, agglutination, and
complement fixation.
Tertiary binding tests actually measure the in vivo
protective effects specific antibodies may have in an animal.
These test measure the biological activity of the antibody to
determine their ability to protect a susceptible animal from
an infectious agent or neutralize the effects of an antigen.
These tests are most useful in experimental trials or in
treatment of certain diseases by passive transfer of antibody.
They are not practical or suitable for the diagnosis of
disease processes.


9
The suitability of a particular type of immunodiagnostic
test depends on many factors such as the use of test (field
versus laboratory use), the sensitivity and specificity
required, the prevalence of the disease, the number of samples
to be tested at any one time, and the speed at which a test
must be performed. In general, selection of a diagnostic test
often involves a compromise trade off between sensitivity,
specificity, and ease of performance. As previously
mentioned, the indirect ELISA is a very sensitive test, easy
to perform, and may be adapted to field use where multiple
samples may be tested simultaneously. It is a primary binding
assay and requires no specialized equipment to perform the
test or interpret the results. This may prove to be an ideal
test for the field detection of A. margnale infected carrier
cattle. Because of the extreme sensitivity of the indirect
ELISA, specificity can be a problem when detecting specific
antibodies in a polyclonal sera. To achieve acceptable
specificity, great care must be taken in selection and
purification of an appropriate test antigen.
Methods for Detection of A. margnale Infected Cattle
Currently an effective vaccine against A. margnale is
not available, and the inability of present serologic tests to
accurately detect persistently infected, carrier cattle
severely compromises efforts to establish disease free herds
and reduce economic losses through testing, isolation, and


Antiqen FL
pfilue
NSP3-12
pfilue
MSP3-12
pel ue
MSP3-12
pel ue
NSP3-12
Antibody Post-
Post-
Post-
Pre-
Pre-
Post-
Post-
Pre-
Pre-
M.U.
FL
FL
FL
FL
FL
VA
VA
VA
VA
200.0 --
97.4 --
69.0 --
30.0
21.5 --
'O
CO


bp
5,090
4,072
3,050
2,036
1,636
1,018
506
396
344
298
Soh I Nde I Sac I Hie II EcoR I
bp
VA SI FL VA SI FL VA SI FL VA SI FL pB 12 CT
-- 23,130
-- 9,416
-- 6,577
-- 4,361
-- 2,332
-- 2,057
564
00
CTi


121
61. Todorovic, R.A. R.F. Long, and B.R. McCallon. 1977.
Comparison of rapid card agglutination test with
complement fixation test for diagnosis of Anaplasma
margnale infection in Colombian cattle. Vet. Microbiol.
2 :167 .
62. Trueblood, E.S., T.C. McGuire, and G.H. Palmer. 1991.
Detection of Anaplasma margnale rickettsemia prior to
onset of clinical signs by using an antigen capture
enzyme-linked immunosorbent assay. J. Clin. Microbiol.
29(7):1542-1544.
63. Uilenberg, G. 1993. Other Ehrlichiosis of ruminants,
p.270. In Z. Woldehiwet and M. Ristic (ed.), Rickettsial
and Chlamydias Diseases of Domestic Animals. Pergamon
Press, Inc., Tarrytown, New York.
64. Van Vliet, A.H.M., F. Jongejan, and B.A.M. Van Der
Zeijst. 1992. Phylogenetic position of Cowdria
ruminantium (Rickettsiales) determined by analysis of
amplified 16S ribosomal DNA sequences. Int. J. of Syst.
Bacteriol., 42:494-498.
65. Van Vliet, A.H.M., F. Jongejan, M. Van Kleef, and B.A.M.
Van Der Zeijst. 1994. Molecular cloning, sequence
analysis, and expression of the gene encoding the
immunodominant 32-kilodalton protein of Cowdria
ruminantium. Infect. Immun., 62:1451-1456.
66. Viseshakul, N. S.M. Kamper, and A.F. Barbet. 1994.
Organization, structure, and expression of the MSP1/3 gene
family of Anaplasma marginale. abstr. 35. Abstr. 75th
Annu. Meet. Conf. Res. Workers An. Dis. 1994.
67. Visser, E.S., T.C. McGuire, G.H. Palmer, W.C. Davis, V.
Shkap, E. Pipano, and D.P. Knowles. 1992. The Anaplasma
marginale msp5 gene encodes a 19-kilodalton protein
conserved in all recognized Anaplasma species. Infect.
Immun., 60:5139-5144.
68. Wanduragala, L. and M. Ristic. 1993. Anaplasmosis, p.
65-74. In Z. Woldehiwet and M. Ristic (ed.), Rickettsial
and Chlamydias Diseases of Domestic Animals. Pergamon
Press, Inc., Tarrytown, New York.
69. Winkler, G.C., G.M. Brown and H. Lutz. 1987. Detection of
antibodies to Anaplasma marginale by an improved enzyme-
linked immunosorbent assay with sodium dodecyl sulfate-
disrupted antigen. J. Clin. Microbiol. 25(4):633-636.
70. Zaugg, J.L., D. Stiller, M.E. Croan, and S.D. Lincoln.
1986. Transmission of Anaplasma marginale Theiler by


21
perform, its lack of sensitivity and questionable specificity
are certainly areas of needed improvement.
Recently, an antigen capture ELISA was developed for the
detection of A. margnale infection (Trueblood et al., 1991).
This ELISA utilizes 2 monoclonal antibodies (MAb) which
recognize 2 different epitopes on the A. margnale surface
protein MSP-la. In this test one MAb is bound to a
polystyrene microtiter plate, antigen (ie. infected whole
blood) is incubated with this monoclonal, and a second
monoclonal conjugated to horseradish peroxidase is added to
the reaction mixture for visualization of any captured
antigen. This assay was sensitive enough to detect infected
animals with parasitemias of <1.0%, however, it is not
sensitive enough to detect carrier cattle which have
parasitemias as low as 104 infected cells per ml of blood
(0.0001%) (Eriks et al., 1993). Thus, although the detection
of A. margnale antigen in the blood of infected cattle would
be the most specific way to determine infection, the
sensitivity of most primary binding assays will not likely be
sufficient to detect the low level of parasitemia encountered
in carrier cattle.
Nucleic Acid Probe Hybridization
One test which has shown excellent results in the ability
to detect A. margnale organisms in infected cattle, even at
a level of sensitivity sufficient to identify chronically


107
does appear to offer at least partial protection against
homologous and heterologous challenge (Palmer et al., 1988b).
Genetic polymorphism between isolates of A. margnale is
seen when comparing isolates after digestion with the same
restriction endonucleases. This is evident by variations in
the length of restriction fragments which contain the MSP3
genes in each isolate. These results are consistent with
previous experiments identifying restriction fragment length
polymorphism between geographic isolates in ethidium bromide-
stained gels (Alleman, et al.f 1993). In addition, our
hybridization studies indicate there are fewer MSP3 copies in
the VA isolate than in the FL or SI strains. This, plus the
fact that serum from animals infected with a VA isolate binds
only one of the MSP3 antigens (pi 5.6), whereas serum from
animals infected with FI or SI strains react with multiple
antigens, may suggest less antigenic variation of MSP3 occurs
within the VA isolate.
Cloning and sequencing of several complete MSP3 genes may
be required to understand the full extent of the genetic and
antigenic polymorphism which exists between expressed copies
of MSP3. The 3 clones made available to us contain partial
gene sequences of MSP3. Although large areas of identity
exist, there is also significant amino acid sequence
variation. Our data suggest the amino acid sequence variation
results in epitope or antigenic variation as well. This is
evidenced by variable reactivity of different immune sera to


Fig. 16 Representation of clone MSP3-12 in the A. margnale genome.
Diagram of clone MSP3-12 indicating the restriction enzyme sites and the resulting
fragments used to map MSP3-12 to the genome. Nucleotide cleavage site for each
restriction enzyme is illustrated diagramatically (top) and by nucleaotide
number(bottom). Size of resulting fragments in base pairs is also provided.


SI
TBE
TE
Tris
TWEEN 20
TX
V
VA
w/v
WA
Mg
Ml
South Idaho
tris-borate-EDTA buffer
tris-EDTA buffer
tris hydroxymethyl aminoethane
polyoxyethylene-sorbitan monolaurate
Texas
volts
Virginia
weight/volume
Washington
microgram (s)
microliter (s)
x


6
regarding the use of radioactivity, the complexity of the test
procedure, and the need for specialized equipment. Because of
their extreme sensitivity, these tests are frequently used to
measure trace amounts of drugs in urine or other body fluids
(Tizard, 1992).
Immunofluorescence assays employ the use of fluorescent
dyes such as fluorescein isothiocyanate (FITC) to measure the
formation of immune complexes. FITC is easily conjugated to
immunoglobulins for detection using dark field microscopy with
an ultraviolet light source, or flow cytometry. Specific
immunoglobulins may be labeled directly and allowed to react
with an antigen as in direct fluorescent antibody tests, or
species specific antiglobulins may be labeled with FITC and
used to bind to antibody/antigen complexes in indirect
fluorescent assays. The indirect tests are usually more
sensitive since each antibody molecule bound to the antigen
may bind several labeled anti-globulin molecules (Tizard,
1992) .
Enzyme-linked immunosorbent assays (ELISA) employ the use
of enzymes for the detection of antigen/antibody interactions.
The three most commonly used enzymes are alkaline phosphatase,
horseradish peroxidase, and /3-galactosidase. In these
techniques the enzymes are chemically linked to
immunoglobulins or anti-globulins. Detection is accomplished
by the addition of an enzyme substrate to the reaction,


34
(AMG75C2, AMG76B1, AMG43/19, & AMG43/23), one of which,
AMG75C2, has been previously described (McGuire et al., 1991).
One of these genes, MSP3-12, contains the N-terminus and 633
bp upstream to the open reading frame (Fig. 1) The C-
terminus is not present in this clone. The remaining 2
clones, MSP3-11 and MSP3-19, are missing the N-terminal
sequence of the open reading frame (Fig. 1) They both
contain the C-terminal end of the gene as well as 1,473 bp and
2,480 bp respectively, downstream to the open reading frame.
Verification of Recombinant MSP3
To verify a cloned MSP3 gene represented a gene which
produced one or more of the 86 kDa antigens seen on 2-D
immunoblots, E. coli cells (Epicurian Coli XLl-Blue)
(Stratagene, Lajolla, CA) were transformed with pBluescript
containing each of the 3 MSP3 genes. This was done according
to manufacturer's instructions. E. coli cells were
transformed with nonrecombinant pBluescript as a negative
control. Transformed cells were plated on Luria agar
containing 50 /xg/ml of ampicillin. Transformants were
selected by blue/white screening and selected colonies of each
transformant were grown overnight in Luria broth containing 50
ig/ml of ampicillin. Transformed E. coli cells were suspended
in PBS, lysed in one half their volume of a 3x sample buffer
containing 0.1 M Tris pH 6.8, 5% SDS (w/v) 50% glycerol, 7.5%
j8-mercaptoethanol, and 0.00125% bromophenol blue, and heat


93
One antigen identified as immunodominant in the acutely
infected animal as well as chronically infected carriers is
the 86 kDa MSP3 protein from a FL isolate of A. marginale
(McGuire et al., 1991; Palmer et al., 1986). The FL isolate
of A. marginale was used for antigen isolation because it has
been found by adsorption studies to contain antigens common to
both morphologic types of A. marginale, the tailed and non-
tailed forms (Goff and Winward, 1985; Kreier and Ristic,
1963). Experiments using affinity purified MSP3 showed
excellent test sensitivity, however, attempts to produce a
recombinant form of MSP3 have resulted in inconsistent
responses to immune cattle sera. Possibilities for these
inconsistencies include 1) an inability to produce an
antigenically similar recombinant form of this protein in the
chosen vector (E. coli), or 2) variants of MSP3 exist within
or between different strains of A. marginale. Antigenic
variability has been previously identified in another major
immunodominant surface protein of A. marginale, MSP2 (McGuire
et al, 1984; Palmer et al., 1994). In these experiments, not
all organisms between strains, or within the same strain
reacted with a given MAb to MSP2.
In addition, the specificity of MSP3 in detecting A.
marginale infected cattle has not been totally established.
Previous experiments have determined there is no cross
reactivity between MSP3 when reacted with sera from animals
infected with B. bovis, B. bigemina, or an unidentified


39
areas of the MSP3 gene (Fig. 2) Digestions of genomic A.
margnale DNA (1.0 /xg) and pBluescript MSP3-12 (0.1 /xg) were
performed according to the manufacturer's specifications.
Digested genomic and plasmid DNA were separated by gel
electrophoresis on 1% agarose gels containing 0.1 /xg ethidium
bromide and photographed.
Southern Blots
Prior to transfer, the gel was incubated at room
temperature for 30 min. in 0.4 N NaOH, 0.6 M NaCl then 30 min.
in 1.5 M NaCl, 0.5 M tris HCL, pH 7.5. Digested DNA was then
transferred to a positively charged, molecular biology nylon
membrane (Boehringer Mannheim Corp., Indianapolis, IN) by
capillary diffusion using lOx SSC (lx SSC is 0.15 M NaCl plus
0.015 M sodium citrate). The filter was washed for 30 sec. in
0.4 N NaOH, then in 0.2 M Tris HC1, pH 7.5, 2X SSC for 2 min.
After air drying, the DNA was cross linked by ultraviolet
radiation and incubated for 3 hours at 65C in
prehybridization solution containing 6x SSC, 0.5% SDS (w/v) -
200 /xg/ml herring sperm DNA, 5x Denhart's [lx Denhart's is
0.02% Ficoll (w/v), 0.02% polyvinylpyrrolidone (w/v), 0.02%
BSA (w/v)]. The digoxigenin-labeled MSP3-12 probe was added
to fresh prehybridization solution at a concentration of 15
ng/ml, and the filter was hybridized overnight at 65C. Bound
probe was detected by enhanced chemiluminescence using
alkaline phosphatase conjugated, anti-digoxigenin IgG


TABLE OF CONTENTS
ACKNOWLEDGMENTS iv
LIST OF FIGURES V
ABBREVIATIONS viii
ABSTRACT X
CHAPTER 1 INTRODUCTION 1
CHAPTER 2 LITERATURE REVIEW 5
General Methods of Serologic Diagnosis 5
Methods for Detection of A. margnale Infected
Cattle 9
The Rapid Card Agglutination Test (CA) ... 11
The Complement Fixation Test (CF) 13
The Indirect Immunofluorescence Test (IIF) 14
The Radioimmunoassay (RIA) 15
Enzyme-Linked Immunosorbent Assays (ELISA) 16
Nucleic Acid Probe Hybridization 21
Previous Experiments 22
CHAPTER 3 MATERIALS AND METHODS 26
Anaplasma margnale Strains 26
Initial Body Preparation 27
Babesia bovis and Babesia bigemina Antigen
Preparation 28
Antisera used for Immunoblots 29
SDS-PAGE 30
Two-Dimensional Gel Electrophoresis 31
Immunoblots with Antisera 32
Purification of A. margnale Genomic DNA 33
MSP3 Clones 33
Verification of Recombinant MSP3 34
Digoxigenin Labeling of pBluescript MSP3-12 ... 37
Representation of pBluescript MSP3-12 38
Restriction Enzyme Digestion 38
Southern Blots 39
Presence of Multiple MSP3 Gene Copies 42
Distribution of MSP3 Copies in the A. margnale
Chromosome 4 3
v


104
proteins expressed by clones MSP3-11 or MSP3-19. The N-
terminus is lacking in these clones, therefore upstream
regulatory sequences of the gene are not available for E. coli
polymerase binding. The clones are in frame with the lacZ
gene in pBluescript, however, either they are not produced in
sufficient quantity without IPTG induction, or they do not
contain the epitope recognized by MAb AMG75C2.
pBluescript MSP3 is an Accurate Representation of
a Genomic Copy of MSP3
Numerous artifacts can occur during the process of
cloning causing disruption of the original form of a gene. In
constructing a library, noncontiguous fragments of DNA may
reanneal prior to ligation into the vector. In addition, many
cloning artifacts may occur once the vector is transformed
into the host cell. Some of these include deletions,
rearrangements, or endonuclease digestion of insert DNA by the
host cell. Repeats or hair-pin structures in insert DNA may
not be well tolerated by some host cells such as E. coli, and
the insert may be omitted or portions rearranged or deleted
during replication. Because of these, and many other
potential cloning artifacts, it must be shown that the
recombinant form of MSP3 is an accurate representation of
genomic MSP3. We have determined that pBluescript MSP3-12 is
an accurate representation of genomic MSP3 by 1) expression of
a recombinant protein which is bound by anti-MSP3 MAb and by
2) demonstration of comigrating bands of predetermined sizes


27
blood from cattle infected with the previously mentioned
isolates was collected with sodium heparin used as an
anticoagulant and washed 3 times in PBS. Erythrocytes were
separated from bovine leukocytes by passing washed blood over
an a-cellulose/microcrystalline cellulose column (Sigmacell
type 50, Sigma Chemical Co., St. Louis, MO) as described by
Beutler (Beutler, 1984), except that 1% bovine serum albumin
(BSA) was included in the wash buffer. Washed cells were
resuspended to a concentration of 5.0 x 106 cells//xl, or
approximately 35% packed cell volume (PCV). Blood films were
prepared from isolated erythrocytes for microscopic detection
of bovine leukocyte contamination and level of parasitemia.
Intact erythrocytes were embedded in 0.7% agarose by
mixing 1 part erythrocyte suspension with 2 parts of 1% FMC
InCert Agarose (FMC Bioproducts, Rockland, ME) in PBS (0.14M
NaCl, 2.68 mM KC1, 8.26 mM K2HP04, pH 7.4), 0.125 M EDTA.
Mixtures of blood and agarose were kept at 37C while
pipetting into molds. Molds containing plugs were placed on
ice to set. Plugs were incubated in 0.5 M EDTA (pH 9.5), 1%
N-lauroylsarcosine, and 2 mg/ml Proteinase K for 48 hours at
37C, then stored at 4C in fresh Proteinase K solution.
Initial Body Preparation
A. margnale initial bodies were isolated from infected
erythrocytes as previously described (Palmer and McGuire,
1984). Briefly, frozen blood was guick-thawed at 37C and


25
However, proteins expressed by these clones were inconsistent
in their reactivity with immune cattle sera. In addition, in
depletion experiments, anti-MSP3 MAbs were reacted with FL A.
margnale initial bodies. The initial bodies still contained
an 86 kDa antigen when reacted with immune cattle sera.
The MSP3 antigen of A. margnale appears to be a strong
candidate as a diagnostic test antigen to detect infected
carrier cattle. However, these previous experiments propose
questions which must be answered to confirm this hypothesis.
Questions must be addressed regarding the specificity of this
protein in detecting A. margnale infection, the conservation
of the MSP3 protein between various geographic isolates of A.
margnale, and explanations for inconsistent reactivity of the
3 MSP3 clones. The purpose of our investigation is to further
characterize the MSP3 protein and determine if it is a
suitable candidate for a diagnostic test antigen to detect A.
margnale infected, carrier cattle.


101
reacted with only a single 86 kDa antigen, rabbit-Anti-MSP3
polyclonal serum reacts with antigens in areas of the pH
gradient identical to those recognized by anti-FL serum, Anti-
VA serum, and the ant-MSP3 MAb. This rabbit sera was made by
injection of purified MSP3, isolated from an affinity column
using MAb AMG75C2. The production and reactivity of this
serum has previously been described (McGuire et al., 1991).
Reactivity of different 86 kDa antigens with rabbit-anti-MSP3
suggests that common epitopes may exist on these antigens.
The above results indicate, similar to the MSP2 protein
of A. margnale, not only size polymorphism, but also
antigenic polymorphism exist between MSP3 antigens of
different isolates. The ability of the organisms to alter
this surface antigen could present a problem for use as a
diagnostic test antigen, resulting in a test with low
sensitivity which could not reliably detect infection in
animals infected with different strains of A. margnale.
There are at least 3 possible explanations for the
multiple 86 kDa antigens present in the FL isolate of A.
margnale. These antigens may arise from, a) post-
translational modification of a protein transcribed from
single copy gene, b) several closely related genes transcribed
and translated from a multigene family, or c) entirely
unrelated genes. Reactivity of the rabbit-anti-MSP3 sera with
multiple 86 kDa antigens suggests at least 3 of the MSP3
antigens share common epitopes. This would indicate the 86


61
Pi
RB-955
M.W.
kD
200 -
97.4 -
69
46
30 -
T
6.68
T
6.42 6.13
5.82 5.66 5.36 5.13 4.16
pi MAb ANAF19E2
6.97
6.9
6.68 6.42 6.13
5.82 5.66 536 5.13 4.16
1 1 1 1 1 1 II 1
200
46
30
21 5 -


11
indirect immunofluorescence test (IIF), the radioimmunoassay
(RIA), and various enzyme-linked immunosorbent assays (ELISA).
The Rapid Card Agglutination Test (CA)
The CA is one of the most widely used in the field
because it is easy to run, requires minimal equipment, and
uses unheated sera or heparinized plasma (Amerault & Roby,
1968). Beinq a secondary binding test, this assay lacks the
sensitivity of primary binding assays. Initial studies using
the CA test reported 100% sensitivity in detecting carrier
cattle and 86% sensitivity in detecting all known infected
cattle (Amerault & Roby, 1968). However, the number of
positive cattle tested was small (22) and subsequent studies
comparing sensitivity and specificity of the CA test with
other serological tests indicates sensitivity to be 84%
(Gonzalez et al., 1978). Yet another study indicated the CA
test was able to detect only 6 of 9 (66.6%) known carrier
cattle (Luther et al., 1980). In addition, 89.3% of
noninfected, A. margnale vaccinated cattle had positive
reactions with the CA test for up to 15 months after
vaccination (Luther et al., 1980).
Specificity of the CA test was initially reported to be
100% with none of the 24 normal cattle sera testing positive
(Amerault & Roby, 1968). This test uses a crude preparation
of A. margnale and erythrocyte antigen, and problems of
specificity are likely to be encountered. Nonspecific


69
regulatory sequences for gene expression. Clones MSP3-11 and
MSP3-19 lack the N-terminus but each contains the C-terminus
and 1,473 bp and 2480 bp respectively, downstream to the 3'
end of the open reading frame. A schematic representation of
each clone shows areas of homology between the 3 genes as well
as areas which are homologous to the MSP2 gene of A. margnale
(Fig. 12).
Homologous areas are distributed throughout the 3 MSP3
genes. Areas common to all 3 genes are at bases 1191 to 1472
and bases 1818 to 2008 of the MSP3-12 clone. An area at the
3' end of MSP3-12 between bases 2090 to 2280 is also found in
MSP3-11, but is absent in MSP3-19. The last 200 bp at the C-
terminal end of genes MSP3-11 and MSP3-19 are homologous.
This area is unavailable for comparison in MSP3-12.
The MSP2 gene encodes a 36 kDa surface protein, and is
known to be expressed by a polymorphic, multigene family
(Palmer et al., 1994). Areas of homology between the N-
terminus of the MSP2 gene and MSP3-12 are indicated in Fig.
12. The N- terminus is absent in the MSP3-11 and MSP3-19
clones, so comparisons cannot be made. The areas of amino
acid sequence homology between MSP3-12 and MSP2 (amino acids
55 through 176) shows 65.6% similarity and 54.9% identity
(Fig. 13). There is a homologous area of over 500 bp between
MSP3-11 and MSP2 (Fig. 12). However, this area is outside of
the open reading frame, in a different reading frame, and a


64
Msp311.Conentry:1 Length:3263 Hay 29, 1991 11:34 Check: 1828 ..
1 GGTGGGGAGATGGTAGGAGTT GAT GAAGGACT AGTTAT ACAAGAACTGAGCAGACCAGAAGAATT AGAAAAGCTACAACAT GAACTAGCMAGCAAGT AA
101 GTAAATTAGCTGAACTTGGAGAACTTAAGTGGTTAGAGCAACTTGAGACACTGGAGACTGAGGAGTTGGAGAAGGTGGCGAAAAGAGCCACACAGAAGCT
201 CAGTATATTGGAGGAGCAGCCAGGAT ACACCT CACTAGAQGAGCT GGAGAAGAAGCTGAAGGAGAT AGAQGAAATAAGGAAACT GAAGGGACCGGAGGCA
301 ATCAGAAAGCTTAGAGACTTGGAGAGGCTGGAAGCTGAGAAGTTGGAGGAGGTGAAGAAGAAGGTAAAGQGTTCAGAGCTTGCAGAGCATTTGGACAAAA
401 CGTGGTTGCTACGCAGGATTGGAAGAGTAQGGCAGGAACTAGCAGCGATAAGGGAGCTGAAGGAATTGGQGCTAGAGGAACGGCTCAGGGrACTAGCTGA
501 GATTAAGGAAGTTAGAGCATT GGCAGAGAAGCGGAAGGCTGGGGGACTAGAGATTCAGGAGGGGTTACAQCT GACT GAGAAGATTAGAGCATTGGGT GGA
601 CAGCTGGAT CT GCTGGAGGCACGAAT ATTATTGGGGCT AGAAGCT GAGAAGAT GAAGGAQGT GGAAGCGQCAAAGGAGGAAGT GGAGATTTTCCT GGGAT
701 T GCTCCACAACAATGACACAGAGACCAAGAAGAT AGAGGAGGT GAAGACTCTGGT AAT GCGGGT GGGT GACAGT CAGGGACTAAGGGGTOCGTTGGT CAA
801 GAAGTTAGAQGAGCTGGCTAAGAAGTT GGAGCCAAAGGT GGGT GGCAATACAGGGT TTCAGGGT CAGGTAGGGCTAGT GAAGGAT CTTAAGAAAAAGCT A
901 GAAGAACTAGCAGCGAT AAGGGAAGCATTQGAGCAACT GAAGGT GGAACCAGCGCT GCT AGAAGGGAT AAAGAGGGAGGTGCT AACACAQCAGAGTACTC
1001 AAGTTACACGTAGGGAGAATAATCCACAAAGCAAGGCAGAAGGTAAACTAGAAGAGATAAGAGCGCTTAGAAAGCTAGGACAGTTGGATGAGCTAATACC
1101 AGAAGAGAGATTGAGAGAATT GT CCAAGGTTAAGGAAACACT GAAGCGGTT GAAGGGT AAGGAGCTAGT AGACCTAGAGAGGAAGCT AGAAACTAT GGGG
1201 GAACT AAAGCAGGAGAT AGAGAAAAT CAAQGGGCAGGAAGACTTGAAGCAGCT AGAGGCTAGGAAGT T GGAGGAGGTGAAGAAGGCAAAAAGGGAAGTGT
1301 TTATTCTGAQGACAAAGGAAAGTCTGGAGAAAGAGGGGAAGCTGGAGGAGTTAAGGGGCTATAAACTCAAGAAGGCGATAAAGAAGCTGGAAACCTTGAT
1401 AGAGAAGATACGAGGGGTCAGCACGCT GAAGGAACAGT GGGAGCCAAAAGTT GAGAAGCTCAAGAGCAAACT AGAAACACTAGCTGCCATAAGGGAGCTG
1501 AAGAAATTGGGGTTTAAAGATTGGTTGAGACTCAGGACACTAGAGGAGCTTACAGAGATAGCTGAGCAGCGTGGAGTTGCAACAGTGATGAAGGCAGCAT
1601 TAGCTAGTGCGATGGAGATAGCTAAGAACAGGGGTTGGACTGATTATTTAAACAGTCTAGATGTAAGTGAGAGAGCTAATGCTGCAAGGGAATTAATTGC
1701 TGCTGAGAAGATTAGAAAATGGGCTAGGGATATTAATAACTTGGATGCTGATGAACGGGCAATGGTCGCTGGGGCCCTAACCCCTTCTACAACAGTGTGA
1801 CTCACCACTCTCCCCCGAAACATCACTCACACTTCCAACTCTACTGGGCACCACCAAGGQCAGCCAGCCTCTCTTGGCACCTCATCTCGTACCCTCTTGG
1901 CCTACCAAATCTCGACCCTQCCACCACAGCCCTCTGCACCGGTGTCACCTGCATAACCCGrCTCCCAGTCrCTTGTACCTCAACACCTGCATCTCATCTT
2001 CCTTACCCAATGTGGCATTOCACAATCTTTGTCCTCTCTACTTACCCCCTACTTCTTACOGGCACCACTCACTGCTTTCCCAGGTCAACTAGTCACCCAA
2101 CCTCTAGCACCTTCAGCATCTCAGCCCAACTCACCACAGCTCTCTGCACCTTCAACAGCTCTAGCAAAAGCCCCGGCAAOVACTGCCTTATCCTCCTTAC
2201 TCAGGTTATTAATGTTATTCGCCATCCCCTGCAAGTTGATGGTGTCTTTGTTCCCCTCAACAGAAAGCAGTGTGTCTGTACCCTCAGTGAACACCTCACT
2301 AATCTTGGTGGTGGCAGTGCTATCGGTGGTACCACACTTCTTAGTGCCATTACTGCCTTCGCCCTTGCCACACACATTCTGGCTCACCTTCTCACCATTA
2401 GTAGTACCCTCTACTGCTTTACCCCACTTCTTGGCTTCACCCTTAGTCATCTTACCTAAAGCAGTGGCAAGACGGTCTACCTGACCTCTTGCTGTATCAT
2501 ATGCTAACTCCTTTCCTAATAAGAATACTGAAGCTGTATCCTCATTAGACTTCTTACCTCCCTTAATGACAAACCTTTCATACCCTACTTCTACTTCAAC
2601 CCTTGCTCCTCCAATACTATACCCAATGCTCCCCTCTAGGGCTGTGAGCATGCTATGGTAGCCCTGATATGGTTATTCATCACAGCAGGTTCTACTAATG
2701 CCGATACTATCAGCAGTACTAACACTGCTGftCGATACAGCTACCAAGGTTATTTGTGACGTCATTGTCTTCGTGCAGAAGCTGGGATTACCCATCATGAC
2801 AGGGGTGGTACTAGGTTCTGGCGTCATGGCTTGGCACACTTGCATGGCCTGCCATGGTAATGCTGGTTGTGTGTACAGCOMAGGTAAGCTTATGGCCAC
2901 TGTATTGAGGGTAATACCAGTTAGTGCAAQGGTGGACGTAGAGGTTTTGQGAGAGTGGTGATGTTGCTAAAACTTAGATTCTTACTCATTGCTGTAGTAT
3001 TAGCATTTGGTATCTTACATGGTGTACCTXTGGGGCTAOTAAGTCCAGCGAGGGTGCTCAGACTGCTGATGATACAGCO\CTGTTGTTATATGTAACGT
3101 TATACGGTTTGTGCAGAAGCTGGGACTACCCATCATGACTGGGGTGATACTAGGTTCTAGCATTATGGCTATCTTTGGTAAGCTTGCATGGCCTGCTATT
3201 GTAATGCTGGTTGTATTTACAGCTATATTCTTTGGTGCTQGTAAGCTTATGGCTAAATTCCCG


55
M.W fl WA SI FL VA
kD
200-
97.4-
69-
46-
Pre Post
Anti-FI A.M. sera (1/300)


EVALUATION OF ANAPLASMA MARGINALE
MAJOR SURFACE PROTEIN 3 (MSP3) AS A DIAGNOSTIC TEST ANTIGEN
By
ARTHUR RICK ALLEMAN
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
1995


66
Msp312.Conentry:1 Length:2337 August21, 1991 15:43 Type:N Check: 1077
1 GGGGGCCTGCCATGGTAATGCTGGTTGTATGTACAGCCATAGGTAAGCTTATGGCCACTGTATTGAGGGTAATACCAGTTAGTGCAAGGGTGGACGTAGA
101 GGTTTTGGGAGAGTGGTGATGTTGCTAAAACTTAGATTCTTACTCATTGCTGTAGTATTAGCATTGGGTATCTTACATGGTGTACCTGCTGGGGCTAGTA
201 AGTCCAGCGAGGGTGCTCAGACTGCTGATGATACAGCGACTATTGTTATATGTAACGTTATACGGTTTGTGCAGAAGCTQGGACTACCCATCATGACTGG
301 AGTGATACTAGGTTCTAGCATTATGGCTATCTTTGGTAAGCTTGCATGGCCTGCTATTGTGATGTTGGTTGTATTTACAQCTATATTCTTTGGTGCTGGT
401 AAGCTTATGQCTAAGTTTGCTGCTGGGTTGAGTGGTGAGGGTGTGAAGGATGCCGGTAGCTTTGACTGTTCCAGTTATAAAGGCACTGCTAGGCAGTGAT
501 GGTGCAGAGAGCTGTGGTGGTCGTGCCAAGTTAGACTGGGATGCCAGGGGTTTATAGATAGTGGCTTTGTTATGTCGGTGGTACCAAGAACTTCAAGTGT
601 GTTGGTGCCCAGTAGAGTTGGAAGTGTGAGTGATGTTTCGAGGGAGAGTGGTGATGATGCCAAAACTTAGATTCTTACTGATTGCTGGGGTATCAGCATT
701 GGGTATCCTOCATAGTATGTCTGCGGGGGCTGCCGATAGAAGTAAGTCACGCACACAGAGGTTGGCTAGTGGACCGGCTTTGGGGAAAGQCAACGGGAGC
801 TTCTACATAGGCCTAGACTATAACCCAACTTTCAACGGTATCAAGGACCTGAAAATCATCGGCGAAACCGATGAGGATGAAATGGATGTTCTCACCGGTG
901 CCAGGGGCCTGTTCCCGATGAACGCTCTTGCTAGCAACGTCACCGATTTTAACTCATACCACTTCGACTQGAGTACCCCACTGCCTGGGCTAGAATTTGG
1001 GAACAGTACCCTGGCTCTTQGAGGGAGCATTGGGTACAGAATTGGAGGAGCCAGGGTTGAAGTAGGGATAGGACATGAGAGGTTTGTTATTAAGGGGGGA
1101 GATGATGCAGCATTCCTACTAGGTAGGGAACTAGCATTGGATACAGCGAQGGGTCAGTTACTATCCAGTGCATTGGGTAQGATGTCCATGGGTGATGTAC
1201 ACAGATTAAAGAAGGAAGTAGTTGATAGTATAGGAAGAGGAACAGCTAGTCCTGTAAGGGCAATGTTTAGTAGAGAGATCTCAGATGGGAATACATTACT
1301 TGCTGGGGAGATGGTAGGAGTTGATGAAGGACTAGTTATACAAGAACTGAGCAGACCAGAAGAATTAGAAAAGCTACAACATGAACTAGCAAAGCAAGTA
1401 AGTAAATTAGCTGAACTTGGAGAACTTAAGTGGTTAGAAGAACTTGAAAAGCTGGAGACTGAGGAGCTGGAGCAAGGACTCGAAGGAGCGCTAAAAGCTT
1501 TGGGCGT GGAAGCAT CAGT GCAGGAGCTGGTACAGAGATTCAAGAAGGAAATTGCT GATQGT AAGACACCGGAAGAGAT AAAGCT GGAGTGGAT CAAGGA
1601 GATAGAAGCTAAGAAGTTAGAGGAGGTGCAGGAGCAGGCTGAGAAAAAGCTGGCGGAGCTGGAACTAGCQGATAAAGGAQGAGCATTGGGAACTAAAGGC
1701 GAAATACGGGAAGTTAGGCAGCTCAAGGAACTAGCGGATAAAGGAGGAGCGTTAGGAATCATGGCCGAGAAGCTCAAGAAGCAGGAAAGTCTCAAGGGGC
1801 T AGGAGGAACAGT AGAAGAATTAGCAGCGAT AAGGGAACTGAAGGAATTGGGGCT AGAAGAGCAGCTGAQGAT GCT GGCTGAGATTAAGGAAGTTAAGAG
1901 ATTGGCAGAGAAGCAGAAATCACAAGGATTAGAGGCTCATGAGGGATTACAACTGACTGAGAAGATTAGAGCATTGGGAGGACAGCTGGATCTGCTGGAA
2001 GAGAGATTGAGTGAAAAGTnAAGAGACTACAGCAGAATGAGCAAGGTATACGTAAGGATCTGAATCCACAAGCTATAACAGCAGTGGCAGGTAAACTAG
2101 ATGAGATTTQGGTACTGGGGACGCTAGGACAGTTAGATGAGCTAGTGCCTGGGAAGAGAGCTCAGACACTGGGGGAAATGAAGGCAACGCrGAAGGAGCT
2201 CGAGAAAATACGGAAGTTAGTAGACCTAGAGAAGAAGCTAGAAACTATGGGAGAACTAAAGAAGGAGATAGAGAAAATCAACGAGGAGGGAGAGCTTGTG
2301 AGGTTTACACGCAAAGAAGTCT CGGAGAT AT CACCCG


99
variation of MSPla among geographic isolates was later
explained by various numbers of tandem repeats within a single
domain (Allred et al.,1990). Despite the marked size
variation of MSPla, many of the surface epitopes, including a
neutralization-sensitive epitope remained conserved between
strains (Allred et al., 1990; Oberle et al., 1988).
The MSP2 protein was found to be encoded by a multigene
family in which substantial nucleotide sequence polymorphism
exists among MSP2 copies. Variability in the expression of
these genes among organisms suggests one potential mechanism
for size polymorphism. This surface protein was found to be
antigenically polymorphic as well (McGuire et al., 1984;
Palmer et al., 1994).
Size variation has been reported in the major
immunodominant surface protein of another rickettsial agent
closely related to A. margnale, the 32 kDa MAPI protein of C.
ruminantium (Barbet, et al., 1994). The conservation of
antigenic epitopes on the protein or the mechanism of size
variation is not known.
Although the mechanism of size variation of the MSP3
protein is not known, this illustrates the organism's ability
to alter important, antigenic surface proteins. The presence
of a complex multigene family for MSP3 or the presence of
variable numbers of tandem repeats are potential mechanisms
for size polymorphism. If tandem repeats are responsible for
the size polymorphism of MSP3, important antigenic epitopes


19
to test the sensitivity of this ELISA in detecting infected
carriers shown to be positive by other proven sensitive tests
such as IIF or calf inoculations.
Another ELISA used a 2 antigen technigue to enhance test
specificity; a negative antigen prepared from a cow prior to
infection, and a positive antigen derived from A. margnale
infected cells (Duzgun et al., 1988). Reactants were
identified using net absorbance values obtained by subtracting
the absorbance value of sera with negative antigen from the
absorbance value of sera with positive antigen. Specificity
of this test was good with only 3% of negative sera, 2% of
sera from animals infected with B. bovis, and 4% of sera from
animals infected with B. bigemina giving positive results.
Other related rickettsial agents were not tested. Sensitivity
also appeared to be good with no false negatives noted in 100
animals confirmed negative by IIF or calf inoculations. A
small number of infected cattle had positive ELISAs for up to
3 years later. This test provided the best sensitivity and
specificity thus far, and was the first ELISA that
demonstrated the ability to detect long term carrier cattle.
However, the crude nature of the test antigen necessitates the
use of the 2 antigen system which is more cumbersome, complex,
and time consuming. If this 2 antigen system had not been
used, 37% of the negative sera would have given false positive
results. In addition, the 2 antigen system reguires the use
of a spectrophotometer to identify reactants. This does not


BIOGRAPHICAL SKETCH
A. Rick Alleman was born in New Orleans, LA, on August
23, 1954. He is married to Mary Alleman, and they have 2
children, Arthur Rick Jr. and Grace Elizabeth.
Rick completed his undergraduate studies at the
University of New Orleans in 1976. He then attended Louisiana
State University, School of Veterinary Medicine in Baton Rouge
where he obtained his Doctor of Veterinary Medicine degree in
1980. At this time he practiced small animal medicine in New
Orleans, LA, with his partner Mariano Guas. In 1986 Rick
became board certified by the American Board of Veterinary
Practitioners (ABVP) as a specialist in companion animal
medicine.
In 1989 he left private practice and, under the guidance
of Drs. Rose Raskin and John Harvey, joined the University of
Florida College of Veterinary Medicine to do a residency in
clinical pathology. In 1992 he completed his residency
training and was board certified by the American College of
Veterinary Pathologists (ACVP), specialty Clinical Pathology.
In 1990, under the guidance of Dr. Anthony Barbet, Rick
entered into a Ph.D. graduate program in the Department of
Pathobiology. During his doctoral research program he
developed a keen interest in the study of rickettsial agents
123


17
most of these tests are good. However, none of these tests
use a single, purified A. margnale antigen, all use a crude
mixture of A. margnale antigens prepared from initial bodies.
Therefore, test specificity has been a problem, as is
frequently encountered using a mixture of proteins as a test
antigen in a highly sensitive technique.
Barry et al., 1986 developed one of the early ELISAs for
diagnosis of A. margnale infection in cattle. Using a crude
mixture of A. margnale initial bodies and cell membranes from
ghost RBC's as a test antigen, the test reportedly was able to
accurately distinguish cattle free from infection from
recently infected animals (up to 8 months duration). However,
no measures were taken to insure the positive or negative
status of most of the animals tested. Also, test accuracy for
identifying carrier cattle was questionable having identified
only 2 of 3 cattle infected > 3 years. In addition, the test
appeared to be fairly insensitive, necessitating the testing
of sera at a dilution of 1:100. Using concentrated serum for
testing may cause a problem with test specificity and 14.6% of
cattle inoculated twice with B. bovis-infected erythrocytes
were positive for A. margnale using this ELISA. Cross
reactions were attributed to antibodies produced against RBC
antigens in the inoculum. Experiments investigating cross
reactivity with other closely related rickettsial or
hematoprotozoal agents were not performed.


MSP3 CLONES
MSP3-12
Start
-or-:
Q-i
\ -
I 613 794 1141 1191 1306 1472
MSP2->
1818 1953 2008 2090 2280 2337
MSP3-11
MSP3-19
1191 1306 1472 1818 1953 Stop
<-MSP2


CHAPTER 3
MATERIALS AND METHODS
Anaplasma margnale Strains
Four isolates of A. margnale were used in this study,
FL, VA, South Idaho (SI), and Washington (WA). These isolates
are designated by their original location of isolation
(McGuire et al., 1984). Isolates were stored in liquid
nitrogen as cryopreserved stabilates (Love, 1972) before being
used to infect splenectomized calves. Thawed stabilate (20ml)
from each isolate was injected intramuscularly into 6-month-
old, male Holstein calves. Calves were monitored daily for
percent parasitemia by blood smear evaluation, and packed cell
volume (PCV). Infected, whole blood was collected in EDTA
from calves during periods of peak parasitemias (FL=70%,
VA=36%, SI=42%, and WA=4 0%), centrifuged at 10,000 x g for 15
min., and the serum and buffy coat was removed. Packed
erythrocytes were washed 3 times in phosphate buffered saline
(PBS) (0.14M NaCl, 2.68 mM KC1, 8.26 mM K2HP04, pH 7.4),
resuspended to a PCV of 50%, and stored at -70C.
A. margnale strains used in clamped homogeneous electric
field electrophoresis (CHEF) studies were prepared as
previously described (Alleman et al., 1993). Briefly, whole
26


62
MSP2 was seen. For example, in a homologous reaction there
was reactivity with antigens with pis from 6.42 to 5.66.
However, when reacted with sera from animals infected with WA
or SI isolates, reactivity with antigens with pis of 6.42 to
5.36 was observed. These spots were confirmed to be the MSP2
antigen by use of an anti-MSP2 MAb (ANF19E2), the reactivity
of which has been previously described (McGuire et al., 1984;
Palmer et al., 1988; Palmer et al., 1994). Multiple spots
with a molecular mass of 36 kDa were recognized by this MAb
(Fig. 8) .
MSP3 Nucleotide Sequence and Translation
Drs. Travis McGuire, Terry McElwain, and Guy Palmer
(Washington State University, Pullman) identified and
sequenced 3 clones of the MSP3 gene by screening a genomic
library of a FL strain of A. margnale DNA with MAbs to MSP3.
These clones were designated MSP3-11, MSP3-12, and MSP3-19.
The 3 clones in pBluescript vectors, along with their
nucleotide and deduced amino acid sequences, were supplied to
us courtesy of the above individuals. The entire nucleotide
sequence of each clone is illustrated in Figs. 9, 10, & 11.
A complete gene sequence is lacking in each clone. However,
MSP3-12 contains the N-terminus of the gene and 633 bp
upstream to the open reading frame. This area upstream to the
5' end of the open reading frame is likely to contain


Fig. 9 Gene sequence of MSP3-11.
The nucleotide sequence of pBluescript MSP3-11 is
illustrated. Numbers on the left indicate the number of
the first nucleotide in each row. Termination codon is
underlined.


Fig. 14 Immunoblots of expressed MSP3 clones.
The reactivity of expressed proteins from clones MSP3-11, MSP3-12, and MSP3-19 are
compared using an anti-MSP3 monoclonal antibody (AMG75C2) and a negative isotype
control (TRP1E1). pBlue is expressed proteins from empty pBluescript plasmid used
as negative control. FL is A. margnale initial body preparation from a FL isolate
as a positive control. Numbers on the left indicate molecular size markers in
kilodaltons.


97
the potential for cross reactivity between A. margnale and
Ehrlichia bovis, particularly since cross-reactive antigens
exists among many Ehrlichia sp (Holland and Ristic, 1993;
Ristic and Holland, 1993; Uilenberg, 1993). This is a
potential problem for serologic testing of cattle in areas
where A. marginale and E. bovis coexist.
Serum from an animal infected with C. ruminantium does
not react with an 8 6 kDa antigen, but does react with a
protein of slightly smaller molecular mass, and an antigen
which has a molecular mass of just above 30 kDa. Recently,
sequence homology has been demonstrated between genes encoding
immunodominant surface proteins of A. marginale and C.
ruminantium. Amino acid sequence analysis revealed 33%
identity and 40 similarity between the C. ruminantium MAPI and
the A. marginale MSP4 (Van Vliet, et al., 1994). Significant
amino acid sequence similarity was also revealed between the
21 kDa MAP2 protein of C. ruminantium and the 19 kDa MSP5
protein of A. marginale (Mahan et al., 1994). In addition,
extensive amino acid sequence homology (59% similarity; 33%
identity) was conserved in multiple oligopeptide sequences
throughout the MSP4 (31 kDa) protein and the MSP2 (36 kDa)
protein of A. marginale (Palmer et al., 1994). Hence,
sequence similarity was also identified between MSP2 of A.
marginale and MAPI of C. ruminantium (Palmer et al., 1994).
Considering the close phylogenetic relationship between these
2 organisms, the known amino acid sequence homology between


Eael
Bso M
FL CL
FL CL
bp
5,090 --
4,072 --
3,050 --
2,036 --
1,636 --
1,018 --
506 --
396 --
344 --
298 --
Nco I
FL
CL
bp
23,130
9,416
6,577
4,361
-- 2,332
-- 2,057
564
00
OJ


103
be identified if these antigens are to be useful as diagnostic
test antigens or vaccine candidates.
Identification of pBluescript MSP3-12 as a
Recombinant Form of an 86 kDa Antigen
Anti-MSP3 MAb AMG75C2 bound expressed protein from an
MSP3 clone, pBluescript MSP3-12. The reactivity of this MAb
has been previously described (McGuire et al., 1991). Because
this MAb also binds to one of the 86 kDa antigens seen on 2-D
gel electrophoresis (pi 5.6), we are able to identify clone
MSP3-12 as a member of the MSP3 gene family. In addition,
immune sera from an animal infected with a VA strain of A.
margnale, but not a FL strain, reacts with recombinant MSP3-
12. This is supportive evidence that MSP3-12 shares common
epitopes with the 86 kDa antigen (pi 5.6) since VA sera
strongly reacts with this antigen on 2-D immunoblots whereas
immune sera from cattle infected with a FL isolate react
poorly if at all.
The MSP3-12 clone contains the N-terminus of the protein
as well as 633 bp upstream to the start codon. The region
upstream to the open reading frame likely contains the
regulatory seguences of the MSP3 gene. This helps insure the
correct transcription and translation of the gene by E. coli
since rickettsial promotors have been shown to be recognized
by E. coli polymerases (Oaks et al., 1987). Genes containing
their own promotor regions will naturally be in the correct
reading frame. Anti-MSP3 MAb AMG75C2 does not bind to


N N ^
S
,0
1
2,33
Restriction
Cleavage site in
Size of co-migrating
Enzyme
pBluescript MSP3-12
plasmid and genomic
fragments
Neo I
12/1,188
1,176
Eae I
52/1,762
1,710
Bsp M
193/2,076
1,886


84
Hie II were chosen because they do not cut within the
sequence of the MSP3-12 gene. This should produce a single
band if only one genomic copy of MSP3 is present. Sac I cuts
once within the open reading frame of MSP3-19 (at nucleotide
498), potentially producing 2 observable bands for a single
copy gene. None of the enzymes cut within the MSP3-11 gene.
Multiple bands were observed on Southern blot
hybridizations, indicating multiple partially homologous MSP3
copies (Fig. 18) The exact number of copies cannot be
determined since restriction site polymorphism may exist in
other MSP3 copies, resulting in more than one band from a
single copy. Similar intensities of many of the bands within
a single isolate may indicate extensive homology exists
between some of the copies. Restriction fragment length
polymorphism (RFLP) is seen when comparing enzyme digests of
each of the 3 isolates of A. margnale.
Distribution of MSP3 Copies in the A. margnale Chromosome
Intact, A. margnale genomic DNA from 3 isolates (FL, SI,
& VA) was digested into large fragments with restriction
enzymes Sfi 1 and Not I, and separated by CHEF
electrophoresis. The gel was stained and photographed (Fig.
19) These fragments are nonoverlapping and have been
previously shown to represent the entire 1,250 kbp A.
margnale genome (Alleman et al., 1993). The Sfi I and Not 1


117
20. Gonzalez, E.F., R.F. Long, and R. A. Todorovic. 1978.
Comparisons of the complement-fixation, indirect
fluorescent antibody, and card agglutination tests for
the diagnosis of bovine anaplasmosis. Am. J. Vet. Res.,
39:1538-1541.
21. Goodger, W.J., T. Carpenter, and H. Reimann. 1979.
Estimation of economic loss associated with anaplasmosis
in California beef cattle. J. Am. Vet. Med. Assoc.
174:1333-1335.
22. Harlow, E. and D. Lane. 1988. Immunoassays, In:
Antibodies: A Laboratory Manual. Cold Springs Harbor
Press, Cold Springs Harbor. p. 557-592.
23. Holland, J. and M. Ristic. 1993. Equine monocytic
Ehrlichiosis, p.219-220. In Z. Woldehiwet and M. Ristic
(ed.), Rickettsial and Chlamydias Diseases of Domestic
Animals. Pergamon Press, Inc., Tarrytown, New York.
24. Jain, N.C. 1986. Hemolytic anemias associated with some
infectious agents, p. 590-600. In Jain, N.C. (ed.),
Schalm's Veterinary Hematology. 4th ed. Lea and Febiger,
Philadelphia, PA.
25. Kieser, S.T., I.S Eriks, and G.H. Palmer. 1990. Cyclic
rickettsemia during persistent Anaplasma margnale
infection of cattle. Infect. Immun. 58:1117-1119.
26. Kocan, K.M. Venable, J.H., Hsu, K.C. and W.E. Brock.
1978a. Ultrastructural localization of anaplasmal
antigens (Pawhuska isolate) with ferritin-conjugated
antibody. Am. J. Vet. Res. 39:1131.
27. Kocan, K.M., Venable, J.H. and W.E. Brock. 1978b.
Ultrastructure of anaplasmal inclusions (Pawhuska
isolate) and their appendages in intact and hemolyzed
erythrocytes and in complement-fixation antigen. Am. J.
Vet. Res. 39:1538.
28. Kreier, J.P., and M. Ristic. 1963. Anaplasmosis. KI.
Immunoserologic characteristics of the parasites present
in the blood of calves infected with the Oregon strain of
Anaplasma margnale. Am. J. Vet. Res. 24:688-696.
29. Kuttler, K.L. 1981. Diagnosis of anaplasmosis and
babesiosis an overview, p. 245. In Hidalgo, R.J. &
Jones, E.W. (eds.), Proceedings of the Seventh National
Anaplasmosis Conference. Mississippi State University.


Copyright 1995
by
Arthur Rick Allexnan


112
infected with Ehrlichia sp. We have also shown that this
antigen is not conserved among various strains of A.
margnale, and that antigenic variation likely exists between
isolates since immune sera from cattle infected with different
strains reacted differently to MSP3.
It has now been shown that 2 major surface antigens of A.
marginale, MPS2 & MSP3, are actually composed of a family of
related proteins. Using an MSP3 clone as a probe in
hybridization studies we concluded the multiple MSP3 antigens
are the result of a complex, multigene family of partially
homologous genes, similar to the MSP2 protein. Although we
cannot state definitively, we estimate that a relatively large
portion of this rickettsial agent's small genome (up to 4%) is
occupied by these 2 gene families. We hypothesize the
organism uses these multigene families to antigenically vary
these major immunogenic surface proteins.
The cross reactivity of this protein with sera from
animals infected with Ehrlichia sp., the size polymorphism of
MSP3 between different geographic isolates, the multiple 86
kDa antigens recognized by various antisera, and the presence
of a multigene family encoding these antigens indicate that in
its native form, a single recombinant MSP3 would not be a
suitable candidate for use as a diagnostic test antigen. In
order to be used as a test antigen, it may be necessary to
define and obtain expressed copies of the MSP3 gene. The
potential for using multiple recombinant MSP3 antigens


Fig. 1 Diagram of MSP3 clones.
A schematic representation of clones MSP3-11, MSP3-12, and MSP3-19. Homologous
regions are indicated by like shaded areas. Nucleotide numbers are indicated on
the bottom. Top numbers of clones MSP3-11 and MSP3-19 indicate corresponding
nucleotides in clone MSP3-12.


Fig. 13 Comparison of MSP3 and MSP2 protein sequences.
BestFit alignment of amino acid sequences of MSP3 (top)
and MSP2 (bottom). Identical amino acids are indicated
by a vertical line. Conservative substitutions are
indicated by an asterisk (*). The symbol is used
to denote a gap used to achieve optimal alignment between
the sequences. Amino acid numbers are indicated at the
beginning and end of each line.


membranes, and probed with a digoxigenin-labeled, cloned gene
of MSP3 (courtesy of Dr. G. Palmer, W.S.U.).
Immunoblots demonstrated cross reactivity between MSP3
and sera from animals infected with A. ovis, E. risticii, and
E. ewingii. Size polymorphism of MSP3 was seen between
different geographic isolates of A. margnale. Two-
dimensional gel electrophoresis revealed at least 3 different
antigens migrating at the 86kDa molecular mass, and sera from
animals infected with different isolates reacted with
different 86kDa antigens. Hybridization studies with a cloned
MSP3 gene identified multiple copies of the gene in the
genome. These results indicate MSP3 is, 1) cross reactive
with A. ovis and some Ehrilichia sp., 2) not conserved between
different isolates of A. margnale, and 3) in at least the FL
isolate, MSP3 is actually a group of 3 or more 86kDa proteins
with different isoelectric points. These data also suggest A.
margnale may antigenically vary this immunodominant protein
by use of a complex multigene family. The variability of MSP3
between isolates, the multiple 86kDa antigens in the FL
isolate, and the multiple copies of the MSP3 gene indicate a
single recombinant form of MSP3 may not be a suitable
diagnostic test antigen. To be used as a diagnostic test
antigen, conserved epitopes between copies of MSP3 genes may
need to be identified and tested for reactivity with immune
sera.
xii


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.

John B. Dame, Chair
Associate Professor of
Vterinary Medicine
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Anthony F. Barbet
Professor of Veterinary
Medicine
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
John W.\ Harv£
Professor cyt Veterihary
Medicine (.
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Rose E. Raskin
Associate Professor of
Veterinary Medicine
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in acope and quality, as
a dissertation for the degree of Doctor of; Philpsophy.
Paul Gulig
Associate Professor
Molecular Genetics
Microbiology


31
Two-Dimensional Gel Electrophoresis
Isoelectric focusing gels were prepared per
manufacturer's instructions (Protean II Slab Cell Instruction
manual, Bio-Rad Laboratories, Richmond, CA) Briefly, an
acrylamide\N,N'-methylene-bis-acrylamide solution containing
9.5 M urea, 2.0% nonidet P-40(v/v), 4.1% acrylamide/bis (30.8%
T/2.6% C) 10 mM (3-[(3-cholamidopropyl)dimethylammonio]-
lpropanesulfonate, CHAPS), and 5.8% Bio-Lyte 5/7 (v/v) was
polymerized in glass tubing 3 mm x 140 mm. Initial body
preparations were incubated for 2 hours at room temperature in
4 times their volume of the previously described lysis buffer
and 5 times their volume of sample buffer containing 9.5 M
urea, 2.0% Triton X-100, 5% /3-mercaptoethanol, 1.6% Bio-Lyte
5/7, and 0.4% Bio-Lyte 3/10. The solution was then
centrifuged at 100,000 x g for 2 hrs. at 25C. A volume of
the supernatant containing 20/ig of protein was loaded onto
each tube gel and overlayed with 50/il of overlay buffer
containing 9.5 M urea, 0.8% Bio-Lyte 5/7, 0.2% Bio-Lyte 3/10,
and 0.0025% bromophenol blue. Tube gels were electrophoresed
at 400 volts for 16 hrs. and 800 volts for 2 hrs. using the
BRL model V16 vertical gel electrophoresis system (Bethesda
Research Laboratories, Gaithersburg, MD). Tube gels were
extruded from the glass tubes and eguilibrated for 5 min. in
buffer containing 0.0625 M Tris-HCl, pH 6.8, 10.0% glycerol,
2.0% SDS (w/v) 5.0% /3-mercaptoethanol, and 0.00125%


10
treatment (Luther et al., 1980; Richey, 1981). Several types
of serological tests have been described for the diagnosis of
A. margnale, including complement fixation (CF), capillary
tube agglutination (CT), card agglutination (CD), indirect
immunofluorescence (IIF), and ELISA (Amerault & Roby, 1968;
Barry et al., 1986; Duzgun et al., 1988; Gonzalez et al. ,
1978; Kuttler, 1981). Error rate with these tests is high,
primarily because antigens used in these tests are a crude
mixture of A. margnale and erythrocyte material (Kocan et
al., 1978; Kocan et al., 1978). False positives as high as
20% are seen due to poor specificity of test antigen (Amerault
& Roby, 1968; Amerault et al., 1973; Barry et al., 1986;
Duzgun et al., 1988; Gonzalez et al., 1978; Luther et al.,
1980; Todorovic et al., 1977). Poor sensitivity of these
tests results in false negatives as high as 21% (Amerault et
al., 1973; Barry et al., 1986; Goff et al., 1990; Gonzalez et
al., 1978; Luther et al., 1980; Maas et al., 1986). Current
technology, which was not available when previous tests were
developed, allows us to economically produce a sensitive and
specific ELISA test for the rapid field diagnosis of A.
margnale using purified, recombinant, A. margnale protein.
As previously mentioned, several types of tests have been
developed for the diagnosis of anaplasmosis. The tests which
have received most attention have been the rapid card
agglutination test (CA), the complement fixation test (CF) ,


57
trypanosome surface protein were used as negative controls and
showed no reactivity to MSP3 (Data not shown).
In a homologous reaction with anti-FL serum, 2 major
areas of reactivity were seen with a molecular mass of 86 kDa,
one with an apparent isoelectric point (pi) of 6.5, and the
other with a pi of approximately 6.2 (Fig. 7). There was
slight reactivity with an antigen at a pi of approximately
5.6. When the initial body preparation from the FL isolate
was reacted with the anti-MSP3 MAb, major reactivity was seen
in the 5.6 area of the pH gradient, but no reactivity was
noted with antigens at a pi of 6.5 or 6.2 (Fig. 7).
Serum from an animal infected with a VA isolate showed
similar reactivity as the MAb. However, serum from an animal
infected with a WA isolate reacted with 2 antigens in an
entirely different area of the pH gradient, having pis of
approximately 5.1 and 5.3 (Fig. 7). When these same initial
body preparations were reacted with antiserum from an animal
infected with a SI isolate reactivity was noted in all 3 areas
of the pH gradient, with approximate pis of 6.5 to 6.2, 5.6
and 5.3 to 5.1 (Fig. 7). When a rabbit-anti-MSP3 polyclonal
sera was used, reactivity was noted in areas of the pH
gradient having pis of 6.5 to 6.2 and 5.6 (Fig. 8).
Although conditions seen here were not optimized to
separate the 36 kDa proteins of A. margnale (MSP2), multiple
spots were visualized in that apparent molecular size (Fig.
7). Some variation in reactivity of the different antisera to


EVALUATION OF ANAPLASMA MARGINALE
MAJOR SURFACE PROTEIN 3 (MSP3) AS A DIAGNOSTIC TEST ANTIGEN
By
ARTHUR RICK ALLEMAN
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
1995

Copyright 1995
by
Arthur Rick Allexnan

This document is dedicated to my beautiful wife Mary, my
wonderful children, Arthur and Grace, and to God who we love
and serve.

ACKNOWLEDGMENTS
I would like to acknowledge my committee members, Dr.
John Dame, Dr. John Harvey, Dr. Paul Gulig, and Dr. Rose
Raskin, for their guidance and assistance in the completion of
this project. I would like to express a special appreciation
to my major professor, Dr. Tony Barbet. He has been a great
instructor, mentor, guidance counselor, and friend. He has
always been there for me as well as for his other graduate
students and staff. His devotion of his time in a completely
unselfish manner has been an inspiration to me on many
occasions, particularly in difficult times. His weekly
individual meetings, and monthly lab meetings were a lesson
well learned and something I hope to continue during my
professional career. His guidance was most valuable to the
completion of these experiments.
I would also like to express my appreciation to some of
my coworkers, Dr. Bill Whitmire, Annie Moreland, Renee
Blentlinger, Anna Lundgren, and Michael Bowie. We all worked
together harmoniously in the lab, each willing to take time
out to assist the other, sharing valuable work experiences.
I am glad friendships last longer than Ph.D. projects.
IV

TABLE OF CONTENTS
ACKNOWLEDGMENTS iv
LIST OF FIGURES V
ABBREVIATIONS viii
ABSTRACT X
CHAPTER 1 INTRODUCTION 1
CHAPTER 2 LITERATURE REVIEW 5
General Methods of Serologic Diagnosis 5
Methods for Detection of A. margnale Infected
Cattle 9
The Rapid Card Agglutination Test (CA) ... 11
The Complement Fixation Test (CF) 13
The Indirect Immunofluorescence Test (IIF) 14
The Radioimmunoassay (RIA) 15
Enzyme-Linked Immunosorbent Assays (ELISA) 16
Nucleic Acid Probe Hybridization 21
Previous Experiments 22
CHAPTER 3 MATERIALS AND METHODS 26
Anaplasma margnale Strains 26
Initial Body Preparation 27
Babesia bovis and Babesia bigemina Antigen
Preparation 28
Antisera used for Immunoblots 29
SDS-PAGE 30
Two-Dimensional Gel Electrophoresis 31
Immunoblots with Antisera 32
Purification of A. margnale Genomic DNA 33
MSP3 Clones 33
Verification of Recombinant MSP3 34
Digoxigenin Labeling of pBluescript MSP3-12 ... 37
Representation of pBluescript MSP3-12 38
Restriction Enzyme Digestion 38
Southern Blots 39
Presence of Multiple MSP3 Gene Copies 42
Distribution of MSP3 Copies in the A. margnale
Chromosome 4 3
v

Restriction Enzyme Digestion in Agarose
Plugs 43
CHEF Gel Electrophoresis 44
CHAPTER 4 RESULTS 46
Specificity Experiments 46
Conservation of MSP3 Between Different
Geographic Isolates of A. marginale 53
Immune Response to MSP3 56
MSP3 Nucleotide Sequence and Translation 62
MAb and Immune Sera Reactivity to Recombinant
MSP3 74
Representation of pBluescript MSP3-12 in the
A. marginale Genome 79
Presence of Multiple MSP3 Gene Copies 79
Distribution of MSP3 Copies in the A. marginale
Chromosome 84
CHAPTER 5 DISCUSSION 92
Rational for Studying the MSP3 Antigen 92
The Specificity of MSP3 in Detecting A. marginale
Infection 94
Size Polymorphism of MSP3 Among Various
Strains of A. marginale 98
Immune Response to MSP3 100
Identification of pBluescript MSP3-12 as a
Recombinant Form of an 86 kDa Antigen 103
pBluescript MSP3 is an Accurate Representation of
a Genomic Copy of MSP3 104
Multiple MSP3 Copies in the A. marginale Genome 105
Distribution of MSP3 Copies in the A. marginale
Genome 110
Summary and Conclusions Ill
REFERENCE LIST 115
BIOGRAPHICAL SKETCH 123
vi

LIST OF FIGURES
Figure page
1 Diagram of MSP3 clones 35
2 Representation of clone MSP3-12 in the A. margnale
genome 40
3 Specificity experiments using immunoblots .... 47
4 Specificity experiments using immunoblots .... 49
5 Specificity experiments using immunoblots .... 51
6 Size polymorphism of MSP3 54
7 2-D gel electrophoresis of A. margnale proteins 58
8 2-D gel electrophoresis of A. margnale proteins 60
9 Gene sequence of MSP3-11 63
10 Gene sequence of MSP3-12 65
11 Gene sequence of MSP3-19 67
12 Diagram of MSP3 clones with MSP2 homology .... 70
13 Comparison of MSP3 and MSP2 protein sequences 72
14 Immunoblots of expressed MSP3 clones 75
15 Immunoblots of expressed MSP3-12 clones 77
16 Representation of clone MSP3-12 in the A. margnale
genome 8 0
17 Genomic representation of MSP3-12 82
18 Presence of multiple copies of MSP3 85
19 Not 1 and Sf I digestion of the A. margnale
genome 88
20 Distribution of MSP3 in the A. margnale genome 90
vii

ABBREVIATIONS
bp
BSA
C
CD
CF
CHEF
CT
DNA
ECL
EDTA
ELISA
Fig.
FL
g
HC1
HRP
I FA
IIF
IPTG
kbp
kDa
base pair (s)
bovine serum albumin
degrees Celsius
card agglutination test
complement fixation test
clamped homogeneous electric field electrophoresis
capillary tube agglutination test
deoxyribonucleic acid
enhanced chemiluminescence
ethylenediaminetetraacetic acid
enzyme-linked immunosorbent assay
figure (s)
Florida
gram (s)
hydrochloric acid
horseradish peroxidase
indirect fluorescent antibody
indirect immunofluorescence test
isopropylthio-/3-galactosidase
kilobase pair (s)
kilodalton
viii

M
molar
MAb
monoclonal antibody (s)
MAPI
major antigenic protein 1
mg
milligram (s)
min.
minute (s)
ml
milliliter (s)
mM
millimolar
MSPla
major surface protein la
MSP1/3
major surface protein 1/3
MSP2
major surface protein 2
MSP3
major surface protein 3
N
normal
NaCl
sodium chloride
NaOH
sodium hydroxide
ng
nanogram (s)
NP-40
nonidet P-40
PBS
phosphate buffered saline
PCV
packed cell volume
PI
post-infection
PMSF
phenylmethylsufonly fluoride
RFLP
restriction fragment length polymorphism
RIA
radioimmunoassay
rRNA
ribosomal ribonucleic acid
SDS-PAGE
sodium dodecyl sulfate, polyacrylamide gel
electrophoresis
sec.
seconds
ix

SI
TBE
TE
Tris
TWEEN 20
TX
V
VA
w/v
WA
Mg
Ml
South Idaho
tris-borate-EDTA buffer
tris-EDTA buffer
tris hydroxymethyl aminoethane
polyoxyethylene-sorbitan monolaurate
Texas
volts
Virginia
weight/volume
Washington
microgram (s)
microliter (s)
x

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
EVALUATION OF ANAPLASMA MARGINALE MAJOR SURFACE
PROTEIN 3 (MSP3) AS A DIAGNOSTIC TEST ANTIGEN
By
Arthur Rick Allemn
August 1995
Chairperson: John Dame, PhD
Major Department: Veterinary Medicine
The immunodominant surface protein, MSP3, has been
proposed as an antigen suitable for the diagnosis of bovine
anaplasmosis. In this study we further characterized MSP3 to
examine its potential as a test antigen for the serological
detection of carrier cattle. The specificity of MSP3 was
evaluated by probing immunoblots of A. margnale proteins with
immune sera from animals infected with related organisms.
Similarly, we used polyacrylamide gel electrophoresis (SDS-
PAGE) and immunoblots to evaluate the conservation of MSP3
between 4 different geographic isolates of A. margnale. In
addition, proteins from a FL isolate were separated by 2-
dimensional gel electrophoresis, and immunoblotted with immune
sera from cattle infected with one of 4 different geographic
isolates of A. margnale. Genomic A. margnale DNA was
digested with restriction endonucleases, transferred to nylon
xi

membranes, and probed with a digoxigenin-labeled, cloned gene
of MSP3 (courtesy of Dr. G. Palmer, W.S.U.).
Immunoblots demonstrated cross reactivity between MSP3
and sera from animals infected with A. ovis, E. risticii, and
E. ewingii. Size polymorphism of MSP3 was seen between
different geographic isolates of A. margnale. Two-
dimensional gel electrophoresis revealed at least 3 different
antigens migrating at the 86kDa molecular mass, and sera from
animals infected with different isolates reacted with
different 86kDa antigens. Hybridization studies with a cloned
MSP3 gene identified multiple copies of the gene in the
genome. These results indicate MSP3 is, 1) cross reactive
with A. ovis and some Ehrilichia sp., 2) not conserved between
different isolates of A. margnale, and 3) in at least the FL
isolate, MSP3 is actually a group of 3 or more 86kDa proteins
with different isoelectric points. These data also suggest A.
margnale may antigenically vary this immunodominant protein
by use of a complex multigene family. The variability of MSP3
between isolates, the multiple 86kDa antigens in the FL
isolate, and the multiple copies of the MSP3 gene indicate a
single recombinant form of MSP3 may not be a suitable
diagnostic test antigen. To be used as a diagnostic test
antigen, conserved epitopes between copies of MSP3 genes may
need to be identified and tested for reactivity with immune
sera.
xii

CHAPTER 1
INTRODUCTION
Anaplasma margnale is an arthropod-borne, rickettsial
hemoparasite which invades the red blood cells of cattle
causing a clinical disease known as anaplasmosis. This
organism is the representative species of the genus Anaplasma,
in the family Anaplasmataceae, order Rickettsiales (Ristic and
Krier, 1974) The genus name, Anaplasma, refers to the
appearance of being devoid of cytoplasm, and the species name,
margnale, refers to the location of the organism in the
margin of infected erythrocytes. A. margnale was originally
discovered in cattle experimentally infected with Babesia
bigemina, and was thought to be a developmental stage of this
organism (Smith and Kilborne 1893) However, several years
later these marginal inclusions noted in the erythrocytes of
cattle infected with B. bigemina were conclusively identified
as agents belonging to another genus, Anaplasma (Theiler
1910).
A. margnale is now known to have a global distribution
which includes the United States of America. World-wide
economic losses are difficult to calculate, but losses in the
U.S. alone are estimated to be over 100 million dollars
annually (Goodger et al., 1979; McCallon, 1973). In certain
1

2
areas of the U.S., the incidence of infected cattle may be as
high as 37% (Maas et al., 1986; McCallon, 1973).
Acute anaplasmosis is usually seen in cattle over 1 year
of age and is characterized by a severe hemolytic anemia,
resulting in weight loss, abortion, decreased milk production,
and often death in infected animals over 3 years of age
(Wanduragala & Ristic, 1993). In acute stages, the disease is
easily diagnosed by finding organisms on routine blood smear
evaluation. However, animals that survive the infection will
remain carriers and maintain a low level of parasitemia which
cannot be detected microscopically (Richey, 1981; Zaugg et
al., 1986). These carrier cattle serve as a perpetual source
of infection for susceptible cattle (Swift & Thomas, 1983).
Cyclic rickettsemia has been detected and quantitated in
carrier cattle using nucleic acid probe hybridization (Eriks
et al., 1993; Kieser et al., 1990). These studies
demonstrated rickettsemia levels in persistently infected
cattle fluctuated at approximately 5 week intervals from a low
of 104 to a high of 107 infected erythrocytes per ml of blood.
Although the level of parasitemia was too low to be detected
microscopically, uninfected Dermacentor andersoni ticks were
able to acquire infection from the cattle at infectivity rates
of up to 80% during the higher rickettsemia levels (107
infected erythrocytes per ml of blood) (Eriks et al., 1993).
Even at extremely low levels of parasitemia 27% of the male
ticks became infected. In addition, once ticks acquire the

3
infection, replication of the organism in the vector allowed
easy transmission of the disease with only a few infected
ticks regardless of the initial infecting dose. This firmly
establishes the important role persistently infected carrier
cattle play in the transmission of the disease. In order to
reduce economic losses associated with anaplasmosis, control
efforts must include an effective way of identifying and
decreasing transmission from carrier cattle.
An inexpensive, sensitive, and specific field test for
the identification of A. margnale infected carriers would
have a tremendous impact on limiting the spread and
consequently the economic losses associated with this disease.
Entire herds could be easily tested and identified carriers
removed or treated with oxytetracycline, thus eliminating the
source of infection for susceptible cattle. A test such as
this would also provide an accurate means of identifying
carrier animals being shipped into nonendemic or noninfected
areas. Ideally, animals infected with the less virulent
species, Anaplasma cntrale, should not contain antibodies
that will cross react with the antigen used in this test.
This would be a marked improvement over other serological
tests which cannot distinguish between the two infections. In
addition, in areas where vaccination with A. cntrale is used
as a means of prevention of A. margnale, this test may
distinguish vaccinated from infected animals.

4
The purpose of this study was to evaluate the MSP3
protein of A. margnale, and determine if a recombinant form
of this protein would be a suitable diagnostic test antigen to
detect A. margnale infection in carrier cattle.

CHAPTER 2
LITERATURE REVIEW
General Methods of Serologic Diagnosis
Serology is the science of detection of specific
antibodies in body fluids, particularly, though not
exclusively, serum. There are 3 broad categories of serologic
techniques, the primary binding tests, secondary binding
tests, and tertiary binding tests. Primary binding tests
allow antigen and antibody to combine, and the resulting
immune complexes are measured using radioisotopes, fluorescent
dye, or enzyme labels. Examples of the primary binding tests
are radioimmunoassays, immunofluorescence assays, and enzyme-
linked immunosorbent assays. Primary binding tests are the
most sensitive of the serologic techniques in terms of ability
to detect smaller amounts of specific antibodies (Tizard,
1992) .
Radioimmunoassays are widely used primarily because of
their extreme sensitivity and ability to detect small amounts
of antigen or antibody. In these assays radioactive isotopes
such as 123I are used to label antigens or antibodies, and the
level of radioactivity is used for quantitation.
Disadvantages of this test are the dangers and restrictions
5

6
regarding the use of radioactivity, the complexity of the test
procedure, and the need for specialized equipment. Because of
their extreme sensitivity, these tests are frequently used to
measure trace amounts of drugs in urine or other body fluids
(Tizard, 1992).
Immunofluorescence assays employ the use of fluorescent
dyes such as fluorescein isothiocyanate (FITC) to measure the
formation of immune complexes. FITC is easily conjugated to
immunoglobulins for detection using dark field microscopy with
an ultraviolet light source, or flow cytometry. Specific
immunoglobulins may be labeled directly and allowed to react
with an antigen as in direct fluorescent antibody tests, or
species specific antiglobulins may be labeled with FITC and
used to bind to antibody/antigen complexes in indirect
fluorescent assays. The indirect tests are usually more
sensitive since each antibody molecule bound to the antigen
may bind several labeled anti-globulin molecules (Tizard,
1992) .
Enzyme-linked immunosorbent assays (ELISA) employ the use
of enzymes for the detection of antigen/antibody interactions.
The three most commonly used enzymes are alkaline phosphatase,
horseradish peroxidase, and /3-galactosidase. In these
techniques the enzymes are chemically linked to
immunoglobulins or anti-globulins. Detection is accomplished
by the addition of an enzyme substrate to the reaction,

7
producing a colored product. The color change may be
estimated visually or determined spectrophotometrically.
There are several variations of the ELISA test; the
direct ELISA which utilizes enzyme-linked immunoglobulins, the
indirect ELISA which uses enzyme-linked anti-globulins, and
the competitive ELISA which employs the use of a labeled
monoclonal antibody to compete with specific antibodies in the
test sample for a single epitope on an antigen molecule. Some
ELISAs bind antibody to a solid phase media to capture a
particular antigen (antigen capture ELISA), while other
techniques bind antigen to the solid phase in order to capture
and detect antibody (antibody capture ELISA) (Harlow & Lane,
1988). Various solid phase media may be used such as
nitrocellulose membranes or polystyrene microtiter plates,
depending on the purpose of the test and the nature of the
material to be tested. The extreme versatility of this
technique, its excellent sensitivity (comparable to RIA), and
its simplicity make it one of the most widely used
immunodiagnostic techniques for many viral, bacterial, and
parasitic infections. In addition, it does not involve the
use of hazardous radioactive isotopes. Because of its extreme
sensitivity, specificity may be a problem with these tests,
particularly if unpurified test antigens are used to detect
specific antibodies in polyclonal sera.

8
Secondary binding tests measure the interaction of
antigen and antibody by in vitro visualization of a secondary
event which occurs as a result of immune complex formation.
Such events include the precipitation of soluble antigens in
solution by immune complex formation, agglutination of
particulate antigen on the surface of bacteria or
erythrocytes, and the activation of the complement pathway,
resulting in cell lysis. Many of these tests require optimal
concentration of antigen and antibody to allow visualization,
and false results may occur in cases of antibody or antigen
excess (Nakamura et al., 1988). As a result of this, and the
gross visual detection required for many of these tests,
secondary binding tests lack the sensitivity of primary
binding assays. Examples of these tests include
immunoprecipitation, immunodiffusion, agglutination, and
complement fixation.
Tertiary binding tests actually measure the in vivo
protective effects specific antibodies may have in an animal.
These test measure the biological activity of the antibody to
determine their ability to protect a susceptible animal from
an infectious agent or neutralize the effects of an antigen.
These tests are most useful in experimental trials or in
treatment of certain diseases by passive transfer of antibody.
They are not practical or suitable for the diagnosis of
disease processes.

9
The suitability of a particular type of immunodiagnostic
test depends on many factors such as the use of test (field
versus laboratory use), the sensitivity and specificity
required, the prevalence of the disease, the number of samples
to be tested at any one time, and the speed at which a test
must be performed. In general, selection of a diagnostic test
often involves a compromise trade off between sensitivity,
specificity, and ease of performance. As previously
mentioned, the indirect ELISA is a very sensitive test, easy
to perform, and may be adapted to field use where multiple
samples may be tested simultaneously. It is a primary binding
assay and requires no specialized equipment to perform the
test or interpret the results. This may prove to be an ideal
test for the field detection of A. margnale infected carrier
cattle. Because of the extreme sensitivity of the indirect
ELISA, specificity can be a problem when detecting specific
antibodies in a polyclonal sera. To achieve acceptable
specificity, great care must be taken in selection and
purification of an appropriate test antigen.
Methods for Detection of A. margnale Infected Cattle
Currently an effective vaccine against A. margnale is
not available, and the inability of present serologic tests to
accurately detect persistently infected, carrier cattle
severely compromises efforts to establish disease free herds
and reduce economic losses through testing, isolation, and

10
treatment (Luther et al., 1980; Richey, 1981). Several types
of serological tests have been described for the diagnosis of
A. margnale, including complement fixation (CF), capillary
tube agglutination (CT), card agglutination (CD), indirect
immunofluorescence (IIF), and ELISA (Amerault & Roby, 1968;
Barry et al., 1986; Duzgun et al., 1988; Gonzalez et al. ,
1978; Kuttler, 1981). Error rate with these tests is high,
primarily because antigens used in these tests are a crude
mixture of A. margnale and erythrocyte material (Kocan et
al., 1978; Kocan et al., 1978). False positives as high as
20% are seen due to poor specificity of test antigen (Amerault
& Roby, 1968; Amerault et al., 1973; Barry et al., 1986;
Duzgun et al., 1988; Gonzalez et al., 1978; Luther et al.,
1980; Todorovic et al., 1977). Poor sensitivity of these
tests results in false negatives as high as 21% (Amerault et
al., 1973; Barry et al., 1986; Goff et al., 1990; Gonzalez et
al., 1978; Luther et al., 1980; Maas et al., 1986). Current
technology, which was not available when previous tests were
developed, allows us to economically produce a sensitive and
specific ELISA test for the rapid field diagnosis of A.
margnale using purified, recombinant, A. margnale protein.
As previously mentioned, several types of tests have been
developed for the diagnosis of anaplasmosis. The tests which
have received most attention have been the rapid card
agglutination test (CA), the complement fixation test (CF) ,

11
indirect immunofluorescence test (IIF), the radioimmunoassay
(RIA), and various enzyme-linked immunosorbent assays (ELISA).
The Rapid Card Agglutination Test (CA)
The CA is one of the most widely used in the field
because it is easy to run, requires minimal equipment, and
uses unheated sera or heparinized plasma (Amerault & Roby,
1968). Beinq a secondary binding test, this assay lacks the
sensitivity of primary binding assays. Initial studies using
the CA test reported 100% sensitivity in detecting carrier
cattle and 86% sensitivity in detecting all known infected
cattle (Amerault & Roby, 1968). However, the number of
positive cattle tested was small (22) and subsequent studies
comparing sensitivity and specificity of the CA test with
other serological tests indicates sensitivity to be 84%
(Gonzalez et al., 1978). Yet another study indicated the CA
test was able to detect only 6 of 9 (66.6%) known carrier
cattle (Luther et al., 1980). In addition, 89.3% of
noninfected, A. margnale vaccinated cattle had positive
reactions with the CA test for up to 15 months after
vaccination (Luther et al., 1980).
Specificity of the CA test was initially reported to be
100% with none of the 24 normal cattle sera testing positive
(Amerault & Roby, 1968). This test uses a crude preparation
of A. margnale and erythrocyte antigen, and problems of
specificity are likely to be encountered. Nonspecific

12
agglutination did occur when fresh serum was used; however,
this problem did not occur when using heparinized plasma.
Additional studies involving large numbers of serum samples
(380) established 98% specificity for the CA test (Gonzalez et
al., 1978). However, none of these studies investigated the
potential for cross reactivity with sera from cattle infected
with other hemoparasites. Current technology, which was not
available when previous tests were developed, allows the
phylogenetic relatedness of various species and genera to be
determined. Using 16s rRNA sequence analysis investigators
have determined A. margnale to be closely related to other
Anaplasma sp., Ehrlichia sp., and Cowdria ruminantium (Dame et
al., 1992; Van Vliet et al., 1992). To accurately determine
the specificity of a test antigen, it is essential to test the
ability of the antigen to distinguish between infections of
related organisms. This was not done for any of the antigens
currently used in serologic diagnosis of A. margnale.
Therefore estimations of specificity for these tests are
unreliable.
Although the CA test is well adapted for field use, its
poor sensitivity (probably because it is a secondary binding
test) and its questionable specificity (because of its use of
a crude test antigen preparation) make it an undesirable test
for the rapid field identification of carrier cattle.

13
The Complement Fixation Test (CF)
The CF is another widely used test for the detection of
antibodies to A. margnale in infected cattle. This test
relies on the ability of A. margnale antibodies in the serum
to fix complement and lyse target cells on which the antigen
is associated (Kuttler & Winward, 1984) The antigen used in
this test is again a crude mixture of A. margnale and
erythrocyte proteins, and since it is a secondary binding
test, sensitivity is not as high as with primary binding
assays. Sensitivity of this assay is reported to be low,
varying from 10% to 79% (Gonzalez et al., 1978; Goff et al.,
1990) In addition, it has been shown that CF antibodies
decline rapidly after natural infections with A. margnale
(Todorovic et al., 1977). CF antibodies decreased to very low
levels (1:10) as soon as 10 weeks post-infection (PI), and
below the sensitivity level of the test (1:5) by 14 weeks PI
(Gonzalez et al., 1978) This may be because CF relies on the
presence of IgM antibodies, which are better able to fix
complement than IgG. IgM levels are elevated in acute
infections, but may decline after acute episodes subside.
These results would indicate this test is unreliable in
detecting carrier cattle. In another study, 65 cattle from a
known A. margnale infected herd were tested by CF and DNA
probe hybridization (Goff et al., 1990). Stained blood smears
prepared from these cattle detected no organisms. However, 64
of the 65 cattle blood samples tested positive by DNA probe

14
hybridization and 60 of the 65 sera tested positive by IIF,
while only 5 of the 65 tested positive by CF.
The specificity of the CF test in detecting non-infected
cattle is very high, up to 100% (Gonzalez et al., 1978). This
is not surprising given the low test sensitivity. However,
the study again looked at specificity involving the detection
of CF antibodies in normal, non-infected cattle, not cattle
which may be infected with other closely related
hemoparasites. Therefore, cross reactivity with cattle
infected with other rickettsial or hematoprotozoal agents may
occur. In addition, as with the CA test, the CF test could
not distinguish between infected animals and animals
vaccinated with the killed A. margnale vaccine (Luther et
al., 1980).
The CF test is a tedious test to perform and reguires
considerable technical skill and knowledge. Therefore, the
complexity of the test and its poor sensitivity, particularly
in detecting infected carriers makes it a poor choice for the
field identification of infected animals.
The Indirect Immunofluorescence Test (IIF)
The indirect immunofluorescence test (IIF), being a
primary binding assay, is much more sensitive than the CA or
CF tests. This test uses A. margnale infected erythrocytes
fixed to a microscope slide and permits reaction to them using
a patient's serum. Antibodies to A. margnale on infected

15
erythrocytes are then visualized using a fluorescein-
conjugated anti-bovine, rabbit immunoglobulin (Gonzalez et
al., 1978). Sensitivity in detecting subclinically infected
animals is reported to be 97% with lower limits of sensitivity
not being reached by 18 weeks PI. In another study involving
64 cattle naturally infected with A. margnale and confirmed
by DNA probe hybridization, 94% of cattle were positively
identified by IIF (Goff et al., 1990). In this study
circulating organisms were not seen in stained blood smears,
but length of infection was undetermined.
Although sensitivity with IIF is much improved over CA
and CF tests, the specificity is somewhat reduced with 10% of
normal, non-infected cattle testing positive by IIF (Gonzalez
et al., 1978). In addition, cross reactivity in cattle
infected with related organisms may also present a
problem. The sensitivity of this test is satisfactory,
however, its questionable specificity, labor intensive nature,
and need for specialized reagents and equipment make it
undesirable as a field test for routine diagnosis and
identification of A. margnale infected carriers.
The Radioimmunoassay CRIA)
Recently a radioimmunoassay (RIA) was developed for the
detection of A. margnale antibodies in sera of infected
cattle (Schuntner & Leatch, 1988). This test initially
demonstrated high specificity and sensitivity (98.8% for each)

16
when testing a large number of A. margnale infected cattle,
normal cattle, and cattle infected with B. bigemina, B. bovis,
and Theileria orientalis. This test utilizes a crude A.
margnale antigen preparation isolated from infected
erythrocytes. Reactants are identified using 125I-labeled,
anti-bovine IgG, rabbit immunoglobulin and an automated gamma
counter. As expected, sensitivity with this primary binding
assay is high; however, substantial numbers of false positives
(up to 37%) occurred unless sera was pre-absorbed with normal
bovine erythrocytes and sonicated B. bovis antigen. Controls
for cross reactivity using pre-absorption should not be
necessary if purified antigens are used, even with tests as
sensitive as RIAs.
Even though the RIA is highly sensitive and specific, it
is too labor intensive to be used as a practical field test
where large numbers of samples need to be processed. In
addition, the use of radioactive material and the need for
specialized equipment limits its use to reference or research
laboratories.
Enzyme-Linked Immunosorbent Assays (ELISA)
Several ELISAs for measuring antibody to A. margnale
have been developed over recent years (Barry et al., 1986;
Duzgun et al., 1988; Montenegro-James et al., 1990; Nakamura
et al., 1988; Trueblood et al., 1991; Winkler et al., 1987).
Because these are primary binding assays, the sensitivity of

17
most of these tests are good. However, none of these tests
use a single, purified A. margnale antigen, all use a crude
mixture of A. margnale antigens prepared from initial bodies.
Therefore, test specificity has been a problem, as is
frequently encountered using a mixture of proteins as a test
antigen in a highly sensitive technique.
Barry et al., 1986 developed one of the early ELISAs for
diagnosis of A. margnale infection in cattle. Using a crude
mixture of A. margnale initial bodies and cell membranes from
ghost RBC's as a test antigen, the test reportedly was able to
accurately distinguish cattle free from infection from
recently infected animals (up to 8 months duration). However,
no measures were taken to insure the positive or negative
status of most of the animals tested. Also, test accuracy for
identifying carrier cattle was questionable having identified
only 2 of 3 cattle infected > 3 years. In addition, the test
appeared to be fairly insensitive, necessitating the testing
of sera at a dilution of 1:100. Using concentrated serum for
testing may cause a problem with test specificity and 14.6% of
cattle inoculated twice with B. bovis-infected erythrocytes
were positive for A. margnale using this ELISA. Cross
reactions were attributed to antibodies produced against RBC
antigens in the inoculum. Experiments investigating cross
reactivity with other closely related rickettsial or
hematoprotozoal agents were not performed.

18
Another diagnostic test utilized the CF test antigen in
particulate and SDS-solubilized form as a test antigen in an
ELISA (Winkler et al., 1987). This study indicated SDS-
solubilization of the proteins decreased background reactivity
in the test and increased test sensitivity in detecting
positive reactants. Complete correlation between the ELISA
and CF test was found when using solubilized test antigen and
CF test positive and negative reference sera. In testing
other infected cattle, false negatives were not observed;
however, cross reactivity ranged from 50% to 80% when testing
sera from cattle infected with or immunized against different
infectious agents.
Nakamura et al., 1988, developed an ELISA test antigen by
nitrogen decompression of infected cells to isolate initial
bodies, and solubilized them in triton X-100. Test
specificity and sensitivity was determined by comparison of
results with the CF test. Specificity of this ELISA was
satisfactory having 100% agreement with sera tested negative
by CF and no cross reactivity with serum from animals infected
with Babesia sp., Theileria sp., or Eperythrozoon sp. Cross
reactivity was detected with A. cntrale. Cross reactivity
with other rickettsial agents such as Ehrlichia sp. or Cowdria
ruminantium was not determined. Sensitivity of this test was
not adequately determined since the results were compared to
a test (CF test) which itself has poor sensitivity,
particularly in detecting carrier cattle. No effort was made

19
to test the sensitivity of this ELISA in detecting infected
carriers shown to be positive by other proven sensitive tests
such as IIF or calf inoculations.
Another ELISA used a 2 antigen technigue to enhance test
specificity; a negative antigen prepared from a cow prior to
infection, and a positive antigen derived from A. margnale
infected cells (Duzgun et al., 1988). Reactants were
identified using net absorbance values obtained by subtracting
the absorbance value of sera with negative antigen from the
absorbance value of sera with positive antigen. Specificity
of this test was good with only 3% of negative sera, 2% of
sera from animals infected with B. bovis, and 4% of sera from
animals infected with B. bigemina giving positive results.
Other related rickettsial agents were not tested. Sensitivity
also appeared to be good with no false negatives noted in 100
animals confirmed negative by IIF or calf inoculations. A
small number of infected cattle had positive ELISAs for up to
3 years later. This test provided the best sensitivity and
specificity thus far, and was the first ELISA that
demonstrated the ability to detect long term carrier cattle.
However, the crude nature of the test antigen necessitates the
use of the 2 antigen system which is more cumbersome, complex,
and time consuming. If this 2 antigen system had not been
used, 37% of the negative sera would have given false positive
results. In addition, the 2 antigen system reguires the use
of a spectrophotometer to identify reactants. This does not

20
allow for eventual visual identification as would be needed
for rapid field diagnosis.
Direct visualization of positive reactants was
accomplished using a Dot-ELISA (Montenegro-James et al.,
1990) In this test whole initial body preparations were
solubilized in SDS and dotted on to nitrocellulose disks.
Test sera were reacted with the antigen and antigen/antibody
complexes were visualized with alkaline phosphatase-conjugated
protein A. Test specificity was good (95%) and cross
reactivity to Babesia or Trypanosome vivax was not observed.
However, other rickettsial agents or related hemoparasites
were not tested. Test specificity was increased by using
protein A-conjugated alkaline phosphatase versus an anti-
globulin-conjugated molecule. This reduced nonspecific
binding of antibodies to the nitrocellulose disk. However,
the use of whole initial bodies isolated from infected
erythrocytes still lends itself to false positive reactions.
Test sensitivity with the Dot-ELISA was fair (92.9%) with
19 false negatives out of 269 true positives (Montenegro-James
et al., 1990). In addition, no effort was made to determine
the ability of this test to detect chronically infected, long
term carrier cattle. A good diagnostic test for
identification of infected cattle for importation or herd
management should have a sensitivity of at least 95% or
higher. Therefore, while this test is easy and convenient to

21
perform, its lack of sensitivity and questionable specificity
are certainly areas of needed improvement.
Recently, an antigen capture ELISA was developed for the
detection of A. margnale infection (Trueblood et al., 1991).
This ELISA utilizes 2 monoclonal antibodies (MAb) which
recognize 2 different epitopes on the A. margnale surface
protein MSP-la. In this test one MAb is bound to a
polystyrene microtiter plate, antigen (ie. infected whole
blood) is incubated with this monoclonal, and a second
monoclonal conjugated to horseradish peroxidase is added to
the reaction mixture for visualization of any captured
antigen. This assay was sensitive enough to detect infected
animals with parasitemias of <1.0%, however, it is not
sensitive enough to detect carrier cattle which have
parasitemias as low as 104 infected cells per ml of blood
(0.0001%) (Eriks et al., 1993). Thus, although the detection
of A. margnale antigen in the blood of infected cattle would
be the most specific way to determine infection, the
sensitivity of most primary binding assays will not likely be
sufficient to detect the low level of parasitemia encountered
in carrier cattle.
Nucleic Acid Probe Hybridization
One test which has shown excellent results in the ability
to detect A. margnale organisms in infected cattle, even at
a level of sensitivity sufficient to identify chronically

22
infected carriers, is a nucleic acid probe derived from a
fragment within the gene coding for an A. margnale surface
protein (Goff et al., 1988; Eriks et al., 1989). This probe
can detect infected animals with parasitemias as low as
0.000025% (Eriks et al., 1989). Sensitivity and specificity
of the test surpasses all previously developed serological
tests, including the IIF (Goff et al., 1990). Although this
test is too complex to be useful for routine field diagnosis,
it has tremendous potential for identification of true
carriers which in the past was done by subinocculation of
splenectomized calves. This could be useful for the
identification of cattle for reference sera, establishing the
effects treatment with tetracycline has on carrier status, or
identification of infected ticks. Considering the alternative
of calf inoculation, a nucleic acid probe is a much more
convenient and economical tool when accurate identification of
infection is essential.
Previous Experiments
The ideal test antigen to detect A. margnale infected
carriers must be antigenic enough to have antibodies present
in sera of cattle during all stages of infection. It must be
species specific so as not to cross react with sera from
animals infected with related organisms, and it must be
conserved between isolates of A. margnale, enough to be
recognized by sera from cattle infected with various

23
geographic isolates. An antigen which shows promise for
meeting the above requirements is the 86kDa surface protein of
A. margnale, MSP3.
In previous experiments, the most antigenic surface
proteins of A. margnale were identified by
immunoprecipitation of radiolabeled initial body proteins from
a Florida (FL) isolate of A. margnale with immune sera from
infected cattle (Palmer et al.f 1986). The FL isolate of A.
margnale was used for antigen isolation because it has been
found by adsorption studies to contain antigens common to both
morphologic types of A. margnale, the tailed and non-tailed
forms (Goff and Winward, 1985; Kreier and Ristic, 1963). An
86kDa protein, MSP3, was identified as the most immunodominant
in all stages of infection from as early as 30 days post
infection (PI) to 255 days PI. Similar reactivity was
observed when initial body preparations were
immunoprecipitated with immune sera from cattle infected with
either of 3 different isolates of A. margnale, FL, Virginia
(VA) and Texas (TX) isolates (Palmer et al., 1986). This
suggested MSP3 from a FL strain was conserved enough to be
recognized by immune sera from animals infected with different
isolates. However, the conservation of MSP3 was not firmly
established since these studies were performed using single
dimensional gel electrophoresis.
These experiments identified for the first time antigenic
surface proteins of A. margnale which could be investigated

24
for potential use as a diagnostic test antigen or vaccine
candidate. Use of a single antigen in a diagnostic test could
markedly increase test specificity over currently available
tests. Since MSP3 was the most immunodominant protein,
further experiments were done to determine its potential as a
diagnostic test antigen. Monoclonal antibodies to MSP3 were
produced and used in a sepharose bead affinity column to
isolate purified MSP3 from FL isolate initial bodies (McGuire
et al., 1991). Affinity purified MSP3 was injected into
rabbits and rabbit-anti-MSP3 immune serum was used on dot
blots to identify epitopes of MSP3 in at least 8 different
geographic isolates of A. marginale (Mcguire et al., 1991).
Using immune sera from infected cattle, purified MSP3
accurately identified long term carriers for up to 5 years PI.
Immune sera from cattle infected with B. bovis, B. bigemina,
or an unidentified rickettsial agent did not cross react with
purified MSP3 (McGuire et al., 1991). However, sera from
animals infected with organism now known to be
phylogenetically related to A. marginale were not tested in
these experiments.
Since MSP3 fulfilled many of the criteria for a good
diagnostic test antigen, attempts were made to clone the MSP3
gene and produce a recombinant protein which might be used in
an ELISA test. A genomic library made from a FL isolate of A.
marginale DNA was screened using a pool of anti-MSP3 MAbs.
Three clones of the MSP3 gene were identified and seguenced.

25
However, proteins expressed by these clones were inconsistent
in their reactivity with immune cattle sera. In addition, in
depletion experiments, anti-MSP3 MAbs were reacted with FL A.
margnale initial bodies. The initial bodies still contained
an 86 kDa antigen when reacted with immune cattle sera.
The MSP3 antigen of A. margnale appears to be a strong
candidate as a diagnostic test antigen to detect infected
carrier cattle. However, these previous experiments propose
questions which must be answered to confirm this hypothesis.
Questions must be addressed regarding the specificity of this
protein in detecting A. margnale infection, the conservation
of the MSP3 protein between various geographic isolates of A.
margnale, and explanations for inconsistent reactivity of the
3 MSP3 clones. The purpose of our investigation is to further
characterize the MSP3 protein and determine if it is a
suitable candidate for a diagnostic test antigen to detect A.
margnale infected, carrier cattle.

CHAPTER 3
MATERIALS AND METHODS
Anaplasma margnale Strains
Four isolates of A. margnale were used in this study,
FL, VA, South Idaho (SI), and Washington (WA). These isolates
are designated by their original location of isolation
(McGuire et al., 1984). Isolates were stored in liquid
nitrogen as cryopreserved stabilates (Love, 1972) before being
used to infect splenectomized calves. Thawed stabilate (20ml)
from each isolate was injected intramuscularly into 6-month-
old, male Holstein calves. Calves were monitored daily for
percent parasitemia by blood smear evaluation, and packed cell
volume (PCV). Infected, whole blood was collected in EDTA
from calves during periods of peak parasitemias (FL=70%,
VA=36%, SI=42%, and WA=4 0%), centrifuged at 10,000 x g for 15
min., and the serum and buffy coat was removed. Packed
erythrocytes were washed 3 times in phosphate buffered saline
(PBS) (0.14M NaCl, 2.68 mM KC1, 8.26 mM K2HP04, pH 7.4),
resuspended to a PCV of 50%, and stored at -70C.
A. margnale strains used in clamped homogeneous electric
field electrophoresis (CHEF) studies were prepared as
previously described (Alleman et al., 1993). Briefly, whole
26

27
blood from cattle infected with the previously mentioned
isolates was collected with sodium heparin used as an
anticoagulant and washed 3 times in PBS. Erythrocytes were
separated from bovine leukocytes by passing washed blood over
an a-cellulose/microcrystalline cellulose column (Sigmacell
type 50, Sigma Chemical Co., St. Louis, MO) as described by
Beutler (Beutler, 1984), except that 1% bovine serum albumin
(BSA) was included in the wash buffer. Washed cells were
resuspended to a concentration of 5.0 x 106 cells//xl, or
approximately 35% packed cell volume (PCV). Blood films were
prepared from isolated erythrocytes for microscopic detection
of bovine leukocyte contamination and level of parasitemia.
Intact erythrocytes were embedded in 0.7% agarose by
mixing 1 part erythrocyte suspension with 2 parts of 1% FMC
InCert Agarose (FMC Bioproducts, Rockland, ME) in PBS (0.14M
NaCl, 2.68 mM KC1, 8.26 mM K2HP04, pH 7.4), 0.125 M EDTA.
Mixtures of blood and agarose were kept at 37C while
pipetting into molds. Molds containing plugs were placed on
ice to set. Plugs were incubated in 0.5 M EDTA (pH 9.5), 1%
N-lauroylsarcosine, and 2 mg/ml Proteinase K for 48 hours at
37C, then stored at 4C in fresh Proteinase K solution.
Initial Body Preparation
A. margnale initial bodies were isolated from infected
erythrocytes as previously described (Palmer and McGuire,
1984). Briefly, frozen blood was guick-thawed at 37C and

28
washed 5 times in PBS with each centrifugation at 16,000 x g
for 25 min. at 4C. After each centrifugation, an upper layer
containing both leukocytes and erythrocytes was removed. The
pellets were resuspended in PBS, sonicated for 2 min. on ice
at 50 W and centrifuged as before. Pelleted material was
again resuspended in PBS, and sonicated for 30 sec. on ice at
50 W and centrifuged a final time. Intact initial bodies were
visualized by Wright-Giemsa stain. The pellets of initial
bodies were resuspended in equal volumes of PBS for use in
SDS-PAGE.
Initial bodies used in 2-D gel electrophoresis were
resuspended in equal volumes of lysis buffer containing 9.5 M
urea, 2% nonidet P-40, 1.6% Ampholyte 5/7 (Bio-Lyte 5/7, Bio-
Rad Laboratories Richmond, CA) 0.4% ampholyte 3/10 (Bio-Lyte
3/7, Bio-Rad Laboratories, Richmond, CA) and 5.0% j8-
mercaptoethanol.
Protein concentrations were determined
spectrophotometrically using the Micro BCA Protein Assay
(Pierce, Rockford, Illinois). Initial body preparations were
stored in small aliquots at -70C. Anaplasma cntrale initial
bodies were also prepared as described above.
Babesia bovis and Babesia biaemina Antigen Preparation
Babesia bovis and B. bigemina antigens were prepared from
organisms maintained in microaerophilic stationary phase
culture as previously described (Levy and Ristic, 1980).

29
Briefly, infected erythrocytes were centrifuged at 10,000 x g
and the supernate was removed. Packed cells were resuspended
to 2 0 times the volume in 10 mM sodium phosphate. The
solution was centrifuged and the supernate was removed. The
pellets were then resuspended in an egual volume of 10 mM
sodium phosphate.
Antisera used for Immunoblots
Six Holstein calves were infected with blood stabilate
containing a FL isolate of A. margnale. Two other Holstein
calves were each infected with either a VA or a SI isolate.
Antisera from all calves were collected at 50 and 70 days PI.
Antiserum from a cow experimentally infected with a WA
isolate, rabbit-anti-MSP3 polyclonal sera (RB-955), an anti-
MSP3 monoclonal antibody (MAb) (AMG75C2), and an anti-MSP2 MAb
(ANAF19E2) were supplied to us courtesy of Travis McGuire and
Guy Palmer (Washington State University, Pullman). The
reactivities of the MAbs (McGuire et al., 1984; McGuire et
al., 1991; Palmer et al., 1988; Palmer et al., 1994) and the
rabbit-anti-MSP3 polyclonal serum (McGuire et al., 1991) have
been previously described. Sera obtained from calves prior to
infection with A. margnale, and a MAb specific for the
variable surface glycoprotein of Trypanosoma brucei (TRYP1E1),
were used as negative controls in immunoblot experiments.
A. cntrale and A. ovis antisera with indirect
fluorescent antibody (IFA) titers of 1:4,000 were supplied

30
courtesy of Susan Oberle (The Salk Institute, San Diego, CA).
Hyperimmune sera from cattle infected experimentally with B.
bovis and B. bigemina were supplied by David Allred
(University of Florida, Gainesville). Sera from cattle
experimentally infected with Cowdria ruminantium were supplied
by Michael Bowie (University of Florida, Gainesville). Equine
sera from animals infected with Ehrlichia equi and Ehrlichia
risticii had I FA titers of 1/1,600, and were obtained from
Ibulaimu Kakoma (University of Illinois, Urbana). Serum from
a dog with canine ehrlichiosis was obtained from Rose Raskin
(University of Florida, Gainesville). This dog was infected
with E. ewingii, but serum from this animal had an IFA titer
of 1/160 for Ehrlichia canis, and 1/64 for E. chaffeensis.
SDS-PAGE
Initial body preparations containing 6.0 10.0 /xg of
protein were solubilized in one half their volume of a 3x
sample buffer containing 0.1 M Tris pH 6.8, 5% SDS (w/v), 50%
glycerol, 7.5% /3-mercaptoethanol, and 0.00125% bromophenol
blue, and heat denatured at 100C for 3 min. Proteins were
electrophoresed on 7.5% to 17.5% (w/v) gradient polyacrylamide
gels. Gels were fixed in 25 mM Tris, 191.8 mM Glycine and 20%
Methanol, and electrophoretically transferred to
nitrocellulose (Hybond ECL, Amersham International pic,
Buckinghamshire, England).

31
Two-Dimensional Gel Electrophoresis
Isoelectric focusing gels were prepared per
manufacturer's instructions (Protean II Slab Cell Instruction
manual, Bio-Rad Laboratories, Richmond, CA) Briefly, an
acrylamide\N,N'-methylene-bis-acrylamide solution containing
9.5 M urea, 2.0% nonidet P-40(v/v), 4.1% acrylamide/bis (30.8%
T/2.6% C) 10 mM (3-[(3-cholamidopropyl)dimethylammonio]-
lpropanesulfonate, CHAPS), and 5.8% Bio-Lyte 5/7 (v/v) was
polymerized in glass tubing 3 mm x 140 mm. Initial body
preparations were incubated for 2 hours at room temperature in
4 times their volume of the previously described lysis buffer
and 5 times their volume of sample buffer containing 9.5 M
urea, 2.0% Triton X-100, 5% /3-mercaptoethanol, 1.6% Bio-Lyte
5/7, and 0.4% Bio-Lyte 3/10. The solution was then
centrifuged at 100,000 x g for 2 hrs. at 25C. A volume of
the supernatant containing 20/ig of protein was loaded onto
each tube gel and overlayed with 50/il of overlay buffer
containing 9.5 M urea, 0.8% Bio-Lyte 5/7, 0.2% Bio-Lyte 3/10,
and 0.0025% bromophenol blue. Tube gels were electrophoresed
at 400 volts for 16 hrs. and 800 volts for 2 hrs. using the
BRL model V16 vertical gel electrophoresis system (Bethesda
Research Laboratories, Gaithersburg, MD). Tube gels were
extruded from the glass tubes and eguilibrated for 5 min. in
buffer containing 0.0625 M Tris-HCl, pH 6.8, 10.0% glycerol,
2.0% SDS (w/v) 5.0% /3-mercaptoethanol, and 0.00125%

32
bromophenol blue. Focused proteins were then electrophoresed
on 7.5% to 17.5% (w/v) gradient polyacrylamide gels and
treated and transferred to nitrocellulose membranes as
described above.
Immunoblots with Antisera
Nitrocellulose membranes containing transferred proteins
were blocked with 5% milk (w/v) in PBS with 0.25%
polyoxyethylene-sorbitan monolaurate (Tween 20) to inhibit
non-specific binding of primary and secondary antibodies. The
membranes were washed with 1% milk (w/v) in PBS with 0.25%
Tween 20, and probed with antisera from animals infected with
one of the following organisms; A. margnale, A. cntrale, A.
ovis, B. bovis, B. bigemina, C. ruminantium, E. equi, E.
risticii, or E. ewingii. Normal sera from respective
uninfected species were used as negative controls. Serum
dilutions (in PBS with 1% milk and 0.25% Tween 20) of 1/100 or
greater were used. Rabbit-anti-MSP3 polyclonal sera was used
at a dilution of 1/5,000. Normal rabbit sera was used at the
same dilution as a negative control. Anti-MSP3, anti-MSP2,
and negative control anti-trypanosome MAbs were used in
concentrations of 5 /^g/ml. The membranes were again washed
with 1% milk (w/v) in PBS with 0.25% Tween 20 and probed with
either species-specific anti-IgG-Horseradish peroxidase (HRP)
conjugated antibody at a dilution of 1/2,000 (Sigma Immuno
Chemicals, St. Louis, MO), or HRP-conjugated Protein G at a

33
dilution of 1/15,000 (Sigma Immuno Chemicals, St. Louis, MO).
Membranes were processed for enhanced chemiluminescence (ECL)
with detection reagents containing luminol as a substrate (ECL
Western Blotting detection reagents) (Amersham International,
pic, Buckinghamshire, England). The membranes were exposed to
Hyperfilm MP (Amersham International, pic, Buckinghamshire,
England) to visualize bound antibody.
Purification of A margnale Genomic DNA
Florida, South Idaho, and Virginia isolates of A.
margnale genomic DNA used in hybridization studies were
purified by phenol/chloroform extraction and ethanol
precipitation as previously described (Barbet et al., 1987;
Barbet and Allred, 1991). A. margnale genomic DNA from FL,
SI and VA isolates used in CHEF studies was prepared in
agarose plugs as previously described above.
MSP3 Clones
Three clones of the MSP3 gene, MSP3-11, MSP3-12, and
MSP3-19 were cloned, sequenced, and supplied to us courtesy of
T. McGuire, G. Palmer, and T. McElwain (Washington State
University, Pullman). These clones were prepared from a
genomic library composed of mechanically sheared A. margnale
DNA, ligated with Eco RI adaptors, and inserted into
pBluescript SK(-) plasmids (Stratagene, Lajolla, CA). Clones
were identified by screening the library with anti-MSP3 MAbs

34
(AMG75C2, AMG76B1, AMG43/19, & AMG43/23), one of which,
AMG75C2, has been previously described (McGuire et al., 1991).
One of these genes, MSP3-12, contains the N-terminus and 633
bp upstream to the open reading frame (Fig. 1) The C-
terminus is not present in this clone. The remaining 2
clones, MSP3-11 and MSP3-19, are missing the N-terminal
sequence of the open reading frame (Fig. 1) They both
contain the C-terminal end of the gene as well as 1,473 bp and
2,480 bp respectively, downstream to the open reading frame.
Verification of Recombinant MSP3
To verify a cloned MSP3 gene represented a gene which
produced one or more of the 86 kDa antigens seen on 2-D
immunoblots, E. coli cells (Epicurian Coli XLl-Blue)
(Stratagene, Lajolla, CA) were transformed with pBluescript
containing each of the 3 MSP3 genes. This was done according
to manufacturer's instructions. E. coli cells were
transformed with nonrecombinant pBluescript as a negative
control. Transformed cells were plated on Luria agar
containing 50 /xg/ml of ampicillin. Transformants were
selected by blue/white screening and selected colonies of each
transformant were grown overnight in Luria broth containing 50
ig/ml of ampicillin. Transformed E. coli cells were suspended
in PBS, lysed in one half their volume of a 3x sample buffer
containing 0.1 M Tris pH 6.8, 5% SDS (w/v) 50% glycerol, 7.5%
j8-mercaptoethanol, and 0.00125% bromophenol blue, and heat

Fig. 1 Diagram of MSP3 clones.
A schematic representation of clones MSP3-11, MSP3-12, and MSP3-19. Homologous
regions are indicated by like shaded areas. Nucleotide numbers are indicated on
the bottom. Top numbers of clones MSP3-11 and MSP3-19 indicate corresponding
nucleotides in clone MSP3-12.

MSP3 CLONES
MSP3-12
Start
MSP3-11
1306 1472 1818 2008 2090 2280 Stop
u>

37
denatured at 100C for 3 min. Bacterial lysates or similarly
prepared FL A. margnale lysates were separated by SDS-PAGE
and transferred to nitrocellulose as previously described
above. Membranes were reacted with anti-MSP3 MAb AMG75C2 or
immune sera from animals infected with a FL or VA isolate of
A. margnale as described above. Nonimmune cattle sera and an
anti-T. brucei MAb were used as negative controls. Antigen
antibody reactions were visualized by ECL with detection
reagents containing luminol as a substrate (ECL Western
Blotting detection reagents), (Amersham International, pic,
Buckinghamshire, England). The membranes were exposed to
Hyperfilm MP (Amersham International, pic, Buckinghamshire,
England) to visualize bound antibody.
Diqoxiqenin Labeling of oBluescript MSP3-12
Empty pBluescript DNA and pBluescript MSP3-12 DNA were
grown in previously prepared transformants. Plasmid DNA was
isolated from bacterial DNA by ion exchange chromatography
using the QIAGEN Plasmid Midi Kit (Qiagen Inc., Chatsworth,
CA) according to the manufacturer's instructions. The
purified plasmid DNA was precipitated in ethanol, dried, and
dissolved in TE (10 mM Tris, pH 7.5, 1 mM EDTA).
A digoxigenin-labeled probe of pBluescript MSP3-12 was
prepared by digestion of 5 fig of pBluescript MSP3-12 with Eco
RI according to manufacturer's instructions (Boehringer
Mannheim Corp., Indianapolis, IN). Empty pBluescript was

38
digested identically for size comparison of digested plasmids.
The 2.3 kbp insert of the MSP3-12 gene, which was previously
inserted using Eco RI adaptors, was separated from plasmid DNA
by electrophoresis on a 1% agarose gel in Tris Borate EDTA
buffer (TBE) (45 mM Tris, 45 mM Boric acid, 1 mM EDTA). The
gel was stained with 0.5 ¡j.g/ml ethidium bromide for 20 min.
and photographed. The 2.3 kbp band representing clone MSP3-12
was cut from the gel and DNA was extracted from the agarose
plug by ion exchange chromatography using QIAquick Gel
Extraction Kit (Qiagen Inc., Chatsworth, CA).
A probe was made by random prime labeling 200 ng of MSP3-
12 DNA with digoxigenin using the Genius System Nonradioactive
DNA Labeling Kit according to the manufacturer's instructions
(Boehringer Mannheim Corp., Indianapolis, IN).
Representation of pBluescript MSP3-12 in
the A. margnale Genome
Restriction Enzyme Digestion
To verify the cloned pBluescript MSP3 was an accurate
representation of genomic MSP3, multiple restriction sites of
pBluescript MSP3-12 and genomic A. margnale DNA from a FL
isolate were compared using restriction enzymes which cut
within the MSP3 gene. Restriction enzymes Neo I (Boehringer
Mannheim Corp., Indianapolis, IN), Bsp M (New England BioLabs,
Beverly, MA) and Eae 1 (New England BioLabs, Beverly, MA) ,
were chosen because they produced large fragments in different

39
areas of the MSP3 gene (Fig. 2) Digestions of genomic A.
margnale DNA (1.0 /xg) and pBluescript MSP3-12 (0.1 /xg) were
performed according to the manufacturer's specifications.
Digested genomic and plasmid DNA were separated by gel
electrophoresis on 1% agarose gels containing 0.1 /xg ethidium
bromide and photographed.
Southern Blots
Prior to transfer, the gel was incubated at room
temperature for 30 min. in 0.4 N NaOH, 0.6 M NaCl then 30 min.
in 1.5 M NaCl, 0.5 M tris HCL, pH 7.5. Digested DNA was then
transferred to a positively charged, molecular biology nylon
membrane (Boehringer Mannheim Corp., Indianapolis, IN) by
capillary diffusion using lOx SSC (lx SSC is 0.15 M NaCl plus
0.015 M sodium citrate). The filter was washed for 30 sec. in
0.4 N NaOH, then in 0.2 M Tris HC1, pH 7.5, 2X SSC for 2 min.
After air drying, the DNA was cross linked by ultraviolet
radiation and incubated for 3 hours at 65C in
prehybridization solution containing 6x SSC, 0.5% SDS (w/v) -
200 /xg/ml herring sperm DNA, 5x Denhart's [lx Denhart's is
0.02% Ficoll (w/v), 0.02% polyvinylpyrrolidone (w/v), 0.02%
BSA (w/v)]. The digoxigenin-labeled MSP3-12 probe was added
to fresh prehybridization solution at a concentration of 15
ng/ml, and the filter was hybridized overnight at 65C. Bound
probe was detected by enhanced chemiluminescence using
alkaline phosphatase conjugated, anti-digoxigenin IgG

Fig. 2 Representation of clone MSP3-12 in the A. margnale genome.
Diagram of clone MSP3-12 indicating the restriction enzyme sites and the resulting
fragments used to map MSP3-12 to the genome. Nucleotide cleavage site for each
restriction enzyme is illustrated diagrammatically (top) and by nucleotide
number(bottom). Size of resulting fragments in base pairs is also provided.

N N ^
S
,0
1
2,33
Restriction
Cleavage site in
Size of co-migrating
Enzyme
pBluescript MSP3-12
plasmid and genomic
fragments
Neo I
12/1,188
1,176
Eae I
52/1,762
1,710
Bsp M
193/2,076
1,886

42
according to the manufacturer's recommendations (The Genius
System Luminescent Detection Kit, Boehringer Mannheim,
Indianapolis, IN) The membranes were exposed to Hyperfilm -
MP (Amersham International, pic, Buckinghamshire, England) to
visualize bound antibody. The molecular sizes of comigrating
cloned and genomic fragments were determined by comparison to
Lambda DNA/Hind III fragments (GIBCO BRL, Gaithersburg, MD)
and 1 kbp DNA Ladder (GIBCO BRL, Gaithersburg, MD) molecular
size standards.
Presence of Multiple MSP3 Gene Copies
A. margnale genomic DNA from either the FL, VA, or SI
isolate was extracted as described above and aliquots of 1 /g
of DNA were digested with Hie II (New England Biolabs,
Beverly, MA) Sac 1 (Boehringer Mannheim Corp., Indianapolis,
IN), Nde I (Boehringer Mannheim Corp., Indianapolis, IN), or
Sph 1 (Boehringer Mannheim Corp., Indianapolis, IN). These
enzymes were chosen because they do not cut within the known
sequence of MSP3-12 gene. Calf thymus DNA was digested
identically as a control. Empty pBluescript and pBluescript
MSP3-12 DNA (0.1 ig) were digested with Eco RI and used as
negative and positive controls, respectively for probe
hybridization. Digested fragments were separated by gel
electrophoresis with 1% agarose gels containing 0.1 /g
ethidium bromide and photographed. Southern blotting was
performed under prehybridization and hybridization conditions

43
as previously described above and blots were probed with
digoxigenin-labeled pBluescript MSP3-12. Bound probe was
detected by ECL as described above. Lamda DNA/Hind III
fragments (GIBCO BRL,Gaithersburg, MD) and 1 kbp DNA Ladder
(GIBCO BRL,Gaithersburg, MD) were used for molecular size
standards.
Distribution of MSP3 Copies in the A. margnale Chromosome
Restriction Enzyme Digestion in Agarose Plugs
The locations of multiple MSP3 copies in the chromosome
were determined by Southern blotting of large A. margnale
genomic fragments separated by CHEF. Infected erythrocytes
containing intact genomic DNA from FL, SI, and VA strains of
A. margnale were embedded in 0.7% agarose plugs and stored in
0.5 M EDTA (pH 9.5), 1% N-lauroy lsarcosine, and 2 mg/ml
Proteinase K at 4 C as previously described above. Agarose
plugs were digested with Not I (Boehringer Mannheim Corp.
Indianapolis, IN) and Sfi I (New England BioLabs, Beverly,
MA) as previously described (Alleman et al., 1993). Briefly,
each plug was washed in 10 times their volume of T.E (10 mM
tris, pH 7.5, 0.1 mM EDTA) over several hours with fresh T.E
replaced each hour, then incubated in T.E plus 1 mM
phenylmethylsulphonyl fluoride (PMSF) for 2 hours at 37C.
Agarose plugs were then equilibrated on ice for 1 hour in 10
volumes lx restriction enzyme buffer plus 0.1 mg/ml BSA. A

44
solution of 100 units of restriction enzyme in lx restriction
enzyme buffer was added and DNA digestions were performed
according to the manufacturer's instructions. Reactions were
stopped by the addition of 0.25 total reaction volume of 0.5
M EDTA, pH 8.0, 0.1% N-lauroylsarcosine (w/v), 1 mg/ml
proteinase K (w/v), incubating reaction mixture at 4C for 1
hour, then 37C for 15 min. Uncut M. bovis chromosomal DNA
(Promega Corp., Madison, WI) was digested identically to serve
as a control for restriction endonuclease activity.
CHEF Gel Electrophoresis
This was done on the CHEF DRII system (Bio-Rad
Laboratories, Richmond, CA) Plugs of digested DNA were
electrophoresed in 1% agarose gels in 0.5x TBE buffer at 14C.
Electrophoretic conditions were set at 180 V, 10 sec. switch
rate, with a 16 hour run time. Delta 39 Lambda Ladders
(Promega Corp., Madison, WI) and Lambda DNA/Hind III Fragments
(GIBCO BRL, Gaithersburg, MD) were used as size standards.
The gels were stained for 30 min. in 0.5 ig/ml ethidium
bromide and photographed. Southern blotting and hybridization
of the separated bands were performed as previously described
except that the gel was depurinated in 0.25 M HC1 for 15 min.
prior to washing and transfer to nylon membranes.
Prehybridization and hybridization conditions were performed
identically to those described above. The membranes were
probed with digoxigenin-labeled pBluescript MSP3-12, and bound

45
probe was detected by ECL as before. The membranes
exposed to Hyperfilm MP (Amersham International,
Buckinghamshire, England) to visualize bound antibody.
were
pic,

CHAPTER 4
RESULTS
Specificity Experiments
SDS-PAGE separated A. margnale initial body proteins
were transferred to nitrocellulose, and probed with antisera
from animals infected with related rickettsial agents or
protozoal hemoparasites. This was done to determine if
animals infected with these organisms contain antibodies which
cross react with the 86 kDa protein (MSP3) of A. margnale.
As a negative control, normal sera from various species were
reacted with A. margnale proteins. Where available, protein
preparations from related organisms were used in a homologous
reaction with respective antisera to serve as a positive
control.
Sera from a sheep infected with A. ovis (Fig. 3), a horse
infected with E. risticii (Fig. 4), and a dog infected with E.
ewingii (Fig. 4) showed strong reactivity with MSP3 of A.
margnale. These sera as well as sera from animals infected
with E. equi and C. ruminantium showed reactivity against
other A. margnale antigens as well. Sera from animals
infected with A. ovis, C. ruminantium, E. ewingii, and E.
equi, showed reactivity against a 36 kDa antigen, possibly
MSP2 (Figs. 3, 4, & 5). Sera from animals infected with C.
46

Fig. 3 Specificity experiments using immunoblots.
A. margnale (AM) or A. cntrale (AC) initial body preparations reacted with normal
sera from non-infected sheep (NSS), non-infected cattle (NBS), or antisera from
animals infected with A. ovis (AO), A. cntrale (AC), or A. margnale (AM) .
Labeling above each lane indicates the serum used (top), the initial body
preparation used (center), and the dilution of the serum (bottom). Molecular size
standards in kilodaltons are illustrated on the left.

200
97.4
69
46
30
NSS
AO
AM
NBS
AC
AC
AM
AM
AM
AM
AM
AC
AM
AM
1/500
1/500
1/500
1/100
1/100
1/100
1/300
.P


Fig. 4 Specificity experiments using immunoblots.
A. margnale (AM) initial body preparations reacted with
normal serum from a non-infected horse (NHS) or anti-sera
from animals infected with A. margnale (AM), E. rstc
(ER), E. equi (EE), or E. ewingii (EC). Labeling above
each lane indicates the serum used (top) the initial
body preparation used (center), and the dilution of the
serum (bottom). Molecular size standards in kilodaltons
are illustrated on the left.

50
30

Fig. 5 Specificity experiments using immunoblots.
A. margnale (AM) initial body preparations, or B. bovis (Bbov) or B. bigemina
(Bbig) antigen preparations reacted with normal serum from a non-infected cow
(NBS), or antisera from animals infected with A. margnale (AM), C. ruminantium
(CR) B. bigemina (Bbig), or B. bovis (Bbov). Labeling above each lane indicates
the serum used (top), the initial body preparation used (center), and the dilution
of the serum (bottom). Molecular size standards in kilodaltons are illustrated on
the left.

NBS AM CR
AM AM AM
1/100 1/500 1/100
200
97.4
69
46
30
Bbig
Bbig
1/100
Bbig Bbov Bbov AM
AM Bbov AM AM
1/100 1/100 1/100 1/500
U1
tsj

53
ruminantium (Fig. 5) and E. equi (Fig. 4) reacted with an
antigen of slightly smaller molecular weight than MSP3. Sera
from animals infected with A. cntrale (Fig. 3) or either
Babesia sp. (Fig. 5) failed to react with any A. margnale
antigens although strong reactivity was demonstrated in
reactions involving homologous preparations.
Conservation of MSP3 Between Different
Geographic Isolates of A. marginale
The conservation of MSP3 between different geographic
isolates of A. marginale was evaluated by SDS-PAGE and
immunoblots using initial body preparations from FL, VA, SI,
and WA isolates of A. marginale. Preparations were separated
on 7.5% to 17.5% (w/v) gradient polyacrylamide gels as
described above.
They
were
then transferred
to
nitrocellulose, and
reacted
with
varying dilutions
of
antiserum from an animal experimentally infected with a FL
isolate and a MAb to MSP3, AMG75C2 (McGuire et al., 1991), for
definitive identification of the MSP3 antigen. An optimal
dilution of this antiserum was established for demonstration
of the immunodominant MSP3 protein (Data not shown). Pre
infection bovine sera and a MAb to a trypanosome surface
protein were used as negative controls and showed no
reactivity to MSP3.
Initial body preparations from these same isolates were
then probed with antiserum at a single dilution previously
established (Fig. 6). The side by side location of the

Fig. 6 Size polymorphism of MSP3.
Immunoblots of initial body preparations from a Florida
(FL) a Washington (WA) a South Idaho (SI) and a
Virginia (VA) isolate of A. margnale (AM), reacted with
normal serum from a non-infected cow, Lane 1 (Pre), or
anti-sera from a cow infected with a FL isolate of AM,
Lanes 2-5. Serum was diluted to 1/300. Molecular size
standards in kilodaltons are illustrated on the left.

55
M.W fl WA SI FL VA
kD
200-
97.4-
69-
46-
Pre Post
Anti-FI A.M. sera (1/300)

56
various isolates clearly demonstrates variation in size of the
MSP3 proteins. The 86 kDa MSP3 antigen is seen in the FL
isolate (Fig. 6) An antigen of similar size is seen when
this same antiserum is reacted with proteins from a VA isolate
(Fig. 6) However, in reactions with the SI isolate, an
antigen of slightly smaller molecular mass is seen, and the WA
isolate has an antigen with a molecular mass greater than 86
kDa (Fig. 6).
Immune Response to MSP3
Realizing MSP3 is not conserved between different
geographic isolates, we then investigated the possibility that
animals infected with different isolates may contain immune
sera that varies in reactivity to MSP3 from a single isolate.
This is an important consideration in attempting to develop a
diagnostic test antigen derived from a single isolate.
Initial body preparations from a FL isolate were separated by
2-D gel electrophoresis as described above in order to
determine if co-migration of antigens of similar molecular
size occurs. Electrophoretically separated proteins were
transferred to nitrocellulose and probed with antisera from
cattle infected with a FL, VA, SI, or WA isolate of A.
margnale as well as an anti-MSP3 MAb (AMG75C2), rabbit-anti-
MSP3 polyclonal serum, and an anti-MSP2 MAb (ANF19E2). Pre
infection bovine sera, normal rabbit sera, and a MAb to a

57
trypanosome surface protein were used as negative controls and
showed no reactivity to MSP3 (Data not shown).
In a homologous reaction with anti-FL serum, 2 major
areas of reactivity were seen with a molecular mass of 86 kDa,
one with an apparent isoelectric point (pi) of 6.5, and the
other with a pi of approximately 6.2 (Fig. 7). There was
slight reactivity with an antigen at a pi of approximately
5.6. When the initial body preparation from the FL isolate
was reacted with the anti-MSP3 MAb, major reactivity was seen
in the 5.6 area of the pH gradient, but no reactivity was
noted with antigens at a pi of 6.5 or 6.2 (Fig. 7).
Serum from an animal infected with a VA isolate showed
similar reactivity as the MAb. However, serum from an animal
infected with a WA isolate reacted with 2 antigens in an
entirely different area of the pH gradient, having pis of
approximately 5.1 and 5.3 (Fig. 7). When these same initial
body preparations were reacted with antiserum from an animal
infected with a SI isolate reactivity was noted in all 3 areas
of the pH gradient, with approximate pis of 6.5 to 6.2, 5.6
and 5.3 to 5.1 (Fig. 7). When a rabbit-anti-MSP3 polyclonal
sera was used, reactivity was noted in areas of the pH
gradient having pis of 6.5 to 6.2 and 5.6 (Fig. 8).
Although conditions seen here were not optimized to
separate the 36 kDa proteins of A. margnale (MSP2), multiple
spots were visualized in that apparent molecular size (Fig.
7). Some variation in reactivity of the different antisera to

Fig. 7 2D gel electrophoresis of A. margnale proteins.
Immunoblots using initial body preparation of a FL isolate of A. margnale
separated by 2-D gel electrophoresis. Letters centered above each immunoblot
indicate the antibody used in the reaction, FL = antiserum from a cow infected with
a Florida isolate of AM, VA = antiserum from a cow infected with a Virginia isolate
of AM, WA = antiserum from a cow infected with a Washington isolate of AM, SI =
antiserum from a cow infected with a South Idaho isolate of AM, and MAb AMG75C2
= anti-MSP3 MAb. Antisera are all used at a dilution of 1/300. Concentration of
MAb = 5/ng/ml. Numbers above each immunoblot indicate the pH taken at 1 cm
distances along the length of the tube gel. Arrows indicate the isoelectric point
of each 86 kilodalton band. On the far right side of each gel, initial body
preparation from a FL isolate of A. margnale was electrophoresed in a single
dimension and immunoblotted along with 2-D focused initial bodies to indicate the
86 kDa, MSP3. Molecular size standards in kilodaltons are illustrated on the left
of each immunoblot.

pi WA
T
7 6 9 6 68 6.42 6.13 5.82 5.66 5 36 5.13 4.56
14.3
VD

Fig. 8 2-D gel electrophoresis of A. margnale
proteins.
Immunoblots using initial body preparation of a FL
isolate of A. margnale separated by 2-D gel
electrophoresis. Letters centered above each immunoblot
indicate the antibody used in the reaction, RB-955 =
rabbit anti-MSP3 polyclonal sera and MAb ANAF19E2 =
anti-MSP2 MAb. Rabbit anti-MSP3 was used at a dilution
of 1/5,000. Concentration of MAb = 5/Ltg/ml. Numbers
above each immunoblot indicate the pH taken at 1 cm
distances along the length of the tube gel. Arrows
indicate the isoelectric point of each 86 kilodalton
band. On the far right side of each gel, initial body
preparation from a FL isolate of A. margnale was
electrophoresed in a single dimension and immunoblotted
along with 2-D focused initial bodies to indicate the 86
kDa, MSP3 and 36 kDa MSP2. Molecular size standards in
kilodaltons are illustrated on the left of each
immunoblot.

61
Pi
RB-955
M.W.
kD
200 -
97.4 -
69
46
30 -
T
6.68
T
6.42 6.13
5.82 5.66 5.36 5.13 4.16
pi MAb ANAF19E2
6.97
6.9
6.68 6.42 6.13
5.82 5.66 536 5.13 4.16
1 1 1 1 1 1 II 1
200
46
30
21 5 -

62
MSP2 was seen. For example, in a homologous reaction there
was reactivity with antigens with pis from 6.42 to 5.66.
However, when reacted with sera from animals infected with WA
or SI isolates, reactivity with antigens with pis of 6.42 to
5.36 was observed. These spots were confirmed to be the MSP2
antigen by use of an anti-MSP2 MAb (ANF19E2), the reactivity
of which has been previously described (McGuire et al., 1984;
Palmer et al., 1988; Palmer et al., 1994). Multiple spots
with a molecular mass of 36 kDa were recognized by this MAb
(Fig. 8) .
MSP3 Nucleotide Sequence and Translation
Drs. Travis McGuire, Terry McElwain, and Guy Palmer
(Washington State University, Pullman) identified and
sequenced 3 clones of the MSP3 gene by screening a genomic
library of a FL strain of A. margnale DNA with MAbs to MSP3.
These clones were designated MSP3-11, MSP3-12, and MSP3-19.
The 3 clones in pBluescript vectors, along with their
nucleotide and deduced amino acid sequences, were supplied to
us courtesy of the above individuals. The entire nucleotide
sequence of each clone is illustrated in Figs. 9, 10, & 11.
A complete gene sequence is lacking in each clone. However,
MSP3-12 contains the N-terminus of the gene and 633 bp
upstream to the open reading frame. This area upstream to the
5' end of the open reading frame is likely to contain

Fig. 9 Gene sequence of MSP3-11.
The nucleotide sequence of pBluescript MSP3-11 is
illustrated. Numbers on the left indicate the number of
the first nucleotide in each row. Termination codon is
underlined.

64
Msp311.Conentry:1 Length:3263 Hay 29, 1991 11:34 Check: 1828 ..
1 GGTGGGGAGATGGTAGGAGTT GAT GAAGGACT AGTTAT ACAAGAACTGAGCAGACCAGAAGAATT AGAAAAGCTACAACAT GAACTAGCMAGCAAGT AA
101 GTAAATTAGCTGAACTTGGAGAACTTAAGTGGTTAGAGCAACTTGAGACACTGGAGACTGAGGAGTTGGAGAAGGTGGCGAAAAGAGCCACACAGAAGCT
201 CAGTATATTGGAGGAGCAGCCAGGAT ACACCT CACTAGAQGAGCT GGAGAAGAAGCTGAAGGAGAT AGAQGAAATAAGGAAACT GAAGGGACCGGAGGCA
301 ATCAGAAAGCTTAGAGACTTGGAGAGGCTGGAAGCTGAGAAGTTGGAGGAGGTGAAGAAGAAGGTAAAGQGTTCAGAGCTTGCAGAGCATTTGGACAAAA
401 CGTGGTTGCTACGCAGGATTGGAAGAGTAQGGCAGGAACTAGCAGCGATAAGGGAGCTGAAGGAATTGGQGCTAGAGGAACGGCTCAGGGrACTAGCTGA
501 GATTAAGGAAGTTAGAGCATT GGCAGAGAAGCGGAAGGCTGGGGGACTAGAGATTCAGGAGGGGTTACAQCT GACT GAGAAGATTAGAGCATTGGGT GGA
601 CAGCTGGAT CT GCTGGAGGCACGAAT ATTATTGGGGCT AGAAGCT GAGAAGAT GAAGGAQGT GGAAGCGQCAAAGGAGGAAGT GGAGATTTTCCT GGGAT
701 T GCTCCACAACAATGACACAGAGACCAAGAAGAT AGAGGAGGT GAAGACTCTGGT AAT GCGGGT GGGT GACAGT CAGGGACTAAGGGGTOCGTTGGT CAA
801 GAAGTTAGAQGAGCTGGCTAAGAAGTT GGAGCCAAAGGT GGGT GGCAATACAGGGT TTCAGGGT CAGGTAGGGCTAGT GAAGGAT CTTAAGAAAAAGCT A
901 GAAGAACTAGCAGCGAT AAGGGAAGCATTQGAGCAACT GAAGGT GGAACCAGCGCT GCT AGAAGGGAT AAAGAGGGAGGTGCT AACACAQCAGAGTACTC
1001 AAGTTACACGTAGGGAGAATAATCCACAAAGCAAGGCAGAAGGTAAACTAGAAGAGATAAGAGCGCTTAGAAAGCTAGGACAGTTGGATGAGCTAATACC
1101 AGAAGAGAGATTGAGAGAATT GT CCAAGGTTAAGGAAACACT GAAGCGGTT GAAGGGT AAGGAGCTAGT AGACCTAGAGAGGAAGCT AGAAACTAT GGGG
1201 GAACT AAAGCAGGAGAT AGAGAAAAT CAAQGGGCAGGAAGACTTGAAGCAGCT AGAGGCTAGGAAGT T GGAGGAGGTGAAGAAGGCAAAAAGGGAAGTGT
1301 TTATTCTGAQGACAAAGGAAAGTCTGGAGAAAGAGGGGAAGCTGGAGGAGTTAAGGGGCTATAAACTCAAGAAGGCGATAAAGAAGCTGGAAACCTTGAT
1401 AGAGAAGATACGAGGGGTCAGCACGCT GAAGGAACAGT GGGAGCCAAAAGTT GAGAAGCTCAAGAGCAAACT AGAAACACTAGCTGCCATAAGGGAGCTG
1501 AAGAAATTGGGGTTTAAAGATTGGTTGAGACTCAGGACACTAGAGGAGCTTACAGAGATAGCTGAGCAGCGTGGAGTTGCAACAGTGATGAAGGCAGCAT
1601 TAGCTAGTGCGATGGAGATAGCTAAGAACAGGGGTTGGACTGATTATTTAAACAGTCTAGATGTAAGTGAGAGAGCTAATGCTGCAAGGGAATTAATTGC
1701 TGCTGAGAAGATTAGAAAATGGGCTAGGGATATTAATAACTTGGATGCTGATGAACGGGCAATGGTCGCTGGGGCCCTAACCCCTTCTACAACAGTGTGA
1801 CTCACCACTCTCCCCCGAAACATCACTCACACTTCCAACTCTACTGGGCACCACCAAGGQCAGCCAGCCTCTCTTGGCACCTCATCTCGTACCCTCTTGG
1901 CCTACCAAATCTCGACCCTQCCACCACAGCCCTCTGCACCGGTGTCACCTGCATAACCCGrCTCCCAGTCrCTTGTACCTCAACACCTGCATCTCATCTT
2001 CCTTACCCAATGTGGCATTOCACAATCTTTGTCCTCTCTACTTACCCCCTACTTCTTACOGGCACCACTCACTGCTTTCCCAGGTCAACTAGTCACCCAA
2101 CCTCTAGCACCTTCAGCATCTCAGCCCAACTCACCACAGCTCTCTGCACCTTCAACAGCTCTAGCAAAAGCCCCGGCAAOVACTGCCTTATCCTCCTTAC
2201 TCAGGTTATTAATGTTATTCGCCATCCCCTGCAAGTTGATGGTGTCTTTGTTCCCCTCAACAGAAAGCAGTGTGTCTGTACCCTCAGTGAACACCTCACT
2301 AATCTTGGTGGTGGCAGTGCTATCGGTGGTACCACACTTCTTAGTGCCATTACTGCCTTCGCCCTTGCCACACACATTCTGGCTCACCTTCTCACCATTA
2401 GTAGTACCCTCTACTGCTTTACCCCACTTCTTGGCTTCACCCTTAGTCATCTTACCTAAAGCAGTGGCAAGACGGTCTACCTGACCTCTTGCTGTATCAT
2501 ATGCTAACTCCTTTCCTAATAAGAATACTGAAGCTGTATCCTCATTAGACTTCTTACCTCCCTTAATGACAAACCTTTCATACCCTACTTCTACTTCAAC
2601 CCTTGCTCCTCCAATACTATACCCAATGCTCCCCTCTAGGGCTGTGAGCATGCTATGGTAGCCCTGATATGGTTATTCATCACAGCAGGTTCTACTAATG
2701 CCGATACTATCAGCAGTACTAACACTGCTGftCGATACAGCTACCAAGGTTATTTGTGACGTCATTGTCTTCGTGCAGAAGCTGGGATTACCCATCATGAC
2801 AGGGGTGGTACTAGGTTCTGGCGTCATGGCTTGGCACACTTGCATGGCCTGCCATGGTAATGCTGGTTGTGTGTACAGCOMAGGTAAGCTTATGGCCAC
2901 TGTATTGAGGGTAATACCAGTTAGTGCAAQGGTGGACGTAGAGGTTTTGQGAGAGTGGTGATGTTGCTAAAACTTAGATTCTTACTCATTGCTGTAGTAT
3001 TAGCATTTGGTATCTTACATGGTGTACCTXTGGGGCTAOTAAGTCCAGCGAGGGTGCTCAGACTGCTGATGATACAGCO\CTGTTGTTATATGTAACGT
3101 TATACGGTTTGTGCAGAAGCTGGGACTACCCATCATGACTGGGGTGATACTAGGTTCTAGCATTATGGCTATCTTTGGTAAGCTTGCATGGCCTGCTATT
3201 GTAATGCTGGTTGTATTTACAGCTATATTCTTTGGTGCTQGTAAGCTTATGGCTAAATTCCCG

Fig. 10 Gene sequence of MSP3-12.
The nucleotide sequence of pBluescript MSP3-12 is
illustrated. Numbers on the left indicate the number of
the first nucleotide in each row. Start codon is
underlined.

66
Msp312.Conentry:1 Length:2337 August21, 1991 15:43 Type:N Check: 1077
1 GGGGGCCTGCCATGGTAATGCTGGTTGTATGTACAGCCATAGGTAAGCTTATGGCCACTGTATTGAGGGTAATACCAGTTAGTGCAAGGGTGGACGTAGA
101 GGTTTTGGGAGAGTGGTGATGTTGCTAAAACTTAGATTCTTACTCATTGCTGTAGTATTAGCATTGGGTATCTTACATGGTGTACCTGCTGGGGCTAGTA
201 AGTCCAGCGAGGGTGCTCAGACTGCTGATGATACAGCGACTATTGTTATATGTAACGTTATACGGTTTGTGCAGAAGCTQGGACTACCCATCATGACTGG
301 AGTGATACTAGGTTCTAGCATTATGGCTATCTTTGGTAAGCTTGCATGGCCTGCTATTGTGATGTTGGTTGTATTTACAQCTATATTCTTTGGTGCTGGT
401 AAGCTTATGQCTAAGTTTGCTGCTGGGTTGAGTGGTGAGGGTGTGAAGGATGCCGGTAGCTTTGACTGTTCCAGTTATAAAGGCACTGCTAGGCAGTGAT
501 GGTGCAGAGAGCTGTGGTGGTCGTGCCAAGTTAGACTGGGATGCCAGGGGTTTATAGATAGTGGCTTTGTTATGTCGGTGGTACCAAGAACTTCAAGTGT
601 GTTGGTGCCCAGTAGAGTTGGAAGTGTGAGTGATGTTTCGAGGGAGAGTGGTGATGATGCCAAAACTTAGATTCTTACTGATTGCTGGGGTATCAGCATT
701 GGGTATCCTOCATAGTATGTCTGCGGGGGCTGCCGATAGAAGTAAGTCACGCACACAGAGGTTGGCTAGTGGACCGGCTTTGGGGAAAGQCAACGGGAGC
801 TTCTACATAGGCCTAGACTATAACCCAACTTTCAACGGTATCAAGGACCTGAAAATCATCGGCGAAACCGATGAGGATGAAATGGATGTTCTCACCGGTG
901 CCAGGGGCCTGTTCCCGATGAACGCTCTTGCTAGCAACGTCACCGATTTTAACTCATACCACTTCGACTQGAGTACCCCACTGCCTGGGCTAGAATTTGG
1001 GAACAGTACCCTGGCTCTTQGAGGGAGCATTGGGTACAGAATTGGAGGAGCCAGGGTTGAAGTAGGGATAGGACATGAGAGGTTTGTTATTAAGGGGGGA
1101 GATGATGCAGCATTCCTACTAGGTAGGGAACTAGCATTGGATACAGCGAQGGGTCAGTTACTATCCAGTGCATTGGGTAQGATGTCCATGGGTGATGTAC
1201 ACAGATTAAAGAAGGAAGTAGTTGATAGTATAGGAAGAGGAACAGCTAGTCCTGTAAGGGCAATGTTTAGTAGAGAGATCTCAGATGGGAATACATTACT
1301 TGCTGGGGAGATGGTAGGAGTTGATGAAGGACTAGTTATACAAGAACTGAGCAGACCAGAAGAATTAGAAAAGCTACAACATGAACTAGCAAAGCAAGTA
1401 AGTAAATTAGCTGAACTTGGAGAACTTAAGTGGTTAGAAGAACTTGAAAAGCTGGAGACTGAGGAGCTGGAGCAAGGACTCGAAGGAGCGCTAAAAGCTT
1501 TGGGCGT GGAAGCAT CAGT GCAGGAGCTGGTACAGAGATTCAAGAAGGAAATTGCT GATQGT AAGACACCGGAAGAGAT AAAGCT GGAGTGGAT CAAGGA
1601 GATAGAAGCTAAGAAGTTAGAGGAGGTGCAGGAGCAGGCTGAGAAAAAGCTGGCGGAGCTGGAACTAGCQGATAAAGGAQGAGCATTGGGAACTAAAGGC
1701 GAAATACGGGAAGTTAGGCAGCTCAAGGAACTAGCGGATAAAGGAGGAGCGTTAGGAATCATGGCCGAGAAGCTCAAGAAGCAGGAAAGTCTCAAGGGGC
1801 T AGGAGGAACAGT AGAAGAATTAGCAGCGAT AAGGGAACTGAAGGAATTGGGGCT AGAAGAGCAGCTGAQGAT GCT GGCTGAGATTAAGGAAGTTAAGAG
1901 ATTGGCAGAGAAGCAGAAATCACAAGGATTAGAGGCTCATGAGGGATTACAACTGACTGAGAAGATTAGAGCATTGGGAGGACAGCTGGATCTGCTGGAA
2001 GAGAGATTGAGTGAAAAGTnAAGAGACTACAGCAGAATGAGCAAGGTATACGTAAGGATCTGAATCCACAAGCTATAACAGCAGTGGCAGGTAAACTAG
2101 ATGAGATTTQGGTACTGGGGACGCTAGGACAGTTAGATGAGCTAGTGCCTGGGAAGAGAGCTCAGACACTGGGGGAAATGAAGGCAACGCrGAAGGAGCT
2201 CGAGAAAATACGGAAGTTAGTAGACCTAGAGAAGAAGCTAGAAACTATGGGAGAACTAAAGAAGGAGATAGAGAAAATCAACGAGGAGGGAGAGCTTGTG
2301 AGGTTTACACGCAAAGAAGTCT CGGAGAT AT CACCCG

Fig. 11 Gene sequence of MSP3-19.
The nucleotide sequence of pBluescript MSP3-19 is
illustrated. Numbers on the left indicate the number of
the first nucleotide in each row. The termination codon
is underlined.

68
Msp319.Corentry:1 Length:4185 October21, 1991 15:01 Type:N Check:2553
1 GGT GATGTACACAGATT AAAGAAGGAAGTAGTTGATAGT AT AGGAAGAGGAACAGCT AGTCCTGT AAGGQCAATGTTTAGT AGAGAGATCTCAGAT GGGA
101 ATACATTACTT GCT GGGGAGAT GGT AGGAGTTGAT GAAGGACT AGT T ATACAAGAACT GAGCAGACCAGAAGAATT AGAAAAGCTACAACAT GAACT AGC
201 AAAGCAAGTAAGTAAATTAGCTGAACTTGGAGAACTTAAGTGGTTAGAAGAACTTGAAAAGCTGGAGACTGAGGAGTTGGGGGAGCTGTTGAGACTGAAA
301 GCGAGAAAAQCCTCACAGGAACT T AGGACGTTGGCGGAAAGT AAAGGACAGCACCT AAATGCT GAT GAGATAGAGGGGGAACTAAAGAGCGCGCT GGGAT
401 TGGAAAGACTGGAAAAGTTGGCAGAGCTGGAGTTGGTCAAGCAGCGGCTTGAGTCGATGAAGGAGCTGGAGAAGAAGATAGAGGAGGCGCAGTTGACAGC
501 AGAAGAGCTCCGGGAAATGAGGGAGAAGCTTAAGGAGTTGGCAGGAAGGGGGGAGGATCCTGCGGGGCTGAAGAAAGCGATAGAGGAGGCAAATGCTGAC
601 GAACTCAAGAAGGGGCTGAAGGAGGTGCTGAAGACGCTGGAGGGAAAGAAGTCAGAGCTGGTGAGGGAAGTAGGAAGGTTGAAGAAGGGAGCCTTGGAGG
701 AGATGAAGGCAGGAAAGGGGTTGGTAGAGGAGGTAAAGGGAGAGCTAGGAAAGCTAGAGQCTAAGAAGATAGAGGAGGTGAAGGCGGCAAAGGGGGAAGT
801 GTCGTTTTTCCCAGGAATAAGAGAACTCACCAAAACTGGAAAGGAGTCGGAGTTGCTGGAAAGGCTTGAGGAGGGGCTGCAGGCGGCTAAGAAGATAAAG
901 AAGGTGGAGGGCCAGGAGAAGCTGTTACAAAATCTCAAGAGCAAGCTAGAAGAACTAGCAGCGATAAGGGAGCTGAAGGAAGGGCCAGAGCTGGTACAGA
1001 TGCTCAAGGAGATAAGACTGGAGCTGGAAGAGCAGCTGGGGATGCTAGCTGAGATTAAGGAAGTTAAGGCACTGGCAGAGAAGCGGCAGGCTGGAGGACT
1101 GGAACTTCAGGAGGTATTACAGCTTACAACTAGAACTGTGGCATTGGAGAGGAAGCTGGTAGAAAAGGTAAGTAAGCTAGATAGGAAGAGCCTGGAAAGT
1201 CAGAAGAAGGTACTGGAGGAGCATAAGAAOGAGCTGGAGCAGGAAAAGCTTAAGCTAACTAAGGAGCTGOCTGAGAAGCGAAAAGCCAAAGGGCTCTTTG
1301 GAGCAGTACTAGACCT GGGACTGGT GGCGCT GGGT GTAGGAGAAACAAAGGACTGGGAGTT AGT GAT ACTGGAGAAAAAACT AGCCAAGAT CAACGCACT
1401 CCTAGAGGGTGGAGATATAGCAAAACTGGGAAGACAGATAAATAGGATCAAGTGGCTAGAGGATCTAGCTGTTAGTAAGAAGCTTAAGGCAGCATTAGCT
1501 AGTGCGATGGAGATAGCTAAGAACAGGGGTTGGACTGATTATTTAAACAGTCTAGATGTAAGTGAGAGAGCTAATGCTGCAAAGGAATTAATTGCTGCTG
1601 AGAAAATTAGAAAATGGGCTAGGGATATTAATAACTTGGATGCTGATGAACGGGCAATGGTCGCTGGGGCCCTAACCCTTTTCCTAACCCCTTCTACAAC
1701 AGTGTGACTCACCACTCTTCCTCGCAACATCGCTCACACTCCTAACCGGCCACTGTACTGAGGGTAATACCAGTTAGTGCAAGGGTGGACGTAGAGGTTT
1801 TGGGAGAGTQGTGATGTTGCTAAAACTTAGATTCTTACTCATTGCTGGAGTATTAGCATTGGGTATCTTACATGGTGTACCTGCAGGGGCTGCTGCTCAG
1901 GCGAGCCCTAGTACTACTGGTACTGAAGCTGATGATACAQCGACTGTTGTTATATGCAATGTTATACGGTTTGTGCAGAAGCTGGGACTACCCATCATGA
2001 CTGGGGTGATACTAGGTTCTAGCATTATGGCTATCTTTGGTAAGCTTGCATGGCCTGCTATTGTAATGCTGGTTGTATTTACAGCTATATTCTTTGGTGC
2101 TGGTAAGCTTATGGCTAAATTTGCTGCTGQGTTGAGTGGTGATGGCATAGGTGATGCCAGTAAGTTTGACTGTTCACAGCATACAGCGGTCAACACCAGC
2201 CAAAGTAACCGATAGAAGATTAGGTGCTGCTGGGCAGTGATGGTGCAGAGAGCTGTGTTGGCATTGGTTQGGAGATGAGQCAGCGGGAGTTCGGAGATAG
2301 GGACTAGGTQCATGGACGATGCAGGTGTGGGCAGAGAGCTGTGTTGGCCTTGGTAGTGAGTTGGGCTGAGGTGCCAAGGQCAGGAGGAGGTACCAGAAGT
2401 TGGGAGATGCTGGTTACCGTTGAAGTGAACCGGTGCAGAGAGCTGTGGCCrTGGGGCCCAGTAGAGTTGGGCTGAGGGGCCAGGGGTAGTGAAGGACACG
2501 ATTTGACTGTTCCAGTTATAAGGAAGTCAAGACCGGCACCAAACCAAGCTAAAGGTTAGGAAATGAAGTGCCGCGTGATGGTGCAGAGAKTGTGGTGGT
2601 AGTACAGGGGTGCTGGTGATAGTTTCTGGTGCTGCAGGGGATCAGAGGACTCCTACTCATGCTGCGGATTATCTCCTTTTGTGTTCCCACAGGTTCACAT
2701 AATTAACCGTACATGGGTAAATGTGCTGGTAGTGCCGGGAACCGCAAGTGGGTAGATGGTGGAGGGGCACCATTGGGGTGGGTGTCCTAGAATCCGAACC
2801 TGATTCCTAGTTCTACCCCTATATTGCGTAGCCCGAAGGATGCTGAGACCTTCTCCTTTGTTTTTCCTAATGGGCTGGCGTCATCGACTQCACGGTGGGC
2901 AGGGGTGTTGTCGTAGTTTGTGCCTAGTGTTTTGCTGGCAGTGCCGCCAATGAATGCTGOCAGGCTGGGTGTGATGCTGTAACTCACACCAGCCTTGACC
3001 TGGCAGGTAAGCTGTGGCTGAAATCGTTGGGCGGTCAGTCCTACAAAGCTTGCACCCAACCCTGCGCAGQCGTAGGGAGACATTCTCCAGTCAACTAGTA
3101 ACCCCACCTGTGGGAAATCATAGCATGCACTCAGGATCGCTGAAACTACCCTGACCCCACTTATTTCCACAATCTCCATGCCCTCTGTGCTTGTGGCAAT
3201 TGCCCGACCAATGACGCCTTTTGCGAGCCGTGAGAGGCCAACAATATCCGTGATAGCTTGATTCCGCTTGCTCTCTTGCGTGTTCTGTGCAGAGGTATCA
3301 GCGCTACCAGCACCGGTTGCAGCAGCACGTTCAGTTTTGGCCTCCTCCCCAGGTATGATQGCATCAAACTTACTTATAAOCTTGTCAGCAAGCGCTTTGC
3401 TTTCCCACGCTTCGTCATGGTCTCTAATATCTTCCAAACGCTTCTTGATGGATCTCAAAOCCGGCTTGGACAACTCCTTCTGCAGCGCGGCCCGAAACCT
3501 TCGACGTCACGCGTCGCACTGGCAGATATTCTTTTCACCAGGGCAAAAGTCGCACTCCCQCCGCTCAGTAAGGCATAATTTTGCGATTTGAAATCATAGT
3601 GCTCATGTCCGAATTCTAACTCAAGCCTGGTGCCGCGTAOCACGCAACCCACGCTTCCCCTGAGGCCCAACAGACTGCTGTTCTCAAACCCCATGGCCGG
3701 ACTCTCCGTAGATTCCCCCQCCCAATCGTAGTTGGCGGAAGAAAGCTCAGCAGAGCCCAACTTCCCAATGTAGGGCAGTATCGCGACGGTTTCTTTTGCA
3801 GCATTAAGCCTAAACCCGCTTATCTTCCCTATGGTGGGCCEGTACCCAAAACTCACGTAAAAATCTTCAOGACCGTTCATCCCATGGGAACTCGCTGGGA
3901 CCAGATAACCAATCGTACTCAGCAATGCAGCAGCTATACATACTACGGGaTCATGACGCACCCAATGACACCGACAACCAACAGGAGGCTAAACGCCCC
4001 TCACTGCACGACAGTGCTATACAACTGTTAAAAAGGTCTTAAAATGAGTTGCATTTTTGAAGGGCTTGCAACTATGCCTAGCTAGAGAATAAGGATTCCG
4101 GGCT GGCGGGCCAAACGGCGCACACT CAGGTT GCCCTTCGTGT GCAT GCAT CAT CAGCGTGGCACTACCTACTTTTTCAGCACCG

69
regulatory sequences for gene expression. Clones MSP3-11 and
MSP3-19 lack the N-terminus but each contains the C-terminus
and 1,473 bp and 2480 bp respectively, downstream to the 3'
end of the open reading frame. A schematic representation of
each clone shows areas of homology between the 3 genes as well
as areas which are homologous to the MSP2 gene of A. margnale
(Fig. 12).
Homologous areas are distributed throughout the 3 MSP3
genes. Areas common to all 3 genes are at bases 1191 to 1472
and bases 1818 to 2008 of the MSP3-12 clone. An area at the
3' end of MSP3-12 between bases 2090 to 2280 is also found in
MSP3-11, but is absent in MSP3-19. The last 200 bp at the C-
terminal end of genes MSP3-11 and MSP3-19 are homologous.
This area is unavailable for comparison in MSP3-12.
The MSP2 gene encodes a 36 kDa surface protein, and is
known to be expressed by a polymorphic, multigene family
(Palmer et al., 1994). Areas of homology between the N-
terminus of the MSP2 gene and MSP3-12 are indicated in Fig.
12. The N- terminus is absent in the MSP3-11 and MSP3-19
clones, so comparisons cannot be made. The areas of amino
acid sequence homology between MSP3-12 and MSP2 (amino acids
55 through 176) shows 65.6% similarity and 54.9% identity
(Fig. 13). There is a homologous area of over 500 bp between
MSP3-11 and MSP2 (Fig. 12). However, this area is outside of
the open reading frame, in a different reading frame, and a

Fig. 12 Diagram of MSP3 clones with MSP2 homology.
A schematic representation of clones MSP3-11, MSP3-12, and MSP3-19. Homologous
regions are indicated by like shaded areas. Regions homologous to the MSP2 gene
are indicated by MSP2->. Arrows indicate the direction of the reading frame.
Nucleotide numbers are indicated on the bottom. Top numbers of clones MSP3-11 and
MSP3-19 indicate corresponding nucleotides in clone MSP3-12.

MSP3 CLONES
MSP3-12
Start
-or-:
Q-i
\ -
I 613 794 1141 1191 1306 1472
MSP2->
1818 1953 2008 2090 2280 2337
MSP3-11
MSP3-19
1191 1306 1472 1818 1953 Stop
<-MSP2

Fig. 13 Comparison of MSP3 and MSP2 protein sequences.
BestFit alignment of amino acid sequences of MSP3 (top)
and MSP2 (bottom). Identical amino acids are indicated
by a vertical line. Conservative substitutions are
indicated by an asterisk (*). The symbol is used
to denote a gap used to achieve optimal alignment between
the sequences. Amino acid numbers are indicated at the
beginning and end of each line.

73
BestFit Allignment of MSP3-12 (top)
and MSP2 (bottom)
55 GSFYIGLDYNPTFNGIKDLKIIGETDEDE 83
I I I I I 11 11 I I III I- I
50 GSFYIGLDYSPAFGSIKDFKV.QEAGGTT 77
84 MDVLTGARGLFPMNALASNVTDFNSYHFD 112
78 RGVFPYKRDAAGRVDFKVHNFD 99
113 WSTPLPGLEFGNSTL ALGGSIGYRIGGA 140
100 WSAPEPKISFKDSMLTALEGSIGYSIGGA 128
141 RVEVGIGHERFVIKGGDDAA FLL 163
129 RVEVEVGYERFVIKGGKKSNEDTASVFLL 157
164 GRELALDTARGQL 176
158 GKELAYHTARGQV 170

74
different direction. A similar, but much smaller area is seen
downstream to the open reading frame of MSP3-19 (Fig. 12).
MAb and Immune Sera Reactivity to Recombinant MSP3
Lysates from E. coli cells transformed with pBluescript
containing each of the 3 MSP3 genes were separated by SDS-
PAGE, transferred to nitrocellulose and reacted with anti-MSP3
MAb (AMG75C2) or immune sera from cattle infected with a VA or
FL strain of A. margnale. This was done to verify that a
cloned MSP3 gene represented a gene which produced one or more
of the 86 kDa antigens seen on 2-D immunoblots. Initial
bodies from a FL strain of A. margnale were used as a
positive control for MAb (AMG75C2) and immune cattle sera.
Lysate from E. coli transformed with empty pBluescript, anti-
Trypanosome brucei-KAb (TYRP1E1), and pre-immune cattle sera
were used as negative controls.
MAb AMG75C2 reacts only with recombinant protein
expressed by clone MSP3-12 (Fig. 14) The recombinant MSP3
has a molecular size of approximately 65.0 kDa. This
corresponds to the calculated size of the protein expressed by
the open reading frame of MSP3-12 which is 1,707 bp (Fig. 10).
No reactivity was observed between MAb AMG75C2 and recombinant
MSP3-11, MSP3-19, or empty pBluescript. Isotype control MAb
TRYP1E1 showed no reactivity as well.
Immune serum from a cow infected with a VA isolate of A.
margnale reacted with recombinant MSP3-12 (Fig. 15).

Fig. 14 Immunoblots of expressed MSP3 clones.
The reactivity of expressed proteins from clones MSP3-11, MSP3-12, and MSP3-19 are
compared using an anti-MSP3 monoclonal antibody (AMG75C2) and a negative isotype
control (TRP1E1). pBlue is expressed proteins from empty pBluescript plasmid used
as negative control. FL is A. margnale initial body preparation from a FL isolate
as a positive control. Numbers on the left indicate molecular size markers in
kilodaltons.

Antiqen
FL
MSP3-11
MSP3-11
MSP3-12
MSP3-12
MSP3-19
MSP3-19
pel Lie
pel ue
Antibody
AMG
Trp
AMG
Trp
AMG
Trp
AMG
Trp
AMG
75C2
1E1
75C2
1E1
75C2
1E1
75C2
1E1
75C2
M.U.
Id)
200.0 --
97.4
46.0 --
30.0
21.5 --
14.3
O)

Fig. 15 Immunoblots of expressed MSP3-12 clones.
The reactivity of pre and post-infection sera from animals infected with a Florida
isolate of A. margnale (Pre and Post-FL) or a Virginia isolate (Pre and Post-VA)
are compared using expressed proteins from clones MSP3-12 and empty pBluescript
plasmid (pBlue). Antigens are listed on top and anti-sera is listed on the bottom.
In lane 1, initial body preparation from a FL isolate of A. margnale (FL) is
reacted with immune sera from an animal infected with a FL isolate as a positive
control. Numbers on the left indicate molecular size markers in kilodaltons.

Antiqen FL
pfilue
NSP3-12
pfilue
MSP3-12
pel ue
MSP3-12
pel ue
NSP3-12
Antibody Post-
Post-
Post-
Pre-
Pre-
Post-
Post-
Pre-
Pre-
M.U.
FL
FL
FL
FL
FL
VA
VA
VA
VA
200.0 --
97.4 --
69.0 --
30.0
21.5 --
'O
CO

79
However, when this same recombinant protein was reacted with
immune sera from an animal infected with a FL isolate, no
reactivity above that which was seen with empty pBluescript
was observed.
Representation of pBluescript MSP3-12 in the
A. margnale Genome
To verify that the cloned pBluescript MSP3-12 faithfully
represented the genomic copy, genomic DNA from a FL isolate of
A. margnale and pBluescript MSP3-12 DNA were digested with
restriction enzymes to yield predicted fragments of specified
lengths. The enzymes used, the restriction sites in
pBluescript MSP3-12, and the predicted size of the fragments
are illustrated (Fig. 16) DNA fragments were Southern
blotted to nylon filters and probed with whole digoxigenin-
labeled MSP3-12. All enzymes used produced pBluescript MSP3-
12 fragments of the predicted size which comigrated with a
fragment in the genomic DNA (Fig. 17). Comigration of large
fragments was seen in digests extending from the 5' terminal
region to the 3' terminal region of the gene.
Presence of Multiple MSP3 Gene Copies
A. margnale genomic DNA from either FL, SI, or VA
isolates was digested with restriction enzymes, Southern
blotted to nylon membranes, and probed with whole,
digoxigenin-labeled MSP3-12. Enzymes Sph I, Nde I, Sac I, and

Fig. 16 Representation of clone MSP3-12 in the A. margnale genome.
Diagram of clone MSP3-12 indicating the restriction enzyme sites and the resulting
fragments used to map MSP3-12 to the genome. Nucleotide cleavage site for each
restriction enzyme is illustrated diagramatically (top) and by nucleaotide
number(bottom). Size of resulting fragments in base pairs is also provided.

> > ^ N S ^
^cV ^ ^ < Restriction Cleavage site in Size of co-migrating
Enzyme pBIuescript MSP3-12 plasmid and genomic
fragments
Neo I 12/1,188 1,176
Eae I 52/1,'762 1,710
BspM 193/2,076 1,886
oo

Fig. 17 Genomic representation of MSP3-12.
Southern blot of genomic DNA from a Florida (FL) isolate of A. margnale or
pBluescript MSP3-12 (CL), digested with restriction enzymes Eae I, Bsp M, or Nco
I and probed with digoxigenin-labeled MSP3-12. Molecular size markers in base
pairs (bp) are indicated to the right and left.

Eael
Bso M
FL CL
FL CL
bp
5,090 --
4,072 --
3,050 --
2,036 --
1,636 --
1,018 --
506 --
396 --
344 --
298 --
Nco I
FL
CL
bp
23,130
9,416
6,577
4,361
-- 2,332
-- 2,057
564
00
OJ

84
Hie II were chosen because they do not cut within the
sequence of the MSP3-12 gene. This should produce a single
band if only one genomic copy of MSP3 is present. Sac I cuts
once within the open reading frame of MSP3-19 (at nucleotide
498), potentially producing 2 observable bands for a single
copy gene. None of the enzymes cut within the MSP3-11 gene.
Multiple bands were observed on Southern blot
hybridizations, indicating multiple partially homologous MSP3
copies (Fig. 18) The exact number of copies cannot be
determined since restriction site polymorphism may exist in
other MSP3 copies, resulting in more than one band from a
single copy. Similar intensities of many of the bands within
a single isolate may indicate extensive homology exists
between some of the copies. Restriction fragment length
polymorphism (RFLP) is seen when comparing enzyme digests of
each of the 3 isolates of A. margnale.
Distribution of MSP3 Copies in the A. margnale Chromosome
Intact, A. margnale genomic DNA from 3 isolates (FL, SI,
& VA) was digested into large fragments with restriction
enzymes Sfi 1 and Not I, and separated by CHEF
electrophoresis. The gel was stained and photographed (Fig.
19) These fragments are nonoverlapping and have been
previously shown to represent the entire 1,250 kbp A.
margnale genome (Alleman et al., 1993). The Sfi I and Not 1

Fig. 18 Presence of multiple copies of MSP3.
Genomic DNA from a Virginia (VA), South Idaho (SI), or Florida (FL) isolate of A.
margnale was digested with the restriction enzymes indicated above the lanes and
Southern blotted. EcoR I digested pBluescript MSP3-12 (pB 12) and undigested
bovine calf thymus DNA (CT) were used as controls. The blots were hybridized with
digoxigenin-labeled pBluescript MSP3-12. Molecular size markers in base pairs (bp)
are indicated to the right and left.

bp
5,090
4,072
3,050
2,036
1,636
1,018
506
396
344
298
Soh I Nde I Sac I Hie II EcoR I
bp
VA SI FL VA SI FL VA SI FL VA SI FL pB 12 CT
-- 23,130
-- 9,416
-- 6,577
-- 4,361
-- 2,332
-- 2,057
564
00
CTi

87
fragments produced here are identical to those previously
reported (Alleman, et al., 1993). Sfi I digestion of the FL
isolate produced 12 bands ranging in size from 170 to 14 kbp.
This same gel was then Southern blotted to nylon filters
and probed with whole, digoxigenin-labeled MSP3-12 (Fig. 20).
The probe hybridized to multiple fragments of Not I and Sfi I
digests of all 3 isolates. Most of the gene copies in the FL
isolate appear to be contained within 2 large Sfi I fragments
with previously reported band sizes of 170 kbp and 137.5 kbp
(Alleman et al., 1993). However, the results do indicate the
MSP3 gene is widely distributed throughout the genome in all
3 isolates tested. Gene copies were present on 6 of the FL
and VA Sfi I digested fragments and 5 of the fragments from
the SI isolate (Fig. 20) Not I digestions yielded 8 bands in
the FL isolate which hybridized to the MSP3-12 probe, 7 bands
in the VA isolate, and 5 bands in the SI isolate (Fig. 20).
RFLP is also observed between isolates on CHEF gels.

Fig. 19 Not I and Sfi I digestion of the A. margnale
genome.
Genomic DNA from a Florida (FL), Virginia (VA), or South
Idaho (SI) isolate of A. margnale was digested with the
restriction enzymes indicated above the lanes, separated
by clamped homogeneous electrical field gel
electrophoresis, and stained with ethidium bromide. Lanes
1 and 2 contain lambda DNA-tfindlll fragments and Promega
Delta 39 markers, respectively, as size markers.
Molecular size markers in kilobase pairs (kbp) are
indicated to the left.

89
S/7 I Not I
1 2 FL VA SI FL VA SI
kbp
i W I
195
156
117
9
*
__
78
* *
if?
39
23.1
n
9.4 --
m
6.6 --
o
4.4 --
to

Fig. 20 Distribution of MSP3 in the A. margnale
genome.
Genomic DNA from a South Idaho (SI), Virginia (VA), or
Florida (FL) isolate of A. margnale was digested with
the restriction enzymes indicated above the lanes, and
separated by clamped homogeneous electrical field gel
electrophoresis. The fragments were then Southern
blotted and probed with digoxigenin-labeled pBluescript
MSP3-12. Molecular size markers in kilobase pairs (kbp)
are indicated to the left.

91
9.4

CHAPTER 5
DISCUSSION
Rational for Studying the MSP3 Antigen
The current serologic tests available for the diagnosis
of A. margnale infection are based on antigens which are a
crude mixture of A. margnale and erythrocyte proteins. This
reduces both the sensitivity and specificity of various tests
to unacceptable levels, particularly in detecting carrier
cattle (Goff et al., 1990; Luther et al., 1980; Maas et al.,
1986; Todorovic et al., 1977). Because of this, and because
A. margnale cannot be successfully maintained in erythrocyte
culture, efforts using molecular techniques have been
attempted to identify an immunodominant protein with
acceptable sensitivity and specificity to be used in
recombinant form as purified test antigen. The ideal test
antigen would be one which is not cross reactive with antigens
from related organisms, and could detect acutely infected
animals as well as carrier cattle infected with any of several
different geographic strains of A. margnale. This antigen
should preferably be conserved in all isolates of A. margnale
and be from a single copy gene to limit genetic recombination
and antigenic variation.
92

93
One antigen identified as immunodominant in the acutely
infected animal as well as chronically infected carriers is
the 86 kDa MSP3 protein from a FL isolate of A. marginale
(McGuire et al., 1991; Palmer et al., 1986). The FL isolate
of A. marginale was used for antigen isolation because it has
been found by adsorption studies to contain antigens common to
both morphologic types of A. marginale, the tailed and non-
tailed forms (Goff and Winward, 1985; Kreier and Ristic,
1963). Experiments using affinity purified MSP3 showed
excellent test sensitivity, however, attempts to produce a
recombinant form of MSP3 have resulted in inconsistent
responses to immune cattle sera. Possibilities for these
inconsistencies include 1) an inability to produce an
antigenically similar recombinant form of this protein in the
chosen vector (E. coli), or 2) variants of MSP3 exist within
or between different strains of A. marginale. Antigenic
variability has been previously identified in another major
immunodominant surface protein of A. marginale, MSP2 (McGuire
et al, 1984; Palmer et al., 1994). In these experiments, not
all organisms between strains, or within the same strain
reacted with a given MAb to MSP2.
In addition, the specificity of MSP3 in detecting A.
marginale infected cattle has not been totally established.
Previous experiments have determined there is no cross
reactivity between MSP3 when reacted with sera from animals
infected with B. bovis, B. bigemina, or an unidentified

94
rickettsial agent isolated from an aborted calf fetus (McGuire
et al., 1991). However, using 16s rRNA sequencing analysis,
A. margnale has been shown to phylogenetically be more
closely related to C. ruminantium and the Ehrlichia sp.,
particularly E. risticii, E. equi, and E. canis (Dame et al.,
1992; Van Vliet et al., 1992).
Initial experiments suggested FL MSP3 was conserved
enough to be recognized by sera from animals infected with
different isolates (Palmer et al., 1986). However, since this
study used single dimension gel electrophoresis, reactivity of
immune sera with different proteins of comigrating molecular
size could not be ruled out.
The specificity of MSP3 in detecting A. marginale
infected carriers, the conservation of MSP3 between various
strains of A. marginale, the reactivity of sera from animals
infected with different strains of A. marginale to MSP3 from
a FL isolate, and the gene(s) which encode the MSP3 antigen
needed to be investigated in order to determine if a
recombinant form would make a suitable diagnostic test
antigen.
The Specificity of MSP3 in Detecting A. marginale Infection
Antigenic similarities between Anaplasma sp. ,
particularly A. marginale and A. cntrale have previously been
identified (Palmer et al., 1988a; Shkap et al., 1990; Shkap et
al., 1991). In our study, A. cntrale antiserum reacted with

95
several A. cntrale proteins, however, this same antiserum
showed no reactivity against A. margnale antigens. In
previous experiments, strong reactivity was seen with the 36
kDa protein (MSP2) of an Israel isolate of A. margnale when
reacted with immune sera from a cow infected with A. cntrale
(Shkap et al., 1991). In addition, epitopes common to both
species were identified using MAbs against the 36 kDa and 105
kDa surface proteins of A. margnale (Palmer et al., 1988a).
Discrepancy between the results of this study and previously
reported data may be explained by differences in geographic
isolates of A. margnale used. Antigenic differences between
different isolates of A. margnale have been well documented
(McGuire et al., 1984). In our study, a FL isolate was used,
whereas an Israel isolate of A. margnale was used in a
previous study which identified common immunodominant proteins
between the 2 species (Shkap et al., 1991). In addition, it
is difficult to compare results of immunoprecipitation
techniques used in previous experiments (Palmer et al., 1988a)
with immunoblots used here. Immunoblotting may alter epitopes
during the process of denaturation and electrophoretic
transfer prior to interaction with antibodies. Our results do
indicate however, that certain epitopes on the MSP3 protein of
the FL isolate of A. margnale are distinctly different from
those recognized by cattle infected with A. cntrale. This
could be a useful distinction since A. cntrale is

96
nonpathogenic and carrier cattle are not a threat to livestock
production.
In contrast, serum from sheep infected with A. ovis
showed a strong positive reaction with MSP3 of A. margnale as
well as a 36 kDa antigen, possibly MSP2. In previous
experiments, a panel of 18 MAbs reactive to various A.
margnale proteins failed to react with A. ovis infected
erythrocytes (McGuire et al., 1984). The difference in
results may be explained by the limited number of epitopes
recognized by these MAbs. Our results confirm cross
reactivity exists between at least 2 of the major antigens of
A. margnale and serum from animals infected with A. ovis.
These results indicate a diagnostic test using MSP3 of A.
margnale may be able to detect A. ovis infected sheep. The
cross reactivity between these two species should not present
a problem in cattle since A. ovis fails to infect either
intact or splenectomized animals (Jain, 1986).
The Ehrlichia sp. and C. ruminantium have been shown to
be phylogenetically closely related to Anaplasma marginale
(Dame et al., 1992; Van Vliet et al., 1992). Our results
confirm cross reactive proteins do exist between these species
when sera from animals infected with any of these organisms is
reacted with initial body preparations from a Florida isolate
of A. marginale. Only sera from animals infected with E.
risticii and E. ewingii react with the MSP3 antigen. These
species do not infect cattle. However, the results illustrate

97
the potential for cross reactivity between A. margnale and
Ehrlichia bovis, particularly since cross-reactive antigens
exists among many Ehrlichia sp (Holland and Ristic, 1993;
Ristic and Holland, 1993; Uilenberg, 1993). This is a
potential problem for serologic testing of cattle in areas
where A. marginale and E. bovis coexist.
Serum from an animal infected with C. ruminantium does
not react with an 8 6 kDa antigen, but does react with a
protein of slightly smaller molecular mass, and an antigen
which has a molecular mass of just above 30 kDa. Recently,
sequence homology has been demonstrated between genes encoding
immunodominant surface proteins of A. marginale and C.
ruminantium. Amino acid sequence analysis revealed 33%
identity and 40 similarity between the C. ruminantium MAPI and
the A. marginale MSP4 (Van Vliet, et al., 1994). Significant
amino acid sequence similarity was also revealed between the
21 kDa MAP2 protein of C. ruminantium and the 19 kDa MSP5
protein of A. marginale (Mahan et al., 1994). In addition,
extensive amino acid sequence homology (59% similarity; 33%
identity) was conserved in multiple oligopeptide sequences
throughout the MSP4 (31 kDa) protein and the MSP2 (36 kDa)
protein of A. marginale (Palmer et al., 1994). Hence,
sequence similarity was also identified between MSP2 of A.
marginale and MAPI of C. ruminantium (Palmer et al., 1994).
Considering the close phylogenetic relationship between these
2 organisms, the known amino acid sequence homology between

98
some surface proteins, and the fact that C. ruminantium
antiserum reacts with several A. margnale antigens on
immunoblots, caution must be taken in the development of a
diagnostic test which can distinguish between these 2
pathogens.
Sera from animals infected with B. bovis or B. bigemina
showed no evidence of cross reactivity with A. margnale
preparations. These organisms are not considered to be
closely related phylogenetically (Dame et al., 1992; Van Vliet
et al., 1992), nor has any similarity in antigens or DNA been
detected in other experiments (Eriks et al., 1989; McGuire et
al., 1984; McGuire et al., 1991; Shkap et al., 1990).
Based on our results, there is no strong evidence which
would exclude MSP3 as a diagnostic test antigen based solely
on the specificity of this protein. However, because there is
strong cross reactivity with sera from animals infected with
Ehrlichia sp. which do not infect cattle, the specificity of
MSP3 still remains questionable, particularly in areas where
E. bovis and A. marginale coexist.
Size Polymorphism of MSP3 Among Various
Strains of A. margnale
Size polymorphism is demonstrated in the MSP3 proteins in
3 of the 4 geographic isolates studied. Marked size
polymorphism has been recognized in other major surface
antigens of A. marginale, the MSPla protein (Oberle et al.,
1988), and the MSP2 protein (Palmer et al., 1988b). Size

99
variation of MSPla among geographic isolates was later
explained by various numbers of tandem repeats within a single
domain (Allred et al.,1990). Despite the marked size
variation of MSPla, many of the surface epitopes, including a
neutralization-sensitive epitope remained conserved between
strains (Allred et al., 1990; Oberle et al., 1988).
The MSP2 protein was found to be encoded by a multigene
family in which substantial nucleotide sequence polymorphism
exists among MSP2 copies. Variability in the expression of
these genes among organisms suggests one potential mechanism
for size polymorphism. This surface protein was found to be
antigenically polymorphic as well (McGuire et al., 1984;
Palmer et al., 1994).
Size variation has been reported in the major
immunodominant surface protein of another rickettsial agent
closely related to A. margnale, the 32 kDa MAPI protein of C.
ruminantium (Barbet, et al., 1994). The conservation of
antigenic epitopes on the protein or the mechanism of size
variation is not known.
Although the mechanism of size variation of the MSP3
protein is not known, this illustrates the organism's ability
to alter important, antigenic surface proteins. The presence
of a complex multigene family for MSP3 or the presence of
variable numbers of tandem repeats are potential mechanisms
for size polymorphism. If tandem repeats are responsible for
the size polymorphism of MSP3, important antigenic epitopes

100
may be conserved, particularly if epitopes are contained
within tandem repeats as with MSPla. If a complex family of
closely related genes is responsible for size variations,
similar to the MSP2 protein of A. margnale, antigenic
variations are more likely to exist as well. Size
polymorphism in MSP3 could pose a problem to its use as a
diagnostic test antigen, if it results in variation of
antigenically important epitopes. This may result in poor
reactivity between the MSP3 antigen of one isolate and immune
sera from an animal infected with a heterologous strain.
Immune Response to MSP3
This study demonstrates variations in reactivity of
immune sera from cattle infected with different geographic
isolates when reacted to the FL MSP3. Multiple 86 kDa
antigens are seen using immunoblots of 2-D separated
preparations. Reactivity of antisera with these antigens
varied depending on which geographic isolate the cattle were
infected with. In a homologous reaction with anti-FI serum
2 major areas of reactivity at pis 6.5 and 6.2 are identified.
These areas are distinctly different from the antigen
recognized by the anti-MSP3 MAb and the anti-VA serum (pi
5.6). The anti-WA serum recognizes 2 entirely different 86
kDa antigens (pi 5.3 and 5.1). Only serum from SI infected
cattle reacts with all the 86 kDa antigens identified with
other antisera and the MAb. Although the anti-MSP3 MAb

101
reacted with only a single 86 kDa antigen, rabbit-Anti-MSP3
polyclonal serum reacts with antigens in areas of the pH
gradient identical to those recognized by anti-FL serum, Anti-
VA serum, and the ant-MSP3 MAb. This rabbit sera was made by
injection of purified MSP3, isolated from an affinity column
using MAb AMG75C2. The production and reactivity of this
serum has previously been described (McGuire et al., 1991).
Reactivity of different 86 kDa antigens with rabbit-anti-MSP3
suggests that common epitopes may exist on these antigens.
The above results indicate, similar to the MSP2 protein
of A. margnale, not only size polymorphism, but also
antigenic polymorphism exist between MSP3 antigens of
different isolates. The ability of the organisms to alter
this surface antigen could present a problem for use as a
diagnostic test antigen, resulting in a test with low
sensitivity which could not reliably detect infection in
animals infected with different strains of A. margnale.
There are at least 3 possible explanations for the
multiple 86 kDa antigens present in the FL isolate of A.
margnale. These antigens may arise from, a) post-
translational modification of a protein transcribed from
single copy gene, b) several closely related genes transcribed
and translated from a multigene family, or c) entirely
unrelated genes. Reactivity of the rabbit-anti-MSP3 sera with
multiple 86 kDa antigens suggests at least 3 of the MSP3
antigens share common epitopes. This would indicate the 86

102
kDa antigens are likely the result of either post-
translational modification of a protein from a single copy
gene, or transcription and translation of several closely
related genes from a multigene family.
Previous studies have shown multigene families encoding
major surface proteins of A. margnale. Between 7 to 10
similar gene copies encode the MSP2 gene of A. margnale
(Palmer et al., 1994). MSP2 is a 36 kDa protein, and our
results illustrate multiple MSP2 antigens using 2-D gel
electrophoresis and immunoblots with an anti-MSP2 MAb
(ANAF19E2). These results support the findings of the MSP2
multigene family, and suggest that transcription and
translation of this family produces variations in the MSP2
protein. In addition, the MSP1/3 subunit of the MSP1 gene of
A. margnale also is encoded by a partially homologous
multigene family (Viseshakul et al., 1994). Genes in this
family were shown to differ from each other by extensive
deletions, insertions and rearrangements of sequences
(Viseshakul et al., 1994).
These previous experiments, along with data presented in
this study, prompted us to investigate the possibility that
the MSP3 antigens could be encoded by a multigene family. If
this is true, it would further demonstrate the ability of the
organism to use multigene families to vary important
immunogenic surface proteins. In this case, immunogenic
epitopes conserved in all strains of A. margnale may need to

103
be identified if these antigens are to be useful as diagnostic
test antigens or vaccine candidates.
Identification of pBluescript MSP3-12 as a
Recombinant Form of an 86 kDa Antigen
Anti-MSP3 MAb AMG75C2 bound expressed protein from an
MSP3 clone, pBluescript MSP3-12. The reactivity of this MAb
has been previously described (McGuire et al., 1991). Because
this MAb also binds to one of the 86 kDa antigens seen on 2-D
gel electrophoresis (pi 5.6), we are able to identify clone
MSP3-12 as a member of the MSP3 gene family. In addition,
immune sera from an animal infected with a VA strain of A.
margnale, but not a FL strain, reacts with recombinant MSP3-
12. This is supportive evidence that MSP3-12 shares common
epitopes with the 86 kDa antigen (pi 5.6) since VA sera
strongly reacts with this antigen on 2-D immunoblots whereas
immune sera from cattle infected with a FL isolate react
poorly if at all.
The MSP3-12 clone contains the N-terminus of the protein
as well as 633 bp upstream to the start codon. The region
upstream to the open reading frame likely contains the
regulatory seguences of the MSP3 gene. This helps insure the
correct transcription and translation of the gene by E. coli
since rickettsial promotors have been shown to be recognized
by E. coli polymerases (Oaks et al., 1987). Genes containing
their own promotor regions will naturally be in the correct
reading frame. Anti-MSP3 MAb AMG75C2 does not bind to

104
proteins expressed by clones MSP3-11 or MSP3-19. The N-
terminus is lacking in these clones, therefore upstream
regulatory sequences of the gene are not available for E. coli
polymerase binding. The clones are in frame with the lacZ
gene in pBluescript, however, either they are not produced in
sufficient quantity without IPTG induction, or they do not
contain the epitope recognized by MAb AMG75C2.
pBluescript MSP3 is an Accurate Representation of
a Genomic Copy of MSP3
Numerous artifacts can occur during the process of
cloning causing disruption of the original form of a gene. In
constructing a library, noncontiguous fragments of DNA may
reanneal prior to ligation into the vector. In addition, many
cloning artifacts may occur once the vector is transformed
into the host cell. Some of these include deletions,
rearrangements, or endonuclease digestion of insert DNA by the
host cell. Repeats or hair-pin structures in insert DNA may
not be well tolerated by some host cells such as E. coli, and
the insert may be omitted or portions rearranged or deleted
during replication. Because of these, and many other
potential cloning artifacts, it must be shown that the
recombinant form of MSP3 is an accurate representation of
genomic MSP3. We have determined that pBluescript MSP3-12 is
an accurate representation of genomic MSP3 by 1) expression of
a recombinant protein which is bound by anti-MSP3 MAb and by
2) demonstration of comigrating bands of predetermined sizes

105
in cloned and genomic DNA when cut with restriction enzymes
Neo I, Bsp M, and Eae I. For example, Nci I digestion of
MSP3-12 produces a 1,176 bp fragment which hybridizes with the
digoxigenin-labeled MSP3-12 probe. Digestion of genomic DNA
from a FL isolate of A. margnale produces a fragment of
identical size which also hybridizes with the MSP3-12 probe.
Digestion of genomic and cloned DNA with Eae 1 and Bsp M
produces similar results. These enzymes cut various places
within MSP3-12 clone, ranging from nucleotides 12 to 2076.
This range covers almost the entire 2337 nucleotide sequence
of the cloned gene.
Multiple MSP3 Copies in the A. margnale Genome
Hybridization studies using digoxigenin-labeled MSP3-12
identified multiple copies of partially homologous MSP3 genes
in the genome of FL, SI, and VA strains of A. margnale.
Genomic DNA was digested with restriction enzymes selected to
cut outside of the MSP3-12 sequence. Hybridization of the
probe with a single fragment would be seen if MSP3 was encoded
by a single copy gene. Multiple fragments homologous to the
MSP3-12 sequence are identified. However, the exact number of
copies cannot be determined because restriction sites may be
polymorphic in other copies of MSP3, causing an exaggerated
estimation of the number of gene copies. In addition, more
than one gene copy may be present on large fragments of
genomic DNA. Although unlikely, it is possible that the

106
uncoded region of clone MSP3-12, which was included in the
probe, could cause hybridization to multiple bands not related
to the MSP3 if this sequence was repeated in the genome. This
region is likely to contain regulatory sequences of the MSP3
gene, but the possibility of sequence homology between this
region and other unnkown multigene families of A. marginale
does exist. To confirm this is not occurring, probes made
from an internal sequence of clone MSP3-12 will need to be
produced and used to probe digested, genomic, A. marginale
DNA.
These data do suggest a copy number of at least 10 to 15
MSP3 genes in the FL and SI isolates with slightly less, 7 to
10, in the VA isolate. With the gene size of MSP3 being
approximately 2.6 kbp, we estimate MSP3 occupies as much as
3.0% of the 1,250 kbp genome of A. marginale (Alleman et al.,
1993). The exact function of this major immunodominant
surface protein is unknown, however, its prevalence in the
small genome of A. marginale suggest a need to antigenically
vary this very immunogenic protein in response to stress from
the host immune system. The A. marginale genome is estimated
to contain 7 to 10 copies of MSP2, another immunodominant A.
marginale surface protein encoded by a multigene family
(Palmer et al., 1994). This gene family occupies > 1% of the
genome. The exact function of this protein is also unknown,
however, immunization of cattle with affinity purified MSP2

107
does appear to offer at least partial protection against
homologous and heterologous challenge (Palmer et al., 1988b).
Genetic polymorphism between isolates of A. margnale is
seen when comparing isolates after digestion with the same
restriction endonucleases. This is evident by variations in
the length of restriction fragments which contain the MSP3
genes in each isolate. These results are consistent with
previous experiments identifying restriction fragment length
polymorphism between geographic isolates in ethidium bromide-
stained gels (Alleman, et al.f 1993). In addition, our
hybridization studies indicate there are fewer MSP3 copies in
the VA isolate than in the FL or SI strains. This, plus the
fact that serum from animals infected with a VA isolate binds
only one of the MSP3 antigens (pi 5.6), whereas serum from
animals infected with FI or SI strains react with multiple
antigens, may suggest less antigenic variation of MSP3 occurs
within the VA isolate.
Cloning and sequencing of several complete MSP3 genes may
be required to understand the full extent of the genetic and
antigenic polymorphism which exists between expressed copies
of MSP3. The 3 clones made available to us contain partial
gene sequences of MSP3. Although large areas of identity
exist, there is also significant amino acid sequence
variation. Our data suggest the amino acid sequence variation
results in epitope or antigenic variation as well. This is
evidenced by variable reactivity of different immune sera to

108
multiple MSP3 antigens. In addition, the epitope recognized
by the anti-MSP3 MAb is present on only one of the 86 kDa
antigens.
It has been proposed that antigenic variation plays a
role in the cyclic rickettsemia and persistent infection
recognized in carrier cattle infected with A. margnale
(Kieser, et al., 1990). The level of rickettsemia varied
markedly at bimonthly intervals from <103 to >105 infected
erythrocytes per ml of blood (Eriks et al., 1989) The number
of infected erythrocytes gradually increased over a 10 to 14
day period, then precipitously decreased (Kieser, et al.,
1990). The length and consistency of the cycles suggests
recurrence is due to continual antigenic variation by the
organism and development of a primary immune response by the
host. Further work is needed to determine if antigenic
variation of MSP2 or MSP3 occurs during rickettsemia cycles in
persistent carriers. This could be done by identifying copy
specific epitopes on expressed MSP3 antigens and monitoring
changes in parasite antigens during these cycles.
Bacteria in the genera Borrelia and Neisseria have been
shown to use multigene families to vary important surface
antigens and aid in the evasion of the host immune system
(Meyer et al., 1990; Barbour, 1990). The genome of Borrelia
hermsii, the causative agent of relapsing fever, contains a
large repertoire of genes encoding variable major proteins
(Vmps) (Barbour, 1990; Barbour, 1991). Multiple silent and

109
active copies of the vmp genes are located on linear,
extrachromosomal, DNA plasmids. The promotor and active copy
of the genes are located at telomeric ends of the linear
plasmids. Switching of a silent copy to the active locus just
downstream from the promotor causes expression of an
antigenically different Vmp and conversion of the organism
from one serotype to another. Antigenic variation allows
Borrelia hermsii to evade the host immune system and avoid
complete clearance from the blood stream. This causes a
persistent illness with a cyclic rise and fall in body
temperature every 4 to 7 days (Barbour, 1991).
Neisseria gonorrhoeae uses multigene families to
antigenically vary 2 important adherence ligands, the pili and
outer membrane opacity proteins (Opa) (Meyer et al., 1990;
Sparling et al., 1990). Variations in pili are accomplished
by multiple, silent, incomplete copies (over 20) of pil genes
termed minicassettes (Sparling et al., 1990). Insertion of
one of these incomplete copies into an expression site can
result in the expression of an antigenically different pilus.
In contrast, 10 12 complete opa genes are present in
Neisseria gonorrhoeae, and more than one may be expressed
simultaneously (Meyer et al., 1990). Expression of individual
opa genes is dependent on a repetitive sequence, the coding
repeat, which encodes the Opa signal peptide. The number of
5-mer repeats present determines if the opa gene is
translationally in frame. Genes in which 5-mer repeats occur

110
in multiples of 3 are expressed while those with any other
multiples are not produced (Meyer et al.f 1990). Variations
in the number of 5-mer repeats occurs frequently during DNA
replication. Antigenic variation allows N. gonorrhoeae to
persist in the host unless appropriate antibiotic therapy is
instituted, and produce repeated infections in the same host
(Sparling et al., 1990)
The exact mechanism by which A. margnale uses multigene
families is not known. However, like Borrelia hermsii and
Neisseria gonorrhoeae, A. marginale does appear to evade the
host immune system and avoid complete clearance. In addition,
as in relapsing fever, fluctuating parasitemia is observed in
cycles consistent with the appearance of antigenic variants in
response to immune pressure from the host (Krieser et al.,
1990). We hypothesize A. marginale could use mechanisms
similar to those seen in B. hermsii and N. gonorrhoeae to vary
important antigenic surface proteins such as MSP2 and MSP3 in
an effort to avoid immune clearance.
Distribution of MSP3 Copies in the A. marginale Genome
Previously, we have shown that by Sfi I digestion and
separation of large fragments using CHEF electrophoresis, we
can separate the entire genome of A. marginale (FL) into 14
fragments ranging in size from 160 to 14 kbp (Alleman et al.,
1993) The MSP3-12 probe hybridizes to 6 of these fragments,
as well as multiple fragments in Not 1 and Sfi I digests of

Ill
other isolates. However, judging from the intensities of the
bands, it appears most of the copies in the FL isolate are
located on two large Sfi I fragments approximately 160 and 130
kbp. The smaller band has been shown to contain a doublet of
comigrating fragments (Alleman, et al., 1993). Therefore,
these two hybridizing bands could represent as much as 25% to
37% of the genome size. Because this is such a large area of
the genome, conclusions regarding the proximity of the genes
are difficult to make. The Not I digest of the FL isolate,
and the Not I and Sfi I digests of the other isolates
indicates a more even distribution of copy numbers throughout
the genome.
This suggests MSP3 copies are widely distributed
throughout the A. marginale genome, similar to the pattern
seen with the MSP2 multigene family (Palmer et al., 1994).
These copies are likely not the result of simple duplication
of a single MSP3 gene since the copies are not present in
tandem along a single stretch of DNA. This assumption is
supported by the partial gene sequences available for 3 of the
MSP3 genes. This would indicate that any coordinated
regulation of MSP3 copies would involve trans-regulation.
Summary and Conclusions
We have determined that the MSP3 antigen of A. marginale
is of questionable specificity as a diagnostic test antigen
with potential for cross reactivity with sera from animals

112
infected with Ehrlichia sp. We have also shown that this
antigen is not conserved among various strains of A.
margnale, and that antigenic variation likely exists between
isolates since immune sera from cattle infected with different
strains reacted differently to MSP3.
It has now been shown that 2 major surface antigens of A.
marginale, MPS2 & MSP3, are actually composed of a family of
related proteins. Using an MSP3 clone as a probe in
hybridization studies we concluded the multiple MSP3 antigens
are the result of a complex, multigene family of partially
homologous genes, similar to the MSP2 protein. Although we
cannot state definitively, we estimate that a relatively large
portion of this rickettsial agent's small genome (up to 4%) is
occupied by these 2 gene families. We hypothesize the
organism uses these multigene families to antigenically vary
these major immunogenic surface proteins.
The cross reactivity of this protein with sera from
animals infected with Ehrlichia sp., the size polymorphism of
MSP3 between different geographic isolates, the multiple 86
kDa antigens recognized by various antisera, and the presence
of a multigene family encoding these antigens indicate that in
its native form, a single recombinant MSP3 would not be a
suitable candidate for use as a diagnostic test antigen. In
order to be used as a test antigen, it may be necessary to
define and obtain expressed copies of the MSP3 gene. The
potential for using multiple recombinant MSP3 antigens

113
produced by the expressed copies could then be evaluated.
Alternatively, conserved epitopes on these genes could be
identified and recombinant or synthetic peptides derived from
gene seguences could be tested with immune cattle sera to
determine their reactivity. However, considering the
apparent ability of A. margnale to antigenically vary this
protein, we feel these attempts may not be practical,
particularly since another surface antigen (MSP5) has shown
some promise for use as a diagnostic test antigen (Ndung'u et
al., 1995). This 19 kDa protein is encoded by a single copy
gene and appears to be conserved between all recognized
Anaplasma species (Visser et al., 1992).
Even though MSP3 may not be an ideal test antigen, the
results of these experiments may provide valuable information
regarding the use of this antigen in a subunit vaccine. It is
not known if response to any of the MSP3 antigens provides
protective immunity to cattle. However, the demonstration of
multiple 86 kDa proteins, and antigenic variation between
these proteins, would indicate that multiple expressed copies
of MSP3 may need to be evaluated for potential use in vaccine
trials. The need for the organism to antigenically vary this
surface protein suggests it serves an important function for
survival in the host. Immune response to an area conserved
between all expressed copies may prove beneficial in
neutralizing infectivity.

114
The information presented in this study may also be
valuable in studying antigenic variation of A. margnale in
persistently infected, carrier cattle. Copy-specific epitopes
on an MSP3 molecule could be defined, and variations in these
epitopes could be monitored in cyclic rickettsemias of carrier
cattle. This would provide much needed information regarding
the mechanisms by which this organism, and possibly other
rickettsial agents, evade the host immune system. Basic
information regarding the means by which organisms adapt to
their host helps establish better ways to diagnose, prevent,
and eventually eradicate these diseases.

REFERENCE LIST
1. Alleman, A.R., S.M. Hamper, N. Viseshakul, and A.F.
Barbet. 1993. Analysis of the Anaplasma margnale genome
by pulsed-field electrophoresis. J. Gen. Microbiol.,
139:2439-2444.
2. Allred, D.R., T.C. McGuire, G.H. Palmer, S.R. Leib, T.M.
Harkins, T.F. McElwain, and A.F. Barbet. 1990. Molecular
basis for surface antigen size polymorphisms and
conservation of a neutralization-sensitive epitope in
Anaplasma margnale. Proc. Natl. Acad. Sci. 87:3220-3224.
3. Amerault, T.E., and T.O. Roby. 1968. A rapid card
agglutination test for bovine anaplasmosis. J. Am. Vet.
Med. Assoc., 153:1828-1834.
4. Amerault, T.E., J.E. Rose, and T.O. Roby. 1973. Modified
card agglutination test for bovine anaplasmosis:
evaluation with serum and plasma from experimental and
natural cases of anaplasmosis. Proceedings of the Annual
Meeting of the U.S. Animal Health Assoc. 76:737.
5. Barbet, A.F. and D.R. Allred. 1991. The msplb multigene
family of Anaplasma margnale: nucelotide sequence
analysis of an expressed copy. Infect. Immun. 59:971-976.
6. Barbet, A.F., L.W. Anderson, G.H. Palmer, and T.C.
McGuire. 1983. Comparison of proteins synthesized by two
different isolates of Anaplasma margnale. Infect. Immun.
40:1068-1074.
7. Barbet, A.F., G.H. Palmer, P.J. Myler, and T.C. McGuire.
1987. Characterization of an immunoprotective protein
complex of Anaplasma margnale by cloning and expression
of the gene coding for polypeptide AM105L. Infect.
Immun. 55:2428-2435.
8. Barbet, A.F., S.M. Semu, N. Chigagure, P.J. Kelly, F.
Jongejan, and S.M. Mahan. 1994. Size variation of the
major immunodomiant protein of Cowdria ruminantium. Clin.
Diagn. Lab. Immunol. 1:744-746.
9. Barbour, A.G. 1990. Antigenic variation of a relapsing
fever Borrela species. Annu. Rev. Microbiol. 44:155-171.
115

116
10. Barbour, A.G. 1991. Molecular biology of antigenic
variation in Lyme borreliosis and relapsing fever: a
comparative analysis. Scand. J. Infect. Dis Suppl.
77:88-93.
11. Barry, D.N., R.J. Parker, A.J. De Vos, P. Dunster, and
B.J. Rodwell. 1986. A microplate enzyme-linked
immunosorbent assay for measuring antibody to Anaplasma
margnale in cattle serum. Aust. Vet. J., 63:76-79.
12. Beutler, E. 1984. The preparation of red cells for assay.
Red Cell Metabolism: A Manual of Biochemical Methods. 3rd
ed. pp 8-19. Edited by Ernest Beutler, MD. Orlando,
Florida: Grue and Stratton, Inc.
13. Dame, J.B., S.M. Mahan, and C.A. Yowell. 1992.
Phylogenetic relationship of Cowdria ruminantium, agent
of heartwater, to Anaplasma margnale and other members
of the order Rickettsiales determined on the basis of 16S
rRNA sequence. Int. J. of Syst. Bacteriol., 42:270-274.
14. Duzgun, A., C.A. Schuntner, I.G. Wright, G. Leatch, and
D.J. Waltishbuhl. 1988. A sensitive ELISA technique for
the diagnosis of Anaplasma margnale infections. Vet.
Parasitol., 29:1-7.
15. Eriks, I.S., G.H. Palmer, T.C. McGuire, D.R. Allred, and
A.F. Barbet. 1989. Detection and quantitation of
Anaplasma margnale in carrier cattle by using a nucleic
acid probe. J. Clin. Microbiol. 27:279-284.
16. Eriks, I.S., D. Stiller, and G.H. Palmer. 1993. Impact of
persistent Anaplasma margnale rickettsemia on tick
infection and transmission. J. Clin. Microbiol. 31:2091-
2096.
17. Goff, W.L., A.F. Barbet, D. Stiller, G.H. Palmer, D.P.
Knowles, K.M. Kocan, J.R. Gorham, and T.C. McGuire. 1988.
Detection of Anaplasma margnale tick vectors by using a
cloned DNA probe. Proc. Natl. Acad. Sci. 85:919-923.
18. Goff, W.L., D. Stiller, R.A. Roeder, L.W. Johnson, D.
Falk, J.R. Gorham, and T.C. McGuire. 1990. Comparison of
a DNA probe, complement fixation, and indirect
immunofluorescence tests for diagnosing Anaplasma
margnale in suspected carrier cattle. Vet. Microbiol.
24:381-390.
19. Goff, W.L., and L.D. Winward. 1985. Detection of
geographic isolates of Anaplasma margnale, using bovine
polyclonal anti-sera and microfluorometry. Am. J. Vet.
Res. 46:2399-2403.

117
20. Gonzalez, E.F., R.F. Long, and R. A. Todorovic. 1978.
Comparisons of the complement-fixation, indirect
fluorescent antibody, and card agglutination tests for
the diagnosis of bovine anaplasmosis. Am. J. Vet. Res.,
39:1538-1541.
21. Goodger, W.J., T. Carpenter, and H. Reimann. 1979.
Estimation of economic loss associated with anaplasmosis
in California beef cattle. J. Am. Vet. Med. Assoc.
174:1333-1335.
22. Harlow, E. and D. Lane. 1988. Immunoassays, In:
Antibodies: A Laboratory Manual. Cold Springs Harbor
Press, Cold Springs Harbor. p. 557-592.
23. Holland, J. and M. Ristic. 1993. Equine monocytic
Ehrlichiosis, p.219-220. In Z. Woldehiwet and M. Ristic
(ed.), Rickettsial and Chlamydias Diseases of Domestic
Animals. Pergamon Press, Inc., Tarrytown, New York.
24. Jain, N.C. 1986. Hemolytic anemias associated with some
infectious agents, p. 590-600. In Jain, N.C. (ed.),
Schalm's Veterinary Hematology. 4th ed. Lea and Febiger,
Philadelphia, PA.
25. Kieser, S.T., I.S Eriks, and G.H. Palmer. 1990. Cyclic
rickettsemia during persistent Anaplasma margnale
infection of cattle. Infect. Immun. 58:1117-1119.
26. Kocan, K.M. Venable, J.H., Hsu, K.C. and W.E. Brock.
1978a. Ultrastructural localization of anaplasmal
antigens (Pawhuska isolate) with ferritin-conjugated
antibody. Am. J. Vet. Res. 39:1131.
27. Kocan, K.M., Venable, J.H. and W.E. Brock. 1978b.
Ultrastructure of anaplasmal inclusions (Pawhuska
isolate) and their appendages in intact and hemolyzed
erythrocytes and in complement-fixation antigen. Am. J.
Vet. Res. 39:1538.
28. Kreier, J.P., and M. Ristic. 1963. Anaplasmosis. KI.
Immunoserologic characteristics of the parasites present
in the blood of calves infected with the Oregon strain of
Anaplasma margnale. Am. J. Vet. Res. 24:688-696.
29. Kuttler, K.L. 1981. Diagnosis of anaplasmosis and
babesiosis an overview, p. 245. In Hidalgo, R.J. &
Jones, E.W. (eds.), Proceedings of the Seventh National
Anaplasmosis Conference. Mississippi State University.

118
30. Kuttler, K.L. and L.D. Winward. 1984. Serologic
comparisons of 4 Anaplasma isolates as measured by the
complement fixation test. Vet. Microbiol. 9:181-186.
31. Levy, M.G., and M. Ristic. 1980. Babesia bovis:
continuous cultivation in a microaerophilus stationary
phase culture. Science. 207:1218-1220.
32. Love, J.N. 1972. Cryogenic preservation of Anaplasma
margnale with dimethylsulf oxide. Am. J. Vet. Res.
33:2557-2560.
33. Luther, D.G., H.U. Cox, and W.O. Nelson. 1980.
Comparisons of Serotests with calf inoculations of
anaplasmosis-vaccinated cattle. Am.J. Vet. Res. 41:2085-
2086.
34. Maas, J., S.D. Lincoln, M.E. Croan, K.L. Kuttler, J.L.
Zaugg, and D. Stiller. 1986. Epidemiologic aspects of
bovine anaplasmosis in semiarid range conditions of south
central Idaho. Am. J. Vet. Res., 47:528-533.
35. Mahan, S.M., T.C. McGuire, S.M. Semu, M.V. Bowie, F.
Jongejan, F.R. Rurangirwa, and A.F. Barbet. 1994.
Molecular cloning of a gene encoding the immunogenic 21
kDa protein of Cowdria ruminantium. Microbiol., 140:2135-
2142.
36. McCallon, B.R. 1973. Prevalence and economic aspects of
anaplasmosis p. 1-3. In E.W. Jones (ed.), Proceedings
of the Sixth National Anaplasmosis Conference. Heritage
Press, Stillwater, Oklahoma.
37. McGuire, T.C., G.H. Palmer, W.L. Goff, M.I. Johnson, and
W.C. Davis. 1984. Common and isolate restricted antigens
of Anaplasma margnale detected with monoclonal
antibodies. Infect. Immun. 45:697-700.
38. McGuire, T.C., W.C. Davis, A.L. Brassfield, T.F.
McElwain, and G.H. Palmer. 1991. Identification of
Anaplasma margnale long-term carrier cattle by detection
of serum antibody to isolated MSP-3. J. Clin. Microbiol.
29:788-793.
39. Meyer, T.F., C.P. Gibbs, and R. Haas. 1990. Variation and
control of protein expression in Neisseria. Annu. Rev.
Microbiol. 44:451-471.
40. Montenegro-James, S., A.T. Guillen, S.J. Ma, P. Tapang,
A. Abel-Gawad, M. Toro, and M. Ristic. 1990. Use of the
dot enzyme-enzyme linked immunosorbent assay with
isolated Anaplasma marginale initial bodies for

119
serodiagnosis of anaplasmosis in cattle. Am. J. Vet. Res.
51(10):1518-1521.
41. Nakamura, Y., S. Shimizu, T. Minami, and S. Ito. 1988.
Enzyme-linked immunosorbent assay using solubilized
antigen for detection of antibody to Anaplasma margnale.
Trop. Anim. Hlth. Prod. 20:259-266.
42. Ndung'u, L.W., C. Aguirre, F.R. Rurangirwa, T.F.
McElwain, T.C. McGuire, D.P. Knowles, and G.H. Palmer.
1995. Detection of Anaplasma ovis infection in goats by
major surface protein 5 competitive inhibition enzyme-
linked immunosorbent assay. J. Clin. Microbiol. 33:675-
679.
43. Oaks, E.V., C.K. Stover, and R.M. Rice. 1987. Molecular
cloning and expression of Rickettsia tsutsugamishi genes
for two major protein antigens in E. coli. Infect. Immun.
55:2428-2435.
44. Oberle, S.M., G.H. Palmer, A.F. Barbet, and T.C. McGuire.
1988. Molecular size variations in an immunoprotective
protein complex among isolates of A. marginale, Infect.
Immun. 56:1567-1573.
45. Palmer, G.H., and T.C. McGuire. 1984. Immune serum
against Anaplasma marginale initial bodies neutralizes
infectivity for cattle. J. Immunol. 133:1010-1015.
46. Palmer, G.H., A.F. Barbet, K.L. Kuttler, and T.C.
McGuire. 1986. Detection of Anaplasma marginale common
surface proteins present in all stages of infection. J.
Clin. Microbiol. 23:1078-1083.
47. Palmer, G.H., A.F. Barbet, A.J. Musoke, J.M. Katende, F.
Rurangirwa, V. Shkap, E. Pipano, W.C. Davis, and T.C.
McGuire. 1988a. Recognition of conserved surface protein
epitopes on Anaplasma cntrale and Anaplasma marginale
isolates from Israel, Kenya, and the United States. Int.
J. Parasitol. 18:33-38.
48. Palmer, G.H., S.M. Oberle, A.F. barbet, W.L. Goff, W.C.
Davis, and T.C. McGuire. 1988b. Immunization of cattle
with a 36-kilodalton surface protein induces protection
against homologous and heterologous Anaplasma marginale
challenge. Infect. Immun. 56:1526-1531.
49. Palmer, G.H., G. Eid, A.F. Barbet, T.C. McGuire, and T.F.
McElwain. 1994. The immunoprotective Anaplasma marginale
major surface protein 2 is encoded by a polymorphic
multigene family. Infect. Immun. 62:3808-3816.

120
50. Richey, E.J. 1981. Bovine anaplasmosis, p. 767-772. In
R.J. Howard (ed.), Current Veterinary Therapy: Food
Animal Practice. The Sanders Co., Philadelphia, PA.
51. Ristic, M and J. Holland. 1993. Canine Ehrlichiosis,
p.172. In Z. Woldehiwet and M. Ristic (ed.), Rickettsial
and Chlamydias Diseases of Domestic Animals. Pergamon
Press, Inc., Tarrytown, New York.
52. Ristic, M. and J.P. Krier. 1974. Family Anaplasmataceae,
p. 906 In: Buchanan, R.E. and N.E. Gibbons (eds.),
Manual of Determinative Bacteriology. Williams and
Wilkins, Baltimore, MD.
53. Schuntner, C.A. and G. Leatch. 1988. Radioimmunoassay for
Anaplasma margnale antibodies in cattle. Am. J. Vet.
Res. 49 (4) :504-507.
54. Shkap, V., H. Bin, H. Ungar-Waron, and E. Pipano. 1990.
An enzyme-linked immunosorbent assay (ELISA) for the
detection of antibodies to Anaplasma margnale and
Anaplasma cntrale. Vet. Microbiol. 25:45-53.
55. Shkap, V., E. Pipano, T.C. McGuire, and G.H. Palmer.
1991. Identification of immunodominant polypeptides
common between Anaplasma cntrale and Anaplasma
margnale. Vet. Immunol. Immunopathol. 29:31-40.
56. Smith, T. and F.L. Kilborne. 1893. Investigations into
the nature, causation, and prevention of Texas or
southern cattle fever. U.S. Dept. Agr., Bur. Animal Ind.
Bull. 1:1-301.
57. Sparling, P.F., J. Tsai, and C.N. Cornelissen. 1990.
Gonococci are survivors. Scand. J. Infect. Dis, Suppl.
69:125-136.
58. Swift, B.L. and G.M. Thomas. 1983. Bovine anaplasmosis:
Elimination of the carrier state with injectable long-
acting oxytetracycline. J. Am. Vet. Med. Assoc. 183:63-
65.
59. Theiler, A. 1910. Anaplasma margnale. The marginal
points in the blood of cattle suffering from a specific
disease, p. 6-64. In: Theilre, A. (ed.), Report of the
Government Veterinary Bacteriologist 1908-1909 Transvaal
Department of Agriculture, Transvaal, South Africa.
60. Tizard, I., 1992. Serology: The detection and measurement
of antibodies. In: Veterinary Immunology. 4th edition,
W.B. Saunders Co., Philadelphia, PA, p 214-236.

121
61. Todorovic, R.A. R.F. Long, and B.R. McCallon. 1977.
Comparison of rapid card agglutination test with
complement fixation test for diagnosis of Anaplasma
margnale infection in Colombian cattle. Vet. Microbiol.
2 :167 .
62. Trueblood, E.S., T.C. McGuire, and G.H. Palmer. 1991.
Detection of Anaplasma margnale rickettsemia prior to
onset of clinical signs by using an antigen capture
enzyme-linked immunosorbent assay. J. Clin. Microbiol.
29(7):1542-1544.
63. Uilenberg, G. 1993. Other Ehrlichiosis of ruminants,
p.270. In Z. Woldehiwet and M. Ristic (ed.), Rickettsial
and Chlamydias Diseases of Domestic Animals. Pergamon
Press, Inc., Tarrytown, New York.
64. Van Vliet, A.H.M., F. Jongejan, and B.A.M. Van Der
Zeijst. 1992. Phylogenetic position of Cowdria
ruminantium (Rickettsiales) determined by analysis of
amplified 16S ribosomal DNA sequences. Int. J. of Syst.
Bacteriol., 42:494-498.
65. Van Vliet, A.H.M., F. Jongejan, M. Van Kleef, and B.A.M.
Van Der Zeijst. 1994. Molecular cloning, sequence
analysis, and expression of the gene encoding the
immunodominant 32-kilodalton protein of Cowdria
ruminantium. Infect. Immun., 62:1451-1456.
66. Viseshakul, N. S.M. Kamper, and A.F. Barbet. 1994.
Organization, structure, and expression of the MSP1/3 gene
family of Anaplasma marginale. abstr. 35. Abstr. 75th
Annu. Meet. Conf. Res. Workers An. Dis. 1994.
67. Visser, E.S., T.C. McGuire, G.H. Palmer, W.C. Davis, V.
Shkap, E. Pipano, and D.P. Knowles. 1992. The Anaplasma
marginale msp5 gene encodes a 19-kilodalton protein
conserved in all recognized Anaplasma species. Infect.
Immun., 60:5139-5144.
68. Wanduragala, L. and M. Ristic. 1993. Anaplasmosis, p.
65-74. In Z. Woldehiwet and M. Ristic (ed.), Rickettsial
and Chlamydias Diseases of Domestic Animals. Pergamon
Press, Inc., Tarrytown, New York.
69. Winkler, G.C., G.M. Brown and H. Lutz. 1987. Detection of
antibodies to Anaplasma marginale by an improved enzyme-
linked immunosorbent assay with sodium dodecyl sulfate-
disrupted antigen. J. Clin. Microbiol. 25(4):633-636.
70. Zaugg, J.L., D. Stiller, M.E. Croan, and S.D. Lincoln.
1986. Transmission of Anaplasma marginale Theiler by

122
males of Dermacentor andersoni stiles fed on an Idaho
field infected, chronic carrier cow. Am. J. Vet. Res.
47:2269-2271.

BIOGRAPHICAL SKETCH
A. Rick Alleman was born in New Orleans, LA, on August
23, 1954. He is married to Mary Alleman, and they have 2
children, Arthur Rick Jr. and Grace Elizabeth.
Rick completed his undergraduate studies at the
University of New Orleans in 1976. He then attended Louisiana
State University, School of Veterinary Medicine in Baton Rouge
where he obtained his Doctor of Veterinary Medicine degree in
1980. At this time he practiced small animal medicine in New
Orleans, LA, with his partner Mariano Guas. In 1986 Rick
became board certified by the American Board of Veterinary
Practitioners (ABVP) as a specialist in companion animal
medicine.
In 1989 he left private practice and, under the guidance
of Drs. Rose Raskin and John Harvey, joined the University of
Florida College of Veterinary Medicine to do a residency in
clinical pathology. In 1992 he completed his residency
training and was board certified by the American College of
Veterinary Pathologists (ACVP), specialty Clinical Pathology.
In 1990, under the guidance of Dr. Anthony Barbet, Rick
entered into a Ph.D. graduate program in the Department of
Pathobiology. During his doctoral research program he
developed a keen interest in the study of rickettsial agents
123

124
at the molecular level. He plans to pursue a career in
academia where he will continue his laboratory investigation
of rickettsial agents as well as provide instructional
services to students and clinical services to the teaching
hospital.

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.

John B. Dame, Chair
Associate Professor of
Vterinary Medicine
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Anthony F. Barbet
Professor of Veterinary
Medicine
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
John W.\ Harv£
Professor cyt Veterihary
Medicine (.
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Rose E. Raskin
Associate Professor of
Veterinary Medicine
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in acope and quality, as
a dissertation for the degree of Doctor of; Philpsophy.
Paul Gulig
Associate Professor
Molecular Genetics
Microbiology

This dissertation was submitted to the Graduate Faculty
of the College of Veterinary Medicine and to the Graduate
School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Pidioso}
August, 1995
Dean ^College of Veterinary
Medicine
Dean, Graduate School

UNIVERSITY OF FLORIDA
3 1262 08554 8179



18
Another diagnostic test utilized the CF test antigen in
particulate and SDS-solubilized form as a test antigen in an
ELISA (Winkler et al., 1987). This study indicated SDS-
solubilization of the proteins decreased background reactivity
in the test and increased test sensitivity in detecting
positive reactants. Complete correlation between the ELISA
and CF test was found when using solubilized test antigen and
CF test positive and negative reference sera. In testing
other infected cattle, false negatives were not observed;
however, cross reactivity ranged from 50% to 80% when testing
sera from cattle infected with or immunized against different
infectious agents.
Nakamura et al., 1988, developed an ELISA test antigen by
nitrogen decompression of infected cells to isolate initial
bodies, and solubilized them in triton X-100. Test
specificity and sensitivity was determined by comparison of
results with the CF test. Specificity of this ELISA was
satisfactory having 100% agreement with sera tested negative
by CF and no cross reactivity with serum from animals infected
with Babesia sp., Theileria sp., or Eperythrozoon sp. Cross
reactivity was detected with A. cntrale. Cross reactivity
with other rickettsial agents such as Ehrlichia sp. or Cowdria
ruminantium was not determined. Sensitivity of this test was
not adequately determined since the results were compared to
a test (CF test) which itself has poor sensitivity,
particularly in detecting carrier cattle. No effort was made


ACKNOWLEDGMENTS
I would like to acknowledge my committee members, Dr.
John Dame, Dr. John Harvey, Dr. Paul Gulig, and Dr. Rose
Raskin, for their guidance and assistance in the completion of
this project. I would like to express a special appreciation
to my major professor, Dr. Tony Barbet. He has been a great
instructor, mentor, guidance counselor, and friend. He has
always been there for me as well as for his other graduate
students and staff. His devotion of his time in a completely
unselfish manner has been an inspiration to me on many
occasions, particularly in difficult times. His weekly
individual meetings, and monthly lab meetings were a lesson
well learned and something I hope to continue during my
professional career. His guidance was most valuable to the
completion of these experiments.
I would also like to express my appreciation to some of
my coworkers, Dr. Bill Whitmire, Annie Moreland, Renee
Blentlinger, Anna Lundgren, and Michael Bowie. We all worked
together harmoniously in the lab, each willing to take time
out to assist the other, sharing valuable work experiences.
I am glad friendships last longer than Ph.D. projects.
IV


45
probe was detected by ECL as before. The membranes
exposed to Hyperfilm MP (Amersham International,
Buckinghamshire, England) to visualize bound antibody.
were
pic,


32
bromophenol blue. Focused proteins were then electrophoresed
on 7.5% to 17.5% (w/v) gradient polyacrylamide gels and
treated and transferred to nitrocellulose membranes as
described above.
Immunoblots with Antisera
Nitrocellulose membranes containing transferred proteins
were blocked with 5% milk (w/v) in PBS with 0.25%
polyoxyethylene-sorbitan monolaurate (Tween 20) to inhibit
non-specific binding of primary and secondary antibodies. The
membranes were washed with 1% milk (w/v) in PBS with 0.25%
Tween 20, and probed with antisera from animals infected with
one of the following organisms; A. margnale, A. cntrale, A.
ovis, B. bovis, B. bigemina, C. ruminantium, E. equi, E.
risticii, or E. ewingii. Normal sera from respective
uninfected species were used as negative controls. Serum
dilutions (in PBS with 1% milk and 0.25% Tween 20) of 1/100 or
greater were used. Rabbit-anti-MSP3 polyclonal sera was used
at a dilution of 1/5,000. Normal rabbit sera was used at the
same dilution as a negative control. Anti-MSP3, anti-MSP2,
and negative control anti-trypanosome MAbs were used in
concentrations of 5 /^g/ml. The membranes were again washed
with 1% milk (w/v) in PBS with 0.25% Tween 20 and probed with
either species-specific anti-IgG-Horseradish peroxidase (HRP)
conjugated antibody at a dilution of 1/2,000 (Sigma Immuno
Chemicals, St. Louis, MO), or HRP-conjugated Protein G at a


ABBREVIATIONS
bp
BSA
C
CD
CF
CHEF
CT
DNA
ECL
EDTA
ELISA
Fig.
FL
g
HC1
HRP
I FA
IIF
IPTG
kbp
kDa
base pair (s)
bovine serum albumin
degrees Celsius
card agglutination test
complement fixation test
clamped homogeneous electric field electrophoresis
capillary tube agglutination test
deoxyribonucleic acid
enhanced chemiluminescence
ethylenediaminetetraacetic acid
enzyme-linked immunosorbent assay
figure (s)
Florida
gram (s)
hydrochloric acid
horseradish peroxidase
indirect fluorescent antibody
indirect immunofluorescence test
isopropylthio-/3-galactosidase
kilobase pair (s)
kilodalton
viii


Fig. 12 Diagram of MSP3 clones with MSP2 homology.
A schematic representation of clones MSP3-11, MSP3-12, and MSP3-19. Homologous
regions are indicated by like shaded areas. Regions homologous to the MSP2 gene
are indicated by MSP2->. Arrows indicate the direction of the reading frame.
Nucleotide numbers are indicated on the bottom. Top numbers of clones MSP3-11 and
MSP3-19 indicate corresponding nucleotides in clone MSP3-12.


M
molar
MAb
monoclonal antibody (s)
MAPI
major antigenic protein 1
mg
milligram (s)
min.
minute (s)
ml
milliliter (s)
mM
millimolar
MSPla
major surface protein la
MSP1/3
major surface protein 1/3
MSP2
major surface protein 2
MSP3
major surface protein 3
N
normal
NaCl
sodium chloride
NaOH
sodium hydroxide
ng
nanogram (s)
NP-40
nonidet P-40
PBS
phosphate buffered saline
PCV
packed cell volume
PI
post-infection
PMSF
phenylmethylsufonly fluoride
RFLP
restriction fragment length polymorphism
RIA
radioimmunoassay
rRNA
ribosomal ribonucleic acid
SDS-PAGE
sodium dodecyl sulfate, polyacrylamide gel
electrophoresis
sec.
seconds
ix


110
in multiples of 3 are expressed while those with any other
multiples are not produced (Meyer et al.f 1990). Variations
in the number of 5-mer repeats occurs frequently during DNA
replication. Antigenic variation allows N. gonorrhoeae to
persist in the host unless appropriate antibiotic therapy is
instituted, and produce repeated infections in the same host
(Sparling et al., 1990)
The exact mechanism by which A. margnale uses multigene
families is not known. However, like Borrelia hermsii and
Neisseria gonorrhoeae, A. marginale does appear to evade the
host immune system and avoid complete clearance. In addition,
as in relapsing fever, fluctuating parasitemia is observed in
cycles consistent with the appearance of antigenic variants in
response to immune pressure from the host (Krieser et al.,
1990). We hypothesize A. marginale could use mechanisms
similar to those seen in B. hermsii and N. gonorrhoeae to vary
important antigenic surface proteins such as MSP2 and MSP3 in
an effort to avoid immune clearance.
Distribution of MSP3 Copies in the A. marginale Genome
Previously, we have shown that by Sfi I digestion and
separation of large fragments using CHEF electrophoresis, we
can separate the entire genome of A. marginale (FL) into 14
fragments ranging in size from 160 to 14 kbp (Alleman et al.,
1993) The MSP3-12 probe hybridizes to 6 of these fragments,
as well as multiple fragments in Not 1 and Sfi I digests of


Restriction Enzyme Digestion in Agarose
Plugs 43
CHEF Gel Electrophoresis 44
CHAPTER 4 RESULTS 46
Specificity Experiments 46
Conservation of MSP3 Between Different
Geographic Isolates of A. marginale 53
Immune Response to MSP3 56
MSP3 Nucleotide Sequence and Translation 62
MAb and Immune Sera Reactivity to Recombinant
MSP3 74
Representation of pBluescript MSP3-12 in the
A. marginale Genome 79
Presence of Multiple MSP3 Gene Copies 79
Distribution of MSP3 Copies in the A. marginale
Chromosome 84
CHAPTER 5 DISCUSSION 92
Rational for Studying the MSP3 Antigen 92
The Specificity of MSP3 in Detecting A. marginale
Infection 94
Size Polymorphism of MSP3 Among Various
Strains of A. marginale 98
Immune Response to MSP3 100
Identification of pBluescript MSP3-12 as a
Recombinant Form of an 86 kDa Antigen 103
pBluescript MSP3 is an Accurate Representation of
a Genomic Copy of MSP3 104
Multiple MSP3 Copies in the A. marginale Genome 105
Distribution of MSP3 Copies in the A. marginale
Genome 110
Summary and Conclusions Ill
REFERENCE LIST 115
BIOGRAPHICAL SKETCH 123
vi


96
nonpathogenic and carrier cattle are not a threat to livestock
production.
In contrast, serum from sheep infected with A. ovis
showed a strong positive reaction with MSP3 of A. margnale as
well as a 36 kDa antigen, possibly MSP2. In previous
experiments, a panel of 18 MAbs reactive to various A.
margnale proteins failed to react with A. ovis infected
erythrocytes (McGuire et al., 1984). The difference in
results may be explained by the limited number of epitopes
recognized by these MAbs. Our results confirm cross
reactivity exists between at least 2 of the major antigens of
A. margnale and serum from animals infected with A. ovis.
These results indicate a diagnostic test using MSP3 of A.
margnale may be able to detect A. ovis infected sheep. The
cross reactivity between these two species should not present
a problem in cattle since A. ovis fails to infect either
intact or splenectomized animals (Jain, 1986).
The Ehrlichia sp. and C. ruminantium have been shown to
be phylogenetically closely related to Anaplasma marginale
(Dame et al., 1992; Van Vliet et al., 1992). Our results
confirm cross reactive proteins do exist between these species
when sera from animals infected with any of these organisms is
reacted with initial body preparations from a Florida isolate
of A. marginale. Only sera from animals infected with E.
risticii and E. ewingii react with the MSP3 antigen. These
species do not infect cattle. However, the results illustrate


15
erythrocytes are then visualized using a fluorescein-
conjugated anti-bovine, rabbit immunoglobulin (Gonzalez et
al., 1978). Sensitivity in detecting subclinically infected
animals is reported to be 97% with lower limits of sensitivity
not being reached by 18 weeks PI. In another study involving
64 cattle naturally infected with A. margnale and confirmed
by DNA probe hybridization, 94% of cattle were positively
identified by IIF (Goff et al., 1990). In this study
circulating organisms were not seen in stained blood smears,
but length of infection was undetermined.
Although sensitivity with IIF is much improved over CA
and CF tests, the specificity is somewhat reduced with 10% of
normal, non-infected cattle testing positive by IIF (Gonzalez
et al., 1978). In addition, cross reactivity in cattle
infected with related organisms may also present a
problem. The sensitivity of this test is satisfactory,
however, its questionable specificity, labor intensive nature,
and need for specialized reagents and equipment make it
undesirable as a field test for routine diagnosis and
identification of A. margnale infected carriers.
The Radioimmunoassay CRIA)
Recently a radioimmunoassay (RIA) was developed for the
detection of A. margnale antibodies in sera of infected
cattle (Schuntner & Leatch, 1988). This test initially
demonstrated high specificity and sensitivity (98.8% for each)


28
washed 5 times in PBS with each centrifugation at 16,000 x g
for 25 min. at 4C. After each centrifugation, an upper layer
containing both leukocytes and erythrocytes was removed. The
pellets were resuspended in PBS, sonicated for 2 min. on ice
at 50 W and centrifuged as before. Pelleted material was
again resuspended in PBS, and sonicated for 30 sec. on ice at
50 W and centrifuged a final time. Intact initial bodies were
visualized by Wright-Giemsa stain. The pellets of initial
bodies were resuspended in equal volumes of PBS for use in
SDS-PAGE.
Initial bodies used in 2-D gel electrophoresis were
resuspended in equal volumes of lysis buffer containing 9.5 M
urea, 2% nonidet P-40, 1.6% Ampholyte 5/7 (Bio-Lyte 5/7, Bio-
Rad Laboratories Richmond, CA) 0.4% ampholyte 3/10 (Bio-Lyte
3/7, Bio-Rad Laboratories, Richmond, CA) and 5.0% j8-
mercaptoethanol.
Protein concentrations were determined
spectrophotometrically using the Micro BCA Protein Assay
(Pierce, Rockford, Illinois). Initial body preparations were
stored in small aliquots at -70C. Anaplasma cntrale initial
bodies were also prepared as described above.
Babesia bovis and Babesia biaemina Antigen Preparation
Babesia bovis and B. bigemina antigens were prepared from
organisms maintained in microaerophilic stationary phase
culture as previously described (Levy and Ristic, 1980).